Polish Journal of Microbiology

POLSKIE TOWARZYSTWO MIKROBIOLOGÓW POLISH SOCIETY OF MICROBIOLOGISTS

Polish Journal of Microbiology formerly

Acta Microbiologica Polonica

2007

POLISH JOURNAL OF MICROBIOLOGY (founded in 1953 as Acta Microbiologica Polonica)

www.microbiology.pl/pjm EDITORIAL OFFICE EDITOR IN CHIEF:

Miros³awa W³odarczyk

EDITORS:

Ryszard Chróst Hanna Dahm Jaros³aw Dziadek Anna Skorupska

EDITORIAL SECRETARY: Anna Kraczkiewicz-Dowjat POSTAL ADDRESS:

Polish Journal of Microbiology Miecznikowa 1 02-096 Warsaw, POLAND

CONTACT:

Phone: (48) 22 554 1318 Fax: (48) 22 554 1402 E-mail: Editorial Office ([email protected]) Editor in Chief ([email protected])

EDITORIAL BOARD

President: Andrzej Piekarowicz (Warsaw, Poland)

Waleria Hryniewicz (Warsaw, Poland) El¿bieta K. Jagusztyn-Krynicka (Warsaw, Poland) Miros³aw Kañtoch (Warsaw, Poland) Donovan P. Kelly (Coventry, UK) Tadeusz Lachowicz (Wroc³aw, Poland) Jadwiga Wild (Madison, USA)

Wanda Ma³ek (Lublin, Poland) Zdzis³aw Markiewicz (Warsaw, Poland) Gerhardt Pulverer (Cologne, Germany) Geoffrey Schild (Potters, Bar, UK) Wac³aw Szybalski (Madison, USA) Torkel Wadstrom (Lund, Sweden)

PUBLISHER: POLISH SOCIETY OF MICROBIOLOGISTS Published quarterly with the financial support of the Ministry of Science and Higher Education SUBSCRIPTION: For information for Polish subscribers contact Secretary of Polish Society of Microbiologists, Che³mska 30/34, 02-725 Warsaw, Poland; phone: (48) 22 841 3367, fax: (48): 22 842 2949, e-mail: [email protected] For information for foreign subscribes contact Ars Polona e-mail: [email protected]; phone: + 48 (22) 509 86 61-5 also www.arspolona.com.pl Cover illustration: Airborne bacterial and fungal colonies (Jerzy Pi¹tkowski, Uniwersity of Wroc³aw, Poland) Typesetting and print: Publishing House Letter Quality, 01-216 Warsaw, Brylowska 35/38 Circulation: 300

Polish Journal of Microbiology formerly Acta Microbiologica Polonica

2007, Vol. 56, No 2

CONTENTS ORIGINAL PAPERS PCR melting profile method for genotyping analysis of vancomycin-resistant Enterococcus faecium isolates from hematological unit patients KRAWCZYK B., LEIBNER J., BRONK M., STOJOWSKA K., SAMET A., KUR J. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Selection and activation of Escherichia coli strains for L-aspartic acid biosynthesis PAPIERZ M., GADOMSKA G., SOBIERAJSKI B., CHMIEL A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Recombinant strains of Escherichia coli for L-aspartic acid biosynthesis GADOMSKA G., P£UCIENNICZAK A., CHMIEL A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Heterogeneity of galF and gnd of the cps region for capsule synthesis in clinical isolates of Klebsiella pneumoniae GIERCZYÑSKI R. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

65 71 77 83

Lipoarabinomannan as a regulator of the monocyte apoptotic response to Mycobacterium bovis BCG Danish strain 1331 infection KRZY¯OWSKA M., SCHOLLENBERGER A., PAW£OWSKI A., HAMASUR B., WINNICKA A., AUGUSTYNOWICZ-KOPEÆ E., NIEMIA£TOWSKI M. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Staphylokinase production by clinical Staphylococcus aureus strains WIÊCKOWSKA-SZAKIEL M., SADOWSKA B., RÓ¯ALSKA B. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

89 97

Indole-3-acetic acid production and effect on sprouting of yam (Dioscorea rotundata L.) minisetts by Bacillus subtilis isolated from culturable cowdung microflora

SWAIN M.R., NASKAR S.K., RAY R.C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

Mercury absorption by Pseudomonas fluorescens BM07 grown at two different temperatures

NOGHABI K.A., ZAHIRI H.S., LOTFI A.S., RAHEB J., NASRI S., YOON S.C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Molecular analysis of temporal changes of a bacterial community structure in activated sludge using denaturing gradient gel electrophoresis (DGGE) and fluorescent in situ hybridization (FISH)

ZIEMBIÑSKA A., RASZKA A., TRUU J., SURMACZ-GÓRSKA J., MIKSCH K. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

Transmission of specific groups of bacteria through water distribution system

GRABIÑSKA-£ONIEWSKA A., WARDZYÑSKA G., PAJOR E., KORSAK D., BORYÑ K. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

Purification and characterization of $-mannonidases from white rot fungus Phlebia radiata

PRENDECKA M., BUCZYÑSKA A., ROGALSKI J. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

INSTRUCTION TO AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148

Polish Journal of Microbiology 2007, Vol. 56, No 2, 6570 ORIGINAL PAPER

PCR Melting Profile Method for Genotyping Analysis of Vancomycin-resistant Enterococcus faecium Isolates from Hematological Unit Patients BEATA KRAWCZYKa, JUSTYNA LEIBNERa, KAROLINA STOJOWSKAa, MAREK BRONKb, ALFRED SAMETb and JÓZEF KURa,* a Gdañsk

b Department

University of Technology, Department of Microbiology, Gdañsk, POLAND; of Clinical Microbiology, Clinical Hospital No 1, Medical University of Gdañsk, POLAND Received 16 February 2007, accepted 10 April 2007 Abstract

A number of Enterococcus strains with high-level inducible resistance to vancomycin have been identified, and the relative incidence of these strains has increased significantly in the last years. The first outbreak caused by vancomycin-resistant enterococci in Poland was reported in 1999. Vancomycin-resistant Enterococcus faecium is known for its propensity to cause infections which are difficult to eradicate. In this study, we determined the genetic similarities between vancomycin-resistant E. faecium isolates consecutively recovered from single patients to assess the duration of infection or colonization. The isolates taken in the study were identified by the conventional methods as E. faecium. PCR melting profile (PCR-MP) and pulsed-field gel electrophoresis (PFGE) typing revealed that the isolates belonged to six distinct genotypes and that two of them were predominant. Consecutive E. faecium isolates with identical genotypes were found in 7 of 12 (58.0%) patients. The delay between the times of recovery of the first and last isolates of identical genotypes from each patient was from 9 days to about 1 year. In six patients, paired blood and non-blood isolates showed identical genotypes. Data presented here demonstrate the complexity of the epidemiological situation concerning vancomycin-resistant enterococci that may occur in a single medical ward. We also show for the first time the evaluation of PCR-MP technique in enterococci strains differentiation and we revealed that there is at least a similar power of discrimination between the present gold-standard REAPFGE and a PCR-MP method. K e y w o r d s: genotyping; PCR fingerprinting; LM PCR; PFGE; vancomycin-resistant enterococci

Introduction Vancomycin-resistant enterococci (VRE) belong nowadays to the most important nosocomial pathogens worldwide (Bonten and Hayden, 1996; Jarvis and Martone, 1992; Moellering, 1992; Murray, 1998), and they usually cause infections in severely debilitated, immunocompromised patients who undergo prolonged and intensive antibiotic therapy (Banerjee et al., 1991; Maki and Agger, 1988; Murray, 1990; Schaberg et al., 1991). In some countries VRE may significantly contribute to enterococcal populations circulating in hospitals. The first reported identification of VRE in Poland occurred at the end of 1996 in a hospital in Gdañsk (Hematological Unit) with the isolation of E. faecium of the phenotype VanA, and it was followed by a large VRE outbreak in this center (Hryniewicz et al., 1999;

Samet et al., 1999; Krawczyk et al., 2003). To determine whether the isolates were epidemiologically related, isolated strains were differentiated by PCR fingerprinting method (Samet et al., 1999). The PCR fingerprinting of VRE from the Hematological Unit demonstrated only small genetic heterogeneity among the isolates over 11 months, with two main genotypes being identified. These strains were genetically closely related. In the next study, 100 VRE strains within a duration of 36 months (between January 1997 and December 1999) taken from 100 patients were examined using ADSRRS-fingerprinting and pulsed field gel electrophoresis (PFGE) methods (Krawczyk et al., 2003). Several lines of evidence obtained in this study with a large number of isolates suggested that the VanA phenotype was selected by most likely by one or two independent events within the ward enterococcal

* Corresponding author: J. Kur, Gdañsk University of Technology, Department of Microbiology, ul. Narutowicza 11/12, 80-952 Gdañsk, Poland; tel/fax: +48 58 3471822; e-mail: [email protected]

66

2

Krawczyk B. et al.

population. This was supported by the observation of the dominant existence of the two closely related ADSRRS-fingerprinting groups. Only minor differences in ADSRRS-fingerprinting patterns observed in both groups revealed the ongoing evolutionary diversification process within their populations. These results confirmed the results described previously (Samet et al., 1999) with also two main closely related genotypes. Up till now, a number of E. faecium strains with high-level inducible resistance to vancomycin still have been isolated from patients of Haematological Unit of an University Hospital in Gdañsk. Analysis of microbiological data based on our collection of clinical E. faecium VRE isolates revealed successive isolates from patients with persistent or recurrent infections. Although biochemical and serological data pointed to a close similarity of at least some of the strains isolated from single patients, information on relatedness at the genetic level was lacking. Here, to study the genetic similarities between these successive E. faecium VRE isolates the PCR-MP and PFGE methods were used.

the pulsed field method of contoured clamped homogenous electric field). The band patterns obtained from the gels were converted and analyzed using the Quantity One software, version 4.3.1 (Bio-Rad). PCR-MP procedure was carried out according to the method described for E. coli isolates (Krawczyk et al., 2006) with slight modifications. The DNA concentration range was about 100200 ng per microliter. Denaturation temperature was calculated during the optimization experiments for several E. faecium isolates using a gradient thermal cycler (Biometra Tgradient Engine) with a gradient range of 79.6 82.5°C for denaturation step. PCRs were performed as follows: 7 min at 72°C to release unligated oligoTable I Patients clinical histories and results of genotyping of the vancomycin-resistant E. faecium isolates Age Patient clinical Isolate PCR MP/ Day(s) Patient a PFGE of isolahistory source no. Sex (yr) genotype tionb myeloid leukaemia myeloid leukaemia

1

M

63

2

M

57

3

M

47

myeloid leukaemia

M

38

myeloid leukaemia

5

F

49

non-Hodgkin lymphoma

6

M

24

multiple myeloma

7

F

56

myeloid leukaemia

8

F

43

9

M

25

10

F

63

11

F

26

12

M

64

Experimental Materials and Methods

Isolates and patients. Clinical samples obtained between April 2003 and April 2005 from patients of the single Hematological Unit at the University Hospital in Gdañsk were examined for the presence of E. faecium isolates. These isolates were tested for the resistance to vancomycin and teicoplanin. A 36 E. faecium VRE isolates (from 12 patients) were chosen for further examination by molecular typing methods. A PCR assay for identification of E. faecium and primers used were the same as in Cheng et al. (1997), with some modifications as described by Samet et al. (1999). A multiplex PCR-restriction fragment length polymorphism (MPCR/RFLP) assay for Van-type identification of E. faecium isolates were carried out according to Patel et al. (1997). Amplification products obtained with Van-specific primers were further analysed by digestion with MspI restriction endonuclease (RFLP). VRE isolates were recovered from blood (21 isolates), stool (12 isolates), urine (1 isolate), sputum (1 isolate) and abscesses (1 isolate). The medical records of the patients from whom these isolates were recovered were examined for information on sex, age, and clinical history (Table I). Genotyping methods. PFGE was performed according to the method described previously (Krawczyk et al., 2003). DNA was digested overnight with 25 U of XbaI (Sigma) at 37°C. DNA was separated on an agarose gel using the GenePath instrument (based on

4

a b

myeloid leukaemia lymphoid leukaemia lymphoid leukaemia myeloid leukaemia myeloid leukaemia

Blood Blood Blood Blood Blood Blood Blood Blood Sputum Abscess Blood Blood Stool Stool Blood Urine Blood Blood Stool Stool Stool Blood Blood Blood Stool Stool Blood Stool Blood Stool Stool Blood Stool Blood Stool Blood

A A B B A A A A A D B B C C A A A B B B A B B B E A F B A A A A B A A A

1 153 1 31 1 2 4 8 24 150 1 3 37 67 1 36 290 1 48 199 1 148 149 150 162 342 1 7 1 9 1 33 1 146 1 12

M, male; F, female. Day(s) of isolation refers to the day of the first day of isolation.

2

67

Enterococcus faecium VRE strains genotyping by PCR-MP

nucleotides and to fill in the single-stranded ends and create amplicons, followed by initial denaturation at 81.2°C for 90 s and 24 cycles of denaturation at 81.2°C for 1 min, annealing and elongation at 72°C for 2 min. After the last cycle, samples were incubated for 5 min at 72°C. PCR products, 15 µl out of 50 µl, were electrophoresed on 6% polyacrylamide gels with TBE buffer, stained in ethidium bromide (0.5 µg/ml aqueous solution) for 1015 min. Images of the gels were analyzed using a Versa Doc Imaging System version 1000 (BioRad). Similarities between fingerprints were calculated by use the Dice band-based similarity coefficient (SD). The patterns with the Dice coefficient of 0.85 were assigned to the same type. Results Patient characteristics and PCR identification of E. faecium and Van-type. Table I shows the characteristics of the patients, the sites and times of recovery of the VRE isolates, and the genotyping analysis. The mean age for the patients was 46.2 ± 22.2 years (age range, 24 to 64 years). All patients had been hospitalized several times, and the mean hospital stay for the patients was 42.9 ± 24.5 days. Ten out of twelve of them died during hospitalization. To confirm the phenotypic identification, the PCR identification of E. faecium and Van-type of antibiotic resistance was

A

carried out. Using the EM1A and EM1B primers (Cheng et al., 1997), a specific 658-bp DNA product, upon PCR amplification of DNA from all isolates identified as E. faecium by standard biochemical assays, was identified (results not shown). These results confirmed that examined isolates in fact belong to E. faecium. Next, the convenient multiplex PCR-restriction fragment length polymorphism (MPCR/ RFLP) assay to detect and discriminate vanA, vanB and vanC-1 genes was applied (Patel et al., 1997). All examined clinical isolates, phenotypically identified as vancomycin-resistant VanA-type of the E. faecium, yielded the 885-bp amplicon, which is characteristic for vanA and vanB genes. The amplified DNAs from all isolates digested with MspI restriction enzyme gave distinct electrophoretic patterns for vanA gene (231, 184, 163, 133 and 131 bp restriction fragments) (results not shown). These experiments confirmed that the isolates belong in fact to VanA-type of the vancomycin resistance. PFGE and PCR MP analysis. PFGE and PCRMP fingerprinting patterns of the 36 VRE found only six unique profiles represented by A to F groups (Figs 1 and 2, Tab. I). Each PFGE (XbaI) pattern consists of approximately 1217 fragments. PCR-MP profile consists of approximately 20 25 fragments in the size range of 1001200 bp. Two groups, A and B, were markedly predominant, as these were represented by 53% and 33%, respectively. Successive isolates

B

Fig. 1. PFGE profiles of the vancomycin-resistant E. faecium strains. Chromosomal DNA was digested with XbaI, and the fragments were fractionated on an 0.9% agarose gel. Numbers 1 12 refer to patients shown in Table I. PFGE genotypes are given above each lane.

68

2

Krawczyk B. et al.

A

B

Fig. 2. PCR MP fingerprints of the vancomycin-resistant E. faecium strains. The amplicons were electrophoresed in 6% polyacrylamide gel by using 1 × Tris-borate EDTA running buffer at a field strength of 12 V/cm. The lane designated M contained molecular mass marker (1008, 883, 615, 517, 466 and 396 bp). PCR MP genotypes are given above e ach lane. Numbers 112 refer to patients shown in Table I.

with identical patterns were recovered from 7 patients. Different genotypes were found in the 5 patients. The delay between the time of recovery of the first E. faecium VRE isolate and the last isolate of the same

A

genotype was in range from 9 days to about 1 year. For patients 1, 2 and 5, the same E. faecium VRE genotype could still be isolated from the blood after 5, 1 and 10 months, respectively. Long-term persis-

B

C

Fig. 3. PCR MP fingerprints of the vancomycin-resistant E. faecium strains at increasing denaturation temperatures (79.9°C, 80.5°C, 81.0°C, and 81.5°C) (representative results). Isolates from patients 10 (panel A), 6 (panel B) and 11 (panel C) are compared. A steady increase in the number of amplified DNA fragments, which is dependent on Td increase, was observed and still produced identical profiles for isolates belonging to the same genotype (panels A and B).

2

Enterococcus faecium VRE strains genotyping by PCR-MP

tence (approximately 1 year) of E. faecium VRE was detected for patient 7. Isolates of that patient were typed into three groups (A, B, E). Genotype A was identified first and at the end (both from stool). Genotype B was detected only from blood (3 isolates). In six patients (patients 3, 5, 6, 9, 10, and 12), the paired blood- and not-blood- related isolates from each patient were of identical genotype. The reproducibility of PFGE and PCR-MP methods was assessed by the duplicate analyses of the isolates. For these duplicates, DNA was extracted separately and proceeded in independent procedures. The resulting products were then electrophoresed on the same gel. All paired PFGE or PCR-MP patterns were found to be identical indicating that the methods were highly reproducible (data not shown). To increase differentiation efficiency of the strain genotyping, isolates from particular patient were tested with PCR-MP at increasing denaturation temperatures. A steady increase in the number of amplified DNA fragments, which is dependent on Td increase, was observed and still produced identical profiles for isolates belonging to the same genotype (for representative results see Fig. 3). Thus, the order of appearance of DNA bands in PCR performed in subsequent increasing temperatures is invariable for a given genomic DNA (genotype) (see Fig. 3, panels A and B). Discussion In the report presented here, we describe the genotypic similarities of multiple E. faecium VRE isolates consecutively recovered from 12 patients. The samples from the majority of the patients contained only a single PFGE/PCR-MP genotype. However, in some cases coinfection with other genotype was found. In addition, in some cases of bacteremic patients, the blood- and non-blood derived E. faecium VRE isolates from the same patient had identical E. faecium VRE genotype. The genotypic similarities of the consecutive isolates from the patients described here indicate the ability of E. faecium VRE isolates to persist as both infecting and colonizing flora for a long time. The present report confirmed the results described previously, where VRE isolates from patients of Hematological Unit between 1996 and 1997 analyzed using PCR fingerprinting (RAPD) technique (Samet et al., 1999) and between 1997 and 1999 analyzed using ADSRRS-fingerprinting and PFGE techniques (Krawczyk et al., 2003) revealed the presence of two main closely related genotypes (groups A and B). The epidemiological studies presented here revealed that those genotypes still exist in Hematological Unit of the hospital. Only minor differences in PCR MP fingerprinting patterns were observed in both groups (not

69

shown). It is due to the ongoing evolutionary diversification process within their populations. PCR-MP analysis of the E. faecium VRE isolates produced reliable, discriminatory, and reproducible typing results; and the existence of distinct PCR-MP types in single patients was supported by differences in PFGE identification. Although PFGE analysis of chromosomal macrorestriction fragments is one of the most commonly used method for the epidemiological typing of bacteria and it is generally recognized to be the method with the highest level of discrimination, we recommend the use of PCR MP technique for routine epidemiological study. The complexity of the PFGE and the costs involved in setting up and using the method may be beyond the capabilities of most laboratory. Besides, PFGE is the method probably not suitable for long-term epidemiology because the evolution of the strains might be too fast. Theoretically, the patterns of the larger bands should have a tendency to evolve into unrelated patterns at a significantly faster rate than that of the small bands. Indeed, the frequency of random genetic events increases with the size of the DNA fragment. These genetic events include point mutation generating a new restriction site, insertion/deletion of a sequence, and rearrangement. In this context, the PCR-MP typing method, in contrast to PFGE, permits the study of genetic relationships based on relatively small restriction fragments. Using PCR-MP method we also have the possibilities to increase the number of amplified DNA restriction fragments by increasing denaturation temperature during PCR. A steady increase in the number of amplified DNA fragments is dependent on denaturation temperature increase. Considering the lower costs and a high discriminatory power of PCR-MP method, we concluded that PCR-MP is a better choice for epidemiological studies (mainly long-term epidemiology analysis) than the use of PFGE analysis of chromosomal macrorestriction fragments obtained with several restriction enzymes. In conclusion, the PCR-MP is generally a simple technique with high discriminatory power and low cost and may be most suitable for epidemiological studies. As shown in this report. a PCR-MP analysis proved to be a suitable technique for determination of the genetic similarities of consecutive E. faecium VRE isolates. Acknowledgements This work was supported by the Polish State Committee for Scientific Research Grant N404 078 31/3454 to B.K.

Literature Banerjee S.N., T.G. Emori, D.H. Culver, R.P. Gaynes, W.R. Jarvis, T. Horan, J.R. Edwards, J. Tolson, T. Henderson and W.J. Martone. 1991. Secular trends in nosocomial primary

70

Krawczyk B. et al.

bloodstream infections in the United States 19801989. National nosocomial infections surveillance system. Am. J. Med. 91: 86S89S. Bonten M.J.M. and M.K. Hayden. 1996. Epidemiology of colonisation of patients and environment with vancomycin-resistant enterococci. Lancet 348: 16151619. Cheng S., F.K. McCleskey, M.J. Gress, J.M. Petroziello, R. Liu, H. Namdari, K. Beninga, A. Salmen and V.G. Del Vecchio. 1997. A PCR assay for identification of Enterococcus faecium. J. Clin. Microbiol. 35: 12481250. Hryniewicz W., K. Szczypa, M. Bronk, A. Samet, A. Hellmann and K. Trzciñski. 1999. First report of vancomycin-resistant Enterococcus faecium isolated in Poland. Clin. Microbiol. Infect. 5: 503505. Jarvis W.R. and W.J. Martone. 1992. Predominant pathogens in hospital infection. J. Antimicrob. Chemother. 29: 1924. Krawczyk B., K. Lewandowski, M. Bronk, A. Samet, P. Myjak and J. Kur. 2003. Evaluation of novel method based on amplification of DNA fragments surrounding rare restriction sites (ADSRRS-fingerprinting) for typing strains of vancomycin-resistant Enterococcus faecium. J. Microbiol. Methods 52: 341351.

2

Krawczyk B., A. Samet, J. Leibner, A. ledziñska and J. Kur. 2006. Evaluation of a PCR melting profile (PCR MP) technique for bacterial strain differentiation. J. Clin. Microbiol. 44: 23272332. Maki D.G. and W.A. Agger. 1988. Enterococcal bacteremia: clinical features, the risk of endocarditis and management. Medicine 67: 248269. Moellering R.C. Jr. 1992. Emergence of Enterococcus as a significant pathogen. Clin. Infect. Dis. 14: 11731178. Murray B.E. 1990. The life and times of Enterococcus. Clin. Microbiol. Rev. 3: 4665. Murray B.E. 1998. Diversity among multidrug-resistant enterococci. Emerg. Infect. Dis. 4: 3747. Patel R., J.R. Uhl, P Kohner, M.K. Hopkins and F.R. Cockerill III. 1997. Multiplex PCR detection of vanA, vanB, vanC-1, and vanC-2/3 genes in enterococci. J. Clin. Microbiol. 35: 703707. Samet A., M. Bronk, A. Hellmann and J. Kur. 1999. Isolation and epidemiological study of vancomycin-resistant Enterococcus faecium from patients of a haematological unit in Poland. J. Hosp. Infect. 41: 137143. Schaberg D.R., D.H. Culver and R.P. Gaynes. 1991. Major trends in the microbial etiology of nosocomial infections. Am. J. Med. 91: 72S75S.


Selection and Activation of Escherichia coli Strains for L-aspartic Acid Biosynthesis MIROS£AW PAPIERZ, GRA¯YNA GADOMSKA, BOGUS£AW SOBIERAJSKI and ALEKSANDER CHMIEL*

Department of Biosynthesis of Drugs, Chair of Biology and Pharmaceutical Biotechnology, Medical University of £ód, £ód, Poland Received 7 February 2007, accepted 20 March 2007 Abstract The strain of Escherichia coli K-12 with high aspartase activity was irradiated with UV. After mutagenesis and selection, the mutant B-715 was isolated which was 4-times more active in L-aspartic acid biosynthesis than parental K-12 strain. The highest productivity was achieved while the strain was cultivated in the ammonium fumarate medium in 37°C for 1830 hours. It was found that better results were obtained when before the main production step of biosynthesis of L-aspartic acid, the cells of E. coli B-715 were incubated in the activation medium with ammonium fumarate. Activation at 37°C was the most advisable for high efficiency of L-aspartic acid biosynthesis. The productivity of E. coli B-715 during 1 hour biosynthesis process was at the range 0.19 0.35 g of L-aspartic acid per 1 gram of dry mass (biomass) per minute. K e y w o r d s: L-aspartic acid biosynthesis, productive strains selection

Introduction Aspartic acid is an amino acid used in pharmacy, cosmetics and food industry (Chibata et al., 1985; Mazo et al., 1999a; 1999b). It is produced from the fumaric acid and ammonia either by chemical synthesis or in biotechnological process, i.e. in enzymatic reaction. In the first case a racemate mixture is produced, while the enzymatic synthesis ensures production of the enantiomeric form L of aspartic acid. For the synthesis of L-(+)-aspartic acid L-aspartate ammonia-lyase EC 4.3.1.1 (aspartase), enzyme from Escherichia coli is applied. The catalytic activity of that enzyme in E. coli was described very early (Quastel and Woolf, 1926). Depending on the pH value two opposite directions of reaction are possible: in acidic conditions aspartic acid is deaminated to fumarate and ammonia, but in alkaline pH L-aspartic acid in solution of fumaric acid and ammonia (or ammonium fumarate) is produced. The first industrial application of aspartase for fermentative L-aspartic acid production was carried out with E. coli strain B by Tanabe Seiyaku Co. in 1960 (Kisumi et al., 1960), and further developments in the L-aspartic acid biotechnology

mostly have been made in Japan. Important improvements of the biosynthesis process were developed using the immobilized cells (Chibata et al., 1974; Nishida et al., 1978; Sato et al., 1979), and the aspartase-hyperproducing strains of E. coli derived from the strain B and K-12 were introduced (Nishimura and Kisumi, 1984; Nishimura et al., 1989). Further technological improvement was made using recombined clones of E. coli (Takano and Kino, 1999; Komatsubara et al., 1986; Nishimura et al., 1987a; 1987b; Mukouyama and Komatsuzaki, 2001). In Poland, aspartic acid (racemate mixture) is produced mainly by means of chemical synthesis. In our Department investigations on the enzymatic biosynthesis of L-aspartic acid have been started. This paper deals with the strain screening, selected strain mutagenesis and the aspartase-active mutants selection. Thereafter, E. coli recombinant strains characterized by L-aspartic acid overproduction were constructed. Regardless of a very wide distribution of aspartase in microbial world, according to the above cited papers E. coli, common organism populating intestines of humans and other mammalians, seems to be the best source of that enzyme. Our search for the aspartase

* Corresponding author: A. Chmiel, Department of Biosynthesis of Drugs, Chair of Biology and Pharmaceutical Biotechnology, Medical Academy of £ód, Muszyñskiego 1, 90-151 £od, Poland; e-mail: [email protected]

72

2

Papierz M. et al.

overproducers involved both the known collection strains of E. coli and the strains newly isolated from humans. To induce the highly active strains mutagenesis was applied and a procedure for the cell cultivation and their subsequent activation before their use for L-aspartic acid production was standardized. Experimental Materials and Methods

Microorganisms. Bacteria used in this work were the collection strains of E. coli B and K-12, as well as new isolates from humans. The isolates were nonpathogenic variants of E. coli received from the Sanitary Epidemiological Laboratory in £ód. Media. For preliminary cultivation of bacteria, LB medium containing Tryptone (Difco) 10 g/l, Yeast Extract (Difco) 5 g/l, NaCl 10 g/l; pH 7.5, was used. A standard growth medium contained: Yeast Extract (Difco) 20 g/l, ammonium fumarate 5 g/l, fumaric acid 10 g/l, KH2PO4 2 g/l, MgSO4 × 7 H2O 0.5 g/l; pH 7.2. For testing different nitrogen sources for active cells multiplication we used media containing (NH4)2SO4 1 g/l, aspartic acid 5 g/l or ammonium fumarate 5 g/l and K2HPO4 7 g/l, KH2PO4 3 g/l, MgSO4 × 7 H2O 0.1 g/l; pH 7.2. Media for both the cells activation and aspartase activity testing contained fumaric acid 5 and 100 g/l respectively, MgSO4 × 7 H2O 0.25 g/l and Triton 0.5 ml/l; pH 8.5. Mutagenesis procedure. Bacteria growing overnight in 250 ml shake flasks containing 50 ml of LB medium in temperature 34°C were centrifuged at 4500 rpm and suspended in 10 ml 0.9% NaCl solution. The cells were mutagenized using UV lamp with 8 = 254 nm and different time of illumination to aim at a different cell survival. Colony activity test. An agar-paper printing technique was developed. After colony culturing on the growth medium solidified with agar the pieces of agar medium (with diameter of 8 mm) containing tested colonies were put on the filter paper for several minutes. The correct time of exposition for sufficient extraction of product from medium pieces to paper was experimentally chosen. Thereafter, the paper was sprayed with 0.1% ethanolic solution of ninhydrin and placed in an air dryer for 57 min. Time of the blue spots appearance and their size and intensity were observed for choosing the best L-aspartic acid producers. Cell suspension activity test. The cells were cultured in 50 ml of the growth medium in 250 ml shak flasks for 18, 24, 30 or 48 hours in 37°C and thereafter centrifuged at 4500 rpm for 20 min.. The pelleted cells were washed 3-times with 50 ml of sterile water. In preliminary investigations non-activated (i.e. non-

permeabilized) cells were used. For further studies the cells were activated in activation medium in 4, or 37°C for 24 or 48 h. After centrifuging and washing 1 g of fresh biomass was incubated in 10 ml of the production medium in 100 ml shake flasks for 34 h; every 1 h samples of 1 ml volume were withdrawn for an aspartic acid analysis. Aspartic acid analysis. For semi-quantitative analysis of the product TLC analysis using DCAlufolien Kiselgel 60 plates (Merck) was applied. Aliquots of 10 ml of the 100-time dilutions of the samples were placed on the plates and chromatographed in a mixture of ethanol and water (67:33 v/v) for 1.5 h. The plates were then sprayed with 0.1% ethanolic solution of ninhydrin and placed in an air dryer for 57 min. The single blue spots presented L-aspartic acid production. For precise estimation of the quantity of L-asparic acid HPLC technique was applied. The samples were deproteinised by with methanol. Methanol served as inactivation reagent of aspartase. Four volumes of methanol was added to the samples and the obtained mixture was centrifuged. After dilution and addition of OPT-thiol reagent, 20 ml sample was injected into HPLC system: Column 250-4 LiChrospher 100 RP-18 (5 microm.)-Merck, Waters pump type 600 and Waters fluorescence detector type 474 were used. A mobile phase was the following: 200 ml methanol (Backer HPLC analysed) in 800 ml 0.05 M sodium phosphate buffer. Retention time for aspartic acid was 8 min, flow-rate 1ml/min, pressure 2100 PSI. The system was operated at temperature 22°C. Cell mass determination. Centrifugal tubes of known dry weight were used for centrifugation of 1 ml cell suspensions at 4500 rpm for 20 min. After removal of the supernatant by decantation the cells were washed with distilled water, centrifuged again, and then dried overnight at 60°C and subsequently in 105°C by 1 h. After cooling to room temperature the samples of dryed cells were weighed. Results Strains selection. For preliminary screening of the aspartase productivity two commonly known strains of E. coli; B and K-12, and ten new nonpathogenic isolates of E. coli were chosen. In 4 h incubation of cell suspension test two strains, i.e. K-12 and an isolate no. 44, presenting over 5-times higher product accumulation were selected (Table I). For further improvement of the productivity by mutagenization E. coli K-12 strain was selected. Selection of the aspartase overproducing mutants for L-aspartic acid biosynthesis. UV-irradiation of the cells of E. coli K-12 was conducted as

2

73

L-aspartic acid biosynthesis by selected strains of E.coli Table I L-aspartic acid production by tested E. coli strain in a cell suspension test in 4 h biosynthesis Strain tested

L-aspartic acid [g/l]

23 38 40 41 42 43 44 93 94 95 B K-12

0.34 0.29 0.31 0.28 0.28 0.22 1.90 0.21 0.29 0.25 0.32 2.18

Table III L-aspartic acid biosynthesis using the cells of E. coli B-715 activated through 48 h at 37°C Biosynthesis time [min]

Table II L-aspartic acid biosynthesis by the activated (permeabilised) cells of E. coli K-12 and their hyperproducing mutants in four-hour-process (through 4 h)

K-12 A-7 B-715

1.3 66.1 103.5 107.4 117.1 121.9

* sample taken directly after mixing the cells with production medium; contact time about 2 min.

Non-activated cells were used.

Strain tested


0* 30 60 120 180 240

L-aspartic acid [g/l] 14.6 50.0 62.4

Before using for a biosynthesis test the cells were permeabilised in the activation medium at 27°C through 18 h.

described in Materials and Methods. After mutagenesis a single colony valuation test for the selection of high-L-aspartic acid-producing mutants was applied. In a ninhydrin test on a filter paper after imprinting of

the colony-agar pieces, about 4000 colonies in four successive mutagenesis processes were selected. On the basis of the blue spots, their size and intensity as well as time of their appearance, dozen of colonies were selected. In comparison to the parental strain of E. coli K-12, the best UV-mutant (designated A-7) selected after the first mutagenesis step, presented more than 3-time higher aspartase activity in the cell suspension test for L-aspartic acid production. This mutant was used for subsequent mutagenesis steps, and as a result of two-stage selection procedure the mutant of E. coli B-715 was isolated which was more than 4-times as active as K-12 (Table II). Standardization of the cell treatment for L-aspartic acid production. The cell activation, i.e. cell membrane permeabilization, apart from the metabolic potential of the strain used, is an important factor in the process of transformation of ammonium fumarate to L-aspartic acid; as it was found earlier by Japanese investigators (Chibata et al., 1974). Because this property is very important in a strain testing procedure as

140 37°C / 48 h


120

37°C / 24 h

100 80

4°C / 48 h

60 4°C / 24 h

40

activation-

20 0 0

30

60

90

120 time [min]

150

180

Fig. 1. Effect of temperature and activated time on L-aspartic acid biosynthesis.

210

74

2

Papierz M. et al. 140


120

activation + keeping

100

activation + keeping +

80 60 40

activation keeping activation keeping +

20 0 0

30

60

90

120 time [min]

150

180

210

Fig. 2. Effect of keeping cells in the culture medium on L-aspartic acid biosynthesis.

well as for further development of a process of L-aspartic acid biosynthesis it was advisable to find the best method for the cell permeabilization. We have applied two different temperatures for the cell incubation in an activation medium: 4° or 37°C. Activation in temperature 37°C was the most advisable for efficiency of L-aspartic acid biosynthesis. The cells activated in 4°C were less productive. Further improvement of cell activity occurred after the prolongation of activation time to 48 h (Fig. 1). The influence of keeping cells in a culture medium on biosynthesis of L-aspartic acid was also tested. The need to keep cells in a culture medium before their

use for biosynthesis process sometimes occurs in technology process. The cells were kept in temperature about 4°C for 24 hours without shaking. This procedure was unprofitable for both the non-activated and activated cells of E. coli. The cells can be activated only in an activation medium with shaking (Fig. 2). In the next experiment the effect of culture duration of mutant E. coli B-715 and also the effect of temperature of the cell activation were investigated. The cells were cultured at 37°C for 18, 24, 30 or 48 hours and subsequent they were suspended in the activated medium at 22, 27, 32 or 37°C for 24 h. The temperature 37°C was again the best activated tempera-

70

Conversion [%]

60 50 40 30 20 10 0

22

27 32 activation temperature [°C]

37

Fig. 3. Effect of culture duration ( 18, 24, 30 and 48 hrs) and temperature of the cell activation (22°C, 27°C, 32°C, 37°C) on L-aspartic acid production.

2

75

L-aspartic acid biosynthesis by selected strains of E.coli 140


120 100 80 60 40 20 0 after substrate adding

30

60

90

120

150

180

210

240

biosynthesis time [min]

Fig. 4. L-aspartic acid biosynthesis using the cells of E. coli B-715, activated at 37°C through 48 h.

ture for L-aspartic acid production. An effect of culture duration on effective biosynthesis for 18 30 h was insignificant, but prolonging culture time to 48 h was unfavorable (Fig. 3). Partial cell autolysis during prolonged culture cannot substitute activation of cells in substrate solution. This study stage was recapitulated following the experiment in which the bacteria were proliferating for 24 h at 37°C and activated at 37°C for 48 h (Fig. 4 and Table III). Discussion Different mutated and recombined strains of Escherichia coli have been used for production of L-aspartic acid from amonium fumarate. Other bacteria strains, for example: Proteus vulgaris OUT 8226, Pseudomonas aeruginosa OUT 8252, Serratia marcescens OUT 8259, Bacterium succinium IAM 1017 have been rarely proposed (Chibata et al., 1974). In our laboratory, selection strains E. coli with high aspartase activity have been undrtaken. Out of ten new nonpathogenic isolates of E. coli and two collection strains of E. coli B and K-12, strain E. coli K-12 was chosen for further study. The strain chosen is very well characterized and was found to have high aspartase activity. The strain E. coli K-12 were irradiated with UV. Mutant of E. coli A-7 and better overproducing mutant of E. coli B-715 were isolated as the result of mutagenesis and investigatied for the yields of L-aspartic acid production. E. coli B strain was used by Chibata et al. (1974). The yields of L-aspartic acid produced by this was

tested. After 60 min and 180240 min duration of the conversion process was at the of level 50% and about 90%, respectively. The strain E. coli ATCC 11303 was used in the study which resulted in the industrial technology. After 1 hour process, the cells of the strain (0.2 g dry mass and 30 ml production medium) produced 11 290 µmol L-aspartic acid. The productivity of 1 g dry cells was 0.125 g/g/min. This result cannot be compared with our results because authors did not determine the starting rate. In order to compare our results with those achieved by Chibata, productivity of strain E. coli B-715 was also calculated after 1 hour duration of biosynthesis. The calculated productivity was 0.190.35 g/g/min in different experiments. It was demonstrated that mutant of E. coli B-715 is considerably more active than the strain E. coli used in the first industrial technology of L-aspartic acid production. Maximum value of biosynthesis velocity obtained in first minutes of a periodic process using mutant of E. coli B-715 was 2.2 g/g/min, so the productivity calculated on 1 g dry biomass of cells was 0.44 g/g/min. The mutant of E. coli B-715 has been used as an aspartase gene source for further strain improvement using gene cloning procedure (Gadomska et al., 2007). Literature Chibata I., T. Tosa and T. Sato. 1985. Aspartic acid. Comprehensive Biotechnology, vol. 3: 633640, Moo-Young M. (ed.) Pergamon Press, Oxfort-New York Chibata I., T. Tosa and T. Sato. 1974. Immobilized aspartasecontaining microbial cells: preparation and enzymatic properties. Appl. Microbiol. 27: 878885.

76

Papierz M. et al.

Gadomska G., A. P³ucienniczak and A. Chmiel. 2007. Development of L-aspatic acid biotechnology. Recombinant strains of Escherichia coli for L-aspartic acid biosynthesis. Pol. J. Microbiol. 56: 7782. Kisumi M., Y. Ashikaga and I. Chibata. 1960. Studies on the fermentative preparation of L-aspartic acid from fumaric acid. Bull. Agric. Chem. Jpn. 24: 296305. Komatsubara S., T. Taniguchi and M. Kisumi. 1986. Overproduction of aspartase of Escherichia coli K-12 by molecular cloning. J. Biotechnol. 3: 281291. Mazo G.Y., J. Mazo, Jr.B. Vallino and R.J. Ross. 1999a. Production of D, L-aspartic acid. United States Patent: 5 872 2 85. Mazo G.Y., J. Mazo, Jr.B. Vallino and R.J. Ross. 1999b. Production of D, L-aspartic acid. United States Pataet: 5 907 057. Mukouyama M. and S. Komatsuzaki. 2001. Method for producing L-aspartic acid. United States Patent: 6 214 589. Nishida Y., T. Sato, T. Tosa and I. Chibata. 1979. Immobilization of Escherichia coli cells having aspartase activity with carrageenan and locust bean gum. Enzyme. Microb. Technol. 1: 9599. Nishimura N. and M. Kisumi. 1984. Aspartase-hyperproducing mutants of Escherichia coli B. App. Environ Microbiol. 48: 10721075.

2

Nishimura N., S. Komatsubara and M. Kisumi. 1987a. Increased production of aspartase in Escherichia coli K-12 by use of stabilized aspA recombinant plasmid. App. Environ. Microbiol. 53: 28002803. Nishimura N., S. Komatsubara, T. Taniguchi and M. Kisumi. 1987b. Hyperproduction of aspartase of Escherichia coli K-12 by the use of runway plasmid vector. J.Biotechnol. 6: 3140. Nishimura N., T. Taniguchi and S. Komatsubara. 1989. Hyperproduction of aspartase by a catabolite repression-resistant mutant of Escherichia coli B harboring multicopy asp A and par recombinant plasmids. J. Ferment. Technol. 67: 107110. Quastel I. H., and B. Woolf. 1926. Equlibrium between L-aspartic acid and fumaric acid and amonia in presence of resting bacteria. Biochem. Z. 250: 193211. Sato T., Y. Nishida, T. Tosa and I. Chibata. 1979. Immobilization of Escherichia coli cells containing aspartase activity with k-carrageenan. Enzymic properties and application for L-aspartic acid production. Biochi. Biophys. Acta 570: 179186. Takano J. and K. Kino. 1999. Process for producing aspartase and process for producing L-aspartic acid. United States Patent: 5 916782.


Recombinant Strains of Escherichia coli for L-aspartic Acid Biosynthesis GRA¯YNA GADOMSKA1, ANDRZEJ P£UCIENNICZAK2, 3 and ALEKSANDER CHMIEL1* 1 Department

of Biosynthesis of Drugs, Chair of Biology and Pharmaceutical Biotechnology, Medical University of £ód, £ód, Poland; 2 Department of Genetic Engineering, PP Terpol, Sieradz, Poland; 3 Institute of Biotechnology and Antibiotics, Warsaw, Poland Received 7 February 2007, accepted 20 March 2007 Abstract The aspartase overproducing mutant B-715 was used as a donor of the aspartase gene for further construction of the aspartasehyperproducing strains by molecular cloning. In preliminary experiments activity of transformants and their efficiency in L-aspartic acid biosynthesis were compared. The conditions for recombinant strain multiplication, biomass activation and L-aspartic acid biosynthesis were optimized. The optimum temperature for cells multiplication, their activation and for product biosynthesis was 37°C. Twostage process of the multiplication of bacteria (first in LB medium, and then in FF medium) eliminates the appearing of the inclusion bodies of aspartase in the cells. The shaking during cell activation improved cells productivity. The change of pH in the course of the biosynthesis process was insignificant but did not influence the process. K e y w o r d s: L-aspartic acid biosynthesis, conditions optimization for biosynthesis of L-aspartic acid

Introduction Aspartic acid can be produced as a racemate mixture using chemical technology (Mazo et al., 1999a; Mazo et al., 1999b) or as its L form using enzymatic process with aspartase (L-aspartate ammonia-lyase, EC 4.3.1.1) (Virtanen and Tarnanen, 1932; Waller, 2001; Chibata et al., 1974). In our laboratory a search for the aspartase-active bacteria among both the collection strains and new isolates of E. coli resulted in selection of a well known E. coli K-12 strain as the best producer of that enzyme (Papierz et al., 2007). Its UV-mutant B-715 with a significant overproduction of aspartase was used as a donor of the aspartase gene for further construction of aspartase-hyperproducing strain by molecular cloning. This paper deals with the clones construction and preliminary investigations in their culturing and stability, as well as their use for L-aspartic acid biosynthesis. Experimental Material and Methods

Bacteria. The strains used in this work were E. coli UV-mutant B-715 and four recombinant strains E. coli P-1, P-2, P-3 and P-4. The strain B-715 was obtained

in our laboratory as a result of UV mutagenesis of parental strain E. coli K-12 (Papierz et al., 2007). Suspensions of bacteria were mixed with 50% glycerol (1:1), frozen and stored in 70°C for further use (stocks). Construction of recombinants of Escherichia coli. Recombinants E. coli P1, P2, P3 and P4 include multiply copy aspartase gene localized in plasmid vector. As a source of the aspartase gene the mutant E. coli B-715 obtained in our laboratory (Papierz et al., 2007) was used (Fig. 1). PCR techniques was applied for the gene multiplication using two primers: ASP1.SEQ[20]: 5 GGT TCA TAT GCC AAA CAA CA 3 and ASP2.SEQ[26]: 5 AAA AAG CTT ACT GTT CGC TTT CAT TC 3, which were synthesized at the Institute of Bioorganic Chemistry of the Polish Academy of Sciences (Poznañ). PCR product was digested using endonucleases NdelI and HindIII, and thereafter inserted into the pBS+ vector. Resulting plasmid pBSASP1 was multiplied in E. coli BLD21 (DE3) and digested with NdelI and HindIII. The ASP gene was then introduced into a plasmid expression pT7-7.This plasmid includes ampicillin resistance gene as a marker and a strong promoter of RNA polimerase of phage T7. Recombinant plasmid pT7ASP1 was introduced into the ampicillin resistance strain E. coli BLD21(DE3). This strain is deficient in both Lon and

* Corresponding author: A. Chmiel, Department of Biosynthesis of Drugs, Chair of Biology and Pharmaceutical Biotechnology, Medical University of £ód, Muszyñskiego 1, 90-151 £od, Poland; e-mail: [email protected]

78

2

Gadomska G. et al.

cultivation conditions the shaken cultures of bacteria were carried out at 27 or 37°C for 16, 18, 20 or 24 hours. (3) Cells activation. The cell culture was centrifuged at 4000 rpm for 20 min and introduced into the activation medium (1 g of wet mass/20 ml). Tween 80 (0.5%) was added to obtain homogeneous cell suspension. The cell suspension was activated with or without shaking for 24 hours at the temperature of 27, 33, 37 or 40°C. Activated cell suspension was centrifuged at 4000 rpm for 20 min and biomass was washed twice with distilled water (1 g wet mass/ 20 ml water). (4) Biosynthesis of L-aspartic acid. Activated cells were mixed with the production medium (1 g wet mass/20 ml) in 100 ml flasks and shaken at the temperature of 27, 33, 37 or 40°C. After 15, 30 and 60 min of incubation the samples of 1 ml were withdrawn for analysis. Aspartic acid analysis. For estimation of L-aspartic acid HPLC technique was applied using column 2504 LiChrospherTM 100 RP-18 (Merck) and Waters fluorence detector type 474. The details were described in proceding paper (Papierz et al., 2007). Results and Discussion

Fig. 1. Construction of expression vector with aspartase gene.

OmpT proteases and include the gene of RNA polymerase originated from phage T7 integrated to chromosome. The T7 RNA polymerase gene is under the control of lac UV5 promoter and is induced by lactose or its analog isopropyl-1-thio-$-D-galactoside (IPTG). Media. (1) LB medium; Trypton 10.0 g/l; Yeast Extract (Difico) 5.0 g/l; NaCl 10.0 g/l; pH 7.5; (2) FF medium for multiplication of biomass (our laboratory): Yeast Extract (Difico); 20.0 g/l, ammonium fumarate 5.0 g/l, KH2PO4 11.4 g/l, MgSO4 × 7 H2O 0.5 g/l, pH 7,2. (3) Medium for cells activation (activate medium): ammonium fumarate 50.0 g/l, MgSO4 × 7 H2O 0.25 g/1; 1% Triton 0.50 ml/l, pH 8.5. (4) Medium for production of L-aspartic acid (productive medium): ammonium fumarate 150.0 g/l; MgSO4×7 H2O 0.25 g/l; pH 8.5. Chemicals, if not indicated otherwise, were purchased from POCh S.A, G (at analytical grade). Conditions of bacteria multiplication and product biosynthesis. (1) Inoculum preparation: 100 µl of cells suspension from a stock were introduced into 100 ml of LB medium and cultivated in shaken 250 ml flask over 6 hours at 37°C. For growing of recombinant cultures the ampicillin at the concentration of 0.1 g/l as selection factor was added. (2) Cell biomass multiplication. 100 ml of FF medium in 250 ml Erlenmayer flasks was inoculated with pre-multiplied inoculum culture or with a cells stock (100 µl per 100 ml). For growing of recombinants cultures ampicillin was added (0.1 g/l). For an optimization of the

Comparison of recombinants. In preliminary experiments aspartase expression and activity in four transformants of E. coli BLD21 (DE3)/pT7ASP1, designed as P1, P2, P3 and P4, was estimated. Bacteria were multiplied in liquid FF medium with ampicillin at 37°C for 24 hours. Electrophoresis of cell protein showed intensive stimulation of cloning gene expression in the presence of IPTG (Fig. 2), however results in the aspartase activity tests were unsatisfied, because the level of L-aspartic acid biosynthesis was low. Microscopic observation showed occurrence of inclusion bodies inside the cells. Under those condiP1

P1+

P2

P2+

P3

P3+

P4

P4+

Fig. 2. Protein electrophoresis of cell lysate of recombinants of E. coli P1, P2, P3 i P4 cultured in FF medium with (+) or without () IPTG (1 mM).

2

79

Biosynthesis of L-aspartic acid by recombinant strains of E. coli 140 P3 P1 P2


120 100

P4

80 60 40 20 0

0

15

30

45 60 time [min]

75

90

Fig. 3. Activity of recombinants E. coli P1, P2, P3 and P4 in productive medium

tions, the expression of the cloned aspartase gene effected synthesis of large amount of aspartase which was inactive and deposited in the inclusion bodies. Conditions to obtain the cells with high aspartase activity were searched. Inoculum multiplication in LB medium was conducted for 38 hours but usually 6 hours incubation was sufficient to obtain a density with the absorption of about 0.250.3 at 8 = 540 nm. FF growth medium without IPTG was inoculated using 2% of the inoculum cultures and incubated at 37°C for 1624 hours. For three clones, i.e.: P1, P2 and P3 the cells cultivated 1618 hours converted over 90% of substrate to L-aspartic acid already after 90 min of incubation (Fig. 3). Aspartase activity in the cells of recombinant P4 was considerably lower with statistical significance of " = 0.05 in Manna-Whitney U-test. In further experiments we present the results obtained for recombinant E. coli P1 which was very active and stable during long-term investigations.

Optimization of recombinants multiplication. In the preliminary experiments we observed the presence of inclusion bodies after prolonging culture duration and especially in the presence of IPTG. The bodies are often present in recombinants cells of E. coli if cloned gene is expressed, and protein deposited in the bodies is biologically inactive. Long-term observations made by one of us show that by lowering temperature of bacteria cultivation it is possible to oppose creating of inclusion bodies in the transformed cells. To confirm this, the strain E. coli P1 was cultivated in FF growth medium without IPTG at 27 and 37°C. In this experiment inclusion bodies inside the cells of bacteria were not found at both temperaturs. Moreover decrease of temperature of bacteria cultivation caused considerable decrease in aspartase activity (Fig. 4). The product synthesis rate in the first 15 minutes of the process for bacteria multiplied at 37°C was 5.5 g/l/min and for bacteria multiplied at 27°C was

Fig. 4. Effect of temperature of biomass multiplication on biosynthesis of L-aspartic acid by activated cells of E. coli P1 (1 g of biomass per 10 ml of substrate solution)

80

2

Gadomska G. et al. 100 90 80 Conversion [%]

70 60 50 40 30 20 10 0

16

18

20 22 culture duration [h]

24

Fig. 5. Effect of culture duration of E. coli P1 on conversion of ammonium fumarate to aspartic acid.

only 3.5 g/l/min. Decrease of cultivation temperature was disadvantageous for aspartase activity. For multiplication of biomass of recombinant P1, FF medium without IPTG was inoculated using the 6 hours culture of this strain in LB medium. Biomass was multiplied at the temperature of 37°C for 16, 18, 20, 22 or 24 hours, and then the cells were activated at 37°C for 24 hours and used for L-aspartic acid biosynthesis. Samples were taken after 60 minutes. The best conversion of ammonium fumarate to aspartic acid was observed after multiplication for 24 hours; however differences between the tested cultivation time variants were not significant (Fig. 5). Preliminary multiplication of bacteria in LB medium efficiently eliminate the problem of inclusion bodies, and prolonging of biomass cultivation even to 24 hours after applying this modification was advantageous. Biomass activation. To optimize the process of cell activation for the biomass multiplied at 37°C for

24 hours both the cell suspension shaking and four variants of activation temperatures (i.e. 27, 33, 37, and 40°C) were tested. For process of biosynthesis of L-aspartic acid at temperature 37°C, samples were taken after 5, 15, 30 and 60 minutes. The shaking during cell activation improved cells productivity by more then a factor of two; product biosynthesis rate in first 15 minutes of process for cells that were not shaken was 1.6 g/l/min, and for shaken cells was 3.4 g/l/min. The highest value of the starting maximum rate of the product synthesis, of more then 5.5 g/l/min, and the most efficient total production of L-aspartic acid were found for the cells activated at 37°C (Fig. 6). Optimization of temperature and pH for L-aspartic acid biosynthesis. As a reference, the process of production of L-aspartic acid was carried out at the temperature of 37°C. An effect of the decrease or increase temperature of 3°C was tested (Fig. 7). In this experiment the best conversion was observed at 37°C.

140 37°C


120

40°C 33°C

100

27°C

80 60 40 20 0

0

15

30

45 time [min]

60

75

Fig. 6. Effect of cell activation temperatures for E. coli P1 on biosynthesis of L-aspartic acid.

2

81

Biosynthesis of L-aspartic acid by recombinant strains of E. coli 140

37°C 40°C 33°C


120 100 80 60 40 20 0 0

15

30

time [min]

45

60

75

Fig. 7. Effect of biosynthesis temperature on L-aspartic acid production by E. coli P1

The starting maximum rate of the product synthesis was 6.0 g/l/min, while at the temperature of 33 or 40°C it was respectively 5.4 and 5.7 g/l/min respectively. Optimum pH for efficient conversion ammonium fumarate to L-aspartic acid in the reaction catalyzed by aspartase is 8.5 10.5 (Chibata et al., 1974; Tosa et al., 1974). In their studies, a production medium with pH 8.5 was used. In our experiment two starting values of pH 8.5 and 9.5 for the production medium were applied (Fig. 8). The change of pH during the process and substrate conversion to L-aspartic acid was investigated. Directly after adding wet mass to the production media pH decreased insignificantly. Constant uniform pH drift during the process was observed in both variants of the experiment but pH

in the investigated range did not influence the conversion process. The application of genetic engineering methods for improving strains is effective manner of rationalization for aspartic acid production. The bacteria with active aspartase were used for researches of L-aspartic acid biotechnology. The most often used were Escherichia coli and Pseudomonas fluorescens as well as representatives of genera Enterobacter and Citrobacter (Mukouyama and Komatsuzaki, 2001). The best producers (E. coli strains with high aspartase activity) were obtained by introducing aspartase gene using diverse plasmids vectors e.g.: pHC18, pSC101, pBR322, pBR325, pKK223-3, pACYC177, pACYC184 (Komatsubara et al., 1986; Nishimura et al., 1987a; 1987b;

10

100

9 8 7 pH

6

60

5 4

40

conversion [%]

80

3 2

20

1 0

0 0

15 30 time of biosynthesis [min]

60

Fig. 8. pH and conversion substrate during biosynthesis of L-aspartic acid by cells E. coli P1 in media of pH 8.5 and 9.5

82

Gadomska G. et al.

Nishimura et al., 1989; Mukouyama and Komatsuzaki, 2001; Mukouyama et al., 2004). Obtained recombinants have significantly better efficiency of L-aspartic acid production; for example the level of conversion of fumaric acid into L-aspartic acid by strain E. coli PUaspE2 was 20 times higher than for wild strain E. coli K-12 (Mukouyama and Komatsuzaki, 2001). E. coli P1 derivative constructed and used in our laboratory are over 12-times more effective than parents strain E. coli K-12 and over 3-times more effective than mutant E. coli B-715. Industrially application of production technology of L-aspartic acid required the immobilization of enzymes or cells. In our laboratory an efficient method for immobilization of E. coli B-715 and P1 strains was developed. The details are given in the proceding paper (Papierz et al., 2007). Literature Chibata I., T. Tosa and T. Sato. 1974. Immobilized aspartasecontaining microbial cells preparation and enzymatic properties. Appl. Microbiol. 27: 878885. Komatsubara S., T. Taniguchi and M. Kisumi. 1986. Overproduction of aspartase of Escherichia coli K-12 by molecular cloning. J. Biotechnol. 3: 281291.

2

Mazo G.Ya, J. Mazo, Jr.B. Valino and R.J. Ross. 1999a. Production of D,L-aspartic acid. United States Patent: 5 907 057. Mazo G.Ya, J. Mazo, Jr.B. Valino and R.J. Ross. 1999b. Production of D, L-aspartic acid. United States Patent: 5 872 285. Mukouyama M. and S. Komatsuzaki. 2001. Method for producing L-aspartic acid. Unitet States Patent: 6 214 589. Mukouyama M., S. Yasuda and S. Komatsuzaki. 2004. Method for producing L-aspartic acid. Unitet States Patent: 6 821 760. Nishimura N., S. Komatsubara, T. Taniguchi and M. Kisumi. 1987a. Hyperproduction of aspartase of Escherichia coli K-12 by the use of a runaway plasmid vector. J. Biotechnol. 6: 3140. Nishimura N., S. Komatsubara and M. Kisumi. 1987b. Increased production of aspartase in Escherichia coli K-12 by use of stabilized aspA recombinant plasmid. App. Environ. Microbiol. 53: 28002803. Nishimura N., T. Taniguchi and S. Komatsubara. 1989. Hyperproduction of aspartase by a catabolite repression-resistant mutant of Escherichia coli B harboring multicopy aspA and par recombinant plasmid. J. Ferment. Technol. 67: 107110. Papierz M., G. Gadomska, B. Sobierajski and A. Chmiel. 2007: Development of L-aspartic acid biotechnology. Selection and activation of Escherichia coli strains for L-aspartic acid biosynthesis. Pol. J. Microbiol. 56: 7176. Tosa T., T. Sato, T. Mori and I. Chibata.1974. Basic studies for continuous production of L-aspartic acid by immobilized Escherichia coli cells. App. Microbiol. 27: 886889. Virtanen A.J., J. Tarnanen. 1932. Enzymatic hydrolysis and synthesis of aspartic acid. Biochem. Z. 250: 193211. Waller A.S. 2001. Process for the production of L-aspartic acid. United States Patent 6 280 980.


Heterogeneity of galF and gnd of the cps Region for Capsule Synthesis in Clinical Isolates of Klebsiella pneumoniae RAFA£ GIERCZYÑSKI

Department of Bacteriology, National Institute of Hygiene, Warsaw, Poland Received 14 March 2007, revised 5 May 2007, accepted 10 May 2007 Abstract The capsular polysaccharide (CPS) plays important role in Klebsiella spp pathogenesis. Capsular types K1 and K2 of Klebsiella pneumoniae are considered most virulent for humans. The capsule biosynthesis region flanking genes galF and gnd from clinical isolates and reference strains of K. pneumoniae were screened for polymorphism. Nucleotide sequence analysis of galF and gnd revealed a high heterogeneity. However, deduced amino acid sequences demonstrated that the majority of mutations were silent implying GalF and Gnd are strongly conserved. This may suggest importance of these loci in the CPS biosynthesis and may argue for their potential usefulness in Klebsiella genotyping. K e y w o r d s: Klebsiella pneumoniae, capsule, heterogeneity of CPS region

Introduction Klebsiella pneumoniae, an important nosocomial pathogen, causes suppurative infection, pneumonia, urinary tract infection and septicaemia in humans, especially immunosuppressed or suffering from underlying diseases like diabetes mellitus (Podschun and Ulmann, 1998; Fung et al., 2002). This bacterium is also known as an etiological agent of community acquired bacterial pneumonia occurring in chronic alcoholics and people of low social status, which is often fatal if untreated (Podschun and Ulmann, 1998). Generally, clinical isolates of K. pneumoniae produce abundant capsular polysaccharide (CPS). The thickness of capsular layer was shown to be an important factor for high virulence, since poorly encapsulated isolates, however appeared into the alveolar epithelial cells, were avirulent in mouse model of pneumonia (Lai et al., 2003; Astorza et al., 2004). Capsule production, is a general prerequisite for virulence since it protects the bacterium from phagocytosis and killing by serum factors. The genomic organisation of the chromosomal K. pneumoniae cps region responsible for capsule K2 synthesis in strain Chedid was determined by Arakawa et al. (1995). Although this cluster contains 19 open reading frames (orfs), 15 were

found to be indispensable for capsule production. These orfs, numbered from 1 to 15, are transcribed at the same direction from two promoters located upstream of orf1 (galF) and orf3. In addition to the nucleotide sequence of the cps region of Chedid which is deposited in GenBank (http://www.ncbi.nlm. nih.gov) under accession number D21242, corresponding sequences for K1 capsule biosynthesis pathway of K. pneumoniae strains DTS and NTUH-K2044 have been recently added: AY762939 and AB198423 respectively. Moreover, the complete genome sequence of strain MGH78578 K. pneumoniae K52 became available at Genome Sequencing Center at Washington University Medical School (http://genome.wustl.edu). The Southern hybridisation experiments performed by Arakawa et al. (1995) shown that upstream section of the cps region from orf1 to orf3 and downstream section from orf12 to orf15 (gnd) were present in all tested Klebsiella capsular reference strains. The results of PCR screening performed during previous study (Gierczyñski et al., 2005a) and results obtained by Brisse et al. (2004) generally confirmed findings reported by Arakawa et al. (1995), however suggested that the downstream section is likely limited to sole gnd. The galF and gnd genes are present in a homologous cps clusters of K. pneumoniae K1 strains DTS

Corresponding author: R. Gierczyñski, Department of Bacteriology National Institute of Hygiene, 00-791 Chocimska 24, Warsaw, Poland; phone: (48) 22 542 12 44, fax: (48) 22 542 13 07; e-mail: [email protected]

84

2

Gierczyñski R.

and NTUH-K2044 as well in K52 strain MGH78578. In contrast, the inner cps cluster flanked by orf4 and orf14 was shown to reveal genetic heterogeneity that correlated to capsular type of Klebsiella reference strains (Brisse et al., 2004). Studies on genetic diversity of genes flanking the cps region in Klebsiella are rather limited. Most of them were conducted on a limited number of laboratory strains and their mutants. Previously a single strand conformation polymorphism (SSCP) in selected fragments of galF and gnd was tested in 56 clinical strains isolated from infants in six nosocomial outbreaks (Gierczyñski et al., 2005b). Eighteen and ten patterns were distinguished for galF and gnd respectively, that suggested a high nucleotide polymorphism. The main purpose of this study was to determine the nucleotide and amino acid diversity of galF and gnd fragments previously tested by the SSCP and to check whether these genes may be useful for genotyping and phylogenetic analyses of K. pneumoniae. In addition, a comparative phylogenetic analysis of the galF and gnd gene variants found in tested strains and data from GenBank was performed to trace eventual co-evolution. Experimental Materials and Methods

Bacterial isolates and database nucleotide sequences. Twelve strains of K. pneumoniae representing different SSCP haplotypes of galF and gnd described previously (Gierczyñski et al., 2005b) were examined (Table I). Biochemical properties (biogroup) of tested strains were determined as described by Ka³u¿ewski (1965). To eliminate occurrence of delayed results of biochemical tests, development period was reduced to 24 hours. The O-antigen serotyping was performed as described by Ka³u¿ewski (1968) using unencapsulated variants of tested strains. Moreover, database nucleotide sequences of galF and gnd from strains Chedid (Genbank accession no.: D21242), DTS (AY762939) and NTUH-K2044 (AB198423), MGH78578 (http://genome.wustl.edu) and their ho-

Table I Characteristic of tested strains of K. pneumoniae Phenotype

Strain

Sample type (localisation)

O-type

A5054 B5055 30 478/02 234 57 11 145/02 222 5 216 273

Reference O1:K1 Reference O1:K2 Blood (Warszawa) Blood (£ód) Blood (Poznañ) CSF (Warszawa) Lung (Warszawa) Nose (Zamoæ) Stool (Warszawa) Stool (Warszawa) Stool (Warszawa) Stool (Poznañ)

O1 O1 O2 O3 O1 O1 O1 NA O3 O1 O3 NA

Bio- galF galF group 355 366 3 10 13 11 8 8 1 18 14 2 3 11

Primer signature

A2 A A1 A1 C B A C1 D1 D D1 E

A2 A B B1 B3 A C C1 A2 C C2 A1

SSCP genotypes distinguished in previous study (Gierczyñski et al., 2005b), NA not assigned

mologues from Escherichia coli strain E69 K30 (AF503613) were used for phylogenetic analysis. PCR analysis. Genomic DNA templates were prepared as described by Gierczyñski et al. (2004) but the lysosyme treating was omitted. Primers listed in Table II and Taq DNA Polymerase (Fermentas, Lithuania) in (NH4)2SO4 reaction buffer with 1.5 mM MgCl2 supplied by the manufacturer were used for amplification. PCR was performed in 25 ml volume using Mastercycler 5333 (Eppendorf, Germany) as follows: 3 min at 94°C and 35 cycles for 30 seconds of each denaturation at 94°C, annealing at 58°C and elongation at 72oC. DNA synthesis was completed at 72°C for 3 min. DNA sequencing. PCR products were subjected for sequence analysis using automated fluorescent DNA sequencer 377 and BigDye Terminator v3.1. (Applied Biosystems, USA) in accordance to manufacturers instructions. Both DNA strands were sequenced. The determined sequences of tested loci were compared to sequences deposited in GenBank database using BLASTN (Altschul et al., 1997) software utility (National Center for Biotechnology Information, USA)

Nucleotide sequence Forward primer

A1 A B A2 B A A1 B1 B1 A1 B B2

gnd

a

Table II List of PCR primers Locus

SSCP-genotypesa

Reward primer

galF or1355 gctgccgatcgttgataagcc gttatagcgcagcgggtcagc galF or1366 tggtcctgccggatatcatcc actgcttcttcgccagttcgg gnd or15343 gttgaatccctcgagacacc caccgatataggtcacacacg

Amplicon Amplified size region 355 366 343

90 444 389 754 181 523

The amplified regions were counted from the initial nucleotide of the coding sequence of appropriate gene of strain Chedid (D21242). The amplicon size was indicated in base pairs.

2

85

Heterogeneity of galF and gnd of the CPS locus

Evolutionary trees. Phylogenetic and molecular evolutionary analyses were conducted using MEGA software version 2.1. (Kumar et al., 2001). The unweighed pair group method with arithmetic means (UPGMA) was used to generate dendrograms of genetic diversity of tested strains. Neighbour-Joining method with p-distance parameter was applied to generate dendrogram reflecting phylogenetic relatedness of tested strains based upon the deduced amino acid sequences of the analysed genes. Kimura 2 parameter model was applied to both UPGMA and NeighbourJoining methods.

of strain Chedid. Nucleotide substitutions within analysed fragments of galF and gnd from tested strains versus strain Chedid are shown in Table III and IV respectively. When compared to strain Chedid, a number of 61 single nucleotide polymprphisms (SNPs) was detected in tested strains for both analysed genes. It is noteworthy the nucleotide polymorphism in gnd was about four times higher than in galF (Tables III and IV). In this case, 43 SNPs were found in 343 base pairs long DNA fragment, whereas seven and eleven SNPs were detected in galF fragments 355 and 366 respectively. In spite of the high nucleotide sequence diversity of gnd, only two amino acid substitutions were found. Three amino acid substitutions were observed in galF fragment 366, whereas no one was detected in galF fragment 355. The nucleotide sequences of galF and gnd determined in this study for selected representatives of tested and reference strains were compared to the homologous database sequences of strains Chedid, DTS, NTUH-K2044, MGH78578 and E. coli E69 to determine phylogenetic relatedness of tested isolates (Fig. 1). Dendrogram generated by the UPGMA based upon combined sequences of galF and gnd

Results All tested strains yielded PCR-products of selected fragments of galF and gnd. The size of amplicons was in agreement to expected size calculated from the nucleotide sequence of the cps region of strain Chedid. The nucleotide sequencing of galF and gnd fragments previously tested by MultitemperatureSSCP confirmed genetic diversity of the SSCP genotypes (Gierczyñski et al., 2005b). Obtained nucleotide sequences were compared to the reference sequence

Table III The nucleotide sequences of SSCP-genotypes of the galF gene fragment 355 bp (A) and 366 bp (B) compared to the galF sequence of K. pneumoniae strain Chedid A Strain (SSCP) Chedid ( NA) A5054 (A1) B5055 (A) 478/02 (A2) 234 (B) 273 (B2) 222 (B1)

Position of single nucleotide polymorphisms 153

210

213

261

297

345

381

c t

c t t

a c c c c c

g a

g a

c t

g a

B Strain (SSCP) Chedid (NA) A5054 (A2) B5055 (A) 478/02 (A1) 145/02 (C1) 234 (C) 216 (D1) 57 (B) 5 (D)

Position of single nucleotide polymorphisms 504

533

537

551

585

609

618

633

709

717

729

c t t

t Ca

c t

c Ab

c t t t

g a a a

t a c c c c c c c

g a

c Tc

g a a a a a

t c c c a c c c c

Position of a single nucleotide polymorphism (SNP) was indicated as the number of nucleotides counted from the initial nucleotide of galF coding sequence of strain Chedid (D21242). Nucleotides identical to sequence D21242 were indicated by dashes. Amino acids substitutions are shown in capitals. a Alanine instead Valine; b Glutamine instead Proline; c Cysteine instead Arginine; NA not assigned.

86

2

Gierczyñski R. Table IV The nucleotide sequences of SSCP-genotypes of the gnd gene fragment 343 bp compared to the gnd sequence of K. pneumoniae strain Chedid Strain (SSCP) Chedid (NA) A5054 (A2) B5055 (A) 478/02 (B1) 145/02 (C1) 30 (B) 5 (C) Chedid (NA) A5054 (A2) B5055 (A) 478/02 (B1) 145/02 (C1) 30 (B) 5 (C) Chedid (NA) A5054 (A2) B5055 (A) 478/02 (B1) 145/02 (C1) 30 (B) 5 (C)

Position of single nucleotide polymorphisms 204 t c c c c c 324 c t 450 a g g g g

207 t c 339 t c 451 c t t t t

233 c Ta 342 a g g g 453 g a a a a

237 c t 345 c a a g g g 456 a t t t t t t

240 c a 351 c a 459 a t t t t t t

243 c t t 375 c a 462 a g g g g

252 t c c c c c c 381 t a 471 g a

261 g t 387 c t t t t t t 474 g Tb Tb

273 t c c c c c 390 c t t t t t t 477 t c c

276 t c 396 g a a 480 g t t t t

282 c t 399 c g g g g 486 g t t t t t t

300 c t 417 c t t 489 t a a

303 c t 426 t a a 498 c t t t t t t

306 c t 432 c g g t t t

309 c t 435 g a

Position of single nucleotide polymorphism (SNP) was indicated by the number of nucleotides counted from the initial nucleotide of gnd coding sequence of strain Chedid (D21242). Nucleotides identical to sequence D21242 were indicated by dashes. Amino acid substitutions are shown in capitals. a Valine instead Alanine; b Histidine instead Glutamine; NA not assigned.

divided tested strains into two main clusters. However, these clusters appear closely related when compared to strains Chedid and MGH78578. The phylogenetic reconstruction (Fig. 1.B) exhibited that strains MGH78578 and E. coli E69 are significantly distant from the other strains tested. Moreover, two topologically significant branches represented by strain B5055 and Chedid were distinguished. The distant position of strains MGH78578 and E69 was also reflected by dendrogram based upon the deduced amino acid sequences of tested strains (Fig. 1.C). The majority of tested strains together with strain Chedid were placed at the same branch while strains (30 and 57) isolated from fatal cases were grouped closely to strain B5055. Discussion In this study we determined the single nucleotide polymorphism (SNP) of galF and gnd in clinical strains of K. pneumoniae belonging to three O-antigen serogroups. Although a relatively high polymorphism was observed, neither specific genotype nor

cluster of genotypes was found to be exclusive for epidemic strains. Nevertheless, the high nucleotide diversity found in analysed fragments of galF and gnd may suggest that these loci are potentially useful targets for the multilocus sequence typing (MLST). The significant disproportion of the SNPs in galF and gnd observed for clinical and reference strains implies that these loci are exposed to different evolutionary pressure. Nelson and Selander (1994) found that the nucleotide diversity in gnd (orf15) is surprisingly high in both K. pneumoniae and E. coli. In the last species recombination at gnd has occurred with such high frequency that the indicated evolutionary relationships among strains are not congruent with those estimated for other housekeeping genes. This was attributed to a horizontal co-transfer of gnd with adjacent locus rfb mediating O-antigen synthesis, whose activities are subject to diversifying selection because of the host immune response (Nelson and Selander, 1994). Moreover, Rahn et al. (1999) observed exchange of large portions of the cps cluster between various strains of Klebsiella as well between Klebsiella and E. coli K1 group strains. In this light,

2

87

Heterogeneity of galF and gnd of the CPS locus

Figure 1. Results of the phylogenetic analysis of the nucleotide and deduced amino acid sequences of galF and gnd from tested strains and selected database sequences. A genetic diversity shown by UPGMA; B based on nucleotide sequence unrooted phylogenetic tree of tested strains; C based on amino acid sequences phylogenetic tree of tested strains.

application of the sole gnd for the phylogenetic analyses of K. pneumoniae may lead to false conclusions. On the other hand, the amino acid sequences of both GalF and Gnd were strongly conserved in tested strains that together with ubiquitous presence of these genes in K. pneumoniae (Gierczyñski et al., 2005a) may argue for their important role in capsule synthesis. This thesis is supported by results of mutagenesis experiments performed by Arakawa et al. (1995). Insertion of Tn5 in gnd abolished capsular synthesis in strain Chedid, while the lack of galF resulted in unstable CPS production characterised as the mixed capsule phenotype. Taken together results presented herein and conclusions made by Nelson and Selander (1994), the high level of nucleotide diversity of galF and gnd appears to be protected from evolutionary bias by the strong evolutionary pressure. Therefore, these loci may be useful markers for strain distinguishing or subtyping. Interestingly, in spite of the apparent nucleotide diversity the majority of tested strains revealed amino acid sequence of GalF and Gnd typical for strain Chedid. Hence, analysis of both the nucleotide and

amino acid sequences demonstrated that two main branches represented by strains B5055 and Chedid can be distinguished among K. pneumoniae K2 isolates. This finding is in accordance with results of the restriction fragment length polymorphism (RFLP) of the cps inner region obtained by Brisse et al. (2004) who distinguished five RFLP profiles (C-patterns) for K. pneumoniae K2 isolates. In contrast, a single profile was observed for all tested K1 isolates. Acknowledgements This work was supported by grants-in-aid for scientific research (3P05D-00225 and 3P05D-05825) from the Ministry of Science and Higher Education of Poland.

Literature Altschul S.F., T.L. Madden, A.A. Schäffer, J. Zhang, Z. Zhang, W. Miller and D.J. Lipman. 1997. Gapped BLAST and PSIBLAST: a new generation of protein database search programs. Nucleic Acids Res. 25: 33893402. Arakawa Y., R. Wacharotayankun, T. Nagatsuka, H. Ito, N. Kato and M. Ohta. 1995. Genomic organization of the Klebsiella pneumoniae cps region responsible for serotype K2 capsular

88

Gierczyñski R.

polysaccharide synthesis in the virulent strain Chedid. Infect. Immun. 177: 17881796. Astorza B., G. Cortes, C. Crespi, C. Saus, J.M. Rojo and S. Alberti. 2004. C3 promotes clearance of Klebsiella pneumoniae by A549 epithelial cells. Infect. Immun. 72: 17671774. Brisse S., S. Issenhuth-Jeanjean and P.A.D. Grimont. 2004. Molecular serotyping of Klebsiella species isolates by restriction of the amplified capsular antigen gene cluster. J. Clin. Microbiol. 42: 33883398. Fung C.-P., F.-Y. Chang, S.-C. Lee, B.-S. Hu, B. I.-T. Kuo, C.-Y. Liu, M. Ho and L.K. Siu. 2002. A global emerging disease of Klebsiella pneumoniae liver abscess: is serotype K1 an important factor for complicated endophthalmitis? Gut 50: 420424. Gierczyñski R., S. Ka³u¿ewski, A. Rakin, M. Jagielski, A. Zasada, A. Jakubczak, B. Borkowska-Opacka and W. Rastawicki. 2004. Intriguing diversity of Bacillus anthracis in eastern Poland the molecular echoes of the past outbreaks. FEMS Microbiol. Lett. 239: 235240. Gierczyñski R., S. Ka³u¿ewski, A.A. Zasada, W. Rastawicki and M. Jagielski. 2005a. Occurrence of selected loci of the cps region for capsule K1 or K2 synthesis in epidemic strains of Klebsiella pneumoniae isolated from infants in Poland. (in Polish) Med. Dow. Mikrobiol. 57: 5163 Gierczyñski R., S. Ka³u¿ewski, A.A. Zasada, W. Rastawicki and M. Jagielski. 2005b. Single strand conformation polymorphism (SSCP) of selected loci of the cps region for capsule synthesis in

2

epidemic and occasionaly isolated strains of Klebsiella pneumoniae in Poland (in Polish). Med. Dow. Mikrobiol. 57: 153161. Ka³u¿ewski S. 1965. Biochemical properties of Klebsiella pneumoniae applied to strain typing. Exp. Med. Microbiol. 17: 18. Ka³u¿ewski S. 1968. Some partial agents of unencapsulated variants of group O2 Klebsiella. I. Characteristics of strains and antigen O preparations. Exp. Med. Microbiol. 20:1632. Kumar S., K. Tamura, I.B. Jakobsen and M. Nei. 2001. MEGA2: Molecular Evolutionary Genetics Analysis software. Bioinformatics 17: 12441245. Lai, Y.-Ch., H.-L. Peng and H.-Y. Chang. 2003. RmpA2, an activator of capsule biosynthesis in Klebsiella pneumoniae CG43, regulates K2 cps gene expression at the transcriptional level. J. Bacteriol. 185: 788800. Nelson K. and R.K. Selander. 1994. Intergeneric transfer and recombination of the 6-phosphogluconate dehydrogenase gene (gnd) in enteric bacteria. Proc. Natl. Acad. Sci. USA 91: 1022710231. Podschun R. and U. Ulmann. 1998. Klebsiella spp. as nosocomial pathogens: epidemiology, taxonomy, typing methods, and pathogenicity factors. Clin. Microbiol. Rev. 11: 589603. Rahn A., J. Drummelsmith and C. Whitfield. 1999. Conserved organisation of the cps gene clusters for expression of Escherichia coli group 1K antigens: relationship to colanic acid biosynthesis locus and the cps genes from Klebsiella pneumoniae. J. Bacteriol. 181: 23072313.


Lipoarabinomannan as a Regulator of the Monocyte Apoptotic Response to Mycobacterium bovis BCG Danish Strain 1331 Infection MA£GORZATA KRZY¯OWSKA1*, ADA SCHOLLENBERGER1, ANDRZEJ PAW£OWSKI2, BESTON HAMASUR2, ANNA WINNICKA3, EWA AUGUSTYNOWICZ-KOPEÆ 4 and MAREK NIEMIA£TOWSKI1 1 Division

of Immunology, Department of Preclinical Sciences, Faculty of Veterinary Medicine, Warsaw Agricultural University, 2 Department of Bacteriology, Swedish Institute for Infectious Disease Control, 3 Laboratory of Clinical Diagnostics, Department of Clinical Sciences, 4 Microbiology Department, National Tuberculosis and Lung Diseases Research Institute Received 5 December 2006, revised 21 March 2007, accepted 26 March 2007 Abstract

The mannosylated lipoarabinomannan (ManLAM) from mycobacterial species possesses strong immunomodulatory effects. Here we examined the ability of Mycobacterium tuberculosis ManLAM to interfere with the apoptotic response of mouse monocyte cell line, RAW 264.7 infected with Mycobacterium bovis BCG Danish strain. Incubation of BCG-infected monocytes with ManLAM decreased production of NO and the numbers of apoptotic cells which synergized with the polarization of mitochondrial membrane. Activities of caspase-1, -3, -8 and 9 followed pattern of apoptosis suppression by ManLAM, except for caspase-1, which showed no significant change in activity. ManLAM also stabilized anti-apoptotic ratio of bcl-2/bax expression in BCG-infected cells and blocked activation of Fas/FasL-induced pathway of apoptosis. Thus, ManLAM, apart from blocking mitochondrial pathway of apoptosis, may induce several other pathways regulating apoptotic response in BCG-infected mouse monocytes. K e y w o r d s: Mycobacterium bovis BCG, apoptosis, Bcl-2, Fas, lipoarabinomannan

Introduction Tuberculosis (TB) is increasing in prevalence in many countries and is now the leading infectious cause of death worldwide, being responsible for three million deaths annually (Raviglione and Uplekar, 2006; Squire et al., 2006). Such a dramatic situation is due, at least in part, to the ability of the airborne bacillus to resist killing by, and to parasitize host alveolar macrophages (Flynn and Chan, 2001; Flynn and Chan, 2003). The mycobacterial cell wall plays an important role in modulation of immune response to mycobacterial infections by specialised molecules such as lipoarabinomannan (LAM) the predominant antigenic lipoglycan of mycobacterial surface (Strohmeier and Fenton, 1999). LAM is a branched form of phosphatidylinositol mannoside, the characteristic cell wall mannophosphoinositide of mycobacteria. LAM

is expressed in a variety of distinct structures, which are generally grouped into two categories: mannosecapped (ManLAM) and uncapped or arabinofuranosylterminated, LAM (AraLAM). ManLAM is abundant in slow-growing, virulent mycobacteria, whereas AraLAM is abundant in fast-growing, avirulent mycobacteria (such as, for example, M. smegmatis) (Briken et al., 2004; Nigou et al., 2004; Nigou et al., 2003). ManLAMs have been shown to inhibit production of proinflammantory cytokines (IL-12, TNF-") by human dendritic and macrophage cells, in contrast to AraLAM (Nigou et al., 2001; Yoshida and Koide, 1997). Furthermore, ManLAM induces production of the antiinflammatory cytokine IL-10. IL-10 treated DCs are not only less efficient in stimulating T-cell responses but also induce a state of antigen-specific tolerance leading to T cell anergy (Geijtenbeek et al., 2003). Purified ManLAM has been shown to be a potent

* Corresponding author: M. Krzy¿owska, Department of Preclinical Sciences, Faculty of Veterinary Medicine, Warsaw Agricultural University, Ciszewskiego 8, 02-786 Warsaw, Poland; phone/fax: + 48-22-59 36066; e-mail: [email protected]

90

Krzy¿owska M. et al.

chemotactic factor for both CD4+ and CD8+ lymphocytes in vitro and an inhibitor of cytokines production by T-cells (Barnes et al., 1992). It has been shown in vitro that attenuated mycobacterial strains, such as M. bovis BCG or M. tuberculosis H37Ra, induced pronounced apoptosis of human and mouse macrophages while fully virulent strain, such as M. tuberculosis H37Rv, were suppressing apoptosis (Nigou et al., 2002; Rojas et al., 1999). Furthermore, the infection of macrophages with attenuated mycobacterial strains resulted in only limited intracellular growth while virulent strains survived longer and allowed for unlimited multiplication of bacilli (Nigou et al., 2002). These observations suggest that apoptosis of infected host cells might constitute an important innate protective mechanism which limits bacterial load in the tissues early in the course of infection. Although specific bacterial factors responsible for induction/evasion of apoptosis are unknown, several reports suggested that in certain systems a 19-kDa mycobacterial lipoprotein may act as apoptotic inducer (Ciaramella et al., 2000) and ManLAM of mycobacterial cell wall may function as inhibitor of apoptosis of macrophages in vitro (Maiti et al., 2001; Rojas et al., 2000). However, the exact mechanism of LAM action upon macrophages remains to be elucidated. Experimental Materials and Methods

Macrophage cell line. The murinemacrophagelike cell line RAW 264.7 (ATCC TIB-71) was supplied by LGC Promochem. The macrophages were cultured in RPMI-1640 supplemented with 10% foetal bovine serum (FBS, Invitrogen), antibiotics and L-glutamine (Sigma-Aldrich). Following three washes with phosphate-buffered saline (PBS) they were cultured for 24 hours before each experiment. Mycobacterium bovis BCG and ManLAM. Nonpathogenic M. bovis BCG Danish strain 1331 was provided by Statens Serum Institute, Copenhagen, Denmark. The bacteria were grown in Middlebrook 7H9 broth (BD Biosciences) supplemented with OADC supplement (BD Biosciences). After 4 weeks of incubation bacteria were washed by centrifugation and suspended in the growth medium supplemented with 10% glycerol (Sigma-Aldrich) and frozen in 1 ml aliquots in 20°C. Before each experiment bacteria were thawed and dispersed using glass beads until no bacterial clumps were present. The number of bacteria (CFU) per ml was checked by plating on Middlebrook 7H9 agar. ManLAM from highly virulent M. tuberculosis H37Rv was obtained in the Department of Bacteriology, Swedish Institute for Infectious Disease Control using a previously described protocol (Hamasur

2

et al., 1999) complemented with a concanavalin A-sepharose and phenyl-sepharose chromatography (Hamasur, unpublished). This procedure yielded a ManLAM preparation which was free of contaminating protein, migrated as single broad band in SDS-PAGE, showed carbohydrate composition with arabinose: mannose ratio characteristic for ManLAM, and reacted with LAM-specific monoclonal antibodies. Infection assay. RAW 264.7 cells were cultured in 24-well plates at 2× 104 cells per well overnight and the infection experiment was performed by incubation with the bacilli suspension at multiplicity of infection (moi) of 5 in RAW-1640 medium without FBS. After 2 hours, cells were washed three times with PBS and kept in RPMI-1640 supplemented with 10% FBS without antibiotics for 3 days. For each experiment, moi was checked by acid-fast staining using TB colour set (Merck). In all experiments ManLAM was added at 7 µg/ml subsequently after washing of unbounded bacilli. Annexin V-FITC/propidium iodide cell death detection. To assess cell viability, a cell death detection assay was used which detects both apoptosis and necrosis. The BD Biosciences annexin V-FITC apoptosis detection kit was used according to the manufacturers directions. Fluorescent green staining of the plasma membrane indicates apoptosis by binding of annexin V to the outer layer of the plasma membrane. Red staining of DNA with propidium iodide, in conjunction with green annexin V staining, indicates a loss of plasma membrane integrity typical of necrosis. Cells were analysed in FACScan using CellQuest programme (BD Biosciences). Assays for caspase-1, -3, -8 and -9 activities. Caspase-1, -3, -8 and -9 activities were measured according to the producers instruction as described elsewhere (Krzyzowska et al., 2005). Substrates were as follows: Ac-YVAD-AMC [N-acetyl-Tyr-Val-AlaAsp-AMC (7-amino-4- methylcoumarin] substrate for caspase-1; Ac-DEVD-AMC [N-acetyl-Asp-Glu-ValAsp-AMC(7-amino-4-methylcoumarin)] substrate for caspase-3; Ac-IETD-AFC [N-acetyl-Val-Glu-Ile-AspAFC (7-amino-4-trifluormethylcoumarin] substrate for caspase-8 (BD Biosciences) and LEHD-AMC [Nacetylo-Leu-Glu-His-Asp-AMC (7-amino-4-methylcoumarin] for caspase-9 (Sigma). Results for this assay are expressed as relative fluorescence unit/50 µg protein/h (RFU) calculated from triplicate numerical data acquired from test and control samples on Fluoroskan Neonate fluorometer by Transmit Software (Labsystems Oy, Finland). JC-1 staining. Following experimental treatments, cells seeded in a 24-well dish were stained with the cationic dye, 5,5,6,6-tetrachloro1,1,3,3-tetraethylbenzimidazolyl-carbocyanine iodide (JC-1; Sigma) as previously described (Salvioli et al., 1997) in order to determine the state of mitochondrial membrane

2

91

Suppression of monocyte apoptosis by LAM

potential. JC-1 is a potentiometric dye which exhibits a membrane potential dependent loss as J-aggregates (polarized mitochondria) transition to JC-1 monomers (depolarized mitochondria) as indicated by fluorescence emission shift from red to green (Salvioli et al., 1997). Therefore, mitochondrial depolarization is indicated by an increase in the green/red fluorescence intensity ratio. Briefly, cell monolayers were incubated with RPMI-1640 containing 10% FBS and 5 µg/ml JC-1 at 37°C for 15 min. Following the incubation, cells were detached from the wells and washed two times in PBS and analysed in FACScan using CellQuest programme (BD Biosciences). Measurement of nitric oxide. The production of reactive nitrogen radicals (RNI) was determined by measuring the amount of nitrite, a metabolic product of NO. Diluted samples (250 µl) were incubated with 2,3-diaminonaphthalene (DAN, 633 mM in 0.67 N HCl, Sigma-Aldrich) at room temperature in plastic containers in the dark for 10 min (Marzinzig et al., 1997). The mixture was adjusted to pH 11.512.0 with 1 N NaOH. The fluorescence was measured with a microtiter Fluoroskan Neonate fluorometer with an excitation of 365 nm and an emission of 405 nm. The nitrite amount was calculated from a NaNO2 standard curve (Marzinzig et al., 1997). Measurement of Fas, FasL, Bcl-2 and Bax expression. For detection of extracellular Fas and FasL expression, cells were stained with polyclonal rabbit anti-FasL (M-20, Santa Cruz Biotechnology) antibody and mouse monoclonal anti-Fas (clone 13) (BD Biosciences) antibody, followed by FITC-conjugated bovine anti-rabbit polyclonal antibody (Santa Cruz Biotechnology) and goat anti-mouse IgG1 PE-conjugated (BD Biosciences) polyclonal antibodies. For double staining of Bcl-2 and Bax, cells were stained with monoclonal mouse anti-Bcl-2 (clone 7) antibody (BD Biosciences) and polyclonal rabbit anti-mouse bax antibody (BD Biosciences) using BD Cytofix/ Cytoperm kit. In the second step rat FITC-conjugated anti-mouse IgG1 polyclonal antibody (BD Biosciences) and PE-conjugated bovine anti-rabbit polyclonal antibody (Santa Cruz Biotechnology) were used, as described above. Cells were analyzed in FACScan (BD Biosciences) by comparing mean fluorescence for Bax and Bcl-2 and percentage of positive cells for Fas and FasL. Statistics. The non-parametric Mann-Whitney U test was carried out for pair-wise comparison of samples using SPSS software. Results RAW 264.7 cells infected with M. bovis BCG showed a statistically significant increase in NO production throughout the course of infection, whereas

BCG-infected, ManLAM treated-cells also showed an increase in NO production, albeit this increase was statistically lower (p ≤0.05) in 1 and 2 d.p.i. (day post infection) in comparison to NO2 concentration in supernatants from BCG-infected cells (Table I). Table I Nitric oxide production in RAW 264.7 cells treated with 7 µg/ml of ManLAM or infected with M. bovis BCG and treated or not with 7 µg/ml of ManLAM day 1

day 2

day 3 59.59 ± 3.87

control

24.32 ± 2.34

49 ± 4.5

ManLAM

26.01 ± 1.72

43.34 ± 3.1

BCG

35.8 ± 4.65

115.1 ± 9.1

*

BCG + ManLAM 24.45 ± 1.82

52.43 ± 2.78

**

78.24 ± 1.99*

69.85 ± 4.32

71.22 ± 2.67*

*

* p ≤ 0.05, ** p ≤ 0.01 in comparison to control, uninfected cells. Results are expressed as mg NO2/ml. Each value represents mean value obtained from 3 separate experiments ± SEM.

Since high levels of NO production may lead to disruption of mitochondrial functioning, the cells were tested for mitochondrial potential. At 2 d.p.i. RAW 264.7 cells infected with BCG showed statistically significant higher number of cells with depolarised mitochondria (increase in green/red fluorescence ratio from JC-1 stain), thus indicating the decrease in mitochondrial potential. At the same time, BCG-infected, ManLAM treated samples showed increased, but statistically unimportant number of cells with depolarised mitochondria. At day 3 of infection the number of cells with green fluorescence was almost equally high in both BCG-infected and BCG-infected, ManLAM treated cells (p≤0.005, Fig. 1). Furthermore, the mitochondrial potential decrease was well reflected by the number of the apoptotic cells at day 2 the percentage of annexin V-positive cells was significantly lower in BCG-infected, ManLAM-treated cells in comparison to BCG-infected RAW 264.7 cells, while at day 3 no significant difference was observed (results not shown). In order to elucidate the apoptotic pathways during BCG infection and the influence of ManLAM upon the outcome of apoptosis, the activities of caspase-1, -3, -8 and 9 were measured. The activity of caspase-1, classified as a pro-inflammatory caspase and involved in maturation of IL-1, was insignificant throughout the whole tested period (Table II), while the activity of caspase-3 followed apoptosis pattern showed by JC-1 and annexin V staining (Table II). The activity of caspase-3 significantly increased at day 1, 2 and 3 in BCG-infected samples and at day 2 in BCG-infected, ManLAM-treated cells (p≤ 0.05). However, at day 2 and 3, the activity of caspase-3 in BCG-infected, ManLAM-treated cells was significantly

92

2

p≤0.05

p≤ 0.05

▼

BCG+ManLAM

p≤ 0.001

▼

BCG

▼

ManLAM

▼

50

control

▼

60

▼

% of cells with green fluorescence


40 30 20 10 0 day 1

day 2

day 3

Fig. 1. Decrease of mitochondrial potential ()R) in RAW 264.7 cells treated with 7 µg/ml of ManLAM or infected with M. bovis BCG and treated or not with 7 µg/ml of ManLAM. Results are expressed as percentage of cells with decreased )R (depolarised mitochondria), i.e. cells with the increase of green/red fluorescence ratio. Results are expressed as mean from three separate experiments ± SEM.

lower then in BCG-infected cells (p ≤0.05) (Table II). Caspase-3 is an effector caspase, and its activity is dependent upon the activation by up-stream, initiator Table II Caspase 1, 3, 8 and 9 activity assays in protein lysates of RAW 264.7 treated with 7 µg/ml of ManLAM or infected with M.bovis BCG and treated or not with 7 µg/ml of ManLAM Caspase -1 activity Day Control

1 15.53± 0.99

2 34.13± 3.22

3 26.79± 1.99

ManLAM

16.33± 1.89

36.70± 3.01

30.73± 3.03

BCG

12.45± 1.98

32.61± 2.09

26.73± 2.06

BCG+ ManLAM

12.61± 2.02

25.89± 2.14

27.26± 1.7

1 87.85± 7.02

2 240.33± 11.69

3 232.53± 12.4

90.36± 5.01

234.26± 13.04

245.36± 11.8

Caspase-3 activity Day Control ManLAM BCG

103.90± 7.19* 299.6± 15.07** 289.63± 10.2*

BCG+ ManLAM

86.88± 6.22

207.76± 10.34* 227.53± 13.89

Caspase-8 activity Day Control

1 132.55± 11.99

2 391 9.92

3 278.86± 7.78

ManLAM

146.45± 8.1

359.56± 9.12

295.33±9.34

BCG

155.7± 9.38

440.8± 21.22**

324.5± 9.67*

BCG+ManLAM

136.85± 9.78

345.5± 18.97

281.76± 10.23

Caspase-9 activity Day Control

1 2 113.37± 6.11 243.23± 10.94

3 222.53± 12

ManLAM

119.15± 10.56 250.03± 12.66

179.66± 11.1

BCG

119.50± 9.28 289.46± 11.89*

255± 10.11*

BCG+ ManLAM

119.85± 9.93 246.23± 10.23

209.8± 10.12

Each value represents mean value obtained from 3 separate experiments ± SEM. Results are expressed as relative fluorescence units. For details see materials and methods. * indicate p ≤ 0.05, ** p ≤ 0.01 in comparison to control cells.

caspases, such as caspase-9 for the mitochondrial pathway and caspase-8 for the receptor pathway of apoptosis. In our experiments the activity of caspase-8 significantly increased only in cells infected with BCG (p ≤ 0.05), whereas ManLAM significantly influenced its activity in the BCG-infected cells (p ≤0.05) (Table II). Since the mitochondrial potential was decreased in BCG-infected cells at all tested time points and it was also decreased in BCG-infected ManLAM-treated cells, we expected that this decrease should have been reflected by the activity of caspase-9, unless any anti-apoptotic mitochondrial proteins were involved in its suppression. Indeed, the results of caspase-9 activity tests showed that caspase-9 activity was significantly increased only in BCG-infected cells and not in ManLAM treated control or infected cells (p≤ 0.05) (Table II). ManLAM also had some influence upon lowering caspase-9 activity in ManLAM treated control cells, however this decrease was statistically insignificant (p≤0.06) (Table II). The activity of caspase-9 is dependent upon the formation of apoptosome, which in turn depends on the action of specific proteins belonging to Bcl-2 family of mitochondrial proteins. The family of Bcl-2 proteins consists of many proteins exerting both pro-apoptotic (for example Bax) and anti-apoptotic (Bcl-2, among many others) action. Since the mitochondrial potential and caspase-9 activity was decreased in BCG-infected, ManLAM treated cells we decided to test expression of Bax and Bcl-2 proteins. Flow cytometry analysis of bcl-2 and bax expression in BCG-infected RAW 264.7 cells showed a significant increase in both antigens expression at day 1 of infection with BCG (p ≤ 0.05) with no change of expression for other groups (Fig. 2A). However, at day 2 bax expression was significantly increased in BCG-infected cells, exceeding significantly (p≤ 0.001) the level of bcl-2

2

93


mean fluorescence

A

450 400 350 300 250 200 150 100 50 0

Bcl-2 and Bax expression at day 1 bcl-2 bax

control

B mean fluorescence

450 400 350 300 250

ManLAM

BCG

BCG+ManLAM

Bcl-2 and Bax expression at day 2 bcl-2 bax

200 150 100 50 0 kon

ManLAM

BCG

BCG+ManLAM

Fig. 2. Flow cytometry analysis of Bcl-2 and Bax expression in RAW 264.7 treated with 7 µg/ml of ManLAM or infected with M. bovis BCG and treated or not with 7 µg/ml of ManLAM at day 1 (2A) and at day 2 (2B). Results are expressed as mean value of mean fluorescence from three separate experiments ± SEM.

expression (Fig. 2B). Addition of ManLAM had a significant influence upon Bax expression at day 1 and 2 in BCG-infected cells (Fig. 2A, B), decreasing it to the level observed in control and ManLAM-treated cells (p ≤0.05). ManLAM alone caused up-regulation of Bcl-2 expression at day 2 in uninfected RAW 264.7 cells (p ≤0.05) (Fig. 2B). The susceptibility of cells to apoptosis depends, among other factors, on the level of Fas and its natural ligand, FasL expression. Fas receptor is present on the surface of all cells, albeit in low number but increases upon induction of its transcription by many pro-apoptotic factors. FasL, although not normally present on the surface of cells, can also become upregulated by many apoptotic factors. Consequently, since the activity of caspase-8 also depends on the Fas and FasL we expected that the expression of Fas and its ligand should have been elevated in BCG-infected cells. Indeed, Fas expression was significantly increased in BCG-infected cells (p≤ 0.05) at day 1 and 2, but also in ManLAM-treated BCG-infected cells at day 1 and 2 (Fig. 3A). Surprisingly, ManLAM alone induced significant up-regulation of Fas expression in all tested period (p≤ 0.05), even at day 3, when Fas expression was significantly down-regulated in BCG-

infected cells (p≤ 0.05) (Fig. 3A). FasL expression was significantly up-regulated in both BCG-infected as well as ManLAM treated cells (Fig. 3B) at day 1 and 2. However, caspase-8 activity was increased only in BCG-infected cells at day 2 and 3. No significant increase in caspase-8 was observed upon ManLAM treatment (Table II). Discussion The results presented here clearly demonstrate that mouse RAW 264.7 monocytes undergo apoptosis after infection with non-virulent M. bovis BCG, which can be inhibited by the addition of ManLAM isolated from virulent M. tuberculosis H37Rv. Apoptosis was consistently demonstrated by quantification of annexin V-positive cells and disruption of mitochondrial potential. Previous reports from other laboratories, using murine and human macrophages have shown that infection with non-virulent M. tuberculosis leads to apoptosis of infected cells, whereas infection with low doses leads to suppression of apoptosis (Flynn and Chan, 2001; Flynn and Chan, 2003; Nigou et al., 2002). There is also evidence that soluble mycobacterial

94

2


A

FAS expression

% of Fas-positive cells

70 60

control ManLAM BCG BCG+ManLAM

50 40 30 20 10 0 day 1

B

day 2

% of FasL positive cells

FasL expression 100 90 80 70 60 50 40 30 20 10 0

day 3

control ManLAM BCG BCG+ManLAM

day 1

day 2

day 3

Fig. 3. Flow cytometry analysis of Fas and FasL expression in RAW 264.7 treated with 7 µg/ml of ManLAM or infected with M. bovis BCG and treated or not with 7 µg/ml of ManLAM at day 1 (3A) and at day 2 (3B). Results are expressed as mean value of percentage of positive cells from three separate experiments ± SEM.

products are able to modulate apoptosis. PPD induces apoptosis of murine macrophages (Rojas et al., 1997) and a sonicate from M. avium (Rojas et al., 1997), as well as the 19-kDa lipoprotein of M. tuberculosis (Ciaramella et al., 2000), have been reported to induce apoptosis of human monocytes. In addition, there are reports indicating that apoptotic macrophages may be found in bronchoalveolar lavages and pulmonary granulomas from patients with TB (Gil et al., 2004; Keane et al., 1997). On the other hand, M. tuberculosis has developed mechanisms to inhibit apoptosis, thus perpetuating the favorable environment for its intracellular growth. The evidence indicates that ManLAM is largely responsible for the inhibition of apoptosis in M. tuberculosis-infected macrophages (Maiti et al., 2001, Nigou et al., 2001). Many events triggered by ManLAM were proposed: (1) preferential induction of IL-10 production, which negatively regulates the production of NO and caspase activation, even in the presence of TNF-" (Rojas et al., 1999); (2) stabilization of Bcl-2 expression and (3) inhibition of the caspase activation cascade (Rojas et al., 2000). Since Bcl-2 is a mitochondria-associated molecule and it stabilizes mitochondrial potential, the

ratio of Bcl-2 protein to apoptosis-inducing mitochondrial protein Bax decides about the outcome of mitochondrial apoptotic challenge (Cory and Adams, 2005; Cory et al., 2003). Our study showed that ManLAM stabilized Bcl-2 expression in RAW 264.7 cells but it also led to a decrease in Bax expression (Fig. 3). Low expression of mitochondrial pore-forming Bax prevented changes in mitochondrial potential and was followed by inhibition of caspase-9 and caspase-3 activation and consequently, suppression of apoptosis (Fig. 1, 2; Table II). In human neutrophils M. tuberculosis-induced apoptosis is also associated with a transient increase in expression of Bax protein, and a more prominent reduction in expression of the antiapoptotic protein Bcl-xL (Perskvist et al., 2002). Another possible pathway is that down-regulation of Bax expression indirectly results from the increase in NO production. NO at high concentration is toxic to mitochondria and its production leads to changes in expression of mitochondrial pro-apoptotic and antiapoptotic proteins (Choi et al., 2002). ManLAM affects the TNF-"/IL-10 balance by upregulating IL-10, which is considered as an antiapoptotic cytokine, influencing NO production.

2


Oddo et al. (1998) showed that M. tuberculosisinfected human macrophages display a reduced susceptibility to FasL-induced apoptosis, together with reduced levels of surface Fas expression. In our study, however, we did not observed a reduction of Fas expression upon M. bovis BCG infection. Surprisingly, ManLAM induced up-regulation of Fas expression at day 1 and 2 of infection, while at day 3 BCGinfected cells showed a down-regulation of Fas (Fig. 4A). FasL expression was up-regulated by BCGinfection and by ManLAM (Fig. 4B). Ligation of FasL upon Fas is an important step preceding the activation of caspase-8, so up-regulation of Fas renders cells more susceptible to apoptosis. This was not observed in our study in BCG-infected, ManLAM treated cells, which indicates that ManLAM, apart from blocking mitochondrial pathway of apoptosis, may induce another pathway blocking death receptor pathway of apoptosis. All these observations suggest that M. tuberculosis has evolved double-edge sword mechanisms influencing macrophage survival and death. The elucidation of the mechanisms that govern macrophage cell death during M. tuberculosis infection will open up new possibilities for understanding host-mycobacteria interactions and manipulating host immune and inflammatory responses. Acknowledgements This work was supported by the State Committee for Scientific Research (grant no 4 P05A 114 18)

Literature Barnes, P.F., D. Chatterjee, J.S. Abrams, S. Lu, E. Wang, M. Yamamura, P.J. Brennan and R.L. Modlin. 1992. Cytokine production induced by Mycobacterium tuberculosis lipoarabinomannan. Relationship to chemical structure. J. Immunol. 149: 541547. Briken, V., S.A. Porcelli, G.S. Besra and L. Kremer. 2004. Mycobacterial lipoarabinomannan and related lipoglycans: from biogenesis to modulation of the immune response. Mol. Microbiol. 53: 391403. Choi, B.M., H.O. Pae, S.I. Jang, Y.M. Kim and H.T. Chung. 2002. Nitric oxide as a pro-apoptotic as well as anti-apoptotic modulator. J. Biochem. Mol. Biol. 35: 116126. Ciaramella, A., A. Martino, R. Cicconi, V. Colizzi and M. Fraziano. 2000. Mycobacterial 19-kDa lipoprotein mediates Mycobacterium tuberculosis-induced apoptosis in monocytes/ macrophages at early stages of infection. Cell Death Differ. 7: 12701272. Cory, S. and J.M. Adams. 2005. Killing cancer cells by flipping the Bcl-2/Bax switch. Cancer Cell 8: 56. Cory, S., D.C. Huang and J.M. Adams. 2003. The Bcl-2 family: roles in cell survival and oncogenesis. Oncogene 22: 85908607. Flynn, J. L. and J. Chan. 2001. Immunology of tuberculosis. Annu. Rev. Immunol. 19: 93129. Flynn, J.L. and J. Chan. 2003. Immune evasion by Mycobacterium tuberculosis: living with the enemy. Curr. Opin. Immunol. 15: 450455.

95

Geijtenbeek, T.B., S.J. Van Vliet, E.A. Koppel, M. SanchezHernandez, C.M. Vandenbroucke-Grauls, B. Appelmelk and Y. Van Kooyk. 2003. Mycobacteria target DC-SIGN to suppress dendritic cell function. J. Exp. Med. 197: 717. Gil, D. P., L.G. Leon, L.I. Correa, J.R. Maya, S.C. Paris, L.F. Garcia and M. Rojas. 2004. Differential induction of apoptosis and necrosis in monocytes from patients with tuberculosis and healthy control subjects. J. Infect. Dis. 189: 21202128. Hamasur, B., G. Kallenius and S.B. Svenson. 1999. A new rapid and simple method for large-scale purification of mycobacterial lipoarabinomannan. FEMS Immunol. Med. Microbiol. 24: 1117 Keane, J., M.K. Balcewicz-Sablinska, H.G. Remold, G.L. Chupp, B.B. Meek, M.J. Fenton and H. Kornfeld. 1997. Infection by Mycobacterium tuberculosis promotes human alveolar macrophage apoptosis. Infect. Immun. 65: 298304. Krzyzowska, M., M. Polanczyk, M. Bas, J. Cymerys, A. Schollenberger, F. Chiodi and M. Niemialtowski. 2005. Mousepox conjunctivitis: the role of Fas/FasL-mediated apoptosis of epithelial cells in virus dissemination. J. Gen. Virol. 86: 20072018. Maiti, D., A. Bhattacharyya and J. Basu. 2001. Lipoarabinomannan from Mycobacterium tuberculosis promotes macrophage survival by phosphorylating Bad through a phosphatidylinositol 3-kinase/Akt pathway. J. Biol. Chem. 276: 329333. Marzinzig, M., A.K. Nussler, J. Stadler, E. Marzinzig, W. Barthlen, N.C. Nussler, H.G. Beger, S.M. Morris and U.B. Bruckner. 1997. Improved methods to measure end products of nitric oxide in biological fluids: nitrite, nitrate, and S-nitrosothiols. Nitric Oxide 1: 177189. Nigou, J., M. Gilleron, T. Brando and G. Puzo. 2004. Structural analysis of mycobacterial lipoglycans. Appl. Biochem. Biotechnol. 118: 253267. Nigou, J., M. Gilleron and G. Puzo. 2003. Lipoarabinomannans: from structure to biosynthesis. Biochimie 85: 153166. Nigou, J., M. Gilleron, M. Rojas, L.F. Garcia, M. Thurnher and G. Puzo. 2002. Mycobacterial lipoarabinomannans: modulators of dendritic cell function and the apoptotic response. Microbes Infect. 4: 945953. Nigou, J., C. Zelle-Rieser, M. Gilleron, M. Thurnher and G. Puzo. 2001. Mannosylated lipoarabinomannans inhibit IL-12 production by human dendritic cells: evidence for a negative signal delivered through the mannose receptor. J. Immunol. 166: 74777485. Oddo, M., T. Renno, A. Attinger, T. Bakker, H.R. MacDonald and P.R.A. Meylan. 1998. Fas Ligand-induced apoptosis of infected human macrophages reduces the viability of intracellular Mycobacterium tuberculosis. J. Immunol. 160: 54485454. Perskvist, N., M. Long, O. Stendahl and L. Zheng. 2002. Mycobacterium tuberculosis promotes apoptosis in human neutrophils by activating caspase-3 and altering expression of Bax/ Bcl-xL via an oxygen-dependent pathway. J. Immunol. 168: 63586365. Raviglione, M.C. and M.W. Uplekar. 2006. WHOs new Stop TB Strategy. Lancet 367: 952955. Rojas, M., L.F. Barrera, G. Puzo and L.F. Garcia. 1997. Differential induction of apoptosis by virulent Mycobacterium tuberculosis in resistant and susceptible murine macrophages: role of nitric oxide and mycobacterial products. J. Immunol. 159: 13521361. Rojas, M., L.F. Garcia, J. Nigou, G. Puzo and M. Olivier. 2000. Mannosylated lipoarabinomannan antagonizes Mycobacterium tuberculosis-induced macrophage apoptosis by altering Ca+2-dependent cell signaling. J. Infect. Dis. 182: 240251. Rojas, M., M. Olivier, P. Gros, L.F. Barrera and L.F. Garcia. 1999. TNF-alpha and IL-10 modulate the induction of apoptosis by virulent Mycobacterium tuberculosis in murine macrophages. J. Immunol. 162: 612261231.

96


Salvioli, S., A. Ardizzoni, C. Franceschi and A. Cossarizza. 1997. JC-1, but not DiOC6(3) or rhodamine 123, is a reliable fluorescent probe to assess delta psi changes in intact cells: implications for studies on mitochondrial functionality during apoptosis. FEBS Lett. 411: 7782. Squire, S.B., A. Obasi and B. Nhlema-Simwaka. 2006. The Global Plan to Stop TB: a unique opportunity to address poverty and the Millennium Development Goals. Lancet 367: 955957.

2

Strohmeier, G.R. and M.J. Fenton. 1999. Roles of lipoarabinomannan in the pathogenesis of tuberculosis. Microbes Infect. 1: 709717. Yoshida, A. and Y. Koide. 1997. Arabinofuranosyl-terminated and mannosylated lipoarabinomannans from Mycobacterium tuberculosis induce different levels of interleukin-12 expression in murine macrophages. Infect. Immun. 65: 19531955.


Staphylokinase Production by Clinical Staphylococcus aureus Strains MARZENA WIÊCKOWSKA-SZAKIEL, BEATA SADOWSKA* and BARBARA RÓ¯ALSKA

Department of Immunology and Infectious Biology, Institute of Microbiology and Immunology, University of £ód, £ód, Poland Received 17 November 2006, revised 15 February 2007, accepted 4 April 2007 Abstract One of virulence factors produced by Staphylococcus aureus is staphylokinase (SAK), which enhances their proteolytic activity leading to tissue damage and improving bacterial invasiveness. In the present study we estimated the ability to produce staphylokinase by 95 S. aureus reference strains and clinical isolates from the airways of cystic fibrosis patients, from skin lesions and from infected bones. We would like to verify any relationship between SAK production and the types of clinical isolates as well as other biochemical properties and activities of these staphylococcal strains, which can be important for their pathogenicity. More than 62% of all tested strains were able to produce secreted type of SAK. Staphylokinase production was significantly more common in the isolates from skin and soft tissue infections than in any other group of tested staphylococci. The general tendencies in the selected properties or activities of both SAK() and SAK(+) isolates were similar. Our data confirm phenotypic dissimilarity in SAK production of S. aureus strains isolated from various types of infections. It is compatible with the biological role of staphylokinase and with hypothetical model of staphylokinase mediated bacterial invasion of host tissues. Thus, the estimation of SAK production by S. aureus isolates may be regarded as the parameter describing potential invasiveness of staphylococci and can be useful as a medical recommendation for the eradication of staphylococci carrier state. K e y w o r d s: Staphylococcus aureus clinical strains, staphylokinase (SAK), virulence

Introduction Staphylococcus aureus is one of the common human pathogens. It permanently colonizes the epithelium of 20% of the population, transiently occurs in more than 60% (Foster, 2005). Thus S. aureus has a simple access to the host organisms and can occasionally cause both acute and chronic infections. These bacteria are responsible for a wide range of illnesses: from skin and soft tissue lesions like ulcers and furuncles, through food poisoning, to life threatening infections such as bacteremia followed by arthritis, osteomyelitis or endocarditis and septic shock (Foster, 2005; Krut et al., 2003). There is no doubt that invasive success of S. aureus is associated with the broad spectrum of its virulence factors. This microorganism expresses a lot of secreted and cell-surface polysacchrides and proteins inclusive of many toxins and enzymes, which promote first the bacterial colonization, then the damage of host tissue, spreading of bacteria through organism, immune evasion and finally

lead to fully-symptomatic disease (Cheung et al., 2004; Foster, 2005; Heyer et al., 2002; Otto et al., 1999). One of these virulence factors is staphylokinase (SAK), also called fibrynolysin. SAK is 136-amino acid extracellular protein produced during the late exponential growth phase by S. aureus strains carrying the prophages which contain the sak gene (Bokarewa et al., 2006; Jin et al., 2004; Lähteenmäki et al., 2001; Rooijakkers et al., 2005). This protein is one of four human specific immune innate modulators, including chemotaxis inhibitory protein of S. aureus (CHIPS), staphylococcal complement inhibitor (SCIN) and enterotoxin A, which form a special cluster (called innate immune evasion cluster IEC) on the conserved 3 end of $-hemolysin converting bacteriophages (van Wamel et al., 2006). Staphylokinase, similarly to streptokinase secreted by some $-hemolytic group of streptococci or Pla surface protease produced by Yersinia pestis, belongs to a group of bacterial plasminogen (PLG) activators (Lähteenmäki et al., 2001; Rooijakkers et al., 2005). PLG is the precursor of

* Corresponding author: B. Sadowska, Department of Immunology and Infectious Biology, Institute of Microbiology and Immunology, University of £ód, Banacha 12/16, 90-237 £ód, Poland; phone: (48) 42 635 45 25; e-mail: [email protected]

98

Wiêckowska-Szakiel M. et al.

fibrinolytic protease: plasmin an enzyme that degrades proteins of the extracellular matrix. SAK does not have an enzymatic activity itself but forms a 1:1 stoichiometric complexes with PLG, which convert other plasminogen molecules to plasmin (Jin et al., 2004; Lähteenmäki et al., 2001; Rooijakkers et al., 2005). Thus staphylokinase enhances proteolytic activity of S. aureus strains against extracellular matrix proteins (ECM). Taking into consideration the fact that such ECM as collagen, laminin, fibronectin or elastin are the main components of tissue barriers and basement membranes, SAK can be regarded as a very important staphylococcal virulence factor, which leads to tissue damage and improves bacterial invasiveness (Bokarewa et al., 2006; Lähteenmäki et al., 2001). In the present study we estimated the ability to produce soluble form of staphylokinase by 95 S. aureus strains selected from our collection. Most of these strains were clinical isolates representing different kinds of staphylococcal infections: from classic skin lesions like abscesses, ulcers or furuncles, through deep, difficult to treat infections of the bones, to specific, very often mixed airways infections of cystic fibrosis patients. Considering staphylokinase as a very important virulence factor of staphylococci, we would like to verify any relationship between SAK production and the types of clinical isolates. In our search we also included the group of reference S. aureus strains as a specific control group. Phenotypic features of these strains, along with their capability of SAK production, could not be modified by the contact with both host organism and other bacteria. The gene coding SAK together with the genes for some other virulence factors form a special cluster IEC. In the light of this information and the fact, that staphylococci possess global regulatory systems of the genes (e.g. agr, sar), the dependence between different, apparently not connected features of these bacteria can be expected. Therefore, we decided to collate our knowledge about some characters and activities of the tested S. aureus strains, which can be important for their pathogenicity, with SAK production by these staphylococci. To this part of our search we chose the group of isolates from cystic fibrosis patients as the most representative, because of their numbers and the percentage of SAK(+) strains similar to those observed for natural and other clinical S. aureus populations. Experimental Materials and Methods

Characterization of S. aureus strains. Four groups of S. aureus strains were set up as: (I) laboratory reference strains of staphylococci (9 strains: Cowan 1

2

overproducer of protein A, Reynolds capsule serotype 5 prototype and Becker capsule serotype 8 prototype, Wood 46 overproducer of "-hemolysin, 8325-4 S. aureus used to genetic manipulations, DU1090 "-hemolysin negative mutant, MRSA 478 MRSA class I prototype, MRSA 479 MRSA class II prototype and MRSA 477 MRSA class III prototype), (II) S. aureus isolated from airways of cystic fibrosis patients (59 strains) [from the Mother and Child Institute of Warsaw and from the Institute for Tuberculosis and Pulmonary Diseases, Rabka, Poland], (III) clinical strains isolated from skin lesions like abscesses, ulcers or furuncles (12 strains) [from the Clinic of Dermatology, Health Care Groupe, £ód, Poland], (IV) clinical strains isolated from infected bones (15 strains) [from M. Copernicus Hospital, £ód, Poland]. All strains were subcultured on sheep blood agar to check their macro- and microscopic (by Gram staining) morphology and their hemolytic activity. Then the isolates were identified as S. aureus using conventional biochemical tests: detection of catalase (slide test with H2O2) and coagulase (tube test with rabbit plasma), decomposition of glucose and mannitol under aerobic and anaerobic conditions (tube test on Hugh-Leifson Medium), novobiocin susceptibility (disc diffusion test on Müller-Hinton Agar) and using latex agglutination assay, which detects clumping factor (CF). Strains were kept frozen ( 80°C) in Triptic Soy Broth (TSB; Difco, USA) with 15% of glycerol. Some special features and activities of S. aureus strains isolated from the airways of cystic fibrosis patients were investigated in our previous studies (Sadowska et al., 2000; Sadowska et al., 2002). The type of capsular polysaccharide (CP) was estimated using monoclonal antibody against CP5 and CP8. Screening technique on Müeller-Hinton Agar with 6 µg/ml oxacillin and 4% NaCl was used to observe the profile of the resistance to methicillin. Also the ability to form small colony variant (SCV) was estimated after the passage of S. aureus strains in tryptic soy broth with 1 µg/ml of gentamicin. Reagents. Human Glu-plasminogen was obtained from American Diagnostica (USA). This reagent was prepared from fresh human citrated plasma by lysineSepharose affinity chromatography in the presence of aprotinin. Recombinant SAK (rSAK) was purchased from ProSpec-Tany TechnoGene LTD (Israel) and the substrate for plasmin H-D-Val-Leu-Lys-pNA× 2HCl (S-2251) from Chromogenix (Italy). Todd-Hewitt Broth (THB), Todd-Hewitt Agar (THA) and sheep blood agar were obtained from BTL (Poland). Culturing of S. aureus strains for SAK production. S. aureus strains were cultured on THA for 24 h at 37°C. One colony of each stahylococcal isolate was transferred into 2 ml of THB and incubated for 18 h at 37°C. The cultures were centrifuged (2600×g, 10 min,

2

99

Staphylokinase production by clinical S.aureus strains

4°C) and the supernatants were collected. The supernatants (in duplicate) were tested for activity of soluble SAK and the selected bacteria for activity of surfacebound SAK. SAK activity measurement in staphylococcal supernatants. Staphylokinase activity was determined by measuring plasmins substrate hydrolysis in the presence of 1 µM plasminogen in Tris/HCl buffer (0.14 M NaCl, 1.5 M Tris/HCl, pH 7.2). Glu-plasminogen was incubated with culture supernatants for 1 h at 37°C to allow the conversion of plasminogen to plasmin. Plasmin formation was evaluated by hydrolysis of 4 mM chromogenic substrate S-2251 for 30 min at 37°C. The standard curve was performed using tenfold dilutions of rSAK (range: 0.0785 µg/ml) preincubated with 1 µM plasminogen in Tris/HCl buffer. The absorbance reading for soluble form of SAK in bacterial supernatants and the standard curve of rSAK was measured at 405 nm on multifunction reader Victor 2 (Wallac, Finland). Mean absorbance values were converted on SAK concentration on the basis of the equation of trend line for the standard curve. Determination of surface-bound SAK activity. The suspensions of staphylococci were prepared in 0.85% NaCl to a density equivalent to the McFarland turbidity standard of 3.0 by Densi-La-Meter (LaChema). Next, the bacteria were centrifuged (2600× g, 10 min, 4°C), resuspended in THB and incubated for 4 h at 37°C with 1 µM Glu-plasminogen. The excess of plasminogen was removed by washing twice with 1 ml PBS. Finally, the bacteria were resuspended in Tris/ HCl buffer (0.14 M NaCl, 1.5 M Tris/HCl, pH 7.2), transferred into 96-well plate and incubated with 4 mM substrate S-2251 for approximately 18 h at 37°C. Positive and negative controls were performed using S. aureus Cowan1 (SAK+) and S. aureus strain notproducing SAK, respectively. Both were prepared as 109 cfu/ml suspensions in Tris/HCl buffer with 4 mM plasmin substrate. The mean absorbance reading for surface-bound SAK of staphylococci was registered spectrophotometrically at 405 nm and then compared with the absorbance values obtained for the controls. Statistical analysis. Chi-square test with Yates correction, Fisher test or V-square test were used to compare SAK production between all four groups of tested staphylococci and to correlate the ability to produce SAK with other properties of strains isolated from patients with cystic fibrosis. A P< 0.05 was considered significant. Results We examined the production of staphylokinase by 95 S. aureus strains divided into four groups: (I) laboratory reference strains of staphylococci (9 strains),

(II) S. aureus isolated from airways of cystic fibrosis patients (59 strains), (III) clinical strains isolated from skin lesions (12 strains), (IV) clinical strains isolated from bones infections (15 strains). Post-culture supernatant samples with SAK level below 0.3 µg/ml were considered as negative (). SAK levels ranging from 0.3 to 2.5 µg/ml were assessed as low production and all samples with SAK level above 2.5 µg/ml as high production. The obtained results are presented in Table I as the percentage of S. aureus strains possessing or not the ability to produce soluble form of staphylokinase. Table I The percentage of S. aureus strains producing or not producing soluble form of staphylokinase (SAK) Group Type of No. S. aureus strains I II III IV

lab reference (9 strains) cystic fibrosis (59 strains) skin lesions (12 strains) infected bones (15 strains)

SAK production (µg/ml) negative low high (below 0.3) (0.32.5) (above 2.5) 33%

22%

45%

41%

42%

17%

8%

67%

25%

53%

14%

33%

More than 62% of all tested strains were able to produce soluble form of SAK: 67% strains from group I; 59% strains from group II; 92% strains from group III and 47% strains from group IV. It is noteworthy, that staphylokinase production was more common in isolates from skin and soft tissue infections (group III) than in any other group of S. aureus strains. However, statistically significant differences only between group III and IV were observed (P = 0.04). More than 22% of all SAK(+) strains secreted high amounts of soluble staphylokinase. Almost 38% of all strains were classified as SAK non-producers, which was verified by a test performed for the detection of surface-bound staphylokinase. A few strains (representatives of all four groups) described as SAK() for a soluble form of this enzyme and S. aureus Cowan 1 as a positive control of soluble SAK producers were tested on cell-associated SAK. All examined strains, which did not secrete of soluble SAK, also did not possess surface-attached form of this enzyme. The search for the correlation of soluble SAK production and some other properties of tested bacterial strains was performed for the isolates from airways of cystic fibrosis patients the most numerous group (59 strains). The percentage of SAK(+) strains from this group was similar as for natural and other clinical

100


Table II The soluble form of SAK production versus other properties and activities of S. aureus strains isolated from airways of cystic fibrosis patients Production of SAK released into the fluid phase

Properties/activities

negative (24 strains)

Type of capsule: CP 5 CP 8 lack or other Profile of resistance to methicillin (MRSA): MSSA (susceptibility) MRSA class I MRSA class II MRSA class III Ability to SCV formation

positive (35 strains)

4 (17%) 12 (50%) 8 (33%)

3 (9%) 15 (43%) 17 (49%)

10 (42%) 10 (42%) 1 (4%) 3 (13%) 6 (25%)

19 (54%) 7 (20%) 0 (0%) 9 (26%) 7 (20%)

populations of staphylococci about 6070% (Bokarewa et al., 2006; Jin et al., 2003; van Wamel et al., 2006). For this reason the isolates from airways of cystic fibrosis patients seem to be the most representative group for this kind of search. First of all we noticed, that hemolytic activity of both SAK() and SAK(+) staphylococci was similar (strong for most strains: 83% and 86% of SAK negative and positive strains, respectively). Little differences in the capability of anaerobic decomposition of mannitol between SAK() and SAK(+) S. aureus strains were observed (respectively 67% and 80% of strains were capable of mannitol fermentation), but the differences were not statistically significant (P = 0.252). Other correlated data are presented in Table II as the number and the percentage of S. aureus isolates possessing or not definite features or activity. Single results received for both SAK() and SAK(+) cystic fibrosis isolates differed insignificantly (P> 0.05) and general tendencies in their properties or activities were similar. Independently of the capability of SAK production, S. aureus strains possessed capsular polysaccharide type 8 (CP 8) more often than type 5 (CP 5), more than 40% of isolates were susceptible to methicillin. Also similar percentage of SAK() and SAK(+) strains SCV under gentamicin pressure was recovered. Discussion The pathogenicity of S. aureus is a complex process involving simultaneously many cell wall components and extracellular products and is very difficult to indicate the importance of their single virulence factor. The observations of the effects of SAK production by S. aureus and the conclusions drawn

2

sometimes seem to be contradictory. SAK-deficient S. aureus isolates happened to be described as more dangerous causing the lethal bacteremia more frequently than staphylococci producing SAK. Moreover, the production of staphylokinase by nasal isolates as one of the adaptive mechanisms of S. aureus symbiosis with the host was suggested (Bokarewa et al., 2006; Jin et al., 2003). On the other hand, it was proved that staphylococcal strains producing SAK were protected against the bactericidal effect of human "-defensins (HNP-1, HNP-2) and against opsonization by both immunoglobulin G and C3b/ C3bi, which could promote the invasion of host tissues by these strains (Jin et al., 2004; Rooijakkers et al., 2005; van Wamel et al., 2006). Such various roles of SAK seem to be dependent on the stage of infection or current needs of staphylococci. It can be presumed, that at the beginning of infection SAK production should be inhibited to prevent the proteolysis of ECM being very important for bacterial adhesion. Then, during the invasion of host tissue, the expression of SAK should be increased allowing degradation of the junctions between host cells or destruction of basement membranes. Based on these considerations, we decided to estimate the ability to the secretion of staphylokinase by 95 S. aureus strains and find the relationships between its production and the types of infection or other biochemical properties and activities of these strains. It was demonstrated that more than 62% of all our tested S. aureus strains were able to produce and release staphylokinase. Van Wamel et al. (2006) also discovered the different IEC types containing sak gene in 76.6% of clinical isolates of staphylococci. The carrying of such genetic mobile elements coding the virulence factors being able to affect human innate immune system (e.g. SEA modulates the functions of chemokine receptors, SAK and SCIN possess anti-opsonic capacity, CHIPS blocks chemotaxis) is very profitable for bacteria (Jarraud et al., 2002; van Wamel et al., 2006). Thereby, the selective distribution of some genes, for instance these coding the superantigens, among S. aureus clinical strains was also described by Ferry et al. (2005), Omoe et al. (2005) and van Belkum et al. (2006). This phenomenon has probably developed during evolutional adaptation of bacteria to the specific micro-environmental conditions appearing in vivo in different kind of infections or even during their stages. On the other hand, taking to consideration our results for control group of staphylococci group I (almost 70% of these strains were SAK+), the ability to staphylokinase production seems to be profitable for these bacteria, even if they dont have contact with host organism and this feature can not be created by special environmental conditions.

2

Staphylokinase production by clinical S.aureus strains

Distribution of SAK production was also described by Jin et al. (2003), whose observations were similar to our results. We noticed significantly more frequent staphylokinase production in isolates from skin and soft tissue infections (group III) than in any other group of tested S. aureus strains. They observed that SAK positive strains were less common (1.7 times) among the isolates from patients with lethal bacteremia than among nasal carriage isolates. It confirms the earlier observations of unexpected lack of SAK production in staphylococci invading internal organs in comparison with these colonizing mucosal tissue and registered SAK production in almost all staphylococcal isolates obtained from skin and mucosa (Bokarewa et al., 2006; Jin et al., 2003). Our results are compatible with one of the models of staphylokinase mediated bacterial invasion. It is suggested, that SAK-PLG complexes may help staphylococci cleave the infectious focus or abscess from the fibrin net, thus enabling these bacteria to enter into the deeper host tissue (Bokarewa et al., 2006). This is a good explanation for our observations, that staphylokinase was produced mainly by strains isolated from skin lesions like abscesses, ulcers or furuncles. In the light of the information about special cluster IEC coding sak gene together with the genes for some other virulence factors and of the fact, that staphylococci possess global regulatory systems of the genes (e.g. agr, sar), the dependence between different features of these bacteria can be expected. Therefore we decided to correlate some other properties and activities of tested S. aureus strains with their ability to produce SAK. In this part of our research, we chose the group of isolates from cystic fibrosis patients as the most representative, because of their number and the percentage of SAK(+) strains similar to those observed for natural and other clinical S. aureus populations. During the collection of the strains and the preparation of their stocks we checked the hemolytic activity of these bacteria. It was proved that the ability of SAK(+) and SAK() strains to produce hemolysin was similar. This observation is interesting with regard to the known effect of inactivation of $-hemolysin gene by the insertion of staphylokinase-carrying bacteriophage to the bacterial genome (Bokarewa et al., 2006; Jin et al., 2003; Lähteenmäki et al., 2001). Although that "-hemolysin is mainly responsible for hemolytic activity of S. aureus strains. In our previous studies (Sadowska et al., 2000; Sadowska et al., 2002) we also estimated some special features and the activity of these S. aureus strains from group II. Now, we noticed that the production of soluble SAK did not correlate with such staphylococci features as type of polysaccharide capsule (P = 0.4292), the profile of resistance to methicillin (P = 0.3409) or small

101

colony variant (SCV) formation under antibiotic pressure (P = 0.6518). Therefore SAK production seems not to have any importance for the strains invading the lungs in cystic fibrosis patients. In conclusion, our data demonstrate that phenotypic differences in secreted SAK production exist among S. aureus strains isolated from various kinds of infections. It is compatible with the biological role of staphylokinase and the theoretical model of staphylokinase mediated bacterial invasion of host tissues. Thus, the simple laboratory method for the estimation of SAK production by S. aureus isolates (e.g. strains isolated from the carriers) may be accepted as the parameter describing potential invasiveness of staphylococci. Such knowledge can be useful as a medical recommendation to eradication of staphylococci carrier state in particular cases. Literature Bokarewa M.I., T. Jin and A. Tarkowski. 2006. Staphylococcus aureus: staphylokinase. IJBCB 38: 504509. Cheung A.L., A.S. Bayer, G. Zhang, H. Gresham and Y-Q. Xiong. 2004. Regulation of virulence determinants in vitro and in vivo in Staphylococcus aureus. FEMS Immunol. Med. Microbiol. 40: 19. Ferry T., D. Thomas, A.-L. Genestier, M. Bes, G. Lina, F. Vandenesch and J. Etienne. 2005. Comparative prevalence of superantigen genes in Staphylococcus aureus isolates causing sepsis with and without septic shock. Clin. Infect. Dis. 41: 771777. Foster T.J. 2005. Immune evasion by staphylococci. Nature Rev. 3: 948957. Heyer G., S. Saba, R. Adamo, W. Rush, G. Soong, A. Cheung and A. Prince. 2002. Staphylococcus aureus agr and sarA functions are required for invasive infection but not inflammatory responses in the lung. Infect. Immun.70: 127133. Jarraud S., C. Mougel, J. Thioulouse, G. Lina, H. Meugnier, F. Forey, X. Nesme, J. Etienne and F. Vandenesch. 2002. Relationships between Staphylococcus aureus genetic background, virulence factors, agr groups (alleles), and human disease. Infect. Immun. 70: 631641. Jin T., M. Bokarewa, L. McIntyre, A. Tarkowski, G.R. Corey, L.B. Reller and V.G. Fowler Jr. 2003. Fatal outcome of bacteraemic patients caused by infection with staphylokinasedeficient Staphylococcus aureus strains. J. Med. Microbiol. 52: 919923. Jin T., M. Bokarewa, T. Foster, J. Mitchell, J. Higgins and A. Tarkowski. 2004. Staphylococcus aureus resists human defensins by production of staphylokinase, a novel bacterial evasion mechanisms. J. Immunol. 172: 11691176. Krut O., O. Utermöhlen, X. Schlossherr and M. Krönke. 2003. Strain-specific association of cytotoxic activity and virulence of clinical Staphylococcus aureus isolates. Infect. Immun. 71: 27162723. Lähteenmäki K., P. Kuusela and T.K. Korhonen. 2001. Bacterial plasminogen activators and receptors. FEMS Microbiol. Rev. 25: 531552. Omoe K, D.-L. Hu, H. Takahashi-Omoe, A. Nakane and K. Shinagawa. 2005. Comprehensive analysis of classical and newly described staphylococcal superantigenic toxin genes in Staphylococcus aureus isolates. FEMS Microbiol. Lett. 246: 191198.

102


Otto M., R. Süßmuth, C. Vuong, G. Jung and F. Götz. 1999. Inhibition of virulence factor expression in Staphylococcus aureus by the Staphylococcus epidermidis agr pheromone and derivatives. FEBS Letters 450: 257262. Rooijakkers S.H.M., W.J.B. van Wamel, M. Ruyken, K.P.M. van Kessel and J.A.G. van Strijp. 2005. Anti-opsonic properties of staphylokinase. Microbes Infect. 7: 476484. Sadowska B., A. Bonar, M. Rzeniczak, I. Solarska, W. Rudnicka and B. Ró¿alska. 2000. Comparative phenotypic characteristics of Staphylococcus aureus isolated from cystic fibrosis patients versus blood and skin-mucosal infections isolates. Bull. Pol. Acad. Sci. (Biol. Ser.) 48: 99109. Sadowska B., A. Bonar, Ch. von Eiff, R.A. Proctor, M. Chmiela, W. Rudnicka and B. Ró¿alska. 2002. Characteristics

2

of Staphylococcus aureus, isolated from airways of cystic fibrosis patients, and their small colony variants. FEMS Immunol. Med. Microbiol. 32: 191197. Van Belkum A., D.C. Melles, S.V. Snijders, W.B. van Leeuwen, H.F.L. Wertheim, J.L. Nouwen, H.A. Verbrugh and J. Etienne. 2006. Clonal distribution and differential occurrence of the enterotoxin gene cluster, egc, in carriage versus bacteremia-associated isolates of Staphylococcus aureus. J. Clin. Microbiol. 44: 15551557. Van Wamel W.J.B., S.H.M. Rooijakkers, M. Ruyken, K.P.M. van Kessel and J.A.G. van Strijp. 2006. The innate immune modulators staphylococcal complement inhibitor and chemotaxis inhibitory protein of Staphylococcus aureus are located on $-hemolysin-converting bacteriophages. J. Bacteriol. 188: 13101315.


Indole-3-acetic Acid Production and Effect on Sprouting of Yam (Dioscorea rotundata L.) Minisetts by Bacillus subtilis Isolated from Culturable Cowdung Microflora MANAS R. SWAIN, SAMIR K. NASKAR and RAMESH C. RAY*

Central Tuber Crops Research Institute (Regional Centre), PO: Dumuduma Housing Board, Bhubaneswar 751019, India Received 16 November 2006, revised 10 March 2007, accepted 15 March 2007 Abstract Bacillus subtilis strains (CM1-CM5) isolated from culturable cowdung microflora were investigated for indole-3-acetic acid (IAA) production in nutrient broth (NB). All the strains tested produced IAA in NB; albeit in very low concentrations (0.09 0.37 mg/l). The addition of L-tryptophan (0.1 1.0 g/l) into NB substantially enhanced IAA production (6.1 31.5 folds) indicating that L-tryptophan was the precursor for IAA biosynthesis by these bacterial strains. Maximum IAA production was observed after 8 days of incubation (in late stationary phase of bacterial growth). The variation in IAA production was attributed to the genetic make up of these strains as evaluated by RAPD analysis of these isolates and B. subtilis type strain MTCC 441. Application of B. subtilis suspension (8 × 109 CFU/ml) on the surface of yam (Dioscorea rotundata L.) minisetts increased the number of sprouts, roots and shoots length, root and shoot fresh weights and root: shoot ratio over those minisetts not treated with bacterial suspension. Fresh cowdung slurry treatment on yam minisetts also produced similar results as obtained with B. subtilis application. K e y w o r d s: Bacillus subtilis, Cowdung, yam (Dioscorea rotundata L.), indole-3-acetic acid, L-tryptophan

Introduction There is firm evidence that indole-3-acetic acid (IAA) and other growth regulators (GA3) produced by plants and essential to their growth and development are produced by various bacteria, which live in soil as well as in association with plants (Glick, 1995). There is also evidence that growth regulators such as IAA produced by bacteria can in some instances increase and improve yields of the host plants. Bacterial production of IAA has been studied not only regarding its physiological effects on plants but also regarding its possible role as a phytohormone in plant microbe interaction (Barbieri and Galli, 1993; Xie et al., 1996; Patten and Glick, 2002). Cowdung is a mixture of faeces and urine in a ratio of around 3:1. It contains crude fibre, crude protein, cellulose, hemicellulose and 24 types of minerals such as N, K, S, traces of P, Fe, Ca, Mg, Co, Mn etc. (Nene, 1999). It is normally used as an organic fertilizer for e.g. enhancing soil fertility, dressing seeds and plastering cut ends of vegetative propagated sugarcane

(Kesavan, 2006). Generally cowdung treated seeds are spared from pathogenic fungal and bacterial attack, because bacteria, and particularly Bacillus spp. in cowdung microflora play a significant role in controlling the growth of pathogenic microorganisms by colonizing the surface area of the seeds (Basak and Lee, 2000 a, b). In a previous study, several strains of Bacillus subtilis isolated from cowdung inhibited the growth of Fusarium oxysporum and Botryodiplodia therobromae, two important post harvest pathogens of yam (Dioscorea rotundata L., Family: Dioscoreaceae) tubers (Swain and Ray, 2007) In India, farmers apply cowdung traditionally on yam tubers before planting, which promotes sprouting and seedling growth, and prevents them from rotting (Naskar et al., 2003; Swain and Ray, 2007). However, the mechanism how cowdung promotes sprouting is not yet understood. In this paper, the production of IAA by B. subtilis strains isolated from culturable cowdung microflora has been explored and the effect of exogenous application of B. subtilis culture and cowdung slurry on sprouting of yam tubers has been studied.

* Corresponding author: R.C. Ray, Central Tuber Crops Research Institute Bhubaneswar 751019, Orissa, India; fax: (91) 674 247 05 28; e-mail: [email protected]

104

Swain M.R. et al.

Experimental Materials and Methods

Bacillus subtilis strains. Five Bacillus subtilis strains (CM1, CM2, CM3, CM4 and CM5), showing antagonistic action against Fusarium oxysporum and Botryodiplodia therobromae, were isolated earlier from the culturable microflora of cowdung (Swain and Ray, in press) and maintained on Nutrient Agar slants. Cowdung. Fresh cowdung was collected from a nearby cow shade and brought to the Microbiology Laboratory in sterile polyethylene bags. Screening B. subtilis strains for IAA production. All five B. subtilis strains (n = 2) were evaluated for IAA production (Bentley, 1962) by inoculating 1 ml of each of cells suspension (1× 106 CFU/ml) in 50 ml of nutrient broth (NB) [peptone, 5.0; yeast extract, 3.0; beef extract, 1.0 and NaCl, 5.0 (g/l)] kept in 250 ml Elenmeyer flasks. L-tryptophan was added at 100 mg/l NB to half (n = 3) of the flasks and the others were kept without L-tryptophan. The flasks were incubated for 8 days at 30°C while being shaken at 120 rpm in an incubator-cum-shaker (Remi India Pvt. Ltd. Bombay, India) and the culture filtrates were analyzed for IAA production as described in the section plant growth hormones extraction and estimation. Experiment 1: Effect of L-tryptophan concentration on IAA production. Strains CM4 and CM5 of B. subtilis were used in this experiment. One ml of each bacterial cell suspension (1× 106 CFU/ml) was inoculated in 250 ml Elenmeyer flasks containing 50 ml of the NB and supplemented with or without L-tryptophan (02 g/l). The flasks (n= 3) were kept shaken at 120 rpm at 30°C and the culture filtrates were analyzed for IAA concentration at the end of 8 days incubation period. The growth of B. subtilis strains was determined by measuring the optical density of the growth medium at 600 nm in a UV-Vis Spectrophotometer (Model No 302, Cecil Instrument, UK). Experiment 2: Effect of incubation period on IAA production. Effect of incubation period (210 days) was studied by inoculating B. subtilis (CM4 or CM5) in NB containing 1 g/l L-tryptophan. The experimental conditions were otherwise the same as in the previous experiment. Plant growth hormones extraction and estimation. After being centrifuged (8000 g for 20 min) bacterial cells were extracted thrice by adding the same amount of ethyl acetate after adjusting pH to 2.5 with 1N HCl (Hansan, 2002). The extracted fractions were mixed together and reduced to 2 ml by evaporation at 45°C using a Rotary Vacuum Evaporator (Model No 102, Strike, Italy). The concentrated extracts were re-dissolved in acetone before thin-layer-chromatography (TLC). TLC was carried out using 0.5 mm-

2

thick preparative silica gel plates and solvent used was a mixture of isopopanol, ammonium hydoxide, and water (10:1:1; v/v/v) to separate plant growth hormones (IAA and GA3). IAA was detected on TLC plates by spraying with Ehrlich reagent (10% p-dimethyl aminobenzyldehyde in 70% perchloric acid) which resulted in development of pink colour visualized under normal light (Bentley, 1962). To detect GA3, extracts were spread on the plates using ethanolic sulphuric acid (90:10; v/v) and heated to induce fluorescence of the compounds in ultraviolet light (MacMilan and Suter, 1963). The identified spots for growth regulators on the TLC plates were eluted in methanol. The colour absorbency for IAA was measured at 565 nm using UV-Vis Spectrophotometer (Model No 302, Cecil Instrument, UK). The IAA content was measured from a standard curve prepared with a known concentration of IAA and expressed as mg/l. The flow-chart for extraction and bioassay of growth regulators from bacterial culture is given in Figure 1. Random Amplified Polymorphic DNA (RAPD) analysis of B. subtilis stains. DNA of B. subtilis strains (CM1-CM5) were isolated from cultures grown overnight in 10 ml NB by the method described by Lampe (1998). A single RAPD primer OPA-15 (5TTCCCGACC3) specific for Bacillus spp. was used for DNA amplification (Lampe, 1998). The reaction mixture (20 ml) contained 40 ng of genomic B. subtilis DNA and 40 pmol of RAPD primer. PCR cycles were performed in a DNA thermal cycle (Model No 200, MJ Research, UK) as follows: four cycles at 94°C for five min, 30°C for five min and 72°C for five min, 30 cycles at 94°C for 30 s, 35°C for 15 s and 72°C for one min and finally extension of 72°C for 10 min. PCR mixture was electrophoresed on 1.5% agarose gel in tris-borate-EDTA buffer to confirm the RAPD typing (Darling et al., 1998) Experiment 3: Effect of B. subtilis culture and cowdung on sprouting of yam minisetts. Healthy and diseases free yam tubers harvested within 2030 days were collected from the farm of Regional Centre of Central Tuber Crops Research Institute, Bhubaneswar during the month of March, 2006 (day temperature, 30 ± 2°C and night temperature, 24 ± 2°C). The tubers were washed under running tap water and its surfaces were sterilized for 10 min in 1% NaOCl followed by a washing step with sterile water. The tubers were dried under laboratory conditions (room temperature, 30± 2°C; relative humidity, 7080%). The tubers were cut into cubes (minisetts) of approximate sizes of 5 × 6 × 8 cm3 weighing between 130150 g each. The yam minisetts (n = 6) were dipped in B. subtilis (CM4 and CM5) culture suspensions separately (8× 109 CFU/ml) for 2 h, planted 5 cm below the sand, and kept for 15 days under laboratory conditions.

2

105

Effect of IAA produced by B.subtilis on sprouting of yam Bacterial cultur broth

Contrifugation at 8000 × g for 20 min to separate cells

Cell free supernatant pH adjusted to 2.5 with 1N HCL

Extracted 3 times with ethyl ecetate

Evaporation under vacuum at 45°C

Re-dissolved in absolute acetone

Thin Layer Chromatography Spraying with Ehrlich reagent

Spraying with ethanolic H2SO4 (90:10 v/v) GA3 (fluoresced under ultraviolet light)

IAA (pink colour spots)

Spectrophotometric bioassay Fig. 1. Procedure used for extraction and bioassay of growth regulators (IAA and GA3) from bacterial cultures.

Cowdung slurry was infused with water (1:1; w/v), and homogenized by shaking for 2 h at 120 rpm. The yam minisetts prepared as above were dipped in cowdung slurry for 2 h and then planted on sand bed for 15 days. Minisetts without any (bacterial or cowdung) treatment served as control in this experiment. Six replicates (n = 6) were maintained for control as well as for each treatment and mean data with standard deviations were calculated for number of sprouting, root and shoot length, root and shoot freshweight, root and shoot dry weight and root: shoot ratio per minisett. Solid-state bioprocessing of cowdung for IAA production by B. subtilis. Twenty grams of ovendried cowdung kept in 250 ml Erlenmeyer flasks was adjusted to 60% water holding capacity (WHC) by the addition of sterile distilled water and autoclaved at 15 lb for 15 min. The sterilized cowdung dust was cooled at room temperature and inoculated with strain of CM4 and CM5 mixture at 10% rate (1×106 CFU/ml) and culture flasks (n = 3) were incubated at 30 ± 2°C for 10 days. The contents in Erlenmeyer flasks were

mixed periodically by gentle tapping and distilled water was supplemented to replenish the loss of water evaporated to maintain WHC of 60%. After 2 days of incubation triplicate flasks were taken out. To determine IAA content, 50 ml distilled water was added to each flask and the mixtures (cowdung and water) were homogenised thoroughly at 200 rpm for 30 min in a shaker. The homogenate was filtered through a cheese cotton cloth and centrifuged at 5000 g for 20 min. The supernatant was then filtered through a Whatmann No 1 filter paper and taken for IAA extraction. The IAA content in cowdung was expressed as mg IAA/gram dry substrate (gds). Result and Discussion In our earlier studies, the microbial composition of cowdung which includes several bacteria (i.e. Bacillus spp., Corynebacterium spp., Lactobacillus spp., etc), fungi (i.e. Aspergillus, Trichoderma, etc) and yeasts (i.e. Saccharomyces, Candida, etc) has been

106

2

Swain M.R. et al.

discussed (Swain et al., 2006; Swain and Ray, 2007). Of these, B. subtilis strains are the predominant microorganisms of agricultural importance (Swain and Ray, 2007). Preliminary studies using TLC and spectrophotometric bioassay showed that B. subtilis strains (CM1-CM5) produce only IAA while GA3 could not be detected (MacMilan and Suter, 1963). In the absence of L-typtophan, B. subtilis strains CM 1, CM2 and CM3 produced negligible amounts (0.120.22 mg/l) while strains CM4 and CM5 produced 0.38 mg/l (62.3% more than CM1-CM3) of IAA. Furthermore, the addition of L-tryptophan (0.1 g/l) enhanced IAA production by CM4 and CM5 strains to 2.12.5 mg/l (5.66.8 fold). Consequently, these two strains were chosen for further studies. Effect of L-tryptophan concentrations on IAA production. L-tryptophan is generally considered as an IAA precursor, because its addition to IAA pro-

ducing bacterial culture promotes an increase in IAA biosynthesis (Costacurta and Vanderleyden, 1995; Tien et al., 1979). Figure 2A shows the effect of L-tryptophan concentrations on IAA production by B. subtilis CM4 and CM5. With the increase in L-tryptophan concentration from 0 to 2 g/l, there was a linear increase in IAA production in the case of strain CM4 (Fig. 2A). However, in the case of CM5, an increase in IAA production was observed up to concentration of tryptophan 1 g/l and there was a slight decrease at higher concentrations. This might be explained by the fact that a single bacterial strain often uses more than one biosynthesis pathway for IAA production (Patten and Glick, 1996). Concomitant with increased IAA biosynthesis due to increasing L-tryptophan concentration, the growth of B. subtilis was also stimulated (Fig. 2B). A similar result was reported in for Azospirillum brasilense (Tien et al.,

25

A

IAA (mg/l)

20

15

10

5

0 0

0.1

0.5 1 2 0 0.1 0.5 CM4 CM5 L-Tryptophan concentration (g/l)

5

2

B

4,5

Growth OD (600 nm)

1

4 3,5 3 2,5 2 1,5 1 0,5 0 0

0,1

0,5

1

2

0

0,1

CM4

0,5

1

2

CM5 L-Tryptophan concentration (g/l)

Fig. 2. Effect of L-tryptophan concentration on (A) IAA production and (B) growth by B. subtilis strains (CM4 and CM5).

2

107

Effect of IAA produced by B.subtilis on sprouting of yam 16

A

14

IAA (mg/l)

12 10 8 6 4 2 0 2

4

6

8

CM4

10

2

4

6

8

10

CM5

Incubation priod (days)

B

4

Growth OD (600 nm)

3.5 3 2.5 2 1.5 1 0.5 0 2

4

6 CM4

8

10

2

4

6

8

10

CM5 Incubation priod (days)

Fig. 3. Effect of incubation period on (A) IAA production and (B) growth by B. subtilis strains (CM4 and CM5).

1979). Likewise, several strains of B. subtilis produced IAA in culture media (Tang, 1994; Ghosh et al., 2003). Tryptophan-dependant IAA synthesis has been also determined in several other bacteria (Patten and Glick, 1996). For example, in Enterobacter cloacae, IAA was synthesized via indole-3pyruvic acid (Koga et al., 1991); in Pseudomonas syringae, IAA biosynthesis occurs mostly from tryptophan via indole-3-acetamide (Hutcheson and Kosuge, 1985; Kosuge and Sanger, 1987) and in Pseudomonas fluorescens, tryptophan bypassing the indole-3pyruvic acid step, is directly converted to indole-3acetaldehyde, which is further converted to IAA (Oberhansli et al., 1991). IAA synthesis has also been found to occur via tryptamine in Agrobacterium tumefaciens and via indole-3-acetonitrile in Alcali-

genes faecalis and A. tumefaciens (Costacurta and Vanderleyden, 1995; Kobayashi et al., 1995). Effect of incubation period on IAA production. Figure 3A shows IAA production by B. subtilis strains CM4 and CM5 in tryptophan (1 g/l) supplemented culture medium. The production of IAA was almost linear from 2 to 8 days; after that the IAA biosynthesis marginally decreased. This was concomitant with the growth of B. subtilis in tryptophan-supplemented medium (Fig. 3B). Unyayar et al., (2000) and Hansan (2002) reported for Pseudomonas spp. that the maximum amount of IAA was synthesized during stationary phase of growth. The reason hypothesized was that during stationary phase the bacterium might be able to get maximum tryptophan from the dead bacterial mass, which could result in more IAA production.

108

2

Swain M.R. et al.

10.0 kb 5.0 kb 4.0 kb

1.0 kb

Fig. 4. RAPD finger printing of DNAs from B. subtilis strains amplified with primer OPA 15. From left: Lane1-molecular size markers (1-kb ladder), Lane 2 CM4, Lane 3 CM1, Lane 4 CM5, Lane 5 CM2, Lane 6 CM3 and Lane 7 Type strain MTCC 441. A photograph of gel was scanned with a HP 800 Scanner.

RAPD amplification. B. subtilis strains (CM1 to CM5) were compared with standard B. subtilis (MTCC 441) by RAPD fingerprinting using a single primer (Fig. 4). Lane 1 represented standard marker, lanes 2 to 6 represented B. subtilis strains (CM1 to CM5) and lane 7 represented type strain B. subtilis (MTCC 441). The band patterns obtained with RAPD were all different except for a single common band detected at 4.0 kb. The result indicated that all the isolates belong to B. subtilis strains but at gene level they were different from each other, as evident from the production of IAA in the present study. Effect of B. subtilis strains and cowdung on sprouting of Dioscorea minisetts. When yam minisetts were dipped in B. subtilis suspension, an increase in root and shoot length as well as fresh and dry weights in comparison with control (minisetts not treated with B. subtilis) could be observed (Table I). For example, yam minisetts treated with B. subtilis strain CM 4 showed 63.5 and 83.3% more root and shoot elongation, respectively, in comparison with those not treated with the bacterial suspension

(Fig. 5A). Similarly, 76.2 and 75.7% more root and shoot fresh weights, respectively, were observed in B. subtilis CM4 treated yam minisetts in comparison with the control minisetts. Similar results were observed also for CM 5 strain treated minisetts. In general, the average root: shoot ratio was higher in B. subtilis-treated yam minisetts in comparison to those not treated with the bacterial culture. In earlier reports, root elongation was found to occur in Sesbania aculeata by inoculation with Azotobacter sp. and Pseudomonas sp. (Ahmad et al., 2005), in Brassica campestris by Bacillus spp. (Ghosh et al., 2003), in Vigna radiata by Pseudomonas putida and in (Patten and Glick, 2002), Pennisetum americanum by Azospirillum brasilense (Tien et al., 1979). Like B. subtilis treatment, increase in shoot and root number and length was observed in the case of cowdung-treated minisetts (Table I). However, the effect was smaller than obtained with the B. subtilis treatment (Fig. 5B). Since B. subtilis strains in the present study did not produce GA3, which could stimulate root and shoot elongation (Hopkins, 1999), this indirectly confirmed the involvement of IAA synthesized by the bacterial strains in enhancing the sprouting of yam minisetts. IAA production by B. subtilis in solid-state fermentation. To examine whether cowdung could serve as a solid substrate for the production of bioinoculants such as B. subtilis that produces IAA, moistened cowdung was inoculated with CM4 and CM5 strains and IAA production was followed for 10 days. IAA production was marginally higher on days 2 4 (42 45 mg/gds) by CM 4 strain and remained more or less stable during the incubation days 6 to 10 (30 36 mg/gds). The synthesis of IAA by CM5 strain was almost the same during the course of study (2 10 days) (38.0 43.0 mg/gds). Conclusion. Cowdung is traditionally used in Asian agriculture as an organic fertilizer. The goal of the evergreen revolution necessarily involves components, which dont adversely affect soil health, water quality, biodiversity, atmosphere and renewable energy sources (Kesavan, 2006). In this context, organic farm-

Table I Effect of application of B. subtilis strains (CM 4 and CM5) and cow dung slurry on sprouting of Dioscorea minisetts Parameters

Control

B. subtilis (CM4)

B. subtilis (CM5)

Cowdung

Root: shoot ratio Root length (cm) Shoot length (cm) Root fresh wt. (g) Shoot fresh wt. (g) Shoot dry wt. (g) Root dry wt. (g)

5.02 15.7 ± 1.23 2.58 ± 0.49 1.24 ± 0.23 1.02 ± 0.03 0.12 ± 0.01 0.11 ± 0.02

6.16 43.08 ± 3.12 15.0 ± 0.90 5.21 ± 2.3 4.19 ± 1.12 0.47 ± 0.04 0.39 ± 0.02

6.2 44.09 ± 3.01 13.8 ± 1.32 5.22 ± 1.80 4.20 ± 0.80 0.47 ± 0.01 0.41 ± 0.1

5.13 34.00 ± 2.02 7.60 ± 1.02 3.54 ± 1.32 3.04 ± 0.42 0.30 ± 0.10 0.23 ± 0.02

2

109

Effect of IAA produced by B.subtilis on sprouting of yam

A

I

II

B

I

II

Fig. 5. Sprouting of Dioscorea minisetts after 15 days planting on sand bed. A. (I) Minisetts with B. subtilis CM4 treatment (II) Minisetts without B. subtilis CM4 treatment. B. (I) Minisetts with cowdung treatment (B) Minisetts without cowdung treatment.

ing using natural substances such as cowdung and effective microorganisms such as B. subtilis would sustain and even increase agricultural productivity without affecting soil health. This is implicit in the activity of cowdung microflora such as B. subtilis by producing growth regulator such as IAA which promoted sprouting (in the present study), biocontrol activity against plant pathogens (Basak and Lee, 2000a, b; Swain and Ray, 2007) and production of agriculturally important enzymes such as "-amylase (Swain et al., 2006) which have manifold beneficial impacts on crop growth and production.

Literature Ahmad F., L. Ahmad and M. S. Khan. 2005. Indole acetic acid production by the indigenous isolates of Azotobacter and fluorescent Pseudomonas in presence and absence of tryptophan. Turk. J. Biol. 29: 2934. Barbieri, P. and E. Galli. 1993. Effect on wheat root development of inoculation with an Azospirillum brasilense mutant with altered indole-3-acetic acid production. Microbiol. Res. 144: 6975. Basak A.B. and M.W. Lee. 2000a. Comparative efficacy and in vitro activity of cow urine and cowdung for controlling Fusarium wilt of cucumber. http://plantpath.snu.ac.kr/ic2001/abstract.html. August 2005

110

Swain M.R. et al.

Basak A.B. and M.W. Lee. 2000b. Efficacy of cowdung in controlling root rot and Fusarium wilt diseases of cucumber plants. http://plantpath.snu.ac.kr/ic2001/abstract.html. August 2005 Bentley J.A. 1962. Analysis of plant hormones methods. Biochem. Anal. 9: 75124. Costacurta A. and J. Vanderleyden. 1995. Synthesis of phytohormones by plant-associated bacteria. Crit. Rev. Microbiol. 21: 118. Darling P., M. Chan, A.D. Cox and P.A. Sokal. 1998. Siderophore production by Cystic fibrosis isolates of Brukholderia cepaciae. Infect. Immuni. 66: 874877. Ghosh S., J.N. Penterman, R.D. Little, R. Chavez and B.R. Glick. 2003. Three newly isolated plant growth-promoting bacilli facilitate the seedling growth of canola, Brassica campestris. Plant Physol. Biochem. 41: 277281. Glick B.R. 1995. The enhancement of plant growth by free-living bacteria. Can. J. Microbiol. 41: 109117. Hansan H.A.H. 2002. Gibberellin and auxin production by plant root fungi and their biosynthesis under salinity-calcium interaction. Rostlinna Vyroba, 48: 101106. Hopkins W.G. 1999. Introduction to Plant Physiology, 2nd ed. John Wiley and Sons, Inc. USA. Hutcheson S. and T. Kosuge. 1985. Regulation of 3-indoleacetic acid production in Pseudomonas syringae pv. savastanoi (purification and properties of tryptophan 2-monooxygenase). J. Biol. Chem. 260: 62816287. Kesavan P.C. 2006. From green revolution to evergreen revolution: pathways and terminologies. Curr. Sci. 90: 45146. Kobayashi M., T. Suzuki, T. Fujita, Masuda, M. and S. Shimizu. 1995. Occurrence of enzymes involved in biosynthesis of indole3-acetic acid from indole-3-acetinitrile in plant associated bacteria, Agrobacterium and Rhizobium. Proc. Natl. Acad. Sci. USA. 92: 714718. Koga J., T. Adachi and H. Hidaka. 1991. Molecular cloning of the gene for indolepyruvate decarboxylase from Enterobacter cloacae. Mol. Gen. Genet. 226: 1016. Kosuge T. and M. Sanger. 1987. Indole acetic acid, its synthesis and regulation: basis for tumorigen city in plant disease. Recent. Adv. Phytochem. 20: 147161. Lampe K.R. 1998. Ph.D. Thesis. Systematics of the entomopathogenic bacteria Bacillus popilliae, Bacillus lentimorbus and Bacil-

2

lus sphaericus., Virginia Polytechnic Institute and State University, Blacksburg, Virginia. MacMilan J and P.J. Suter. 1963. Thin layer chromatography of the gibberellins. Nature (London). 97: 790. Naskar S.K., P. Sethuraman and R.C. Ray. 2003. Sprouting in yam by cowdung slurry. Validation of Indigenous Technical Knowledge in Agriculture. Division of Agricultural Extension, Indian Council of Agricultural Research, New Delhi, India, pp. 197201. Nene Y.L. 1999. Seed health in ancient and medieval history and it relevance to present day agriculture. Asian Agri. His. 153184. Oberhansli T., G. Defago and D. Haas. 1991. Indole-3-acetic acid (IAA) synthesis in the biocontrol strain CHAO of Pseudomonas fluoresces: role of tryptophan side chain oxidase. J. Gen. Microbial. 137: 22732279. Patten C.L. and B.R. Glick. 2002. Role of Pseudomonas putida indo lactic acid in development of the host plant root system. Appl. Environ. Microbial. 68: 37953801. Patten C.L. and B.P. Glick. 1996. Bacterial biosynthesis of indole-3-acetic acid. Can. J. Microbiol. 42: 207220. Swain M.R. and R.C. Ray. 2007. Biocontrol and other beneficial activities of Bacillus subtilis isolated from cowdung microflora. Microbiol. Res. 162: (in press). Swain M.R., S. Kar, G. Padmaja and R.C. Ray. 2006 Partial characterization and optimization of production of extracellular "-amylase from Bacillus subtilis isolated from culturable cowdung microflora. Polish J. Microbiol. 55: 289296. Tang W.H. 1994.Yield-Increasing Bacteria (YIB) and biocontrol of sheath blight of rice. In: Improving Plant Productivity with Rhizosphere Bacteria (M.H. Ryder, ed.), CSIRO Division of Soils, Glen Osmond, Australia, pp. 267273. Tien T.M., M.H. Gaskins and D.H. Hubbell. 1979. Plant growth substrates produced by Azospirillum brasilense and their effect on growth of pearl millet (Pennisetum americanum L.). Appl. Environ. Microbiol. 37: 10161024. Unyayar S., A. Unyayar and E. Unal. 2000. Production of auxin and abscisic acid by Phanerochaete chrysosporium ME446 immobilized on polyurethane foam. Turk J. Biol. 24: 769774. Xie H., J.J. Pasternak and B.R. Glick. 1996. Isolation and characterization of mutants of the plant growth-promoting rhizobacterium Pseudomonas putida GR12-2 that over produce indole acetic acid. Curr. Microbiol. 32: 6771.


Mercury Absorption by Pseudomonas fluorescens BM07 Grown at Two Different Temperatures KAMBIZ A. NOGHABI*1, HOSSEIN S. ZAHIRI1, ABBAS S. LOTFI1, JAMSHID RAHEB1, SIMA NASRI2 and SUNG C. YOON3 1 National

Institute for Genetic Engineering and Biotechnology (NIGEB), Tehran, IRAN, of Biology, Azad University, Parand Branch, Parand New City, IRAN, 3 Environmental Biotechnology National Core Research Center; Gyeongsang National University, Chinju, S. KOREA 2 Department

Received 27 January 2007, revised 15 April 2007, accepted 17 April 2007 Abstract Pseudomonas fluorescens BM07 was characterized as a producer of cold-induced biopolymer by decreasing the temperature down to as low as 10°C. It was previously shown that the synthesis of BM07 biopolymer was inhibited at 30°C. The present study was conducted to investigate the biosorption of mercury (Hg2+) ions on the BM07 cells grown on M1 minimal medium at two temperatures (10°C and 30°C). The effects of various factors including pH, contact time, initial concentration of metal and cell biomass on the biosorption yield were also studied. Study of the effect of pH on mercury removal indicated that the metal biosorption increased with increasing pH from 3.0 to 7.0. The optimum adsorption pH value was found to be 7.0. Our results showed that, at optimum pH, BM07 cells were able to uptake the mercury up to 102 and 60 mg Hg2+/g dry biomass for 10°C and 30°C grown cells respectively. The removal capacity of cells increased when the cell biomass concentrations increased. The maximum removal efficiency was obtained when cells concentration was 0.83 mg dry biomass/ml for both conditions. The initial metal ion concentration significantly influenced the equilibrium metal uptake and adsorption yield. The equilibrium data were analyzed using Langmuir adsorption model. The qmax was 62.9 and 82.25 mg Hg2+ /g dry biomass for cells grown at 30°C and 10°C respectively. The results suggest that, the existence of residual cold-induced biopolymer on the external surface of cells may play an important role in biosorption efficiency, as P. fluorescens BM07 cells which were grown at 10°C under similar conditions showed higher efficiency to biosorbe mercury than non-polymer producing cells grown at 30°C K e y w o r d s: Pseudomonas fluorescens BM07, biosorption, mercury

Introduction Contamination of soil, water and groundwater by heavy metals in many areas in the world is a matter of high concern and constituting serious environmental problems. Among these heavy metals, mercury is well known for its high toxicity and strong affinity toward the thiol groups in proteins. Purification of areas polluted by heavy metals such as mercury is difficult because they cannot be transformed to harmless elements. Conventional methods such as: chemical precipitation, chemical oxidation or reduction, ion exchange, filtration, electrochemical treatment, reverse osmosis, membrane technologies are generally used for removing metals from aqueous solutions (Grau and Bisang, 1995; Gray, 1999). These chemical pro-

cesses may be ineffective or extremely expensive and need enormous input of chemicals leading to secondary pollution (Habashi, 1978). Therefore removal of toxic heavy metals by an environmentally friendly manner is of great importance. Far works has been done on the heavy metals absorption by microorganisms. The microbial processes for bioremediation of toxic metals from waste streams employ living cells, non-living biomass or biopolymers as biosorbents (Volesky and Holan, 1995). Different bioproducts and non-living biomass types have been used to adsorb heavy metal ions from the environment. Chitosan, wool, peanut skins, seaweed, mold, bacteria, crab shells and yeast are among the different kinds of biomass, which have been tested for metal biosorption or removal (de Rome and Gadd,

* Corresponding author: K.A. Noghabi, National Institute for Genetic Engineering and Biotechnology (NIGEB), P.O. BOX 14155-6343, Tehran, Iran; phone: +98-21-44580532; fax: +98-21-44580399; e-mail: [email protected], [email protected]

112

2

Noghabi K.A. et al.

1991; Friedman and Waiss, 1972; Peniche-Covas et al., 1992; Tiemann et al., 1999; Volesky and Holan, 1995; Volesky, 1990). Pseudomonas fluorescens is nonpathogenic, saprophyte which can be found in soil, water and plant surface environments. This bacterium has simple nutritional requirements and grows well in mineral salt media supplemented with various types of carbon sources (Palleroni, 1989). BM07 strain was capable to produce a lot of slime material when the temperature was decreasing down to as low as 10°C. Maximum production of the cold-induced biopolymer was obtained when cells were grown aerobically at 10°C and its synthesis was inhibited at 30°C (Noghabi, unpublished). It has been reported the sorption of mercury ion by inactivated cells of mercury resistant bacterium, Pseudomonas aeruginosa PU21 (Chang and Hong, 1994). No reports have been mentioned about the mercury biosorption by P. fluorescens. The current study was undertaken to assess the mercury sorption efficiency of living biomass of P. fluorescens BM07 which was grown at two different temperatures (10°C and 30°C) and the probable role of residual biopolymer on the external surface of cells grown at 10°C as well as its efficacy for Hg2+ ions absorption rate. Experimental Materials and Methods

Microorganism and culture conditions. Pseudomonas fluorescence BM07 was isolated from activated sludge in a municipal wastewater treatment plant in the south of Korea and maintained in nutrient rich agar medium containing 1% yeast extract, 1.5% nutrient broth, 1% ammonium sulfate and 2% agar, at 4°C (Lee et al., 2001). A 5 ml NR broth medium was inoculated with a single colony of P. fluorescens BM07 and incubated at 30°C, 180 rpm for 12 h as seed culture and then transferred to a 2-liter flask containing 500 ml of modified M1 minimal medium of the same composition as reported earlier (Choi and Yoon, 1994). Fructose (70 mM) and ammonium sulfate (1 g/l) were used as carbon and nitrogen sources and initial pH of the medium was adjusted to 7.0. P. fluorescens BM07 was cultivated in M1 minimal medium at two different temperature, 30°C and 10°C. Bacterial cells at late exponential phase (144 h and 72 h for cells grown at 10°C and 30°C respectively) were harvested by centrifugation (10 000× g, 20 min). Afterward the cells were washed with distilled water twice and freeze dried. Time course samples of culture medium were withdrawn in appropriate time intervals and monitored for optical density at 660 nm (OD 600).

Metal sorption assay. A batch equilibrium method was used to determine sorption of mercury by P. fluorescens BM07. A set of 100 ml Erlenmeyer flasks containing 30 ml of the tested mercury solution was used in the experiments. Powdered dried cells (10 mg) were exposed to metal solution for 2 hours at 25°C on a rotary shaker at 170 rpm. The dried powdered cells were separated by centrifugation at 12 000×g for 10 min, and supernatants were analyzed for residual mercury concentration on ICP-AES model Optima 4300 DV. Metal absorbtion by the tested dried cells (mg metal/g dry cells biomass) was calculated according to the Volesky and May-Phillips method (Volesky, 1990). The mercury sorption efficiency of the dried cells was determined by the following equation in all of the experiments, unless otherwise stated: Q = V (Ci Cf)/ 1000 M; where: Q is specific mercury uptake (mg/g biosorption), V is volume of mercury solution (ml), Ci is initial concentration of mercury in the solution (mg/l), Cf is final concentration of mercury in the solution (mg/l), M is mass of the powdered dried cells (g). The Langmuir model as sorption model was used to evaluate the sorption behavior of BM07 cells grown at two different temperatures. It served to estimate the maximum mercury uptake values. Its constant b can serve as an indicator of the isotherm which reflects quantitatively the affinity between the sorbent and the sorbate (equations) (Hughes and Poole, 1989). Q=

QmaxbC ; 1 + bC

Q=

QmaxbC ; K+C

Q is the amount of mercury bound per unit weight biomass, b is the equilibrium constant and K (mg/l) is dissociation constant related to the stability of metalbiosorbent complexes, reciprocal to the equilibrium constant b. C is the concentration of mercury remaining in solution at equilibrium (free), Qmax (mg Hg2+/g biosorbent) is the maximum metal uptake under the given conditions. Sorption study as a function of pH. In order to evaluate the effect of pH on Hg2+ uptake, 10 mg of dried powdered living cells of P. fluorescens BM07 were suspended in a volume of 30 ml of mercury solution in a 100 ml conical flask for 2 hours on a rotary shaker at 170 rpm. The pH of solution was adjusted to be in the range 38. After shaking the flasks for 2 h, the suspension was centrifuged at 12 000×g for 10 min. The supernatant was collected in separate clean test tubes and analyzed for residual mercury content. Sorption study as a function of metal concentration. To determine the absorptive capacity of P. fluorescens BM07, the initial metal ion concentrations varied from 7 750 mg/l while the dry cell weight in each sample was constant at 0.33 mg/l. Equilibrium batch experiments resulted in points which were approximated by the Langmuir model. This model was used

2

113

Mercury absorption by P. fluorescens

to evaluate the sorption behavior of the materials examined and served to estimate the maximum level of metal uptake values (qmax) when they could not be reached in the experiments. A volume of 30 ml of metal ion solution Hg2+ (as HgCl2) was placed in a 100 ml conical flask with varying initial metal ion concentrations in duplicates. An accurately weighed P. fluorescens BM07 biomass sample (10 mg) was then added to the solution to obtain a suspension. The suspensions were adjusted to pH 7.0. A series of such conical flasks was then shaken at a constant speed 170 rpm at temperature of 25°C. After shaking the flasks for 2 hours, the suspension was centrifuged at 12 000× g for 10 min. The supernatants were collected in separate clean test tubes. To evaluate the biosoption rate of mercury as a function of cell biomass quantities, various dried cells weight of 0.16 to 0.83 mg/l was used. The residual metal content at each condition was determined using ICP-AES. Effect of exposure time. To better examine the mercury biosorption mechanism, 10 mg of dried powdered cells of bacteria were contacted with 30 ml of aliquots of mercury solution (52 mg/l) in 100 ml of conical flasks. Shaking flasks were incubated at 25°C for different time intervals (0 to 120 min) and analyzed for residual mercury content. Scanning electron microscopic analysis of BM07 cells grown at 10°C. P. fluorescens BM07 cells in the late exponential phase were centrifuged at 5000 rpm

for 20 min and washed sufficiently with distilled water. The cells were fixed with 0.1 M phosphate buffer (pH 7.2) containing 1% glutaraldehyde for 2 h, washed with distilled water. Fixed cells were then dehydrated through a graded ethanol series (25, 50, 75, 95 and 100%) for 5 min each. The final dehydration process was repeated two times. The dried cells were spatter-coated with gold. SEM observation was carried out using a JEOL JEM-2010 scanning electron microscope. Results and Discussion Effect of pH value. Removal of metals from aqueous solutions is significantly influenced by pH of the medium. Experimental results are presented in Figure 1. In both conditions the maximum mercury biosorption occurred at around neutral pH. The amount of adsorbed mercury on dry biomass at pH 7 were respectively about 3% and 58% for cells grown at 30°C and 10°C. There was an increase in Hg2+ adsorption per unit weight of biomass with pH 7. Biosorption kinetics and effect of biomass quantity on mercury uptake. The uptake of mercury takes place at a high reaction rate and is completed after few minutes. A comparison between maximum amounts of uptake concentration and theoretical maximum amounts with a total saturation of the 120

70 30°C 10°C 10°C 30°C

60

100

80 40 60 30 40

qeq (mg/g dry biomass)

Mercury removal (%)

50

20 20

10

0

0 2

3

4

5

6 pH value

7

8

9

Fig. 1. Effect of initial pH on equilibrium absorption capacity of mercury ion and its removal by dried cells of P. fluorescens BM07 grown on M1 minimal medium at 30°C and 10°C (C0: 63 mg/l, temperature 25°C)

114

90

50

80 30

70 60

20

50 10 40 0

0.16 0.33 0.5 0.66 0.83 Cell biomass concentration (mg/ml)

30

Q (mg/gr dry biomass) Metal removal (%)

100

90

40 80 30 70 20 60

10 0

Metal removal (%)

40

60

Q (mg/gr dry biomass)

Q (mg/gr dry biomass) Metal removal (%)

100

Metal removal (%)

50

Q (mg/gr dry biomass)

2

Noghabi K.A. et al.

0.16 0.33 0.5 0.66 0.83 Cell biomass concentration (mg/ml)

50

Fig. 2. Effect of BM07 cell biomass quantity on specific uptake and mercury removal at 30°C (A) and 10°C (B).

surface leads to the assumption that mercury is bound to definite sites. The biomass concentration remarkably influenced the equilibrium metal uptake and specific uptake of mercury. With an increase in the concentration of BM07 cell biomass the larger amount of mercury was taken up. Maximum removal and specific uptake of Hg2+ was achieved using 0.83 mg/ml of cell biomass quantity for both conditions (Fig. 2). Time of contact of adsorbent and adsorbate is of great importance in biosorption, because it depends on the nature of system used. Mercury uptake by BM07

living cells was a rapid process and occurred within a few minutes. This has well supported our observation of mercury-bacterium adsorption system equilibrium where maximum adsorption was achieved within 20 and 40 min respectively for cells grown at 10°C and 30°C (Fig. 3). Effect of the initial mercury concentration on biosorption rate. The concentration of both the metal ions and biosorbent is a significant factor to be considered for effective biosorption. It determines the sorbent/sorbate equilibrium of the system. The rate of adsorption is a function of the initial concentration of

70 10°C 30°C

qeq (mg/g dry biomass)

60

50

40

30

20 0

20

40 60 80 Contact time (min)

100

120

140

Fig. 3. Time course of mercury uptake by dried cells of P. fluorescens BM07 (C0: 52 mg/l, temperature 25°C).

2


Table I Equilibrium adsorbed quantities and adsorption yields of mercury ion obtained at different initial metal ion concentrations* Cells grown at 10°C C0 (mg/l)

Ad%

Cells grown at 30°C

qeq (mg/g) C0 (mg/l) 0.60

Ad%

qeq (mg/g)

7.04

2.84

7.04

30.57

6.46

29.15

34.30

30.0

29.15

27.65

24.2

110.3

9.50

31.5

110.3

9.32

30.9

721

3.64

85.5

721

3.50

74.5

* Temperature 25°C; agitation rate 170 rpm, at pH 7.0.

metal ions. The initial mercury ion concentration remarkably influenced the equilibrium metal uptake and absorption yield as shown in Table I. When the initial mercury ion concentration varied from 7 to 721 mg/l, the loading capacity of dried cells of P. fluorescens BM07 increased from 1 to 85 and 6 to 78 mg/g dry biomass of cells grown at 10°C and 30°C respectively. The increase of loading capacity of biosorbents with the increase of metal ion concentration is probably due to higher interaction between metal ions and each of biosorbent. As can be deduced from Table I, higher adsorption yields were observed at lower concentrations of metal ion. Isotherms of mercury biosorption by cells. Basing on the data obtained here, the mercury ion uptake capacity of the living dried cells of P. fluorescens BM07 grown at two different temperatures (30°C and 10°C) was calculated using the Langmuir isotherms at fixed biosorbent mass of 0.33 mg/ml. As mentioned above the Qmax (mg/g dry biosorbent) is the maximum metal uptake under the given conditions, b is the equi-

115

librium constant and K (mg/l) is dissociation constant related to the stability of metal-biosorbent complexes, reciprocal to the equilibrium constant, b. Thus, a plot of 1/Q versus 1/C was used to obtain the Qmax (intercept), and K (slope) as shown in Fig. 4. The data on conversion to Langmuir adsorption isotherms model resulted in a straight line (Fig. 4). The values of qmax calculated from the linearized Langmuir plot for P. fluorescens BM07 cells. These values are very close to the experimental values. This shows that the experimental biosorption data perfectly fit the Langmuir isotherms equation. The regression coefficient (r2) for cells was respectively 0.988 and 0.922 for cells grown at 30°C and 10°C which further support the goodness of fit to the Langmuir model. The Langmuir parameters of living biomass of P. fluorescens BM07 grown at 30°C and 10°C were estimated and Qmax was 62.9 and 82.25 mg/g dry biomass and K value was 1.49 and 0.45 mg/l for 30°C and10°C respectively. Scanning electron micrograph (Fig. 5) shows cells of P. fluorescens BM07 grown to late exponential phase at 10°C under condition of induced extracellular biopolymer production with a uniquely structured sheath surrounding them. Conclusions. It is very important to study the metal-removing characteristics of biomass to identify possible individual differences and exploit them. Therefore we aimed to investigate the biosorption characteristic of P. fluorescens BM07 strain in the removal of mercury ions with emphasis on its unique ability of producing biopolymer which is thought to affect its sorption abilities. P. fluorescens BM07 were found to be efficient for adsorption of mercury from solutions. The characterization of mercury uptake

Fig. 4. Langmuir adsorption isotherm for mercury biosorption by living BM07 cells at 10°C (A) and 30°C (B).

116

2

Noghabi K.A. et al.

Fig. 5. Scanning electron micrograph of P. fluorescens BM07 cells grown in fructose-containing M1 minimal medium for 144 h (late exponential phase) at 10°C under condition of induced extracellular biopolymer production. Micrographs showing groups of cells with a uniquely structured sheath surrounding them. Magnification 40 K, Bar 500.

showed that, mercury biosorption is dependent on initial pH, initial mercury ion concentration and biomass quantity. The Langmuir model of mercury biosorption by P. fluorescens BM07 was successfully applied to describe the biosorption equilibrium. Under similar conditions mercury adsorption rate by cells grown at 10°C was higher than those which were grown at 30°C grown cells. Considering that the biopolymer is synthesized at 10°C and with respect to the presence of a variety of many functional groups such as carboxyl, amine, hydroxyl, phosphate and sulfhydryl groups in the biopolymer, which potentially are capable of ion exchanging with metal cations, it seems reasonable to ascribe the elevated mercury adsorption to the production of the biopolymer. These metal ions may be adsorbed with negatively charged reaction sites on the cell surface (Beveridge and Murray, 1980; Gadd, 1990; Gupta et al., 2000). On the other hand, BATH test data showed that the surface of cells grown at 10oC was more hydrophobic than the surface of cells grown at 10oC (unpublished data). It might be due to the release of cold-induced biopolymer to out of the cells. Furthermore the role of van der Waals interaction between the cations and hydrophobic biopolymer can not be ignored (Cotton and Wilkinson, 1972; Frausto da Silva and Williams, 1991). Microorganisms generally produce an extracellular network of polysaccharides and proteins, such as capsules, slime, and more-structured sheaths. A coating of extracellular biopolymer may provide some capacity to adsorb cat-

ionic metal species, and its ability to sequester such cations will be provided by the cell wall, in which case the cationic species must be able to migrate through the sheath. The sheaths permeability is sufficient to allow a flow of metal ions. The existence of residual biopolymer on the external surface of BM07 cells grown at 10°C may be a pivotal factor to augment the mercury biosorption rate in comparison with 30°C grown cells. Considering all of the parameters, it seems likely that the cold-induced biopolymer production plays an important role in biosorption efficiency, as P. fluorescens BM07 cells which were grown at 10°C under similar condition showed higher efficiency to biosorbe mercury than non-polymer producing cells grown at 30°C. Further investigations are required to reveal how the dual-layered distribution of reactive sites of BM07 cells grown at 10oC will affect its ability to adsorb mercury more efficiently than non-polymer producing cells grown at 30°C. Literature Beveridge T.J. and R.G.E. Murray. 1980. Sites of metal deposition in the cell walls of Bacillus subtilis. J. Bacteriol. 141: 876887 Chang J.S. and J. Hong. 1994. Biosorption of mercury by the inactivated cells of Pseudomonas aeruginosa PU21. Biotechnol. Bioeng. 44: 9991006. Choi M.H. and S.C.Yoon. 1994. Polyester biosynthesis characteristics of Pseudomonas citronellolis grown on various carbon sources, including 3-methyl-branched substrates. Appl Environ. Microbiol. 60: 32453254.

2


Cotton F.A. and G. Wilkinson. 1972. Advanced Inorganic Chemistry. John Wiley & Sons, London, pp. 504527. de Rome L. and G.M. Gadd. 1991. Use of pelleted and immobilized yeast and fungal biomass for heavy metal and radionuclide recovery. J. Ind. Microbiol. 7: 97104. Friedman M. and A. Waiss. 1972. Mercury uptake by selected agricultural products and by-products. Environ. Sci. Technol. 6: 457458. Frausto da Silva J.J.R. and R.J.P. Williams. 1991. The Biological Chemistry of the Elements: the Inorganic Chemistry of Life. Clarendon Press, Oxford, pp 531552 Gadd G.M. 1990. Heavy metal accumulation by bacteria and other microorganisms. Experientia 46: 834840. Grau J.M. and J.M. Bisang. 1995. Removal and recovery of mercury from chloride solutions by contact deposition on iron felt. J. Chem. Tech. Biotechnol. 62: 153158. Gray N.F. 1999. Water Technology. John Wiley & Sons, New York, pp.473474. Gupta R., P. Ahuja, S. Khan, R.K. Saxena and M. Mohapatra. 2000. Microbial biosorbents: meetings challenges of heavy metals pollution in aqueous solution. Curr. Sci. 78: 967973. Habashi, F. 1978. Metallurgical plants: how mercury pollution is abated. Environ. Sci. Technol.12: 13721376.

117

Hughes, M.N and R.K. Poole. 1989. Removal or recovery of metal ions and compounds from solution by microbiological methods. In: Metals and Microorganisms. London: Chapman and Hall p. 328. Lee H.J., M.H. Choi, T.U. Kim, and S.C. Yoon. 2001. Accumulation of polyhydroxyalkanoic acid containing large amounts of unsaturated monomers in Pseudomonas fluorescens BM07 utilizing saccharides and its inhibition by 2-bromooctanoic acid. Appl. Environ. Microbiol. 67: 49634974. Palleroni, N.J. 1989. Pseudomonadaceae. In Bergeys Manual of Systematic Bacteriology. Kreig NR and Holt JG (eds). Baltimore: The Williams and Wilkins Co., pp.141199. Peniche-Covas, C., L.W. Alvarez and W. Argüles-Monal. 1992. The adsorption of mercuric ions by chitosan. J. Appl. Polym. Sci. 46: 11471150. Tiemann, K.J., J.L. Gardea-Torresdey, G. Gamez, K. Dokken and S. Sias. 1999. Use of X-ray absorption spectroscopy and esterification to investigate chromium (III) and nickel (II) ligand in alfalfa biomass. Environ. Sci. Technol. 33:150154. Volesky, B., and Z. Holan. 1995. Biosorption of heavy metals. Biotechnol. Prog. 11: 235250. Volesky, B. 1990. Biosorption of Heavy Metals. CRC Press, Boca Raton, FL.

118

Noghabi K.A. et al.

2


Molecular Analysis of Temporal Changes of a Bacterial Community Structure in Activated Sludge Using Denaturing Gradient Gel Electrophoresis (DGGE) and Fluorescent in situ Hybridization (FISH) ALEKSANDRA ZIEMBIÑSKA1*, ANNA RASZKA1, JAAK TRUU2, JOANNA SURMACZ-GÓRSKA1 and KORNELIUSZ MIKSCH1 1 Environmental

Biotechnology Department, Faculty of Power and Environmental Engineering, The Silesian University of Technology, Poland; 2 Institute of Molecular and Cell Biology, Faculty of Biology and Geography, Tartu University, Tartu, Estonia Received 25 September 2006, revised 30 January 2007, accepted 14 February 2007 Abstract Wastewater treatment based on activated sludge is known to be one of the most effective and popular wastewater purification methods. An estimation of microbial community variability in activated sludge allows us to observe the correlation between a particular bacterial groups appearance and the effectiveness of the removal of chemical substances. This research is focused on microbial community temporal changes in membrane bioreactors treating wastes containing a high level of ammonia nitrogen. Samples for this study were collected from two membrane bioreactors with an activated sludge age of 12 and 32 days, respectively. The activated sludge microbial community was adapted for the removal of ammonia nitrogen up to a level of 0.3 g NH 4+ N g/VSS/d (VSS volatile suspended solids). The methods denaturing gradient gel electrophoresis (DGGE) based on 16S rRNA gene PCR products and fluorescent in situ hybridization (FISH) with 16S rRNA gene probes revealed significant differences in the microbial community structure in the two bioreactors, caused mainly by a difference in sludge age. According to the results obtained in this study, a bioreactor with a sludge age of 12 days is characterized by a much higher microbial community diversity than a bioreactor with a sludge age of 32 days. Interestingly, the appearance of particular species of nitrifying bacteria was constant throughout the experiment in both bioreactors. Changes occured only in the case of the Nitrosomonas oligotropha lineage bacteria. This study demonstrates that the bacterial community of bioreactors operating with different sludge ages differs in total community structure. Nevertheless, the changeability of the bacterial community structure did not have any influence on the efficiency of nitrification. K e y w o r d s: activated sludge, ammonia nitrogen removal effectiveness, PCR-DGGE, FISH

Introduction Activated sludge, as a mixture of microorganisms, is an excellent research material both for microbiology and technology, which allows researchers to find a method to effectively utilize different chemical substances. In the case of wastewater treatment, the removal of nitrogen compounds is one of the priorities. Despite the fact that studies in the field of deammonification as a process connected with nitrogen removal have recently expanded, nitrification is still the most commonly examined and used process. Nitrification consists of two steps: nitritation (biological oxidation of ammonia) and nitratation (biological oxidation of nitrite). Ammonia oxidizing bacteria (AOB)

and nitrite oxidizing bacteria (NOB) respectively carry out these steps. Many studies of nitrifying bacteria revealed that Nitrosomonas sp. and Nitrobacter sp. are the most common bacteria encountered in wastewater systems (Chain et al., 2003; Duddleston et al., 2000; Hommes et al., 2001; Kelly et al., 2005). Nowadays, due to the common use of molecular methods, it is known that the nitrifiers group in activated sludge is much more diverse. Traditional cultivation techniques underestimate the number and the diversity of the nitrifiers groups, which are known to be fastidious and slow growing in the laboratory (Amman and Kuhl, 1998; Luxmy et al., 2000). Research based on cultivation-independent methods enabled researchers to extend the knowledge about ammonia and nitrite oxidizers.

* Corresponding author: A. Ziembiñska, Environmental Biotechnology Department, Faculty of Power and Environmental Engineering, The Silesian University of Technology, Akademicka 2A, 44-100 Gliwice, Poland; phone: +48 32 2371717; fax: +48 32 2372946; e-mail: [email protected]

120

2

Ziembiñska A. et al.

Descriptions and comparisons of activated sludge bacterial communities have been carried out since the early 90s. Molecular methods based on PCR reaction were introduced due to their sensitivity and the possibility of avoiding difficulties attributable to pure culture obtainment (Blackall et al., 1997; Snaidr et al., 1997; Juretschko et al., 1998). DNA isolation directly from bacterial cells in a bacterial community accelerates the identification procedures, assessment of composition and changeability of individual microbial groups without previous pure culture isolation (Holben et al., 1998; Kowalchuk et al., 1997). Currently, the basic tools used in a comparative analysis of bacterial communities without previous cultivation are DGGE (Denaturing Gradient Gel Electrophoresis) and FISH (Fluorescent in situ Hybridization). DGGE is based on the separation of DNA fragments of the same length, but with a different nucleotide sequence. The use of increasing gradient of denaturing factors (formamide and urea) makes the obtainment of detailed characteristics for the sample fingerprint possible (Curtis and Craine, 1998; van de Gast et al., 2006). FISH uses fluorescently labeled oligonucleotides to study the presence and estimate the abundance of particular bacteria (Schuppler et al., 1998). In addition to the difficulties in the pure culture isolation of nitrifiers, the most problematic issues in studies of this group of bacteria in laboratory-scale experiments are a low bacterial growth rate and sensitivity to toxic shocks, such as pH and temperature swings (Luxmy et al., 2000; Rowan et al., 2003). Due to low bacterial growth (long doubling time), nitrifiers are more sensitive to washing out from the wastewater treatment plant than heterotrophic bacteria, while the system is operated on short sludge retention times. Sludge retention time (SRT), or sludge age, is a technological parameter determining the time during which the activated sludge stays in the reactor. SRT is calculated on the basis of the amount of suspended solids in the reactor and the amount of activated sludge removed from the reactor as excessive sludge and the amount of suspended solids in the effluent (Metcalf and Eddy, 1991). Membrane bioreactors were used in this study. Such reactors have become very commonly used due to their advantages over settler-operated systems. Membrane reactors possess a filtration module (membrane) instead of a settler. The membranes are able to retain particles of different sizes, depending on the membrane type within the reactor (Bodzek et al., 1997; Charcosset, 2006). Application of membranes in activated sludge systems enables a much more effective separation of solids from wastewater and can protect wastewater treatment plant from the problems connected with bulking. Membranes can be installed inside (submerged membrane reactors) or outside of

the reactor. A membrane located inside the reactor reduces the space requirements and expenses connected with activated sludge pumping (Charcosset, 2006). Due to the lack of solids in the effluent in membrane reactors, sludge retention time would be the only crucial parameter responsible for bacterial withdrawal, because activated sludge flocs are removed from the system only during sludge age regulation. Thus, sludge age is an important parameter influencing the community structure because of the differences in the nitrifier species growth time. The aim of the study was to compare activated sludge communities in two bioreactors containing activated sludge adapted to different sludge ages and to observe temporal changes in the bacterial communities. Such analysis was undertaken in order to check whether the communities performing a stable process of nitrification show any changeability on the bacterial level. An analysis performed to estimate the abundance of the main species and the structure of the bacterial communities pointed mainly to ammonia and nitrite-oxidizing bacteria. Experimental Materials and Methods

Reactor details and operational data. Bacteriological material from two completely mixed, laboratory-scale membrane bioreactors was used in this study. The membrane (pore size of 0.4 µm) was submerged in the reactor (Fig. 1). Activated sludge from a municipal wastewater treatment plant performing nutrient removal was used for seeding. The reactors were operated on with nitrification and fed with a synthetic medium containing high ammonia concentrainfluent

effluent

filtration module membrane bioreactor

aeration Fig. 1. The scheme of the membrane bioreactor.

2

Molecular analysis of bacterial community structure in activated sludge

tions. The wastewater was composed of 500700 mg NH4+ N/l, 250 mg COD/l (COD chemical oxygen demand) coming from CH3COONa and broth extract and an amount of phosphorus (Na2HPO4) enabling 100:1 C:P ratio obtainment. The pH was maintained at a level of 7.58.0 using NaHCO3. Sludge age was calculated typically for activated sludge systems as follows: SRT = Vr × X / QW × XW Since in the MBR reactor: XW = X than: SRT = Vr / QW where: Vr reactor volume, l; Q W amount of activated sludge removed daily from the reactor, l/d; X biomass concentration within the reactor, g VSS/l; XW biomass contained in the waste sludge, g VSS/l. In order to obtain a particular sludge age, a constant and appropriate (QN = Vr /SRT) volume of activated sludge was removed daily from each reactor (Sponza, 2002; Sponza, 2003). The performance of the reactors was monitored by analysis of the influent and effluent. Nitrogen compound concentrations were determined colorimetrically ammonia with Nessler reagent according to PN-C-04576-4:1994, nitrite with alfanaftyloamine reagent according to Hermanowicz and Dojlido (1999) and nitrate with dimethylphenol reagent according to ISO 7890-1. Due to daily sludge removal (in order to obtain a particular sludge age) and activated sludge suspended solids variations during the adaptation phase of the experiment, the substrate load was 0.3 g NH4+ N g/VSS/d for reactor A and 0.24 g NH4+ N g/VSS/d for reactor B (all the technological parameters of the investigated bioreactors are shown in Table I). The experiment was carried out for 8 months. Activated sludge samples were taken from the bioreactors within a period of 4 months each 2 weeks, Table I Technical parameters of two membrane bioreactors used in the study Technical parameter Bioreactor volume, l Sludge age, days Activated sludge removed daily for sludge age control, l Substrate load, g NH 4+ N g/VSS/d Volatile suspended solids, g VSS/l Flow speed, l/d

Bioreactor A Bioreactor B 25 12

36 32

2.08

1.13

0.3 0.6 9

0.24 1 9

after 4 months of adaptation to the high load of ammonia and upon achieving a particular sludge age. Activated sludge samples for DGGE. Activated sludge samples 1A and 2A-6B (volume of 10 ml) were collected from both bioreactors at 2-week intervals (samples 6A and 6B were obtained at the 4 week

121

mark), pelleted by centrifugation (10 000× g, 10 min, 4°C) and stored at 20°C. DNA extraction and PCR conditions. Total genomic DNA was extracted from 0.3 g of the activated sludge samples using an Ultra Clean Soil Isolation DNA Kit (MoBio Laboratories Inc., USA) according to the manufacturers instructions and stored at 20°C until PCR amplification. Primers: 968F with a GC clamp (5 CGC CCG GGG CGC GCC CCG GGC GGG GCG GGG GCA CGG GGG GAA CGC GAA GAA CCT TAC 3) and 1401R (5 CGG TGT GTA CAA GGC CC 3) were used for partial 16S rRNA bacterial gene PCR amplification (Felske et al., 1996). PCR was carried out in a 25 µl (total volume) reaction mixture containing 17 µl sterile MiliQ water, 2.5 µl 10× PCR buffer containing (NH4)2SO4, 2.5 µl MgCl2 (2 mM), 0.5 µl BSA (Bovine Serum Albumin, 3 mg/µl), 0.5 µl of both primers (20 pmol), 0.5 µl dNTPs (2.5 mM), 0.5 µl of genomic DNA and 0.5 µl Taq DNA polymerase (1U). All of the components were delivered by Fermentas, Lithuania. PCR amplification was performed using an Eppendorf thermal cycler and the following steps: (1) the initial denaturation step (5 min at 94°C); (2) 30 cycles, each single cycle consisting of denaturation (1 min at 94°C), annealing (1 min at 53°C), and elongation (1 min at 72°C); and (3) the final extension step (10 min at 72°C). Products were evaluated in agarose gel (0.8% w/vol agarose, 1× TAE buffer), stained with ethidium bromide (1% w/vol) in MiliQ water and photographed under UV light. DGGE denaturing gradient gel electrophoresis. The DGGE of PCR products obtained in reaction with 968F-GC and 1401R primers were performed using the Dcode Universal Mutation Detection System (BioRad). Polyacrylamide gel (6%, 37:1 acrylamidebisacrylamide) with a gradient of 2850% denaturant was prepared with a gradient former (Amersham Bioscience) according to the manufacturers guidelines. The gel was run for 11 h at 80 V in a 1× TAE buffer at a constant temperature of 60°C. The gel was stained with ethidium bromide (1% w/vol) in MiliQ water for 20 min and washed in MiliQ water twice for 15 min, then visualized under UV light and photographed. Numerical analysis of the DGGE fingerprints. The DGGE banding patterns with 16S rDNA PCR products were analyzed using GelCompar II software (Applied Maths, Ghent, Belgium) in order to compare the fingerprint patterns obtained from the separation of the PCR products from samples 1A-6B. The clustering was done using the Pearson correlation coefficient and the UPGMA method (Unweighted Pair Group Method with Arithmetic mean). The principle of moving window analysis was used in order to evaluate the stability of the bacterial community (Possemiers et al., 2004; Wittebolle et al., 2005).

122

2

Ziembiñska A. et al. Table II rRNA-targeted oligonucleotide probes used in the study Probe

Nso 1225 NEU CTE Cluster 6a192 Comp-Cluster 6a192 Ntspa-662 Comp-Ntspa662 NIT3 Comp NIT3 Nsv443 EUB338 EUB338 II EUB338 III a b c d

Target organisms

Sequence (5 3)

B-proteobacterial ammoniaCGC CAT TGT ATT ACG TGT GA oxidizing bacteria most halophilic and CCC CTC TGC TGC ACT CTA halotolerant Nitrosomonas sp. a TTC CAT CCC CCT CTG CCG Nitrosomonas oligotropha CTT TCG ATC CCC TAC TTT CC lineage

Target site

% formamide

Reference

1224 1243

35

Mobarry et al., 1996

653 670

40

Wagner et al., 1995

653 670

a

192 211

35

Wagner et al., 1995 Adamczyk et al., 2003 Adamczyk et al., 2003 Daims et al., 2001a Daims et al., 2001a Wagner et al., 1996 Wagner et al., 1996 Mobarry et al., 1996 Amann et al., 1990 Daims et al., 1999 Daims et al., 1999

b

CTT TCG ATC CCC TGC TTC C

192 211

b

Nitrospira sp.

GGA ATT CCG CGC TCC TCT GGA ATT CCG CTC TCC TCT CCT GTG CTC CAT GCT CCG CCT GTG CTC CAG GCT CCG CCG TGA CCG TTT CGT TCC G GCT GCC TCC CGT AGG AGT GCA GCC ACC CGT AGG TGT GCT GCC ACC CGT AGG TGT

662 679 662679 1035 1052 1035 1052 444 462 338 355 338 355 338 355

35 c 40 d 30 35 35 35

c

Nitrobacter sp. d

Nitrosospira sp. most bacteria Planctomycetales Verrucomicrobiales

used as unlabeled competitor together with probe S-*-Nsm-0651-a-A-18 used as unlabeled competitor together with probe Cluster 6a192 used as unlabeled competitor together with probe S-G-Ntspa-662-a-A-18 used as unlabeled competitor together with probe S-G-Nbac-1035-a-A-18

The structural diversity of the bacterial community was estimated on the basis of the Shannon-Weaver diversity index, H (Eichner et al., 1999; Nübel et al., 1999; Luxmy et al., 2000), estimated on the relative band intensities obtained from the DGGE fingerprints. FISH Fluorescent in situ hybridization: sample preparation, oligonucleotide probes, confocal microscopy and cell quantification. Activated sludge samples were fixed with a paraformaldehyde solution (4% paraformaldehyde in phosphate-buffered saline, PBS, pH 7.2) at 4°C for 3 hours and subsequently washed in PBS. Fixed samples were stored in PBS: ethanol (1:1) solution at 20°C. In situ hybridization was performed as described previously by Daims (Daims et al., 2005). 16S rRNA targeted fluorescently labeled oligonucleotide probes, the sequences and targeted sites are listed in Table II. The probes EUB338, EUB338 II and EUB338 III were mixed together (EUB338 mix) in the proportion 1:1:1 in order to detect all bacteria. Details on the chosen oligonucleotide probes are available at probeBase (Loy et al., 2003). The probes were 5 labeled with the dye FLUOS (5(6)-carboxyfluorescein-N-hydroxysuccinimide ester), Cy3 or Cy5. Both the probes and unlabeled competitor oligonucleotides were obtained from Biomers, Ulm, Germany. Prior to microscope observations, samples were embedded in Citifluor (Citifluor Ltd, UK) to reduce fluorochrome fading. A scanning confocal micro-

scope (Zeiss LSM 510) equipped with an Ar-ion laser (488 nm) and two HeNeLasers (543 nm and 633 nm) was used to examine the microbial community. Image processing was performed using the standard software package delivered with the instrument (Zeiss LSM version 3.95). For cell quantification DAIME software was used. Results and Discussion Ammonia oxidation effectiveness in the experiment. The measurements of the ammonia, nitrite and nitrate in bioreactors A and B pointed to a high nitrification efficiency (Fig. 2). In both bioreactors the effectiveness of ammonia removal was maintained at a level of 99%. The effluent quality confirmed full nitrification obtainment. Such nitrification efficiency suggests that the bacterial community contains a considerable amount of ammonia and nitrite-oxidizing bacteria. DGGE analysis of the bacterial community. The fingerprints obtained from DGGE separation of 16S rRNA gene fragments are shown in Figure 3. The DNA-based DGGE pattern had changed in both bioreactors during the experiment. The dendrogram (Fig. 4) obtained from DGGE fingerprint analysis presents the value of similarity among activated sludge samples. This could be calculated by summarizing the length of branches connecting particular samples of the dendro-

2

123


300

700

250

600 500

200 400 150 300

N-NO3, mg/l

ammonia load, mg NH4+-N/gd; ammonia removal efficiency, % N-NH4+-/NO2-N, mg/l

reactor B 32d

100 200 50

100

0

0 1

2

3

4

5

sampling time amonia load N-NH4+-N in bioreactor effluent ammonia removal effeciency

N-NO3-N in bioreactor effluent N-NO2-N in bioreactor effluent

600

300

500

250

400

200 300 150

N-NO3, mg/l

ammonia load, mg NH4+-N/gd; ammonia removal efficiency, % N-NH4+-/NO2-N, mg/l

reactor A 12d 350

200

100

100

50 0

0 1

2

3

4

5

sampling time amonia load N-NH4+-N in bioreactor effluent ammonia removal effeciency

N-NO3-N in bioreactor effluent N-NO2-N in bioreactor effluent

Fig. 2. The temporal dynamics of ammonia-nitrogen removal from bioreactors A and B.

gram. The lower this value is, the higher the similarity the samples present. This analysis revealed higher similarity values among samples collected from bioreactor B (shorter dendrogram branches connecting particular samples), which suggests a lower temporal variation in this community. Nevertheless, sample A6 seems to be more congruent to the bioreactor B cluster than to its own group. There is a possibility that at the end of the experiment, the bioreactor A community reached a level of homogeneity, which characterized the older sludge from bioreactor B.

The variability of the bacterial community according to the results obtained in Shannon-Weaver index estimations is higher in bioreactor A (Fig. 5). The diversity of the bacterial community changed slightly during the experiment. Interestingly, in the case of bioreactor A, the diversity increases (in the period of sampling time for samples 2A-3A) and then decreases after the third measurement (samples 3A-5A), while the diversity of bacteria decreases constantly at a similar rate in bioreactor B (between measurements for samples 2B-5B). The reason for such situation

124


2

1A 2A 2B 3A 3B 4A 4B 5A 5B 6A 6B

Fig. 3. DGGE pattern of 16S rRNA genes fragments with a size ca. 500 bp amplified using DNA obtained from activated sludge samples.

could be the difference in sludge age. In order to maintain the shorter age of the activated sludge, a larger volume of the sludge was removed from the bioreactor. In this situation, the ammonia and nitrite-oxidizers community is removed from the environment of bioreactor A faster, while the bacteria in bioreactor B have a longer time to multiply and change. Results obtained from both bacterial communities underwent the moving window analysis (Fig. 6) in order to estimate bacterial community stability (Wittebolle et al., 2005). This analysis revealed that differences in sludge age could be the cause of the dissimilarities

in the value of correlation coefficient. The higher the sludge age is, the higher the coefficient of correlation. The results obtained correlate with the performance of the bioreactors leading nitrification with high efficiency and stability. The analysis proved that bioreactors achieved stability by the first experimental sampling time, but in the case of bioreactor A, the microbial community steadily changed until the fourth measurement. Diversity of nitrifying bacteria analyzed by fluorescent in situ hybridization (FISH). The appearance of the ammonia and nitrite oxidizers was identified

Fig. 4. DGGE tree based on Pearson correlation coefficient and UPGMA (Unweighted Pair Group Method with Arithmetic mean) clustering method

2

125

Molecular analysis of bacterial community structure in activated sludge Shannon-Weaver diversity values

diversity index H

4 3 bioreactor A

2

bioreactor B

1 0 2

3

4

5

6

activated sludge samples Fig. 5. Temporal dynamics of the Shannon-Weaver diversity values (H) for activated sludge samples (activated sludge samples from 2A to 6B) Moving window analysis

correlation coefficient (%)

120 100 80 bioreactor A bioreactor B

60 40 20 0 2

3

4

5

6

activated sludge samles Fig. 6. Moving window analysis of DGGE fingerprints of sludge samples from bioreactors A and B (2A-6B) presents the difference of the correlation coefficient due to the differences in the sludge age

using molecular probes characteristic for " and $ proteobacterial representatives of these functional groups as well as for the Nitrospira sp. group. The EUB mix probe detects most of the bacteria and was used as a control. Tables III and IV gather the results of the FISH investigation for the groups: Nitrobacter sp., Nitrospira sp., halophilic and halotolerant Nitrosomonas sp. Bacteria belonging to these groups appear in both bioreactors, while no representatives of Nitrosospira sp. group were found. The only difference is

the presence of Nitrosomonas oligotropha lineage members in bioreactor A during the time of the experiment, while in bioreactor B such bacteria appear at the end of the experimental period. The obtained results confirm previous studies where Nitrosomonas sp. appearance was usually exhibited in the engineered high ammonia environments (Juretschko et al., 1998). Interestingly, these bacteria were absent in bioreactor B, which had activated sludge age of 32-days, during most of the experiment, which could mean that the

Table III Structure of nitrifiers community in the activated sludge of reactor A (obtained by FISH analysis) Activated Nitrosomonas Most halophilic Nitrosospira Nitrobacter sludge oligotropha and halotolerant sp. sp. sample lineage (clad) Nitrosomonas sp.

Nitrospira sp.

Amount of nitrifiers* (%)

2A

+

+

+

+

46.9

3A

+

+

+

+

37.9

4A

+

+

+

+

36.7

* amount of nitrifiers present in activated sludge as a percent of the total amount of bacteria

126

2

Ziembiñska A. et al. Table IV Structure of nitrifiers community in the activated sludge of reactor B (obtained by FISH analysis) Activated Nitrosomonas Most halophilic Nitrosospira Nitrobacter sludge oligotropha and halotolerant sp. sp. sample lineage Nitrosomonas sp. 2B

Nitrospira sp.

Amount of nitrifiers* (%)

+

+

+

64.6

3B

+

+

+

60.8

4B

+

+

+

+

48.6

* amount of nitrifiers present in activated sludge as a percent of the total amount of bacteria

presence of Nitrosomonas sp. in a well-working engineered system is not obligatory. The level of ammonia can be a relevant factor of selection in bacterial appearance in the environment (Pommerening-Röser et al., 1996). It is important to note that lab-scale artificial environments are known to abound in one group of ammonia oxidizers such as Nitrosomonas-like, Nitrosospira-like bacteria or a mixture of these groups (Schramm et al., 1998). In this case no representatives of Nitrosospira sp. were found in either bioreactor. This could suggest the existence of an unknown environmental factor that eliminates these bacteria from the environment and disturbs the disposition of the other bacterial groups. We might suspect that bioreactor conditions are not suitable for the desired bacterias adaptation. Nevertheless, it was previously shown (Daims et al., 2001b) that a large diversity of nitrifying bacteria appear in spite of a high ammonia and salt concentration known to be extreme conditions. It was also noted previously that different wastewater systems support different groups of bacteria as well as population richness (Rowan et al., 2003). The total amount of nitrifiers decreased during the experiment in both cases. However, the rate of the decline is slightly higher in the bioreactor with a shorter activated sludge age (bioreactor A) which could be caused by the larger volume of sludge removed from the bioreactor in order to maintain a particular sludge age. There is also a possibility that some groups of nitrifiers can adapt better to the environment than others. It could mean that the nitrifiers remaining in the bioreactor adapted to the environment gaining higher efficiency of the process. In such a case, the effectiveness of nitrification is maintained at a high level even in a situation where a high amount of bacteria is removed from the bioreactor. Conclusions. This study found that two lab-scale environments of bioreactors dealing with the same wastewater vary in the homogeneity of the bacterial groups and that bioreactor A (sludge age of 12 days) is characterized by a much higher diversity of the genotypes than bioreactor B (sludge age of 32 days). Such differences probably occur due to the differences in sludge age. There were no disturbances of

nitrification during the experiment and the efficiency of the process in both environments was very high. The total amount of nitrifiers decreased in both bioreactors during the experiment, but changes in the appearance of particular groups occurred only in the case of the Nitrosomonas oligotropha lineage bacteria. Bacteria belonging to this group appear in bioreactor B in the final stage of the experiment, while in bioreactor A they are constantly present. It can be concluded that the effectiveness of the bioreactors performance does not depend on the level of the total bacterial diversity or the presence of a particular species. Acknowledgements The research was supported by the Centre of Excellence Environmental Biotechnology Research Centre DEMETER (contract number: EVK1-CT-2002-80009). We are grateful to the authorities of the Department of Genetics, Institute of Molecular and Cell Biology, Tartu University, Estonia and the Department of Microbial Ecology, the Vienna Ecology Centre, University of Vienna, Austria for access to laboratory equipment and also to the laboratory staff for their excellent technical assistance. We would like to thank Kilian Stoecker and Holger Daims from the Vienna Ecology Centre for their essential technical help.

Literature Adamczyk J., M. Hesselsoe, N. Iversen, M. Horn, A. Lehner, P.H. Nielsen, M. Schloter, P. Roslev. and M. Wagner. 2003. The isotope array, a new tool that employs substrate-mediated labeling of rRNA for determination of microbial community structure and function. Appl. Environ. Microbiol. 69: 68756887. Amann R.I., B.J. Binder, R.J. Olson, S.W. Chisholm, R. Devereux and D.A. Stahl. 1990. Combination of 16S rRNAtargeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol. 56: 19191925. Amann R. and M. Kuhl. 1998. In situ methods for assessment of microorganisms and their activities. Curr Opin Microbiol. 1: 352358. Blackall L.L., S. Rossetti, C. Christenson, M. Cunningham, P. Hartman, P. Hugenholtz and V. Tandoi. 1997. The characterization and description of representatives of G bacteria from activated sludge plants. Lett. Appl. Microbiol. 25: 6369. Bodzek M., J. Bohdziewicz and K. Konieczny. 1997. Membrane techniques in the environmental protection (in Polish). Wydawnictwo Politechniki l¹skiej, Gliwice.

2


Chain P., J. Lamerdin, F. Larimer, W. Regala, V. Lao V., M. Land, L. Hauser, A. Hooper, M. Klotz, J. Norton, L. Sayavedra-Soto, D. Arciero, N. Hommes, M. Whittaker and D. Arp. 2003. Complete genome sequence of the ammonia-oxidizing bacterium and obligate chemolithoautotroph Nitrosomonas europaea. J. Bacteriol. 85: 27592773. Charcosset C. 2006. Membrane process in biotechnology: an overview. Biotechnol. Adv. 24: 482492. Curtis T.P. and N.G. Craine. 1998. The comparison of the diversity of activated sludge plants. Wat. Sci. Technol. 37: 7178. Daims H., J.L. Nielsen, P.H. Nielsen, K.H. Schleifer and M. Wagner. 2001a. In situ characterization of Nitrospira-like nitriteoxidizing bacteria active in wastewater treatment plants. Appl. Environ. Microbiol. 67: 52735284. Daims H., A. Brühl, R. Amann, K.H. Schleife and M. Wagner. 1999. The domain-specific probe EUB338 is insufficient for the detection of all Bacteria: Development and evaluation of a more comprehensive probe set. Syst. Appl. Microbiol. 22: 434444. Daims H., U. Purkhold, L. Bjerrum, E. Arnold, P.A. Wilderer. and M. Wagner. 2001b. Nitrification in sequencing batch biofilm reactors: lessons from molecular approach. Wat. Sci. Technol. 43: 918. Daims H., K. Stoecker. and M. Wagner. 2005. Fluorescence in situ hybridization for the detection of prokaryotes, pp: 213239. In: Osborn A.M. and C.J. Smith (eds), Advanced Methods in Molecular Microbial ecology. Bios-Garland. Abingdon, UK. Duddleston K.N., P.J. Bottomley, A. Porter and D.J. Arp. 2000. Effects of soil and water content on methyl bromide oxidation by the ammonia-oxidizing bacterium Nitrosomonas europaea. Appl. Environ. Microbiol. 66: 26362640. Eichner C.A., R.W. Erb, K.N. Timmis. and I. Wagner-Döbler. 1999. Thermal gradient gel electrophoresis analysis of bioprotection from pollutant shock in the activated sludge microbial community. Appl. Environ. Microbiol. 65: 102109. Felske A., B. Engelen., U. Nübel and H. Backhaus. 1996. Direct ribosome isolation from soil to extract bacterial rRNA for community analysis. Appl. Environ. Microbiol. 62: 41624167. Hermanowicz W. and J. Dojlido. 1999. Physical and Chemical Methods in Water and Wastewater Analysis (in Polish). Arkady, Warszawa. Holben W.E, K. Noto, T. Sumino. and Y. Suwa. 1998. Molecular analysis of bacterial communities in a three-compartment granular activated sludge system indicates community-level control by compatible nitrification processes. Appl. Environ. Microbiol. 64: 25282532. Hommes N.G., L.A. Sayavedra-Soto and D.J. Arp. 2001. Transcript analysis of multiple copies of amo (encoding ammonia monooxygenase) and hao (encoding hydroxylamine oxidoreductase) in Nitrosomonas europaea. FEMS Microbiol. Ecol. 43: 195206. ISO regulation: ISO 78901. Juretschko S., G. Timmermann, M. Schmid, K.H. Schleifer, A. Pommerening-Röser, H.P. Koops and M. Wagner. 1998. Combined molecular and conventional analyses of nitrifying bacterium diversity in activated sludge: Nitrosococcus mobilis and Nitrospira-like bacteria as dominant populations. Appl Environ Microbiol. 64: 30423051. Kelly J.J., S. Siripong., J. McCormack, L.R. Janus, H. Urakawa, S. El Fantroussi, P.A. Noble, L. Sappelsa., B.E. Rittmann and D.A. Stahl. 2005. DNA microarray detection of nitrifying bacterial 16S rRNA in wastewater treatment plant samples. Water Res. 39: 32293238. Kowalchuk G.A., J.R. Stephen, W. De Boer, J.I. Prosser, T.M. Embley. and J.W. Woldendorp. 1997. Analysis of ammonia-oxidizing bacteria of the $ subdivision of the class Proteobacteria in coastal sand dunes by denaturing gradient gel electro-

127

phoresis and sequencing of PCR-amplified 16S ribosomal DNA fragments. Appl. Environ. Microbiol. 63: 14891497. Loy A., M. Horn and M. Wagner. 2003. ProbeBase an online resource for rRNA-targeted oligonucleotide probes. Nucleic Acids Res. 31: 514516. Luxmy B.S., F. Nakajima and K. Yamamoto. 2000. Analysis of bacterial community in membrane-separation bioreactors by fluorescent in situ hybridization (FISH) and denaturing gradient gel electrophoresis (DGGE) techniques. Wat. Sci. Technol. 41: 259268. Metcalf and Eddy inc. 1991. Wastewater Engineering: Treatment, Disposal and Reuse, 3rd ed. McGrow Hill Inc., New York. Mobarry B.K., M. Wagner, V. Urbain, B.E. Rittmann and D.A. Stahl. 1996. Phylogenetic probes for analyzing abundance and spatial organization of nitrifying bacteria. Appl. Environ. Microbiol. 62: 21562162. Nübel U., F. Garcia-Pichel, M. Kuhl and G. Muyzer. 1999. Quantify microbial diversity: morphotypes, 16S rRNA genes and carotenoids of oxygenic phototrophs in microbial mats, Appl. Environ. Microbiol. 62: 422430. Polish regulation: PN-C-04576-4:1994. Pommerening-Röser A., G. Rath and H.P. Koops. 1996. Phylogenetic diversity within the genus Nitrosomonas. Syst. Appl. Microbiol. 19: 344351. Possemiers S.,K. Verthe, S. Uyttendaele and W. Verstraete. 2004. PCR-DGGE-based quantification of stability of the microbial community in a simulator of the human intestinal microbial ecosystem. FEMS Microbiol. Ecol. 49: 495507. Rowan A.K., J.R. Snape, D. Fearnside, M.R. Barer, T.P. Curtis and I.M. Head. 2003. Composition and diversity of ammoniaoxidising bacterial communities in wastewater treatment reactors of different design treating identical wastewater. FEMS Microbiol. Ecol., 43: 195206. Schramm A., D. de Beer., M. Wagner and R. Amman. 1998 Identification and activity in situ of Nitrosospira and Nitrospira sp. as dominant populations in a nitrifying fluidized bed reactor. Appl. Environ. Microbiol. 67: 13511362. Schuppler M., M. Wagner, G. Schon and U.B. Gobel. 1998. In situ identification of nocardioform actinomycetes in activated sludge using fluorescent rRNA-targeted oligonucleotide probes. Microbiology 144: 249259. Snaidr J., R. Amann, I. Huber, W. Ludwig. and K.H. Schleifer. 1997. Phylogenetic analysis and in situ identification of bacteria in activated sludge. Appl Environ Microbiol. 63: 28842889. Sponza D.T. 2002. Extracellular polymer substances and physicochemical properties of flocs in steady- and unsteady-state activated sludge systems. Process Biochem. 37: 983998. Sponza D.T. 2003. Investigation of extracellular polymer substances (EPS) and physicochemical properties of different activated sludge flocs under steady-state conditions. Enzyme Microb. Technol. 32: 375385. Van der Gast Ch.J., B. Jefferson, E. Reid, T. Robinson, M.J. Bailey, S.J. Judd and I.P. Thompson. 2006. Bacterial diversity is determined by volume in membrane bioreactors. Environ. Microbiol. 8: 10481055. Wagner M., G. Rath, R. Amann, H.P. Koops and H.K. Schleifer. 1995. In situ identification of ammonia-oxidizing bacteria. Syst. Appl. Microbiol. 18: 251264. Wagner M., G. Rath, H.P. Koops and H.K. Schleifer. 1996. In situ identification of nitrifying bacteria in sewage treatment plants. Wat. Sci. Tech. 34: 237244. Wittebolle L., N. Boon, B. Vanparys, K. Heylen K., P. De Vos. and W. Verstraete. 2005. Failure of the ammonia oxidation process in two pharmaceutical wastewater treatment plants is linked to shifts in the bacterial communities. J. Appl. Microbiol. 99: 9971006.

128


2


Transmission of Specific Groups of Bacteria through Water Distribution System ANNA GRABIÑSKA-£ONIEWSKA1*, GRA¯YNA WARDZYÑSKA1, EL¯BIETA PAJOR1, DOROTA KORSAK2 and KRYSTYNA BORYÑ3 1 Institute

of Environmental Engineering Systems, Warsaw University of Technology, Warsaw, Poland; of General Microbiology, Institute of Microbiology, Faculty of Biology, University of Warsaw, Warsaw, Poland; 3 Municipal Water Supply and Wastewater Company, Warsaw, Poland

2 Department

Received 21 November 2006, revised 1 March 2007, accepted 23 March 2007 Abstract Microbial contamination of a water distribution system was examined. The number and the taxonomy of non-pigmented and pigmented heterotrophic bacteria (HB), number of bacteria (Pseudomonas sp., Enterococcus sp., Campylobacter sp., Yersinia sp., representatives of the Enterobacteriaceae, coagulase-positive staphylococci, and C. perfringens) in the bulk water phase, biomass of zoogloeal aggregates of bacteria, fungi, algae, protozoa and rotifers (ZABFAPR) (separated from the above on 5 µm pore size filters) and in pipe sediments was determined. An increased number of HB occurred at the sampling sites situated as close as 4.2 km to the Water Treatment Plant (WTP), and was especially significant at 10.3 km. It was shown that the main reservoir of hygienically relevant bacteria did not occur in the water phase which is monitored in routine control analyses carried out by the WTP laboratories, but in the ZABFAPR biomass not monitored so far. K e y w o r d s: Campylobacter sp., Yersinia sp., bacterial regrowth, opportunistic pathogenic bacteria, water distribution system

Introduction Risk assessment of the transmission of potentially pathogenic microorganisms in drinking water distribution systems has been investigated with special commitment by epidemiologists and sanitary engineers over the last few years. This interest is prompted by the number of epidemic diseases of the digestive tract caused by unidentified microorganisms, greatly exceeding that caused by typical pathogenic microorganisms such as Salmonella sp., Shigella sp., Vibrio sp., (Anderson et al., 1997) as well as the recent rapid increase in the population of the immunosuppressed particularly susceptible to infection. Those include, for instance, cancer and AIDS patients undergoing chemotherapy, individuals suffering from diabetes and people with various implants. Numerous bacterial, cyanobacterial and fungal species described as opportunistic microorganisms are causative agents of these infections. Their role in transmitting epidemic diseases through water distribution systems was described by Grabiñska-£oniewska (2005).

As the review of literature on the subject provided in it shows, the scope of relevant examinations conducted so far does not include the influence of the quality of intake waters delivered to the Water Treatment Plant (WTP), treatment effectiveness, distance from the WTP and the number of microorganisms occurring both in the bulk water phase as well as in the biomass of zoogloeal aggregates of bacteria, fungi, algae, protozoa and rotifers (ZABFAPR) suspended in it, in the pipe wall biofilm and in pipe sediments on the contamination degree of water distribution systems with potentially pathogenic microorganisms. Therefore, the aim of this study was to determine the occurrence of selected groups of pathogenic and opportunistic pathogenic bacteria in intake waters and after various stages of their treatment as well as in sections of the water distribution system located at different distances from the WTP, in samples collected from the water phase, the biomass of ZABFAPR and pipe sediments. For the purposes of the study, numbers of pigmented and non-pigmented heterotrophic bacteria (HB),

* Corresponding author: A. Grabiñska-£oniewska, Institute of Environmental Engineering Systems, Warsaw University of Technology, Nowowiejska 20, 00-653 Warsaw, Poland; e-mail: [email protected]

130

Grabiñska-£oniewska A. et al.

bacteria of the genus Pseudomonas growing at 7°C, 30°C and 42°C, bacteria of the family Enterobacteriaceae, of the genera Yersinia and Campylobacter, faecal streptococci, Enterococcus sp. coagulase-positive staphylococci, C. perfringens were determined in the samples. The quantitative composition of ZABFAPR was determined in the biomass separated from the water phase on 5 µm pore size filters. Experimental Material and Methods

Collection of samples. Water samples were collected between December (2000) and November (2002) from different sites of municipal water supply of the city of Warsaw. These included intake waters supplying the Water Treatment Plant (WTP), water treated in technological lines I and II of WTP, a mixture of the above waters, disinfected with a mixture of ClO2 and Cl2 transferred into the distribution system and waters from 7 sites within the distribution system located on 4.2; 4.3; 5.4; 5.7; 8,7; 10.0 and 10.3 km from the WTP. Description of water treatment in technological lines I and II and methods of sample collection are given elsewhere (Grabiñska£oniewska et. al., 2007). Samples preparation. Water samples freed from chlorine (using 10% solution of sodium thiosulphate) were used for examinations. The biomass of ZABFAPR was separated from 20 l water samples collected from the system using the filtration method through a cellulose nitrate membrane filter (Whatman Schleider and Schuell), pore size 5 µm, with a vacuum pump (Merck ME 2). The biomass retained on the filter was eluted for 30 min in a shaker (Elpan, type 357), with the amplitude 6 and 200 cpm/min, to 50 ml of 0.28% solution of sodium pyrophosphate. The prepared suspension was used for microscopic examinations. The suspension (as above), disrupted in an ultrasound disintegrator UD 20, vibration amplitude A = 20 µm for 40 s, was used for microbiological examinations. Using this method of sample preparation, the obtained suspension contained both microorganisms occurring inside zoogloeal aggregates of bacteria, as well as inside and on the surface of cells of algae and protozoa colonising the water phase in the water distribution system. Samples from which the biomass of ZABFAPR was removed using the method described above were used to prepare the suspension of microorganisms collected from the water phase in the water distribution system. They were concentrated using the filtration method (as above) through a membrane filter, pore size 0.22 µm (Campylobacter sp.) and 0.45 µm (other

2

bacteria). Depending on degree of water contamination, filtration was performed from: 5.0; 1.0; 0.5; 0.1 litre as well as 10 and 1 ml. The filter was then placed in 10 ml of a 0.28% solution of sodium pyrophosphate and shaken for 30 min in a shaker (as above) to wash the microorganisms off the filter. The suspension prepared was used to inoculate culture media. Sediments were scraped from 62.8 80 cm2 of the inner surface of sample sections, weighed, ground in a mortar, suspended in 0.28% solution of sodium pyrophosphate and shaken in shaker (as above) to elute microorganisms from the mineral fraction of the sediments. Non-concentrated samples were used for inoculation. Microbiological determinations. The number of pigmented and non-pigmented heterotrophic bacteria (HB) was determined in culture on broth agar medium (MPA) at 26°C after 7 days of incubation. Bacteria of the genus Pseudomonas were incubated on King B medium at 7°C for 10 days, 30°C 4 days and 42°C 2 days. VRBG agar medium was used to enumerate bacteria of the family Enterobacteriaceae, and the Enteroplast EPL 21 test (temp. 37°C) was used for confirmatory re-examination (Burbianka et al., 1983). Yersinia sp. bacteria were first enumerated on Endo MLCe agar medium after 25 days of incubation at 30°C. Colonies characterised by typical morphology (round, 12 mm in diameter, dark red), non-cytochrome oxidase producing were re-examined for confirmation using Enteroplast EPL 21 (Krogulska and Maleszewska, 1992). The occurrence of Campylobacter sp. was initially determined on Oxoid Brucella liquid enrichment medium supplemented with compounds lowering E h and after incubation at 37°C for 4 h, also with a mixture of antibiotics. After 20 h of incubation at 42°C, the culture was transferred to Oxoid Brucella agar supplemented with 5% of blood and incubated for 2 days at 42°C in anoxic conditions. Characteristic colonies (pink point, beige coloured) were re-examined for confirmation using the PCR method (Or and Jordan, 2003). The density of faecal streptococci was determined according to PN-82/C-04615/25, coagulase-positive staphylococci PN-75/A-04024, C. perfringens PN-77/C-04615. The determination (Enterococcus sp.) results of the number of HB, bacteria of the genus Pseudomonas growing at 7, 30 and 42°C, bacteria of the family Enterobacteriaceae, bacteria of the genera Yersinia and Enterococcus sp. are reported as cfu/l, cogulase-positive staphylococci as MPN per 100 ml, and bacteria of the genus Campylobacter as a titre. The biomass of ZABFAPR of the water distribution system was observed under a phase contrast microscope (Opton). Determinations results are reported as the density of individual groups of organisms expressed per litre of the sample.

2

Transmission of specific groups of bacteria trough water distribution system

Microbiological examinations of pipe sediments collected from the inner surface of sample sections of water distribution pipes were conducted at the same time. Their scope was as above; however, the density of bacteria of the family Enterobacteriaceae and Cl. perfringens was not determined due to technical reasons. The results are reported as cfu/100 g. Results and Discussion The examinations conducted show different degrees of microbiological contamination of waters supplying the WTP. Infiltration waters supplying technological line I were characterised by low density of non-pigmented and pigmented HB, bacteria Pseudomonas sp. growing at 7°C, 30°C and 42°C (11.7× 104; 2.8× 104; 350; 13.8× 103 and 86 cfu/l, respectively) and C. perfringens (33 cfu/l). Bacteria of the family Enterobacteriaceae, Yersinia sp., Campylobacter sp. and Enterococcus sp. were not found, and the MPN of coagulase-positive staphylococci was