The NO Answer for Autism Spectrum Disorder

Abstract

Autism spectrum disorders (ASDs) include a wide range of neurodevelopmental disorders. Several reports showed that mutations in different high-risk ASD genes lead to ASD. However, the underlying molecular mechanisms have not been deciphered. Recently, they reported a dramatic increase in nitric oxide (NO) levels in ASD mouse models. Here, they conducted a multidisciplinary study to investigate the role of NO in ASD. High levels of nitrosative stress biomarkers are found in both the Shank3 and Cntnap2 ASD mouse models. Pharmacological intervention with a neuronal NO synthase (nNOS) inhibitor in both models led to a reversal of the molecular, synaptic, and behavioral ASD-associated phenotypes. Importantly, treating iPSC-derived cortical neurons from patients with SHANK3 mutation with the nNOS inhibitor showed similar therapeutic effects. Clinically, they found a significant increase in nitrosative stress biomarkers in the plasma of low-functioning ASD patients. Bioinformatics of the SNO-proteome revealed that the complement system is enriched in ASD. This novel work reveals, for the first time, that NO plays a significant role in ASD. Their important findings will open novel directions to examine NO in diverse mutations on the spectrum as well as in other neurodevelopmental disorders. Finally, it suggests a novel strategy for effectively treating ASD.

Keywords: S-nitrosylation; Shank3; autism spectrum disorder; behavior; contactin-associated protein-like2; nitric oxide.

Figure 1

Illustrative figure describing the aberrant role of NO in ASD and the therapeutic potential following the inhibition of nNOS. Under normal conditions, NO acts as a signaling molecule, playing a crucial role in the development and formation of dendritic spines, which are essential for proper cognitive function. However, mutations in the SHANK3 or CNTNAP2 (CASPR2) genes elevate the levels of NO production, which can lead to the disruption of dendritic spines and neuronal functions; this results in behavioral deficits. Pharmacological inhibition of nNOS restores the ASD phenotype.

Figure 2

NO donor administration led to nitrosative stress and synaptic pathology in the cortex of WT mice. A) Representative western blots of an indicator of nitrosative stress, 3‐Ntyr, and the synaptic proteins Syp, NR1, GAD1, VGAT, Homer, and PSD95. Β‐actin was used as a reference for protein loading. B) Statistical analysis of the relative abundance of proteins shown in (A). C) Representative confocal images of the NR1, NeuN, DAPI, and their respective secondary ab control fluorescence. The image was captured at 40× magnification and the scale bar = 50 µm. D) Representative confocal images of the GAD1, NeuN, and DAPI fluorescence. The image was captured at 60× magnification and the scale bar = 50 µm. E,F) Statistical analysis of the mean fluorescence intensity of NR1, and GAD1 protein. G) Left—representative images of the dendritic spines; right—statistical analysis of the dendritic spine density. Student's two‐tailed t‐test was used for two‐group comparisons. *p < 0.05, **p < 0.01, ***p < 0.001. Groups of mice: WT (n = 8); SNAP (WT mice treated with the NO donor compound SNAP, n = 8) for western blot analysis and WT (n = 9); SNAP (n = 9) for the immunofluorescence experiments. In the IF images, red denotes NR1/GAD1, green denotes neuronal marker NeuN, and blue denotes DAPI.

Figure 3

NO inhibition led to a reversal in the molecular and synaptic deficits in the Shank3^Δ4‐22 mouse model. A) Representative western blots of synaptic proteins Shank3, Syp, NR1, GAD1, VGAT, Homer, and PSD95 in four groups of mice: WT1, WT1+7‐NI, M1, and M1+7‐NI. β‐actin was used as a reference for loading control. B) Representative western blots of an indicator of nitrosative stress; 3‐Ntyr and b‐actin are used as loading control. C) Statistical analysis of the relative abundance of synaptophysin/b‐actin, D) NR1/b‐actin, E) GAD1/b‐actin, F) VGAT/b‐actin, G) Homer/ b‐actin, H) PSD95/b‐actin, I) 3‐Ntyr/b‐actin. J) Representative confocal images of the NR1, NeuN, DAPI, and their relative secondary ab control fluorescence. The image was captured at 40× magnification and the scale bar = 50 µm. K) Representative confocal images of the GAD1, NeuN, and DAPI fluorescence. The image was captured at 60× magnification and the scale bar = 50 µm. L,M) Statistical analysis of the mean fluorescence intensity of NR1 and GAD1 protein shown in (J) and (K), respectively. N) Left—representative images of the dendritic spines; right—statistical analysis of the dendritic spine density. The mean and standard deviation (SD) were calculated for the western blots, immunofluorescence, and spine density counting. A one‐way ANOVA (for IF and spine density) and a two‐way ANOVA test, along with the Bonferroni multiple comparison tests were used for the western blot analyses. *p < 0.05, **p < 0.01, ***p < 0.001, ns = non‐significant. Abbreviations, WT1 (Shank3 WT littermates), WT1+7‐NI (Shank3 WT littermates treated with 7‐NI), M1 (Shank3 mutant mice‐Shank3^Δ4‐22 ), M1+7‐NI (Shank3^Δ4‐22 mice treated with the nNOS inhibitor 7‐NI). The number of mice: WT (n = 6), WT1+7‐NI (n = 6), M1 (n = 6), and M1+7‐NI (n = 6). In IF images, red denotes NR1/GAD1, green denotes neuronal marker NeuN, and blue denotes DAPI.

Figure 4

NO inhibition in the Cntnap2^(‐/‐) mutant (M2) mouse model led to a reversal of the molecular and synaptic deficits. A) Representative western blots of synaptic proteins Cntnap2, Syp, NR1, GAD1, VGAT, PSD95, and Homer in 4 groups of mice: WT2, WT2+7‐NI, M2, M2+7‐NI (M2 mice treated with 7‐NI). β‐actin was used as loading control. B) Representative western blots of an indicator of nitrosative stress, 3‐Ntyr; b‐actin was used as loading control here. C) Statistical analysis of the relative abundance of Synaptophysin/b‐actin. D) NR1/b‐actin, E) GAD1/b‐actin, F) VGAT/b‐actin, G) Homer/b‐actin, H) PSD95/b‐actin, I) 3‐Ntyr /b‐actin. J) Representative confocal images of NR1, NeuN, DAPI, and their relative secondary ab control fluorescence. The image was captured at 40× magnification and the scale bar = 50 µm. K) Representative confocal images of the GAD1, NeuN, and DAPI fluorescence. The image was captured at 60× magnification and the scale bar = 50 µm. L,M) Statistical analysis of the mean fluorescence intensity of NR1, and the GAD1 protein shown in (J) and (K), respectively. N) Left—representative images of the dendritic spines; right—statistical analysis of the dendritic spine density. The mean and standard deviation (SD) were calculated for the western blots, immunofluorescence, and spine density counting. A one‐way ANOVA (for IF and spine density) and a two‐way ANOVA test, along with the Bonferroni multiple comparison tests, were used for the western blot analyses. *p < 0.05, **p < 0.01, ***p < 0.001, ns = non‐significant. Abbreviations, WT2 (C57bL/6 mice were used as a control), WT2+7‐NI (WT2 mice treated with 7‐NI), M2 (Cntnap2^(‐/‐) mutant mice), M2+7‐NI (Cntnap2^(‐/‐) mutant treated with the nNOS inhibitor 7‐NI). The number of mice: WT2 (n = 6), WT2+7‐NI (n = 6), M2 (n = 6), and M2+7‐NI (n = 6). In IF images, red denotes NR1/GAD1, green denotes neuronal marker NeuN, and blue denotes DAPI.

Figure 5

7‐NI administration reversed the ASD‐like behavior in the Shank3 male mice. Behavioral tests analysis was conducted for the following groups of male mice: 1. WT1 (Shank3 littermates treated with vehicle), M1 (Shank3^Δ4‐22 mice treated with vehicle), and M1+7‐NI (Shank3^Δ4‐22 treated with 7‐NI (i.p. injection of 80 mg kg⁻¹). A) Left—an illustration of Novel object recognition (NOR) test platform. Right—statistical analysis of the object exploration time of either a novel (N) or a familiar (F) object. The WT1 mice spent significantly more time exploring the novel object than the familiar one (n = 8, *p = 0.0107). The M1 mice exhibited no significant preference for the novel object over the familiar one, indicating a lack of novelty seeking (n = 9, p > 0.05). However, the 7‐NI‐treated mutant mice (M1+7‐NI) exhibited a significant preference for the novel object over the familiar object (n = 8, **p = 0.0059). B) Upper panel—an illustration of three‐chamber sociability test platform (the first session) and the lower panel—the statistical analysis of the interaction time with either an empty cage (E) or a stranger mouse (S). The WT1 mice spent significantly more time interacting with the stranger mouse than with the empty cage (n = 15, **p = 0.0037). The M1 mice did not show any particular interest in engaging in social interaction, indicating reduced sociability among the M1 male mice (n = 19, ns = not significant). However, M1+7‐NI mice spent significantly increased time interacting with the stranger mouse than with the empty cage (n = 15, ***p < 0.0001). C) Upper panel—an illustration of three‐chamber sociability test platform (the second session) and the lower panel—the statistical analysis of the interaction time with either a familiar mouse (S1) or a novel mouse (S2). The WT1 mice spent significantly more time interacting with S2 than with S1 (n = 16, ***p = 0.0004), and the M1 mice showed no particular interest in interacting with S2 or S1(n = 15, ns = not significant), whereas the M1+7‐NI mice spent significantly more time interacting with S2 than with S1 (n = 8, *p = 0.0432). D) Left—an illustration of the elevated plus maze test platform. Right—statistical analysis of the time spent in the closed arms or in the open arms. The M1 mice spent significantly less time in the open arms compared with the WT1 mice (*p‐value, M1‐WT1 = 0.0484) and WT1+7‐NI (***p‐value M1‐M1+7‐NI = 0.0003). The M1 mice spent significantly more time in the closed arms than the WT1 mice (*p‐value M1‐WT1 = 0.0206). However, the M1+7‐NI mice spent significantly (***p‐value M1‐M1+7‐NI < 0.0001) less time in the closed arms compared with their vehicle‐treated littermates‐M1 (N = 9, 8, and 15 for WT1, M1, and M1+7‐NI, respectively). In all tests, the time is presented as the percentage of the total time. The data are presented as the mean ± SD. Statistical significance was determined using a two‐way ANOVA with Bonferroni's multiple comparisons tests. *p < 0.05, **p < 0.001, ***p < 0.0001, and ns = non‐significant.

Figure 6

NO inhibition reversed the autistic behavior abnormalities in Cntnap2^(‐/‐) mutant male mice, whereas NO donor treatment led to behavioral abnormalities in WT male mice. A–C) Behavioral test analyses for the following groups of male mice: 1) WT2 (wild‐type mice treated with vehicle), 2) M2 (Cntnap2^(‐/‐) mice treated with vehicle), and 3) M2+7‐NI (Cntnap2^(‐/‐) treated with 7‐NI) (i.p. injection of 80 mg kg⁻¹). D–G) Analysis of the WT (C57BL/6J or black6) male mice and WT mice treated with SNAP (20 mg/kg). A) Novel Object recognition test (NOR). The WT2 mice spent significantly more time exploring the novel (N) object than the familiar (F) one (n = 11, **p = 0.0090). The M2 mice did not show any particular interest in exploring either the novel object or the familiar one, indicating a lack of novelty seeking and interest (n = 15, ns = not significant). However, M2+7‐NI exhibited significantly increased time exploring the novel object than the familiar one (n = 17, ***p = 0.0002). B) Three‐chamber sociability test. The WT2 mice spent a significantly prolonged time interacting with the stranger mouse than with the empty cage (n = 11, *p = 0.0202). The M2 mice did not show any particular interest in engaging in social interaction, indicating reduced sociability among the M2 male mice (n = 16, p = 0.3313). However, the M2+7‐NI mice spent significantly increased time interacting with the stranger mouse than with the empty cage (n = 18, **p = 0.0024). C) Elevated plus maze test. The M2 mice (n = 9) spent significantly more time in the closed arms than did the WT2 mice (n = 9, *p = 0.0371), showing increased anxiety among the M2 mice. However, the M2+7‐NI mice spent significantly (n = 13, **p = 0.0067) less time in the closed arms than did the M2 mice. In all tests, the time is presented as the percentage of the total time. The data are presented as the mean ± SD. Statistical significance was determined using a two‐way ANOVA with Bonferroni's multiple comparisons tests. *p < 0.05, **p < 0.001, ***p < 0.0001, and ns = non‐significant. D) Novel object recognition test showing the normal spontaneous behaviors of the WT mice in exploring the novel object rather than the familiar one (n = 11, *p = 0.0139). The WT+SNAP mice failed to display a significant preference either for the novel object or for the familiar one (n = 10, p = 0.6910). E) The time interacting with either E or S. The WT mice spent significantly more time with S than with E (n = 16, *p = 0.0051). In contrast, the WT+SNAP mice did not show any particular interest in interacting with S or E (n = 15, p = 0.2387). F) The WT mice spent more time interacting with the S2 than with S1 (n = 11, ***p  =  0.0001), whereas the WT+SNAP mice did not show any significant preference to engage in social interaction with S1 or S2 (n = 15, p = 0.7452). G) Left—the time spent in closed arms. The WT+SNAP mice (n = 8) spent significantly more time in the closed arms compared with their WT counterparts (n = 9, *p = 0.0221). G) Right—the time spent in open arms shows that the WT+SNAP mice spent significantly less time in the open arms than did the WT mice (*p = 0.0310). In all tests, the time is presented as the percentage of the total time. The data are presented as the mean ± SD. A Two‐tailed t‐test was conducted to determine the statistical significance. *p < 0.05, **p < 0.001, ***p < 0.0001, and ns = non‐significant.

Figure 7

NO contributes to nitrosative stress and reduced synaptogenesis in the primary cortical neurons of the Shank3^Δ4‐22 and Cntnap2^(‐/‐) mutant groups. A) Representative confocal images of the MAP2 (green), Syp (red), DAPI (blue), and their respective secondary ab control in the WT1, M1, and M1+7‐NI groups. B) Representative confocal images of the 3‐Ntyr (green), MAP2 (red), and DAPI (blue) fluorescence in the WT1, M1, and M1+7‐NI groups. C) Statistical analysis of the mean fluorescence intensity of Syp protein. D) Statistical analysis of the mean fluorescence intensity of 3‐Ntyr. E) Representative confocal images of the MAP2 (green), Syp (red), and DAPI (blue) in the WT, M2, and M2+7‐NI mice. F) Representative confocal images of the 3‐Ntyr (green), MAP2 (red), and DAPI (blue) fluorescence in the WT2, M2, and M2+7‐NI groups. G) Statistical analysis of the mean fluorescence intensity of Syp protein. H) Statistical analysis of the mean fluorescence intensity of 3‐Ntyr. All IF images were captured at 60×. The scale bar = 50 µm in scale. A one‐way ANOVA test with the Tukey post hoc test was used for multiple comparisons in all groups. *p < 0.05, **p < 0.01, ***p < 0.001. Abbreviations, WT1 (primary neurons isolated from Shank3 WT embryos), M1 (primary neurons isolated from Shank3^Δ4‐22 embryos), M1+7‐NI (primary neurons isolated from Shank3^Δ4‐22 embryos treated with the nNOS inhibitor 7‐NI). WT2 (primary neurons isolated from black6 embryo), M2 (primary neurons isolated from Cntnap2^(‐/‐) embryos), M2+7‐NI (primary neurons isolated from Cntnap2^(‐/‐) embryos treated with the nNOS inhibitor 7‐NI). WT1 (n = 6), M1 (n = 6), and M1+7‐NI (n = 6), WT2 (n = 6) M2 (n = 6), and M2+7‐NI (n = 6).

Figure 8

Molecular reprogramming in iPSC‐derived cortical neurons from patients with SHANK3 mutation as well as in human cell lines and blood samples from ASD children. A) Representative confocal images of MAP2 (green), Syp (red), and DAPI (blue) in SHSY5Y (n = 10); SHSY5Y+siSHANK3 (n = 10); and SHSY5Y+siSHANK3+si‐nNOS (n = 10). Image captured at 60×; the scale bar = 50 µm. B) Statistical analysis of the mean fluorescence intensity of Syp protein. Stealth RNAi Negative Control Duplexes was used for the RNA interference (RNAi) experiments as a control for sequence‐independent effects. C) Representative western blot of 3‐Ntyr in human plasma samples in TD (n = 8), and ASD (n = 8) with statistical analysis. D) Left—a western blot of Syp and NR1 from iPSC‐derived cortical control (C), iPSC‐derived cortical neurons from patients with SHANK3 mutations (Shank3^C.3679insG) (M), and M+7‐NI (Shank3^C.3679insG+7‐NI) right—statistical analysis of NR1/actin and Syp/actin. E) Representative confocal image of Ntyr (red) and MAP2 (green) in human iPSC‐derived cortical neurons from control, mutant, and mutant+7‐NI groups. Images were captured at 60× and the scale bar = 50 µm. F) Statistical analysis of the mean fluorescence intensity of 3‐Ntyr shown in (E). G) Volcano plot displaying log 2 (FC) on the x‐axis plotted against the −log 10 (p‐value) on the y‐axis for all the identified proteins that were differentially expressed in the plasma of TD (n = 11) and ASD (n = 11) individuals. The horizontal line represents a significance level of p‐value =  0.05. SNO proteins that were significantly upregulated in ASD (p < 0.05) are denoted in orange on the right upper side of the plot, whereas the proteins that were significantly downregulated are denoted in orange on the left upper side of the plot (p < 0.05). H) BP and pathways analysis were conducted on the identified significant SNO proteins using the STRING database, version 11.5. Each bar represents the −log 10 (FDR). I) Clustering analysis of the plasma SNO‐proteins. “The complement system in neuronal development and plasticity” was enriched (the number of proteins = 28, FDR = 6.26e‐28). The red nodes are the proteins that were significantly changed between the ASD and TD individuals (p < 0.05). STRING, version 11.5, and Cytoscape software version 3.3.0 was used to generate this figure. The data are presented as the mean ± SD. A Two‐tailed t‐test was conducted to determine the statistical significance of the comparison between two groups. A one‐way ANOVA test with the Tukey post hoc test was used for multiple comparisons in all groups *p < 0.05, **p < 0.001, ***p < 0.0001, and ns = non‐significant.