Phenanthrene biodegradation by halophilic Martelella sp. AD-3


  • T.-c. Feng,

    1. State Environmental Protection Key Laboratory of Environmental Risk Assessment and Control on Chemical Process, School of Resources and Environmental Engineering, East China University of Science and Technology, Shanghai, China
    Search for more papers by this author
  • C.-z. Cui,

    Corresponding author
    • State Environmental Protection Key Laboratory of Environmental Risk Assessment and Control on Chemical Process, School of Resources and Environmental Engineering, East China University of Science and Technology, Shanghai, China
    Search for more papers by this author
  • F. Dong,

    1. State Environmental Protection Key Laboratory of Environmental Risk Assessment and Control on Chemical Process, School of Resources and Environmental Engineering, East China University of Science and Technology, Shanghai, China
    Search for more papers by this author
  • Y.-y. Feng,

    1. State Environmental Protection Key Laboratory of Environmental Risk Assessment and Control on Chemical Process, School of Resources and Environmental Engineering, East China University of Science and Technology, Shanghai, China
    Search for more papers by this author
  • Y.-d. Liu,

    1. State Environmental Protection Key Laboratory of Environmental Risk Assessment and Control on Chemical Process, School of Resources and Environmental Engineering, East China University of Science and Technology, Shanghai, China
    Search for more papers by this author
  • X.-m. Yang

    1. State Environmental Protection Key Laboratory of Environmental Risk Assessment and Control on Chemical Process, School of Resources and Environmental Engineering, East China University of Science and Technology, Shanghai, China
    Search for more papers by this author


Chang-zheng Cui, State Environmental Protection Key Laboratory of Environmental Risk Assessment and Control on Chemical Process, School of Resources and Environmental Engineering, East China University of Science and Technology, Shanghai 200237, China. E-mail:



To investigate the phenanthrene-degrading abilities of the halophilic Martelella species AD-3 under different conditions and to propose a possible metabolic pathway.

Methods and Results

Using HPLC and GC-MS analyses, the phenanthrene-degrading properties of the halophilic strain AD-3 and its metabolites were analysed. This isolate efficiently degraded phenanthrene under multiple conditions characterized by different concentrations of phenanthrene (100–400 mg l−1), a broad range of salinities (0·1–15%) and varying pHs (6·0–10·0). Phenanthrene (200 mg l−1) was completely depleted under 3% salinity and a pH of 9·0 within 6 days. The potential toxicity of phenanthrene and its generated metabolites towards the bacterium Vibrio fischeri was significantly reduced 10 days after the bioassay. On the basis of the identified metabolites, enzyme activities and the utilization of probable intermediates, phenanthrene degradation by strain AD-3 was proposed in two distinct routes. In route I, metabolism of phenanthrene was initiated by the dioxygenation at C-3,4 via 1-hydroxy-2-naphthoic acid, 1-naphthol, salicylic acid and gentisic acid. In route II, phenanthrene was metabolized to 9-phenanthrol and 9,10-phenanthrenequinone. Further study indicated that strain AD-3 exhibited a wide spectrum of substrate utilization including other polycyclic aromatic hydrocarbons (PAHs).


The results suggest that strain AD-3 possesses a high phenanthrene biodegradability and that the degradation occurs via two routes that remarkably reduce toxicity.

Significance and Impact of the Study

To the best of our knowledge, this work presents the first report of phenanthrene degradation by a halophilic PAH-degrading strain via two routes. In the future, the use of halophilic strain AD-3 provides a potential application for efficient PAH-contaminated hypersaline field remediation.


In nature, many marine and coastal sites are frequently polluted by polycyclic aromatic hydrocarbons (PAHs) as a result of occasional accidents, such as oil spills during oil exploration, transport and use and industrial processes. PAHs have attracted significant attention for decades because of their prolonged persistence, recalcitrance and potential mutagenic and carcinogenic properties (Haritash and Kaushik 2009). The microbial degradation of PAHs plays a significant role in treating these contaminated marine environments. When there is little to no indigenous microflora, bioaugmentation may be a potential strategy to enhance the processing efficiency of bioremediation (Mohan et al. 2006). Therefore, it is extremely vital to select appropriate halophilic or halotolerant PAH-degraders to aid in the clean-up of contaminated marine areas. Compared with nonhalophilic degraders, halophilic degraders have dual capabilities of both being able to maintain an osmotic balance with their external environment and treat contaminants (Zhuang et al. 2010).

Numerous species of halophilic microbes that are capable of degrading PAHs have been isolated from marine or saline water and sediment (Kasai et al. 2003; Li and Bai 2005). To date, most studies on halophilic micro-organisms have concentrated on strain identification, characterization and PAH biodegradability evaluation (Hedlund et al. 1999; Melcher et al. 2002; Zhao et al. 2009). However, to the best of our knowledge, relatively little attention has been paid to the metabolic products and pathways involved in the biodegradation by halophilic bacteria (Arulazhagan and Vasudevan 2009, 2011). Moreover, decreased concentrations of parent contaminants do not always mean the reduction in overall toxicity of biodegradation microcosms because by-products can potentially be toxic (Song et al. 2007; Elgh-Dalgren et al. 2011). It is imperative to undertake risk assessments to measure the total concentration of target PAHs and to monitor the toxicity of residuals and degradation of intermediate products. A standardized ecotoxicity test (Microtox®), performed on the marine bacterium Vibrio fischeri, is a sensitive biological assay that is suitable for assessing the toxicity of PAHs and their degradation products (Pagnout et al. 2006). According to Kazunga et al. 2001, PAH-degraders with complete degradation pathways and no potentially toxic metabolites are more efficient and thus are valuable in in situ bioremediation.

This study was aimed at evaluating the phenanthrene degradation capacity of the halophilic strain AD-3, a soil isolate that was enriched from a petroleum-contaminated site with high salinity. Another goal was to identify the possible metabolites of phenanthrene degradation under optimal conditions and to determine toxicity dynamics in strain AD-3's metabolism process. In addition, some organic compounds, including possible phenanthrene intermediates and other PAHs (naphthalene, anthracene, pyrene and benzo[a]pyrene), were selected as growth substrates for strain AD-3. Enzyme assays were conducted to examine key enzymes involved in phenanthrene metabolism by strain AD-3. On the basis of the previous studies, we propose a possible metabolic pathway for the degradation of phenanthrene in the halophilic Martelella sp. strain AD-3.

Materials and methods

Chemicals and media

Phenanthrene, naphthalene, anthracene, pyrene and benzo[a]pyrene were purchased from Fluka (Germany) with a purity above 96%. Salicylic acid, succinic acid, o-phthalic acid, gentisic acid, 1-hydroxy-2-naphthoic acid, 1-naphthol, 2-hydroxy-1-naphthoic acid and catechol were purchased from Sigma-Aldrich (St Louis, MO, USA) and were of the highest purity available. All other solvents and chemicals used were of reagent grade or better.

Each PAHs stock solution was prepared by dissolving the appropriate amount of phenanthrene, naphthalene, anthracene, pyrene and benzo [a] pyrene in 1,2-dichloroethane at concentrations of 100, 50, 10, 10 and 5 g l−1, respectively. Stock solutions of salicylic acid, succinic acid, o-phthalic acid and catechol were prepared by dissolving compounds individually in distilled water at an equal concentration of 10 g l−1. Gentisic acid, 1-hydroxy-2-naphthoic acid, 1-naphthol and 2-hydroxy-1-naphthoic acid stock solutions were made by dissolving the compounds individually in anhydrous diethyl ether at an equal concentration of 50 g l−1. All stock solutions were filtered through a 0·22-μm syringe filter.

Both the growth and biodegradation liquid media were carbon-free Mineral Salt Medium (MSM) with varying concentrations of salinity and were made with slightly modifications to the standard protocol as described by Zhao et al. 2009;. One litre of MSM with 5% total salinity contains the following: 1·0 g of NH4Cl, 39·73 g of NaCl, 3·85 g of MgCl2∙6H2O, 5·23 g of MgSO4∙7H2O, 0·38 g of CaCl2∙7H2O, 1·1 g of KCl, 0·05 g of NaHCO3, 0·13 g of NaBr, 0·25 g of K2HPO4 and 2 ml of trace-element solution. The trace-element solution was prepared as described by Zhao et al. 2009. The pH of the MSM was adjusted with NaOH (0·1 mol l−1) or HCl (0·1 mol l−1), as required. Assessment of cell growth was performed based on colony forming unit (CFU) plate-counting on nutrient agar consisting of 10 g l−1 peptone, 10 g l−1 yeast extract, 5 g l−1 NaCl and 3·0% agar. All of the plates were maintained at 30°C for 2 days.

Experimental strain

The PAH-degrading bacterium Martelella sp. AD-3 (CTCCM 2011218) used in the study was previously isolated by our laboratory from petroleum-contaminated soil with high salinity by enrichment on anthracene as the sole carbon and energy source (Cui et al. 2012). To prepare seed cultures, a pure bacterial culture was cultivated in MSM containing phenanthrene at a concentration of 200 mg l−1 for 2 days. Cultured cells were harvested by centrifugation (14 000 g, 3 min) and washed three times with sterile MSM. The cells were subsequently resuspended in the same sterile MSM, stored at 4°C and used as the strain source in the biodegradation experiments.

Phenanthrene biodegradation procedure

To study the biodegradation of phenanthrene by the halophilic strain AD-3, the required amount of phenanthrene stock solution was transferred to a 100-ml sterilized Erlenmeyer flask. After 1,2-dichloroethane evaporated in a fume hood, leaving a thin film of phenanthrene on the flask bottom, autoclaved MSM (29 ml) was added to each flask, and the samples were inoculated with 1 ml cell suspension to bring the final bacterial population to approximately 107 CFU ml−1. The experiments were replicated three times and were conducted to investigate the effects of several factors (phenanthrene concentration, salinity and pH) on phenanthrene degradation. Optimization was performed for each condition. The concentration levels of phenanthrene were 100, 200, 300 and 400 mg l−1, and MSM ranged in salinity from 0·02% to 20%. To optimize the initial pH of MSM, nine different pH values ranging from 6·0 to 10·0, adjusted with NaOH (0·1 mol l−1) or HCl (0·1 mol l−1), were tested. The biodegradation curve of phenanthrene by strain AD-3, as well as cell growth (CFU), was measured under the optimal degradation conditions of this strain. All treatments were performed in triplicate for 6 days and incubated aerobically in a dark incubator at 30 ± 0·5°C with agitation at the speed of 150 rev min−1. A control containing all of the materials without an inoculum was used to detect any abiotic loss of phenanthrene. During the 6-day incubation time, each phenanthrene degradation percentage was tested.

Analytical methods

At specific time intervals during the 6-day incubation period, triplicate flasks and the relevant control were removed, and 1 ml of culture from each flask was withdrawn for cell growth estimation.

The pH of the remaining aqueous fraction was lowered to 2·0 with 5 mol l−1 HCl. Analysis of the residual phenanthrene concentration in the culture fluid was performed by extracting the acidified culture twice with same amount of ethyl acetate. The two ethyl acetate extracts were combined, dried with anhydrous Na2SO4 and evaporated under gentle nitrogen flow at 40°C. The residues were dissolved in methanol and stored at 4°C prior to analysis. The residual phenanthrene was analysed by reversed-phase HPLC (Shimadzu LC-2010HT, Kyoto, Japan) equipped with a UV detector and a 4·6 × 250 mm XDB-C18 column with its corresponding guard column (Agilent, Palo Alto, CA, USA). A methanol-water mixture (80/20, v/v) was used as the mobile phase at a flow rate of 1 ml min−1. The UV absorbance spectra were obtained on-line at 254 nm. The biodegradation percentage of phenanthrene was calculated as the difference in residual phenanthrene concentration between the control and the inoculated flasks.

For the analysis and identification of phenanthrene metabolites, bacterial cells and culture medium were thoroughly extracted six times with equal volumes of ethyl acetate. Three extractions were performed at a neutral pH and three were performed at a pH of 2·0 after acidification with 5 mol l−1 HCl. The neutral and acidic extracts were pooled, dried on anhydrous Na2SO4 and the solvents were evaporated to dryness under gentle high-purity nitrogen flow at 40°C. The residuals were dissolved in 0·25 ml acetonitrile. The silylation prior to GC-MS analysis was performed with bis-(trimethylsilyl) trifluoroacetamide (BSTFA; Sigma, St Louis, MO, USA) using a method described by Moody et al. 2004. Analysis of phenanthrene metabolites was performed using gas chromatography (PerkinElmer Clarus 500, Waltham, MA, USA) coupled to a Turbo mass spectrophotometer (PerkinElmer Clarus 500, USA) with a HP-5MS (30 m × 0·25 mm i.d × 0·25 μm film thickness) fused silica capillary column. The column temperature procedure was as follows: maintained at 80°C for 1 min, increased to 160°C at a increase in 15°C min−1, then increased by 5°C min−1 to 300°C and held for 10 min. The mass spectrometer was operated in the electron impact (EI) ionization mode at an energy of 70 eV. Ultra-pure helium was used as the carrier gas with a flow rate of 1 ml min−1. The injection volume was 1 μl, and the injection mode was splitless. The injector, ion source and detector temperature were set at 280, 250 and 300°C, respectively.

Enzyme assays

Enzyme activities in selected cell-free extracts were applied to confirm the route of metabolism of phenanthrene by strain AD-3. Reported spectrophotometric methods were used to monitor the activities of 1-hydroxy-2-naphthoic acid hydroxylase (Deveryshetty and Phale 2010), 1-hydroxy-2-naphthoic acid dioxygenase (Deveryshetty and Phale 2009), 1-naphthol-2-hydroxylase (1-NH) (Swetha et al. 2007), salicylate 5-hydroxylase (S5H) (Zhou et al. 2002), gentisate 1,2-dioxygenase (GDO) (Fu and Oriel 1998), catechol 1,2-dioxygenase (Doddamani and Ninnekar 2000) and catechol 2,3-dioxygenase (Doddamani and Ninnekar 2000) in cell-free extracts of Martelella sp. strain AD-3 grown on various substrates which included phenanthrene, salicylic acid and glucose. The preparation of the cell-free extracts was carried out using a previously described procedure (Fu and Oriel 1998). A published Bradford assay method (1976), with bovine serum albumin as the standard, was used to determine protein concentrations. One unit of enzyme activity was defined as the amount of enzyme that catalysed the disappearance of 1 μmol of NADH or the formation of 1 μmol of product per min. The specific activities were expressed as μmol min−1 mg−1 protein.

Toxicity assay

The toxicity generated by phenanthrene, its intermediates and dead-end metabolites in the process of phenanthrene degradation under optimum conditions was estimated using the luminescent bacterium Vibrio fischeri. Sample preparation and ecotoxicity test procedures were followed according to the method described by Pagnout et al. 2006. Briefly, extractions were performed with dichloromethane (DCM). The six extracts (three neutral and three acidic extracts) were pooled and dried on anhydrous Na2SO4, followed by a DCM removal under gentle high-purity nitrogen flow at 40°C. The residuals that were dissolved in dimethyl sulphoxide (DMSO) with a concentration factor of 10 were then subjected to the ecotoxicity test.

Utilization of other organic compounds

The ability of the strain to grow on alternative organic compounds, including naphthalene, anthracene, pyrene, benzo[a]pyrene, salicylic acid, succinic acid, o-phthalic acid, gentisic acid, 1-hydroxy-2-naphthoic acid, 1-naphthol, 2-hydroxy-1-naphthoic acid and catechol, was evaluated. The test was performed using MSM (100 ml, pH 7·5) containing a given compound with a concentration of 100 mg l−1 in a sealed 250-ml flask. Exceptions to this were the concentrations of naphthalene, anthracene, pyrene, benzo[a]pyrene added to flasks, as concentrations were set at 200, 50, 50 and 10 mg l−1, respectively. Cometabolism of benzo[a]pyrene by the halophilic Martelella sp. AD-3 was performed using MSM supplemented with yeast extract at 0·02% (w/v). The uninoculated control and duplicate samples were prepared for each treatment. After several days of incubation at 30 ± 0·5°C and an agitation speed of 150 rev min−1, the reaction solutions and controls were sampled and examined for their OD at 600 nm (OD600).

Statistical analysis

Statistical analysis was carried out using SPSS version 17.0 software. Parametric one-way analysis of variance (anova) was utilized to compare the effects of factors on phenanthrene degradation. Multiple comparison analysis was performed using the least significant difference (LSD) test at the significant level of P < 0·05. The results were expressed as the mean values and corresponding standard errors.


Effect of culture conditions on phenanthrene degradation

Figures 1–3 showed the degradation of phenanthrene by the halophilic micro-organism AD-3 under different incubation conditions. The effect of phenanthrene concentration on degradation by strain AD-3 was examined in terms of biological rate of degradation. Figure 1 demonstrated that the phenanthrene degradation rate in the 200 mg l−1 treatment group was significantly higher than the other treatments (F = 127·5, P = 0·000). The degradation rate within 6 days was 13·7, 21·5, 14·6, 12·9 mg l−1 day−1 for treatments with 100, 200, 300, 400 mg l−1, respectively.

Figure 1.

Effect of initial phenanthrene concentration on degradation by the halophilic micro-organism Martelella sp. AD-3. Cells were cultivated in MSM (pH 7·5, salinity 5%) at 30°C and shaken at 150 rev min−1 in the dark for 6 days.

Figure 2.

Effect of salinity on phenanthrene degradation by the halophilic micro-organism Martelella sp. AD-3. Cells were cultivated in MSM (pH 7·5) containing 200 mg l−1 phenanthrene at 30°C and shaken at 150 rev min−1 in the dark for 6 days.

Figure 3.

Effect of pH on phenanthrene degradation by the halophilic micro-organism Martelella sp. AD-3. Cells were cultivated in MSM (salinity 3%) containing 200 mg l−1 phenanthrene at 30°C and shaken at 150 rev min−1 in the dark for 6 days.

Regarding the salinity of the culture media, as shown in Fig. 2, a salinity of 3% would be considered as the most efficient salinity for phenanthrene degradation (F = 98·48, P < 0·001), although the means are not significantly different between the 3% and 2% salinity treatments based on multiple comparison results (P > 0·05). In the case of low salinity (0·1%) or high salinity (15%), biodegradation still occurred when the cells were exposed to 200 mg l−1 phenanthrene. However, no significant degradation occurred at a salinity above 15% or below 0·1%, even after 6 days of incubation (data not shown). This indicates that the growth on phenanthrene required a minimal salt concentration of 0·1%, whereas it could be fully inhibited by a salt concentration exceeding 15%. Therefore, the strain exhibited a very broad salinity profile.

Figure 3 indicated that, under media pH ranging from 6·0 to 10·0, strain AD-3 could utilize phenanthrene as its sole carbon and energy source. Additionally, a majority of phenanthrene (>60%) was removed. The highest depletion percentage (100%) was achieved when the initial pH value was 9·0, and phenanthrene degradation in the pH 9·0 treatment group was significantly superior to that in the other treatments based on manova results (F = 7·549, P < 0·05). No significant difference was observed between the pH 9·0 and 8·5 treatments (P > 0·05).

Growth and biodegradation of phenanthrene under optimum conditions

The time-courses of cell growth and phenanthrene degradation with an initial concentration of 200 mg l−1 were determined under optimized conditions (salinity 3%, pH 9·0). As shown in Fig. 4, phenanthrene was rapidly degraded by strain AD-3. After 6 days of incubation, the concentration of this compound decreased to undetectable levels in the culture medium. A rapid increase of the cell density, corresponding to the sharp decrease in residual phenanthrene, was observed during the first 36 h. However, there was a gradual decrease in cell density from 36 to 144 h. The colour changes of the culture media over the incubation time were as follows: yellowish-orange colour, followed by a deeper yellow and then the colouration subsided gradually.

Figure 4.

Growth of Martelella sp. AD-3 on utilization of phenanthrene as the sole carbon and energy source. Cells were cultivated in MSM (salinity 3%, pH 9·0) containing 200 mg l−1 phenanthrene at 30°C and shaken at 150 rev min−1 in the dark for 6 days. Values represent the average of triplicate determinations and standard deviations. (▷) Residual phenanthrene (CK); (▼) Residual phenanthrene (AD-3) and (▲) Cell density (AD-3).

Phenanthrene degradation kinetics was estimated in a first-order degradation rate model. The first-order rate coefficient k1 (day−1) can be calculated according to the equation Xt = X0 ekt, where Xt is the phenanthrene concentration (mg l−1) in liquid cultures at time t, and X0 is the initial phenanthrene concentration. The k1 value for phenanthrene degradation by strain AD-3 under optimized conditions was equal to 0·84 per day.

Phenanthrene metabolic products

The GC-MS analysis of derivatized culture extracts demonstrated that four metabolites (Table 1), designated as Peak I, Peak II, Peak III and Peak IV, were formed during the degradation of phenanthrene by strain AD-3. Peak I had a mass spectrum with a molecular ion (M+) at m/z 216 and fragment ions at m/z 201, 185, 170, 115 and 73 and was confirmed as derivatized 1-naphthol by comparing its retention time and mass spectrum with the use of an authentic standard. Peak II had a molecular ion (M+) at m/z 332 and major fragment ions at m/z 317, 185, 147 and 73, which was consistent with those of derivatized 1-hydroxy-2-naphthoic acid reported by Zhong et al. 2011;. Therefore, Peak II was identified as derivatized 1-hydroxy-2-naphthoic acid. The mass spectrum of Peak III showed a molecular ion (M+) at m/z 208 and fragment ions at m/z 180, 151 and 126, indicating that the compound could be 9,10-phenanthrenequinone on the basis of its m/z values and comparison of the mass spectra with the NIST library available in GC-MS. Peak IV was identified as derivatized forms of 9-phenanthrol, as its mass spectrum had an identical fragmentation to that of 9-phenanthrol reported by Zhong et al. 2011, which had a M+ of 266 and loss of fragments of 15 (-CH3) and 31 (-OCH3).

Table 1. Mass spectra of identified phenanthrene metabolites formed from its degradation by a pure culture of the halophilic strain AD-3
Peakm/z of fragment ions (% relative intensity)Identified metabolite
I216 (M+, 73), 201 (-CH3, 100), 185 (-OCH3, 61), 170 (-OCH3-CH3, 12), 115 (25), 73 (TMS, 42)Derivatized 1-naphthol
II332 (M+, 13), 317 (-CH3, 100), 243 (-CH3-OCH3-CO-CH3, 15), 185 (36), 147 (31), 73 (TMS, 78)Derivatized 1-hydroxy-2-naphthoic acid
III208 (M+, 47), 180 (–CO, 100), 151 (–CO-CHO, 69), 126 (–CO-CHO–C2H, 16)9,10-Phenanthrenequinone
IV266 (M+, 100), 251 (-CH3, 54), 235 (-OCH3, 20), 73 (TMS, 45)Derivatized 9-phenanthrol

Enzyme activities in cell-free extracts

An enzyme extract prepared from phenanthrene-grown cells showed 1-hydroxy-2-naphthoic acid hydroxylase, 1-NH and S5H activity, but failed to show the activity of 1-hydroxy-2-naphthoic acid dioxygenase, catechol-1,2-dioxygenase, catechol-2,3-dioxygenase or GDO. Conversely, salicylic acid-grown cells showed comparatively reduced activities for 1-hydroxy-2-naphthoic acid hydroxylase and 1-NH and no activity of 1-hydroxy-2-naphthoic acid dioxygenase or catechol-1,2- and 2,3-dioxygenase. However, significantly higher activities of S5H and GDO were observed. The lack of activity of these enzymes in glucose-grown cell-free extracts indicated that the enzymes responsible for phenanthrene metabolism were inducible (Table 2).

Table 2. Specific activity of various enzymes in the cell-free extract of the halophilic strain AD-3 grown on different substrates to the mid-exponential phase
EnzymesSpecific activitya (nmol min−1 mg−1 protein)
GlucoseSalicylic acidPhenanthrene
  1. a

    The values are the average values measured in triplicate.

  2. b

    Activity could not be detected.

1-Hydroxy-2-naphthoic acid hydroxylaseNDb34·190·7
1-Hydroxy-2-naphthoic acid dioxygenaseNDNDND

Toxicity evolution

As reflected in Fig. 5, the dynamics of toxicity evolution was inconsistent with that of a phenanthrene degradation curve, showing a temporary increase in toxicity and the highest luminescence emission inhibition achieved at the 12th hour. The results also indicated that the toxicity of the liquid cultures during the first 6 days (except at the 3rd day) was higher than that measured in the control experiments. In addition, toxicity was still present when phenanthrene was completely degraded during 6 days of incubation. The overall toxicity during phenanthrene consumption by strain AD-3 underwent complex changes, but was significantly reduced by the end of the assay (10 days).

Figure 5.

Dynamics of toxicity evolution in the phenanthrene degradation process of the halophilic micro-organism Martelella sp. AD-3. Cells were cultivated in MSM (salinity 3%, pH 9·0) containing 200 mg l−1 phenanthrene at 30°C and shaken at 150 rev min−1 in the dark for 10 days. Values represent the average of triplicate determinations and standard deviations. (□) CK and (▶) AD-3.

Growth on other organic compounds

Halophilic Martelella sp. strain AD-3 was found to have the ability to utilize salicylic acid, succinic acid, gentisic acid, 1-hydroxy-2-naphthoic acid, 1-naphthol, 2-ring PAHs naphthalene, 3-ring PAHs anthracene or 4-ring PAHs pyrene as the sole carbon source, which was evidenced by each significant cell density increments of strain AD-3 in the test flasks after 3 days of incubation. The exception to this was pyrene, which exhibited significant increments after 10 days. In addition, the strain degraded 5-ring PAHs benzo[a]pyrene via cometabolism in the presence of yeast extract (0·02%). The strain exhibited a very broad substrate profile. The detailed results are shown in Table 3.

Table 3. Profile of utilization of various substrates by strain AD-3
SubstratesConcentration (mg l−1)Growth
  1. (++) Good growth: OD600 > 0·2; (+) growth: 0·1 < OD600 <0·2; (−) no growth: OD600 < 0·02. Cells were cultivated in MSM (salinity 3%, pH 7·5) containing each substrate individually at specific concentration at 30°C and 150 rev min−1 in the dark for 3 days incubation except pyrene and benzo[a]pyrene for 10 days.

Salicylic acid100++
2-Hydroxy-1-naphthoic acid100
1-Hydroxy-2-naphthoic acid100+
o-Phthalic acid100
Succinic acid100++
Gentisic acid100++


It is important to emphasize the optimization of abiotic factors of contaminants during the biodegradation process because of their effect on microbial biodegradation efficiencies. This study analysed three experimental factors, namely initial phenanthrene concentration, salinity and pH on phenanthrene biodegradation by the halophilic strain AD-3. With regards to initial phenanthrene concentration, in this study, the low phenanthrene concentration (<200 mg l−1) may not be sufficient to support cell growth. However, the high concentration of phenanthrene (>200 mg l−1), together with its metabolites, might show increased potential toxicity to strain AD-3. Thus, the phenanthrene degradation rates were decreased with both treatments. This is consistent with the findings of Ling et al. 2011; who also found the decreased PAHs efficiencies with the lower and higher dosages. Salinity is another abiotic factor influencing the osmotic pressure in bacterial cells (Tang et al. 2005). The ability of halophilic strain AD-3 tolerant to a wide range of salinities is very beneficial for its use in the treatment of PAH-contaminated marine environments. The impact of pH on PAHs degradation has been widely studied (Shin et al. 2004; Kim et al. 2005; Simarro et al. 2011). The previous works demonstrated that PAHs microbial degradation favoured a pH range between 5·5 and 7·8. However, our results were inconsistent with the previous studies, showing an optimum pH of 9·0. It is likely that this was because of the unique strain AD-3 derived from a petroleum-contaminated site of a highly saline and alkaline (average pH 7·3–7·8) environment in China.

There are published data on phenanthrene degradation efficiency in other PAH-degrading isolates. Tao et al. 2007 reported a phenanthrene-degrading strain, Sphingomonas sp. GY2B, that was isolated from contaminated soils was able to degrade 99·8% of phenanthrene (100 mg l−1) in 2 days. They also reported strain GY2B could grow on other aromatic compounds, such as naphthalene and 1-hydroxy-2-naphthoic acid. However, the test was conducted under nonsaline conditions. Tam et al. 2002 reported that a bacterium enriched from marine mangrove sediment in Hong Kong could remove 40% of phenanthrene (50 mg l−1) within 6 days at 35 g l−1 salinity. Zhao et al. 2009 evaluated the potential of phenanthrene degradation by a halophilic bacterial consortium at different salinity concentrations. They observed complete degradation of phenanthrene (80 mg l−1) within 5–8 days under 5%, 10% and 15% salinity. Arulazhagan and Vasudevan 2009, 2011 also obtained a moderately halophilic bacterial consortium and a halotolerant bacterial strain Ochrobactrum sp. VA1, which could deplete 89% of phenanthrene (50 mg l−1) and 92% of phenanthrene (3 mg l−1) within 4 days at 30 g l−1 salinity, respectively. In contrast, strain AD-3 showed high phenanthrene biodegradation potential despite relatively high concentrations of PAH. The higher biodegradability of strain AD-3 could possibly be attributed to its unique genetic make-up and active enzymes responsible for degradation. The molecular mechanism responsible for such a high capacity of phenanthrene removal will be further investigated in the future.

The toxicity assessment performed on phenanthrene and its metabolites after degradation by Martelella sp. AD-3 showed a high reduction in the overall toxicity potential, indicating the transformation of phenanthrene to less toxic or harmless by-products. Each detected metabolite, other than 1-hydroxy-2-naphthoic acid, produced from the degradation of phenanthrene appears more toxic in respect to the parent PAH towards Vibrio fischeri (McConkey et al. 1997; Parikh et al. 2004; Capotorti et al. 2005). On the basis of this finding, it is logical that the toxicity of the liquid cultures, especially during the first 6 days of incubation (except at day 3), was higher than that measured in the control experiments. It is well known that the characteristic yellow colour of the medium was caused by the occurrence of the accumulated metabolite 1-hydroxy-2-naphthoic acid (Guerin and Jones 1988). A possible explanation for the lower toxicity than the control at day 3 was the major amount of 1-hydroxy-2-naphthoic acid accumulated in cultivation media, which might be evidenced by the deeper yellow colouration of the culture at day 3 followed by gradual fading away. The data presented in the toxicity bioassay seem to be favourable to the use of Martelella sp. AD-3 in bioremediation. Phenanthrene degradation by strain AD-3 was recognized as a detoxification process, and this finding was similar to results reported by Molina et al. 2009. To conclude, more concern should be given to toxicity evolution during the biodegradation process to establish a feasible and efficient remediation strategy.

Phenanthrene was metabolized by Martelella sp. AD-3 via at least two distinct pathways (Fig. 6). The formation of 1-hydroxy-2-naphthoic acid indicates that dioxygenation occurred at the C-3 and C-4 position of phenanthrene to form 3,4-dihydroxyphenanthrene as the initial oxidation product followed by its meta-cleavage. However, 3,4-dihydroxyphenanthrene was not detected in the present study. It is well known that 1-hydroxy-2-naphthoic acid is degraded through two different routes, namely the ‘naphthalene’ or the ‘phthalic acid’ route (Deveryshetty and Phale 2009). The fact that strain AD-3 was able to grow on salicylic acid rather than o-phthalic acid as its sole carbon source showed the strain AD-3 followed the ‘naphthalene’ route, which could be further supported by the presence of 1-hydroxy-2-naphthoic acid hydroxylase and the absence of 1-hydroxy-2-naphthoic acid dioxygenase in the cell-free extract of strain AD-3. Previous studies have reported a transformation sequence from 1-hydroxy-2-naphthoic acid to salicylic acid via the formation of 1-naphthol as the intermediate (Samanta et al. 1999; Prabhu and Phale 2003; Tao et al. 2007). Interestingly, the identification of 1-naphthol from the culture confirmed the metabolic diversity reported. In the lower metabolic pathway, salicylic acid can be further converted to either catechol or gentisic acid (Woo and Park 2004). Both catechol and gentisic acid were then subject to ring fission to form TCA-cycle intermediates (Houghton and Shanley 1994). The catechol pathway appears to be more common than the gentisate pathway (Ishiyama et al. 2004). The gentisate pathway has been studied in phenanthrene metabolism by Burkholderia sp. C3 (Seo et al. 2007). The cells showed good growth on gentisic acid, but no growth on catechol, suggesting that the gentisic acid rather than the catechol pathway was involved in the phenanthrene metabolic pathway. Failure to detect the activities of catechol-1,2- and 2,3-dioxygenase associated with catechol metabolism, together with the existence of S5H, which is responsible for the conversion of salicylic acid to gentisic acid, could also rule out the possibility of a catechol pathway. However, it is noted that the activity of GDO could not be detected even in the late exponential and stationary phases of the phenanthrene-grown cells. This could be due to a very low concentration of this enzyme. The production of 9-phenanthrol and 9,10-phenanthrenequinone was identical to a previous report (van Herwijnen et al. 2003) for Sphingomonas sp. LB126. The observation of these two metabolites in samples suggested a second pathway existed, going through hydroxylation at the 9,10-positions. However, it is uncertain that whether the 9-phenanthrol is formed by a monohydroxylation of phenanthrene or by dehydration of 9,10-dihydroxyphenanthrene, which was not detected in our study. Initial attack on phenanthrene at C-9,10 positions is consistent with previous report on Mycobacterium strain S1 growing on anthracene in the presence of phenanthrene (Tongpim and Pickard 1999), a mixed culture of Mycobacterium sp. and Sphingomonas sp. (Zhong et al. 2011).

Figure 6.

Proposed pathway for the degradation of phenanthrene by the halophilic strain Martelella sp. AD-3 on the basis of identified metabolites, enzyme activity and utilization profile of different substrates. Compounds in brackets are hypothetical intermediates, but not detected.


In summary, we described the capability the halophilic strain Martelella sp. AD-3 to degrade phenanthrene and other PAH compounds and the pathway that is involved. Phenanthrene degradation under multiple conditions and its complete removal was observed in our laboratory experiments. Moreover, the overall toxic potential of degradation products was significantly reduced, as determined by an assay using the marine bacterium Vibrio fischeri. Strain AD-3 extended our knowledge of halophilic bacteria that can initiate an attack on phenanthrene through two different pathways: via the C-3,4 or C-9,10 positions. This finding warrants further extensive investigations of halophilic PAH-degrading bacteria. The strain AD-3 may have a potential for bioremediation of PAH-contaminated hypersaline sites.


This project is supported by special fund of the Fundamental Research Funds for the Central Universities (no. WB1114031), State Key Joint Laboratory of Environment Simulation and Pollution Control (no. 11K02ESPCT) and the Specialized Research Fund for the Doctoral Program of Higher Education (no. 20110074130002).