OUP user menu

Importance of chemical structure on the development of hydrocarbon catabolism in soil

Jacqueline L. Stroud, Graeme I. Paton, Kirk T. Semple
DOI: http://dx.doi.org/10.1111/j.1574-6968.2007.00750.x 120-126 First published online: 1 July 2007

Abstract

A soil was amended with 14C-analogues of naphthalene, phenanthrene, pyrene, B[a]P or hexadecane at 50 mg kg−1 and the development of catabolic activity was assessed by determining the rate and extent of 14CO2 evolution at time points over 180 days. The catabolic potential of the soil was hexadecane>naphthalene>phenanthrene>pyrene>B[a]P, determined by the decrease in lag time (as defined by the time taken for 5%14CO2 to be evolved from the minerialization of the 14C-labeled hydrocarbons). The results clearly showed the difference between constitutive and inducible biodegradation systems. The 0 day time point showed that hexadecane minerialization was rapid and immediate, with a 45.4 ± 0.6% mineralization extent, compared with pyrene minerialization at 1.0 ± 0.1%. However, catabolism for pyrene developed over time and after a 95 days soil-pyrene contact time, mineralization extent was found to be 63.1 ± 7.8%. Strong regression was found (r2>0.99) between the maximum rates of mineralization and the partioning coefficient between the mineralized hydrocarbons, which may indicate linearity in the system.

Keywords
  • hexadecane
  • PAHs
  • indigenous catabolic activity
  • ageing

Introduction

Hydrocarbons are widespread in the environment due to natural and anthropogenic processes; however, high concentrations of hydrocarbons in soils are often associated with anthropogenic activities. It is estimated that oil contamination accounts for over 25% of all chemical contamination incidents in the UK (Environment Agency, 2003). Oils are mixtures of both aliphatic and aromatic hydrocarbons, and the level of hydrocarbon contaminants in soils is a cause for concern, as these contaminants may persist in soils and can exert toxic and carcinogenic affects on biological receptors (Cerniglia, 1992).

Biodegradation is central to hydrocarbon removal from soils and the physico-chemical characteristics of the hydrocarbon directly impacts biodegradation. For example, microorganisms readily attack the n-alkane open chain structure via constitutive β-oxidation. In comparison, PAHs are composed of fused benzene rings, which are relatively resistant to microbial attack, often requiring the induction of suitable enzymes for catabolism, and are degraded by either ortho or meta cleavage of the ring structure (van der Meer et al., 1992). Larger PAHs, such as benzo[a]pyrene (B[a]P), are often unsuitable sole carbon sources and a cometabolism mechanism may be the dominant process for biodegradation (Kanaly et al., 1997).

Biodegradation of contaminants requires adaptation responses within the microbial community, and may involve either the enrichment in the numbers of degrading microorganisms or the induction of catabolic enzymes leading to the degradation of the target contaminant(s) (Leahy & Colwell, 1990). A range of microorganisms are able to degrade hydrocarbons, and the ubiquitous nature of hydrocarbon contamination means that competent hydrocarbon degraders are common (Macleod & Semple, 2002, 2006; Lee et al., 2003). Frequency of contamination and the length of exposure to contaminants have been shown to enhance catabolism for hydrocarbons in soils (Macleod & Semple, 2002, 2006). However, catabolism is complicated by the decreasing bioaaccessability of contaminants over time due to associations with soil components (Hatzinger & Alexander, 1995). Macleod & Semple (2002) showed that catabolism for pyrene was limited by pyrene sorption to soil organic matter in a woodland soil. The extent to which a hydrocarbon associates with organic matter is described by the Koc, but this is described by the Kow to a large extent. The hydrocarbons used in this study had a range of Kow values from 3.36 (naphthalene) to 9.1 (hexadecane).

The development of catabolism by soil microbial communities can be measured using 14C-labelled substrate respirometry (Reid et al., 2001; Macleod & Semple, 2002, 2006; Lee et al., 2003). The production of 14CO2 is a result of microbial degradation, making it an ideal technique to detect microbial adaptation (Spain et al., 1980). The aims of this study were (1) to quantify the catabolic activity of soil microbial communities to hydrocarbons with different physico-chemical properties and (2) to compare the differences between constitutive and inducible biodegradation systems of ageing hydrocarbons. To address theses aims, naphthalene, phenanthrene, pyrene, B[a]P and hexadecane were selected for investigation in this study.

Materials and methods

Materials

All the nonlabelled and labelled [1-14C] naphthalene (specific activity=2–10 mCi mmol−1, radiochemical purity > 95%); [9-14C] phenanthrene (specific activity=40–60 mCi mmol−1, radiochemical purity >95%); [14C-4,5,9,10] pyrene (specific activity=10–30 mCi mmol−1, radiochemical purity >95%); [7-14C] B[a]P (specific activity = 1–30 mCi mmol−1, radiochemical purity >95%) and [1-14C] n-hexadecane (specific activity=1–15 mCi mmol−1, radiochemical purity >95%) were obtained from Sigma Aldrich (UK). The liquid scintillation cocktail, Goldstar and 7 mL and 20 mL glass scintillation vials were obtained from Meridian (UK). Bibby (UK) supplied the Erlenmeyer flasks with Teflon™-lined screw caps. RS (UK) supplied metal fittings used to make the respirometers. Sodium hydroxide was obtained from VWR International (UK). Packard Canberra (UK) supplied the sample oxidizer scintillants, Carbosorb and Permafluor as well as combustion cones and caps.

Soil preparation and microcosm set-up

A river alluvium of the Enborne series was collected from the top 7 to 25 cm of a pasture field at Myescough, Lancashire. This soil was characterized as a clay loam, pH 6.5, organic matter content 2.7 ± 0.04%. The field-sampled moisture content of 28 ± 2% (w/w) was maintained for the duration of the study. Before spiking, the soil was passed through a 2 mm sieve to remove stones, roots and facilitate mixing. The soil was then rehydrated to field conditions using de-ionized water and stored in the dark at 4°C for 4 days. Soil samples (500 g) were amended with either naphthalene, phenanthrene, pyrene, B[a]P or hexadecane in a carrier solvent of acetone to give a concentration of 50 mg kg−1 and c. 100 Bq g−1 using a glass bowl, stainless steel spoon and hand-mixing method. This method has been shown to give the most homogenous distribution of hydrocarbons throughout the soil, with minimal disruption to the microbial community in comparison with other methods (Doick et al., 2003). The acetone was allowed to volatilize over 24 h, and the contaminated soils were placed in duplicate amber glass microcosms (500 g capacity). One set of soils was sterilized using γ-irradiation (35.1 Kgy; Isotron, Bradford, UK) shortly (<24 h) after spiking. The sterility of each microcosm was ensured by aseptic handling techniques and checked at each timepoint using standard microbiological techniques. All of the microcosms were set up in triplicate and stored in the dark at room temperature during the ageing period.

Determination of total 14C-hydrocarbon activity in soil

The total 14C-hydrocarbon activity in the soil was assessed at each timepoint (0, 30, 60, 90 and 180 days) by combusting 1 g of soil (n=3) from each microcosm using a sample oxidizer (Model 307, Packard). The initial recovery (0 day) of the 14C-hydrocarbon-associated activity in the soil was normalized to 100%. At each subsequent timepoint, the 14C-hydrocarbon recovery is compared with the initial amendment (%), to monitor 14C-losses from each of the microcosms.

Minerialization of aged 14C-hydrocarbon-associated activity in soil

At 0, 30, 60, 90 and 180 days, subsamples (n=3) from each of the microcosms were taken and 14C-respirometry was used following the procedure developed by Reid et al., (2001). Briefly, 10 ± 0.2 g soil and 30 mL sterile mineral salts basal medium (MBS) was added to a 250 mL Erlenmeyer flask to form a slurry (ratio 1 : 3 w/v). This was sealed with a modified screw lid incorporating a stainless steel crocodile clip, which enabled a CO2 trap of 1 mL NaOH (1 M) in a 7 mL glass scintillation vial, to be securely attached inside the vessel. The sealed flasks were placed in an incubator shaker at 21 ± 2°C and shaken at 100 r.p.m. Each CO2 trap was replaced every 24 h and 5 mL of liquid scintillation fluid added to the sampled vial. Following overnight storage, the level of 14C-activity was determined by liquid scintillation counting and blank corrected (Canberra Packard Tri-Carb 2250Ca). This method enabled the comparison between the aged 14C-hydrocarbon-associated activity in soil to the measured evolution of 14CO2 from the 14C-labelled hydrocarbons at each timepoint, thus detecting the development of catabolism.

Statistical analysis

To investigate the development of catabolism in the soil, the lag phases (14CO2 evolution >5%), maximum rates of minerialization (calculated from the kinetic response observed) and the total cumulative extents of minerialization were compared at each time point and statistically tested using anova after normality testing (sigma stat, Version 2.03, SPSS Inc., Tukey test, P≤0.05) following blank correction, catabolic profiles have been presented using Sigma Plot (Version 6.10, SPSS).

Results

Catabolism of 14C-naphthalene in soil

Loss of 14C-naphthalene was measured in the soil at each of the sampling points. There was no removal of the PAH from the nonsterile microcosm after 30 days incubation with 99.6 ± 0.4%; however, there was a decrease of c. 10% of the initial amendment after 60 days contact time. After 180 days incubation, c. 70% of the initial 14C-activity remained in the soil (Fig. 1a). In comparison, the sterile microcosm showed smaller losses from 60 days ageing, c. 4% of the initial amendment. However, after 180 days ageing, a decrease of c. 31% of the initial amendment was found (Fig. 1b).

Figure 1

14C-Recovery compared with the 14C-initial amendment of (•) 14C-napthalene, (□)14C-phenanthrene, (▲)14C-pyrene, (▿)14C-B[a]P, (◆)14C-hexadecane (50 mg kg−1) from soil over 180 days incubation in (a) nonsterile and (b) sterile microcosms. SE of the mean where visible (n=3).

Despite no appreciable removal of 14C-naphthalene from the soil incubations at time 0, 14C-naphthalene was significantly (P≤0.05) degraded by the indigenous microbial communities after a short lag time of 2 days compared with the other hydrocarbons, with a plateau at 58.3 ± 1.4% at the end of the 14 days incubation, and had the highest rate of mineralization of 1.4%14CO2 h−1 (Table 1; Fig. 2a). The cumulative mineralization extent at the end of the 14 days assay remained significantly high at both the 30 and 60 days soil-contact time, reaching a maximum extent of mineralization of 60.7 ± 3.2% at 30 days. Similarly, the maximum rate of mineralization at 30 days was 1.3%14CO2 h−1 and the lag time was 2 days (Table 1, Fig. 2a). However, after ageing for 95 days, the extent of mineralization significantly (P≤0.05) decreased to just 13.8 ± 1.8%. Similarly, the maximum rate of mineralization significantly (P≤0.05) decreased by 86–0.2%14CO2 h−1 when compared with the other time points (Table 1, Fig. 2a).

View this table:
Table 1

Catabolic profiles of 14C-labelled hydrocarbons in soil after 14 days respirometric assay

NaphthalenePhenanthrenePyreneHexadecane
Ageing timeLag (days)Max rate (%14CO2 h−1)Extent (%)Lag (days)Max rate (%14CO2 h−1)Extent (%)Lag (days)Max rate (%14CO2 h−1)Extent (%)Extent (%)B[a]PLag (days)Max rate (%14CO2 h−1)Extent (%)
021.458.3 ± 1.441.260.1 ± 1.6>14<0.11.0 ± 0.1<110.445.4 ± 0.6
3021.360.7 ± 3.240.957.1 ± 2.6120.3334.5 ± 0.1<130.446.0 ± 0.3
6011.152.6 ± 2.130.660.0 ± 3.5110.4245.4 ± 1.5<130.117.4 ± 0.5
9520.213.8 ± 1.811.062.6 ± 1.850.5863.1 ± 7.8<130.117.7 ± 1.3
18020.316.0 ± 7.110.658.5 ± 12.411.0958.1 ± 4.4<120.123.9 ± 5.0
Figure 2

Development of 14CO2 from the catabolism of 14C-hydrocarbons (50 mg kg−1) in soil for naphthalene (a), phenanthrene (b), pyrene (c) and hexadecane (d) at (•) 0 days, (□) 30 days, (▲) 60 days, (▿) 95 days, (♦) 180 days. SE of the mean (n=3).

Catabolism of 14C-phenanthrene in soil

Loss of 14C-phenanthrene was measured in the soil at each of the sampling points. There was no removal of the PAH after 60 days incubation in the nonsterile microcosm; however, there was a decrease of c. 11% after 90 days. At 180 days incubation, 49% of the initial 14C-activity added to the soil was recovered (Fig. 1a). In comparison, the sterile microcosm showed no loss of 14C-activity in the soil, with 99.4 ± 0.5% after 180 days (Fig. 1b).

Despite no appreciable removal of 14C-phenanthrene from the soil for the first 60 days, degradation of 14C-phenanthrene was measured in the soil at 0, 30 and 60 days. For 14C-phenanthrene, the initial mineralization profile was similar to naphthalene (P≥0.05), with decreasing lag time and a consistently extensive mineralization over time (Table 1, Fig. 2b). The extent of phenanthrene mineralization reached 60.1 ± 1.6% and a lag time of 4 days in soil sampled at 0 day incubation. The maximum rate of mineralization for this contaminant was found at this initial time point at a significant (P≤0.05) 1.2%14CO2 h−1 (Table 1, Fig. 2b). The lag time decreased by 75% during the 180 days incubation period to just 1 day. There was no significant (P≥0.05) difference in the extent of phenanthrene mineralization over the 180 days incubation, with a mean of 59.7% and reaching a maximum of 62.6 ± 1.8% at 95 days (Table 1, Fig. 2b).

Catabolism of 14C-pyrene in soil

Loss of 14C-pyrene was measured in the soil at each of the sampling points. There was virtually no loss of pyrene in the nonsterile soil after 95 days with 99.4 ± 0.3% recovered. Over the 180 days incubation period, 84.2 ± 0.7% of the initial 14C-activity remained in the soil (Fig. 1a). No loss of 14C-pyrene was detected over 180 days in the sterile microcosm, with an activity of 100.4 ± 0.3% recovered at 180 days (Fig. 1b).

The extent of mineralization of 14C-pyrene was low at 0 day, at just 1.0 ± 0.1% and a lag time greater than the 14 days incubation (Table 1, Fig. 2c). However, after the 30 days incubation period, a lag time of 12 days and a statistically significant (P≤0.05) pyrene mineralization of 34.5 ± 0.1% was measured. The maximum rate of mineralization was estimated at 0.3%14CO2 h−1 (Table 1, Fig. 2c). The mineralization profile at each successive time point showed increasing extents of pyrene mineralization and reductions in lag time, with the maximal mineralization extent reaching at 63.1 ± 7.8% after 95 days of incubation (Table 1, Fig. 2c). The lag time reduced by 12-fold over the 180 days incubation period to just 1 day; furthermore, the maximum rate of mineralization increased between the 30 and 180 days timepoints to 1.1%14CO2 h−1 (Table 1).

Catabolism of 14C-benzo[a]pyrene in soil

Despite measuring the loss of 14C-B[a]P in the soil over the 180 days incubation, there was no appreciable degradation in the microcosms over the 180 days ageing period (Fig. 2a and b). The catabolism of B[a]P was significantly different from all the other hydrocarbons in this study (Table 1). B[a]P was not mineralized over the duration of the study, with a cumulative mineralization extent of<1% at all time points.

Catabolism of 14C-hexadecane in soil

Hexadecane was rapidly degraded in the nonsterile microcosms (Fig. 1a), 18.2% loss was detected after the 30 days incubation period. This degradation continued and by 180 days, the recovery of 14C-hydrocarbon-associated activity was found to be 42.1 ± 3.4%. The sterile microcosms showed no significant losses, with the level of 14C-hydrocarbon-associated activity found to be 98.6 ± 1.5% at 180 days (Fig. 1b).

Hexadecane had a very different mineralization profile compared with the PAHs (Table 1, Fig. 2d). Initially this hydrocarbon was rapidly mineralized, with a lag time of 1 day and was the shortest lag time of all of the hydrocarbons. The mineralization extent of 14C-hexadecane at time 0 was 45.4 ± 0.6%. However, unlike PAH mineralization, the lag time increased to 3 days at the 30 days ageing period, while the mineralization extent remained at 46.0 ± 0.3% and the maximum rate of mineralization was consistent with the previous time-point, reaching the maximum of 0.4%14CO2 h−1 (Table 1, Fig. 2d). From the 60 days ageing period, the extent of mineralization significantly (P≤0.05) decreased and remained constant, with an average of 18.6% (Table 1, Fig. 2d).

Comparison of the catabolic behaviour of the hydrocarbons in soil

There is correlation between the maximum rate of mineralization and log Kow (Fig. 3). With increasing log Kow, lower maximum rates of mineralization were detected. B[a]P was not considered due to the lack of measurable mineralization. Naphthalene has lowest log Kow in the study of 3.36, but the fastest rate of mineralization at 1.4%14CO2 h−1. In comparison, phenanthrene has a log Kow 4.16, and a lower maximum rate of mineralization of 1.2%14CO2 h−1. The maximum rate of pyrene mineralization was determined to be 1.1%14CO2 h−1 and has a log Kow of 5.19. Hexadecane, the hydrocarbon with the highest log Kow in this study, had the lowest maximum rate, which was up to four times smaller than the PAHs at 0.4%14CO2 h−1. There was a strong negative trend with a quality of fit, r2 value of 0.99 and correlation coefficient of 0.97 (P≤0.05).

Figure 3

Correlation between the maximum rate of mineralization of the 14C-labelled hydrocarbons and the partioning coefficient (log Kow) of the 14C-labelled hydrocarbons.

Discussion

It is widely accepted that the rate and extent of the degradation of organic contaminants differs between classes of chemicals and even between individual chemicals with similar structures. This catabolic behaviour is controlled by the physico-chemical properties of the contaminants and the soil, with organic matter playing an important role (Macleod & Semple, 2002). Differences in degradation of both aromatic and aliphatic hydrocarbons were observed in this study. For example, hexadecane was removed most rapidly, followed by the PAHs, with the number of aromatic rings influencing the persistence and the development of catabolism in the soil, with naphthalene > phenanthrene > pyrene>B[a]P (Fig. 1a).

Lag time has been reported as a measure of the catabolic potential of microorganisms for hydrocarbons (Macleod & Semple, 2002, 2006; Reid et al., 2002; Lee et al., 2003; Kanaly & Watanabe, 2004). In this study, the evolution of 14CO2 through the action of the soil microbial communities on the 14C-hexadecane occurred immediately as described by the length of the lag time in the respirometers at the beginning of the study. Hexadecane is an n-alkane, similar in structure to low molecular weight soil organic matter and usually degraded through β-oxidation of the terminal methyl group (Koma et al., 2001). Furthermore, the enzymes required for this biotransformation are constitutive, thus catabolism is inherent in the soil microbial community. Constitutive biodegradation is often characterized by rapid and immediate degradation aliphatic hydrocarbon degraders known to be common in soil (Bouchez-Naitali et al., 1999), and perhaps even more numerous than aromatic hydrocarbon degraders (Gogoi et al., 2003). This would explain the consistency of n-hexadecane biodegradation over the timescale of this study (Table 1, Figs 1 and 2d).

The soil microbial communities were able to rapidly degrade the lower molecular weight 14C-PAHs, naphthalene and phenanthrene, as shown by the relatively short lag time and relatively high extents of mineralization found at the beginning of the study (Table 1), indicating that catabolic activity was already present in the soil. This is not surprising as PAHs are ubiquitous contaminants and are known to be present in all soils at very low concentrations. Macleod & Semple (2003) reported that the background PAH burdens of two rural soils ranged between 200 and 400 µg kg−1. Furthermore, it has been widely reported that both naphthalene and phenanthrene may be readily degraded by indigenous soil microbial communities (Reid et al., 2001; Lee et al., 2003). Given the sigmoidal nature of the catabolic data (Fig. 2a and c), it is likely that the microorganisms capable of degradation were increasing in biomass to biodegrade this carbon source, causing the lag time to decrease over time (Grosser et al., 1991). Mineralization of 14C-pyrene was not measured until well into the study as the indigenous soil microbial communities required more time to develop the capacity to degrade the four-ring PAH; this is in agreement with other studies (Macleod & Semple, 2002, 2006).

As a point of note, there were no significant losses of 14C-naphthalene in either the sterile and nonsterile microcosms until 95 days (Figs 1a and b). However, the 14C-respirometric data showed that the soil microbial communities were able to mineralize the 14C-naphthalene before this, after a 2 days lag period (Table 1, Fig. 2a), suggesting that the microorganisms were readily able to degrade this simple PAH, as discussed previously. Interestingly, at the same time as the abiotic loss of naphthalene (after 95 days), there was also a significant decrease in the extent to which 14C-naphthalene was mineralized (Table 1, Fig. 2a). A similar observation was noted by Reid et al., (2002), who were investigating the development in phenanthrene catabolic activity in Phase II mushroom compost, although the authors were unable to provide a definitive reason for this behaviour. Although there is no definitive evidence for this observation in this study, there may be several reasons as to why the abiotic loss of 14C-naphthalene had a significant impact on the ability of the indigenous microbial communities to mineralize the 14C-PAH. This could have been due to the amount of the PAH remaining in the soil coupled with putative decreases in the accessibility of the naphthalene to the soil microbial communities. The sequestration of hydrocarbons into soil over time has been shown to significantly affect bioaccessibility, causing limited biodegradation (Hatzinger & Alexander, 1995). Finally, the degrading microbial population(s) may have decreased their activity, perhaps through the production of a toxic metabolite during degradation. It is possible that a combination of these possibilities contributed to the decrease in mineralization of the 14C-naphthalene after 95 days incubation; however, a decrease in the bioaccessible concentration seems a more likely explanation.

Finally and unsurprisingly, the indigenous soil microbial communities were not able to catabolize B[a]P in this study. This was expected as B[a]P is known to be highly resistant to microbial degradation and it has been reported that cometabolism may be involved in biodegradation of this high molecular weight PAH (Kanaly et al., 1997).

An interesting finding of this study was the correlation between the maximum rates of mineralization and the octanol-water partition coefficient (log Kow) of the hydrocarbons, which was found to be highly significant (P≤ 0.01). This correlation shows that the higher the log Kow value, the slower the rates of mineralization. Given the trend in physico-chemical properties, partitioning of the PAHs within the soil may be the rate-limiting step in the biodegradation of hydrocarbons. However, caution must be applied with regard to the experimental design of the study. For hexadecane, it is known that desorption to the aqueous phase may not be important for biodegradation to proceed in soil (Huesemann et al., 2003). In this study, respirometry was conducted under slurry conditions, maximizing the interactions between the soil, the contaminant, the indigenous microbial communities and the aqueous phase. Therefore, the conditions in the respirometric set-up may have impacted the measured rates and extents of 14C-hexadecane mineralization.

The value of this study is that it investigated the catabolism of aliphatic and aromatic hydrocarbons with significantly different physico-chemical properties. This has, to the authors knowledge, not been investigated in this way before. The study showed that a single soil may possess the catabolic potential for a wide range of hydrocarbon contaminants and that biodegradation is an important loss mechanism for organic contaminants in soil.

Acknowledgements

The authors would like to thank the Natural Environment Research Council, UK and Remedios, for financially supporting this work.

Footnotes

  • Editor: Hans-Peter Kohler

References

View Abstract