Sex-specific autistic endophenotypes induced by prenatal exposure to valproic acid involve anandamide signalling

Abstract

Background and purpose: Autism spectrum disorder (ASD) is more commonly diagnosed in males than in females. Prenatal exposure to the antiepileptic drug valproic acid (VPA) is an environmental risk factor of ASD. Male rats prenatally exposed to VPA show socio-emotional autistic-like dysfunctions that have been related to changes in the activity of the endocannabinoid anandamide. Here, we have investigated if prenatal VPA induced sex-specific autistic endophenotypes involving anandamide signalling.

Experimental approach: We studied sex-specific differences in the ASD-like socio-emotional, cognitive and repetitive symptoms displayed during development of Wistar rats of both sexes, prenatally exposed to VPA. The involvement of anandamide was followed by Western blotting of cannabinoid CB₁ receptors and by inhibiting its metabolism.

Key results: Female rats were less vulnerable to the deleterious effects of prenatal VPA exposure on social communication, emotional reactivity and cognitive performance than male rats. Conversely, as observed in male rats, prenatal VPA exposure induced selective deficits in social play behaviour and stereotypies in the female rat offspring. At the neurochemical level, prenatal VPA exposure altered phosphorylation of CB₁ receptors in a sex-specific, age-specific and tissue-specific manner. Enhancing anandamide signalling through inhibition of its degradation reversed the behavioural deficits displayed by VPA-exposed animals of both sexes.

Conclusions and implications: These findings highlight sexually dimorphic consequences of prenatal VPA exposure that may be related to sex-specific effects of VPA on endocannabinoid neurotransmission in the course of development and introduce a new therapeutic target for reversing autistic-like symptoms in both sexes.

Figure 1

Timeline of the behavioural (A) and biochemical (B) experiments.

Figure 2

Sex‐specific effects of prenatal VPA exposure on social communication and social discrimination in the infant rat offspring. (A) VPA‐exposed male but not female pups vocalized significantly less compared with SAL‐exposed pups (male: SAL, n = 13 and VPA, n = 17; female: SAL, n = 13 and VPA, n = 16). When tested in the homing behaviour test, the male but not the female offspring prenatally exposed to VPA displayed a lower latency to reach the home cage bedding (B) and spent less time in the nest area (C) compared with SAL‐exposed male animals (male: SAL, n = 24 and VPA, n = 26; female: SAL, n = 10 and VPA, n = 14). Data are means ± SEM. *P < 0.05, significantly different from SAL group; two‐way ANOVA with Student–Newman–Keuls post hoc test.

Figure 3

Effects of prenatal VPA exposure on core and secondary autistic‐like features in the male and female rat offspring at PND 35. VPA‐exposed male and female rats responded to play solicitation with an increased frequency of evasion (A) and partial rotation (B) compared with SAL‐exposed animals (male: SAL, n = 13 and VPA, n = 13; female: SAL, n = 13 and VPA, n = 15). VPA‐exposed males but not females showed stereotypic behaviours in the hole board test (C) compared with SAL‐exposed male animals (male: SAL, n = 12 and VPA, n = 12; female: SAL, n = 15 and VPA, n = 15). VPA‐exposed males but not females spent less time in the open arms (D) and made less open entries (E) in the elevated plus maze test compared with SAL‐exposed male animals (male: SAL, n = 23 and VPA, n = 21; female: SAL, n = 13 and VPA, n = 13). No differences among groups were found in the object recognition test (F) (male: SAL, n = 7 and VPA, n = 11; female: SAL, n = 10 and VPA, n = 11). Data are means ± SEM. *P < 0.05, significantly different from SAL group; ^# P < 0.05, significantly different from SAL group; two‐way ANOVA with Student–Newman–Keuls post hoc test.

Figure 4

Effects of prenatal VPA exposure on core and secondary autistic‐like features in the male and female rat offspring at PND 90. Prenatal VPA exposure reduced sociability of male but not female rats in the three‐chamber test (A) (male: SAL, n = 9 and VPA, n = 9; female: SAL, n = 8 and VPA, n = 7) while it induced stereotypic behaviour in the hole board test both in male and female animals (B) (male: SAL, n = 12 and VPA, n = 12; female: SAL, n = 15 and VPA, n = 15). VPA‐exposed male but not female rats spent less time in the open arms (C) and made less open‐arm entries (D) in the elevated plus maze test (male: SAL, n = 13 and VPA, n = 10; female: SAL, n = 16 and VPA, n = 16). No differences among groups were found in the object recognition test (E) (male: SAL, n = 12 and VPA, n = 10; female: SAL, n = 15 and VPA, n = 16). Male but not female rats prenatally exposed to VPA showed impaired social discrimination (F) (male: SAL, n = 11 and VPA, n = 11; female: SAL, n = 10 and VPA, n = 10). No differences among groups were found in the acquisition trial of the inhibitory avoidance test (G). However, male but not female rats prenatally exposed to VPA showed impaired memory consolidation during the retention session (H) (male: SAL, n = 15 and VPA, n = 15; female: SAL, n = 12 and VPA, n = 14). Data are means ± SEM. *P < 0.05, significantly different from SAL group; ^# P < 0.05, significantly different from SAL group; two‐way ANOVA with Student–Newman–Keuls post hoc test.

Figure 5

Sex‐specific, age‐specific and tissue‐specific changes in phosphorylation of CB₁ receptors induced by prenatal VPA exposure. VPA‐exposed male but not female rats displayed altered phosphorylation of CB₁ receptors in dorsal striatum [PND 35 (A), male: SAL, n = 5 and VPA, n = 5; female: SAL, n = 5 and VPA, n = 6; PND 90 (E), male: SAL, n = 5 and VPA, n = 5; female: SAL, n = 5 and VPA, n = 6] and hippocampus [PND 35 (C), male: SAL, n = 10 and VPA, n = 8; female: SAL, n = 5 and VPA, n = 6; PND 90 (G), male: SAL, n = 8 and VPA, n = 9; female: SAL, n = 5 and VPA, n = 6]. VPA‐exposed male but not female rats displayed reduced phosphorylation of CB₁ receptors at PND 90 in the amygdala (H) (male: SAL, n = 6 and VPA, n = 10; female: SAL, n = 5 and VPA, n = 6). Conversely, VPA‐exposed female rats displayed increased phosphorylation of CB₁ receptors in the prefrontal cortex only at PND 35 (B) (male: SAL, n = 6 and VPA, n = 6; female: SAL, n = 4 and VPA, n = 5). Data are means ± SEM. *P < 0.05, significantly different from SAL group; ^# P < 0.05, significantly different from SAL group; two‐way ANOVA with Student–Newman–Keuls post hoc test.

Figure 6

Pharmacological interference with anandamide hydrolysis corrects the partial behavioural alterations displayed by VPA‐exposed female rats and the cognitive deficit found in VPA‐exposed male rats. The administration of URB597 normalized the altered pattern of social play behaviour (A, B) displayed by VPA‐exposed female rats at PND 35 [SAL‐vehicle (VEH), n = 5; SAL‐URB, n = 7; VPA‐VEH, n = 6; and VPA‐URB, n = 8] and their stereotypic behaviour at PND 90 (C) (SAL‐VEH, n = 7; SAL‐URB, n = 8; VPA‐VEH, n = 9; and VPA‐URB, n = 10). URB597 also reversed the deficits displayed by VPA‐exposed male rats in the social discrimination (D) (SAL‐VEH, n = 8; VPA‐VEH, n = 9; SAL‐URB, n = 8; and VPA‐URB, n = 8) and inhibitory avoidance (E, F) (SAL‐VEH, n = 14; VPA‐VEH, n = 12; SAL‐URB, n = 8; and VPA‐URB, n = 7) tests. Data are means ± SEM. *P < 0.05, significantly different from SAL‐VEH group; ^$ P < 0.05, significantly different from VPA‐VEH group; two‐way ANOVA with Student–Newman–Keuls post hoc test.