Identification and characterization of

0 downloads 1 Views 212KB Size Report
been used as a cell-free system for performing photo- synthesis [11,12]. ... changed from dark green to reddish purple in color. This strain was selected for ...

Biomolecular Engineering 18 (2001) 49 – 56 www.elsevier.com/locate/geneanabioeng

Identification and characterization of Rhodopseudomonas spp., a purple, non-sulfur bacterium from microbial mats Sharifeh Mehrabi a,*, Udoudo M. Ekanemesang a, Felix O. Aikhionbare b, K. Sean Kimbro a, Judith Bender a b

a Department of Biological Sciences, Clark Atlanta Uni6ersity, Atlanta, GA 30314, USA Department of Microbiology/Biochemistry/Immunology, Morehouse School of Medicine, Atlanta, GA 30310, USA

Received 2 March 2001; received in revised form 10 June 2001; accepted 10 June 2001

Abstract A species of facultative photo-organotrophic, purple, non-sulfur bacterium was isolated from mixed-species microbial mats, characterized and examined for metal tolerance and bioremediation potential. Contributing mats were natural consortia of microbes, dominated by cyanobacteria and containing several species of bacteria arranged in a laminar structure, stabilized within a gel matrix. Constructed microbial mats were used for bioremediation of heavy metals and organic chemical pollutants. Purple, non-sulfur bacteria are characteristically found in lower strata of intact mats, but their contributing function in mats survival and function by mediating the chemical environment has not been explored. The gram-negative rod-shaped bacterium, reported here, produced a dark red culture under phototrophic conditions, reproduced by budding and formed a lamellar intracytoplasmic membrane (ICM) system parallel to cytoplasmic membrane, which contained bacteriochlorophyll a and carotenoids. This strain was found to have multiple metal resistances and to be effective in the reductive removal of Cr(VI) and the degradation of 2,4,6-trichlorophenol. Based on the results obtained from morphology, nutrient requirements, major bacteriochlorophyll content, GC content, random amplified polymorphic DNA-polymerase chain reaction (RAPD-PCR) profile and 16S-rDNA phylogenetic analysis, this member of the microbial mats may be identified as a new strain of the genus Rhodopseudomonas. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Microbial mat; Rhodopseudomonas, non-sulfur purple bacteria; Intracytoplasmic membrane; RAPD-PCR; 16S-rDNA; TCP; Cr(VI)

1. Introduction Species of the genus Rhodopseudomonas constitute the majority of phototrophic purple non-sulfur bacteria, and are characterized as rod-shaped, motile cells that show polar growth and asymmetrical division or budding as a mode of reproduction [1]. They form lamellar intracytoplasmic membranes that are adjacent and parallel to cytoplasmic membrane and contain photosynthetic pigments under phototrophic conditions [2]. The colors of cell suspensions of many species grown under phototrophic conditions vary from species to species, indicating the presence of major carotenoid * Corresponding author. Tel.: + 1-404-8806297; fax: +1-4048806756. E-mail address: [email protected] (S. Mehrabi).

types (maximum absorption at 450 –550 nm) and bacteriochlorophylls [3]. Absorption spectra of cell suspensions yield primary information on predominant bacteriochlorophylls and carotenoids. This color trait is useful in initial characterization of these species [2]. The major carotenoids are lycopene, rhodopin (463, 490, 524 nm), spirilloxanthin (486, 515, 552 nm), okenone (521 nm) and spheroidene (450, 482, 514 nm). The spirilloxanthin series (lycopene, rhodopin, spirilloxanthin) as the major component gives a pink or red color, increasing amounts of rhodopin turn the color to red – brown, okenone (521 nm) results in purple –red. The spheroidenone series gives a green or brownish red and greenish brown under reducing conditions [2,3]. Rhodopseudomonas species have shown a high degree of morphological and structural similarities, however, they are phylogenetically quite diverse. Comparisons of rRNA sequences, DNA –DNA and DNA –RNA hy-

1050-3862/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 1 3 8 9 - 0 3 4 4 ( 0 1 ) 0 0 0 8 6 - 7

50

S. Mehrabi et al. / Biomolecular Engineering 18 (2001) 49–56

bridization have demonstrated that these bacteria belong to several different lines of descent within the alpha-2 subclass of Proteobacteria [3]. Morphological, cytological and physiological properties of the genus Rhodopseudomonas have been summarized by Imhoff et al. [4]. Recent phylogenetic studies of Rhodopseudomonas species demonstrated that Rhodopseudomonas palustris is more closely related to some chemotrophic taxa, such as Afipia felis, Bradyrhizobium japonicum, Blastobacter denitrificans and Nitrobacter winogradskyi, than to any of the other Rhodopseudomonas species for which rDNA sequences are available [5,6]. Phototrophic non-sulfur bacteria have been used in sewage treatment processes [7,8], for biomass and vitamin production [9] and for production of molecular hydrogen [10]. In bioenergetic research, they have been used as a cell-free system for performing photosynthesis [11,12]. Biodegradation of aromatic compounds by Rhodopseudomonas species such as benzoate [13], dihydroxylates, methyoxylate [14] and 4-hydrobenzoate [15] have been reported. Non-sulfur bacteria display a high resistance to heavy metals [16], giving these cells another potential application in bioreclamation of rare-earth metals. In attempts to isolate the bacterial members of mats and elucidate their roles in bioremediation, we isolated and partially characterized a phototrophic, purple, non-sulfur bacterium reported in this study. The mats were developed for bioremediation and constructed by stimulating the formation of a microbial consortium with specific substrates (ensilaged grass) and inocula composed of Oscillatoria spp. and allowing volunteer heterotrophic groups from the soil to colonize the preparation [17]. When the mat was used for treatment of organic [18,19] and/or inorganic [20– 23] chemicals, the population of purple bacteria within the mat significantly increased and the mat changed from dark green to reddish purple in color. This strain was selected for identification and characterization because of its dominance during the sewage treatment process. The isolate is characterized according to its phenotypic, genotypic and phylogenetic properties and compared with other members of Rhodopseudomonas spp. The contribution of this strain in microbial mats bioremediation properties will be reported in future.

2. Materials and methods

2.1. Bacterium source and isolation technique The bacterium was isolated from a mature microbial mat used for metal sequestration and organic pollutant degradation experiments. The development and

application of microbial mats have been reported previously [17,18,20–22]. Samples of microbial mat were blended in sterile saline solution (NaCl 8.5 g/l) and serially diluted. Colony count pour plates were prepared from each dilution in nutrient agar medium (Dickerson Microbiology Systems, Cokeysville, MD), incubated at room temperature in the dark and anaerobically under 60 W incandescent light in an anaerobic chamber (Forma Scientific, Marietta, OH). Several red or red–brown colonies of different sizes and textures appeared after five to seven days incubation under anaerobic/light conditions. The most dominant red colony was selected for purification. Several representatives of single red colonies were individually transferred into screw cap tubes filled with nutrient broth medium and incubated under the light at room temperature for enrichment. To ensure the purity of cell culture, the streak dilution plates were prepared from each tube containing the originally isolated colony on nutrient agar medium, and checked for purity.

2.2. Determination of nutritional and physiological characteristics The stock solutions of different carbon sources were filter-sterilized and added to the AT salt medium [2] to a final concentration of 1–2 g/l. The pH of the media was adjusted to 7 with HCl or NaOH for routine analysis or to 5, 6, 7 or 8 for pH profile analysis. Heterotrophic cultures were grown in 100 ml of medium in 250 ml flasks incubated in a rotary shaker at 150 rpm in the dark. Phototrophic cultures were grown in completely filled 250 ml screw cap culture bottles and incubated under 60 W incandescent bulbs at a distance of 60 cm in an anaerobic chamber. The growth was monitored turbidimetrically, using a Beckman DU 650 UV/VIS scanning spectrophotometer (Beckman, Fullerton, CA) at 600 nm. The number of colony forming units (CFU) was determined by the spread plate count method using serial dilutions. The growth rate constant (K), the number of generations (n) and the mean generation time (g) were calculated using standard mathematical growth equations [24]. The measurement of optimum growth temperature was carried out in nutrient broth medium, in a water bath set to the desired temperature. Assay for formation of intracellular sulfur globules was performed according to the method described by Balows et al. [2]. Briefly, the cells were incubated on a microscopic slide for 20 min with neutralized Na2S solution and monitored for intracellular sulfur globules formation. Chromatium 6inosum (ATCC 17899), a sulfur purple bacterium was used as a positive control. All experiments were performed in triplicate.

S. Mehrabi et al. / Biomolecular Engineering 18 (2001) 49–56

2.3. Absorption spectra and electron microscopy Determination of the predominant bacteriochlorophyll and the carotenoid composition of the isolate were performed as described by Balows et al. [2]. The absorption spectra of cell suspensions in 4 M sucrose were recorded against a blank of 4 M sucrose solution using a Beckman DU 650 UV/VIS scanning spectrophotometer. The approximate maximum absorption of major peaks was determined. A transmission electron microscope (TEM, Hitachi H-6000), set at 0.3 nm point-to-point resolution and 50 000× magnification, was used to examine the isolate. Standard microscopic methods were used for slide preparation and Gram staining.

2.4. Reducti6e remo6al of Cr(VI) Cells were grown to the late log phase in nutrient broth harvested, washed and resuspended in 100 ml of minimal salt medium at cell density equivalent to optical density reading of 1.5 at 600 nm. Varying concentrations of dichromate solution were added to the cells. Aliquots were removed at time intervals, filtered and analyzed for hexavalent chromium. The concentration of Cr(VI) was determined colorimetrically by reaction with diphenylcarbazide in acid solution as described in section 3500 of the Standard Method for Examination of Water and Wastewater (American Public Health Association, 1995). An aliquot of 2 ml of sample was added to 9.8 ml of 0.2 N H2SO4, followed by the addition of 0.5 ml of diphenylcarbazide reagent (0.5% in acetone, Labchem Inc., Pittsburgh, PA). The absorbance of the solution was measured at 540 nm using a Beckman DU 650 UV/VIS scanning spectrophotometer.

51

Peaks were quantified by external standards using system Gold.

2.6. Sample source and GC content analysis Seven species of Rhodopseudomonas that share phenotypic characteristics with the isolate were purchased from ATCC (American Type Culture Collection) (Table 1). These strains were cultured in media recommended by ATCC under phototrophic conditions. Cells were harvested at their exponential growth phase. The G+ C base content of the isolate was determined by the thermal denaturarion method as described by Marmur and Doty [25].

2.7. DNA preparation and 16S-rDNA sequence analysis DNA extraction was performed using the methods described by Maniatis et al. [26]. DNA was then precipitated with chilled ethanol, washed with 70% ethanol, dried and finally dissolved in TE (10:1) and stored at − 20 °C until diluted to about 100 ng/ml prior to use. Amplification of 16S ribosomal DNA (rDNA) was performed on the isolated DNA using primers (5%CAGCCAGCCGCGGTAATAC-3%) and (5%-CCGTCAATTCCTTTGAGTT-3%). The PCR reaction mix was performed as previously described [27,28]. Amplified products were resolved on 1.5% agarose gel, then excised from the gel and purified. The purified products were cloned into pGEM-T (Promega) according to manufacturer’s instructions and subsequently sequenced using Big Dye Terminator chemistry (ABI, Foster, CA). The sequences were assembled using Autoassembler and sequence navigator (ABI, Foster, CA). A 407 bp 16S-rDNA sequence was obtained from this study.

2.5. Degradation of trichlorophenols 2.8. RAPD-PCR analysis Cells were grown in nutrient broth to the late log phase, harvested, washed and resuspended in 100 ml of minimal salt medium at cell density equivalent to optical density reading of 1.5 at 600 nm. 2,4.6Trichlorophenol (100 mg/ml) was added to the culture flasks and incubated in an anaerobic chamber at 30 °C on a rotary shaker at 150 rpm. At time intervals aliquots of the culture (2.0 ml) were removed; cells were removed by centrifugation. Equal volume of methanol was added to the supernatant and filtered through a 0.45 mm pore-size filter (Millipor, Bedford, MA) into a HPLC sample vial. Samples were analyzed using a Beckman System Gold HPLC equipped with a model 502 autosampler, 126 solvent module, diode array detector and c18 column (15 cm× 4.6 mm; particle size 5 mm). Injection volume was 20 ml. The mobile phase was methanol/water (60:40) at a flow rate of 1 ml/min.

A total of 20 random and previous 16S-rRNA sequences of Rhodopseudomonas species were used in the selection of RAPD primers, then obtained from Genosys Biotechnologies Inc. (Woodlands, TX). All Table 1 Rhodopseudomonas species used in RAPD-PCR Sample

Species

ATCC

Reference

1 2 3 4 5 6 7

Rps. palustris (RPS p) Rps. acidophila (RPS a) Rps. rutila (RPS r) Rps. rosea (PRS ro) Rps. 6iridis (RPS v) Rps. marina (RPS m) Isolate

17001 25092 33872 49724 19567 35675

[31] [31] [32] [33] [31] [34] This study

S. Mehrabi et al. / Biomolecular Engineering 18 (2001) 49–56

52

Fig. 1. Scanning electron micrograph of the isolate; circle shows a budding cell; bar, 1 mm.

RAPD primers used in this study were 10-mers, with a 60% or 70% G+C content. RAPD-PCR reactions were obtained in 25 ml volume similar to previously described [29]. Amplified PCR products and controls were resolved in 1.5% agarose gels. All the resultant bands were scored in reference to the 100 bp molecular weight ladder (GibcoBRL, Gaitherburg, MD) as either present or absent across the lanes.

2.9. Phylogenetic analysis 407 bp of the16S-rDNA sequence of the isolate was matched to other Rhodopseudomonas spp. obtained from GenBank databases. Names and accession numbers are as follows: Rhodopseudomonas spp. B29, AB027692; Uncultured alpha proteobacterium A, AB035489; Uncultured alpha proteobacterium, AB035490; Bradyrhizobium spp. Mml-3, AF159438; Bradyrhizobium spp. Pel-3; AF159436; Bradyrhizobium spp. Pe4, AF159437; Bradyrhizobium spp. DesB1, AF178436; Bradyrhizobium spp. jwc91-2, AF178437; Bradyrhizobium spp. th-b2, AF178438; Bradyrhizobium spp. ORS2012, AF230721; Bradyrhizobium spp. ORS285, AF230722; Bradyrhizobium elkanii strain SEMIA 5019, AF237422; Bradyrhizobium spp. STM461, AF239254; Bradyrhizobium spp.

ORS278, AF239255; Bradyrhizobium elkanii PRY52, AF239846; Bradyrhizobium japonicum, AF293382; Bradyrhizobium spp. strain Cj3-3, AF321215; Bradyrhizobium liaoningense strain 2281, AF363132; Bradyrhizobium japonicum strain USDA 123, AF363151; Afipia genosp. U87771; Rhizospere soil bacterium isolate RSI-33, AJ252600; Bradyrhizobium spp. MSDJ5726, AF363149; Bradyrhizobium elkanii strain USDA 94, AF363152; Bradyrhizobium genospp. J, Z94820; Rhodopseudomonas palustris, D89811; Rhodopseudomonas spp. g–c 16S AF123086; R. palustris, X87279. The program Clustal W [30] was used to create a multiple-sequence alignment of the sequences and to construct the tree from the alignment. Bootstrapped distance matrix analysis was performed using Clustal W with 1000 bootstrapping trials. We used the PHYLIP program DRAWTREE to draw the tree from the output of the Clustal W program.

3. Results

3.1. Morphology and phototrophic properties The isolate was a gram-negative rod-shaped motile bacterium with a length of 2.0–2.4 mm. The cells divided

S. Mehrabi et al. / Biomolecular Engineering 18 (2001) 49–56

53

Table 2 Physiological properties of the isolate

Fig. 2. Thin section electron micrograph of the isolate showing intracytoplasmic membrane (ICM); bar, 0.3 mm.

asymmetrically by budding. Bud formation at the end of the mother cell gives a dumbbell-shaped appearance before separation of the daughter cell [1] (Fig. 1). The electron micrographs of thin sections of phototrophically grown cells showed a lamellar intracytoplasmic membrane (ICM) system (Fig. 2). The bacteria form smooth, round and dark red colonies with a diameter of 2 –3 mm on agar media and red cell suspensions in liquid media under phototrophic conditions. The strain formed photosynthetic pigments when grown under anoxygenic phototrophic conditions. Absorption spectra of cells suspended in sucrose had three major peaks at 593, 805 and 871 nm, indicating the presence of bacteriochlorophyll a (590, 800– 900 nm), and three peaks at 463, 490 and 530 nm, indicating the presence of lycopene and rhodopin. There were no absorption peaks for bacteriochlorophylls b (400, 605, 840 and

Substrate, reaction or enzyme

Phototrophic growth

Succinate Ethanol Pyruvate Malate Acetate Valerate Benzoate Tween 80 hydrolysis Nitrate reduction Catalase 3% NaCl Arginine Lysine Ornithine Sodium citrate Sodium thiosulfate Urea Tryptophan Gelatinase Galactosidase Glucose Mannitol Starch hydrolysis Sucrose Inositol Rhamnose Melibiose Amygdalin Arabinose

+ + + − + + − − + − − − − − − − − − 9 − + − − + + + − − −

+, Positive growth or reaction; −, negative growth or reaction; 9, partially positive.

1025–1035 nm) and c (660–668 nm), which are characteristic of purple non-sulfur bacteria (Fig. 3) [2,3].

3.2. Physiological and metabolic properties Optimal growth occurred at 30–35 °C and at pH 6–8 under both heterotrophic and phototrophic conditions. Phototrophic growth was observed on several carbon and electron donor sources (Table 2). The growth of the bacteria in nutrient broth was significantly higher than in minimal salt media with a single carbon source. The K values (mean growth rate constants) of the bacterial growth in nutrient broth and other carbon substrates under phototrophic and heterotrophic conditions are presented in Table 3.

3.3. Strain identification Fig. 3. Absorption spectrum of cell suspension of isolate grown under phototrophic conditions showing major absorption peaks at 590, 805 and 871 nm, indicating the presence of bacteriochlorophyll a and 463, 490 and 530 nm, indicating the presence of carotenoid (lycopene and rhodopin).

The isolate was sent to two commercial organizations, ATCC (Bethesda, MD) and Analytical Services Inc. (Essex Junction, VT) for their independent confirmation of our identification. Analytical Services was unable to obtain a match on its microbial identification

S. Mehrabi et al. / Biomolecular Engineering 18 (2001) 49–56

54

system (MIS) database using the fatty acid profile. Based on microscopic examination, cultural requirements and metabolic properties, ATCC identified the bacterium as an unknown species of Rhodopseudomonas genus.

3.4. Bioremediation property The bacterium reduced Cr(VI) to Cr(III). The reduction was rapid at lower concentrations of Cr(IV) and decreased at higher concentrations as toxicity of Cr(IV) increased. The reduction rates were as follows: 10 mg/l per 2 h and 80 mg/l per 2 days. TCP degradation rate was about 10 mg/l per day. The metabolic products of 2,4,6-TCP and 2,4-DCP, 4-CP also appeared in approximately 3 h and 2 days and disappeared in 4 and 8 days, respectively. These products were identified using known standards. 4,6-DCP, 2,6-DCP, 2-CP and 6-CP were not detected, suggesting that groups at positions 6 and 2 are most labile.

3.5. RAPD While 20 RAPD primers produced multiple amplification products for the seven bacterial species (Fig. 4), three primers: SMO1 (5%-ACGGTGCCTG-3%), SMO2 (5%-GAGATCCGCG-3%) and SMO3 (5%-CGGGTCGATC-3%) produced the most variable banding profile. RAPD-PCR produced distinguishable bands with the studied species (lane 7) from all other six species (lanes 1, 2, 3, 4, 5, 6).

3.6. Phylogeny To classify the isolate, we used Clustal W [29] to align the 407 bp sequence obtained from the isolate with several non-sulfur alpha-2 Proteobacteria groups. From this alignment, we construct a rooted phylogenetic tree (Fig. 5). Phylogenetic analysis of this isolate using the 16S-rDNA sequence data suggests that the isolate was an unidentified species that is closely related Table 3 Growth parameters of the isolate in different nutrients Media

Ta (h)

g (h)

Nutrient broth Acetate Succinate Ethanol Glucose Pyruvate

95 164 168 234 234 270

4:45 16 21 18 23 19

Value Ta is end of exponential phase in hours determined directly from growth curves; g is mean generation time in hours calculated from growth equations [24].

Fig. 4. RAPD-PCR profiles obtained with primers: SMO1 (a); SMO2 (b); SMO3 (c) show variations among the DNA of the seven Rhodopseudomonas spp. (lane 1, Rps. palustris; lane 2, Rps. acidophila; lane 3, Rps. rutila; lane 4, Rps. rosea; lane 5, Rps 6iridis; lane 6, Rps marina; lane 7, isolate). Lanes M, Kb ladder.

to Bradyrhizobium palustris.

spp.

and

Rhodopseudomonas

4. Discussion One of the objectives of this study was to identify and characterize some of the dominant species in a synthetic microbial mat designed for bioremediation applications. Results from this investigation provided information on a heretofore-unidentified strain of Rhodopseudomonas spp., a dominant bacterium in the mats. In order to assess the bioremediation potential of the isolate for metal removal from contaminated water, an initial scan of tolerance to various metals was performed (data not shown). Based on these results, a selection of Cr(VI) was made for further studies in metal removal rates from contaminated water. Although Cr(VI) was the only metal included in these studies, metal tolerances indicate that this microbe may

S. Mehrabi et al. / Biomolecular Engineering 18 (2001) 49–56

be important in the bioremediation of other metals as well. The identification of this bacterium as a putative new member of non-sulfur purple bacteria was based on its morphological, physiological characteristics, and DNA analyses. These results suggest that this isolate is a mesothermophilic gram-negative rod-shaped bacterium that shares general characteristics with the Rhodopseudomonas a genus of Rhodospirillaceae family. Cellular and morphological similarities of the isolate to the previously described species in the genus Rhodopseudomonas are: the presence of a lamellar intracytoplasmic membrane system, the bactochlorophyll a and reproduction by budding [1 – 3]. Electron micrographs of the isolate also exhibited these characteristics (Figs. 1 and 2). The isolate did not grow on sulfide-containing medium and failed to form intracellular sulfur globules that distinguished this strain from the sulfur purple bacteria. The general identification of the isolate as a Rhodopseudomonas spp. is consistent with results from ATCC. The RAPD-PCR profile of the isolate had a pattern

55

that was clearly distinguishable from other known Rhodopseudomonas species. The physiological and morphological properties of the isolate are also different from those of the other known Rhodopseudomonas species [2], which was consistent with the RAPD-PCR results. Comparison of the partial 16S-rDNA sequence (407 bp) of this isolate with the sequences found in the GenBank database yielded 97% homology to Nitrobacter and several Bradyrhizobia and Rhodopseudomonas palustris strains. Using SEQUENCE-MATCH from the Ribosomal Database Project II we determined that this isolate belongs to the purple bacteria, alpha subdivision of proteobacteria based on the fact that sequence homology corresponded with the rhizobium-agrobacterium group, bradyrhizobium subgroup. This study and other phylogenetic studies using 16S-rRNA sequence homologies [5,6] have demonstrated that species Rhodopseudomonas palustris is more closely related to some chemotrophic taxa, such as Bradyrhizobium species, Nitrobacter winogradskyi and Afipia felis than to any of the other Rhodopseudomonas species. Although data obtained from this study clearly provide reliable evidence for close relationships between this isolate and Rhodopseudomonas palustris, it would be unduly optimistic to hope that analysis of 407 nucleotide positions of 16S-rDNA, many of which are invariant, could allow an unambiguous reconstruction of all deeper branches in the phylogeny of this isolate to already known R. species. However, our results show that the isolate, Bradyrhizobium spp. and Rhodopseudomonas palustris belong to the same lineage, and are distinct from all other Rhodopseudomonas species. Based on the results of this study, we suggest that this isolate is an unidentified strain of Rhodopseudomonas species. The bioremediation properties of this isolate will be presented in another report.

Acknowledgements This project was partially supported by the US Environmental Protection Agency, Office of Exploratory Research, Washington, DC 20460; Assistance ID NO R 821301-01-0 and a grant of the Department of Energy, No. DE-Fco4-90AL66158.

References Fig. 5. A rooted tree generated from an alignment of 407 bp of the16S-rDNA of the isolated with the phototrophic purple non-sulfur bacteria obtained from the GenBank database, using the Clustal W [30] and the PHYLIP program DRAWTREE. The tree was rooted with Bradyrhizobium japonicum strain, AF363151. The names and accession numbers of the species used in this phylogenetic tree are given in Section 2.

[1] Pfennig N, Tru¨ per HG. In: Buchanan RE, Gibbons NE, editors. Bergey’s Manual of Determinative Bacteriology, eighth edition. Baltimore, MD: Williams and Wilkins, 1974:24 – 33. [2] Balows A, Tru¨ per HG, Dworkin M, Hander W, Schleifer K, editors. The Prokaryotes. New York: Springer-Verlag, 1992 pp. 2141 – 2155 and 3200 – 3222. [3] Hiraishi A, Ueda Y. Int J Syst Bacteriol 1994;44(4):665 –73.

56

S. Mehrabi et al. / Biomolecular Engineering 18 (2001) 49–56

[4] Imhoff JF, Tru¨ per HG, Pfennig N. Int J Bacteriol 1984;34(3):340 – 3. [5] Kawasaki H, Hoshino Y, Yamasato K. FEMS Microbiol Lett 1993;112:61 – 6. [6] Vinuesa P, Rademarker JLW, Bruijn FJ, Werner D. Appl Environ Microbiol 1998;64(6):2096 –104. [7] Irgens RL, Pfennig N. Appl Environ Microbiol 1978;35:38 – 44. [8] Kobayashi MP. In: Schlegel HG, Barnea J, editors. Microbial Energy Conversion. Oxford: Pergamon Press, 1977:443 –53. [9] Sasaki K, Noparatnaraporn N, Hayashi M, Shizowa Y, Najai S. J Ferm Technol 1981;59:471 –7. [10] Matsunaga T, Matsunaga N, Tsubaki K. J Bacteriol 1986;168(1):460 – 3. [11] Svensson B, Etchebest C, Tuffery P, Van Kan P, Smith J, Styring S. Biochemistry 1996;35(46):14486 –502. [12] Varga AR, Staehelin LA. Arch Microbiol 1985;141(4):290 – 6. [13] Dutton FL, Evans C. Biochem J 1967;104:30 – 1. [14] Krooneman J, Akker SV, Pedro T, Forney J, Gottschal J. Appl Environ Micribiol 1999;65(1):131 –7. [15] Gibson J, Dispensa M, Harwood CS. J Bacteriol 1997;179(3):634 – 42. [16] Moore MD, Kaplan S. ASM News 1994;60(1):17 – 23. [17] Bender J, Vatcharapijarn Y, Russell A. Aquacult Eng 1989;8:407 – 19. [18] Bender J, Murray R, Phillips P. In: Means JL, Hinchee RE, editors. Emerging Technology for Bioremediation of Metals. Boca Raton, FL: CRC Press, 1994. [19] Phillips, P., Bender, J. 1993. Report to Battelle Pacific Northwest Laboratories.

[20] Bender J, Lee RF, Phillips P. J Industrial Microbiol 1995;14:113 – 8. [21] Bender J, Archibold E, Ibeanusi B, Gould JP. Water Sci Technol 1989;21:1661 – 7. [22] Bender J, Rodriguez-Eaton S, Ekanemesang UM, Phillips P. Appl Environ Microbiol 1995;60:2311 – 5. [23] Bender J, Vatcharapijarn Y, Linenmeyer C. Abstract, 94th Annual Meeting of American Society of Microbiology, Las Vegas, 1994. [24] Prescott LM, Harley JP, Klein Donald A, editors. Microbiology, second edition. Dubuque: William C. Brown, 1993:112 –22. [25] Marmur J, Doty P. J Mol Biol 1962;5:109 – 18. [26] Manials T, Fritsch EF, Sambrook J. Molecular Cloning A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1982. [27] Lane DJ, Pace B, Olsen GJ, Stahl DA, Sogin M, Pace NR. Proc Natl Acad Sci USA 1985;82:6955 – 9. [28] Brosius J, Palmer ML, Kennedy PJ, Noller HF. Proc Natl Acad Sci USA 1978;75(10):4801 – 5. [29] Aikhionbare FO, Pruess KP, Mayo ZB. Gen Anal Biomol Eng 1998;14:105 – 8. [30] Thompson JD, Higgins DG, Gibson TS. Nucleic Acids Res 1994;22:4673 – 80. [31] Skerman VB, McGowan V, Sneath PH. Int Syst Bacteriol 1980;30:225 – 32. [32] Akiba TR, Usami R, Horikoshi K. Int J Syst Bacteriol 1983;33:551 – 6. [33] Janssen PH, Harfoot CG. Int J Syst Bacteriol 1991;41:26 –30. [34] List No. 44. Int J Syst Bacteriol 1985;43:188 – 9.