Species-specific regulation of PXR/CAR/ER-target genes in the mouse and rat liver elicited by o, p'-DDT

Abstract

Background: Dichlorodiphenyltrichloroethane (DDT) is a persistent estrogenic organochlorine pesticide that is a rodent hepatic tumor promoter, with inconclusive carcinogenicity in humans. We have previously reported that o, p'-DDT elicits primarily PXR/CAR-mediated activity, rather than ER-mediated hepatic responses, and suggested that CAR-mediated effects, as opposed to ER-mediated effects, may be more important in tumor promotion in the rat liver. To further characterize species-specific hepatic responses, gene expression analysis, with complementary histopathology and tissue level analyses were investigated in immature, ovariectomized C57BL/6 mice treated with 300 mg/kg o, p'-DDT, and compared to Sprague-Dawley rat data.

Results: Rats and mice exhibited negligible histopathology with rapid o, p'-DDT metabolism. Gene expression profiles were also similar, exhibiting PXR/CAR regulation with the characteristic induction of Cyp2b10 and Cyp3a11. However, PXR-specific target genes such as Apoa4 or Insig2 exhibited more pronounced induction compared to CAR-specific genes in the mouse. In addition, mouse Car mRNA levels decreased, possibly contributing to the preferential activation of mouse PXR. ER-regulated genes Cyp17a1 and Cyp7b1 were also induced, suggesting o, p'-DDT also elicits ER-mediated gene expression in the mouse, while ER-mediated effects were negligible in the rat, possibly due to the inhibitory effects of CAR on ER activities. In addition, o, p'-DDT induced Gadd45a, Gadd45b and Cdkn1, suggesting DNA damage may be an additional risk factor. Furthermore, elevated blood DHEA-S levels at 12 h after treatment in the mouse may also contribute to the endocrine-related effects of o, p'-DDT.

Conclusion: Although DDT is known to cause rodent hepatic tumors, the marked species differences in PXR/CAR structure, expression patterns and ligand preference as well as significant species-specific differences in steroidogenesis, especially CYP17A1 expression and activity, confound the extrapolation of these results to humans. Nevertheless, the identification of potential modes of action as well as species-specific responses may assist in the selection and further development of more appropriate models for assessing the toxicity of DDT to humans and wildlife.

Figure 1

Relative liver weight and tissue level analysis. (A) Relative liver weight. Immature ovariectomized C57BL/6 mice were orally administered 300 mg/kg o, p'-DDT or sesame oil vehicle at time 0, 24 and 48 h. Mice were sacrificed 2, 4, 8, 12, 18, 24 or 72 h after the initial dose. The relative liver weight was significantly increased at 72 h. (B) Hepatic concentration of o, p'-DDT and o, p'-DDD. (C) Hepatic concentration of o, p'-DDE. Hepatic tissue levels of o, p'-DDT, o, p'-DDD and o, p'-DDE were determined using high-resolution gas chromatograph/HRMS from three randomly selected mice orally gavaged with 300 mg/kg o, p'-DDT. The data are presented as mean ± SE. The asterisk (*) indicates a significant (p < 0.05) difference from the vehicle controls.

Figure 2

Microarray and QRT-PCR results. (A) Identification of differentially regulated genes. (B) Number of differentially expressed genes following o, p'-DDT treatment in the mouse liver. Differentially expressed genes were selected based on a p1(t) ≥ 0.999 at two or more time points and an absolute fold change ≥ 1.5 at one or more time points relative to time-matched vehicle controls. All differentially expressed genes are listed in Additional file 3. (C) Verification of microarray results by QRT-PCR. QRT-PCR results relative to time-matched vehicle controls are shown as bar and presented as mean ± SE. Microarray results are represented as lines. The dashed line indicates the expression level of the time-matched vehicle control. The asterisk (*) indicates a significant (p < 0.05) difference from the time-matched vehicle controls for QRT-PCR, n = 5. (D) Hepatic expression of PXR/CAR-target genes in o, p'-DDT-treated mouse. A heat map of o, p'-DDT elicited microarray expression profiles for selected PXR-, CAR-specific and PXR/CAR-shared target genes identified in the literature [20-23]. While some CAR-regulated genes such as Cyp1a1, Fmo5, Sult1d1 or Abcc2 were moderately induced, several PXR-target genes, including ApoA4, Ces2, Gstm2 or Insig2, exhibited strong induction. However, other PXR-target genes such as Hmgcs1 and Hmgcs2 were down-regulated.

Figure 3

Comparative analysis of global gene expression profiles elicited by o, p'-DDT. (A). Comparative gene expression analysis between EE-treated mouse, o, p'-DDT-treated rat and o, p'-DDT-treated mouse. A total of 996 orthologs were represented on the rat cDNA microarray, mouse cDNA microarray and mouse Agilent oligonucleotide microarrays determined by HomoloGene . 538 of these orthologs showed a |fold change| ≥ 1.5 for at least one time point in either species. These 538 differentially expressed orthologs were subjected to hierarchical clustering. The dendrogram illustrates that mouse o, p'-DDT gene expression profiles are more similar to rat o, p'-DDT gene expression profiles than the mouse EE gene expression profiles. (B) Correlation analysis using differentially expressed orthologous genes. The temporal profiles of o, p'-DDT-treated mouse liver (current study) and those of the o, p'-DDT-treated rat liver [11] were compared by determining the Pearson's correlation of the temporal gene expression (fold change) and significance (p1 [t] value) between orthologs. Both studies used comparable study designs and data analysis methods, although different platforms were used (i.e., rat cDNA microarray and mouse Agilent oligonucleotide microarray). 140 genes were identified as differentially expressed orthologs. (C) Scatter plot of the 140 differentially expressed orthologous genes. Correlations for gene expression and significance approaching 1.0 indicate that the behavior or the orthologous genes are similar and would fall in the upper right quadrant. Orthologs tended to localize in upper- or lower-right quadrant (32.9% and 47.9% of total number of spots, respectively), indicating that temporal gene expression changes for o, p'-DDT-treated mouse and rat liver are comparable. However, poor correlations between the temporal p1(t) values and gene expression fold changes would fall within the lower left quadrant. For example, Cyp17a1 fell into this quadrant suggesting that significant differences exist between the rat and mouse ortholog expression profiles.

Figure 4

Species-specific regulation of steroid hormone metabolism elicited by o, p'-DDT. (A) Overview of the role of CYP17A1 and CYP7B1 in steroid metabolism. CYP17A1 metabolizes pregnenolone and progesterone to produce DHEA and androstenedione, respectively. Hepatic CYP7B1 is involved in bile acid biosynthesis, and also responsible for 7α-hydroxylation of DHEA. (B) Hepatic Cyp17a1 and (C) Cyp7b1 mRNA levels in the o, p'-DDT-treated mouse and rat. QRT-PCR results relative to time-matched vehicle controls are shown as bars and presented as mean ± SE. Microarray results are represented as lines. o, p'-DDT induced Cyp17a1 and Cyp7b1 mRNAs in the mouse liver, while it did not affect in the rat liver [11]. The dashed line indicates the expression level of the time-matched vehicle control. The asterisk (*) indicates a significant (p < 0.05) difference from the time-matched vehicle controls for QRT-PCR, n = 5. (C) Representative Western analysis result for hepatic CYP17A1 protein in o, p'-DDT-treated mouse liver. CYP17A1 protein levels were induced at 18 and 24 h. Western analyses were performed on 3 independent biological replicates to verify the consistency of the results. C, control; T, 300 mg/kg o, p'-DDT. (D) Blood DHEA-S levels. DHEA-S level was significantly higher at 12 h following o, p'-DDT treatment compared to time-matched controls in the mouse, while it did not change in rats.

Figure 5

Potential species-specific responses to o, p'-DDT. In the rat liver, o, p'-DDT elicits PXR/CAR-mediated response, with negligible ER-mediated effects despite the established estrogenicity of o, p'-DDT in other target tissues (e.g., uterus). The absence of ER-mediated gene expression may be due, in part, to the inhibitory effects of CAR on ER. In contrast, o, p'-DDT elicits not only PXR/CAR-mediated responses but also ER-mediated effects such as the induction of Cyp17a1 and Cyp7b1 in the mouse liver. Gene expression profiling suggests that o, p'-DDT preferentially activates PXR in the mouse liver. Therefore, the inhibitory effects of CAR on ER-mediated signaling would be reduced in the mouse liver, allowing the activation of ER-mediated gene expression by o, p'-DDT. In addition, only mouse DHEA-S levels were increased, which could affect both endocrine and liver physiology. The induction of Cyp17a1 and Cyp7b1 may be involved in either the metabolism or disposition of DHEA, or the conversion of DHEA into other metabolites including neurosteroids.