World Journal of Microbiology and Biotechnology

World Journal of Microbiology and Biotechnology Volume 21, Number 3 Production of single cell protein through fermentation of a perennial grass grown on saline lands with Cellulomonas biazotea

(207 - 211)

M. Ibrahim Rajoka DOI: 10.1007/s11274-004-2889-6 Chromate reduction by Bacillus megaterium TKW3 isolated from marine sediments

(213 - 219)

K. H. Cheung and Ji-Dong Gu DOI: 10.1007/s11274-004-3619-9 Optimization of redox reactions employing whole cell biocatalysis

(221 - 227)

Abhishek A. Chakraborty, Ravindra P. Phadke, Fauzia A. Chaudhary, Prakash S. Shete, Bhalchandra S. Rao, Kushan D. Jasani DOI: 10.1007/s11274-004-3620-3 Formation of Yersinia pseudotuberculosis biofilms on multiple surfaces on Caenorhabditis elegans

(229 - 231)

Marc A. Nascarella and Steven M. Presley DOI: 10.1007/s11274-004-5299-x Xylanase production under solid-state fermentation and its characterization by an isolated strain of Aspergillus foetidus in India

(233 - 243)

Amita R Shah and Datta Madamwar DOI: 10.1007/s11274-004-3622-1 Use of RAPD to investigate the epidemiology of Staphylococcus aureus infection in Malaysian hospitals

(245 - 251)

V. Neela, N. S. Mariana, S. Radu, S. Zamberi, A. R. Raha, R. Rosli DOI: 10.1007/s11274-004-3624-z Isolation of Enterobacteria able to degrade simple aromatic compounds from the wastewater from olive oil extraction

(253 - 259)

Emna Ammar, Moncef Nasri, Khaled Medhioub DOI: 10.1007/s11274-004-3625-y Effects of Zn supplementation on the growth, amino acid composition, polysaccharide yields and anti-tumour activity of Agaricus brasiliensis

(261 - 264)

Xiang Zou DOI: 10.1007/s11274-004-2614-5 Isolation and characterization of a new carbendazimdegrading Ralstonia sp. strain

(265 - 269)

Gui-Shan Zhang, Xiao-Ming Jia, Tian-Fan Cheng, Xiao-Hang Ma, Yu-Hua Zhao DOI: 10.1007/s11274-004-3628-8 Survey of plasmid profiles of Shigella species isolated in Malaysia during 1994–2000

(271 - 278)

C. H. Hoe, R. M. Yassin, Y. T. Koh, K. L. Thong DOI: 10.1007/s11274-004-3631-0 Physico-chemical characterization of exomannan from Rhodotorula acheniorum MC

(279 - 283)

K. Pavlova, I. Panchev, Ts. Hristozova DOI: 10.1007/s11274-004-3632-z Isolation and characterization of Bacillus thuringiensis strains from different grain habitats in Turkey

(285 - 292)

Özgür Apaydin, A. Fazil Yenidünya, Şebnem Harsa, Hatice Güneş DOI: 10.1007/s11274-004-3633-y Preliminary studies on chorote – a traditional Mexican fermented product

(293 - 296)

Marisol Castillo-Morales, María del Carmen Wacher-Rodarte, Humberto HernándezSánchez DOI: 10.1007/s11274-004-3634-x Biodegradation of pentachorophenol by tropical basidiomycetes in soils contaminated with industrial residues

(297 - 301)

Kátia Maria Gomes Machado, Dácio Roberto Matheus, Regina Teresa Rosim Monteiro, Vera Lúcia Ramos Bononi DOI: 10.1007/s11274-004-3693-z Solid-state fermentation of wood residues by Streptomyces griseus B1, a soil isolate, and solubilization of lignins

(303 - 308)

Anju Arora, Lata Nain, J. K. Gupta DOI: 10.1007/s11274-004-3827-3 Detecting the heavy metal tolerance level in ectomycorrhizal fungi in vitro

(309 - 315)

Prasun Ray, Richa Tiwari, U. Gangi Reddy, Alok Adholeya DOI: 10.1007/s11274-004-3572-7 Potential of Rhodococcus erythropolis as a bioremediation organism

(317 - 321)

Alena Čejková, Jan Masák, Vladimír Jirků, Martin Veselý, Miroslav Pátek, Jan Nešvera DOI: 10.1007/s11274-004-2152-1 Production of the ligninolytic enzymes by immobilized Phanerochaete chrysosporium in an air atmosphere

(323 - 327)

Guoce Yu, Xianghua Wen, Yi Qian DOI: 10.1007/s11274-004-3571-8 Use of cereals as basal medium for the formulation of alternative culture media for fungi

(329 - 336)

A. O. Adesemoye and C. O. Adedire DOI: 10.1007/s11274-004-3907-4 Investigation of the active site of the extracellular β-Dglucosidase from Aspergillus carbonarius

(337 - 343)

Szilvia Jäger and László Kiss DOI: 10.1007/s11274-004-2609-2 Effect of frozen storage and aging on the Kashkaval cheese starter culture

(345 - 350)

Zh. I Simov and G. Y. Ivanov Characteristics of the bacteriocin produced by Lactococcus lactis subsp. cremoris CTC 204 and the effect of this compound on the mesophilic bacteria associated with raw beef R. Bromberg, I. Moreno, R. R. Delboni, H. C. Cintra, P. T. V Oliveira DOI: 10.1007/s11274-004-2610-9

(351 - 358)

The effect of salinity on trichloroethylene co-metabolism by mixed cultures enriched on phenol

(359 - 365)

Chi-Yuan Lee, Yu-Chia Chan, Chin-Lung Lin DOI: 10.1007/s11274-004-2611-8 Microbial decolorization of reactive azo dyes under aerobic conditions

(367 - 370)

K. M. Kodam, I. Soojhawon, P. D. Lokhande, K. R. Gawai DOI: 10.1007/s11274-004-5957-z Rapid method for the affinity purification of thermostable α-amylase from Bacillus licheniformis M. Damodara Rao, B. V. V. Ratnam, Dasari VenkataRamesh, C. Ayyanna DOI: 10.1007/s11274-004-3908-3

(371 - 375)


Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 207–211 DOI 10.1007/s11274-004-2889-6

Production of single cell protein through fermentation of a perennial grass grown on saline lands with Cellulomonas biazotea M. Ibrahim Rajoka National Institute for Biotechnology and Genetic Engineering, P.O. Box 577, Faisalabad, Pakistan (Tel.:+92-41-651475, +92-041-550815, Fax: +92-41-651472, E-mail: [email protected]) Received 12 May 2004; accepted 30 June 2004

Keywords: Cellulomonas biazotea, growth of bacteria, kinetics of production, Leptochloas fusca

Summary Microbial protein from alkali-treated Leptochloa fusca (kaller grass) was produced by growing Cellulomonas biazotea in shake flasks and in an aerated 6-l fermentor. Single cell protein, produced in the fermentor contained 56.10 ± 4.64, 60.00 ± 5.04, 11.50 ± 1.34, 12.95 ± 1.24, 3.50 ± 0.24 and 1.00 ± 0.44% true protein, crude protein, crude fibre, ash, cellulose and RNA content respectively. Maximum values compared favourably with published data. The biomass contained all desired amino acids with isoleucine as limiting acid. The dried biomass showed a gross metabolizable energy value of 3500 kcal kg)1 and indicated that it might serve as energy as well as a protein source particularly when fed to poultry.

Nomenclature X S Qx Qs YX/S qs l QCP YCP/X SCP DO v/v/m qCP LC

cell mass (g l)1) substrate (g l)1) rate of cell mass formation (g cells l)1 h)1) rate of substrate consumption (g l)1 h)1) cell yield coefficient (g cells g)1 substrate utilized) specific rate of substrate consumption (g g)1cells h)1) specific growth rate (h)1) rate of crude protein formation (g l)1 h)1) specific yield of crude protein (g crude protein g)1 cells) single cell protein: Leptochloa fusca (kallar grass) grown cell mass dissolved oxygen (%) air flow rate (volume of air per volume of fermentation medium per min.) specific rate of crude protein productivity (g g)1 cells h)1) lignocellulosic

Introduction Single cell protein (SCP) has attracted commercial interest due to the possible substitution of microbial protein for the conventional protein supplements currently being used in the dairy and poultry industries (Rosenberg 1993; Chanda & Chakrabarti 1996). Microorganisms have the ability to upgrade low protein plant

material to high protein feed. Large-scale fermentation of methanol, starch and molasses-based media have proved economically viable for the production of animal feed and human food (Rosenberg 1993; Hongpattarakere & H-Kittikun 1995; Paul et al. 2002). Production of SCP from lignocellulosic (LC) agricultural biomass is of immense importance to get high quality protein in developing countries (Rajoka 1990; Bajpai & Bajpai 1991). The product can be fed to poultry and livestock as cheaper rations for the production of eggs, milk and meat (Rosenberg 1993). Moreover, SCP will save cereals for human consumption in third world countries (Hongpattarakere & H-Kittikun 1995). At the moment, cereals are being used in livestock and poultry feeds. Efforts are still being made to develop a process to produce low cost SCP through fermentation of abundantly available agro-industrial wastes (Rosenberg 1993; Chanda & Chakrabarti 1996). Many workers have used pure Cellulomonas strains or mixed culture of bacteria for production of SCP from cellulose-based raw materials (Hitchner & Leatherwood 1980; Enriquez & Rodriquez 1983; Rodriguez-Vesquez et al. 1992; Pece et al. 1994). Leptochloa fusca (kallar grass) can be inexpensively raised on highly saline lands (one-third of cultivable land in Pakistan) without application of fertilizer and can produce biomass (up to 50 metric tonnes ha)1 year)1) for biotechnical applications (Latif et al. 1994; Rajoka & Malik 1997). The cost of pre-treatment is also very low ($0.03 kg)1). C. biazotea effectively utilized this substrate and produced more cell mass than that produced on wheat straw, bagasse, Sesbania aculeate, Panicum maximum, cotton

208 stalks and Atriplex lentiformis (Rajoka & Malik 1997). SCP produced on kallar grass is expected to possess a high nutritive value as reported earlier (Enriquez & Rodriquez 1983; Rajoka 1990; Zayed & Mostafa 1992; Pece et al. 1994). Cell mass and product formation kinetic parameters need to be studied for scaling-up to large bioreactors (Tobajas & Garcia-Calvo 1999). Use of kallar grass for enzyme production has already been studied (Rajoka & Malik 1997; Rajoka et al. 1997) but no detailed information is available on its use for production of single cell protein. Moreover, information on the detailed kinetics of SCP production from LC substrates is not available. In this work, it was planned to study the kinetics of SCP production in greater detail by culturing C. biazotea on alkali-pre-treated kallar grass which contained 6.5 ± 0.45, 11.82 ± 1.04, 57.37 ± 4.13, 55 ± 4.54 and 23.00 ± 2.00% crude protein, crude fibre, nitrogen-free extract, cellulose and hemicellulose respectively and evaluate it for crude protein, RNA, metabolizable energy and amino acid content, so as to work out the suitability of the single cell protein product as a poultry and livestock feed supplement.

Materials and methods

M. Ibrahim Rajoka carbon and energy source was examined in optimized salt medium containing (g l)1): KH2PO4 1.0, NaNO3 4.0 (instead of 0.5), KCl 1.0, MgSO4 0.5, FeSO4 0.1; yeast extract 2.0 and alkali-treated straw (1.5 g total carbohydrates). All media were adjusted to pH 7.3 with 1 M NaOH or 1 M HCl. The data of batch fermentation were gained by performing experiments (three runs) on above media under optimized conditions in shake flasks (Rajoka et al. 1997). Sample flasks in triplicate were withdrawn periodically to follow the assay of dry cell mass, crude protein, true protein and RNA. Fermentor studies Fermentation was carried out in 6-l Eyela fermentor (Model M-160 Eyela, Tokyo, Japan) fitted with automatic pH, temperature, dissolved oxygen tension (DOT), agitation and airflow rate controls. The optimized medium (4 l) was steam sterilized in an autoclave. The medium was inoculated with seed culture (10% v/v inoculum) prepared as above. Standard temperature, agitation speed, airflow rate and pH were 30 C, 400 rev min)1, 2 v/v/m and 7.3 respectively. Silicone oil was used as an antifoaming agent. Samples in triplicate were collected periodically to follow the assay of dry cell mass, crude protein, true protein and RNA.

Organism Analytical methods Cellulomonas biazotea NIAB 442 was maintained on Dubos-Sigmacell 100-agar slants. The inoculum was prepared by transferring a loopful of cells to 50 ml seed culture medium containing (g l)1) KH2 PO4 1.0, NaNO3 0.5, KCl 1.0, MgSO4, 0.5, yeast extract 0.2 (pH 7.0 ± 0.1) (Rajoka & Malik 1997) and grown at 30 C on an orbital shaker (150 rev min)1 for 24 h). Dry cell mass was estimated using the pre-determined conversion factor of 1 g dry cell weight per l for unit absorbance at 610 nm. The conversion factor was calculated from the relationship between absorbance and dry cell mass (g per l). Concentration of the culture was adjusted to contain 2.5 ± 0.14 g dry cells l)1 of the inoculum medium. Substrate Leptochloa fusca (kallar grass) was collected from Lahore Biosaline Research Substation of Nuclear Institute for Agriculture and Biology (NIAB), Faisalabad, where it is grown as an energy crop from highly saline waste lands in Pakistan (Latif et al. 1994). It contained 55 ± 1.2 and 23 ± 2.4% cellulose and hemicellulose respectively and were determined as described previously (Latif et al. 1994). Shake flask studies The ability of the organism to produce cell mass and protein from pre-treated kallar grass straw as a sole

Samples in triplicate were taken at different time intervals (h) to assay concentration of cell, solid substrate, crude protein, true protein, crude fibre, nitrogen free extract and RNA (Pacheco et al. 1997; Paul et al. 2002). Culture samples (100 ml) were centrifuged (7000 · g at 10 C for 10 min) to remove substrate. The substrate was washed twice with saline and dried to estimate unutilized solid material. The supernatant (100 ml) was also centrifuged (10,000 · g, 10 min). The cell mass was washed twice with saline, suspended in 10 ml distilled water and dried at 70 C. The remaining 100 ml sample, containing cell mass and unutilized substrate was also dried (called dry biomass). It was routinely analysed for crude protein, true protein, and RNA as described previously (Pacheco et al. 1997; Paul et al. 2002). The micro-Kjeldahl nitrogen of dry biomass was multiplied by 6.25 to calculate crude protein. For determination of true protein, 0.5 g homogenized and dry biomass was treated with 20 ml (5% v/v) trichloroacetic acid for 5 min with shaking, placed at 90 C (in an oven) for 15 min and shaken occasionally. It was filtered while hot and the residue was rinsed with hot water three times and dried to constant weight. Its nitrogen content was determined by the micro-Kjeldahl method and true protein was calculated as above. Soluble proteins in fermentation broth were determined by the Lowry method. Total RNA content was determined using orcinol-HCl reagent as described previously (Pacheco et al. 1997; Paul et al.

209

Production of SCP from a grass 2002) and its content was reduced by heat-treatment as mentioned earlier (Abouzeid et al. 1995). The calorific value of dry biomass was determined using the Parr method (Hill & Anderson 1958) with a Parr oxygen bomb calorimeter. The calorific value was calculated by the amount of heat generated by the combustion of a known weight of the sample in the presence of 20 atm oxygen. Moisture, dry matter, ash, ether extract, crude fibre, nitrogen free extract, carbon and cellulose were analysed according to AOAC (1984) methods. Amino acid analysis and chemical score The amino acid profile of the SCP was determined by automatic amino acid analyser (Evans Electroselenium Limited, UK) on an HCl-hydrolysate of SCP. The chemical score of the biomass product was calculated following method of FAO/WHO (1957). Determination of kinetic parameters All kinetic parameters were determined as described earlier (Aiba et al. 1973; Zayed & Mostafa 1992). Volumetric rate of crude and true protein production (QP) was determined from a plot between protein (g l)1) and time of fermentation (h). Process product yield (YP/S) was determined from dP/dS, specific product yield (YP/x, g g)1 cells) was determined using relationship dP/dX, and volumetric rate of substrate consumption was determined from a plot between solid substrate (g l)1) present in the fermentation medium and time of fermentation (h). Cell mass productivity expressed as g dry cells l)1h)1 was determined from a plot of g dry cells l)1 and time of fermentation. Specific growth rate was determined from the relationship lt ¼ ln Xt/Xo while specific productivity was a multiple of l and YP/X.

Table 1. Comparative kinetic parameters of C. biazotea for kallar grass consumption and protein production following growth under optimized cultural conditions in shake flask and 6-l fermentor studies. Kinetic parameter lmax (h)1) Yx/s (g cells g)1) Qs (g l)1h)1) Qx (g cells l)1h)1) QCP (g l)1 h)1) QTP (g l)1h)1) qCP (g g)1cell h)1) qTP (g g)1cell h)1) YCP/S (g g)1 substrate) YCP/X (g g)1 cells) YTP/X (g g)1 cells) RNA (after heat treatment)

Shake flask 0.12 0.48 0.51 0.31 0.48 0.45 0.11 0.085 0.35 0.88 0.71 1.12

± ± ± ± ± ± ± ± ± ± ± ±

0.011 0.04 0.05 0.06 0.03 0.02 0.01 0.01 0.02 0.05 0.04 0.12

6-l fermentor 0.21 0.56 0.59 0.36 0.56 0.52 0.20 0.16 0.36 1.00 0.86 1.0

± ± ± ± ± ± ± ± ± ± ± ±

0.014 0.05 0.071 0.082 0.090 0.025 0.010 0.011 0.02 0.05 0.06 0.08

Each value is a mean of three independent experiments. ± Stands for standard deviation among replicates.

were found to hold good to synthesize more crude protein, true protein, cell mass productivities and cell yield and were adopted for detailed studies. There was enhanced substrate consumption and product formation in fermentor when the organism was grown on 1.9% kallar grass in optimized Dubos salts medium (controlled pH 7.3) inoculated with 10% seed culture. When agitation and aeration rates were investigated in the 6-l fermentor it was found that C. biazotea grew faster with

Results and discussion Cultural conditions in shake flask To maximize the total crude or true protein productivity, the cultural conditions for the batch culture process were optimized in shake flasks. Among concentrations of kallar grass and nitrogen sources employed, 1.9% kallar grass and NaNO3 (0.40 g 100 ml)1) favoured maximum crude protein productivity, crude protein yield, crude protein’s specific yield, cell mass yield and RNA accumulation in the cells respectively at 30 C. The cell yield (Table 1) obtained following growth on kallar grass was higher than that reported by Hitchner & Leatherwood (1980), Meyer et al. (1993) and Pece et al. (1994). Cultural conditions in 6-l fermentor Various experiments were conducted in a 6-l fermentor to test the above-optimized cultural conditions. They

Figure 1. Kinetics of crude protein (CP) (s), true protein (TP) (n), cell mass (X) ð(Þ and solid kallar grass (,) present in the fermentation broth in shake flask (a) and continuously stirred tank fermentor (b). The initial pH of the medium containing 1.9% (w/v) substrate and inoculated with 10% (v/v inoculum size) was regulated at 7.3, and temperature 30 C. Error bars show standard deviation among three replicates.

210

M. Ibrahim Rajoka

a high aeration rate. Maximum growth rate and protein productivities were realized at an agitation speed of 400 rev min)1 and an aeration rate of 2.0 v/v min at 30 C throughout the fermentation period of 3 days. For comparative studies of both fermentation processes, C. biazotea was permitted to grow (in triplicate) in shake flasks, and 6-l fermentor respectively in optimized fermentation conditions. The kinetics of crude protein production, cell mass formation and solid substrate in the medium in shake flasks (a), and 6 l fermentor (b) is shown in Figure 1. All values of fermentation attributes were higher in the fermentor than in the shake flasks (Table 1) and were significantly higher than those reported by Nigam & Vogel (1991), Table 2. Nutrient composition (percent) of alkali-treated kallar grass and the Cellulomonas biazotea biomass. Nutrients

Kallar grass

C. biazotea biomass

Moisture Dry matter Crude protein True protein Ether extract Ash Crude fibre Nitrogen free extract Cellulose Hemicellulose RNA Calorific value (kcal kg)1)

2.50 ± 0.15 97.50 ± 0.04 6.5 ± 0.45 0.2 ± 0.0 3.01 ± 0.24 17.8 ± 1.14 11.82 ± 1.04 57.37 ± 4.13 55.0 ± 4.54 23.0 ± 2.00 0.25 ± 0.03 100 ± 9.25

0.2 ± 0.04 100.00 ± 8.65 60.0 ± 5.04 56.1 ± 4.64 3.5 ± 0.24 12.95 ± 1.24 11.50 ± 1.34 0.5 ± 0.04 1.50 ± 0.06 0.54 ± 0.02 1.00 ± 0.44 3500 ± 21.23

Each value is a mean of three replicates. ± Stands for standard deviation among replicates. The product was rich in true protein and had low cellulose, hemi cellulose and RNA content.

Lee & Kim (2000), Nigam (2000) and Paul et al. (2002). There was enhanced substrate metabolism by the aerobic pathway, resulting in build-up of high cell mass (maximum cell mass is equal to 8.9 g l)1) (Figure 1). The values of the above process variables were higher than those reported previously (Zayed & Mustafa 1992; Nigam 2000) and were attributed to the presence of high cellulase and xylanase activities in the organism (Rajoka & Malik 1997; Rajoka et al. 1997) and in the case of the fermentor, due to the higher aeration and mass transfer rates. The improvement of crude protein from 6.3 to 60% with 56.1% as true protein indicated a great deal of nitrogen source utilization by the cells. Single cell protein product reported by Singh et al. (1991) contained 30.4% crude protein and Kluyveromyces fragilis biomass grown on deproteinized when supplemented with 0.8% diammonium hydrogen phosphate contained 37% crude protein (Paul et al. 2002). Similarly yeast cell mass reported by Chanda & Chakrabarti (1996) and Meyer et al. (1993) contained 54.3%, and 47% crude protein respectively. Biochemical evaluation of SCP Compositional analysis of single cell protein obtained with C. biazotea in the fermentor (Table 2) revealed that dry biomass was rich in crude protein, true protein and low in RNA and was found superior to those reported by Singh et al. (1991), Abouzeid et al. (1995), Nigam (2002), and Paul et al. (2002). The RNA content was found to be 1.0% which is significantly lower than those reported by other workers (Nigam & Vogel 1991; Singh et al. 1991; Paul et al. 2002). The dry biomass showed a

Table 3. Chemical score of single cell protein obtained through fermentation of alkali-treated kallar grass with C. biazotea using FAO/WHO (1957) amino acid pattern. Amino Acid (AA)

SCP AA content (g 100 g)1)

Amino acid patterna (mg g)1)

SCP AA (mg g)1)

Available (%)

Lysine Leucine Isolecucine Phenylalanine Methionine Threonine Valine Alanine Arginine Aspartic acid Cystein/cystine Glutamate Glutamine Histidine Proline Serine Threonine Tyrosine

3.7 5.9 3.2 3.6 1.9 2.5 4.5 2.1 1.5 3.4 5.5 5.5 1.8 1.6 4.8 1.6 1.8 1.5

42 48 42 28 22 28 42 – – – – – – – – – – –

36.59 58.32 31.25 35.72 19.18 24.98 45.19 – – – – – – – – – – –

87.10 121.50 74.40 127.58 87.18 89.21 107.59 – – – – – – – – – – –

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.21 0.31 0.21 0.24 0.14 0.15 0.24 0.14 0.20 0.25 0.36 0.45 0.21 0.12 0.34 0.15 0.17 0.21

Test samples were hydrolysed with HCl and analysed by automatic amino acid analyser in triplicate. Each value is a mean of three replicates. ± Stands for standard deviation among replicates. a FAO/WHO amino acid pattern. The protein product was deficient in isoleucine whose chemical score was 74.4.

Production of SCP from a grass gross metabolizable energy value of 3500 kcal kg)1. The calorific value clearly indicated that the biomass could serve as energy source when it may be fed to poultry and livestock. Single cell protein contained all the amino acids and the chemical score of seven selected essential amino acids (Table 3) indicated favourably comparative values to FAO/WHO (1957) reference protein in mg g)1 and confirmed the findings of Nigam (2000).

Conclusions Kallar grass has shown excellent potential as alternative energy crop. It gives a massive biomass yield per ha of saline land. On the basis of SCP production kinetic data using C. biazotea, a yield of 1000 kg per ha could be obtained. This product contains fairly good quality protein. The NIAB Laboratories have already introduced kallar grass in our province and is providing assistance to farmers. More saline lands in the country are being identified which are not suitable for other crops and can be used for the cultivation of this perennial grass for bulk production of SCP economically for fortification of livestock or poultry feed.

References Abouzeid, A.Z.A., Khan, J.A. & Abulanaja, K.O. 1995. On methods for reduction of nucleic acids content in a single-cell protein from gas oil. Bioresource Technology 52, 21–24. Aiba, S., Humphrey, A.E. & Millis, N.F. 1973 Biochemical Engineering. 2nd edn. pp. 92–127. New York: Academic Press. ISBN 0120450526. Association of Official Analytical Chemists, 1984 Official Methods of Analysis. 14th edn. Washington, DC. ISBN 0935584242. Bajpai, P.K. & Bajpai, P. 1991 Cultivation and utilization of Jerusalem artichoke for ethanol, single cell protein and high-fructose syrup production. Enzyme and Microbial Technology 13, 359–363. Chanda, S. & Chakrabarti, S. 1996 Plant origin liquid waste. A resource for single cell protein production by yeasts. Bioresource Technology 57, 51–54. Enriquez, A. & Rodriquez, H. 1983 High productivity and good nutritive values of cellulolytic bacteria grown on sugarcane bagasse. Biotechnology and Bioengineering 25, 877–880. FAO/WHO, 1957 Ad hoc Expert ‘Committee, Energy and Protein Requirements’. pp. 35–36. World Health publishing Co. Haddadin, M.S., Abdulrahim, S.M., Alkhawaldeh, G.Y. & Robibson, R.K. 1999 Solid-state fermentation of waste pomace from olive processing. Journal of Chemical Technology and Biotechnology 74, 613–618. Hill, F. & Anderson, D.L. 1958 Comparison of metabolizable energy and productive energy determinations with growing chicks. Journal of Nutrition 64, 587–603.

211 Hitchner, E.V. & Leatherwood, J.M. 1980 Use of a cellulasederepressed mutant of Cellulomonas in the production of a single cell protein. Applied and Environmental Microbiology 39, 382. Hongpattarakere, T. & H-Kittikun, A. 1995 Optimization of single cell-protein production from cassava starch using Schwanniomyces castelli. World Journal of Microbiology and Biotechnology 11, 607– 609. Latif, F., Rajoka, M.I. & Malik, K.A. 1994 Saccharification of Leptochloa fusca (kallar grass straw) by thermostable cellulases. Bioresource Technology 50, 107–111. Lee, B.K. & Kim, J.K. 2000 Production of Candida utilis biomass on molasses in different culture types. Aquacultural Engineering 25, 111–124. Meyer, P.S., du Preez, J.C. & Kilian, S.G. 1993 Chemostat cultivation of Candida blankii on sugarcane bagasse hemicellulose hydrolysate. Biotechnology and Bioengineering 40, 353–358. Nigam, J.N. 2000 Cultivation of Candida langeronii in sugar cane bagasse hemicellulosic hydrolyzate for the production of single cell protein. World Journal of Microbiology and Biotechnology 16, 367– 372. Nigam, P. & Vogel, M. 1991 Bioconversion of sugar-industry byproducts molasses and sugar-beet pulp for single cell protein production by yeasts. Biomass and Bioenergy 1, 339–345. Pacheco, M.T., Caballero-Cordoba, G.M. & Sgarbieri, V.C. 1997 Composition and nutritive value of yeast biomass and yeast protein concentrates. Journal of Nutrition Science and Vitaminology (Tokyo) 43, 610–612. Paul, D., Mukhopadhyay, R., Chatterjee, B.P. & Guha, A.K. 2002 Nutritional profile of food yeast Kluyveromyces fragilis biomass grown on whey. Applied Biochemistry and Biotechnology 97, 209– 218. Pece, N., Perotti, N. & Molina, O. 1994 Bacterial protein production from corn cob pretreated with NaOH at room-temperature. World Journal of Microbiology and Biotechnology 10, 593–594. Rajoka, M.I. 1990 Bioconversion of lignocellulosic materials raised from saline lands for production of biofuels using Cellulomonas species (as cellulolytic organisms). PhD thesis, University of the Punjab, Lahore, Pakistan. Rajoka, M.I. & Malik, K.A. 1997. Cellulase production by Cellulomonas biazotea cultured in media containing different cellulosic substrates. Bioresource Technology 59, 21–27. Rajoka, M.I., Bashir, A. & Malik, K.A. 1997 Mutagenesis of Cellulomonas biazotea for enhanced production of xylanases. Bioresource Technology 62, 99–108. Rodriguez-Vasquez, R.R., Villanueva-Ventura, G. & Rios-Leal, E. 1992 Sugarcane bagasse pith dry pretreatment for single cell protein-production. Bioresource Technology 39, 17–22. Rosenberg, E., 1993 Exploiting microbial-growth on hydrocarbons: new markets. Trends in Biotechnology 11, 419–424. Singh, A., Abidi, A.B., Agarwal, A.K. & Darmwal, N.S. 1991 Single cell protein production by Aspergillus niger and its evaluation. Zentralblatt fu¨r Mikrobiologie 146, 181–184. Tobajas, M. & Garcia-Calvo, E. 1999 Determination of biomass yield of Candida utilis on glucose: Black box and metabolic description. World Journal of Microbiology and Biotechnology 15, 431– 438. Zayed, G. & Mostafa, N. 1992 Studies on the production and kinetic aspects of single cell protein from sugar-cane bagasse saccharified by Aspergillus niger. Biomass and Bioenergy 3, 363–367.


Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 213–219 DOI 10.1007/s11274-004-3619-9

Chromate reduction by Bacillus megaterium TKW3 isolated from marine sediments K.H. Cheung1 and Ji-Dong Gu1,2,* 1 Laboratory of Environmental Toxicology, Department of Ecology & Biodiversity, The University of Hong Kong, Pokfulam Road, Hong Kong SAR, PR China 2 The Swire Institute of Marine Science, The University of Hong Kong, Shek O, Cape d’Aguilar, Hong Kong SAR, PR China * Author for correspondence: Tel.: +852-2299-0605, Fax: +852-2517-6082, E-mail: [email protected] Received 8 April 2004; accepted 10 July 2004

Keywords: Bacillus megaterium, bioremediation, chromate reduction, chromium, metal resistance, selenium

Summary Bacillus megaterium strain TKW3 was isolated from multiple-metal-contaminated marine sediments of Tokwawan, Hong Kong SAR. This facultative aerobe utilized arabinose, mannitol, N-acetylglucosamine, maltose, caprate, citrate, butyrate or lactate as the sole source of carbon and energy for growth. B. megaterium TKW3 reduced highly 3+ toxic and soluble Cr6+ (as CrO2
under aerobic conditions. Complete 4 ) into almost non-toxic and insoluble Cr 6+ reduction of 0.20 mM Cr by B. megaterium TKW3 was achieved within 360 h. Initial Cr6+ concentration below 0.90 mM or inoculum less than 107 cells ml)1 did not have significant effect on Cr6+ reduction, while the residue Cr6+ concentration was the lowest at 107 cells ml)1. Cr6+ reduction by this strain was inhibited by high levels of NaCl (55&). B. megaterium TKW3 was also resistant to other oxyanions including 0.34 mM Cr2O2
7 0.32 mM 4+ 2
2
2
AsO3
, 0.58 mM SeO and 0.53 mM SeO , and reduced soluble Se (as SeO ) to insoluble red amorphous Se0. 3 4 3 4 B. megaterium TKW3 might have potential application in bioremediation of Cr-laden sediments associated with other oxyanions.

Introduction Chromium (Cr) is used extensively in industries like electroplating, stainless steel production and wood preservation (Cheung & Gu 2002; Ryan et al. 2002). Although Cr is an essential trace metal to humans, its potential teratogenicity, mutagenicity, carcinogenicity and toxicity render it hazardous at very low concentrations and it has been classified as priority pollutant by the US Environmental Protection Agency (US EPA) (1998) (Fernando & Flora 1977; Sharma et al. 1995; Gibb et al. 2000). The valency state of Cr ranges from )2 to +6, with the hexavalent state (Cr6+) like 2
chromate (CrO2
4 ) and dichromate (Cr2O7 ) being the dominant species in natural seawater, the trivalent state (Cr3+) prevalent in wastewater inundated with organics (Fukai 1967; Jan & Young 1978). In contrast to the highly toxic and soluble Cr6+, Cr3+ compounds, e.g. Cr(OH)3 are almost non-toxic and insoluble which can be precipitated readily. In wastewater treatment, Cr6+ reduction is traditionally being catalysed by physicochemical methods, which involve initial pH adjustment with strong acid (e.g. H2SO4) followed by the addition

of reducing agent (e.g. SO2) and re-adjustment of pH with alkali (e.g. NaOH) for precipitation (Mahajan 1985). Bioremediation of Cr6+ using microorganisms has been studied, where periplasmic biosorption, intracellular bioaccumulation and biotransformation (mainly as dissimilatory reduction) through direct enzymatic reaction or indirectly with metabolites are the major processes (Lovley 1993; Lee et al. 2000; Valls et al. 2000). Cr6+-reducing bacteria have been isolated, a number of them belong to the group of anaerobic sulphate-reducing bacteria (SRB), while studies on species of Pseudomonas, Bacillus, Escherichia and Enterobacter have also been conducted (Lovley & Phillips 1994; McLean & Beveridge 2001; Cheung & Gu 2003). Investigations on Bacillus megaterium were previously focused on its endospore and plasmids, no reports on multiple-metal resistance and reduction by this species have been reported (Slepecky & Hemphill 1999). In the present study, B. megaterium TKW3 has been isolated from marine sediments and was further investigated for the reduction of Cr6+ under laboratory conditions.

214 Materials and methods Enrichment and isolation Sediments heavily polluted with various metals, including Cr, were collected from a sheltered bay Tokwawan, Hong Kong SAR in the vicinity of industrial effluent outlets (HK EPD 2000). 5 ml sediment slurry was transferred into 50 ml nutrient broth (NB) (Difco Laboratory, Detroit, USA) containing (g l)1) 1.0 beef extract, 2.0 yeast extract, 5.0 peptone and 5.0 NaCl, amended with 1.0 mM Cr6+ (as K2CrO4; Fisher Scientific). The inoculated culture was incubated in a rotary shaker (150 rev/min) at 30 C for 7 days, serving as the initial enrichment culture. Subsequent enrichment transfer cultures were established in a similar manner with inoculum from the preceding culture. Pure culture was obtained from the third enrichment transfer culture with standard spreading plates and streaking techniques on nutrient agar (Difco Laboratory, Detroit, Michigan) plates with 1.5% agar. All the procedures above were conducted aseptically. Characterization and identification After Gram staining of the pure bacterial isolate, API 20NE Multiple Kit (bioMerieux, Marcy l’Etoile, France) was adopted to preliminarily characterize the isolates following standard protocol according to the manufacturer’s instructions (bioMerieux 07615C 1998). The biochemical utilization of the strain was further studied, 5 ml pure culture was inoculated into 60 ml minimal salt medium (MSM) supplemented with 20 mM of various carbon substrates. The MSM was composed of (g l)1) 0.8 K2HPO4, 0.2 KH2PO4, 0.05 CaSO4 Æ 2H2O, 0.5 MgSO4 Æ7H2O, 0.01 FeSO4 Æ 7H2O, and 1.0 (NH4)2SO4 (Gu et al. 1998). The carbon sources under investigation included sodium salts of succinate, propionate, pyruvate, butyrate, caproate, formate, acetate, citrate, fumarate, lactate and benzoate; D (+)glucose, phenol, indole, ethanol, propanol, butanol and quinoline. The optimum growing conditions of the isolate were studied at temperatures of 10, 20 and 30 C, and salinities of 5, 35 and 55& NaCl. Bacterial growth was quantified by measuring the protein contents of liquid culture with the modified Bradford assay (Daniels et al. 1994) as follows. 200 ll sub-sample was added to 2.5 ml Coomassie Brilliant Blue reagent (100 mg Coomassie Brilliant Blue G250, 50 ml 95% ethanol, 100 ml 85% H3PO4, 1 l distilled water). The absorbance at wavelength 595 nm (A595) was measured using spectrophotometer (UV-1201V, Shimadzu, Kyoto) after 5-min reaction. Concentration of standard protein (Albumin from Sigma) showed linear relationship between 2.5 and 30 lg ml)1. For anaerobic growth of the strain, 50 ml NB (Difco Laboratory) in a serum bottle was flushed with pure N2 for 30 min to purge out O2, 0.25 g l)1 Na2S Æ 9H2O was

K.H. Cheung and Ji-Dong Gu added to reduce the residual amount of dissolved oxygen, then capped with a tert-butyl rubber stopper and crimp sealed with aluminium. Resazurin (1.0 lg m)1) was added as the redox indicator showing any potential contamination of O2. 5 ml pure culture was inoculated into the anaerobic NB and incubated with shaking at 150 rev/min and 30 C for the assessment of bacterial growth. All experiments above were conducted aseptically in duplicate with uninoculated controls. Bacterial 16S rDNA extraction and analysis One ml overnight culture of the bacterium grown in NB (Difco Laboratory) was centrifuged and the cell pellet was suspended in 200 ll extraction buffer (100 mM Tris, 100 mM EDTA, 200 mM NaCl, 1% polyvinylpyrrolidone (w/v), 2% CTAB pH 8.0). Then 200 ll SDS buffer consisting of 2% SDS (w/v), 10 mM Tris, and 200 mM NaCl (pH 8.0) was added. The suspension was extracted with 400 ll phenol/Tris-HCl, 400 ll phenol and chloroform–isoamyl alcohol (25:24:1) and 400 ll chloroform/isoamyl alcohol (24:1) with centrifugation at 16,000 rev/min for 5 min, respectively. Bacterial DNA was precipitated in 400 ll isopropanol at )20 C for 1 h and centrifuged at 14,000 rev/min for 20 min. DNA was vacuum-dried, dissolved in 50 ll sterile distilled water and stored at )20 C before further experiment. Concentration of DNA was determined using an Eppendorf BioPhotometer. The PCR mixture (50 ll) contained the bacterial DNA, 0.5 lM primers (for the first 527-bp fragment), PCR buffer (10 mM Tris-HCl at pH 8.3, 50 mM KCl, 2 mM MgCl2, 0.01% gelatin), a 200 lM concentration of each dNTP, and 1.0 U of Taq polymerase. Primers were 5¢-GAGAG TTTGA TCCTG GCTCA G-3¢ (forward) and 5¢-CTACG GCTAC CTTGT TACGA-3¢ (reverse) (Cello et al. 1997). The mixture was amplified for 30 cycles at 95 C for 30 s, 60 C for 30 s, and 72 C for 45 s with a final extension at 72 C for 10 min, in an automated thermal cycler. Purified PCR product (Qiagen PCR Purification Kit) was cloned in pGEM-T vector and E. coli DH 5a competent cells for sequencing. DNA nucleotide sequence was analysed using the database of GenBank. For the construction of a phylogenetic tree, sequences of the 10 most closely related organisms with that of Escherichia coli were included in the comparison. The similarity values for these sequences were transformed into phylogenetic distance values that compensate for multiple substitutions at any given site in the sequence. The phylogenetic dendrogram was constructed with the neighbour-joining method using the PHYLIP package. Bacteria used in construction of the phylogenetic tree with their GenBank accession numbers included Bacillus megaterium Strain C1 (AJ491841), Bacillus flexus (AY422986), Bacillus asahii (AB109209), Bacillus psychrosaccharolyticus (AB021195), B. circulans (X60613), Bacillus cereus strain G3317 (AY138278), Bacillus thuringiensis serovar

Chromate reduction by Bacillus sp.

215 Strain TKW3

100 79

Bacillus megaterium

29

Bacillus flexus Bacillus circulans

55

Bacillus subtilis 55

Bacillus thuringiensis

96 100

Bacillus cereus

74 Bacillus anthracis

Bacillus asahii Bacillus psychrosaccharolyticus 60

Bacillus simplex Escherichia coli

0.05 Figure 1. Phylogenetic relationships by a neighbouring-joining analysis of 16S rDNA sequences, showing the position of strain TKW3 isolated from marine sediments. E. coli was used as the out-group. Scale bar = 10 nucleotides substitutions per 100 nucleotides of 16S rDNA sequence.

israelensis (AY461762), B. simplex (X60638), Bacillus subtilis (AF545570), Bacillus anthracis (AB116124), E. coli (J01859). 16S rDNA sequence of isolate TKW was deposited in GenBank with the accession number AY524978.

were conducted in duplicate with non-inoculated controls.

Metal resistance and chromate reduction

Enrichment culture and the isolate

To investigate heavy metal resistance of the isolate on other environmentally important metals under aerobic conditions, 500 ll pure culture was inoculated into 20 ml NB (Difco Laboratory) amended with 0.34 mM 3
dichromate (Cr2O2
7 ), 0.32 mM arsenate (AsO4 ), 2
0.53 mM selenate (SeO4 ) or 0.58 mM selenite (SeO2
3 ). One ml sample was extracted at regular time intervals for the measurement of bacterial growth. For the effect of initial cell density on Cr6+ reduction, 1 ml pure culture with an initial cell density of 105, 106 or 107 cells ml)1 was inoculated into 100 ml NB (Difco Laboratory) amended with 0.48 mM Cr6+ (as CrO2
4 ). Two ml samples were taken at regular time intervals for the assessment of bacterial growth, and Cr6+ concentration with colorimetric diphenylcarbazide (DPC) analysis as described previously (Cheung & Gu 2003). 300 ll sub-sample was pipetted into 9.7 ml 0.4 M H2SO4 buffered with 25 ll H3PO4, 0.5 ml DPC reagent (0.025 g 1,5-diphenylcarbazide dissolved in 10 ml acetone) was added for 5 min reaction before measuring the A540. For the effect of initial concentration of Cr6+ and NaCl, inoculum (1 ml) with initial cell density of 105 cell ml)1 were added to NB (Difco Laboratory) amended with different concentrations of Cr6+ (0.05, 0.20, 0.45 or 0.90 mM CrO2
4 ) and NaCl (5, 35 or 55%, plus 0.4 mM CrO2
4 ). Two ml samples were taken at regular time intervals for the assessment of bacterial growth and Cr6+ concentration. All experiments above

A pure culture capable of reducing Cr6+ was isolated from marine sediments after enrichment. The isolate was a Gram-positive facultative aerobe of size 4.0 · 1.0 lm, with limited growth under anaerobic condition (data not shown). The phylogenetic identity of the isolate was determined by 16S rDNA sequencing, which revealed that it was most closely related to B. megaterium (designated as strain TKW3) of family Bacillaceae with 99.9% similarity (Figure 1). A broad biochemical utilization profile was found for B. megaterium TKW3, where arabinose, mannitol, N-acetylglucosamine, maltose, caprate, citrate, butyrate and lactate could be utilized as sole source of carbon and energy (Table 1) (Slepecky & Hemphill 1999). Among the tested temperature values, the highest growth for B. megaterium TKW3 was at 30 C with a rate of 0.067 lg (protein) s)1, lower temperatures retarded bacterial growth (deferred by 96 h at 10 C) yet the maximum growth yield was similar at about 200 lg (protein) ml)1 (after 188 h at 10 C, while 96 h at 20 and 30 C) (Figure 2). At the tested salinity levels, B. megaterium TKW grew best at 5& NaCl with a growth rate of 0.080 lg (protein) s)1, higher salinity retarded both the growth rate and maximum yield (Figure 3). Cr6+ reduction by B. megaterium TKW3 was retarded by 35& NaCl, with Cr6+ reduction rate of 5.26 · 10)12 lmol (Cr6+) cell)1 h)1 compared with 1.39 · 10)11 lmol (Cr6+) cell)1 h)1 for 5& NaCl.

Results

216

K.H. Cheung and Ji-Dong Gu

Result

Cell type Cell size Gram staining Nitrate reductase Oxidase Tryptophan Esculine hydrolase Gelatine p-Nitrophenyl-D-galactopyranoside D-Glucose L-Arabinose D-Mannose D-Mannitol N-acetylglucosamine D-Maltose Gluconate Caprate Adipate Malate Citrate Phenylacetate Succinate Propionate Pyruvate Butyrate Formate Acetate Fumarate Lactate Phenol Indole Benzoate Ethanol Butanol Propanol Quinoline Glucosea Arginine dihydrolasea Ureasea

Rod 4.0 lm · 1.0 lm + ) ) ) + + + + + + (weak) + + + + (weak) + + (weak) + + + (weak) ) ) + (weak) + ) + (weak) + (weak) + ) ) ) ) ) ) + (weak) ) ) )

0.3 200 0.2 100 0

0.1 0

40

80 Time (h)

120

160

0

Figure 3. Effect of salinity on the growth of and Cr6+ reduction by B. megaterium TKW3. Bacterial growth was investigated with the Bradford assay, while the reduction of Cr6+ was quantified with colorimetric diphenylcarbazide method. 1 ml inoculum of B. megaterium TKW3 was added into 100 ml nutrient broth amended with 0.40 mM Cr6+of salinity 5& (proteins: n, Cr6+: h); 35& (proteins: r, Cr6+: e) and 55& (proteins: d, Cr6+: s) NaCl, then incubated with shaking (150 rev/min) at 30 C. Vertical bars indicate standard deviation from duplicate from each treatment.

inhibition of Cr6+ reduction was observed with 55& NaCl (Figure 3). Metal resistance and chromate reduction

a

Tested under anaerobic conditions.

Protein analyses revealed that B. megaterium TKW3 maintained metabolic activity in 0.34 mM Cr2O2
7 , 2
2
0.32 mM AsO3
, 0.58 mM SeO and 0.53 mM SeO 3 4 , 4 while the magnitude of bacterial growth was in 3
2
descending order: SeO2
and Cr2O2
4 , AsO4 , SeO3 7 (Figure 4). Precipitates were found in cultures amended 2
with Cr2O2
7 (white precipitates) and SeO3 (red precipitates), while no visible precipitates were observed in uninoculated or non-viable cell controls (data not shown), illustrating that the transformation of soluble metal ions to insoluble states was biologically mediated. Although initial cell density did not significantly affect the Cr6+ reduction rate, the residual Cr6+ (0.126 mM, 73.47% reduced) was the lowest at 107 cells ml)1 as the 200 Proteins (ug ml-1)

Cr6+ reduction by B. megaterium TKW3 ceased after 96 h with residue concentration of 0.28 mM (Cr6+) (31.7% reduced) at 35& NaCl level, while complete 250 Proteins(ug ml-1)

0.4

300

Cr(VI) (mM)

Test / Characteristic

0.5

400 Proteins (ug ml-1)

Table 1. Phenotypic (morphological and biochemical) expression of B. megaterium TKW3.

200 150

150 100 50

100

0

50 0

0

40

80

120 Time (h)

160

0

40

80

120 Time (h)

160

200

240

200

Figure 2. Effect of temperature on the growth of B. megaterium TKW3. Bacterial growth was determined by measuring the protein contents of the culture with Bradford assay. 1 ml inoculum of B. megaterium TKW3 was added into 20 ml nutrient broth and incubated in a rotary shaker (150 rev/min) at temperatures of 10 C (n), 20 C (s) and 30 C (h).

Figure 4. Growth of B. megaterium TKW3 in the presence of other metals. Bacterial growth was determined by measuring the protein contents of the culture with the Bradford assay. 500 ll culture of B. megaterium TKW3 was inoculated into 20 ml nutrient broth amended with Cr2O2
(0.34 mM; e), AsO3
(0.32 mM; ·), SeO2
7 3 4 (0.58 mM; h) and SeO2
4 (0.53 mM; n), then incubated with shaking (150 rev/min) at 30 C. Vertical bars indicate standard deviation from duplicate from each treatment.

Chromate reduction by Bacillus sp.

217

substrates; whereas arabinose, mannitol, N-acetylglucosamine, maltose, caprate, citrate, butyrate and lactate 0.5 were firstly reported to be utilized by this species 0.4 (Table 1) (Slepecky & Hemphill 1999). The broadened spectrum of utilizable substrates might enhance the 0.3 flexibility of culturing B. megaterium TKW3. Although 0.2 studies on the cytology and biochemistry of B. megate0.1 rium were conducted, strain TKW3 has other interesting characteristics of multiple-metal resistance and Cr6+ 0 0 40 80 120 160 200 reduction compared with other strains (Hu & Boyer Time (h) 1996; England et al. 1997; Adam et al. 2001; Tang et al. 6+ Figure 5. Effect of initial cell density on Cr reduction by 2001). Similar to B. megaterium TKW3, high NaCl level 6+ was quantified of salinity at 58.5& caused a concentration-dependent B. megaterium TKW3. The concentration of Cr with the colorimetric diphenylcarbazide method. 1 ml inoculum of growth delay and decreased growth yield in B. megateB. megaterium TKW3 with initial cell density of 105 (e), 106 (h) and rium strain 27, related to the transformation of stable 107 (n) cell ml)1 was added into 100 ml nutrient broth amended with long-lived cell proteins (LLP) into quickly degraded 0.48 mM Cr6+, then incubated with shaking (150 rev/min) at 30 C. (SLP) (Nekolny & Chaloupka Vertical bars indicate standard deviation from duplicate from each short-lived proteins 6+ 2000). Since Cr reduction by B. megaterium TKW3 treatment. was inhibited by a high level of NaCl, direct application of this strain in Cr6+ bioremediation at high salinity 1 over 55& NaCl is limited (Figure 3). Cr6+ reduction by B. megaterium TKW3 was depen0.8 dent on the initial cell density, while the initial Cr6+ 0.6 concentration did not have significant effect on Cr reduction rate when below 0.9 mM in our study. Similar 0.4 phenomena were observed in Microbacterium sp. MP30 and Enterobacter cloacae HO1, where the Cr6+ reduc0.2 tion rate was similar with initial Cr6+ concentration 0 below 0.5 and 10 mM, respectively, illustrating the 0 40 80 120 160 200 240 280 320 360 presence of other limiting factors like the availability of Time (h) carbon substrates (Komori et al. 1990; Ohtake et al. 6+ were indepenFigure 6. Effect of initial Cr6+ concentration on its reduction by 1990). Resistance and reduction of Cr 6+ B. megaterium TKW3. The concentration of Cr reduction was dent processes in Pseudomonas fluorescens, where Cr6+ quantified with the colorimetric diphenylcarbazide method. 1 ml reduction was mediated by a non-specific enzyme that inoculum (105 cell ml)1) of B. megaterium TKW3 was added into naturally consumes other substrate(s) (Cervantes & 100 ml nutrient broth amended with Cr6+ of concentration 0.05 (s), 6+ reduction rate below 0.9 0.20 (e), 0.45 (n) and 0.90 (h) mM, then incubated with shaking silver 1992). The similar Cr 6+ by B. megaterium TKW3 reflected the normal (150 rev/min) at 30 C. Vertical bars indicate standard deviation from mM Cr duplicate from each treatment. functioning of its detoxification mechanism(s) to Cr6+. B. megaterium TKW3 was found to be resistant to 2
2
3
inoculum, and the residual Cr6+ was similar between Cr2O2
7 , SeO3 , SeO4 and AsO4 , so it is worthwhile to 5 6 )1 inocula of 10 and 10 cells ml with 0.195 and investigate the interaction (neutral, synergistic or antag0.186 mM (Cr6+), respectively (Figure 5). The Cr6+ onistic) between these metals/metalloids on Cr6+ reducreduction rate was similar with initial Cr6+ concentra- tion by this strain. The formation of precipitates only in 2
tions between 0.05 and 0.90 mM in our study. Complete cultures amended with Cr2O2
7 and SeO3 illustrated the 6+ reduction of 0.05 and 0.20 mM (Cr ) by B. megaterium transformation of these soluble oxyanions by B. megaTKW3 was observed within 24 and 360 h, respectively. terium TKW3. Cr6+ was reduced to Cr3+ that forms The residue Cr6+ was 0.13 mM (71.1% reduced) and insoluble hydroxide Cr(OH)3 as a white precipitate 0.58 mM (35.6% reduced) respectively for initial con- (Lovley & Phillips 1994; Chardin et al. 2002). The centrations of 0.45 and 0.90 mM Cr6+ after incubating detection of Cr5+ as a short-lived intermediate in Cr6+ for 360 h (Figure 6). No significant Cr6+ reduction was reduction to Cr3+ by Pseudomonas ambigua G-1 detected in all uninoculated or non-viable controls. indicated that the reduction involved at least two reaction steps (Suzuki et al. 1992). Aerobic Cr6+ reduction involved a soluble reductase using NADH or Discussion endogenous electron reserves as electron donor; while under anaerobic conditions, Cr6+ might serve as termiB. megaterium TKW3 was isolated from marine sedi- nal electron acceptor through respiratory chains ments contaminated with high level of metals, such as mediated by soluble reductase, membrane-bound reduc160 mg Cr kg)1 (HK EPD 2000). B. megaterium was tase or both (Wang & Shen 1995). SeO2
3 was reduced commonly found in soil and could utilize a range of by B. megaterium TKW3 to elemental Se0 as insoluble Cr(VI) (mM)

Cr(VI) (mM)

0.6

218 red amorphous selenium, which was reported to be a detoxification mechanism which is not associated with energy conservation in bacteria (Blake II et al. 1993; White et al. 1997). B. megaterium TKW3 reduced toxic and soluble Cr6+ to non-toxic and insoluble Cr3+ readily under moderate conditions, with a wide range of simple carbon substrates, and no supplement of chemicals like flocculants, reducing agents, acid and alkali were required. Resistance of this strain to other heavy metals is also of advantage for in situ and ex situ Cr6+ bioremediation, which might be associated with other heavy metals. Other techniques like the dialysis sac bioreactor, anionexchange membrane reactor, continuous stirred tank bioreactor, artificial biofilms and rotating biological contactor could be applied complementarily with cells of B. megaterium TKW3 to enhance the overall efficiency in Cr6+ reduction (Ganguli & Tripathi 2002). Current investigations are being conducted to study the fraction(s) of protein (periplasmic, cytoplasmic or both) of B. megaterium TKW3 responsible for Cr6+ reduction to delineate the pathway and mechanism involved, as well as the interaction with other heavy metals on Cr6+ reduction by this strain.

Acknowledgements This project was supported by an ITF grant (276/00) from the Innovative Technology Commission of Hong Kong SAR, and industrial partnerships with Kou Hing Hong Scientific Supplies, Peako Engineering Co. We thank Jessie Lai for laboratory assistance and Paula Wang and Li Pan for the DNA sequence alignment. References Adam, W., Lukacs, Z., Kahle, C., Saha-Moller C.R. & Schreier, P. 2001 Biocatalytic asymmetric hydroxylation of hydrocarbons by free and immobilized Bacillus megaterium cells. Journal of Molecular Catalysis 11, 377–385. Blake II, R.C., Choate, D.M., Bardhan, S., Revis, N., Barton, L.L. & Zocco, T.G. 1993 Chemical transformation of toxic metals by a Pseudomonas strain from a toxic waste site. Environmental Toxicology and Chemistry 12, 1365–1376. Cello, F.D., Bevivino, A., Chiarini, L., Fani, R., Paffetti, D., Tabacchioni, S. & Dalmastri, C. 1997 Biodiversity of a Burkholderia cepacia population isolated from the maize rhizosphere at different plant growth stages. Applied and Environmental Microbiology 63, 4485–4493. Cervantes, C. & Silver, S. 1992 Plasmid chromate resistance and chromate reduction. Plasmid 27, 65–71. Chardin, B., Dolla, A., Chaspoul, F., Fardeau, M.L., Gallice, P. & Bruschi, M. 2002 Bioremediation of chromate: thermodynamic analysis of the effects of Cr(VI) on sulfate-reducing bacteria. Applied Microbiology and Biotechnology 60, 352–360. Cheung, K.H. & Gu, J.-D. 2002 Bacterial color response to hexavalent chromium, Cr6+. The Journal of Microbiology 40, 234–236. Cheung, K.H. & Gu, J.-D. 2003 Reduction of chromate (CrO2
4 by an enrichment consortium and an isolate of marine sulfate-reducing bacteria. Chemosphere 52, 1523–1529.

K.H. Cheung and Ji-Dong Gu Daniels, L., Hanson R.S. & Phillips. J.A. 1994 Chemical analysis. In Methods for general and Molecular Bacteriology, Eds Gerhardt, P. Murray, R.G.E., Wood, W.A. & Kried, N.R. pp. 512–554. Washington DC: American Society for Microbiology. ISBN 1-55581-048-9. England, D.F., Penfold, R.J., Delaney S.F. & Rogers, P.L. 1997 Isolation of Bacillus megaterium mutants that produce high levels of heterogenous protein, and their use to construct a highly mosquitocidal strain. Current Microbiology 35, 71–76. Fernando, L.P. & Flora, S.D. 1977 Toxicity and mutagenicity of hexavalent chromium on Salmonella typhimurium. Applied and Environmental Microbiology 33, 805–809. Fukai, R. 1967 Valency state of chromium in seawater. Nature (London) 213, 901. Ganguli, A. & Tripathi, A.K. 2002 Bioremediation of toxic chromium from electroplating effluent by chromate-reducing Pseudomonas aeruginosa A2Chr in two bioreactors. Applied Microbiology and Biotechnology 58, 416–420. Gibb, H.J., Lees, P.S.J., Pinsky, P.F. & Rooney, B.C. 2000 Lung cancer among workers in chromium chemical production. American Journal of Industrial Medicine 38, 115–126. Gu, J.-D., Mitchell, R., Mitton, B. & Ford, T.E. 1998 Microbial degradation of polymeric protective coatings determined by the electrochemical impedance spectroscopy. Biodegradation 9, 39–45. HK Environmental Protection Department (HK EPD) 2000 Marine water quality in Hong Kong in 2000. Hong Kong SAR, PR China: Environmental Protection Department. Hu, X. & Boyer, G.L. 1996 Siderophore-mediated aluminum uptake by Bacillus megaterium ATCC 19213. Applied and Environmental Microbiology 62, 4044–4048. Jan, T.-K. & Young, D.R. 1978 Chromium speciation in municipal wastewaters and seawater. Journal of Water Pollution Control Federation 50, 2327–2336. Komori, K., Rivas, A., Toda, K. & Ohtake, H. 1990 A method for removal of toxic chromium using dialysis-sac culture of a chromate-reducing strain of Enterobacter cloacae. Applied Microbiology and Biotechnology 33, 117–119. Lee, D.-C., Park, C.-J., Yang, J.-E., Jeong, Y.-H., & Rhee, H.-I. 2000 Screening of hexavalent chromium biosorbent from marine algae. Applied Microbiology and Biotechnology 54, 445–448. Lovley, D.R. 1993 Dissimilatory metal reduction. Annual Review of Microbiology 47, 263–290. Lovley, D.R. & Phillips, E.J.P. 1994 Reduction of chromate by Desulfovibrio vulgaris and its c3 cytochrome. Applied and Environmental Microbiology 60, 726–728. Mahajan, S.P. 1985 Removal of chromium. In Pollution Control in Process Industries, pp. 82–89. New Delhi, India: Tata McGrawHill Publishing Company Limited. ISBN 0-07-451772-4. McLean, J. & Beveridge, T.J. 2001 Chromate reduction by a Pseudomonas isolated from a site contaminated with chromated copper arsenate. Applied and Environmental Microbiology 67, 1076–1084. Nekolny, D. & Chaloupka, J. 2000 Protein catabolism in growing Bacillus megaterium during adaptation to salt stress. FEMS Microbiology Letters 184, 173–177. Ohtake, H., Fujii, E. & Toda, K. 1990 Reduction of toxic chromate in an industrial effluent by use of a chromate-reducing strain of Enterobacter cloacae. Environmental Technology 11, 663–668. Ryan, M.P., Williams, D.E., Chater, R.J., Hutton, B.M. & McPhail, D.S. 2002 Why stainless steel corrodes. Nature (London) 415, 770–774. Sharma, D.C., Chatterjee, C. & Sharma, C.P. 1995 Chromium accumulation and its effects on wheat ðTriticum aestivum L. cv. HD 2204) metabolism. Plant Science 111, 145–151. Slepecky, R.A. & Hemphill, H.E. 1999 The genus Bacillus – nonmedical. In The Prokaryotes: An Evolving Electronic Resource for the Microbiological Community, 3rd edn., ed. Dworkin, M. et al., release 3.7, 21 May 1999, New York: Springer-Verlag (www. prokaryotes.com).

Chromate reduction by Bacillus sp. Suzuki, T., Miyata, N., Horitsu, H., Kawai, K., Takamizawa, K., Tai, Y. & Okazaki, M. 1992 NAD(P)H-dependent chromium(VI) reductase of Pseudomonas ambigua G-1: a Cr(V) intermediate is formed during the reduction of Cr(VI) to Cr(III). Journal of Bacteriology 174, 5340–5345. Tang, Y.-F., Xu, J.-H., Ye, Q. & Schulze, B. 2001 Biocatalytic preparation of (S)-phenyl glycidyl ether using newly isolated Bacillus megaterium ECU1001. Journal of Molecular Catalysis 13, 61–68. US Environmental Protection Agency (US EPA) 1998 Toxicological review of hexavalent chromium. Washington DC: US Environmental Protection Agency.

219 Valls, M., Atrian, S., de Lorenzo, V. & Fernandez, L.A. 2000 Engineering a mouse metallothionein on the cell surface of Ralstonia eutropha CH34 for immobilization of heavy metals in soil. Nature Biotechnology 18, 661–665. Wang, Y.-T. & Shen, H. 1995 Bacterial reduction of hexavalent chromium. Journal of Industrial Microbiology 14, 159–163. White, C., Sayer, J.A. & Gadd, G.M. 1997 Microbial solubilization and immobilization of toxic metals: key biogeochemical processes for treatment of contamination. FEMS Microbiology Reviews 20, 503–516.

 Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 229–231 DOI 10.1007/s11274-004-5299-x

Formation of Yersinia pseudotuberculosis biofilms on multiple surfaces on Caenorhabditis elegans Marc A. Nascarella* and Steven M. Presley The Institute of Environmental & Human Health, Department of Environmental Toxicology, Texas Tech University, MS 41163, Lubbock, TX 79409-1163, USA *Author for correspondence: Tel.: +1-806-885-4567, Fax: +1-806-885-4577, E-mail: [email protected] Received 1 October 2004; accepted 25 October 2004

Keywords: Biofilm, Caenorhabditis elegans, endotokia matricide, Yersinia pseudotuberculosis

Summary A liquid-based assay was used to evaluate the ability of Yersinia pseudotuberculosis to form a bacterial biofilm on the nematode Caenorhabditis elegans. After 3 days of incubation in the liquid assay a biofilm was clearly visible by light microscopy on both the head and vulva region of the worms. At times, the biofilm formation on the vulva appeared to prevent the laying of eggs by the adult hermaphrodite; the eggs would later hatch inside of the worm. One possible explanation for the biofilm formation observed on the vulva may be the increased motion of the cuticle surrounding the vulva when the worm is immersed in a liquid culture. This is the first report of biofilm formation on the vulva of C. elegans.

Introduction

Materials and methods

Bubonic plague is a zoonotic disease transmitted by fleas infected with the bacterium Yersinia pestis. The disease is spread rapidly due to a Y. pestis-induced extracellular mass that obstructs the flea’s alimentary canal, and results in repeated attempts to feed by the flea (Bacot & Martin 1914). Recently (Hinnebusch 2003), this extracellular matrix has been termed a biofilm (Costerton et al. 1978). Similarly, Yersinia pseudotuberculosis will form an extracellular biofilm on the anterior cuticle of the nematode Caenorhabditis elegans and inhibit feeding (Darby et al. 2002). Thus, the Y. pseudotuberculosis–C. elegans model of biofilm formation provides an experimental system in which to study the obstruction of the fleas digestive tract by Y. pestis (Darby et al. 2002; Joshua et al. 2003; Tan & Darby 2004). Biofilm accumulation on C. elegans was previously determined to be dependent upon movement through a bacterial lawn (Tan & Darby 2004). The formation of this movement-dependent biofilm was also limited to the anterior portion of the worms. Here we show that worms infected in a liquid-based assay produce biofilms on both the anterior (head) and vulva of the worm. To our knowledge, this is the first report of a biofilm on the vulva of C. elegans.

Bacterial strains and growth media Yersinia pseudotuberculosis YPIII (a gift of Dr. Creg Darby, University of Alabama, Birmingham, AB, USA) was cultured in 5 ml of Luria-Bertani broth (LB). The culture was incubated overnight (17 h) at 28
C. The C. elegans wild-type N2 (var. Bristol) and E. coli OP50 strains that were used in this study were obtained from the Caenorhabditis Genetics Center (CGC, Biological Sciences Center, University of Minnesota, 1445 Gortner Ave., St. Paul, MN, USA). Nematodes were cultured at 20
C on nematode growth medium (NGM) that had been previously seeded with E. coli OP50 and cultured overnight at 37
C (Sulston & Hodgkin 1988). To maintain the nematode colony, cultures were transferred to freshly seeded plates every 7 days. Synchronizing C. elegans cultures Age-synchronized cultures of C. elegans were obtained following exposure to a sodium hypochlorite (bleach)/ sodium hydroxide solution that has been previously described (Sulston & Hodgkin 1988). Four days prior to the assay, overgrown plates of C. elegans were washed three times with 5 ml of K-medium (3.075 g NaCl,

230 2.42 g KCl, and 1 l of distilled water). The rinsing liquid following each wash was combined in a 15 ml centrifuge tube. The worms were then pelleted by centrifugation at 3000 g for 7 min and resuspended in 10 ml of the sodium hypochlorite/sodium hydroxide solution. The suspension was allowed to sit for 7 min and received careful shaking every few minutes, killing all the nematodes except for the eggs. The suspension was then pelleted again (2000 g for 7 min), and washed three times in K-medium. The final pellet was resuspended in approximately 1 ml of K-medium and incubated at 20
C for 1 day on an NGM plate. On day 2 the worms were washed off the plate with K-medium and transferred to a seeded NGM plate until the worms had reached adulthood.

M.A. Nascarella and S.M. Presley mitted light. Photomicrographs (Figure 1) were taken using a digital camera after viewing C. elegans on 1% agarose pads (Sulston & Horvitz 1977) with a 20· objective, on a compound light microscope (Olympus BX-51) using transmitted light. In some instances (Figure 1b & c), worms were first stained with 0.1% crystal violet.

Results

Microscopy

After approximately 3 days of incubation in the test solution a biofilm was observed on both the head and vulva of the worms (Figure 1). Worms that had formed anterior biofilms were easily identified as the worms would violently thrash in an apparent attempt to free themselves from the obstruction. Often times, multiple worms would become joined at the head or vulva and form an asterisk (*) appearing structure (Figure 1a). At times, the biofilm formation on the vulva appeared to prevent the laying of eggs by the adult hermaphrodite; the eggs would later hatch inside of the worm (Figure 1b & c). After 7 days the biofilm on the vulva became increasing thick with material accumulating on top of the initially formed layer (Figure 1d). The accumulation of biofilm appeared similar to what has been observed with C. elegans infected with agar lawn-based Yersinia infections (Darby et al. 2002; Joshua et al. 2003; Tan & Darby 2004).

Worms were observed daily for biofilm formation using a stereomicroscope with epi-illumination and/or trans-

Discussion

Figure 1. Biofilm growth on adult Caenorhabditis elegans hermaphrodites following infection by Yersinia pseudotuberculosis in a liquid medium. (a) A group of larvae that have become joined at the head by a biofilm mass. (b, c) The vulva of this single worm has become obstructed (see arrow on inset b) and the eggs have hatched into larvae that are trapped within the hermaphrodite. (d) After 5 days of infection the vulva has become infected with a large biofilm mass and additional worms have become trapped.

The biofilm formation on the vulva may not have led to the hermaphrodite progeny hatching inside of the adult worm. Internal hatching, or endotokia matricida as it has been called in the general nematode literature (Johnigk & Ehlers 1999), has previously been observed in C. elegans. This phenomenon, generally referred to as producing ‘bags of worms’ (Trent et al. 1983) or more simply as ‘bagging’, will usually occur as a response to a defect in vulval cell development, inadequate motility of the vulval muscles, or will occasionally occur in older animals (Kornfeld 1997). However, bagging has also been observed in response to stressors such as antimicrobials, high salt, and antagonistic bacteria (Chen & Caswell-Chen 2003, 2004). The cited benefits of bagging under stressful or food-limited situations have been to provide adequate nutrition (by way of the adult’s internal organs) as to allow larvae to become dauers. It has therefore been suggested (Chen & Caswell-Chen 2004) that bagging may actually be a type of facultative vivipary and a normal important life-history trait of C. elegans where parental resources are offered to their progeny. Therefore, it is difficult to say with complete certainty that the bagging observed in these experiments was not a secondary response to biofilm formation that followed the food-limiting situation created by the anterior biofilm blocking food intake. We feel confident

Infection assay Two 5 ml borosilicate tubes containing an overnight (17 h) culture of Yersinia pseudotuberculosis in LB Broth was centrifuged at 2000 g for 15 min and the supernatant was removed. The pellet was resuspended in 10 ml of K-medium and 1 ml of the resulting solution was transferred into a standard 24-well tissue culture plate (Falcon 3047). Approximately 10 adult worms were pipetted into each well. The plate was then placed at 20
C, in the dark.

Biofilm growth on C. elegans that this is not a secondary response, as bagging has not been reported in similar lawn-based assays (Darby et al. 2002; Joshua et al. 2003; Tan & Darby 2004). It seems likely that the bagging is a primary response to biofilm formation, whereby biofilm formation on the vulva prevented the laying of eggs, either by physically obstructing the vulval opening or by preventing the motility of vulval muscles. It is unlikely that the bagging observed here was due to the age of the worms, as similar experiments have been conducted in this lab with E. coli OP50 and no bagging was observed. Why a worm’s vulva would become preferentially infected in a liquid-based assay and not on an agarbased assay is not known. Tan & Darby (2004) have shown that C. elegans movement is necessary in order for a Yersinia spp. biofilm to form. They also speculate that the necessary movement in the biofilm-mediated transfer of Y. pestis in an infected flea is provided by the peristaltic action of the flea’s digestive and feeding muscles. One possible explanation for the dichotomy observed in the two infection phenotypes (liquid versus agar-based infection assays) may be the increased motion of the cuticle surrounding the vulva when the worm is immersed in a liquid assay. The increased thrashing of the worm may provide the necessary movement around the centre of the worm’s cuticle for the biofilm to form. In fact, the cuticle is located almost exactly at the apex of the thrashing ‘V’ that is formed when a worm is added to a liquid solution (see Figure 1d). However, the physiology of the vulva has some apparent interaction as the infection is specifically located on the vulva and is not observed on the worms cuticle located 180
from the vulva. Future work involves the use of lectins as in situ probes for earlier detection of biofilm binding (Tan & Darby 2004). Future work will also evaluate host resistance factors by evaluating the response of C. elegans mutants; mutations leading to aberrations in feeding, surface recognition, and movement (Joshua et al. 2003).

Acknowledgements The authors gratefully acknowledge Dr. Creg Darby, University of Alabama, Birmingham, for suggestions on laboratory methods, for providing a copy of a manu-

231 script prior to publication, and for providing the Yersinia pseudotuberculosis III strain used in this study; and Dr. Phil Williams, University of Georgia, Athens, for his assistance with the cultivatoin of C. elegans. Partial support for this work was provided by a ColgatePalmolive/Society of Toxicology Award for Research Training in Alternative Methods. The nematode strain used in this work was provided by the Caenorhabditis Genetics Center, which is funded by the National Institutes of Health (NIH) National Center for Research Resources (NCRR).

References Bacot, A.M. & Martin, C.J. 1914 Observations on the mechanism of the transmission of plague by fleas. Journal of Hygiene Plague Supplement 3, 423–439. Costerton, J.W., Geesey, C.G. & Cheng, K. 1978 How bacteria stick. Scientific American 238, 86–95. Chen, J. & Caswell-Chen, E.P. 2003 Why Caenorhabditis elegans adults sacrifice their bodies to progeny. Nematology 5, 641–645. Chen, J. & Caswell-Chen, E.P. 2004 Facultative Vivipary is a lifehistory trait in Caenorhabditis elegans. Journal of Nematology 36, 107–113. Darby, C., Hsu, J.W., Ghori, N. & Falkow, S. 2002 Caenorhabditis elegans: plague bacteria biofilm blocks food intake. Nature 417, 243–244. Hinnebusch, B.J. 2003 Transmission factors: Yersinia pestis genes required to infect the flea vector of plague. Advances in Experimental Medicine and Biology 529, 55–62. Johnigk, S.-A. & Ehlers, R.-U. 1999 Endotokia matricida in hermaphrodites of Heterorhabditis spp. and the effect of the food supply. Nematology 1, 717–726. Joshua, G.W., Karlyshev, A.V., Smith, M.P., Isherwood, K.E., Titball, R.W. & Wren, B.W., 2003 A Caenorhabditis elegans model of Yersinia infection: biofilm formation on a biotic surface. Microbiology. 149, 3221–3229. Kornfeld, K. 1997 Vulval development in Caenorhabditis elegans. Trends in Genetics 13, 55–61. Sulston, J. & Hodgkin, J. 1988. Methods, In The Nematode Caenorhabditis elegans, ed. Wood, W.B. pp. 587–606 Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, ISBN 0-87969-433-5. Sulston, J.E. & Horvitz, H.R. 1977 Post-embryonic cell lineages of the nematode, Caenorhabditis elegans. Developmental Biology 56, 110– 156. Tan, L. & Darby, C. 2004 A movable surface: formation of Yersinia sp. biofilms on motile Caenorhabditis elegans. Journal of Bacteriology 186, 5087–5092. Trent, C., Tsung, N., & Horvitz, H.R. 1983 Egg-laying defective mutants of the nematode Caenorhabditis elegans. Genetics 104, 619–647.

World Journal of Microbiology & Biotechnology (2005) 21: 245–251 DOI 10.1007/s11274-004-3624-z


Springer 2005

Use of RAPD to investigate the epidemiology of Staphylococcus aureus infection in Malaysian hospitals V. Neela1,*, N.S. Mariana1, S. Radu3, S. Zamberi1, A.R. Raha3 and R. Rosli2 1 Department of Clinical Laboratory Sciences 2 Department of Human Growth and Development, Faculty of Medicine and Heath Sciences 3 Department of Biotechnology, Faculty of Food Science and Biotechnology, University Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia *Author for correspondence: Tel.: +603-89468497, Fax: +603-89412787, E-mails: [email protected], [email protected] Received 5 April 2004; accepted 12 July 2004

Keywords: Dendrogram, MRSA, non-MRSA, randomly amplified polymorphic DNA, Staphylococcus aureus

Summary Randomly amplified polymorphic DNA (RAPD) with four different decamer oligonucleotide primers was performed on 50 clinical Staphylococcus aureus isolates obtained from different hospitals in Malaysia. All the four primers generated polymorphisms in all 50 isolates of S. aureus studied, revealing DNA markers with sizes ranging from 100 to 7000 bp. The dendrogram generated from the RAPD analysis revealed two major groups (Groups I–II) with three clusters each in one group. S. aureus strains isolated from the same hospital were found to be genetically closely related and most of them were placed in the same cluster. In addition RAPD differentiated between MRSA and non-MRSA based on the clustering, where all MRSA and non-MRSA were placed in their respective clusters. The RAPD analysis showed that there could be four to five clones of S. aureus spreading around Malaysia, of which two clones may be MRSA. The overall genetic distances ranged from 0.088235 to 0.954545 among the isolates. This technique was found to be a simple and effective method for epidemiological investigation. Because of these efficient features, this technique may have more general application for the study of S. aureus infections in hospitals and the community.

Introduction Members of the bacterial genus Staphylococcus are commonly found in and on the human and animal body, and are frequently isolated as etiological agents of infectious processes. Among the staphylococci Staphylococcus aureus is considered as the most important human pathogen of this group (Kloos & Bannerman 1999). Although S. aureus has remained a prime pathogen of nosocomial and community-acquired infections, the rising prevalence of methicillin-resistant Staphylococcus aureus (MRSA) globally has become a major clinical problem (Chambers 1997). Today MRSA strains have become resistant to a wide range of other antibiotics and are the major cause of nosocomial infections throughout the world (Brumfitt & HamiltonMiller 1989; Boyce 1994). In Malaysia the incidence of MRSA increased from 28% in 1992 to 40% in 1999 (Cheong et al. 1994; Rohani et al. 2000). Specific and rapid epidemiological typing is necessary to study the epidemiology of S. aureus strains. The limitations of phenotypic methods have stimulated the development of DNA-based techniques. The first genotypic method

identified to study the epidemiology of S. aureus was plasmid profile analysis (Locksley et al. 1982; Tenover et al. 1994). Currently, analysis of chromosomal DNA with a variety of techniques to study the epidemiology of S. aureus has been documented. These include restriction fragment length polymorphism analysis, ribotyping, binary typing (van Leeuwen et al., 2001), pulsed-field gel electrophoresis (Bannerman et al. 1995) and PCR-based methods (DelVecchio et al. 1995; Van Belkum et al. 1995). Randomly amplified polymorphic DNA (RAPD) typing (Williams et al. 1990) also called arbitrarily primed polymerase chain reaction (AP-PCR) (Welsh & McClelland 1990) is a technique suited for rapid detection of genomic polymorphism. This technique is based on the amplification of the genomic DNA with a single short oligonucleotide primer of arbitrary or random sequence. This technique plays a vital role in epidemiological studies. RAPD has frequently been used to study the epidemiological spread of nosocomial infections with MRSA strains (Hojo et al. 1995; Van Belkum et al. 1995; Kobayashi et al. 1997; Trzcinski et al. 1997; Tambic et al. 1999; Telecco et al. 1999). This method is

246

V. Neela et al.

attractive, because of its ease of performance and the ability to be applied to any organism (Power 1996). In this study we used the RAPD molecular typing method to investigate the epidemiology of S. aureus in Malaysia.

Materials and methods Bacterial isolates Fifty isolates of Staphylococcus aureus (Table 1) were obtained from different patients visiting Hospital Tunku Ampuan Afzan in Kuantan (HTAA), Hospital Seremban, Hospital Miri in Sarawak, Hospital University Sains Malaysia Kota Bharu (HUSM) and from University Malaya Medical Center (UMMC). Of the 50 isolates, 21 isolates were MRSA and 29 were nonMRSA (Table 2). All the isolates were stored in Luria Bertanii broth with 30% glycerol. All the isolates were cultured on Blood Agar plates to isolate single colonies. DNA preparation Genomic DNA from each isolate was extracted from an overnight culture by using the bacterial Genomic DNA extraction (GeniSpin, Biosyntech Sdn. Bhd, Malaysia) protocol. The purity and quality of the DNA were determined by u.v. absorption with a u.v. spectrophotometer (Shimadzu UV-1601). 100 ng of genomic DNA was used for RAPD analysis. Primers DNA primers OPAE-06, 10, 14 and 15 were obtained from Operon 10-mer Kit (Kit AE, Operon Technologies

Table 1. Isolates obtained from different hospitals in Malaysia. Source of isolates

Isolate No:

Hospital Miri, Sarawak

1,2,3,4,5,6,7,8,9,10,11,12,13,14,15, 16,17,18,19,20 N1,N2,N3,N4,N5,N6,N7,N8,N9, N10 N11,N12,N13,N14,N15,N16,N17

Hospital Tunku. Ampuan Afzan (HTAA), Kuantan Hospital University Science Malaysia (HUSM), Kota Bharu University Malaya Medical Center (UMMC), Kuala Lumpur Hospital Seremban, Seremban

Inc.,) containing 10-base oligonucleotides, primers were selected based on G+C content ranging from 60 to 70%. RAPD finger printing Amplification reaction was performed in 25 ll volume mixtures containing 80 mM MgCl2, PCR buffer, 3.75 mM dNTP mix (MBI Fermentas), 10 picomoles of each RAPD primers, 100 ng of template and 1Unit of Taq polymerase (BST Taq, Biosyntech Sdn. Bhd, Malaysia). Amplifications were carried out by using a thermal cycler (Biometra-Trio Thermoblock) programmed for 1 min at 94 C followed by 30 cycles, each consisting of 1 min at 94 C, 1 min at 36 C and 2 min at 72 C and a final extension period of 7 min at 72 C. Ten microlitre of amplified PCR products were separated by electrophoresis through 1.4% agarose gels in 1 · TBE buffer for 2.30 h at 65 V. A1 kb DNA ladder (MBI Fermentas) was used as a DNA size standard. The gel was then stained with ethidium bromide and photographed under u.v. illumination. The DNA from each isolate was subjected to the RAPD assay at least three times to ensure reproducibility of the results. RAPD data analysis The fingerprint was analysed both by visual inspection and by computer-aided methods. The RAPD analysis was based on the manual scoring for the presence or absence of bands, as observed for the RAPD banding profile obtained. Only the clear, prominent and reproducible bands from repeats of the experiment (at least three times) were given consideration showing the true polymorphism. Bands which appeared in the RAPD primer amplification are denoted by 1 while the absence of a band is denoted by 0. An input matrix was produced by entering the data regarding presence (1) or absence (0) of an amplified fragment for Jaccards’s pairwise genetic distance analysis and to generate the neighbour-joining tree calculated by the unweighted pair group method with arithmetic averages. The genetic relationships between the isolates are represented graphically by the dendrogram.

Results S12, S13

S1,S2,S3,S4,S5,S6,S7,S8,S9,S10,S11

Table 2. MRSA and non-MRSA isolates used in this study. MRSA

Non-MRSA

S1,S13,1,3,4,5,6,7,8,9,10,11,12, 13,14,15,16,17,18,20 and N7

S2, S3,S4,S5,S6,S7,S8,S9,S10,S11, S12, N1,N2,N3,N4,N5,N6,N8,N9, N10,N11,N12,N13,N14,N15,N16, N17, 2,19

Fifty isolates of S. aureus obtained from different hospitals in Malaysia were fingerprinted by the RAPD technique. RAPD analysis with four primers yielded 1– 11 bands per isolate, revealing DNA markers ranging from 100 to 7000 bp. Representative profiles of the reproducible bands with primers OPAE 10, 14 and 15 for the isolates used in this study are presented in Figures 1–3. Isolates that are closely related to each other will have a lower genetic distance value, whereas isolates that are not closely related tend to have higher relative genetic distance value. The genetic distances between the 50 S. aureus strains calculated by the

Epidemiological investigation of Staphylococcus aureus by RAPD

247

Figure 1. RAPD genetic profile obtained with primer OPAE 10. Lanes 1–20: 1–20 Staphylococcus aureus isolates from Hospital Miri, lanes 21– 37: N1-N17 isolates from Hospital Tengku Ampuan Afzan (HTAA) Kuatan, Pahang, lanes 38–48: Hospital Seremban and lanes 49–50: University Malaya Medical Center (UMMC). Lane M represents 1kb molecular weight marker (MBI Fermentas).

Figure 2. RAPD genetic profile obtained with primer OPAE 14. Lanes 1–20: 1–20 Staphylococcus aureus isolates from Hospital Miri, lanes 21– 37: N1–N17 isolates from Hospital Tengku Ampuan Afzan (HTAA) Kuatan, Pahang, lanes 38–48: Hospital Seremban and lanes 49–50: University Malaya Medical Center (UMMC). Lane M represents 1kb molecular weight marker (MBI Fermentas).

Figure 3. RAPD genetic profile obtained with primer OPAE 15. Lanes 1–20: 1–20 Staphylococcus aureus isolates from Hospital Miri, lanes 21– 37: N1–N17 isolates from Hospital Tengku Ampuan Afzan (HTAA) Kuatan, Pahang, lanes 38–48: Hospital Seremban and lanes 49–50: University Malaya Medical Center (UMMC). Lane M represents 1 kb molecular weight marker (MBI Fermentas).

Jaccard’s index ranged from 0.088235 for isolate no: 7 from Hospital Miri to 0.954545 for isolate S12 from UMMC. The dendrogram obtained with the combined data of four RAPD primers generated two main groups, groups I and II with three main clusters (A, B, C and D, E, F) for each group. Cluster A comprised of 65% of Hospital Miri isolates (isolate number 1–13). Cluster B consisted of 40% Kuantan isolates (N2, N3, N6 and N7) and finally cluster C also contained 40% Kuantan isolates (N4, N5, N8 and N9). Cluster B and C of group I contained 80% of Kuantan isolates. Cluster D of

group II is divided into two sub-clusters (Di and Dii). The sub-cluster Di contained 10% of Hospital HTAA, Kuantan isolates (isolate no N10) and 71.4% of HUSM (N11, N12, N13, N14, N15). The sub-cluster Dii contained 28.5% of HUSM isolates (N16 and N17), 100% of Hospital Seremban isolates (S1–S11) and 100% of UMMC isolates (S12 and S13). The Cluster E of Group II is comprised of 35% of Miri isolates (isolate number 14–20). Finally the Cluster F of Group II contained 10% of HTAA isolates (N1). The dendrogram obtained although showing that the isolates are

248 widely disseminated, most of the isolates were grouped according to their geographical locations. Cluster A of group I (65%) and cluster E (35%) of group II covered all the Hospital Miri isolates (65 + 35% ¼ 100%) which shows that the isolates may have emerged from two different clones from Miri. In the case of Kuantan isolates N1–N10, cluster B contained N2, N3 N6 and N7 and cluster C contained N4, N5, N8 and N9 (80% in cluster B and C). The remaining two other isolates N1 and N10 are in group II (20%). The Kuantan isolates except for N1 and N10 belong to the same clone although subdivided into two clusters. Among the seven HUSM isolates, five isolates (71.4%) N11, N12, N13, N14 and N15 belonged to cluster D and two isolates (28.6%) N16 and N17 belonged to cluster E together with Hospital Seremban isolates. All the eleven Seremban isolates are grouped into the same cluster E showing that they have emerged from the same clone. The dendrogram obtained clearly shows that most of the isolates were grouped according to their geographical location. RAPD was also able to distinguish MRSA and non-MRSA isolates. The dendrogram shows that all of the isolates placed in cluster A, E and F were MRSA, whereas isolates belonging to cluster B, C and D except for isolate N7 in cluster B, S1 and S13 from cluster D were non-MRSA. The isolates that were closely related to each other will have a higher percentage of similarity and vice versa. The scale above the dendrogram (Figure 4) indicates the similarity index. From the dendrogram the highest percentage of similarity was found between isolates S12 and S13, which showed that these two isolates were 68% similar.

Discussion AP-PCR is widely used for the comparison of genomes from eucaryotes (Welsh et al. 1991) or bacteria (Ralph et al. 1993). From the results obtained in this investigation, it was found that different banding patterns of the amplified products generated by different primers, had allowed the genotyping of the Staphylococcus aureus isolates. The fingerprints generated by the four different primers (OPAE-06, 10, 14 and 15) revealed unique profiles for each strain in terms of number and position of RAPD bands. The dendrogram (Figure 4) obtained in the current investigation carried out with 50 isolates of S. aureus from five different hospitals in Malaysia, showed distinct clustering of these isolates corresponding to the geographical location. The dendrogram generated two main groups with three main clusters each, whereby S. aureus strains isolated from the same hospital (same geographical location) were found to be closely related, and were placed in the same cluster and vice versa. All the isolates obtained from Hospital Seremban were placed in the subcluster Dii of the main cluster D of group II. The two UMMC isolates S12 and S13 and two HUSM isolates N16 and N17 were also clustered together with the Hospital Seremban

V. Neela et al. isolates. The two UMMC isolates, although clustered together with Hospital Seremban isolates, are shown by the dendrogram to be very closely related. The reason for clustering of HUSM isolates N16 and N17 with Hospital Seremban isolates could be that the patient might have been infected in Seremban and has taken treatment in HUSM. However, unless the patient history is known, it is difficult to trace the origin of the isolate. In the case of the Hospital Miri isolates, the isolates were clustered into two different groups nos; 1–13 in cluster A of group I and nos: 14–20 in cluster E of group II. The clustering of the Miri isolates into two different groups shows that the Miri isolates might have emerged from two different clones. On the other hand, the clustering of Kuantan (HTAA) isolates in Group I, except for two isolates N1 and N10, which were contained in group II shows that HTAA isolates other than N1 and N10 are closely related. In conclusion, when the number of possible clones spreading around Malaysia is considered, it can be assumed that there are 4–5 different clones, 2 clones from Miri, and one clone each from Kuantan, Kelantan and Seremban. Monitoring the geographic expansion of such clones is important in understanding why certain clones were spread over considerable distances, whereas others were limited to a single state or hospital. RAPD analysis, apart from genotyping the isolates according to the geographical location, was also able to discriminate between MRSA and non-MRSA. Although there was a slight discrepancy for five isolates (2, 19 N1, S1 and S13) which were not placed according to their type, most of isolates (90%) were clearly distinguished as MRSA and non-MRSA (Figure 4). The reason for clustering of two (isolate number 2 and 19) non-MRSA isolates from Hospital Miri with MRSA could be that these two isolates in future due to the extensive pressure by the usage of the antibiotic Oxacillin or Cloxacillin (the common antibiotic used in Malaysia for treating S. aureus infection) may result in exhibiting methicillin resistance. On the other hand the reason for MRSA to cluster with non-MRSA, could be that these three isolates may have very different Staphylococcal cassette chromosome (SCCmec) types from the other isolates. The RAPD analysis on differentiating MRSA and non-MRSA shows that there might be two MRSA clonal lineages emerging from Sarawak (West Malaysia) and spreading all over Malaysia. However the spread of 2 MRSA clones in Malaysia has to be confirmed by typing and studying a greates number of S. aureus isolates from other hospitals in Malaysia. The higher relative genetic distance between the two isolates (7 and S12) shows that these two isolates are very distantly related to each other, whereas, the highest percentage of similarity of 68% between isolates S12 and S13 from UMMC, indicate that they are closely related to each other. This could be because the two isolates S12 and S13 were obtained from the same hospital and may have originated from the same clone. The visual observation of the RAPD banding pattern was not able to distinguish between isolates from

Epidemiological investigation of Staphylococcus aureus by RAPD

249

Figure 4. Dendrogram of genetic relationship between fifty isolates of Staphylococcus aureus obtained with four RAPD primers. The scale above indicates the similarity index.

different hospitals, but the dendrogram generated based on the sharing of bands between the isolates was able to clearly distinguish between isolates from different geographical locations as well as differentiating between MRSA and non-MRSA. Similar observations were made by Van Belkum et al. (1995), in their study on multi-centre evaluation of arbitrarily primed PCR for

typing S. aureus strains and they reported that AP-PCR banding, although it differed markedly between laboratories for the same strain, the clustering of strains into related patterns was more reproducible. As S. aureus is one of the most frequently occurring nosocomial pathogens, and the emergence of MRSA has become a problem worldwide (Doebbling et al. 1995;

250 Chambers 1997) specific and rapid epidemiological typing is necessary for tracking inter-hospital spread and evolution of strains of MRSA. The high degree of genetic relatedness between the MRSA isolates has proven to be an obstacle for epidemiological analysis. Thus strict use of microbiological monitoring and epidemiological investigation are essential for controlling MRSA in hospitals (Brumfitt & Hamilton-Miller 1989; Boyce 1994). Although an antibiogram provides useful information for routine surveillance (Mulligan & Arbeit 1991; Rossney et al., 1994; Tenover et al., 1994), additional typing should be performed to investigate the outbreaks suspected of being caused by MRSA. There are several phenotypic and genotypic methods available for typing MRSA, but PFGE of chromosomal digests with infrequently cutting enzymes has been considered as the gold standard for typing MRSA (Schwartz & Cantor 1984; Prevost et al., 1992; Schlichting et al., 1993). However, in the current study RAPD was chosen to study the epidemiology of S. aureus in Malaysia, as the RAPD technique has already been validated by many researchers (Hojo et al., 1995; Van Belkum et al., 1995; Kobayashi et al., 1997; Trzcinski et al., 1997) in typing S. aureus isolates, moreover it is simple, rapid and easy to perform compared to PFGE. Tambic et al. (1997) reported that when RAPD and PFGE were applied to type MRSA, both the techniques gave similar results, but the PFGE technique is tedious and timeconsuming. The information obtained from typing S. aureus isolates can help the infection control teams to understand the epidemiology of this organism in the respective hospitals. By comparing the fingerprints or the clustering pattern of the S. aureus isolates, the spread of MRSA can be monitored within the hospital and between the hospitals as previously mentioned by Norazah et al. (2001). Thus the overall results obtained from this study demonstrated the practical value of RAPD for studying the genetic profile of S. aureus isolates for molecular epidemiology purposes. This study gives us the clear indication that RAPD analysis is a rapid, accurate and highly reliable tool that can show the genetic relatedness among the S. aureus strains and can group the isolates according to the geographical location and determine the genetic diversity of strains that are otherwise impossible by biochemical analysis. RAPD improves the understanding of the epidemiology of S. aureus isolates and thus aids the formulation of effective control measures. As RAPD analysis can correctly type S. aureus isolates, this technique would be of great use in preventing nosocomial S. aureus infections and thus could be applied in hospitals.

Acknowledgements The authors thank Hospital Tunku Ampuan Afzan, Kuantan (HTAA); Hospital Seremban; Hospital Miri, Sarawak; Hospital University Sains Malaysia Kota Bharu (HUSM); and Prof Dr Sazaly Abu Bakar from University

V. Neela et al. Malaya Medical Center (UMCC) for providing the S. aureus isolates. This study was supported by the Malaysian Government through the IRPA grant mechanism.

References Bannerman, R.L., Hancock, G.A., Tenover, F.C. & Miller, J.M. 1995 Pulsed-field gel electrophoresis as a replacement for bacteriophage typing of Staphylococcus aureus. Journal of Clinical Microbiology 33, 551–555. Boyce, J.M. 1994 Methicillin-resistant Staphylococcus aureus: a continuing infection control challenge. European Journal of Clinical Microbiology and Infectious Diseases 13, 45–49. Brumfitt, W. & Hamilton-Miller, J. 1989 Methicillin-resistant Staphylococcus aureus. New England Journal of Medicine 320, 1188–1196. Chambers, H.F. 1997. Methicillin resistance in staphylococci: molecular and biochemical basis and clinical implications. Clinical Microbiology Reviews 10, 781–791. Cheong, I., Tan, S.C., Wong, Y.H., Zainudin, B.M.Z. & Rahman, M.Z.A. 1994 Methicillin resistant Staphylococcus aureus (MRSA) in a Malaysian hospital. Medical Journal of Malaysia 49, 24–28. DelVecchio, V.G., Petroziello, J.M., Gress, M.J., McCleskey, F.K., Melcher, G.P., Crouch, H.K. & Lupski, J.R. 1995 Molecular genotyping of methicillin-resistant Staphylococcus aureus via fluorophore-enhanced repetitive-sequence PCR. Journal of Clinical Microbiology 33, 2141–2144. Doebbling, B.N. 1995 The epidemiology of methicillin-ressitant Staphylococcus aureus infection and colonization. Journal of Chemotherapy 7(Suppl. 3), 99–103. Hojo, S., Fugita, J., Negayama, K., Ohnishi, T., Xu, G., Yamaji, Y., Okada, H. & Takahara, J. 1995 Clinical Utility of DNA Fingerprinting by Arbitrarily-Primed Polymerase Chain Reaction (AP-PCR) in Nosocomial infection caused by Methicillin-resistant Staphyloccus aureus. Kansenshogaku Zasshi (Japanese article) 69, 1272–1277. Kloos, W.E. & Bannerman, T.L. 1999 Staphylococcus and Micrococcus, pp. 264–282. In Manual of Clinical Microbiology 7th edn. eds. Murray, P.R., Baron, E.J., Pfaller, M.A., Tenover, F.C. & Yolken (s), R.C. Washington D.C. ASM Press. ISBN 1-66681-126-4. Kobayashi, M., Ohara-Nemoto, Y. & Kaneko, M. 1997 Genotyping of clinical isolates of MRSA by PFGE and AP-PCR. Kansenshogaku-Zasshi (Japanese article) 71, 620–627. Locksley, R.M., Cohen, M.L., Quinn, T.C., Tompkins, L.S., Coyle, M.B., Kirishra, J.M. & Counts, G.W. 1982 Multiply antibioticresistant Staphylococcus aureus: introduction, transmission and evolution of nosocomial infection. Annals of Internal Medicine 97, 317–324. Mulligan, M.E. & Arbeit, R.D. 1991 Epidemiologic and Clinical utility of typing systems for differentiating among strains of methicillinresistant Staphylococcus aureus. Infection Control and Hospital Epidemiology 12, 20–28. Norazah, A., Liew, S.M., Kamel, A.G.M. & Lim, V.K.E. 2001 DNA fingerprinting of methicillin-resistant Staphylococcus aureus by Pulsed-Field Gel electrophoresis (PFGE): Comparison of strains from 2 Malaysian hospitals. Singapore Medical Journal 42, 015–019. Power, E.G.M. 1996 RAPD typing in microbiology–a technical review. Journal of Hospital Infection 34, 247–265. Prevost, G., Jaulhae, B. & Piemont, Y. 1992 DNA fingerprinting by pulsed-field gel electrophoresis is more effective than ribotyping in distinguishing among methicillin-resistant Staphylococcus aureus isolates. Journal of Clinical Microbiology 30, 967–973. Ralph, D., McClelland, M. & Welsh, J. 1993 Leptospira species categorized by arbitrarily primed polymerase chain reaction (PCR) and by mapped restriction polymorphisms in PCR-amplified rRNA genes. Journal of Bacteriology 175, 973–981. Rohani, M.Y., Raudzah, A., Lau, M.G., Zaidatul, A.A.R., Salbiah, M.N., Keah, K.C., Noraini, A. & Zainuldin, T. 2000 Susceptibility

Epidemiological investigation of Staphylococcus aureus by RAPD pattern of Staphylococcus aureus isolated in Malaysian hospitals. International Journal of Antimicrobial Agents 13, 209–213. Rossney, A.S., Coleman, D.C & Keane, C.T. 1994 Evaluation of an antibiogram-resistogram typing scheme for methicillin-resistant Staphylococcus aureus. Journal of Medical Microbiology 41, 441–447. Schlichting, C., Branger, C., Fournier, J.M., Witte, W., Boutonnier, A., Wolz, C., Goullet, P. & Doring, G. 1993 Typing of Staphylococcus aureus by pulsed-field gel electrophoresis, zymotyping, capsular typing and phage typing: resolution of clonal relationships. Journal of Clinical Microbiology 31, 227–232. Schwartz, D.C. & Cantor, C.R. 1984 Separation of yeast chromosomesized DNAs by Pulsed field gradient gel electrophoresis. Cell 37, 67–75. Tambic, A., Power, E.G., Talsania, M., Anthony, R.M. & French, G.L. 1997 Analysis of an outbreak of non-phage-typeable methicillin-resistant Staphylococcus aureus by using a randomly amplified polymorphic DNA assay. Journal of Clinical Microbiology 35, 3092–3097. Tambic, A., Power, E.G.M., Tambic, T., Snur, I. & French, G.L. 1999 Epidemiological analysis of methicillin-resistant Staphylococcus aureus in a Zagreb Trauma hospital using a randomly amplified polymorphic DNA-typing method. European Journal of Clinical Microbiology and Infectious Diseases 18, 335–340. Telecco, S., Barbarini, D., Carretto, E., Comoncini, S., Emmi, V. & Marone, P. 1999 Typing of methicillin-resistant Staphylococcus aureus (MRSA) strains from an intensive care unit by random amplified polymorphic DNA (RAPD). New Microbiology 22, 323–329. Tenover, F.C., Arbeit, R., Archer, G., Biddle, J., Byrne, S., Goering, R., Hancock, G., Hebert, G.A., Hill, B., Hollis, R., Javis, W.R.,

251

Kreiswrith, B., Eisner, W., Maslow, J., MacDougat, L.K., Miller, J.M., Mulligan, M. & Pfaller, M.A. 1994 Comparison of traditional and molecular methods of typing isolates of Staphylococcus aureus. Journal of Clinical Microbiology 32, 407–415. Trzcinski, K., Hryniewicz, W., Kluytmans, J., van Leeuwen, W., Sijmons, M., Dulny, G., Verbrugh, H. & Van Belkum, A. 1997 Simultaneous persistence of methicillin-resistant and methicillinsusceptible clones of Staphylococcus aureus in a neonatal ward of a Warsaw hospital. Journal of Hospital Infection 36, 291–303. Van Belkum, A., Kluytmans, J., van Leeuwen, W., Bax, R., Quit, W., Peters, E., Fluit, A., Vandenbroucke-Grauls, C., van den Brule, A., Koeleman, H., Melchers, W., Meis, J., Elaichouni, A., Vaneechoutte, A., Moonens, F., Maes, N., Struelens, M., Tenover, F. & Verbrugh, H. 1995 Multi center evaluation of arbitrarily primed PCR for typing of Staphylococcus aureus strains. Journal of Clinical Microbiology 33, 1537–1547. van Leeuwen, N., Librregts, C., Schalk, M., Veuskens, J., Verbrugh, H. & van Belkum, A. 2001 Binary typing of Staphylococcus aureus strains through reverse hybridization using digoxigenin-universal linkage system-labeled bacterial genomic DNA. Journal of Clinical Microbiology 39, 328–331. Welsh, J. & McClelland, M. 1990 Fingerprinting genomes using PCR with arbitrary primers. Nucleic Acids Research 18, 7213–7218. Welsh, J., Petersen, C. & McClelland, M. 1991 Polymorphisms generated by arbitrarily primed PCR in the mouse: applications to strain identification and genetic mapping. Nucleic Acids Research 19, 303–306 Williams, J.G.K., Kubelik, A.R., Livak, K.J., Rafalski, J.A. & Tingey, S.V. 1990 DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Research 18, 6531–6535.

World Journal of Microbiology & Biotechnology (2005) 21: 253–259 DOI 10.1007/s11274-004-3625-y


Springer 2005

Isolation of Enterobacteria able to degrade simple aromatic compounds from the wastewater from olive oil extraction Emna Ammar1,*, Moncef Nasri2 and Khaled Medhioub1 1 UR Etude et Gestion des Environnements Urbains et Coˆtiers – LARSEN, Ecole Nationale d’Inge´nieurs de Sfax, B.P. ‘‘W’’–3038 Sfax, Tunisia 2 Laboratoire de Ge´nie Enzymatique et de Microbiologie, Ecole Nationale d’Inge´nieurs de Sfax, B.P. ‘‘W’’–3038 Sfax, Tunisia *Author for correspondence: Tel.: +216-98-412-364, Fax: +216-74-275-595, E-mail: [email protected] Received 12 February 2004; accepted 13 July 2004

Keywords: Biodegradation, Citrobacter diversus, Enterobacteria, Klebsiella oxytoca, olive oil mill, simple aromatic compounds, wastewater

Summary Enterobacteria growing on wastewater from olive oil extraction were selected. Among this microflora, strains of Klebsiella oxytoca and Citrobacter diversus able to degrade simple monomeric aromatic compounds were isolated by enrichment culture of the effluent lacking simple sugars. In this preliminary investigation, the phenolic acids tested on solid and liquid media were gentisic, protocatechuic, p-hydroxybenzoic, benzoic, vanillic and ferulic. It was shown that the biodegradation of an aromatic acid is tightly dependent on both the type and the position of the radical substituted on the aromatic ring. Citrobacter was the most efficient strain in metabolizing ferulic acid in liquid medium at a concentration of 1.5 g/l. The substrate biodegradation yield achieved exceeded 86%.

Introduction Olive oil extraction yield is generally around 22% (v/w). This amount is extracted after grinding 100 kg of olives mixed with a variable quantity of water ranging from 40 to 130%, depending on the chosen process, respectively either classical discontinuous pressing or a continuous system (Bosku 1996). This industrial activity generates two main by-products. About 40 kg of press-cake are obtained from 100 kg of processed olives (Karapinar & Worgan 1983). This solid phase contains the remaining oil: 5–8% and water: 25–55%. The major components of this solid waste are kernel cellulose, hemicelluloses and lignin at a rate of 42–54% and skin at 10% (Fedeli 1997). The press-cake is generally incorporated into food for animals after the removal of residual oil by hexane extraction (Ercoli & Ertola 1983; Tan & Gill 1984). The second by-product is the wastewater, which causes an important environmental problem related to olive oil extraction in Mediterranean countries. In fact, the volume of olive oil mill wastewater (OMW) varies from 70% in the discontinuous process to 170% in the continuous system. This effluent contains sugars: about 2.8%, nitrogen derivatives: 1.2–2.4%, organic acids: 0.5– 1.5%, poly-hydroxyl compounds: 1.0–1.5%, salts: 1.2– 1.8% and residual oil: 0.03–1.00%. Its biological oxygen demand (BOD5: 40–80 g/l) and chemical oxygen demand (COD: 70–210 g/l) are relatively high (Bosku 1996).

Furthermore, in many countries including olive oil producers, the maximum COD value allowed for discharge into the sewage or natural receptor media is about 50 mg/l (Japan International Cooperation Agency 1993; Ammar & Ueno 1999; Ben Rouina et al. 2001). Production of single cell protein on OMW was considered on a small scale (Fiestas 1966; Ercoli & Ertola 1983). More recently, the biodegradation of OMW aromatic compounds by four Pleurotus species showed the ability of those fungi to detoxify the effluent by oxidation of aromatic compounds by laccases (Sanjust et al. 1991; Tsioulpas et al. 2002; Aggelis et al. 2003). Aspergillus niger and Aspergillus terrus degrade the aromatic content of OMW with efficiency (RamosCormanzana et al. 1982; Hamdi et al. 1991). Furthermore, the degradation of phenol was performed by Candida tropicalis (Chang et al. 1998) and Trichosporon cutaneum (Shivarova et al. 1999). Also, the white-rot fungi were reported to be able to degrade a variety of persistent environmental pollutants, notably phenolic compounds and toxic substances with potential application for remediation purposes (Cameron et al. 2000). Despite the existence of some research on the efficiency of fungi in the degradation of aromatic compounds, investigations dealing with bacteria able to grow on such compounds, especially those abundant in OMW remain scarce. Indeed, apart from the isolation of

254 Pseudomonas strains able to degrade phenolic acids (Perez et al. 1990; Hinteregger et al. 1992), enrichment to yield an aerobic consortium (Zouari & Ellouz 1996), the screening of Pseudomonas metabolizing eugenol (Rabenhorst 1996) and the role of a TOL-like plasmid and a gene encoding a novel phenol hydroxylase in Klebsiella strain (Heesche-Wagner et al. 1999) no other studies have been made on this subject. Therefore, the microbial diversity of the isolates and aromatic compound biodegradation in the environment are not yet well understood (Ammar et al. 1986; Watanabe et al. 1998). It would be advantageous to treat the effluent from the olive oil extraction process using bacteria that are efficient in degrading the aromatic compounds responsible for the toxic effects on the aerobic consortium of activated sludge. The aim of this investigation was to isolate bacteria which can grow on OMW and which are especially able to degrade monomeric aromatic compounds, commonly found in OMW and having a toxic effect on microflora in the sludge plant. For this purpose, bacteria from the Enterobacteriaceae group were isolated, to our knowledge, for the first time. This work tried to assess the capacities of these bacteria in degrading some monomeric aromatic compounds.

Material and methods Materials The OMW used was collected from a storage tank of an olive oil extraction plant in Sfax (Tunisia), using Chemlali olive species (Argenson et al. 1999). This wastewater is produced by a continuous press system and has a COD of 70 g/l. The effluent containing initially 8.5 g total sugars/l was fermented by a yeast: Saccharomyces rouxii (Gharsallah 1993). Its sugar content was thereby reduced to less than 1.0 g/l. It was then centrifuged at 62,537 g at 4 C for 15 min and the supernatant was collected. Actidione (0.5 g/l) was added to this supernatant of OMW used as the culture media to inhibit the growth of any possible fungal spores during bacterial enrichment. For the characterization of different OMW, samples of effluent collected and analysed yearly over a decade were presented. Media and cultures The supernatant previously described was supplemented with 0.5 g vanillic acid (Fluka)/l and used as a culture medium. The purpose of vanillic acid use was to adapt the microorganisms to aromatic compounds with structures similar to those of lignin derivatives, which are abundant in the effluent. The pH was adjusted to 7.1 with NaOH before autoclaving. An aliquot of 0.1 l was sterilized in a 0.5 l baffled flask by autoclaving at 121 C for 20 min. The inocula used were sampled from natural environments in relation with the effluent: they were a mixture

Emna Ammar et al. in equal proportions (5%) of fermented OMW as well as press-cake, both collected from the olive-processing factory, and also from the soil of an olive tree field, where OMW has been used for years for ferti-irrigation. This type of soil is expected to exhibit many different types of microbial associations adapted to OMW, discharged for years in the field. Inoculated flasks were placed on an orbital incubator (Stuart Scientific S150) at 200 rev/min and at 25 C. Total aerobic mesophilic flora was enumerated on plate count agar medium after 72 h of incubation at 30 C and Enterobacteria on a specific culture medium: eosin methylene blue incubated at 30 C for 48 h. Identification of the bacterial strains The presumptive identification of the isolated strains degrading the OMW lacking sugars was determined using purified colonies according to the Bergey’s Manual of Systematic Bacteriology (Krieg & Holt 1983). The main characteristics for identification of these strains are listed and discussed. Routine biochemical tests and API 20E strips (Smith et al. 1972) were used in duplicate as a rapid system of identification and complementary tests were achieved using standard bacteriological methods developed by Dr Richard whose biotype classification was used as an efficient identification tool (Richard 1982). Modification of simple aromatic constituents in the medium Considering the OMW used in this study, we were first unable to confirm precisely whether the isolated organisms (bacteria) could grow only on the aromatic part of the effluent or on other substrates present in the medium. To demonstrate the aromatic nucleus degradation capacity of these isolates, this study had a 2-fold aim: first to show the capacity of bacteria to degrade and transform the aromatic structure of OMW, second to assess the ability of these bacteria to degrade the simple acid aromatic monomers present in OMW or obtained on biodegradation of the macroaromatic polymer. Low molecular weight aromatic compounds were separated on Silica gel G plates (Merck) with benzene/dioxan/ acetic acid (90: 25: 4, by vol.) as the whole phase and revealed with the Folin–Ciocalteu reagent (Randerath 1971). A volume of 1 ml of the medium was first extracted in methanol (1:2, v/v). Then, 20 ll of extracted organic compounds in the methanolic phase were used as a deposit for thin layer chromatography. The quality characterization of the OMW was made according to Japanese standards (JIS-Handbook 1995). Catabolism of aromatic acids The metabolism of aromatic compounds considered as OMW-related monomers was tested using a basal medium with 0.15% (w/v) of aromatic acid, as the sole

255

Isolation of bacteria degrading phenols carbon source. The basal mineral medium (MM) had the following composition (w/v): (NH4)2SO4, 0.25%; K2HPO4, 0.1%; KH2PO4, 0.05%; MgSO4, 0.02%; CaCl2, 0.001%; FeSO4, 0.0001% and yeast extract 0.01%. All compounds were sterilized separately by filtration with a Millipore syringe after adjustment of the pH to 7.1 and were then added to the MM solidified with 16 g agar/l in the case of test growth on solid medium. For growth in liquid medium, 100 ml of this broth were poured in 500 ml flask shaken at 200 rev/ min. Isolated strains were used separately to inoculate the medium and incubation were made at 25 C for 24 h. These tests were made simultaneously on solid medium and in shaking culture. On the former, the bacterial growth was assessed visually by comparing the growth to the observed biomass on the same solid medium lacking the carbon source and including the yeast extract. In liquid medium, it was based on turbidimetric measurements of optical density at 600 nm (Spectrophotometer Hitachi U 2000), the remaining aromatic acid concentration was determined at 280 nm.

Results and discussion Isolation of Enterobacteria During the enrichment, total aerobic mesophilic flora evolution was almost constant. The initial concentration was 2 · 108 c.f.u./ml; it decreased after 5 days of incubation to about 107 c.f.u./ml and then increased to reach the initial concentration. This preliminary cell number reduction would be related to the selectivity of the medium composition, where only bacteria were able to utilize the residual sugars and the OMW aromatic compounds, and therefore proliferated. The first culture was carried out with a heterogeneous population present in the different inocula. It is likely that further transfers promoted the selection of a microflora with high tolerance to the aromatic constituents of the medium (Ramos-Cormanzana 1986). The concentration of Enterobacteria varied in the same manner as total aerobic flora, starting by decreasing. Only acclimated strains were developed during the enrichment. After three

successive transfers with about one month interval, Enterobacteria became predominant. Polyphenols (2–5%) are the main constituents of the OMW; these include also simple aromatic acids such as protocatechuic, gallic, cinnamic, vanillic, ferulic and syringic acids, present at low levels. After 2 days of incubation, the disappearance of high molecular weight aromatic compounds in inoculated OMW was observed. This was demonstrated on TLC plates (Rf ¼ 0.632). Consequently, after 4 days of incubation, the capacity of the acclimated flora to transform the aromatic compounds in the inoculated OMW was evidenced by the disappearance of compounds with Rf of 0.628 and 0.712. Furthermore, TLC demonstrated the complete degradation of four low molecular weight phenolic compounds. According to their respective Rf values, those would be gallic, protocatechuic, ferulic and syringic acids. Those modifications were accompanied by a COD reduction yield of 18%. In pure culture medium (MM) with 20 g of OMW/l as the sole substrate, all the isolates had a rapid and good growth when they were mixed together. Bacterial strains were purified and identified after the enrichment. Their main bacteriological characteristics are listed in Table 1. All the isolates had typical characteristics of Enterobacteria. They were facultative anaerobic bacteria having both respiratory and fermentative types of metabolism. Half of the isolates belonged to the Klebsiella genus. The other half had the characteristics of Citrobacter. The strains had the following common properties: glucose fermentation positive, production of H2S negative, catalase positive and oxidase negative. Acid from sugars was positive for every sugar tested: glucose, lactose, mannitol, sorbitol, arabinose, rhamnose, maltose, xylose, trehalose and cellobiose. All the strains assimilated citrate and malonate. The strains differed principally from each other in their biotype. The main different properties of the Klebsiella isolates are listed in Table 2. These metabolized D -adonitol, mucate and urea. However, they did not have ornithine decarboxylase. It is noteworthy that Klebsiella strains can grow readily on all kinds of media since they have no particular growth requirements (Krieg & Holt 1983). While the Klebsiella selected were all of the K. oxytoca species, the other Enterobacteria (strains M13b, M14,

Table 1. Phenotypic characteristics of the strains. Strains

M13

M13b

M14

M15

M16

M16b

M17

M18

M19

M20

Cell morphology Biotype Biochemical tests Glucose fermentation Gelatin hydrolysis Indole Arginine dihydrolase Lysine decarboxylase

wB C

B C

sB C

B C

stB Dd

B C

lB Dd

B C

B Dd

lB Dd

+G + + ) +

+G ) + ) )

+ ) + ) )

+G + + ) +

+G + + ) +

+G ) + ) )

+G ) + ) +

+G ) + ) )

+G + + ) +

+G + + ) )

B, bacilli; w, wide; l, long; st, short; G, gas production.

256

Emna Ammar et al.

Table 2. Metabolic characterization of Klebsiella oxytoca isolated. Strains

M13

M16

M17

M19

M20

Tests Voges–Proskauer test Methyl-red Dulcit&

) + )

+ + +

+ ) +

+ ) +

+ + )

be concluded that the ratio COD/BOD5 increased with COD and hence the biodegradability of OMW was negatively affected. The evolution of BOD of such mixed OMW increased till day 20 and hence BOD5 would not be informative on the biodegradability of the stored effluent (Figure 1).

Batch process

BOD (g/l)

M15, M16b and M18) were identified as Citrobacter diversus. It is noticeable that the genera Citrobacter and Klebsiella occur not only in intestinal flora as commensals without causing disease but also in aquatic environments and in soil (Krieg & Holt 1983). The equal frequency of isolates may be explained by the diversity of the inocula initially used where both kinds of genus are present. These could have been acclimated to the substrate during the enrichment and hence could grow easily.

70 60 50 40 30 20 10 0

COD = 80 g/l COD = 100 g/l COD = 200 g/l

0

10

20

30

Time (Days)

Continuous process

COD = 80 g/l COD = 100 g/l COD = 200 g/l

0

10

20

30

Time (Days)

Storage basin BOD (g/l)

A high polluting load, with relatively high COD and BOD5, characterizes the OMW. It contains polyphenols and a large quantity of potassium giving an important salinity to the effluent (Table 3). The quality of the effluent varies exceedingly depending on the time of its production (Table 3). It was found that the COD/BOD5 ratio varied according to the initial COD of the OMW. The study of BOD in general showed that the evolution depended on the type of the system used for oil extraction and initial effluent COD value. For the batch process, a constant BOD was reached after 5 days. However, the oxygen demand increased with the passage of time. Sample from the continuous process exhibited a different evolution, which was related to initial COD value. In fact, the biological oxygen demand was constant after 5 days for OMW initial COD of 100 g/l. However, BOD increased to reach a constant value after 15 days for the effluent with initial COD of 80 g/l and 200 g/l. It might

BOD (g/l)

Biodegradability of olive oil mill wastewater

70 60 50 40 30 20 10 0

70 60 50 40 30 20 10 0

COD = 80 g/l COD = 100 g/l COD = 200 g/l

0

10

20

30

Time (Days) Figure 1. BOD evolution of olive oil mill wastewater according to different conditions.

Table 3. Quality of olive mill effluent. Year

1991

1995

1996

1997

1998

2000

2002

Process Basic index PH COD (g/l) BOD (g/l) Oil (g/l) Suspended solid (g/l) Polyphenols (%) Other constituents (g/l) K+ Total P Total N Ash content (g/l)

Batch

Batch

Continuous

Batch

Batch

Continuous

Basin storage

5.3 72 60 2.5 ND 1.5

4.9 149 30 3.7 17 ND

5.6 120 40 2.0 ND 1.9

5.1 160 70 2.3 ND 2.1

4.9 126 42 0.3 11 2.2

4.9 147 55 3.9 50 2.5

5.2 60 20 4.9 40 1.3

ND 0.42 0.97 ND

7.37 0.33 1.24 12.5

5.10 0.31 0.84 ND

7.60 0.27 1.30 ND

ND ND ND 7.7

3.36 0.18 2.00 9.9

2.63 0.35 0.25 9.6

ND, Non determined.

257

Isolation of bacteria degrading phenols Table 4. Assimilation of aromatic acids by the different isolates on solid media. Strains

M13

M13b

M14

M15

M16

M16b

M17

M18

M19

M20

Acids Protocatechuic p-hydroxybenzoic Gentisic Benzoic Vanillic Ferulic

+ + + + ) )

+ + + ) ) )

) ) + ) ) )

) ) ) ± +

+ + + ) ) )

+ + + ) ) )

+ + + ) ) +

) ) + + ) +

+ + + ) ) +

+ + + ) ) )

It is obvious that the variability in OMW composition observed is a result of many parameters such as harvesting period, olive drupe maturity and the oil extraction process. Utilization of simple aromatic compounds Aromatic acids are among the most important natural compounds and are widely distributed in higher plants such as olive trees and found in OMW. On solid media, the catabolism of simple aromatic acids by the isolates showed that all the strains were able to metabolize gentisic acid, while they showed some differences in their capacity to metabolize other aromatic compounds. Indeed, out of 10 isolates, 7 strains used both protocatechuic and p-hydroxybenzoic acids (Table 4). The latter could be metabolized after its hydroxylation, as is known in bacteria such as Pseudomonas and fungi. This oxidation would be catalysed by a mono-oxygenase (Kent-Kirk et al. 1981). Since all the isolates can grow on gentisic acid

as a sole carbon source, this would be synthesized during biotransformation of complex aromatic compounds in OMW and would be completely catabolized. Benzoic acid was only used by two strains: Klebsiella oxytoca referred to as M13 and Citrobacter diversus referred to as M18. In spite of the use of vanillic acid (4-hydroxy-3methoxybenzoic acid) during the enrichment cultures, only Klebsiella oxytoca M15 seemed to degrade this compound. The methoxylated nucleus was not biodegraded by the other isolates. While comparing the behaviour of the different isolates on aromatic acids, some similarities could be noticed. Strains M16 and M20 exhibited the same abilities in metabolizing such compounds. Also, the strains M17 and M19 showed the same potentialities in assimilating the aromatic acids tested. Moreover, the first group (M16 and M20) differed from the second one (M17 and M19) in ferulic metabolism. Strain M15 seemed to be different from the others: it can grow on methoxylated aromatic rings.

1

1

0 0

2

4

6

8

2

1

1

0

0 0

10 12 16 20 24

2

1

0 4

6

8 10 12 16 20 24

Time incubation (h)

Absorbance at 600 nm

1

Concentration (g/l)

Absorbance at 600 nm

2

2

8

10 12 16 20 24

Klebsiella oxytoca M19

Klebsiella oxytoca M17 2

0

6

Time incubation (h)

Time incubation (h)

0

4

2

2

1

1

0

Concentration (g/l)

0

Absorbance at 600 nm

2 Concentration (g/l)

Absorbance at 600 nm

2

2

Concentration (g/l)

Citrobacter diversus M18 Citrobacter diversus M15

0 0

2

4

6

8

10 12 16 20 24

Time incubation (h)

Figure 2. Assimilation of ferulic acid by isolated Citrobacter diversus and Klebsiella oxytoca. (s): Growth on mineral medium based on yeast extract; (d): Growth on mineral medium with ferulic acid; and (n): Concentration of ferulic acid (g/l).

258 On liquid media, the growth kinetics of four isolates M15, M17, M18 and M19 on ferulic acid are presented in Figure 2. While the strains of Citrobacter diversus M15 and M18 exhibited high degradation yields, respectively 91 and 86%, Klebsiella oxytoca strains M17 and M19 showed lower degradation yields respectively equal to 47 and 51%. When growing on ferulic acid (4-hydroxy-3-methoxycinnamic acid), a typical smell of eugenol (4-allyl-2-methoxyphenol) was noticed in the culture medium. This may indicate that the bacteria converts ferulic acid to eugenol by reduction. This phenomenon of bioconversion was observed with all the studied strains (M15, M17, M18 and M19) while growing on ferulic acid as sole carbon source.

Conclusion The isolation of bacteria by an enrichment culture on OMW lacking partially the carbohydrate content and in which vanillic acid at a constant rate was added as a selective pressure, showed that Enterobacteria were able to use such an effluent rich in aromatic compounds. The isolates were principally Klebsiella oxytoca and Citrobacter diversus at an equal rate. The ability of Klebsiella to carryout aromatic ring degradation has previously been shown (Ammar et al. 1986). Citrobacter degradation of aromatic acids is newly revealed and hence these Enterobacteria will act actively in the first stage of natural OMW treatment. The fact that these strains were mostly isolated on fermented wastewater may explain their aromatic compound-assimilating abilities. However, these capacities were dependent on the type of substitutions on the ring as well as their positions and their numbers. The most efficient strains in biodegrading aromatic acids were two strains of Klebsiella oxytoca (M13 and M19) and one strain of Citrobacter diversus referred as M15. On solid media, all of these strains were able to metabolize four different phenolic acids used at a concentration of 1.5 g/l. In liquid media, the strain M15 was the most efficient in degrading ferulic acid at a concentration of 1.5 g/l. This preliminary study would help in understanding the role of bacteria in biological treatment of wastewaters including aromatic compounds such as those of olive oil processing. Since 1998, in Sfax (Tunisia), all the OMW produced in different plants located in the area is collected and stored in a huge basin of 32 ha. Effluents resulting from both systems of olive extraction are mixed and naturally evaporated. Further investigation is needed.

Acknowledgements We thank Dr Jose´ Menaia from Laboratorrio de Engenharia Sanitaria in Losboa for his helpful contri-

Emna Ammar et al. bution in the identification of the strains and M. Hajji Ayadi for his help with English. References Aggelis, G., Iconomou, D., Christou, M., Bokas, D., Kotzailias, S., Christou, G., Tsagou, V. & Papanikolaou, S. 2003 Phenolic removal in a model olive oil mill wastewater using Pleurotus ostreatus in bioreactor cultures and biological evaluation of the process. Water Research 37, 3897–3904. Ammar, E., Deschamps, A.M. & Lebault, J.M. 1986 Biodegradation of ammonium sulfite spent liquor by pure bacterial culture. Applied Microbiology and Biotechnology 24, 122–127. Ammar, E. & Ueno, S. 1999 Connaissances de base pour la lutte contre la pollution des eaux use´es. pp. 10–77. Ammar et Ueno: Sfax. ISBN 9973-31-106-X. Argenson, C., Re´gis, S., Jourdain, J.M. & Vaysse, P. 1999 L’olivier, pp. 163–181. CTIFL: Paris. ISBN 2-87911-86-6. Ben Rouina, B., Gargouri, K. & Taamallah, H. 2001 L’utilisation des margines comme fertilisant en agriculture. In Journe´es Me´diterrane´eenes de l’Olivier: Nimes (France) 6–8 Avril. Bosku, D. 1996 Olive Oil Chemistry and Technology, pp. 1–53. AOCS Press: Champaign. ISBN 0-935315-73-X. Cameron, M.D., Timofeevski, S. & Aust, S.D. 2000 Enzymology of Phanerochaete chrysosporium with respect to the degradation of recalcitrant compounds and xenobiotics. Applied Microbiology and Biotechnology 54, 751–758. Chang, Y.H., Li, C.T., Chang, M.C. & Shieh, W.K. 1998 Batch phenol degradation by Candida tropicalis and its fusant. Biotechnology and Bioengineering 60, 391–395. Ercoli, E. & Ertola, R. 1983 SCP production from olive black water. Biotechnology Letters 5, 457–462. Fedeli, E. 1997 Encyclope´die mondiale de l’olivier. pp. 306–312. Conseil Ole´icole International, Plazaa et Jane´s: Barcelona. ISBN 84-0161883-5. Fiestas Ros de Ursinos, J.A. 1966 Estudio del alpechin para su approvechamiento industrial. Grasas y Aceite 2, 41–47. Gharsallah, N. 1993 Production of single cell protein from olive mill waste water by yeasts. Environmental Technology 14, 391–395. Hamdi, M., Khadir, A. & Garcia, J.L. 1991 The use of Aspergillus niger for bioconversion of olive oil mill wastewaters. Applied Microbiology and Biotechnology 34, 828–831. Heesche-Wagner, K., Schwarz, T. & Kaufmann, M. 1999 Phenol degradation by an enterobacterium: a Klebsiella strain carries a TOL-like plasmid and a gene encoding a novel phenol hydroxylase. Canadian Journal of Microbiology 45, 162–171. Hinteregger, C., Leitner, R., Loidl, M., Ferschl, A. & Streichsbier, F. 1992 Degradation of phenol and phenolic compounds by Pseudomonas putida EKII. Applied Microbiology and Biotechnology 37, 252–259. Japan International Cooperation Agency, 1993 The Study on Waste Treatment Recycling Plan of Selected Industries in the Region of Sfax in The Republic of Tunisia, pp. 4-29–4-41. JICA, Tokyo. Japan Standards Association, 1995 JIS-Handbook: Quality control, pp. 2347. JIS, Tokyo. ISBN 4-542-13541-1 C3050 P20000E. Karapinar, M. & Worgan, J.T. 1983 Bioprotein production from the waste products of olive oil extraction. Journal of Chemical Techology and Biotechnology 33B, 185–188. Kent-Kirk, T., Higuchi, T. & Chang, H. 1981 Lignin Biodegradation: Microbiology, Chemistry and Potential Applications, Vol 1, pp. 104–144. CRC Press: Florida. ISBN 0-8493-5459-5. Krieg, N.R. & Holt, J.G. 1983 Bergey’s Manual of Systematic Bacteriology, vol 1, pp. 458–465. Williams & Wilkins: Baltimore. ISBN 0-683-04108-8. Perez, J., Ramos-Cormenzana, A. & Martinez, J. 1990 Bacteria degrading phenolic acids on a polymer phenolic pigment. Journal of Applied Bacteriology 69, 38–42. Rabenhorst, J. 1996 Production of methoxyphenol-type natural aroma chemicals by biotransformation of eugenol with a new

Isolation of bacteria degrading phenols Pseudomonas sp. Applied Microbiology and Biotechnology 46, 470–474. Ramos-Cormanzana, A. 1986 Physical, chemical, microbiological and biochemical characters of vegetation water. In International Symposium on Valorization of Olive By-products: pp. 19–40, FAO, La Caja de Ahorros, Madrid. ISBN 84-505-4814-4. Ramos-Cormanzana, A., Garcia-Pareja, M.P., Martinez-Nietol, E. & Garrido-Hoyos, S.E. 1982 Phenolic compounds biodegradation of olive mill waste water with Aspergillus terrus. Grasas y Aceites 43, 75–81. Randerath, K. 1971 Chromatographie sur couches minces, pp. 208–220. Gauthier-Villars: Paris. Richard, C. 1982 Bacte´riologie et e´pide´miologie du genre Klebsiella. Bulletin de l’Institut Pasteur 80, 127–142. Sanjust, E., Pompei, R., Rescigno, A., Rinaldi, A. & Ballero, M. 1991 Olive milling waste water as a medium for growth of four Pleurotus species. Applied Biochemistry and Biotechnology 1, 223–235. Shivarova, N., Zlateva, P., Atansov, B., Christov, A., Peneva, N., Guerginova, M. & Alexieva, Z. 1999 Phenol utilization by

259 filamentous yeast Trichosporun cutaneum. Bioprocess Engineering 20, 325–328. Smith, B.P., Tomfohrde, K.M., Rohden, D.L. & Balows, A. 1972 Api system: a multitube micromethod of identification of Enterobacteriacea. Applied Microbiology 24, 449–452. Tan, K.J. & Gill, C.O. 1984 Effect of culture conditions on batch growth of Saccharomycopsis lipolytica on olive oil. Applied Microbiology and Biotechnology 20, 201–206. Tsioulpas, A., Dimou, D., Iconomou, D. & Aggelis, G. 2002 Phenolic removal in olive mill wastewater by strains of Pleurotus spp. in respect to their phenol oxidase (laccase) activity. Bioresource Technology 84, 251–257. Watanabe, K., Teramotoo, M., Futama, H. & Harayama, S. 1998 Molecular detection, isolation and physiological characterization of functionally dominant phenol-degrading bacteria. Applied and Environmental Microbiology 64, 4396–4402. Zouari, N. & Ellouz, R. 1996 Microbial consortia for the anaerobic degradation of aromatic compounds in olive oil mill effluent. Journal of Industrial Microbiology 16, 155–162.


Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 261–264 DOI 10.1007/s11274-004-2614-5

Effects of Zn supplementation on the growth, amino acid composition, polysaccharide yields and anti-tumour activity of Agaricus brasiliensis Xiang Zou Institute of Modern Biopharmaceuticals, Southwest China Normal University, Chongqing 400715, P.R. China Tel.: +86-23-68332475, Fax: +86-23-68252365, E-mail: [email protected] Received 6 April 2004; accepted 19 August 2004

Keywords: Agaricus brasiliensis, amino acid, anti-tumour activity, polysaccharide, Zn biosorption

Summary The effects of Zn supplementation on the growth, amino acid composition, polysaccharide yields and anti-tumour activity of Agaricus brasiliensis were studied. An initial Zn concentration within the range of 0–300 mg/l had a significant effect on the cell growth and Zn biosorption. At an initial Zn concentration of 300 mg/l, a maximal extracellular polysaccharide (EPS) yield of 5.08 ± 0.25 g/l was obtained, as well as a maximal intracellular polysaccharide (IPS) of 12.25 ± 0.31 mg/g DW. Amino acid analysis results showed that the total amino acid contents of mycelia and culture filtrate decreased from 1090.08 ± 0.76 (233.62 ± 0.06) to 1077.40 ± 0.77 mg/ 100 g DW (229.52 ± 0.05 mg/l), respectively, while the total essential amino acid contents of mycelia and culture filtrate markedly increased from 429.51 ± 0.86 (58.84 ± 0.05) to 476.9 ± 0.85 mg/100 g DW (59.99 ± 0.04 mg/l), respectively. The anti-tumour activity of Zn-enriched mycelial powder against sarcoma 180 in mice showed that the tumour inhibition ratio was 61.5% and was enhanced markedly as compared to normal mycelial powder of 30.8%. The fundamental information obtained in this study will be useful for efficient production of Zn-enriched foods or drugs.

Introduction Higher fungi are abundant sources of a wide range of useful natural products and new products with interesting biological activities (Lorenzen & Anke 1998). Agaricus brasiliensis (Japanese name: Himemtsutke or Agarikusutake) has been traditionally used as a health food source in Brazil for the prevention of cancer, diabetes, hyperlipidaemia, arteriosclerosis and chronic hepatitis. Agaricus brasiliensis is used by
300,000– 500,000 persons in Japan for the prevention of cancer and as an adjuvant with cancer chemotherapy drugs after the removal of malignant tumours (Takaku et al. 2001). Polysaccharides isolated from Agaricus brasiliensis have stronger anti-tumour activity against sarcoma 180 in mice than those from Ganoderma lucidum, Lentinus edodes, and Coriolus versicolor (Mizuno et al. 1990; Mizuno et al. 1999; Chin et al. 2003). It has been reported that 10,000–30,000 kg of the dried body of Agaricus brasiliensis is produced every year in China. Zn is a necessary microelement for humans, and plays a very important physiological role in metabolism. It is used in production of sex and growth hormones and is needed to activate multiple chemical reactions throughout the body. The average adult needs approximately 25 mg of zinc daily. However, many people in China

lack Zn due to their diet. There is a great need to supply the market with a large amount of high quality, safe and avirulent foods fortified with zinc ions. Zn biosorption by higher fungi (Fan et al. 1995; Zhou 2002; Deng & Chen 2003) is viewed as a promising alternative for efficient production of Zn-enriched foods or drugs, and the form of organic Zn can produce some new physiological functions as compared to inorganic zinc ions. It is obviously necessary and important to study the metabolic effects and bioactivity after Zn biosorption. However, as far as we know, until now there have been no reports on the metabolic effects and anti-tumour activity after Zn-enriched culture of Agaricus brasiliensis. In this article, we have investigated the effects of Zn-enriched culture of Agaricus brasiliensis on these properties in order to obtain useful information for the production of Zn-enriched foods or drugs.

Materials and methods Maintenance and preculture of Agaricus brasiliensis The strain of Agaricus brasiliensis AB-2 was purchased from China General Microbiogical Fermentation Collection Center (Beijing, China). It was maintained on

262 potato-agar-dextrose slants. The slant was inoculated with mycelia and incubated at 25 C for 9 days, then stored at 4 C for about 2 weeks. Preculture medium consisted of the following components(g/l): sucrose 40, peptone 2, yeast extract 2.5, KH2PO4  H2O 3, MgSO4  7H2O 2, and Vitamin B1 0.05. For the first preculture, 50 ml medium with an initial pH of 5.5 was prepared in a 250 ml flask, and then 10 ml mycelial suspension from a slant culture was inoculated, and followed by 7 days incubation at 25 C on a rotary shaker (160 rev/min). For the second preculture, 50 ml medium was prepared in a 250 ml flask, and inoculated with 5 ml of the first preculture broth (with ca. 100 mg DW of cells/l), then followed by 4 days incubation at 25 C on a rotary shaker (160 rev/ min) (Fang & Zhong 2002). Methods of enriched Zn in liquid culture The effect of Zn on the fungus culture was studied by using various ZnSO4 concentrations, i.e. 100, 200, 300, 400, and 500 mg/l. The fermentation medium (with an initial pH of 5.5) was composed of corn powder 20 g/l, sucrose 20 g/l, peptone 4 g/l, KH2PO4 Æ H2O 3 g/l, MgSO4 Æ 7H2O 2 g/l, Vitamin B1 0.05 g/l and various Zn concentrations as investigated. The medium was inoculated with 5 ml second preculture broth (with ca. 600–650 mg DW of cells/l) in 45 ml medium in a 250 ml flask. The fermentation was conducted in the dark at 25 C in a rotary shaker at 160 rev/min. Multiple flasks were run at the same time, and three flasks were used at each sampling point. Measurements of extracellular and intracellular polysaccharide (Fang & Zhong 2002) For the determination of extracellular polysaccharide (EPS), after removal of mycelium by filtration, the fermentation filtrate was dialysed, and the crude extracellular polysaccharide was precipitated by addition of 4 volumes of 95% (v/v) ethanol, then separated by centrifugation (10,000 · g, 10 min). The insoluble components were suspended in 1 M NaOH at 60 C (1 h), then the content of EPS in supernatant was measured by the phenol–sulphuric acid method (Dubois et al. 1956). For the analysis of intracellular polysaccharide (IPS), the dried mycelia (100 mg) by filtration were extracted by 1 M NaOH at 60 C (1 h), then the content of IPS in the supernatant was assayed by the phenol–sulphuric acid method (Dubois et al. 1956). Sample, determination of dry weight (Fang & Zhong 2002) For sampling, three flasks were taken each time. For measurement of cell dry weight, the cells were obtained by centrifuging a sample at 12,000 · g for 10 min, washing the precipitated cells for three times to a constant weight, and the fresh cells were dried at 50 C

Xiang Zou for sufficient time until a constant dry weight (DW) was obtained. Assay of Zn concentration in mycelia For the analysis of Zn concentration in mycelia, after the mycelia had been obtained by centrifuging a sample at 12,000 · g for 10 min, washing the precipitated cells for three times, the biosorbed Zn concentration was analysed with a SpectrAA 220 atomic absorption photometer after digestion of samples in dense HNO3 at 500 C for 8 h in a muffle furnace. Assay of amino acids in mycelia and culture filtrate Amino acids were analysed with a Hitachi 835 amino acid analyser after hydrolysis of samples in 6 M HCl at 110 C for 20 h in sealed, evacuated tubes. Assay for anti-tumour activity Anti-tumour activity was measured by intraperitoneal (i.p.) injection of mice bearing sarcoma 180. Five-weekold male ICR/Slc mice (25 ± 2 g) were housed in cages in an air-conditional room and supplied with commercial diet and water ad libitum, sarcoma 180 solid tumour was initially supplied and maintained by the Center of Disease Control, ZheJiang province of China. A fragment, about 3 mm in diameter, of the 14-day-old tumour was implanted subcutaneously into the right groin of mice by a trocar. The number of mice was 10 in each group. The samples of normal mycelia and Znenriched mycelial powder dissolved or suspended in saline were injected intraperitoneally after the tumour implantation. In the control group, mice were injected with saline only in the same dose and were maintained for the same time period as those in the treated groups, tumour growth was indicated by measuring the tumour weight 3 weeks after implantation. The inhibition ratios were calculated by the following formula: inhibition ratio (percent) ¼ {1) (average of tumour weight in treated group/average of tumour weight in control group)} · 100. Complete regression of tumour and mortality were compared with those of the control mice 6 weeks after tumour implantation.

Results and discussion Zn biosorption level of Agaricus brasiliensis by submerged culture It is necessary to investigate the effect of Zn ions added to the medium on the cell growth and Zn biosorption level of Agaricus brasiliensis. Effect of different Zn concentrations added i.e. 100, 200, 300, 400 and 500 mg/l on the cell growth and Zn biosorption level were investigated. As shown in Table 1, A. brasiliensis could grow at the different Zn concentrations. More biomass and Zn

Zinc effects on Agaricus brasiliensis

263

Table 1. Zn biosorption level of Agaricus brasiliensis by submerged culture. Zn concentration in medium (mg/l)

Cell dry weight (g/l)

Zn concentration in mycelia (mg/l)

0 100 200 300 400 500

10.21 10.68 11.26 11.47 10.17 9.13

2.26 2.56 13.50 27.49 25.11 24.32

± ± ± ± ± ±

0.30 0.42 0.24 0.35 0.45 0.30

± ± ± ± ± ±

0.07 0.02 0.02 0.07 0.11 0.08

biosorption capacity could be obtained at relatively lower Zn concentrations between 100 and 300 mg/l. When the initial Zn concentration was above 300 mg/l, the biomass and Zn biosorption accumulation decreased sharply. The relatively higher initial Zn level may have inhibited the cell growth and Zn biosorption capacity. It was clear that an initial Zn concentration of 300 mg/l was favourable to cell growth and Zn biosorption. There were very little difference between the values of Zn biosorption in unsupplemented cultures and those with 100 mg/l Zn as shown in Table 1, Zn biosorption in unsupplemented cultures may mainly come from the component in medium such as corn powder obtained from local market. Effect of Zn biosorption on the production of polysaccharide Based on the above information as obtained, the effects of initial Zn concentrations on the polysaccharide production were investigated (Figure 1). The maximum production of extracellular polysaccharide (EPS) and intracellular polysaccharide (IPS) was 3.05 ± 0.09 (7.1 ± 0.12), 3.96 ± 0.15 (7.27 ± 0.15), 4.25 ± 0.24 (8.20 ± 0.35), 4.66 ± 0.14 (9.03 ± 0.27), 5.08 ± 0.25 (12.25 ± 0.31), 3.45 ± 0.22 g/l (9.57 ± 0.18 mg/g DW) at an initial Zn concentration 50, 100, 200, 300 and 400 mg/l, respectively. The results indicated that the EPS and IPS content simultaneously enhanced at a relatively lower Zn concentration between 50 and 300 mg/l. When the initial Zn concentration was above 300 mg/l, the FPS and IPS production also markedly inhibited. The above results indicated that an initial Zn concentration of 300 mg/l was desirable for optimal polysaccharide (EPS and IPS) production. Effect of Zn biosorption on the amino acid components The effect of Zn-enriched culture on the amino acid components varieties of mycelia and culture filtrate is shown in Table 2. After Zn biosorption, the total amino acid contents of mycelia and culture filtrate both decreased slightly. The total amino acid contents of mycelia and culture filtrate decreased from 1090.08 ± 0.76 (233.62 ± 0.06) to 1077.40 ± 0.77 mg/ 100 g DW (229.52 ± 0.05 mg/l), respectively, while the total essential amino acids contents of both mycelia and culture filtrate markedly increased from 429.51 ± 0.86 (58.84 ± 0.05) to 476.9 ± 0.85 mg/

Figure 1. Effect of initial Zn concentration on the production of polysaccharide and biomass symbol: (j) EPS, (d) IPS, (m) DW.

100 g DW (59.99 ± 0.04 mg/l), respectively. The results indicated that Zn biosorption might have changed the protein biosynthesis of Agaricus brasiliensis in this case. Anti-tumour activity of Zn-enriched mycelial powder The anti-tumour activity of Zn-enriched mycelial powder against sarcoma 180 in mice was examined. As shown in Table 3, the tumour inhibition ratio of normal mycelia and Zn-enriched mycelial powder showed 30.8 and 61.5% at the end of week 3 after sarcoma 180 implantation as compared to the control, respectively. Hence the tumour inhibition ratio of Zn-enriched mycelial powder was much higher than that of normal mycelial powder. The difference of inhibition ratio against sarcoma 180 in mice indicated that new bioactivity components may be produced and enhanced the anti-tumour activity. This is also the first report that Zn biosorption by higher fungi, Agaricus brasiliensis, improves the anti-tumour activity against sarcoma 180 in mice.

Conclusion In this work, Zn biosorption of Agaricus brasiliensis has a significant effect on the growth, amino acid composition, polysaccharide yields. Zn-enriched mycelial powder has exhibited strong anti-tumour activity and markedly

264

Xiang Zou

Table 2. Amino acid composition varieties of mycelia and culture filtrate. Amino acid

Amino acid content Control culture filtrate (mg/l)

Asn 33.12 Cys 5.70 Ser 14.91 Glu 49.82 Gly 23.63 Ala 11.31 Arg 9.12 Pro 21.94 His 5.23 Thr 11.51 Val 14.04 Met 1.84 Ile 5.20 Leu 9.42 Phe 8.42 Lys 8.41 Tyr —* Total content 233.62 Essential amino acids content 58.84

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.04 0.02 0.05 0.12 0.07 0.05 0.06 0.1 0.07 0.05 0.06 0.04 0.06 0.05 0.06 0.03

± 0.06 ± 0.05

Zn-enriched culture filtrate (g/l)

Control mycelia (mg/100 g DW)

Zn-enriched mycelia (mg/100 g DW)

32.13 5.91 12.14 47.13 24.64 11.50 10.33 20.73 5.02 10.83 15.12 1.73 5.21 9.14 7.52 10.44 — 229.52 59.99

103.84 15.81 52.33 204.7 82.53 79.58 47.66 60.69 13.40 47.78 77.32 35.32 52.07 97.97 50.97 40.13 27.95 1090.08 429.51

103.79 0.00 53.99 179.3 66.31 76.49 50.57 56.26 13.79 48.16 76.53 91.50 57.46 86.84 48.71 40.93 26.77 1077.40 476.9

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.05 0.03 0.04 0.10 0.08 0.04 0.05 0.08 0.04 0.06 0.04 0.02 0.03 0.03 0.04 0.05

± 0.05 ± 0.04

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

1.8 0.05 0.12 2.1 0.45 0.38 0.36 0.65 0.07 0.72 1.0 0.8 1.1 1.4 0.8 0.7 0.4 0.76 0.86

± 1.5 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.10 1.7 0.5 0.54 0.42 0.80 0.08 0.8 1.2 1.0 1.0 1.2 0.7 0.6 0.3 0.77 0.85

* ¼ not found.

Table 3. Anti-tumour activity of Zn-enriched mycelia powder against sarcoma 180 in mice. Name

Dose (mg)

Times

Route

CR/n

Tumuor weight mean ± SD (g)

% Inhibition

Control Zn-enriched mycelia powder Normal mycelia powder

13 13 13

10 10 10

i.p. i.p. i.p.

0/10 0/10 0/10

1.3 ± 0.5 0.5 ± 0.2* 0.9 ± 0.2*

00.00 61.50 30.80

*P < 0.05, significantly different from the saline-treated control group. Dose:/mouse, CR/n: number of tumour free mice/total mouse.

improved the anti-tumour activity against sarcoma 180 as compared to normal mycelial powder. Increased Zn levels in mycelium increase the anti-tumour effects, and the cause of this is currently under examination. Zn biosorption by Flamulina velutipes also improves the learning ability and immunological function of mice (Fan et al. 1995), and Zn biosorption by higher fungi may help to produce some new physiological functions. The fundamental information obtained in this study will be useful for efficient production of Zn-enriched foods or drugs.

Acknowledgements The authors would like to express their grateful thanks to Po Huo (Department of Chemistry and Life, ZheJiang University of Science and Technology) for her kind guidance in the test of anti-tumour activity. References Chin, H.S., Bor, W. & Ko, J.L. 2003 Monitoring the polysaccharide quality of Agaricus blazei in submerged culture by examining molecular weight distribution and TNF-a release capability of

macrophage cell line RAW 264.7. Biotechnology Letters 25, 2061– 2064. Deng, B.W. & Chen, W.Q. 2003 Study on accumulation of zinc in Polyporus umbellate in liquid culture. Edible Fungi of China 22, 33–35. Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers P.A. & Smith, F. 1956 Colorimetric method for determination of sugars and related substances. Analytical Chemistry 28, 350–356. Fan, Q.S., Wei, H. & Xie, J.J. 1995 Effect of Flamulina velutipes cultured in zinc enriched medium on the learning ability and immunological function of mice. Acta Nutrimenta Sinica 17, 89–91. Fang, Q.H. & Zhong, J.J. 2002 Effect of initial pH on production of ganoderic acid and polysaccharide by submerged fermentation of Ganoderma lucidum. Process Biochemistry 37, 769–774. Lorenzen, K. & Anke, T. 1998 Basidiomycetes as a source for new bioactive natural products. Current Organic Chemistry 2, 329–364. Mizuno, T., Hagiwara, T., Nakamura, T., Ito, H., Shimura, K. & Asakura, A. 1990 Antitumor activity and some properties of water-soluble polysaccharides from ‘Himematsutake’, the fruiting body of Agaricus brasiliensis. Agricultural and Biological Chemistry 54, 2889–2896. Mizuno, M., Ichino, K., Ito, H., Kawade, M. & Teral, H. 1999 Antitumor polysaccharide from the mycelium of liquid-cultured Agaricus brasiliensis. Biochemistry and Molecular Biology International 47, 707–714. Takaku, T., Kimura, Y. & Okuda, H. 2001 Isolation of an anti-tumour compound from Agaricus brasiliensis and its mechanism of action. Journal of Nutrition 21, 1409–1413. Zhou. J.L. 2002. Zn biosorption by Rhizopus arrhizus and other fungi. Applied Microbiology and Biotechnology 51, 686–693.

 Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 265–269 DOI 10.1007/s11274-004-3628-8

Isolation and characterization of a new carbendazim-degrading Ralstonia sp. strain Gui-Shan Zhang, Xiao-Ming Jia*, Tian-Fan Cheng, Xiao-Hang Ma and Yu-Hua Zhao College of Life Science, Zhejiang University, Hangzhou 310029, China *Author for correspondence: Tel.: +86-571-8697-1962, E-mail: [email protected] Received 24 February 2004; accepted 13 July 2004

Keywords: 16S rDNA, carbendazim, phenotypic features, phylogenetic analysis, Ralstonia

Summary A bacterial strain 1-1 capable of utilizing carbendazim was isolated from carbendazim-treated Qiyang red soils Hunan Province, China. It is gram-negative, rod-shaped, motile with peritrichous flagella, which formed round, smooth, convex and transparent colonies of about 1.1 mm diameter after 3 days of incubation on the isolation and purification medium using carbendazim as the sole carbon and energy sources. The degradation ratios of carbendazim by strain 1-1 were 19.16 and 95.96% in the carbendazim (500 mg/l)-degrading medium and the carbendazim (500 mg/l)-degrading medium supplemented with yeast extract (150 mg/l) within 24 days, respectively. Strain 1-1 was identified as Ralstonia sp. (b-Proteobacteria) based on the results of phenotypic features, G+C mol% and phylogenetic analysis of 16S rDNA. Strain 1-1 could become a new bacterial resource for biodegrading carbendazim and might play a bioremediation role for soils contaminated by carbendazim. Introduction Carbendazim (methyl benzimidazol-2-ylcarbamate) is the most widely used benzamidazole fungicide and is also the major degradation product of other systemic fungicides, benomyl and thiophanate-methyl (Fleeker et al. 1974; Mongomery et al. 1997). (Figure 1). It is a stable compound with a long half-life (Carbendazim is decomposed in the environment with half-lives of 6–12 months on bare soil, 3–6 months on turf, and half-lives in water of 2 and 25 months under aerobic and anaerobic conditions, respectively.) in the environment (WHO 1993). Hence, it can persist at application sites and easily induce cumulative effects; its residue in fruits, plants and soils could be harmful to human health through food chains. Some studies indicated that carbendazim could do harm to the liver to some extent (WHO 1993) and it has been documented as mutagenic and has teratogenic effects on mammals at single, lowlevel doses (Sarrif et al. 1994;Nakai et al. 1998). The degradation of carbendazim in the environment has received extensive attention. Carbendazim may be degraded by way of both photolysis and biodegradation (Fleeker & Lacy 1977; Yarden et al. 1990). Helweg (1977) reported that carbendazim in soil was decomposed mainly by microorganisms. The isolation and screening of efficient carbendazim-degrading microorganisms is a useful approach in bioremediation of carbendazim contamination. However, the isolation and identification of bacterial pure cultures capable of degrading carbendazim have been seldom reported so far. Fuchs & de Vries. (1978) reported that carbendazim

could be degraded to 2-aminobenzimidazole by Pseudomonas spp., whereas Holtman & Kobayashi (1997) found carbendazim could be degraded by Rhodococcus erythropolis. This paper reports the isolation and characterization of a strain of carbendazim-degrading bacterium from carbendazim-contaminating Qiyang red soil from Hunan Province, China.

Materials and methods Enrichment, isolation and purification of carbendazim-degrading bacteria The soil which had been treated with carbendazim for 3 months at the time of sampling was collected from Qiyang District, Hunan Province, China. For the isolation of potential carbendazim-degrading bacteria, a continuous enrichment culture method was used. 7.5 g sample of soil was added to 75 ml sterilized enrichment medium (g/l): NaCl 1.0, K2HPO4 1.0, MgSO4Æ7H2O 1.0, CaCO3 1.0, carbendazim (94.6% pure) was firstly dissolved in 0.01 M hydrochloric acid, at five concentrations (g/l): 0.2, 0.4, 0.5, 0.6, 0.8, distilled water, pH 7.0 and incubated with shaking at 30
C in the dark. After 3 days, 1 ml aliquot from each conical flask was transferred to a new flask containing fresh medium, then was still incubated at 30
C with shaking in the dark. This process was repeated five times before each culture was dilution-plated (five repeats) onto sterilized isolation and purification medium (g/l): NaCl 1.0, K2HPO4 1.0, MgSO4Æ7H2O 1.0, CaCO3 1.0, (NH4)2SO4 2.0,

266

Zhang et al. CONH(CH2)3CH3 N NHCOOCH3

O

NHCNHCOOCH3

N

NHCNHCOOCH3

Benomyl 6 5

7 4

1

O

Thiophanate-Methyl

H N

H N

2 3

N

Benzimidazole

O C* NHCOCH3

N

Carbendazim H N C* NH2 N

2-Aminobenzimidazole *CO2

Figure 1. Structural formulae of benzimidazoles and a suggested biotransformation pathway. Benomyl, thiophanate-methyl and carbendazim ( ¼ MBC, ¼ methyl benzimidazol-2-ylcarbamate) are systemic fungicides; 2-AB ( ¼ 2-aminobenzimidazole) and benzimidazole are non-fungicides. *C ¼ labelled carbon.

carbendazim (five concentrations are the same as those of enrichment medium), agar 20, distilled water, pH 7.0. All plates were incubated at 28±1
C in the dark for bacterial isolation. After five subcultures, the pure bacterial cultures that could grow on the isolation and purification medium were selected. One of these pure cultures, strain 1-1 which was from the highest carbendazim concentration (0.5 g/l) of isolation and purification medium , was maintained on LB agar medium and used to carry out further studies. Carbendazim degradability of strain 1-1 For the study of bacterial carbendazim degradability, five 500 ml conical flasks containing 200 ml carbendazim-degrading medium (g/l): NaCl 1.0, K2HPO4 1.0, MgSO4Æ7H2O 1.0, CaCO3 1.0, (NH4)2SO4 2.0, carbendazim 0.5, distilled water, pH 7.0, five 500 ml flasks containing 200 ml carbendazim-degrading medium supplemented with yeast extract (150 mg/l) and five control flasks containing carbendazim-degrading medium were designed. The strain 1-1 cells were inoculated into each sample flask. Five control flasks were inoculated with aseptic distilled water instead of strain 1-1. All flasks were incubated at 28±1
C with shaking in the dark. The concentration of carbendazim was determined as follows: 5 ml samples (1 ml each flask, three repeats) were drawn periodically (every 3 days) under aseptic conditions. The 5 ml solution was placed in a 50 ml conical flask, to which was then added 20 ml 1,4dioxane, heated for 5 min in boiling water, then 1 ml acetic acid was added to and cooled to room temperature, then filtered into a 50 ml volumetric flask by G4 sand core funnel. About 10 ll of filtered solution was taken and samples spotted on GF254 glass-backed

plates (10cm · 20 cm) and dried with the aid of a hair drier. The plate was then developed in the solvent system benzene-acetone-acetic acid (70:30:5, v/v) and then dried. Carbendazim could be easily identified as violet spots under u.v. light and the Rf of the carbendazim spots were about 0.65. The content and the standard curve of carbendazim were determined with the same process mentioned above. Concentrations of carbendazim were calculated according to the standard curve of carbendazim. The recovery rate of this method was 94.56%, the lowest limit of determination was 3.5 mg/l. The carbendazim degradation of strain 1-1 was analysed by using a Shimadzu CS930 dual-wavelength thin-layer chromatogram (TLC) scanner at 280 nm (Braithwaite & Smith 1985). Identification of strain 1-1 The morphology and motility of strain 1-1 were determined by conventional methods. Flagella were observed using JEM-1200EX electron microscope (JEOL Inc., USA). Physiological and biochemical characterization was performed by biochemical test tubes (Hangzhou Tianhe Microorganism Reagent Co., Ltd., China) according to the protocol supplied by the manufacturer. DNA G+C content (Tm method) was analysed on Shimadzu UV-2550 spectrophotometer (Shimadzu Corporation, Japan) equipped with water cycle heating-up system. PCR amplification and sequencing of 16S rDNA Strain 1-1 cells that had been cultured on LB medium for 24 h were transferred to an Eppendorf tube containing 200 ll aseptic double-distilled water. The mixture was centrifuged for 5 min after it was kept in boiling water for 3 min, the supernatant of which was directly used amplification as template DNA. Amplification was done by PCR with primers (Devereux & Willis 1995) named BSF8/20 (5¢-AGAGT TTGAT CCTGG CTCAG-3¢) and BSRl541/20 (5¢AAGGA GGTGA TCCAG CCGCA-3¢). The reactions were performed in a final reaction mixture of 50 ll containing 5 ll 10 · PCR buffer (containing Mg2+), 1 ll dNTP (5 mM),1 ll (0.2 lM) each primer, l.5 ll template DNA,0.5 ll Taq DNA polymerase (5 U/ll),40 ll ddH2O and 30 ll liquid paraffin. The amplification reactions were performed with the following cycles of parameters: 94
C for 5 min, followed by three cycles of 45 s at 94
C, 2 min at 50
C and 1 min at 72
C and 29 cycles of 1 min at 94
C, 1 min at 50
C and 1 min at 72
C, with a final extension at 72
C for 7 min. The amplification products were checked by 2% agarose gel electrophoresis and staining with ethidium bromide. PCR products were purified and sequenced by Shanghai Bioasia Biologic Technology Co., Ltd., China.

267

Isolation and characterization Phylogenetic analysis The 16S rDNA gene sequence (605 bp, GenBank accession number AY447043) of strain 1-1 aligned with all the sequences available from the GenBank database by BLAST and all sequences were retrieved from Genbank database individually and aligned using ClustalX 1.8.1 with default settings (Thompson et al. 1997). Phylogenetic analysis was performed by means of MEGA version 2.1 (Kumar et al. 2001) software using UPGMA method and selecting Kimura 2-parameter distance model, which was tested by Bootstrap method (1000 repetitions). The 16S rDNA sequences included in the phylogenetic analysis can be seen in Figure 4.

Results and discussion Isolation and taxonomic characteristics of bacteria Bacterial colonies could be observed on the isolation and purification medium using carbendazim as the sole carbon and energy sources after 3–5 days except at 0.6 and 0.8 g/l (carbendazim concentration). All the colonies were identical morphologically, from which strain 1-1 was isolated. The cells of strain 1-1 were 0.5–0.6 lm · 0.6–2.7 lm, gram-negative, short rod-shaped, motile with peritrichous flagella (Figure 2), non-spore-forming. It could form round, smooth, convex and transparent colonies of about 1.1 mm diameter on the isolation and purification medium after 3 days. However, ivory-white, opaque colonies were formed on LB medium after 1–2 days. It was observed that yeast extract could accelerate growth of strain 1-1. Strain 1-1 could grow at 20–41
C, but not at 42
C, and could also grow on 3% NaCl LB medium, but not on 5% NaCl LB medium. The following enzyme reactions were positive: oxidase, catalase, urease, arginine dihydrolase, lysine decarboxylase, alkaline (acid) phosphatase and ornithine decarboxylase. Nitrate was reduced, nitrite was not. No indole was produced from

Figure 2. Electron micrograph of negative-stained 1-1 cell with peripheral flagella (12000·).

Figure 3. Degradation kinetic curves of carbendazim by cells of strain 1-1 within 24 days expressing that carbendazim (500 mg/l)-degrading medium supplemented with yeast extract (150 mg/l) can distinctly accelerate the degradation of carbendazim and carbendazim cannot be degraded without cells of strain 1-1 in the dark.

tryptophan. Glucose was not fermented. D -ribose, D gluconate, L -malate and citrate were assimilated. No assimilation of galactose, D - & L -arabinose, lactose, maltose, mannose, mannitol, melezitose and raffinose was detected. The DNA G+C content was 64.9 mol% (Tm). These phenotypic characteristics are very close to those of four species of Ralstonia spp. (Yabuuchi et al. 1995; Goris et al. 2001).

Carbendazim degradation of strain 1-1 The results of carbendazim degradation by strain 1-1 (Figure 3) showed that the concentration of carbendazim fell from 500 to 404.2 mg/l in carbendazim-degrading medium and to 20.2 mg/l in carbendazim-degrading medium supplemented with yeast extract within 24 days, respectively. The degradation ratios and the average degradation rates were 19.16%, 3.99 mg/(lÆd) and 95.96%, 19.99 mg/(lÆd) , respectively. The addition of yeast extract accelerated the degradation of carbendazim by strain 1-1. It is most probable that the compound is a poor microbial energy source and not the most suitable substrate for strain 1-1, which indicated that the degradation may be a co-metabolic process similar to that suggested in the earlier studies of Helweg (1977). Holtman & Kobayashi (1997) have reported that five strains of Rhodococcus erythropolis could completely degrade 16 mg carbendazim/l in M9 salts medium within 15 days, the average degradation rate of which was 1.07 mg/(lÆd). However, the average degradation rate by strain 1-1 was 3.99 mg/(lÆd) in carbendazimdegrading medium. Hence, the carbendazim degradability of strain 1-1 exceeded that of Rhodococcus erythropolis.

268

Zhang et al. species (Euze´by 1997) that have been validly published including R. eutropha JS705 (Van der Meer et al. 1998) and R. basilensis RK1 (Steinle et al. 1998) that could degrade chlorobenzene and 2,6-dichlorophenol, respectively. Furthermore, the isolated strain 1-1 in this study is a carbendazim-degrading bacterium. To date, pure cultures of bacteria capable of degrading carbendazim that have been documented have been only Pseudomonas spp. (Fuchs & de Vries 1978) and Rhodococcus erythropolis (Holtman & kobayashi1997). The former and strain 1-1 belong to the Proteobacteria, so they were comparatively close in taxonomic status, which could be seen on the phylogenetic tree, however, the Rhodococcus sp. belongs to an unrelated group: Nocardiaceae, Actinomycetales, Actinobacteria. In this study, strain 1-1 that we have isolated is a new bacterial resouce for biodegrading carbendazim and might have a possible bioremediation role for soils contaminated by carbendazim.

Figure 4. UPGMA phylogenetic tree of strain 1-1, validly described Ralstonia species, carbendazim degradation bacteria Pseudomonas spp. based on the 16S rDNA sequences comparisons. Bootstrap values obtained with 1000 repetitions are indicated as percentages at all branches. The 16S rDNA sequences of Alcaligenes faecalis, Burkholderia cepacia and Burkholderia andropogonis were included as closely related species. Scale bar indates evolutionary distance. Abbreviation: A., Alcaligenes; B., Burkholderia; P., Pseudomonas; R., Ralstonia.

Phylogenetic analysis of 16S rDNA sequence A phylogenetic tree including all known representatives of validly described Ralstonia species and other correlative species is given in Figure 4. Strain 1-1 and Ralstonia species constituted one big cluster on the phylogenetic tree. The 16S rDNA gene of strain 1-1 showed high sequence similarity (more than 97%) to the 16S rDNA genes of all Ralstonia species including the type species: R. pickettii. The highest sequence homology between strain 1-1 and R. campinensis with heavymetal-resistance (Goris et al. 2001) was 99.8% though they have slightly difference in phenotypic features. The 16S rDNA gene sequences of Burkholderia cepacia, Burkholderia andropogonis and Alcaligenes faecalis apparently formed a big cluster, which was close to that of Ralstonia spp. and the three species were suggested to be emended to Ralstonia spp. by Yabuuchi et al. (1995). According to some phenotypic characteristics mentioned above, G + C content, phylogenetic analysis, strain 1-1 should be assigned to Ralstonia sp. (bProteobacteria). DNA–DNA hybridizations with related species should be performed for species taxa. The genus Ralstonia (Yabuuchi et al. 1995) has been created for a group of organisms from ecologically diverse niches to accommodate bacteria that were formerly classified as members of Burkholderia (Yabuuchi et al. 1992) and Alcaligenes, which so far has 13

Acknowledgements We are grateful to Professor Ren-lin Cao at Institute of Agro-Environment Protection, Ministry of Agriculture, PR China, for his providing us with carbendazim (94.6% pure) samples. This project was supported by the Research Fund for Public Welfare of Ministry of Science and Technology, PR China (Grant number 2001-177).

References Braithwaite, A. & Smith, F.J. 1985 Chromatographic Methods. 4th edn. London, New York: Chapman and Hall. ISBN 0-4122-6770-5. Devereux, R. & Willis, S.G. 1995 Amplification of ribosomal RNA sequences. In ed. Molecular Microbial Ecology Manual. Akkermans A.D.L., et al. The Netherlands: Kluwer Academic Publishers, 3.3.1-3.3.11. ISBN 0-7923-3698-4. De Vos, P., Kersters, K., Falsen, E., Pot, B., Gillis, M., Segers, P. & De Ley, J. 1985 Comamonas Davis and Park 1962 gen. nov. nom. rev. emend and Comamonas terrigena Hugh 1962 sp. nov. nom. Rev. International Journal of Systematic Bacteriology 35, 443–453. Euze´by, J.P. 1997 List of Bacterial Names with Standing in Nomenclature (http://www.bacterio.cict.fr/, viewing date: Feb. 12. 2004). Fleeker, J.R. & Lacy, H.M. 1977 Photolysis of methyl-2-benzimidazole-carbamate. Journal of Agricultural and Food Chemistry 25, 51–55. Fleeker, J.R., Lacy, H.M., Schultz, I.R. & Houkom, E.C. 1974 Persistence and metabolism of thiophanate-methyl in soil. Journal of Agricultural and Food Chemistry 22, 592–595. Fuchs, A. & de Vries, F.W. 1978 Bacterial breakdown of benomyl. I. Pure cultures. Antonie van Leeuwenhoek 44, 283–292. Goris, J., De Vos, P., Coenye, T., Hoste, B., Janssens, D., Brim, H., Diels, L., Mergeay, M., Kersters, K. & Vandamme, P. 2001 Classification of metal-resistant bacteria from industrial biotopes as Ralstonia campinensis sp. nov., Ralstonia metallidurans sp. nov. and Ralstonia basilensis Steinle et al. 1998 emend. International Journal of Systematic and Evolutionary Microbiology 51, 1773– 1782. Helweg, A. 1977 Degradation and adsorption of carbendazim and 2-aminobenzimidazole in soil. Pesticide Science 8, 71–78.

Isolation and characterization Holtman, M.A. & Kobayashi, D.Y. 1997 Identification of Rhodococcus erythropolis isolates capable of degrading the fungicide carbendazim. Applied Microbiology and Biotechnology 47, 578–582. Kumar, S., Tamura, K., Jakobsen, I.B. & Nei, M. 2001 MEGA2: Molecular Evolutionary Genetics Analysis software, Arizona State University, Tempe, Arizona, USA. Nakai, M., Toshimori, K., Yoshinaga, K., Nasu, T. & Hess, R.A. 1998 Carbendazim-induced abnormal development of the acrosome during early phases of spermiogenesis in the rat testis. Cell and Tissue Research 294, 145–152. Mongomery, J.H. 1997 Benomyl In Agrochemicals Desk Reference. 2nd edn. New York: Lewis Publishers. ISBN: 1-56670-167-8. Sarrif, A.M., Arce, G.T., Krahn, D.F., O’Neil, R.M. & Reynolds, V.L. 1994 Evaluation of carbendazim for gene mutations in the Salmonella/Ames plate-incorporation assay: the role of aminophenazine impurities. Mutation Research 321, 43–56. Steinle, P., Stucki, P., Stettler, R. & Hanselmann, K.W. 1998 Aerobic mineralization of 2,6-dichlorophenol by Ralstonia sp. strain RK1. Applied and Environmental Microbiology 64, 2566–2571. Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F. & Higgins, D.G. 1997 The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research 24, 4876–4882.

269 Van der Meer, J.R., Werlen, C., Nishino, S.F. & Spain, J.C. 1998 Evolution of a pathway for chlorobenzene metabolism leads to natural attenuation in contaminated groundwater. Applied and Environmental Microbiology 64, 4185–4193. WHO. 1993 Environment Health Criteria 149: Carbendazim. Geneva: World Health Organization. (http://www.inchem.org/documents/ ehc/ehc/ehc149.htm, viewing date: Jan. 19. 2004) Yabuuchi, E., Kosako, Y., Yano, I., Hotta, H. & Nishiuchi, Y. 1995 Transfer of two Burkholderia and an Alcaligenes species to Ralstonia gen. nov.: proposal of Ralstonia pickettii (Ralston, Palleroni and Doudoroff 1973) comb. nov., Ralstonia solanacearum (Smith 1896) comb. nov. and Ralstonia eutropha (Davis 1969) comb. nov. Microbiology and Immunology 39, 897–904. Yabuuchi, E., Kosako, Y., Oyaizu, H., Yano, I., Hotta, H., Hashimoto, Y., Ezaki, T. & Arakawa, M. 1992 Proposal of Burkholderia gen. nov. and transfer of seven species of the genus Pseudomonas homology group II to the new genus, with the types species Burkholderia cepacia (Palleroni and Holmes 1981) comb. nov. Microbiology and Immunology 36, 1251–1275. Yarden, O., Salomon, R., Katan, J. & Aharonson, N. 1990 Involvement of fungi and bacteria in enhanced and nonenhanced biodegradation of carbendazim and other benzimidazole compounds in soil. Canadian Journal of Microbiology 36, 15–23.


Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 271–278 DOI 10.1007/s11274-004-3631-0

Survey of plasmid profiles of Shigella species isolated in Malaysia during 1994–2000 C.H. Hoe1, R.M. Yassin2, Y.T. Koh2 and K.L. Thong1,* 1 Microbiology Division, Institute of Biological Sciences, Faculty of Science, University of Malaya, 50603 Kuala Lumpur, Malaysia 2 Bacteriological Unit, Institute for Medical Research, Kuala Lumpur, Malaysia *Author for correspondence: Tel.: 603-79674437, Fax: 603-79675908, E-mail: [email protected] Received 2 March 2004; accepted 13 July 2004

Keywords: Antimicrobial susceptibility, Malaysia, multi-drug resistance, plasmid profile, Shigella

Summary Plasmid profiling was used to characterize 219 strains of Shigella species isolated from sporadic cases of shigellosis in Malaysia during the period 1994–2000. Heterogeneous plasmid patterns were observed in all Shigella spp. There was a correlation between plasmid patterns and serotypes of S. flexneri, S. dysenteriae and S. sonnei. Five common small plasmids (>20.0 kb) were observed in S. flexneri 1b and 2a, whereas six common small plasmids were found in serotype 3a. Some of these plasmids appeared to maintain their existence stably in each individual serotype. Plasmids of size 11.40 and 4.20 kb were present only in S. flexneri 2a isolates, whereas the 4.40 kb plasmid was unique for serotype 3a. Large (>150 kb) or mid-range plasmid (20.0–150 kb) was not observed from any S. flexneri 1b isolates. Eighty-nine percent of S. flexneri of various serotypes harboured the plasmid of 3.20 kb. All S. dysenteriae type 2 isolates harboured the 9.00 kb plasmid, while four common small plasmids were found in S. sonnei isolates. The 2.10 kb plasmid was only seen in S. sonnei. Streptomycin resistance in S. dysenteriae type 2 and multi-drug resistance in S. sonnei may be associated with the 9.00 and 14.8 kb plasmids, respectively. Plasmid profiling provided a further discrimination beyond serotyping and a useful alternative genotypic marker for differentiation of Shigella species. To the best of our knowledge, this is the first report on the plasmid prevalence of the Malaysian Shigella species.

Introduction Shigellosis or bacillary dysentery is an acute diarrhoeal disease caused by members of the bacterial genus Shigella, which is composed of four species; S. dysenteriae, S. flexneri, S. boydii, and S. sonnei. It is one of the major causes of morbidity and mortality in children with diarrhoea in developing countries (Chiou et al. 2001). Worldwide, the annual number of Shigella episodes was estimated to be 164.7 million, of which 163.2 million were in developing countries and 1.5 million in industrialized countries. Shigellosis is responsible for about 1.1 million deaths per year throughout the world, and two-thirds of the patients are children under 5 years of age (Kotloff et al. 1999). Shigellosis is endemic in Malaysia. From 1978 to 1997, 26,320 stool specimens were collected by the University of Malaya Medical Center (UMMC) from children aged less than 16 years with clinical diarrhoea. Among these, 386 isolates tested were positive for Shigella species, representing 1.4% of the 26,320 total stool specimens and 13% of 2986 isolates positive for bacterial patho-

gens (Lee & Puthucheary 2002). The actual incidence of shigellosis in this country could be higher, as the disease is self-limiting in adults and the affected individuals would usually go to different general practitioner clinics for medical care or employ self-medication. Thus, the vast majority of the cases are not notified. In recent years, traditional typing techniques based on phenotypic characteristics e.g. biotyping, serotyping, antibiotic susceptibility, are increasingly challenged by the use of DNA-based techniques. Approaches at molecular level have been used to assess the relatedness of closely related bacterial isolates. Plasmid profiling was found to be useful for the characterization or differentiation of bacterial strains harbouring plasmids, including Shigella (Chiou et al. 2001; Gebre-Yohannes & Drasar 1997; Litwin et al. 1991; Liu et al. 1995). Plasmid profiling is rapid, easy to perform and cost effective. Thus, the aim of this study was to evaluate this method for subtyping of our Malaysian Shigella spp. To the best of our knowledge, this is the first study to report the plasmid profile of a large collection of Malaysian Shigella spp.

272

C.H. Hoe et al.

Material and methods Bacterial strains A total of 219 clinical isolates of Shigella spp., isolated during 1994–2000 from sporadic cases of endemic shigellosis in different parts of Malaysia, were studied. Of the 219 clinical isolates, 146 (66.7%) were S. flexneri, 9 (4.1%) were S. dysenteriae, 4 (1.8%) were S. boydii, and 60 (27.4%) were S. sonnei. S. flexneri was represented by 9 serotypes: 1a (n ¼ 1), 1b (n ¼ 25), 2a (n ¼ 72), 3a (n ¼ 40), 3b (n ¼ 1), 4a (n ¼ 3), 4b (n ¼ 1), 6 (n ¼ 1), X (n ¼ 1), and Y (n ¼ 1). All S. dysenteriae isolates were of serotype 2, while each of the four isolates of S. boydii was of a different serotype (serotype 1, 2, 5, and 6). The Shigella spp. were identified by biochemical and serological tests (Mast Diagnostics, UK) at Institute for Medical Research, Kuala Lumpur, Malaysia. Repeated sub-culturing was avoided. Bacterial isolates were cultured in Luria Broth (LB) and stored in final concentration of 25% glycerol at )80 C.

Figure 1. Plasmid profiles of representative strains of Shigella spp. Lanes 1–10, S. flexneri serotype 1a, 1b, 2a, 3a, 3b, 4a, 4b, 6, Y and X, respectively; lane 11, S. dysenteriae type 2; lanes 12-14, S. boydii type 1, 2, 5, respectively; lane 15, S. sonnei; lane 16, S. flexneri 2b ATCC 12022; lane 17, S. sonnei ATCC 11060; lane 18, E. coli 39R861; lane 19, E. coli V517.

Antimicrobial susceptibility test Susceptibility was determined by the disk diffusion method (Bauer et al. 1966). The antimicrobial disks were obtained from Oxoid (UK), and the six antimicrobial agents tested included: ampicillin (AMP, 10 lg), tetracycline (TET, 30 lg), chloramphenicol (CHL, 30 lg), streptomycin (STR, 10 lg), kanamycin (KAN, 30 lg), and trimethoprim-sulfamethoxazole (SXT, 25 lg). Escherichia coli ATCC 25922, Staphylococcus aureus ATCC 25923, and Pseudomonas aeruginosa ATCC 27853 were used as quality control organisms. Zones of inhibition were recorded in mm and interpreted as sensitive, intermediate, or resistant according to the recommendations of the National Committee for Clinical Laboratory Standards (NCCLS, 2000).

Results

Plasmid profile analysis

All Shigella isolates were typeable and reproducible plasmid patterns were obtained when the analysis was repeated twice (data not shown). The isolates were divided into different groups based on the presence of small common plasmids in each serotype of Shigella spp.. The isolates in each group were then sub-divided into different profiles or patterns, according to the presence or absence of the less common plasmids and unique plasmids (Tables 2–4). Overall, heterogeneous plasmid patterns were found among the Malaysian Shigella spp. as the majority of the patterns were represented by only one isolate. Analysis of plasmid DNA showed that all Shigella spp. harboured 1 to 11 plasmids, ranging in size from 1.35 to 230 kb. A 3.20 kb plasmid was observed in 89% S. flexneri isolates. This plasmid was not found in other Shigella species (Table 2). S. flexneri 1b isolates displayed very different plasmid profiles compared to the other S. flexneri serotypes in this study. No large

Plasmid DNAs were extracted from 1.5 ml of overnight cell cultures, according to the method described by Olsen (1990) with some modifications. Approximately 2 ll of lysozyme (10 mg/ml) was added into lysis buffer prior to the 35 min of incubation. Extracted DNA was dissolved in 50 ll of pre-heated Tris-EDTA buffer (60 C). Plasmid DNAs were then separated by electrophoresis in 0.8% horizontal agarose gel at a constant voltage of 100 V for 6 h in cold 0.5 · Tris-borate buffer. Plasmid extraction of all Shigella strains was repeated twice to determine the reproducibility. Only bright bands were used in the analysis and faint bands were interpreted as relaxed forms of brighter bands. Reference plasmids carried in E. coli 39R861 and E. coli V517 were used as molecular weight markers. Plasmid sizes were calculated by using Gel-Compar II software (Applied Maths, Kortrijk, Belgium) (Figure 1).

Antimicrobial susceptibility Test Resistance rates of Shigella species to various antimicrobial agents are shown in Table 1. Twenty-eight percent of the S. flexneri isolates were susceptible to all six antimicrobial agents, whereas the remaining isolates were resistant to one to six antimicrobial agents. Fifty percent of the isolates were susceptible to all six antimicrobial agents, whereas the other 50% were resistant to one to four antimicrobial agents. Plasmid profile analysis

Plasmid profiling of Shigella species

273

Table 1. Antimicrobial resistance rates of Shigella species. Antimicrobial agent

Ampicillin Kanamycin Streptomycin Tetracycline Chloramphenicol Trimethoprim-sulfamethoxazole

Percent Shigella resistance S. flexneri (n = 146)

S. dysenteriae (n = 9)

S. boydiia (n = 4)

S. sonnei (n = 60)

66.4 0 71.9 66.4 64.4 42.5

0 0 100 0 0 0

0 0 25.0 25.0 0 25.0

6.67 0 43.3 33.3 1.67 38.3

a Resistance rate of S. boydii was contributed by S. boydii type 1 only. All other serotypes of S. boydii isolates showed full sensitivity to all antimicrobial agents tested.

(>150 kb) or mid-range (20.0–150 kb) plasmid above the chromosomal band was observed. Plasmids of 4.50, 3.20, 2.65, 2.25, and 1.55 kb were commonly found in this serotype. The 1.55 and 2.65 kb plasmids were observed in all S. flexneri 1b isolates, while the 4.50 kb plasmid was found in 96% of the isolates (Table 2). Five common small plasmids were observed in S. flexneri 2a: 11.4 (67.1%), 9.00 (76.4%), 5.00 (86.1%), 4.20 (89%), and 3.20 kb (98.6%). Both plasmids, 11.4 kb and 4.20 kb were unique for S. flexneri 2a and were not found in other serotypes of Shigella species. Other less common plasmids observed were the 230 kb (17.8%), and 163 kb (38.4%) plasmids. Occasionally, unique plasmids of different sizes, ranging from 2.70 to 215 kb were also observed. The large 230 kb was absent from the S. flexneri 2a strains isolated in 1994 but present in 9.1% (1 of 11) of strains isolated in 1996 (Table 2). Six of the most common small plasmids found in S. flexneri 3a isolates were the 7.40, 6.50, 4.85, 4.40, 3.50, and 3.20 kb plasmids, with prevalence rates ranging from 42.5 to 92.5%. The 4.40 kb plasmid was only observed in S. flexneri 3a isolates. A mid-range plasmid of 90.5 kb was observed in 45% of the isolates. A greater diversity of plasmids was observed in S. flexneri 3a compared to that of S. flexneri 2a. Fourty-three plasmids of different sizes were found among the 40 strains of S. flexneri 3a. Unique plasmids ranging from 1.50 to 200 kb in size were observed. Similarly, the 230 kb plasmid was only present in the strains isolated after 1996 (Table 2). The 60 isolates of S. sonnei were subtyped into 10 groups comprising of 53 profiles, with 44 different plasmids of different sizes ranging from 1.55 to 230 kb (Table 3). The 163 kb plasmid, which was found in S. flexneri and S. dysenteriae, was absent in S. sonnei. Four of the most common small plasmids found were 10.4, 8.20, 2.65, and 2.10 kb, with the prevalence ranging from 31.7 to 60%. Most of the isolates (60%) harboured the 2.10 kb plasmid, which was absent in other Shigella serotypes (Table 3). All the nine S. dysenteriae type 2 isolates were subtyped into 4 groups with a predominant 9.00 kb plasmid (Table 4). This plasmid was also prevalent in S. flexneri 2a isolates but less frequent in S. flexneri 1b (2 of

25), and serotype 3a (5 of 39). Each of the S. boydii isolates had a distinct plasmid profile, comprising of 3–7 plasmids, ranging in size from 1.90 to 216 kb.

Discussion Few published data are available on the association of the plasmid patterns of Shigella species with their serotypes. Haider et al. (1985) reported that 2.0 MDa (3.0 kb) and 2.7 MDa (4.1 kb) plasmids were found in 94 and 97% of 125 S. flexneri strains isolated in Dhaka, Bangladesh, respectively. S. flexneri 2a isolates only harboured two of these plasmids, whereas plasmids of various other sizes were found in the other serotypes. Litwin et al. (1991) reported similar finding in which a number of small plasmids ranging in size from 1.50 to 7.00 kb, were commonly seen in S. flexneri serotype 1, 2 and 4, including the 4.10 and 3.20 kb plasmid which were observed in 54 and 77% of the isolates, respectively. In this study, we also found several small plasmids common to S. flexneri 1b, 2a, and 3a. 44.0% of S. flexneri 1b isolates had plasmid profile F1.1b.P6. All 25 of the S. flexneri 1b isolates harboured the 1.55 (1.0 MDa) and 2.65 kb (1.75 MDa) plasmids, whereas 96% harboured the 4.50 kb (3.0 MDa) plasmid. Each of the plasmids, 3.20 (2.1 MDa) and 2.25 kb (1.5 MDa), was present in 84% of the isolates. These five plasmids were observed in both the susceptible and resistant strains. Thus, these plasmids might be considered as the core plasmids of serotype 1b, and might be useful in identification of the isolates. A similar result was reported by Talukder et al. (2003), in which plasmids of 2.8, 2.1, 1.8, and 1.0 MDa were found in all the 144 strains of S. flexneri serotype 1 isolated in Bangladesh. Interestingly, no large or mid-range plasmid was observed in the S. flexneri 1b isolates in this study. In contrast to our finding, Talukder et al. (2003), found that all the S. flexneri 1a and 1b strains and about 88% of the 1c strains in their study harboured the 140 MDa plasmid. Therefore, this unique characteristic could be useful to differentiate our Malaysian strains of S. flexneri 1b from isolates of other countries.

274

C.H. Hoe et al.

Table 2. Plasmid profiles of Shigella flexneri. Group

Common small plasmids (kb)

S. flexneri 1b F1.1b

4.50

3.20

2.65

2.25

1.55

F2.1b F3.1b F4.1b F5.1b F6.1b S. flexneri 2a F1.2a

4.50 4.50 4.50 4.50 –

3.20 – – – 3.20

2.65 2.65 2.65 2.65 2.65

– 2.25 2.25 – –

1.55 1.55 1.55 1.55 1.55

11.4

9.00

5.00

4.20

3.20

F2.2a F3.2a

11.4 –

9.00 9.00

– 5.00

4.20 4.20

3.20 3.20

F4.2a

–

9.00

–

–

3.20

F5.2a

–

–

5.00

4.20

3.20

F6.2a F7.2a

– –

– –

– –

4.20 –

3.20 3.20

S. flexneri 3a F1.3a

–

6.50

4.85

4.40

3.50

3.20

F2.3a

7.40

6.50

–

4.40

3.50

3.20

F3.3a F4.3a F5.3a F6.3a F7.3a

7.40 7.40 7.40 7.40 –

6.50 6.50 6.50 6.50 6.50

– – – – –

– 4.40 – – –

3.50 – – – –

3.20 – 3.20 – 3.20

Pattern

Other plasmids present (kb)

No. Strain

P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11

10.9, 2.15 9.00, 2.15 9.00 2.90 2.15 – – – 2.15 3.55 –

1 1 1 1 3 11 2 2 1 1 1

P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15 P16 P17 P18 P19 P20 P21 P22 P23 P24 P25 P26 P27 P28 P29 P30 P31 P32 P33 P34

a

230, 163, 136, 54.5 230, 163, 159, 66.0 230, 163, 54.5 230, 163 230 163 163, 68.0 215 200 190 142 72.0 47.0 2.70 – – 230, 163 – 163 – 230, 163, 80.0, 72.0, 54.5 230, 163, 80.0, 54.5 230, 163 230, 136 230 163 155 136, 63.0 6.70 – 119 163, 4.80 163 127

1 1 1 1 3 13 1 1 1 1 2 2 1 1 18 1 1 1 1 3 1 1 1 1 1 1 2 1 1 2 1 1 2 1

P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15

230, 230, 230, 230, 200, 200, 163, 230, 163, 163, 163, 163, 163, 230, 230, 3.85 163, 159, 1.75

P16 P17 P18

163, 90.5 90.5, 61.0 90.5, 1.75 90.5 120, 90.5 90.5 130, 90.5 163, 86.5 90.5, 69.0, 11.3, 8.60, 2.45 90.5 90.5 90.5 86.5 163, 86.5, 80.0 163, 90.5, 9.00, 5.30, 4.30,

2 1 1 2 1 1 1 1 1 3 1 1 1 1 1

106, 1.75 1.75

1 1 1

Plasmid profiling of Shigella species

275

Table 2. (Continued.) Group

Common small plasmids (kb)

F8.3a F9.3a F10.3a F11.3a

– – – –

– – – –

4.85 4.85 4.85 4.85

4.40 4.40 – –

3.50 – 3.50 –

3.20 3.20 3.20 3.20

F12.3a F13.3a

– –

– –

– –

4.40 4.40

3.50 –

3.20 3.20

F14.3a F15.3a

– –

– –

– –

– –

3.50 –

3.20 3.20

a

Pattern

Other plasmids present (kb)

P19 P20 P21 P22 P23 P24 P25 P26 P27

90.5 163, 1.75 9.80, 9.00 230,163, 9.80, 9.00, 4.30 163, 66.0, 9.80, 9.15, 9.00, 4.30 163, 9.00, 4.30 159, 1.75 163, 130, 86.5 173, 95.5, 8.30, 6.70, 5.00, 3.75, 2.05, 1.50 163, 90.5, 80.0, 3.10 230, 163, 6.25, 1.35 230, 159, 6.25, 1.75 230, 2.05 200, 130, 6.25 163, 16.5, 13.0, 5.80, 5.30, 4.30 159, 1.75

P28 P29 P30 P31 P32 P33 P34

No. Strain 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2

The 230 kb plasmid shown in the table was not included into plasmid profiling due to its instability on long storage and subculture.

S. flexneri 2a is the predominant serotype encountered in developing countries (60%) (WHO 2003), including Malaysia (Kotloff et al. 1999). In our study, S. flexneri 2a constituted about 33% of the total Shigella isolates. Plasmid profile analysis showed that five small plasmids ranging in size from 3.20 to 11.4 kb were commonly seen in S. flexneri 2a isolates. The 3.20 kb plasmid was observed in all the isolates, whereas the 4.20 and 5.00 kb plasmids were found in 88.9 and 86.1% of the isolates, respectively. These three plasmids were also found in both the susceptible and resistant strains and therefore, could be considered as the core plasmids of serotype 2a. The 11.4 and 4.20 kb plasmids were not observed in other serotypes of Shigella species. This finding suggests that these two plasmids could be used as potential extrachromosomal markers in identification of S. flexneri 2a isolates. Although Kotloff et al. (1999) cited that S. flexneri 3a is the third most prevalent serotype of S. flexneri isolated in developing countries, there were very few published reports on this serotype. In this study, six common small plasmids were observed among 40 strains of S. flexneri 3a isolates. About 92.3% of the isolates harboured the 3.20 kb plasmid, whereas the prevalence of the five other common plasmids was relatively low (42.5 to 52.5%). Apparently, the existence of these plasmids was not stable. The 4.40 kb plasmid was only present in S. flexneri 3a (47.5%). Nevertheless, due to its low prevalence the usefulness of this plasmid as an extra-chromosomal marker is limited. The low prevalence of these common plasmids, along with the presence of numerous plasmids of different sizes in the S. flexneri 3a isolates suggested that these strains have multiple unrelated origins. Most of the S. flexneri isolates in this study harboured the 3.20 kb plasmid (89%). Several previous studies also showed the high prevalence of this 3.20 kb plasmid in different serotypes of S. flexneri. The presence of the

3.20 kb plasmid in different serotypes of S. flexneri isolates from widely separated geographic regions such as Malaysia, Bangladesh (Talukder et al. 2003), Arizona (Litwin et al. 1991), and Taiwan (Chiou et al. 2001) suggests that it may carry important housekeeping genes that are important to the survival of this species. A correlation between resistant phenotypes with plasmids was described in a number of previous studies (Talukder et al. 2002, 2003), in which resistance to tetracycline, ampicillin, and trimethoprim-sulfamethoxazole in S. flexneri was associated with mid-range plasmids. However, such an association was not observed in this study. Twenty eight percent of S. flexneri 1b isolates were resistant to multiple antibiotics, including tetracycline, ampicillin, and trimethoprimsulfamethoxazole. As mentioned earlier, all isolates of serotype 1b harboured only small plasmids. Multi-drug resistant strains were also observed in other serotypes of S. flexneri isolates, with and without mid-range plasmids. Genes conferring antibiotic resistance traits could be located on chromosomal DNA or other small plasmids. This assumption is supported by the study of Casalino et al. (1994), who observed that genes for resistance to ampicillin, chloramphenicol, spectinomycin, and tetracycline formed a linkage group located on the chromosome of the strains of all serotypes of S. flexneri isolated in Somalia. However, further study is needed in order to confirm this assumption. Plasmid profiling appeared to be insufficient to characterize multiple antibiotic resistance in our Malaysian strains of S. flexneri. Hence, other molecular subtyping techniques such as PFGE, PCR based technologies, are currently being applied to further characterize these S. flexneri strains. S. dysenteriae type 2, also known as Schmitz bacillus, is rarely encountered worldwide. Nevertheless, it could be the predominant serotype of S. dysenteriae circulating in Malaysia, as more than 93% of the S. dysenteriae

276

C.H. Hoe et al.

Table 3. Plasmid profiles of Shigella sonnei. Group

Common small plasmids (kb)

S1

10.4

8.20

2.65

2.10

S2

10.4

8.20

2.65

–

S3

10.4

–

2.65

2.10

S4

10.4

–

–

2.10

S5

10.4

–

–

–

S6

–

8.20

2.65

2.10

S7 S8 S9

– – –

8.20 8.20 –

2.65 – –

– – 2.10

S10

–

–

–

–

a

Pattern

Other plasmids present (kb)

No. strain

P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15 P16 P17 P18 P19 P20 P21 P22 P23 P24 P25 P26 P27 P28 P29 P30 P31 P32 P33 P34 P35 P36 P37 P38 P39 P40 P41 P42 P43 P44 P45 P46 P47 P48 P49 P50 P51 P52 P53

117, 107, 11.5, 6.40, 5.00 1 83.0, 14.8, 11.5, 6.40 1 14.8, 11.5, 6.40 6 a 230,155, 14.8, 11.5, 6.40 1 230, 88.0, 14.8, 11.5, 6.40 1 14.8, 11.5 1 14.8, 11.5, 6.40 1 14.8, 3.60, 2.55, 2.45 1 11.5, 8.90 1 11.5 1 8.50, 5.00, 4.80 1 8.50, 2.50, 2.45 1 – 1 230, 129, 88.0, 13.3, 11.5, 8.50 1 129, 101, 8.50 1 8.50, 2.30 1 8.50 1 230, 14.8, 6.40 2 14.8, 6.40 1 14.8 1 – 1 14.8 1 129, 80.0, 11.0 1 230, 101, 6.40 1 200, 133, 105 1 102, 8.60 1 93.0 1 83.0 2 80.0, 8.60 1 80.0, 3.10 1 80.0 1 11.9, 2.50, 2.45 1 8.60 1 8.50, 5.00, 4.50, 2.50, 2.45 1 8.50, 2.50, 2.45 1 3.10 1 2.50, 2.45 1 – 1 230, 129, 97.0, 14.4, 1.60 1 230, 129, 11.9, 11.2, 8.50, 7.30, 1.55 1 230, 6.40 1 230, 1.60 1 135, 11.9, 11.2 1 133, 11.0 1 129, 97.0, 5.00, 1.60 1 129, 11.9, 11.2 1 129, 11.0 1 129, 8.50, 4.80, 4.50, 1.60 1 129, 8.50 1 129, 1.60 1 129 1 71.0 1 8.50, 2.50, 2.45 1

The 230 kb plasmid shown in the table was not included into plasmid profiling due to its instability on long storage and subculture.

isolates submitted to the Institute for Medical Research, Malaysia during 1994–2000 were serotype 2. Each of the S. dysenteriae type 2 isolates had a distinct plasmid pattern, with the 9.00 kb plasmid as its core plasmid. All these strains were isolated in Ipoh, Malaysia, and were resistant to streptomycin only. As resistance to streptomycin in S. dysenteriae is either chromosomal or smaller plasmid-mediated (Gebre-Yohannes & Drasar, 1990),

there is a possibility that this resistance is linked to the 9.00 kb plasmid. The uniformity in antimicrobial resistance pattern, and geographical distribution, along with the presence of the 9.00 kb core plasmid might suggest a common source of infection. Numerous plasmids of different sizes were observed in 60 strains of S. sonnei isolates. Four of the most common plasmids (10.4, 8.20, 2.65, and 2.10 kb) were

Plasmid profiling of Shigella species

277

Table 4. Plasmid profiles of Shigella dysenteriae type 2. Group

Common small plasmids (kb)

D2.1

9.00

6.20

2.55

D2.2 D2.3

9.00 9.00

6.20 –

– 2.55

D2.4

9.00

–

–

a

Pattern

Other plasmids present (kb)

No. Strain

P1 P2 P3 P4 P5 P6 P7 P8 P9

a

1 1 1 1 1 1 1 1 1

230, 163 230 216 163 – 216, 163, 2.45 230, 5.80 216 163, 5.80, 2.45

The 230 kb plasmid shown in the table was not included into plasmid profiling due to its instability on long storage and subulture.

observed at relatively low prevalence (31.7 to 60%). The 2.10 kb plasmid, which was found in 60% of S. sonnei isolates, was not observed in other Shigella species. Although previous studies (Lee et al. 2003) reported the association of multi-drug resistance in S. sonnei with a mid-range plasmid of 110 kb, the Malaysian multi-drug resistant (trimethoprim-sulfamethoxazole, streptomycin, tetracycline) strains of S. sonnei may be associated with the 14.8 kb small plasmid, as all the isolates that possessed this resistant phenotype harboured this plasmid. However, this plasmid was absent from other S. sonnei isolates that had additional resistance to ampicillin. The genes associated with virulence in Shigella are found on a large 230 kb plasmid, often referred as virulence plasmid or invasive plasmid. The size of the virulence plasmid ranged from 180 kb (S. sonnei) and 210 to 240 kb (S. flexneri) (Sasakawa 1995). It was noted that the 230 kb virulence plasmid was absent from all the Shigella strains isolated in 1994, and had a very low prevalence rate in strains isolated in 1996 compared to the later years (1997–2000). This indicated that the 230 kb plasmid is unstable on long storage or subculture, as suggested by Litwin et al. (1991). Thus, this plasmid was not included in the plasmid profiling. The detailed analysis of this plasmid is beyond the scope of this study and will be further investigated in future. Our study showed that plasmid profiling is a valuable typing technique in subtyping Shigella species. Heterogeneous plasmid patterns were seen in different serotypes of Shigella in this study. In this instance, plasmid profiling offered a refinement beyond serotyping. Technically, plasmid profile analysis is one of the oldest DNA-based methods and can be efficiently performed with basic electrophoretic equipment. In conclusion, our study has demonstrated that plasmid profiling is useful in differentiating isolates of Shigella in our population. We found the correlation between the serotypes and plasmid patterns in Shigella species, which may be useful for the detection of the epidemic strains. Unique plasmid characteristic observed in S. flexneri 1b isolates was a good indicator in differentiating our Malaysian strains of serotype 1b isolates from those of other countries. Several plasmids

in this study were found to have potential to be used as extrachromosomal markers for identification of new serotypes and for differentiation of existing serotypes as well. To the best of our knowledge, this is the first report on plasmid analysis of Malaysian strains of Shigella species, and the data generated will be useful for the future surveillance of this genus. Acknowledgement This work was supported by the IRPA grants 06-02-031007 and 06-02-03-0750 from Ministry of Science, Technology and Environment, Malaysia, Vote-F F0107/2003B and PASCA from University of Malaya. References Bauer, A.W., Kirby, W.M., Sherris, J.C. & Turek, N. 1966 Antibiotic susceptibility testing by a standardized single disk method. American Journal of Clinical Pathology 45, 493–496. Casalino, M., Nicoletti, M., Salvia, A., Colonna, B., Pazzani, C., Calconi, A., Mohamud, K.A. & Maimone. F. 1994 Characterization of endemic Shigella flexneri strains in Somalia: antimicrobial resistance, plasmid profiles, and serotype correlation. Journal of Clinical Microbiology 32, 1179–1183. Chiou, C.S., Hsu, W.B., Wei, H.L. & Chen, J.H. 2001 Molecular epidemiology of a Shigella flexneri outbreak in a mountainous township in Taiwan, Republic of China. Journal of Clinical Microbiology 40, 1048–1056. Gebre-Yohannes, A. & Drasar, B.S. 1990 Plasmid profiles of antibiotic-resistant Shigella dysenteriae types 2, 3, 4, 6 and 7 isolates in Ethiopia during 1976–85. Epidemiology and Infection 105, 65–72. Gebre-Yohannes, A. & Drasar, B.S. 1997 Plasmid profiles of drug resistant Shigella boydii types 1–5, 8, 10, 12–14 from Ethopia (1974–85). Epidemiology and Infection 119, 293–98. Haider, K., Huq, M.I., Samadi, A.R. & Ahmad, K. 1985 Plasmid characterization of Shigella spp. isolated from children with shigellosis and asymptomatic excretors. Journal of Antimicrobial Chemotherapy 16, 691–698. Kotloff, K.L., Winickoff, J.P., Ivanoff, B., Clemens, J.D., Swerdlow, D.L., Sansonetti, P.J., Adak, G.K. & Levine, M.M. 1999 Global burden of Shigella infections: Implications for vaccine development and implementation of control strategies. Bulletin of the World Health Organization 77, 651–666 Lee, T.M., Chang, C.Y., Chang, L.L., Chen, W.M., Wanf, T.K. & Chang, S.F. 2003 One predominant type of genetically closely related Shigella sonnei prevalent in four sequential outbreaks in school children. Diagnostic Microbiology and Infectious Disease 45, 173–181.

278 Lee, W.S. & Puthucheary, S.D. 2002 Species distribution and antibiotic resistance of Shigella isolates in an urban community in Malaysia. Medical Journal of Malaysia 58, 262–267. Liu, P.Y., Lau, Y.J., Hu, B.S., Shry, M.J.Z., Shi, Y., Tsai, W.S., Lin, Y.H. & Tseng, C.Y. 1995 Analysis of clonal relationships among isolates of Shigella sonnei by different typing methods. Journal of Clinical Microbiology 33, 1779–1783. Litwin, C.M., Storm, A.L., Chipowsky, S. & Ryan, K.J. 1991 Molecular epidemiology of Shigella infections: Plasmid profiles, serotype correlation, and restriction endonuclease analysis. Journal of Clinical Microbiology 29, 107–108. National Committee for Clinical Laboratory Standards. 2000 Performance standards for antimicrobial disk susceptibility tests. Approved standard M2-A7. NCCLS, Wayne PA. Olsen, J.E. 1990 An improved method for isolation of plasmid DNA from wild type Gram negative bacteria for plasmid restriction profile. Letters in Applied Microbiology 10, 209–212.

C.H. Hoe et al. Sasakawa, C. 1995 Molecular basis of pathogenicity of Shigella. Reviews in Medical Microbiology 6, 257–266. Talukder, K.A., Islam, M.A., Dutta, D.K., Hassan, F., Safa, A., Nair, G.B. & Sack, D.A. 2002 Phenotypic and genotypic characterization of serologically atypical strains of Shigella flexneri type 4 isolated in Dhaka, Bangladesh. Journal of Clinical Microbiology 40, 2490–2497. Talukder, K.A., Islam, Z., Islam, M.A., Dutta, D.K., Safa, A., Ansaruzzaman, M., Faruque, A.S.G., Shahed, S.N., Nair, G.B. & Sack, D.A. 2003 Phenotypic and genotypic characterization of provisional serotype Shigella flexneri 1c and clonal relationships with 1a and 1b strains isolated in Bangladesh. Journal of Clinical Microbiology 41, 110–117. World Health Organization. 2003 State of the art of new vaccines research and development. W.H.O./IVR

 Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 279–283 DOI 10.1007/s11274-004-3632-z

Physico-chemical characterization of exomannan from Rhodotorula acheniorum MC K. Pavlova1,*, I. Panchev2 and Ts. Hristozova1 1 Laboratory of Applied Microbiology, Institute of Microbiology, Bulgarian Academy of Sciences, 26 Maritza Blvd, 4002 Polvdiv, Bulgaria 2 University of Food Technologies, 26 Maritza Blvd., 4002 Plovdiv, Bulgaria *Author for correspondence: Tel.: +359-32-603-831, Fax: +359-32-440-102/2-700-109, E-mail: [email protected] Received 8 March 2004; accepted 16 July 2004

Keywords: Exomannan, mannose, physicochemical properties, Rhodotorula acheniorum MC, yeast

Summary The batch fermentation of Rhodotorula acheniorum MC on a culture medium containing 5% sucrose, mineral salts and yeast extract at 26
C for 96 h, with aeration at 0.75 v/v/m and agitation at 500 rev min)1 resulted in the synthesis of an exopolysaccharide (6.2 g l)1) which formed two fractions upon precipitation. The fractions were purified to a carbohydrate content of 98.2% for fraction I and 87.3% for fraction II. Mannose was the main monosaccharide component in a 92.8% concentration in fraction I and a 90.6% concentration in Fraction II. The exopolysaccharide was thus a mannan. The gel chromatograms confirmed the chemical composition of both fractions. The molecular weight of mannan I was 310 kD, whereas that of mannan II was 249 kD. The mannan I intrinsic viscosity [g] ¼ 6.23 dl g)1 was higher than that of mannan II [g] ¼ 2.73 dl g)1. The water-binding capacity of the mannan samples was established within the 1.2–3.5 g g)1 range. The multiplicative model [g] ¼ 387.22. D
0:1913 . T)1.095. C1.814 describing the effect of the velocity gradient Dr, the exomannan r concentration C and the temperature T on the dynamic viscosity values g of polymer solutions was obtained. Introduction Microbial polysaccharides are water-soluble macromolecules, which increase the medium viscosity under the influence of different physical and chemical agents. Because of the wide diversity in their composition, structure and physical properties, they can change the rheology and texture of the products in which they are supplemented. Microbial exopolysaccharides have found application in the food, pharmaceutical and other industries (Sandford et al. 1984; Crescenzi 1995; Vandamme et al. 1996). Biopolymers with industrial application are bacterial and fungal products like xanthan, dextran and scleroglucan (Margaritis & Pace 1985; Paul et al. 1986; Sutherland 1998). Yeast belonging to the different Cryptococcus, Hansenula, Rhodotorula, Lipomyces, Bullera, Sporobolomyces genera can synthesize exopolysaccharides. The polymer types reported for yeast producers include mannans, glucans, glucomannans, galactomannans and phosphomannans (Elinov et al. 1979, 1992; Chiura et al. 1982a, 1982b; Heald & Kristiansen 1985; Peterson et al. 1989, 1990; Vitovskaya et al. 1989; Adami & Cavazzoni 1990; Vorotynskaya et al. 1992). The authors of this paper have not come across any reports in the literature concerning the investigation of the yeast Rhodotorula acheniorum in terms of exopolysaccharide production. The strain was

first repoted as an exomannan producer in one of our previous works (Pavlova & Grigorova 1999). Yeast polysaccharide applications in the food, fodder, pharmaceutical and other industries require detailed knowledge of the physico-chemical properties of the polysaccharides synthesized. The objectives of this paper were to identify the exopolysaccharide produced by R. acheniorum MC and investigate its physico-chemical properties. Materials and methods Strains The R. acheniorum MC yeast strain was selected as a suitable exopolysaccharide producer (Pavlova & Grigorova 1999). It was registered in the National Bank for Industrial Microorganism and Cell Cultures, Bulgaria.

Exopolysaccharide production The inoculum of R. acheniorum MC was prepared by cultivation in 500 ml Erlenmeyer flasks containing 50 ml medium with 20 g malt extract l)1 (Fluka Chemie AG, Buchs, Switzerland) on a rotary shaker (220 rev min)1) at 26
C for 48 h. The culture medium was inoculated with 1% (w/v) of inoculum. The batch fermentation was

280 carried out in a 10-l laboratory bioreactor with 7.5-l working volume (Chemap AG, Mannedorf, Switzerland). The fermentation medium contained (g l)1): sucrose, 50; (NH4)2SO4, 2.5; KH2PO4, 1.0; MgSO4.7H2O, 0.5; NaCl, 0.1; CaCl2.2H2O, 0.1; yeast extract, 1.0. The initial pH was adjusted to pH 5.3 and medium was sterilized at 112
C for 30 min. The fermentation occurred at an aeration rate of 0.75 v/v/ m and agitation at 500 rev min)1 at 26
C for 96 h. Isolation of crude exopolysaccharide The cell cultures were centrifuged at 6000 · g for 30 min. The exopolysaccharides in the supernatant were precipitated with two volumes of 96% ethanol and two fractions were formed: one at the time of precipitation, and the other one after storage for 24 h at 4
C. The precipitates were recovered by centrifugation at 6000 · g for 10 min, washed with ethanol, dried and weighed. Polysaccharide purification The crude exopolysaccharides (2.0 g) were dissolved in distilled water (100 ml) and treated with Fehling’s reagent (12.5 ml). Mannan precipitation by copper salts is based on the formation of a water-insoluble complex. The precipitated mannan–copper complex was washed with 2.0% KOH, then with ethanol and dissolved in distilled water (100 ml). The solution was treated with a Wofatit KPS cation exchanger to eliminate the copper ions. Then the solution was precipitated with two volumes of ethanol. The purified mannans were washed with ethanol and dried at 50
C under vacuum. Analytical methods The total amount of carbohydrate in the exopolysaccharides was determined using the phenol–sulphuric acid method (Dubois et al. 1956). The amount of total protein was determined by means of a Kjeltec Auto 1030 Analyzer (Tecator, Sweden). The ash content was estimated after heating for 2 h in a muffle furnace at 550
C for 3 h. The carbohydrate composition was determined by gas chromatography using Fractovap 2407 (Carbo Erba) after hydrolysis of the exopolysaccharides with 2M H2SO4 at 105
C for 8 h, neutralization with barium hydroxide, followed by centrifugation and using a Wofatit KPS (VEB Chemiecombinat Bitterfeld, Germany) cation exchanged for the elimination of the Ba ions. For the purpose of the gas chromatographic analysis, the monosaccharides were transformed into volatile derivatives by silanization using hexamethyldisilazane in the presence of the catalyst trifluoracetic acid in a solution of pyridine and then gas chromatographed under the following conditions: column of length 2000 mm with diameter 4 mm, packed with 2% SE-54 on Chromosorb W (silanizated) 80/100 mesh; flame ionization detector; carrier gas: nitrogen with a flow rate of 35 ml min)1; temperature of the injector and detector 350
C; initial

K. Pavlova et al. temperature of the programme 120
C with an increase of 4
C min)1. The purified fractions designated as mannan I and mannan II were subjected to gel filtration with Sephadex G-200 in a 25 · 250 mm column. 0.2% mannan solution was passed through the column in 5 ml samples. Distilled water was used as the eluant at an elution rate of 12 ml h)1 and 7 ml samples were collected. The carbon content in the samples was measured by an LI-3 laboratory interferometer (Karl Zeiss, Germany). The water-binding capacity (WBC) of the mannan samples was measured following Baumann’s capillary suction method (Wallingford & Labuza 1983). The mean molecular weight (Mn) of the mannan samples was determined according to the osmometric method by a Knauer type 73101 (Germany) membrane osmometer (Morawetz 1967). The intrinsic viscosity [g] of the water–mannan solutions was measured at 25
C by a VPG 2 Ubelode type capillary viscosimeter with a capillary diameter of 0.56 · 10)3 m. The experimental data processing was made by means of a computer program written in FORTRAN 77 on the basis of a linear regression equation algorithm according to the Huggins equation (Morawetz 1967). The dynamic viscosity g of the water–mannan solutions was measured by a Rheotest-2, Rheoviscosimeter (VEB MLW Pru¨fgera¨tewerk Medingen/Dresden, Germany) using measuring cylinder N of the device. The numerical values of the velocity gradient Dr and the modulus of shearing were processed according to Oswald de Waele’s equation (Lapasin & Pricl 1994) using a REO computer program written in FORTRAN 77. The least squares method was used for the purpose of finding a simple and linear regression equation (Draper & Smith 1981). The numerical procedures were carried out by computer programs compiled by Johnson (1980).

Results and discussion The scheme for exomannan production by R. acheniorum MC is shown in Figure 1. The parameters established during the batch fermentation of the strain indicated that a temperature of 26
C, aeration at 0.75 v/v/m and agitation at 500 rev min)1 proved to be the most suitable conditions for polysaccharide synthesis. As a result, 6.2 g l)1 of crude exopolysaccharide were produced after 96 h. Two fractions were obtained after precipitation. Fraction I, formed at the time of precipitation, was compact and fibrous in texture. Fraction II was formed after 24-h storage of the precipitated polysaccharide at 8
C and was a fine-grained, noncompact mass. Data leading to the selection of the cultivation conditions were published in previous our article (Pavlova & Grigorova 1999). After the fractions had been purified, their composition was analyzed. The chemical composition of fraction I and fraction II is shown in Table 1.

Physico-chemical characterization of exomannan

281

Figure 1. Scheme for obtaining of exomannan by R. acheniorum MC.

Table 1. Chemical composition of the pure exomannan synthesized by Rh. acheniorum MC. Polysaccharide

Mannan I

Mannan II

Carbohydrate content (%) Protein (%) Ash (%) Monosaccharides (% of carbohydrate content) Mannose Glucose

98.2 ± 1.3 1.2 ± 0.1 0.4 ± 0.1

87.3 ± 1.1 8.0 ± 0.8 4.7 ± 0.3

92.8 ± 1.4 7.2 ± 0.9

90.6 ± 0.1 9.4 ± 0.9

The values represent two determinations per analysis.

The carbohydrate content of the purified polysaccharide fractions was 98.2% for mannan I and 87.3% for mannan II. Mannose was the main component and its concentration was 92.8% for fraction I and 90.6% for fraction II. The high mannose content of the two polysaccharide fractions served as a basis for their

designation as mannan I and mannan II and the subsequent investigation of their physico-chemical properties. The water-binding capacity of the mannan samples was measured by means of the capillary suction method at 25
C. The WBC quantitative values ranged between 1.2 and 3.5 g g)1 and depended on the sample dispersiveness. As seen from the gel chromatograms in Figures 2 and 3, both fractions contained a high-molecular weight component accompanied by low-molecular weight components present in different proportions in the two fractions. The data intrinsic viscosity confirmed the result of the osmometric investigations, i.e. that the molecular weight of mannan I was higher. The viscosimetric data on purified mannan demonstrated a 2.6-times increase of [g] in mannan I and 1.7-times increase of [g] in mannan II (Table 2). The application of polysaccharides such as xanthans, galactomannans and glucomannans as thickeners,

282

K. Pavlova et al.

Figure 4. Dynamic viscosity of water–mannan solutions vs. shear rate at 25
C.

Figure 2. Gel chromatogram of exomannan – Fraction I.

Figure 3. Gel chromatogram of exomannan – Fraction II.

Figure 5. Dynamic viscosity of water–mannan solutions vs. shear rate at 80
C.

Table 2. Molecular weight and intrinsic viscosity of mannan fractions. Mannan fractions

Molecular Weight(kD)

Intrinsic viscosity (dl g)1)

Crude mannan I Crude mannan II Pure mannan I Pure mannan II

ND ND 310 ± 25 249 ± 20

2.40 1.54 6.23 2.73

± ± ± ±

0.08 0.09 0.12 0.21

ND – No data.

emulsifiers, stabilizers, gelling and suspending agents is defined by the rheological properties of their solutions. Viscosity is one of the most important rheological characteristics and depends on different factors. The use of mannan as a thickener requires good knowledge of its rheological characteristics at increased concentrations (> 0.1%). Therefore, the effect of mannan I concentration (with [g] ¼ (6.23 ± 0.12) dl g)1), the shear rate Dr and the temperature on the dynamic viscosity value g was investigated. The change in the dynamic viscosity of mannan solutions at 25 and 80
C is shown in Figures 4 and 5. As a result of the

experiments conducted within a temperature range of 25 < T < 80
C for 1.0–4.0% mannan concentrations and Dr values of 1.5 < Dr < 1312 s)1, the following multiplicative model was obtained: ½g ¼ 387:22:D
0:1913 :T
1:095 :C1:814 r

ð1Þ

The statistical regression analysis of the experimental data for evaluation of the adequacy of Equation 1 obtained by means of the computer programs complied by Johnson (1980) established the multiple correlation coefficient at Rm ¼ 0.9854, which suggested that the equation could be used to predict [g] in relation to the changes in Dr, T and C. The water-mannan solutions showed lower absolute dynamic viscosity values than similar microbial polysaccharide solutions owing to their lower molecular weight. In this study, the production of exomannan by R. acheniorum MC and its physico-chemical properties have been explored. The strain demonstrated biosynthesis of exomannan when sucrose was used as a carbon

Physico-chemical characterization of exomannan source. Mannose was the main component in the monosaccharide composition of the polysaccharide. The molecular weight, the intrinsic and dynamic viscosity and the water-binding capacity of mannan have been established for further application in the food industry.

References Adami, A. & Cavazzoni, V. 1990 Exopolysaccharides produced by some yeast strains. Annali di Microbiologia ed Enzimologia 40, 245– 253. Chiura, H., Iizuka M. & Yamamoto, T. 1982a A glucomannan as an extracellular product of Candida utilis. I. Production and characterization of a glucomannan. Agricultural and Biological Chemistry 46, 1723–1733. Chiura, H., Iizuka M. & Yamamoto, T. 1982b A glucomannan as an extracellular product of Candida utilis. II. Structure of a glucomannan and characterization of oligosaccharides obtained by partial hydrolysis. Agricultural and Biological Chemistry 46, 1733– 1742. Crescenzi, V. 1995 Microbial polysaccharides of applied interest: ongoing research activities in Europe. Biotechnology Progress 11, 251–259. Draper, N.R. & Smith, H. 1981 Applied Regression Analysis. New York: Wiley. ISBN 0-471-02995-5. Dubois, M., Gilles, K., Hamilton, Y., Roberts, P. & Smith, F. 1956 Colorimetric method for determination of sugars and related substances. Analytical Chemistry 28, 350–356. Elinov, N.P., Ananyeva, E.P. & Vitovskaya, G.A. 1992 Peculiarities of biosynthesis and characteristics of exoglucanes in yeasts of Sporobolomyces genus. Microbiologiya 4, 615–621. Elinov, N.P., Vitovskaya, G.A., Marikhin, V.A. & Koslova, T.V. 1979 Mannan produced by Rhodotorula rubra strain 14. Carbohydrate Research 75, 185–190. Heald P.J. & Kristiansen, B. 1985 Synthesis of polysaccharide by yeast-like forms of Aureobasidium pullulans. Biotechnology and Bioengineering 27, 1516–1519.

283 Johnson, K.J. 1980 Numerical methods in chemistry. New York: Marcel Dekker Inc. ISBN 0-82476818-3. Lapasin, R. & Pricl, S. 1994 Rheology of industrial polysaccharides theory and application. London: London Blackie Academic and Professional. ISBN 0-75140211-7. Margaritis, A. & Pace, G.W. 1985 Microbial polysaccharides. Comprehensive biotechnology, vol. 3, pp. 1005–1041. Oxford, UK: Pergamon Press. ISBN. Morawetz, H. 1967 Macromolecules in solution. New York: Interscience Publishers. Pavlova, K. & Grigorova, D. 1999 Production and properties of exopolysaccharide by Rhodotorula acheniorum MC. Food Research International 32, 473–477. Paul, F., Morin, P. & Monsan, P. 1986 Microbial polysaccharides with actual potential industrial applications. Biotechnology Advances 4, 245–259. Peterson, G.R., Nelson, G.A., Cathey, C.A. & Fuller, G.G. 1989 Rheologically interesting polysaccharides from yeasts. Applied Biochemistry and Biotechnology 20/21, 845–867. Peterson, G., Schubert, W.W., Richards, G.F. & Nelson, G.A. 1990 Yeasts producing exopolysaccharides with drag-reducing activity. Enzyme and Microbial Technology 12, 255–259. Sandford, P.A., Cotterell, L.W. & Pettitt, D.J. 1984 Microbial polysaccharides: new products and their commercial applications. Pure and Applied Chemistry 56, 879–895. Sutherland, I.W. 1998 Novel and established applications of microbial polysaccharides. Trends in Biotechnology 16, 41–45. Vandamme, E., Bruggeman, G., Baets, S. & Vanhooven, P. 1996 Useful polymers of microbial origin. Agro Food Industry Hitech 7, 21–25. Vitovskaya, G.A., Samarkina, G.M., Ananyeva, E.P. & Sinitskaya, I.A. 1989 Synthesis of extracellular heteroglycans by Cryptococcus species.Microbiologiya 58, 240–245. Vorotynskaya, S.L., Vitovskaya, G.A. & Ananyeva, E.P. 1992 Studies on the properties of polysaccharides produced by the yeasts Cryptococcus luteolus (Saito) skinner. Microbiologiya i Fitopatologiya 26, 367–371. Wallingford, L. & Labuza, T.P. 1983 Evaluation of the water binding properties of food hydrocolloids by physical/chemical methods and in a low fat meat emulsion. Journal of Food Science 48, 1–5.


Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 285–292 DOI 10.1007/s11274-004-3633-y

Isolation and characterization of Bacillus thuringiensis strains from different grain habitats in Turkey O¨zgu¨r Apaydin1, A. Fazil Yenidu¨nya2, S¸ebnem Harsa3 and Hatice Gu¨nes¸ 2,* 1 _ _ Biotechnology Program, Izmir Institute of Technology, Izmir TR-35430, Turkey 2 _ _ Department of Biology, Izmir Institute of Technology, Izmir TR-35430, Turkey 3 _ _ Department of Food Engineering Izmir Institute of Technology, Izmir TR-35430, Turkey *Author for correspondence: Tel.: +90-232-750-7544, Fax: +90-232-750-7509, E-mail: [email protected] Received 8 March 2004; accepted 17 July 2004

Keywords: Bacillus thuringiensis, cry genes, grain habitats, isolation, plasmid, PFGE profiles

Summary Bacillus thuringiensis (Bt) is a gram-positive, spore-forming bacterium and it produces insecticidal crystal (cry) proteins during sporulation. Because the genetic diversity and toxic potential of Bt strains differ from region to region, strains have been collected and characterized all over the world. The aim of this study is to isolate Bt strains in grain-related habitats in Turkey and to characterize them on the basis of crystal morphology, cry gene content, and chromosomal and plasmid DNA profiles. Four approaches were taken: analysis with phase contrast (PC) microscopy, polymerase chain reaction (PCR), pulsed field gel electrophoresis (PFGE) and plasmid isolation. Ninety-six samples were collected from Central Anatolia and the Aegean region. Bt was isolated from 61 of 96 samples (63.5%) and 500 Bt-like colonies were obtained. One hundred and sixty three of the colonies were identified as Bt based on cry protein formation using PC microscopy. Among the examined colonies, the overall proportion identified (as Bt index) was 0.33. We found that 103 isolates were positive for the five different cry genes (cry1, cry2, cry3, cry4 and cry9) examined with PCR. In addition, plasmid profiling of 37 cry gene-positive isolates indicated that the 15 kb plasmid band was present in all isolates; however, 11 of 37 isolates had more than one plasmid band at different sizes. Finally, chromosomal DNA profiling by PFGE gave rise to different DNA patterns for isolates containing the same cry gene which suggests a high level of diversity among the Bt strains isolated.

Introduction Bacillus thuringiensis (Bt) is a gram-positive, facultative anaerobe and spore-forming bacterium. It produces different insecticidal toxic proteins in parasporal crystals during the stationary phase of its growth cycle (Rowe et al. 1987). The genes coding for cry proteins are mostly carried on plasmids ranging from 3–4 to 150 Mda (Gonzales & Carlton 1980; Aronson 2002). Up to now, many cry protein genes have been cloned, sequenced and named cry genes. Over 100 cry gene sequences are organized into 32 groups and different subgroups based on nucleotide similarities and range of host specificity (Crickmore et al. 1998; Bravo et al. 1998). Insecticidal activity of Bt depends mostly on cry proteins and varies with insect type. Natural isolates of Bt have been used as biological pesticides against different insect orders such as Lepidoptera, Diptera, Coleoptera, Hymenoptera, Homoptera and Acari (Cannon 1993; Fieltson et al. 1992). In addition, some strains of Bt have been found to be toxic to nematodes

and protozoa (Feitelson et al. 1992; Edwards et al. 1988). The lack of mammalian toxicity of cry proteins has resulted in an increase in the use of Bt as an insecticide and intensified the search for new strains with different toxic activities. It has been reported that Bt can be present in many different habitats such as soil, stored product dust, insect cadavers, grains, agricultural soils, olive tree related-habitats, different plant and aquatic environments (Martin & Travers 1989; Meadows et al. 1992; Ben-Dov et al. 1997; Theunis et al. 1998; Bravo et al. 1998; Bel et al. 1997; Mizuki et al. 1999; Iriarte et al. 2000). Bt strains show genetic diversity with different toxic potential mostly due to plasmid exchange between strains (Thomas et al. 2001). In fact, each habitat may contain a novel Bt strain awaiting discovery which has a toxic effect on a target insect group. Therefore, Bt strains have been collected from different environments and characterized to evaluate their toxic potential against various insect orders (Chak et al. 1994; Theunis et al. 1998; Bravo et al. 1998; Forsyth & Logan 2000; Uribe et al. 2003). Different methods are used for the

O¨. Apaydin et al.

286 characterization of Bt isolates such as polymerase chain reaction (PCR), southern blotting, serotyping and bioassay; however, PCR is the most widely used, efficient and rapid technique for screening of a large number of isolates (Juarez-Perez et al. 1997; Porcar & Juarez-Perez 2002). Because the use of Bt products as an alternative to chemical insecticides is increasing rapidly, many research centres have focused on isolation of the native strains in order to establish Bt strain collection worldwide. Therefore, the purpose of this study is to initiate the establishment of a native Bt strain collection from different regions of Turkey and to determine its diversity. We isolated and characterized 103 Bt isolates from grain-related habitats of Central Anatolia and the Aegean regions of Turkey based on crystal formation,

cry gene content, and plasmid and chromosomal DNA profiles.

Materials and methods Sample collection Soil, grain, stored product dust, straw, insect cadavers and various residues were collected from grain silos, crop fields, farms, caves, haylofts where Bt preparation have not been applied, in central Anatolia (Ere
gli/ Konya, Takale/Karaman) and the Aegean region (Nikfer/Denizli, Bozbu¨k/So¨ke) as shown in Table 1. Samples were taken from the places not exposed to sunlight or at 5 cm depth from the surface and were placed into plastic

Table 1. Distribution of Bacillus thuringiensis based on sample types and locationa. Location

Type of sample

No. of sample

No. of sample yielding Bt

No. of isolates obtained

No. of isolates producing crystals

No. of isolates positive for cry genes

Bt Index

Ayranli/Ere
gli-Konya (CA) Eregli/Konya (CA)

Soil Grain Soil Stored product dust

9 3 3 2

7 1 3 1

70 4 19 6

42 1 13 3

33 0 8 3

0.60 0.25 0.68 0.50

Ivriz/Ere
gli-Konya (CA) U¨c¸harman/Ere
gli-Konya (CA)

Soil Soil Various residues

7 7 1

8

5 7 6 0

29 57 52 1

17 15 17 0

11

0.59b 0.26 0.33 0.00

7

14 7 0

Animal faeces Soil Stored product dust Various residues

1 9 5

0 5 2

0 37 22

0 8 4

0 3 1

0.32b 0.22 0.18

5

1

23

1

0

0.04

Natural Grain Silos (NGS) Tas¸ kale-Karaman (CA)

Animal faeces Grain Stored product dust

1 9 16

Bozbu¨k/So¨ke (AR)

Animal faeces Dead insect Grain Soil Straw

1 1

0 0

4 0

0 0

0 0

1 5 1

0 4 0

0 25 0

0 11 0

0 6 0

8 Manazan Caves (MC) Tas¸ kale-Karaman (CA)

6

20

8 1 2 13

26

Soil Stored product dust Straw

3 4 2

Total a

96

61

11 10 3

16 8 500

20

0.27b -

6

2 16

163

0.16b 0.33 0.18 0.28

3 3

3 62

4 1 1 18

32

29 26 20

2 9

13 2 2 28

118

4 3 3

17

82 6 11 101

16

9 Nikfer/Denizli (AR)

53

8 103

0.44 0.38b 0.38 0.15 0.19 0.26b 0.33

Isolates were examined with PC microscope for crystal formation and cry gene content of crystal positive isolates was screened by PCR. CA: Central Anatolia, AR: Aegean Region. Bt index is the ratio of Bt isolates producing crystal to all isolates in each sample group. b Indicates the total Bt index in each geographical location.

Bacillus thuringiensis from grain habitats bags aseptically. All samples were stored at 4 C until processed. Bacterial strains B. thuringiensis subsp. kurstaki (HD1), B. thuringiensis subsp. Aizawai (HD133), B. thuringiensis subsp. kumamotoensis (HD867), B. thuringiensis biovar. tenebrionis (tenebrionis), B. thuringiensis biovar. israelensis (HD500) were kindly supplied by Dr Daniel R. Zeigler from the Bacillus Genetic Stock Center (Ohio, USA) and used as reference strains. Bacillus thuringiensis isolation and crystal morphology analysis Bt strains were isolated based on the acetate selection method described by Travers et al. (1987). Briefly, 0.25 g of the environmental sample were suspended in 10 ml nutrient broth (Applichem) medium containing 0.25 M sodium acetate and left for microbial growth at 37 C overnight. Heat treatment was then applied for 5 min at 80 C to kill vegetative cells. After that, they were plated on nutrient agar plates and allowed to grow overnight at 37 C. Bt-like colonies which are usually described as cream-coloured and have the appearance of a fried egg on the plates (Travers et al. 1987) were labelled and subcultured. Subculturing from an individual colony was repeated three times to obtain a pure culture. Finally, each pure culture was grown on T3 agar plates and colonies dispersed in sterile distilled water were examined with a PC microscope for crystal production and morphology. Duplicate stock samples were prepared from each of the isolates in 25% glycerol and kept at )80 C. Cry gene identification Polymerase chain reaction (PCR) was used to identify cry gene content. All isolates producing crystal proteins were screened by five pairs of universal primers for the cry1, cry2, cry3, cry4 and cry9 genes described by Ben-Dov et al. (1997, 1999). DNA isolation was performed by the method of Bravo et al. (1998). Briefly, a loopful of cells from overnight Bt cultures was transferred into 0.2 ml of water and suspended. After freezing at )80 C for 20 min, they were transferred into boiling water for 10 min. The cell lysate was centrifuged (Henttich, Micro 12-24 Eppendorf Model) for 10 s at 11,000 · g and 15 ll of supernatant were used as DNA template. PCR reactions were carried out in 50 ll reaction volumes. DNA template was mixed with reaction buffer containing 200 lM deoxynucleotide triphosphate mix, 0.5 lM each primer (synthesized by Integrated DNA Technologies), 3 mM MgCl and 2 U of Taq DNA polymerase (Fermentas). Amplifications were carried out in a DNA thermal cycler (Techne Progen, England). For all cry genes, an initial denaturation step

287 was applied at 94 C for 1 min and followed by denaturation at 94 C for 1 min, annealing at 54 C (for cry1) and 60 C (for cry2, cry3, cry4 and cry9) for 1 min, then extension at 72 C for 1 min. Thirty-five cycles were carried out for the amplification of cry gene fragments. Finally, an extra extension step was applied at 72 C for 10 min. After amplifications, 10 ll of each PCR product were electrophoresed on 1% agarose– ethidium bromide (Sigma) gel in TAE buffer (0.04 M Tris-Acetate, 0.001 M EDTA [pH 8]) at 95 V for 40 min. Gels were visualized in a gel documentation system (Vilber Lourmat, France). Plasmid profiling Plasmid isolation was performed with minor modifications of the method described by O’Sullivan & Klaenhammer (1993). Bacterial cultures were grown on nutrient agar plates overnight and transferred into eppendorf tubes by scraping gently with the help of sterile distilled water. After pelleting the cells, they were resuspended in 200 ll of a solution containing 25% sucrose and 30 mg lysozyme/ml (Applichem) and incubated at 37 C for 15 min. The sample was mixed with 400 ll alkaline SDS solution (3% SDS, 0.2 N NaOH) and incubated for 7 min at room temperature. Then 300 ll ice-cold 3 M sodium acetate (pH 4.8) was added, mixed and centrifuged at 11,000 · g for 20 min (4 C). Supernatants were transferred into new eppendorf tubes, mixed with 650 ll of isopropanol (Sigma) and centrifuged again at 11,000 · g for 20 min (4 C). After discarding all liquid, pellets were resuspended in 320 ll sterile distilled water. They were mixed with 200 ll 7.5 M ammonium acetate containing 0.5 mg/ml ethidium bromide and 400 ll phenol/chloroform, then centrifuged at 11,000 · g for 10 min, at room temperature. Upper phases were transferred into new eppendorf tubes and mixed with 1 ml ethanol at )20 C. After centrifugation at 11,000 · g for 20 min (4 C), pellets were washed with 70% ethanol. All liquid was discarded and the pellets were dissolved in 25 ll TER (TE, pH 7.8 and RNase, 0.1 mg/ml). After incubation at 37 C for 20 min, plasmid samples were electrophoresed in 0.8% agarose–ethidium bromide gel at 80 V for 3 h and visualized with the gel documentation system. PFGE analysis PFGE analysis was carried out according to Rivera & Priest (2003) with some modifications. Bacterial strains were grown in 10 ml nutrient broth overnight and cells were harvested by centrifugation at 4 C for 2 min at 4500 rev/min. Cells were washed with 500 ll TE (50 mM Tris, 1 mM EDTA, pH 8.0) and SE (10 mM NaCl, 30 mM EDTA, pH 7.5) buffer, respectively. The cells were then resuspended in 50 ll SE buffer, mixed with 50 ll 2% agarose (Low Melt) at 50 C, and dispensed into the slots of plug mould. The plugs were

O¨. Apaydin et al.

288

Results Bacillus thuringiensis distribution shown by sample types and locations In total 96 samples, 78 from Central Anatolia region and 18 from the Aegean region were examined in this study (Table 1). Sample types consist of 43 soil, 27 stored product dust, 13 grain and 13 other samples including straw, animal faeces, various residues and an insect cadaver. After acetate selection, no microbial growth was observed in six grain samples and six other samples in different groups. According to colony morphology and PC microscopy analysis, Bt was isolated from 61 of the 96 samples which corresponds to 63.5% of the whole number of samples (Table 1). Five hundred isolates were obtained from these 61 samples. Bt index, reflecting the ratio of Bt colonies in total colonies isolated, was found to vary between 0.00 and 0.68 through origins with the average value of 0.33 (Table 1). Compared to all locations, Ere
gli/Konya was the richest area for Bt occurrence with an 0.59 Bt index. Crystal composition of the isolates

Figure 1. Photomicrography of a Bt isolate, 39Ya. The Bt isolate was grown for 48 h and examined with the PC microscope for spore formation and crystal protein production. Some cells were lysed and spores and crystals released into the medium whereas the others were intact. Arrow C and S indicate crystal protein and spore, respectively. Bar represents 2.5 lm.

Crystal morphology of Bt can give information about target insect spectra (Maeda et al. 2000). Therefore, in order to determine the crystal morphology of each Bt isolate, all isolates were grown for 48 h and examined with the PC microscope. Five different crystal shapes were observed in 163 isolates. Although only one type of crystal morphology was observed in some isolates, more than one type of crystal morphology was present in others (Figure 2). Distribution of crystal shapes in 163 isolates was 36% spherical (S), 5% cubic (C), 9% irregular pointed (IP), 2% bipyramidal (B), 19% cubic and spherical (C&S), 22% spherical and irregular pointed (S&IP), 1% cubic and irregular pointed (C&IP), 2% irregular shaped (IS), and 6% not defined (Figure 2).

40

Percent of Bt isolates

allowed to set at room temperature. The cells embedded into agarose were allowed to lyse in lysis buffer (30 mM Tris, 50 mM NaCl, 5 mM EDTA, pH 8.0) containing 2 mg/ml lysozyme at 37 C for 18 h. Bacterial plugs were then washed three times with 5 ml of buffer containing 20 mM Tris, 50 mM EDTA, pH 8.0. Proteins were digested with 2 ml proteinase K solution (0.5 mg proteinase K/ml and 0.1% N-lauroylsarcosine– EDTA, 50 mM, pH 8.0) at 50 C overnight. Then plugs were washed twice with 5 ml of buffer containing 20 mM Tris, 50 mM EDTA, 1 mM NaCl, pH 8.0; once with buffer containing Tris, 50 mM EDTA, 1 mM PMSF, pH 8.0, and once with buffer containing 20 mM Tris, 50 mM EDTA, pH 8.0. After the plugs were equilibrated with 1 ml restriction enzyme buffer, they were digested with 40 U of SmaI (Fermentas) at 30 C for 20 h. Then the plugs were electrophoresed in 1% agarose in TBE buffer in a CHEF-DRII system at 14 C for 40 h at 4 V/cm with pulse times of 15 s rising to 60 s. After staining the gel in ethidium bromide (1 ll/ml) for 45 min and destaining in distilled water for 1 h, DNA profiles were recorded in a gel documentation system (Vilber Lourmat, France).

35 30 25 20 15 10

ND

IS

C&IP

S&IP

B

IP

C

0 S

Five hundred isolates were examined with the PC microscope for spore formation and crystal production (Figure 1). Among them, 163 isolates produced crystals (Table 1). Even though 99 other isolates had Bt-like spore and colony morphology, they did not show crystal formation. The remaining 238 isolates did not exhibit any morphological similarities to Bt nor did they produce crystals.

C&S

5

crystal shape

Figure 2. Crystal shape distribution of Bt isolates. After growing the isolates for 48 h, crystal protein formation was observed using a PC microscope. Description of crystal shapes is: C: cubic, B: bipyramidal, S: spherical, IP: irregular pointed, IS: irregular shaped, ND: not defined.

Bacillus thuringiensis from grain habitats

289

Characterization of cry gene content of the isolates Because crystal proteins are encoded by cry genes and one Bt strain can contain more than one cry protein, the cry gene content of each isolate had to be determined. PCR reactions for each isolate were carried out with universal primers specific for cry1, cry2, cry3, cry4 and cry9 genes. One hundred and three of 163 isolates were positive for the cry genes examined. Some examples of PCR products amplified with different cry gene primers are shown in Figure 3. Even though most of the isolates gave only one DNA band with a specific cry gene primer, some of them showed two or three DNA bands with the same cry gene primer (Figure 3). PCR analysis of each isolate with five different cry gene primers indicated that 63 of the isolates had only one type of cry gene; however, 40 of them contained more than one type of cry gene (Figure 4). The number of isolates carrying one type of cry gene are 17 for cry1, 6 for cry2, 10 for cry3, 7 for cry4 and 21 for cry9. On the other hand, 28 isolates contained two different cry genes. In addition, 8 isolates were positive for 3 different

cry genes and 4 isolates for 4 different cry genes (Figure 4). No amplification of DNA template was observed for 60 isolates producing crystal protein indicating that they have cry genes different from the genes examined in this study. Plasmid and PFGE profiles of the isolates Bt has been known to have several circular/linear plasmids, and cry genes are generally found in these plasmids (Carson et al. 1996). Therefore, in the present study plasmids were isolated from 33 cry gene-positive isolates as well as four different Bt reference strains. A major plasmid band at 15 kb in size was obtained in all isolates (Figure 5). In addition, plasmid bands varying in length between 15 and 22 kb were observed in some of the isolates: cry2-positive (lanes 11, 12, 13, 14, 16), cry3-positive (lanes 18, 19), and cry9-positive (lanes 30, 31). Pulsed field gel electrophoresis (PFGE) of chromosomal DNA digested with a restriction enzyme is an

Figure 3. PCR analysis of crystal protein positive isolates. DNA template from each isolate was amplified with PCR in the presence of each cry gene primer. PCR products of some of the isolates are for cry1, lanes 1–5; cry2, lanes 6–14; cry3, lanes 15–19; cry4, lanes 20–24; cry9, lanes 25–36. Identity of isolates in each lane is LaneM: 1kb DNA MW marker; Lane1: Bacillus thuringiensis subsp. aizawai; Lane2: 39Yb; Lane3: 43Db; Lane4: 48Ra; Lane5: 71Na; Lane6: Bacillus thuringiensis subsp. kurstaki; Lane7: 18Fa; Lane8: 93Ha; Lane9: 93Da; Lane10: 93FFa; Lane11: 27Fb; Lane12: 19Rb; Lane13: 19Hb; Lane14: 85PPb; Lane15: Bacillus thuringiensis biovar. tenebrionis; Lane16: 21KB; Lane17: 71Lb; Lane18: 98Lb; Lane19: 86Db; Lane20: Bacillus thuringiensis biovar. israelensis; Lane21: 19Pb; Lane22: 2Ja; Lane23: 28Da; Lane24: 113Ya; Lane25: Bacillus thuringiensis subsp. aizawai; Lane26: 82YYb; Lane27: 36Ba; Lane28: 24Ca; Lane29: 25Ca; Lane30: 94YYb; Lane31: 24Nb; Lane32: 93Da; Lane33: 25Aa; Lane34: 29Fa; Lane35: 53Yb; Lane36: 62PPa.

25

number of isolates

20

15

10

5

cry1,cry3,cry4,cry9

cry3,cry4,cry9

cry2,cry4,cry9

cry2,cry3,cry9

cry1,cry4,cry9

cry4,cry9

cry1,cry2,cry9

cry3,cry9

cry1,cry2,cry4,cry9

cry genes

cry3,cry4

cry2,cry9

cry2,cry4

cry2,cry3

cry1,cry9

cry1,cry4

cry1,cry3

cry1,cry2

cry9

cry4

cry3

cry2

cry1

0

Figure 4. cry gene distribution of Bt isolates. Crystal protein-producing isolates were screened by PCR to find out their cry gene contents by using five primer pairs for cry1, cry2, cry3, cry4 and cry9. Figure shows number of isolates and their cry gene profiles.

290

O¨. Apaydin et al.

Figure 5. Plasmid patterns of Bt isolates. Plasmid DNA was prepared from 33 cry gene-positive isolates and subjected to electrophoresis in 0.8% agarose gel with ethidium bromide. Name of the isolate or reference strain in each lane is as follows: Lane M: 1 kb DNA Ladder; Lane1: Bacillus thuringiensis subsp. aizawai; Lane2: 48Ra; Lane3: 39Ya; Lane4: 39Yb; Lane5: 43Db; Lane6: 71Na; Lane7: 55Ka; Lane8: Bacillus thuringiensis subsp. kurstaki; Lane9: 18Fa; Lane10: 93FFa; Lane11: 93Ha; Lane12: 19Rb; Lane13: 27Fb; Lane14: 93Da; Lane15: 19Hb; Lane16: 85PPb; Lane17: Bacillus thuringiensis biovar. tenebrionis; Lane18: 71Lb; Lane19: 2Ja; Lane20: 98Lb; Lane21: 86Db; Lane22: Bacillus thuringiensis biovar. israelensis; Lane23: 19Pb; Lane24: 28Da; Lane25: 113Ya; Lane26: Bacillus thuringiensis subsp. aizawai; Lane27: 82Yb; Lane28: 24Nb; Lane29: 25Aa; Lane30: 25Ca; Lane31: 36Ba; Lane32: 29Fa; Lane33: 93Da; Lane34: 94YYb; Lane35: 53Yb; Lane36: 24Ca; Lane37: 62PPa.

Figure 6. PFGE profiles of cry 9-positive isolates. PFGE analysis was carried out as described in Materials and Methods section. Identity of the isolate in each lane is as follows: LaneM: 5 kb DNA ladder; Lane1: 62PPa; Lane2: 24Ca; Lane3: 28Aa; Lane4: 53Yb; Lane5: 94YYb; Lane6: 93Da; Lane7: 29Fa; Lane8: 25Aa; Lane9: 82YYb; Lane10: Bacillus thuringiensis subsp. aizawai.

accurate typing method for bacteria (Tenover et al. 1995). In order to see if Bt isolates carrying the same cry gene show similar PFGE patterns, chromosomal DNA from 6 cry1, 8 cry2, 5 cry3, 4 cry4 and 11 cry9 positive isolates was subjected to PFGE analysis. Even though there are some similarities among the PFGE patterns of the isolates (data not shown except for cry9), no identical patterns were obtained within each of the cry gene groups (Figure 6).

Discussion In this study, Bt occurrence was examined in grainrelated habitats of Central Anatolia and in the Aegean region where no Bt products have been applied before.

Bt occurence in all soil samples collected from Konya was found to be relatively high compared to other soil _ samples (Table 1). Especially in Ivriz, Bt was isolated from all of the samples. These places are crop fields and this suggests the abundant presence of Bt in agricultural lands. The percentage of samples yielding Bt from Nikfer was also high, 89%. This is because the sampled haylofts had been used for 65 years. In addition, natural grain silos (NGS) have been used for grain storage for more than 500 years and the percent of samples yielding Bt was 62%. In fact, Bt indexes of NGS and Nikfer are very similar with the values of 0.27 and 0.26, respectively. This shows a similar degree of occurrence of Bt in two places with similar background. The Bt index serves as a measurement of success in isolating Bt. After acetate selection for Bt isolation, no growth was observed in six of the grain samples. In all regions, the percentage of grain samples yielding Bt was relatively low in grain samples (23%) when compared with those of soil (81%), stored product dust (70%) and straw (67%) samples (Table 1). This indicates that grain is not as good a source as the others for Bt. An average Bt index was found to be 0.33 for all samples but the index changes according to sample types and origins (Table 1). The abundance of Bt was the highest in all soil samples, with a Bt index of 0.40. It decreases to 0.26 in all stored dust product samples and to 0.20 in all grain and animal faeces. Unlike our study, Bravo et al. (1998) collected soil samples from cultivated fields in Mexico and obtained a Bt index of about 0.24, nearly twofold lower than our index. However, Martin & Travers (1989) found the highest Bt index as 0.85 in the soil samples collected from Asia, nearly twofold greater than ours. This may be related to climate and geographic conditions. In addition, Hongyu et al. (2000) and Bernhard et al. (1997) reported that Bt is more abundant in stored product environments than in soil. Taken together, these studies show that the level of Bt index changes from region to region and between types of samples.

Bacillus thuringiensis from grain habitats Because there is a relationship between toxic activity and crystal shape of Bt strains (Maeda et al. 2000), observation of crystal morphology by PC microscope can provide valuable information about toxic activity of Bt isolates. In fact, observation of crystal morphology is the first step for establishing Bt strain collection (Ohba & Aizawa 1986; Bernhard et al. 1997). Therefore, when the crystal shape of the isolates was examined using a PC microscope, it was found that 42% of the isolates had more than one crystal shape; however, 58% of them had only one crystal shape (Figure 2). More definitive results about toxic activity of the isolates will be obtained from bioactivity assay of each isolate in the future study because a discrepancy exists between predicted cry gene type and its insecticidal activity (Shisa et al. 2002). For example, even though the cry1 gene product is toxic against Lepidoptera, Shisa et al. have reported that the cry1 gene product of native Bt strains was toxic to only Diptera. In addition, it was observed that a Bt isolate was sometimes positive for two or more cry genes even though it had only one type of crystal morphology. This may be due to lack of expression of all different cry genes at protein level. Moreover, 99 isolates exhibited spore and colony morphology similar to that of Bt whereas no crystal formation was observed by PC microscope. On the other hand, when PCR analysis was performed for nine of them, seven isolates were positive for cry genes examined. This is also related to the absence of gene expression at the protein level. In fact, crystal protein synthesis in Bt is controlled by a variety of mechanisms at the transcriptional, post-transcriptional or post-translational levels (Agassie & Lereclus, 1995). PCR screening of 163 crystal-forming isolates indicated that 103 of them were positive with primers for the five different cry genes examined. The number of isolates containing the cry9 gene was the greatest (21) compared to that of isolates containing the cry1 gene (17), cry2 gene (7) and cry3 gene (10) and cry4 gene (8). However, Bravo et al. (1998) have found cry1 genes the most frequent (49.5%), then cry3 gene as highly abundant (21.7%) and cry9 gene less abundant (2.6%). These results show how different geographic regions affect diversity of cry gene content of Bt strains. In addition, it is probable that the remaining 60 isolates negative for the observed cry genes may contain different cry genes from the ones examined in this study because 32 different cry gene groups and many subgroups have been defined in the literature (Schnepf et al. 1998; Crickmore et al. 1998). PFGE patterns of restriction enzyme-digested genomic DNA is known to be a useful technique to identify closely related bacterial isolates (Bygraves & Maiden 1992; Tenover et al. 1995). A recent study by Rivera & Priest (2003) has indicated that PFGE is a better technique than H-serotyping for discriminative typing of Bt strains. In the present study, we carried out PFGE analysis in order to see if isolates carrying the same cry

291 gene are identical. Although there were some similarities among PFGE patterns of the isolates (Figure 6), none of them were the same. Based on Rivera & Priest (2003), if PFGE patterns differed by changes up to 3 bands and more than three bands, strains are described as closely related and unrelated, respectively. Therefore, our isolates carrying the cry9 gene could be unrelated strains. In addition, results show that cry9-positive isolates may be heterogenous because Rivera & Priest (2003) have reported that serovars canadensis and entomocidus exhibited unique patterns and were described as heterogenous. Similar to their results, our findings also showed that there is no exact correlation between cry gene content and PFGE patterns. This is possible because cry genes are often carried on plasmids and plasmid exchange between strains as well as recombination between cry genes from different backgrounds occur in Bt strains (De Maagd et al. 2001). As a result, extensive genetic characterization and PFGE patterns will give more definite results about diversity of Bt strains with different cry genes. In conclusion, this study is the first for isolation and characterization of Bt native strains in Turkey. Different PFGE patterns of isolates carrying the same cry gene indicates wide range of biodiversity among Bt strains in Anatolia. Planned further studies related to Bt isolation from different parts of Anatolia and detailed genetic characterization as well as toxic activity will give more comprehensive results about biodiversity of Bt strains.

Acknowledgements This work was partially supported by grants from IYTE (2002 IYTE43) and from DPT (2002K-1207390).

References Aronson, A. 2002 Sporulation and d-endotoxin synthesis by Bacillus thuringiensis. Cellular and Molecular Life Sciences 59, 417–425. Agaisse, H. & Lereclus, D. 1995 How does Bacillus thuringiensis produce so much insecticidal crystal protein? Journal of Bacteriology 177, 6027–6032. Bel, Y., Granero, F., Alberola, T.M., Martinez-Sebastian, M.J. & Ferre, J. 1997 Distribution, frequency and diversity of Bacillus thuringiensis in olive tree environments in Spain. Systematic Applied Microbiology 20, 652–658. Ben-Dov, E., Zaritsky, A., Dahan, E., Barak, Z., Sinal, R., Manasherob, R., Khamraev, A., Troitskaya, E., Dubitsky, A., Berezina, N. & Margalith, Y. 1997 Extended screening by TAB;PCR for seven cry-group genes from field-collected strains of Bacillus thuringiensis. Applied and Environmental Microbiology 63, 4883–4890. Ben-Dov, E., Wang, Q., Zaritsky, A., Manasherob, R., Barak, Z., Schneider, B., Khamraev, A., Baizhanov, M., Glupov, V. & Margalith, Y. 1999 Multiplex PCR screening to detect cry9 genes in Bacillus thuringiensis strains. Applied and Environmental Microbiology 65, 3714–3716. Bernhard, K., Jarrett, P., Meadows, M., Butt, J., Ellis, J., Roberts, G.M., Pauli, S., Rodgers, P. & Burges, H.D. 1997 Natural isolates of Bacillus thuringiensis: worldwide distribution, characterization and activity against insect pests. Journal of Invertebrate Pathology 70, 59–68.

292 Bravo, A., Sarabia, S., Lopez, L., Ontiveros, H., Abarca, C., Ortiz, A., Ortiz, M., Lina, L., Villalobos, V., Pena, G., Nunez-Valdez, M., Soberon, M. & Quintero, R. 1998 Characterization of cry genes in a Mexican Bacillus thuringiensis strain collection. Applied and Environmental Microbiology 64, 4965–4972. Bygraves, J. & Maiden, M.C. 1992 Analysis of the clonal relationships between strains of Neisseria meningitidis by pulsed field gel electrophoresis. Journal of Genomic Microbiology 138, 523–531. Carson, C.R., Johenson, T., Lecadet, M.-M. & Kolstø, A.-B. 1996 Genomic organization of the entomopathogenic bacterium Bacillus thuringiensis subsp. berliner 1715. Microbiology 142, 1625–1634. Cannon, R.J.C. 1993 Prospects and progress for Bacillus thuringiensisbased pesticides. Pesticide Science 37, 331–335. Chak, K.F., Chao, D.C., Tseng, M.Y., Kao, S.S., Tuan, S.J. & Feng, T.Y. 1994 Determination and distribution of cry-type genes of Bacillus thuringiensis isolates from Taiwan. Applied and Environmental Microbiology 60, 2415–2420. Crickmore, N., Zeigler, D.R., Feitelson, J., Schnepf, E., Van-Rie, J., Lereclus, D., Baum, J. & Dean, D.H. 1998 Revision of nomenclature for the Bacillus thuringiensis pesticidal crystal proteins. Microbiology and Molecular Biology Review 62, 807–813. De Maagd, R.A., Bravo, A. & Crickmore, N. 2001 How Bacillus thuringiensis has evolved specific toxins to colonize the insect world. Trends in Genetics 17, 193–199. Edwards, D.L., Payne, J. & Soares, G.G. 1988 Novel isolates of Bacillus thuringiensis having activity against nematodes. European Patent Application, EP 0 303 426 A2. Forsyth, G. & Logan, N.A. 2000 Isolation of Bacillus thuringiensis from Northern Victoria Land, Antarctica. Letters in Applied Microbiology 30, 263–266. Feitelson, J.S., Payne, J. & Kim, L. 1992 Bacillus thuringiensis: insects and beyond. Bio/Technology 10, 271–275. Gonzales, J.M. & Carlton, B.C. 1980 Patterns of plasmid DNA in crystalliferous strains of B. thuringiensis. Plasmid 3, 92–98. Hongyu, Z., Ziniu, Y. & Wangxi, D. 2000 Isolation, distribution and toxicity of Bacillus thuringiensis from warehouses in China. Crop Protection 19, 449–454. Iriarte, J., Porcar, M., Lecadet, M.M. & Caballero, P. 2000 Isolation and characterization of Bacillus thuringiensis strains from aquatic environments in Spain. Current Microbiology 40, 402–408. Juarez-Perez, V.M., Ferrandis, M.D. & Frutos R. 1997. PCR-based approach for detection of novel Bacillus thuringiensis cry genes. Applied and Environmental Microbiology 63, 2997–3002. Maeda, M., Mizuki, E., Nakamura, Y., Hatano, T. & Ohba, M. 2000 Recovery of Bacillus thuringiensis from marine sediments of Japan. Current Microbiology 40, 413–422. Martin, P.A.W. & Travers, R.S. 1989 Worldwide abundance and distribution of Bacillus thuringiensis isolates. Applied and Environmental Microbiology 55, 2437–2442. Meadows, M.P., Ellis, D.J., Butt, J., Jarrett, P. & Burges, D. 1992 Distribution, frequency and diversity of Bacillus thuringiensis in an

O¨. Apaydin et al. animal feed mill. Applied and Environmental Microbiology 58, 1344–1350. Mizuki, E., Ichimatsu, T., Hwang, S.H., Park, Y.S., Saitoh, H., Higuchi, K. & Ohba, M. 1999 Ubiquity of Bacillus thuringiensis on phylloplanes of arboreous and herbaceous plants in Japan. Journal of Applied Microbiology 86, 979–984. Ohba, M., & Aizawa, K. 1986 Distribution of Bacillus thuringiensis in soils of Japan. Journal of Invertebrate Pathology 47, 277–282. O’Sullivan, D.J. & Klaenhammer, T.R. 1993 Rapid mini-prep. isolation of high-quality plasmid DNA from Lactococcus and Lactobacillus spp. Applied and Environmental Microbiology 59, 2730–2733. Porcar, M. & Juarez-Perez, V. 2002 PCR-based identification of Bacillus thuringiensis pesticidal crystal genes. FEMS Microbiology Reviews 757, 1–4. Rivera, A.M.G. & Priest, F.G. 2003 Pulsed field gel electrophoresis of chromosomal DNA reveals a clonal population structure to Bacillus thuringiensis that relates in general to crystal protein gene content. FEMS Microbiology Letters 223, 61–66. Rowe, G.E., Margaritis, A. & Dulmage H.T. 1987 Bioprocess developments in the production of bioinsecticides by Bacillus thuringiensis. Critical Reviews in Biotechnology 6, 87–127. Schnepf, E., Crickmore, N., Van-Rie, J., Lereclus, D., Baum, J., Feitelson, J., Zeigler, D.R. & Dean, D.H. 1998 Bacillus thuringiensis and its insecticidal proteins. Microbiology and Molecular Biology Reviews 62, 774–806. Shisa, N., Wasano, N. & Ohba, M. 2002 Discrepancy between cry gene-predicted and bioassay-determined insecticidal activities in Bacillus thuringiensis natural isolates. Journal of Invertebrate Pathology 81, 59–61. Tenover, F.C., Arbeit, R.D., Goering, R.V., Mickelson, P.A., Murray, B.E., Persing, D.H. & Swaminathan, B.A. 1995 Interpreting chromosomal DNA restriction patterns produced by pulse-field gel electrophoresis: criteria for bacterial typing. Journal of Clinical Microbiology 33, 2233–2239. Theunis, W., Aguda, R.M., Cruz, W.T., Decock, C., Peferoen, M., Lambert, B., Bottrell, D.G., Gould, F.L., Litsinger, J.A. & Cohen, M.B. 1998 Bacillus thuringiensis isolates from the Philippines: habitat distribution, d-endotoxin diversity and toxicity to rice stem borers (Lepidoptera: Pyralidae). Bulletin of Entomological Research 88, 335–342. Thomas, D.J.I., Morgan, A.W., Whipps, J.M. & Saunders, J.R. 2001 Plasmid transfer between Bacillus thuringiensis subsp. israelensis strains in laboratory culture, river water and dipteran larvae. Applied and Environmental Microbiology 67, 330–338. Travers, R.S., Martin, P.A.W. & Reichelderfer, C.F. 1987. Selective process for efficient isolation of soil Bacillus spp. Applied and Environmental Microbiology 53, 1263–1266. Uribe, D., Martinez, W. & Ceron, J. 2003 Distribution and diversity of cry genes in native strains of Bacillus thuringiensis obtained from different ecosystems from Colombia. Journal of Invertebrate Pathology 82, 119–127.

 Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 293–296 DOI 10.1007/s11274-004-3634-x

Preliminary studies on chorote – a traditional Mexican fermented product Marisol Castillo-Morales1, Marı´ a del Carmen Wacher-Rodarte2 and Humberto Herna´ndez-Sa´nchez1,* 1 Departamento de Graduados e Investigacio´n en Alimentos, Escuela Nacional de Ciencias Biolo´gicas, Instituto Polite´cnico Nacional, Carpio y Plan de Ayala, Me´xico, DF CP 11340, Me´xico 2 Facultad de Quı´mica, Universidad Nacional Auto´noma de Me´xico, Me´xico, DF, Me´xico *Author for correspondence: Tel.: +525729-6000 ext. 62458, Fax: +52-5729-6000 ext. 62359, E-mail: [email protected] Received 19 March 2004; accepted 16 July 2004

Keywords: Cacao, chorote, fermentation, maize, pozol

Summary The purpose of this work was to study some biochemical and microbiological changes occurring during the fermentation of chorote, an important indigenous fermented food in the Mexican state of Tabasco. The chorote solid state fermentation is similar to that for the production of pozol, though fermented and roasted cacao beans are added to the maize dough. Changes in pH, lactic acid, total and soluble protein, available lysine and tryptophan as well as in some microbial groups were recorded during a 9-day fermentation period. Microbiological analysis showed an increase in moulds, yeasts, amylolytic lactic acid and nitrogen-fixing bacteria during the fermentation. There was a decrease in pH (to 4.2) and a significant net increase in protein content. There was also an increase in soluble protein and tryptophan.

Introduction World-wide, there are many indigenous and traditional fermented foods which have not been studied and that are very interesting from the technological and cultural point of view. This is the case for chorote, an important fermented food elaborated from maize and cacao in the Mexican states of Tabasco, Chiapas and Oaxaca (JavierQuero 2000). Chorote is a homogeneous fermented dough made out of lime-treated maize and fermented cocoa beans. This dough is moulded into balls with a diameter from 5 to 10 cm, fermented and later suspended in water to be consumed as a refreshing drink. There are no previous studies on chorote fermentation, even though in some places it is as popular as pozol which has been thoroughly studied. Pozol is a refreshing beverage prepared with fermented Nixtamal, i.e. maize cooked in lime water, which is drunk raw as a staple food by the Indian and mestizo populations in several southeastern states of Mexico like Tabasco, Campeche, Yucata´n, Quintana Roo, Chiapas and Oaxaca (Can˜as et al. 1993). In spite of the great variation in the microflora of pozol due to the uncontrolled manner of preparation, there are several species of bacteria, yeasts and moulds which are always present in pozol from different places and prepared at different times (Ulloa 1974). Usually, lactic acid bacteria account for 90–97% of the microflora. The predominant genera are Lactobacillus, Streptococcus, Leuconostoc and Weisella (Ampe

et al. 1999). The presence of Bifidobacterium, Enterococcus, and enterobacteria has also been reported, and may indicate a human origin of some pozol microorganisms (ben Omar & Ampe 2000). The importance of the phenomenon of nitrogen fixation in pozol production has been pointed out (Quintero-Ramı´ rez et al. 1999). Two of the main nitrogen-fixing bacteria isolated from pozol are Agrobacterium azotophilum and Enterobacter aerogenes (Salinas & Herrera 1974; Herrera & Ulloa 1975). A. azotophilum is reported to be antagonistic to several pathogenic moulds and bacteria (Herrera & Ulloa 1975). Some amino acids and oxygen may inhibit nitrogen fixation by this microorganism (Taboada & Herrera 1972). Alcaligenes pozolis and Klebsiella pneumoniae have also been reported as nitrogen-fixing organisms present in pozol (Quintero-Ramı´ rez et al. 1999). The purpose of this work was then to study some biochemical and microbiological changes occurring during the chorote fermentation.

Materials and methods Manufacture of chorote The chorote solid state fermentation is similar to that of pozol though 6% fermented, roasted and ground cacao beans are added to the maize dough. A fermentation time of 9 days at room temperature (28
C) was used

294 and 10% of a previous chorote fermentation brought from Villahermosa, Tabasco, Me´xico was used as starter culture for the process. The dough was formed into 150g greyish balls, and changes in pH, titratable acidity (as lactic acid), total and soluble protein, zein, ash, starch, nitrogen fixation, lysine and tryptophan as well as changes in some microbial groups i.e. aerobic mesophilic bacteria, coliforms, amylolytic lactic acid bacteria, moulds and yeasts, and nitrogen-fixing bacteria were recorded during the fermentation, according to the methods described below. Five fermentation trials were included in this study. Enumeration and isolation of microorganisms Ten grams of sample from the core of chorote balls were mixed with 100 ml portions of sterile 0.9% sodium chloride solution in a Waring blendor and then serial dilutions were prepared in the same diluent and used for microbial enumeration with the following media: plate count agar (Bioxon – Becton Dickinson, Cuatitla´n, Mexico) to estimate the number of total aerobic mesophilic bacteria (30
C for 72 h); violet red bile agar (Bioxon – Becton Dickinson, Cuatitla´n Mexico) for coliform organisms (35
C for 24 h); MRS-starch medium containing 2% soluble starch instead of glucose for amylolytic lactic acid bacteria. The medium was prepared from the ingredients indicated in the original formula described by De Man et al. (1960), the inoculated plates were incubated under anaerobic conditions at 30
C for 72 h. After incubation, the plates were flooded with iodine solution and the colonies with a colourless area around the growth were selected as amylolytic lactic acid bacteria. The strains so recovered were examined microscopically after Gram staining, since lactic acid bacteria are Grampositive. Potato dextrose agar (Bioxon de Me´xico, Cuautitla´n, Mexico) acidified with tartaric acid to pH 3 was used for cultivating yeasts and moulds (25
C for 120 h); nitrogen-free 77 medium containing 2.5 g glucose/l for nitrogen-fixing bacteria (the medium was prepared from the ingredients according to the formula described by Ulloa et al. (1971) and the inoculated plates were incubated under anaerobic conditions at 28
C for 120 h). Thelco incubators (model 31480, Precision Scientific, Chicago, USA) were used for incubating the plates and anaerobic jars with GasPak (BBL, Cuautitla´n, Mexico) systems were used when necessary. In all cases, the results presented later are the means of five determinations. Chemical analyses Moisture (method 14.002), fat (method 14.018) and protein content (method 2.049, microKjeldahl with a nitrogen to protein conversion factor of 6.25), pH (method 14.022) and titratable acidity (expressed as lactic acid, method 22.060) were determined by the standard AOAC methods (Horwitz et al. 1975). To measure soluble protein, a 4-g sample was suspended in 50 ml of distilled water and stirred for 30 min.

M. Castillo-Morales et al. The slurry was filtered through Whatman 2 filter paper and soluble protein was determined by the Lowry method in 1 ml of the filtrate (Loaeza-Cha´vez & Wacher-Rodarte 1993). Ash and starch contents were determined by the AOAC methods 14.006 and 14.069, respectively (Horwitz et al. 1975). Available lysine was determined by the method of Hurrel & Carpenter (1976) using the dye Acid Orange 12 which selectively binds to the e-amino group of the lysine in the protein forming a precipitate. Dye-binding by lysine is first determined at 482 nm, after which, in a new sample, the e-amino group of the amino acid is blocked by treatment with propionic anhydride and absorbance at 482 nm is read again. Lysine content was then calculated by difference. Tryptophan content was determined by the method of Spies & Chambers (1950) in which the colour formed by oxidation of a complex formed with tryptophan and p-dimethylaminobenzaldehyde with nitrite ions is measured at 590 nm. The rationale to assay for lysine and tryptophan is the fact that maize protein is limiting in both essential amino acids. The zein content was determined by the turbidity formed by the addition of a 10 g/kg sodium chloride solution to a 700 g ethanol/l extract of the sample (Paulis et al. 1974). In order to differentiate the increase in nitrogen content due to fixation from that due to concentration by organic matter consumption, the variation of the ratio of nitrogen to ash concentrations was evaluated. This ratio can be used as an index of nitrogen fixation since the ash content remains constant during the fermentation (Aguilera 1989). All the chemicals were of analytical grade and were purchased from Sigma-Aldrich Fine Chemicals (St. Louis, USA) and in all cases, the results presented later are the averages of three determinations.

Results and discussion Microbiological examination Figure 1 shows the time course of the changes in the different microbial groups in the chorote from day zero (the dough immediately after inoculation) through the day nine of fermentation. These analyses showed an increase from the number at day zero to higher final numbers of aerobic mesophilic bacteria (1.5 · 109 c.f.u./g), moulds (2.1 · 108 c.f.u./g), yeasts (2 · 107 c.f.u./g), amylolytic lactic acid (4.5 · 108 c.f.u./g) and nitrogenfixing bacteria (4.4 · 104 c.f.u./g) after the 9 days of fermentation and especially during the third day. Coliforms were only detected in very low numbers (100 c.f.u./g) in the inoculated dough and disappeared completely after 6 days of fermentation. The amylolytic lactic acid bacteria were the predominant group in the product after three days of fermentation, possibly

295

Chorote maize fermentation 12 Amylolytic lactic acid bacteria

log (c.f.u./g) , pH

10

Aerobic mesophillic bacteria

8

Table 1. Changes in chemical composition during chorote fermentation1. Component

Day 0

Day 9

Ash (g/kg) Total Protein (g/kg) Soluble Protein (g/kg) Zein (g/kg) Fat (g/kg) Starch (g/kg) Moisture (g/kg)

30.7 103.0 6.1 13.9 34.0 448.1 610.0

40.8 152.72 31.9 15.1 54.6 295.0 527.0

Coliform organisms

6

Moulds

4

Yeasts

2

Nitrogen fixing bacteria pH

0 0

2

4 6 Time (days)

8

10

1

Figure 1. Growth of the different microbial groups and kinetics of pH decrease during chorote fermentation at 28
C from lime-treated maize with 6% roasted and ground cacao beans.

creating good conditions for the growth of the nitrogenfixing organisms as described by Mun˜oz & Viniegra (1981). Nitrogen-fixing microorganisms increased their population 200-fold after 3 days reaching, however, very low numbers. The lactic acid fermentation is very important in controlling some pathogenic and deteriorative microorganisms. Pozol and chorote are classified as lactic acid-fermented cereals (Steinkraus 2002). The results obtained are similar to pozol, where lactic acid bacteria (Lactobacillus brevis, Lactobacillus amylophilus, Lactobacillus amylovorus, and Lactobacillus fermentum) are dominant in the first stage of the fermentation of the maize dough (Steinkraus 1995). The counts of these microorganisms were close to those reported for pozol (ben Omar & Ampe 2000). After day 3, the nitrogen-fixing bacterial count remained constant, whereas moulds and yeasts started to increase in number, reaching a maximum at day 9. The counts of moulds are higher than those reported for pozol (Steinkraus 1995; ben Omar & Ampe 2000). It is possible that the differences in the raw materials may be responsible for this observation. The counts of yeasts are very similar to those reported for pozol (ben Omar & Ampe 2000). Yeasts of various kinds have been isolated from pozol (Ulloa 1974), including strains of Candida parapsilosis and Trichosporon cutaneum which have been reported to be amylolytic (Linardi & Machado 1990) and therefore potentially important in the fermentation of starchy substrates like pozol and chorote. In the case of the nitrogen-fixing bacteria, the maximal number was reached after 72 h (6.0 · 104 c.f.u./g), a quantity close to that reported for pozol (6.1 · 104 c.f.u./g) after 144 h of fermentation (Aguilera 1989). It may be concluded that the essential difference between chorote and pozol – the addition of cacao beans to chorote – has no effect on the microbiology of the fermentation.

All the components (except moisture) are in dry basis and the results are the average of five determinations. 2 The predicted concentration based only on the decrease in organic matter would be 136.4 g/kg, so the actual increase was possibly due to nitrogen fixation.

acid production by the amylolytic lactic acid bacteria, a similar phenomenon to that observed in pozol. It can also be observed that even though the amount of nitrogen-fixing organisms was relatively low, a significant net increase in protein content (detected by a oneway analysis of variance, p £ 0.05) due to nitrogen fixation and not attributable to a decrease in organic matter (sugar consumption) was detected. This increase (50%) was higher than the one reported by LoaezaChavez & Wacher-Rodarte (1993) for pozol, even though the amount of nitrogen-fixing bacteria in chorote is similar. In this study, the identification of these bacteria was not performed. However, Quintero-Ramı´ rez et al. (1999) reported the presence of the nitrogenfixing bacteria Agrobacterium azotophilum, Alcaligenes pozolis, and Klebsiella pneumoniae in pozol, so it is possible that these microorganisms are also present in chorote. It has been previously reported that lactic acid promotes the growth of nitrogen-fixing bacteria (Mun˜oz & Viniegra 1981). There was also an increase in soluble protein very similar to that reported by Loaeza-Chavez & Wacher-Rodarte (1993) for pozol. The increase in soluble protein is assumed to be explained by the presence of proteolytic organisms capable of hydrolysing zein and other proteins present in chorote. Starch was also consumed partially during the fermentation by amylolytic fungi and lactic acid bacteria. Available lysine remained constant during the chorote fermentation (21 g/kg of protein). In the case of tryptophan, there was a significant increase (p £ 0.05) from 5 to 9.8 g/kg of protein due to the fermentation process, i.e. this amino acid was synthesised de novo by the microflora. This is important since maize is deficient in this essential amino acid. In the case of pozol, Ramı´ rez (1987) found an increase in both amino acids during the fermentation.

Chemical analyses The results of the chemical analyses are given in Table 1. There was a dehydration of the product during the fermentation probably due to the evolving of metabolic heat and exposure of the surface of the balls to the air. There was also a decrease in pH (to 4.2) due to lactic

Conclusions This preliminary study indicates that the chorote fermentation is similar, from the chemical and microbiological point of view to the pozol fermentation.

296 However, a higher increase (50%) in the net protein content and the fact that the available lysine remained constant give the chorote some unique characteristics that make it different from other fermented maize products. It is also possible to say that this important increase in net protein due to the presence of nitrogenfixing bacteria, offers the possibility of obtaining products with a high nutritive value (at least in terms of protein quantity) by means of maize fermentations like chorote and pozol. This is especially important since these products are consumed mainly by people on low wages.

References Aguilera, F.D. 1989 Estudio Microbiolo´gico y Bioquı´ mico de la Fermentacio´n del Pozol. M.Sc. thesis in Food Science, Universidad Iberoamericana, Me´xico, pp. 84–86 Ampe, F., ben Omar, N., Moizan, C., Wacher, C. & Guyot, J.P. 1999 Polyphasic study of the spatial distribution of microorganisms in Mexican pozol, a fermented maize dough, demonstrates the need for cultivation-independent methods to investigate traditional fermentations. Applied and Environmental Microbiology 65, 5464–5473. ben Omar, N. & Ampe, F. 2000 Microbial community dynamics during production of the Mexican fermented maize dough pozol. Applied and Environmental Microbiology 66, 3664–3673. Can˜as, A.O., Ba´rzana, E., Owens, J.D. & Wacher, M.C. 1993 Study of the variability in the methods of pozol production in the highlands of Chiapas. In Alimentos Fermentados Indı´genas de Me´xico, eds. Wacher, M.C. & Lappe, P. pp. 69–74. Me´xico: Universidad Nacional Auto´noma de Me´xico. ISBN 968-36-3248-3. De Man, J.C., Rogosa, M. & Sharpe, M.E. 1960 A medium for the cultivation of Lactobacilli. Journal of Applied Bacteriology 23, 653–658. Herrera, T. & Ulloa, M. 1975 Antagonismo del pozol y de Agrobacterium azotophilum sobre diversas especies de bacterias y hongos, algunas pato´genas del hombre. Revista Latinoamericana de Microbiologı´a 17, 143–147. Horwitz, W., Senzel, A., Reynolds, H. & Park, D.L. 1975 Official Methods of Analysis of the Association of Official Analytical Chemists. 12th edn. pp. 15–16, 222, 225, 233, 401. Washington: AOAC.

M. Castillo-Morales et al. Hurrel, R.F. & Carpenter, K.J. 1976 An approach to the rapid measurement of ‘‘reactive lysine’’ in foods by dye binding. Proceedings of the Nutrition Society 35, 23A–24A. Javier-Quero, J.C. 2000 Bebidas y Dulces Tradicionales de Tabasco. pp. 34–36. Me´xico: CONACULTA. ISBN 970-18-5124-2. Linardi, V.R. & Machado, K.M.G. 1990 Production of amylases by yeasts. Canadian Journal of Microbiology 36, 751–753. Loaeza-Cha´vez, N.A. & Wacher-Rodarte, M.C. 1993 The fermentation of pozol: the effect of thermal and alkaline treatment of maize. In Alimentos Fermentados Indı´genas de Me´xico, eds. Wacher, M.C. & Lappe, P. pp. 103–108. Me´xico: Universidad Nacional Auto´noma de Me´xico. ISBN 968-36-3248-3. Mun˜oz, G. & Viniegra, G. 1981 Fijacio´n de nitro´geno atmosfe´rico por un cultivo mixto de una bacteria la´ctica y Azotobacter chrococcum. Revista Latinoamericana de Microbiologı´a 23, 213–217. Paulis, J.W., Wall, J.S., Kwolek, W.F. & Donaldson, G.L. 1974 Selection of high lysine corns with varied kernel characteristics and compositions by a rapid turbidimetric assay for zein. Journal of Agricultural and Food Chemistry 22, 318–323. Quintero-Ramı´ rez, R., Lorence-Quin˜ones, A. & Wacher-Rodarte, C. 1999 Cereal fermentations in Latin American countries. In Fermented Cereals. A Global Perspective, eds. Haard, N.F., Odunfa, S.A., Lee, C.H., Quintero-Ramı´ rez, R., Lorence-Quin˜ones, A. & Wacher-Rodarte, C. pp. 99–144. Rome: FAO. ISBN 92-5-104296-9. Ramı´ rez, J.F. 1987 Biochemical studies on a Mexican fermented corn food: pozol. PhD thesis, Cornell University, Ithaca, USA. Salinas, C. & Herrera, T. 1974 Aislamiento de Aerobacter aerogenes del pozol del Estado de Campeche. Revista Latinoamericana de Microbiologı´a 16, 95–98. Spies, J.R. & Chambers, D.C. 1950 Determination of tryptophan with p-dimethyl amino benzaldehyde using photochemical development of color. Analytical Chemistry 22, 1209–1210. Steinkraus, K.H. 1995 Handbook of Indigenous Fermented Foods. 2nd edn. pp. 226–233, 252–259. New York: Marcel Dekker Inc. ISBN 08-2-479352-8. Steinkraus, K.H. 2002 Fermentations in world food processing. Comprehensive Reviews in Food Science and Food Safety 1, 1–9. Taboada, J. & Herrera, T. 1972 Efecto de aminoa´cidos sobre la fijacio´n de nitro´geno por Agrobacterium azotophilum. Anales del Instituto de Biologı´a de la Universidad Nacional Auto´noma de Me´xico 43, 35–42. Ulloa, M. 1974 Mycofloral succesion in pozol from Tabasco, Mexico. Boletı´n de la Sociedad Mexicana de Micologı´a 8, 17–48. Ulloa, M., Herrera, T. & De La Lanza, G. 1971 Fijacio´n de nitro´geno atmosfe´rico por microorganismos del pozol. Revista Latinoamericana de Microbiologı´a 13, 113–124.


Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 297–301 DOI 10.1007/s11274-004-3693-z

Biodegradation of pentachorophenol by tropical basidiomycetes in soils contaminated with industrial residues Ka´tia Maria Gomes Machado1,*, Da´cio Roberto Matheus2, Regina Teresa Rosim Monteiro3 and Vera Lu´cia Ramos Bononi2 1 Universidade Cato´lica de Santos, Santos, and Fundac¸a˜o Centro Tecnolo´gico de Minas Gerais, Belo Horizonte, Brazil 2 Instituto de Botaˆnica, Secretaria do Meio Ambiente do Estado de Sa˜o Paulo, Brazil 3 Centro de Energia Nuclear na Agricultura, Universidade de Sa˜o Paulo, Sa˜o Paulo, Brazil *Author for correspondence: UniSantos, Av. Conselheiro Ne´bias 300, Santos, Sa˜o Paulo, CEP11015-002, Brazil. Tel.: +55-13-32055555, Fax: +55-13-32844146, E-mail: [email protected]

Keywords: Non-lignicolous basidiomycetes, organochlorine compounds, pentachloroanisole, pentachlorophenolmineralization, soil bioremediation, soil pollution, tropical fungi

Summary The ability of tropical Brazilian basidiomycetes to degrade pentachlorophenol (PCP) in soils from areas contaminated with organochlorine industrial residues was studied. Thirty-six basidiomycetes isolated from different tropical ecosystems were tested for tolerance to high PCP concentrations in soil. Peniophora cinerea and Psilocybe castanella, two strains of Trametes villosa, Agrocybe perfecta, Trichaptum bisogenum and Lentinus villosus were able to colonize soil columns containing up to 4600 mg pentachlorophenol kg)1 soil. The first four species were inoculated into soil containing 1278 mg pentachlorophenol kg)1 soil supplemented with gypsum and sugar cane bagasse. P. cinerea, P. castanella, T. villosa CCB176 and CCB213 and Agrocybe perfecta reduced the PCP present in the contaminated soil by 78, 64, 58, 36 and 43%, respectively, after 90 days of incubation. All fungi mineralized [14C] pentachlorophenol, mainly P. cinerea and T. villosa with the production of 7.11 and 8.15% 14CO2, respectively, during 120 days of incubation. All fungi produced chloride ions during growth on soil containing PCP, indicating dehalogenation of the molecule. Conversion of PCP to pentachloroanisole was observed only after 90 days of incubation in soils inoculated with A. perfecta, P. cinerea and one of T. villosa strain. The present study shows the potential of Brazilian fungi for the biodegradation of toxic and persistent pollutants and it is the first to report fungal growth and PCP depletion in soils with high pentachlorophenol concentrations.

Introduction Pentachlorophenol (PCP), synthesized for the first time in 1872, has been used since the late 30s as a wood preservative together with its salt, sodium pentachlorophenate, due to its broad spectrum and low cost. Its esters have also been used as biocides. This generalized use has led to the contamination of many ecosystems, with PCP currently considered to be a product of priority for decontamination studies according to the European Community and the American Environmental Protection Agency. In Brazil, PCP has been widely used as a herbicide to clean pasture fields and in deforestation, for the construction of hydroelectric plants and dams for water reservoirs. In the State of Sa˜o Paulo, the most industrialized region of Brazil, there is widespread concern over locating, monitoring and treating soil areas that were contaminated with organic

pollutants, particularly in the sixties and seventies, among them residuals containing organochlorine compounds such as hexachlorobenzene and PCP (Matheus et al. 2000). The capacity of basidiomycetes to degrade PCP, as well the possibility of using these organisms in processes of reclamation of PCP-contaminated soils, have been the subject of much previous research (Lamar & Dietrich 1992; Liang & McFarland 1994; Reddy & Gold 2000; Leontievsky et al. 2002). An enormous diversity of basidiomycetes exists, with 29,914 species known worldwide (Kirk et al. 2001). Thus it is clear that a selection must be made of species with greater potential to degrade xenobiotics, and that at the same time are able to grow in contaminated environments or in materials that must be treated. Moreover, in many countries, including Brazil, there are restrictions about the use of certain allochthonous microorganisms in

298 treatment systems open to the environment. Thus it is desirable to screen indigenous species. In the present study, we assessed the potential use of basidiomycetes from Brazilian ecosystems in the bioremediation of soils contaminated with organochlorine compounds in terms of (i) tolerance to organochlorine compounds in the soil, and (ii) capacity to biodegrade PCP.

Materials and methods

K.M.G. Machado et al. Biodegradation of pentachlorophenol A 25 g (dry weight) mixture of soil–CaSO4–sugar cane bagasse (95:2.5:2.5, dry weight) was placed in bottles of 800 ml. The moisture content was adjusted to 75% of the mixture’s water retention capacity and corrected weekly. A 2.5 g amount of inoculum was used. In the control treatment, the fungi were replaced with sterilized wheat grains. All treatments were carried out in quadruplicate. Free chlorides were determined in aqueous extract using a silver electrode.

Fungi Chromatographic analyses The 36 fungal strains studied were previously selected for Remazol brilliant blue R decolorization and a growth rate exceeding 1.0 cm day)1 in a solid medium (Faleiro et al. 1996 ; Okino et al. 2000). All cultures were from Brazilian ecosystems and were maintained in 2% (w/v) malt extract agar (MEA) at 4 C. They are deposited in the culture collection of basidiomycetes (CCB) of the Instituto de Botaˆnica, Sa˜o Paulo, SP, Brazil.

The compounds were extracted from the total content of each bottle with a soxhlet apparatus and the concentrations of organochlorines were determined by high-resolution gas chromatography (Matheus et al. 2000). The analyses were standardized for: PCP, pentachloroanisole (PCA), hexachlorethane, hexachlorobutadiene, tetrachlorobenzene, pentachlorobenzene and hexachlorobenzene. The factor of recovery ranged from 98.9 to 100.7%. The detection limit was 10 mg kg)1 of soil.

Soil Mineralization of The soil was collected from an area situated in Sa˜o Vicente, SP, in a region where organochlorine industrial residues with high concentrations of PCP had been inadequately disposed. The soil contained 98.0% (w/w) sand, pH 3.65, had a cation exchange capacity of 5.5 mEq 100 g)1 soil, and contained 2.3% (w/w) organic matter, 0.06 g nitrogen per 100 g)1 soil, 1.0 lg phosphorus per g)1 soil, and 0.01 mEq potassium per 100 ml)1. No bacteria or fungi were isolated from this soil. Before use, the soil was air dried, sieved through a 2 mm mesh, diluted in non-contaminated soil and autoclaved for 1 h at 100 C for three consecutive days. Fungal tolerance This test was performed as described by Matheus et al. (2000). Fungi were inoculated into non-inclined test tubes containing MEA and incubated at 30 C for about one week. The soil, with 46,300 mg of PCP kg)1 of soil, was diluted with non-contaminated soil. When fungal growth had started, a column of 15 g sterilized soil was placed over each colony, in quadruplicate, for 10, 50 and 100% (w/w) of PCP-soil. Non-contaminated soil was used as control. The fungi that could colonize more than 25% of the soil column were considered to be tolerant to organochlorine compounds. Fungal inoculum For the experiments of PCP degradation and mineralization, the inoculum of each fungus was prepared using wheat grains (Matheus et al. 2000).

14

C-PCP

An amount of 3.0 · 106 d.p.m. of [U-14C] pentachlorophenol (Sigma, St. Louis, MO) per 100 g of soil was placed in 300 ml bottles with 100 g of the mixture of soil–gypsum–bagasse. The culture conditions and the treatment control were the same as described to the PCP biodegradation. Incubation was carried out in a dark chamber at 23 ± 2 C. The initial radioactivity applied was determined by combustion and the 14CO2 produced was periodically captured in a trap of soda lime and extracted as described by Matheus et al. (2000). Thin layer chromatography (TLC) The extracts were evaporated in a rotary evaporator and resuspended in 2 ml of acetonitrile. PCP and PCA were identified by coelution of the 14C-labelled products with added standards on TLC plates (60F254 Merck). Mass balance analysis Soil samples were extracted for the activity remaining in the flasks. Five ml of acetonitrile was added to 10 g of sample in a stainless steel container, shaken in a Vortex apparatus (5 min) and later centrifuged (14,000 rpm, 20 min). The supernatant was collected and the extraction operation repeated. The activity was measured in 1 ml from combined supernatants. The residual soil was air dried at room temperature and then homogenized using a ceramic pulverizer. The bound residues was determined by combustion of 1 g of these material in a biological oxidizer (R.J. Harvey, model OX-500) and the 14CO2 produced was determined. The total radioactivity recovered was calculated by summing the

299

Pentachlorophenol biodegradation by basidiomycetes radioactivity in the acetonitrile extract, in the trapping solution and bound in the soil matrix.

14

CO2

Table 1. Colonization of PCP contaminated soil by basidiomycetes. Basidiomycetes

PCP concentration in column of soil (mg PCP kg)1 soil)

Agrocybe perfecta CCB1611 Antrodiella semidupina CCB446 Antrodiella sp. CCB426 Coprinus jamaicensis CCB318 Gymnopillus chrysopellus CCB381 Gymnopillus eartei CCB375 Hydnopolyporus fimbriatus CCB289 Irpex lacteusCCB196 Lentinus crinitus CCB217 Lentinus crinitus CCB356 Lentinus strigellus CCB380 Lentinus velutinus CCB268 Lentinus villosus CCB271 Lentinus zeyheri CCB274 Oudemansiella canarii CCB179 Oudemansiella canarii CCB241 Peniophora cinerea CCB204 Peniophora cremea CCB286 Phanerochaete chrysosporium CCB478 Phelinus lividus CCB305 Psathyrella tenuis CCB376 Psilocybe castanella CCB444 Pycnoporus sanguineus CCB175 Pycnoporus sanguineus CCB277 Pycnoporus sanguineus CCB458 Ripartitella brasiliensis CCB467 Stereum ostrea CCB267 Trametes versicolor CCB202 Trametes versicolor CCB372 Trametes villosa CCB165 Trametes villosa CCB213 Trametes villosa CCB291 Trichaptum byssogenum CCB203 Trogia buccinalis CCB390 Tyromyces pseudolacteus CCB193

4600.0 ng2 0.0 ng 0.0 0.0 ng ng 0.0 ng 0.0 ng 4600.0 0.0 ng ng 0.0 ng 0.0 0.0 ng 4600.0 0.0 ng 0.0 0.0 0.0 0.0 ng 0.0 4600.0 ng 4600.0 ng ng

Statistical analysis

Results and discussion Tolerance to organochlorine compounds in the soil The soil used contained 46,300 mg PCP kg)1 soil, the concentrations in the diluted soils being proportional to the executed dilutions, corresponding to 4600 and 23,000 mg of PCP kg)1 soil, respectively. Among the 36 fungi, Agrocybe perfecta (Rick) Sing. CCB161, Trametes villosa (Fr.) Kreisel CCB176 and CCB213, Trichaptum bisogenum (Jungh.) Riv. CCB203, Lentinus villosus Klotzsch CCB271 and Psilocybe castanella Peck CCB444 were capable of colonizing the columns of soil with 4600 mg PCP kg)1 soil (Table 1). Although the majority of the isolates studied are lignicolous species (Okino et al. 2000), the results showed that many of them developed well in the soil column. Due to the toxic effect of PCP as an inhibitor of oxidative phosphorylation, the growth of these fungi at the higher PCP concentration was unexpected. P. castanella, a saprotrophic species, was isolated from sandy organochlorine-contaminated soil in an area where industrial dumping had occurred. A. perfecta, a non-lignicolous species, was isolated from a sugar cane bagasse stack in Sa˜o Paulo city. T. villosa CCB176 and CCB213 are lignicolous fungi and both strains were isolated from forests distant from the contaminated areas. In another study, Peniophora cinerea (Fr.) Cooke CCB204 a lignicolous fungus isolated from the Atlantic forest, Sa˜o Vicente, SP, grew in soil with 1300 mg PCP kg)1 soil (results not shown) and was also evaluated. L. villosus and T. bissogenum are lignicolous species isolated from ‘restinga’ forests near the contaminated areas, but were not screened for the experiments of PCP biodegradation, because they did not show good growth in the incubation system. PCP depletion PCP depletion was evaluated on the basis of the residual concentrations of the compound in relation to the initial concentration of 1278 mg PCP kg)1 of soil. P. cinerea caused the greatest decrease of PCP (77.97%), followed by P. castanella (64.46%) and T. villosa CCB176 (58.07%), after 90 days of incubation. These results

1 2

Depletion of PCP in soil (%)

This was done according to a fully randomized experimental design. Analyses of variance (ANOVA) were conducted (a ¼ 0.05) using the ANOVA program of Excel 5.0 for Windows. Whenever a significant effect of treatments was observed, differences among treatment means were determined by the Tukey test (P £ 0.05).

CCB = code in the culture collection of basidiomycetes. ng = no growth.

65 days of incubation 90 days of incubation

100 90 80 70 60 50 40 30 20 10 0 CCB161

CCB204

CCB444

CCB176

CCB213

Control

Figure 1. Depletion of PCP by basidiomycete in organochlorinecontaminated soil (the bars indicate the standard deviation of the mean). CCB161 ¼ Agrocybe perfecta, CCB213 and CCB176 ¼ Trametes villosa, CCB444 ¼ Psilocybe castanella, CCB204 ¼ Peniophora cinerea, control ¼ without fungus.

(Figure 1) were significantly greater (P < 0.05) than those observed in control treatment (27.74%). The other fungi did not differ statistically from the control treatment. The biotic loss of 996.73, 824.07 and

300

K.M.G. Machado et al.

742.4 mg PCP kg)1 of soil incubated with the strains P. cinerea, P. castanella and T. villosa CCB 176, respectively, are close to those reported in other studies (44–89%) using Phanerochaete spp. and Lentinula edodes, in soils with initial concentration between 200 and 700 mg kg)1 of PCP (Lamar & Evans 1993; Okeke et al. 1993). Several factors can influence organochlorine depletion in soils (soil type, availability of nutrients, oxygenation, moisture content, pH and compound concentration). The conditions employed in the present experiment were compatible with the good colonization of the fungus in the soil and the optimization of manageable parameters was not evaluated. At this study, the soil had high concentrations of organochlorines, since this is a frequently encountered condition in the region. Chloride ion production The initial concentration of PCP could produce 851 mg Cl) kg)1 soil. The largest Cl) concentration was encountered in the soil incubated with P. cinerea and A. perfecta (440 and 418 mg Cl) kg)1 soil) corresponding to 51.7 and 51.3% of the initial concentration, respectively (Figure 2). The other treatments did not differ from each other, although all of them were significantly different (P £ 0.05) from the control. The results indicate transformation of PCP and molecule dehalogenation. In soil inoculated with A. perfecta and T. villosa CCB213, the levels of chloride ions did not show a direct relation to the loss of PCP. One explanation is that degradation by these fungi resulted in a transformation product that had more chlorine atoms than those produced by the other fungi. PCA production In soil treated with T. villosa CCB213, P. cinerea and A. perfecta, 16, 25.5 and 35 mg PCA kg)1 soil were detected, respectively, only at 90 days of incubation. The rates of PCP conversion to PCA (Figure 3) were lower than those presented by Phanerochaete chrysosporium,

Control 65 days of incubation

CCB213

90 days of incubation CCB176 CCB444 CCB204 CCB161 0

0,5

1

1,5

2

2,5

3

PCP to PCA conversion (%)

Figure 3. Transformation of PCP to PCA in organochlorinecontaminated soil after incubation with basidiomycetes. CCB161= Agrocybe perfecta, CCB213 and CCB176 ¼ Trametes villosa, CCB444 ¼ Psilocybe castanella, CCB204 ¼ Peniophora cinerea, control ¼ without fungus.

P. sordida and L. edodes (Lamar & Dietrich 1990; Okeke et al. 1993). PCA was not detected in the control treatment nor in soil with P. castanella and T. villosa CCB176 as described for other fungi as Trametes hirsuta and Ceriporiopsis subvermispora (Lamar & Dietrich 1990). Mineralization and mass balance of

14

C-PCP

The initial concentration of PCP in this test was 1180 mg PCP kg)1 soil. The extent of PCP mineralization was 3.07, 3.17, 4.95, 7.11 and 8.15% when the soil was incubated for 120 days with T. villosa CCB213, A. perfecta, P. castanella, P. cinerea and T. villosa CCB176, respectively (Figure 4). These rates are comparable to those described in the literature (Lamar et al. 1990; Liang & Mcfarland 1994). Indeed, the effect of the C/N ratio on these organisms could also have produced differences in the amounts of PCP degraded, mineralized or converted to PCA by the several fungi, as observed by Lamar & White (2001). Table 2 presents the 14C balance of masses. The low recovery of 14C from treatments with P. castanella,

9,00

400 350 300 250

7,00 CO2 (%)

65 days of incubation 90 days of incubation

450

14

-1

mg of inorganic chloride kg of soil

8,00 500

6,00 5,00 4,00

200

3,00

150

2,00

100

1,00

50

0,00

0 CCB161

CCB204

CCB444

CCB176

CCB213

Control

Figure 2. Inorganic chloride in organochlorine-contaminated soil after incubation with basidiomycetes. Initial concentration of 851 mg Cl) kg)1 of soil (the bars indicate the standard deviation of the mean). CCB161 ¼ Agrocybe perfecta, CCB213 and CCB176 ¼ Trametes villosa, CCB444 ¼ Psilocybe castanella, CCB204 =Peniophora cinerea, control =without fungus.

0

20

40 60 80 Time of incubation (days)

100

120

Figure 4. Mineralization of 14C-PCP by basidiomycetes in organochlorine-contaminated soil. (r) Agrocybe perfecta CCB161, (·) Trametes villosa CCB176, (m) T. villosa CCB213, (n) Psilocybe castanella CCB444, (d) Peniophora cinerea CCB204; (s) control= without fungus.

301

Pentachlorophenol biodegradation by basidiomycetes

Table 2. Mass balance of radioactivity from [14C] pentachlorophenol in soil bioaugmented with basidiomycetes during 120 days of incubation. Basidiomycetes

Agrocybe perfecta CCB1611 Peniophora cinerea CCB204 Psilocybe castanella CCB444 Trametes villosa CCB176 Trametes villosa CCB213 Control

Radioactivity recovery (%) Mineralization

Acetonitrile extract

Soil combustion

Total balance

3.17 7.11 4.95 8.15 3.07 0.53

47.84 nd2 28.21 16.16 37.93 67.30

31.41 nd 38.08 37.37 24.41 17.64

82.42 nd 71.24 61.68 65.41 85.47

1

CCB = code in the culture collection of basidiomycetes. nd = not determined.

2

T. villosa CCB213 and CCB176 may be justified by the lack of determination of volatilized compounds and adsorption to the bottles, which can correspond to as much as 10% (Liang & Mcfarland 1994). The main fate of PCP in soils inoculated with basidiomycetes is the transformation to extractable metabolites (as PCA) and bound residues, depending on the kind of soil. These residues can be formed by oxidative reactions mediated by a variety of biotic and non-biotic catalysts, producing polymers that become incorporated into humus matter (Bollag & Dec 1995; Ullah et al. 2000). In the soil bioremediation process, the most desirable outcome is the conversion from PCP to CO2 and bound residues, reducing the bioavailability and consequently the toxicity of this pollutant. Among the fungi evaluated, P. castanella and T. villosa CCB176 showed 43.03 and 45.52% of the recovered radioactivity as 14CO2 from PCP mineralization and soil combustion. This is the first study in which initial PCP levels of more than 1200 mg kg)1 soil were used. The performance of P. castanella, also emphasizes the possibility of use non-lignicolous basidiomycetes in bioremediation.

Acknowledgements This project was the result of an agreement between Rhodia Brazil Ltda and Universidade Estadual Paulista, together with Instituto de Botaˆnica of Sa˜o Paulo, Brazil. We are grateful to FUNDUNESP, CNPq and FAPEMIG for financial support.

References Bollag, J.-M. & Dec, J. 1995 Incorporation of halogenated substances into humic material. In Naturally-produced organohalogens, eds. Grimwall, A. & Leer, E.W.B. pp. 161–169. Amsterdam: Kluwer Academic Publishers. ISBN 0-7923-3435-3. Faleiro, C.M., Machado, K.M.G. & Bononi, V.L.R. 1996 Screening of ligninolytic fungi for soil bioremediation I: mycelial extension rates of ligninolytic basidiomycetes isolated in Brazil. In Biodegradation and Biodeterioration: Proceedings of the Second Latin American Symposium held at Porto Alegre, eds. Gaylarde, C.C., Sa´, E.L.S. &

Gaylarde, P.M. pp. 136–137. Porto Alegre: MIRCEN, UNEP/ UNESCO/ICRO – FEAGRO/UFRGS. Kirk, P.M., Cannon, P.F., David, J.C. & Stalpers, J.A. 2001 Ainsworth & Bisby’s Dictionary of the Fungi. Wallingford: C.A.B. International. ISBN 0-8519-9377-X. Lamar, R.T. & Dietrich, D.M. 1990 In situ depletion of pentachlorophenol from contaminated soil by Phanerochaete chrysosporium. Applied and Environmental Microbiology 56, 3093–3100. Lamar, R.T. & Dietrich, D.M. 1992 Use of lignin-degrading fungi in the disposal of pentachlorophenol-treated wood. Journal of Industrial Microbiology 9, 181–191. Lamar, R.T. & Evans, J.W. 1993 Solid-phase treatment of a pentachlorophenol-contaminated soil using lignin-degrading fungi. Environmental Science and Technology 27, 2566–2571. Lamar, R.T., Glaser, J.A. & Kirk, T.K. 1990 Fate of pentachlorophenol in sterile soils inoculated with the white-rot basidiomycete Phanerochaete chrysosporium: mineralization, volatilization and depletion of PCP. Soil Biology and Biochemistry 22, 433–440. Lamar, R.T. & White, R.B. 2001 Mycorremediation – Commercial status and recent developments. In Ex Situ Biological Treatment Technologies, vol 6, eds. Magar, A. & von Fahnestock, F.M. pp. 263–278. Columbus: Battelle Press. Leontievsky, A.A., Myasoedova, N.M., Golovleva, L.A., Sedarati, M. & Evans, C.S. 2002 Adaptation of the white-rot basidiomycete Panus tigrinus for transformation of high concentration of chlorophenols. Applied Microbiology and Biotechnology 59, 599– 604. Liang, R.L. & Mcfarland, M.J. 1994 Biodegradation of pentachlorophenol in soil amended with the white rot fungus Phanerochaete chrysosporium. Hazardous Waste and Hazardous Materials 11, 411–421. Matheus, D.R., Bononi, V.L.R. & Machado, K.M.G. 2000 Biodegradation of hexachlorobenzene by basidiomycetes in soil contaminated with industrial residues. World Journal of Microbiology and Biotechnology 16, 415–421. Okeke, B.C., Smith, J.E., Paterson. A. & Watson-Craik, I.A. 1993 Aerobic metabolism of pentachlorophenol by spent sawdust culture of ‘‘shiitake’’mushroom (Lentinus edodes) in soil. Biotechnology Letters 15, 1077–1080. Okino, L.K., Machado, K.M.G., Fabris, C. & Bononi, V.L.R. 2000 Ligninolytic activity of tropical rainforest basidiomycetes. World Journal of Microbiology and Biotechnology 16, 889–893. Reddy, G.V.B. & Gold, M. 2000 Purification and characterization of glutathione conjugate reductase: a component of the tetrachlorohydroquinone reductive dehalogenase system from Phanerochaete chrysosporium. Archives of Biochemistry and Biophysics 391, 271– 277. Ullah, M.A., Kadhim, H., Rastall, R.A. & Evans, C.S. 2000 Evaluation of solid substrates for enzyme production by Coriolus versicolor for use in bioremediation of chlorophenols in aqueous effluents. Applied Microbiology and Biotechnology 54, 832–837.

World Journal of Microbiology & Biotechnology (2005) 21: 303–308 DOI 10.1007/s11274-004-3827-3


Springer 2005

Solid-state fermentation of wood residues by Streptomyces griseus B1, a soil isolate, and solubilization of lignins Anju Arora1,*, Lata Nain2 and J.K. Gupta3 1 Centre for Conservation and Utilization of Blue Green Algae, Indian Agricultural Research Institute, N. Delhi 110012, India 2 Division of Microbiology, Indian Agricultural Research Institute, N. Delhi 110012, India 3 Department of Microbiology, Panjab University, Chandigarh 160014, India. *Author for correspondence: Tel.: +91-11-25848431, Fax: +91-11-25741648, E-mail: [email protected]

Keywords: Acid precipitable polymeric lignins (APPL), bioconversion, lignocellulosics, partial decomposition, solid-state fermentation, Streptomyces, wood residues

Summary The actinomycete strain Streptomyces griseus B1, isolated from soil, when grown on cellulose powder as submerged culture produced high levels of all the three components i.e. filter paper lyase (FPase), CMCellulase and bglucosidase of the cellulolytic enzyme system. FP activity and CMCellulase were present only extracellularly, while b-glucosidase was both intra- and extra-cellular. It produced highest FPase activity when grown on hardwood powder under submerged culture. It was unable to use lignin monomers (ferulic acid, vanillic acid and syringic acid) as carbon source. While growing on hardwood and softwood powders under solid-state conditions, it depleted them of cellulose (36.3% in the case of softwood and 14.4% in the case of hardwood). It also caused partial loss of lignin content in both the substrates by solubilizing them. These solubilized lignins could be recovered as acid precipitable polymeric lignins (APPL) from extracts of wood powders upon acidification. Extracts of inoculated wood powders yielded higher amounts of APPL than uninoculated controls. Also, the APPLs from Streptomyces-treated wood powders differed from control APPLs in their molecular weight distribution, as observed from their elution pattern using Sephadex G-100.

Introduction Increasing energy and environmental crises have once again focused attention on obtaining energy from renewable and non-polluting resources. Advancing agricultural operations and urbanization produce huge amounts of biomass wastes, which pose disposal problems and have to be recycled. Lignocellulose biotechnology offers significant opportunities to developing countries. Solid-state fermentation of lignocellulosics is an attractive process for developing countries, as it requires low capital and infrastructure and is practical for complex substrates including agricultural, forestry and food-processing residues (Howard et al. 2003). Among various alternatives, energy from biomass produced by microbiological processes in the form of bioethanol, methane, hydrogen and biodiesel have attracted interest. Bioethanol from sugar cane has been successfully used by Brazil and in the US; bioethanol is mainly produced from corn (Ward & Singh 2002). Because of the problems associated with conversion of lignocellulosics to fermentable sugars, ethanol from lignocellulosics has remained economically unattractive

and up to now, most of the plants have relied on starchy and sugary substrates (Classen et al. 1999). Each lignocellulosic substrate has its own mix of sugar polymers in the form of cellulose and hemicellulose, cemented together by lignin. The intractable nature of cellulose and the presence of recalcitrant lignin make the conversion of lignocellulosics difficult. Plant biomass is an abundant, underutilized resource which can be used as starting material for producing new value-added products and the development of additional biotechnological uses of lignocellulose requires an understanding of its biodegradation (Deobald & Crawford 2002). Major improvements/breakthroughs in the efficiency of hydrolysis of cellulose can be achieved by using enzymes. Higher yields of enzymes are required to allow pretreatment to make a major impact on production costs of bioethanol. Many organisms including bacteria, fungi, yeasts and actinomycetes have been known for their lignocellulolytic capabilities and use for biogas production, composting and SCP production (Howard et al. 2003). Among saprophytic organisms, which play a major role in bioconversion and recycling of these lignocellulosic wastes in the form of compost, are fungi and actinomy-

304 cetes. These have served for production of cellulases, for protein enrichment of lignocellulosic residues and production of value-added lignin byproducts. Soil actinomycetes like Streptomyces spp., Thermonospora and Thermoactinomyces have been known for production of cellulases and degradation of lignin. It is thought that native lignin degradation by actinomycetes is associated with primary growth and their main activity is lignin depolymerization and solubilization rather than mineralization. Further developments in high value applications for lignins and other byproducts could go a long way towards offsetting the high cost of cellulose hydrolysis (Thomas & Crawford 1998). In nature, microbes rarely act alone and various organisms including wood-rotting fungi, bacteria and actinomycetes attack, solubilize and catabolize lignins to varying degrees (Reid et al. 1982). These solubilized lignins are still polymeric and can be recovered by acid precipitation and can be used for important applications as antioxidants, surfactants and components in adhesives (Crawford et al. 1984). This report presents results of growth of a soil actinomycete on wood powders under solid-state fermentation and partial degradation and solubilization of lignins.

A. Arora et al. inoculating it on plates with basal medium containing 1.1% CMcellulose and incubating at 37 C for 5 days. Plates were flooded with aqueous Congo red solution (1 mg/ml). After 15 min dye was drained and plates were washed three times with 1 M NaCl and observed for production of yellow zones. For quantitative assay, the organism was grown in 50 ml basal medium dispensed in 250 ml Erlenmeyer flasks with 1% cellulose powder or wood powders as C source under shake culture at 30±2 C. 1 ml of spore suspension (showing 0.2 absorbance at 540 nm) in distilled water with 0.06% Tween 80 was used as inoculum. About 5 ml aliquots of culture were aseptically withdrawn on days 7 and 15 and filtered. The filtrate was used as source of extracellular enzymes. Culture solids (cell mass and residual cellulose), were washed three times with 0.05 M cold citrate buffer (pH 5) and suspended in 15 ml of the same buffer. The suspension was disrupted using Vibronics Ultrasonic Processor at 150 mA for 10 min at low temperature and centrifuged. Supernatants were used for checking intracellular enzyme activity. Enzyme assays

Materials and methods Organism Streptomyces griseus B1 was isolated from soil samples collected from composting leaf litter using an isolation medium consisting of basal medium with the composition as described by Sylvester & Kluepfel (1979), 3% macerated Whatman paper (as C source) and ampicillin 25 lg ml)1 and maintained on nutrient agar or yeast extract-tryptone agar. It was identified on the basis of sugar utilization pattern described by Pridham & Gottlieb (1948). Rhodotorula glutinis was obtained from the culture collection of the Department of Microbiology, Panjab University, Chandigarh. Carbon substrates for growth Hardwood (mango) and softwood (deodar) powders were prepared from wood shavings obtained from a local timber store. Wood shavings, dried at 40 C, were ground to fine powder to pass through 30-mesh sieve. They were extracted with hot water and dried in oven at 40 C till constant weight was obtained. Lignin model compounds ferulic, syringic and vanillic acids and vanillin were obtained from Fluka chemicals, USA. The [carboxyl-14C]vanillic acid and [carboxyl-14C]syringic acid were gifts from Dr. Kondrad Haider (Braunschweig, Germany). Cellulolytic activity The cellulolytic ability of S. griseus B1 was ascertained by the Congo red test (Teather & Wood 1982) by point-

Filter paper lyase (Fpase) activity was assayed at 50 C as described by Mandels (1975) with slight modifications and the reducing sugars released were measured by DNSA (Miller 1959). Activity was expressed as units ml)1 supernatant. CMCellulase assay was done according to the procedure described by Reese & Mandels (1963) and the reducing sugars released were measured by DNSA. CMCellulase activity was expressed as units ml)1 supernatant. One unit of Filter Paper activity or CMCellulase corresponded to 1 lmol of reducing sugar released ml)1 of enzyme solution h)1. B-Glucosidase (cellobiase) assay was performed using 4-paranitrophenyl-b-D -glucopyranoside as substrate (Wood & Bhat 1988). The b-glucosidase activity was expressed as units ml)1 and one unit was equivalent to 1 lmol of 4-paranitrophenol released ml)1 of enzyme solution h)1. Proteins in the supernatants were determined by the Lowry method using BSA as standard. Using lignin monomers as C source The Streptomyces B1 was grown in basal medium containing ferulic or vanillic acid as sole C source or along with 0.5% glucose. Ferulic and vanillic acid stock solutions were prepared initially as alkaline solutions and pH was later brought down to 7.2 with 1 M HCl. These were filter sterilized and added aseptically to the medium to give 0.5 mM final concentration. The inoculation of the medium was done with spore suspension and incubation was carried out on a rotary shaker at 30 ± 2 C. Depletion of ferulic and vanillic acids with growth of actinomycete was monitored regularly by measuring the fall in absorbance of centrifuged culture supernatant at 320 and 254 nm respectively. Rhodotorula

Partial degradation of wood residues by Streptomyces griseus

305

served as positive control for monitoring the fall in absorbance and depletion of lignin model compounds. It was grown in the same medium and conditions except that the pH of the basal medium was 5.0.

given by Updegraff (1969) and sugars formed upon hydrolysis of cellulose by the anthrone method.

Utilization of radiolabelled vanillic acid and syringic acid

The molecular weight of APPLs was determined by molecular sieve chromatography (Crawford et al. 1983) on a Sephadex G-100 column (void volume 35 ml) using proteins of known molecular weight like cytochrome c, chymotrypsinogen A and bovine serum albumin as standards. The lignin origin of APPLs was confirmed by subjecting the samples to acidolysis as done by Pometto & Crawford (1985) and determining the products by TLC on Silica Gel GF using the solvent system benzene-ethyl acetate-formic acid (85:15:1) and detecting the spots under UV illumination. All treatments were run in triplicate and the results were statistically analysed.

In these experiments [carboxyl-14C]vanillic acid and [carboxyl-14C]syringic acid were fed into the medium in addition to unlabelled substrates. The 14CO2 liberated was trapped in 1 M NaOH and counted at the end of 5 days incubation in a liquid scintillation counter. The experiment was performed using 250 ml Erlenmeyer flasks with small glass cups (containing 1 ml 1 M NaOH to trap 14CO2) hanging through the rubber corks fitted on neck of the flasks (Bakshi 1988). Each flask contained 50 ml of the basal medium with filter-sterilized (unlabelled) vanillic acid or syringic acid aseptically added to the medium to give 0.5 mM final concentration. The amount of labelled vanillic and labelled syringic acids incorporated into flasks gave radioactivity of 1.4 · 104 and 3.5 · 104 dpm respectively. In another set of flasks, 0.5% glucose was also added in addition to the lignin monomers. The inoculation of actinomycete into the medium was done with spore suspension as described earlier and incubation was carried out at 37 C with intermittent shaking in a metabolic shaker water bath. Uninoculated controls and flasks inoculated with Rhodotorula (to serve as positive control) were maintained and treated similarly except that the pH was 5. At the end of incubation, the alkali containing 14CO2 was transferred to scintillation vials by washing twice with distilled water and 8 ml of counting fluid POPOP was added to each vial. Solid-state fermentation of wood powders by S. griseus B1 S. griseus B1 was grown on 5 g portions of hardwood or softwood powders (defatted with a benzene/alcohol mixture 1:1 in a soxhlet apparatus for 6 h and dried at 40 C in an oven till constant weight was obtained) in 500 ml Erlenmeyer flasks (autoclaved at 121 C for 1 h for two consecutive days). Inoculation was done with S. griseus B1 cultures grown for 48 h in 50 ml basal medium and transferring all the contents to 500 ml flasks containing dried wood substrates under aseptic conditions. Uninoculated and inoculated flasks were incubated for 4 weeks and occasionally rolled and shaken. After incubation, acid-precipitable polymeric lignins (APPLs) were harvested by extracting partially degraded substrates with 100 ml 0.1 M NaOH for 2 h as described by Pometto & Crawford (1986). The precipitates were washed twice with acidified distilled water, redissolved in 25 ml of 0.1 M NaOH, reprecipitated by acidification and lyophilized. The spent lignocellulosic residues after APPL extraction were dried and analysed for lignin content by the Klason gravimetric method (Effland 1977) and cellulose content analysed by method

Characterization of APPLs

Results and discussion Streptomyces griseus B1 formed colonies tenaciously adhering to solid screening agar as described by Sylvester & Kluepfel (1979) and produced spores. Upon observing a smear, dyed with crystal violet, under the microscope, very thin filamentous hyphae were observed. When grown in submerged culture the isolate did not give uniform turbidity but formed clumps and grew well at 37 and 42 C. It produced very good zones of cellulolytic activity on CMC agar. when grown on 1% cellulose powder under shake culture, it expressed all the three components of the cellulase system extracellularly and specific activities mg)1 of protein after 7 days growth on cellulose were 18.3, 37.2 and 140 units of FPase, CMCellulase and cellobiase respectively. At harvest of cell mass on day 15, no intracellular FPase or CMCellulase was detected while high levels of b-glucosidase was present intracellularly. When native wood powders were provided as C substrate good yields of both components CMCase & FP activity were formed (Table 1). The enzyme levels were higher in case of hardwood than softwood and

Table 1. Levels of extracellular cellulases expressed by Streptomyces griseus B1 under shake culture. Carbon source

Cellulose powder Hardwood (mango) powder Softwood (deodar) powder CD (at 5%) 0.13 S EM 0.05

FP-lyase activity (IUml)1)

CMCellulase (IUml)1)

days of incubation

days of incubation

7 1.83 3.0

15 3.77 4.0

7 3.72 2.94

15 4.44 3.38

3.0

2.50

0.44

1.22

306

A. Arora et al.

FP activity even exceeded that expressed on cellulose powder. Ability to catabolize lignin monomers S. griseus B1 was unable to utilize lignin model compounds like ferulic acid, syringic acid or vanillic acid as exclusive C source or in combination with 0.5% glucose and there was no decline in UV absorption of the culture supernatants while Rhodotorula culture supernatants showed sharp decline in absorbance within 12 hrs in presence of glucose. S. griseus B1 did not evolve any 14 CO2 while growing on [carboxyl-14C]vanillic acid or syringic acid (Table 2). Solid-state fermentation of wood powders Growth of S. griseus B1 on hardwood and softwood substrates under solid-state fermentation at 37 C, caused higher depletion of cellulose (36.3%) from softwood than hardwood substrate (14.4%) while uninoculated controls showed no depletion (see Table 3). Hardwood substrates showed higher lignin loss and higher yield of APPL in inoculated as well as uninoculated treatments as compared to softwood substrates. Higher APPL yields from hardwood substrates correlated well with higher lignin loss. In case of hardwood

Table 2. Catabolism of [carboxyl-14C]vanillic acid and syringic acid by Streptomyces griseus in 5 days at 37 C. 14

CO2 evolved (dpm)

Carbon substrate 14

COOH Vanillic acid Control Streptomyces 14 COOH Syringic acid Control Streptomyces 14 COOH Syringic acid + 0.5% glucose Control Streptomyces Rhodotorula (positive control)

16 87 6 15 12 30 1.9 · 103

Table 3. Production of APPLs and change in cellulose and lignin contents of wood substrates by solid state fermentation with Streptomyces griseus for 4 weeks at 37 C. Treatment

Loss in content (%)

APPL yield (mg/5 g substrate)

Klason lignin Cellulose Softwood Control S. griseus Hardwood Control S. griseus CD (at 5%) 1.07 SEM 0.36 ND – Not detected.

0.5 10.5

ND 36.3

20 34

0.5 23.4

ND 14.4

28 43

10–13% and in case of softwood 16–17% of lost Klason lignin was recoverable as APPL. The uninoculated controls in both cases showed higher APPL recovery than Klason lignin lost. This discrepancy probably resulted from inaccuracies inherent in the Klason procedure (Crawford 1981). The molecular weight of APPLs recovered from both types of substrates as determined by molecular sieve chromatography showed that the APPLs recovered from inoculated treatments differed from those from uninoculated ones. The APPLs from inoculated substrates eluted in a sharp peak with a molecular weight of about 30,000 while APPLs from uninoculated controls were a heterogeneous mixture in the case of hardwood. In the case of softwood their molecular weight was lower (13,000–18,000) and the mixture was heterogeneous, showing that polymerization of APPLs occurred differently in inoculated and uninoculated substrates. TLC of acid-hydrolysed APPLs showed vanillic acid and vanillin as constituents of APPLs from softwood while those from hardwood had vanillic acid and syringic acid. This confirmed their origin from lignin. Cellulase production including b-glucosidase has been reported in actinomycetes. The FPase activity and CMCase production from this strain were comparable to yields reported by Crawford & McCoy (1972) for S. diastaticus and Thermonospora fusca. However, S. griseus B1 showed high constitutive level of b-glucosidase. Low b-glucosidase activity has been reported for S. flavogriseus (Ishaque & Kluepfel 1980) and most of it was cell associated. Stutzenberger & Kahler (1986) reported low b-glucosidase activity in Thermonospora curvata. Hardwood and softwood powder served as good substrate for FPase activity and hardwood was even better than cellulose powder, but less CMCase activity was produced on native lignocellulosics than on cellulose powder. Stutzenberger (1972) got maximum cellulase yields from T. fusca while growing on relatively difficult substrates like cotton fibres as compared to other soluble and insoluble substrates like Avicel, fibrous cellulose powder (Whatman) and cellobiose. Deobald & Crawford (1987) also assayed cellulase production by S. viridosporus growing on corn stover and found induced levels of CMCase, FPase activity and xylanases. S. griseus B1 was unable to metabolize lignin monomers as sole C source or in the presence of glucose. Crawford & Olson (1978) also reported that Streptomyces sp. grew poorly with 0.05% vanillic acid as sole C substrate but could grow well when 0.05% yeast extract was also present and converted it to guiacol. However, Pometto et al. (1981) and Sutherland et al. (1981) have given catabolic pathways of lignin monomers in S. albulus and S. viridosporus. McCarthy & Broda (1984) indicated that ligninolytic ability of an isolate could not be inferred from degradation of native lignin or other less complex substrates. When Streptomyces are grown on lignocellulosics, both lignin and carbohydrate contents are lost from

Partial degradation of wood residues by Streptomyces griseus substrates as it uses carbohydrates as energy source and dissolves lignins. S. viridosporus depleted 19.7% lignin and 44.4% carbohydrate from corn stover (Crawford et al. 1983). These workers obtained an APPL yield of 11.5 mg/g from spruce (softwood) and 17.2 mg/g from maple (hardwood). APPL yields appeared to be correlated with biodegradability of lignocellulosic. They proposed actinomycete mediation in polymerization of low molecular weight aromatic compounds into APPLS, which are chemically, the lignin fragments having an increased number of free phenolic, hydroxyl, carbonyl and carboxylic groups as compared to the native lignin, which makes them soluble at neutral pH and insoluble at lower pH values. During decay, lignin is not completely mineralized to CO2 but degraded partly to water-soluble intermediates. The kraft lignin was solubilized and depolymerized by the white-rot fungus Phanerochaete chrysosporium (Leisola et al. 1985). Reid et al. (1982) reported that soluble lignins produced by P. chrysosporium in aqueous medium had an increased number of carboxylic groups but were apparently not precipitable by acidification. Crawford et al. (1983) studied APPLs from actinomycete-degraded corn stover. The control APPLs had lower molecular weight but the acidification of extracts caused them to condense under low pH. Regalado et al. (1997) studied lignin degradation and modification by the soil-inhabiting fungus Fusarium proliferatum and found lignin degradation was maximal during primary metabolism and elution profiles of non-labelled industrial lignins and natural pine milled wood lignins (MWL) after fungal attack showed significant changes in molecular mass distribution pattern after 30 days incubation resulting in an increase in the high molecular mass lignin fraction. The role of extracellular enzymes produced by woodrotting organisms is stated to be in both degradation and polymerization of lignin. The laccase produced by Coriolus versicolor, a wood-rotting fungus, could degrade milled wood lignin to low molecular weight fractions. But gel permeation chromatography of original milled wood lignin and the MWL incubated with laccase showed that the enzyme had shifted the molecular weight of MWL towards the higher side by increasing the chain length (Ishikawa et al. 1963). Evans (1987) also stated that laccase, a polyphenol oxidase could act both in polymerization as well as degradation of lignins. Borgmeyer & Crawford (1985) reported that Streptomyces viridosporus produced APPLs by oxidative depolymerization while S. badius produced APPLs from repolymerization of lower molecular weight intermediates of lignin degradation in the presence of extracellular phenol oxidase. To adjudge the intermediary role of APPL in lignin degradation Pometto & Crawford (1986) studied its catabolism by ligninolytic Streptomyces species and Phanerochaete chrysosporium. They found that P. chrysosporium and S. setonii and all ligninolytic species of Streptomyces metabolized APPL very slowly and to a

307

limited extent showing that they released APPLs to gain access to polysaccharides. It appears that the actinomycete caused some chemical modification in the lignin fragments released into medium and made them to precipitate out from extracts, in larger amounts than the controls. Crawford et al. (1983) after analysing the elemental and methoxyl content of APPLs, have reported that in addition to b ether cleavage by Streptomyces, there are also other extensive modifications of the solubilized polymer. It was hypothesized earlier that APPLs are intermediates in lignin degradation but Pometto & Crawford (1986) have concluded that APPLs are probably a terminal product of lignin degradation by Streptomyces. From these findings it appears that Streptomyces griseus B1 attacked and modified lignin only to a limited extent and released them as soluble fragments but did not use them as C source. While growing on wood powders, the high levels of cellulases produced by it resulted in considerable loss of the cellulose component. Thus, this process can be further exploited for production of cellulases and solubilization of lignins from forestry residues.

References Bakshi, A. 1988 Lignocellulolytic activity of a soil actinomycete and its mutant. M.Sc Thesis, Panjab University, Chandigarh, India, 47–50. Borgmeyer, J.R. & Crawford, D.L. 1985 Production and characterization of polymeric lignin degradation intermediates from two different Streptomyces spp. Applied and Environmental Microbiology 49, 273–278. Classen, P.A.M., Van Lier, J.B., Lopez Contreras, A.M., van Neil, E.W.J., Sijtsma, L., Stams, A.J.M., de Vries, S.S. & Weusthuis, R.A. 1999 Utilization of biomass for the supply of energy carriers. Applied Microbiology and Biotechnology 52, 741–755. Crawford, D.L. 1981 Microbial conversion of lignin to useful chemicals using a lignin degrading Streptomyces. Biotechnology and Bioengineering Symposium 11, 275–291. Crawford, D.L. & McCoy, E. 1972 Cellulases of Thermonospora fusca and Streptomyces thermodiastaticus. Applied Microbiology 24, 150–152. Crawford, R.L. & Olson, P. 1978 Microbial catabolism of vanillate: decarboxylation to guaiacol. Applied and Environmental Microbiology 36, 539–543. Crawford, D.L., Pettey, T.M., Thede, B.M. & Deobald, L.E. 1984 Genetic manipulation of lignolytic Streptomyces and generation of improved lignin to chemical bioconversion strains. Biotechnology and Bioengineering Symposium 14, 241–255. Crawford, D.L., Pometto, A.L., III & Crawford, R.L. 1983 Lignin degradation by Streptomyces viridosporus, isolation and characterization of a new polymeric lignin degradation intermediate. Applied and Environmental Microbiology 45, 898–904. Deobald, L.A. & Crawford, D.L. 1987 Activities of cellulase and other extracellular enzymes during lignin solubilization by Streptomyces viridosporus. Applied Microbiology and Biotechnology 26, 158–163. Deobald, L.A. & Crawford, D.L. 2002 Lignocellulose biodegradation. In Manual of Environmental Microbiology, 2nd edn., eds. Hurst, J.C., Crawford, R.L., Knudsen, G.R., McInrey, J. & Stetzenbach, L.D. pp. 925–933. Washington DC: ASM Press, ISBN 155581199X. Effland, M. J. 1977 Modified procedure to determine acid–insoluble lignin in wood and pulp. TAPPI 60, 143–144.

308 Evans, C. S. 1987 Lignin degradation. Process Biochemistry 22, 102–105. Howard, R.L., Abotsi, E., Jansen van Rensburg & Howard, S. 2003 Lignocellulose biotechnology: issues of bioconversion and enzyme production. African Journal of Biotechnology 2, 602–619. Ishaque, M. & Kluepfel, D. 1980 Cellulase complex of a mesophilic streptomyces strain. Canadian Journal of Microbiology 26, 183–189. Ishikawa, H., Schubent, W.J. & Nord, F.F. 1963 Investigations on Lignins and lignification. The enzymic degradation of softwood lignins by white rot fungi. Archives of Biochemistry and Biophysics 100, 131–139. Leisola, M.S.A., Thanei-Wyss, U. & Feichter, A. 1985 Strategies for production of high ligninase activity by Phanerochaete chrysosporium. Journal of Biotechnology 3, 97–107. Mandels, M. 1975 Microbial sources of cellulases. Biotechnology and Bioengineering Symposium 5, 81–105. McCarthy, A.J. & Broda, P. 1984 Screening for lignin degrading actinomycetes and characterization of their activity against 14Clabeled wheat lignocellulose. Journal of General Microbiology 130, 2905–2913. Miller, G.L. 1959 Use of dinitrosalicylic acid reagent for determination of reducing sugars. Analytical Chemistry 31, 426–428. Pometto, A.L. & Crawford, D.L. 1985 Simplified procedure for recovery of lignin acidolysis products for determining the lignin degrading abilities of microorganisms. Applied and Environmental Microbiology 49, 879–881. Pometto, A.L., III & Crawford, D.L. 1986 Catabolic fate of Streptomyces viridosporus T7A produced acid precipitable polymeric lignins upon incubation with ligninolytic Streptomyces species and Phanerchaete chrysosporium. Applied and Environmental Microbiology 51, 171–179. Pometto, A.L., Sutherland, J.B. & Crawford, D.L. 1981 Streptomyces setonii: Catabolism of vanillic acid via guaicol and catechol. Canadian Journal of Microbiology 27, 636–638. Pridham, T.G. & Gottlieb, G. 1948 The utilization of C compounds by some actinomycetes as an aid for species determination. Journal of Bacteriology 56, 107–114.

A. Arora et al. Reese, E.T. & Mandels, M. 1963 Enzymatic hydrolysis of cellulose and its derivatives. Methods in Carbohydrate Chemistry 3, 139– 142. Reid, I.D., Abrams, G.D. & Pepper, J.M. 1982 Water soluble products from the degradation of aspen lignin by Phanerochaete chrysosporium. Canadian Journal of Botany 60, 252–260. Regalado, V., Rodriguez, A., Perestelo, F., Carnicero, A., Fuente, G. & Falcon, M.A. 1997 lignin degradation and modification by soil inhabiting fungus Fusarium proliferatum 63, 3716–3718. Stutzenberger, F.J. 1972 Cellulolytic activity of Thermonospora curvata: nutritional requirements for cellulase production. Applied Microbiology 24, 77–82. Stutzenberger, F.J. & Kahler, G. 1986 Cellulase biosynthesis during degradation of cellulose derivatives by Thermonospora curvata. Journal of Applied Bacteriology 61, 225–233. Sutherland, J.B., Crawford, D.L. & Pometto A.L. 1981 Catabolism of substituted benzoic acids by Streptomyces species. Applied and Environmental Microbiology 41, 442–448. Sylvester, N.D. & Kluepfel, D. 1979 Method for rapid screening of cellulolytic Streptomycetes and their mutants. Canadian Journal of Microbiology 25, 858–860. Teather, R.H. & Wood, P.J. 1982 Use of Congo red polysaccharide interactions in enumeration and characterization of cellulolytic bacteria from bovine rumen. Applied and Environmental Microbiology 43, 777–780. Thomas, T. & Crawford, D.L. 1998 Cloning of clustered Streptomyces viridosporus T7A lignocellulose catabolism genes encoding peroxidase and endoglucanase and their extracellular expression in Pichia pastoris.Canadian Journal of Microbiology 44, 364– 372. Updegraff, D.M. 1969 Semimicro determination of cellulose in biological materials. Analytical Biochemistry 32, 420–424. Ward, O.P. & Singh, A. 2002 Bioethanol technology: developments and perspectives. Advances in Applied Microbiology 51, 53–80. Wood, T.M. & Bhat, K.M. 1988 Methods for measuring cellulase activities. Methods in Enzymology 160, 87–112.


Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 309–315 DOI 10.1007/s11274-004-3572-7

Detecting the heavy metal tolerance level in ectomycorrhizal fungi in vitro Prasun Ray1, Richa Tiwari2, U. Gangi Reddy1 and Alok Adholeya1,* 1 Centre for Mycorrhizal Research, The Energy and Resources Institute, India Habitat Centre, D S Block Lodhi Road, New Delhi -110003, India 2 Department of Biotechnology, Barkatullah University, Bhopal -462026. Madhya Pradesh, India *Author for Correspondence: Tel.: +91-11-24682100/111, Fax: +91-11-24682144/145, E-Mail: [email protected] Received 17 February 2004; accepted 8 June 2004

Keywords: Coal ash, ectomycorrhiza, heavy metal tolerance, in vitro

Summary Eight isolates of ectomycorrhizal fungi namely, Laccaria fraterna (EM-1083), Laccaria laccata (EM-1191), Pisolithus tinctorius (EM-1081), Pisolithus tinctorius (EM-1293), Scleroderma cepa (EM-1233), Scleroderma flavidum (EM-1235), Scleroderma verucosum, (EM-1283) and Hysterangium incarceratum (EM-1185) were grown on specially designed cocktail media prepared by adding various concentrations of different heavy metals namely Al, As, Cd, Cr, Ni and Pb. The heavy metals were selected keeping in view their relative abundance in coal ash and potential toxicity. The fungal isolates were grown on such designed cocktail media. The colony diameter was used for the measurement of the fungal growth. Total heavy metal accumulated in the mycelia was assayed by atomic absorption spectrophotometry. In relation to metal tolerance ability in general, Hysterangium incarceratum (EM-1185) showed maximum tolerance with respect to growth, Laccaria fraterna (EM-1083) and Pisolithus tinctorius (EM-1293) also showed considerable tolerance to the heavy metals tested. In relation to metal uptake in particular, Pisolithus tinctorius (EM-1293), has reported maximum uptake of Al (34642.58 ppm), Cd (302.12 ppm) and Pb (3501.96 ppm). In Laccaria fraterna (EM-1083), As (130.57 ppm) and Cr (402.38 ppm) uptake was recorded maximum; and Hysterangium incarceratum (EM-1185) has recorded maximum Ni (2648.59 ppm) uptake among the three suitable isolates documented here.

Introduction Coal-based thermal power stations generate over 70% of the total electric power in India, using high ash content (40–50%) coal resulting in the production of huge amount of coal ash. Large areas of land are required for the disposal of these ashes (26,000 hectares nation-wide). Besides requiring large tracts of land for disposal, ash particles by virtue of their fineness becomes airborne, and this results in air pollution. Developing green cover extensively on and around disposal sites will arrest the fugitive dust and further prevent contamination of ground water and soil otherwise caused by the downward movement of metals. However, phytoremediation of such disturbed land, with the plants representing the original biodiversity of the area is a major challenge. Disturbance greatly alters the chemical, physical and biological characteristic of the soil. As a result revegetation frequency fails. Revegetation of disturbed soils should aim to reestablish a stable ecosystem with a fully functioning,

nutrient cycling process. Mycorrhizal fungi help in achieving revegetation by increasing nutrient uptake by plants and restoration of soil structure. Hence inoculation of forest tree seedlings with ecologically adapted ectomycorrhizal fungi could be the best option for the future. Several ectomycorrhizal fungi can protect their host plants against the toxicity of heavy metals in soil (Colpaert & Vanassche 1993). There is a considerable proof that mycorrhizae protect the roots against heavy metal toxicity (Gildon & Tinker 1983; Dueck et al. 1986; Jones & Huchinson 1988). Ectomycorrhizal fungi can increase the capacity of plants to grow in sites with toxic metals (Brown & Wilkins 1985). Recent studies have indicated that colonization of tree roots by ectomycorrhizal fungi can increase the tolerance of their hosts to the presence of metals in toxic concentrations (Turner 1994; Wilkinson & Dickinson 1995; Leyval et al. 1997). Ectomycorrhizae confer metal tolerance by binding metals to electronegative sites on the cell walls of the hyphae, or binding to the phosphates and sulphydryl compounds within the cells (Galli

310 et al. 1994; Godbold et al. 1998). In view of their metal accumulation capacities, it might be possible to use ectomycorrhizae in the disturbed soils. From very few investigations, it is clear that, in general, the metal-ameliorating effect strongly depends on the fungal species. This suggests that some fungi are effective ameliorators and others are not. Given the broad spectrum of fungi forming ectomycorrhizal associations it is thus most likely that different fungi have different mechanisms towards metal tolerance and varied sensitivity. Hence screening for selection of suitable isolates for reclamation of ash-affected sites remains an essential criterion. Although binding of metals to the external mycelium is an important mean, direct evidence for the significance of this process, either in artificial substrate or in soils is still lacking. Thus it is necessary to evaluate the role of the external mycelium in the immobilization of metals and experiments should be set up which quantify metal fluxes in the growth units and estimate the absolute amount of metals immobilized on fungal structures (Jentschke & Godbold 2000). Responses of ectomycorrhizal fungi to toxic metals are of importance in the view of interest in the reclamation of polluted sites and influence on plant growth and productivity (Blaudez et al. 2000 a, b). The purpose of the present study was to determine the tolerance of ectomycorrhizal fungi towards six metals: aluminium, arsenic, cadmium, chromium, nickel and lead and studying their metal uptake ability. Suitable isolate(s) thus selected could be eventually applied to ashoverburdened sites in association with their respective hosts for phytoremediation-based reclamation purposes.

Materials and methods

P. Ray et al. (10 ppm), Pb (10 ppm), and Ni (20 ppm), respectively in MMN media (1.0x). Gradients of such metalimpregnated media were made ranging from 0.1x to 1.0x. Agar was replaced by phytagel (Sigma Chemicals Co) as solidifying agent. Measurement of growth Three fungal discs of 7 mm diameter were cut from the edge of an actively growing colony and placed equidistantly from each other in the petri dishes containing the cocktail media with increasing concentrations of metals. They were incubated in the dark at 24 C until the margin of the adjacent colony touched each other. Fungal growth in terms of colony diameter was measured as a function of time (Sundari & Adholeya 2001) using the software Image Proplus. Heavy metal analysis After growth, the mycelium in each of the petri dishes was harvested by washing with 10 mM sodium citrate buffer, pH 6.0 (Doner & Becard 1991) at 37 C, and the mycelial dry weights of triplicate samples were determined after overnight air-drying. Colonies were digested with 5 ml of HNO3 and 1 ml of HF in closed vessel at 170 C (Kalra et al. 1989) using MARS 5 (microwave accelerated reaction system 5), CEM Corp, for heavy metal analysis. The digested fungal samples were analysed for absorbed Al, As, Cd, Cr, Pb and Ni concentration within the mycelia. The buffer after washing was collected and analysed for adsorbed heavy metals on the mycelia. The analysis was carried out by TJA Solutions AAS software version 1.14 (Unicam) Model SOLAAR M Series with GF 95 graphite furnace equipped with FS 95 auto sampler having Software SOLAAR M Data station version 8.12.

Selection of isolates Eight isolates of ectomycorrhizal fungi namely Laccaria laaccata (EM-1191), Laccaria fraterna (EM-1083), Pisolithus tinctorius (EM-1081), Pisolithus tinctorius (EM-1293), Scleroderma verucosum (EM-1283), Scleroderma flavidum (EM-1235), Scleroderma cepa (EM1233) and Hysterangium incarceratum (EM-1185) were obtained from the Centre for Mycorrhizal Culture Collection (CMCC), TERI, New Delhi. They were maintained and subcultured regularly in modified Melin-Norkrans (MMN) agar (Marx 1969) till they were used for the present investigation. Preparation of cocktail media In this experiment specially designed cocktail media were prepared. One thousand ppm (1000 mg/l) standard of each of Al, As, Cd, Cr, Pb and Ni were prepared using their oxides following a standard protocol (Dean 1985). Cocktail media were then prepared by adding Al (400 ppm), As (1 ppm), Cd (10 ppm), Cr

Results Growth Growth of the isolates on the heavy metal amended media is represented in Figures 1 and 2. Although assessment of fungal biomass is also used for measurement of growth, it only gives the absolute value and does not take account for the intermediate growth kinetics. So for the present work we have used colony diameter as the parameter for measurement of fungal growth. Three isolates, namely Laccaria fraterna (EM-1083), Hysterangium incarceratum (EM-1185) and Pisolithus tinctorius (EM-1293) showed very good growth in the entire range of metal amendment levels tested in this experiment (Figure 1). Laccaria fraterna (EM-1083) and Pisolithus tinctorius (EM-1293) showed a decrease in growth with increasing metal concentrations. Growth of Hysterangium incarceratum (EM-1185) however did not change significantly with increasing concentration of

311

Heavy metal-tolerant ectomycorrhizal fungi 6 Laccaria fraterna [EM-1083] Pisolithus tinctorius [EM-1293] Hysterangium incarceratum [EM-1185]

Growth (Colony diameter in c.m)

a 5

4

b c

a

cd

cd 3

b

a

cd

d

d

e

b ab

ab cd

ab

ab

cdb

ab

db

cdb

d

2

ef b

b

ab f bc b

tration of metal amendment levels (Figure 2). We also tested one more isolate named Lacterious controversus (EM 1197), which however failed to grow even on the lowest concentration of the cocktail media, thus proving it to be highly susceptible to metal concentrations. Heavy metal uptake The total metal uptake (absorbed and adsorbed) profiles of three isolates, namely Laccaria fraterna (EM-1083), Hysterangium incarceratum (EM-1185) and Pisolithus tinctorius (EM-1293) are represented in figures 3–8.

1

0 Control 0.1X

0.2X

0.3X

0.4X

0.5X

0.6X

0.7X

0.8X

0.9X

1.0X

a

Laccaria fraterna [(0.01)=386.94] Hysterangium incarceratum [(0.01)=228.54] Pisolithus tinctorius [(0.01)=386.13] c d

30000

Aluminum uptake (ppm)

Figure 1. Growth of Laccaria fraterna (EM-1083), Pisolithus tinctorius (EM-1293) and Hysterangium incarceratum (EM-1185) in heavy metalamended cocktail media. x represents concentrations of different metals amended in the cocktail media, where 1x constitutes Al (400 ppm), As (1 ppm), Cd (10 ppm), Cr (10 ppm), Pb (10 ppm), and Ni (20 ppm), respectively. Letters above the histogram bars represents Analysis of Variance (ANOVA). Bars with different letters indicate values with significant difference. Bars represent means of three replicates at P £ 0.01.

a

a

35000

Concentration of metals.

b a

25000

20000

a

e f

15000 g

i

10000

b

d

e

f

g

h

b

c

c

c d

e g

f

f

f

h h

ji 5000 i

metal amendments with respect to control (Figure 1). Laccaria laccata (EM-1191) and Pisolithus tinctorius (EM-1081) showed growth up to 0.4x concentration, growth of Scleroderma verucosum (EM-1283), Scleroderma flavidum (EM-1235) and Scleroderma cepa (EM-1233) decreased very sharply with increasing concentration of heavy metals. Scleroderma verucosum (EM-1283) showed growth of up to 0.3x while Scleroderma flavidum (EM-1235) and Scleroderma cepa (EM-1233) restricted their growth up to 0.2x concen7 a

b

a

a

c

a a b

3

b

d

b

2

b

b c c c

c cd

1

c

cd

0.3X

0.4X

0.5X

0.6X

0.7X

0.8X

0.9X

1.0X

Figure 3. Aluminum uptake in ppm of fungal biomass, by Laccaria fraterna (EM-1083), Hysterangium incarceratum (EM-1185) and Pisolithus tinctorius (EM-1293) at different concentrations of metalamended cocktail media. x represents concentrations of different metals amended in the cocktail media, where 1x constitutes Al (400 ppm), As (1 ppm), Cd (10 ppm), Cr (10 ppm), Pb (10 ppm), and Ni (20 ppm), respectively. Letters above the histogram bars represents Analysis of Variance (ANOVA). Bars with different letters indicate values with significant difference. Bars represent means of three replicates at P £ 0.01. 140

a

Hysterangium incarceratum [LSD(0.01)=0.42] Pisolithus tinctorius [LSD(0.01)=3.50] Laccaria fraterna [(0.01)=11.48] c d

120 100 80

4

0.2X

Aluminum concentrations

Arsenic uptake (ppm)

Growth (Colony diameter in cm)

Control 0.1X

Scleroderma verucosum [EM-1283] Scleroderma flavidum [EM-1235] Pisolithus tinctorius [EM-1081] Scleroderma cepa [EM1233] Laccaria laccata [EM-1191]

6

5

0

b c

60

a

40

e

20

hf

ge

hf

e

e

d

b

cd

c

c

d

c

a

b

0.7X

0.8X

0.9X

1.0X

f

i f f f g h h

0 Control

0.1X

0.2X

0.3X

i

0.4X

Concentration of metals

0 Control 0.1X

0.2X

0.3X

0.4X

0.5X

0.6X

Arsenic concentrations

Figure 2. Growth of Scleroderma verucosum (EM-1283), Scleroderma flavidum (EM-1235), Scleroderma cepa (EM-1233), Pisolithus tinctorius (EM-1081) and Laccaria laccata (EM-1191) in heavy metal-amended cocktail media. x represents concentrations of different metals amended in the cocktail media, where 1x constitutes Al (400 ppm), As (1 ppm), Cd (10 ppm), Cr (10 ppm), Pb (10 ppm), and Ni (20 ppm), respectively. Letters above the histogram bars represent Analysis of Variance (ANOVA). Bars with different letters indicate values with significant difference. Bars represent means of three replicates at P £ 0.01.

Figure 4. Arsenic uptake in ppm of fungal biomass, by Laccaria fraterna (EM-1083), Hysterangium incarceratum (EM-1185) and Pisolithus tinctorius (EM-1293) at different concentrations of metal-amended cocktail media x represents concentrations of different metals amended in the cocktail media, where 1x constitutes Al (400 ppm), As (1 ppm), Cd (10 ppm), Cr (10 ppm), Pb (10 ppm), and Ni (20 ppm), respectively. Letters above the histogram bars represent Analysis of Variance (ANOVA). Bars with different letters indicate values with significant difference. Bars represent means of three replicates at P £ 0.01.

312

P. Ray et al. 2500

b

b

c d e f

c

100 e

g g h f

h

egh

cd

eg

d

0.2X

0.3X

b

c

d a

1500 b d

1000

g

500

b

b

0.9X

1.0X

h g j

Control 0.1X

i

d

e

f

g

0.4X

0.5X

0.6X

0.7X

0.8X

ab

c

d

a

b

c

Control 0.1X 0.2X 0.3X 0.4X 0.5X 0.6X 0.7X 0.8X 0.9X 1.0X

Nickel concentrations

Figure 5. Cadmium uptake in ppm of fungal biomass, by Laccaria fraterna (EM-1083), Hysterangium incarceratum (EM-1185) and Pisolithus tinctorius (EM-1293) at different concentrations of metalamended cocktail media x represents concentrations of different metals amended in the cocktail media, where 1x constitutes Al (400 ppm), As (1 ppm), Cd (10 ppm), Cr (10 ppm), Pb (10 ppm), and Ni (20 ppm), respectively. Letters above the histogram bars represent Analysis of Variance (ANOVA). Bars with different letters indicate values with significant difference. Bars represent means of three replicates at P £ 0.01.

Hysterangium incarceratum [(0.01)=6.97] Pisolithus tinctorius [(0.01)=18.55] Laccaria laccata [(0.01)=15.37]

Figure 7. Nickel uptake in ppm of fungal biomass, by Laccaria fraterna (EM-1083), Hysterangium incarceratum (EM-1185) and Pisolithus tinctorius (EM-1293) at different concentrations of metalamended cocktail media. x represents concentrations of different metals amended in the cocktail media, where 1x constitutes Al (400 ppm), As (1 ppm), Cd (10 ppm), Cr (10 ppm), Pb (10 ppm), and Ni (20 ppm), respectively. Letters above the histogram bars represent Analysis of Variance (ANOVA). Bars with different letters indicate values with significant difference. Bars represent means of three replicates at P £ 0.01.

a

3500

a

Hysterangium incarceratum [(0.01)=23.20] Laccaria laccata [(0.01)=51.23] Pisolithus tinctorius [(0.01)=100.76]

a

a

b

b

300

b cd

d

c

b

c

c

e

200

b

b

b

d ef

e fg

ef gh

b

100 g

fg

e

f

d

d

d

a

c

Lead uptake (ppm)

Chromium uptake (ppm)

g

gh

Cadmium concentrations

fh

c

f

0

0

400

b

b

e

e f

f h

a

a

f

f

50

d

de

c

2000

Nickel uptake (ppm)

Cadmium uptake (ppm)

200

b

b

Laccaria laccata [(0.01)=17.44] Pisolithus tinctorius [(0.01)=30.73] Hysterangium incarceratum [(0.01)=50.69]

a

b

250

150

a

a

Laccaria fraterna [(0.01)=3.78] Hysterangium incercaratum [(0.01)=12.20] Pisolithus tinctorius [(0.01)=17.89]

300

3000

e

1000

0 0.3X

0.4X

0.5X

0.6X

0.7X

0.8X

0.9X

1.0X

Chromium concentrations

a

b

f g

g h

0 0.2X

d

1500

500

h Control 0.1X

c

2000

j

i

h

hh

g g

f f

e

c

d

e

e d

c

b

a

cd

i

Control 0.1X 0.2X 0.3X 0.4X 0.5X 0.6X 0.7X 0.8X 0.9X 1.0X

Lead concentrations

Figure 6. Chromium uptake in ppm of fungal biomass, by Laccaria fraterna (EM-1083), Hysterangium incarceratum (EM-1185) and Pisolithus tinctorius (EM-1293) at different concentrations of metal amended cocktail media. x represents concentrations of different metals amended in the cocktail media, where 1x constitutes Al (400 ppm), As (1 ppm), Cd (10 ppm), Cr (10 ppm), Pb (10 ppm), and Ni (20 ppm), respectively. Letters above the histogram bars represents Analysis of Variance (ANOVA). Bars with different letters indicate values with significant difference. Bars represent means of three replicates at P £ 0.01.

Figure 8. Lead uptake in ppm of fungal biomass, by Laccaria fraterna (EM-1083), Hysterangium incarceratum (EM-1185) and Pisolithus tinctorius (EM-1293) at different concentrations of metal-amended cocktail media. x represents concentrations of different metals amended in the cocktail media, where 1x constitutes Al (400 ppm), As (1 ppm), Cd (10 ppm), Cr (10 ppm), Pb (10 ppm), and Ni (20 ppm), respectively. Letters above the histogram bars represent Analysis of Variance (ANOVA). Bars with different letters indicate values with significant difference. Bars represent means of three replicates at P £ 0.01.

Hysterangium incarceratum (EM-1185) Heavy metal uptake was highest in 0.9· concentration of cocktail media, while the least uptake was noticed in the control. However this isolate showed a steady growth in all the amendment levels. No significant difference in the growth was noticed in the metalamended media with respect to control. This isolate recorded maximum Ni uptake (2648.59 ppm) among the three most suitable isolates documented here.

Laccaria fraterna (EM-1083) Al (19312.97 ppm) and Cd (62.30ppm), uptake was highest at 0.7x and 0.8x concentration of cocktail media. Ni uptake (518.17 ppm) was recorded maximum in the 0.9x concentrations, and As (130.57 ppm), Cr (402.38 ppm) and Pb (1428.24 ppm) uptakes were highest in 0.7x concentration of metal-amended media. Cd (60.46 ppm) and Ni (509.19 ppm) uptake was also very high in 0.7x concentration of metal amended media. A decrease in growth was recorded beyond 0.7x with an

Heavy metal-tolerant ectomycorrhizal fungi increase in subsequent metal uptake. Maximum growth was recorded in the control where the subsequent metal uptake was least. As (130.57 ppm) and Cr (402.38 ppm) uptake was recorded maximum among the three suitable isolates reported here. Pisolithus tinctorius (EM-1293) Al (34642.58 ppm) and As (44.42 ppm) uptake was maximum at 0.7x, Cd (302.12 ppm), Cr (337.53 ppm) and Pb (3501.96 ppm) uptake were maximum in the 1.0x concentrations and Ni (1678.56 ppm) uptake was highest in the 0.9x concentration of the cocktail media. The growth of the isolate was comparatively more in the higher concentrations of metal-amended media where subsequent metal uptake was also high. The growth however did not change significantly due to metalamendment with respect to control. This isolate has reported maximum uptakes of Al (34492.50 ppm), Cd (302.12 ppm) and Pb (3501.96 ppm) among the three isolates represented here.

Discussion Ectomycorrhizae can increase the capacity of plants to grow in sites of toxic metals (Brown & Wilikins 1985; Jones & Hutchinson 1986; Dixon 1988) by accumulating them in the extramatricular hyphae (Galli et al. 1994) and extrahyphal slime (Tam 1995), thereby reducing uptake into the plant (Marchner & Dell 1994). However, the mechanisms involved in conferring this increase in tolerance have proved difficult to resolve. They may be quite diverse and show considerable metal specificity since large differences in response to metals have been observed, both between fungal species and to different metals within a species (Hartley et al. 1997; Huttermann et al. 1999). Tam (1995) showed considerable variation between the ability of five ectomycorrhizal fungi to grow in a culture with a range of nine different heavy metals. This is also in broad concordance with our findings. Out of eight isolates we tested, three isolates showed a very good growth, while rest of the isolates showed less or no growth on higher concentrations of metal-impregnated media. This supports the fact that despite the similar artificial conditions, different fungi have different degree of tolerance towards heavy metals. This is definitely because each ectomycorrhizal species has its own physiology and therefore its own specificity with respect to the metal tolerance levels. Most of the mechanisms involved in metal tolerance (Jentschke & Godbold 2000) include absorption of metal by the hyphal sheath, reduced access to apoplast due to the hydrophobicity of the fungal sheath, chelation by fungal exudates, and adsorption onto the external mycelium. Moreover, adsorption largely depends upon the relative affinity of different metals to bind to the fungal cell wall (Kapoor & Viraraghavan 1995). Hence the metal in particular, the chemical

313 composition of the fungal cell wall and the fungal species involved remains an important consideration. Thus this could be the reason why high uptake or tolerance to a particular metal by one isolate, did not confer tolerance to all the metals simultaneously. Further, the variation between the species with respect to metal uptake and subsequent growth observed in our experiment supports the fact that adsorption largely depends upon the metal and is likely to vary in significance between different fungal species as well. Al, Cd and Pb uptake was reported maximum in Pisolithus tinctorius, (EM-1293) with a significant growth with respect to control. This hints that this isolate could be possibly tolerant to Al, Cd and Pb. However the exact mechanism behind such possible tolerance cannot be anticipated. It could be due to the formation of aluminium–polyphosphate complexes inside the fungal cell (Martin et al. 1994). Cd is a nonessential element and can be toxic at very low concentrations (Blaudez et al. 2000). However the effects of Cd on the in vitro growth of ectomycorrhizal fungi are very varied (Mccreight & Schroeder 1982; Colpaert & Vanassche 1992) In our present work, Cd uptake was high in Hysterangium incarceratum (EM-1185) and Pisolithus tinctorius (EM-1293), thus hinting them to be Cd-tolerant. In Hysterangium incarceratum (EM1185), a consistent growth and simultaneous high Ni uptake was observed. Such possible tolerance could possibly due to the detoxification of the nickel by the ectomycorrhizal fungi (Jones & Hutchinson 1988). Similarly increased growth and subsequent higher uptake of Al, Cd and Pb in Pisolithus tinctorius (EM1293) hints towards its ability to detoxify the metals (Tam 1995). Vare (1990) demonstrated presence of aluminium polyphosphate granules in ectomycorrhizal fungi and suggested them to be responsible for such detoxification of aluminium. Although the mechanisms of tolerance of various heavy metals on the growth of ectomycorrhizal fungi are well documented, we did not come across many papers mentioning the mechanism of chromium tolerance on the growth of ectomycorrhizal fungi. In our present work, Laccaria fraterna (EM-1083) showed maximum Cr (402.38 ppm) uptake in 0.7x concentration of metalamended media. This suggests that this isolate could be tolerant to chromium in particular. However the cause behind such high uptake is beyond scope of the current study. Aggangan et al. (1998) have documented the effect of chromium on ectomycorrhizal fungi, but have not however mentioned the possible mechanism of chromium uptake or tolerance. Interestingly, for all the three isolates a critical metal concentration limit was identified beyond which uptake decreased with increase in metal concentration. In Laccaria fraterna (EM-1083), this was found to be 0.7x except for nickel. For Hysterangium incarceratum (EM-1185), this critical limit was 0.9x except for chromium, and in case of Pisolithus tinctorius (EM1293) such a limit was detected only for Al and As (0.7x)

314 and Ni (0.9x). For the rest, an increasing trend in metal uptake was recorded till 1.0x. This trend gives an indication that Pisolithus tinctorius (EM-1293) is more tolerant to heavy metals in general, followed by Hysterangium incarceratum (EM-1185) and Laccaria fraterna (EM-1083). At present it is not clear why results from field (Ashford et al. 1988) and laboratory-grown mycorrhizae (Behrmann & Heyser 1992) are not consistent. Possible explanations include, the effects of the rhizosphere, as presumably metals had to be first mobilized in the rhizosphere before entering into the mycorrhizae, methodological problems, etc. Difference in moisture regimes between laboratory and field (wet conditions in laboratory and dry conditions in field) that may influence the actual degree of water repellency by the hydrophobic fungi (Wessels 1997) may provide another likely explanation. To summarize, on the basis of preferential tolerance and uptake ability towards heavy metals in particular, Pisolithus tinctorius (EM-1293), has a reported maximum uptake of Al (34642.58 ppm), Cd (302.12 ppm) and Pb (3501.96 ppm). In Laccaria fraterna (EM1083), As (130.57 ppm) and Cr (402.38 ppm) uptakes were maximum and Hysterangium incarceratum (EM1185) has recorded maximum Ni (2648.59 ppm) uptake. We conclude, on the basis of the growth and the heavy metal sorption profile in vitro that Hyterangium incarceratum (EM-1185), Pisolithus tinctorius (EM1293), and Laccaria fraterna (EM-1083) can survive and as well as tolerate high metal concentrations and hence are the most suitable among all the isolates tested.

Acknowledgements The present study was supported by funding from IndoFrench Centre for the promotion of advanced research (IFCPAR), Project No. 2109-1. The authors wish to thank Dr R.K. Pachauri Director General, The Energy and Resources Institute, India for providing the infrastructure facilities. References Aggangan, N.S., Dell, B. & Malajczuk, N. 1998 Effects of chromium and nickel on growth of the ectomycorrhizal fungus Pisolithus and formation of ectomycorrhizas on Eucalyptus urophylla S.T. Blake. Geoderma 84, 15–27. Ashford, A.E., Peterson, C.A., Carpenter, J.L., Cairney, J.W.G. & Allaway, W.G. 1988 Structure and permeability of the fungal sheath in the Pisonia Mycorrhiza. Protoplasma 147, 149–161. Behrmann, P. & Heyser, W. 1992 Apoplastic transport through the fungal sheath of Pinus sylvestris Suillus bovinus ectomycorrhizae. Botanica Acta 105, 427–434. Blaudez, D., Botton, B. & Chalot, M. 2000a Cadmium uptake and subcellular compartmentation in the ectomycorrhizal fungus Paxillus involutus. Microbiology 146, 1109–1117.

P. Ray et al. Blaudez, D., Jacob, C., Turnau, K., Colpaert, J., Ahonen-Jonnarth, U., Finlay, R.D., Botton, B. & Chalot, M. 2000b. Differential responses of ectomycorrhizal fungal isolates to heavy metals in vitro. Mycological Research 104, 1366–1371. Brown, M.T. & Wilikins, D.A. 1985 Zinc tolerance of mycorrhizal Betula. New Phytologist 99, 101–106. Colpaert, J.V. & Vanassche, J.A. 1992 The effects of cadmium and the cadmium–zinc interactions on axenic growth of ectomycorrhizal fungi. Plant and Soil 145, 201–211. Colpaert, J.V. & Vanassche, J.A. 1993 The effects of cadmium on ectomycorrhizal Pinus sylvestris L. New Phytologist 123, 325–333. Dean, J.A. 1985 Lange’s Handbook of Chemistry, 13th edn. New York: McGraw-Hill Book Company. Dixon, R.K. 1988 The response of ectomycorrhizal Quercus rubra to soil cadmium, nickel and lead. Soil.Biology.Biochemistry. 20, 555– 559. Doner, L.W. & Becard, G. 1991 Solublization of gellan gels by chelation of cations. Biotechnology Techniques 5, 25–28. Dueck, T.A., Visser, P., Ernst, W.H.O. & Schat, H. 1986 Vesiculararbuscular mycorrhiza decrease zinc toxicity to grasses in zinc polluted soil. Soil Biology & Biochemistry 18, 331–333. Galli, U., Schuepp, H. & Brunold, C. 1994 Heavy metal binding by mycorrhizal fungi. Physiologia. Plantarium. 92, 364–368. Gildon, A. & Tinker, P.B. 1983 Interactions of vesicular arbuscular mycorrhizal infection and heavy metals on the development of vesicular-arbuscular mycorrhizas. New Phytologist 95, 247–261. Godbold, L., Jentschke, G., Winter, S. & Marschner, P. 1998 Ectomycorrhizas and amelioration of metal stress in forest tress. Chemosphere 36, 757–762. Hartley, J., Cairney, J. W. G. & Meharg, A.A. 1997 Do ectomycorrhizal fungi exhibit adaptive tolerance to potentially toxic metals in the environment? Plant and Soil 189, 303–319. Huttermann, A., Arduini, I. & Godbold, D.L. 1999 Metal pollution and forest decline. In Heavy Metal Stress in Plants, from Molecules to Ecosystems, eds. Prasad, N.M.V.& Hagemeyer, J.pp. 253–272 Berlin: Springer-Verlag. Jentschke, G. & Godbold, D.L. 2000 Metal toxicity and ectomycorrhizas. Physiologia Plantarum 109, 107–116. Jones, M.D. & Hutchinson, T.C. 1986 The effect of mycorrhizal infection on the response of Betula papyrifera to nickel and copper. Plant and Soil 102, 429–442. Jones, M.D. & Hutchinson, T.C. 1988 Nickel toxicity in mycorrhizal birch seedlings infected with Lacterius rufus or Scleroderma flavidum. II. Uptake of nickel, calcium, magnesium, phosphorous and iron. New Phytologist 108, 461–470. Jones, M.D., Dainty, J.D. & Hutchinson, T.C. 1988 The effect of infection by Lacterius rufus or Scleroderma flavidum on uptake of 63 Ni by Paper birch. Canadian Journal of Botany 66, 934–940. Kalra, Y.P., Maynard, D.G. & Radford, F.G. 1989 Microwave digestion of tree foliage for multielement analysis. Canadian Journal of Forest Research 19, 981–985. Kapoor, A. & Viraraghavan, T. 1995 Fungal biosorption – an alternative treatment option for heavy metal bearing wastewaters: A review. Bioresource Technology 53, 195–206. Leyval, C., Turnau, K. & Haselwandter, K. 1997 Effect of heavy metal pollution on mycorrhization colonization and function: physiological, ecological and applied aspects. Mycorrhiza 7, 139–153. Marchner, H. & Dell, B. 1994 Nutrient uptake in mycorrhizal symbiosis. Plant and Soil 159, 89–102. Martin, F., Rubini, P., Cote, R. & Kottke, I. 1994 Aluminum phosphate complexes in the mycorrhizal basidiomycete Laccaria bicolor: A27 Al-nuclear magnetic resonance study. Planta 194, 241– 246. Marx, D.H. 1969 The influence of ectotrophic mycorrhizal fungi on the resistance of pine roots to pathogenic interactions. An antagonism of mycorrhizal fungi and soil bacteria. Phytopathology 59, 153–163. Mccreight, J.D. & Schroeder, D.B. 1982 Inhibition of growth of nine ectomycorrhizal fungi by cadmium, lead and nickel in vitro. Environmental and Experimental Botany 22, 1–7.

Heavy metal-tolerant ectomycorrhizal fungi Sundari, S.K. & Adholeya, A. 2001 Functional integrity and structural stability of freeze dried ectomycorrhizal fungi established through viability assays. Canadian Journal of Microbiology 47, 172–177. Tam, P.C.F. 1995 Heavy metal tolerance by ectomycorrhizal fungi and metal amelioration by Pisolithus tinctorius. Mycorrhiza 5, 181–187. Turner, A.P. 1994 The responses of plants to heavy metals. In Toxic metals in Soil–Plant Systems, ed. Ross, S.M. pp. 153–187. Chichester, UK. John-Wiley.

315 Vare, H. 1990 Aluminum polyphosphate in ectomycorrhizal fungus Suillus variegatus (Fr.) O. kunze as revealed by energy dispersive spectrometry. New Phytologist 116, 663–668. Wessels, J.G.H. 1997 Hydrophobins: proteins that change the nature of fungal surface. Advances in Microbial Physiology 38, 1–45. Wilkinson, D.M. & Dickinson, N.M. 1995 Metal resistance in trees: the role of mycorrhizae. Oikos 72, 298–300.

Ó Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 317–321 DOI 10.1007/s11274-004-2152-1

Potential of Rhodococcus erythropolis as a bioremediation organism Alena Cˇejkova´1, Jan Masa´k1, Vladimı´ r Jirku˚1,*, Martin Vesely´2, Miroslav Pa´tek2 and Jan Nesˇ vera2 1 Department of Fermentation Chemistry and Bioengineering, Institute of Chemical Technology, CZ-166 28 Prague 6, Czech Republic 2 Institute of Microbiology, Academy of Sciences of the Czech Republic, CZ-14220 Prague 4, Czech Republic *Author for correspondence: E-mail: [email protected] Received 21 May 2004; accepted 21 July 2004

Keywords: Biofilm, conjugation, humic acid, monoaromatic compounds, Rhodococcus erythropolis, 16S rRNA

Summary The capability of Rhodococcus erythropolis CCM 2595 (ATCC 11048) to utilize phenol, pyrocatechol, resorcinol, p-nitrophenol, p-chlorophenol, hydroquinone and hydroxybenzoate, respectively, or as respective binary mixtures with phenol, was described. This capability was found to depend on the substrate and its initial concentration. Some monoaromatic compounds had a suppressive effect on the strain’s ability to utilize phenol in a binary mixture and easily utilizable monoaromatics were strong inducers of the phenol 2-monooxygenase (EC 1.14.13.7). The capacity of R. erythropolis to colonize a synthetic zeolite was demonstrated and the enhancement of phenol tolerance of biofilms utilizing phenol was observed. The effect of humic acids on phenol killing was described and discussed as well. To allow use of recombinant DNA technology for strain improvement, methods of genetic transfer (transformation and conjugation) in R. erythropolis were established.

Introduction The capacity of single cell organisms to utilize (degrade) naturally occurring organic compounds, evolved over millions of years, has been challenged with anthropogenic chemicals introduced into the environment deliberately, or through accidental spills and other releases (Semple & Cain 1996). In this connection, monoaromatic compounds have become common environmental contaminants. Thirty monoaromatics are on the ‘EPA Priority Pollutant List’ (EPA 1996), and some of them are listed on the ‘Priority List of Hazardous Substances’ published by the ‘Agency for Toxic Substances and Disease Registry’ (ASTDR 1997). Bioremediation of these compounds has become an attractive alternative to the traditional physical and chemical decontamination methods that can be costly and can produce hazardous products (Singleton 1994). However, a limited capability of the degrading microorganism to utilize a wider range of monoaromatic homologues (including binary or complex mixtures), as well as a limited tolerance of degraders towards the cytotoxicity of these substrates are, among others, fundamental limitations on the biodegradation of these compounds. Therefore, efforts are focused on selecting strains with a disposition to tolerate the effect of both substrate interactions and substrate toxicity. In this context, the bacterial genus Rhodococcus has been shown to be a Gram-positive group manifesting not only a wide

range of catabolic versatility (Hughes et al. 1998), but also some cell properties required for technological application of microbial degraders (Lang & Philip 1998). In order to control these diverse capabilities properly, the development of both physiological and genetic tools enabling the construction of a tailor-made (rhodococcal) phenotype, is very important. In this context, the capability to utilize model monoaromatics, the susceptibility to the effect of cell attachment and specific additives, as well as introduction of the methods for genetic transfer providing a basis for aimed gene manipulations, were investigated to show the applicability of Rhodococcus erythropolis in bioremediation.

Materials and methods Organisms and culture conditions Rhodococcus erythropolis CCM 2595 (ATCC 11048), CCM 2597, CCM 4426; Rhodococcus rhodochrous CCM 2751 and Rhodococcus equi CCM 3429 were kindly provided by the Czech Collection of Microorganisms, Masaryk University Brno, Czech Republic. Escherichia coli XL1-Blue MRF’ (Stratagene) and E. coli S17-1 (Simon et al. 1983) were used for gene cloning and transfer, respectively. Precultivations of R. erythropolis were performed using a rotary shaker (90 rev min)1,

A. Cˇejkova´ et al.

318 20 °C) and basic salt medium (Masa´k et al. 1997). The degradative function of suspended cells was tested using a bioreactor (Braun Biotech International, Germany) with an operating volume of 2–l, and with control of temperature (20 °C), pH (6.7) and rotation speed (120 rev min)1). Samples were withdrawn periodically from the culture vessel via the side-neck. If necessary, samples were subsequently filtered through a 0.2 l Millipore Membrane Filter (Millipore Corporation, MA). Cell growth/ growth yield analysis was performed using Bioscreen C analyser (Labsystem Oy, Finland). Carrier and cell attachment Particles of synthetic zeolite MS 3A (Grace, USA) were colonized using a modified procedure according to Masa´k et al. (1997). The degradative function of rhodococcal biofilms was investigated using a jacketed, tubular reactor (30 cm inner diameter; 70 cm length), enabling temperature (20 °C) control and a countercurrent, continuous circulation of medium (pH 6.7) and air. Humic acids Preparations used were isolated from the oxihumolites (oxidatively altered lignite analogues) of a brown coal deposit in the north part of the Czech Republic, and kindly provided by the Institute of Inorganic Chemistry, U´stı´ nad Labem, Czech Republic. Analytical methods Protein quantity monitoring was performed according by the Bradford method. Phenol concentration in the cell-free medium was assayed using the 4-amino-antipyrine colorimetric method (Martin 1949). Phenol hydroxylase (phenol 2-monooxygenase, EC 1.14.13.7) was measured by assaying the oxidation of NADPH at 340 nm [U ¼ 0.1DA min)1] according to Neujahr & Goal (1973).

sequences from five Rhodococcus type strains: R. erythropolis ATCC 4277T (GenBank accession no. X81929), R. fascians ATCC 12974T (GenBank accession no. X81930), R. rhodochrous ATCC 271T (GenBank accession no. X81921), R. equi DSM 20307T (GenBank accession no. AF490539) and R. opacus DSM 43206T (GenBank accession no. X80631). The sequence of 881-bp 16S rDNA fragment was determined using the ABI Prism 2100 sequencer (PerkinElmer) and deposited in GenBank under Accession No. AJ 620506. Conjugation was done essentially according to van der Geize et al. (2001), using the E. coli S17-1 (Simon et al. 1983) as donor. The cells of R. erythropolis and E. coli were mixed and cultivated on LBP plates (van der Geize et al. 2001) at 30 °C for 24 h. Transconjugants were selected on LBP plates containing nalidixic acid (50 lg/ml) and kanamycin (200 lg/ml).

Results and discussion Capability of the strain to utilize monoaromatic compounds Rhodococcus erythropolis CCM 2595 was selected from a screening of several other rhodococcal strains (see Materials and methods), examined because of their supposed biodegradative versatility and different capability to colonize solid surfaces. The potential of this strain to utilize a number of monoaromatic compounds, when added singly as sole carbon and energy source, is shown in Figure 1. It was observed that the decrease of specific growth rates of exponential growth had no uniform pattern on increasing the concentration of respective compounds. The results obtained show that the inhibitory effect of the respective substrate concentrations starts in the range from 0.1 to 0.5 g l)1. In view of the fact that the capacity of a cell population to utilize a substrate is also affected by its biodegradation time and metabolic and growth adaptability of a given taxon,

Microscopy The assessment of the colonization of zeolite surfaces was based on the image analysis of the carrier, using LUCIA 4.20 for Windows 2000 for light microscopy image processing. Genetic manipulations DNA isolation, PCR, transformation of E. coli, DNA cloning and DNA analysis were done by standard methods (Sambrook & Russell 2001). Chromosomal DNA of R. erythropolis was isolated as described previously (Treadway et al. 1999). Oligonucleotide primers 16SREF (TTGGATCCACCTCACGGTCTCG)and 16SRER (CAGGATCCTACGGGAGGCAGCAG) were used for PCR. The primers were designed according to the conserved regions of 16S rDNA

Figure 1. Effect of substrate concentration on the specific growth rate of R. erythropolis.

Rhodococcus erythropolis in bioremediation

319 resorcinol or phenol and hydroxybenzoate, respectively. More rapid utilization of these mixtures could be achieved by a pre-exposure to these substrates. Other monoaromatic compounds have a suppressive effect on the ability of R. erythropolis to utilize phenol in a binary mixture. These suppressions suggest a relatively low tolerance of this strain to the cumulation of cytotoxic effects of the respective monoaromatics, or a susceptibility to the specific regulatory roles of pyrocatechol (the effect of pyrocatechol accumulation in the culture medium during phenol degradation, among others, Paller et al. 1995). Phenol hydroxylase activity

Figure 2. Effect of utilized substrate on R. erythropolis growth yield (j) / lag period (h); (lag period interval was calculated using the lag time [Tl] equation).

among others, the individual effect of monoaromatic compounds on the lag period and growth yield was compared (Figure 2) using an initial substrate concentration of 0.3 g l)1. This comparison proves as well that there is no uniform capacity of R. erythropolis to utilize the range of substrates studied. Significant differences were found, especially among the growth effects of p-nitrophenol, p-chlorophenol, hydroquinone and hydroxybenzoate.

For further characterization of R. erythropolis, the ability of the monoaromatic compounds studied to affect the level of intracellular FAD-dependent phenol 2-monooxygenase (phenol hydroxylase, EC 1.14.13.7), catalyzing the initial step of phenol mineralization, was described (Figure 4). These data show that easily utilizable monoaromatics are strong inducers of this enzyme. The induction capacity of others is much weaker and nonutilizable p-nitrophenol has no induction capacity at all. The enzyme inducibility was found to be a changeable marker as far as the effect of a long-term strain storage in the presence of phenol is concerned. The combination of phenol and resorcinol (see section utilization of binary mixtures 3.2.) enhances the induction capacity of respective substrates (data not shown).

Utilization of binary mixtures Cell attachment effect In order to obtain preliminary information on substrate interaction, the utilization of binary mixtures combining phenol with an additive (monoaromatic) substrate was investigated. The results for respective substrate combinations (Figure 3) show promoted growth yield only under the growth conditions combining phenol and

Figure 3. Effect of substrate interaction on R. erythropolis growth yield during the utilization of a phenol (0.3 g l)1) + monoaromatic compound (0.3 g l)1) mixture.

The comparison between suspended and attached bacteria or yeasts have already provided results supporting the hypothesis that a multipoint, cell-support/cell–cell physical contact brings positive (complex) alterations in both cellular types, enhancing cell capabilities to tolerate both the cytotoxic effects of xenobiotics and the

Figure 4. Capacity of respective substrates to induce phenol hydroxylase.

320

Figure 5. Rate of phenol uptake in suspended (h) and attached (j) R. erythropolis cells. Error bars represent total variations of two replicates.

stressing conditions of the environment, among others (Jirku˚ et al. 2001; Junter et al. 2002). The enhancement of phenol tolerance to those phenol concentrations, which almost or significantly inactivate the R. erythropolis growth in suspended cultures, was investigated comparing the specific uptake of phenol in suspended and biofilm cultures (Figure 5). The experiments performed suggest that the phenol-utilizing biofilm can tolerate higher phenol concentrations. We were able to demonstrate biofilm formation on the synthetic zeolite surface (Figure 6) by most Rhodococcus strains tested (see Materials and methods), however, the colonization of zeolite was found to be fully reproducible in R. erythropolis only. Effect of humic additives The precipitations associated with the outer part of the cell wall of the yeast cell exposed to Na-humate, resulting in the resistance of yeasts to nystatin (Jirku˚ et al. 1998), encouraged the idea that such an attachment (supplementary structure) of a polydisperse, macromolecular compound could also enhance the tolerance

Figure 6. Scanning electron micrograph of R. erythropolis biofilm formed on a zeolite MS 3A (right panel). (Left panel, control).

A. Cˇejkova´ et al.

Figure 7. Effect of phenol concentration on R. erythropolis growth yield in the presence (0.05 g l)1) / absence (h) of humic acid S4 ( ), S5 (j) and PAB68 ( ) ).

of bacterial degraders towards cytotoxic monoaromatics. Comparison of biomass yield in the cell populations exposed and not exposed to three dissolved humic acids, respectively, (Figure 7) suggests that the enhancement of phenol tolerance in rhodococcal cells varies as a function of humic acid preparation and the initial concentration of carbon source (R. erythropolis has no capacity to utilize humic acids as sole carbon source). The different effect of the humic acid preparations tested corroborates the fact that any interaction of both cell surface and small (organic) molecule with a macromolecular humic substance is determined by the binding capacity of a humic compound, i.e. its macromolecular structure/weight, composition, complex/micelle forming capacity, surface electrical charge and potential etc. (Dec & Bollag 1997). Development and use of genetic transfer Establishment of methods for DNA manipulations in the characterized strain would allow purposeful modification of the strain characteristics, particularly the abilities to degrade the studied substrates. We have already optimized electrotransformation of R. erythropolis CCM 2595 with plasmid DNA (Vesely´ et al. 2003). Conjugation represents another possibility for genetic transfer. We have used this method for testing the system for integration of DNA into the chromosome. The integrative system extends the cloning potential and allows construction of plasmid-free, self-cloning strains devoid of the unwelcome vector sequences. Moreover, gene deletions and replacements may be achieved by fragment-directed chromosomal integration (Schwarzer & Pu¨hler 1991). The gene coding for 16S rRNA, which is usually present in several copies in bacteria, was chosen as a target for the homologous recombination. Since the genes for 16S rRNA are highly conserved within the group of related bacteria (Rhodococcus strains), oligonucleotide primers delimiting the internal fragment of

Rhodococcus erythropolis in bioremediation 881 bp could be designed. Using the primers 16SREF and 16SRER (with the attached BamHI sites) and chromosomal DNA isolated from R. erythropolis as the template, the fragment of 881 bp was amplified by PCR. The fragment was ligated into the BamHI site of the vector pK18mobsacB [22] and the resulting plasmid pRE16SmobsacB gained by cloning was isolated from the E. coli transformant. The sequence of the fragment was determined and compared with the sequences of 16S rDNA from the related Rhodococcus strains available in the GenBank database. The 881-bp sequence showed 99.7% identity with the respective sequence of 16S rDNA from R. erythropolis ATCC 4277T (a single-nucleotide mismatch), whereas the other sequences showed lower similarity. The percentage of identity was 96.4% for R. opacus, 95.5% for R. fascians, 95.5% for R. equi and 94.1% for R. rhodochrous. This analysis confirmed the taxonomic identity of the strain under study. The plasmid pRE16SmobsacB was transferred by transformation into E. coli S17-1. The selected transformant was then used as a donor for conjugative transfer (mobilization) into R. erythropolis. The transconjugants were selected on the plates with kanamycin and nalidixic acid and integration of pRE16SmobsacB (which is unable to replicate in R. erythropolis) into the chromosome was confirmed by PCR. The system for positive selection of a double recombination event based on the conditional lethal effect of the sacB gene (Scha¨fer et al. 1994) was also tested. The double recombination may result in completely removing the vector sequence from the chromosome and thus losing the respective phenotypic markers (sucrose sensitivity and kanamycin resistance). The selection of clones growing on sucrose (appearing with the frequency of approximately 10)4) and sensitive to kanamycin, confirmed that this system may be used for manipulations within the R. erythropolis chromosome. In view of the fact that a good bioremediation system usually requires the engagement of more microbial species, the strain described offers properties that can be physiologically and genetically enhanced/modulated, to provide a phenotype complementing a degradative function of mixed (technological) cultures. Acknowledgement This work was supported by the Grant Agency of the Czech Republic, projects: 526/01/0177 and 526/04/0542; EUREKA: E! 3100 CAWAB, and by Institutional Research Concept no. AV0Z5020903. References ASTDR 1997 Priority list of hazardous substances. Agency of toxic substances and disease registry.

321 Bradford, M.A. 1972 A rapid and sensitive method for the quantitation of microgram quantities of protein using the principle of dye binding. Analytical Biochemistry 72, 248–252. Dec, J. & Bollag, J.M. 1997 Determination of covalent and noncovalent binding interactions between xenobiotic chemicals and soil. Soil Science 162, 858–874. EPA 1996 Priority pollutants. Code of Federal Regulations, Title 40, Part 423, Appendix A. Hughes, J., Armitage, Y.C. & Symes, K.C. 1998 Application of whole cell rhodoccocal biocatalysts in acrylic polymer manufacture. Antonie van Leeuwenhoek 74, 107–118. Jirku˚, V., Masa´k, J. & Cˇejkova´, A. 2001 Reduced susceptibility of a model S. cerevisiae biofilm to osmotic upshifts. Journal of Microbiology and Biotechnology 11, 17–20. Jirku`, V., Zˇizˇka, Z., Sedla´rˇ ova´, R. & Pospı´ sˇ il, F. 1998 A modulatory effect of yeast cell-humic salt interaction. Microbiological Research 153, 149–152. Junter, G.A., Coquet, L., Vilain, S. & Jouenne, T. 2002 Immobilized-cell physiology: current data and the potentialities of proteomics. Enzyme and Microbial Technology 31, 201–212. Lang, S. & Philip, J.C. 1998 Surface-active lipids in rhodococci. Antonie van Leeuwenhoek 74, 59–70. Martin, R.W. 1949 Rapid colorimetric estimation of phenol. Nature 21, 1419–1420. Masa´k, J., Cˇejkova´, A. & Jirku˚, V. 1997 Isolation of acetone/ethylene glycol utilizing and biofilm forming strains of bacteria. Journal of Microbiological Methods 30, 133–139. Neujahr, H.Y. & Goal, A. 1973 Phenol hydroxylase from yeast. Purification and properties of the enzyme from Trichosporon cutaneum. European Journal of Biochemistry 35, 386–400. Paller, G., Hommel, R.K. & Kleber, H.P. 1995 Phenol degradation by Acinetobacter calcoaceticus NCIB 8250. Journal of Basic Microbiology 35, 325–335. Sambrook, J. & Russell, D.V. 2001 Molecular Cloning. A Laboratory Manual, 3rd edn. Cold Spring Harbor Laboratory Press, New York, Cold Spring Harbor, New York. ISBN 0879695773. Scha¨fer, A., Tauch, A., Jager, W., Kalinowski, J., Thierbach, G. & Pu¨hler, A. 1994 Small mobilizable multi-purpose cloning vectors derived from the Escherichia coli plasmids pK18 and pK19: selection of defined deletions in the chromosome of Corynebacterium glutamicum. Gene 145, 69–73. Schwarzer, A. & Pu¨hler, A. 1991 Manipulation of Corynebacterium glutamicum by gene disruption and replacement. Bio/Technology 9, 84–87. Semple, K.T. & Cain, R.B. 1996 Biodegradation of phenols by the alga Ochromonas danica. Applied and Environmental Microbiology 62, 1265–1273. Simon, R., Priefer, U. & Pu¨hler, A. 1983 A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram negative bacteria. Bio/Technology 1, 784– 791. Singleton, I. 1994 Microbial metabolism of xenobiotics: fundamental and applied research. Journal of Chemical Technology and Biotechnology 59, 9–23. Treadway, S.L., Yanagimachi, K.S., Lankenau, E., Lessard, P.A., Stephanopoulos, G. & Sinskey, A.J. 1999 Isolation and characterization of indene bioconversion genes from Rhodococcus strain I24. Applied Microbiology and Biotechnology 51, 786–793. van der Geize, R., Hessels, G.I., van Gerwen, R., van der Meijden, P. & Dijkhuizen, L. 2001 Unmarked gene deletion mutagenesis of kstD, encoding 3-ketosteroid Delta1-dehydrogenase, in Rhodococcus erythropolis SQ1 using sacB as counter-selectable marker. FEMS Microbiology Letters 205, 197–202. Vesely´, M., Pa´tek M., Nesˇ vera J., Cˇejkova´ A., Masa´k J. & Jirku˚ V. 2003 Host-vector system for phenol-degrading Rhodococcus erythropolis based on Corynebacterium plasmids. Applied Microbiology and Biotechnology 61, 523–527.

World Journal of Microbiology & Biotechnology (2005) 21: 323–327 DOI 10.1007/s11274-004-3571-8


Springer 2005

Production of the ligninolytic enzymes by immobilized Phanerochaete chrysosporium in an air atmosphere Guoce Yu*, Xianghua Wen and Yi Qian State Key Joint Laboratory of Environment Simulation and Pollution Control, Department of Environmental Science and Engineering, Tsinghua University, Beijing 100084, P.R. China *Author for correspondence: Tel.: 86-10-62772838, Fax: 86-10-62771472, E-mail: [email protected] Received 17 February 2004; accepted 4 August 2004

Keywords: Immobilized culture, in air, ligninolytic enzymes, non-immersed culture, Phanerochaete chrysosporium, production

Summary The production of the ligninolytic enzymes by Phanerochaete chrysosporium immobilized on polyurethane foam cubes in air was investigated by adopting different sizes and amounts of the carriers, different medium C/N ratios and different glucose-feeding strategies. No lignin peroxidase (LiP) activity was observed under nitrogen limitation (C/N ratio, expressed as glucose/NHþ 4 , 56/2.2 mM) with two sizes and three amounts of the carriers, while comparable levels of manganese peroxidase (MnP) activities were detected only in non-immersed cultures with two sizes of the carriers. A non-immersed state also stimulated LiP formation under carbon limitation (C/N ratio 28/44 mM). High peak activities of LiP, 197 and 164 U/l, were obtained in non-immersed cultures under carbon limitation at the C/N ratios of 28/44 and 56/44 mM, respectively, the occurrence of the activities coinciding with the complete consumption of glucose. A very low level of MnP was measured at the C/N ratio of 28/44 mM compared with the similar activities at 56/2.2 and 56/44 mM. An addition of 2 g glucose/l after its complete depletion improved both the production of LiP and MnP markedly in non-immersed culture at the initial C/N ratio of 28/44 mM, whereas a replenishment of 5 g/l, still enhancing the formation of MnP, inhibited the production of LiP first before the later reactivation. It is suggested that non-immersed liquid culture under carbon limitation reinforced by a suitable glucose feeding strategy is one potential way to realize high production of the ligninolytic enzymes by P. chrysosporium in air.

Introduction White rot fungi can secrete the enzymes capable of degrading lignin as well as a variety of persistent environmental pollutants (Bumpus et al. 1985; Kirk & Farrell 1987; Barr & Aust 1994). The ligninolytic enzymes are expected to be of much importance in biopulping and biobleaching in the paper industry, water pollution control and soil bioremediation. The basidiomycete Phanerochaete chrysosporium is the most extensively characterized white rot fungus and commonly adopted in the production of the ligninolytic enzymes, its ligninolytic system consisting mainly of lignin peroxidase (LiP, EC1.11.1.14), manganese peroxidase (MnP, EC1.11.1.13) and H2O2-producing enzymes such as glyoxal oxidase. Triggered by nitrogen, carbon or sulphur limitations, the expression of these enzymes is active at high oxygen tensions (Kirk et al. 1978; Faison & Kirk 1985; Dosoretz et al. 1990a). A high partial pressure of oxygen has been proposed to help overcome difficulties of oxygen transfer into fungal mycelia, especially caused by the accumulation of extracellular

polysaccharides, to induce the synthesis of the ligninolytic enzymes (Dosoretz et al. 1990a; Michel et al. 1992). However, a high oxygen level may also lead to a rapid decay of the ligninolytic enzymes due to the increased protease activity (Dosoretz et al. 1990a,b). Most studies on the production of the ligninolytic enzymes in liquid cultures of P. chrysosporium have been conducted in pure oxygen or in an oxygen-enriched environment, though various kinds of culture systems have been employed. Rothschild et al. (1995) first demonstrated the formation of LiP and glyoxal oxidase by P. chrysosporium in a non-immersed culture system in an air atmosphere, and later it was proposed that the high oxygen level required for LiP formation by P. chrysosporium can be substituted by the absence of manganese in shallow stationary culture (Rothschild et al. 1999). Zacchi et al. (2000) obtained high LiP activities (200–400 U/l) in submerged agitated liquid culture in air by using cellulose instead of glucose as the carbon source. The effective synthesis of the ligninolytic enzymes under an air condition implies lower costs and greater feasibility, which are essential for enzyme production on a large

324 scale. More efforts on enhancing formation of the ligninolytic enzymes in air are required to facilitate their realistic production. Immobilized culture can improve the availability of oxygen to fungal mycelia by providing a great surface area and enabling efficient mass transfer, and thus may be taken as a possible culture mode for enzyme fermentation in air. By using an immobilized culture system of P. chrysosporium, this work investigated the effects of sizes and amounts of the carriers, medium C/N ratios and fed-batch operation on the formation of the ligninolytic enzymes, and proposed a possible way to realize high enzyme production in air.

Materials and methods Strain and medium P. chrysosporium strain BKM-F-1767 was maintained at 30 C on PDA (200 g potato extract/l, 20 g glucose/l and 20 g agar/l) plates. The culture medium was based on that described by Tien & Kirk (1988), containing 0.02 M acetate buffer (pH 4.4) instead of dimethyl succinate buffer. The glucose concentration and the nitrogen (supplied as diammonium tartrate) concentration were modified as indicated. 1.5 mM veratryl alcohol was introduced from the beginning of cultures and no surfactant was added.

G. Yu et al. Chemicals Veratryl alcohol and nitrilotriacetate (used in medium) were from Fluka (Buchs, Switzerland). The glucose assay kit was from Shanghai Institute of Biological Products (Shanghai, China). All other chemicals used were of analytical grade.

Results Effect of sizes and amounts of the carriers No LiP activity was observed under nitrogen limitation (C/N ratio, expressed as glucose/NHþ 4 , 56/2.2 mM) throughout the cultures, no matter what sizes (0.5 and 1.5 cm) and amounts (0.4, 0.8 and 1.6 g) of the carriers were used (data not shown). Not detected with a less amount (0.4 g and 0.8 g) of the carriers, MnP activity could be measured with 1.6 g carriers corresponding to a non-immersed state, both occurring on day 2, peaking at comparable levels (67 U/l for the 0.5 cm carriers and 93 U/l for the 1.5 cm) on day 4 with two sizes of the carriers (Figure 1). Under carbon limitation (C/N ratio 28/44 mM) in another experiment, LiP activity was also (a) 70

carrier amount

60

Culture conditions

Assays LiP activity was measured as described by Tien & Kirk (1988), with 1 U defined as 1 lmol of veratryl alcohol oxidized to veratraldehyde per min. A molar extinction coefficient of 9300 M)1 cm)1 was used for veratraldehyde. MnP activity was measured as described by Paszczynski et al. (1988) with Mn2+ as the substrate, with an extinction coefficient of 6500 M)1 cm-1 used for the Mn3+-tartaric acid complex. One unit of activity was defined as 1 lmol of Mn2+ oxidized per min. Glucose was measured by using a glucose assay kit.

MnP (U/1)

40 30 20 10 0 -10 0

1

2

3

4 5 Time (day)

6

7

8

(b) 100 carrier amount 0.4 g 0.8 g 1.6 g

80 MnP (U/1)

Cubes of polyurethane foam (average pore diameter 0.031 cm, density 0.039 g/cm3), 0.5 and 1.5 cm per side, respectively, were adopted as the carriers after boiling for 30 min, rinsing three times and drying. Their amounts were organized so that the carriers corresponded to an immersed (0.4 g), a non-immersed (1.6 g) and the critical state (the liquid level was largely equal to the height of the piled cubes) (0.8 g), respectively, in a 250 ml Erlenmeyer flask containing 100 ml medium. The spore concentration after inoculation was 1 · 105 spores/ml. Cultures were incubated in air at 37 C in a rotary shaker with agitation at 160 rev/min with a 2.5-cm-diameter throw. Experiments were carried out in triplicate and results are expressed as the mean values.

0.4 g 0.8 g 1.6 g

50

60 40 20 0 0

1

2

3

4 5 Time (day)

6

7

8

Figure 1. MnP production by immobilized P. chrysosporium in air at a C/N ratio of 56/2.2 mM with different amounts of (a) the 0.5 cm cube and (b) the 1.5 cm cube carriers. Cultures were incubated at 37 C in a rotary shaker with agitation at 160 rev/min. Experiments were performed in triplicate and values are means ± standard deviations.

325

Ligninolytic enzyme production in air

The effect of medium C/N ratio on the production of the ligninolytic enzymes by immobilized P. chrysosporium in air in a non-immersed culture is presented in Figure 2. In contrast with no production of LiP at the C/N ratio of 56/2.2 mM, high LiP activities were detected under carbon limitation conditions at the C/N ratios of 28/44 and 56/44 mM (Figure 2a). The activities occurred on day 2 and day 4, respectively, coinciding with the complete consumption of glucose (Figure 2c), and achieved the maxima, 197 U/l on day 3 at 28/44 mM and 164 U/l on day 5 at 56/44 mM. Though MnP was produced at the three C/N ratios, the peak activity was very low at 28/44 compared with the similar levels at 56/2.2 mM and 56/44 mM (Figure 2b). The formation of MnP also coincided with the depletion of glucose under carbon limitation conditions (Figure 2c).

56/2.2 mM 56/44 mM 28/44 mM

200

LiP (U/1)

Effect of C/N ratio

C/N ratio

(a) 250

150 100 50 0 0

(b)

1

100

2

3

4 5 Time (day)

6

7

8

4 5 Time (day)

6

7

8

C/N ratio 56/2.2 mM 56/44 mM 28/44 mM

80 MnP (U/1)

only found in non-immersed culture states and not in immersed states when adopting the 0.5 cm carriers, and no MnP was produced during the cultures with any carrier amount used at this time (data not shown).

60 40 20

Effect of glucose feeding

0

1

2

3

(c) 12 C/N ratio 56/2.2 mM 56/44 mM 28/44 mM

10 Glucose (g/1)

In non-immersed cultures with 1.6 g 1.5 cm carriers at an initial C/N ratio of 28/44 mM, two strategies of glucose feeding, i.e. restoring the glucose concentration to 2 and 5 g/l, respectively after the complete consumption of glucose, were adopted to check their effect on the formation of the ligninolytic enzymes. Figure 3 depicts the effect of glucose feeding as well as the utilization of glucose. 2 g glucose/l was fed on day 3 and day 5 and depleted within one day, and 5 g/l on day 3 and day 6 and consumed within two days (Figure 3c). Compared with no glucose feeding, an addition of 2 g/l reactivated the synthesis of LiP and MnP immediately, as seen clearly from the elevated enzyme activities on day 4 and day 6 (Figure 3a and b). A replenishment of 5 g/l first decreased the activity of LiP, and then restored it to the approximately original level on day 5 and day 6 with the depletion of added glucose, and in the meantime it improved the MnP production on day 4 and day 7 apparently, as somewhat similar to the effect of the 2 g/l feeding (Figure 3a and b).

0

8 6 4 2 0 0

1

2

3

4

5

6

7

8

Time (day) Figure 2. Production of the ligninolytic enzymes by P. chrysosporium in air in non-immersed culture with 1.6 g 1.5 cm carriers at different C/N ratios. (a) LiP, (b) MnP, (c) glucose. Cultures were incubated at 37 C in a rotary shaker with agitation at 160 rev/min. Experiments were performed in triplicate and values are means ± standard deviations.

Discussion The production of the ligninolytic enzymes by P. chrysosporium has rarely been reported in liquid cultures in an air environment (Rothschild et al. 1995; 1999; Zacchi et al. 2000). The present work showed some detailed characteristics of the production of the ligninolytic enzymes by immobilized P. chrysosporium in air concerning the effects of sizes and amounts of the carriers, medium C/N ratios and glucose feeding. The results obtained imply a great possibility for the

accomplishment of high ligninolytic enzyme production by P. chrysosporium in liquid culture in an air atmosphere. It was shown in this work that the carrier size did not cause distinct effects on the formation of the ligninolytic enzymes by P. chrysosporium immobilized on a porous support, while the carrier amount played a critical role. Non-immersed culture (Dosoretz et al. 1993), in which a higher quantity of the carriers were used for the

326

G. Yu et al.

(a) 250 glucose feeding no to 2 g/1 to 5 g/1

200

LiP (U/1)

150 100 50 0 0

1

2

3

4 5 Time (day)

6

8

9

glucose feeding no to 2 g/1 to 5 g/1

(b) 50 40 MnP (U/1)

7

30 20 10 0 0

1

2

3

4 5 Time (day)

6

7

8

9

(c) 5

glucose feeding no to 2 g/1 to 5 g/1

Glucose (g/1)

4 3 2 1 0 0

1

2

3

4 5 Time (day)

6

7

8

9

Figure 3. Production of the ligninolytic enzymes by P. chrysosporium in air in non-immersed culture with 1.6 g 1.5 cm carriers at an initial C/N ratio of 28/44 mM with different glucose feeding strategies. (a) LiP, (b) MnP, (c) glucose. Cultures were incubated at 37 C in a rotary shaker with agitation at 160 rev/min. Experiments were performed in triplicate and values are means ± standard deviations.

exposure of the immobilized mycelia to air, was demonstrated to be advantageous for enzyme production. Resembling the wild state of cell growth to an extent, a non-immersed culture system may strongly enhance the oxygen availability to the fungi by generating much more air contact area and thus can satisfy the high oxygen demand for the expression of the ligninolytic enzymes. In the meantime, good mass transfer of liquid

nutrients may be guaranteed with agitation close to that in submerged culture. Somewhat similar to semi-solid state culture (Rodriguez et al. 1997; 1998), non-immersed culture is usually characterized by holding a far larger proportion of free liquid medium which can also enable an easy harvest of the fermented product. Medium C/N ratio exerted different influence on the production of LiP and MnP in non-immersed culture of P. chrysosporium in air. LiP was synthesized under carbon limitation and not under nitrogen limitation, indicating that a low C/N ratio was advantageous for LiP production. On the other hand, the significant production under nitrogen limitation or high carbon conditions (56 mM/44 mM) and the repressed formation at a low carbon concentration (28 mM/44 mM) suggested that the MnP synthesis preferred a high carbon level. These results were consistent with some observations by Rothschild et al. (1995). The discrepancy between the two enzymes may result from the different regulatory mechanisms in the expression of LiP and MnP, which were closely related to the physiological effects caused by carbon and nitrogen starvation, such as nitrogen or carbon catabolic repression. In addition, the possible formation of extracellular polysaccharides under nitrogen limitation may result in lower oxygen availability and affect enzyme synthesis to some degree (Rothschild et al. 1995). It is deduced that C/N ratio control can be taken as a measure to regulate the relative production of LiP and MnP in non-immersed culture in air. Enzyme instability seriously damaged the high production of the ligninolytic enzymes, for which protease was thought to be responsible (Dosoretz et al. 1990b,c). The production of protease was closely related to glucose metabolism. The addition of a suitable amount of glucose upon its complete depletion may prevent cell autolysis, inhibit protease activity (Dosoretz et al. 1990b; Chen et al. 1992), and thus stabilize and enhance the production of the ligninolytic enzymes. An excess supplement, however, may greatly destroy the carbon starvation condition required for enzyme expression and reduce its production during a short time as seen for LiP with an addition of 5 g glucose/l. The improvement of MnP activity at this time (5 g/l addition) reflected its different response to the same carbon level from LiP as indicated above. An optimal glucose feeding strategy, probably one continuous feeding mode, simultaneously maintaining carbon limitation and inhibiting protease activity, should exist for the maximum and stable production of LiP and MnP. In conclusion, this paper has presented some characteristics of the production of the ligninolytic enzymes by immobilized P. chrysosporium in air. Non-immersed liquid culture under carbon limitation reinforced by a suitable glucose feeding strategy may be one promising way to accomplish high production of the ligninolytic enzymes by P. chrysosporium in air. Further optimization of the culture parameters as well as scale up in a suitable bioreactor is required for its practice.

Ligninolytic enzyme production in air Acknowledgments This work was supported by a grant from the National High Technology Research and Development Program of China (863 Program) (No. 2002AA649100). References Barr, D.P. & Aust, S.D. 1994 Mechanisms white rot fungi use to degrade pollutants. Environmental Science and Technology 28, 78A–87A. Bumpus, J.A., Tien, M., Wright, D. & Aust, S.D. 1985 Oxidation of persistent environmental pollutants by a white rot fungus. Science 228, 1434–1436. Chen, A.H.C., Dosoretz, C.G. & Grethlein, H.E. 1992 Strategy for the production and stabilization of lignin peroxidase from Phanerochaete chrysosporium in air-lift and stirred tank bioreactors. In Frontiers in Bioprocessing II, eds. Todd, P., Sikdar, S.K. & Bier, M. pp. 181–187. Washington, DC: American Chemical Society. ISBN 0-8412-2181-2. Dosoretz, C.G., Chen, A.H.C. & Grethlein, H.E. 1990a Effect of oxygenation conditions on submerged cultures of Phanerochaete chrysosporium. Applied Microbiology and Biotechnology 34, 131–137. Dosoretz, C.G., Chen, H.C. & Grethlein, H.E. 1990b Effect of environmental conditions on extracellular protease activity in ligninolytic cultures of Phanerochaete chrysosporium. Applied and Environmental Microbiology 56, 395–400. Dosoretz, C.G., Dass, S.B., Reddy, C.A. & Grethlein, H.E. 1990c Protease-mediated degradation of lignin peroxidase in liquid cultures of Phanerochaete chrysosporium. Applied and Environmental Microbiology 56, 3429–3434. Dosoretz, C.G., Rothschild, N. & Hadar, Y. 1993 Overproduction of lignin peroxidase by Phanerochaete chrysosporium (BKM-F-1767) under nonlimiting nutrient conditions. Applied and Environmental Microbiology 59, 1919–1926.

327 Faison, B.D. & Kirk, T.K. 1985 Factors involved in the regulation of a ligninase activity in Phanerochaete chrysosporium. Applied and Environmental Microbiology 49, 299–304. Kirk, T.K. & Farrell, R.L. 1987 Enzymatic ‘‘combustion’’: the microbial degradation of lignin. Annual Review of Microbiology 41, 465–505. Kirk, T.K., Schultz, E., Connors, W.J., Lorenz, L.F. & Zeikus, J.G. 1978 Influence of culture parameters on lignin metabolism by Phanerochaete chrysosporium. Archives of Microbiology 117, 277– 285. Michel, F.C., Grulke, E.A. & Reddy, C.A. 1992 Determination of the respiration kinetics for mycelial pellets of Phanerochaete chrysosporium. Applied and Environmental Microbiology 58, 1740–1745. Paszczynski, A., Crawford, R.L. & Huynh, V.-B. 1988 Manganese peroxidase of Phanerochaete chrysosporium: purification. Methods in Enzymology 161, 264–270. Rodriguez, C.S., Santoro, R., Cameselle, C. & Sanroman, A. 1997 Laccase production in semi-solid cultures of Phanerochaete chrysosporium. Biotechnology Letters 19, 995–998. Rodriguez, S., Santoro, R., Cameselle, C. & Sanroman, A. 1998 Effect of the different parts of the corn cob employed as a carrier on ligninolytic activity in solid state cultures by P. chrysosporium. Bioprocess Engineering 18, 251–255. Rothschild, N., Hadar, Y. & Dosoretz, C. 1995 Ligninolytic system formation by Phanerochaete chrysosporium in air. Applied and Environmental Microbiology 61, 1833–1838. Rothschild, N., Levkowitz, A., Hadar, Y. & Dosoretz, C.G. 1999 Manganese deficiency can replace high oxygen levels needed for lignin peroxidase formation by Phanerochaete chrysosporium. Applied and Environmental Microbiology 65, 483–488. Tien, M. & Kirk, T.K. 1988 Lignin peroxidase of Phanerochaete chrysosporium. Methods in Enzymology 161, 238–249. Zacchi, L., Burla, G., Zuolong, D. & Harvey, P.J. 2000 Metabolism of cellulose by Phanerochaete chrysosporium in continuously agitated culture is associated with enhanced production of lignin peroxidase. Journal of Biotechnology 78, 185–192.

Ó Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 329–336 DOI 10.1007/s11274-004-3907-4

Use of cereals as basal medium for the formulation of alternative culture media for fungi A.O. Adesemoye1,* and C.O. Adedire2 1 Department of Microbiology, Adekunle Ajasin University, P.M.B. 001, Akungba-Akoko, Ondo State, Nigeria 2 Department of Biology, Federal University of Technology, P.M.B. 704, Akure, Ondo State, Nigeria * Author for correspondence: Tel.:+234-803-718-6133, E-mail: [email protected] Received 23 January 2004; accepted 17 September 2004

Keywords: Cereals, culture media, fungi, growth, potato dextrose agar

Summary The feasibility of developing alternative media to different culture media particularly potato dextrose agar was assessed using local cereal species as the basal media. Three cereal meal extracts – corn, sorghum and millet – were prepared, using them as substitute for the potato in potato dextrose agar. Potato dextrose agar (PDA) was the standard set up with which the performances of the formulated media were compared. Eight genera of fungi (Aspergillus niger, Fusarium moniliforme, Penicillium sp., Cercospora sp., Curvularia palescens, Botryodiplopodia sp., Rhizopus sp. and Rhodotorula rubra) were isolated and pure cultures of each species aseptically inoculated onto the three different formulated media including PDA and allowed to grow. Their growths were measured at 24, 48, 72, and 96 h after inoculation, using diameter of growth as an index. The set up was repeated thrice for each species on the three formulated media and the control (PDA). Growth of all the fungal species were observed to be about the same or sometimes better in the formulated media relative to those on the standard set up, except for Rhodotorula rubra. The radius of growth of F. moniliforme had an average of 15 + 0.58 mm on corn-dextrose agar relative to 12 mm on PDA at 96 h while Cercospora sp. measured 30 + 0.58 mm on millet-meal dextrose agar relative to 37 + 1.16 mm at 48 h. Botryodiplopodia sp. grew through the whole diameter of the plate (covering the total length of the radius of 45 mm) in both sorghum-meal and PDA at 96 h.

Introduction Microbiological studies depend on the ability to grow and maintain microorganisms under laboratory conditions by providing suitable culture media that offer favourable environmental conditions (Prescott et al. 2002). The conditions needed for growth include good carbon source, nitrogen source such as protein, availability of enzymes, vitamins, mineral elements such as phosphorus and sulphur, suitable pH, suitable temperature, relative humidity, inorganic salt and water. The knowledge of the conditions is useful in the control of the growth of microbes that cause diseases and food spoilage but also in the effort to encourage the growth of helpful microbes and those to be studied (Tortora et al. 1997; Cappuccino & Sherman 1998). Efforts to provide suitable conditions for microorganisms, started with van Leeuwenhoek in 1675 using a fluid obtained by soaking peppercorn in water. Solid medium was developed thereafter by Koch in 1881 when he reported the use of boiled potatoes, sliced with a flame-sterilized knife in culturing bacteria (Khaled et al. 1996; Olutiola et al. 2000; Prescott et al. 2002). One of the most important agar media in modem days is potato

dextrose agar (PDA), a general purpose agar. Ali et al. (2002) reported that it comprised 200 g diced potato/l, 20 g dextrose (glucose)/l and 15 g agar/1. Glucose is the source of carbon and energy. However, information on recent studies in media formulation using locally available materials are scanty. Such studies will therefore continue for some time to come, especially in developing countries such as Nigeria, where research in microbiology is hindered by the high cost and scarcity of culture media (Poopathi et al. 2002). This study therefore investigates the possibility of using locally available cereals as substitutes for potato in PDA. Characteristics of cereals such as suitability for the growth of fungi, richness in protein, resistance to drought etc informed the choice in this study (Blum & Sullivan 1974; Brimble et al. 1982; Roni et al. 1996).

Materials and methods Proximate analysis of cereal samples The three cereal samples, namely maize (corn), millet and sorghum used were purchased from Erekesan

330

A.O. Adesemoye and C.O. Adedire

Market in Akure, Ondo State, Nigeria. Whole grains were pulverised into fine powder in a large and dry mortar. Proximate analysis of each of the cereal samples were carried out according to AOAC (1980) for ash, crude fibre, moisture, fats and protein. Carbohydrate content was calculated by subtraction method. Media formulation Three different media were formulated, namely, cornmeal dextrose agar (CDA), millet-meal dextrose agar (MDA) and sorghum-meal dextrose agar (SDA). The method first recommended by Martins et al. (1955) but modified by Olutiola et al. (2000) to formulate corn– meal dextrose agar was used for all the three samples. Approximately 6 g of the pulverized yellow maize was measured into a flat bottom flask, 160 ml of clean water was added and heated in a water bath for 1 h. It was filtered through a muslin cloth. Then 6 g of dextrose and 4 g of agar–agar were added to the filtrate. The volume of the mixture was made to 200 ml with distilled water and heated on a hot plate with steady stirring until the solution boiled. The stirring was meant to achieve homogeneity. The resultant suspension was sterilized in the autoclave for about 15 min at 121 °C. The content of the flask was thereafter poured aseptically into eight pre-sterilized Petri dishes. The media were allowed to cool and solidify before inoculation. The procedure was repeated in formulating millet-meal dextrose agar and sorghummeal dextrose agar by substituting corn with millet and sorghum, respectively.

Test organisms The test organisms were chosen to cover fungi of agricultural and medical importance. Except where otherwise stated, eight fungal genera used were isolated from the following sources using malt extract agar. Aspergillus niger (onion), Fusarium moniliforme (carrot), Penicillium sp. (orange), Cercospora sp. (yam), Curvularia palescens (maize), Botryodiplopodia sp. (cassava), Rhizopus sp. (moist bread) and Rhodotorula rubra (isolated from man but the isolate was obtained from the laboratory of the State Specialist Hospital, Akure, Ondo State). Inocula were drawn from the pure cultures of the isolates just before 96th h of growth and aseptically inoculated onto all the formulated media and the control (PDA). The inocula were taken at that time to ensure that growth was still high at logarithmic phase when cells would have uniform physiological characteristics (Salmon et al. 1989; Nwachukwu & Akpata 2003). Inoculation of media

In preparing the control medium, 6.8 g of potato dextrose agar was mixed with 200 ml of water, heated and stirred until boiling before autoclaving. The physico-chemical properties of the PDA was compared with those of the formulated media and presented in Table 1. The pH test readings showed that all the formulated media were more alkaline than PDA and all contained particles. Corn-meal dextrose agar contained the least particles because it was easier to grind more finely due to its low fibre content relative to the two other cereals.

The suitability of the formulated media was estimated by culturing the isolated fungal species on them through the following procedure. The working table was maintained under aseptic condition by disinfecting with 70% alcohol (i.e 7:3 by vol. of alcohol/vol. of water). Cork borer (No. 1) and platinum wire inoculating loop were also dipped into the alcohol and flamed. The cork borer, with size approximately 0.37 cm in diameter was used to bore holes on the pure cultures of organisms. It was resterilized again before being used to bore holes in successive pure cultures. The mycelia agar plugs were then removed with the sterilized inoculating loop and transferred top down onto the centre of the formulated media. Each of the eight organisms from the pure cultures was inoculated on the plates of formulated CDA, MDA, SDA and PDA in like manner. The plates were then incubated at 31 °C inside the incubator for 96 h. The growth of the organisms were evaluated by Colony Size Method (Fawole & Oso 1995). The radius of growth was measured at 24 hourly intervals using a Venier caliper. The whole inoculation process was repeated thrice for each sample with each of the test organisms. The mean value of readings and the standard deviation were then estimated.

Table 1. Comparison of the physico-chemical properties of the formulated media and PDA

Results

Characteristics of PDA compared with formulated media

Medium

pH Value

Colour

Particles

Clarity

PDA CDA MDA SDA

5.90 7.20 7.30 7.35

Yellow Light yellow Cream Pale yellow

) + ++ ++

CC CC C C

Key: – )absent; + – present; ++ – more present; C – turbid; CC – clear; PDA – potato dextrose agar; CDA – corn dextrose agar; MDA – millet dextrose agar; SDA – sorghum dextrose agar.

The physico-chemical characteristics of the formulated media and potato dextrose agar (PDA) showed that sorghum dextrose agar (SDA) had the highest pH value (7.35) while PDA had the lowest pH value (5.90). Particles were most in SDA followed by millet dextrose agar (MDA) and least in corn dextrose agar (CDA) but absent in PDA. The media showed the following properties in terms of colour and clarity: PDA (yellow

331

Alternative culture media for fungi and clear), CDA (light yellow and clear), MDA (cream turbid) and SDA (pale yellow and turbid).

(2.00 ± 0.45) (3.60 ± 0.45)

Proximate analysis of corn, millet and sorghum used in the formulation

Growth of the selected fungi on the formulated media and the control

The results of the proximate analyses of the cereals are presented in Figure 1. Corn showed the highest value in moisture content (10.36 ± 0.29) but least in ash content (1.36 ± 0.00); millet was highest in carbohydrate (76.85 ± 0.70), fat (4.50 ± 0.05) and ash contents (2.00 ± 0.45) but least in moisture content (4.83 ± 0.00) and sorghum showed the highest value for protein (13.61 ± 0.50) and fibre content

Figures 2–9 are graphical representations of the data obtained from culturing eight test fungi on the three formulated alternative culture media and PDA (radius of growth taken every 24 through 96 h). The figures are in the following order: Figure 2 (Aspergillus niger), Figure 3 (F. moniliforme) Figure 4 (Penicillium sp.), Figure 5 (Cercospora sp), Figure 6 (C. palescens), Figure 7 (Botryodiplopodia sp.), Figure 8 (Rhizopus sp.)

Figure 1. Proximate analysis of corn, millet and sorghum used in media formulation.

Figure 2. The growth of A. niger on the formulated media and PDA through 96 h.

but

least

values

in

fat

content

332

A.O. Adesemoye and C.O. Adedire

Figure 3. The growth of F. moniliforme on the formulated media and PDA through 96 h.

Figure 4. The growth of Penicillum sp. on the formulated media and PDA through 96 h.

and Figure 9 (R. rubra). The growth of the test organisms on the media are compared below. Comparison of growth on corn-meal dextrose agar with potato dextrose agar Observation at the end of the first 24 h shows that all the organisms have adjusted to the culture environment and have started growing on all the formulated media and PDA. The only exception was R. rubra, which showed growth within 24 h only on CDA (Figure 9). This

period could be said to be within the lag phase of R. rubra on all the other media except for CDA where it has passed that stage. All the other organisms showed comparable growth on PDA and CDA. There was a higher rate of growth for Cercospora sp. on PDA (45.0 ± 0.29) by the 72nd h while on CDA, it was 34.0 ± 1.16 mm (Figure 5). A. niger, though grew faster on PDA than CDA within the first 24 h (9.0 ± 0.58 and 5.0 ± 0.0 respectively), the rate of growth within 48th and 72nd h was higher on CDA (Figure 2).

333

Alternative culture media for fungi

Figure 5. The growth of Cercospora sp. on the formulated media and PDA through 96 h.

Figure 6. The growth of C. Palescens sp. on the formulated media and PDA through 96 h.

Comparison of growth on millet-meal dextrose agar with PDA F. moniliforme showed high rate of growth within the first 24 h on PDA relative to MDA (Figure 3). However, from the 48th h to the old culture at 96 h, the growth of the organism was better on MDA than for PDA (20.0 ± 1.16 and 12.0 ± 0 respectively). The growth rate of Curvularia sp. was higher on the MDA relative to that on PDA throughout the growth phases (Figure 6). Observations on Penicillium sp. was not significantly different on MDA and PDA (Figure 4), at

one time it was slightly higher on MDA (24 h), at another time it was slightly higher for PDA (48 h).

Comparison of growth on sorghum-meal dextrose agar with PDA Botryodiplopodia sp. grew faster on SDA between 24 and 48 h than on PDA (Figure 7), in like manner, F. moniliforme grew better up to the 96th h on SDA (21.0 ± 0.58) relative to (12.0 ± 0) on PDA (Figure 3). However, the growth of A. niger, Penicillum sp.,

334

A.O. Adesemoye and C.O. Adedire

Figure 7. The growth of Botryodiplopodia sp. on the formulated media and PDA through 96 h.

Figure 8. The growth of Rhizopus sp. on the formulated media and PDA through 96 h.

C. palescens, and R. rubra were better on PDA than on SDA at the end of 96 h.

Performance comparison of each test organism on the four media The growth of the eight test organisms on all the formulated media and PDA were presented in Figure 2–9 to enhance easy comparison. A. niger for instance grew averagely well on all the media and when a graph of growth radius was plotted against incubation time, the graph reflected features similar to the logarithmic phase

(Nwachukwu & Akpata 2003) except for PDA. The growth between 48 and 72 h was insignificant on the plate but the reason for that could not be deduced. The result of the first set of experiments for PDA arouse curiosity but after the other two replicates, the suspicion of error was proved wrong as the standard deviation of (± 1.16) shows. Other organisms performed well on all the media with most showing a minimum average growth of 40 mm radius in 72 h and 45 mm radius in 96 h. F. moniliforme showed its characteristic attractive red colour on the agar but only showed a maximum growth of 21.0 ± 0.58 in 96 h recorded on SDA. R. rubra responded least on all the media among all the organism.

335

Alternative culture media for fungi

Figure 9. The growth of R. rubra on the formulated media and PDA through 96 h.

Discussion Development of media using locally available materials has become imperative (Poopathi et al. 2002), since conventional media are either in short supply or very expensive in developing countries, particularly Nigeria. The high price can be traced to the poor economy, weak purchasing power of the local currency, hence forcing microbiologists to abandon much ongoing research. It was not surprising that the test microorganisms in this study performed well on the formulated media considering that cereals easily support the growth of a wide variety of flora without any additive. For instance maize naturally carries up to 43 pathogens and rice 50, the dominant organism being fungi (Kommedhl & Lang 1971; Brimble et al. 1982; Campbell 1985; Larran et al. 2002). Fungi comprise almost half of the known 157,000 species of microorganisms and show substantial variation in nutritional and biochemical attributes (Hawksworth 1994; DalBello et al. 2002). Avoidance of errors in inoculation or enumeration (Adams & Moss 1999; Boulter et al. 2002) and the impact of hydrogen ion concentration (Fawole & Oso 1995; Alcano 1997) are very important to the success of this study. There is need to emphasize that the growth of Rhizopus sp. and Rhodotorula rubra in this study may not be good bases for estimating the suitability of the formulated media because they were at the extremes relative to other test organisms. Owing to the easy growth of Rhizopus sp. on almost every medium, some microbiologists have suggested that it is not a very good test organism. Species in the genus Rhodotorula are commensals in their natural environment and in humans (Po-Ren et al. 2003), this could have accounted for the poor growth on the media. Corn-meal dextrose agar was observed to be the best among all the formulated media.

On the average it supported the growth of all the test organisms better than the other formulated media. Corn-meal dextrose agar is viewed with high prospects, more so that corn (maize) is readily available across the globe, presenting it as a base for teaching and further research with the possibility of large scale production at economic benefits. Acknowledgements The authors are grateful to Dr E.O. Igbosuah, Department of Botany & Microbiology, University of Lagos, Nigeria, for her useful criticisms of the revised manuscript and Prof. Femi Mimiko, Adekunle Ajasin University, Akungba-Akoko, Nigeria for his assistance towards the initial manuscript.

References Adams, M.R. & Moss, M.O. 1999 Food Microbiology, pp. 303–310. Royal, Society of Chemistry. ISBN 0-85404-509-0. Alcano, I.E. 1997 Fundamentals of Microbiology, 5th edn, pp. 727. New York, U.S.A: Benjamin/Cummings Publishing Co. ISBN 0-8053-0532-7. Ali, S., Ikram-ul-Haq, Qadeer, M.A. & Iqbal, J. 2002 Production of citric acid by Aspergillus niger using cane molasses in a stirred fermentor. Electronic Journal of Biotechnology 5, 1. AOAC (Association of Official Analytical Chemists) 1980 Official Methods of Analytical Chemists, 12th edn. Washington D.C: AOAC. Blum, A. & Sullivan, Y. 1974 Leaf water potential and stomatal activity in sorghum as influenced by soil moisture stress. Israel Journal of Botany 23, 1–2. Boulter, J.I., Trevors, J.T. & Boland, G.J. 2002 Microbial studies of compost: bacteria identification, and their potential for turfgrass pathogen suppression. World Journal of Microbiology and Biotechnology 18, 661–671.

336 Brimble, L.J., Williams, S. & Bond, G. 1982 Intermediate Botany, pp. 38–346. London: Macmillan Press Ltd. ISBN 0-333-08403-9. Campbell, R. 1985 Plant Microbiology. pp. 42–43. London: Edwards Arnold Publishers Ltd. ISBN 0-7131-2892-5. Cappuccino, J.G. & Sherman, N. 1998 Microbiology: A Laboratory Manual, 5th edn pp. 83–84. California, USA: Benjamin/Cummings Publishing Co. ISBN 0-8053-7646-1. Dal-Bello, G.M., Monaco, G.I. & Simon, M.R. 2002 Biological control of seedling blight of wheat caused by Fusarium graminearum with beneficial rhizosphere microorganisms. World Journal of Microbiology and Biotechnology 18, 627–628. Fawole, M.O. & Oso, B.A. 1995 Laboratory Manual of Microbiology, pp. 49–52. Ibadan, Nigeria: Supreme Books Ltd. ISBN 978-246032-X. Hawksworth, D.L. 1994 Biodiversity in microorganisms and its role in ecosystem function. Biodiversity and Global Change 3, 85–93. Khaled, K.A., Mahmoud, A.H. & Roger, J.M. 1996 Nisin resistance distinguishes Mycoplasma spp from Acholeplasma spp and provides a basis for selective growth media. Applied and Environmental Microbiology 52, 3107–3111. Kommedhl, T. & Lang, D.S. 1971 Seedling with corn from kernels naturally infected with Helminthosporum maydis in Minnesoba. Plant Disease 55, 371–372. Larran, S., Perello, A., Simon, M.R. & Moreno, V. 2002 Isolation and analysis of endophytic microorganisms in wheat (Triticum aestivum L.) leaves. World Journal of Microbiology and Biotechnology 18, 683–686. Levy, J., Campbell & Blackburn. 1973 Introductory Microbiology. International edn, 111. pp. New York, USA: John Wiley and sons Inc. ISBN 0-471-53155-3.

A.O. Adesemoye and C.O. Adedire Nwachukwu, S.C.U. & Akpata, T.V.I. 2003 Principles of Quantitative Microbiology, pp. 30–37. Lagos, Nigeria: University of Lagos Press. ISBN 978-017-636-5. Olutiola, P.O., Famurewa, O. & Sonntag, H.G. 2000 An Introduction to General Microbiology: A Practical Approach. Reprinted edn. pp. 50, 196, 223. Heidelberg, Germany: Hygiene-Institute Der Universitat Heiderberg. ISBN 3-89426-042-4. Pelczar, M.J., Chan, E.C.S & Kreig, N.R. 1986 Microbiology, 5th edn. pp. 104–107. London: McGraw Hill Publishers. ISBN 0-07049234-4. Poopathi, S., Kumar, K.A., Kabilan, L. & Sekar, V. 2002 Development of low-cost media for the culture of mosquito larvicides, Bacillus sphaericus and Bacillus thuringiensis serovar. Israelensis. World Journal of Microbiology and Biotechnology 18, 209–216. Po-Ren, H., Lee-Jene, Shen-Wu, H. & Kwen-Tay, L. 2003 Catheterrelated sepsis due to Rhodotorula glutinis. Clinical Microbiology 41, 857–859. Prescott, L.M., Harley, J.P. & Klein, D.A. 2002 Microbiology, 5th edn. pp. 105–106. London: McGraw Hill Publishers. ISBN 0-07232041-9. Roni, S., Nachman, P., Osnat, E., Mazal, M., Anait, M. & Raffael, S. 1996 Detection of aflatoxin molds in grains by PCR. Applied and Environmental Microbiology 62, 3271–3276. Salmon, J., Pinon, R. & Gancedo, C. 1989 Isolation and characterisation of mutants of Saccharomyces cerevisae able to sporulate in the presence of glucose. General Microbiology 135, 203–209. Tortora, G.J., Funke, B.R. & Case, C.L. 1997 Microbiology: An Introduction. 6th edn. pp. 154. California, USA: Benjamin/Cummings Publishing Co. ISBN 0-8053-8446-4.


Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 337–343 DOI 10.1007/s11274-004-2609-2

Investigation of the active site of the extracellular b-D -glucosidase from Aspergillus carbonarius Szilvia Ja¨ger and La´szlo´ Kiss* Institute of Biochemistry, Faculty of Sciences, University of Debrecen, P.O. Box 55, H-4010 Debrecen, Hungary *Author for correspondence: Tel.: +36-52-512900-2734, Fax: +36-52-512913, E-mail: [email protected] Received 3 February 2004; accepted 10 August 2004

Keywords: Affinity label, Aspergillus carbonarius, chemical modification, b-D -glucosidase, N-bromoacetyl-b-D glucopyranosylamine

Summary The catalytic amino acid residues of the extracellular b-D -glucosidase (b-D -glucoside glucohydrolase, EC 3.2.1.21) from Aspergillus carbonarius were investigated. The pH dependence curves gave apparent pK values of 2.8 and 5.93 for the free enzyme, and 2.24 and 6.14 for the enzyme–substrate complex using p-nitrophenyl-b-D -glucoside as substrate. Carbodiimide- and Woodward reagent K-mediated chemical modifications suggested that a carboxylate residue, located in the active centre, was fundamental in the catalysis. The pH dependence of inactivation revealed the involvement of a group with pK value of 4.61 in the modification reaction, proving that a carboxylate residue was modified. The A. carbonarius b-glucosidase was irreversibly inactivated by N-bromoacetyl-b-D -glucopyranosylamine. The active site specificity of the inactivation was proved by using the competitive inhibitor p-nitrophenyl-1-thio-b-D -glucopyranoside. pH Dependence studies of inactivation revealed that modification by N-bromoacetyl-b-D -glucopyranosylamine could be directed toward the carboxylate group acting as the catalytic nucleophile, as in the case of the carbodiimide and Woodward reagent K modifications. Abbreviations: pATP-Glc – p-aminophenyl-1-thio-b-D -glucopyranoside; EDAC – 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide; NBAGA – N-bromoacetyl-b-D -glucopyranosylamine; pNP-Glc, p-nitrophenyl-b-D -glucopyranoside; pNTP-Glc – p-nitrophenyl-1-thio-b-D -glucopyranoside; WRK – Woodward’s reagent K (N-ethyl-5-phenylisoxazolium-30 -sulphonate)

Introduction Current research on b-glucosidases has significant scientific, medical and economic implications (Esen 1993; Izstein & Colman 1996; Taylor 1996). Enzymatic hydrolysis of lignocellulose materials involves the interaction of three principal components of the cellulase enzyme system: endo-1,4-b-D -glucanase (EC 3.2.1.4), exo-1,4-b-D -glucanase (EC 3.2.1.91) and 1,4-b-D -glucosidase. In fungi and bacteria b-glucosidases are involved in the hydrolysis of cellobiose into two molecules of glucose. It is interesting that Trichoderma species, seemingly the best sources of solubilizing activities, are poor producers of b-glucosidases. On the other hand, Aspergillus species have been shown to be better producers of b-glucosidases (Bothast & Saha 1997). The cellulolytic b-glucosidases are retaining enzymes which use a double displacement mechanism to hydrolyse their substrates. These enzymes operate via transition states with substantial oxocarbenium ion

character and form covalent glycosyl–enzyme intermediates. The hydrolysis requires two catalytic groups. One of them functions as a general acid and general base while the other acts as a nucleophile and a leaving group. Asp and Glu residues are commonly found to be catalytic in glycosyl hydrolases, either as proton donors in their protonated form or as nucleophile or oxocarbonium–stabilizing agents in their charged form (Baird et al. 1990; Sinnott 1990; Clarke et al. 1993). Understanding the mechanisms by which these enzymes hydrolyse glucosidic bonds has been the focus of much research. Investigations have involved sequencing glucosidase genes (Hazlewood et al. 1988; Liebl et al. 1994; Dan et al. 2000), kinetic analysis of mutants (Lawson et al. 1998; Ly & Withers 1999; Li et al. 2001), studies on the effect of pH on the enzyme kinetics, inhibition studies and selective chemical modification (Lundblad & Noyes 1985; Keresztessy et al. 1994a; Skoubas & Georgatsos 1997). An alternative approach is the application of irreversible inhibitors, like the

338 mechanism–based inhibitors (Withers et al. 1988, 1990) and the affinity labels. Examples of affinity labels include glycosyl epoxide derivatives (Legler 1990), glycosyl isothiocyanates (Shulman et al. 1976) and N-bromoacetyl-glycosylamines. The successful application of N-bromoacetyl-glycosylamines as affinity labels has already been proved in the case of several glycosidases (Black et al. 1993; Keresztessy et al. 1994b; Tull et al. 1996; Kiss et al. 2002). We have already reported the production, purification and biochemical properties of the b-glucosidase from Aspergillus carbonarius (Ja¨ger et al. 2001). The present paper describes both kinetic and chemical modification studies that provide experimental evidence for the essential role of carboxyl groups in the mechanism of action of the b-glucosidase produced by Aspergillus carbonarius.

Materials and methods Enzyme purification and assays b-D -Glucosidase from Aspergillus carbonarius was produced and purified as described in our previous paper (Ja¨ger et al. 2001). The b-D -glucosidase activity was assayed by using 100 ll of 5 mM pNP-Glc2 as substrate in 1 ml final volume of 20 mM Na–acetate buffer (pH 4.0), for 10 min at 50 C (unless otherwise indicated) as described earlier (Ja¨ger et al. 2001). Inhibition studies In the inhibition experiments the reaction rate was determined in the presence of different inhibitors such as glucose, glucono(1–5)lactone, p-nitrophenyl-1-thio-b-D glucopyranoside (pNTP-Glc) and p-aminophenyl-1thio-b-D -glucopyranoside (pATP-Glc), respectively as described above. The reactions were performed at 6 different inhibitor concentrations between 0.5 and 3 Ki and 4 different substrate concentrations in the range of 0.5–3 Km. Kinetic and pH dependence studies The maximal velocity (Vmax) and the Michaelis constant (Km) for the hydrolysis of pNP-Glc were determined at 6 different substrate concentrations, which were selected in the range of 0.5–5 Km, at 9 different pH levels between pH 2.5 and 6.5 in 20 mM citrate phosphate buffer. Chemical modification of carboxylate groups Selective modification of exposed carboxyl groups of b-glucosidase was performed using the water-soluble carbodiimide, EDAC, in the presence of nucleophile glycine methyl ester and Woodward’s reagent K (Lundblad & Noyes 1985).

S. Ja¨ger and L. Kiss In experiments employing EDAC, b-glucosidase (40 lg) was incubated in 2 ml water together with 7– 40 mM EDAC and 0.5 mM glycine methyl ester at pH 4.75, adjusted manually with HCl, at 25 C for kinetic investigations. To estimate residual hydrolytic activities, aliquots were withdrawn and added to 1 ml of 20 mM Na–acetate buffer, pH 4.0, containing 4.2 mM pNP-Glc (6 Km) as substrate and incubating it at 50 C for 10 min. Kinetic constants of inactivation were calculated as described by Keresztessy et al. (1994a). In a set of parallel experiments the enzyme was treated with 30 mM EDAC and 0.5 mM glycine methyl ester in the presence of 10 mM (5 Ki) pNTP-Glc as a competitive inhibitor. Aliquots were assayed periodically for residual activity against 4.2 mM pNP-Glc in 1 ml of 20 mM Na–acetate buffer for 10 min at 50 C. In pH dependence studies of the inactivation by 30 mM EDAC and 0.5 mM glycine methyl ester, 6 different pH values between 3.5 and 6.5 were applied. In the WRK-mediated inactivation b-glucosidase (43 lg) was treated with 20–50 mM WRK in 2.5 ml of 20 mM Na–acetate buffer, pH 5.0, at 50 C. In order to determine the residual activity of the modified enzyme, aliquots were transferred to tubes, each containing 1 ml of 20 mM Na–acetate buffer, pH 4.0, and 4.2 mM pNP-Glc (6 Km) as substrate and were incubated at 50 C for 10 min. The kinetics of inactivation was evaluated as described by Keresztessy et al. (1994b). To analyse protection by a competitive inhibitor, 20 mM (10 Ki) pNTP-Glc was present during the incubation of the enzyme with 25 mM WRK. To estimate the change in reaction kinetic parameters (Vmax and Km) during modification, 256 lg enzyme was treated with 50 mM WRK in 3 ml of 20 mM Na– acetate buffer, pH 5.0. Samples were removed and were subjected to gel filtration on a Sephadex G-25 column (1 · 5 cm), equilibrated with 20 mM Na–acetate buffer (pH 5.0), to remove the excess reagent. Kinetic constants of the hydrolysis of pNP-Glc were determined using 6 different substrate concentration in the range of 0.5–5 Km.

Inactivation studies with N-bromoacetylb-D -glucopyranosylamine N-Bromoacetyl-b-D -glucopyranosylamine was synthesized in our laboratory according to Thomas (1977). Inactivation of the enzyme (85 lg) was carried out in the presence of 5 different concentrations of NBAGA in the range of 1.0–3.0 mM, in 2.5 ml of 20 mM Na–acetate buffer (pH 4.0) at 50 C. Protection of b-glucosidase from inactivation was carried out by the inclusion of the competitive inhibitor, pNTP-Glc (8.0 mM, 4 Ki) in the reaction mixtures containing 3.0 mM NBAGA. pH Dependence studies of the inactivation process were carried out between pH 3.0 and 7.5 in 20 mM citrate–phosphate buffer in the presence of 3 mM NBAGA.

Active site studies of A. carbonarius b-D -glucosidase

339

Results and discussion

Chemical modification of carboxylate groups

Investigation of the substrate-binding site

In order to further investigate the type of the essential ionizable groups in A. carbonarius b-glucosidase, on the basis of the above results the water-soluble carbodiimide, EDAC, and Woodward reagent K were used as carboxyl group-selective modifying agents for these functional residues (Lundblad & Noyes 1985). Modification of the enzyme with EDAC in the presence of glycine methyl ester (25 C, pH 4.75) resulted in the loss of 50–80% of the initial activity. A plot of residual activity against time was biphasic (Figure 1) and the course of inactivation can be resolved into two first-order processes with the slope in the first 8 min determining the rate of the process. At the first 8 min of inactivation, enzymatic activity decreased by 70–50% while at the second phase by only 20–30%, depending on EDAC concentration. The explanation might be that at the first phase essential carboxylate groups were modified causing rapid loss of activity, while in the slower phase such carboxylate groups were modified which are not catalytic but needed in maintaining the conformation of the enzyme. The pesudofirst-order rate constants, obtained as described by Keresztessy et al. (1994a), were kapp1 ¼ 0.44 min)1 and kapp2 ¼ 0.0164 min)1 using 40 mM EDAC and 0.5 mM glycine methyl ester. The plot of the inverse of kapp1 against the inverse of EDAC concentration provided a second-order rate constant k1 ¼ 0.56 min)1M)1 for the process (Figure 1 inset) (Clarke 1990). Analysis of the order of the inactivation (the double logarithmic plots of kapp against the concentration of the inactivator) yielded a slope of 0.71, indicating that one molecule of EDAC

pH Dependence of reaction kinetic parameters The plot of the maximal velocity for the hydrolysis of pNP-Glc against pH (Cornish-Bowden 1995) showed that two ionizable amino acid residues are involved in the hydrolysis of pNP-Glc (data not shown) and the hydrolysis of the substrate can be described on the basis of the diprotic enzyme model. The plot of Vmax against pH provided pKES1 ¼ 2.24 and pKES2 ¼ 6.14 values for essential ionizable groups in the enzyme– substrate complex. These data permit us to conclude that an ionized carboxylate group and a protonated carboxyl may participate in the hydrolysis of pNP-Glc in the active site in the A. carbonarius b-glucosidase. In the case of pH dependence of Vmax/Km for pNPGlc the corresponding pKE values were 2.8 and 5.93, respectively for the free enzyme. The very low pK value of the ionized carboxylate suggests that this is an unusually acidic group. On the basis of the results, we suppose that similar amino acid side chains participate in the formation of the enzyme–substrate complex, that are either glutamic or aspartic acid residues. Similar results have been obtained for other b-glucosidases, such as cassava linamarase (Keresztessy et al. 1994a), barley b-glucosidase (Skoubas & Georgatsos 1997) or Schizophyllum commune b-glucosidase (Clarke 1990). On the basis of pK values, early workers also invoked the participation of a His residue but chemical modification experiments with group specific reagents have precluded their direct role in catalysis (Clarke et al. 1993).

10

1

1/kapp1 (min)

Inhibition studies were carried out to obtain information about the substrate-binding site of the A. carbonarius b-glucosidase. This enzyme has narrow substrate specificity, as it hydrolyses only b-glucosidic bonds. The good inhibitory effect of glucose (Ki ¼ 8.5 mM) reveals that the interaction with the carbohydrate moiety of the substrate is as important as with the hydrophobic aglycone. This is in contrast to other b-glucosidases, which have wide substrate specificity, where glucose did not have an effect on substrate hydrolysis and the interaction with the hydrophobic aglycone was dominant, e.g. b-glucosidase from pig kidney (Po´csi & Kiss 1988), b-glucosidase from white clover (Po´csi et al. 1989), b-glucosidase from cassava (Keresztessy et al. 1994a). The strong inhibitory effect of glucono(1–5)lactone (Ki ¼ 24 lM) proves that in the transition state the substrate is distorted into half chair conformation. Using 1-thio-glucoside inhibitors, pNTP-Glc (Ki ¼ 2 mM) proved to be a better inhibitor than pATP-Glc (Ki ¼ 50 mM) indicating that the aglyconebinding site has hydrophobic character.

0.8

8 6 4 2 0 0

50

100

150

1/[EDAC] (1/M)

0.6

A/A0 0.4

0.2

0

0

20

40

60

Time (min) Figure 1. Inactivation of b-glucosidase by the water-soluble carbodiimide, EDAC. Enzyme (20 lg/ml) was treated with EDAC in the presence of 0.5 mM glycine methyl ester at pH 4.75, 25 C. At the times indicated samples were taken and residual enzymatic activity was determined using 4.2 mM pNP-Glc as substrate (at pH 4.0 for 10 min at 50 C). EDAC concentrations were: 7 mM (+), 10 mM (d), 20 mM (s), 30 mM (j) and 40 mM (h). (Inset) Plot of the inverse of the pseudo-first-order rate constants obtained for the first phase of inactivation against the inverse of EDAC concentration.

340

S. Ja¨ger and L. Kiss

reacts with one molecule of enzyme when inactivation occurs. The inclusion of pNTP-Glc as a competitive inhibitor in the reaction medium appeared to decrease the rate of activity loss. Thus, after incubation with 30 mM EDAC for 8 min in the absence of added ligand, the enzyme retained 43% of its catalytic activity, whereas 68% residual activity remained in the presence of 10 mM pNTP-Glc. This demonstrates that the catalytically active group is located in the active centre of the enzyme. In order to determine the acid dissociation constant of the detected catalytic carboxylate group, we investigated the pH dependence of the inactivation. The best-fit curve of the plot of kapp1 vs. pH could be reached using KA ¼ 2.45 · 10)5 resulting in pKA ¼ 4.61 (Keresztessy et al. 1994a). This is similar to the pKA values obtained for other b-glycosidases (Keresztessy et al. 1994a; Kiss et al. 2002) further supporting the hypothesis that a carboxylate residue of Asp or Glu plays a catalytic role in the active centre of the enzyme. Considering that the rate of inactivation was the highest at the enzyme’s optimum pH value (pH 4.5–5.5), and that at this pH the nucleophile is deprotonated and the acid catalyst group is protonated we can assume that the EDAC modified the catalytic nucleophile. The b-glucosidase was also rapidly inactivated by WRK. Depending on the reagent concentration the residual activity against pNP-Glc was only 7–30% after 40 min incubation (Figure 2). While the two reagents inactivated the enzyme with similar efficiency we observed different kinetics in the two cases; WRK inactivated the enzyme according to pseudo-first-order kinetics. In contrast to the EDAC modification, the WRK-mediated inactivation occurred only at a higher temperature (50 C). The differences can be explained by the different modification mechanism of the 2 1,8 lg kapp

1,6

ln A0/A

1,4 1,2 1

-1,2 -1,25 1,2 -1,3 -1,35 -1,4 -1,45 -1,5 -1,55 -1,6 -1,65

1,4

1,6

1,8

lg WRK

0,8 0,6 0,4 0,2 0 0

5

10

15

20

25

30

Time (min)

Figure 2. Inactivation of b-glucosidase by Woodward reagent K; semilogarithmic plot of the residual activity against time. Enzyme (17 lg/ml) in Na–acetate buffer, pH 5.0, at 50 C, was treated with 20 mM (d), 30 mM (s), 40 mM (j), 50 mM (h) WRK. At indicated times aliquots were removed and remaining enzyme activity was determined against pNP-Glc as described in the legend to Fig. 1. (Inset) Plot of the logarithm of pseudo-first-order rate constants against the logarithm of inactivator concentration.

reagents. In the case of the WRK the inactivation is more specific for the carboxylate groups located in the active site due to the reagent’s hydrophobic character and structural similarity to the substrate. This might also be an explanation for the different temperatures of the modification reactions. The temperature optimum of the enzyme is 60 C, and high temperature may promote the binding of the substrate-like WRK into the active centre. In the case of the EDAC-mediated inactivation, the glycine methyl ester (which is activated by the EDAC) makes a nucleophile attack modifying not only the active site carboxylate but other carboxylates, too, which are easily accesible for the reagent. Semilogatrithmic plots of residual activity as a function of time of inactivation were linear, indicating that inactivation exhibits pseudo-first-order kinetics (Figure 2). The experimental values correlated well with the lines fitted according to the method described in Keresztessy et al. (1994b). The apparent inhibition constant (Ki) and the first-order rate constant (ki) were obtained; Ki ¼ 105.9 mM and ki ¼ 0.154 min)1. The slope (0.82) determined from the double logarithmic plots of kapp against the concentration of the inactivator was close enough to 1 to be consistent with a reaction order in which one molecule of inactivator reacts with one molecule of enzyme (Figure 2 inset). Protection by pNTP-Glc against inactivation by WRK was shown for pNP-Glc-ase activity demonstrating the active site specificity of the modification reaction. The presence of the competitive inhibitor pNTP-Glc (20 mM) reduced the pseudo-first-order rate constant (kapp) of inactivation from 0.0276 to 0.019 min)1. The effect of blocking of the essential carboxylate group on the reaction kinetic constants was also investigated. After incubation with 50 mM WRK for 30 min Vmax decreased to half of that of the unmodified enzyme (80 mkatal/kg – 35 mkatal/kg) while Km did not change significantly (0.56 mM – 0.63 mM). These results clearly indicate that the carboxylate group plays a role in substrate cleavage and not binding. Chemical modification experiments employing carbodiimide and WRK have confirmed the essential role of only carboxylic acids/carboxylates in the catalytic mechanism of other glycosidases while discounting any role for His residues. In the case of Schizophyllum commune b-glucosidase (Clarke 1990), cellulase (Clarke & Yaguchi 1985) and xylanase (Bray & Clarke 1990) and A. carbonarius b-xylosidase (Kiss et al. 2002) the essential role of carboxylic acids at the active site has been demonstrated by carbodiimide modifications. Similar results were obtained also in the WRK-mediated inactivation reactions as in the case of cassava b-glucosidase (Keresztessy et al. 1994a), S. commune b-glucosidase (Clarke 1990), A. carbonarius b-xylosidase (Kiss et al. 2002), Trichoderma reesei cellobiohydrolase I (Tomme & Claeyssens 1989) or N-acetyl-b-D -glucosaminidase (Amutha et al. 1999).

Active site studies of A. carbonarius b-D -glucosidase

341

Inactivation of the A. carbonarius b-glucosidase by N-bromoacetyl-b-D -glucopyranosylamine

2 1/Kapp (min)

1,8 1,6 1,4

ln A0/A

On the basis of the results of the inactivation studies with specific chemical modifications (EDAC with glycine methyl ester and WRK), we supposed that a carboxylate nucleophile is involved in the catalytic process of A. carbonarius b-glucosidase. To obtain further support for this idea, an affinity label, N-bromoacetyl-b-D -glucopyranosylamine (NBAGA) was used, whose affinity labelling character has been already been proved in the case of other b-glycosidases (Black et al. 1993; Keresztessy et al. 1994b; Kiss et al. 2002). After longer incubation times the enzyme was totally inactivated by NBAGA at 50 C according to pseudofirst-order kinetics (Figure 3). This is a characteristic of active-site-directed modifying agents which form a covalent enzyme–reagent complex via a noncovalent intermediate. The kinetic parameters of inactivation (Ki and ki) were determined similarly to the WRKmediated modification. The apparent inhibition constant (Ki) and the first-order rate constant (ki) were: Ki ¼ 1.06 mM and ki ¼ 0.019 min)1, while the efficiency constant of inactivation ki/Ki ¼ 17.92 min)1 M)1, which is larger with two order of magnitude than in the case of cassava linamarase (ki/Ki ¼ 0.108 min)1 M)1) (Keresztessy et al. 1994b) and one order of magnitude than in the case of A. carbonarius b-xylosidase (ki/Ki ¼ 1.9 min)1 M)1) (Kiss et al. 2002).

150 100 50 0

1,2

0

0,5

1

1/[I] (1/mM)

1 0,8 0,6 0,4 0,2 0 0

20

40

60

HO HO

O

NH

OH

O CH2

HO

OH Br

O

HO HO

+

O O

NH

OH

O O

CH2 (+)

O

-

O

-

C

Br (-)

O

C Br-

C OH

HO

O

NH

HO HO

O O

OH CH2

O

120

The enzyme was partially protected against NBAGAmediated inactivation in the presence of 8 mM pNTPGlc, reducing the pseudo-first-order rate constant (kapp) from 0.0144 to 0.0062 min)1. This result demonstrates that the modified carboxylate is located in the active site of the enzyme.

C

HO

100

Figure 3. Inactivation of b-glucosidase by N-bromoacetyl-b-D -glucopyranosylamine. Semilogarithmic plot of residual activity against time at the following inactivator concentrations: 1.0 mM (h), 1.5 mM (j), 2.0 mM (s), 2.5 mM (d) and 3.0 mM (+). Inactivation reactions were carried out with 35 lg/ml enzyme in pH 4.0 at 50 C. Kinetic parameters of inactivation were determined as described in the case of WRK-mediated modification. (Inset) Double reciprocal plot of pseudo-first-order rate constants calculated from Fig. 2 vs. NBAGA concentration.

C OH

80

Time (min)

O C

Figure 4. Proposed mechanism of inactivation of A. carbonarius b-glucosidase by N-bromoacetyl-b-D -glucopyranosylamine.

342 The pseudo-first-order rate constant (kapp) showed dependence on the proton concentration of the reaction solution (data not shown). When the inactivator reacts simultaneously with two forms of the enzyme (the nucleophile is either in carboxylate form or in protonated form) at conditions approaching saturation inactivator concentration (3 mM, 3 Ki), the observed rate of inactivation can be described according to Naider et al. (1972). The best-fit curve indicated the participation of a group with pKA ¼ 4.5 in the inactivation process. The rate of inactivation was the highest at the optimum pH value of the enzyme (pH 4.5) when the catalytic nucleophile is deprotonated and the acid catalyst group is protonated. Our results suggest that the catalytic nucleophile was modified by NBAGA similarly to the case of the WRK and EDAC-mediated inactivation (where a similar pKA value was obtained for the modified carboxylate and also the rate of inactivation was the highest at the optimum pH of the enzyme). These facts permit us to conclude that modification by NBAGA could be directed toward the carboxyl group acting as the catalytic nucleophile in the mechanism of action of this enzyme. However in some cases the modification of the acid catalyst proton donor group was reported (Keresztessy et al. 1994b; Chir et al. 2002). According to our results Figure 4 shows the proposed mechanism of action of inactivation of A. carbonarius b-glucosidase by NBAGA. The affinity label N-bromoacetyl-b-D -glucopyranosylamine has the potential for use as a general tool for the identification of catalytic amino acid residues. Therefore investigations are in preparation for the identification of the catalytic nucleophile by using this affinity label molecule. Acknowledgements The Hungarian Scientific Research Fund (OTKA T 032106) is gratefully acknowledged for the financial support. Sz. J. acknowledges the financial support of the PhD programme of the University of Debrecen. The authors thank the Department of Microbiology and Biotechnology for the A. carbonarius strain.

References Amutha, B., Jayant, M., Khire, M. & Khan, I. 1999 Active site characterization of the exo-N-acetyl-b-D -glucosaminidase from thermotolerant Bacillus sp. NCIM 5120: involvement of tryptophan, histidine and carboxylate residues in catalytic activity. Biochimica et Biophysica Acta 1427, 121–132. Baird, S.D., Hefford, M.A., Johnson, D.A., Sung, W.L., Yagucgi, M. & Seligy, V.L. 1990 The Glu residue in the conserved Asn-Glu-Pro sequence of two highly divergent endo-b-1,4-glucanases is essential for enzymatic activity. Biochemical and Biophysical Research Communications 169, 1035–1039. Black, T.S., Kiss, L., Tull, D. & Withers, S.G. 1993 N-Bromoacetylglycopyranosylamines as affinity labels for a b-glucosidase and a cellulase. Carbohydrate Research 250, 195–202.

S. Ja¨ger and L. Kiss Bothast, R.J. & Saha, B.C. 1997 Ethanol production from agricultural biomass substrates. Advances in Applied Microbiology 44, 261–286. Bray, M.R. & Clarke, A.J. 1990 Essential carboxy groups in xylanase A. Biochemical Journal 270, 91–96. Chir, J., Withers, S., Wan, C.-F. & Li, Y.-K. 2002 Identification of the two essential groups in the family 3 b-glucosidase from Flavobacterium meningosepticum by labelling and tandem mass spectrometric analysis. Biochemical Journal 365, 857–863. Clarke, A.J. & Yaguchi, M. 1985 The role of carboxyl groups in the function of endo-b-1,4-glucanase from Schizophyllum commune. European Journal of Biochemistry 149, 233–238. Clarke, A.J. 1990 Chemical modification of a b-glucosidase from Schizophyllum commune: evidence for essential carboxyl groups. Biochimica et Biophysica Acta 1040, 145–152. Clarke, A.J., Bray, M.R. & Strating, H. 1993 b-Glucosidases, b-glucanases and xylanases. Their mechanism of catalysis. In b-Glucosidases: Biochemistry and Molecular Biology, ed. Esen, A. Chap. 3, pp. 27–41. Washington: American Chemical Society. ISBN 0-8412–2697–0. Cornish-Bowden, A. 1995 Fundamentals of Enzyme Kinetics. pp. 187– 192. London: Portland Press. ISBN 1-85578072-0. Dan, S., Marton, M., Dekel, M., Bravdo, B.-A., He, S., Withers, S.G. & Shoseyov, O. 2000 Cloning, expression, characterization and nucleophile identification of family 3, Aspergillus niger b-glucosidase. Journal of Biological Chemistry 275, 4973–4980. Esen, A. 1993 b-Glucosidases. In Esen, A. (Ed.), b-Glucosidases: Biochemistry and Molecular Biology, ed. Esen, A. Chap. 1, pp. 1–14. Washington: American Chemical Society. ISBN 0-8412-2697-0. Hazlewood, G.P., Romaniec, M.P.M., Davidson, K., Gre´pinet, O., Be´guin, P., Millet, J., Raynaud, O. & Aubert, J.-P. 1988 A catalogue of Clostridium thermocellum endoglucanase, b-glucosidase and xylanase genes cloned in Escherichia coli. FEMS Microbiology Letters 51, 231–236. Izstein, M. & Colman, P. 1996 Design and synthesis of carbohydratebased inhibitors of protein-carbohydrate interactions. Current Opinion in Structural Biology 6, 703–709. Ja¨ger, Sz., Brumbauer, A., Fehe´r, E., Re´zcey, K. & Kiss, L. 2001 Production and characterization of b-glucosidases from different Aspergillus strains. World Journal of Microbiology and Biotechnology 17, 455–461. Keresztessy, Zs., Kiss, L. & Hughes, M.A. 1994a Investigation of the active site of the cyanogenic b-D -glucosidase (linamarase) from Manihot esculenta crantz (cassava). I. Evidence for an essential carboxylate and a reactive histidine residue in a single catalytic center. Archives of Biochemistry and Biophysics 314, 142–152. Keresztessy, Zs., Kiss, L. & Hughes, M.A. 1994b Investigation of the active site of the cyanogenic b-D -glucosidase (linamarase) from Manihot esculenta crantz (cassava). II. Identification of Glu-198 as an active site carboxylate group with acid catalytic function. Archives of Biochemistry and Biophysics 315, 323–330. Kiss, T., Erdei, A. & Kiss, L. 2002 Investigation of the active site of the extracellular b-D -xylosidase from Aspergillus carbonarius. Archives of Biochemistry and Biophysics 399, 188–194. Lawson, S.L., Warren, R.A.J. & Withers, S.G. 1998 Mechanistic consequences of replacing the active-site nucleophile Glu-358 in Agrobacterium sp. b-glucosidase with a cysteine residue. Biochemical Journal 330, 203–209. Legler, G. 1990 Glycoside hydrolases: mechanistic information from studies with reversible and irreversible inhibitors. Advances in Carbohydrate Chemistry and Biochemistry 48, 319–384. Li, K., Chir, J. & Chen, F.-Y. 2001 Catalytic mechanism of a family 3 b-glucosidase and mutagenesis study on residue Asp-247. Biochemical Journal 355, 835–840. Liebl, W., Gabelsberger, J. & Schleifer, K.-H. 1994 Comparative amino acid sequence analysis of Thermotoga maritima b-glucosidase (BglA) deduced from the nucleotide sequence of the gene indicates distant relationship between b-glucosidases of the BGA family and other families of b-1,4-glycosyl hydrolases. Molecular and General Genetics 242, 111–115.

Active site studies of A. carbonarius b-D -glucosidase Lundblad, R.L. & Noyes, C.M. 1985 The modification of carboxyl groups. In Chemical Reagents for Protein Modification. Vol. II, pp. 105–122. Florida: CRC Press. ISBN 0-84935087-5. Ly, H. & Withers, S.G. 1999 Mutagenesis of glycosidases. Annual Review of Biochemistry 68, 487–522. Naider, F., Bohak, Z. & Yariv, J. 1972 Reversible alkylation of a methionyl residue near the active site of b-galactosidase. Biochemistry 11, 3202–3207. Po´csi, I. & Kiss, L. 1988 Kinetic studies on the broad-specificity b-D glucosidase from pig kidney. Biochemical Journal 256, 139–146. Po´csi, I., Kiss, L., Hughes, M.A. & Na´na´si, P. 1989 Kinetic investigation of the substrate specificity of the cyanogenic b-D glucosidase (linamarase) of white clover. Archives of Biochemistry and Biophysics 272, 496–506. Shulman, M.L., Shiyan, S.D. & Khorlin, A.Y. 1976 Specific irreversible inhibition of sweet-almond b-glucosidase by some b-glycopyranosylepoxyalkanes and b-D -glucopyranosyl isothiocyanate. Biochimica et Biophysica Acta 445, 169–181. Sinnott, M.L. 1990 Catalytic mechanisms of enzymic glycosyl transfer. Chemical Reviews 90, 1171–1202. Skoubas, A. & Georgatsos, J.G. 1997 Identification of essential amino acids for the catalytic activity of barley b-glucosidase. Phytochemistry 46, 997–1003.

343 Taylor, G. 1996 Sialideses: structures, biological significance and therapeutic potential. Current Opinion in Structural Biology 6, 830– 837. Thomas, E.W. 1977 Carbohydrate binding sites. Methods in Enzymology 46, 362–368. Tomme, P. & Claeyssens, M. 1989 Identification of a functionally important carboxyl group in cellobiohydrolase I from Trichoderma reesei. FEBS Letters, 243, 239–243. Tull, D., Burgoyne, D.L., Chow, D.T., Withers, S.G. & Aebersold, R. 1996 A mass spectrometry based approach for probing enzyme active sites: identification of Glu 127 in Cellulomonas fimi exoglycanase as the residue modified by N-bromoacetyl cellobiosylamine. Analytical Biochemistry 234, 119–125. Withers, S.G., Rupitz, K. & Street, I.P. 1988 2-Deoxy-2-fluoroD -glycosyl fluorides. A new class of specific mechanism-based glycosidase inhibitors. Journal of Biological Chemistry 236, 7929– 7932. Withers, S.G., Warren, R.A.J., Street, I.P., Rupitz, K., Kempton, J.B. & Aebersold, R. 1990 Unequivocal demonstration of the involvement of a glutamate residue as a nucleophile in the mechanism of a ‘‘retaining’’ glycosidase. Journal of the American Chemical Society 112, 5887–5889.

 Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 345–350 DOI 10.1007/s11274-004-4800-x

Effect of frozen storage and aging on the Kashkaval cheese starter culture Zh.I. Simov and G.Y. Ivanov* Department of Milk and Dairy Products Technology, University of Food Technologies, 4000 Plovdiv, Bulgaria *Author for correspondence: Tel.: +359-889-700-278, Fax:+359-32-644-102, E-mail: [email protected] Received 30th June 2004; accepted 10th August 2004

Keywords: aging, freezing, frozen storage, Kashkaval cheese, starter culture

Summary The changes in the number of the starter microorganisms Lb. delbrueckii subsp. bulgaricus and Str. thermophilus were followed in frozen-stored Kashkaval cheese made from cow’s milk. Kashkaval samples of various aging times were produced industrially, frozen at T ¼ )16
C and stored at T ¼ )10 to )12
C for 12 months. It was found that the number of Lb. delbrueckii subsp. bulgaricus and Str. thermophilus decreased considerably during frozen storage. The decrease was more substantial for Lb. delbrueckii subsp. bulgaricus, which was evidence for its greater sensitivity to the impact of low temperatures. The aging time of Kashkaval did not influence the changes in the starter culture during frozen storage but is important for its amount in the product aged after defrosting. There was an increase in the Str. thermophilus: Lb. delbrueckii subsp. bulgaricus ratio in samples with shorter aging time subjected to frozen storage and aged after defrosting. The changes in the starter culture in frozen stored Kashkaval cheese can be controlled by an appropriate combination of the two factors: aging time and period of frozen storage. Introduction Frozen storage has been found to be an effective method of prolonging the durability and, in turn, the shelf life of a number of cheeses. Along these lines good results were obtained with Mozzarella (Oberg et al. 1992; Kuo et al. 2003) and Cheddar (Kasprzak et al. 1994; Johnston 2000) hard cheeses as well as with some soft cheeses (Sendra et al. 1999; Tejada et al. 1999l; Verdini & Rubiolo 2002). The investigations revealed that the preservation of the quality of cheeses during frozen storage depended on a number of factors such as freezing and storage conditions used (Bertola et al. 1996; Graiver et al. 2004), composition and aging time of the product (Verdine et al. 2002; Kuo & Gunasekaran 2003), etc. Understanding the effect of these factors on the characteristics of the cheeses is of vital importance for defining the optimum parameters of the product that is going to be frozen-stored. An important characteristic that changes during the processes of freezing and storage of cheese is the amount and composition of their microflora. Fontecha et al. (1996) found a significant decrease in the number of starter microorganisms in semi-hard frozen-stored cheeses. During the aging of defrosted samples, the number of those microorganisms increased, but did not reach the control values. Alichanidis et al. (1981) established similar changes in the microflora of young Teleme cheese stored for 6 months at T ¼ )20
C. Different microorganism species decrease to a different extent during storage. Proteolytic microorganisms decrease gradually until the second month of storage followed by a sharp

drop by the sixth month. Lactobacilli decrease at a faster rate than proteolytic microorganisms until the second month, but diminish to a lesser extent by the sixth month. The remaining lactic acid microorganisms decrease to a smaller degree and more gradually during storage. Psychrophilic and lipolytic microorganisms decrease mainly between the first and second months, while yeasts decrease mainly during the first month of their frozen storage. Portman (1971) studied the changes in the microflora of cheese curd stored at T ¼ )20
C to )40
C for 6–7 months. He found significant changes in the correlations between various bacterial groups depending on their ability to endure the long-time impact of low temperatures. A major phenomenon– is the significant decrease in the lactic acid microflora, which is largely dominating. At the same time the amount of the other bacterial species remains at starting level, while micrococci and yeasts decrease significantly (50–100 times), and coliforms tend to disappear altogether. The purpose of the present study was to establish the effect of frozen storage and aging on the growth of the starter micro organisms in cow’s milk Kashkaval cheese.

Materials and methods Materials Kashkaval loaves were produced at a local dairy plant (Filipopolis-RK Ltd, Plovdiv) from a single vat of milk

346 according to the following procedure. Cow’s milk of 3.9 % fat was heat-treated at T ¼ 65
C for 15 s and cooled to T ¼ 33
C, pumped into a cheese vat, and inoculated with a thermophilic culture which consisted of Streptococcus thermophilus and Lactobacillus delbrueckii subsp. bulgaricus. Subsequently, calcium chloride and commercial animal rennet were also added. Following a 30 min set, the curd was cut and allowed to heal for 5 min. Next, the curd was stirred gently without heat for another 20min, followed by heating to 38–40
C for 40 min with continuous agitation. After that, the whey was drained from the curd, which was partially moulded. The curd was kept at cooking temperature for cheddaring until the pH reached 5.3 (about 2 h). Then the curd was milled and stretched on a blade mixer under a concentrated salt solution (13%) at 72
C. The Kashkaval loaves were moulded into 0.5 kg parallelepiped forms. After 15 h the Kashkaval loaves were vacuum-packaged in polyethylene bags under 90–99.8 Pa and left to ripen at T ¼ 6–8
C for 45 days. For the purposes of the present study, the ripening process was divided into three stages lasting 5, 25 and 45 days, respectively. In this way Kashkaval cheese of three different aging times was obtained – young (5 days), semi-ripened (25 days), and ripened (45 days). After each stage of ripening, the Kashkaval loaves with the respective degree of ripeness were frozen. Freezing and defrosting Kashkaval samples of various aging times were frozen in a freezer (T ¼ )16
C) until a core temperature of T ¼ )10
C was reached. The frozen samples were stored at T ¼ )10 to )12
C for 1, 3, 6, 9 and 12 months, respectively. The samples were defrosted at T ¼ +8
C, and the young and semi-ripened samples were subsequently aged at T ¼ 6–8
C for 40 and 20 days, respectively. Physicochemical analyses The Kashkaval loaves were analysed for moisture content (heat at 105
C to constant weight), pH, fat content according to Gerber-Van Gulik (1955) and NaCl content in the aqueous phase according to Mohr (1992). Non-casein nitrogen, non-protein nitrogen and total nitrogen content were determined by the Vakaleris & Price method (1959) modified to suit the specific conditions of the analysis. For non-casein nitrogen (NCN) determination, approximately 5 g of kashkaval cheese was extracted in 100 ml sodium acetate buffer (pH ¼ 4.6), the homogenate was agitated at ambient temperature for 2 h and filtered. Nitrogen fraction soluble in 12% trichloracetic acid was considered the non-protein nitrogen (NPN). To determine the NPN content, approximately 5 g of kashkaval cheese were homogenized in 40 ml sodium acetate buffer (pH ¼ 4.6), the homogenate was agitated at ambient temperature for 2 h, then 10 ml of 60% trichloracetic acid was added and homogenate was filtered. Nitrogen determination

Zh.I. Simov and G.Y. Ivanov was performed in duplicate by the Kjeldahl method using Kjeltec Auto 1030 Analyzer (Tecator Sweden) combined with the Digestion System 20.

Microbiological analyses The total number of viable cells Streptococcus thermophilus and Lactobacillus delbrueckii subsp. bulgaricus in the studied Kashkaval samples was determined by cultivation on synthetic culture media (M17 and MRS). The methodology described in IDF Standard 149A: 1997 was followed. The samples were prepared according to IDF Standard 122C: 1996. 10 g of the test kashkaval sample were transferred into the container of a peristaltic-type blender. 90 ml diluent (20 % sodium citrate solution) was added and the mixture was blended until the kashkaval was thoroughly dispersed. Appropriate dilutions were mixed with the molten and cooled (47 ± 1
C) medium (M17 for Streptococcus thermophilus and MRS for Lactobacillus delbrueckii subsp. bulgaricus). After solidification the Petri dishes were inverted and incubated at 30 ± 1
C for 48 h under aerobic conditions for Streptococcus thermophilus and at 37 ± 1
C for 72 h under anaerobic conditions for Lactobacillus delbrueckii subsp. bulgaricus. After incubation all colonies were counted. The percentage ratio of Streptococcus thermophilus and Lactobacillus delbrueckii subsp. bulgaricus was calculated by means of the following equations: PRL ¼

NL
100% NL þ NS

PRS ¼

NS
100% NL þ NS

where PRL is the percentage ratio of Lactobacillus delbrueckii subsp. bulgaricus; PRS – the percentage ratio of Streptococcus thermophilus; NL – the total number of viable cells Lactobacillus delbrueckii subsp. bulgaricus in the tested samples; NS – the total number of viable cells Streptococcus thermophilus in the tested samples. Statistical analysis Statistical analyses were carried out on the averages of the triplicate results. Two-way multivariate analysis of variance (MANOVA) and multiple comparison tests were carried out to study the effect of both freezing procedures and ripening time on the physicochemical characteristics and the count of Streptococcus thermophilus and Lactobacillus delbrueckii subsp. bulgaricus in Kashkaval loaves (Box et al. 1978). Differences in the averages and F-tests were considered significant when the computed probabilities were less than 0.05. All

347

Frozen storage Kashkaval cheese starter culture Table 1. Changes in the pH, non-casein nitrogen, non-protein nitrogen and total nitrogen contents during the aging of Kashkaval cheese. Aging time

pH

TN % NCN % NPN % NCN/ NPN/ TN % TN %

Young Kashkaval 5.44a 3.10a (5 days) Semi-ripened 5.30c 3.10a Kashkaval (25 days) Ripened Kashkaval 5.17b 3.10a (45 days)

0.180c

0.129a

5.81a

4.16b

0.336a

0.204b

10.84c 6.58c

0.504b

0.284c

16.26b 9.16a

a, b, c – means within same column bearing a common superscript did not differ significantly (P < 0.05). TN–total nitrogen. NCN– non-casein nitrogen. NPN– non-protein nitrogen.

statistical procedures were computed using the Microsoft Excel software.

Results The Kashkaval loaves had a moisture content of 48 ± 1%, fat in dry matter of 51 ± 0.5% and salt in

moisture of 5 ± 0.6%. The pH values and the noncasein nitrogen, non-protein nitrogen and total nitrogen contents are shown in Table 1. It can be seen that the pH values slightly decreased when the aging time of Kashkaval cheese was increased. That was due to the lactic acid fermentation, which took place in Kashkaval cheese during the first stages of ripening. The non-casein nitrogen and non-protein nitrogen content increased significantly (P < 0.05) during ripening as a result of the proteolytic activity of the starter microorganisms of Kashkaval cheese (Table 1). There were no significant differences (P < 0.05) in the physicochemical characteristics between the controls and the frozen stored Kashkaval loaves. During the processes of freezing and frozen storage the amount of starter culture in Kashkaval cheese decreased significantly (P < 0.05) (Figure 1). The number of viable cells Lb. delbrueckii subsp. bulgaricus decreased during the 12-month storage from 6.4 · 104 to 5.7 · 101 c.f.u./g for young Kashkaval cheese, from 4.3 · 107 to 1.6 · 104 c.f.u./g for semi-ripened Kashkaval cheese, and from 7 · 108 to 7.4 · 105 c.f.u./g for the ripened Kashkaval cheese. The decrease of Str. thermophilus for the same period was from 107 to 105 c.f.u./g

Figure 1. Changes in the numbers of Str. thermophilus (a) and Lb. delbrueckii subsp. bulgaricus (b) of frozen stored young, semi-ripened and ripened Kashkaval cheese. Each plot represents an average of three determinations.

348 regardless of the aging time of the samples prior freezing. It is evident that the decrease in the starter culture during storage does not depend on the aging time of Kashkaval cheese (Figure 1). During the aging process before freezing and after defrosting, the amount of starter culture in Kashkaval cheese increased significantly (P < 0.05) (Figure 2). During the aging process of the control Kashkaval sample the number of viable cells Str. thermophilus increased slightly and remained in the order of 107 c.f.u./g, while Lb. delbrueckii subsp. bulgaricus grew intensively and increased from 6.4 · 104 c.f.u./g at the beginning to 7 · 108 c.f.u./g at the end of the aging process. In samples frozen at various aging times and aged after defrosting (up to 45 days) there was a decrease in the amount of starter microflora with longer frozen storage times (Figure 2). That decrease was greater for Lb. delbrueckii subsp. bulgaricus, whose amount decreased from 7 · 108 to 106–107 c.f.u./g compared to Str. thermophilus, which decreased from the order of 108 to the order of 107 c.f.u./g. The changes in the percentage ratio of Lb. delbrueckii subsp. bulgaricus and Str. thermophilus are shown in

Zh.I. Simov and G.Y. Ivanov Figure 3. It is evident that the percentage ratio of Str. thermophilus increased and the percentage ratio of Lb. delbrueckii subsp. bulgaricus decreased during frozen storage of the tested samples. In samples with shorter aging times subjected to frozen storage and aged after defrostation there was an increase in the Str. thermophilus: Lb. delbrueckii subsp. bulgaricus ratio.

Discussion The changes established in the starter culture of Kashkaval cheese are similar to those observed during storage of other kinds of cheeses. Alichanidis et al. (1981) and Fontecha et al. (1996) also found a significant decrease in the number of starter microorganisms during frozen storage of semi-hard sheep milk cheeses. According to Tejada et al. (2002) lactobacilli counts in sheep milk cheeses frozen for 9 months were significantly (P < 0.05) lower than those detected in control cheeses and in cheeses frozen for 3 months. There was also a decrease in lactococci counts after 6 months of

Figure 2. Changes in the numbers of Str. thermophilus (a) and Lb. delbrueckii subsp. bulgaricus (b) of frozen stored young and semi-ripened Kashkaval cheese fully ripened after thawing. Each plot represents an average of three determinations.

Frozen storage Kashkaval cheese starter culture

349

Figure 3. Changes in the percentage ratio of Str. thermophilus (a) and Lb. delbrueckii subsp. bulgaricus (b) in frozen stored Kashkaval cheese.

frozen storage. Portman (1971) found that the size of the lactic acid bacteria population during frozen storage of cheese curd contracted approximately 10 times as compared with its initial amount contained in the fresh curd. The results obtained (Figures 1 and 2) show that the decrease in the number of viable cells during frozen storage is more substantial for Lb. delbrueckii subsp. bulgaricus, which was evidence for its greater sensitivity to the impact of low temperatures. We suggest that the greater sensitivity of the Lb. delbrueckii subsp. bulgaricus to the impact of low temperatures is due to the smaller amounts of lipids and free amino acids on the cell wall compared to the Str. thermophilus. Other authors (Alichanidis et al. 1981; Tejada et al. 2002) also established a greater decrease in the number of lactobacilli during frozen storage of sheep milk cheeses as compared to the other lactic acid microorganisms. We suggest that the increase in the Str. thermophilus: Lb. delbrueckii subsp. bulgaricus ratio in samples with shorter aging time aged after thawing was due to the different sensitivity of these microorganisms to the impact of low temperatures and their different growth

intensity during aging. Str. thermophilus grew poorly during aging of Kashkaval loaves and for that reason the differences in the number of its viable cells in samples of different aging times were small. After defrosting, however, the samples of shorter aging time underwent a longer period of aging than those of longer aging time, allowing Str. thermophilus to grow in them to a larger extent. Lb. delbrueckii subsp. bulgaricus grew intensively during aging resulting in a significant increase in its amount with longer aging time of Kashkaval samples. After the stored samples were defrosted, its amount was much less in young Kashkaval cheese than in semi-ripened Kashkaval cheese. Compared to semi-ripened Kashkaval cheese, the longer period of aging after defrosting of young Kashkaval cheese does not succeed in neutralizing these differences in the amount of Lb. delbrueckii subsp. bulgaricus and they remain the same in aged samples. The significant decrease in the amount of the starter microorganisms and the changes in the Str. thermophilus: Lb. delbrueckii subsp. bulgaricus ratio could disrupt the normal aging path after defrosting frozen-stored Kashkaval cheese. To avoid that, it is recommended

350 that young Kashkaval cheese frozen storage should not exceed 12 months. When circumstances demand extended storage, Kashkaval cheese of longer aging time should be frozen. It is recommended also that the Lb. delbrueckii subsp. bulgaricus strains with improved resistance to freezing be used (Monnet et al. 2003). Conclusions The starter culture of Kashkaval cheese decreases significantly (P < 0.05) during freezing and frozen storage. Lb. delbrueckii subsp. bulgaricus is more sensitive to the impact of low temperatures than Str. thermophilus. The aging time of Kashkaval cheese does not influence the changes in its starter culture during frozen storage but is important for its amount in the product aged after defrosting. With the decrease in the aging time of frozen stored samples the Str. thermophilus: Lb. delbrueckii subsp. bulgaricus ratio increases in Kashkaval aged after defrostation. The percentage ratio of Str. thermophilus increased and the percentage ratio of Lb. delbrueckii subsp. bulgaricus decreased during frozen storage of tested samples. By an appropriate combination of the two factors, aging time and frozen storage period, the changes in the starter culture of Kashkaval cheese can be controlled, which is important for the quality of the final product. References Alichanidis, E., Polychroniadou, A., Tzanetakis, N. & Vafopoulou, A. 1981 Teleme cheese from deep frozen curd. Journal of Dairy Science 64, 732–739. Bertola, N.C., Califano, A.N., Bevilacqua, A.E. & Zaritzky, N.E. 1996 Effect of freezing conditions on functional properties of low moisture Mozzarella cheese. Journal of Dairy Science 79, 185–190. Box, G.E.P., Hunter, W.G. & Hunter, J.S. 1978 Statistics for Experimenters: An introduction to Design, Data Analysis, and Model Building. New York: J. Wiley & Sons. ISBN 0471093157 BSI (1955) British Standard No 696. Gerber method for the determination of fat in milk and milk products. British Standards Institution, London. Fontecha, J., Kalab, M., Medina, J.A., Pelaez, C. & Juarez M. 1996 Effects of freezing and frozen storage on the microstructure and

Zh.I. Simov and G.Y. Ivanov texture of ewe‘s milk cheese. Zeitschrift fu¨r Lebensmittelen-Unterzuch und -Forschung 203, 245–251r. Graiver, N. G., Zaritzky, N.E. & Califano, A.N. 2004 Viscoelastic behaviour of refrigerated and frozen low-moisture Mozzarella cheese. Journal of Food Science 69, 123–128. IDF-Standard 122C: 1996 – Milk and milk products – Preparation of samples and dilutions for microbiological examination. IDF-Standard 149A: 1997 – Dairy starter cultures of lactic acid bacteria – Standard of identity. Johnston, D.E. 2000 The effects of freezing at high pressure on the rheology of Cheddar and Mozzarella cheeses. Milchwissenschaft 55, 545–549. Kasprzak, K., Wendorff, W.L. & Chen, C.M. 1994 Freezing qualities of Cheddar- type cheeses containing varied percentages of fat, moisture, and salt. Journal of Dairy Science 77, 1771 1782. Kuo, M.I. & Gunasekaran, S. 2003 Effect of frozen storage on physical properties of pasta filata and non-pasta filata mozzarella cheeses. Journal of Dairy Science 86, 1108–1117. Kuo, M.I., Anderson, M.E. & Guanasekaran, S. 2003 Determining effects of freezing on pasta filata and non-pasta filata Mozzarella cheeses by nuclear magnetic resonance imaging. Journal of Dairy Science 86, 2525–2536. Marshall, R.T. 1992 Standard Methods for the Examination of Dairy Products, pp. 446–447 Washington, DC: American Public Health Association:. ISBN 0–87553–210–1. Monnet, C., Beal, C. & Corrieu, G. 2003 Improvement of the resistance of Lb. delbrueckii ssp. bulgaricus to freezing by natural selection. Journal of Dairy Science 86, 3048–3053. Oberg, C.J., Merril, R.K., Brown, R.J. & Richardson, G.H. 1992 Effects of freezing, thawing, and shredding on low moisture, partskim Mozzarella cheese. Journal of Dairy Science 75, 1161 Portmann, A. 1971 Revue Generale Du Froid 11, 1043–1047. Sendra, E., Mor-Mur, M., Pla, R. & Guamis, B. 1999. Evaluation of freezing pressed curd for delayed ripening of semi-hard ovine cheese. Milchwissenschaft 54, 550–553. Tejada, L., Gomez, R. & Vioque, M. 2000 Effect of freezing and frozen storage on the sensorial characteristics of Los Pedroches, a Spanish ewe cheese. Journal of Sensory Studies 15, 251–262. Tejada, L., Sanchez, E., Gomez, R., Vioque, M. & FernandezSalguero, J. 2002 Effect of freezing and frozen storage on chemical and microbiological characteristics in sheep milk cheese. Journal of Food Science 67, 126–129. Vakaleris, D.G. & Price, W.V. 1959 A rapid spectrometric method for measuring cheese ripening. Journal of Dairy Science 42, 246. Verdini, R.A. & Rubiolo, A.C. 2002 Effect of frozen storage time on the proteolysis of soft cheeses studied by principal component analysis of proteolytic profiles. Journal of Food Science 67, 963– 967. Verdini, R.A., Zorrilla, S.E. & Rubiolo, A.C. 2002 Free amino acid profiles during ripening of Port Salut Argentino cheese after frozen storage. Journal of Food Science 67, 3264–3270.

World Journal of Microbiology & Biotechnology (2005) 21: 351–358 DOI 10.1007/s11274-004-2610-9

Ó Springer 2005

Characteristics of the bacteriocin produced by Lactococcus lactis subsp. cremoris CTC 204 and the effect of this compound on the mesophilic bacteria associated with raw beef R. Bromberg*, I. Moreno, R.R. Delboni, H.C. Cintra and P.T.V. Oliveira Instituto de Tecnologia de Alimentos, Av. Brasil 2880, 13070-178, P.O. Box 139, Campinas, Sa˜o Paulo, Brazil *Author for correspondence: Tel.: +55-19-3743-1880, Fax: +55-19-3743-1882, E-mail: [email protected] Received 12 February 2004; accepted 17 August 2004

Keywords: Bacteriocin, characterization, lactic acid bacteria, Lactococcus lactis subsp. cremoris, meat, mesophilic bacteria, preservation

Summary Screening for the bacteriocin production of strains of lactic acid bacteria from various meat and meat products resulted in the detection of a bacteriocin-producing Lactococcus lactis subsp. cremoris CTC 204, isolated from chicken. The bacteriocin inhibited not only closely related lactic acid bacteria (Lactobacillus helveticus), but also pathogenic microorganisms (Staphylococcus aureus, Listeria monocytogenes, Bacillus cereus, and Clostridium perfringens). It was inactivated by a-chymotrypsin, ficin, papain, and pronase E, but not by lipase or pepsin. This compound was heat stable even at autoclaving temperature (121 °C for 10 min) and was produced during refrigerated storage. It was also active over a wide pH range (2–10), but the highest activity was observed in the lower pH range. The results indicated that dipping raw beef in the bacteriocin produced by strain CTC 204 could contribute to the extension of the shelf life of refrigerated bovine meat.

Introduction Bacteriocins are widely produced by lactic acid bacteria. These compounds have received much attention mainly because of their potential use as ‘natural’ food preservatives (Cleveland et al. 2001). Furthermore, they show an interesting potential application as food additives in the control of food spoilage and pathogenic foodborn microorganisms (Yang & Ray 1994; Muriana 1996). Bacteriocins have traditionally been defined as proteinaceous compounds produced by bacteria, which inhibit (bacteriostatic) or kill (bactericidal) closely related species (Tagg et al. 1976). However, this definition has been extended to include similar compounds that act on a broader range of species and contain other functional moieties, such as carbohydrates or lipids (Eckner 1992). Bacteriocin-producing strains can be used as part of, or adjuncts to, starter cultures for fermented foods, in order to improve safety and quality (Caplice & Fitzgerald 1999; Lu¨cke 2000). Bacteriocin-like substances produced by lactic acid bacteria from dairy products, fruits and vegetables are well known (Eckner 1992). However, bacteriocins produced by lactic acid bacteria associated with meat, such as Pediococcus, Leuconostoc, Carnobacterium, Lactobacillus, and Lactococcus genera, are likely to have a much greater potential as preservatives in their

application to meat (McMullen & Stiles 1996). According to Schillinger & Lu¨cke (1989), lactic acid bacteria originally isolated from meat and meat products are probably the best candidates for improving the microbiological safety of these foods, because they are well adapted to the conditions in meats and should therefore be more competitive than lactic acid bacteria from other sources. Some bacteriocins have been tested for the biopreservation of meat (Hugas 1998). Nisin, produced by Lactococcus lactis subsp. lactis, has not been very successful because of its low solubility, uneven distribution and lack of stability. Moreover the required effective dose is uneconomical and exceeds the acceptable daily intake for a consumption of 100 g meat/day and an average weight of 60 kg. Pediocin, produced by Pediococcus acidilactici, is more suitable for use in meat and meat products than nisin. However, this microorganism is not an indigenous meat strain and is not able to grow and thus produce bacteriocin at retrigeration temperatures. An extensive screening of nearly 800 bacteriocinproducing lactic acid strains isolated from a variety of meat and meat products led to the selection of several strains which demonstrated antibacterial activity (R. Bromberg et al. 2004). According to the results of the well diffusion assay 128 strains inhibited the growth of the indicator strains, Staphylococcus aureus CTC 033

352

R. Bromberg et al.

and/or Listeria innocua Lin 11. The activity of the inhibitory agent was tested under conditions that eliminated the possible effect of organic acids, hydrogen peroxide and bacteriophages. Of these colonies, one designated as CTC 204 secreted inhibitory substance into the culture medium. In the present investigation, some characteristics of a bacteriocin produced by strain CTC 204, such as its antibacterial activity, stability at different pH values and temperature conditions, growth characteristics and bacteriocin production, were studied. Also, the effect of this compound on the mesophilic population present in minced raw beef was evaluated.

for the other microorganisms Trypticase Soya Broth (TSB, Oxoid) was used. Both media were supplemented with 15.0% glycerol. Working cultures were prepared as slants on MRS Agar (Oxoid) for the producer bacterium or TSA Agar with 0.6% yeast extract supplement (Oxoid) for the others, and stored at 4 °C. Cultures for experiments were streak-plated once a week, inoculated into media from a single colony and incubated for 24 h. Before use, the microorganisms were transferred twice into MRS Broth for lactic acid bacteria or TSB Broth for the other bacteria, and incubated according to the growing conditions presented in Table 1. Unless otherwise stated, Bacillus cereus CTC 001 was used as the indicator in bacteriocin activity assays.

Materials and methods Microorganisms and their maintenance

Identification of the isolate producing the antimicrobial substance

The bacteriocin-producing bacterium, strain CTC 204, used in this study was obtained from the microbiological culture collection at the Centro de Tecnologia de Carnes, Instituto de Tecnologia de Alimentos, Campinas, Sa˜o Paulo, Brazil. This strain was isolated from raw chicken giblets. The indicator bacterium, the spoilage and foodborn microorganisms used in the experiments are listed in Table 1. The stock cultures of lactic acid bacteria were maintained at )80 °C in de Man Rogosa Sharpe Broth (MRS, Oxoid Ltd., Basingstoke, UK) and

The selected strain CTC 204 was further characterized and identified on the basis of the Gram stain, catalase reaction, reduction of nitrate to nitrite, salt tolerance, ability to grow at different temperatures and pH values (assayed according to Harrigan & McCance 1976) and glucose metabolism (Elortondo et al. 1999). Carbohydrate fermentation and other tests were determined using the BBL CrystalÔ Identification Systems, GramPositive ID Kit (Becton & Dickinson Microbiology Systems, Maryland, USA). Bacteriocin preparation

Table 1. Microorganisms, strains and growth conditions. Microorganisms

Growth conditions

Bacillus cereus ATCCa 14579 B. cereus CTCc 001 Clostridium perfringens CTC 042 CI. sporogenes CTC 006 Enterococcus faecalis ATCC 19433 Escherichia coli ATCC 25422 Lactobacillus helveticus (Wiesbyd) Lb. plantarum TECNOLATf 434 Leuconostoc mesenteroides ATCC 10830 Listeria innocua Lin 11 (INRAg) L. monocytogenes CTC 021 Micrococcus sp. ATCC 4698 Pseudomonas sp. CTC 032 Salmonella typhimurium ATCC 14028 Staphylococcus aureus CTC 033 Streptococcus sp. ATCC 25175 Sulphite-reducing clostridium CTC 005 Weissella viridescens CCTf 0849

TSBb 24 h/30 °C TSB 24 h/30 °C TSB 24 h/37 °C TSB 24–48 h/37 °C TSB 24 h/30 °C TSB 24 h/37 °C MRSe 24–48 h/45 °C MRS 24 h/30 °C MRS 24 h/30 °C TSB 24 h/37 °C TSB 24 h/37 °C TSB 24 h/30 °C TSB 24 h/37 °C TSB 24 h/37 °C TSB 24 h/37 °C TSB 24 h/30 °C TSB 24 h/37 °C MRS 24 h/30 °C

a

ATCC – American Type Culture Collection, Rockville, MD, USA. TSB – Trypticase Soya Broth. c CTC – Centro de Tecnologia de Carnes, Instituto de Tecnologia de Alimentos, Campinas, SP, Brazil. d Wiesby GmbH & Co. KG, Germany. e MRS – de Man Rogosa Sharpe Broth. f TECNOLAT – Centro de Tecnologia de Laticı´ nios, Instituto de Tecnologia de Alimentos, Campinas, SP, Brazil. g INRA – Institute National de Recherches Agronomiques, Jouyen-Josas, France. h CCT – Fundac¸a˜o Tropical Andre´ Tosello, Campinas, SP, Brazil. b

The bacteriocin-producing strain was grown in MRS Broth for 24 h at 30 °C. Cell-free supernatants were collected by centrifugation (7500 g, 10 min, 4 °C) of overnight MRS broth cultures. The supernatant fluid was neutralised to pH 6.5 with 10 M NaOH and sterilized by heating at 95 °C for 5 min. This crude bacteriocin preparation was kept at 5 °C until use. Inhibitory activity spectrum The inhibitory spectrum of strain CTC 204 was tested against the bacteria shown in Table 1. The test organism was assayed by the serial twofold dilution assay of Mayr-Harting et al. (1972). The titre was defined as the reciprocal of the highest dilution showing an inhibition of the indicator strain multiplied by 100 to express the results as activity units per millilitre (AU/ml). Sensitivity of the bacteriocin-like substance to enzymes Cell-free supernatant at pH 6.5 was treated with the following enzymes (0.2 mg/ml): ficin (3.4.22.3., Sigma Chemical Co., Dorset, England) in 20 mM sodium phosphate, pH 7.0; trypsin (EC 3.4.21.4., Sigma) in 40 mM Tris–HCI, pH 8.2; a-chymotrypsin (EC 3.4.21.1., Sigma) in 20 mM Tris–HCl, pH 8.0; pronase E (EC 3.4.24.4., Sigma) in 20 mM Tris–HCI, pH 7.8;

353

Bacteriocin: properties and use in meat pepsin (EC 3.4.23.1., Merck Darmstad, Germany) in 0.002 M HCI; lipase (3.1.1.3., Merck) in 0.1 M potassium phosphate, pH 6.0; papain (3.4.22.2., Sigma) in 0.05 M sodium phosphate, pH 7.0. All these solutions were filter-sterilized through Millex GV 0.22 lm filters (Millipore S.A., St. Quentin-en-Yvelines, France) and then added to sterile cell-free supernatants (v/v, 1/1) at pH 6.5. The controls consisted of enzyme solutions without bacteriocin and cell-free supernatant alone in 0.1 M sodium phosphate buffer. The samples and controls were incubated at 37 °C for 2 h and heated in boiling water for 5 min to denature the enzymes. Cell-free supernatant fluid from the bacteriocin-producing strain was obtained as described before. The remaining bacteriocin-like substance activity was determined by the serial twofold dilution assay (Mayr-Harting et al. 1972). The titre was defined as mentioned before.

(Seward Laboratory, Model ‘400’, London, England). Further decimal dilutions were prepared using 0.1% peptone water as the diluent. The dilutions were plated in duplicate on Agar Plate Count (APC, Merck 1.05463) for total mesophilic counts. Inoculated plates were incubated aerobically at 35 °C for 48 h. Results were expressed as log c.fu./g. The pH values of the samples were monitored with a pH meter at the moment of sampling. The antibacterial activity of the bacteriocinlike substance was determined by the twofold dilution assay, using L. innocua Lin 11 as indicator. Specific count values were calculated by linear regression (95% confidence interval).

Resistance of the antibacterial substance under different temperature conditions and pH values

Strain CTC 204 was identified as a Gram-positive, nonmotile, catalase-negative coccus, producing no gas from glucose. The strain reduced nitrate to nitrite, grew at 10 °C, at pH 4.4 and in media containing 6.5% NaCl but not at 45 °C, pH 9.6 or in 18% NaCl. Based on these characteristics, the analysis of the carbohydrate fermentation pattern and other reactions using the BBL CrystalÔ Identification System (Table 2), strain CTC 204 was tentatively identified as Lactococcus lactis subsp. cremoris (0.9917%). Strains of Lc. lactis ssp., organisms traditionally associated with dairy and vegetable products, have been isolated from fermented sausage (Rodriguez et al. 1995), raw pork (Garver & Muriana 1993), vacuum-packed seafood (Mauguin & Novel 1994) and cooked poultry meat (Barakat et al. 2000). Lc. lactis ssp. has been employed in meat starter cultures for a variety of fermented meat products (Cleveland et al. 2001; Scannell et al. 2001).

To determine the thermal stability at different pH values, the cell-free. supernatant, adjusted to different pH values (2.0, 4.0, 6.0, 7.0, 8.0, 9.0, 10.0, and 12.0) was heated at 65 °C for 30 min, 100 °C for 10 min, and 121 °C for 10 min, or cooled at 4 °C for 24 h. The treated and untreated samples were assayed for bacteriocin-like substance activity by the critical dilution assay (Mayr-Harting et al. 1972). Bacteriocin production at different temperatures In order to study the effect of incubation temperature on growth and bacteriocin production, stationary phase cells of strain CTC 204 were inoculated at 1% (v/v, 105– 106 c.f.u./ml) into buffered MRS Broth and incubated at its optimum temperature (37 °C), abusive temperature for meat storage (25 °C), refrigeration temperature (4 °C) and freezing temperature ()20 °C). At appropriate intervals the following determinations were performed: biomass by absorbance at OD600 nm in an u.v–visible spectrophotometer (Varian, Cary 1 E, USA), pH (Toledo Mettler, MP125, Switzerland), and the antibacterial activity of the bacteriocin-like substance by the twofold dilution assay (Mayr-Harting et al. 1972). Meat preparation and bacteriocin addition Twenty-five-gram portions of raw minced beef were dipped into the bacteriocin preparation and maintained there for 5 min following by draining for 2 min. The samples were packed into sterile plastic bags and stored at 4 °C. Samples without added bacteriocin, but dipped into sterile distilled water, were used as controls. The tests were repeated three times at selected times with three different beef samples. For the microbiological determinations, each sample was homogenized with 225 ml of 0.1% peptone water, using a stomacher

Results and discussion Identification of the bacteriocin-producing strains

Antimicrobial activity According to the inhibitory spectrum of activity of the bacteriocin-like substance produced by Lc. lactis subsp. cremoris CTC 204, it presented activity against one closely related bacterium (Lb. helveticus), pathogenic strains of S. aureus CTC 033, L. monocytogenes CTC 021, Cl. perfringens CTC 042, B. cereus CTC 001 and other contaminants such as L. innocua Lin 11 (not shown). Like other bacteriocins produced by lactic acid bacteria (Piard et al. 1990), it was not effective (0 AU/ ml) against the Gram-negative bacteria tested (E. coli ATCC 25422, Pseudomonas sp. CTC 032, and Salm. typhimurium ATCC 1402). The mentioned bacteriocin activity was based on the choice of an arbitrary endpoint, the last dilution showing complete inhibition of the indicator strains. Bacteriocin produced by strain CTC 204 was two times more potent over Lb. helveticus (Wiesby) (800 AU/mt) than against L. innocua Lin 11, L. monocytogenes CTC 021, and S. aureus

354

R. Bromberg et al.

Table 2. Growth, morphological and biochemical profile of strain CTC 204. Tests employed

Strain characteristics

Morphology

Gram positive, coccus, non-motile ) Homofermentative + + ) + ) + ) + ) + ) + ) ) ) + + ) + ) ) ) +

Catalase reaction Glucose metabolism Nitrate reaction Growth at 10 °C Growth at 45 °C Growth in 6.5% NaCl Growth in 18% NaCI Growth at pH 4.4 Growth at pH 9.6 4MU-b-D -g1ucoside L -Valine L -Phenylalanine 4MU-a-D -glucoside L -Pyroglutamic acid–AMC L -Tryptophan L -Arginine 4 MU-N-acetyl-b-D -glucosaminide 4 MU-phosphate 4 MU-b-D -glucuronide L -Isoleucine p-Nitrophenyl-b-D -glucoside p-Nitrophenyl-b-D -cellobioside p-Nitrophenyl-phosphate p-Nitrophenyl a-D -maltoside o-Nitrophenyl-b-D -galactoside (ONPG) and p-nitrophenyl-a-D -galactoside Proline and Leucine-p-nitroanilide Urea Esculin Trehalose Lactose Methyl-a- and b-glucoside Sucrose Mannitol Maltotriose Arabinose Glycerol Fructose Arginine

) ) ) ) ) ) ) ) ) ) ) + )

CTC 033, that presented activity of 400 AU/ml over each one. The bacteriocin was less potent over B. cereus CTC 001 and Cl. perfringens CTC 042 (200 AU/ ml). The other strains tested (B. cereus ATCC 14578, Cl. sporogenes CTC 006, Ent. faecalis ATCC 19433, Lb. plantarum TECNOLAT 434, Leuc. mesenteroides ATCC 10830, Micrococcus sp. ATCC 4698, Streptococcus sp. ATCC 25175, sulphite-reducing clostridia CTC 005, and W. viridescens CCT 0849) were resistant (0 AU/ml) to the bacteriocin-like substance produced by this culture. Bactericidal activity against Listeria has also been reported for other bacteriocins produced by Lc. lactis ssp. (Muriana 1996; Moreno et al. 2000). Many bacteriocins produced by lactic acid bacteria act against bacteria closely related to the producer organisms and also inhibit Listeria. According to Lu¨cke (2000), only few are effective against Bacillus, Clostridium, and Staphylococcus, as verified for strain CTC 204.

Sensitivity to proteolytic and lipolytic enzymes Bacteriocins possess a protein moiety, which is responsible for the inhibition of the target organisms. The loss of the antimicrobial activity after treatment with enzymes indicated the sensitivity of the active compounds secreted by the lactic acid strains. After treatment with a-chymotrypsin, ficin, papain and pronase E, the activity of the antibacterial compound produced by Lc. lactis subsp. cremoris CTC 204 decreased from 800 to 0 AU/ml. Trypsin partially inactivated (75%) this compound, since the activity was reduced from 800 to 200 AU/ml. It was completely resistant to degradation by pepsin and lipase (800 AU/ml). Nisin is an extensively characterized bacteriocin produced by some strains of Lc. lactis subsp. lactis, isolated from dairy products and applied internationally to many food products as a preservative (Hurst 1983). In a comparison of nisin with the bacteriocin of strain CTC 204, it is important to note that this compound was also inactivated by a-chymotrypsin (Jarvis & Mahoney 1969). Pepsin affects neither the activity of nisin (Hurst 1983) nor that of the bacteriocin produced by strain CTC 204. Strains of Lc. lactis ssp. isolated from vegetables were resistant to pepsin and trypsin, but were inactivated by a-chymotrypsin (Uhlman et al. 1992). The bacteriocins produced by Lc. lactis ssp. BFE 1500 isolated from cheese (Olasupo et al. 1999) and Lc. lactis subsp. lactis A164 isolated from fermented vegetables (Choi et al. 2000) were also sensitive to achymotrypsin and resistant to pepsin. Sensitivity to ficin and pronase E was reported by Moreno et al. (2000) for the bacteriocin produced by Lc. lactis subsp. lactis ATCC 11454. Lc. lactis subsp. diacetylactis S50 produces a bacteriocin-designated sensitive to pepsin, trypsin, and pronase E (Kojic et al. 1991). Lacticin 481, a bacteriocin produced by Lc. lactis 481, was completely inactivated following digestion with ficin, partly inactivated by a-chymotrypsin and pronase, but unaffected by trypsin (Piard et al. 1990). Sensitivity to trypsin is a characteristic of diplococcins but not of nisin (Davey & Richardson 1981). According to these authors, the proteolytic enzymes trypsin, pronase, and a-chymotrypsin completely inactivated diplococcin 346, produced by Lc. lactis subsp. cremoris. It is interesting to note that the pancreatic enzymes (trypsin and a-chymotrypsin) inactivated the bacteriocin-like substance produced by strain CTC 204. This is an important aspect with respect to food safety, since the digestive enzymes can destroy the bacteriocins (Caplice & Fitzgerald 1999). The destruction of the antimicrobial activity by proteases suggested that this compound could be a peptide or bacteriocin. Effect of temperature and pH on the antibacterial compound Hurdle technology combines different preservation methods to inhibit microbial growth. Bacteriocins often

Bacteriocin: properties and use in meat have synergies with other treatments and can be used as a hurdle to improve food safety (Cleveland et al. 2001). Bacteriocins present different sensitivities to changes in temperature and pH. The effects of different pH values (from 2.0 to 12.0) and refrigeration (4 °C/24 h), pasteurization (65 °C/30 min, 100 °C/10 min), and sterilization (121 °C/10 min) temperatures on the activity of Lc. lactis subsp. cremoris CTC 204 were studied. The effects of different pH values and treatment at 4 °C for 24 h on the activity of the bacteriocin under test showed that the highest level of activity presented by this strain was 3200 AU/ml at pH 2.0 and 4.0. It lost 50% of its activity from pH 6.0 to 9.0 and 75% at pH 10.0. It was completely destroyed at pH 12.0 (0 AU/ml). In comparison to refrigeration temperatures, heat treatment at 65 °C for 30 min caused a slight loss of activity of the antimicrobial compound produced by strain CTC 204. The maximum activity value produced after heat treatment was at least 50% lower than that produced at 4 °C. At pH 2.0 it produced 1600 AU/ml; from pH 4.0 to 9.0 the production was 800 AU/ml; at pH 10.0 it was 400 AU/ml and at pH 12.0 no activity was detected. After heat treatment at 100 °C for 10 min the highest level of activity presented by the bacteriocin was 1600 AU/ml at pH 2.0. Up to pH 6.0 it presented the same pattern of production as that produced after heat treatment at 65 °C for 30 min: 800 AU/ml. After that, from pH 7.0 to 10.0 it lost 50% of its activity (400 AU/ ml) and was subsequently inactivated at pH 12.0. The effect of pH and sterilization temperature (121 °C for 10 min) on the activity of the antimicrobial compound showed that even at high temperatures it was still active in the pH range from 2.0 to 10.0. It maintained maximum bacteriocin production (1600 AU/ml) from pH 2.0 to 6.0 and lost 50% of its activity (800 AU/ml) from pH 7.0 to 10.0, being completely destroyed at pH 12.0. The activity level of this bacteriocin was maintained during all heat treatments under acid conditions. Many bacteriocins are active at acidic, neutral and alkaline pH values, which may reflect the adaptation of these substances to the environmental conditions in which bacteriocin-producer bacteria develop (Kojic et al. 1991; Olasupo et al. 1999). Nevertheless, nisin and lactostrepcins (produced by lactococci) are exceptions to this rule, since their antimicrobial activities are dependent on this parameter. Solubility and stability of the former decrease from optimal at pH 2.0 to considerably reduced at 6.0 and irreversibly inactivated at pH 7.0 (Hurst 1981); on the other hand, lactostrepcins are stable and active within the pH range from 4.2 to 5.0 and reversibly inactivated at pH 7.0 and 8.0 (Kozak et al. 1978). However, many bacteriocin-producing lactic acid bacteria withstand exposure to a wide range of pH values (3.0–9.0) (Uhlman et al. 1992; Cintas et al. 2001). Tolerance to more extreme pH values (between 1.0–2.0 and 10.0–11.0) has been reported for acidocin B (produced by Lb. acidophilus M46) (Ten Brink et al. 1994), bavaricin A (produced by Lb. bavaricus MI401)

355 (Larsen et al. 1993), and the bacteriocin produced by Lc. lactis subsp. diacetylactis (Kojic et al. 1991). It is interesting that the applicability of these compounds in foods, especially in meat products, depends on some of the characteristics found in the bacteriocin produced by this strain. The bacteriocin produced by strain CTC 204 was active in a wide range of pH values and temperatures of refrigeration, pasteurization and sterilization. Bacteriocin production and bacterial growth at different temperatures and pH values The potential for stability at different pH values and temperature conditions was evaluated for cell growth and bacteriocin production by strain CTC 204 in buffered MRS Broth (Table 3). Cell growth and bacteriocin production properties were examined for 162 h in flask cultures at
20, 4, 25 and 37 °C under noncontrolled pH conditions. The freezing temperature did not allow for growth and subsequent bacteriocin production by the strain tested. At this temperature, frozen food could be preserved by the direct addition of the bacteriocin produced by this strain. At the optimum temperature for the growth of lactic acid bacteria (37 °C), cell growth was 58.9% greater after 16 h. After 48 h of incubation at this temperature, it presented the highest biomass production. After this, until 162 h cell density decreased by 12.6%. At the abusive temperature for food storage (25 °C), the bacterial density increased 58.7% after 16 h of incubation. The maximum growth at this temperature was achieved after 24 h of incubation and it was 17.1% lower than at 37 °C. At 4 °C, cell density presented a slower growth rate during the first 24 h. From that point on, accentuated growth was observed, reaching a peak at 162 h of incubation. At 0 °C, the initial density was reduced by 9.5% after the first 24 h, gradually decreasing until 162 h. Cell growth increased with temperature, reaching a maximum at 37 °C. At 4, 25, and 37 °C the maximum bacteriocin production by strain CTC 204 was 800 AU/ml. At 25 and 37 °C, maximum bacteriocin production was detected after 16 h when the bacterial growth was approximately 2.4 times higher than the initial value. At )20 °C the maximum bacteriocin production occurred after 20 h of incubation when the cell growth was 1.2 times higher than at the beginning. Bacteriocin production decreased 50% after 24 h of incubation at 4 and 37 °C and after 48 h of incubation at 25 °C. During the period of incubation the pH decreased about 2 units at 25 and 37 °C, while at 4 °C it decreased 1.3 units. Due to an absence of growth at )20 °C, no variation in pH value was observed. According to Hugas et al. (1998), the maximum sakacin K production by Lb. sake CTC 494 was detected after 7 days at 4 °C, while at 25 °C it was detected after 9 h. The rate of cell growth increased according to the temperature, reaching a maximum at 30 °C, but bacteriocin production was higher at the lower temperatures.

356

R. Bromberg et al.

Table 3. Growth, bacteriocin-like substance production and pH resistance of Lc. lactis subsp. ceremoris CTC 204 incubated in different temperatures. 37 °C

Time (h)

0 16 18 20 22 24 48 114 162

25 °C

4 °C

0 °C

ODa

BAb

pH

OD

BA

pH

OD

BA

pH

OD

BA

pH

0.281 0.684 0.772 0.768 0.742 0.726 0.808 0.738 0.706

0 800 800 800 800 400 400 200 0

6.43 4.60 4.58 4.54 4.52 4.49 4.8 4.47 4.46

0.264 0.640 0.636 0.624 0.616 0.670 0.644 0.562 0.546

0 800 800 800 800 800 400 400 0

6.43 5.08 5.00 4.94 4.86 4.80 4.68 4.52 4.46

0.292 0.343 0.354 0.362 0.398 0.412 0.588 0.636 0.666

0 0 400 800 800 400 400 400 0

6.46 6.30 6.25 6.20 6.14 6.10 5.82 5.31 5.16

0.326 – – – – 0.295 0.276 0.243 0.240

0 – – – – 0 0 0 0

6.34 – – – – 6.34 6.32 6.36 6.35

– not detected. Absorbance at 600 nm. b Bacteriocin activity (AU/ml). a

However, Yang & Ray (1994) reported that the amount of bacteriocin produced by a lactic acid bacterium isolated from vacuum-packaged meat products was 2–3 times higher at the abusive temperature (25 °C) than at refrigeration temperature (4 °C). Effect of the bacteriocin on the mesophilic population present in meat Raw meat stored aerobically under chilled conditions spoils predominantly due to Gram-negative bacteria, especially Pseudomonas (Lu¨cke 2000). Moreover, bacterial pathogens of greater significance to the consumer of raw meat (Salmonella sp., Campylobacter sp., Yersinia enterocolitica, Escherichia coli 0157:H7) are also Gramnegative. Under these conditions, lactic acid bacteria compete poorly and the use of the bacteriocin extract would be an alternative to control undesirable microorganisms. Figure 1 shows the inhibitory effect on aerobic mesophilic bacteria during the storage of minced beef submitted to treatment with the bacteriocin produced by strain CTC 204, confirming the results of three experiments. In these tests, ca. 500 AU/ml of bacteriocin produced by strain CTC 204 were added to the minced beef samples. The initial mesophilic counts in the meat samples varied from 2.2 to 7.9 log10 c.f.u./g (data not shown). It means that each beef sample constitutes a ‘closed system’ in which the natural microbiota (types and concentrations of the microorganisms presented) cannot be reproducible, so it is important to consider

each sample analysed as a different one. Based on this fact, a linear regression curve to express the way the mesophilic bacterium population was affected by the bacteriocin produced by strain CTC 204 was obtained. The correlation coefficient (R2) showed a value of 0.95 and the effectivity of the treatment was 7% (1)0.934) considering P ¼ (0.899 < 0.934 < 0.970) ¼ 95% as confidence interval. The bacteriocin activity was not influenced by the pH value once it decreased approximately 0.75 units in both control and test samples during the experiment. According to Muriana (1996), the effectiveness of bacteriocins may also depend on the amount of bacteriocin inactivated by interaction with food components. Because of their relatively large size (4–8 kDa), bacteriocins may also be considered a finite population of macromolecular inhibitors and, therefore, the relative amount of bacteriocin required to inactivate target cells may depend on the population of cells that may be present. There is little information on the effect of lactic acid culture treatments and their bacteriocins on the mesophilic aerobic plate counts of meat during refrigerated storage. Cell suspensions of Lc. lactis subsp. lactis biovar. diacetylactis have been used to inhibit the growth of gram-negative bacterial populations in refrigerated ground beef (Daly et al. 1972). Meat stored in air is rapidly spoiled by bacteria, which are responsible for discoloration and production of off-odours, resulting in consumer rejection (Labadie 1999). In our study, the colour of both the treated and control samples was

Figure 1. Effect of the bacteriocin produced by Lc. lactis subsp. cremoris CTC 204 on the mesophilic counts of minced beef during refrigerated storage.

Bacteriocin: properties and use in meat verified during the storage period. The appearance of a green colour was detected in the samples dipped in water at the end of the storage time. The green fluorescent pigment (pyoverdin) is produced by many fluorescent species of Pseudomonas (Labadie 1999). Generally, the storage life depends largely on the bacterial counts of the meat at the beginning of the storage period, notably the proportion of Pseudomonas spp. within the flora (Dainty & Mackey 1992). The results of this study revealed that the bacteriocin produced by Lc. lactis subsp. cremoris CTC 204 was inhibitory towards the mesophilic aerobic bacteria associated with red meat. A low concentration of bacteriocin reduced the number of bacteria present by 1 cycle log. This difference was found in some heavily contaminated beef samples, being considered satisfactory. Considering its bactericidal activity, proteinaceous nature and heat resistance, the antimicrobial compound produced by Lc. lactis subsp. cremoris CTC 204 can be classified as a bacteriocin. It presented characteristics of resistance to proteolytic enzymes and temperatures similar to those of nisin. It was thermostable and could thus be used in pasteurized food products. It was also stable over a wide range of pH values and could be employed in acid and non-acid foods. As it was stable at low temperatures, the microorganism could act as a potential barrier to inhibit the growth of psychrotrophic or mesophilic spoilage and foodborn pathogens, such as Lactobacillus ssp., L. monocytogenes, S. aureus, B. cereus, and Cl. perfringens, frequently found in foods stored under refrigeration. Further work to evaluate the potential of this compound in meat after purification is in progress.

Acknowledgements The authors thank FAPESP (Fundac¸a˜o de Amparo a Pesquisa do Estado de Sa˜o Paulo) for their financial support (Process 99/12314-0) and Homero F. Gumerato for the statistic analysis.

References Barakat, R.K., Griffiths, M.W. & Harris, L.J. 2000 Isolation and characterization of Carnobacterium, Lactococcus, and Enterococcus spp. from cooked, modified atmosphere packaged, refrigerated, poultry meat. International Journa1 of Food Microbiology 62, 83–94. Caplice, E. & Fitzgerald, G.F. 1999 Food fermentations: role of microorganisms in food production and preservation. International Journal of Food Microbiology 50, 131–149. Choi, H.J., Cheigh, C.I., Kim, S.B. & Pyun, Y.R. 2000 Production of a nisin-like bacteriocin by Lactococcus lactis subsp. lactis A164 from Kimchi. Journal of Applied Microbiology 88, 563–571. Cintas, L.M., Casaus, M.P., Herranz, C., Nes, I.F. & Hera´ndez, P.E. 2001 Review: bacteriocins of lactic acid bacteria. Food Science and Technology International 7, 281–305. Cleveland, J., Montville, T.J., Nes, I.F. & Chikindas, M.L. 2001 Bacteriocins: safe, natural antimicrobials for food preservation. International Journal qf Food Microbiology 71, 1–20.

357 Dainty, R.H. & Mackey, B.M. 1992 The relationship between the phenotypic properties of bacteria from chilled-stored meat and spoilage processes. Journal of Applied Bacteriology (Symposium Supplement) 73, 103S–114S. Daly, C., Sandine, C.E. & Elliker, P.R. 1972 Interactions of food starter cultures and foodbome pathogens: Streptococcus diacetylactis versus food pathogens. Journal of Food Technology 35, 349–357. Davey, G.P. & Richardson, B.C. 1981 Purification and some properties of diplococcin from Streptococcus cremoris 346. Applied and Environmental Microbiology 41, 84–89. Eckner, K.F. 1992 Bacteriocins and food applications. Dairy, Food and Environmental Sanitation 12, 204–209. Elortondo, F.J.P., Albisu, M. & Barcina, Y. 1999 Changes in the microflora of Idiaza´bel cheese with the commercial lactic starters. Australian Journal of Dairy Technology 48, 10–13. Garver, K.I. & Muriana, P.M. 1993 Detection, identification and characterization of bacteriocin-producing lactic acid bacteria from retail food products. International Journal of Food Microbiology 19, 241–258. Harrigan, W.F. & McCance, M.E. 1976 Basic methods In Laboratory Methods in Food and Dairy Microbiology eds. Harrigan, W.F. & McCance, M.E. pp. 1–115. London: Academic Press. ISBN 0-12326040-X Hugas, M. 1998 Bacteriocinogenic lactic acid bacteria for the biopreservation of meat and meat products. Meat Science 49 (Suppl. 1), S139–S150. Hugas, M., Pages, F., Garriga, M. & Monfort, J.M. 1998 Application of the bacteriocinogenic Lactobacillus sakei CTC494 to prevent growth of Listeria in fresh and cooked meat products packed with different atmospheres. Food Microbiology 15, 639–650. Hurst, A. 1981 Advances in Applied Microbiology Vol. 27. pp. 85–123. New York: Academic Press. Inc. ISBN 0-12-002627-9. Hurst, A. 1983 Antimicrobials in Foods. pp. 327–350. New York: Marcel Dekker Inc. ISBN 0-78-377397-8. Jarvis, B. & Mahoney, R.R. 1969 Inactivation of nisin by alpha chymotrypsin. Journal of Dairy Science 52, 1448–1450. Kojic, M., Svircevic, J., Banina, A. & Topisirovic, L. 1991 Bacteriocinproducing strain of Lactococcus lactis subsp. diacetylactis S50. Applied and Environmental/Microbiology 57, 1835–1837. Kozak, W., Bardoniski, J & Dobrzanski, W.T. 1978 Lactostrepcinsacid bacteriocins produced by lactic streptococci. Journal of Dairy Research 45, 247–257. Labadie, J. 1999 Consequences of packaging on bacterial growth. Meat is an ecological niche. Meat Science 52, 299–305. Larsen, A.G., Vogensen, F.K. & Josephsen, J. 1993 Antimicrobial activity of lactic acid bacteria isolate from sour doughs: purification and characterization of bavaricin A, a bacteriocin produced by Lactobacillus bavaricus MI401. Journal of Applied Bacteriology 75, 113–122. Lu¨cke, F.-K. 2000 Utilization of microbes to process and preserve meat. Meat Science 56, 105–115. Mauguin, S. & Novel, G. 1994 Characterization of lactic acid bacteria isolated from seafood. Journal of Applied Bacteriology 76, 616–625. Mayr-Harting, A., Hedges, AJ. & Berkeley, R.C.W. 1972 Methods for studying bacteriocins. In Methods in Microbiology. Vol. 7A. eds. Norris, J.R. & Ribbons, D.W. pp. 315–422. New York: Academic Press Inc. ISBN 0-12-521546-0. McMullen, L.M. & Stiles, M.B. 1996 Potential for use of bacteriocinproducing lactic acid bacteria in the preservation of meats. Journal of Food Protection (Supplement), 64–71. Moreno, I., Lerayer, A.L.S., Baldini, V.L.S. & Leita˜o, M.F.F. 2000 Characterization of bacteriocins produced by Lactococcus lactis strains. Brazilian Journal of Microbiology 31, 184–192. Muriana, P.M. 1996 Bacteriocins for control of Listeria spp. in food Journal of Food Protection Supplement. 54–63. Olasupo, N.A., Schillinger, U., Narbad, A., Dodd, H. & Holzapfel, W.H. 1999 Occurrence of nisin Z production in Lactococcus lactis BFE 1500 isolated from wara, a traditional Nigerian cheese product. International Journal of Food Microbiology 53, 141–152.

358 Piard, J.C., Delorme, F., Giraffa, G., Commissaire, J. & Desmazeaud, M. 1990 Evidence for a bacteriocin produced by Lactococcus lactis CNRZ 481. Netherlands Milk Dairy Journal 44, 143–158. Rodriguez, J.M., Cintas, L.M., Casaus, P., Horn, N., Dodd, H.M., Hernandez, P.E. & Gasson, M.J. 1995 Isolation of nisin-producing Lactococcus lactis strains from dry fermented sausages. Journal of Applied Bacteriology 78, 109–115. Scannell, A.G.M., Schwarz, G., Hill, C., Ross, R.P. & Arendt, E.K. 2001 Prehy inoculation enrichment procedure enhances the performance of bacteriocinogenic Lactococcus lactis meat starter culture. International Journal of Food Microbiology 64, 151–159. Schillinger, U. & Lu¨cke, F.-K. 1989 Antibacterial activity of Lactobacillus sake isolated from meat. Applied and Environmental Microbiology, 55, 1901–1906.

R. Bromberg et al. Tagg, J.R., Dajani, A.S. & Wannamaker, L.W. 1976 Bacteriocins of Gram-positive bacteria. Bacteriological Reviews 40, 722–756. Ten Brink, B., Minekus, M., Van Der Vossem, J.M.B.M., Leer, R.J. & Huis In’T Veld, J.H.J. 1994 Antimicrobial activity of lactobacilli: preliminary characterization and optimization of production of acidocin B, a novel bacteriocin produced by Lactobacillus acidophilus M46. Journal of Applied Bacteriology 77, 140–148. Uhlman, L., Schillinger, U., Rupnow, J.R. & Holzapfel, W.H. 1992 Identification and characterization of two bacteriocin-producing strains of Lactococcus lactis isolated iTom vegetables. International Journal of Food Microbiology 16, 141–151. Yang, R. & Ray, R. 1994 Prevalence and biological control of bacteriocin-producing psychrotrophic Leuconostocs associated with spoilage of vacuum-packaged processed meats. Journal of Food Protection 57, 209–217.

World Journal of Microbiology & Biotechnology (2005) 21: 359–365 DOI 10.1007/s11274-004-2611-8


Springer 2005

The effect of salinity on trichloroethylene co-metabolism by mixed cultures enriched on phenol Chi-Yuan Lee*, Yu-Chia Chan and Chin-Lung Lin Water Resources and Environmental Engineering Program, Department of Harbor and River Engineering, National Taiwan Ocean University, Keelung 20224, Taiwan, R. O. C. *Author for correspondence: Tel.:+886-2-2462-2192 ext. 6147, Fax: +886-2-24624770, E-mail: cylee@mail. ntou.edu.tw Received 19 March 2004; accepted 18 August 2004

Keywords: Biodegradation, phenol, phenol-oxidizing microorganisms, salinity, tolerance, transformation capacity, trichloroethylene

Summary This work examines the effects of salinity on the biodegradation of trichloroethylene (TCE) by four chemostatcultivated cultures: LHPO-3, LHPO-6, HHPO-3 and HHPO-6, all of which had been enriched on phenol but grown under different conditions. Cultures LHPO-3 (with hydraulic retention time [HRT] of 3.1 days) and LHPO-6 (6.5day HRT) were cultivated with fresh water, whereas cultures HHPO-3 (3.3-day HRT) and HHPO-6 (6.1-day HRT) were cultivated with seawater. Batch tests of TCE degradation by the four bacterial consortia in the absence of phenol were undertaken in solutions with salinities in the range 0–3.28% (w/v). Moreover, the effect of adding phenol on TCE degradation by LHPO-3 in 1.64% salinity solution was investigated. The results showed that the observed bacterial yields for the cultures LHPO-3, LHPO-6, HHPO-3 and HHPO-6 were 0.66, 0.47, 0.58 and 0.33 mg volatile suspended solids/mg phenol, respectively. In the absence of phenol, the extents of TCE degradation by cultures LHPO-3 and LHPO-6 increased with salinity stress, reaching 0.052 mg TCE/mg VSS for LHPO-3 and 0.033 mg TCE/mg VSS for LHPO-6, and then declined as salinity increased further. The tolerance of TCE degradation to salinity for culture LHPO-3 was around 3.28% and that for LHPO-6 was 1.64–2.33%. In the presence of phenol, the rate and extent of TCE degradation by LHPO-3 were enhanced when an optimal dosage of phenol of 10 mg phenol/mg TCE was applied. Degradation of TCE by cultures HHPO-3 and HHPO-6 was not observed.

Introduction Trichloroethylene (TCE), a chlorinated chemical widely found in industrial discharges and groundwater, represents a serious human health risk because it is carcinogenic. Most biological mechanisms for the removal of TCE usually employ co-metabolic biodegradation, in which a growth substrate, such as phenol, toluene or methane, is used to cultivate bacteria and thus induce the required nonspecific enzyme to catalyse TCE oxidation. The use of phenol as the inducing substrate is beneficial, primarily because higher rates of TCE transformation can be achieved than can be obtained using other substrates (Hopkins et al. 1993; Shurtliff et al. 1996; Semprini 1997). TCE undergoing co-metabolic transformation in the saline solutions frequently occurs in engineered and natural systems. For instance, when TCE vapour is treated using a biofilter, salinity accumulates in the liquid phase, because of acid–base reactions between externally added sodium hydroxide (NaOH) and hydrogen chloride (HCl) that is released during TCE breakdown, a phenomenon similar to that

observed during the treatment of dichloromethane vapour (Diks et al. 1994). Additionally, during biofiltration of the waste gas, sodium chloride may be added to prevent excessive accumulation of biomass (Scho¨nduve et al. 1996). Other phenomena include groundwater contamination with TCE in coastal aquifers, where seawater intrudes. The effects of salinity on environmental biological processes have recently attracted much attention, but the emphasis has been on the metabolism of growth substrates (Hinteregger & Streichsbier 1997, BromleyChallenor et al. 2000; Margesin & Schinner 2001). Woolard & Irvine (1994) reported that heterotrophic halophilic organisms in sequencing batch biofilms removed 99% of phenol from 15% NaCl saline wastewater. Kargı` & Dinc¸er (1999) stated that when saline organic wastewater was treated in rotating biological contactors, salt inhibition occurred at 2% NaCl. However, the detrimental effects of high salinity were greatly improved using salt-tolerant microorganisms (Halobacter) mixed with activated sludge cultures (Kargı` & Dinc¸er 1996a, b). Dan et al. (2003) compared the performance

360 of the aerobic treatment of organic-high salinity wastewater with yeast to that by bacterial cultures, and found that the yeast culture was more efficient. Diks et al. (1994) found that the inhibition of microbial growth increased with salt concentration, but that the degradation of the substrate was much less inhibited by NaCl during dichloromethane degradation. To the best of our knowledge, no study has addressed the co-metabolic behaviour of microorganisms in response to salinity. This work studies the effect on TCE degradation of transient salinity to phenol-oxidizing microorganisms cultivated with fresh water. Additionally, cultures cultivated with seawater, representing those under longterm salinity stress, were investigated to evaluate their ability to degrade TCE.

Materials and methods Chemicals TCE (GR grade) was purchased from MercK Co. (Merck Taiwan Ltd.). TCE-saturated water solution was prepared by injecting 10 ml of pure TCE into a 120 ml vial, which contained 80 ml of distilled water and was capped with a Teflon-lined rubber septum. The vial was vigorously shaken for 1 min and the contents allowed to settle for 2 h before use. The saturated TCE concentration in the vial was 1100 mg/l. Phenol with purity over 99% was purchased from Yakuri Pure Chemical Co., Ltd. (Japan). The phenol-saturated stock solution was prepared by placing 35 ml of liquid phenol in distilled water in a 500 ml serum bottle. The bottle was sealed, vigorously shaken for 5 min, and the contents allowed to settle for 2 h before use. The average phenol concentration in the bottle was 8100 mg/l. Sodium chloride (NaCl purity of 99.63%) used to establish the salinity of artificial feeding seawater was obtained from the Taiwan Salt Industrial Corporation. Cultivation of mixed cultures with fresh water Mixed bacterial consortia enriched on phenol were cultivated using a chemostat reactor (Reactor A) fed with fresh water. These cultures were used for investigating the transient effects of salinity on TCE cometabolic transformation. The reactor was incubated at room temperature and had a volume of 1.75 l; it was 30 cm high and 10 cm in diameter, and was equipped with a magnetic bar used to stir the contents at 600 rev/ min to ensure complete mixing. The solution fed to the reactor that contained 1000 mg of phenol/l as the sole carbon source was stored in a refrigerator at 4 C. The artificial influent was prepared by diluting phenolsaturated stock solution with distilled water, and supplemented with nutrients and trace elements (measured per g phenol): K2HPO3 Æ 3H2O 4.25 g, NaH2PO4 Æ H2O 1.00 g, NH4Cl 2.00 g, MgSO4 Æ 7H2O 0.20 g, FeSO4 Æ

C.-Y. Lee et al. 7H2O 0.012 g, MnSO4 Æ H2O 0.003 g, Zn SO4 Æ 7H2O 0.003 g and CoSO4 Æ 7H2O 0.001 g. Cultivations were initiated using inocula from a biofilm reactor (Lee & Cheng 1998; Lee & Lan 1999) that had been operated in the authors’ laboratory for 5 years. In the first phase of the investigation, the reactor was discharged at an average flow rate of 561 ml/day, yielding a 3.1-day hydraulic retention time (HRT). The reactor was maintained in a steady state for a period of 4 months while TCE degradation tests were conducted. The cultivated bacterial consortia in this phase were called low-halotolerant phenol-oxidizing microorganisms, LHPO-3. To examine the effect of operating HRT on bacterial activity in degrading TCE, the reactor was subsequently changed to a 6.5-day HRT operation, after batch tests on TCE degradation by culture LHPO-3 had been finished. In the second phase of the study, the reactor was operated by the sane procedure as used to cultivate the culture LHPO-3, except that the flow rate was reduced to about 280 ml/day. The enriched bacteria in this phase are denoted by LHPO-6. Cultivation of mixed cultures with saline water To examine the behaviour of phenol-oxidizing microorganisms in degrading TCE when the cultures adapted to the saline environment, a parallel reactor (Reactor B) fed with seawater was operated. The seeding inocula of Reactor B were taken from Reactor A. In the first phase of the study, the reactor was discharged with saline influent prepared by placing phenol-saturated stock solution in seawater, to yield a phenol concentration of 1000 mg/l. The supplemental nutrients to the influent included (per g of phenol) K2HPO3 Æ 3H2O 4.25 g, NaH2PO4 Æ H2O 1.00 g, NH4Cl 2.00 g. The average flow rate to the reactor was 526 ml/day, yielding an HRT of 3.3 days; the cultures thus enriched were called HHPO-3. In the second phase of the study, the feeding solution was changed to artificial seawater with 3.28% salinity and 1000 mg of phenol/l, which was prepared by adding minerals (as for cultivating LHPO cultures), sodium chloride and phenol-saturated solution to distilled water. The flow rate in this phase was reduced to 293 ml/day, yielding a culture called HHPO-6. The procedure for cultivating HHPO cultures in both phases was the same as that for cultivating LHPO cultures. During the growth of the four cultures, the reactor performance was generally monitored weekly by sampling from the influent and the effluent. The measurement parameters included phenol, biomass, dissolved oxygen, pH and salinity. TCE biodegradation Batch tests of TCE biodegradation were started without adding phenol as a growth substrate. The tests were performed using serum bottles, each with a total volume of 122 ml, a liquid volume of 80 ml and a headspace of

361

Salinity and trichloroethylene co-metabolism 42 ml. Prior to each test, fresh bacterial solution was taken from a reactor that had been maintained at the steady state. The bacterial solution was concentrated by centrifugation; the supernatant was decanted, and the settled bacteria were diluted with a solution of nutrients but without phenol, to achieve a bacterial concentration of 500 mg volatile suspended solids (VSS) per liter. A volume of 8 ml of this prepared cultural solution was poured into the serum bottle, and was immediately diluted with a mixture of distilled water and seawater, until the liquid volume was 80 ml, yielding a bacterial concentration of 50 mg VSS/l. The proportion of distilled water and seawater in the mixtures depended on the salinity required. A series of salinities (w/v) for the test set were 0, 0.33, 0.98, 1.64, 2.3 and 3.28%. Then, 175 ll of TCE-saturated solution was injected into each serum bottle that had been sealed with 3 mm teflonlined rubber septa and crimp-top caps, yielding an initial TCE concentration of 2 mg/l. After the TCE had been injected, the serum bottle was then mixed vigorously using a magnetic stir bar at 600 rev/min, to ensure that TCE equilibrium existed between the liquid and the headspace. These serum bottles were incubated at 25 C in a temperature-controlled room. Duplicate headspace samples were taken during the incubation. Phenol was added to culture LHPO-3 at 1.64% salinity to study the improvement of TCE degradation in saline water. The series of phenol dosages was 0, 0.1, 1, 5, 10, 20, 50 and 100 mg/l. The experimental procedure was the same as that used for testing in the absence of phenol.

matter, or VSS, as is commonly used in environmental engineering (Gaudy & Gaudy 1980; Rittmann & McCarty 2001). Values of VSS were measured using an ultraviolet spectrophotometer (Cecil CE1021) at a wavelength of 610 mm, calibrated with dry cell mass. The dry biomass was determined gravimetrically from the difference between the cell mass at 103 C overnight and that after combustion at 550 C for 1 hour (APHA 1998). The concentration of dissolved oxygen and salinity were obtained using a YSI 52 meter (Yellow Springs Instruments, USA). Phenol was analysed by HPLC (Shimadzu, LC10A) equipped with an ODS-80 TM column (Tosoh) and a u.v. detector (Shimadzu, SPD-10A). The mobile phase was a mixture of methanol:distilled water (3:1) at a flow rate of 1 ml/min. TCE concentration was determined from the headspace, and a 10 ll TCE gas sample was withdrawn from the vial using a gas-tight syringe and injected into a gas chromatograph (Hewlett Packard 6890). TCE was detected using an electron capture detector (ECD), in which the capillary column was Supelcowax 10 (30 m, 5.3 mm, and 1.00 lm), and the carrier gas was nitrogen with a flow rate of 5 ml/min. The temperature of the oven was maintained at 85 C for 3 min, the injection port was maintained at 250 C, and the ECD was at 300 C. The dimensionless Henry’s constants at 25 C and the various salinities were 0.36 (for salinities 0– 0.33%), 0.38 (0.98%), 0.40 (1.64%). 0.41 (2.3%) and 0.44 (3.28%), which values were used to compute the TCE aqueous concentrations (Schwarzenbach et al. 1993; Dewulf et al. 1995).

Calculations of the rate and extent of TCE biodegradation

Results

The rate of TCE degradation is expressed as the mass of TCE removed during the first 12 h of incubation per initial mass of microorganisms present in the serum bottle. The extent of TCE biodegradation is determined from the transformation capacity (mg TCE/mg VSS), Tc, as proposed by Alvarez-Cohen & McCarty (1991). The transformation capacity is defined as Tc ¼

DS DX

ð1Þ

where DS is the mass of TCE transformed and DX is the biomass inactived . This expression indicates that cells are killed by the TCE transformation, because the degradation intermediates of the TCE epoxide are toxic to the cultures. In Equation (1) the cells killed by bacterial decay are neglected since under a batch system the product toxicity is considerably greater than cell decay (Chang & Alvarez-Cohen 1995). Analyses In this study, the mass of microorganisms was determined by measuring the amount of suspended organic

Chemostat operations During TCE degradation testing, Reactors A and B were maintained in steady states. Table 1 presents the average steady-state performances of the two reactors in each study phase. The effluent phenol concentrations were in the range 2–4 mg/l, indicating that a high removal efficiency had been achieved, regardless of HRT and the salinity of the solutions. The observed bacterial yields for LHPO-3 and HHPO-3 were 0.66 and 0.58 mg VSS/mg phenol, respectively, and those for LHPO-6 and HHPO-6 were reduced to 0.47 and 0.33 mg VSS/mg phenol, respectively. The above results revealed that biomass production decreased as the salinity increased and that, at a given salinity, the biomass production decreases as HRT increases. TCE degradation in the absence of phenol Figure 1 plots TCE degradation by LHPO-3 at an initial concentration of approximately 2 mg/l at salinities of 0– 3.28%. The disappearance of TCE varies with salinity, and all residual TCE concentrations levelled off within 80 h of incubation. This result reveals that in each set of

362

C.-Y. Lee et al.

Table 1. Steady-state performance of chemostat reactors for cultivating phenol-oxidizing microorganisms in two phases of study. Item

Unit

Reactor A

Reactor B

Phase I Culture Flow Rate Influent phenol Effluent phenol Effluent biomass PH HRT Yobs c

ml/day mg/l mg/l mg VSS/l unit days mgVSS/mg phenol

LHPO-3 561(43) 1074(31) 4(3) 714(73) 6.6(0.5) 3.1(0.2) 0.66(0.06)

HHPO-3 526 (24) 1083(158) 3(1) 621(63) 6.0(0.2) 3.3(0.1) 0.58(0.08)

Phase II Culture Flow rate Influent phenol Effluent phenol Effluent biomass PH HRT Yobs c

ml/day Mg/l Mg/l Mg VSS/l unit days mgVSS/mg phenol

LHPO-6 280 (48) 1012(18) 2.4(1.0) 481(38) 5.9(0.2) 6.5(2.1) 0.47(0.04)

HHPO-6 293 (45) 1026(33) 2.0(1.0) 342(21) 5.7(0.3) 6.1(1.1) 0.33(0.02)

a

Abbreviations for item are: LHPO-3 (low halotolerant phenol oxidizers cultivated with fresh water in Phase I), HHPO-3 (highly halotolerant phenol oxidizers cultivated with seawater in Phase I), LHPO-6 (low halotolerant phenol oxidizers cultivated with fresh water in Phase II), HHPO-6 (highly halotolerant phenol oxidizers cultivated with artificial seawater in Phase II), and HRT (hydraulic retention time). b Numbers in parenthesis are standard errors. c Parameter Yobs represents observed bacterial yield coefficient.

tests, 50 mg VSS/l bacteria that were originally present in the serum bottle were dead, because co-metabolic transformation of TCE could not continue. Figure 2 further plots the computed TCE degradation rate and transformation capacity Tc as functions of salinity. The figure shows that the highest Tc of 0.052 mg TCE/mg VSS was observed at a salinity of 1.64%. The value of Tc declined as salinity increased further, falling finally to 0.02 mg TCE/mg VSS at a salinity of 3.28%. Similarly, the degradation rates also increased with salinity, to a maximum of 0.067 mg TCE/mg VSS-day at 1.64% salinity, before decreasing to 0.04 mg TCE/mg VSS-day as the salinity increased to 3.28%. The values of Tc and the removal rate at a salinity of 3.28% are close to the levels at 0% salinity, so the tolerance of culture LHPO-3 to transient salinity stress is around 3.28%. The degradation by culture LHPO-6 of TCE was less stimulated by applied salinity stress than was culture LHPO-3, so the former culture was less tolerant to osmolarity. The values of Tc increased from 0.029 mg TCE/mg VSS at 0% salinity to the highest value of only 0.033 mg TCE/mg VSS at 0.98% salinity. It fell to 0.03 mg TCE/mg VSS at 1.64% salinity. In a similar trend, the TCE removal rate at zero salinity was 0.053 mg TCE/mg VSS-day, increasing to a maximum of 0.063 mg TCE/mg VSS-day at a salinity of 1.64%, but decreasing to 0.04 mg TCE/mg VSS-day at a salinity of 2.3%. At salinities 1.64–2.3%, the values of Tc and the rate of TCE degradation equaled those at 0% salinity, suggesting that culture LHPO-6 can tolerate salinity levels of 1.64–2.3%.

When cultures LHPO-3 and LHPO-6 were placed in seawater and allowed to adapt to such a saline environment, they lost their ability to degrade TCE, although they remained active in consuming phenol. The experimental results revealed that the salt-tolerant HHPO-3 (Figure 3) and HHPO-6 microorganisms gave negligible TCE degradations within 96 h at an initial TCE concentration of 2 mg/l. TCE degradation in the presence of phenol Figure 4 depicts TCE degradation by LHPO-3 in the presence of phenol. The experiment was conducted in 1.64% salinity because the bacteria exhibited a very high capacity at this salinity to degrade TCE in the absence of phenol. When less than or equal to 1 mg of phenol/l was added, the initial TCE concentration of 2.5 mg/l was reduced effectively but not completely within 20 h. The value of Tc (Figure 5) obtained in the range 0.048– 0.054 mg TCE/mg VSS remained almost the same as that obtained in tests in which phenol was absent. This finding implied that adding a little phenol slightly promoted TCE degradation. When the dosages of added phenol were increased to 5, 10 or 20 mg/l, TCE was completely degraded within 15 h, and the second spike in the degradation of TCE remained significant, greatly increasing Tc values. For example, adding 20 mg of phenol/l increased the degradation rate in the first 12 h to 0.11 mg TCE/mg VSS-day and increased the transformation capacity to 0.1 mg TCE/mg VSS. Adding more phenol delayed TCE degradation. For instance, when 50 mg of phenol/l was added, TCE degradation was delayed for 4 h, and when 100 mg of phenol/l was added, it was delayed for 12 h (Figure 4H, I). The delay of TCE degradation was attributable to competitive inhibition by phenol (Bielefeldt & Stensel 1999).

Discussion The phenol-oxidizing microorganisms cultivated herein exhibited quite different capacities to degrade phenol and TCE in response to salinity. The bacteria grown in seawater degraded phenol as effectively as those grown in fresh water. An important effect of saline influent on bacterial metabolism is the reduction in the production of biomass, as reported by Diks et al. (1994) and Varela et al. (2003). Diks et al. (1994) noted a 50% reduction of the specific maximum growth rate of Hyphomicrobium GJ21 at 200–250 mM Nacl. Varela et al. (2003) indicated that the biomass production of Corynebacterium glutamicum decreased from 1.0 to 0.7 g/l-h as the salinity increased from 270 to 1800 mOsmol/kg. Such a decrease in bacterial production with increasing salinity can be attributed to the energy spent in maintaining microbial cytoplasm isoosmosis with the medium (Oren 1999). The hydraulic retention time of the chemostat reactors affected the formation of biomass. When the chemostat reactors were operated at about 3-day HRT, they

Salinity and trichloroethylene co-metabolism

363

Figure 1. TCE biodegradation by cultures LHPO-3 (top) and LHPO-6 (bottom) in different salinity solutions in the absence of phenol. These two cultures were cultivated by chemostat reactors with fresh water, where LHPO-3 at hydraulic retention time (HRT) of 3.1 days, and LHPO-6 at HRT of 6.5 days.

Figure 2. TCE transformation capacities (Tc) and removal rates for cultures LHPO-3 (top) and LHPO-6 (bottom) as functions of salinity.

produced more biomass than when operated at approximately 6-day HRT, yielding higher values of Yobs, because the 3-day HRT in the chemostat reactors caused the bacteria to spend less energy in self respiration, leaving more energy for synthesizing new cells (Grady et al. 1999; Rittmann & McCarty 2001). Except at a very high biomass production rate, the culture LHPO-3 was even more active in degrading TCE than LHPO-6. These findings are consistent with those obtained by Lee & Cheng (1998), who compared three cultures, all cultivated with freshwater but with different HRTs. Of three cultures, the one with lowest HRT (3.8 days) had the highest Tc (0.026 mg TCE/mg VSS). Adding phenol improved TCE degradation, as indicated by the transformation capacity, Tc. However, this Tc value was high, chiefly because the added phenol led to the synthesis of extra biomass. Therefore, Tc should be adjusted, by considering the newly synthesized biomass, to reveal the true extent of TCE degradation. Table 2 provides adjusted values of Tc for the culture LHPO-3. Consider the addition of 20 mg of phenol/l for comparison: the adjusted Tc was reduced from 0.1 to 0.079 mg TCE/mg VSS; the value was overestimated by approximately 25% when the additional biomass was neglected in Tc computation.

364

C.-Y. Lee et al.

Figure 3. TCE biodegradation by culture HHPO-3 in different salinity solutions in the absence of phenol. The culture HHPO-3 was cultivated by chemostat reactor with sea water at hydraulic retention time of 3.3 days.

Figure 4. Effects of phenol supplements on TCE biodegradation by culture LHPO-3 at 1.64% salinity. Abiotic test without adding LHPO-3 was shown in (a). Phenol supplements, mg/l: (a) 0, (b) 0, (c) 0.1, (d) 1, (e) 5, (f) 10, (g) 20, (h) 50, and (i) 100.

Figure 5. TCE transformation capacities (Tc) and removal rates for culture LHPO-3 as functions of phenol supplement in the 1.64% salinity solution.

Based on current knowledge, the TCE transformation mechanisms in response to salinity stress cannot easily be elucidated. However, the findings from this study shed some light on the development of novel methods for removing TCE from saline solutions. They demonstrate that the phenol-oxidizing microorganisms grown with fresh water, such as LHPO, could degrade TCE under transient salinity stress. Thus, adding active LHPO cultures is more effective than cultivating indigenous bacteria in removing TCE in saline solutions. This approach is similar to the process of bioaugmentation, which has been proposed to detoxify hazardous chemicals (Lee 1997; Steffan 1999; Atagana 2003). However, future research efforts should seek to elucidate the changes of bacterial activity during the evolution of

Salinity and trichloroethylene co-metabolism Table 2. Adjusted TCE transformation capacities due to bacterial growth from consuming phenol during TCE degradation by culture LHPO-3. Supplemented phenol concentrationa (mg/l)

Tcb (mg TCE/mg VSS)

Adjusted Tcc (mg TCE/mg VSS)

0 0.1 1 5 10 20 50 100

0.052 0.048 0.054 0.070d 0.079d 0.100d 0.057 0.055

0.052 0.048 0.053 0.066d 0.071d 0.079d 0.034 0.023

a

The salinity in the solutions during TCE biodegradation by LHPO-3 was 1.64%. b Parameter Tc denotes transformation capacity, equal to the total TCE removal mass divided by the biomass originally placed in the serum bottles. c Parameter Tc denotes adjusted transformation capacity, equal to the total TCE removal mass divided by adjusted biomass, the combined biomass that originally added and those synthesized from the phenol supplements. The observed bacterial yield coefficient was assumed to be 0.66 mg VSS/mg phenol. d The total TCE removal mass includes second spike.

microbial adaptation to saline environments, and so explore the potential of salt-tolerant phenol-oxidizing microorganisms to perform co-metabolic transformation of trichloroethylene in saline solutions. Acknowledgement The authors would like to thank the National Science Council, Taiwan, Republic of China, for financially supporting this research under Contract No. NSC882211-E-019-012. References Alvarez-Cohen, L. & McCarty, P.L. 1991 A cometabolic biotransformation model for halogenated aliphatic compounds exhibiting product toxicity. Environmental Science and Technology 25, 1381– 1387. APHA. 1998 Standard Methods for the Examination of Water and Wastewater, 20th edn. pp. 2-54–2-60. American Public Health Association, Washington, DC, USA. ISBN 0-87553235-7 Atagana, H.I. 2003 Bioremediation of creosote-contaminated soil: a pilot-scale landfarming evaluation. World Journal of Microbiology and Biotechnololgy 19, 571–581. Bromley-Challenor, K.C.A., Caggiano, N. & Knapp, J.S. 2000 Bacterial growth on N,N-dimethylformamide: implications for the biotreatment of industrial wastewater. Journal of Industrial Microbiology and Biotechnology 25, 8–16. Bielefeldt, A.R. & Stensel, H.D. 1999 Biodegradation of aromatic compounds and TCE by a filmentous bacteria-dominated consortium. Biodegradation 10, 1–13. Chang, H.L. & Alvarez-Cohen, L. 1995 Model for the cometabolic biodegradation of chlorinated organics. Environmental Science and Technology 29, 2357–2367. Dan, N.P., Visvanathan, C. & Biswadeep Basu 2003 Comparative evaluation of yeast and bacterial treatment of high salinity wastewater on biokinetic coefficients. Bioresource Technology 87, 51–56.

365 Dewulf, J., Drijvers, D., & Langenhove, H.V. 1995 Measurement of Henry’s Law constant as function of temperature and salinity for the low temperature range. Atmospheric Environment 29, 323–331. Diks, R.M.M., Ottengraf, S.P.P. & van den Oever, A.H.C. 1994 The influence of NaCl on the degradation rate of dichloromethane by Hyphomicrobium sp. Biodegradation 5, 129–141. Gaudy, A.F. Jr. & Gaudy, E.T. 1980 Microbiology for Environmental Scientists and Engineers. pp. 225–230. McGraw-Hill Companies, Inc. ISBN 0-47-023035-8. Grady, C.P.L. Jr., Daigger, G.T. & Lim, H.C. 1999 Biological Wastewater Treatment. pp. 149–156. Marcel Dekker, Inc. ISBN 0-8247-8919-9. Hopkins, G.D., Munakata, L., Semprini, L. & McCarty, P.L. 1993 TCE concentration effects on pilot field-scale in situ groundwater bioremediation by phenol-oxidizing microorganisms. Environmental Science and Technology 27, 2542–2547. Hinteregger, C. & Streichsbier, F. 1997 Halomonas sp., a moderately halophilic strain, for biotreatment of saline phenolic waste-water. Biotechnology Letters 19, 1099–1102. Kargı` , F. & Dinc¸er, A.R. 1996a Enhancement of biological treatment of saline wastewater by halophilic bacteria. Bioprocess Engineering. 15, 51–58. Kargı` , F. & Dinc¸er, A.R. 1996b Effects of salt concentration on biological treatment of saline wastewater by fed-batch operation. Enzyme and Microbial Technology 19, 529–537. Kargı` , F. & Dinc¸er, A.R. 1999 Salt inhibition effects in biological treatment on saline wastewater in RBC. Journal of Environmental Engineering. 125, 966–971. Lee, C.Y. & Cheng, S.Z. 1998 Trichloroethylene biodegradation by phenol-oxidizing cultures grown from various conditions. Journal of Environmental Science and Health B33, 705–721. Lee, C.Y. & Lan, Y.W. 1999 Competitiveness evaluation for individual growth in hybrid processes. Journal of Environmental Engineering. 125, 146–152. Lee, C.Y. 1997 Discussion: how input active biomass affects sludge age and process stability. Journal of Environmental Engineering. 123, 101–103. Margesin, R. & Schinner, F. 2001 Potential of halotolerant and halophilic microorganisms for biotechnology. Extremophiles 5, 73–83. Oren, A. 1999 Bioenergetic aspects of halophilism. Microbiology and Molecular Biology Reviews 63, 334–348. Rittmann, B.E. & McCarty, P.L. 2001 Environmental Biotechnology: Principles and Applications. pp. 13–14, pp. 130–132. McGraw-Hill Companies, Inc. ISBN 0-07-234553-5. Schwarzenbach, R.P., Gschwend, P.M. & Imboden, D.M. 1993 Environmental Organic Chemistry. pp. 120–121. John Wiley & Sons, Inc. ISBN 0-471-83941-8. Scho¨nduve, P., Sa´ra, M. & Friedl, A. 1996 Influence of physiologically relevant parameters on biomass formation in a trickle-bed bioreactor used for waste gas cleaning. Applied Microbiology Biotechnology 45, 286–291. Semprini, L. 1997 Strategy for the aerobic co-metabolism of chlorinated solvents. Current Opinion in Biotechnology 8, 296–308. Shurtliff, M.M., Parkin, G.F., Weathers, L.J. & Gilson, D.T. 1996 Biotransformation of trichloroethylene by a phenol-induced mixed culture. Journal of Environmental Engineering 122, 581–589. Steffan, R.J., Walsh, M.T., Vainberg, S. & Condee, C.W. 1999 Fieldscale evaluation of in situ bioaugmentation for remediation of chlorinated solvents in groundwater. Environmental Science and Technology 33, 2771–2881. Varela, C., Agosin, E., Baez, M., Klapa, M. & Stephanopoulos, G. 2003 Metabolic flux redistribution in Corynebacterium glutamicum in response to osmotic stress. Applied Microbiology Biotechnology 60, 547–555. Woolard, C.R. & Irvine, R.L. 1994 Biological treatment of hypersaline wastewater by a biofilm of halophilic bacteria. Water and Environmental Research 66, 230–235.


Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 367–370 DOI 10.1007/s11274-004-5957-z

Microbial decolorization of reactive azo dyes under aerobic conditions K.M. Kodam*, I. Soojhawon, P.D. Lokhande and K.R. Gawai Biochemistry Section, Department of Chemistry, University of Pune, Pune 411 007, India *Author for correspondence: Tel.: +91-20-25691728, Fax: +91-20-25691728, E-mail: [email protected] Received 21 August 2004; accepted 6 November 2004

Keywords: Aerobic, bioremediation, decolorization, sulfonated azo dyes

Summary Kodam et al. reported a 100% decolorization of the sulfonated azo dyes Reactive Red 2, Reactive Red 141, Reactive Orange 4, Reactive Orange 7 and Reactive Violet 5 by an unidentified bacterium, KMK 48. High effectiveness was attained within 36 h of incubation at room temperature and neutral pH. Optimum decolorization took place strictly under aerobic conditions, which is contrary to other well-documented reports. Thus, this microorganism seems to be potentially effective for bioremediation of textile-dyeing industry effluents.

Introduction Azo dyes are the largest group of man-made chemicals used in the textile and paper-printing industries, and they are usually recalcitrant to conventional wastewater treatment. Physico-chemical treatment methods are not economically feasible and produce large amounts of sludge. Thus, more studies are now focused on their biodegradation (Yang et al. 2003). The prerequisite for the complete mineralization of azo dyes is a combination of reductive and oxidative steps (Tan et al. 1999). Decolorization of azo dyes normally begins with initial reduction or cleavage of azo bond anaerobically, which results into colorless compounds. This is followed by complete degradation of aromatic amines strictly under aerobic conditions (Sponza & Isik 2002). Therefore, anaerobic/aerobic processes are crucial for complete mineralization of azo dyes (Tan et al. 1999). However, not all bacteria have both anaerobic and aerobic properties. Usually consortia are routinely used for the degradation of azo dyes. Decolorization of azo dyes by bacteria is initiated by azoreductase, which is responsible for the reduction or cleavage of azo bonds in an anaerobic environment (Zimmermann et al. 1982a). Moreover, conventional aerobic waste water treatment processes, such as activated sludge, cannot usually efficiently remove the color of azo dyes since these compounds are often recalcitrant aerobically (Chung & Stevens 1993). So there is still a need to develop novel biological processes leading to a more effective cleanup of azo dye contamination. Hence, the objective of this study was to assess the potential of the unknown microorganism to decolorize and mineralize sulfonated azo dyes under aerobic conditions only, as most of the textile azo dyes are

normally decolorized strictly under anaerobic conditions and mineralized under aerobic conditions. Such potential of the microorganism will definitely prove costeffective in waste water treatment.

Materials and methods Microorganism and culture media The microorganism present in the soil from the effluent disposal site of a textile-dyeing industry located in Solapur, India was enriched in nutrient broth medium containing 50 mg of the dye l)1. The morphologically distinct bacterium was isolated and used for the decolorization studies. After obtaining the pure culture, it was grown aerobically in conical flasks, containing 100 ml (pH 7.0) nutrient broth medium (g l)1): yeast extract, 5.0, peptone, 5.0, along with mineral sources (mg l)1): KH2PO4, 170.0, Na2HPO4, 980.0, (NH4)2SO4, 100.0, MgSO4, 4.87, MgO, 0.1, FeSO4, 0.05, CaCO3, 0.20, ZnSO4, 0.08, CuSO4, 0.016, CoSO4, 0.015, H3BO3, 0.006 at room temperature. The pure culture of the microorganism was white, coccoid in shape and Gram-negative. It was designated KMK 48. Dyes Commercially available azo dyes viz. Reactive Red 2, Reactive Red 141, Reactive Orange 4, Reactive Orange 7 and Reactive Violet 5 were purchased from a local market and used for this study without any further purification. The chemical structures of the dyes used in this study are shown in Table 1.

368

K.M. Kodam et al.

Table 1. Chemical structures and absorption maxima of reactive azo dyes. Name & C.I. no.

kmax nm

Reactive Red 2 (C.I. no.- 18200)

538

Chemical Structure of azo dye NaO3S

SO3Na

N=N NH

OH

N

Cl

Reactive Red 141 (C.I. no.- not known)

N

N

Cl

544 NaO S 3

SO Na 3

SO Na 3

NaO S 3

N=N

N=N

NaO S

SO Na 3

3

HO

Cl

Cl

OH

N

N N

N

N

N

N

N

H

H

N

N

H

H

SO Na 3

NaO S 3

Reactive Orange 4 (C.I. no.- 18260)

490

Cl

OH

SO3Na

N N

N=N

N N

CH3

NaO3S

Cl

SO3Na

Reactive Orange 7 (C.I. no.-17757)

480

O

NaO3S-OCH2-CH2- S

N=N

SO3Na

O HO

H3COCHN

Reactive Violet 5 (C.I. no.- 18097)

530

OH

NaO3S-O-CH2-CH2- O2S

OH

NHCOCH3

N=N

NaO3S

SO3Na

369

Microbial decolorization of reactive azo dyes Decolorization of azo dyes by growing cells The decolorization experiments were performed in two sets. The microbial culture isolated from the effluent disposal site was cultivated aerobically for 24 h in conical flasks containing 100 ml nutrient broth. After 24 h, azo dye was added at a concentration of 200 and 1000 mg l)1. The conical flasks were placed on a rotary platform incubator shaker at 200 rev min)1 at room temperature. After different time intervals aliquot (5 ml) of the culture media was withdrawn, centrifuged at 10000 · g for 15 min in a refrigerated centrifuge (Dupont Sorvall RC-5B) to separate the bacterial cell mass and supernatant. Analytical methods Absorbance of the supernatant withdrawn at different time intervals were measured at the maximum absorption wavelength (kmax) in the visible region for each dye (Table 1) on a Shimadzu double beam spectrophotometer (UV 1601). The percentage of decolorization was calculated from the difference between initial and final absorbance values. All experiments were repeated three times. Extraction of probable biotransformed products The supernatants, which might contain biotransformation products of the dyes, were extracted with ethyl

acetate and dried over anhydrous magnesium sulphate. The solvent was evaporated and the residue was then chromatographed on silica gel. The H1NMR spectra of the purified samples were taken on Varian Mercury spectrometer (YH 300). The H1NMR spectra of the parent compounds were compared with that of the products.

Results and discussion Decolorization of Reactive Red 2, Reactive Red 141, Reactive Orange 4, Reactive Orange 7 and Reactive Violet 5 was observed by the unidentified bacterium, KMK 48. The percent decolorization of each dyes are shown in Figures 1 and 2 at two different concentrations (200 and 1000 mg l)1), respectively. As shown in Figures 1 and 2, complete decolorization (100%) of Reactive Red 2, at the concentration of 200 and 1000 mg l)1, was observed after 30 h. However, 200 mg l)1 of Reactive Red 141 was decolorized within 24 h, whereas decolorization of 1000 mg l)1 was achieved after 30 h. Moreover, 200 mg l)1 of Reactive Orange 4 was decolorized completely after 30 h whereas, with 1000 mg l)1 decolorization was observed by 36 h. The time taken for total decolorization of Reactive Orange 7, having the concentration of 200 mg l)1 was 18 h whereas that of 1000 mg l)1 was found to be 24 h. Reactive Violet 5 was decolorized

% Decolorization

100 80 Reactive Red 2

60

Reactive Red 141 Reactive Orange 4

40

Reactive Orange 7 Reactive Violet 5

20 0 0

6

12

18

24

30

Time (h) Figure 1. Decolorization assay of Reactive azo dyes at 200 mg l)1.

% Decolorization

100 80 Reactive Red 2

60

Reactive Red 141 Reactive Orange 4

40

Reactive Orange 7 Reactive Violet 5

20 0 0

6

12

18

Time (h)

Figure 2. Decolorization assay of Reactive azo dyes at 1000 mg l)1.

24

30

36

370

K.M. Kodam et al. can definitely solve the problem of maintaining anaerobic conditions for bioremediation of textile dye effluents.

Acknowledgments The authors would like to acknowledge the assistance provided by Anita Chacko, Yogendra Risbud, Amit Makone & Santosh Misal.

References Figure 3. H1NMR spectrum of biotransformed product of Reactive Orange 7.

completely after 12 h (200 mg l)1) and that of 1000 mg l)1 was within 30 h. These results indicate that almost 100% decolorization is achieved in a short duration (36 h), suggesting the effectiveness of KMK 48 in dye removal. It was also observed that this strain could decolorize sulfonated azo dyes, which are normally considered to be more recalcitrant than carboxylated azo dyes (Haug et al. 1991). Sulfonated azo dyes are not decolorized easily, as the permeation through the cell membrane is the ratelimiting step during bacterial reduction of azo dyes (Mechsner & Wuhrmann 1982). Such decolorization could have occured by an oxygen-insensitive azoreductase (Zimmermann et al. 1982b). Moreover, the H1NMR spectra of the products extracted after 36 h did not show any peaks corresponding to the aromatic protons (Figure 3). Thus, in addition to azo bond cleavage, ring fission had occurred. This clearly indicates that the dyes could have been catabolized and used as carbon and nitrogen source by KMK 48. Thus, the advantage of KMK 48 is that it can decolorize sulfonated azo dyes even under aerobic conditions, which is in contrast to other reports (Rafii et al. 1990; Nigam et al. 1996). Secondly, high effectiveness in decolorization is attained. Such characteristic properties of the microorganism happen to be a new finding among bacterial decolorization of reactive azo dyes. Such dual properties of the microorganism

Chung, K-T. & Stevens, S.E. 1993 Degradation of azo dyes by environmental microorganisms and helminths. Environmental Toxicology and Chemistry 12, 2121–2132. Haug, W., Schmidt, A., No¨rtemann, B., Hempel, D.C., Stolz, A. & Knackmuss, H.J. 1991 Mineralization of the sulfonated azo dye mordant yellow 3 by a 6-aminonaphthalene-2-sulfonate degrading bacterial consortium. Applied and Environmental Microbiology 57, 3144–3149. Mechsner, K. & Wuhrmann, K. 1982 Cell permeability as a rate limiting factor in the microbial reduction of sulfonated azo dyes. European Journal of Applied Microbiology and Biotechnology 15, 123–126. Nigam, P., Banat, I.M., Singh, D. & Marchant, R. 1996 Microbial process for the decolorization of textile effluent containing azo, diazo and reactive dyes. Process Biochemistry 31, 435–442. Rafii, F., Franklin, W. & Cerniglia, C.E. 1990 Azoreducatase activity of anaerobic bacteria isolated from human intestinal micro flora. Applied and Environmental Microbiology 56, 2146–2151. Sponza, D.T. & Isik, M. 2002 Decolorization and azo dye degradation by anaerobic/aerobic sequential process. Enzyme and Microbial Technology 31, 102–110. Tan, N.C.G., Prenafeta-Boldu´, F.X., Opsteeg, J.L., Lettinga, G. & Field, J.A. 1999 Biodegradation of azo dyes in cocultures of anaerobic granular sludge with aerobic aromatic amine degrading enrichment cultures. Applied Microbiology and Biotechnology 51, 865–871. Yang, Q.X., Yang, M., Pritsch, K., Yediler, A., Hagn, A., Schloter, M. & Kettrup, A. 2003 Decolorization of synthetic dyes and production of manganese-dependent peroxidase by new fungal isolates. Biotechnology Letters 25, 709–713. Zimmermann, T., Kulla, H.G. & Leisinger, T. 1982a Properties of purified orange II azoreductase, the enzyme initiating azo dye degradation by Pseudomonas luteola. Bioresource Technology 49, 47–51. Zimmermann, T., Kulla, H.G. & Leisinger, T. 1982b Properties of purified orange II azoreductase, the enzyme initiating azo dye degradation by Pseudomonas KF46. European Journal of Biochemistry 129, 197–203.

 Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 371–375 DOI 10.1007/s11274-004-3908-3

Rapid method for the affinity purification of thermostable a-amylase from Bacillus licheniformis M. Damodara Rao1,*, B.V.V. Ratnam2, Dasari VenkataRamesh3 and C. Ayyanna2 1 Department of Pediatrics Infectious Diseases, Johns Hopkins University School of Medicine, 720 Rutland Avenue, Ross 1135B, Baltimore, MD-21205, USA 2 Centre for Biotechnology, Department of Chemical Engineering, Andhra University, Visakhapatnam 530 003, India (E-mail: ayyanna123@rediffmail.com) 3 Department of Animal Biology, School of Veterinary Medicine, University of Pennsylvania, 3800 Spruce Street, Philadelphia PA-19104, USA [Tel: +1(215-898-6830. E-mail: [email protected])] * Author for correspondence: Tel: 410 614 0058, Fax: 410 614 1315, E-mail: [email protected] Received 19 March 2004; accepted 16 September 2004

Keywords: Affinity adsorption, a-amylase, Bacillus licheniformis

Summary A rapid and efficient method the exploiting affinity of a-amylase for its substrate starch is described. a-amylase from Bacillus licheniformis was purified to homogeneity by ammonium sulphate precipitation and affinity chromatography with 230-fold purification. The a-amylase adsorption to various starches was examined in order to screen its ability for highest binding to starch. The a-amylase was bound to starch more tenaciously, hence various eluants like maltose, soluble starch and high salts could not elute the bound a-amylase. However, the bound a-amylase was instantly eluted using 2% (w/v) dextrin. The purified enzyme showed a single polypeptide on SDS-PAGE, with a molecular weight of 58 kD. Western blot analysis confirmed the specificity of antibody raised against purified a-amylase.

Introduction Thermostable a-amylases (EC 3.2.1.1) have had many commercial applications for several decades. These enzymes are used in textile and paper industries, starch liquefaction, food, adhesives, and sugar production. Traditionally a-amylase from microbial sources was purified by ion-exchange, and gel filtration chromatography with low recovery (Igarashi et al. 1998; Aguilar et al. 2000; Pierce et al. 2002; Aquino et al. 2003; Das et al. 2004). The specific binding of an enzyme to its substrate has been exploited in several cases to constitute a useful purification step (Wilchek et al. 1984). aamylases from various species have been purified to homogeneity by either binding to starch (Shaw et al. 1995; Ben Ali et al. 2001) or by using cross-linked starch (Somers et al. 1991; Li & Geng 1992; Stredansky et al. 1993). Not only the binding of an enzyme to the affinity column but also its subsequent elution constitutes an important step in purification. The disadvantage of crosslinked starch was its slight biodegradation by the amylolytic enzyme and desorption of bound enzyme at high temperatures (50 – 70
C). Various methods were used for eluting the a-amylase bound to starch such as soluble starch gradients (Markovitz et al. 1956), pH shifts (Rozie et al. 1991) or with maltose (Mondal et al. 2003; Safarikova et al. 2003) with less efficiency. Here

we report a rapid, instant and reproducible method for the purification of a-amylase from microbial sources.

Materials and methods Culture, media, and growth conditions Bacillus licheniformis strain MTCC1483 was obtained from the Institute of Microbial Technology, Chandigarh, India. The culture was maintained in nutrient starch agar slants. The production media consisted of g l)1: corn starch 10.0; yeast extract 2.0; peptone 5.0; MgSO4 Æ 7H2O 0.5; CaCl2 0.15; (NH4)2HPO4 1.0. The medium pH was adjusted to 7.0 and 50 ml medium was transferred into 250 ml Erlenmeyer flasks. Cell growth continued at 40
C for 24 h in a rotary shaker at 300 revol/min, after which the cells were separated by centrifugation. The clear supernatant was used as the crude enzyme preparation. Analytical methods a-amylase was assayed according to Bernfeld (1955). In brief, an aliquot of enzyme was incubated with a reaction mixture containing 50 mM phosphate buffer pH, 7.0 and 1% soluble starch at 37
C for 30 min. The

372 reaction was stopped and the liberated maltose was measured by the addition of DNS (3,5-dinitrosalicylic acid) reagent. One enzyme unit is defined as the amount liberate required to 1 lmol of reducing sugar under the specified conditions. Protein was estimated by the Lowry method. Native starch from various sources was isolated according to Schoch (1957). Adsorption of a-amylase to starch a-amylase fractions obtained after ammonium sulphate precipitation (20–80%) were pooled to perform batch experiments of a-amylase binding to polymeric carbohydrates . Effect of starch concentration on a-amylase binding a-amylase enzyme (22 lmol) was separately incubated with different amounts of starch (0 – 600 mg) in 5 ml of 50 mM phosphate buffer, pH 6.8. The enzyme–substrate complex was left overnight at 4
C. The adsorption % was determined by assaying enzyme activity in the supernatant. A control experiment was conducted under similar conditions without enzyme. Enzyme activity (%) was determined using the following equation: Initial activity final activity  100 Initial activity Effect of time on the adsorption of a-amylase to starch The enzyme (5 lmol) was incubated with 0.5 g of starch in 50 mM phosphate buffer, pH 6.8 at 4
C from 0, 10, 20, 30, 40, 50, 60, 120, 240 and 300 min. At the end of each incubation period, the enzyme–substrate complex was centrifuged. The enzyme activity was assayed in the supernatant.

M.Damodara Rao et al. to bind a-amylase (maltose and white dextrin). aamylase bound to starch in batch experiments was incubated with various eluants like (i) 1 M NaCl (ii) 1 M KCl (iii)1 M CaCl2. The percent recovery of aamylase has been calculated by taking the enzyme activity bound to starch as 100%. Elution of a-amylase by soluble starch solution a-amylase (125 lmol) was applied to column packed with 1 g of starch. The bound enzyme was eluted with a stepwise gradient of soluble starch solution (0–500 mg/ ml in 50 mM phosphate buffer pH 6.8) and 2.5 ml fractions were collected at a flow rate of 10 ml/h. The fractions were assayed for a-amylase activity and the percent recovery of enzyme at each step of the gradient was calculated. Purification of a-amylase on starch column The crude enzyme was salt (ammonium sulphate) precipitated from (20–80%) and 60% salt precipitate was dialysed against 50 mM Na-phosphate buffer, pH 6.8. Two grams of native starch was suspended in 50 mM Na-phosphate buffer, pH 6.8, and the resulting slurry was packed into a column (1.5 · 10 cm) after repeated washing with 50 mM Na-phosphate buffer, pH 6.8 to remove fine starch particles. After equilibration, dialysate was loaded onto a column at a flow rate of 20 ml/h at 4
C. The bound enzyme was eluted by 2% (w/v) dextrin (molecular weight is 154) solution in equilibrated buffer at room temperature (25
C). The fractions enriched in a-amylase activity were pooled. The dextrin present in the fractions was removed by dialysis. SDS-PAGE and Western blot analysis

Effect of temperature and pH on the adsorption of a-amylase to starch The enzyme (5 lmol) in 20 mM Na-phosphate buffer pH 6.8 was incubated with 0.5 g of starch for 30 min at different pH values ranging from 5.0 to 9.0 and temperatures from 5–20
C. After 30 min, the supernatant was assayed for a-amylase activity. The buffers employed were 20 mM acetate pH 5.0 – 6.0, 20 mM phosphate pH 6.0 – 7.0, 20 mM Tris–HCl pH 7.0 – 9.0. Elution of a-amylase bound to starch a-amylase, a starch-degrading enzyme upon incubation with its substrate starch results in a decrease in its activity in the supernatant as a result of its tenacious binding to starch, hence attempts were made to elute bound enzyme from starch. The elution of bound aamylase to starch was systematically attempted by employing various eluants such as salts, soluble starch, products of a-amylase on starch that strongly competes

Sodium dodecyl sulphate (SDS) – PAGE was performed according to Laemmli (1970). The enzyme was purified according to Damodara Rao et al. (2002), injected into rabbits to obtain a-amylase-specific antiserum. The immunoglobulin G (IgG) fraction from the antiserum was purified by using Protein A Sepharose chromatography. The purified a-amylase was resolved in SDSPAGE (10%). The protein was then transferred to a 0.22 lm nitrocellulose membrane by electroblotting. Following transfer, the nitrocellulose membrane was blocked overnight in 5% non-fat dried milk powder in Tris buffered saline containing 0.1% Tween 20 (TTBS) followed by incubation with primary antibodies diluted in TTBS (a-amylase antiserum diluted 1:2500) for 1 h. The blots were then washed three times with TTBS (each wash for 5 min) and incubated with anti-rabbit horseradish peroxidase antibody (Santa Cruz Biotechnology) for 1 h. The blots were again washed three times with TTBS and the signal was detected between 1 and 15 min using ECL chemiluminiscent detection reagents.

373

Rapid method for the affinity purification of a-amylase Results Adsorption of a-amylase to native starches In order to standardize a batch method of a-amylase purification, the relative binding of various starches from potato, wheat, corn, pearl millet, sorghum and ragi were evaluated. Wheat (78%) and millet starches (71%) caused a more significant reduction in a-amylase activity in the supernatant than corn (64%), sorghum (52%), and ragi (56%) starches. Based upon the relative binding capacities, wheat starch was chosen for further studies. However, no a-amylase activity could be detected in the pelleted starch on incubating it with assay mixture, indicating a tenacious binding of enzyme to starch.

shift from 5.0 to 9.0, it was found to be ineffective in release the bound a-amylase from starch and 1% (w/v) soluble starch could elute only about 4% of bound enzyme. The bound enzyme could, nevertheless be most efficiently eluted by white dextrin using a linear gradient (Figure. 1). Furthermore, in comparison to all eluants investigated, the dextrin-mediated elution was not time dependent and bound enzyme could be eluted almost instantaneously (Figure. 2, Table 1). The molecular weight of the purified enzyme was found to be 58 kD on SDS-PAGE (Figure. 3) and western blot shows the specificity of antibody against purified a-amylase (Figure. 4).

Discussion Optimization of a-amylase adsorption to starch Optimization of a-amylase binding to starch depends on factors such as starch concentration, time dependence etc. These data shows that no concomitant decline in aamylase activity present in the supernatant with increasing starch concentration (0–600 mg of starch) and 38% a-amylase was adsorbed to 600 mg of starch. Since aamylase adsorption to starch is time dependent, further studies were done using various time intervals. The binding capacity was not changed over a period ranging from 30 min to 5 h with no further significant increase. In view of the above, a 30 min incubation time was routinely employed to investigate a-amylase adsorption to starch. The time dependent adsorption was 78% at 5
C, 72% at 10
C, 68% at 15
C and 59% at 20
C. The adsorption capacity increased with decrease in temperature. At high temperatures, starch will be hydrolysed more due to decreased adsorption of the enzyme on the granules than to increased swelling of the granules. Experiments were conducted to find whether the variation in adsorption with pH could be similarly explained. The a-amylase adsorption to starch depends on pH. The extent of adsorption increased as the pH was raised. Adsorption of a-amylase was 80% at pH 6.0, 78% at pH 7.0, 71% at pH 5.0 and 47% at pH 5.0, which shows that adsorption was also dependent on pH. The adsorption % at pH 6.0 and 7.0 was almost similar, and hence further experiments were conducted at pH 6.8.

The affinity of starch-metabolizing enzymes towards their natural substrate starch has been extensively exploited for obtaining highly purified enzyme preparations (Dube & Nordin 1961; Ishizaki et al. 1983). Preliminary experiments revealed the tight binding of a-amylase to starch because of the reduction of aamylase activity in the supernatant results from a starch mediated-inactivation of the enzyme. The experiments concerning the elution of likely bound a-amylase from

Figure 1. Elution by dextrin of a-amylase bound to starch 26 lmol of a-amylase was bound to 1 g of the starch column. The bound enzyme was eluted using a linear gradient of 1–5% of dextrin in 50 mM phosphate buffer pH 6.8 at a flow rate of 20 ml/h.

Elution of a-amylase bound to starch The elution of a-amylase bound to starch was systematically attempted by employing various eluants such as maltose, salts, starch, pH shifts and white dextrin. The maltose-mediated elution (500 mg/ml) of a-amylase eluted 45% of bound a-amylase at 24 h. Thereafter, further incubation did not improve the enzyme elution. In comparison to maltose, the incubation with high salt concentrations such as 1 M NaCl, KCl, CaCl2 could elute only 10, 6 and 16% of bound enzyme activity, respectively. Similarly, on subjecting the starch to a pH

Figure 2. Affinity purification of a-amylase on a starch column (-d-, A280 nm; - -, enzyme activity).

374

M.Damodara Rao et al.

Table 1. Affinity purification of a-amylase on starch column. Purification stage

Volume (ml)

Activity (lmol)

Protein (mg)

Crude enzyme Ammonium sulfate fractionation and dialysis Dextrin eluate

140 20

1184 880

532 210

50

609

1.2

Figure 3. Polyacrylamide gel electrophoresis of a-amylase at different purification Stages. (A) Crude extract, (B) Ammonium sulphate precipitation, (C) Starch affinity chromatography, (D) Molecular weight markers.

Figure 4. Western analysis of purified a-amylase from Bacillus licheniformis: Identity was confirmed by immunological cross reactivity with an antibody raised against a-amylase from Bacillus licheniformis. (A) Crude extract, (B) Purified fraction from starch affinity chromatography.

Specific activity (lmol mg)1) 2.2 4.1 507

Yield (%)

100 74 51

Purification-fold

1 1.8 230

pelleted starch favoured the former view. Since a prolonged incubation of pelleted starch with the product of a-amylase action, maltose leads to a partial release (45%) of a-amylase in the supernatant, it is evident that enzyme is firmly bound to starch. The a-amylase binding to starch depends on substrate concentration and incubation time. The incubation of pelleted starch with a soluble starch solution could elute only 3.8% of enzyme after a prolonged incubation period. Similarly, use of acidic to alkaline pH shifts and high salt concentrations could not lead to an appreciable release of bound enzyme from starch. Probably the change in structural configuration of a-amylase induced by the above pH and salt treatment is not strong enough to alter the configurations of the active site firmly binding to starch molecules. The release of bound enzyme from starch would directly depend on subjecting the starch aamylase complex to an equivalent or excess of nonreducing end eluant polymers to compete with starch. Since the a-amylase bound to starch can be rapidly eluted with high degree of recovery by using a dextrin solution, the availability of total number of nonreducing groups on the eluant polymer plays a major role in the elution of bound enzyme. Since dextrins are extremely fragmented starch derivatives it is obvious that even though both the starch and dextrin concentrations employed for elution are identical, the latter has a greater number of non-reducing ends leading to more efficient elution of bound a-amylase. In view of the variable efficiency of batch type experiments for binding a-amylase to starch, affinity chromatography on a column was attempted. In contrast to batch experiment, a-amylase application to a starch column led to almost 100% binding of enzyme unless all available binding sites of starch were saturated with enzyme (binding capacity 380 lmol/g starch). The bound a-amylase from the starch column was eluted by 50 mM Na-phosphate buffer pH 6.8, containing 2% dextrin (w/v). The fractions obtained on dextrin elution contained purified a-amylase with the single polypeptide on SDS-PAGE indicating the absence of subunits which is in conformity with Morgan & Priest (1981). Traditional purification methods like ion exchange and gel filtration chromatography used by several groups, purified a-amylase with 6.1-fold purification by Igarashi et al. (1998), 9.12-fold purification by Sumitani et al. (2000), 4.2-fold purification by Pierce et al. (2002), 18.2fold purification by Aquino et al. (2003), 16.1-fold purification by Li & Peeples (2004). These data suggest

Rapid method for the affinity purification of a-amylase looking for an efficient alternate method to traditional multi-step chromatography. Recently Mondal et al. (2003) and Safarikova et al. (2003) purified a-amylase from Bacillus species by affinity partitioning chromatography using 1 M Maltose as an eluant, with 5.5-and9.0 fold purification. Our method instantly elutes the a-amylase from the starch column using 2% (w/v) white dextrin with 230-fold purification, which is superior to any other affinity chromatographic methods so far described. Even though affinity purification is not new, there is lack of study of the systematic elution of bound a-amylase from starch. We have exploited the affinity interactions between a-amylase and its products on starch competes to bind a-amylase rather than to starch. We have developed a method with highest purity and recovery than the recently reported methods. Moreover this method is rapid, reproducible and efficiently elutes the bound aamylase from the starch columns. In conclusion, it is evident that starch columns have stronger affinity towards bacterial a-amylase, which could be purified to homogeneity in a single step by binding to a starch column and elution by dextrin solution. References Aguilar, G., Morlon-Guyot, J., Trejo-Aguilar, B. & Guyot, J.P. 2000 Purification and characterization of an extracellular alpha-amylase produced by Lactobacillus manihotivorans LMG 18010(T), an amylolytic lactic acid bacterium. Enzyme and Microbial Technology 27, 406–413. Aquino, A.C., Jorge, J.A., Terenzi, H.F. & Polizeli, M.L. 2003 Studies on a thermostable alpha-amylase from the thermophilic fungus Scytalidium thermophilum. Applied Microbiology and Biotechnolology 61, 323–328. Ben Ali, M., Mhiri, S., Mezghani, M. & Bejar, S. 2001 Purification and sequence analysis of the maltohexose-forming a-amylase of the B. stearothermophilus US100. Enzyme and Microbiol Technology 28, 536–542. Bernfeld, P. 1955 Enzymes of carbohydrate metabolism. Amylases a and b. Methods in Enzymology 1, 149–158. Damodara Rao, M., Purnima, A., Ramesh, D.V. & Ayyanna, C. 2002 Purification of a-amylase from Bacillus licheniformis by chromatofocusing and gel filtration chromatography. World Journal of Microbiology and Biotechnology 18, 547–550. Das, K., Doley, R. & Mukherjee A.K. 2004 Purification and biochemical characterization of a thermostable, alkaliphilic, extracellular alpha-amylase from Bacillus subtilis DM-03, isolated from the traditional fermented food of India. Biotechnology and Applied Biochemistry.(in Press) Dube, S.K. & Nordin, P. 1961 Isolation and properties of Sorghum aamylase. Archives of Biochemistry and Biophysics 94, 121–127. Igarashi, K., Hatada, Y., Hagihara, H., Saeki, K., Takaiwa, M., Uemura, T., Ara, K., Ozaki, K., Kawai, S., Kobayashi, T. & Ito, S.

375 1998 Enzymatic properties of a novel liquefying alpha-amylase from an alkaliphilic Bacillus isolate and entire nucleotide and amino acid sequences. Applied and Environmental Microbiology 64, 3282–3289. Ishizaki, Y., Taniguchi, H., Maruyama, Y. & Nakamura, M. 1983 Debranching enzymes of potato tubers (Solanum tuberosum L.) 1. Purification and some properties of potato isoamylase. Agricultural and Biolology Chemistry 47, 771–779. Laemmli, U.K. 1970 Cleavage of structural proteins during the assembly of the head of Bacteriophage T4. Nature 277, 680– 685. Li H. & Geng, X. 1992 Preparation of high performance affinity chromatography packings for specific adsorption of a-amylase and purification of crude a-amylase. Journal of Liquid Chromatography 15, 707–714. Li, M & Peeples T,L. 2004 Purification of hyperthermophilic archaeal amylolytic enzyme (MJA1) using thermoseparating aqueous twophase systems. Journal of Chromatography B Analytical Technology Biomedical Life Sciences. 807, 69–74. Markovitz,A., Klein, H.P. & Fischer. E. H.1956. Purification, crystallization, and properties of the alpha-amylase of Pseudomonas saccharophila. Biochimcal Biophysica Acta. 19, 267–273. Mondal, K., Sharma, A., Lata, L. & Gupta, M.N. 2003 Macroaffinity ligand-facilitated three-phase partitioning (MLFTPP) of alphaamylases using a modified alginate. Biotechnology Progress. 19, 493–494. Morgan, F.J. & Priest, F.G. 1981 Characterization of a thermostable a-amylase from Bacillus licheniformis NCIB 6346. Journal of Applied Bacteriology 50, 107–114. Pierce, J.J., Robinson, S.C., Ward, J. M., Keshavarz-Moore, E. Dunnill P. 2002 A comparison of the process issues in expressing the same recombinant enzyme periplasmically in Escherichia coli and extracellularly in Streptomyces lividans. Journal Biotechnology. 18, 205–215. Rozie, H., Somers, W., Van’t Reit, K., Rombouts, F.M. & Visser, J. 1991 Crosslinked potato starch as an affinity adsorbent for bacterial a-amylase. Carbohydrate Polymers 15, 349–365. Safarikova, M., Roy, I., Gupta, M.N. & Safarik, I. 2003 Magnetic alginate microparticles for purification of alpha-amylases. Journal of Biotechnology 105, 255–260. Shaw, J.F., Lin, F.P., Chen, S.C. & Chen, H.C. 1995 Purification and properties of an extracellular a-amylase from Thermus species. Botanica Bullehin of Academia Sinica 36, 195–200. Somers, W., Rozie, H., Bonte, A., Visser, J., Rombouts, F.M. & Van’t Riet, K. 1991 Isolation of a-amylase on crosslinked starch. Enzyme and Microbial Technology 13, 997–1006. Stredansky, M., Kremnicky, L., Sturdik, E. & Feckova, A. 1993 Simultaneous production and purification of Bacillus subtilis aamylase. Applied Biochemistry and Biotechnology 38, 269–276. Sumitani, J., Tottori, T., Kawaguchi, T. & Arai, M. 2000 New type of starch-binding domain: the direct repeat motif in the C-terminal region of Bacillus sp. no. 195 alpha-amylase contributes to starch binding and raw starch degrading. Biochemical Journal 350, 477– 484. Wanderley, K.J., Torres, F.A., Moraes, L.M. & Ulhoa, C.J. 2004 Biochemical characterization of alpha-amylase from the yeast Cryptococcus flavus. FEMS Microbiology Letters 231, 165– 9.7. Wilchek, M., Miron, T. & Kohn, J. 1984 Affinity chromatography. Methods in Enzymology 104, 3–55.

Ó Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 377–380 DOI 10.1007/s11274-004-3627-9

Short communication

Bioleaching of uranium from low grade black schists by Acidithiobacillus ferrooxidans Moon-Sung Choi1, Kyung-Suk Cho1,*, Dong-Su Kim1, and Hee-Wook Ryu2 1 Department of Environmental Science and Engineering, Ewha Womans University, 11-1 Daehyun-dong, Seodaemungu, Seoul 120-750, Korea 2 Department of Chemical and Enviromental Engineering, Soongsil University, 1-1 Sangdo-5dong, Dongjak-gu, Seoul 156-743, Korea *Author for correspondence: Tel.: +82-2-3277-2393, Fax: +82-2-3277-3275, E-mail: [email protected] Received 24 February 2004; accepted 10 July 2004

Keywords: Acidithiobacillus ferrooxidans, bioleaching, black schist, uranium

Summary The feasibility of bacterial recovery of uranium from the low grade black schists occurring in the Okcheon district, South Korea, was investigated. Following the introduction of Acidithiobacillus ferrooxidans, 80% of the uranium could be extracted from the schists, which contain 0.01% U3O8 by weight, within 60 h at a pulp density of 100 gore/l. Only 18% of the uranium was extracted without microbial activity. The uranium-leaching efficiency was not greatly affected by the addition of Fe2+ in the range of 5–9 g/l, and the leaching efficiency of uranium from the schists by A. ferrooxidans could be efficiently maintained at high pulp densities (up to 500 g-ore/l).

Introduction Uranium is an important natural resource used for the generation of nuclear energy, as the raw material for the production of uranium oxide (U3O8), as a component for anti-corrosive alloys, and as a colouring agent for glass and porcelain. The quantity of uranium-bearing ore in the Daejeon and Okcheon districts of South Korea is estimated to be over 100 million tons. Uranium is conventionally extracted using a process that employs strong acids as leaching agents. This method often creates environmental problems, requires large amounts of energy, and involves a complex operational plant (Agate 1996; Bosecker 1997). It is not, however, economical to extract uranium from lowgrade ores by chemical leaching (Agate 1996; Bosecker 1997; Munoz et al. 1995c). Because the content of U3O8 in the Korean ore is below 0.1% by weight, it is necessary to develop an alternative process to enable the efficient and economic recovery of the uranium. In contrast to the chemical leaching process, the bioleaching process, which employs microorganisms such as bacteria and fungi as the leaching catalysts, is known to be economical and environmentally acceptable (Agate 1996; Bosecker 1997; Munoz et al. 1995c). In addition, bioleaching is readily adaptable to low grade ores and does not require a complex plant, such that several countries have already adopted bacterial

leaching as a new extractive metallurgical process (Agate 1996; Bosecker 1997; Munoz et al. 1995c; Rawlings 1998). However, uranium bioleaching plants have not been opened for a couple of decades. One of the most widely employed microorganisms in the bioleaching process is A. ferrooxidans (Agate 1996; Bosecker 1997; Krebs et al. 1997; Rawlings 1998). A. ferrooxidans is an acidophilic and chemolithotrophic bacterium, which can oxidize reduced iron and sulphur compounds. The mechanism of uranium extraction facilitated by the indirect oxidation function of this microbe is probably as follows (Munoz et al. 1995c): 2þ UO2 þ 2Fe3þ þ SO2
4 ! UO2 SO4 þ 2Fe

ðU4þ þ 2Fe3þ ! U6þ þ 2Fe2þ Þ Uranium is barely soluble in an aqueous environment when it is in the +4 oxidation state; however, in an acidic medium the ferric ion oxidizes U4+ to U6+, which is easily dissolved. As a conjugate reaction to the oxidation of U4+, the ferric ion reduces to the ferrous ion, and through the oxidation function of A. ferrooxidans it is re-oxidized back to the ferric state, which is then able to continue oxidizing U4+ to U6+ (Munoz et al. 1995c). In this study, the characteristics of uranium leaching from black schist mined in the Okcheon district, employing A. ferrooxidans as the bioleaching microorganism,

378

Microorganisms and culture The medium for A. ferrooxidans (ATCC19859) growth was 9K medium of Silverman & Lundgren (1959), which is a mixture of mineral salts ((NH4)2SO4 3.0 g/l, K2HPO4 0.5 g/l, MgSO4 7H2O 0.5 g/l, KCl 0.1 g/l, Ca(NO3)2 0.01 g/l). FeSO4 7H2O (45 g/l) was added as an energy source. The pH of the medium was adjusted to 2.0 using 0.5 M H2SO4. The culture broth was cultivated at 30 °C for 3–4 days before centrifugation at 2000 g for 10 min to remove precipitates, followed by another centrifugation at 7600 g for 20 min to recover the microorganism. The harvested cells of A. ferrooxidans were suspended in a fresh solution of the mineral salt medium for the preparation of the bacterial concentrate, which was used as the admixture for the uranium leaching experiments.

Results and discussion The variations of the pH, ORP, and Fe2+ and Fe3+ concentrations with reaction time in the leaching slurry were determined; the results are shown in Figure 1. It can be seen that the introduction of A. ferrooxidans into the reaction mixture caused the ORP of the slurry to increase from 344 to 515 mV. In contrast, the pH remained approximately constant over time in both the microbe-containing and control systems. Figure 1 additionally shows that, when A. ferrooxidans was inoculated, most of the ferrous ions were oxidized to the ferric state within 60 h, whereas little oxidation of ferrous ions occurred in the absence of the biological agent.

4

Analytical procedures The pH of the leaching slurry was measured using a pH meter (ORION, Model 420A, USA) and the changes in

500

3

400

2

300 200

1

U leaching efficiency (%)

Fe2+ concentration (g·L-1)

Bioleaching experiments Ten grams of ground and sieved black schist was placed in a 250 ml conical flask and 100 ml of 9 K medium was added. After inoculating A. ferrooxidans, the bioleaching of uranium was carried out, taking the amount of Fe in solution and the quantity of ore as the two process variables. The leaching characteristics of uranium without A. ferrooxidans, under the same conditions that in bioleaching tests, were also examined as a control experiment, to distinguish the biological and chemical outcomes. The concentration of Fe was set at 3, 5, 7, and 9 g/l using FeSO4 7H2O, and the amount of ore was adjusted to 10, 50, 100, 200, 300, 500, and 700 g-ore/l. Each experiment was carried out twice under the same standard conditions, at 30 °C and 180 rev/min shaking speed.

600

ORP (mV)

The uranium-bearing black schist ore used in the experiments was supplied by the Korea Institute of Geoscience and Mineral Resources. The uranium content in the black schist is 0.01% by weight, and the other principal components are silicates. In preparation for the experiment, the schist was ground using an agate mortar and then sieved. The particle size of the sieved material ranged from 300 to 2800 lm, with an average particle size of 500 lm.

100

0

0

10

10

8

8

6

6

4

4

2

2

0

0

(g·L-1)

Uranium ore

Fe3+ concentration

Materials and methods

oxidation/reduction potential (ORP) were monitored using an ORP meter (ORION, Model 420, USA). The concentration of ferrous ions was analysed using the o-phenanthroline method (Ryu et al. 1995). To prevent interference by jarosite particles, the slurry was initially centrifuged at 12000 rpm for 5 min. Then, the ophenanthroline reagent was added and the absorptivity was measured at 510 nm using a spectrophotometer (Milton-Roy, Spectronic 20, USA). The uranium concentration in the leachate was analysed by ICP-MS (Plasma Quad, VG, England).

pH

have been investigated to evaluate the feasibility of uranium recovery from low-grade ore through the bioleaching process.

M.-S.Choi et al.

80 60 40 20 0

0

12

24

36

48

60

72

Time (h) Figure 1. Time profiles of pH, ORP, Fe2+, and Fe3+ concentration during uranium leaching from black schist ore. Open symbols, without inoculation; closed symbols, with injection of A. ferrooxidans. s d, pH; h j, ORP; n m, Fe2+ concentration; , ., Fe3+ concentration; e r uranium leaching efficiency.

379

Uranium from low grade black schists

U leaching efficiency (%)

0.14 0.12

80

0.10 60 0.08 0.06

40

0.04 20 0.02 0

Iron oxidation rate (g L-1 h-1)

0.16 100

0.00

3

5

7

9

Fe2+ concentration (g L-1) Figure 2. Effect of Fe2+ concentration on iron oxidation rate and uranium-leaching efficiency (Pulp density ¼ 100 gl)1, Leaching time60 h). Bar, U leaching efficiency; Circle symbol, Iron oxidation rate.

100

U leaching efficiency (%)

The amount of uranium leached from the schist was observed to increase as Fe2+ ions were oxidized. The highest uranium-leaching efficiency increased from 18% in the absence of A. ferrooxidans to almost 80% following the introduction of A. ferrooxidans, which oxidized Fe2+ to Fe3+, transforming the uranium in the schist into the soluble form (Bhatti et al. 1997). These results show that in the leaching of uranium from lowgrade ore, the biological leaching process is superior to chemical leaching. Previous reports (Munoz et al. 1995a; Munoz et al. 1995b; Bhatti et al. 1997; Mathur et al. 2000) regarding the bioleaching of uranium by A. ferrooxidans have come to similar conclusions. The quantitive effect of Fe2+ on uranium extraction was also investigated. As shown in Figure 2, the uranium-leaching efficiency was not greatly changed by addition of Fe2+ in the range of 5–9 g/l, but it was significantly reduced (62%) by addition of 3 g/l Fe2+. These results indicate that, to maintain efficiency in the practical leaching process of uranium by A. ferrooxidans, the concentration of Fe2+ should not be allowed to fall below 5 g/l. The uranium-leaching efficiency was relatively insensitive to the ore pulp density for pulp densities up to 500 g/l, but decreased significantly with increasing pulp density on further increase of the pulp density (Figure 3). At high pulp density, the iron-oxidation activity of A. ferrooxidans was observed only after a lag of 4–5 days, the activity of microorganisms was inhibited, and the pH of the slurry was observed to rise slightly (data not shown). A. ferrooxidans is a bacterium that grows using carbon dioxide as a carbon source and oxygen as a final electron acceptor. Under conditions of high pulp density, the aggregation of fine ore particles occurs, blocking the mass transfer of carbon dioxide and oxygen in the slurry (Andrews et al. 1988; Ryu et al. 1995). This would explain the decreased iron-oxidation activity of A. ferrooxidans at high pulp density. Although some

80

60

40

20

0 0

10

50

100

200

300

500

700

Pulp density (g·L-1) Figure 3. Effect of pulp density on uranium leaching efficiency (Fe2+ concentration ¼ 9 gl)1, Leaching time ¼ 60 h).

decrease in uranium extraction efficiency occurs at pulp densities higher than 500 g/l, control of the pulp density at around this reasonably high concentration should permit adequate efficiency to be maintained in the bioleaching process of uranium from black schist by A. ferrooxidans.

Acknowledgements This work was supported by a research grant from Korea Institute of Geoscience and Mineral Resources.

References Agate, A.D. 1996 Recent advances in microbial mining. World Journal of Microbiology and Biotechnology 12, 487–495. Andrews, G.F., Darroch, M. & Hansson, T. 1988 Bacterial removal of pyrite from concentrated coal slurries. Biotechnology and Bioengineering 32, 813–820. Bhatti, T.M., Vuorinen, A., Lehtiene, M.K. & Tuoviene, O.H. 1997 Acid dissolution of uranophane and carnotite. Journal of Environmental Science and Health A32, 1827–1835. Bosecker, K. 1997 Bioleaching: metal solubilization by microorganisms. FEMS Microbiology Reviews 20, 591–604. Krebs, W., Brombacher, C., Bosshard, P.P., Bachofen, R. & Brandl, H. 1997 Microbial recovery of metals from solids. FEMS Microbiology Reviews 20, 605–617. Mathur, A.K., Viswamohan, K., Mohanty, K.B., Murthy, V.K. & Seshardrinath, S.T. 2000 Uranium extraction using biogenic ferric sulfate: a case study on quartz chlorite ore from Jaduguda, Singhbhum Thrust Belt (STB), Bihar, India. Minerals Engineering 13, 575–579. Munoz, J.A., Ballester, A., Gonzalez, F. & Blazquez, M.L. 1995a A study of the bioleaching of a Spanish uranium ore. Part II: orbital shaker experiments. Hydrometallurgy 38, 59–78. Munoz, J.A., Blazquez, M.L., Ballester, A. & Gonzalez, F. 1995b A study of the bioleaching of a Spanish uranium ore. Part III: column experiments. Hydrometallurgy 38, 79–97. Munoz, J.A., Gonzalez, F., Blazquez, M.L. & Ballester, A. 1995c A study of the bioleaching of a Spanish uranium ore. Part I: a review

380 of the bacterial leaching in the treatment of uranium ores. Hydrometallurgy 38, 39–57. Rawlings, D.E. 1998 Industrial practice and the biology of leaching of metals from ores. Journal of Industrial Microbiology and Biotechnology 20, 268–274. Ryu, H.W., Cho, K.S., Chang, Y.K., Kim, S.D. & Mori, T. 1995 Refinement of low-grade clay by microbial removal of sulfur and

M.-S.Choi et al. iron compounds using Thiobacillus ferrooxidans. Journal of Fermentation and Bioengineering 80, 46–52. Silverman, M.P. & Lundgren, D.G. 1959 Studies on the chemoautotrophic iron bacterium Ferrobacillus ferrooxidans I. An improved medium and a harvesting procedure for securing high cell yields. Journal of Bacteriology 77, 642–647.

 Springer 2005

World Journal of Microbiology & Biotechnology (2005) 21: 381–384 DOI 10.1007/s11274-004-3630-1

Short communication

Characterization of a diesel-degrading bacterium, Pseudomonas aeruginosa IU5, isolated from oil-contaminated soil in Korea Ji Hye Hong1, Jaisoo Kim1, Ok Kyoung Choi1, Kyung-Suk Cho1,* and Hee Wook Ryu2 1 Department of Environmental Science and Engineering, Ewha Womans University, 11-1 Daehyun-dong, Seodaemun-gu, Seoul 120-750, Korea 2 Department of Environmental and Chemical Engineering, Soongsil University, 1-1 Sangdo-5-dong, Dongjak-gu, Seoul 156-743, Korea *Author for correspondence: Tel.: +82-2-3277-2393, Fax: +82-2-3277-3275, E-mail: [email protected] Received 24 February 2004; accepted 13 July 2004

Keywords: Biodegradation, bioremediation, diesel, petroleum hydrocarbons, Pseudomonas aeruginosa

Summary A diesel-degrading bacterium (strain IU5) isolated from oil-contaminated soil was characterized in this study. Fatty acid and 16s rDNA sequence analysis identified IU5 as a strain of Pseudomonas aeruginosa, and growth curve experiments identified the bacterium’s optimum conditions as pH 7 and 30
C. P. aeruginosa IU5 degraded up to 60% of applied diesel (8500 mg/kg) over 13 days in a soil-slurry phase. In addition, this strain was able to grow on many other petroleum hydrocarbons as sole carbon sources, including crude oil, gasoline, benzene, toluene, xylene, and even PAHs such as naphthalene, phenanthrene and pyrene. Therefore, P. aeruginosa IU5 may be useful for bioremediation of soils and groundwater contaminated with a variety of hydrocarbons.

Introduction Diesel contains many highly concentrated toxic materials (Dillard et al. 1997; Gold-Bouchot et al. 1997), and diesel contamination can negatively influence soil microbes and plants, as well as contaminate groundwater, which may be used for drinking or agriculture. Because of this, a variety of methods have been developed to treat diesel contamination. While many of the established physical and chemical methods are efficient, they are also expensive and can cause recontamination by secondary contaminants. Bioremediation is the microbial degradation of organic pollutants such as petroleum in soil and groundwater. This technique has the benefits of high treatment efficiency, low cost, relatively quick action, in site and ex site application, and compatibility with other techniques. Nevertheless, the application of hydrocarbon-degrading bacteria in oil-contaminated sites does not guarantee to remove all components of crude oil because some components still remain difficult to degrade, such as alkanes of shorter and longer chains (