Intact maternal buffering of stress response in infant rats despite altered responsivity towards maternal olfactory cues in the valproic acid model of autism-like behavior

Abstract

Disrupted processing of social cues and altered social behaviors are among the core symptoms of autism spectrum disorders (ASDs), and they emerge as early as the first year of life. These differences in sensory abilities may affect the ability of children with ASDs to securely attach to a caregiver and experience caregiver buffering of stress. Prenatal exposure to valproic acid (VPA) has been used to model some aspects of ASDs in rodents. Here, we asked whether prenatal VPA exposure altered infant rats' behavioral responsivity to maternal olfactory cues in an Odor Preference Test (OPT) and affected maternal buffering of infants' stress responsivity to shock. In the odor preference test, 1-week old rats treated with VPA during pregnancy appeared to have impaired social recognition and/or may be less motivated to approach social odors in early infancy. These effects were particularly prominent in female pups. In 2-week old rats, VPA-exposed pups and saline-exposed pups showed similar preferences for home cage bedding. Although VPA-exposed pups may initially have a deficit in this attachment-related behavior they do recover typical responses to home cage bedding in later infancy. Both control and VPA-exposed pups showed robust stress hormone responses to repeated shocks, an effect which was blocked when a calm mother was present during shock exposure. No sex differences in the effect of maternal presence on the stress response to shock and no interactions between sex and prenatal drug exposure were observed. Although VPA-exposed pups may show impaired responsivity to maternal cues in early infancy, maternal presence is still capable of regulating the stress response in VPA-exposed pups. In this study we demonstrate the importance of utilizing multiple batteries of tests in assessing behavior, dissecting the behavior on one test into different components. Our results inform about the underlying behavioral characteristics of some of the ASD phenotypes, including sex differences reported by clinical studies, and could shed light on potential opportunities for intervention.

Keywords: autism; infancy; maternal buffering; rat; social behavior; valproic acid.

Figure 1

(A) P6–7: latency to enter zones. Data expressed as mean ± standard error of the mean. We observed significant main effects of bedding type, p < 0.0001, and drug exposure, p = 0.007. VPA-exposed pups showed a longer latency to enter the clean bedding zone, p = 0.008. (B) P6–7: time spent in zones. Data expressed as mean ± standard error of the mean. We observed significant main effects of bedding type, p < 0.0001, drug exposure, p < 0.0001, and a significant interaction, p = 0.014. VPA-exposed pups spent significantly less time in the SB zone, p < 0.0001. (C) P6–7: number of zone entries. Data expressed as mean ± standard error of the mean. We observed significant main effects of bedding type, p < 0.04, drug exposure, p < 0.0001, and a significant interaction, p = 0.02. VPA-exposed pups made significantly fewer entries into the social bedding zone, p = 0.01. (D) P6–7: distance traveled during test. Data expressed as mean ± standard error of the mean. VPA-exposed pups did not differ from saline-exposed pups in their distance traveled during the test. *p = 0.01–0.05; **p = 0.001–0.01; ***p = 0.0001–0.001; ****p < 0.0001; ns = p > 0.05.

Figure 2

(A,B) P6–7 Sex differences: latency to enter zones. Data expressed as mean ± standard error of the mean. We observed a significant main effect of bedding type in females (A), p < 0.0001, and males (B), p < 0.0001. There was a significant main effect of drug exposure in males alone, p = 0.003. (C,D) P6–7 Sex differences: time spent in zones. Data expressed as mean ± standard error of the mean. We observed a significant main effect of bedding in both females (C), p < 0.0001, and males (D), p < 0.0001. In females, we also observed a significant main effect of drug exposure, p = 0.015, and a significant interaction, p = 0.02. Female VPA-exposed pups spent significantly less time in the SB zone than female saline-exposure pups, p = 0.005. (E,F) P6–7 Sex differences: number of zone entries. Data expressed as mean ± standard error of the mean. We observed significant main effects of bedding type and drug exposure in females (E), p < 0.0001 and p = 0.008, and males (F), p < 0.0001 and p = 0.0013. There was also a significant interaction in females, p = 0.04. Both female, p = 0.006, and male, p = 0.0075, VPA-exposed pups made significantly fewer entries into the SB zone. *p = 0.01–0.05; **p = 0.001–0.01; ***p = 0.0001–0.001; ****p < 0.0001; ns = p > 0.05.

Figure 3

(A) P13: latency to enter zones. Data expressed as mean ± standard error of the mean. We observed a significant main effect of bedding type, p < 0.0001. Both VPA-exposed and saline exposed-pups showed a shorter latency to enter the social bedding zone than the clean bedding zone, p’s < 0.0001. (B) P13: time spent in zones. Data expressed as mean ± standard error of the mean. We observed a significant main effects of bedding type, p < 0.0001. Both VPA-exposed pups and saline-exposed pups spent significantly more time in the social bedding zone than the clean bedding zone, p’s < 0.0001. (C) P13: number of zone entries. Data expressed as mean ± standard error of the mean. We observed a significant main effects of bedding type, p < 0.00001. VPA-exposed pups and saline-exposed pups made significantly fewer entries into the social bedding zone relative to the clean bedding zone p’s < 0.00001. (D) P13: distance traveled during test. Data expressed as mean ± standard error of the mean. VPA-exposed pups did not differ from saline-exposed pups in their distance traveled during the test, p = 0.64. ****p < 0.0001; ns = p > 0.05.

Figure 4

(A,B) P13 Sex differences: latency to enter zones. Data expressed as mean ± standard error of the mean. We observed a significant main effect of bedding type in females (A), p < 0.0001, and males (B), p < 0.0001. (C,D) P13 Sex differences: time spent in zones. Data expressed as mean ± standard error of the mean. We observed a significant main effect of bedding in both females (C), p < 0.0001, and males (D), p < 0.0001. (E,F) P13 Sex differences: number of zone entries. Data expressed as mean ± standard error of the mean. We observed a significant main effect of bedding in both females (C), p < 0.0001, and males (D), p < 0.0001. ***p = 0.0001–0.001; ****p < 0.0001; ns = p > 0.05.

Figure 5

(A) Maternal buffering of stress response to shock at P6–7. Data expressed as mean ± standard error of the mean. We observed a significant main effect of experimental condition, p < 0.0001 on pup corticosterone levels. Both saline and VPA-treated pups exposed to shock showed significantly higher corticosterone levels than baseline groups (p’s < 0.0001) and the groups exposed to shock in the presence of a calm mother (p’s < 0.0001). (B) Maternal buffering of stress response to shock in females at P6–7. Data expressed as mean ± standard error of the mean. We observed a significant main effect of experimental condition, p < 0.0001 on female pup corticosterone levels. Both saline and VPA-treated female pups exposed to shock showed significantly higher corticosterone levels than baseline groups (p’s < 0.0001) and the groups exposed to shock in the presence of a calm mother (p’s < 0.0001). (C) Maternal buffering of stress response to shock in males at P6–7. Data expressed as mean ± standard error of the mean. We observed a significant main effect of experimental condition, p < 0.0001 on male pup corticosterone levels. Both saline and VPA-treated male pups exposed to shock showed significantly higher corticosterone levels than baseline groups (p’s < 0.0001) and the groups exposed to shock in the presence of a calm mother (saline p = 0.0001, VPA p = 0.0003). ***p = 0.0001–0.001; ****p < 0.0001; ns = p > 0.05.