Physiological characterization of Mycobacterium sp. strain 1B isolated from a bacterial culture able to degrade high-molecular-weight polycyclic aromatic hydrocarbons


Richard Bentham, Department of Environmental Health, Flinders University of South Australia, GPO Box 2100, Adelaide SA 5001, Australia (e-mail:


Aim:  The aim of this study was to further characterize a bacterial culture (VUN 10,010) capable of benzo[a]pyrene cometabolism.

Methods and Results:  The bacterial culture, previously characterized as a pure culture of Stenotrophomonas maltophilia (VUN 10,010), was found to also contain another bacterial species (Mycobacterium sp. strain 1B), capable of degrading a similar range of PAH substrates. Analysis of its 16S rRNA gene sequence and growth characteristics revealed the strain to be a fast-growing Mycobacterium sp., closely related to other previously isolated PAH and xenobiotic-degrading mycobacterial strains. Comparison of the PAH-degrading characteristics of Mycobacterium sp. strain 1B with those of S. maltophilia indicated some similarities (ability to degrade phenanthrene and pyrene), but some differences were also noted (S. maltophilia able to degrade fluorene, but not fluoranthene, whereas Mycobacterium sp. strain 1B can degrade fluoranthene, but not fluorene). Unlike the S. maltophilia culture, there was no evidence of benzo[a]pyrene degradation by Mycobacterium sp. strain 1B, even in the presence of other PAHs (ie pyrene) as co-metabolic substrates. Growth of Mycobacterium sp. strain 1B on other organic carbon sources was also limited compared with the S. maltophilia culture.

Conclusions:  This study isolated a Mycobacterium strain from a bacterial culture capable of benzo[a]pyrene cometabolism. The Mycobacterium strain displays different PAH-degrading characteristics to those described previously for the PAH-degrading bacterial culture. It is unclear what role the two bacterial strains play in benzo[a]pyrene cometabolism, as the Mycobacterium strain does not appear to have endogenous benzo[a]pyrene degrading ability.

Significance and Impact of the Study:  This study describes the isolation and characterization of a novel PAH-degrading Mycobacterium strain from a PAH-degrading culture. Further studies utilizing this strain alone, and in combination with other members of the consortium, will provide insight into the diverse roles different bacteria may play in PAH degradation in mixed cultures and in the environment.


Polycyclic aromatic hydrocarbons (PAHs) are ubiquitous environmental pollutants. Although there are significant natural sources of PAHs in the environment (i.e. bushfires), high concentrations are generally derived from anthropogenic sources such as gas manufacturing plants, combustion processes, petroleum spillage and timber treatment plants (Juhasz and Naidu 2000). PAHs are considered priority pollutants by the US Environmental Protection Agency and other environmental authorities due to their toxicity, carcinogenicity, teratogenicity and recalcitrance in the environment (Kanaly and Harayama 2000). PAHs are known to be recalcitrant in the environment, mainly due to low water solubility and sorption to organic matter of the compounds limiting their bioavailability. High-molecular-weight PAHs (four to seven ring) have a lower water solubility and are more recalcitrant than low molecular weight PAHs such as naphthalene (two ring) and phenanthrene (three ring; Cerniglia 1992; Wilson and Jones 1993; Juhasz and Naidu 2000).

Many bacterial strains have been isolated with the ability to degrade or mineralize a range of PAHs containing up to four benzene rings, including phenanthrene, fluoranthene and pyrene (Cerniglia 1992; Juhasz and Naidu 2000). PAH degradation appears to be associated with certain phylogenetic groups of bacteria, particularly Sphingomonas, Burkholderia and Pseudomonas amongst Gram-negative species, and Mycobacterium and Rhodococcus amongst Gram-positive species (Mueller et al. 1997).

Benzo[a]pyrene, a five-ring PAH, is considered one of the most important PAH pollutants due to its potent carcinogenicity and recalcitrance in the environment. Both bacterial and fungal strains and consortia have been isolated with the ability to metabolize benzo[a]pyrene and in most cases metabolism occurs cometabolically with other PAHs or other hydrocarbons acting as primary carbon sources [i.e. pyrene (PAH) or diesel; Boonchan et al. 2000; Kanaly et al. 2000]. Whilst there have been reports of benzo[a]pyrene mineralization as a sole carbon source (Rafin et al. 2000), the ability to degrade or mineralize significant amounts of benzo[a]pyrene by single isolates remains elusive.

Previous work by Boonchan et al. (2000) described the isolation and characterization of a bacterial-fungal co-culture with the ability to degrade and mineralize benzo[a]pyrene in liquid culture and soil microcosms. The co-culture comprised a single bacterial and a single fungal strain that in combination mineralized significantly higher amounts of benzo[a]pyrene than either strain in isolation. The bacterial and fungal components were isolated from separate manufactured gas plant-contaminated soils. The bacterial strain (VUN 10,010) was identified as Stenotrophomonas maltophilia by 16S rRNA gene analysis. Stenotrophomonas maltophilia was capable of degrading a range of PAHs (fluorene, phenanthrene and pyrene) and cometabolizing high molecular weight PAHs when pyrene was also present (benz[a]anthracene, benzo[a]pyrene and dibenz[a,h]anthracene). No significant mineralization of benzo[a]pyrene by VUN 10,010 was observed when used as sole carbon source. When pyrene was added as co-substrate, 32·4% of radiolabelled benzo[a]pyrene was recovered as 14CO2 after 56 days. When VUN 10,010 was grown in combination with the fungal strain, VUO 10,201 Penicillium janthinellum, the co-culture mineralized significantly higher amounts of radiolabelled benzo[a]pyrene (58·1% recovered as 14CO2 in 56 days). It was clear from the work performed by Boonchan et al. (2000) that the combination of bacterial and fungal activity increased the mineralization of high molecular weight PAHs in liquid culture and in soil.

The aim of the present study was to further characterize the actions and abilities of the bacterial component of the co-culture for degradation of PAHs and other xenobiotics. The isolation, identification and characterization of an additional and previously unreported strain (Mycobacterium sp. strain 1B) from the bacterial culture is reported.

Materials and methods

Chemicals and media

Naphthalene, phenanthrene and fluorene were purchased from Supelco (Sigma Aldrich, Castle Hill, NSW, Australia), fluoranthene was obtained from Koch-Light Laboratories (Colnbrook, Bucks, UK), pyrene was purchased from Sigma Chemical Company (St Louis, MO, USA) and benzo[a]pyrene was purchased from Aldrich Chemical Company (Milwaukee, WI, USA). For growth studies, all chemicals used (toluene, benzene, glucose, maleic acid, n-decane, n-hexadecane, salicylic acid, pentachlorophenol, 4-cholorophenol, d-pantothenic acid, 3-chlorobenzoate, 2,4,6-tricholorophenol, sodium pyruvate, iso-phthalic acid, succinic acid, p-aminobenzoic acid, catechol, 2,4,5-trichlorophenol, sodium benzoate, 4-nitrophenol, naphthalene, Tween-20, Tween-80) were at least reagent grade. Dichloromethane was obtained in analytical grade (Omnisolv) from Merck KgaA (Darmstadt, Germany) and dimethylformamide (DMF) was obtained from BDH Chemicals Ltd (Poole, UK). The composition of basal salts medium (BSM) containing trace elements, vitamins and Mg/Ca solutions has been described previously (Juhasz et al. 1996). BSMY was prepared by addition of 50 mg l−1 yeast extract to standard BSM. Bacteriological media including nutrient agar (CM3), R2A agar (CM906), yeast extract (L21), tryptone (L42) and bacteriological agar (L11) were obtained from Oxoid (Unipath Ltd, Basingstoke, Hampshire, UK). PAH stock solutions were prepared in DMF at 0·5% (w/v) (benzo[a]pyrene) and 2% (w/v) (all other PAHs).

Bacterial culture

Stock cultures of VUN 10,010, previously characterized as S. maltophilia (Boonchan et al. 2000), were kindly provided by Prof. Grant Stanley (Victoria University of Technology) in freeze-dried, plate and liquid culture forms. Aliquots of stock cultures were inoculated into BSM broth containing 250 mg l−1 pyrene and incubated at 28°C on a rotary shaker (150 rev min−1) in the dark until the solid pyrene crystals were removed from suspension (7–10 days). Broths were subcultured in this medium a further two times, after which bacteria were spread-plated onto BSM agar plates containing 100 mg l−1 pyrene. Single colonies were picked from the BSM pyrene plates and streaked onto nutrient agar, R2A agar and BSM pyrene agar plates to obtain pure cultures of pyrene-degrading strains. PAH-degrading mixed and pure cultures were maintained by subculture in BSMY containing pyrene (250 mg l−1) and in 25% glycerol at −80°C.

Fatty acid methyl ester (FAME) analysis

Bacterial cells were grown on Tryptic Soy Agar (TSA; BBL, Becton Dickinson, North Ryde, NSW, Australia) at 28°C for 24–48 h before being harvested for analysis. FAME analysis was performed by Bruce Hawke, CSIRO Land and Water, Adelaide, to aid in identification of the newly isolated strain. Fatty acid profiles were compared with the MIDI Microbial Identification System database (MIDI, Newark, DE, USA; Sasser 1990).

DNA extraction and 16S rRNA PCR

Chromosomal DNA was extracted from bacterial cells using the Puregene® DNA Isolation Kit (Gentra Systems, Minneapolis, MN, USA). 16S rRNA DNA was amplified using the following primers: fD1 (5′-AGA GTT TGA TCC TGG CTC AG-3′) and rD1 (5′-aag gag gtg atc cag cc-3′; Weisburg et al. 1991), 765r (5′-CTG TTT GCT CCC CAC GCT TTC-3′) and 704f (5′-gta gcg gtg tgc gta ga-3′; Damiani et al. 1997). Specific reaction conditions were as follows: primers 1 μm, MgCl2 3 mm, template DNA 1 μl. All other conditions were as specified in the AmpliTaqTM Gold instruction manual (Applied Biosystems, Melbourne, Victoria, Australia). Cycling conditions were as follows: one cycle of 95°C/10 min, 35 cycles of 94°C/30 s, 50°C/30 s, 72°C/90 s, one cycle of 72°C for 10 min. Amplified DNA (5 μl) was separated by gel electrophoresis on 1·2% (w/v) agarose (in 1× TAE) at 80 V. The PCR products obtained were purified using Boehringer Mannheim (Roche Diagnostics, Castle Hill, Victoria, Australia) high pure PCR product purification kit. PCR products were cloned into pGEM-T or pGEM-T Easy PCR cloning vectors (Promega Corporation Australia, Annandale, NSW, Australia) according to the manufacturers’ instructions. Ligation mixtures were transformed into competent E. coli DH5α cells by heat shock at 42°C (Ausubel et al. 1993) and after recovery for 1 h in added Luria–Bertani (LB) broth, plated onto LB agar with ampicillin (Amp), IPTG (isopropyl-β-thiogalactopyranoside) and X-gal (5-bromo, 4-chloro, 3-indolyl-β-d-galactopyranoside). White colonies were screened for insert-containing plasmids by colony cracking (Sambrook et al. 1989) and several positive clones were grown overnight at 37°C in LB broth + Amp. Plasmid DNA was isolated using a modified alkaline lysis/polyethylene glycol precipitation method. Plasmids containing inserts of the correct size were quantified and sent to the Australian Genome Research Facility (AGRF, Brisbane) for sequencing. DNA was sequenced using both forward and reverse primers. Forward and reverse sequences were aligned manually or using the Gap program (Devereux et al. 1984) and a partial consensus sequence obtained. This consensus sequence was entered into the Ribosomal Database Project II aaa(RDP-II; Cole et al. 2003) database and the nearest relatives identified. Phylogenetic trees were constructed using the Phylip interface on the RDP-II page.

Growth on and degradation of PAHs and other carbon sources

Bacterial cultures were grown at 30°C, 150 rev min−1 in either R2A or BSMY containing 250 mg l−1 pyrene until growth reached late exponential phase, as determined by clearing of solid pyrene suspension from the culture broth. Cells were harvested by centrifugation, washed twice in sterile BSM and resuspended in one-tenth volume of medium. This concentrated cell suspension was used as inoculum for subsequent experiments. Killed bacterial inocula were prepared by the addition of mercuric chloride (0·2 g l−1) and incubation for 24 h before use. Growth media were prepared by adding PAH or other carbon source stock solutions in DMF to 10 ml of broth in 30-ml serum bottles. Bacteria were inoculated to achieve a starting population of approx. 106 viable cells per millilitre. PAHs were individually added to broths to give final concentrations of 50 mg l−1 (benzo[a]pyrene), 100 mg l−1 (fluoranthene) and 250 mg l−1 (phenanthrene and pyrene). Alternative carbon sources where used were added at 100 mg l−1. The experimental cultures were conducted in triplicate and controls were in duplicate, with incubation at 30°C on a rotary shaker at 150 rev min−1 in the dark. Replicate samples were sacrificed for the determination of bacterial numbers and PAH concentrations. Growth on alternative carbon sources was monitored by O.D.600 nm and visual determinations. Effect of salicylic acid on PAH degradation was analysed by the addition of 100 mg l−1 salicylic acid (pH adjusted to 6·9–7·0) to broth cultures containing individual PAHs. Indigo production from indole was monitored to determine presence of dioxygenase enzymes. Indigo production was analysed by the addition of indole (0·2 mm) to broths containing PAHs. Indigo (blue) colour formation was monitored visually by the presence of insoluble blue crystals or by measurement of O.D.610 nm from ethyl acetate extractions (1 ml extracted twice with equal volumes of ethyl acetate) of culture fluids (Boonchan 1998).

PAH extraction and analytical procedures

The PAHs were extracted from culture fluids using dichloromethane (DCM). Before extractions, internal standard (benzo[b]fluorene, 0·1 ml of 1000 μg ml−1 stock solution) was added to each serum bottle. Each extraction consisted of the addition of DCM, vigorous shaking of the serum bottles, 2 × 30 min sonications and subsequent freezing (−20°C) of the bottles to facilitate separation of the two phases. Triplicate extractions were performed with 3–4 ml DCM each time and DCM was removed from the serum bottles after thawing at room temperature. DCM from the three extractions was pooled and concentrated to 1 ml using a rotary concentrator (Büchi Rotavapor-R; Büchi Labortechnik, AG, Flawil, Switzerland). The 1 ml volume of concentrated extract was dried over anhydrous sodium sulphate and placed in a 2·0 ml glass screw cap vial (Varian Australia, Kilkenny, SA, Australia) and stored at −20°C before analysis by gas chromatography–flame ionization detector (GC–FID).

GC–FID analysis

Analyses of DCM extracts and PAH standards were performed on a Varian Star 3800 Gas Chromatograph equipped with FID and Saturn software Version 5·51 (Varian Australia). The following conditions were standard for all analyses: EC-5 ECONO-CAP capillary column (Alltech Associates Australia, Box Hill, Victoria, Australia; 30 m × 0·25 mm 25 μm film, P/N 104305); carrier gas, nitrogen; injector temperature, 320°C; detector temperature, 320°C; column flow rate, 1·5 ml min−1; split flow ratio, 10 : 1. The starting oven temperature was 50°C, followed by a linear increase at 15°C min−1 to 320°C, holding at 320°C for up to 10 min. The concentration of PAHs was calculated using benzo[b]fluorene (1000 μg ml−1) as the internal standard. Standard solutions (1 ml) were prepared with benzo[b]fluorene and PAHs ranging in concentration from 50 to 2500 mg l−1. Detector response factors were averaged over the range of concentration of PAHs and the standard deviation calculated. A percentage error of <5% was considered acceptable. The concentration of PAHs in extracted samples was calculated using the ratio between the PAH and the internal standard peak area and the respective detector response factor.

Genbank accession numbers

The Genbank accession numbers of the two copies of 16S rRNA gene sequence of strain 1B are AY227355 and AY227356.


Isolation of an axenic pyrene-degrading strain from the PAH-degrading culture

A pure bacterial culture capable of growing on and degrading pyrene as a sole source of carbon and energy was isolated from the bacterial culture (VUN 10,010) previously described by Boonchan et al. (2000). The strain was isolated after serial enrichment with pyrene as a sole C source in flask cultures. Initial subcultures took longer to remove pyrene and did not have high numbers of the pyrene-degrading strain, whereas subsequent subcultures removed pyrene more quickly and had high numbers of pyrene-degrading colonies when plated on BSM pyrene plates. The dominant colony type obtained on BSM pyrene plates (named 1B) were small (1–2 mm diameter after 10 days growth), orange, smooth and had zones of clearing 1–2 mm around them where pyrene appeared to have been removed/degraded from the agar. Other colony morphologies were seen on agar plates, but did not show pyrene-clearing characteristics. The isolated strain stained Gram-positive when freshly grown, but Gram-variable in cells from older colonies. The organism grew well on R2A agar (colonies visible in 4–5 days) and BSMY agar containing pyrene (250 mg l−1, colonies visible in 6–7 days), but not nutrient agar (colonies visible in 9–10 days), where growth obtained was patchy and inconsistent. Under no circumstances was a Gram-negative strain isolated showing pyrene-degrading characteristics similar to those previously described for the S. maltophilia pure culture.

FAME analysis.  Fatty acid analysis did not reveal high similarity of the 1B strain with any in the database. The result obtained (0·026 similarity to Rhodococcus equi GC subgroup A) was not sufficient for a positive identification. Slow growth of the strain on TSA (several days compared with the optimal 24–48-h growth period) is likely to have affected the result obtained. The presence of 10Me18:0 (tuberculostearic acid) in FAME chromatograms (data not shown) was an indication that the strain may be a Mycobacterium.

16S rRNA sequence analysis.  Analysis of the 16S rRNA gene sequence of the 1B strain revealed two distinct 16S rRNA sequences within the strain (Genbank accession numbers: AY227355, AY227356). Significant differences (21 bp total) were obtained between the two sequences, mostly within the hypervariable regions common to the 16S gene (Fig. 1). Phylogenetic analysis of both sequences indicated that the strain was most closely related to fast-growing Mycobacteria within the Mycobacterium tuberculosis subgroup. The full sequence was compared with sequences in the Ribosomal Database Project (Cole et al. 2003) separately for the two copies obtained. Highest similarity for both sequences was obtained with Mycobacterium sp. strain SM7.6.1 (Friedrich et al. 2000), a mycobacterial strain isolated with phenanthrene-degrading capabilities under low bioavailability conditions. All sequences obtained with high similarity matches or good alignments to the 1B isolate were from fast-growing mycobacterial strains (Fig. 2). Many of the fast-growing mycobacterial strains with high similarity in their 16S sequences were also capable of degradation of xenobiotic compounds such as 1,2-dibromoethane (Poelarends et al. 1999), PAHs (Wang et al. 1996), pentachlorophenol (Briglia et al. 1994) and petroleum hydrocarbons (Churchill et al. 1999).

Figure 1.

Sequence alignments showing base-pair differences between two copies of 16S rRNA from Mycobacterium sp. strain 1B. Sequence numbering from Genbank sequence (add 48 to give numbers in E. coli numbering system). Sequence 1 = AY227355 and sequence 2 = AY227356 shows the same bases for the two sequences, letters show different bases between sequences

Figure 2.

Neighbour-joining tree for two sequences of Mycobacterium sp. strain 1B, strains of known Mycobacterium spp. and Nocardia asteroides used as an outgroup. The Mycobacterium and Nocardia (accession numbers) are as follows: M. anthracenicum (Y15709), Mycobacterium sp. strain WF2 (U90877), M. chubuense (X55596), M. chlorophenolicum PCP-1 (X81926), Mycobacterium sp. strain GP1 (AJ012626), Mycobacterium sp. strain PAH135 (X84978), M. aichiense (X55598), M. chitae (X55603), Mycobacterium sp. strain CH-1 (AF054278), M. ratisbonense (AJ271863), Mycobacterium sp. strain HE5 (AJ012738), Mycobacterium sp. strain LB307T (AJ245703), Mycobacterium sp. strain BB1 (X81891), Mycobacterium sp. strain SM7.6·1 (AF247497), M. farcinogenes (AF055333), M. smegmatis (AJ131761), Mycobacterium sp. strain T103 (U62889) and Mycobacterium sp. strain PYR-1 (U30662). Asterisk indicates that the strain is a known degrader of xenobiotic chemicals

PAH-degrading capabilities of Mycobacterium sp. strain 1B

The original culture described by Boonchan et al. (2000) degraded or cometabolized a range of PAHs including fluorene, phenanthrene, pyrene and benzo[a]pyrene. Mycobacterium sp. strain 1B was isolated from that culture on its ability to degrade pyrene, but it was still necessary to determine the further PAH-degrading characteristics of the strain. Phenanthrene and pyrene (250 mg l−1) were completely degraded within 5–6 days, but fluoranthene biodegradation (100 mg l−1) was slower, taking 18–20 days to be fully removed from solution as determined by GC–FID analyses (Fig. 3a–c). In all cultures, significant growth of the culture was concurrent with PAH removal. No significant growth was obtained in cultures inoculated with Mycobacterium sp. strain 1B with no PAH added as carbon source (data not shown). When benzo[a]pyrene was added as sole carbon source to broth cultures, no removal (Fig. 4b) or growth (data not shown) was observed. Similar results were also obtained when pyrene was added as a co-substrate, as no removal of benzo[a]pyrene was observed (Fig. 4a). Pregrowth of the strain in these experiments used pyrene as the growth substrate.

Figure 3.

PAH degradation by Mycobacterium sp. strain 1B (bsl00066). Culture broth comprised BSMY and either 250 mg l−1 phenanthrene (a), 250 mg l−1 pyrene (b) or 100 mg l−1 fluoranthene (c) as carbon and energy source. The most probable number (MPN) of bacteria (•) and BSMY-PAH media inoculated with HgCl2 killed inoculum are also shown (bsl00001). Error bars where shown are standard deviation of three replicates

Figure 4.

Benzo[a]pyrene (BaP; bsl00066) or pyrene (PYR; ○) degradation by Mycobacterium sp. strain 1B. Culture broth comprised BSMY and either (a) BaP and PYR (50 mg l−1 and 250 mg l−1) or (b) BaP alone as carbon and energy source. BSMY-PAH media inoculated with HgCl2 killed inoculum are also shown (bsl00001 or •, respectively). Error bars where shown are standard deviation of three replicates

Effect of salicylic acid on growth on PAHs.  Previous work with S. maltophilia VUN 10,010 had shown improved degradation of low molecular weight PAHs (i.e. phenanthrene and fluorene) when salicylic acid was added as co-substrate. In the case of Mycobacterium sp. strain 1B, salicylic acid did not enhance removal of any of the PAHs tested (phenanthrene, fluoranthene and pyrene). When salicylic acid was provided as a sole source of carbon and energy, no significant growth was obtained (Table 1).

Table 1.  Comparison of various characteristics of Mycobacterium sp. strain 1B with Stenotrophomonas maltophilia VUN 10,010 (from Boonchan 1998 and Boonchan et al. 2000)
Characteristic1BVUN 10,010
  1. *Growth tests conducted at 30°C.

  2. †NT, not tested.

  3. ‡Indigo (blue) produced from indole in presence of PAHs.

  4. §Effect of salicylic acid on growth on PAHs: yes, enhanced growth; no, inhibition.

Growth on*
Growth/degradation of PAHs
 Benzo[a]pyrene (+Pyrene)No (No)No (Yes)
Indigo production‡
Effect of salicylic acid§
Growth on
 Succinic acidYes (minimal)Yes
 ρ-Hydroxybenzoic acidNTYes
 Maleic acidNoYes
 Protocatechuic acidNTYes
 d-pantothenic acidNoYes
 Gentisic acidNTYes
 Cinnamic acidNTYes
 Salicylic acidNoYes
 Phthalic acidNTYes
 Iso-phthalic acidNoNT
 3-Chlorobenzoic acidNoNT
 ρ-Aminobenzoic acidNoNT

Production of indigo from indole in PAH cultures.  For Mycobacterium sp. strain 1B, only phenanthrene induced significant amounts of indigo production when indole was included in BSMY–PAH broths (Table 1). This is different to previous findings for VUN 10,010, where phenanthrene, fluorene and pyrene all induced production of indigo (Boonchan 1998).

Growth on other carbon sources

Growth of Mycobacterium sp. strain 1B was only obtained on Tween-20 and -80, sodium pyruvate, succinic acid, glucose and straight chain hydrocarbons n-decane and n-hexadecane. The highest growth was obtained on the two hydrocarbon substrates (n-decane and n-hexadecane), with O.D.600 nm more than 0·5 after 3 weeks. For the other carbon sources tested, O.D.600 nm was between 0·1 and 0·2 after 3-week growth. Mycobacterium sp. strain 1B was not able to grow on or degrade any of the chlorinated substrates tested. Appreciable differences were observed between the utilizable substrate range of Mycobacterium sp. strain 1B and that of the original S. maltophilia culture (Table 1).


There has been growing interest in mycobacterial strains as potential bioremediation agents and as important components of indigenous PAH- and other xenobiotic-degrading populations in the environment (Cheung and Kinkle 2001). Many bacterial isolates capable of degradation of high molecular weight PAHs are actinomycetes (Kanaly and Harayama 2000). In this case, a Mycobacterium strain has been isolated from a previously characterized bacterial culture capable of benzo[a]pyrene cometabolism when grown on pyrene.

Identification of the isolate obtained in this study was complicated by the presence of two distinct copies of the 16S rRNA gene. Previous studies of Mycobacterium and other types of bacteria have also shown diversity in the number of copies of 16S genes and the sequence of these copies (Ninet et al. 1996; Reischl et al. 1998). Ninet et al. (1996) found 1.2% (18 bp over 1540 nucleotides total length) differences between two copies of 16S rRNA within a single strain of Mycobacterium. The sequence differences were obtained within regions 177–200 and 450–480, which is also where the main regions of difference between sequences were obtained in this study. Reischl et al. (1998) obtained similar results when comparing copies of 16S sequence from a strain of M. celatum, although a smaller number of mismatches were found in this case. The strain isolated in this study belongs to the fast-growing group of mycobacteria within the M. tuberculosis subgroup. Fast-growing mycobacterial strains have two copies of the 16S rRNA gene, as opposed to slow-growing strains which generally have only a single copy. The copies in fast-growing strains are generally thought to be identical, apart from the above exceptions. Although previous work has shown different 16S sequences from fast-growing mycobacterial clinical strains, this is the first instance of this occurring from an environmental mycobacterial isolate.

The Mycobacterium strain in this study was not readily isolated from an initial mixed culture, but required serial enrichment with pyrene as sole carbon and energy source. This is similar to results obtained by Poelarends et al. (1999) and Sutherland et al. (2002) who isolated Mycobacterium sp. strains capable of growth on 1,2-dibromoethane and endosulfan, respectively. The strains were only able to be isolated from mixed soil enrichment cultures after repeated subculturing with 1,2-dibromoethane or endosulfan as sole sources of carbon and energy. Attempts to isolate 1,2-dibromoethane- or endosulfan-degrading strains from the mixed cultures without any prior adaptation were unsuccessful. It is likely that the Mycobacterium strain isolated in this study was always part of the S. maltophilia bacterial culture isolated by Boonchan et al. (2000), but only became a more prominent component of the culture after repeated culturing with pyrene as sole source of carbon and energy. We suggest that VUN 10,010 is a mixed culture containing at least two strains (S. maltophilia and Mycobacterium sp. strain 1B) that are involved in PAH degradation. It is likely that although the Mycobacterium strain does not appear to have inherent benzo[a]pyrene-degrading characteristics under the conditions tested here, it does contribute some metabolic capabilities to benzo[a]pyrene and other PAH degradation by the fungal-bacterial co-culture previously described (Boonchan et al. 2000). Addition of pyrene inhibits benzo[a]pyrene degradation in several strains of Mycobacterium (McClellan et al. 2002; Bogan et al. 2003). It is possible that pregrowth on pyrene and addition of pyrene to benzo[a]pyrene degradation trials has inhibited benzo[a]pyrene degradation in Mycobacterium sp. strain 1B in this study. Further experiments with other co-substrates (i.e. phenanthrene and diesel; Juhasz et al. 1997; Kanaly et al. 2000; McClellan et al. 2002) and in combination with other microbes are required to determine whether this strain plays a significant role in benzo[a]pyrene cometabolism or degradation.

Comparison of the characteristics of the S. maltophilia VUN 10,010 original culture with the newly isolated Mycobacterium sp. strain 1B indicated differences in not only PAH degradation but other aspects of microbial metabolism. Salicylic acid is known to be an inducer of PAH catabolism in some bacterial strains (Boonchan 1998). It was interesting to note that addition of salicylic acid did not enhance degradation of any of the PAHs tested. Salicylic acid was also not used as a growth substrate by the Mycobacterium strain. Salicylic acid is a known metabolite in naphthalene degradation and induces gene expression in some (mainly Gram-negative) PAH-degrading strains. Previous studies have isolated PAH degradation products from Mycobacterium strains, and it is clear that several different mechanisms (mono-oxygenase, dioxygenase and cytochrome P-450-dependent) are functioning in these strains (Heitkamp et al. 1988; Moody et al. 2001). In mycobacterial PAH degradation pathways, compounds such as protocatechuic acid, cinnamic acid and phthalic acid (Heitkamp et al. 1988; Rehmann et al. 1998; Vila et al. 2001) have been identified as metabolic intermediates from pyrene and phenanthrene degradation. It is possible that the presence of salicylic acid inhibits activity in these mycobacterial PAH-degradation pathways or is toxic to the strain at the concentration tested (100 mg l−1). Naphthalene is also known to be toxic to some bacteria. The Mycobacterium strain isolated here could not grow on or degrade naphthalene under the conditions tested. Unpublished observations from our laboratory also indicate toxicity of naphthalene to the Mycobacterium strain, as evidenced by cell death after inoculation into broths containing naphthalene as sole carbon and energy source.

Production of indigo from indole is indicative of bacterial dioxygenase activity (Ensley et al. 1983). Only phenanthrene induced dioxygenase activity in Mycobacterium sp. strain 1B, providing evidence that other mechanisms of degradation (i.e. mono-oxygenase) of fluoranthene and pyrene are potentially functioning in this strain. This is consistent with the findings of Heitkamp et al. (1988), who identified mono-oxygenase activity during pyrene degradation by a Mycobacterium sp. Further experiments utilizing radiolabelled substrates and isolation of metabolites are required to determine the mechanism of enzymatic attack on pyrene and other PAHs by Mycobacterium sp. strain 1B.

The ability of Mycobacterium sp. strain 1B to grow on straight chain hydrocarbons as well as polycyclic aromatic hydrocarbons appears to be a common feature of environmental Mycobacterium isolates. Churchill et al. (1999) isolated a Mycobacterium strain (CH-1) from contaminated sediment. The strain was capable of growing on phenanthrene, fluoranthene and pyrene as well as alkanes (hexadecane, octacosane, etc.). Mycobacterium strains LB501T and LB307T, with anthracene- and phenanthrene-degrading abilities, respectively, can also grow on alkanes (decane, dodecane and hexadecane; Wick et al. 2002) as sole carbon and energy sources. Bogan et al. (2003) described a strain of M. austroafricanum also capable of growth on and mineralization of several PAHs (phenanthrene, fluoranthene and pyrene) and straight chain hydrocarbons (dodecane and hexadecane). The isolate of Bogan et al. (2003) was also able to degrade benzo[a]pyrene, although no mineralization of benzo[a]pyrene was observed under any circumstances.

It is clear from previous studies and the work performed here that it is unlikely that rapid degradation of significant amounts of high molecular weight PAHs such as benzo[a]pyrene can be achieved by single isolates of either bacteria or fungi. Although the original study by Boonchan et al. (2000) showed benzo[a]pyrene cometabolism by the S. maltophilia 10,010 strain, the results of this study question whether the whole process was performed by one strain, or whether the Mycobacterium isolated here also contributed to the process. The combined enzymatic activity of several microbial components appears to enhance the rate of degradation of recalcitrant contaminants such as PAHs (Boonchan et al. 2000). The use of bacterial or bacterial/fungal consortia as inoculants for bioremediation shows promise, provided stable activity of the consortia can be maintained over time. Further research into these consortia may be able to determine the necessary components and their functions, to further understand the role of different strains in degradation of polycyclic aromatic hydrocarbons in the environment.


The authors would like to acknowledge Grant Stanley and Belinda Davis from Victoria University of Technology for supply of the bacterial PAH-degrading cultures and Bruce Hawke from CSIRO, Adelaide, for FAME analysis. Three anonymous reviewers, Chris Lease and Albert Juhasz are acknowledged for critical reading of the manuscript. C.E.D. was supported by a Flinders University postgraduate scholarship.