Fetal blockade of nicotinic acetylcholine transmission causes autism-like impairment of biological motion preference in the neonatal chick

Abstract

Several environmental chemicals are suspected risk factors for autism spectrum disorder (ASD), including valproic acid (VPA) and pesticides acting on nicotinic acetylcholine receptors (nAChRs), if administered during pregnancy. However, their target processes in fetal neuro-development are unknown. We report that the injection of VPA into the fetus impaired imprinting to an artificial object in neonatal chicks, while a predisposed preference for biological motion (BM) remained intact. Blockade of nAChRs acted oppositely, sparing imprinting and impairing BM preference. Beside ketamine and tubocurarine, significant effects of imidacloprid (a neonicotinoid insecticide) appeared at a dose ≤1 ppm. In accord with the behavioral dissociations, VPA enhanced histone acetylation in the primary cell culture of fetal telencephalon, whereas ketamine did not. VPA reduced the brain weight and the ratio of NeuN-positive cells (matured neurons) in the telencephalon of hatchlings, whereas ketamine/tubocurarine did not. Despite the distinct underlying mechanisms, both VPA and nAChR blockade similarly impaired imprinting to biological image composed of point-light animations. Furthermore, both impairments were abolished by postnatal bumetanide treatment, suggesting a common pathology underlying the social attachment malformation. Neurotransmission via nAChR is thus critical for the early social bond formation, which is hindered by ambient neonicotinoids through impaired visual predispositions for animate objects.

Keywords: autism spectrum disorder; biological motion; imprinting; neonicotinoid; valproic acid.

Fig. 1

Suppression of the spontaneous fetal movement by VPA and other chemicals. A, B) Ballistography recording (experiment 1) and blockade by sodium valproate (VPA). Record stylus placed against the eggshell surface detected the miniature vibration caused by spontaneous fetal movement. C) The dose was experimentally searched for each agent so that fetal movement was similarly suppressed on E14, wherein the treatment gave rise to normal hatch rate and successful training of the hatchlings. Asterisks indicate the significant difference from the control by multiple regression analysis; ^*, P < 0.05, ^**, P < 0.01, ^***, P < 0.001 in this and the following figures. Number in parenthesis indicates the sample size.

Fig. 2

BM preference and imprinting; training and test procedures (experiment 2). Chicks were trained by an artifact object (rotating red toy) and tested by binary choices for BM (walking motion vs. linear motion) and imprinting (red toy vs. yellow); stay time difference (s) was used for the scores.

Fig. 3

BM preference and imprinting; VPA and nAChR blockers (experiment 2). A) BM and B) imprinting score of the treated chicks are shown in box-plot of the median and quadrants with superimposed individual data. Horizontal colored lines indicate the critical levels of the control data determined by bootstrap computation; average (blue), 95% (brown), and 99% (red) confidence levels; see the Supplementary Fig. S4 for details. C) Imprinting scores were plotted against BM scores. Distinct spectrums were found between VPA and other agents. Asterisks (^*, ^**, ^***) indicate significance levels by multiple regression analysis. Sharp marks (#, ##) indicate significance levels based on the bootstrapping analysis.

Fig. 4

BM preference and imprinting; IMI (experiment 2). A) IMI suppressed BM if injected on E0 or E14. B, C) Although not significant, imprinting was slightly suppressed by IMI 50 μg in both injection days.

Fig. 5

Imprinting of BM image; training and test procedures (experiment 3). After the first (P1) training using an artifact (rotating red toy) and BM test, chicks received the second day (P2) training, wherein two images of point light animations were presented with yellow in walking (Wp[yellow]) and red in linear rigid motion (Lp[red]).

Fig. 6

Imprinting of BM image was similarly impaired by both VPA and nAChR blockade. The 2-step imprinting paradigm revealed a common hypoplasia by E14 injection of VPA and nAChR blockade (experiment 3). A) BM scores confirmed the experiment 2. B, D) Both VPA and ketamine chicks showed the impaired formation of preference to Wp[yellow]. Notably, the tubocurarine chicks showed similar patterns of imprinting versus BM plots to the ketamine. C, E) VPA chicks showed the impaired imprinting to artifact red toy as in experiment 2.

Fig. 7

Distinct effects of VPA on histone acetylation. VPA but not ketamine enhanced the acetylation level of histone H3K27 in primary cell culture made from E14 telencephalon (experiment 4).

Fig. 8

Distinct effects of VPA on brain size, and matured neuron ratio. Aa) VPA but not ketamine/tubocurarine reduced the brain weight and Ba) the ratio of NeuN positive cells measured by isotropic fractionation; Ab) body weight and Bb) total cell number were not affected. Male brains were significantly heavier, but no interaction was found between the sex and the agents. No significant effects of sex were detected on the cell number and the NeuN ratio.

Fig. 9

Scenario of the adaptive socialization through imprinting.