![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Previous Article | Next Article 
Microbiology and Molecular Biology Reviews, December 2003, p. 503-549, Vol. 67, No. 4
1092-2172/03/$08.00+0 DOI: 10.1128/MMBR.67.4.503-549.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Department of Biological Sciences, The University College of the Cariboo, Kamloops, British Columbia V2C 5N3,1 Petrozyme Technologies, Inc., Guelph, Ontario N1H 6H9,2 Department of Biology, University of Waterloo, Waterloo, Ontario N2L 3G1, Canada3
SUMMARY INTRODUCTION METABOLISM Aerobic Alkane Metabolism Aerobic PAH Metabolism Anaerobic Hydrocarbon Metabolism BEHAVIORAL AND PHYSIOLOGICAL RESPONSES TO HYDROCARBONS Membrane Alterations, Uptake, and Efflux Mechanisms of tolerance. Taxis MICROBIAL COMMUNITY DYNAMICS Culture-Based Methods Culture-Independent Approaches MICROBIAL TREATMENT OF PETROLEUM WASTE Treatment of Contaminated Soils and Sludges Factors affecting bioremediation. Passive bioremediation processes. Landfarming of oily wastes. Bioreactor-based processes. Biofiltration of Volatile Organic Compounds Removal of H2S and SOX MICROBIAL PROCESSES FOR RECOVERING AND UPGRADING PETROLEUM Microbial Enhanced Oil Recovery Microbial Deemulsification Microbial Desulfurization Microbial Denitrogenation Enzymatic Upgrading of Petroleum Fractions and Pure Hydrocarbons BACTERIAL BIOSENSORS CONCLUSIONS AND FUTURE PROSPECTS ACKNOWLEDGMENTS REFERENCES
| SUMMARY |
|---|
 Top NextReferences |
|---|
| INTRODUCTION |
|---|
|
TopPreviousNextReferences |
|---|
A wide range of studies have dealt with biotransformation, biodegradation, and bioremediation of petroleum hydrocarbons (30, 31, 48, 415, 490, 523), and interest in exploiting petroleum-degrading organisms for environmental clean-up has become central to petroleum microbiology (29). A common theme of early reviews focused on the examination of factors, including nutrients, physical state of the oil, oxygen, temperature, salinity, and pressure, influencing petroleum biodegradation rates, with a view to developing environmental applications (29). Metabolic studies were implemented on the aerobic pathways for alkane, cycloalkane, and aromatic and polycyclic aromatic hydrocarbon (PAH) biodegradation (103, 104, 294, 301, 479, 572, 596, 656), for transformations of nitrogen and sulfur compounds (55, 74, 75, 299, 352, 417), and, more recently, the microbial mechanisms of anaerobic hydrocarbon catabolism (203, 243, 250, 581, 390, 482, 664).
Most significantly, through the developments and applications of molecular techniques, our understanding of the processes of hydrocarbon catabolism has advanced substantially, and many novel catalytic mechanisms have been characterized. A molecular approach is also contributing to a more detailed characterization of bacterial membrane structure. We are learning a great deal about cellular and other physiological adaptations to the presence of hydrocarbons, as well as the biochemical mechanisms involved in hydrocarbon accession and uptake (143, 251, 566). The use of genetically engineered microbes for bioremediation has also been considered (210).
The vast range of substrates and metabolites present in hydrocarbon-impacted soils surely provides an environment for the development of a quite complex microbial community. Culture-based methods and culture-independent methods are being developed and implemented to improve our understanding of these microbial communities. Isolating and identifying microorganisms responsible for hydrocarbon transformations have long been recognized as important from a fundamental and applied viewpoint, and lists of hydrocarbon-degrading organisms (bacteria, yeasts, fungi, and algae) are available (30, 33, 366, 522). Leahy and Colwell (366) discussed colony hybridization and dot blot assays in their review and cited molecular tools as revolutionary for describing microbial communities. Magot et al. (398) recently reviewed the current state of knowledge of microorganisms from petroleum reservoirs, including mesophilic and thermophilic sulfate-reducing bacteria, methanogens, mesophilic and thermophilic fermentative bacteria, and iron-reducing bacteria. Again, molecular tools were called upon to provide more detailed community characterizations. These and related studies should provide us with new information on the long-term ecological effects of petroleum pollution and give us directions, for example, regarding the development of new remedial approaches and methods to control some of the deleterious microbial activities occurring during petroleum production.
Current applied research on petroleum microbiology encompasses oil spill remediation (490, 492, 598), fermentor- and wetland-based hydrocarbon treatment (212, 281, 336, 530, 569), biofiltration of volatile hydrocarbons (176), microbial enhanced oil recovery (42, 153), oil and fuel upgrading through desulfurization (417, 554) and denitrogenation (55), coal processing (102), fine-chemical production (412, 415), and microbial community-based site assessment (394). The roles and practical applications of chemical and biological surfactants have been widely reviewed (260, 454, 529, 643).
Oil spill treatment on shorelines and problems associated with open-ocean remediation have been discussed through case histories in numerous reviews (30, 31, 44, 489, 599). Other practical applications include land- and reactor-based refinery waste treatment, in situ tanker ballast cleaning, and subsurface remediation (31, 44).
Heavy crude oil recovery, facilitated by microorganisms, was suggested in the 1920s and received growing interest in the 1980s as microbial enhanced oil recovery (153). As of 1998, only one productive microbial enhanced oil recovery project was being carried out in the United States (613), although in situ biosurfactant and biopolymer applications continue to garner interest (42).
A limited number of studies have been carried out on biological methods of removing heavy metals such as nickel and vanadium from petroleum distillate fractions, coal-derived liquid shale, bitumens, tars, and synthetic fuels (188, 429, 487, 488, 673). In one approach, cytochrome c reductase and chloroperoxidase enzymes have shown potential for metal removal from petroleum fractions. However, further characterization on the biochemical mechanisms and bioprocessing issues involved in heavy metal removal are required in order to develop a reliable biological process.
Bacteria with selected petroleum-metabolizing enzymes amenable to being linked to electronic interfaces are being engineered and developed as biosensors (142). These systems have applications in monitoring environmental contaminant concentrations and toxicities during implementation of remedial processes and also have potential applications in control of environmental processes.
This review deals with developments in our knowledge of petroleum microbiology and in the application of microorganisms in oil bioprocesses and as biosensors. Advances in our understanding of microbial catabolism are presented, including an evaluation of the biochemical mechanisms that control microbial responses to hydrocarbon substrates. These aspects include changes in membrane architecture, active uptake and efflux of hydrocarbons and chemotaxis, and the potential for coordinate control of some of these systems to allow metabolism to take place. Developments in oil bioprocessing focus on transformation of wastes and on the production and upgrading of petroleum and petrochemicals, with emphasis placed on maximizing the rates and extents of microbial growth, hydrocarbon accession, and transformation. Sections dealing with desulfurization and fine-chemical synthesis additionally illustrate the potential benefits of recombinant strains containing enzymes with enhanced activity and/or altered substrate specificity. The possible use of biosensors for online monitoring of pollutants is also addressed.
| METABOLISM |
|---|
|
TopPreviousNextReferences |
|---|
From a regulatory genetic standpoint, the most extensively characterized alkane degradation pathway is encoded by the OCT plasmid carried by Pseudomonas putida Gpo1 (formerly Pseudomonas oleovorans) (626, 627). Here, a membrane-bound monooxygenase and soluble rubredoxin and rubredoxin reductase serve to shunt electrons through NADH to the hydroxylase for conversion of an alkane into an alcohol. The alcohol can be further oxidized to an aldehyde and acid prior to proceeding into the ß-oxidation and tricarboxylic acid cycles. Recently, van Beilen et al. (626, 627) studied the OCT plasmid, while Canosa et al. (98) and Panake et al. (470) examined expression of the AlkS regulator, and Yuste et al. (683, 684) studied the catabolite repression system.
A model for alkane metabolism, including the locations of the Alk proteins and regulation of the alk genes, is shown in Fig. 1 (627). Here, the alkBFGHJKL operon encodes the enzymes necessary for converting alkanes into acetyl-coenzyme A (CoA), while alkST encode a rubredoxin reductase (AlkT) and the positive regulator for the alkBFGHJKL operon (AlkS). These two operons are located end to end, separated by 9.7 kb of DNA, within which lies alkN, a gene coding for a methyl-accepting transducer protein that may be involved in alkane chemotaxis. Note that of all the genes described, the function of alkL remains unknown, although it is suspected to be involved in transport. Comparative analysis of insertion sequences in P. putida P1 and the previous observation that the G+C content of the alk genes is lower than that of both the host strain and the OCT plasmid suggest that the genes are part of an integrated mobile element. Two other plasmid systems have been partially characterized: the OCT plasmid in Pseudomonas maltophilia has an alkA gene distinct from that of P. putida (374), and the unique pDEC plasmid in Pseudomonas sp. strain C12B (347).
|
The alkM, rubA, and rubB genes in Acinetobacter sp. strain M1 are homologous to those in Acinetobacter sp. strain ADP1. Interestingly, two alkane hydroxylase complexes (alkMa and alkMb) whose expression is controlled by n-alkane chain length are present in this strain. Conversely, the rubredoxin and rubredoxin reductase are constitutively expressed. Hydropathy plots of AlkMa and AlkMb suggest that the proteins are similar to AlkB in P. putida in that they are membrane bound. AlkMa appears to be similar to AlkM of Acinetobacter sp. strain ADP1. The first of two transcriptional regulators in Acinetobacter sp. strain M1 (AlkRa) is related to AraC-XylS type regulators, which includes that of Acinetobacter sp. strain ADP1. The second regulator (AlkRb) is similar to OruR of P. aeruginosa. The two regulators are induced by different n-alkanes in this strain. alkMa responds to solid, long-chain alkanes (>C22), while alkMb responds to liquid alkanes (C16 to C22). Unlike the case in P. putida, neither acetate nor hexadecanol induces alkMa and alkMb (602).
The presence of multiple alkane hydroxylase genes in a single strain does not appear to be a unique phenomenon. Two distinct monooxygenases, a Cu-containing monooxygenase and an integral-membrane, binuclear-iron monooxygenase similar to that of P. putida GMo1 have been described in Nocardiodes sp. strain CF8 (233). While the Cu-containing monooxygenase is expressed in response to a wide range of alkanes, only those with more than six carbons induce the binuclear-iron monooxygenase. Once again, the genes encoding alkane metabolism in Acinetobacter sp. strain M1 and Nocardiodes sp. strain CF8 are not clustered together as in the OCT plasmid (275, 602). Other enzymes involved in Acinetobacter sp. strain M1 alkane metabolism have been characterized. Ishige et al. (275) isolated a soluble long-chain NAD+-dependent aldehyde dehydrogenase whose activity increased with increasing aldehyde chain length (tetradecanal preferred) that is encoded by the chromosomal ald1 gene. This enzyme plays a role in both alkane degradation and biosynthesis, depending on the conditions. The NAD+-dependent aldehyde dehydrogenase in strain HD1 is also reported to prefer long-chain aldehydes (462). A thermostable NADP+-dependent medium-chain alcohol dehydrogenase, encoded by alrA, has also been isolated but is not believed to participate in the main alkane oxidation pathway due to its cytosolic location and greater activity towards medium-chain alcohols (603).
Despite the importance of alkane degradation systems, little information is available for pathways other than the aerobic monooxygenase-mediated pathway found on the OCT plasmid. Evidence for the Finnerty pathway, where a dioxygenase converts alkanes to aldehydes through n-alkyl hydroperoxides without an alcohol intermediate, has been described for Acinetobacter sp. strain M1 (397, 534). The dioxygenase requires molecular oxygen to catalyze the oxidation of n-alkanes (C10 to C30) and alkenes (C12 to C20) without the production of oxygen radicals. A flavin adenine dinucleotide chromophore was detected, and the enzyme is thought to contain Cu2+. Unlike the case for the 1-monooxygenase in P. putida, rubredoxin and NAD(P)H are not required.
Another novel metabolic pathway has been observed in a Rhodococcus mutant (338). In this case, aliphatics are cis-desaturated, producing products with double bonds mainly at the ninth carbon from the terminal methyl group. It is postulated that a coenzyme A-independent cis-desaturase may be involved in this activity. Dutta and Harayama (159) recently noted that the degradation of the long side chains of n-alkylbenzenes and n-alkylcyclohexanes by Alcanivorax sp. strain MBIC 4326 proceeds mainly by ß-oxidation (Fig. 2). However, minor products suggest the possibility of other degradative routes. For example, 4-cyclohexylbutanoic acid was metabolized through 4-cyclohexyl-2-butenoic acid (ß-oxidation) and other intermediates not believed to be formed by ß-oxidation (4-cyclohexyl-3-butenoic acid and cyclohexylcarboxylic acid).
|
The low water solubility and high sorbtion capacity of PAHs are often found to greatly influence biodegradation, but other factors, including production of toxic or dead-end metabolites, metabolite repression, the presence of preferred substrates, and the lack of cometabolic or inducer substrates, must be considered when PAH persistence is evident (433, 295). Understanding how these factors affect the transformation of and determining any given PAH is difficult; understanding the processes in natural environments when mixtures of PAHs and their myriad metabolites are present is more difficult, especially as the majority of work has focused on a narrow selection of species. Indeed, the cited reviews generally conclude by calling for more study into the regulation of PAH biodegradation, biodegradation of PAH mixtures, and interactions within microbial consortia.
Until recently, the majority of information on the genetics of PAH metabolism has come from studying naphthalene catabolic plasmids such as NAH7 from Pseudomonas putida strain G7. In this well-characterized system, the first operon (nahAaAbAcAdBFCED) encodes the pathway for naphthalene conversion to salicylate (upper pathway), and the second (nahGTHINLOMKJ) codes for the conversion of salicylate via catechol meta-cleavage to acetaldehyde and pyruvate (lower pathway) (164, 485, 568, 679). The regulator for both operons is encoded by a third operon containing nahR, which is induced by salicylate (547). Here, molecular oxygen is introduced into the aromatic nucleus via naphthalene dioxygenase, a multicomponent nonheme iron oxygenase enzyme system consisting of a reductase, a putative Rieske [2Fe-2S] iron sulfur center in a ferredoxin, and an iron-sulfur flavoprotein. The initial reaction results in the formation of cis-naphthalene dihydrodiol, which is subsequently converted to salicylate and then to tricarboxylic acid intermediates (for more detail, see references 104, 220, and 679). As will be discussed below, naphthalene dioxygenase is now known to be a versatile enzyme, able to catalyze a wide variety of reactions. Molecular and biochemical evidence that the naphthalene plasmid degradative enzyme system could mineralize other PAHs, such as phenanthrene and anthracene, was first provided by two research groups in 1993 (423, 540).
As more PAH-degrading bacteria were isolated and characterized, and as molecular methods to study microbial communities developed, the diversity of PAH metabolic genes was discovered. Examples of bacteria with unknown, nonhomologous genes to the naphthalene NAH7-like catabolic plasmids have been reported recently (318, 528). At the same time, a variety of new isofunctional gene sequences have been reported in different bacterial species, most notably in Nocardia, Rhodococcus, and Mycobacterium spp., some of which are capable of using high-molecular-weight PAHs such as pyrene as carbon and energy sources (Table 1).
|
90%) of homology and a conserved gene arrangement are observed in the nah, ndo, pah, and dox sequences (63, 64, 147, 333, 355, 601). In fact, it has been proposed that the dox plasmid, which encodes a dibenzothiophene (DBT) metabolic pathway analogous to the naphthalene catabolic pathway, may in fact be a naphthalene catabolic plasmid (163). High homology, however, does not necessarily translate to similar substrate specificities, as Yang et al. (677) found that P. putida NCIB 9816 clones were able to produce metabolites from naphthalene, fluorine, and phenanthrene, while only naphthalene metabolites were detected from a homologous NAH7 clone. In addition, with respect to PAH metabolism, novel gene sequences and gene orders have been observed in a variety of strains, including Burkholderia sp. strain RP007, phnFECDAcAdB (364); Pseudomonas sp. strain U2, nagAaGHAbAcAdBF (205); Rhodococcus sp. strain I24, nidABCD (615); Mycobacterium sp. strain PYR1, nidDBA (318); and Nocardiodes sp. strain KP7, phdABCD (542). Sequence diversity, and the fact that naphthalene catabolic genes have now been found on the chromosome as well as on plasmids indicate that lateral gene transfer and genetic recombination may have played an important role in the development of these versatile metabolic pathways (63, 64, 205, 364, 542). For example, the phn locus has similarities to both nah and bph genes in Burkholderia sp. strain RP007 (364), while the chromosomally encoded nah upper and lower pathways in Pseudomonas stutzeri AN10 appear to have been recruited from other organisms and recombined. In fact, two entire nah upper pathways may exist in this strain (63, 64).
Thus, not only are new gene sequences being found for PAH metabolism, but strains possessing multiple genes for similar enzymes are being detected. Ferrero et al. (189) recently showed, while studying Pseudomonas spp. isolated from the western Mediterranean, that single strains can have two distinct nahAc-like genes as well as other genes of the upper nah pathway. With respect to the lower pathway, Bosch et al. (63) found two distinct genes for salicylate 1-hydroxylase, the flavoprotein monooxygenase that converts salicylate to catechol, on the chromosome of P. stutzeri AN10. While the nahG gene was found in the meta-cleavage pathway transcriptional unit, the novel nahW was found close to but outside of this unit. Both are induced upon exposure to salicylate and have broad substrate specificities, but nahW is missing the conserved flavin adenine dinucleotidebinding site (GxGxxG) normally found in these hydroxylases. This is the first example of two isofunctional salicylate hydroxylases in one strain, and it will be interesting to discover if the combination of genes from various catabolic routes is a widespread phenomenon.
This type of metabolic expansionism is exemplified by Sphingomonas yanoikuyae B1, which has recruited, modified, and reorganized genes to obtain catabolic pathways for naphthalene, phenanthrene, anthracene, biphenyl, toluene, and m- and p-xylene. In this case, nah, bph, and xyl genes are present but are not arranged in three distinct operons (215, 330, 692). Indeed, this gene clustering may be typical of Sphingomonas spp. capable of degrading aromatic compounds. Romine et al. (519, 520) sequenced the pNL1 (184 kb) plasmid of Sphingomonas aromaticivorans F199, which is capable of degrading toluene, xylenes, salicylate, biphenyl, dibenzothiophene, flourene, and benzoate. In this plasmid, at least 13 gene clusters are predicted to encode all of the necessary enzymes. In addition, seven three-component oxygenases with components spread over six gene clusters have been predicted.
Beyond the genes known to participate directly in PAH metabolism, genes that may provide important support functions are being described. Sphingomonas paucimobilis var. EPA500, a strain able to use fluoranthene, naphthalene, and phenanthrene as sole carbon and energy sources, has pbhD, a gene encoding pyruvate phosphate dikinase homologous to ppdK that is known to be involved in glucose uptake in prokaryotes and plants. If pbhD is disrupted, fluoranthene metabolism is interrupted. While the gene function is not clear, it is possible that it is involved in the uptake of fluoranthene catabolites that leak from the cell (587). Another example is the katG gene in Mycobacterium sp. strain PYR-1, which encodes an 81-kDa catalase-peroxidase induced upon exposure to pyrene (651). This enzyme may protect the dioxygenase from oxidative inactivation by exogenous oxidation or by removing H2O2 generated endogenously during PAH metabolism (375, 426, 651). Grimm and Harwood (226, 227) recently found nahY on the NAH7 catabolic plasmid of P. putida G7, which encodes a membrane protein that may be a chemoreceptor for naphthalene or naphthalene metabolites.
In order to move towards a better understanding of the diversity of PAH metabolism in the ecosystem, research should be directed towards genera other than mesophilic pseudomonads. This will allow a variety of research questions to be addressed: what impact different genera have on PAH metabolism in the evironment; what and how pathways should be encouraged in active bioremedation systems; and what relationship exists between ecosystem properties and PAH metabolism.
To start, synergistic and antagonistic interactions between PAHs of both high and low molecular weights are being investigated. For example, Molina et al. (433) observed that, for both a mixed culture and Mycobacterium sp. strain M1, cross-acclimation occurred between phenanthrene and pyrene metabolism in that pyrene-grown cells did not require new protein synthesis to degrade phenanthrene. On the other hand, neither naphthalene nor anthracene resulted in induction or inhibition of pyrene mineralization. Samanta et al. (537) found that phenanthrene mineralization increased in two strains when fluorine, fluoranthene, and pyrene mixtures were added, while mineralization was not affected in two other strains. In this case, a consortium of the four strains did not enhance phenanthrene mineralization, as has been observed in other studies with defined bacterial and bacterial-fungal consortia (61, 67, 101, 616).
Inhibition may also occur, presumably due to competition for enzymes involved in oxidation or transport, accumulation of by-products resulting in cytotoxicity, and blockage of enzyme induction (66, 295, 590, 564). Determining which mechanism is important in any given situation can be complicated by the presence of metabolites from the different PAHs. The pyrene metabolite cis-4,5-dihydro-4,5-dihydroxypyrene inhibited phenathrene metablism in Pseudomonas saccharophila strain P15 and Sphingomonas yanoikuyae R1 but had little effect on Pseudomonas stutzeri P16 and Bacillus cereus P21 (313). In addition, the above metabolite and its oxidation product, pyrene-4,5-dione, inhibited benzo[a]pyrene mineralization in the sensitive strains. In a follow-up study, the strains were found to form the dead-end product fluoranthene-2,3-dione as a cometabolic product of flouranthene when grown on phenanthrene. Phenanthrene removal was inhibited by this metabolite in Sphingomonas sp. strain R1 but not in the three other strains studied. Mineralization of benz[a]anthracene, benzo[a]pyrene, and chrysene was also inhibited in R1, while only benzo[a]pyrene metabolism in P15 was affected. Cytotoxicity was partly responsible for the observed inhibition (314). Thus, depending on the strains, transformation products from one PAH may affect the removal of other PAHs (295, 112). Overall, induction effects in complex mixtures may be as important as diauxic effects (49, 304, 305, 418).
Understanding how a metabolite may interact with a specific receptor or enzyme requires knowledge of what metabolites are formed and how persistent they are in the environment. Indeed, the number of known metabolites from both low- and high-molecular-weight PAHs is increasing as more researchers apply techniques such as high-resolution gas chromatography-mass spectroscopy and nuclear magnetic resonance in their studies. Recent studies with members of the mycobacteria, ubiquitous soil microorganisms with versatile metabolic abilities, illustrate the diversity of PAH metabolic pathways.
For example, Grund et al. (230) noted that Rhodococcus sp. strain B4, whose naphthalene metabolic pathway was not induced by salicylate, the normal inducer of the NAH7 pathway, oxidized salicylate to gentisate rather than catechol. More recently, Dean-Ross et al. (144) described a Rhodococcus sp. that metabolizes anthracene to 1,2-dihydroxyanthracene and then to either 3-(2-carboxyvinyl)naphthalene-2-carboxylic acid or 6,7-benzocoumarin. The second product is from the meta-cleavage pathway found in both gram-positive and gram-negative bacteria, while the first product is from a novel ortho-pathway, to date only identified in gram-positives (22, 437, 641). In gram-negatives, novel metabolic pathways for low-molecular-weight PAHs, such as phenanthrene and fluorene, have been recently described as well (100, 537).
The number of strains known to utilize four-ring PAHs as sole carbon and energy sources, even in the absence of cofactors or surfactants, and those known to cometabolize PAHs with more than four rings has increased greatly in the last 10 years. Along with this, a myriad of metabolic pathways have been proposed, as documented by Kanaly and Harayama (301) for a variety of high-molecular-weight PAHs in bacteria, and by Juhasz and Naidu (294), who focused on microbial metabolism of benzo[a]pyrene. In the short time since these reviews appeared, more examples of novel metabolic pathways and cooxidation products have been described. For example, Rehmann et al. (507) outlined a new pathway for fluoranthene metabolism in Mycobacterium sp. strain KR20, whereby initial dioxygenation commences at the 2,3 position (Fig. 3). Kazunga et al. (314) identified fluoranthene-2,3-dione and fluroanthene-1,5-dione as dead-end metabolites from fluoranthene during growth on phenanthrene in Pseudomonas saccharophila strain P15, Sphingomonas yanoikuyae strain R1, Pseudomonas stutzeri P16, and Bacillus cereus strain P2. These metabolites are not likely to be intermediates of fluoranthene metabolism, but instead are probably autooxidation products of the corresponding o-dihydroxy metabolites.
|
Overall, the broad PAH-degrading capabilities in many strains may be attributed to relaxed initial enzyme specificity for PAHs (low and high molecular weight and methyl substituted), the presence of multiple oxygenases, and the presence of multiple metabolic pathways or multiple genes for isofunctional pathways (83, 112, 160, 249, 220, 330, 396, 399, 418, 437, 519, 520, 532, 641, 677). Finally, the presence of both alkane and aromatic compound-degrading genes within single strains appears to be common (120, 301, 576, 578, 641, 662).
How these various metabolic routes are controlled at the genetic level and how they compete for a substrate is still a major question. This is especially evident when novel dead-end metabolites, such as the methoxylated 1-methoxy-2-hydroxyanthracene from anthracene metabolism (641) and the dicarboxylic acid 6,6'-dihydroxy-2,2'-biphenyl dicarboxylic acid from pyrene metabolism (437), are detected with strains simultaneously employing multiple degradative routes for a single substrate. This is also the case in strains that have degradative pathways for multiple aromatic substrates (588, 519). For example, in Sphingomonas aromaticivorans strain F199, induction studies have indicated that naphthalene and toluene mineralization may be higher in the presence of both substrates, as greater gene expression can be achieved (519).
Work with microbial consortia in the field, in enrichment cultures, and in microcosms has illustrated that hydrocarbons such as toluene (171, 358), alkylbenzenes including m-, o-, and p-xylene and trimethylbenzenes (39, 111, 235, 481), benzene (90, 312, 521), naphthalene and phenanthrene (50, 124, 421, 686), methylnaphthalene and tetralin (20, 23), >C6 n-alkanes (18, 96, 168, 575), branched alkanes (72, 73), and hydrocarbon mixtures (228) can be metabolized under anaerobic conditions. These reactions may take place under Fe(III)-reducing, denitrifying, and sulfate-reducing conditions, by anoxygenic photosynthetic bacteria, or in syntrophic consortia of proton-reducing and methanogenic bacteria. Other terminal electron acceptors shown to be used during anaerobic hydrocarbon metabolism include manganese oxides (357, 358), soil humic acids and the humic acid model compound anthraquinone-2,6-disulfonate (105), and fumarate in a fermentative oxidation process (420). Mixed-culture work continues as enhanced bioremediation strategies are tested (17, 530) and new metabolites are described (23, 172, 421, 687).
More recently, the number of pure cultures shown to metabolize various hydrocarbons with different electron acceptors has increased (Table 2). This diverse set of bacteria (no fungi have been studied to date), including members of the 
-, ß-, 
-, and 
-subclasses of the proteobacteria, form an excellent framework from which to elucidate the underlying biochemical and molecular mechanisms driving anaerobic hydrocarbon metabolism.
|
Benzylsuccinate synthase has been purified from Azoarcus sp. strain T and T. aromatica strain K172 and is characterized as a 2ß22 heterohexamer with a flavin cofactor but no iron-sulfur clusters (54, 378) and represents a new class of glycyl radical-containing enzymes (350). Succinyl-CoA:(R)-benzylsuccinate CoA-transferase, which activates (R)-benzylsuccinate to 2-(R)-benzylsuccinyl-CoA, has also been purified from strain Thauera aromatica K172 (380).
The genes encoding benzylsuccinate synthase have been cloned and sequenced in Azoarcus sp. strain T (2), T. aromatica strain K172 (378), and T. aromatica strain T1 (135, 136, 137, 378). In strain T. aromatica K172, the bbs (beta-oxidation of benzylsuccinate) operon contains bbsDCABE, with bbsCAB encoding the , , and ß subunits of benzylsuccinate synthase, a region with significant homology to the tutFDG genes in strain T1 (136, 255, 378). The genes encoding the putative activating enzyme (bssD and tutE) are found upstream and also show homology in the two strains. BssE in K172 may be an ATP-dependent chaperone for assembly or deactivation of benzylsuccinate synthase (255). In contrast to K172 and T1, strain T mineralizes both toluene and m-xylene. In this case, expression of the bssDCABE operon is required for growth on both substrates (2).
Similar operons may be present in other strains, as the novel benzylsuccinate synthase reaction, catalyzing the addition of fumarate to toluene (110, 181), may also be involved in the metabolism of xylenes (349, 444, 445), alkylnaphthalenes (20, 23), n-hexadecane (497), and n-dodecane (351). For example, dodecylsuccinic acids were detected from a sulfate-reducing enrichment culture growing on n-dodecane (351), and an n-hexane-utilizing denitrifying bacterium with a protein similar to BssC has been isolated from the toluene-degrading denitrifying bacteria (664). In addition, the metabolites (1-methylpentyl)succinate and (1-ethylbenzyl)succinate from the anaerobic metabolism of n-hexane by a denitrifying strain indicate a C-2 and a C-3 addition of fumarate, analogous to the toluene activation reaction (497). The (1-methylpentyl)succinate is then converted to a CoA thioester prior to rearrangement to (2-methylhexyl)malonyl-CoA and degradation by conventional ß-oxidation (666). Thus, it appears that the fate of the alkylsuccinates produced is probably fatty acid metabolism (5, 574, 666).
For ethylbenzene, oxidation under denitrifying conditions appears to commence with a dehydrogenation by ethylbenzene dehydrogenase to produce 1-phenylethanol followed by oxidation to acetophenone (39, 108, 291, 495, 496). Ethylbenzene dehydrogenase has been isolated from both Azoarcus sp. strains EB1 (292) and EbN1 (335). In both cases, the enzyme is an ß-Mo-Fe-S heterotrimer. Johnson et al. (292) sequenced ebdA, encoding the -subunit containing a molybdopterin-binding domain; ebdB, encoding the ß-subunit containing several 4Fe-4S binding domains; and ebdC, encoding the -subunit, a potential membrane anchor subunit. Kniemeyer and Heider (334) isolated the NAD+-dependent secondary alcohol dehydrogenase (S)-1-phenylethanol dehydrogenase, which catalyzes acetophenone formation in Azoarcus sp. strain EbN1. Analogous reactions are believed to occur for n-propylbenzene (495), while for sulfate-reducing bacteria the metabolic pathway may be similar to that of toluene metabolism, as (1-phenylethyl)succinate has been detected in enrichment cultures (172). It is of interest that Azoarcus sp. strain EbN1 also degrades toluene, but via benzylsuccinate (496).
Two- and three-ring PAHs may also be metabolized under anaerobic conditions. For naphthalene, activation proceeds via carboxylation to form 2-naphthoate in sulfate-reducing (208, 438) and denitrifying (517) bacteria. Carboxylation has also been observed for phenanthrene added to a sulfidogenic culture (686). Alkylnaphthalenes appear to be activated by a mechanism similar to that of toluene, as naphthyl-2-methylsuccinate has been detected in sulfate-reducing enrichment cultures exposed to 2-methylnaphthalene (20).
Recently, Annweiler et al. (23) proposed that, with a sulfate-reducing enrichment culture, naphthalene, 2-methylnaphthalene, and tetralin (1,2,3,4-tetrahydronaphthalene) are all degraded, with 2-naphthoic acid being the central intermediate in a pathway analogous to the benzyl-CoA pathway for monoaromatic compounds. Further degradation occurs through saturated compounds with cyclohexane ring structures (also see 687). They have also found that a sulfate-reducing enrichment culture cometabolized benzothiophene when grown with naphthalene. While activity was not very high, perhaps because of inhibition, toxicity of benzothiophene or metabolites, or benzothiophene being a poor substrate, the products formed (2- and 5-carboxybenzothiophene) indicated that the initial enzyme could nonspecifically attack either the benzene or thiophene ring. As for naphthalene, the C1 unit was derived from bicarbonate, as revealed in 13C radiolabeling experiments (22). In similar experiments with [13C]bicarbonate and 2-[14C]methylnaphthalene, the formation of 2-naphthoic acid via methyl group oxidation was observed in a sulfate-reducing consortium. Also, the presence of 2-methynaphthalenes suggests an alternative metabolic pathway (594).
To date, the mechanism of benzene activation leading to its anaerobic degradation has not been elucidated because no pure cultures have yet been isolated for study. Recently, two Dechloromonas strains (RCB and JJ) of the ß-proteobacteria that mineralize benzene with nitrate as the electron acceptor have been isolated (123), and elucidating the genetics and biochemistry of this metabolism is an area that deserves attention.
The diversity and unique properties of the anaerobic hydrocarbon-utilizing bacteria are areas that are in need of more work. While difficult, greater focus on isolating and characterizing the enzymes involved in anaerobic hydrocarbon metabolism is required. Futhermore, uptake, efflux, and chemotaxis, areas only recently explored for aerobes, are topics so far untouched in the anaerobic realm. A balanced shift from molecular biology back to enzymology and protein biochemistry is a move that would benefit the understanding of hydrocarbon metabolism in all areas.
| BEHAVIORAL AND PHYSIOLOGICAL RESPONSES TO HYDROCARBONS |
|---|
|
TopPreviousNextReferences |
|---|
In terms of general stress responses, bacteria may form biofilms, alter their cell surface hydrophobicity to regulate their partitioning with respect to hydrocarbon-water interfaces or, in gram-negative bacteria, gain protection from hydrophilic lipopolysaccharide components that offer high transfer resistance to lipophilic compounds. In addition, energy-dependent repair mechanisms may be used to compensate for losses in membrane integrity resulting from the partitioning of lipophilic compounds. For example, membrane fluidity can be decreased through increased membrane ordering by affecting cis/trans phospolipid isomerizations, by decreasing unsaturated fatty acid content, and by altering phospholipid head groups (297, 501, 566, 617, 659). These changes may be associated with an overall increase in phospholipid content and increased phospholipid biosynthesis in solvent-stressed cells (484).
These alterations serve to produce a physical barrier to the intercalation of hydrocarbons in membranes, thus offsetting the passive influx of hydrocarbons into the cell. It is generally believed that hydrocarbons interact with microorganisms nonspecifically and move passively into the cells (45). Of course, hydrocarbon-degrading microorganisms must necessarily come in contact with their substrates before any transport, either active or passive, may take place. Traditionally, three modes of hydrocarbon uptake are cited to describe how hydrocarbon-metabolizing organisms come in contact with their substrates. However, since uptake implies an active movement of substrate across the cell membrane, a more accurate nomenclature for the initial stages of cell-substrate interaction may be hydrocarbon access (631). While microorganisms may contact water-solubilized hydrocarbons, decreasing solubility with increasing molecular weight is restrictive (91). Two additional, perhaps more widespread modes of hydrocarbon accession are direct adherence to large oil droplets and interaction with pseudosolubilized oil (67). For example, Van Hamme and Ward (631) described a Rhodococcus strain that grew directly on crude oil droplets and could be removed with the addition of exogenous chemical surfactant, while a Pseudomonas strain required surfactant-solubilized oil to efficiently access hydrocarbons. In P. aeruginosa, hydrocarbon solubilization and micellar transport control hexadecane biodegradation during biosurfactant-enhanced growth (552). Similarly, encapsulating solid n-C18 and n-C36 in liposomes increased growth and biodegradation by a Pseudomonas sp., indicating that cell-liposome fusion may deliver encapsulated hydrocarbons to membrane-bound enzymes (427).
Only a limited number of studies conclusively indicate that active hydrocarbon uptake into bacterial cells occurs. Naphthalene uptake by P. putida PpG1 appears to be nonspecific, as there is no inhibition by protein inhibitors or iodacetamine and no requirement for specific naphthalene degradation gene expression (45). Similarly, phenanthrene uptake by Pseudomonas fluorescens LP6a appears to be passive, in contrast to the observed energy-dependent phenanthrene efflux (84). With respect to active transport, proton motive force uncouplers have been shown to apparently decrease both n-hexadecane (46) and naphthalene (660) uptake, which could indicate that energy-dependent uptake is important in some strains. In these two studies, the fact that the strains being studied could metabolize the substrates over the long incubation times complicates the separation of phenomena related to transport, metabolism, and growth. Probably the best observational evidence for energy-dependent alkane uptake is the case of Rhodococcus erythropolis S+14He, which preferentially accumulates n-hexadecane from hydrocarbon mixtures (327).
Recently, Story et al. (587) identified a gene (pbhD) in Sphingomonas paucimobilis var. EPA505 that is necessary for fluoranthene metabolism and has homology to the gene pyruvate phosphate dikinase (ppdK), a gene involved in glucose uptake in prokaryotes and plants. The authors postulated that pbhD may be involved in the uptake of flouranthene catabolites that leak from the cell, although no experiments were performed to verify this. Even though direct molecular evidence for active uptake has not been presented, it would not be surprising to find energy-dependent pumps that transport hydrocarbons into the cell. The presence of hydrocarbon inclusions, of both pure and partially oxidized alkanes, for example (46, 274), indicates that these substrates can be accumulated against a concentration gradient, presumably an energy-dependent process. In addition, as has been observed for 2,4-dichlorophenoxyacetate (244) and 4-hydroxybenzoate (245) metabolism, uptake and chemotaxis may be coordinately controlled at the molecular level.
Mechanisms of tolerance. While an undisputed molecular mechanism for active hydrocarbon uptake is not yet available, excellent descriptions of active hydrocarbon efflux from bacterial cells have been presented in the last 7 years. In their review, Sikkema et al. (566) stated that "there is no precedent why active excretion systems should not play a role in lowering the concentrations in the cytoplasmic membrane (and cytoplasm) of toxic lipophilic molecules." Since that time, two Pseudomonas putida strains (DOT-T1E and S12) have been characterized in great detail, both physiologically and genetically, with respect to their ability to thrive in the presence of hydrocarbons. The most notable advance in this area has been the molecular characterization of active solvent efflux pumps for aromatic hydrocarbons (322, 332, 382, 441, 518).
Ramos et al. (501) isolated P. putida DOT-T1E, which metabolizes toluene and is capable of growing in the presence of 90% (vol/vol) toluene. In early studies, DOT-T1E was found to increase membrane rigidity by converting cis-9,10-methylene hexadecanoic acid to 9-cis-hexadecanoic acid and subsequently to the corresponding trans isomer. This short-term response typically occurs in less than 1 min upon exposure to toluene. P. putida S12, which does not grow on toluene but can tolerate high levels of organic solvents such as styrene (658) and toluene (659), also exhibits cis/trans isomerizations (659). In the long-term (15 to 20 min) exposure, DOT-T1E decreased the amount of phospatidylethanolamine in the phospolipid polar head groups and increased cardiolipid levels, again increasing membrane rigidity (501). These changes increase lipid ordering to restore membrane integrity and reduce organic solvent partitioning in the membrane. A gene encoding a cis/trans isomerase, cti, which catalyzes the isomerization of esterified fatty acids in phospholipids (mainly cis-oleic acid [C16:1,9] and cis-vaccenic acid [C18:1,11]) has been cloned and sequenced in DOT-T1E.
Null mutants exhibited lower survival rates upon toluene shock. In addition, while a longer lag time was observed when mutants were exposed to toluene in the vapor phase, the growth rates for the mutant and the wild-type strain were similar. Thus, the cis/trans isomerization helped prevent cell damage but was apparently not the most important element in solvent resistance. Cti is constitutively expressed in DOT-T1E and, as expected, is located in the membrane. The cti gene is also found in nonresistant P. putida strains and other Pseudomonas species (297).
Toluene tolerance in DOT-T1E was found to be inducible by exposure to toluene in the vapor phase, which led the group to postulate that an active solvent exclusion system and metabolic toluene removal afforded some protection (501). Similarly, resistance to antibiotics and solvents such as ethanol was found to increase in S12 with exposure to toluene but not antibiotics (279). In [14C]toluene influx studies, an energy-dependent efflux system was proposed, as less influx was observed in adapted cells, while greater influx was observed in the presence of potassium cyanide, a respiratory chain inhibitor, and m-chlorophenylhydrazone, a proton conductor (276). The interruption of toluene metabolism through mutation of the tod genes did not affect toluene tolerance in DOT-T1E, suggesting that some other mechanism of tolerance was involved (440). Indeed, active solvent exclusion systems, have been characterized in these two strains.
The srpABC (solvent resistance pump) genes of P. putida S12 were the first to be cloned and unambiguously shown to be responsible for toluene efflux (322). The pump consists of SrpB (inner membrane transporter), SrpC (outer membrane channel), and SrpA (periplasmic linker protein) and is homologous to the proton-dependent multidrug efflux systems of the resistance/nodulation/cell division (RND) family of pumps, which export antibiotics metals, and oligosaccharides. These pumps have been well reviewed by Paulsen et al. (477).
Induced by aromatic and aliphatic solvents and alcohols, the efflux system encoded by srpABC is proton dependent and does not pump antibiotics or other substrates of multidrup resistance pumps (277). Unlike cis/trans isomerisations, which can be a general stress response (251), the srpABC genes are not induced by extremes of pH, temperature, salt, organic acids, or heavy metals (323). These adaptation mechanisms are energy consuming and have been shown to decrease growth rates and yields while increasing maintenance energy and lag times (278). Presumably, the increased energy consumption may also result from solvent-mediated membrane uncoupling and disruption of energy-transducing proteins.
The first efflux pump in DOT-T1E was found by producing a toluene-sensitive, octanol-tolerant mutant (DOT-T1E-18) by Tn5-phoA mutagenesis with a gene knockout homologous to the drug exclusion gene mexB, which is a member of the efflux pump family of the resistant modulator type (502). The gene was named ttgB for toluene tolerance gene. Solvent exclusion testing with 1,2,4-[14C]trichlorobenzene showed that increasing toluene concentrations increased the amount of radiolabel in the membranes. In addition, the pump was shown to be specific, as DOT-T1E is sensitive to benzene but not m-xylene. Given the fact that the mutant exhibited low levels of survival when toluene was delivered in the vapor phase, it was postulated that at least two efflux pumps were present, one constitutive and one inducible.
Indeed, three toluene efflux pumps have ultimately been found in DOT-T1E (441, 518). This is not without precedent, as P. aeruginosa has at least three RND antibiotic efflux pumps, which also accommodate organic solvents: MexAB-OprM, MexCD-OprJ, and MexEF-OprN (381, 382). The first pump in DOT-T1E, ttgABC, is a constitutive efflux pump controlled by ttgR, which produces a transcriptional repressor for the ttgABC operon, which in turn is controlled by another repressor belonging to the Lrp family of global regulators. In this case, TtgR is expressed at high levels in the presence of toluene, which in turn reduces TtgABC expression (158). The second pump, ttgDEF, is found adjacent to the tod genes and is expressed in response to toluene and styrene. Unlike ttgABC, ttgDEF does not appear to efflux antibiotics and is closely related but not identical to the toluene efflux pump srpABC of P. putida S12.
The third pump, ttgGHI, is expressed constitutively at high levels from a single promoter and, if grown with toluene, is expressed at higher levels from two promoters: one a constitutive promoter and a second, overlapping, inducible promoter (518). ttgG encodes the periplasmic lipoprotein that is anchored to the inner membrane and, along with the inner membrane pump encoded by ttgG, forms the putitive translocase. ttgI encodes the outer membrane protein that may form a channel into the periplasmic space (518). In order to make DOT-T1E sensitive to toluene shock and to eliminate its ability to grow with toluene in the gas phase, mutations had to be introduced in all three pumps. Mutation studies showed that TtgABC and TtgGHI pump toluene, styrene, m-xylene, ethylbenzene, and propylbenzene. TtgDEF only removes tolene and styrene.
Overall, it appears that efflux pumps in Pseudomonas spp. can be divided into three general groups: those that pump organic solvents, those that pump antibiotics, and those that pump both. Kieboom et al. (321) recently described an active antibiotic efflux pump in S12 (ArpABC) which does not pump solvents. This is in contrast to the MepABC pump in P. putida KT2442 (206) and the Mex pumps in P. aeruginosa (382), which pump both solvents and antibiotics. Furthermore, much will be gained if efflux pumps for other hydrocarbons and for other microorganisms are studied in detail and compared to known systems. Further research at the protein level will be required for many systems, as comparative studies will help to unravel the factors affecting pump specificity, to understand what forces govern substrate recognition, and to see if and how pump receptors are able to regulate other behaviors such as taxis, the final behavior to be discussed here.
One can imagine that movement away from a hydrocarbon plume could reduce toxic effects or that movement towards a water-insoluble substrate such as naphthalene could be advantageous in poorly mixed field situations. Indeed, Marx and Aitken (410) used a capillary assay (409) to show that Pseudomonas putida G7 catalyzed naphthalene degradation at faster rates in unmixed, heterogeneous systems than did mutants deficient in either motility or naphthalene chemotaxis. In mixed systems, the naphthalene degradation rate was identical for the wild-type and mutant strains.
P. putida G7 possesses the NAH7 catabolic plasmid for the meta-cleavage of aromatic hydrocarbons (226, 227). The plasmid includes the nahY gene, encoding a 538-amino-acid membrane protein whose C terminus resembles that of chemotaxis transducer proteins (i.e., methyl-accepting chemotaxis proteins). This indicates that NahY may be a chemoreceptor for naphthalene or naphthalene metabolites (227), but neither the molecular nature of binding nor the cascade of responses that occur following binding has been studied.
Pseudomonas putida RKJ1 possesses an 83-kb plasmid for naphthalene metabolism through salicylate (538). A Nap- Sal+ mutant was chemotactic towards only salicylate, while a Nap- Sal- mutant exhibited no chemotaxis. This suggests the presence of a metabolism-dependent energy taxis in this strain. Thus, a change in the redox potential or cellular energy level in the cell probably provides the signal for chemotaxis. Alternatively, a membrane-bound or intracellular chemoreceptor may recognize naphthalene or salicylate degradation products.
To date, no reports describing the molecular basis for alkane chemotaxis have appeared. However, van Beilen et al. (627) detected alkN in the 9.7-kb region between alkBFGHJKL and alkST in P. putida GPo1, which encodes a protein with 30% sequence similarity to methyl-accepting transducers such as the one found in strain G7 (227). As GPo1 is not very motile, the functionality of the gene is difficult to study.
Overall, taxis in relation to petroleum hydrocarbons has been neglected, and the area is ripe for study. First of all, more examples of tactic behavior to hydrocarbons are required in other genera and with different hydrocarbons in order to appreciate the diversity of responses. Second, when putative chemoreceptors are detected by gene sequencing, systematic studies of purified proteins are required in order to understand the key molecular interactions that take place to allow a cell to detect a particular chemical. Third, the mechanisms by which chemoreceptors translate signals induced by hydrocarbons into cellular responses and their impact on overall cellular biochemistry would allow the integration of this behavior, and all of the behaviors discussed here, into a larger picture of hydrocabon-metabolizing organisms. Recent developments for the large-scale and nearly real-time monitoring of gene expression in live cells with green fluorescent protein promoter fusions (300, 579) will allow this type of integrating study. Finally, understanding the true role of chemotaxis during remediation needs more attention if we are going to understand the impact of taxis on biofilm formation, substrate access, and avoidance of toxic substances. Recent developments in tracking live bacterial cells with advanced imaging technologies (559) could be combined with gene expression technologies and traditional measurements of hydrocarbon degradation (258) to study these questions.
| MICROBIAL COMMUNITY DYNAMICS |
|---|
|
TopPreviousNextReferences |
|---|
Approaches to cataloguing microbial diversity and community function can be broadly divided into culture-dependent and culture-independent methods, both of which may include genetic characterization techniques. Traditional culture-dependent methods are the most familiar and are based on differential morphological, metabolic, and physiologic traits. These include isolation and cultivation on solid media, most-probable-number (MPN)-style liquid assays, and more recently, Biolog substrate utilization plates. Culture-independent methods for community analysis began with direct examination of metabolically active microorganisms with differential stains such as 4',6'-diamidino-2-phenylindole, (INT)-formizan and CTC, fluorescence in situ hybridization, and bulk analysis of total protein banding and phospholipid fatty acid analysis.
With rapid expansions in the field of molecular genetics, a host of PCR-based approaches have emerged to study specific microorganisms or groups of microorganisms and specific genes and to evaluate overall community profiles. Methods to evaluate community profiles include denaturing and temperature gradient gel electrophoresis, ribosomal intergenic spacer analysis, single-strand conformation polymorphism, internal transcribed spacer-restriction fragment length polymorphism, random amplified polymorphic DNA, and amplified ribosomal DNA restriction analysis (317). Recently, developments in the use of DNA microarrays have attracted the attention of environmental microbiologists for more rapid throughput to allow the tracking of thousands of genes at one time (146).
A few examples of community studies involving petroleum applications are discussed here in order to highlight the utilities and limitations of the various methods (Table 3).
|
If one is interested either in reporting an isolated microorganism as having hydrocarbon-metabolizing abilities or in performing enumerations of hydrocarbon-degrading microorganisms, it is essential to include proper controls. Ample evidence is available to illustrate that non-hydrocarbon-degrading microorganisms will develop on agar plates prepared with solid, liquid, or volatile hydrocarbons due to the presence of utilizable carbon even in purified agarose (60, 504). In an evaluation of mineral agar plates with and without toluene-xylene fumes, it was revealed that little selection was provided against non-toluene- and non-xylene-degrading bacteria. Despite the caution to incubate plates with and without hydrocarbon, studies with oil agar to enumerate hydrocarbon-degrading bacteria without reporting proper controls can still be found. This type of report should be examined with care.
In an attempt to overcome the problem with trace carbon in agar preparations, some researchers turned to the use of silica gel as a solidifying agent. However, this tedious procedure has not enjoyed widespread use. If isolates are not required, a rapid MPN test (sheen-screen) with tissue culture plates can be employed for nonvolatile hydrocarbons based on the formation of emulsions, avoiding the problem of trace carbon contamination altogether (77). A similar assay to screen for hydrocarbon degraders based on a redox indicator has been described (236) and combined with the sheen-screen to produce an MPN assay based on both emulsification and respiration (633).
Numerous studies have attempted to describe microbe-microbe and microbe-hydrocarbon interactions by extrapolating from detailed laboratory studies with isolates from hydrocarbon-contaminated environments. For example, evaluations of functional and physiological isolate groupings have been carried out in an effort to quantify the oil emulsification abilities and type of hydrocabon accession mode used by environmental isolates (67). Researchers have also constructed s
