Use of Genetic
Markers as a Screening Tool for Ecological Risk Assessment at the
Idaho National Engineering and Environmental Laboratory:
Microsatellite Mutation Rate of Burrowing Mammals
The purpose of this research was to explore the utility of molecular genetic
techniques as screening tools for evaluating the risk to natural populations
from contaminant exposure. These tools can be used to help evaluate the need for
site remediation. If remediation is implemented, genetic characterization of
populations can provide insight on the effectiveness of the remediation through
This was a three-year study to determine if radiological contamination
affects the genome of deer mice (Peromyscus maniculatus). Radiological
and genetic analysis were performed on deer mice collected outside the
Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA)
sites of the stationary low-power reactor number 1 (SL-1) and the Radioactive
Waste Management Complex (RWMC), and two control sites (Burn and Atomic City).
Radiological and hazardous wastes have been disposed of at the INEEL since
1952. Escape of radionuclides and hazardous constituents from uncontained
wastes, deterioration of waste containers, and waste disposal practices have
resulted in contamination of the subsurface soils at the INEEL's Subsurface
Disposal Area (SDA) and at other facilities. To assess the risks to human and
environmental health, the potential impacts of contaminant exposure on
identified receptors must be determined.
Burrowing and excavation of the soil by small mammals, including deer mice,
has been shown to be responsible for some radionuclide transport through the SDA
environment; however, the genomic effects of exposure to contaminants at the
INEEL was not known. Research was needed to develop new techniques to determine
the effects of that exposure to the genome of individuals, so that environmental
and remedial actions can be properly implemented. The objectives of this
research were to:
- Perform comparative analysis of mother/offspring genotypes across the
study areas using allelic data compiled from two consecutive field seasons (14
genetic markers: 147 females; 529 embryos; and 9464 Polymerase Chain
- Identify mutant alleles in all samples and perform statistical analysis.
- Perform microsatellite genetic analysis of the four study area populations
to determine population structure using a variety of genetic software
- Analyze radiological data obtained from all the females collected during
fiscal years 2001 and 2002 and perform statistical analysis.
Laboratory Experimental Procedures - Fourteen genetic markers were
used to perform microsatellite genetic analysis. Data were organized by
population and families; allelic maternal segregation of the offspring was
determined, paternal alleles inferred, and mutations identified. Only mutations
of alleles segregated from the mother were used for this analysis. Not all the
families provided useful data and some had to be removed for the genetic
analysis (36 percent of the families were removed).
Mutation Rate Analysis Results - The ratio of microsatellite mutant
alleles versus the non-mutant alleles was used as a direct assessment of
mutation using parent/offspring comparison of allele differences. Allele scoring
was performed with 14 microsatellite markers for females and offspring from the
four study areas. The proportion of mutant alleles from each population was
pair-wise compared between populations and tested for significant differences
using the Fisher's exact test. The statistical analysis suggests that there is
no significant difference when comparing the two contaminated sites (RWMC and
SL-1) with the two control sites (Burn and Atomic City).
Population Genetic Structure Analysis - Four different approaches were
used to test for population genetic structure:
- The allele frequency differences among populations were tested using the
Fisher's exact test, where an unbiased estimate of the p-value of the
probability test is obtained. The null hypothesis (Ho) "the allelic
distribution is uniform across populations" was tested for each locus on a
- The differences in genotype distribution across the populations were based
on estimates of Wright's F-statistics. The null hypothesis, Fst = 0
"the genotypic distribution is uniform across
populations" was tested for all loci using a chi-square goodness-of-fit test.
- The estimate of Rhost statistics is a measure analogous
to Fst but incorporates allele size estimates and assumes a
step-wise mutation model.
- Finally, an assignment test was performed using GeneClass. This test uses
a Bayesian approach to detect immigrants by using multilocus genotypes.
The null hypothesis was rejected in the first two methods; that is, the
differences in allele frequency distribution and the genotype distribution
differences between populations are statistically significant. The third method
indicates that some level of gene flow occurs between populations but not enough
that the distributions of alleles or genotypes between them are homogenized. The
assignment test also supports population genetic structure because 98.61 percent
of the individuals were correctly assigned to their population of origin.
Radiological Analysis - Samples from 80 females were used for this
analysis: 43 from Atomic City; 21 from SL-1; and 16 from RWMC (the Burn site was
excluded from the analysis due to small available sample size). Female carcasses
were processed and analyzed for presence of radiological contaminants. None of
the females showed exposure at levels higher than background. Statistical
analysis of the radiological data indicates that the differences between the
populations are not significant. That is, female mice collected from RWMC and
SL-1 had similar levels of radiological (background) contamination compared to
the Atomic City control site.
The populations have different allelic frequency composition, and even though
there is evidence that some level of gene flow occurs, and that they are not
isolated completely, the migration rate is not strong enough to create a
panmictic population (a population with no genetic structure). This study
indicates that the four populations are distinctive when comparing the two
control sites with SL-1 and RWMC population.
It is assumed that natural populations share the same level of mutation rate,
because this is an evolutionary force that is stochastic. This means that it is
not affected by any of the evolutionary selective forces, and it occurs at a
very low frequency. If there are no other contributing factors such as
anthropogenic activities, it is predicted that in all of the studied populations
the mutation rates should be similar. This study shows that there is no
difference in the rate of mutation in populations exposed to anthropogenic
activities in comparison with populations that have had little or no exposure.
This suggests that no external forces (outside from evolutionary forces) are
acting on this population.
The population genetic analysis and the mutation rate data support each
other. The population genetic analysis suggests that there is a geographical
component that plays a role in population differentiation. The radiological
analysis indicated that the animals collected at the RWMC site have the same
level of background exposure as the ones from the other three sites. Based on
this study, we cannot conclude that exposure to radiological contaminants is an
issue of concern for mice collected at RWMC or SL-1. One observation that is
important to make, however, is that samples were not allowed to be collected
from areas of known soil contamination. Therefore, the question of whether
microsatellites are a good tool for identifying mutation differences caused by
exposure still needs to be tested.
Further research should be pursued using this species as a biological
indicator for environmental monitoring of contaminants, as well as long-term
stewardship and ecotoxicological studies.
This research is part of a long-term plan for building a capability at INEEL
in the use of genetic markers to address environmental issues. Once the
technology is established, numerous applications using other species as
environmental monitoring indicators can be explored. This information will help
researchers focus resources and efforts on environmental monitoring at sites
where there is a high probability of adverse biological impact caused by the
presence of contaminants.
Investigators and Affiliations
Angela Stormberg, Principal Scientist, Idaho National Engineering and
Environmental Laboratory, Idaho Falls, ID
Joseph Cook, Department of Biological Sciences, Idaho State University,
Laboratory Directed Research and Development (LDRD)