Abstract

The immediate effects of petroleum oil exposure following an oil spill are so dramatic that it is relatively straightforward to record, monitor, and study the short-term impacts on individuals and on populations in the spill area. The acute effects of oil exposure in marine mammals are varied and include disturbances in thermoregulation, reproduction, and possibly damage to specific organ systems such as lung, liver and kidney. While some of these acute effects can be detected clinically by hematological and serum chemical analyses or at necropsy, the more chronic and/or low-grade effects of oil exposure are more difficult to identify. Despite this difficulty several studies indicate that the long-term impact of petroleum oil exposure is significant. These long-term effects may be a result of sub-lethal pathology in individuals exposed to oil at the time of the spill, or of chronic physiological stresses from continued exposure to oil remaining in the environment. The key problem regarding these low-grade, yet significant, oil-induced pathophysiological changes in susceptible individuals, is the difficulty in detecting them by conventional diagnostic methodology. This difficulty has precluded the effective use of physiological measures in studying the long-term impact of oil spills on exposed individuals. Instead, the majority of information regarding the effects of chronic oil exposure has been generated using statistical techniques to identify changes in population demographics, patterns of mortality, reproductive efficiency, and survivability. While these ecological studies have been extremely informative, the results can be confounded by factors, unrelated to oil exposure, that also impact ecology and individual survival. There is, therefore, a need to develop specific and sensitive methods for detecting lingering, low-grade pathophysiological changes in oil-exposed individuals. These tools could be used to complement the population ecology studies, by reliably identifying individuals with ongoing, subtle oil-associated physiological perturbations.

The major problem in developing assays for measuring subtle toxin-induced physiological changes in marine mammals is the logical restriction on performing controlled exposure studies in the majority of these species. This hurdle can be over-come by the use of surrogate species. In marine mammal toxicology, the most successful of these has been the use of the American mink (Mustela vison) as a surrogate for the Southern sea otter (Enhydra lutris). In this study 8-month old male American mink were used to study the effects of long-term petroleum oil-exposure. Mink were divided into two groups: one group of animals was fed once a day a food formulated for mink that had mixed into it 500 ppm of Bunker C fuel oil (provided by Dr. Jonna Mazet, Wildlife Heath Center, UCD). This dose of petroleum oil was readily ingested by the mink and it corresponds to the petroleum oil concentrations measured in invertebrates sampled in the oiled Prince William Sound region one year after the Exxon Valdez oil spill. The other group of mink was fed food that was mixed with an equivalent amount of mineral oil (placebo) and served as controls. In this study we examined using alterations in gene expression as a sensitive and specific assay for detecting oil-associated changes in mink tissues. In marine mammal toxicology there has been a heavy reliance on the identification of foreign chemicals (xenobiotics) within individual tissues as an indicator of a toxic insult. Unfortunately these assays are limited in the information they provide, since they do not measure how the chemicals influence the health of an individual. In addition, the effect of some toxic insults may persist beyond metabolism and excretion of the potentiating toxin. The absence of detectable oil or its metabolites from an individual is therefore not evidence for the absence of oil-related effects. The advantage of using a gene expression assay in marine mammal toxicology is its ability to measure the physiologic responses (acute or chronic) of an individual to the metabolic insult independent of the continued presence of the original toxin or its metabolites. The utility of the methodology proposed in this study relies on the hypothesis that oil-induced sub-lethal pathology is accompanied by changes in gene expression in affected individuals. Such an assay could potentially be adapted for use in free ranging animals, not only in monitoring the long-term effects of an oil spill on individuals, but also to identify ecosystems effected by low-grade oil contamination. In addition these methods could provide insight into the mechanisms by which oil can deleteriously effect an individual animal over a long period, and therefore help in the design of therapeutic and preventative strategies to treat and protect susceptible individuals and populations at risk of oil exposure.

In this study we use a human microarray for detecting changes in gene expression in the livers of petroleum oil-fed mink. Microarray or biochip assays offer a systematic way of analyzing nucleic acid variation that work on similar principles of those in Southern blots where labeled nucleic acids can be used to probe a nucleic acid of interest attached to a solid support. After 4 months of petroleum oil exposure the animals were euthanized, examined by necropsy, and samples taken for multiple analyses. Liver samples were snap-frozen and stored in liquid nitrogen pending further analysis. The liver samples of three mink pairs (petroleum oil exposed versus control) were selected for microarray analysis, and messenger RNA (mRNA) was isolated using silica-based gel membranes combined with microspin technology (Qiagen, Santa Clarita, CA). The human microarray used in this study contained nucleic acid sequences from 8,000 important genes (UniGemV^TM, Incyte Genomics, Palo Alto, CA). The principle of the assay is that an array of nucleic acids, representing important functional genes from the species of interest, attached to a solid phase is used to hybridize to fluorescently-labeled cDNA from a sample of interest. The power of this technology is in comparing two similar samples, in this study the gene expression in an oil-exposed animal versus a non-exposed individual. The technology uses a color-coding technique to discover the differences in gene expression between two mRNA samples. Despite using a microarray from a different species, there was a significant cross-hybridization in over 40% (>3500) of the genes on the human microarray. In addition the oil-fed mink showed a significant (more than twofold) increase in expression in more than 10, and a significant decrease in expression in more than 11, biologically-relevant genes. These differences involved genes associated with immunomodulation, inflammation, cytoprotection, calcium regulation, stress responses, heavy metal metabolism and enzyme induction.

It is important to emphasize that the application of microarrays in this study is for gene discovery. There are serious concerns regarding the sensitivity and specificity of this technique because of the cross-species application. Consequently, studies are underway to validate the findings of this analysis by designing species-specific primers and probes for the genes identified as differentially expressed on the human microarray. These primers and probes tools will be used to perform real-time quantitative RT-PCR on the remaining mink (n>20).

The advantage of using nucleic acid microarrays as screens for oil-induced changes results from their ability to detect changes in expression in a vast number of genes. Since petroleum oil has multiple constituents, the toxic effects of exposure and ingestion are likely to be diverse and widespread within the body. For this reason, the utility of a single marker of sub-lethal oil-induced pathology will be limited. It is possible that this study will identify a number of oil-induced markers of altered cellular metabolism, and that these could be used to define toxin-specific patterns of altered gene expression in the future. The application of species-specific assays for detecting these patterns of altered gene expression has the exciting potential of providing a method for monitoring the long-term effects of oil exposure in individual, free-ranging sea otters. In addition, since sea otters are top predators in coastal marine systems, and consume invertebrate species that bioaccumulate contaminants from substrate and water environment, a sensitive means of measuring their physiological health would be a valuable method for examining the health of various marine ecosystems.

Acknowledgements

The authors wish to thank the Oiled Wildife Care Network (OWCN) for providing financial support for this project.

References

1.  Mazet JK, Gardner IA, Jessup DA, Lowenstine LJ, Boyce WM, 2000. Evaluation of changes in hematologic and clinical biochemical values after exposure to petroleum products in mink (Mustela vison) as a model for assessment of sea otters (Enhydra lutris). Am J Vet Res; 61(10):1197-203

2.  Mazet JA, Gardner IA, Jessup DA, Lowenstine LJ, 2001. Effects of petroleum on mink applied as a model for reproductive success in sea otters. J Wildl Dis; 37(4):686-92

3.  Monson DH, Doak DF, Ballachey BE, Johnson A, Bodkin JL, 2000. Long-term impacts of the Exxon Valdez oil spill on sea otters, assessed through age-dependent mortality patterns. Proc Natl Acad Sci 6; 97(12):6562-7

4.  Parker GA, Bogo V, Young RW, 1986. Acute toxicity of petroleum- and shale-derived distillate fuel, marine: light microscopic, hematologic, and serum chemistry studies. Fundam Appl Toxicol; 7(1):101-5.