Novel marine carbazole-degrading bacteria

Abstract

Eleven carbazole (CAR)-degrading bacterial strains were isolated from seawater collected off the coast of Japan using two different media. Seven isolates were shown to be most closely related to the genera Erythrobacter, Hyphomonas, Sphingosinicella, Caulobacter, and Lysobacter. Meanwhile, strains OC3, OC6S, OC9, and OC11S showed low similarity to known bacteria, the closest relative being Kordiimonas gwangyangensis GW14-5 (90% similarity). Southern hybridization analysis revealed that only five isolates carried car genes similar to those reported in Pseudomonas resinovorans CA10 (car_CA10) or Sphingomonas sp. strain KA1 (car_KA1). The isolates were subjected to GC-MS and the results indicated that these strains degrade CAR to anthranilic acid.

Introduction

Previous studies have indicated that the Gram-negative bacteria Pseudomonas resinovorans CA10 and Sphingomonas sp. strain KA1 and the Gram-positive bacterium Nocardioides aromaticivorans IC177 are carbazole (CAR)-degrading bacteria, and the CAR degradation pathway and genes have been studied in detail in these organisms (Ouchiyama et al., 1993; Sato et al., 1997a, b; Nojiri et al., 2001; Habe et al., 2002; Inoue et al., 2004, 2005, 2006; Urata et al., 2006). These strains degrade CAR to 2′-aminobiphenyl-2,3-diol via angular dioxygenation catalyzed by CAR 1,9a-dioxygenase (CARDO), which is encoded by the carAa, carAc, and carAd genes. CARDO transforms not only CAR but also dioxin compounds and polycyclic aromatic hydrocarbons (PAHs) (Nojiri et al., 1999; Habe et al., 2001a; Inoue et al., 2006; Urata et al., 2006). These properties make CARDO an excellent tool for bioremediation, and CAR-degrading bacteria have been shown to have the ability to degrade CAR and various other contaminants in vivo (Habe et al., 2001b, 2002; Widada et al., 2002). Most CAR-degrading bacteria have been isolated from onshore or freshwater sites such as soils, river water, and activated sludge, and thus are not useful for bioremediation in marine environments. A search for new isolates from marine environments using molecular techniques could be useful for aromatic compound bioremediation in such environments. Specifically, marine CAR-degrading bacteria that have a novel unique CARDO of marine bacterial origin are expected to be capable of degrading dioxins and PAHs. Although a few CAR-degrading marine bacteria have been isolated from marine environments (Fuse et al., 2003; Inoue et al., 2005), more information regarding marine CAR-degrading bacteria is required to implement effective bioremediation programs in marine environments.

In this study, we isolated CAR-utilizing bacteria from seawater collected off the coast of Japan. Two different media were used to isolate CAR-utilizing bacteria belonging to two distinct genera from the same sample. Sequence analysis using 16S rRNA gene revealed that most of the genera identified here have not been reported previously as CAR-utilizing bacteria. The isolates were also subjected to PCR and Southern hybridization analysis to detect previously identified car genes from strains CA10, KA1, and IC177. GC-MS revealed that the isolates produced CAR degradation intermediates.

Materials and methods

Isolation of CAR-degrading bacteria from seawater

We collected 10 L of seawater from 15 areas in Japan, including Niigata (Koshinetu region), Hyogo (Kinki region), Toyama (Hokuriku region), Yamaguchi (Chugoku region), Shizuoka (Tokai region), Fukuoka (Kyushu region), and Chiba and Kanagawa (Kanto region). Marine bacteria were collected from the seawater samples by filtration and then suspended in 30 mL of filtered seawater. A 1 mL suspension of each sample was inoculated into 100 mL of artificial seawater medium ONR7a (Dyksterhouse et al., 1995) supplemented with 0.1% (w/v) CAR and incubated at 30 °C on a rotary shaker at 150 r.p.m. Positive strains were chosen based on their ability to color the medium yellow; the chosen strains were enriched fourfold by growth in the same medium. The culture was diluted and spread onto ONR7a agar plates supplemented with CAR. Colonies that were surrounded by clear zones, as a result of CAR degradation, were transferred onto Marine Broth 2216 (Difco Laboratories, Detroit, MI) agar plates supplemented with CAR to isolate the CAR-degrading bacteria completely. Instead of ONR7a, natural seawater medium (Fuse et al., 2003) (designated SEA medium) was also used for isolation of CAR-degrading bacteria.

Oxygenase and meta-cleavage enzyme tests

Oxygenase activity was examined by monitoring the conversion of indole to indigo. All isolates were grown on marine broth agar plates supplied with indole vapor and a change in medium color to indigo blue was sought. The meta-cleavage enzyme activity was examined by monitoring the conversion of 2,3-dihydroxylbiphenyl to 2-hydroxy-6-oxo-6-phenyl-hexa-2,4-dienoic acid. All isolates were grown on marine broth agar plates and sprayed with 2,3-dihydroxylbiphenyl-acetone solution. A change in medium color to yellow indicating meta-cleavage activity was sought.

PCR, sequencing, and phylogenetic analysis

Total DNA was extracted from isolates grown on marine broth medium supplemented with CAR at 30 °C using standard protocols (Sambrook & Russell, 2001). The partial 16S rRNA gene sequence of isolates was amplified by PCR using the primers 10f (Escherichia coli numbering system: 5′-AGTTTGATCCTGGCTC-3′) as the forward primer and 1500r (E. coli numbering system: 5′-GGCTACCTTGTTACGA-3′) as the reverse primer. The partial 16S rRNA gene sequence was amplified by PCR as follows: 30 s at 94 °C; 30 cycles of 30 s at 94 °C, 30 s at 60 °C, and 2 min at 72 °C, and a final extension at 72 °C for 5 min.

The genes of the terminal oxygenase of CARDO were amplified by PCR using carAa sharing primers (designated CS primer) designed from carAa of strain CA10 and Sphingomonas sp. strain CB3 (Habe et al., 2002). Additionally, an AJ primer set (AJ025/26) (Armengaud et al., 1998) was used for the large subunit of the terminal oxygenase that was related to aromatic compound degradation. The genes were amplified by Touchdown PCR using CS and AJ primer sets. Touchdown PCR was conducted as follows: 5 min at 95 °C; 30 cycles of 30 s at 95 °C, 30 s at 58–48 °C (the initial annealing temperature was 58 °C and the temperature was decreased by 1 °C each cycle), and 45 s at 72 °C; 20 cycles of 30 s at 95 °C, 30 s at 48 °C, and 45 s at 72 °C; and a final extension at 72 °C for 5 min.

The PCR products were cloned into the pT7Blue T-vector (Novagen, Madison, WI) for sequence analysis. The nucleotide sequences were determined using the Big Dye terminator cycle sequencing kit (Applied Biosystems, Tokyo, Japan) and an ABI PRISM 310NT genetic analyzer according to the manufacturer's instructions. dnasis-pro software (version 2.8; Hitachi Software Engineering, Kanagawa, Japan) was used to analyze the nucleotide sequences and to produce phylogenetic trees of homologous sequences (Figs 1 and 2); the DDBJ/EMBL/GenBank databases were searched using the blast program.

Southern hybridization

Southern hybridization was performed using the digoxigenin-DNA labeling and detection kit (Roche Diagnostics) according to the manufacturer's instructions. For all isolates, the total DNA was digested with EcoRI. Electrophoresis was conducted in 0.9% (w/v) agarose gels, and DNA was transferred onto Biodyne B nylon membranes (Pall, NY). The car gene clusters of strains CA10, KA1, and IC177 were digoxigenin labeled and used as probes. The car_CA10 probe containing carAaBaBbCAcAd and partial carD was prepared from a 6.9-kb EcoRI fragment from pUCA1 (Sato et al., 1997b). The car_KA1 probe containing carRAaBaBbCAc was prepared from a 4.5-kb XhoI–HindIII fragment from pBKA102 (Inoue et al., 2004). The 4.6-kb car_IC177 probe containing carAaCBaBbAcAd was prepared from plasmid pSC1770033 (Inoue et al., 2006) by PCR using custom primers (5′-ATGAGCACCTCTCAGGAAAT-3′ and 5′-CTATGCGATCAGCGACGTCA-3′). The PCR product was cloned into the pT7Blue T-vector and sequenced. Prehybridization and hybridization were carried out at 55 °C (hybridizing conditions for similarity >60%) in hybridization buffer containing 5 × SSC (0.75 M NaCl+0.075 M sodium citrate), 1% (w/v) blocking reagent (Roche Diagnostics), 0.2% sodium dodecyl sulfate (SDS), and 0.1% lauroyl sarcosine. After hybridization, the membrane was washed twice with 2 × SSC containing 0.1% SDS for 5 min at room temperature and twice with 0.1 × SSC containing 0.1% SDS for 15 min at 55 °C.

Characterization of CAR degradation intermediates

The isolates were analyzed for CAR degradation intermediates by GC-MS. Each isolate was grown on marine broth agar plates supplied with CAR. The colony of each isolate was inoculated in 5 mL of SEA medium containing 0.03% CAR (w/v) and incubated at 30 °C for 3 days. Each culture was extracted with ethyl acetate after acidification to pH 2 with 1 N HCl and dried over anhydrous sodium sulfate. Each extract was derivatized by trimethylsilylation with 2,2,2-trifluoro-N-methyl-N-trimethylsilyl-acetamide (MSTFA) at 70 °C for 20 min and analyzed using a Finnigan TRACE DSQ GC-MS quadrupole mass spectrometer (Thermo Electron, Kanagawa, Japan) equipped with a splitless injector. A capillary column INERTCAP-5 (0.25 mm inside diameter by 15 m, 0.25 μm film thickness; GL Science, Tokyo, Japan) was used as the analytical column. Each sample was injected into the column at 80 °C in the splitless mode. After 2 min at 80 °C, the column temperature was increased to 280 °C at 16 °C min⁻¹. The head pressure of the helium carrier gas was 65 kPa.

Nucleotide sequence accession numbers

The nucleotide sequences reported here were registered in the DDBJ, EMBL, and GenBank nucleotide sequence databases. The nucleotide sequences of the 16S rRNA gene are OC3, AB429066; OC4, AB429067; OC5, AB429068; OC5S, AB429069; OC6, AB429070; OC6S, AB429071; OC7, AB429072; OC8S, AB429073; OC9, AB429074; OC10, AB429075; and OC11S, AB429077. The nucleotide sequence of the PCR product amplified from strain OC5S using the CS primer set is AB429285. The nucleotide sequences of the PCR products amplified from strains OC5, OC6, and OC7 using the AJ primer set are AB429286, AB429287, and AB429288, respectively.

Results and discussion

Isolation of novel CAR-degrading bacteria from seawater

We isolated 11 CAR-utilizing bacterial strains from seawater using ONR7a and SEA as basal media (Table 1). Five strains (OC3, OC4, OC5, OC6, and OC10) were isolated using ONR7a medium, and four strains (OC5S, OC6S, OC8S, and OC11S) were isolated using SEA medium. Two isolates from a seawater sample from Toyama that were cultured in ONR7a and SEA medium were assumed to be the same strain because of identical colony color and morphology; furthermore, the repetitive sequence-based PCR rep-PCR (de Bruijn, 1992) band patterns were identical (data not shown). Similarly, it was also assumed that strains obtained from two isolates from a sample from Fukuoka were the same. The CAR-degrading bacteria isolated by ONR7a medium were chosen for further analysis, and isolates from Toyama and Fukuoka were designated as strains OC7 and OC9, respectively. Rep-PCR also indicated that two distinct CAR-utilizing bacteria, strains OC5 and OC5S, were isolated from the same seawater sample using ONR7a and SEA media, respectively. These results suggested that strains OC5 and OC5S were dominant in ONR7a and SEA media, respectively, as also observed for samples collected from Hyogo (strains OC6 and OC6S). Thus, the use of different media is an effective method to isolate distinct CAR-degrading bacteria from the same source. It is known that most marine microorganisms are unculturable (Amann et al., 1995), and there may be several unculturable CAR-degrading bacteria in marine environments. A previous study has shown that medium plays a significant role in these strain isolations (Sørheim et al., 1989; Aagot et al., 2001; Janssen et al., 2002). It is likely that additional CAR-degrading bacteria may be isolated using other types of media.

In terms of 16S rRNA gene similarity, the isolates, except strains OC3, OC6S, OC9, and OC11S, were most closely related to the genera Erythrobacter, Hyphomonas, Sphingosinicella, Caulobacter, and Lysobacter (>97% similarity; Table 1). Genera Erythrobacter, Hyphomonas, Caulobacter, and Lysobacter have been isolated from marine environments (Abraham et al., 1999; Weiner et al., 2000; Du et al., 2006; Romanenko et al., 2008), although there is no known report of Sphingosinicella isolated from marine sample. As for CAR-degrading bacteria, only Erythrobacter are known to be CAR-utilizing bacteria (Inoue et al., 2005). To our knowledge, this is the first report showing that the genera Hyphomonas, Sphingosinicella, Caulobacter, and Lysobacter belong to CAR-degrading bacteria. Strains OC3, OC6S, OC9, and OC11S showed no similarity to previously identified genera; it was impossible to classify these isolates, although the closest-known genus was Kordiimonas gwangyangensis GW14-5 isolated from a marine sediment (Kwon et al., 2005), and sequence analysis indicated that these four isolates were closely related to each other (>97% similarity).

Various terrestrial bacteria belonging to Pseudomonas and Sphingomonas show CAR-degrading ability (Ouchiyama et al., 1993; Hisatsuka & Sato, 1994; Gieg et al., 1996; Shepherd & Lloyd-Jones, 1998; Habe et al., 2002; Inoue et al., 2004, 2005). Our phylogenetic analysis showed that the marine isolates (strains OC3, OC6S, OC9, and OC11S) formed a new group of CAR-degrading bacteria (Fig. 1). There may be many other CAR-degrading bacteria belonging to the new genus formed by the four isolates in marine environments. This new genus may also play an important role in the degradation of aromatic compounds such as biphenyl, CAR, dibenzofuran, naphthalene, and phenanthrene (Dyksterhouse et al., 1995; Geiselbrecht et al., 1998; Hedlund et al., 1999; Kasai et al., 2002; Fuse et al., 2003), and the isolates described here were not similar to Cycloclasticus and Neptunomonas (Fig. 2). Therefore, further research on this novel genus may help develop bioremediation strategies for marine environments.

Oxygenase and meta-cleavage enzyme activity

Indole and 2,3-dihydroxylbiphenyl were used to detect oxygenase and meta-cleavage enzyme activity, respectively (Table 2). In the indole test, most isolates changed the color of the medium to indigo blue. Oxygenase activity for strains OC5S and OC10 could not be determined because the medium turned brown. In the 2,3-dihydroxylbiphenyl test all strains, except OC4 and OC8S, turned the medium yellow. For strains OC4 and OC8S, GC-MS analysis revealed that CAR was degraded to anthranilic acid (described in Characterization of CAR degradation intermediates), indicating angular dioxygenation and meta-cleavage (Ouchiyama et al., 1993). However, it is assumed that 2,3-dihydroxylbiphenyl could not penetrate cell membrane of strains OC4 and OC8S, which resulted in the medium not turning yellow. Whether these two isolates possess meta-cleavage enzymes has not been determined at this stage.

PCR analysis for aromatic dioxygenases

PCR products were amplified using the CS primer set for strains CA10, CB3, KA1, and IC177, which produced amplicons of 810, 918, 1100, and 991 bp, respectively (Habe et al., 2002; Inoue et al., 2005). In this study, the CS primer set amplified a product of 1103 bp from strain OC5S alone, and sequence analysis revealed that this amplified fragment contained two potential ORFs. The deduced amino acid sequence of the first ORF showed 98% similarity to the terminal oxygenase of CARDO, and the second ORF showed 98% similarity to the small subunit of 2′-aminobiphenyl-2,3-diol 1,2-dioxygenase from strain KA1.

PCR was then performed using the AJ primer set for the dibenzo-p-dioxin-degrading bacterium Sphingomonas sp. strain RW1, which produces an amplicon of 434 bp (Armengaud et al., 1998). In this study, amplicons of 422 bp were obtained from strains OC5, OC6, and OC7. The deduced amino acid sequences of OC5 and OC6 products showed 100% similarity to each other and 73% similarity to the large subunit of biphenyl 2,3-dioxygenase from the biphenyl-degrading bacterium Pseudomonas sp. strain KKS102. Strains OC5 and OC7 showed only 86% similarity to each other, and strain OC7 showed 78% similarity to the large subunit of biphenyl 2,3-dioxygenase from the biphenyl-degrading bacterium Comamonas testosteroni TK102. Also, PCR products from strains OC5, OC6, and OC7 may be related to CAR degradation because the terminal oxygenase of the CAR-degrading bacterium from strain CB3 showed similarity to biphenyl dioxygenase (Shepherd & Lloyd-Jones, 1998). In addition, the deduced amino acid sequences of products from OC5 and OC6 showed 46% similarity, while products from OC7 showed 48% similarity to the terminal oxygenase of strain CB3. Further studies are necessary to determine whether these genes function in CAR degradation or facilitate metabolism of other compounds.

Southern hybridization analysis

Southern hybridization was performed at 55 °C (hybridization conditions for similarity >60%; Table 2). Strain OC3 hybridized with the car_CA10 probe, strain OC6 hybridized with the car_KA1 probe, and strains OC4, OC5S, and OC8S hybridized with both car_CA10 and car_KA1 probes. These results suggested that the car_CA10 and car_KA1 genes are widespread in marine bacteria. None of the isolates hybridized with the car_IC177 probe. The remaining six isolates (strains OC5, OC6S, OC7, OC9, OC10, and OC11S) did not hybridize with any of the probes, suggesting that these strains carry genes such as car_CB3 (Shepherd & Lloyd-Jones, 1998) or other unknown car genes. Research on the novel car gene sequences that did not hybridize with any of the probes used is currently in progress.

Characterization of CAR degradation intermediates

The isolates were analyzed for CAR degradation intermediates by GC-MS. Each isolate produced a yellow color change in the medium when supplemented with CAR, indicating meta-cleavage activity, and CAR degradation intermediates, the trimethylsilylation derivative of the COOH group and both the COOH and the NH₂ groups of anthranilic acid, were detected in each culture, similar to those found from strain IC177 (Inoue et al., 2005). The first retention time for compounds trimethylsilylated at the COOH group was 5.84 min, with a mass-fragmentation pattern of 209 (M⁺, 99), 194 (74), 176 (44), 150 (51), 119 (100), 92 (52), 75 (25), and 65 (44). The second retention time, for compounds trimethylsilylated at both the COOH and the NH₂ groups, was 6.88 min, with a mass-fragmentation pattern of 281 (M⁺, 7), 266 (100), and 73 (49). These results suggest that the degradation pathway for CAR to anthranilic acid by the isolates is similar to strain CA10 (Ouchiyama et al., 1993). It is important that the isolates degrade CAR to anthranilic acid because anthranilic acid is an easily degradable and harmless substrate, and anthranilic acid is assimilated for the tryptophan biosynthesis pathway by various organisms (Gibson & Pittard, 1968).

In this study, we obtained 11 marine CAR-degrading bacterial strains from seawater, and two distinct CAR-degrading bacteria were isolated from the same seawater sample using two different media. This finding revealed that the basal medium plays a significant role in strain selection. Our isolates were most closely related to the genera Erythrobacter, Hyphomonas, Sphingosinicella, Caulobacter, and Lysobacter, indicating that CAR-degrading bacteria from several genera can be found in marine environments. With the exception of Erythrobacter, this is the first report that these genera are CAR-degrading bacteria. Notably, strains OC3, OC6S, OC9, and OC11S belong to a new genus and may play an important role in the degradation of aromatic compounds in marine environments. The results of PCR and Southern hybridization showed that most isolates might have new CAR-degradative genes, which are distinct from previously reported genes, and indicate that there may be a greater diversity for microbial degradation of other xenobiotics in the sea. Further research on these isolates is necessary to fully understand how to use these bacteria for marine bioremediation.

Acknowledgements

This research was supported by a Grant-in-Aid for Scientific Research (17380056 to T.O.) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan. We thank Kengo Inoue of the University of Tokyo for providing N. aromaticivorans IC177.

References