Correspondence: Zongze Shao, Key Laboratory of Marine Biogenetic Resources, The Third Institute of Oceanography, State of Oceanic Administration, Daxue Road178, Xiamen 361005, Fujian, China. Tel.: (+86)592 2195321; fax: (+86)592 2085376; e-mail: email@example.com
Many bacteria have been reported as degraders of long-chain (LC) n-alkanes, but the mechanism is poorly understood. Flavin-binding monooxygenase (AlmA) was recently found to be involved in LC-alkane degradation in bacteria of the Acinetobacter and Alcanivorax genera. However, the diversity of this gene and the role it plays in other bacteria remains unclear. In this study, we surveyed the diversity of almA in marine bacteria and in bacteria found in oil-enrichment communities. To identify the presence of this gene, a pair of degenerate PCR primers were was designed based on conserved motifs of the almA gene sequences in public databases. Using this approach, we identified diverse almA genes in the hydrocarbon-degrading bacteria and in bacterial communities from the surface seawater of the Xiamen coastal area, the South China Sea, the Indian Ocean, and the Atlantic Ocean. As a result, almA was positively detected in 35 isolates belonging to four genera, and a total of 39 different almA sequences were obtained. Five isolates were confirmed to harbor two to three almA genes. From the Xiamen coastal area and the Atlantic Ocean oil-enrichment communities, a total of 60 different almA sequences were obtained. These sequences mainly formed two clusters in the phylogenetic tree, named Class I and Class II, and these shared 45–56% identity at the amino acid level. Class I contained 11 sequences from bacteria represented by the Salinisphaera and Parvibaculum genera. Class II was larger and more diverse, and it was composed of 88 sequences from Proteobacteria, Gram-negative bacteria, and the enriched bacterial communities. These communities were represented by the Alcanivorax and Marinobacter genera, which are the two most popular genera hosting the almA gene. AlmA was also detected across a wide geographical range, as determined by the origin of the bacterial host. Our results demonstrate the diversity of almA and confirm its high rate of occurrence in hydrocarbon-degrading bacteria, indicating that this gene plays an important role in the degradation of LC alkanes in marine environments.
Petroleum hydrocarbons are important pollutants of marine environments, although large amounts of oil enter the oceans via natural seafloor seeps (NRC, 2003). Alkanes are saturated hydrocarbons representing the main constituents of crude oil, and they can be linear (n-alkanes), circular (cycloalkanes), or branched (isoalkanes). Among the n-alkanes, short chains tend to evaporate or dissolve in seawater, and both short and medium-chain alkanes found in oil are degraded quickly after spill occurrence. However, the long-chain (LC) alkanes, with more than 18 carbon atoms, do not dissolve and degrade efficiently and thus are present in various environments for longer periods of time. Moreover, LC alkanes are usually in the solid phase at room temperature and may cause problems, such as clogging oil pipes, during oil transportation. Recently, the biodegradation of LC alkanes has started to attract more attention (Wentzel et al., 2007; Yakimov et al., 2007; Wang & Shao, 2010; Liu et al., 2011), and advances in the field have recently been reviewed by Wentzel et al. (2007).
In marine environments, species of the genus Alcanivorax are ubiquitous and have proven to be obligate alkane degraders (Head & Jones, 2006; Yakimov et al., 2007). For instance, Alcanivorax borkumensis SK2 can grow on LC n-alkanes with up to 32 carbon atoms (C32) (Kasai et al., 2001, 2002; Syutsubo et al., 2001), and Alcanivorax dieselolei B-5 and Alcanivorax hongdengensis A-11-3 can utilize n-alkanes with up to 36 carbon atoms (C36) (Liu & Shao, 2005; Wu et al., 2008).
In addition, many other LC-degrading bacteria have been found, and some of these are summarized by Wentzel et al. (2007). Other LC-degrading bacteria have recently been described, including Thalassolituus oleivorans, which utilizes LC n-alkanes obligately with a substrate range of up to C20 (Yakimov et al., 2004); strains of the genus Acinetobacter, which utilizes alkanes with chain lengths ranging from 10 to 44 carbon atoms (Makula et al., 1975; Kleber et al., 1983; Amund & Higgins, 1985; Sakai et al., 1994; Lal & Khanna, 1996; Yuste et al., 2000; Koma et al., 2001; Throne-Holst et al., 2006, 2007); strains of the genus Rhodococcus, degrading LC alkanes up to C36 (Sorkhoh et al., 1990; van Beilen et al., 2002; Kunihiro et al., 2005; Peng & Liu, 2007); and Marinobacter hydrocarbonoclasticus 617, which utilizes n-alkanes from C16 to C30 (Doumenq et al., 2001; Duran, 2010).
Characterization of the genes involved in the utilization of n-alkanes has led to an improved understanding of the bacterial metabolism used to break down these hydrocarbons. Hydroxylases involved in the degradation of short- and medium-length alkanes have been investigated in more detail (van Beilen & Funhoff, 2007). The bacterial cytochrome P450 enzyme and the AlkB-related alkane hydroxylases have been found to be involved in the degradation of C5 to C24 with overlapping ranges (van Beilen et al., 2003; Vaneechoutte et al., 2006; van Beilen & Funhoff, 2007; Wang et al., 2010; Liu et al., 2011). However, these genes are not specific enough to estimate the oil-degradation capacities, as reported by Widdel & Rabus (2001), Kuhn et al. (2009) and Paisse et al. (2011). Thus, the search for specific genes that can determine LC hydrocarbons degradation capacities from a petroleum origin is currently an important research area.
Recently, a gene designated, almA, which encodes a putative monooxygenase of the flavin-binding family, was identified from Acinetobacter sp. DSM 17874 (Throne-Holst et al., 2007; Wentzel et al., 2007). This gene encodes the first experimentally confirmed enzyme to be involved in the metabolism of LC n-alkanes of C32, and longer. almA homologues have also been found in other isolates of the genus Acinetobacter (Wentzel et al., 2007), including Acinetobacter sp. RAG-1 (Reisfeld et al., 1972), Acinetobacter sp. M-1 (Sakai et al., 1994), and Acinetobacter baylyi ADP1 (Vaneechoutte et al., 2006).
To confirm the role of almA in Alcanivorax strains, two homologues of almA in strain A. hongdengensis A-11-3 and strain A. dieselolei B-5 were analyzed using quantitative real-time PCR and heterogenic expression, respectively. Both strains expressed almA at sufficient levels, to facilitate the efficient degradation of LC n-alkanes (Wang & Shao, 2010; Liu et al., 2011). This was the first evidence in Alcanivorax bacteria that almA encoded an enzyme that functions as an alkane hydroxylase.
In addition, similar genes can be found from other genera in GenBank, such as Marinobacter aquaeolei VT8, Oceanobacter sp. RED65, Ralstonia spp., Mycobacterium spp., Photorhabdus sp., Psychrobacter spp., and Nocardia farcinica IFM10152. However, few of these genes have been functionally characterized. In the case of A. borkumensis SK2, a putative almA was identified in the genome sequence (Sabirova et al., 2006; Schneiker et al., 2006), but its function has not been experimentally confirmed.
Hundreds of millions of gallons of petroleum enter the marine environment from both natural and anthropogenic sources every year (NRC, 2003). Marine hydrocarbon-degrading bacteria are quite diverse and widespread (Head & Jones, 2006). To investigate the possibility that bacteria function to degrade oil in a wide geographical range of oceans, surface seawater was collected form the South China Sea, the Indian Ocean, the Pacific Ocean Xiamen coastal areas, and the Atlantic Ocean, and these samples were enriched with oil to select for the presence of oil-degrading bacteria. In this report, a pair of degenerate primers was designed to detect and amplify almA gene sequences of various bacteria. To understand the diversity and role of almA in oil degradation, we surveyed almA in various marine hydrocarbon-degrading bacteria and from a variety wide of geographical locations.
Materials and methods
Sample collection, enrichment, and isolation of oil-degrading bacteria
Seawater samples were collected using conductivity, temperature, and depth (CTD) water sampler (Hydro-Bios, Germany) from the South China Sea, the Indian Ocean, the Pacific Ocean Xiamen coastal areas located at the west bank of the Taiwan Straits (Wanpeng et al., 2010) and from various sites across the Atlantic Ocean (Wang et al., 2010). Once collected, 500 mL of seawater was aliquoted into sterilized bottles. To enrich the capability of the oil-degrading bacteria, the seawater was supplemented with 2 mL of a 1 : 1 mixture of sterilized crude and diesel oils (as a carbon source) and with N, P, and Fe sources to reach similar concentrations as those found in NH medium (Wang et al., 2010). The crude oil come from South China Sea oil well, and it was mainly composed of C15–C44 alkanes (about 60%) and polycyclic aromatic hydrocarbon (about 30%)(unpublished). The crude oil and diesel oil were sterilized, using high-temperature sterilization technology (121 °C, 15 min). The sampling sites and chemical–physical parameters are shown in Supporting Information, Fig. S2 and Table S1.
The preliminary enrichments were promptly transferred to 100 mL of fresh NH medium (with 1% crude oil mixture as a carbon source) with 1% inoculum in 250-ml Erlenmeyer flasks, and the cultures were incubated at 25 °C with shaking (150 r.p.m.) for 10 days. The transfer of enriched cultures (2% v/v) was repeated two times every 10 days. An oil treatment without inoculum was conducted simultaneously as a control. Serial dilutions of enrichments were streaked on M2 agar plates and incubated at 25 °C. Representative colonies were picked and restreaked onto the 216L plates (Wang & Lai, 2008), to obtain pure cultures.
Bacteria and oil-degrading communities
Together, 137 bacteria isolated from the surface seawater of the South China Sea (data not shown) were tested to determine the presence of almA. In addition, DNA of the above oil-degrading communities was also tested to determine the presence of almA.
Additionally, two type strains of different species of Alcanivorax were also used to clone almA, including Alcanivorax jadensis T9T (AJ001150) and Alcanivorax venustensis ISO4T (AF328765). All of these bacteria had not yet been examined for the presence of almA.
Assays for oil biodegradation
The isolates were tested for their ability to utilize crude oil or paraffin. They were first cultivated in 216L liquid medium (Wang & Lai, 2008), cells were collected and washed twice with sterilized ddH2O, and then, approximately 5–7 × 108 cells were inoculated into a 100-ml Erlenmeyer flask containing 25 mL of sterile MM medium (Wang & Lai, 2008) supplemented with 1% (v/v) sterilized diesel oil (mainly composed of C9–C18 straight alkanes) or 1% (w/v) sterilized paraffin wax (mainly composed of C18–C48 straight alkanes) and incubated at 25 °C with shaking (150 r.p.m.) for approximately 1 week. The oil-degrading capability was reflected by the amount of cell growth, which was determined by measuring the OD600 cell densities and comparing these to controls that include the absence of a carbon source and a noninoculated control.
Genomic DNA preparation
To prepare the genomic DNA from the isolates, the cells were grown overnight, harvested by centrifugation, and extracted with the TIANamp Bacteria DNA kit (TianGen).
For preparation of the total DNA from oil-degrading communities, a 3-mL culture of each community was centrifuged, and the cell pellets were resuspended in 120 μL 1× TE (10 mM Tris pH 7.5; 1 mM EDTA) and incubated for 1 h. After the addition of 80 μL of 10% SDS, the samples were incubated at 65 °C for 1.5 h. The final extraction was carried out using the TIANamp Bacteria DNA kit.
PCR detection of AlmA genes
Degenerate primers were used to detect almA genes in both isolates and community DNA. The PCR mix contained a final volume of 50 μL, that is, 5 μL of 10× buffer (provided with Taq polymerase), 1.5 μL MgCl2 (50 mM), 4–5 μL dNTP mix (2.5 mM of each nucleotide), 5–6 μL of each primer (10 μM), 10–15 ng of purified DNA of cultured strains or 15–20 ng of total community DNA, and 2.5 U Taq DNA polymerase (Invitrogen, Karlsruhe, Germany). Cycling was performed with an initial denaturation for 5 min at 94 °C, 30 cycles with 30 s at 94 °C, 30 s at 50 °C, and 60 s at 72 °C, followed by a final elongation step for 10 min at 72 °C. PCR products were separated on 1.0% agarose gels.
PCR product sequencing
The amplification product from a single isolate was purified using the E.Z.N.A. cycle-pure kit (Omega) or excised from the gel and purified with the E.Z.N.A. Gel Extraction kit (Omega). The purified PCR products were ligated into a PMD-19T vector (TaKaRa Bio, China), transformed into Escherichia coli DH5α cells, and rapidly screened by colony PCR for positive transformants.
PCR products from community DNA were inserted into the vector pMD19-T vector and then transformed into E. coli DH5α. The libraries were screened via restriction analysis by digestion with HaeIII and MspI (TakaRa Bio, China), and clones with a unique restriction pattern were sequenced on an automated DNA sequencer (model 377), using a BigDye Terminators Cycle Sequencing kit (Applied Biosystems).
The obtained sequences were aligned using dnaman (version 5.1; Lynnon Biosoft) with AlmA sequences retrieved from GenBank. A phylogenetic tree of the derived protein sequences was constructed with the neighbor-joining method (Saitou & Nei, 1987), using the phylip and dnaman packages. Bootstrapping analysis was used to evaluate the tree topology of the neighbor-joining data by performing 1000 resamplings.
Nucleotide sequence accession numbers
Sequence data were submitted to the GenBank database. The AlmA gene sequences are deposited under accession numbers FJ263098–FJ263196.
Designing degenerate PCR primers of almA
To search for conserved motifs for primer design, several putative, full-length AlmA sequences were retrieved from GenBank, and these were aligned with clustalw using the identity matrix for pairwise alignment and the Gonnet matrix for multiple alignment (Thompson et al., 1997) (Fig. S1). Among the 20 sequences used, those from Acinetobacter sp. DSM 17874 (AlmA gene accession No. ABQ18224), Acinetobacter sp. ADP1 (CAG69876), Acinetobacter sp. M-1 (ABQ18228), and Acinetobacter sp. RAG-1 (ABQ18226) have been confirmed as encoding enzymes that function in n-alkane degradation (Throne-Holst et al., 2007). The other sequences considered were putative AlmA genes, including A. borkumensis SK2 (CAL15730, almA accession No.), A. borkumensis SK2 (CAL15638), M. aquaeolei VT8, (ABM20899), Marinobacter algicola (EDM48565), Mycobacterium leprae strain TN (CAC29573, Mycobacterium gilvum PYR-GCK (ABP43446), N. farcinica IFM 10152 (BAD57540), Oceanobacter sp. RED65 (EAT10988), Psychrobacter cryohalolentis K5 (ABE76067), Ralstonia eutropha JMP134 (AAZ64503), Limnobacter sp. MED105 (EDM84915), Parvibaculum lavamentivorans DS-1 (ABS62476), Pseudomonas aeruginosa PA7 (ABR84379), and Sphingopyxis alaskensis RB2256 (ABF52436). In addition, two putative flavin-binding monooxygenases were also included from two fungi: Neosartorya fischeri NRRL 181 (EAW20007) and Aspergillus clavatus NRRL 1 (EAW13597).
Sequence alignment identified six regions, named A to F, that were well conserved within the alignment of the AlmA protein amino acid sequences (Fig. S1). These were I(V/I)GAGXXG, GGTWLF(R/K)YPGIRSDSD, IGSGATA, TMLQR(S/T)P, ADI(I/V)(V/I)(T/S)ATGL, and GY(I/T)N(A/I)(S/P)WTL. The N-terminal GxGxxG sequence was indicative of the βαβ-fold (or Rossmann fold) that binds the ADP moiety of FAD (Weienga et al., 1986). Regions in the second and last conserved motifs were selected for developing a pair of degenerate PCR primers. The oligonucleotide sequences that were designed are as follows: AlmAdf5′-GGNGGNACNTGGGAYCTNTT-3′ and AlmAdr 5′-ATRTCNGCYTTNAGNGTCC-3′. The expected size of the PCR product was approximately 1100 bp. The specificity of the primers was examined using blast to query public databases, and the primers were further tested using reference strains of known genotypes, including Acinetobacter sp. M-1, Acinetobacter sp. ADP1 and RAG-1, and A. borkumensis SK2 (Sakai et al., 1994; Schneiker et al., 2006; Throne-Holst et al., 2007; Wentzel et al., 2007). All of the preliminary tests supported a successful design.
AlmA genes from marine bacteria
In total, 169 bacterial isolates, comprising 15 genera, were obtained from 10 oil-degrading communities enriched from seawater samples collected from the South China Sea, the Pacific Ocean, and the Indian Ocean. To analyze the diversity of almA in these marine bacteria, PCR was performed using the degenerate primers described earlier. As a result, one to three copies of almA were positively detected in 35 bacterial species belonging to four genera. Moreover, three almA sequences (approximately 1100 bp) were obtained from the two strain types of the genus Alcanivorax, two of which were from A. jadensis T9T, while one was from A. venustensis ISO4T (Table 1). In total, 28 unique almA sequences were derived from 26 strains isolated from the South China Sea, and these strains mainly belonged to the α- and γ-proteobacteria. In addition, eight almA sequences were obtained from 5 of 28 isolates from the surface saewater of the Indian and Pacific Oceans (Table 1). The ability to degrade alkane was initially predicted by detecting the presence of genes encoding AlmA(s) and then confirmed by the assessment of the ability of the bacteria encoding AlmA to grow on diesel oil or paraffin wax (Table 1).
Table 1. Marine bacteria used in this report and their AlmA genes
++++, +++, ++, +, and +/−, indicating the growth capability from strong to weak with diesel oil or crude oil as sole carbon and energy source, measured by optical density at 600 nm.
NT, not tested; ++++, growth (OD600 > 1) after a 10-day incubation at 25 °C; +++, growth (OD600 > 0.6) after a 10-day incubation at 25 °C; ++, growth (0.6 > OD600 > 0.2) after a 10-day incubation at 25 °C; +, growth (OD600 < 0.2) after a 10-day incubation at 25 °C; −, no growth; MCCC, Marine Culture Collection of China.
Sources of ‘bacteria': (a)–(h) indicate the marine areas. (a) Bohai Sea (China); (b) the Straits of Malacca and Singapore; (c) German North Sea coast; (d) Santa Pola coast; (e) South China Sea; (f) Austalia Sea; (g) Pacific Ocean; (h) Indian Ocean.
Growth capability with paraffinic wax or crude oil in MM medium.
More almA sequences were detected directly from oil-degrading communities using community DNA as the PCR template. These communities were previously enriched with surface seawater from the Xiamen coastal areas (Taiwan Straits) (Wanpeng et al., 2010b) and from across the Atlantic Ocean (Wang et al., 2010). A total of 26 different sequences were obtained from the Xiamen coastal area, and these were named AlmA-1, AlmA-4, and so on. A total of 34 sequences were obtained from the Atlantic Ocean, and these were named AlmAS1-2, AlmAS2-3, AlmAS3-3, AlmAS8-8, and so on (AlmS1-2 indicates clone No. 2 that was derived from sampling site S1 of the Atlantic Ocean). All of the sequences from the isolates and community DNA shared homology to almA as determined using blast, to compare the sequences to known genes in public databases. In addition, these sequences possessed the conserved motifs that we defined for AlmA. Some of these sequences were more divergent, with shared amino acid sequence identities as low as 50%, compared to the AlmA of Acinetobacter sp. DSM 17874.
Multiple AlmA genes in one bacterium
As mentioned earlier, occasionally multiple almA genes occurred in a single host (Table 1). Although this phenomenon was believed to be rare, all strains containing multiple almA genes were assessed. Consequently, five isolates were confirmed to harbor two to three AlmA genes. This phenomenon appeared to occur more frequently in Alcanivorax isolates, including isolate L52-11-25B (A. jadensis, 98%, from the South China Sea), isolate 407-1 (Alcanivorax sp., from the Indian Ocean), the type strain A-11-3 of A. hongdengensis (isolated from the Straits of Malacca), and the type strain of A. jadensis T9T (from the German North Sea coast). In contrast, the type strain B-5 and ISO4 strains carried only one almA gene.
Phylogenetic analysis of AlmA genes
A phylogenetic tree was constructed based on the deduced amino acid sequences of the AlmA sequences together with those retrieved from GenBank (Fig. 1). The almA genes obtained in this study were quite diverse. As shown in Fig. 1, all of the sequences formed two major clades, named Class I and Class II. The deduced peptides in Classes I and II shared 45–56% identity with each other at the amino acid level. Two AlmAs that originated in fungus formed a distantly related branch adjacent to the bacterial group (Fig. 1), and these were used as the outgroup reference.
Class I contained 11 sequences from isolates Salinisphaera sp. w510-1, Parvibaculum sp. 322-6, Alcanivorax sp. 407-1 and the sequences from communities of the Xiamen coastal area and the Atlantic Ocean. They were closely related to the AlmAs found in P. lavamentivorans DS-1 (ABS2476) and Sphingobium alaskensis RB2256 (ABF52436) that were retrieved from GenBank.
Class II was a larger group that was more diverse than Class I. It was composed of 88 sequences obtained from Proteobacteria, Gram-positive bacteria, and the enriched communities. Within Class II, the sequences formed three major groups, namely Groups I to III. For instance, nine sequences obtained from our nine Marinobacter isolates and two putative monooxygenases of M. aquaeolei VT8 (ABM20899) and of M. algicola (EDM48565) formed a highly concentrated cluster, named Group II. Our isolates were phylogenetically related to five species of the genera, including Marinobacter koreensis (isolates L53-1-2, L54-11-46, L54-11-32, and L54-11-13), Marinobacter alkaliphilus (isolates L52-11-18 and L53-10-4), Marinobacter flavimaris (isolate L53-10-3), Marinobacter segnegenens (isolate L52-11-17), and Marinobacter lipolyticus (isolate 324-2). In addition, Group II also contained several sequences obtained from other genera, such as isolate Bacillus flexus L51-10-5, as well as those from the enrichments of the Xiamen Island coast and the Atlantic Ocean. Sequences in this group shared similarities of approximately 84–100% with each other at amino acid level.
Also in Class II, a cluster named Group I occurred, which consisted of sequences mainly from Alcanivorax strains, including A. borkumensis SK2, A. dieselolei B-5, A. hongdengensis A-11-3, A. venustensis ISO4, and A. jadensis T-9 (Fig. 1a and b). The AlmAs in this Alcanivorax group shared approximately 66%–96% identity at the amino acid level with each other. Within this group, the biggest subgroup was Cluster III, which constituted 20 homologues from nine isolates and the communities (Fig. 1b). Sequences in this cluster shared a high identity (> 96%) and thus formed a very tightly condensed branch.
In addition to Cluster III, four other clusters were included in Group I (Fig. 1a and b). Within Cluster IV, 13 AlmA sequences that were closely related (> 95% identity) to AlmA2 of A. hongdengensis A-11-3 (Fig. 1a) and six AlmA sequences closely related to AlmA1 of strain A-11-3 (> 90% identity) formed two clades (Fig. 1b). Ten AlmA homologues closely related to A. borkumensis SK2 putative AlmA (CAL15730) and five homologues closely related to another putative AlmA (CAL15638) of SK2 formed the two ‘SK2' clusters; in Fig. 1a and b, respectively.
Also within Class II, the Group III clade formed, but it was closely related to the above groups. It contained the two AlmA homologues (AlmA-28 and AlmA-29) that were identical to that of Acinetobacter sp. DSM 17874 (Throne-Holst et al., 2007) in addition to those from Acinetobacter sp. DSM 17874, Acinetobacter sp. ADP1, and Acinetobacter sp. M1.
In this study, we surveyed the diversity of the flavin-binding monooxygenase-encoding gene, almA, in marine oil-degrading bacteria either as pure isolates or directly from community DNA. The almA gene was quite diverse and most frequently occurred in bacteria of two genera: Alcanivorax and Marinobacter. However, this gene was present most frequently in bacterial species of the Alcanivorax genus that have been recognized as ubiquitous obligate marine bacteria (Head & Jones, 2006).
AlmA was detected across a wide geographical range, as determined by the origin of the bacterial hosts that were detected in a variety of oceanic environments (Fig. S2). Similar or identical genes were detected in bacteria from different geographical locations. For instance, Cluster III (Fig. 1b) contained 20 AlmA genes that were detected from the Atlantic Ocean, the South China Sea, and the coastal area of Xiamen, as well as some type strains that were isolated from the Bohai Sea of China (strain B-5) and along the coast of the Mediterranean Sea (strain ISO4). Similarly, genes in the two SK2 clusters and two A-11-3 clusters also had a wide geographical origin. For example, the 15 AlmA genes that formed Cluster IV (Fig. 1b) originated from the Atlantic Ocean, the South China Sea, the Xiamen Island coast, the Indian Ocean, the German North Sea coast (strain T9), and the Straits of Malacca (strain A-11-3). Thus, it is apparent that the AlmA genes in Clusters III and IV are more prevalent in marine environments than others. It may be that these genes contribute to the host's ability to spread globally by oxidizing the ubiquitous alkanes, especially the LC alkanes.
Together, 22 isolates of the Alcanivorax genus were determined to have the almA gene. These strains were closely related to A. dieselolei, A. borkumensis, A. venustensis, A. hongdengensis, and A. jadensis, in addition to several unclassified novel species of this genus. The growth test proved that these isolates were degraders of diesel oil and paraffin wax (Table 1). Moreover, the AlmAs of these strains were similar to, and in some cases even identical to, the AlmAs of Alcanivorax type strains, including A. borkumensis SK2T, A. jadensis T9T, A. venustensis ISO4T, A. hongdengensis A-11-3T, or A. dieselolei B-5T. Interestingly, many Marinobacter spp. have been reported as hydrocarbon degraders in marine environments, such as M. hydrocarbonoclasticus 617 and M. aquaeolei Vt8, and have been shown to harbor alkB genes (Doumenq et al., 2001; Duran, 2010); however, the alkane monooxygenase genes of these bacteria have not yet been examined for functionality. In this report, almA was found in nine isolates of this genus. Moreover, growth tests proved that these isolates degraded LC alkanes (Table 1). To confirm the role of the almA gene in the degradation of n-alkanes, the expression of almAs in M. koreensis L53-1-2, M. alkaliphilus L53-10-4, M. alkaliphilus L52-11-18, and M. lipolyticus 324-2 was subjected to analysis with quantitative real-time polymerase chain reaction (Q-RT-PCR). Significantly enhanced expression of the AlmA gene was observed in four strains when grown in the presence of LC alkanes (C24–C36) and pristane, but not observed in the case of C12–C22 alkanes (Fig. S3).
AlmA was detected for the first time in some novel bacterial species. For example, the isolate Salinisphaera sp. w510-1 represents a potential novel species of Salinisphaera with highest similarity to Salinisphaera dokdonensis (97%). The growth test proved that this isolate had the ability to degrade oil and paraffin wax (Table 1). In addition, it contained the AlmA gene, which shared 63% similarity with that of P. lavamentivorans DS-1 (ABS2476). Other potential novel species, including Marinobacter sp. (97%) and Parvibaculum sp. (95%), also harbored AlmA genes.
Genes involved in LC-alkane degradation seem to have evolved specifically for the degradation of long chains. In Acinetobacter sp. M1, three genes involved in LC alkane degradation were found: alkMa, which was induced by LC n-alkanes > C22; alkMb, which was preferentially induced by n-alkanes with chain lengths of C16 to C22 (Tani et al., 2001); and a flavin-containing n-alkane dioxygenase, with a proposed substrate range from C10 to C30 (Maeng et al., 1996). Acinetobacter sp. DSM 1784 was first reported to encode an enzyme involved in the degradation of n-alkanes longer than C30, which was named AlmA (Throne-Holst et al., 2007). This bacterium could degrade n-alkanes with a chain length of C32 or longer. The Q-PCR results in our previous report confirmed that the AlmA genes of A. hongdengensis A-11-3 and A. dieselolei B-5 were involved in the degradation of LC alkanes (Wang & Shao, 2010; Liu et al., 2011). In this report, the diversity and biogeography of almA genes were surveyed in marine hydrocarbon-degrading bacteria across a wide geographic distance. Furthermore, the role of almA was confirmed in alkane degradation in Marinobacter spp. (Data S1). In addition to AlmA, other alkane hydroxylases involved in LC n-alkane degradation were also reported, such as a thermophilic soluble LC-alkane monooxygenase (LadA), which exerts terminal oxidation on LC-alkane substrates ranging from C15 to C36 in Geobacillus thermodenitrificans NG80-2 (Feng et al., 2007).
In contrast, some isolates that we tested did not encode an AlmA gene, but these did grow in the presence of LC alkanes, such as isolate Acinetobacter sp. 326-6 (from the pelagic area of the Indian Ocean), Alcanivorax sp. L52-1-16 and L52-11-24 (South China Sea). Similarly, other previously reported LC-alkane degraders were poorly understood about the degradation mechanism, including Alcaligenes (Lal & Khanna, 1996), Bacillus (Kato et al., 2001; Chaerun et al., 2004), Arthrobacter (Radwan et al., 1996), Burkholderia (Yuste et al., 2000), Pseudomonas (Chaerun et al., 2004), and Rhodococcus (van Beilen et al., 2002). Thus, further studies that focus upon searching for and characterizing the related alkane monooxygenases will help to further our understanding.
In summary, a diverse population of almA genes was found in various marine bacteria using a pair of degenerate primers designed in this report. The gene sequences we identified formed specific groups according to the almA gene phylogeny, and these were mainly represented by several genera, including Alcanivorax, Marinobacter and Acinetobacter as well as Salinisphaera and Parvibaculum. The Alcanivorax genus contains the largest group of bacterial species that encode the almA gene. The results of this report demonstrate that the almA gene is widely distributed in marine alkane-degrading bacteria and implies this gene plays an important role in the degradation of LC alkanes in marine environments.
This work was financially supported by the National Science Foundation of China (41176154, 41106151), International Sci & Tech Cooperation Program of China (2010DFB23320), COMRA project (No. DY125-15-1), and the Project sponsored by the Scientific Research Foundation of Third Institute of Oceanography, SOA (2011036).