OUP user menu

Hydrocarbon utilization within a diesel-degrading bacterial consortium

Lena Ciric, James C. Philp, Andrew S. Whiteley
DOI: http://dx.doi.org/10.1111/j.1574-6968.2009.01871.x 116-122 First published online: 1 February 2010


Diesel fuel is a common environmental pollutant comprised of a large number of both aromatic and aliphatic hydrocarbons. The microbial degradation of individual hydrocarbons has been well characterized, however, the community dynamics within a system degrading a complex pollutant such as diesel fuel are still poorly understood. The growth capabilities of a diesel-degrading consortium, along with organisms isolated from a contaminated site, were investigated using molecular profiling, isolation, and physiological methods using 10 of the fuel's most abundant constituents as sole carbon sources. The results indicated that the degradation of the fuel's constituents may be shared among the diverse microbial community. Some organisms were capable of growth on the majority of the hydrocarbons tested, whereas others seemed specialized to only a few of the substrates.

  • hydrocarbon
  • degradation
  • community
  • diesel
  • isolate


Diesel fuel, a complex and common pollutant, is well characterized in terms of its main components (Bacha et al., 1998; Wang et al., 2005). It consists mainly of aliphatic hydrocarbons ranging from C9 to C23 as well as a number of aromatic compounds (Bacha et al., 1998). The susceptibility of hydrocarbons to microbial degradation is well documented, dating to the 1940s (Zobell, 1946), and varies according to their chemical structure. This chemical structure also affects the compounds' solubility and therefore bioavailability. Mid- to high-chain-length alkanes, C10–C24, all have very low water solubilities, however, are degraded with varying efficiency by many microorganisms despite this (Atlas, 1981; Singer & Finnerty, 1984; de Carvalho & da Fonseca, 2005). Aromatic compounds, including naphthalene, are more water soluble and are also readily degraded by microorganisms (Atlas, 1981; Gibson & Subramanian, 1984; Harayama, 1997; Samanta et al., 2002; Diaz, 2004). However, only limited research has focussed on the division of labour in a single system, in terms of the degradation of the constituent compounds.

For complex pollutants such as diesel, two scenarios could exist, independently or in combination: the presence of generalist degraders, which remediate a wide spectrum of compounds; or the presence of multiple, and potentially cooperative, degraders specialized to particular chemical species. The current study had two main aims: to investigate to what extent organisms found at a diesel-contaminated site undergoing remediation were capable of utilizing the fuel's constituents; and to determine carbon substrate specificity or preference. This was performed using a combination of molecular biology, isolation, and physiological analyses of the microbial consortium in order to better understand degradation processes and aid subsequent optimization of natural or engineered attenuation strategies.

Materials and methods

Field site

The study site was situated on an undisclosed oil rig building and maintenance site in the United Kingdom, where a remediation company, ERS Ltd (http://www.ersremediation.com/), had set up a recirculating pump system in order to remediate a large-scale diesel fuel spill. The volume of water pumped around the system was 600 000 L day−1. Approximately 500 L of diesel were physically skimmed off and recovered from the contaminated water daily. The treatment involved the application of a diesel-degrading multispecies consortium obtained from a series of enrichments performed on organisms indigenous to the site. The augmentation treatment ran for approximately 4 months when samples were taken from around the site in order to identify isolates capable of diesel degradation.

Isolation of diesel-degrading organisms from the study site

Diesel-contaminated water samples were collected from the groundwater bioremediation system situated at an undisclosed industrial site in the United Kingdom. For randomized isolation, 100 μL of the water samples taken were cultured for 96 h at room temperature on M9 agar (Maniatis et al., 1982) sprayed with 15 μL diesel fuel sterilized using a 0.2-μm PTFE filter (Nalgene). Representative single colonies were picked and frozen in 30% glycerol at −70 °C. In total, 47 organisms were isolated from samples taken from the site. The organisms were then screened by denaturing gradient gel electrophoresis (DGGE) to reveal replicates and 12 different species were finally identified.

16S rRNA gene sequence analysis of diesel-degrading isolates

The isolated organisms were identified by full-length 16S rRNA gene sequencing using universal primers, 27F and 1492R (Lane, 1991), and an ABI sequencer using the ABI Prism® BigDyeTM Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems) according to the manufacturer's instructions. The resulting sequence reads were assembled using sequencher software (Gene Codes), manually checked and edited, and finally identified on the basis of similarity using blastn protocols (http://www.ncbi.nlm.nih.gov/BLAST).

Culture of site-degrading consortium on diesel constituents

The multispecies consortium used as the inoculum at the on-site groundwater bioremediation system was obtained following a series of batch culture enrichments performed on indigenous organisms previous to the commencement of the present study. A sample of the bacterial consortium was taken from the site and frozen at −70 °C in 30% glycerol.

The consortium was cultured in triplicate using the top 10 diesel constituents individually under aerobic conditions at 28 °C with agitation at 200 r.p.m. in liquid M9 minimal medium (Maniatis et al., 1982) supplemented with 2 g L−1. of the individual carbon sources. The concentration of diesel fuel at the study site was found to be approximately 1 g L−1. and the slightly higher concentration was used in order to enrich for the degraders of specific diesel components. The carbon sources used were nine n-alkanes (C13–C21) and naphthalene, representing the top 10 constituents of the site-derived diesel determined by GC-MS (Fig. 1). The profile was shown to be slightly different in the aged and nonaged diesel fuel. Although the same pattern can be observed, showing a normal distribution, the C13–C17 alkanes were less abundant in the aged diesel fuel taken from the study site. The ranking of the compounds in terms of abundance (high to low) was as follows: C18, C17, C15, C16, C19, C14, C13, C20, C21, and naphthalene. After 1 week of growth, total community DNA was extracted from 1 mL culture.

Figure 1

(a) Abundance of the top 10 diesel constituents by area percentage is shown according to their carbon number obtained by GC-MS analysis of the diesel fuel obtained from the study site. (b) The mass percent constitution of nonaged diesel fuel is shown according to the carbon number (Bacha et al., 1998).

16S rRNA gene DGGE analysis of communities utilizing diesel constituents

DNA extraction followed by 16S rRNA gene PCR amplification and DGGE was carried out according to the methods of Griffiths (2000). The resulting DGGE profiles were analysed using principal component analysis (PCA) (Pearson, 1901; Griffiths et al., 2003), generating scatter plots. PCA aims to quantify the variability within a sample set resulting from particular components within the samples. The components of samples, in this case bands within each DGGE profile, are ranked and similarities identified. The resulting scatter plot shows these relationships graphically, where groupings along the two-component axes represent similarity. Separation along axis 1 is indicative of higher variability than that along axis 2.

Culture of diesel-degrading isolates on diesel constituents

Diesel-degrading site isolates were subcultured on M9 and diesel agar as above, transferred to M9 broth containing 1 g L−1 diesel and grown at room temperature for 48 h. Although the hydrocarbons are not entirely water soluble at this concentration, it was chosen to reflect that found at the study site. These cultures were then used to inoculate triplicate M9 broths containing one of 11 carbon sources (nine n-alkanes, C13–C21; naphthalene; and diesel) at 1 g L−1 and for 1 week at room temperature, agitated at 100 r.p.m. The increase in biomass was quantified by measuring OD600 nm at the start and the end of the week. A reading of OD600 nm is frequently used in studies characterizing the physiology of hydrocarbon utilization (Peng et al., 2007; Zeinali et al., 2007; Bouchez-Naitali & Vandecasteele, 2008; Binazadeh et al., 2009; Isaza & Daugulis, 2009). OD600 nm readings of negative controls containing only hydrocarbons were subtracted from the final reading to allow for any OD600 nm difference caused by factors other than microbial growth.

Results and discussion

Hydrocarbon utilization within the enriched consortium

The two main aims of the study were to ascertain to what extent site organisms were able to utilize diesel fuel constituents and to investigate whether there was any carbon source preference or specificity among the organisms. In order to address the latter aim, the diesel-degrading consortium used in the remediation system at the study site was cultured on the diesel constituents separately in order to identify the communities responsible for the utilization of each compound. The subsequent DGGE profiles and their corresponding PCA scatter plot clearly showed community variation according to the carbon source. This was seen in the scatter plot through the separation along the axes (Fig. 2). Specifically, three distinct groups emerged during PCA analyses of DGGE profiles. The community profiles indicated that despite the uniform diversity present within the starting consortium inocula, consistent enrichment occurred for subpopulations that were dependent upon carbon source type. The DGGE community profile of the site-derived multispecies consortium (data not shown here) used as the inoculum showed a very diverse community with little hierarchy. Overall, three distinct sets could be identified, which all derived from the diesel-degrading consortium obtained from the study site: naphthalene utilizers, mid-chain alkane (C13–C18) utilizers, and long-chain alkane (C19–C21) utilizers. DGGE analysis not shown in this work showed that the three utilizer sets were very different from the diesel-degrading consortium. The most commonly found group utilized mid-chain alkanes, previously reported to be most easily degraded by microorganisms (Atlas, 1981). Shorter chain length alkanes and metabolites resulting from their degradation can be toxic to organisms, while very long-chain alkanes are rather resistant to degradation (Singer & Finnerty, 1984; Vestal et al., 1984; Atlas & Unterman, 2002). The profiling data also revealed that the two groups of alkane degraders showed some intergroup and low intragroup variability through highly similar DGGE profiles and the separation seen in axis 2 of the PCA scatter plot. However, those communities degrading naphthalene exhibited a larger inter- and intragroup (seen through the separation on axis 1 of the PCR scatter plot) variation in diversity over replicate enrichments, suggesting more stochastic events occurring within these microbial communities. These results suggested a potential cooperative effect in terms of community-based diesel degradation.

Figure 2

16S rRNA gene-based DGGE community profiles are shown in triplicate for enrichment cultures grown on C13–C21n-alkanes and naphthalene (Naph) as a sole carbon source with the corresponding PCA scatter plot. Separation into C13–C18 and C19–C21 classes as well as the variation among replicates for naphthalene is shown.

Hydrocarbon utilization capabilities of single isolates

In order to investigate the extent to which site isolates could utilize diesel constituents and whether they exhibited any carbon source preference, each isolate was cultured individually on each hydrocarbon.

16S rRNA gene sequence analysis of the site isolates resulted in the recovery of 12 taxa consisting of five Pseudomonas spp., three Psychrobacter spp., two Achromobacter spp., one Rhodococcus sp., and an Acinetobacter sp. (Table 1). All of the genera fell into the phylum Proteobacteria, with the exception of Rhodoccocus belonging to the Actinobacteria, and have frequently been associated with hydrocarbon degradation (Venkateswaran et al., 1991; Prince, 1993; Cutright & Lee, 1994; Baldi et al., 1999; de Carvalho & da Fonseca, 2005; de Carvalho et al., 2009).

View this table:
Table 1

Results of taxonomic classification of organisms isolated from the study site from sequence analysis of the 16S rRNA gene

IsolateSequence length (bp)Identity (%)GenBank accession no.
Rhodococcus erythropolis149399.93AP008957
Pseudomonas sp. 1151399.67AB365062
Achromobacter xylosoxidans 1150799.14AJ278451
Psychrobacter sp. 1150791.04AB016059
Acinetobacter sp.151499.00EU341175
Achromobacter xylosoxidans 2149898.46Y14908
Pseudomonas anguilliseptica151399.21AF439803
Pseudomonas veronii151396.56AY179328
Psychrobacter sp. 2151090.83NR_025205
Pseudomonas sp. 21152100.00AY576004
Psychrobacter sp. 3149491.03AF441202
Pseudomonas sp. 3151399.74AY091598

OD600 nm measurements showed that all 12 organisms were capable of utilizing some or all of the diesel constituents (Table 2). Although the values were relatively low, they were not unlike those seen in previous studies (Peng et al., 2007; Zeinali et al., 2007; Bouchez-Naitali & Vandecasteele, 2008); taking into account that the organisms in the present study were cultured using lower nutrient concentrations, agitation, and temperature in order to better reflect environmental conditions. Overall, the physiological response was variable, ranging from Pseudomonas sp. 3, which was capable of growth on only two and Pseudomonas sp. 1, which could utilize all 10 hydrocarbons. Relatively high growth was observed for six of the isolates (Table 2), including Rhodococcus erythropolis, Psychrobacter sp. 1, Pseudomonas sp. 1, two Achromobacter xylosoxidans, and an Acinetobacter sp., but only in relation to mid-chain length alkanes (C13–C17). Preferential utilization of lower chain length alkanes within a community has been described previously (Richard & Vogel, 1999). This observation also ties in with the GC-MS data, which showed that C13–C17 alkanes were not as abundant in the aged diesel fuel found at the site as they are in nonaged diesel fuel (Fig. 1). Of the above, two isolates (Acinetobacter sp. and A. xylosoxidans 2) displayed appreciable growth on C19–C21 alkanes, and hence probably represented more generalist degraders. For long-chain degradation one isolate consistently displayed a higher affinity for long-chain length over mid-chain length (Pseudomonas anguilliseptica), again indicating probable compartmentalization of physiologies within the community. Of the remaining five isolates only low growth on all substrates was observed across a range of chain lengths, suggesting that these strains were generalist degraders with a relatively low degradation capability and low specialization.

View this table:
Table 2

Ability of site isolates to utilize each of the diesel fuel components (C13–C21n-alkanes, naphthalene, and diesel) by absorbance (OD600 nm)-based growth assays

Rhodococcus erythropolis++++++++++++++++++++++
Pseudomonas sp. 1+++++++++++++++++++++++
Achromobacter xylosoxidans 1+++++++++++++++++++
Psychrobacter sp. 1++++++++++++++++++
Acinetobacter sp.+++++++++++++++++++++
Achromobacter xylosoxidans 2+++++++++++++++++
Pseudomonas anguilliseptica+++++++++
Pseudomonas veronii++++++
Psychrobacter sp. 2+++++++++
Pseudomonas sp. 2++++++++++
Psychrobacter sp. 3+++++++++++
Pseudomonas sp. 3+++
  • Hatched boxes denote strains capable of mid- and long-chain alkane utilization, as well as highlighting the lack of growth on C18 and naphthalene by any strain.

  • −, OD600 nm 0.00–0.019; +, OD600 nm 0.02–0.099; ++, OD600 nm 0.1–0.2; +++, OD600 nm >0.2.

Interestingly, no degrader displayed a large growth capability on C18 or naphthalene as a sole carbon source. Despite a single carbon chain length difference between C17 and C19, C18 degradation seemed to be problematic, even for organisms that grew well on either mid- or long-chain alkanes. The same was true for naphthalene. Lack of naphthalene degradation could be explained by its higher toxicity, due to its relatively high solubility of 30 mg L−1 (Atlas, 1981; Bouchez et al., 1995), as well as previous reports of naphthalene degraders being recalcitrant to culture (Huang et al., 2009). However, the compound's degradation (Cerniglia, 1984; Gibson & Subramanian, 1984; Yu & Chu, 2005) and the isolation of organisms that utilize it is well documented (Cerniglia & Shuttleworth, 2002). The lack of naphthalene-degrading isolates may also be an artefact of the isolation method, which did not select for them specifically at such high concentration. In the case of C18 degradation, previous studies have reported both efficient and slow degradation rates by individual organisms and microbial consortia (Abed et al., 2002; Grotzschel et al., 2002; Radwan et al., 2002). In the present study, the results suggest that C18n-alkanes and naphthalene are more than likely remediated at low levels by a range of organisms overlapping in their abilities in situ. This hypothesis is supported by the GC-MS analysis of the site diesel fuel, which showed C18n-alkanes to be the overall most abundant constituents and naphthalene the most abundant aromatic compound (Fig. 1).

At this stage, it is important to consider the bioavailability of the 10 compounds for microbial utilization. The compounds were added to media at a relatively high concentration of 1000 p.p.m. (or 1 g L−1) in order to mimic the concentration of diesel fuel at the study site. In reality, however, only a fraction of the hydrocarbon added would have been available to the organisms. The water solubility of mid- to long-chain length alkanes is notoriously difficult to measure as well as predict. A number of studies have estimated the solubility of C13–C21 alkanes to range between a mole fraction value of 4 × 10−10 and 7 × 10−11 at 25 °C (Sutton & Calder, 1974; Ferguson et al., 2009). This is translated as ranging between approximately 3 μg L−1 for tridecane (C13) and 0.9 μg L−1 for hexadecane (C16) and is equivalent to 0.03–0.009 p.p.m. The very low water solubility of these compounds would have made their utilization by the 12 field isolates difficult. However, although not at high levels, growth was observed through changes in the OD600 nm measurements. Some microbial organisms, such as some Pseudomonas, Acinetobacter, and Rhodococcus species, produce biosurfactants, which effectively make the hydrocarbons more available for microbial utilization (Beal & Betts, 2000; Chang et al., 2009; Henry & Abazinge, 2009). Pseudomonas and Rhodococcus species, in particular, are well known for their production of biosurfactants. In the current study, both achieved relatively high growth on all of the alkane substrates, and principally the mid-chain length alkanes.

In summary, results suggest that members of the same community showed preference for specific carbon sources shown through their ability to utilize various diesel constituents, potentially leading to a cooperative hypothesis within the community. Some are likely to be competitive in a broader range of scenarios, while others may be more suited to specific conditions and habitats. The site isolates could be categorized into two classes of microorganisms, which have previously been identified in terms of their survival strategy: the K-strategists and the r-strategists (Winogradsky, 1924; Kuznetsov et al., 1979; Andrews & Harris, 1985). The r-strategists exist mostly in a resting phase demonstrating brief periods of activity stimulated by the appearance of an available substrate. Examples in the present study could be R. erythropolis, Pseudomonas sp. 1, and A. xylosoxidans 1. In contrast, the K-strategists are continually and slowly active: for example Pseudomonas sp. 2 and 3, and Psychrobacter sp. 3. It was observed that, in general, organisms that were particularly good at degrading diesel were likely to fall into the r-strategists. Previous studies of communities utilizing a mixed hydrocarbon source have observed either antagonism and competition between the organisms or cometabolism (Bouchez et al., 1999; Mariano et al., 2008). The investigation demonstrated that high community diversity may allow for the coexistence of both K- and r-strategists and the compartmentalization of functions among key organisms resulting in the utilization of the whole spectrum of diesel fuel components.


This work was supported by the Natural Environment Research Council and Napier University, Edinburgh. We would like to thank CORUS UK for the GC-MS analysis of the site diesel fuel and ERS Ltd (http://www.ersremediation.com/index.php) for access to the study site.


  • Editor: Andreas Stolz


View Abstract