BRCA1 protein dose-dependent risk for embryonic oxidative DNA damage, embryopathies and neurodevelopmental disorders with and without ethanol exposure

Abstract

Although widely known as a tumor suppressor, the breast cancer 1 susceptibility protein (BRCA1) is also important in development, where it regulates fetal DNA repair pathways that protect against DNA damage caused by physiological and drug-enhanced levels of reactive oxygen species (ROS). We previously showed that conditional heterozygous (+/-) knockout (cKO) mouse embryos with a minor 28% BRCA1 deficiency developed normally in culture, but when exposed to the ROS-initiating drug, alcohol (ethanol, EtOH), exhibited embryopathies not evident in wild-type (+/+) littermates. Herein, we characterized a directBrca1 +/- knockout (KO) model with a 2-fold greater (58%) reduction in BRCA1 protein vs. the cKO model. We also characterized and compared learning & memory deficits in both the cKO and KO models. Even saline-exposed Brca1 +/- vs. +/+ KO progeny exhibited enhanced oxidative DNA damage and embryopathies in embryo culture and learning & memory deficits in females in vivo, which were not observed in the cKO model, revealing the potential pathogenicity of physiological ROS levels. The embryopathic EtOH concentration for cultured direct KO embryos was half that for cKO embryos, and EtOH affected Brca1 +/+ embryos only in the direct KO model. The spectrum and severity of EtOH embryopathies in culture were greater in both Brca1 +/- vs. +/+ embryos, and direct KO vs. cKO +/- embryos. Motor coordination deficits were evident in both male and female Brca1 +/- KO progeny exposed in utero to EtOH. The results in our direct KO model with a greater BRCA1 deficiency vs. cKO mice provide the first evidence for BRCA1 protein dose-dependent susceptibility to developmental disorders caused by physiological and drug-enhanced oxidative stress.

Keywords: Alcohol (ethanol, EtOH); Breast cancer 1 susceptibility gene (Brca1); Conditional knockout (cKO) mouse model; DNA damage and repair; Developmental disorders; Fetal alcohol spectrum disorders (FASD); Knockout (KO) mouse model; Reactive oxygen species (ROS).

Graphical abstract

Fig. 1

Enhanced oxidative DNA damage in embryos dependent upon decreasing BRCA1 protein levels in two knockout mouse models following in utero exposure to saline or ethanol (EtOH). ConditionalBrca1 +/- knockout embryos (cKO) with a minor BRCA1 protein deficiency are compared to direct knockouts (KO) with a more severe BRCA1 deficiency. The data for the KO embryos are new, while those for the cKO embryos were previously published [30] and are provided here for comparison. EmbryonicBRCA1 protein levels in the Brca1cKO model on gestational day (GD) 10 (Panel A), and the KO model on GD 12 (Panel B), were determined by western blot analysis at 220 kDa, normalized to GAPDH at 37 kDa for the loading control. (Panel A) +/- Brca1cKOembryos exhibited a minor 28% reduction in BRCA1 protein compared to +/+ embryos (p < 0.05). (Panel B) +/- Brca1KOembryos exhibited a 58% reduction in BRCA1 protein compared to +/+ KO littermates (p < 0.0001), and 2-fold greater than the BRCA1 reduction in +/- cKO embryos. Significant differences were determined by Student's t-test. The number of mice is shown in parentheses above each bar. (Panels C&D) Relative 8-oxoguanine (8-oxoG)lesion levels in +/+ and +/- Brca1cKOembryos(Panel C), and in +/+ and +/- Brca1KOembryos(Panel D), sampled 6 h after maternal i.p. injection with saline or EtOH (4 g/kg). 8-OxoG levels were determined by ELISA and normalized to μg of DNA loaded in the assay. (Panel C) In the cKOmodel, EtOH-exposed Brca1 +/+ embryos demonstrated enhanced 8-oxoG levels compared to saline +/+ controls (p < 0.05), which were further elevated in EtOH-exposed +/- littermates (p < 0.001). A genotype effect was also evident, where EtOH-exposed +/- cKO embryos had 13% higher 8-oxoG levels compared to +/+ littermates (p < 0.05). (Panel D) Saline-exposed Brca1 +/- KOembryos exhibited 50% higher 8-oxoG levels compared to +/+ littermates (p < 0.05), in contrast to saline-exposed cKO embryos, in which there was no genotype effect. 8-OxoG levels were enhanced in both saline-exposed (p < 0.01) and EtOH-exposed (p < 0.01) Brca1 +/- KO embryos compared to their respective saline- and EtOH-exposed Brca1 +/+ littermates. However, there was no increase in +/- Brca1 KO embryos exposed to EtOH vs. saline, unlike in KO embryos, possibly because of an already maximal level of DNA damage in the saline-exposed +/- embryos. Significant differences were determined using a two-way ANOVA with a Bonferroni post-hoc test. The number of mice is shown in parentheses above each bar, where results represent primarily female embryos (plotted by sex in the Supplementary Materials, Fig. S1). (Panels E&F) Relative γH2AX levels in Brca1cKO embryos (Panel E), and in Brca1KO embryos (Panel F), 6 h after saline or EtOH exposure, were determined by western blot analysis at 17 kDa, normalized to GAPDH at 37 kDa for the loading control. (Panel E) EtOH exposure in Brca1 +/- cKOembryos resulted in a 4.2-fold increase in γH2AX levels compared to saline-exposed +/- controls (p < 0.001). A genotype effect was also evident, where EtOH-exposed +/- cKO embryos had 75% higher γH2AX levels compared to +/+ littermates (p < 0.05). (Panel F) Saline-exposed Brca1 +/- KOembryos exhibited 46% higher γH2AX levels compared to +/+ littermates (p < 0.05), whereas no genotype effect was evident in saline-exposed cKO embryos. A larger genotype effect was observed in KO +/- embryos with EtOH exposure, with a 3.5-fold increase in γH2AX levels compared to +/+ littermates (p < 0.0001). A drug effect also was evident in +/- Brca1 embryos, which exhibited a 3-fold increase in γH2AX levels when exposed to EtOH vs. saline (p < 0.0001). Significant differences were determined using two-way ANOVA with a Bonferroni post-hoc test. The number of embryos is shown in parentheses above each bar, with roughly equal numbers of males and females, and no sex difference (plotted by sex in the Supplementary Materials, Fig. S2). Embryos were selected from at least 3 litters to minimize any impact of a litter effect.

Fig. 2

Effect of ethanol on embryonic survival and heart rate in culture comparing Brca1 conditional knockouts (cKO) and direct knockouts (KO). For cKO embryos, Cre^+/- male mice were mated overnight with Brca1^LoxP/LoxP female mice (plug designated GD 1) (see Methods), and Brca1 +/- direct KO mice were mated at about 09:00 for 2–4 h (plug designated GD 1). Breeding for both cKO and KO models resulted in +/+ and +/- Brca1 littermates respectively exhibiting normal and deficient levels of BRCA1 protein, with a greater BRCA1 deficiency in KO vs. cKO embryos (see Fig. 1A&B). On GD 9, Brca1 +/+ and +/- 7-8-somite cKO embryos were explanted and incubated for 24 h with ethanol (EtOH (4 mg/ml for cKO embryos) or its saline vehicle. On GD 9.5, Brca1 +/+ and +/- 6-8-somite KO embryos were explanted and incubated for 24 h with EtOH (2 mg/ml for direct KO embryos, as 4 mg/mL caused lethality, Panel A) or its saline vehicle. Embryos were evaluated under a microscope for morphological and functional parameters. (Panel A) In the direct KO model, embryonic survival throughout the 24 h culture period was not significantly affected by EtOH at 2 mg/mL, while a concentration of 4 mg/mL caused 100% lethality. (Panel B) In the cKO model, independent of Brca1 genotype, EtOH enhanced the heart rate relative to saline controls of the same genotype (p < 0.05). The data for the cKO embryos were previously published [30] and are provided here for comparison. (Panel C) Unlike in the cKO embryos, EtOH exposure had no effect on embryonic heart rate in the KO embryos of either Brca1 genotype, possibly due to the lower EtOH concentration used in the KO model. Significant differences were determined for binomial data using Fisher's exact test and for continuous data using two-way ANOVA with a Bonferroni post-hoc test. The numbers in parentheses above each bar indicate the number of embryos, which were selected from at least 3 litters to minimize any impact of a litter effect.

Fig. 3

BRCA1 protein dose-dependent susceptibility to morphological developmental disorders in cultured embryos exposed to saline or ethanol (EtOH). The mating procedures for the conditional knockout (cKO) model and the Brca1 +/- direct knockout (KO) model are detailed in the legend for Fig. 2 and in the Methods. Breeding for both the cKO and KO models resulted in +/+ and +/- Brca1 littermates respectively exhibiting normal and deficient levels of BRCA1 protein, with a greater BRCA1 deficiency in KO vs. cKO embryos (see Fig. 1A&B). On GD 9, Brca1 +/+ and +/- 7-8-somite cKO embryos were explanted and incubated for 24 h with EtOH (4 mg/ml for cKO embryos) or its saline vehicle. On GD 9.5, Brca1 +/+ and +/- 6-8-somite KO embryos were explanted and incubated for 24 h with EtOH (2 mg/ml for direct KO embryos, as 4 mg/mL caused lethality [Panel A]) or its saline vehicle. Embryos were evaluated under a microscope for morphological and functional parameters. The data for the KO embryos are new, while those for the cKO embryos were previously published [30] and are provided here for comparison. (Panels A&B) EtOH-exposed Brca1cKO embryos exhibited decreased anterior neuropore closure (p < 0.001) and embryonic turning (p < 0.05), relative to saline-exposed controls of the same genotype. EtOH-exposed BRCA1-deficient embryos exhibited decreased anterior neuropore closure (p < 0.05) and embryonic turning (p < 0.05) relative to EtOH-exposed +/+ embryos. (Panels G&H) In the direct KO model, an EtOH-induced decrease in anterior neuropore closure and embryonic turning was observed in both genotypes, as distinct from only in the cKO +/- embryos (p < 0.05). EtOH-exposed BRCA1-deficient cKO embryos exhibited increased susceptibility compared to +/+ littermates (p < 0.005), while a similar trend in EtOH-exposed Brca1 +/- KO embryos was not significant, likely due in part to the already significant deleterious impact of EtOH in Brca1 +/+ littermates. (Panels C&D) No genotypic or treatment effects were observed for yolk sac diameter and crown rump length measurements in the cKO model. (Panels I&J) In +/- Brca1direct KO embryos, unlike the cKO embryos, EtOH decreased the yolk sac diameter (p < 0.05) and crown rump length (p < 0.0005) relative to saline controls of the same genotype. The crown rump length of +/- Brca1direct KO embryos was also decreased compared to EtOH-exposed +/+ embryos (p < 0.05). (Panels E&F) EtOH-exposed Brca1cKO embryos exhibited decreased head length (p < 0.001) and rate of somite development (p < 0.05) relative to saline-exposed controls of the same genotype. EtOH-exposed BRCA1-deficient embryos also exhibited decreased head length (p < 0.05), and rate of somite development (p < 0.05) relative to EtOH-exposed +/+ embryos. (Panels K&L) In the direct KO model, EtOH exposure decreased head length (p < 0.05) and the rate of somite development (p < 0.0001) in both genotypes, in contrast to the cKO +/+ embryos, which were unaffected. The impact of EtOH on head length and somite development also was greater in +/- than ++ Brca1direct KO embryos (p < 0.005). Significant differences were determined for binomial data using Fisher's exact test and for continuous data using a two-way ANOVA with a Bonferroni post-hoc test. The numbers in parentheses above each bar indicate the number of embryos, which were selected from at least 3 litters to minimize any impact of a litter effect.

Fig. 4

Decreased learning and memory in BRCA1-deficient female progeny measured via passive avoidance testing following in utero exposure to a single 4 g/kg maternal dose of ethanol (EtOH) on GD 17. Brca1 cKO and KO fetuses were exposed in utero to saline or EtOH (4 g/kg) via maternal i.p. injection on GD 17. At 6, 9 and 12 weeks of age, mouse progeny were assessed in three daily trials for their latency to enter the dark chamber, where a mild footshock was received. The data for both the KO and cKO progeny have not been previously published. Latency to enter the dark chamber after the footshock for (Panel A) female Brca1conditional knockout (cKO) progeny, and (Panel B) female direct knockout (KO) progeny, is shown for the third trial of the final week. (Panels C&D) The performance of male vs. female Brca1KO progeny is shown from 9 to 12 weeks. (Panel A) In the final trial at 12 weeks of age, EtOH-exposed female Brca1cKO progeny exhibited a decreased latency to enter the dark chamber relative to +/+ littermates (p < 0.05). (Panel B) In the final trial at 12 weeks of age, female +/- Brca1KO progeny exposed to saline vehicle have a 75% decreased latency to enter the dark chamber relative to +/+ littermates (p < 0.0001). A similar 72% reduction in latency was observed in female +/- Brca1 KO progeny exposed to EtOH (p < 0.0001). (Panels C&D) The Brca1 genotypic differences for female +/- Brca1 progeny over 12 weeks is plotted (****p < 0.0001). No Brca1 genotypic differences were observed for male KO progeny. Statistical differences were determined by two-way ANOVA with a Bonferroni post-hoc test at each timepoint. The number of progeny is shown in parentheses above each bar (Panels A&B) or in the figure key (Panels C&D). Progeny were selected from at least 3 litters to minimize any impact of a litter effect.

Fig. 5

In utero EtOH exposure impaired motor coordination in Brca1 direct KO progeny.Brca1directKO fetuses were exposed in utero to saline or EtOH (4 g/kg i.p.) via maternal injection on GD 17, and motor coordination was assessed at postnatal week 5 using a rotarod apparatus consisting of a rotating rod that accelerated from 4 to 40 rpm linearly, over 5 min. Three trials were carried out with a 3-min break between. The mean latency to fall from, or hold on to, the rotating rod for 3 rotations during trial 2 and 3 was plotted. Male and female mice were analyzed separately based on sex differences observed in other behavioural tests (i.e., learning & memory, passive avoidance). There was no significant difference between the sexes (see Supplementary Materials, Fig. S3), so the data were combined. Brca1 +/- progeny exposed in utero to EtOH exhibited a 16% reduction in latency to fall compared to saline-exposed +/- controls (p < 0.001) (drug effect), and a 19% reduction in latency compared to EtOH-exposed +/+ littermates (p < 0.01) (Brca1 genotype effect). Statistical analyses were performed using two-way ANOVA with a Bonferroni post-test. The number of mice assessed is shown in parentheses. Progeny were selected from at least 3 litters to minimize any impact of a litter effect. Motor coordination was not evaluated in the cKO model.