INTRODUCTION

Top Abstract Introduction Materials and Methods Results and Discussion References

The anaerobic biodegradation of n-alkanes has recently been demonstrated under nitrate-reducing (6, 11, 19), sulfate-reducing (1, 2, 7, 9, 20-22), and methanogenic (3, 25) conditions. Pure bacterial cultures that couple the reduction of the first two electron acceptors to the oxidation of alkanes have been isolated (1, 2, 11, 20-22), but little is known about the initial mechanism(s) of anaerobic activation and subsequent metabolism of these hydrocarbons. Previous studies of the total cellular fatty acid composition of pure sulfate-reducing cultures grown on alkanes have suggested that activation does not occur via dehydrogenation to the corresponding 1-alkene (2) but does occur by addition of an unknown organic carbon fragment to the C-2 position of the original molecule (22).

Recent research has shown that the anaerobic bacterial metabolism of toluene and xylenes is initiated by addition of the methyl group to the double bond of fumarate, forming optically pure (R)-(+)-benzylsuccinic acid and methylbenzylsuccinic acids, respectively (4, 5, 12-14). These addition reactions occur with retention of the proton abstracted from the methyl group in the succinyl moiety of the metabolite (4, 5). The detection of 3-phenyl-1,2-butanedicarboxylic acid from ethylbenzene (L. M. Gieg and J. M. Suflita, Abstr. 99th Gen. Meet. Am. Soc. Microbiol., abstr. Q-109, p. 554, 1999) may result from an analogous fumarate addition reaction of the methylene group. Alternatively, and as was suggested for toluene, this metabolite may be formed by successive additions of two carbon units from acetyl coenzyme A (acetyl-CoA), a mechanism that would result in the loss of two H atoms (8). Partially purified benzylsuccinate synthase from the denitrifying Azoarcus sp. strain T was shown to catalyze addition to fumarate by the methyl groups of toluene and all three isomers of xylene, as well as of 1-methyl-1-cyclohexene but not of 4-methyl-1-cyclohexene or methylcyclohexane (5). The nonreactivity of the latter two substrates led the researchers to conclude that the resonance stabilization offered to the radical transition state in the form of a conjugated  system by the adjacent aromatic ring or conjugated double bond was necessary for this bioconversion (5). The dodecylsuccinic acid metabolites observed in the present study are formed with retention of the deuterium from D₂₆-dodecane on the succinyl moiety, suggesting that the bond between the dodecane and succinate was formed by addition across the double bond of fumarate. This is a novel bioconversion for alkanes; the nature of the enzymes that catalyze this conversion, as well as how it occurs without resonance stabilization offered by a conjugated  system, remains to be clarified.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results and Discussion References

Growth of enrichment cultures. Enrichment cultures were grown in a sulfate-containing brackish mineral medium (24) that was amended with resazurin (1 mg/liter) as a redox indicator and cysteine hydrochloride and Na₂S · 9H₂O (50 mg/liter each) as reductants. All culture bottles were sealed with Teflon-lined serum stoppers and incubated at 32°C in an inverted position. Uninoculated sterile controls remained reduced throughout the incubation period. Initially, 50 ml of this medium was inoculated with 10 ml of an oily sludge collected from a naval wastewater storage facility as previously described (15). This was incubated under an N₂:CO₂ (4:1) atmosphere with 1 ml of an alkane mixture containing decane, dodecane, hexadecane, and octadecane (1:1:1:1.5, vol/vol). The alkane mixture, and later pure liquid alkane feedstocks, was flushed with N₂, sealed under Teflon-lined stoppers, and autoclaved before use. In the enrichment cultures, sulfide production was monitored colorimetrically (23) and sulfate reduction was measured by ion chromatography (7). Due to the abundance of labile substrates in the inoculum, it was not until the third 25% transfer that an enhancement of sulfide production and sulfate reduction were measured in triplicate alkane-amended cultures relative to the substrate-unamended controls (data not shown). After the fifth transfer, individual alkanes were provided to the culture. The dodecane-amended incubation grew best and therefore was selected for further study. The culture has since been repeatedly transferred for over 2 years (bimonthly; 25%, vol/vol) in 35 ml of medium given 50 µl of dodecane (>99%; Aldrich, Milwaukee, Wis.) as the sole carbon source. This is well above the aqueous solubility of dodecane (3.4 µg/liter) (10), so throughout this report the amount of dodecane given to the cultures is presented as an absolute value rather than as an aqueous concentration.

[1-¹⁴C]dodecane mineralization. Cultures were incubated with 25 µl of unlabeled dodecane and 0.82 µCi of [1-¹⁴C]dodecane (98%; Sigma, St. Louis, Mo.). After 9 weeks of incubation, the cultures were acidified with 2 ml of H₂SO₄ (9 M) and ¹⁴CO₂ was recovered and quantified in a trapping train designed for this purpose (17). After flushing of the cultures, 10 ml of pentane was used to extract the remaining dodecane and [1-¹⁴C]dodecane, and the residual radioactivity was determined by analyzing a 0.5-ml portion of the extract in a liquid scintillation counter.

Dodecane loss measurements. The loss of dodecane (25 µl) in cultures relative to sterile controls was quantified by acidifying the medium with 0.2 ml of HCl (6 M), giving 10 µl of n-tetradecane as the internal standard and extracting three times with dichloromethane (15 ml). The extracts were pooled, dried over Na₂SO₄, concentrated on a rotary evaporator to 2-ml volumes, and analyzed with a Hewlett-Packard 6890 GC by using an HP-5 column (length, 30 m; inside diameter, 0.32 mm; film thickness, 0.25 µm) and a flame ionization detector. The injector and detector temperatures were 200 and 250°C, respectively, and the oven was held at 90°C for 2 min before its temperature was increased by 4°C/min to 200°C. Helium was used as the carrier gas at a flow rate of 2 ml/min.

Metabolites of H₂₆- and D₂₆-dodecane. Cultures were inoculated with a H₂₆-dodecane-grown culture and incubated with 25 µl of either H₂₆- or D₂₆-dodecane (>98%; Cambridge Isotope Laboratories, Andover, Mass.) or a mixture of 12 µl of each, for periods ranging from 4 to 8 weeks. Cultures were then acidified with 2 ml of HCl (6 M) and extracted twice with 20-ml volumes of ethyl acetate. During earlier extractions, cultures were given 0.75 ml of NaOH (6 M) before acidification and left for 30 min to hydrolyze prospective CoA thioesters. However, this base treatment was not necessary to recover the metabolites. The pooled extracts were dried over Na₂SO₄ before concentration on a rotary evaporator and subsequently under a flow of N₂, to 0.1 ml. The extracts were then given 0.1 ml of N,O-bis(trimethylsilyl)trifluoroacetimide (BSTFA; Pierce, Rockford, Ill.) and heated at 65°C for 10 min before analysis by gas chromatography-mass spectrometry (GC-MS). To help clarify the nature of the metabolites, extracts were also derivatized with 1 ml of an ethereal solution of diazomethane. GC-MS analyses used a DB-5 column (length, 30 m; inside diameter, 0.25 mm; film thickness, 0.1 µm), in a Hewlett-Packard 5890 GC with a 5970 mass-selective detector. The injector and detector temperatures were both 250°C, and the temperature program for the oven was the same as that given above, except that the final temperature was 240°C. Helium was used as the carrier gas at a flow rate of 0.8 ml/min. Chemical ionization GC-MS with ammonia as the reagent gas was conducted at the University of Alberta (Mass Spectrometry Laboratory, Department of Chemistry). Tandem GC-MS (GC-MS-MS) analyses were conducted at the University of Oklahoma (Analytical Services Laboratory, Department of Chemistry). An authentic standard of n-dodecylsuccinic acid was prepared by base hydrolysis of n-dodecylsuccinic anhydride (TCI America, Portland, Oreg.). The anhydride (50 mg) was suspended in 100 ml of NaOH (3 M) and 10 ml of methanol and heated at 65°C. After 2 h, the remaining crystals of undissolved anhydride were removed by filtration, and the filtrate was acidified to a pH of <2 with HCl and extracted twice with 20-ml portions of ethyl acetate. The pooled extracts were dried over sodium sulfate, concentrated, and derivatized.

	RESULTS AND DISCUSSION

Top Abstract Introduction Materials and Methods Results and Discussion References

[1-¹⁴C]dodecane mineralization. In enrichment cultures, 54% of the added radioactivity was recovered as ¹⁴CO₂ and 25% as unoxidized dodecane in a pentane extract (79% total recovery) (Table 1). When sulfate was excluded from the medium, except for a small amount transferred with the inoculum, only 6% mineralization was observed and 78% of the added [1-¹⁴C]dodecane was recovered unaltered in the pentane extract (84% total recovery) (Table 1). According to equation 1, this 6% mineralization would theoretically require 1.75 mM sulfate. The sulfate transferred with the inoculum resulted in an initial concentration of 2.5 mM sulfate in the sulfate-free medium, and during the 9-week incubation this was reduced to 0.5 mM. Thus, the low amount of sulfate transferred with the inoculum accounts for the 6% mineralization observed in the sulfate-free medium. With incubations that contained sulfate and equimolar amounts of sodium molybdate (30 mM), an inhibitor of sulfate reduction, there was no ¹⁴CO₂ produced (Table 1). Thus, dodecane mineralization was dependent on sulfate reduction. In uninoculated sterile controls, the total recovery of [1-¹⁴C]dodecane was 88% and there was no ¹⁴CO₂ produced (Table 1). The remaining 12 to 21% radioactivity that was not accounted for in the sterile control and cultures, respectively, was likely lost due to sorption to the incubation flask closures. In nonsterile incubations, the unrecovered radioactivity may also be partly due to the presence of ¹⁴C-labeled metabolites or ¹⁴C incorporation into cell material, which did not partition to the pentane extract. Collectively, these experiments argue that anaerobic mineralization of dodecane by the enrichment was coupled to the reduction of sulfate.

Stoichiometry of dodecane oxidation coupled to sulfate reduction. The depletion of dodecane in the cultures relative to sterile controls extracted both at the start of the experiment and at the end was determined by the decrease in the ratio of the peak area of the dodecane to the tetradecane internal standard. In these experiments, 8% abiotic loss of dodecane was observed in the sterile controls over an 11-week incubation. An additional 70 to 73% degradation of dodecane was observed in triplicate nonsterile incubations. The amount of sulfate reduction theoretically expected, assuming complete dodecane mineralization, was determined according to equation 1:

(1)

After correction for sulfate reduction in controls lacking dodecane, the amount of sulfate reduction measured in the cultures was 94% ± 5% of the theoretically expected amount.

Dodecane degradation in bicarbonate-free medium. In another experiment, cells of the dodecane-degrading enrichment were washed and incubated in bicarbonate-free medium under an N₂ headspace. The medium was buffered with HEPES (25 mM, pH 7.5). Four replicate cell suspensions were incubated with dodecane, two of which were amended with bicarbonate (2 g/liter). After 7 weeks of incubation, analyses indicated that the extent of dodecane degradation and sulfate reduction was equal in all the cultures (data not shown). Thus, anaerobic alkane biodegradation by these cells was not bicarbonate dependent. This suggests that the mechanism for alkane metabolism was unlikely to involve carboxylation and is consistent with previous observations for strain AK-01 (22).

TMS-esterified metabolites of H₂₆- and D₂₆-dodecane. When ethyl acetate extracts of acidified cultures grown on H₂₆-dodecane were derivatized to form trimethylsilyl (TMS) esters and analyzed by GC-MS, two equally abundant trace metabolites eluted at 30.0 and 30.2 min. They gave nearly identical mass spectra, so only one is shown in Fig. 1a. Similarly, when cultures were grown on D₂₆-dodecane, two metabolites with nearly identical mass spectra eluted at 29.6 and 29.8 min, so a single example is shown in Fig. 1b. All four metabolites were detected in cultures amended with a mixture of H₂₆- and D₂₆-dodecane, while none were detected in sterile controls. All four metabolites showed abundant ions at m/z 73, which is due to the TMS group in these BSTFA-derivatized metabolites, and m/z 147, which is the (CH₃)₂Si==OSi(CH₃)₃ ion (Fig. 1). This latter ion indicates that the metabolites are di-TMS-derivatized compounds containing two BSTFA-reactive functional groups (18). These two ions are also the most abundant ions in the mass spectrum of an authentic standard of BSTFA-derivatized succinic acid (data not shown).

View larger version (21K): [in this window] [in a new window]   FIG. 1.   The mass spectra of both dodecylsuccinic acid metabolites of H26-dodecane (a) and D26-dodecane (b) analyzed as TMS esters.

Methyl-esterified metabolites of H₂₆- and D₂₆-dodecane. When extracts of cultures grown on H₂₆-dodecane were methyl esterified by treatment with diazomethane and analyzed by GC-MS, a single metabolite peak eluted at a retention time of 26.2 min that gave the mass spectrum shown in Fig. 2a. Similarly, cultures grown on D₂₆-dodecane gave one metabolite peak at a retention time of 25.8 min with the mass spectrum shown in Fig. 2b. Neither of these metabolites were detected in the corresponding sterile controls. The ion with the largest mass in the spectra of these metabolites occurs at m/z 283 with H₂₆-dodecane (Fig. 2a) and m/z 309 with D₂₆-dodecane (Fig. 2b). The latter ion is larger by 26 AMU than the former, accounting for all 26 of the deuterium atoms in the labeled parent substrate. However, these ions are not the M⁺ ions of the metabolites but rather the (M-31)⁺ ions that result from the loss of a methoxy group from the methyl-esterified succinyl moiety. This fragmentation pattern is typical of methyl esters (16) and gives the base peak in the mass spectrum of an authentic standard of methyl-esterified succinic acid for which the M⁺ ion was also not detected (data not shown). The molecular weights of these metabolites were verified as 314 and 340, respectively, by chemical ionization GC-MS with ammonia as the reagent gas. These analyses gave quasimolecular ions of (M+1)⁺ and (M+18)⁺, corresponding to (M+H)⁺ and (M+NH₄)⁺, respectively, at m/z 315 and 332 for the protonated metabolite and m/z 341 and 358 for the deuterated metabolite (data not shown). These molecular weights are consistent with the proposed dodecylsuccinic acid structures shown in Fig. 2 as the fumarate addition products of H₂₆- and D₂₆-dodecane. As discussed above, this assumes that the fumarate molecules were not significantly deuterated, even in cultures grown on D₂₆-dodecane.

View larger version (17K): [in this window] [in a new window]   FIG. 2.   The mass spectra of dodecylsuccinic acid metabolites of H26-dodecane (a) and D26-dodecane (b) analyzed as methyl esters.

Authentic standard of n-dodecylsuccinic acid. A synthesized authentic standard of protonated n-dodecylsuccinic acid eluted from the GC-MS as a single sharp peak at a retention time of 31.5 min when it was analyzed as a TMS ester. The mass spectrum of the derivatized standard (data not shown) showed the same fragmentation pattern as the metabolites from H₂₆-dodecane (Fig. 1a). However, the retention time of the standard was just over 1 min longer than that for the TMS-derivatized metabolites and only gave a single chromatographic peak, whereas the culture extract had two equally abundant, nearly coeluting metabolites. This confirms that the succinyl moiety in the metabolites cannot be attached to the terminal carbon atom in the dodecyl moiety. The structures depicted in Fig. 1 assume a subterminal attachment point at position 2 of the alkane, based on results of previous research (22). The detection of two equally abundant TMS-esterified metabolites suggests that the GC analysis is resolving two diastereomers which have the succinyl moiety located at the same subterminal carbon atom in at least two different absolute configurations. Any subterminal location of the succinyl moiety would give a metabolite that has two chiral carbon atoms. The two methine carbons, which form the bond between the dodecyl and succinyl moieties, would each be chiral and could each exist in either the R or the S configuration. Thus, there exist two possible pairs of enantiomers which have a diasteromeric relationship to each other. While enantiomers differ only in the ability to rotate the plane of polarized light, diastereomers can have different physical properties, which might allow their resolution by GC-MS analysis. Previous studies with benzylsuccinate synthase using toluene and xylenes have yielded the metabolites benzylsuccinic and methylbenzylsuccinic acids, respectively, which have only one chiral carbon atom eachthe methine carbon in the succinyl moiety (5). Thus, there is no possibility for these metabolites to exist as diastereomers. Benzylsuccinate synthase has been shown to form exclusively the (R)-(+)-enantiomer from toluene (14). If a similar enzyme specificity exists for the chiral methine carbon in the succinyl moiety of the dodecylsuccinic acid metabolites observed in this study, the other chiral methine carbon in the dodecyl moiety could conceivably still exist in either the R or the S configuration, giving two possible diasteromers (R,R- and R,S- referring to the methine carbons in the succinyl and dodecyl moieties, respectively). The authentic standard of n-dodecylsuccinic acid bears the succinic acid moiety on the terminal methylene group of the dodecyl moiety. Thus, because of symmetry in the dodecyl moiety, there is only one chiral carbon atom in the standardthe methine group in the succinyl moiety. With only one chiral carbon atom, there is not a possibility for diastereomers to exist and hence the standard gives only a single peak. Our results do not allow us to conclusively rule out the possibility that duplicate metabolite peaks exist as TMS esters, because there may be more than one methylene group in the alkane that undergoes the proposed fumarate addition reaction, giving at least two structural isomers. However, the fact that the two TMS-derivatized metabolites were always observed in equal abundance from each of H₂₆- and D₂₆-dodecane lends support to the stereochemistry argument.