Differential Susceptibility to Benzo[a]pyrene Exposure during Gestation and Lactation in Mice with Genetic Variations in the Aryl Hydrocarbon Receptor and Cyp1 Genes

Abstract

Polycyclic aromatic hydrocarbons are ubiquitous air pollutants, with additional widespread exposure in the diet. PAH exposure has been linked to adverse birth outcomes and long-term neurological consequences. To understand genetic differences that could affect susceptibility following developmental exposure to polycyclic aromatic hydrocarbons, we exposed mice with variations in the aryl hydrocarbon receptor and the three CYP1 enzymes from gestational day 10 (G10) to weaning at postnatal day 25 (P25). We found unexpectedly high neonatal lethality in high-affinity Ahr^bCyp1b1(-/-) knockout mice compared with all other genotypes. Over 60% of BaP-exposed pups died within their first 5 days of life. There was a significant effect of BaP on growth rates in surviving pups, with lower weights observed from P7 to P21. Again, Ahr^bCyp1b1(-/-) knockout mice were the most susceptible to growth retardation. Independent of treatment, this line of mice also had impaired development of the surface righting reflex. We used high-resolution mass spectrometry to measure BaP and metabolites in tissues from both dams and pups. We found the highest BaP levels in adipose from poor-affinity Ahr^dCyp1a2(-/-) dams and identified three major BaP metabolites (BaP-7-OH, BaP-9-OH, and BaP-4,5-diol), but our measurements were limited to a single time point. Future work is needed to understand BaP pharmacokinetics in the contexts of gestation and lactation and how differential metabolism leads to adverse developmental outcomes.

Keywords: CYP1A1; CYP1A2; CYP1B1; aryl hydrocarbon receptor; benzo[a]pyrene; neonatal lethality; polycyclic aromatic hydrocarbons.

Figure 1

Dam weights at the beginning and end of BaP treatment. There was a significant effect of genotype at baseline (G10) for poor-affinity Ahr^dCyp1a2(-/-) dams and at weaning for high-affinity Ahr^bCyp1a1(-/-) and Ahr^bCyp1b1(-/-) mice. N = 10–15 litters per group; a = significantly different from C57BL/6J controls; b = significantly different from C57BL/6J; p < 0.05.

Figure 2

Neonatal lethality in BaP-treated wild type and knockout lines. Gestational and lactational treatment with BaP resulted in high litter losses for the Ahr^bCyp1b1(-/-) line. N = 10–15 litters per group; a = significantly different from C57BL/6J controls; p < 0.01.

Figure 3

(A) Weight gain in offspring of the control and BaP-exposed offspring. There was a significant main effect of BaP treatment on weight in pups and a slightly slower growth rate in BaP treatment groups. All genotypes are pooled into one treatment group. N = 10–15 litters per genotype; *** p < 0.001. (B) Susceptible genotypes: there was a significant gene x treatment interaction for weight gain. Ahr^bCyp1a1(-/-), poor-affinity Ahr^dCyp1a2(-/-) and Ahr^bCyp1b1(-/-) knockout mice all weighed less than the controls at P7. At P14 and P21, the effects persisted in the Ahr^bCyp1a1(-/-) and Ahr^bCyp1b1(-/-) knockout mice. N = 10–15 litters per group; a = significantly different from controls at P7; b = significantly different from controls at P14; c = significantly different from controls at P21; p < 0.001 for all time points.

Figure 3

(A) Weight gain in offspring of the control and BaP-exposed offspring. There was a significant main effect of BaP treatment on weight in pups and a slightly slower growth rate in BaP treatment groups. All genotypes are pooled into one treatment group. N = 10–15 litters per genotype; *** p < 0.001. (B) Susceptible genotypes: there was a significant gene x treatment interaction for weight gain. Ahr^bCyp1a1(-/-), poor-affinity Ahr^dCyp1a2(-/-) and Ahr^bCyp1b1(-/-) knockout mice all weighed less than the controls at P7. At P14 and P21, the effects persisted in the Ahr^bCyp1a1(-/-) and Ahr^bCyp1b1(-/-) knockout mice. N = 10–15 litters per group; a = significantly different from controls at P7; b = significantly different from controls at P14; c = significantly different from controls at P21; p < 0.001 for all time points.

Figure 4

Spleen weights at P25. There was a significant gene x treatment interaction, with BaP-exposed C57BL/6J offspring having lower spleen weights, whereas all other genotypes showed no evidence of immune suppression. N = 10–15 litters per group; a = significantly different from corn oil-treated controls; p < 0.05.

Figure 5

(A) Surface righting reflex at P5: impairments were seen in BaP-exposed high-affinity Ahr^bCyp1a1(-/-), Ahr^bCyp1a2(-/-), and corn-oil treated Ahr^bCyp1b1(-/-) mice. N = 10–15 litters per group; a = significantly slower than BaP-treated C57BL/6J mice, p < 0.05; b = significantly slower than corn oil-treated C57BL/6J, p < 0.01. (B) Surface righting reflex at P7: impairments were seen in BaP-exposed high-affinity Ahr^bCyp1a1(-/-) and all Ahr^bCyp1b1(-/-) mice. Both groups of Ahr^dCyp1a2(-/-) mice were faster than the controls. N = 10–15 litters per group; a = significantly slower than all other genotypes, p < 0.001; b = significantly slower than BaP-treated C57BL/6J, p < 0.001; c = significantly faster than BaP-treated C57BL/6J, p < 0.001. (C) Surface righting reflex at P10: impairments were seen in BaP-exposed high-affinity Ahr^bCyp1a1(-/-) and corn-oil treated Ahr^bCyp1b1(-/-) mice. In contrast, BaP-exposed Ahr^bCyp1a2(-/-) mice outperformed controls. N = 10–15 litters per group; a = significantly slower than all other genotypes, p < 0.001; b = significantly faster than BaP-treated C57BL/6J, p < 0.001; c = significantly slower than corn oil-treated Ahr^bCyp1a1(-/-), p < 0.01.

Figure 5

(A) Surface righting reflex at P5: impairments were seen in BaP-exposed high-affinity Ahr^bCyp1a1(-/-), Ahr^bCyp1a2(-/-), and corn-oil treated Ahr^bCyp1b1(-/-) mice. N = 10–15 litters per group; a = significantly slower than BaP-treated C57BL/6J mice, p < 0.05; b = significantly slower than corn oil-treated C57BL/6J, p < 0.01. (B) Surface righting reflex at P7: impairments were seen in BaP-exposed high-affinity Ahr^bCyp1a1(-/-) and all Ahr^bCyp1b1(-/-) mice. Both groups of Ahr^dCyp1a2(-/-) mice were faster than the controls. N = 10–15 litters per group; a = significantly slower than all other genotypes, p < 0.001; b = significantly slower than BaP-treated C57BL/6J, p < 0.001; c = significantly faster than BaP-treated C57BL/6J, p < 0.001. (C) Surface righting reflex at P10: impairments were seen in BaP-exposed high-affinity Ahr^bCyp1a1(-/-) and corn-oil treated Ahr^bCyp1b1(-/-) mice. In contrast, BaP-exposed Ahr^bCyp1a2(-/-) mice outperformed controls. N = 10–15 litters per group; a = significantly slower than all other genotypes, p < 0.001; b = significantly faster than BaP-treated C57BL/6J, p < 0.001; c = significantly slower than corn oil-treated Ahr^bCyp1a1(-/-), p < 0.01.

Figure 6

(A) Negative geotaxis at P7: high-affinity Ahr^bCyp1a1(-/-) and Ahr^bCyp1a2(-/-) knockouts had the shortest latencies to complete the 180° turn. N = 10–15 litters per group; a = significantly different from all other genotypes, p < 0.001. (B) Negative geotaxis at P10: high-affinity Ahr^bCyp1a1(-/-) and Ahr^bCyp1a2(-/-) knockouts had the shortest latencies to complete the 180° turn. N = 10–15 litters per group; a = significantly different from all other genotypes, p < 0.001. (C) Negative geotaxis at P14: high-affinity Ahr^bCyp1a1(-/-) and Ahr^bCyp1a2(-/-) knockouts had the shortest latencies to complete the 180° turn. N = 10–15 litters per group; a = significantly different from all other genotypes, p < 0.001.

Figure 6

(A) Negative geotaxis at P7: high-affinity Ahr^bCyp1a1(-/-) and Ahr^bCyp1a2(-/-) knockouts had the shortest latencies to complete the 180° turn. N = 10–15 litters per group; a = significantly different from all other genotypes, p < 0.001. (B) Negative geotaxis at P10: high-affinity Ahr^bCyp1a1(-/-) and Ahr^bCyp1a2(-/-) knockouts had the shortest latencies to complete the 180° turn. N = 10–15 litters per group; a = significantly different from all other genotypes, p < 0.001. (C) Negative geotaxis at P14: high-affinity Ahr^bCyp1a1(-/-) and Ahr^bCyp1a2(-/-) knockouts had the shortest latencies to complete the 180° turn. N = 10–15 litters per group; a = significantly different from all other genotypes, p < 0.001.

Figure 7

(A,B) Sample chromatograms for benzo[a]pyrene. (A) Chromatograms from the BaP mass channel (252.097 m/z) for all studied genotypes. All samples are from adipose tissue. (B) Chromatograms from the BaP mass channel (252.097 m/z) from a BaP in acetonitrile standard (purple) and samples from all four tissue types (adipose in black, liver in green, cerebellum in red, and cortex in blue) from the same Ahr^dCyp1a2(-/-) pup.

Figure 8

(A) Example chromatograms showing the probable detection of BaP metabolites in Ahr^dCyp1a2(-/-) liver (red) and adipose (orange) samples. Standards for BaP-7-OH (blue) and BaP-9-OH (black) are shown. (B) Example chromatograms showing the probable detection of the BaP-4,5-diol metabolite in a C57B/6J cerebellum (CB) sample (red) and the standards for the metabolite (black).