Diversity Outbred Mice Identify Population-Based Exposure Thresholds and Genetic Factors that Influence Benzene-Induced Genotoxicity

Abstract

Background: Inhalation of benzene at levels below the current exposure limit values leads to hematotoxicity in occupationally exposed workers.

Objective: We sought to evaluate Diversity Outbred (DO) mice as a tool for exposure threshold assessment and to identify genetic factors that influence benzene-induced genotoxicity.

Methods: We exposed male DO mice to benzene (0, 1, 10, or 100 ppm; 75 mice/exposure group) via inhalation for 28 days (6 hr/day for 5 days/week). The study was repeated using two independent cohorts of 300 animals each. We measured micronuclei frequency in reticulocytes from peripheral blood and bone marrow and applied benchmark concentration modeling to estimate exposure thresholds. We genotyped the mice and performed linkage analysis.

Results: We observed a dose-dependent increase in benzene-induced chromosomal damage and estimated a benchmark concentration limit of 0.205 ppm benzene using DO mice. This estimate is an order of magnitude below the value estimated using B6C3F1 mice. We identified a locus on Chr 10 (31.87 Mb) that contained a pair of overexpressed sulfotransferases that were inversely correlated with genotoxicity.

Conclusions: The genetically diverse DO mice provided a reproducible response to benzene exposure. The DO mice display interindividual variation in toxicity response and, as such, may more accurately reflect the range of response that is observed in human populations. Studies using DO mice can localize genetic associations with high precision. The identification of sulfotransferases as candidate genes suggests that DO mice may provide additional insight into benzene-induced genotoxicity.

Figure 1

MN‑RET measurements were consistent between cohorts. (A) Preexposure blood MN‑RETs. Boxes represent the median and interquartile range, and whiskers cover the entire data range. (B) Postexposure blood MN‑RETs increased in the 100-ppm group. (C) Postexposure bone marrow MN‑RETs increased in a dose-dependent manner. *Approximately 1-unit increase in MN-RET/1,000 with each order of increase in benzene concentration.

Figure 2

BMC modeling of bone marrow MN‑RET in DO mice using the 0-, 1-, 10-, and 100-ppm exposure groups to estimate a BMR_1SD (A) and a BMR₁₀ (B). Values shown are the mean of each exposure group ± SE. Curved black lines represent the model fit, dashed blue lines represent the BMC, and red dot and dash lines indicate the BMCL. Insets show details of low concentration range. (C) BMC modeling of bone marrow MN‑RET in DO mice using the 0-, 1-, and 10-ppm exposure groups using a BMR₁₀. BMC modeling of bone marrow PCE in B6C3F1 mice using a BMR_1SD (D) and a BMR₁₀ (E).

Figure 3

Linkage mapping of bone marrow MN‑RET in the 100-ppm exposure group revealed a significant QTL on Chr 10. (A) Plot of the LOD at each marker; the red line indicates the permutation-derived significance threshold of p = 0.05. (B, top) Plots of the effects of each of the eight DO founder alleles on Chr 10 (top) and the LOD score on Chr 10 (bottom); the CAST allele (green) is associated with lower MN‑RETs. (C) MN‑RET values by DO genotype at the marker with the maximum LOD score on Chr 10 (31.868 Mb). Data points indicate BM MN‑RET values for individual DO mice, and red lines show the mean ± SE of each genotype group. Genotypes are listed on the x-axis, with each DO founder represented by a letter; genotypes containing the CAST allele are shown in green. (D) Association mapping within the Chr 10 QTL interval. (D, top) Each data point shows the LOD score at one SNP; red data points indicate scores above the p < 0.01 threshold. (D, bottom) Genes in the QTL interval. Dashed vertical lines show the QTL support interval. Sult3a1 and GM4794 are highlighted in red to indicate their location.

Figure 4

Linkage mapping of liver Sult3a1 (A,C) and Gm4794 (B,D) expression revealed a QTL on Chr 10 in the same location as the MN‑RET QTL. (A,B) Plot of the LOD at each marker; the red line indicates the permutation-derived significance threshold of p = 0.05. (C,D) Plots of the effects of each of the eight DO founder alleles on Chr 10 (top) and plots of the LOD score on Chr 10 (bottom). The CAST allele (green) is associated with higher liver Sult3a1 and Gm4794 expression.