Assessment of polycyclic aromatic hydrocarbon biodegradation potential in mangrove sediment from Don Hoi Lot, Samut Songkram Province, Thailand

Authors

  • C. Muangchinda,

    1. Bioremediation Research Unit, Department of Microbiology, Faculty of Science, Chulalongkorn University, Bangkok, Thailand
    Search for more papers by this author
  • R. Pansri,

    1. Bioremediation Research Unit, Department of Microbiology, Faculty of Science, Chulalongkorn University, Bangkok, Thailand
    Search for more papers by this author
  • W. Wongwongsee,

    1. Bioremediation Research Unit, Department of Microbiology, Faculty of Science, Chulalongkorn University, Bangkok, Thailand
    2. Microbiology Program in Science, Graduate School, Chulalongkorn University, Bangkok, Thailand
    Search for more papers by this author
  • O. Pinyakong

    Corresponding author
    1. Center of Excellence for Environmental and Hazardous Waste Management (EHWM), Bangkok, Thailand
    • Bioremediation Research Unit, Department of Microbiology, Faculty of Science, Chulalongkorn University, Bangkok, Thailand
    Search for more papers by this author

Correspondence

Onruthai Pinyakong, Department of Microbiology, Faculty of Science, Chulalongkorn University, Bangkok 10330, Thailand.E-mail: onruthai@gmail.com

Abstract

Aims

To assess the biodegradation potential of mixed polycyclic aromatic hydrocarbons (PAHs) in mangrove sediments.

Methods and Results

Sediment microcosms were constructed with sediment collected from Don Hoi Lot, Samut Songkram Province, Thailand, by supplementation with a mixture of acenaphthene, phenanthrene and pyrene. At the end of 8 weeks, low molecular weight PAHs, acenaphthene and phenanthrene were completely degraded. PCR-denaturing gradient gel electrophoresis profile suggests that Marinobacter, Enterobacter and Dethiosulfatibacter play important roles in PAH degradation in mangrove sediment. Furthermore, six PAH-degrading bacteria were isolated consisting of novel phenanthrene-degrading Dyella sp. and Luteibacter sp., phenanthrene-degrading Burkholderia sp., acenaphthene-degrading Alcaligenes sp. and pyrene-degrading Ochrobactrum sp. Moreover, dioxygenase genes could be detected both in sediment microcosms as well as in all of the isolated strains.

Conclusions

These results demonstrated that indigenous bacteria in the mangrove sediment had the ability to degrade phenanthrene in the presence of mixture PAHs with high efficacy.

Significance and Impact of the Study

Culture and nonculture methods were combined to assess PAH biodegradation in mangrove sediment. Two novel phenanthrene-degrading bacteria were isolated. Three genera of bacteria that play important roles in PAH degradation were indicated by nonculture approach. Moreover, dioxygenase genes could be detected. This information is useful for further bioremediation of PAH-contaminated mangrove sediments.

Introduction

Mangroves are important intertidal wetlands along the coastlines of tropical and subtropical regions and have been considered significant sinks for pollutants from tidal water, river water and land-based sources (Bernard et al. 1996). Don Hoi Lot is one of the most famous tourist attractions of Samut Songkram Province, Thailand. As a major urban area, it is a community with a dense population and is an important industrial zone for the country. For these reasons, the mangrove area of Don Hoi Lot may become contaminated with pollutants, including polycyclic aromatic hydrocarbons (PAHs).

PAHs are widespread in the environment and affect the health of aquatic life and humans (Johnsen et al. 2005). Acenaphthene and phenanthrene, which are low molecular weight (LMW) PAHs, and pyrene, which is a high molecular weight (HMW) PAH, are considered priority pollutants by the US Environmental Protection Agency (US EPA) (Cerniglia 1992). In general, the PAHs in the environment are found as a mixture of different PAH compounds. Due to environmental problems and health concerns, it is essential to remove PAH compounds from a contaminated site quickly and effectively. A mixture of acenaphthene, phenanthrene and pyrene is a good model to be used for the study of the biodegradation of mixed PAHs because these chemicals share structures with those of possible carcinogenic PAHs, such as benzo[b]fluoranthene and benzo[a]pyrene (Haritash and Kaushik 2009).

The principal process for the successful removal of PAHs from contaminated environments is microbial transformation and degradation (Lu et al. 2011). Sphingomonas strains have been suggested to be the dominant genera of indigenous PAH-degrading microbes in mangrove sediments (Ho et al. 2000; Daane et al. 2001). Furthermore, several bacteria have been isolated from the environment and have been shown to have the ability to degrade PAHs. These bacteria include Mycobacterium sp. (Churchill et al. 2008), Rhodococcus sp. (Guo et al. 2005) and Marinobacter sp. (Tian et al. 2008).

The biodegradation of PAHs is catalysed by several enzymes from micro-organisms. The key enzyme used to attack the aromatic ring structure of PAHs under aerobic conditions is the initial dioxygenase. The enzyme consists of three components, including ferredoxin, ferredoxin reductase and terminal ring-hydroxylating dioxygenase, which itself is composed of an α- and a β-subunit. The α-subunit of terminal dioxygenase is more conserved than its other components and is essential for substrate specificity, which makes the α-subunit of terminal dioxygenase-encoding genes a useful marker for PAH degradation activity (Andreoni and Gianfreda 2007).

Therefore, the understanding of the microbial community and its catabolic activity are essential for the assessment of its biodegradation potential and for the effective remediation of contaminated areas. Molecular biological techniques have been employed for the detection, quantification and identification of the diversity and structure of microbial communities. For instance, denaturing gradient gel electrophoresis (DGGE) analysis of 16S rRNA genes is typically used to study the composition of bacterial communities (Ferguson et al. 2007; Tian et al. 2008; Guo et al. 2011). In addition, conventional PCR and quantitative real-time PCR have been applied for the detection of catabolic genes in many environmental samples (Cébron et al. 2008; Guo et al. 2010).

This study therefore aims to assess the biodegradation potential of mixed PAHs in mangrove sediments with both culture-independent methods, such as PCR-DGGE and real-time PCR, and a culture-dependent method that requires the isolation of PAH-degrading bacteria. The results from both approaches could provide a greater understanding of the microbial structures that are present in mangrove sediments and may be applied to the further bioremediation of PAH-contaminated mangrove sediments.

Materials and methods

Sediment sample collection and analysis of PAHs

Five surface sediment samples were collected from Don Hoi Lot, Samut Songkram Province, Thailand. The sampling locations (1–5) are shown in Fig. S1. The dominant mangrove plant species in this area are Rhizophora apiculata and Avicennia alba. For PAH analysis, 1 g of sediment was extracted with 4 ml of n-hexane (Merck, Darmstadt, Germany) and 1·5 ml of 15% TritonX-100 (Sigma-Aldrich, Steinheim, Germany) and mixed by shaking at 200 rev min−1 for 24 h. The samples were stored at −20°C for 24 h and then dehydrated by adding anhydrous Na2SO4 (Merck) prior to being sieved through a 0·2-μm PTFE filter. PAH concentrations were quantified by gas chromatography with a flame ionization detector (GC-FID), model 6890 (Agilent Technologies, Palo Alto, CA, USA), using a 30 m × 0·32 mm ID × 0·25 μm – 5% phenyl methyl siloxane capillary column. The recoveries of acenaphthene, phenanthrene and pyrene by using this extraction method were about 40, 100 and 70%, respectively. These recoveries (%) were comparable with those of standard methods (Song et al. 2002). The detection limits of acenaphthene, phenanthrene and pyrene are 0·958, 1·064 and 1·344 mg l−1, respectively.

Sediment microcosm construction

The sediment sample from location no.1 (Fig. S1) used in the construction of the microcosms was analysed for its chemical and physical properties. The water-holding capacity was measured by a comparison between the wet weight and the dry weight. Organic matter, organic carbon, total nitrogen and phosphorus measurements were carried out according to the Walkley–Black method (Walkley and Black 1934), the wet oxidation method (McLeod 1973), the Kjeldahl method (Jackson 1958) and the Bray II method (Bray and Kurtz 1945), respectively. The pH of the sediment samples was measured using a pH metre (Mettler Toledo, Schwerzenbach, Switzerland). Sediment texture was analysed using the hydrometer method (Gee and Bauder 1979).

Sediment microcosms were prepared by transferring 20 g of sediment into glass bottles supplemented with a mixture of acenaphthene, phenanthrene and pyrene (Sigma-Aldrich) at a final concentration of 30 mg kg−1 for each PAH. Stock of each PAH was prepared by dissolving PAH in acetone at the final concentration of 1000 mg l−1, and 600 μl of each PAH stock solution was added into each bottle and was mixed with sterilized spatula. Sterilized sediment that was autoclaved (121°C, 20 min) for three times was used as a control. Microcosms were set up in triplicates at each time point (in total 30 bottles). Three bottles of both sterilized control and experimental sets were sampled every 2 weeks for 8 weeks. Samples from each bottle were used for analysis of residual PAHs (1 g), enumeration of total bacteria (1 g) and PAH-degrading bacteria (1 g) isolation of PAH-degrading bacteria (5 g), extraction of DNA (1 g) for assessment of bacterial community dynamics and detection of PAH catabolic genes (by using combined triplicate DNA sample).

Enumeration of total bacteria and PAH-degrading bacteria

Total bacterial numbers were determined by the most probable number (MPN) method in Luria Bertani medium. The population sizes of PAH-degrading bacteria were also examined by MPN, as described by Johnsen et al. (2002). A 10-fold serial dilution was prepared for each retrieved sample. The 96-well plates for total bacteria and PAH-degrading bacteria population estimates were incubated at 30°C for 2 days and 4 weeks, respectively. The number of total bacteria was enumerated by positive growth in Luria Bertani medium, while the number of PAH degraders was counted based on their growth in a carbon-free mineral medium (CFMM) (Mahanty et al. 2006) supplemented with 500 ppm of individual PAHs in 96-well plates. After incubation at 30°C for 3 weeks, detection was performed using a respiration indicator. The population sizes of total bacteria and PAH-degrading bacteria were retrieved from an MPN table.

Enrichment and isolation of PAH-degrading bacteria

A total of 5 g of sediment taken from each microcosm at different sampling time points was added to 50 ml of CFMM supplemented with individual PAH solutions, either acenaphthene, phenanthrene or pyrene, at a final concentration of 100 mg l−1. The culture flasks were incubated at 30°C with agitation at 200 rev min−1. After 7 days of incubation, the suspensions were left without shaking for 2 h to settle the large sediment particles, after which 5 ml of supernatant was transferred into 45 ml of new CFMM supplemented with individual PAH solutions at the same concentrations and incubated under the same conditions. The cultures were examined for turbidity and colour, which indicated the degradation of PAHs. Then, 5 ml of enrichment culture was transferred into new CFMM containing the same PAHs at the same concentrations and incubated under the same conditions. After serial transfer had been performed five times to enrich the PAH-degrading bacteria, 0·1 ml of culture was spread onto CFMM agar sprayed with 2% PAH solutions in diethyl ether. After incubation, the colonies surrounded by clear zones were selected and purified using 10-fold diluted Luria Bertani medium supplemented with same PAHs.

Identification of the isolated bacterial strains by 16S rDNA analysis

DNA from cells grown on Luria Bertani medium was extracted according to the method described by Ausubel et al. (1993). The 16S rDNA was amplified using primers 27F and 1492R (Lane 1991). The PCR product was separated by agarose gel electrophoresis, and a band of the expected size was purified using the Gel/PCR DNA Fragments Extraction Kit (Geneaid, Taipei, Taiwan). The DNA sequences were determined directly using the primers 27F, 350F (5′-TACGGGAGGCAGCAG), 1100R (5′-AGGGTTGCGCTCGTTG) and 1492R. The 16S rDNA nucleotide sequences were compared with sequences available in GenBank using the Blastn program.

Analysis of bacterial community dynamics

DNA was extracted from sediment samples using the method described by Zhou et al. (2006). Extracted DNA was separated by electrophoresis in a 0·9% agarose gel in 1× TAE buffer and visualized under UV light by staining with ethidium bromide. The genomic DNA was purified using the Gel/PCR DNA Fragments Extraction kit (Geneaid) according to the manufacturer's instructions. Purified DNA was stored at −20°C prior to PCR.

The PCR amplification targeting bacterial 16S rDNA was accomplished using primers 341F with a GC clamp and 520R (Muyzer et al. 1993). PCR was performed in a total volume of 30 μl containing 1× PCR buffer (New England Biolabs, Beverly, MA, USA), 0·5 U Taq DNA polymerase (New England Biolabs), 20 μmol l−1 dNTPs, 20 pmol of each primer and 100 ng of template DNA. Reactions were carried out using an MJ Mini Thermal Cycler (Bio-Rad Laboratories Inc., Hercules, CA, USA). The conditions for PCR amplification were as follows: initial denaturation at 94°C for 5 min, followed by 30 cycles of 94°C for 1 min, 55°C for 1 min and 72°C for 2 min, with a final extension at 72°C for 10 min. The PCR amplification targeting Sphingomonas was amplified using primers Sphingo 108f with a GC clamp and Sphingo 420r (Leys et al. 2004). The conditions for PCR amplification were as follows: initial denaturation at 94°C for 2 min, followed by 30 cycles of 94°C for 15 s, 65°C for 30 s and 74°C for 30 s, with a final extension at 74°C for 10 min. A total of 5 μl of PCR product was run on a 2% agarose gel to verify the size of the PCR product. DGGE was performed using a DCodeTM system (Bio-Rad Laboratories Inc.) maintained at a constant condition of 60°C and 130 V for 5 h in 7 l of 1× TAE buffer. The acrylamide concentration in the gel was 8% with a 30–70% denaturant gradient; 100% denaturant was defined as 7 mol l−1 urea plus 40% (v/v) formamide. The gels were run, then stained with ethidium bromide dissolved in purified water (0·5 mg l−1) for 20 min and visualized under a UV transilluminator using a Gel Doc 2000™ documentation system with Quantity One software (Bio-Rad Laboratories Inc.). The bands excised from the DGGE gels were eluted in distilled water overnight at 4°C. Eluted DNA (1 μl) was used as the template for PCR using primers without the GC clamp. The PCR products were purified from the agarose gel and ligated to a pGEM-T Easy Vector (Promega, Madison, WI, USA). The ligation mixture was transformed into Escherichia coli JM109 competent cells, and clones with inserts were sent to be sequenced using Applied Biosystems 3730 DNA analyzers (Applied Biosystems, Foster city, CA, USA).

Construction of phylogenetic tree

The phylogenetic tree of 16S rDNA sequence of all isolates and dominant DGGE bands with those representative known PAH-degrading bacteria was constructed in MEGA 5.1 using the aligned sequences by the neighbour joining using kimura-2 parameter distances (Tamura et al. 2011).

Detection of dioxygenase genes

The detection of ring-hydroxylating dioxygenase genes involved in PAH degradation in sediment samples was conducted using a PCR technique with five primer sets: PAH-RHDα GPF/GPR, PAH-RHDα GNF/GNR (Cébron et al. 2008), NMR331f/NMR1134r (Marcos et al. 2009), arhA1F/arhA1R (Klankeo et al. 2009) and nidAF/nidAR (Khan et al. 2001). After amplification, the PCR products were purified and cloned using the method described previously. Recombinant plasmids from sediment microcosms that had the correct DNA insert fragments were analysed using restriction fragment length polymorphisms (RFLPs) to group the plasmids that had the same patterns of DNA restriction fragments. The order of usage for the restriction enzymes for RFLPs was BsuRI, HinFI and then RsaI. Selected plasmids were sent to be sequenced and analysed with the Blastx program.

PCR amplification with GNF/GNR primers was also performed to detect ring-hydroxylating dioxygenase genes in PAH-degrading isolates. The PCR products were cloned and analysed as described previously.

Quantification of the dioxygenase genes by real-time PCR

Purified PAH-RHDα GN and 16S rRNA PCR products from Burkholderia sp. FP2-1 and Sphingobium sp. P2 were cloned using the method described previously. After plasmid purification, the plasmid DNA was used to prepare standard curves for the determination of the gene copy numbers of the dioxygenase genes and 16S rRNA genes, respectively. Plasmid concentrations were measured using Quant-iT™dsDNA BR Assay Kits with a Qubit fluorometer (Invitrogen, Carlsbad, CA, USA). A series of 10-fold dilutions of the plasmids was prepared, and these dilutions of the plasmid were amplified along with the DNA samples. Linear regression equations for the cycle threshold values (Ct) were calculated as functions of known plasmid copy numbers. Real-time PCR experiments were performed in a MiniOpticon Real-Time PCR detector with MJ Opticon Monitor Analysis Software (Bio-Rad Laboratories Inc.). The reaction was performed in 0·2-ml-thin wall, clear PCR strip tubes with 25 μl reaction volumes containing Maxima™SYBR Green qPCR Master Mix (Fermentas Life Sciences, Hanover, MD, USA), 0·3 μmol l−1 of primers and 2 μl of template DNA. The PCR conditions were as follows: initial denaturation at 95°C for 10 min, followed by 40 cycles of 1 min of denaturation at 94°C, 1 min at the primers' specific annealing temperature (56 and 57°C for 16S rRNA amplification using 968F/1401R (Felske et al. 1998) and PAH-RHDα GN primer sets, respectively), 1 min of elongation at 72°C and 10 min of a final extension at 72°C. The real-time PCR efficiencies for the primers 968F/1401R and PAH-RHDα GN primer were 97·43 and 94·06%, respectively.

Nucleotide sequence accession numbers

The sequence data for the 16S rRNA genes of isolated bacteria and uncultured bacteria have been deposited in the GenBank nucleotide sequence database under accession numbers JX910135-JX910140 and JX966116-JX966132, respectively. The sequence data of the dioxygenase genes of isolated bacteria and uncultured bacteria have been deposited in the GenBank nucleotide sequence database under accession numbers JX966106-JX966111 and JX966112-JX966115, respectively.

Results

PAH contamination and sediment characterization

Using the method described previously, no contamination of particular PAHs (acenaphthene, phenanthrene and pyrene) was observed in any of the five sediment samples. One sediment sample was used to construct sediment microcosms with artificial PAH contamination to study the biodegradation potential of the mangrove sediment. The ability to degrade mixed PAHs was investigated by the detection of residual PAHs, total bacteria and PAH-degrading bacteria enumeration, analysis of bacterial community dynamics, isolation of PAH-degrading bacteria and detection of dioxygenase genes. The selected sediment was characterized for its properties and was identified as a loam soil. The water-holding capacity was 58·02%. The organic matter, organic carbon and total nitrogen contents were 3·32, 1·926 and 0·166%, respectively, with a phosphorus content of 61 ppm. The pH of the sediment sample was 7.

Biodegradation of PAHs

Sediment microcosms with a mixture of acenaphthene, phenanthrene and pyrene were assessed for their ability to biodegrade these PAHs. Compared with an abiotic control, the acenaphthene and phenanthrene concentrations decreased faster in sediment containing indigenous micro-organisms. The reduction was observed beginning at week 2, and complete reduction of both PAHs was achieved at the end of incubation (Figs 1a,b). The rate of phenanthrene removal in experimental microcosm set was 0·549 mg kg−1 per day. No degradation of pyrene was observed within 8 weeks (Fig. 1c).

Figure 1.

Biodegradation of acenaphthene (a), phenanthrene (b) and pyrene (c) in sediment microcosms. The initial concentration of each polycyclic aromatic hydrocarbons was 30 mg l−1. (image_n/jam12128-gra-0001.png) Control and (image_n/jam12128-gra-0002.png) Experiment.

Enumeration of total bacteria and PAH-degrading bacteria

The population sizes of total bacteria and PAH-degrading bacteria were determined by the MPN method. The number of total bacteria ranged from 9·2 × 107 to 5·0 × 107 MPN g−1 dry weight. For the number of PAH-degrading bacteria, the population of PAH-degrading bacteria was enhanced by spiked PAHs. The acenaphthene-degrading bacterial numbers tended to increase until the end of the experiment, and the values ranged from 3·1 × 103 to 1·3 × 105 MPN g−1 dry weight. The numbers of phenanthrene-degrading bacteria ranged from 2·8 × 103 to 5·3 × 104 MPN g−1 dry weight, and the increase was observed beginning at week 2 and became relatively stable until the final week. For pyrene-degrading bacteria, the numbers slightly increased early in the incubation period, reached the highest number in the final week and the values ranged from 2·8 × 103 to 5·5 × 104 MPN g−1 dry weight.

Isolation and identification of PAH-degrading bacteria

Six PAH-degrading bacteria designated FP0, FP2-1, FP2-2, FP8, FA4 and FPY8 were isolated by the enrichment approach. All of the isolated strains were identified based on 16S rDNA with entries in the GenBank database. Based on the analysis of 16S rDNA, strains FP2-1 and FP8 were identified as Burkholderia sp., and strains FP0, FP2-2, FA4 and FPY8 were identified as Dyella sp., Luteibacter sp., Alcaligenes sp. and Ochrobactrum sp., respectively (Table 1).

Table 1. Isolation and identification of polycyclic aromatic hydrocarbons (PAH)-degrading bacteria from sediment microcosms
StrainSubstrateaWeekbClosest organismAccession no.% SimilarityTaxonomic group
  1. a

    Substrate was used as the sole carbon and energy source to enrich PAH-degrading bacteria from sediment microcosms.

  2. b

    PAH-degrading bacteria were isolated from samples taken after the incubation period of the indicated time.

FP0Phenanthrene0 Dyella japonica AB110496 100γ-Proteobacteria
FP2-1Phenanthrene2Burkholderia sp. SBI-19 AB366313 99β-Proteobacteria
FP2-2Phenanthrene2Luteibacter sp. IMP04 FR727720 99γ-Proteobacteria
FP8Phenanthrene8Burkholderia sp. J62 EF555575 99β-Proteobacteria
FA4Acenaphthene4 Alcaligenes faecalis FN997611 100β-Proteobacteria
FPY8Pyrene8Ochrobactrum sp. TSH93 AB508896 99α-Proteobacteria

Analysis of bacterial community dynamics

The analyses of total bacterial and Sphingomonas communities were conducted using the DGGE technique with samples collected from microcosms every 2 weeks. The results showed that the total bacterial and Sphingomonas community structures changed after exposure to PAHs. For the total bacterial community, DGGE profiles from week 2 to week 8 were different from those of week 0 (Fig. 2a). Bands NP6, NP7 and NP8 were detected beginning at week 2, persisted until week 8 and had 98, 95 and 95% similarity to sequences of Marinobacter sp., Enterobacter sp. and Dethiosulfatibacter sp., respectively. Moreover, bands NP1, NP2, NP3, NP4 and NP5 were present for all weeks and were identical to sequences of Lysinbacillus sp., Sedimentibacter sp., Ochrobactrum sp. and Rhodococcus sp., respectively (Table 2). For the Sphingomonas community, the PCR product could be detected beginning at week 2 but not at the start of the experiment (week 0). DGGE profiles from week 4 contained numerous bands, and the number of bands from week 6 to the final week tended to decrease (Fig. 2b). The sequences of predominant DGGE bands from Sphingomonas were identical to those of Marinobacterium sp., Sphingomonadaceae bacterium, Erythrobacter sp., Novosphingobium sp., Porticoccus sp. and Porphyrobacter sp. (Table 2).

Table 2. Identification of 16S rDNA sequences obtained from denaturing gradient gel electrophoresis bands
StrainsWeekClosest organismAccession no.% SimilarityTaxonomic group
NP10–8Lysinibacillus sp. JN416563 100Firmicutes
NP20–8Sedimentibacter sp. EF059533 99Firmicutes
NP30–8Ochrobactrum sp. AB695227 100α-Proteobacteria
NP40–8Rhodococcus sp. HQ659583 100Actinobacteria
NP50–8Rhodococcus sp. AB183440 100Actinobacteria
NP62–8Marinobacter sp. DQ768635 98γ-Proteobacteria
NP72–8Enterobacter sp. JQ917108 95γ-Proteobacteria
NP82–8Dethiosulfatibacter sp. AM933654 99Firmicutes
SP212–6Marinobacterium sp. GQ245906 96γ-Proteobacteria
SP222–4Marinobacterium sp. FJ161301 96γ-Proteobacteria
SP434–8Porphyrobacter sp. JN547333 99α-Proteobacteria
SP444–8Erythrobacter sp. JN594622 99α-Proteobacteria
SP454Novosphingobium sp. EU430056 97α-Proteobacteria
SP464Novosphingobium sp. EF424403 99α-Proteobacteria
SP474Marinobacterium sp. FJ716699 96γ-Proteobacteria
SP484Porticoccus sp. EF468719 98γ-Proteobacteria
SP694–6Porphyrobacter sp. HQ588835 99α-Proteobacteria
Figure 2.

Denaturing gradient gel electrophoresis (30–70% denaturant) of total bacterial communities (0, 2, 4, 6 and 8 weeks) (a) and Sphingomonas communities (2, 4, 6 and 8 weeks) (b) in microcosms.

Detection of dioxygenase genes

Detection of ring-hydroxylating dioxygenase genes involved in PAH degradation in the sediment was conducted using a PCR technique with primers specific for the arh and nid genes, which are involved in acenaphthene and pyrene biodegradation, respectively, and the PAH-RHDα genes from Gram-positive and Gram-negative bacteria. The expected PCR products could be detected in samples taken from week 2 to week 8 (Fig. 3a) when using the GNF/GNR primers, which are specific to ring-hydroxylating dioxygenase genes of Gram-negative bacteria, and these PCR products were cloned. A total of 17 clones were compared according to their patterns of digestion by all three RFLP restriction enzymes, and these clones were sorted into six groups. Representatives of each group were selected for sequence analysis. Cloned GN2-2 was related to phnAc of Burkholderia glathei [AAN74945] with 42% similarity. Cloned GN2-3 was related to phnAc of Burkholderia sp. RP007 [AAD09872] with 45% similarity and related to nahAc of Burkholderia sp. S1-17 [AAL46983] with 41% similarity. Cloned GN2-9 was related to the naphthalene dioxygenase of an uncultured bacterium [AAK85517] and Comamonas testosteroni [ABX89315] with 46 and 43% similarity, respectively. Cloned GN6-2 was related to phnAc of Burkholderia sp. Eh1-1 [AAQ84686] and Burkholderia phenazinium [AAN74946] with 44 and 43% similarity, respectively (Table 3). However, analysis of the other two representative clones showed no relation to any ring-hydroxylating dioxygenases (data not shown).

Table 3. Identities of amino acid sequences from cloned PCR products obtained via PCR with the polycyclic aromatic hydrocarbons-RHDα GN primers and DNA from the sediment microcosms and isolated strains
DNA sampleClone name% amino acid sequence identityClosest relative gene, bacteria and corresponding amino acid sequence accession number
Sediment microcosmGN2-242phnAc, Burkholderia glathei [AAN74945]
GN2-345phnAc, Burkholderia sp. RP007 [AAD09872]
41nahAc, Burkholderia sp. S1-17 [AAL46983]
GN2-946Naphthalene dioxygenase, uncultured bacterium [AAK85517]
43Naphthalene dioxygenase, Comamonas testosteroni [ABX89315]
GN6-244phnAc, Burkholderia sp. Eh1-1 [AAQ84686]
43phnAc, Burkholderia phenazinium [AAN74946]
Isolated strainDFP040Naphthalene dioxygenase, uncultured bacterium [AAK85517]
39naphthalene dioxygenase, Herbaspirillum sp. HG1 [AAN74943]
DFP2143Naphthalene dioxygenase, Burkholderia phenazinium [AAN74946]
DFP2242phnAc, Burkholderia sartisoli [AAD09872]
42phnAc, Burkholderia sp. Eh1-1 [AAQ84686]
DFP842Naphthalene dioxygenase, Burkholderia glathei [AAN74944]
41Naphthalene dioxygenase, Burkholderia phenazinium [AAN74947]
DFA445Dioxygenase, Alcaligenes faecalis [BAA76323]
44phnAc, Burkholderia sp. Ch1-1[AAQ84683]
DFPY836Naphthalene dioxygenase, Pseudomonas sp. CY13 [AAV33335]
36Naphthalene dioxygenase, Pseudomonas sp. SOD-3 [AAL46987]
Figure 3.

Detection of polycyclic aromatic hydrocarbons (PAH)-RHDα dioxygenase genes in microcosms using PAH-RHDα GN (a) and GP (b) primers and in isolated strains using PAH-RHDα GN primers (c).

Moreover, in week 8, the expected PCR products could be detected when using GPF/GPR primers, which are specific to ring-hydroxylating dioxygenase genes of Gram-positive bacteria (Fig. 3b). In addition, dioxygenase genes in the isolated strains were examined with GNF/GNR primers, and the expected PCR products could be detected in all strains (Fig. 3c). These PCR products were also cloned, and their sequences were analysed. The results showed that the Dyella sp. strain FP0 contained a gene related to the naphthalene dioxygenase of an uncultured bacterium [AAK85517] and Herbaspirillum sp. HG1 [AAN74943] with 40 and 39% similarity, respectively. The Burkholderia sp. DFP21 gene was related to naphthalene dioxygenase in Burkholderia phenazinium [AAN74946] with 43% similarity. The Luteibacter sp. DFP22 gene was related to phnAc in Burkholderia sartisoli [AAD09872] and Burkholderia sp. Eh1-1 [AAQ84686] with 42% similarity. The Burkholderia sp. DFP8 gene was related to naphthalene dioxygenases in Burkholderia glathei [AAN74944] and Burkholderia phenazinium [AAN74947] with 42 and 41% similarity, respectively. The gene of Alcaligenes sp. DFA4 was related to the dioxygenase of Alcaligenes faecalis [BAA76323] with 45% similarity and related to phnAc in Burkholderia sp. Ch1-1[AAQ84683] with 44% similarity. Furthermore, Ochrobactrum sp. DFPY8 contained a gene related to naphthalene dioxygenases in Pseudomonas sp. CY13 [AAV33335] and Pseudomonas sp. SOD-3 [AAL46987] with 36 and 36% similarity, respectively (Table 3).

Real-time PCR quantitative detection of PAH catabolic genes

Real-time PCR was used to quantify PAH-RHDα genes targeted in sediment microcosms. This study used primer set GNF and GNR to detect and quantify PAH-RHDα genes and primer set 968 F and 1401 R to detect and quantify 16S rRNA genes. The 16S rRNA gene copy number was decreased from 9·7 to 8·6 log copies number g−1 sediment dry weight. PAH-RHDα GN genes could not be detected in sediment microcosms at the initial sampling, but the gene copy number was appeared and stable between weeks 2 and 8 at about 7 log copies number g−1 sediment dry weight (Fig. 4).

Figure 4.

16S rRNA gene and polycyclic aromatic hydrocarbons-RHDα GN gene copy numbers by real-time PCR from sediment microcosms. (image_n/jam12128-gra-0003.png) 16S rRNA genes and (image_n/jam12128-gra-0004.png) PAH-RHDα GN genes.

Discussion

Mangrove ecosystems are subject to various types of contamination including PAHs from anthropogenic activities (Klekowski et al. 1994). It is believed that micro-organisms that are capable of effectively degrading xenobiotic pollutants can be found in contaminated sediments (Koutny et al. 2003). In this study, using a culture-dependent method, six PAH-degrading bacteria were isolated from sediment microcosms that were artificially contaminated with acenaphthene, phenanthrene and pyrene. These isolates belong to the genera Dyella, Burkholderia, Luteibacter, Alcaligenes and Ochrobactrum. Arulazhagan and Vasudevan (2011) showed that Ochrobactrum sp. VA1 was able to degrade both LMW and high molecular weight PAHs under saline conditions. Species of the genus Burkholderia are also known for degrading aromatic hydrocarbons (Kang et al. 2003; Tae Jung et al. 2003) and chlorine compounds (Rehmann and Daugulis 2006) in the environment. Bacteria in the genus Alcaligenes were shown to degrade acenaphthene in the present study. This indicates that not only Sphingomonas species (Pinyakong et al. 2004) but also Alcaligenes species play an important role in acenaphthene degradation (Selifonov et al. 1993). There have also been reports that several Alcaligenes species have PAH-degrading activities, including phenanthrene-, fluorene- and fluoranthene-degrading activities (Walter et al. 1990; Weissenfels et al. 1990; Moller and Ingvorsen 1993). Furthermore, for the genus Dyella, Li et al. (2009) reported that Dyella ginsengisoli LA-4 removed biphenyl removal with great efficiency and could use other aromatic compounds such as benzoic acid, naphthalene and toluene as a sole source of carbon and energy. Interestingly, this is the first report of phenanthrene degradation by Dyella sp. Furthermore, this is also the first report indicating that a member of the genus Luteibacter isolated from mangrove sediments is capable of utilizing PAH. Furthermore, PAH-degrading bacteria isolated from sediment microcosms in this study were the different genus with those isolated from mangrove sediments in various study reported so far. Several reports presented the isolation of Rhodococcus (Guo et al. 2005; Luo et al. 2005; Yu et al. 2005; Zhou et al. 2008; Huijie et al. 2011), Sphingomonas (Guo et al. 2005; Zhou et al. 2008; Huijie et al. 2011) and Pseudomonas (Yu et al. 2005; Brito et al. 2006) as PAH degraders from mangrove sediments. However, it has been shown that many studies could isolate different PAH-degrading bacteria from mangrove sediments (Huijie et al. 2011). They suggested several reasons including (i) the use of different enrichment, screening and isolation methods for isolation and (ii) the use of different source and location of mangrove sediments (Huijie et al. 2011).

To study bacterial community dynamic changes, this study employed PCR-DGGE. For the total bacterial community, the number of bands in the profile increased beginning at week 2, and these bands remained present until week 8. In addition, these bands were thought to represent bacteria related to PAH degradation. We suggest that PAHs induced an increase in indigenous PAH-degrading bacteria and favored their growth in sediment microcosms. Bands NP6, NP7 and NP8 were examples in this group. NP6 and NP7 were identical to sequences from Marinobacter sp. and Enterobacter sp., respectively. Both were related to bacterial groups known for their capacity to degrade PAHs in marine and mangrove sediments (Gauthier et al. 1992; Arulazhagan et al. 2010). NP8 had identity with Dethiosulfatibacter sp., which had not been known to degrade petroleum hydrocarbons. Moreover, it was found that bands NP1, NP2, NP3, NP4 and NP5 were present for all weeks. Thus, the results suggested that these bacteria were the dominant genera in this sediment microcosm, and they tolerated and were alive in the presence of PAHs. NP1 and NP2 were similar to Lysinbacillus sp. and Sedimentibacter sp., respectively, and both NP4 and NP5 were similar to Rhodococcus sp. Josic et al. (2008) reported that Lysinibacillus fusiformis was isolated from oil-polluted soil near a gas station and from stabilized sludge from a petrochemical plant. This species demonstrated the ability to tolerate heavy metals and grow on toluene. Winderl et al. (2008) reported that Sedimentibacter sp. was found in a tar oil-contaminated aquifer. Members of the genus Rhodococcus are present in many soils and are able to degrade PAHs such as naphthalene, phenanthrene, anthracene, fluoranthene and pyrene (Walter et al. 1991; Grund et al. 1992; Bouchez et al. 1996; Dean-Ross et al. 2001). Furthermore, NP3 had similarity to Ochrobactrum sp., which was also isolated as PAH-degrading bacteria in this study and was dominant in the DGGE profiles.

For the Sphingomonas community, a PCR product could be detected beginning at week 2. Its induction after PAH exposure indicated that the Sphingomonas community may play a role in PAH degradation. The bacterial genus Sphingomonas was found to be the major degrader of PAHs in contaminated soil, and it is well known as a degrader of various PAHs, such as naphthalene and phenanthrene (Pinyakong et al. 2003; Stolz 2009). Although sequenced DGGE bands were not similar to data for the genus Sphingomonas, some of the bands (SP43, SP44, SP45, SP46 and SP69) were related to aromatic hydrocarbon-degrading species of the family Sphingomonadaceae. For example, bands SP43 and SP69 had sequences similar to that of Porphyrobacter sp. Band SP45 and SP46 had similarity to Novosphingobium sp., and band SP44 had similarity to Erythrobacter sp. Hiraishi et al. (2002) reported that Porphyrobacter sanguineus is capable of degrading biphenyl and dibenzofuran. Novosphingobium aromaticivorans strain F199 was reported to be a naphthalene–fluorene-utilizing bacterium (Romine et al. 1999). Erythrobacter sp. has recently been reported to degrade PAHs and was isolated from an actual crude-oil-polluted Indonesian sea site (Harwati et al. 2007).

Zhou et al. (2009) studied on the bacterial community structure in mangrove sediment and indicated that marine bacteria including Vibrio, Roseobacter and Ferrimonas were the dominant genera after exposure to PAHs by using culture-independent method. Although the dominant genera found in our study are not the same groups with that report which may have resulted from the use of different source of mangrove sediment and type and amount of PAH, our DGGE profiles also suggest the role of marine bacteria in PAH degradation in mangrove sediment. Moreover, it is in accordance with previous works that culture-dependent method usually allowed to isolate the terrestrial bacteria from mangrove sediments (Guo et al. 2005; Zhou et al. 2008, 2009). Furthermore, interestingly, some PAH-degrading bacteria were isolated in this study but were not dominant in the DGGE profiles. We suggest that this was due to the different conditions under which the sediment microcosms had a mixture of PAHs, while isolated strains were further enriched and isolated with individual PAHs. However, there were Sphingomonas sp. and other bacteria that could be detected by molecular techniques but could not be cultivated by the culture method. We suggest that these bacteria were more difficult to isolate using the traditional culture method of this experiment. Therefore, in this study, we used both culture-independent methods and culture-dependent methods to analyse the microbial community. The combination of these two methods has enhanced the ability to detect and identify micro-organisms in the environment because culture-dependent methods do not provide comprehensive information on the composition of the microbial community, and culture-independent methods may detect only species that predominate in the environment. Phylogenetic tree of 16S rDNA sequences from all the isolates and dominant DGGE bands and those representative known PAH-degrading bacteria is shown in Fig. 5.

Figure 5.

Phylogenetic tree of the isolates and dominant denaturing gradient gel electrophoresis bands with those representative known polycyclic aromatic hydrocarbons-degrading bacteria based on 16S rDNA sequences. Numbers at the node indicate the bootstrap values as a percentage of 1000 replicates.

In comparisons between the bacterial community and PAH biodegradation, acenaphthene and phenanthrene were degraded beginning at week 2, and the DGGE profiles also changed beginning at week 2. These results suggest that the bacterial communities were changed after exposure to PAHs. In sediment microcosms, acenaphthene and phenanthrene were completely degraded, while no degradation of pyrene was observed. This can be explained because pyrene is a high molecular weight PAH that may be degraded more slowly than LMW PAHs such as acenaphthene and phenanthrene. Juhasz and Naidu (2000) reported that the LMW PAHs are relatively easier to degrade, while the HMW PAHs are persistent. However, pyrene-degrading bacteria could be enumerated, and the number was highest in the final week when pyrene-degrading Ochrobactrum sp. was isolated. We thought that for the degradation of HMW PAH, extended incubation times may be required for the adaptation and population growth (Venkata et al. 2008; Nopcharoenkul et al. 2011). Therefore, the degradation of pyrene may possibly be observed after a prolonged incubation period. For acenaphthene, the decrease in this PAH was also found in the control with autoclaved sediment due to its volatilization ability. The decrease in acenaphthene in the experimental set was observed earlier than that in the control, but slopes were almost identical. This can be thought that indigenous bacteria might start to degrade acenaphthene at the beginning. However, the amount of acenaphthene applied in this study was too low and easy to volatile, and the difference between rate of bacterial degradation and physico-chemical reduction was then not observed. Therefore, it is hard to conclude the impact of bacterial degradation of acenaphthene in this experiment. However, we could isolate acenaphthene-degrading Alcaligenes faecalis FA4 from sediment microcosm of during 4 weeks of incubation by using enrichment method, suggesting that the potential of acenaphthene biodegradation in mangrove sediment is promising for further studies.

PAH degradation is mostly carried out by dioxygenase enzymes from micro-organism under aerobic conditions (Andreoni and Gianfreda 2007). In the present study, the dioxygenase gene was detected in sediment microcosms beginning at week 2 when using the GNF/GNR primers, which are specific to ring-hydroxylating dioxygenase genes of Gram-negative bacteria. This result revealed a pattern similar to that of PAH biodegradation. The finding of the dioxygenase gene is a good indicator of the PAH degradation potential of the bacterial community present in the environment. In addition, dioxygenase genes could also be detected in all isolated strains, and they were moderately similar to those of known naphthalene and phenanthrene dioxygenases. The presence of the dioxygenase gene further confirmed the ability of the isolates to degrade PAHs.

Moreover, the quantification of PAH-RHDα GN genes in sediment microcosms was consistent with phenanthrene biodegradation. Similarly, some studies have shown positive relationships between PAH biodegradation/pollution and the abundance of PAH-dioxygenase genes (Dionisi et al. 2004; Tuomi et al. 2004). Interestingly, real-time PCR showed the decrease in 16S rRNA gene copy number after PAH exposure and those increase and stability of PAH-RHDα GN gene copy number. The ratio of PAH-RHDα GN gene relative to 16S rRNA gene copy number increased up to 0·80 at week 8, indicating that PAH-degrading bacteria carrying PAH-RHDα GN gene trended to become a higher proportion in total bacterial community after PAH exposure.

In conclusion, phenanthrene biodegradation was observed in sediment microcosms constructed with mangrove sediment and mixed PAHs. Culture-independent approaches revealed the change in bacterial community structures after exposure to PAHs and demonstrated that bacteria in the genera Marinobacter, Enterobacter and Dethiosulfatibacter have the potential to play important roles as PAH degraders. However, six PAH-degrading bacteria were isolated from the sediment microcosms. They were identified as Dyella sp., Burkholderia sp., Luteibacter sp., Alcaligenes sp. and Ochrobactrum sp. Interestingly, PAH-RHDα GN genes could be detected from both the DNA extracted from the sediments and from all of the isolated bacteria. These results demonstrated that indigenous bacteria in mangrove sediment had ability to degrade PAHs.

Acknowledgements

This work was supported by the Thai Government Stimulus Package 2 (TKK2555) under the Project for the Establishment of the Comprehensive Center for Innovative Food, Health Products and Agriculture.

Ancillary