Perinatal exposure to pesticides alters synaptic plasticity signaling and induces behavioral deficits associated with neurodevelopmental disorders

Abstract

Increasing evidence from animal and epidemiological studies indicates that perinatal exposure to pesticides cause developmental neurotoxicity and may increase the risk for psychiatric disorders such as autism and intellectual disability. However, the underlying pathogenic mechanisms remain largely elusive. This work was aimed at testing the hypothesis that developmental exposure to different classes of pesticides hijacks intracellular neuronal signaling contributing to synaptic and behavioral alterations associated with neurodevelopmental disorders (NDD). Low concentrations of organochlorine (dieldrin, endosulfan, and chlordane) and organophosphate (chlorpyrifos and its oxon metabolite) pesticides were chronically dosed ex vivo (organotypic rat hippocampal slices) or in vivo (perinatal exposure in rats), and then biochemical, electrophysiological, behavioral, and proteomic studies were performed. All the pesticides tested caused prolonged activation of MAPK/ERK pathway in a concentration-dependent manner. Additionally, some of them impaired metabotropic glutamate receptor-dependent long-term depression (mGluR-LTD). In the case of the pesticide chlordane, the effect was attributed to chronic modulation of MAPK/ERK signaling. These synaptic alterations were reproduced following developmental in vivo exposure to chlordane and chlorpyrifos-oxon, and were also associated with prototypical behavioral phenotypes of NDD, including impaired motor development, increased anxiety, and social and memory deficits. Lastly, proteomic analysis revealed that these pesticides differentially regulate the expression of proteins in the hippocampus with pivotal roles in brain development and synaptic signaling, some of which are associated with NDD. Based on these results, we propose a novel mechanism of synaptic dysfunction, involving chronic overactivation of MAPK and impaired mGluR-LTD, shared by different pesticides which may have important implications for NDD.

Keywords: Contaminants; Kinase signaling; MAPK; Neurodevelopment; Psychiatric disorders; mGluR LTD.

Fig. 1

Chronic effects of pesticides on plasticity-regulated signaling. a Western blots of P-ERK1/2 vs. total ERK1/2 in organotypic hippocampal slices exposed chronically to DMSO or to 1–100 nM of pesticides. Mean ± SEM, n = 3–6, *P < 0.05, **P < 0.01 vs. DMSO, one-way ANOVA + Dunnett’s post-test. b Western blots of P-ERK1/2 vs. total ERK1/2 in organotypic hippocampal slices exposed chronically to contaminants in the absence or presence of 100 µM AP-5. Mean ± SEM, n = 3–9, *P < 0.05, **P < 0.01, ***P < 0.001 vs. DMSO, two-way ANOVA + Bonferroni’s post-test. c, d Western blots of P-ERK1/2 vs. total ERK1/2 and P-Akt vs. total Akt 15 min after chemical LTP (LTP) and DHPG (LTD) treatments in slices exposed chronically to pesticides. Mean ± SEM, n = 4–12, *P < 0.05, **P < 0.01, ***P < 0.001 vs. control, ^#P < 0.05, ^###P < 0.001 vs. LTP/DHPG alone, two-way ANOVA + Bonferroni’s post-test. e Puromycin labelling after chemical LTP (LTP) and DHPG (LTD) treatments in slices exposed chronically to pesticides. Slices non-treated with puromycin (-Pur) were used as negative control. Mean ± SEM, n = 3–8, **P < 0.01, ***P < 0.001 vs. control, ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 vs. DHPG alone, two-way ANOVA + Bonferroni’s post-test

Fig. 2

Chronic effects of pesticides on basal transmission and synaptic plasticity. Patch-clamp recordings of CA1 pyramidal neurons from slices exposed chronically to DMSO, pesticides, and/or 100 nM atropine. Non-treated slices (untreat) were used as control. a Representative traces (above) and quantification (below) of AMPAR vs. NMDAR currents (I). Mean ± SEM, n = 9–15. Scale bars = 40 pA/20 ms. b Representative traces (above) and quantification (below) of GABA_AR vs. AMPAR currents (I). Mean ± SEM, n = 9–17, *P < 0.05, one-way ANOVA + Dunnett’s post-test. Scale bars = 100 pA/20 ms. c Representative traces (above) and cumulative frequency distribution (below) of miniature inhibitory postsynaptic currents (mIPSC) amplitude. n = 13–20 neurons and 1270–1599 events; ***P < 0.001 for CPO and atropine as compared to DMSO, Kolmogorov–Smirnov test. Scale bars = 20 pA/10 ms. d Representative traces (above) and quantification (right) of AMPAR currents before (baseline, dark line) and 35–40 min after (light line) induction of mGluR-LTD. Results are expressed as % compared to baseline responses. *P < 0.05, ***P < 0.001 vs. baseline (one-sample t-test); ^#P < 0.05, ^##P < 0.01 vs. DMSO (one-way ANOVA + Dunnett’s post-test). Scale bars = 40 pA/20 ms. e Representative traces (above) and quantification (right) of AMPAR currents before (baseline, dark line) and 40–45 min after (light line) induction of TEA-LTP. Slices were treated with 100 µM AP-5 only during the induction. Results are expressed as % compared to baseline responses. *P < 0.05, **P < 0.01, ***P < 0.001 vs. baseline (one-sample t-test). Scale bars = 50 pA/20 ms. f Representative traces (above) and quantification (right) of AMPAR currents before (baseline) and after (25–30 min) induction of NMDA-LTD. Results are expressed as % compared to baseline responses. *P < 0.05, **P < 0.01, ***P < 0.001 vs. baseline (one-sample t-test). Scale bars = 40 pA/20 ms

Fig. 3

Chronic MAPK inhibition restores mGluR-LTD in chlordane-treated slices. a Western blots of P-ERK1/2 vs. total ERK1/2 and P-S6K vs. total S6K in organotypic hippocampal slices exposed chronically to 100 nM chlordane in the absence or presence of different concentrations of PD98059 (PD). Mean ± SEM, n = 4–8, *P < 0.05, **P < 0.01 vs. DMSO; ^#P < 0.05, two-way ANOVA + Bonferroni’s post-test. b Western blots of P-ERK1/2 vs. total ERK1/2 in organotypic hippocampal slices exposed chronically to 100 nM chlordane or 10 nM CPO in the absence or presence of 1 µM PD 15 min after DHPG treatment. Mean ± SEM, n = 3–8, *P < 0.05, **P < 0.01 vs. control, ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 vs. chlordane alone, two-way ANOVA + Bonferroni’s post-test. c Representative traces (above) and quantification (right) of AMPA currents before (baseline) and after (35–40 min) induction of mGluR-LTD in slices treated chronically with 100 nM chlordane or 10 nM CPO in the presence of 1 µM PD. ***P < 0.001 vs. baseline, one-sample t-test; ^#P < 0.05, two-way ANOVA + Bonferroni’s post-test. Scale bars = 40 pA/20 ms

Fig. 4

Perinatal exposure to chlordane and CPO impairs mGluR-LTD. Field recordings of CA1 area of hippocampus in acute slices from 0.05-0.5 mg/Kg/day CPO-, chlordane-, and DMSO-treated (vehicle) rats. a, c, e Input/output curves of field excitatory postsynaptic potentials (fEPSP) slope. Mean ± SEM, n = 14–17. b, d, f Representative traces and quantification of fEPSP slope before (baseline) and after (55–60 min) induction of mGluR-LTD in PND14-30 juvenile (b) and PND55-77 adult (d, f) rats. Results are expressed as % of fEPSP slope compared to baseline responses. Scale bars = 0.2 mV/ms /20 ms. **P < 0.01, ***P < 0.001 vs. baseline (one-sample t-test); b Mean ± SEM, n = 10–18, ^#P < 0.05, two-tailed t-test. d, f Mean ± SEM, n = 12–18, ^#P < 0.05, one-way ANOVA + Dunnett’s post-test

Fig. 5

Perinatal exposure to pesticides alters motor development and locomotor activity. a Somatic growth of rat pups as measured by weight gain during the lactation period. *P < 0.05, **P < 0.01, ***P < 0.001 vs. vehicle, two-way ANOVA + Bonferroni’s post-test. b Latency to righting all four limbs during the first days of postnatal period. Mean ± SEM, values from 3 consecutive trials were averaged for each rat pup. *P < 0.05, **P < 0.01 vs vehicle, two-way ANOVA + Bonferroni’s post-test. c Hind limb clasping score at weaning (PND21), measured as previously described (Dodge et al. 2020). *P < 0.05, one-way ANOVA + Dunnett’s post-test. d Total distance travelled in the open field test divided by 1-min intervals. *P < 0.05 vs. vehicle, **P < 0.01 vs. CPO 0.5, two-way ANOVA + Bonferroni’s post-test. e Time in the center of the arena in the open field test. *P < 0.05, one-way ANOVA + Dunnett’s post-test. f Number of marbles buried during a 30-min period

Fig. 6

Perinatal exposure to pesticides causes social and cognitive deficits. a Percentage of time spent sniffing the unfamiliar rat vs. total time sniffing in the first phase of SPSN. **P < 0.01, **P < 0.001 vs. chance value = 50, one-sample t-test; ^##P < 0.01, one-way ANOVA + Dunnett’s post-test. b Percentage of time spent sniffing the second unfamiliar rat vs. total time sniffing in the second phase of SPSN. *P < 0.05 vs. chance value = 50, one-sample t-test; ^#P < 0.05, one-way ANOVA + Dunnett’s post-test. c Total time of activity in the juvenile social dyad. Mean ± SEM, n = 10–20, *P < 0.05, **P < 0.01, ***P < 0.001 vs. vehicle, two-way ANOVA + Dunnett’s post-test. d Re-exposure; percentage of time exploring the objects during the re-exposure phase as compared to phase I (familiarization) of NORT. Test; percentage of time exploring the novel object vs. total object exploration time. **P < 0.01, *** P < 0.001 vs. reference value = 100 (re-exposure) or 50 (test), one-sample t-test; ^#P < 0.05, one-way ANOVA + Dunnett’s post-test. e Percentage of time exploring the novel location vs total object exploration time during the test phase of NOLT. *P < 0.05, **P < 0.01 vs. chance value = 50, one-sample t-test; ^#P < 0.05, one-way ANOVA + Dunnett’s post-test

Fig. 7

Perinatal exposure to pesticides reprograms developmental signaling networks. a, b Volcano plots of all the proteins detected in the hippocampus of CPO or chlordane-treated rats (n = 4, for all groups) by proteomic analysis as a function of their fold change (FC) and adjusted P value as compared to vehicle-treated rats (n = 4). Cut-off values were |Log₂ FC|> 0.28 and P < 0.05. Name of relevant, significantly regulated proteins have been highlighted in the plots. c, e Canonical signaling pathways and physiological functions and disease into which significantly regulated proteins were categorized by Ingenuity pathways analysis. Cut-off values were P < 0.05 and P < 0.01 for canonical pathways and physiological functions and disease, respectively. d Venn diagram of the number commonly regulated proteins for CPO (blue) and chlordane (red). Examples of these proteins are noted below and arrows indicate whether they were upregulated or downregulated by each pesticide