anaerobic biodegradation of longer-chain n -alkanes coupled to methane production in oil sands...

Published: June 06, 2011

r 2011 American Chemical Society 5892 dx.doi.org/10.1021/es200649t | Environ. Sci. Technol. 2011, 45, 5892–5899

ARTICLE

pubs.acs.org/est

Anaerobic Biodegradation of Longer-Chain n-Alkanes Coupled toMethane Production in Oil Sands TailingsTariq Siddique,*,† Tara Penner,‡ Kathleen Semple,§ and Julia M. Foght*,§

†Department of Renewable Resources, University of Alberta, Edmonton, Alberta T6G 2E3, Canada‡Syncrude Canada Ltd. Research and Development, Edmonton, Alberta T6N 1H4, Canada§Department of Biological Sciences, University of Alberta, Edmonton, Alberta T6G 2E9, Canada

bS Supporting Information

’ INTRODUCTION

Oil sands tailings, a slurry of slightly alkaline water, sand, silt,clay, and residual hydrocarbons, are byproducts of bitumenextraction from surface mining and processing of oil sandsores.1 Vast oil sands operations in northern Alberta, Canadaproduce ∼1.31 million barrels of bitumen and generate∼262,000 m3 of tailings per day (http://www.energy.gov.ab.ca)that are deposited into settling basins (tailings ponds). Tailingsaccumulate because the producing companies operate under azero discharge policy. Currently, more than 170 km2 in the oilsands region are covered by tailings ponds containing ∼840million m3 of fine tailings (http://www.ercb.ca). Mildred LakeSettling Basin (MLSB), the largest tailings pond operated bySyncrude Canada Ltd., currently contains >400millionm3 of finetailings.2 Challenges associated with tailings ponds include thepresence of inorganic and organic contaminants (metals, salts,petroleum hydrocarbons, naphthenic acids, etc.), emission ofbiogenic greenhouse gases (CH4 and CO2), and very slowconsolidation (settling) of fine tailings solids (which segregatefrom the sand component after deposition). The fine tailingssettle by gravity most rapidly during the first 3�4 years afterdeposition, forming mature fine tailings (MFT; also called fluidfine tailings, FFT) which can then require decades for significantincremental settling. Additional challenges faced by the oil sands

industry include dewatering ofMFT (i.e., recovery of water in thetailings slurry for reuse in processing), reducing the storedtailings volumes, and using the consolidated MFT in landscapereconstruction.

Biogenic methane emissions from MLSB alone have beenestimated at 43,000 m3 day�1.3 Paradoxically, rather than inter-fering with gravitational settling of MFT, methane production isassociated with accelerated settling of tailings solids both in situand in the laboratory.4 To understand and possibly mitigatemethane emissions and to provide a scientific rationale forengineered tailings management, long-term laboratory experi-ments have been conducted by incubating MFT with hydro-carbons known to be present in the tailings ponds. MFT con-tains unrecovered bitumen (∼5 wt %) and a small proportion(e0.5 wt %) of fugitive solvent used in bitumen froth treatmentsuch as naphtha, a mixture of aliphatic and aromatic hydrocar-bons in the range of C3�C14.

5 We previously reported thatspecific components of the residual naphtha that enters MLSBwith fresh tailings were metabolized by the indigenous microbial

Received: February 24, 2011Accepted: May 24, 2011Revised: May 16, 2011

ABSTRACT: Extraction of bitumen from mined oil sands oresproduces enormous volumes of tailings that are stored insettling basins (current inventory g840 million m3). Ourprevious studies revealed that certain hydrocarbons (short-chain n-alkanes [C6�C10] and monoaromatics [toluene,o-xylene, m-xylene]) in residual naphtha entrained in thetailings are biodegraded to CH4 by a consortium of micro-organisms. Here we show that higher molecular weightn-alkanes (C14, C16, and C18) are also degraded under metha-nogenic conditions in oil sands tailings, albeit after a lengthylag (∼180 d) before the onset of methanogenesis. Gaschromatographic analyses showed that the longer-chain n-alkanes each added at∼400mg L�1 were completely degradedby the resident microorganisms within ∼440 d at ∼20 �C. 16S rRNA gene sequence analysis of clone libraries implied that thepredominant pathway of longer-chain n-alkane metabolism in tailings is through syntrophic oxidation of n-alkanes coupled withCO2 reduction to CH4. These studies demonstrating methanogenic biodegradation of longer-chain n-alkanes by microbes nativeto oil sands tailings may be important for effective management of tailings and greenhouse gas emissions from tailings ponds.

5893 dx.doi.org/10.1021/es200649t |Environ. Sci. Technol. 2011, 45, 5892–5899

Environmental Science & Technology ARTICLE

communities to CH4.6 Paraffinic (nC6�C10) and aromatic

(toluene and xylene isomers) components of naphtha werebiodegraded during one year of incubation, but the branched(isoparaffins) and cyclic (naphthenes) aliphatics remained un-degraded after one year.5,6

The mass of bitumen-associated n-alkanes entering MLSB canbe roughly estimated by using the following values: acyclicalkanes of length <C19 comprise ∼0.04 wt % of Athabascabitumen;7 unextracted bitumen comprises e5 wt % of MFT inMLSB;5 and∼178 million m3 of MFT with density∼1.28 kg L�1

are deposited in MLSB annually,2 representing a projected mass of∼4560 tonnes of bitumen-associated linear alkanes (<C19) enteringthe pond in 2011. Thus, there is a substantial and constant input oflonger-chain n-alkanes that become exposed to potential microbialbiodegradation in the tailings pond. This study was conducted toassess the potential of tailings pond microbes to biodegrade n-C14,C16, and C18 under methanogenic conditions. Active microbialcommunities were characterized using 16S rRNA gene clonelibraries to indicate the dominant organisms and infer thealkane degradation pathway to CH4. The results will helpdefine variables for an existing kinetic model to predict CH4

flux from oil sands tailings ponds.8 More generally, becausereports of methanogenic biodegradation of longer-chainalkanes are sparse,9 the current study may help predict biode-gradability of crude oil in subsurface petroleum reservoirs aswell as methanogenic biodegradation of petroleum hydrocar-bons in anaerobic environments undergoing bioremediation.

’EXPERIMENTAL SECTION

Chemicals andMaterials. n-Tetradecane (nC14, 99% purity),n-hexadecane (nC16, 99%), and n-octadecane (nC18, 99%) werepurchased from Sigma-Aldrich, Milwaukee, WI, USA. “Hexanes”(HPLC grade H302�4, g95% n-hexane), hereafter calledn-hexane, and other chemicals (analytical reagent grade) werepurchased from Fisher Scientific, Ontario, Canada. The MFTwas collected by Syncrude Canada Ltd. fromMLSB at 6 m depthin July 2005 and stored at 4 �C until used as inoculum in 2006.The MFT contained solids (∼40% by wt.), bitumen (4.4%by wt.), and naphtha (0.4% by wt.) at pH 7.8 and electricalconductivity of 4200 μS cm�1. The complete MFT analysis hasbeen reported.5

Quantitation of n-Alkane (nC14, nC16, nC18) Biodegrada-tion. Anaerobic experimental microcosms of 120-mL capacityreceived 40 mL each of MFT and methanogenic medium lackingorganic carbon, prepared as described previously5 with a head-space of 30% O2-free CO2, balance N2. Single microcosms werespiked with either nC14 and nC16 (“2-alkanes”) or nC14, nC16,and nC18 (“3-alkanes”), at final concentrations of ∼400 mg L�1

of each compound. Duplicate viable ‘baseline’ control micro-cosms were prepared, consisting of MFT and medium withoutadditional alkanes, to account for metabolism of endogenoushydrocarbons in the MFT. Heat-killed control microcosms wereprepared in parallel with the test microcosms, autoclaved eachday for 4 consecutive days to ensure sterility, and finally spikedwith the three alkanes. The microcosms were incubated in thedark without shaking at ∼20 �C, ca. in situ temperature in thetailings pond, for 438 d until a plateau ofmethane production wasapproached. Four-milliliter samples of the cultures were with-drawn, after vigorous shaking to ensure representative samplingof the tailings, for residual alkane analysis at intervals dictated bythe rate of methane production, placed into 8-mL glass vials

capped with Teflon-coated septa (B7990-3; National Scientific,Rockwood, TN, USA), and stored at �20 �C before analysis.To compensate for lack of replication of the two initial test

cultures (2-alkanes and 3-alkanes), at∼500 d, 10-mL portions ofthe 3-alkane culture were transferred to duplicate microcosmscontaining 40 mL of fresh methanogenic medium plus the threealkanes to demonstrate continued production of CH4 fromlonger-chain n-alkanes in conjunction with degradation of thealkanes. This general procedure was followed in three additionalsuccessive transfers of the enrichment culture.Chemical Analyses. Methane production in the microcosms

was measured by using a sterile needle and syringe to remove0.1 mL of headspace for analysis by gas chromatography (GC)with a flame ionization detector.4 At intervals excess headspacepressure in the microcosms was reduced by aseptically removinga known volume of gas and replenishing with known volume of30% O2-free CO2, balance N2. The volumes of gases removedand replenished were accounted for when calculating the totalmethane production.Hydrocarbons sorbed onto MFT can be extracted for analysis

using appropriate organic solvents, as demonstrated in our previousstudies5,6 where methanol was effectively used to extract short-chain n-alkanes, naphtha and BTEX compounds from MFTincubated for ∼46 weeks. In the present study, an extractionmethod optimized and calibrated in our laboratory used amixture of n-hexane and acetone to effectively extract longer-chain n-alkanes from culture samples collected and storedat �20 �C. Briefly, 2-mL samples were placed in 20 mL EPAglass vials capped with Teflon-coated septa (03-339-14C; FisherScientific), and pristane (final concentration 300 ppm) wasadded as a surrogate extraction standard. Then 5 mL of acetone(Spectranalyzed grade A119-4; Fisher Scientific) and 5 mL ofn-hexane were added into the vials followed by shaking for30 min at room temperature. The acetone-hexane phase wasremoved to a new vial, and the MFT-medium slurry was re-extracted with 10 mL of fresh acetone and n-hexane. The twoextracts were combined, dried by passage through anhydroussodium sulfate (Na2SO4), concentrated under a stream of N2 to<1 mL, and transferred to a gas chromatography autosamplervial. The final extract volume was adjusted to 1.5 mL with hexaneand analyzed using an Agilent 5673 GC-mass spectrometry (GC-MS) system with a DB5-MS capillary GC column. The columnwas held at 65 �C for 2 min and then increased at 4 �C min�1 to280 �C. Helium was used as a carrier gas with a flow rate of1.1 mLmin�1 under splitless conditions. Peak areas of individualcompounds were determined using a quantitation programderived from standard calibration curves created using appro-priate external and extraction standards.The masses of nC14, nC16, and nC18 biodegraded during 438 d

incubation were used to calculate the maximum theoreticalmethane yield using stoichiometric equations derived from theSymons and Buswell equation10

n-Tetradecane ðnC14Þ : C14H30 þ 6:5H2O

f 3:25CO2 þ 10:75CH4 ð1Þ

n-Hexadecane ðnC16Þ : C16H34 þ 7:5H2O

f 3:75CO2 þ 12:25CH4 ð2Þ

n-Octadecane ðnC18Þ : C18H38 þ 8:5H2O

f 4:25CO2 þ 13:75CH4 ð3Þ



Characterization of Microbial Communities. To prepare16S rRNA gene clone libraries, after 438 d incubation duplicatealiquots of 300 μL were collected from each of the two baselinecontrols (unamended MFT) and the 3-alkane enrichment cul-ture (amendedMFT). To enhance DNA recovery from theMFTwhich has a high clay content, total community DNA wasextracted by subjecting each aliquot to two sequential lysis stepsby bead beating (FastPrep Bead Beater, Bio101 Systems) in thepresence of SDS, chloroform, and isoamyl alcohol as previouslydescribed.11 Cell debris was precipitated with ammonium acetateand removed by centrifugation, and the DNA was purified byprecipitation with isopropanol. The DNA extracted from the twosequential lysis steps was pooled for each aliquot. To average thevariability associated with amplification, three replicated inde-pendent PCR amplifications were conducted for each pooledextract using either bacterial primers (PB36 and PB38; ref 12) orarchaeal primers (21F and 958R; ref 13) and Phusion HighFidelity DNA Polymerase (Finnzymes OY, Finland) to yieldnear full-length 16S rRNA genes. PCR conditions for bacterial16S rRNA gene amplification consisted of a 3 min initialdenaturation at 98 �C, followed 98 �C for 30 s, 54 �C for 30 s,and 72 �C for 1 min for 30 cycles, followed by a final extension at72 �C for 10 min. The PCR conditions for archaeal genes weresimilar except that 27 PCR cycles were used with annealing at51 �C. After amplification, the PCR products from the baseline

control (3 PCR � 2 aliquots � 2 microcosms) and 3-alkaneculture (3 PCR � 2 aliquots � 1 microcosm) were pooledseparately for construction of a bacterial and an archaeal clonelibrary. Appropriate negative (reagent only) extraction controlswere included during DNA extraction, and positive and negativePCR reagent controls were included during gene amplification.PCR products were purified from agarose using the QIAquick

Gel Purification Kit (Qiagen Inc., Missisauga, ON, Canada) andcloned into the pJET1/blunt cloning vector using the manufac-turer’s protocol (Fermentas Life Sciences, Burlington, ON,Canada). Cloned inserts were reamplified from 1 μL of cellculture (grown in 100 μL LB broth overnight at 37 �C) usingplasmid-based primers pJETF and pJETR (Fermentas). Theproducts were subjected to amplified rDNA restriction analysis(ARDRA) by digestion with the restriction endonucleasesHaeIIIor CfoI separately under conditions specified by the supplier(Fermentas), and the fragments were resolved by electrophor-esis. Digestion pattern analysis was performed using GelProAnalyzer Software version 4.5 (Media Cybernetics, Inc.), andrestriction patterns were grouped based on both visual inspectionand comparison of fragment sizes. Initially, only ARDRA pat-terns comprising g2 clones were sequenced, but subsequentlyadditional singleton clones were selected randomly and sequencedto expand the phylogenetic information in the libraries. Theadditional sequences confirmed that the singleton clones were

Table 1. Sequences Affiliated with Deltaproteobacterial 16S rRNA Genes Detected in Microbial Communities Enriched afterIncubation of MFT with Three Longer-Chain n-Alkanes or in a Viable Baseline Control

number of clones

clone sequence

accession no. best sequence matches (BLASTn)a sequence similaritybbaseline control

(total 87)

3-alkane culture

(total 76)

Syntrophobacterales

EU522631 Syntrophus sp. (AJ133795) 1400/1419 (98%) 3 12

EU522637 uncultured eubacterium (AF050534) 1312/1363 (96%) 1 1

uncultured Deltaproteobacterium (EF420213) 1311/1365 (96%)

EU522632 uncultured Deltaproteobacterium (FN429791) 1297/1364 (95%) 1 5

Syntrophus sp. clone B2 (AJ133795) 1296/1365 (94%)

EU522634 uncultured bacterium (AM086113) 1336/1377 (97%) 2 3

Syntrophus aciditrophicus (GU993263) 1296/1383 (93%)

HM992528 uncultured bacterium (GU996558) 595/613 (97%) 1 5

Syntrophus sp. clone (AJ133795) 593/615 (96%)

EU522636 uncultured eubacterium (AF050534) 1345/1404 (95%) 1

Syntrophus sp. (AM933651) 1330/1421 (93%)

JF305753 Syntrophus sp. (AJ133795) 1317/1376 (95%) 3

EU522635 uncultured Syntrophus sp. (EU050697) 1221/1313 (92%) 2

HM992529 uncultured bacterium (AM086113) 722/765 (94%) 1

uncultured Syntrophaceae bacterium (EU043564) 715/765 (93%)

Desulfobacterales

EU522638 enrichment culture bacterium (AM933657) 1431/1435 (99%) 1

Algidimarina propionica strain (AY851291) 1338/1428 (93%)

EU522639 sulfate-reducing bacterium TRM1 (Desulfobulbaceae) (GU133208) 1356/1370 (99%) 2

uncultured Deltaproteobacterium (EU266811) 1363/1368 (99%)

Desulfuromonadales

EU522641 uncultured bacterium (GQ339175) 1371/1410 (97%) 5

uncultured Desulfuromonas sp. (EF205265) 1331/1425 (93%)aGenBank accession date July, 2010. bNumber of nucleotides in query sequence with identity to GenBank sequence.



not rare OTUs but rather represented microdiversity of taxaalready detected in the libraries. Inserts from the clones selectedfor sequencing were reamplified using pJETF and pJETR pri-mers, purified using the High Pure PCR Purification Kit (RocheDiagnostics) following the manufacturer’s protocol, sequencedusing BigDye Terminator mix (Applied Biosystems Inc., USA),and resolved on an Applied Biosystems 373A automated DNAsequencer. The forward primers PB36 and 21F were used forpartial sequencing of bacteria and archaea, respectively. Sequencesthat were less than 3% divergent were grouped into operationaltaxonomic units (OTUs), and a consensus clone from eachOTUwas selected for near full length sequencing using previouslydescribed primers.14 Chimeric sequences were detected with theprograms Mallard and Pintail15,16 and were excluded from furtheranalyses. The 16S rRNA gene sequences from clone librarieswere compared to sequences in GenBank17 using BLAST-n todetermine related taxa. Presumptive identities were supported byphylogenetic tree construction (not shown). Sequences havebeen deposited inGenBank (accession numbers given in Table 1,Table 2, and Supporting Information Table S1).

’RESULTS

Methane Production.Methanogenic biodegradation of threelonger-chain n-alkanes was monitored at intervals by measuringthe mass of CH4 generated and by quantifying residual n-alkanesin cultures. The cumulative CH4 production byMFT spiked withtwo alkanes (nC14, nC16) or three alkanes (nC14, nC16, nC18)during 438 d incubation is shown in Figure 1A. By ∼180 dincubation, CH4 production from the 3-alkane microcosmexceeded that of the baseline control, whereas the 2-alkanemicrocosm exhibited a lag of ∼280 d. Total CH4 productionby 438 d was calculated to be 2.8 and 4.2 mmol for the 2-alkaneand 3-alkane microcosms, respectively, compared with the dupli-cate viable baseline controls that produced only 0.5 ( 0.1 mmol

CH4 from the endogenous organic material in the MFT(Figure 1A). The heat-killed control microcosms did notproduce any methane (data not shown).n-Alkane Biodegradation and Stoichiometric Methane

Production. n-Alkane biodegradation was monitored at intervals(Figure 1B). GC-MS analyses of recovered hydrocarbonsshowed that the initial masses of n-alkanes added to the micro-cosmswere 354 and 408mg L�1 of nC14 and nC16 in the 2-alkanemicrocosm and 354, 415, and 317 mg L�1 of nC14, nC16, andnC18, respectively, in the 3-alkane microcosm. Little biodegrada-tion of n-alkanes was observed in either test microcosm by 255 d,but thereafter concentrations of all n-alkanes in both test micro-cosms decreased sharply (Figure 1B). By 438 d, the residualconcentrations of n-alkanes were 89 and 24 mg L�1 for nC14 andnC16, respectively, in the 2-alkane microcosm, representing 75%and 94% depletion of the initial masses. In the 3-alkane micro-cosm, 62, 28, and 45 mg L�1 of nC14, nC16, and nC18,respectively, were measured at 438 d, representing 82%, 93%,and 85% depletion. n-Alkane concentrations in the heat-killedcontrol microcosms did not change during the incubation period(Figure 1B, inset); recovery was >90% of the initial mass added.The mass losses of individual alkanes biodegraded during the

incubation period were fit into the respective stoichiometricequations (eqs 1 to 3) to estimate maximum theoretical methaneproduction resulting from their metabolism. The stoichiometriccalculations predicted production of 3.5 and 5.0 mmol CH4 inthe 2-alkane-and 3-alkane microcosms, respectively. Thus, themeasured CH4 (Figure 1A) represented 80% and 84% of thesetheoretical maxima at 438 d.Sustainability of Methane Production. At ∼500 d, the

3-alkane culture was transferred in duplicate to fresh mediumplus ∼300 mg L�1 each of nC14, nC16, and nC18 and monitoredfor CH4 production and alkane biodegradation. The culturetransfers sustained their activity of producing CH4 with thebiodegradation of added n-alkanes (data not shown). These

Table 2. Archaeal 16S rRNA Gene Sequences Detected in Microbial Communities Enriched after Incubation of MFT with ThreeLonger-Chain n-Alkanes or in a Viable Baseline Control

number of clones

clone sequence

accession no.

best sequence matches

(BLASTn)a sequence similaritybbaseline control

(total 68)

3-alkane culture

(total 58)

Methanomicrobiales

EU522625 uncultured euryarchaeote (AF374276) 825/838 (98%) 11 21

uncultured archaeon clone (EU662674) 818/838 (98%)

Methanomicrobiales archeon (AB479390) 815/838 (97%)

EU522626 uncultured archaeon clone (EU481619) 629/632 (99%) 3

uncultured Methanomicrobiales (CU917177) 608/631 (96%)

EU522630 uncultured Methanoculleus sp. (EU721744) 765/771 (99%) 4

Methanosarcinales

EU522627 Methanosaeta sp. clone A1 (AJ133791) 860/866 (99%) 32 27

Methanosaeta harundinacea (AY970347) 845/867 (97%)

EU522628 uncultured Methanosarcinales (CU916297) 832/839 (99%) 19 3

uncultured Methanosaeta clone A12 (EU888812) 824/828 (99%)

Unclassified Archaea

EU522629 uncultured euryarchaeote clone F3 (EU910627) 846/859 (98%) 6

uncultured archaeon clone OK3 (GQ406364) 843/859 (98%)aGenBank accession date July, 2010. bNumber of nucleotides in query sequence with identity to GenBank sequence.



duplicate transfer cultures have themselves been transferredsequentially an additional two times since 2007. Biodegradationlosses of the n-alkanes in all transfer cultures followed the sametrends of CH4 production and n-alkane depletion shown by theoriginal culture in Figure 1B, thus demonstrating reproducibilityand sustainability of the activity.Microbial Community Structure. The bacterial and archaeal

communities in the 3-alkane microcosm and baseline controlcultures were analyzed after 438 d incubation by construction of16S rRNA gene clone libraries. For the bacterial libraries, 87clones (of 135 in the library) and 76 clones (of 117) weresequenced for the baseline control and 3-alkane culture, respec-tively. Because of the lower apparent biodiversity of the archaeallibraries, smaller libraries were constructed; 68 clones (of 81) and58 clones (of 75) were sequenced from the baseline control and3-alkane cultures, respectively.The sequenced bacterial clones from the baseline control and

3-alkane microcosms comprised sequences related to Deltapro-teobacteria, Firmicutes, Chloroflexi, Betaproteobacteria, andBacteroidetes (Figure 2A,B, Table 1, and Supporting Informa-tion Tables 1S, 2S, and 3S). Deltaproteobacteria dominated the3-alkane culture library (45% of sequenced clones) versus 17% of

sequenced clones in the viable baseline control culture. Withinthis taxon, clones related to Syntrophaceae were dominant, withsequence EU522631 (98% similar to Syntrophus sp.) being mostabundant (Table 1). Clones related to Bacteroidetes also in-creased from 4% (control) to 13% (3-alkane culture) of thesequenced clones. Clones categorized as “Unclassified” lackedsubstantial homology (<90%) to named sequences in GenBank.As a result, clones related to other bacterial taxa (Firmicutes,Chloroflexi, and Betaproteobacteria) decreased proportionally inthe n-alkane culture versus the baseline control (Figure 2A,B andSupporting Information Tables 1S, 2S, and 3S).The archaeal 16S rRNA gene clone libraries contained sequences

similar to uncultured clones from two major functional groups ofmethanogens: presumptive Methanomicrobiales (hydrogeno-trophic methanogens) increased from 16% in the baselinecontrol to 48% of the sequenced clones from the 3-alkaneculture, whereas presumptive Methanosarcinales (acetoclasticmethanogens) decreased from 75% to 52% (Table 2). Sequenceanalysis revealed that clones affiliated with Methanomicrobialeswere most closely related (g96%) to uncultured sequencesincluding Methanoculleus. Methanosaeta spp. represented themajority of the clones in the Methanosarcinales (Table 2). Afew sequences (9% of sequenced clones) related to unclassifiedEuryarchaeota were detected only in the baseline control culture(Figure 2C, Table 2).

’DISCUSSION

Biodegradation of longer-chain n-alkanes under methano-genic conditions is not extensively documented; instead moststudies have been conducted under sulfate- or thiosulfate-reducing conditions.9 Therefore, the documented presence ofshort-chain alkane-degrading consortia in tailings ponds 5,6 pro-vided the possibility of enrichingmethanogenic cultures capable oflonger-chain alkane biodegradation. In addition, in previousmethanogenic studies with MFT we observed preferential utiliza-tion of individual short-chain n-alkanes (provided both asmixturesof pure alkanes5 and as native components of naphtha6) in theorder nC10 > nC8 > nC7 > nC6, and we wished to determinewhether a similar pattern of utilization also pertained to longer-chain alkanes (nC14, nC16, and nC18). Finally, a predictive modeldeveloped to estimate CH4 flux from MLSB8 was based onanaerobic biodegradation of short-chain n-alkanes (nC6-nC10)and monoaromatic (BTEX) constituents of residual naphtha inthe tailings. Because longer-chain alkanes contributed by unex-tracted bitumen entering the pondsmay also be a source of CH4 inMFT, it was necessary to examine the potential contribution ofsuch n-alkanes to tailings pond CH4 emissions. The microbialcommunity in MLSB MFT is now shown to be capable ofmethanogenic biodegradation of longer-chain alkanes as well asshort-chain alkanes.5,6 In contrast, Zengler et al.18 observedmethanogenic degradation of nC16 incubated with anoxic ditchsediment after a lag phase of 4 months but did not observemethanogenesis from the short-chain alkanes nC6 or nC10.

Long acclimation phases were observed before significantCH4 production occurred from longer-chain alkanes, with the3-alkane culture enduring a shorter lag phase (∼180 d) than the2-alkane culture (∼280 d). These results contrast with ourprevious study5,6 in which the same source of MFT incubatedwith short-chain alkanes (nC6�C10) consistently produced CH4

after <10 d lag time. The difference could be attributed to prioracclimation of the microbes in MFT to utilization of short-chain

Figure 1. Methane (CH4) production coupled to biodegradation oflonger-chain n-alkanes (C14, C16, and C18) by MFT incubated for 438 dat∼20 �C. (A) CH4 production from 2-alkane culture, 3-alkane-culture,and viable baseline control culture (no alkane addition). Horizontal linesindicate the theoretical maximum CH4 calculated using stoichiometricequations (see text). Error bars (where visible) represent 1 standarddeviation for duplicate baseline control cultures. (B) Biodegradation ofn-alkanes in 2-alkane culture (solid symbols), 3-alkane culture (opensymbols), and heat-killed control (inset).



n-alkanes, which are significant components of residual naphthapresent in the oil sands tailings, versus the low proportions oflonger chain alkanes associated with bitumen. In that case, theprotracted lag time exhibited by the longer-chain alkane culturesmight reflect a requirement either for enrichment of competentmicrobial consortia or, less likely, for induction of specific geneexpression. However, a similar lag phase of∼180 d was observedfor a subsequent transfer of the 3-alkane culture, arguing againstchanges in community structure during the lag phase. Similarly,the difference in lag time between 2-alkane and 3-alkane culturescannot be explained by shifting community structure, since alllonger-chain n-alkanes were consumed at approximately the samerate once biodegradation began (data not shown). Additional studyis required to understand the factors controlling lag phases asso-ciated with anaerobic biodegradation of longer-chain alkanes.

Methane production in this study correlated with the loss ofn-alkanes quantified by GC-MS. No preferential pattern of longer-chain n-alkane degradation was observed, in contrast to ourprevious study with short-chain alkanes5 showing preferential useof nC10 > nC8 > nC7 > nC6. However, the overall degradation oflonger-chain n-alkanes in MFT agrees with the recent reports ofmethanogenic degradation of crude oil containing longer-chainalkanes. Townsend et al.19 reported consumption of the n-alkanefraction (C13�C34) of a weathered Alaskan North Slope crudeoil by anoxic aquifer samples incubated under methanogenicconditions in laboratory microcosms. Similarly, Jones et al.20

incubated North Sea crude oil with river sediment undermethanogenic conditions for 686 d and observed loss of C7�C34 n-alkanes with corresponding CH4 production, and Gieget al.21 documented methanogenic conversion of crude oilalkanes by an enrichment culture derived from gas condensate-contaminated subsurface sediments.

In the present study, experimental production of CH4 during438 d incubation approached 80�84% of the theoretical maximumCH4 predicted by stoichiometry. The difference between mea-sured and predicted CH4 values might be accounted for bycarbon assimilation into microbial biomass and inefficiency ofinterspecies H2 transfer in a diverse consortium of anaerobicmicroorganisms in the MFT. The results support our earliercalculations that 77�79% of predicted CH4 was produced fromshort-chain n-alkanes,5 whereas Zengler et al.18 observed only64% of predicted CH4 from metabolism of nC16 in an enrich-ment culture.

Although efforts were made to construct the 16S rRNA geneclone libraries by independently extracting and amplifying re-plicate subsamples of cultures, interpretation of the clonelibraries is speculative because they were derived from singlecultures. However, the general community compositions de-tected are consistent with metagenomic 16S rRNA gene se-quences from analogous cultures (unpublished data) and clonelibrary analysis of uncultivated MFT.22 Presumptive identifica-tion of sequenced clones pointed to taxa that may be importantfor anaerobic degradation of longer-chain alkanes, inferred fromincreased clone abundance in the 3-alkane culture versus theviable baseline control clone libraries. For example, the propor-tions of clones affiliated with Deltaproteobacteria and unclassifiedBacteria significantly increased after incubation with n-alkanes.Of the nine OTUs affiliated with the Deltaproteobacteria, allwere presumptively identified as Syntrophus spp., and the domi-nant sequence (EU522631) was closely related to a Syntrophussp. associated with methanogenic degradation of hexadecane inan enrichment culture.18 The archaeal clone library constructedfrom the 3-alkane culture exhibited an increased proportion ofhydrogenotrophic methanogens and decreased acetoclastic

Figure 2. Relative abundance of sequenced clones in bacterial (A, B) and archaeal (C,D) 16S rRNA gene clone libraries constructed from viable baselinecontrol culture (A, C) or 3-alkane culture (B, D). N indicates the number of clones sequenced in each clone library.



methanogens versus the baseline control where sequences affiliatedwith acetoclastic methanogens dominated the clone library. Thedominant sequence (EU522625) associated with the hydroge-notrophic Methanomicrobiales was most closely related tomethanogens in the brackish and marine ends of an estuary.23

The dominant cloned sequence (EU522627) affiliated with theMethanosarcinales was closely related to acetoclastic Methano-saeta sp. associated with longer-chain alkane degradation.18

The dominance of Syntrophus spp. in the 3-alkane culturesupports their proposed role in the activation and fermentationof n-alkanes to acetate and H2 or further acetate conversion toCO2 and H2 as suggested by other recent reports.24,25 Extra-polating the current community structure results from this studysuggests that the principal pathway of longer-chain n-alkanemetabolism is through syntrophic oxidation of n-alkanes toacetate and H2 mediated by Syntrophus spp., likely followed byCO2 reduction with H2 utilization to CH4 (hydrogenotrophicmethanogenesis). This postulated pathway is consistent with theMADCOR process proposed by Jones et al.20 for methanogenicdegradation of n-alkanes in crude oil. Using isotopic analysis ofCO2 and CH4 produced in degraded oils from the Peace RiverOil Sands area of western Canada and laboratory microcosmdata, they estimated that 75�92% of methanogenesis in situoccurred through CO2 reduction after syntrophic oxidation of n-alkanes by Syntrophus spp. However, because half of the archaealclone library in our study comprised clones affiliated withacetoclastic methanogens even after incubation with n-alkanes,we cannot discount the potential role of the acetoclastic pathwayin methanogenic MFT cultures. This important pathway issupported by the findings of Zengler et al.18 who proposed thatacetoclastic CH4 production from nC16 degradation was thedominant pathway in an enrichment culture derived from anoxicsediment. A recent review of anaerobic alkane degradation9 citedstudies supporting both hydrogenotrophic and acetoclasticmethanogenic degradation of n-alkanes and concluded that thedominantmethanogenic pathwaywas likely dictated by the environ-ment and the provenance of the microbial community.

The current study shows that the microbial community in oilsands tailings is capable of utilizing longer-chain n-alkanes undermethanogenic conditions. This activity may affect the predictionof CH4 flux from a kinetic model,8 depending on the input oflonger-chain alkanes into tailings ponds with unrecovered bitu-men. More broadly, the results contribute to knowledge aboutthe potential for methanogenic degradation of longer-chain n-alkanes present in other anaerobic environments, such as hydro-carbon-contaminated aquifers and petroleum reservoirs.

’ASSOCIATED CONTENT

bS Supporting Information. Tables 1S-3S. This material isavailable free of charge via the Internet at http://pubs.acs.org.

’AUTHOR INFORMATION

Corresponding Author*E-mail: [email protected] (T.S.) and [email protected] (J.M.F.).

’ACKNOWLEDGMENT

The authors gratefully acknowledge funding fromNSERC(Post-Doctoral Fellowship to T.S.; Discovery Grants to J.M.F. and T.S.)

and Syncrude Canada Ltd. for providing tailings. We also thankJonathan L. Klassen for technical assistance.

’REFERENCES

(1) Schramm, L. L.; Stasiuk, E. N.; MacKinnon, M. Surfactants inAthabasca oil sands slurry conditioning, flotation recovery, and tailingsprocesses. In Surfactants, Fundamentals, and Applications in the PetroleumIndustry; Schramm, L. L., Ed.; Cambridge University Press: Cambridge,UK, 2000; pp 365�430.

(2) Syncrude Canada Limited. Annual tailings plan submission,Syncrude: Mildred Lake. Submitted to Energy Resources ConservationBoard, Alberta. 2010; pp 5�15.

(3) Holowenko, F. M.; MacKinnon, M. D.; Fedorak, P. M. Metha-nogens and sulfate-reducing bacteria in oil sands fine tailings waste. Can.J. Microbiol. 2000, 46, 927–937.

(4) Fedorak, P. M.; Coy, D. L.; Dudas, M. J.; Simpson, M. J.;Renneberg, A. J.; MacKinnon, M. D. Microbially-mediated fugitive gasproduction from oil sands tailings and increased tailings densificationrates. J. Environ. Eng. Sci. 2003, 2, 199–211.

(5) Siddique, T.; Fedorak, P. M.; Foght, J. M. Biodegradation ofshort-chain n-alkanes in oil sands tailings under methanogenic condi-tions. Environ. Sci. Technol. 2006, 40, 5459–5464.

(6) Siddique, T.; Fedorak, P. M.; MacKinnon, M. D.; Foght, J. M.Metabolism of BTEX and naphtha compounds to methane in oil sandstailings. Environ. Sci. Technol. 2007, 41, 2350–2356.

(7) Strausz, O. P.; Morales-Izquierdo, A.; Kazmi, N.; Montgomery,D. S.; Payzant, J. D.; Safarik, I.; Murgich, J. Chemical composition ofAthabasca bitumen: the saturate fraction.Energy Fuels 2010, 24, 5053–5072.

(8) Siddique, T.; Gupta, R.; Fedorak, P. M.; MacKinnon, M. D.;Foght, J. M. A first approximation kinetic model to predict methanegeneration from an oil sands tailings settling basin. Chemosphere 2008,72, 1573–1580.

(9) Mbadinga, S. M.; Wang, L.-Y.; Zhou, L.; Liu, J.-F.; Gu, J.-D.; Mu,B.-Z. Microbial communities involved in anaerobic degradation ofalkanes. Int. Biodeterior. Biodegrad. 2011, 65, 1–13.

(10) Roberts, D. J. Methods for assessing anaerobic biodegradationpotential. InManual of Environmental Microbiology, 2nd ed.; Hurst, C. J.,Crawford, R. L., Knudson, G. R., McInerney, M. J., Stetzenbach, L. D.,Eds.; ASM Press: Washington, DC, 2002; pp 1008�1017.

(11) Foght, J. M.; Aislabie, J.; Turner, S.; Brown, C. E.; Ryburn, J.;Saul, D. J.; Lawson, W. Culturable bacteria in subglacial sediments andice from two Southern Hemisphere glaciers. Microbial Ecol. 2004,47, 329–340.

(12) Saul, D. J., Aislabie, J. M.; Brown, C. E.; Harris, L.; Foght, J. M.2005. Hydrocarbon contamination changes the bacterial diversity of soilfrom around Scott Base, Antarctica. FEMS Microbiol. Ecol. 2005, 53,141-155.

(13) DeLong, E. F. Archaea in costal marine environments. Proc.Natl. Acad. Sci. U.S.A. 1992, 89, 5685–5689.

(14) Cheng, S. M.; Foght, J. M. Cultivation-independent and �dependent characterization of Bacteria resident beneath John EvansGlacier. FEMS Microbiol. Ecol. 2007, 59, 318–330.

(15) Ashelford, K. E.; Chuzhanova, N. A.; Fry, J. C.; Jones, A. J.;Weightman, A. J. At least 1 in 20 16S rRNA sequence records currentlyheld in public repositories is estimated to contain substantial anomalies.Appl. Environ. Microbiol. 2005, 71, 7724–7736.

(16) Ashelford, K. E.; Chuzhanova, N. A.; Fry, J. C.; Jones, A. J.;Weightman, A. J. New screening software shows that most recent large16S rRNA gene clone libraries contain chimeras. Appl. Environ. Micro-biol. 2006, 72, 5734–5741.

(17) Benson, D. A.; Karsch-Mizrachi, I.; Lipman, D. J.; Ostell, J.;Wheeler, D. L. GenBank. Nucleic Acids Res. 2007, 35, D21–D25.

(18) Zengler, K.; Richnow, H. H.; Rossello-Mora, R.; Michaelis, W.;Widdel, F. Methane formation from long-chain alkanes by anaerobicmicroorganisms. Nature 1999, 401, 266–269.

(19) Townsend, G. T.; Prince, R. C.; Suflita, J. M. Anaerobicoxidation of crude oil hydrocarbons by the resident microorganisms



of a contaminated anoxic aquifer. Environ. Sci. Technol. 2003, 37,5213–5218.(20) Jones, D. M.; Head, I. M.; Gray, N. D.; Adams, J. J.; Rowan,

A. K.; Aitken, C. M.; Bennett, B.; Huang, H.; Brown, A.; Bowler, B. F. J.;Oldenburg, T.; Erdmann, M.; Larter, S. R. Crude oil biodegradation viamethanogenesis in subsurface petroleum reservoirs. Nature 2008,451, 176–180.(21) Gieg, L. M.; Duncan, K. E.; Suflita, J. M. Bioenergy production

via microbial conversion of residual oil to natural gas. Appl. Environ.Microbiol. 2008, 74, 3022–3029.(22) Penner, T. J.; Foght, J. M. Mature fine tailings from oil sands

processing harbor diverse methanogenic communities. Can. J. Microbiol.2010, 56, 459–470.(23) Purdy, K. J.; Munson, M. A.; Nedwell, D. B.; Embley, T. M.

Comparison of the molecular diversity of the methanogenic communityat the brackish and marine ends of a UK estuary. FEMS Microbiol. Ecol.2002, 39, 17–21.(24) Gray, N. D.; Sherry, A.; Hubert, C.; Dolfing, J.; Head, I. M.

Methanogenic degradation of petroleum hydrocarbons in subsurfaceenvironments: Remediation, heavy oil formation and energy recovery.Adv. Appl. Microbiol. 2010, 72, 137–161.(25) Head, I. M.; Larter, S. R.; Gray, N. D.; Sherry, A.; Adams, J. J.;

Aitken, C. M.; Jones, D. M.; Rowan, A. K.; Huang, H.; Roling, W. F. M.Hydrocarbon degradation in petroleum reservoirs. In Handbook ofHydrocarbon and Lipid Microbiology; Timmis, K. N., Ed.; Springer-Verlag: Berlin, Heidelberg, 2010; pp 3097�3109.

anaerobic biodegradation of longer-chain n -alkanes coupled to methane production in oil sands...

Documents