Altered sensitivity to environmental stimuli is a core feature of autism spectrum disorder (ASD) and may increase vulnerability to stress- and anxiety-related disorders. The autonomic nervous system (ANS) mediates rapid physiological responses to stress that often translate into observable behaviors such as colonic motility (fecal pellet release), freezing, and emission of 22-kHz ultrasonic vocalizations (USVs). The valproic acid (VPA) prenatal exposure rat model reproduces several neural and behavioral features of ASD, but its relevance for studying stress susceptibility across different contexts required further examination. This study evaluated autonomic and behavioral responses to tactile, nociceptive, and social stressors in juvenile VPA-exposed and saline-treated control (CTL) rats.
Timed-pregnant Wistar females received a single intraperitoneal injection on gestational day 12.5 of sodium valproate (500 mg/kg in saline; VPA group, n = 8 females) or saline (CTL group, n = 6 females). An additional cohort of naive females produced demonstrator animals for social experiments. After birth, litters remained with dams until weaning at postnatal day (P) 21. VPA-treated offspring consistently displayed the characteristic embryological tail kink. Behavioral testing occurred across juvenile ages: tactile tests P33–P37, and nociceptive and social tests P38–P42. Cohort composition and litter clustering were accounted for in analyses.
Four experimental paradigms were used to probe stress reactivity: (a) gentle handling sessions across four days as a tactile manipulation/habituation paradigm; (b) an electro-tactile sensitivity test using incremental foot-shocks to determine tactile/nociceptive thresholds and acute responses; (c) a contextual fear conditioning protocol with repeated unpredictable footshocks to assess nociceptive stress and freezing dynamics; and (d) an emotional contagion protocol where an observer rat witnessed a cagemate demonstrator receiving shocks, to evaluate responses to negative social stimuli.
Gentle handling consisted of four 1-minute sessions per day for four days (P33–P36) during which each rat was stroked while held by the experimenter; fecal pellet counts were recorded after each session. Electro-tactile testing occurred one day after the last handling session. Rats were placed in a shock chamber with a metal grid floor and, after 90 s exploration, received 1-s rectangular electrical pulses with increasing amplitude (50–550 µA in 50 µA steps) and randomized inter-stimulus intervals (30–90 s). Detection behaviors (grid investigation, paw retraction) defined tactile threshold; jumping/fleeing defined nociceptive threshold. The test ended when the animal jumped or fled in two consecutive stimuli. Video and ultrasonic recordings were synchronized to stimuli for later analysis.
At P38, rats underwent a contextual fear conditioning session in an operant chamber. Following a 4-min baseline (B0), five 1-s footshocks of 750 µA (two times the nociceptive threshold defined in the electro-tactile test) were delivered with random inter-shock intervals up to 180 s. The session finished 4 minutes after the last shock. Freezing episodes were defined as immobility exceeding 3 s and were quantified as relative freezing time per block. Fecal pellets were counted immediately after the session. Video and high-sampling-rate ultrasonic recordings captured freezing and USV production.
The emotional contagion apparatus had adjacent shock and neutral chambers separated by a clear plexiglass wall with openings permitting visual contact. Observers (OBS) habituated to the neutral chamber on P39–P40. On P41, DEM animals in the shock chamber received five 1-s shocks of 750 µA (inter-shock intervals up to 360 s) while OBS animals could view them from the neutral chamber. Shock delivery was timed when the OBS faced the DEM to maximize attention. Audio and video were recorded under dim light during the lights-off phase. To assign call origin, vocalization power thresholds were derived from separate recordings: calls > −76 dB were considered demonstrator-origin, calls < −85 dB considered observer-origin; calls between these cutoffs (≈9.8% of calls) were excluded to minimize ambiguity.
Primary behavioral endpoints were fecal pellet counts, relative freezing time (immobility > 3 s), and production and acoustic features of ultrasonic vocalizations—particularly 22-kHz USVs, which are widely associated with stress. Video and audio files were randomly coded and analyzed blind. USV detection, curation, segmentation, and feature extraction used laboratory routines and a custom-modified DeepSqueak pipeline. Blocks following baseline (B1–B5) were defined relative to shock timing for time-resolved analysis of behavior.
During gentle handling, VPA-exposed rats displayed sustained hyperdefecation compared to CTLs. In the electro-tactile test, VPA rats showed a higher prevalence of defecation during stimulation, without corresponding changes in freezing. In contextual fear conditioning, VPA animals had a delayed onset of freezing but later exhibited enhanced freezing; moreover, the temporal coordination between freezing and defecation behaviors was disrupted in VPA rats. Across groups, rates of 22-kHz USVs correlated positively with freezing. In the emotional contagion paradigm, observing a distressed conspecific increased freezing prevalence and suppressed vocalization rates in both VPA and CTL observers; however, these effects were prolonged in VPA rats, with persistent freezing and an earlier reduction in vocalization rate.
The data indicate that prenatal VPA exposure in rats leads to heightened stress reactivity, impaired habituation to repeated tactile manipulation, and altered coordination among canonical fear-related outputs (freezing, defecation, and 22-kHz USVs). The prolonged freezing and altered vocal suppression during social observation suggest intensified or prolonged processing of social aversive cues in VPA animals. These features support using the VPA model to investigate neurobiological mechanisms linking sensory sensitivity, stress vulnerability, and social dysfunction relevant to ASD. The authors emphasize the model’s translational utility for studying stress-related comorbidity in autism.
All procedures were approved by the Ethics Committee on the Use of Animals (CEUA-UFRN, protocol nº 017/2020). Data underlying the findings are publicly available at Zenodo (https://doi.org/10.5281/zenodo.20600919). Funding sources included CAPES and CNPq; funders did not influence study design or reporting.