Org. Geochem. Vol. IS, No. I, pp. 17-22, 1992 Printed in Great Britain. All rights reserved
0146-6380192 $5.00+0.00 Copyright © 1992PergamonPress pie
Biodegradation studies of unresolved complex mixtures of hydrocarbons: model UCM hydrocarbons and the aliphatic UCM M. A. GOUOH,* M. M. RHEAD and S. J. R O W L A N D t Petroleum and Environmental Geochemistry Group, Department of Environmental Sciences, Polytechnic South West, Drake Circus,Plymouth PLI 8AA, England (Received 3 September 1990; returned for revision 23 October 1990; accepted 13 June 1991)
Abstract--An unresolved complex mixture (UCM) of hydrocarbons isolated from a commercially available lubricating oil, and a synthetic mixture of C25 hydrocarbons were subjected to biodegradation by an aerobic bacterium, Pseudomonasfluorescens, for 0-25 days. The rate and extent of biodegradation was influenced by molecular structure. Straight chain and monomethyl-alkanes were rapidly degraded in the first 10 days in the approximate order n-C25> 2-methyl-> 9-methyl-tetracosane. The remaining alkanes were degraded more slowly, initially at a rate comparable to the UCM (ca 1% day -l up to day 14) but thereafter more quickly (ca 3% day-'). The UCM was degraded as a whole, i.e. no reduction in the proportion of resolved vs unresolved alkanes was observed. Key words--UCM, biodegradation, hydrocarbons, lube oil
INTRODUCTION
surprising since many structurally simple hydrocarbons are relatively easily degraded by microbes. There are many reports of the preferential microbial oxidation of n-alkanes relative to branched alkanes, both in laboratory studies using pure bacterial strains (e.g. McKenna and Kallio, 1964; Pirnik et al., 1974; Robson and Rowland, 1989) and in environmental oil spills (e.g. Atlas et al., 1981; Jones et al., 1986). Simply branched (e.g. monomethyl) alkanes are also susceptible to attack by microbes, as observed with individual hydrocarbons (e.g. Thijsse and Van der Linden, 1961; Mckenna and Kallio, 1964; Pirnik et al., 1974) and in crude oil biodegradation studies (Connan et al., 1980; Connan, 1984). It is therefore necessary to assess the susceptibility of simple compounds such as 7-n-hexylnonadecane (3), 9-(2-cyclohexylethyl)-heptadecane (4) and 9-(2-phenylethyl)heptadecane (5) to microbial degradation; and to compare the results with degradation studies of the UCM.
The degradation of petroleum by bacteria often results in the progressive depletion of chromatographically resolved hydrocarbons (e.g. n-alkanes, acyclic isoprenoid alkanes; alkyl benzenes, naphthalenes and phenanthrenes) relative to the unresolved hydrocarbon mixture (UCM). This has been noted in recent sediments affected by oil spills (e.g. Blumer et ai., 1973; Atlas et al., 1981; Oudet et al., 1981); in laboratory degraded crude oils (e.g. Bailey et al., 1973; Rubinstein et al., 1977; Jones et al., 1986) and in crude oils biodegraded in the reservoir (e.g. Deroo et al., 1974; Volkman et al., 1983, 1984; Connan, 1984). Hence, the UCM is thought to comprise compounds which are relatively inert to microbial degradation. The general consensus is that the UCM is a mixture of many structurally complex isomers and homologues of branched and cyclic hydrocarbons (e.g. Eglinton et al., 1975; Alexander et al., 1982; Sanders and Tibbetts, 1987). Recently, oxidative and other studies (Killops and A1-Juboori, 1990; Gough and Rowland, 1990) have suggested that some UCMs consist, in part, of mixtures of fairly simple compounds comprising unsubstituted alkyl chains up to Cm9. Cough and Rowland (1990) suggested that these may include isomeric monoalkyl substituted "T'-branched alkanes [e.g. 7-n-hexylnonadecane (3)]. If this supposition is correct, the resistance of the UCM to microbial degradation is perhaps somewhat
EXPERIMENTAL
An experiment was devised to measure the biodegradation rates of the synthetic model hydrocarbons. Also included in the test mixture were compounds which are not thought to be representative of the aliphatic UCM, but which were useful for comparison [n-pentacosane (8), 2-methyltctracosane (7), 9-methyltetracosane (6), 2,6,10,14,18-pentamethyleicosanc (2), and 2,6,10,14-tetramethyl-7(3-methylpentyl)-pentadecane (1) (Fig. 5)]. The organism chosen for the study was a pure strain of Pseudomonas fluorescens, obtained by enrichment from a used metal working fluid (Dr D. Gaylarde, City of London Polytechnic, personal communication).
*Present address: National Rivers Authority, Southern Region Laboratory, 4 The Meadows, Waterberry Drive, Waterlooville, Portsmouth PO7 7XX, England. tTo whom all correspondence should be addressed. IS/I--B
17
M.A. GOUGHet
18
al.
i30
i29 Ld 0'3 Z 0 O.. CO U.I re"
I
[illf)t!
2 3,
i25
10I
2 I0
i35
310
4I0
510
TIME (minutes) Fig. l. Gas chromatogram of aliphatic hydrocarbons isolated from a lubricating oil base stock showing the occurrence of resolved components identified as principally acyclic isoprenoid alkanes (i24-i35), superimposed on the envelope of unresolved components referred to as the unresolved complex mixture (UCM). P s e u d o m o n a s species are well known hydrocarbon degraders, they are widespread and are often the most dominant of the hydrocarbon-utilizing microbes in the marine environment (Karrick, 1977). A pure strain was used so that microbial degradation rates could be measured, and all hydrocarbons contained 25 carbon atoms so that any variations could be ascribed solely to molecular structure. The procedures used were based on those reported by Robson and Rowland (1989), and involved addition of a known amount of the hydrocarbon mixture to flasks containing sterilized minimal salts medium and bacterial broth. The aliphatic UCM was treated in an identical manner in separate flasks at the same sample loading (i.e. c a 80#gem-3). The flasks, and hydrocarbon contents, were incubated for varying periods of time in the dark on a shaking water bath at 20°C. Residual hydrocarbons were recovered by extraction with DCM, the extracts were dried (anhydrous Na2SO4), the solvent removed, and made up to volume with DCM prior to analysis by GC. The synthetic hydrocarbon mixture was quantified by reference to an external standard composed of all eight hydrocarbons at known concentrations. Replicate analysis ( x 10) indicated a GC reproducibility of ca 6% per compound. Variation in detector
response was assessed by monitoring integrated peak areas of each component daily. Throughout the course of the experiment these did not fall outside the limits imposed by experimental error. The aliphatic UCM was quantified by reference to a calibration curve constructed from the FID responses of known concentrations of UCM, For this an automated time slice area integral measurement was used which at the same time provided a measure of the percentage resolved alkanes vs percentage unresolved alkanes (Fig. 1). Replicate analyses ( x 10) indicated a GC precision of 5.4%. Daily monitoring of detector response variations showed that during the experiment integrated standard UCM areas did not fall outside the experimentally determined error limits. RESULTS AND DISCUSSION The daily concentration of individual synthetic hydrocarbons expressed as a percentage of their starting concentrations is presented in Table 2. Also presented are changes in the concentration of the aliphatic UCM monitored over the same period, and daily variations in the percentage resolved alkanes vs percentage unresolved alkanes obtained by the time slice area measurement.
Table 1. Composition and concentrationat day 0 of the syntheticbiodegradationmixture Compound Concentrationat day 0 No. Compound (#gcm- 3) 1 2 3 4 5 6 7 8
2,6,10,14-Tetramethyl-7-(3-methylpcntyl)-pcntadecane 2,6,10,14,18-Pentamethyleicosane 7-n -Hexylnonadccane 9-(2-Cyclohexylethyl)-heptadccane 9-(2-Phenylethyl)-heptadecane 9-Mcthyltetracosane 2-Methyltetracosane n-Pentacosane
10.4 10.6 10.8 10.0 10.4 10.0 10.8 9.8
Biodegradation studies of UCM of hydrocarbons
19
Table 2. Daily proportions (%) of individual hydrocarbons and aliphatic U C M as a percentage of concentrations at day 0
Day
1
2
3
0 1 2 3 4 5 6 7 8 9 10 11 12 14 16 18 20 25 25sct
100 99 93 96 100 95 89 89 ND~ 93 90 83 93 90 85 75 69 60 98
100 100 94 96 101 96 89 89 ND 93 90 82 92 91 85 74 69 59 99
100 100 94 97 101 97 89 88 ND 94 90 83 94 93 87 77 72 61 100
Compound No.* 4 5 100 101 96 98 101 97 90 88 ND 93 91 83 95 93 89 78 73 62 101
100 100 96 99 101 97 90 88 ND 93 90 83 95 92 86 77 72 61 101
6
7
8
UCM
Resolved (%)
Unresolved (%)
100 82 70 42 33 26 16 15 ND 17 11 6 10 8 10 7 6 5 98
100 75 60 33 27 19 15 13 ND 15 8 5 7 6 9 7 5 5 100
100 75 46 29 26 14 16 14 ND 17 10 5 7 7 9 9 5 5 101
100 100 100 97 93 91 92 88 86 81 84 88 83 86 87 83 76 78 95
9.6 10.5 9.7 10.2 9.7 10.3 10.2 10.2 9.7 10.1 10.4 10.3 10.0 9.0 9.2 9.2 9.2 9.5 10.5
90.4 89.5 90.3 89.8 90.3 89.7 89.8 89.8 90.3 89.9 89.6 89.7 90.0 91.0 90.8 90.8 90.8 90.5 89.5
*Compound identity given in Table 1. tsc, Sterile control. :~ND, Not determined.
Figure 2 shows the gas chromatograms of the model C25 hydrocarbon mixture after 5, 10 and 25 days and the day 25 sterilized control. Figure 3 depicts the changes observed in the concentration of n- and monomethyl branched C25 alkanes during the course of the study. It is evident that under the conditions used, the normal and monomethyl substituted alkanes were rapidly degraded within the first 10 days in the approximate order n-C25 > 2-methyltetracosane > 9-methyltetracosane. These changes could be attributed to microbial alteration, since the sterilized control (Fig. 2) revealed little reduction in the concentrations of alkanes caused by abiotic factors (e.g. evaporation). After approx. 14 days the rates of degradation of these compounds began to level off (Fig. 3). In contrast the remaining hydrocarbons in the mixture were degraded at a relatively slow rate up to day 14. Thus, Fig. 4 depicts the observed changes in daily concentrations for the proposed UCM hydrocarbons (3,4,5) compared to the changes observed for the aliphatic UCM. The behaviour of the three candidate hydrocarbons was similar throughout the 25 day period as shown by coplotting the day 0--day 25 data sets (r = 0.997). Up to day 14 a relatively slow rate of decrease was observed. This was approximated as linear and regression of the 7-n-hexylnonadecane data points (day 0--day 14) provided a measure of the rate of decrease as - 0 . 7 4 % day -~ (r = 0.619). After 14 days the rate of decrease of the candidate UCM hydrocarbons was more rapid, and linear regression of the 7-n-hexylnonadecane data points (days 14, 16, 18, 20 and 25) provided a daily rate decrease of - 2 . 9 4 % day -~ (r =0.985). It is interesting to note that after day 14 the rate of decrease of the n- and monomethyl alkanes (Fig. 3) slowed and became near linear (r = 0.652) with a rate decrease of - 0 . 3 % day -1. It appears that at this
stage in the biodegradation sequence the bacteria began to utilize hydrocarbons (3-5) in preference to the small remaining amounts of the initially more labile n- and monomethyl C2s alkanes. Over the first 14 days the aliphatic UCM was also observed to decrease in concentration (Fig. 4) ( - 1.36% day-l; r = 0.890), i.e. slightly greater than that observed for the candidate UCM hydrocarbons (3-5). Thereafter, however, the rate of decrease did not parallel that observed for the model UCM hydrocarbons, and remained fairly constant ( - 0 . 9 3 % day- 1, r = 0.807). An examination of the values of percentage resolved alkanes vs percentage unresolved alkanes provided by the time slice area measurement (Table 1) did not show any increase in the proportion of unresolved alkanes within the limits of experimental error ( + 8%). It appears therefore that the UCM was degraded as a whole and that individual resolved alkanes were degraded at the same rate as unresolved components. Also of interest was the behaviour of two of the candidate aliphatic UCM alkanes (3,4) compared to the acyclic isoprenoid alkanes. Though it is generally recognized that acyclic isoprenoid alkanes are relatively resistant to microbial degradation compared to n- and monomethyl alkanes, certain isoprenoids have been shown to be degraded by both pure and mixed cultures (e.g. McKenna, 1971; Cox et al., 1974; Pirnik et al., 1974; Rontani et al., 1986). In particular the regular acyclic isoprenoids pristane (C~9) and phytane (C20) have been observed to degrade before any observable change in the aliphatic UCM profile (Deroo et al., 1974, 1977; Connan, 1984). Therefore, if the monoalkyl substituted acyclic (3) and monocyclic (4) alkanes are good models for the alkane components of the UCM, they should be at least as resistant, and possibly more resistant, to microbial degradation than the acyclic isoprenoid alkanes (1,2).
20
M.A. GOUGH et al.
In fact, all three branched alkanes were observed to degrade at a relatively slow rate up to day 14 followed by a more rapid decrease up to day 25.
A
3
4
Though a slightly greater resistance was noted for the monoalkyl "T"-branched alkane, the difference did not exceed the limits imposed by experimental error. In fact, the data sets for the regular acyclic isoprenoid alkane and the "T"-branched alkane covaried (r = 0.997). No appreciable difference was therefore noted for these two alkane types.
CONCLUSIONS
(1) Laboratory biodegradation of a mixture of
eight C25 hydrocarbons with the aerobe Pseudomonas ~7
I L IlL D
Fig. 2. Partial gas chromatograms of synthetic C25 hydrocarbon mixture at (A) day 5, (B) day 14, (C) day 25 and (D) day 25 sterilized control (for peak identity see Table 1 and Fig. 5).
fluorescens showed that the rate and extent of degradation was influenced by molecular structure. (2) The normal and monomethyl alkanes were rapidly degraded within the first 10 days of the experiment in the approximate order n-pentacosane > 2-methyltetracosane > 9-methyltetracosane. This is in accordance with previous studies of alkane biodegradation, both in the laboratory and from field observations (e.g. Thijsse and Van der Linden, 1961; McKenna and Kallio, 1964; Pirnik et al., 1974; Connan et al., 1980). (3) The remaining synthetic hydrocarbons were also observed to degrade but to a limited extent in comparison to normal and monomethyl alkanes. A relatively slow decrease was noted up to day 14 (ca - 0 . 7 % d a y ~), followed by a more rapid decrease in concentration (ca - 3 % day -1) to a value of ca 60% of the starting material. (4) The aliphatic UCM was also observed to partially degrade in the biodegradation experiment, initially at a comparable rate to that observed for the candidate UCM alkanes (ca 1% d a y -~, up to day 14). Thereafter however the rate of decrease remained essentially constant, whereas the candidate UCM alkanes were observed to degrade more rapidly (ca 3% day-~). Of particular interest was the observation that the UCM appeared degraded "as a whole", i.e. no significant reduction in the proportion of resolved vs unresolved alkanes was noted. (5) A significant proportion of the resolved alkanes were identified previously (Gough, 1989) as regular acyclic isoprenoids (C24-C35), If the monoalkyl branched acyclic and monocyclic aikanes are good models for the aliphatic UCM, then these should degrade at a comparable rate to the regular acyclic isoprenoids. This was in fact observed in the degradation of the synthetic hydrocarbon mixture, the concentration of 7-n-hexylnonadecane, 9-(2-cyclohexylethyi)-heptadecane and 2,6,10,14,18-pentamethyl-eicosane covaried throughout the study with a high degree of linearity (r = 0.997). The applicability of these results to sedimentary environments comprising a variety of bacterial strains and with other degradable cosubstrates is difficult to assess. For instance, Connan and Chossan (1990) showed that of 73 pure strains of bacteria assessed for their ability to degrade steranes and
Biodegradation studies of UCM of hydrocarbons
II
110 100
21
11o
t,c
SC
100 90 90 Legend
80
A - o------
70
o
A
.......
9-MT 2-MT n-C25
80
70
.~ 60 .c o
E
50
(X
40
.~ 60 o
E n-
i
40
3o
20
30
Legend
20
o - 7-n-HN & - - - - - - 9-(2-CHE)o ....... 9-(2-PE)-H • - - ' UCM
".\
'.~
10 -O
50
10 I
I
I
L
I
5
10
15
20
25
Biodegrodation t i m e (days)
0
I 5
I 10
I 15
I 20
I 25
Biodegradation time (days)
Fig. 3. Relative biodegradation rates of straight chain and monomethyl C25 alkanes (sc, sterilized control; ©, n-pentacosane; Fq, 2-methyltetracosane; A, 9-methyltetracosane).
Fig. 4. Relative biodegradation rates o f candidate U C M hydrocarbons (sc, sterile control; © , 7-n-hexylnonadecan¢; A , 9-(2-cyclohexylethyl)-heptadccan¢; D , 9-(2-phenylethyl)heptadecane) compared to the aliphatic U C M (I-]).
cyclic triterpanes, only 7 gram-positive strains, belonging to the Nocardia and Arthrobacter genera, were able to produce noticeable effects. This emphasizes the need for continued experimentation with
UCM and model hydrocarbons, under conditions more relevant to natural environments, and with other bacterial strains. Further studies addressing these questions are currently being undertaken.
1
2
5
6
7
8
4 Fig. 5. Structures of model C25 hydrocarbons.
22
M.A. GOUGH et al.
Acknowledgements--We are grateful to Mr A. Walker (Silkolene Lubricants, U.K.) for the lubricating base oil and to Ms J. Eddy (Department of Biological Sciences, Polytechnic South West) for assistance with culturing the bacteria.
REFERENCES
Alexander R., Cumbers M., Kagi R., Offer M. and Taylor R. (1982)Petroleum contamination of Cockburn Sound, Western Australia. Toxicol. Environ. Chem. 5, 251 275. Atlas R. M., Boehm P. D. and Calder J. A. (1981) Chemical and biological weathering of oil from the Amoco Cadiz spillage, within the littoral zone. Estuarine Coastal Shelf Sci. 12, 589-608. Bailey N. J. L., Jobson A. M. and Rogers M. A. (1973) Bacterial degradation of crude oil: comparison of field and experimental data. Chem. Geol. 11, 203-221. Blumer M., Ehrhardt H. and Jones J. H. (1973) The environmental fate of stranded oil. Deep-Sea Res. 20, 239-259. Connan J. (1984) Biodegradation of crude oils in reservoirs. In Advances in Petroleum Geochemistry (Edited by Brooks J. and Welte D.), Vol. 1, pp. 299-336. Academic Press, London. Connan J. and Chosson P. (1990) In vitro biodegradation of steranes and terpanes. Presented at the 4th Workshop on the Chemistry and Analysis of Environmental Hydrocarbons, Strasbourg, France. Connan J., Restle A. and Albrecht P. (1980) Biodegradation of crude oil in the Aquitaine basin. In Advances in Organic Geochemistry 1979(Edited by Douglas A. G. and Maxwell J. R.), pp. 1-17. Pergamon Press, Oxford. Cox R. E., Maxwell J. R., Ackman R. G. and Hooper S. N. (1974) Stereochemical studies of acyclic isoprenoid compounds. IV. Microbial oxidation of 2,6,10,14-tetramethylpentadecane (pristane). Biochem. Biophys. Acta 360, 166-173. Deroo G., Powell T. G., Tissot B. and McCrossan R. G. (1977) The origin and migration of petroleum in the Western Canadian sedimentary basin, Alberta--a geochemical and thermal maturation study. Geol. Surv. Can. Bull. 262. Deroo G., Tissot B., McCrossan R. G. and Per F. (1974) Geochemistry of the heavy oils of Alberta. In Off Sands-Fuel of the Future (Edited by Hills L. V.), Memoir 3, pp. 148-167. Canadian Society of Petroleum Geologists. Eglinton G., Maxwell J. R. and Philp R. P. (1975) Organic geochemistry of sediments from contemporary aquatic environments. In Advances in Organic Geochemisto, 1973 (Edited by Tissot B. and Bienner F.), pp. 941-961. Editions Technip, Paris. Gough M. A. (1989) Characterisation of unresolved complex mixtures of hydrocarbons. Ph. D. thesis, Polytechnic South West, Plymouth, England. Gough M. A. and Rowland S. J. (1990) Characterisation of unresolved complex mixtures of hydrocarbons in petroleum. Nature 344, 648-650. Jones D. M , Rowland S. J., Douglas A. G. and Howells S. (1986) An examination of the fate of Nigerian crude oil
in surface sediments of the Humber Estuary by gas chromatography and gas chromatography-mass spectrometry. Int. J. Environ. Anal. Chem. 2,4, 227-247. Karrick N. L, (1977) Alterations in petroleum resulting from physico-chemical and microbiological factors. In Effects of Petroleum on Arctic and Subarctic Marine Environments and Organisms (Edited by Malins D. C.), Vol. 1, pp. 225-278. Academic Press, New York. Killops S. D. and AI-Juboori M. A. H. A. (1990) Characterisation of the unresolved complex mixture (UCM) in the gas chromatograms of biodegraded petroleums. Org. Geochem. 15, 147-160. McKenna E. J. (1971) Microbial metabolism of normal and branched chain alkanes. In Degradation of Synthetic Organic Molecules in the Biosphere, pp. 73-98. National Academy of Sciences, Washington, D.C. McKenna E. J and Kallio R. E. (1964) Hydrocarbon structure: its effect on bacterial utilisation of alkanes. In Principles and Applications in Aquatic Microbiology (Edited by Heukelekia H. and Dondero D.), pp. I 14. Wiley, New York. Oudet J., Fusey P., Van Praet M., Feral J. P. and Gaill G. ( 1981 ) Hydrocarbon weathering in seashore invertebrates and sediments over a two-year period following the Amoco Cadiz oil spill: influence of microbial metabolism. Environ. Pollut. Set. A 26, 93 110. Pirnik M. P., Atlas R. M. and Bartha R. (1974) Hydrocarbon metabolism by Brevibacterium erythrogenes, normal and branched alkanes. J, Bacteriol. 119, 868 875. Robson J. N. and Rowland S. J. (1989) Biodegradation of highly branched isoprenoid hydrocarbons: a possible explanation of sedimentary abundance. In Advances in Organic Geochemistry 1987 (Edited by Mattivelli L. and Novelli L.). Org. Geochem. 13, 691-695. Pergamon Press, Oxford. Rontani J. F., Bertrand J. C., Blanc F. and Giusti G. (1986) Gas chromatography and gas chromatography-mass spectrometry applied to the determination of a new pathway of pristane degradation by a mixed bacterial population. Mar. Chem. 18, 9-16. Rubinstein I., Strausz O. P., Spyckerelle C., Crawford R. J. and Westlake D. W. S. (1977) The origin of the oil sand bitumens of Alberta: a chemical and microbiological simulation study. Geochim. Cosmochim Acta 41, 1341 1353. Sanders P~ F. and Tibbetts P. J. C. (1987) Effects of discarded drill muds on microbial populations. Philos. Trans. R Soc. London (B) 316, 567 585. Thijsse G. J. E. and Van der Linden A. C. (1961) Iso-alkane oxidation by a pseudomonas. Part I--metabolism of 2-methylhexane. Antonie van Leeuwenhoek J. Microbiol. Serol. 27, 171 ~179. Volkman J. K., Alexander R., Kagi R. 1. and Woodhouse G. W. (1983) Demethylated hopanes in crude oils and their applications in petroleum geochemistry. Geochim. Cosmochim. Acta 47, 785-794. Volkman J. K., Alexander R., Kagi R. I., Rowland S. J. and Sheppard P. N. (1984) Biodegradation of aromatic hydrocarbons in crude oils from the Barrow sub-basin of Western Australia. Org. Geochem. 6, 619--632,