Phosphoproteomic profiling reveals reversal of dysregulated kinase signaling by nitric oxide inhibition in the Shank3 mouse model of autism Authors: Shelly Ginzburg, Igor Khaliulin, Maryam Kartawy, Wajeha Hamoudi, Haitham Amal Published: 2026 ABSTRACT AbstractAutism spectrum disorder (ASD) is a wide-ranging neurodevelopmental disorder characterized by behavioral deficits, including difficulties in social communication and interaction, restricted or repetitive behaviors, and limited interests. The SHANK3 gene encodes a scaffolding protein in the postsynaptic density of glutamatergic excitatory synapses and is one of the highest-confidence ASD risk genes. Our previous studies showed abnormally elevated levels of nitric oxide (NO) in Shank3 models of ASD, and inhibition of the neuronal NO synthase (nNOS) reversed ASD-like behavioral and synaptic phenotypes in these models. Protein phosphorylation is involved in all biological processes in the body, but it is still unknown how protein phosphorylation is affected by the NO signaling in ASD. We studied the global proteome and phosphoproteome of the cortical tissue of the Shank3Δ4–22 mouse model of autism, followed by system biology analysis. First, we compared Shank3 KO to the WT mice, revealing altered phosphorylation states in proteins enriched in synaptic, developmental, and RNA metabolism biological processes. Following treatment with nNOS inhibitor, 7-nitroindazole , we observed reversal of the phosphorylation state in proteins participating in the same biological processes, indicating a link between NO signaling and protein phosphorylation patterns. Importantly, Shank3 mutation revealed altered phosphorylation in several ASD SFARI high-risk genes, and many of these changes were reversed by the NO inhibition. This work is the first to investigate the cross-talk between NO and kinase signaling in a Shank3 ASD model. Finally, it might provide new insights into the NO-mediated protein phosphorylation mechanisms and help identIfying new therapeutic targets for ASD.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-57345-0.Subject terms: Biochemistry, Cell biology, Neuroscience Introduction Autism spectrum disorder (ASD) is a major neurodevelopmental disorder whose prevalence rate grows continuously. Epidemiological studies have found a nearly 52% increase in ASD prevalence in the US between 2017 and 20201. According to the latest data from the US Centers for Disease Control and Prevention (CDC), 1 in 31 children has been diagnosed with this disorder2. Despite extensive studies, no effective treatment has been found for ASD3. Therefore, there is an urgent need to study the molecular mechanisms of this disorder, which may reveal novel, robust biomarkers and therapeutic targets for ASD. One of the essential aspects of pathogenesis is aberrant post-translational modifications (PTMs) of proteins. PTMs are involved in the regulation of protein localization, function, translocation, interaction with other molecules, and degradation. S-nitrosylation (SNO) is a form of nitric oxide (NO)-dependent PTM4,5. NO is a gas molecule synthesized by three isoforms of NO synthase (NOS): neuronal (nNOS), inducible (iNOS), and endothelial (eNOS). Reaction of NO with sulfhydryl groups of Cys in proteins leads to protein SNO4,5. SNO is a mediator of NO signaling under both physiological and pathological conditions. At low levels, this PTM has been shown to regulate the normal localization and activity of many key enzymes and receptors5–7, and modulate various physiological processes in the brain5,8, including synaptic plasticity9,10, axonal elongation, and neuronal survival5,11. However, prolonged and extensive generation of NO during brain disease can triggerSNO, which is not observed in physiological conditions. Such abnormal SNO can cause protein misfolding, synaptic damage, mitochondrial fission, or apoptosis12,13. Another kind of PTM, protein phosphorylation, regulates nearly all biological processes in the cell. Protein phosphorylation or dephosphorylation can mediate protein functions in every possible way, increasing or decreasing activity, targeting a protein for degradation, promoting its translocation to subcellular compartments, or regulating its interaction with other proteins14. This PTM is mediated by protein kinases, which convert the protein from a hydrophobic apolar to a hydrophilic polar form, changing the protein conformation necessary for interaction with other molecules15. In physiological conditions, a constant balance is maintained between phosphorylation and dephosphorylation ,carried out by protein kinases and phosphatases, respectively . Genetic mutations or disruptions of regulatory mechanisms may result in abnormal kinase/phosphatase signaling, causing dysregulation of protein function16. Interestingly, a strong connection has been found between these two forms of PTM. In most cases, this link is manifested by SNO-induced inhibition of kinases17,18. Using the SNOTRAP technology developed by us and colleagues19, we found a reprogramming of the SNO-proteome in the brains of ASD mouse models, which appears to play a crucial role in the pathogenesis of this disorder13,19–26. Our discovery of the excessive levels of NO in mouse models of ASD is consistent with a previous postmortem study showing the accumulation of 3-nitrotyrosine, a marker of nitrosative stress, and another form of modification caused by NO that might affect the protein function and is associated with protein degradation27,28, in the brains of individuals with ASD29. In this study, we determined which phosphoproteins are affected by the increase of nitrosative stress in ASD. We also investigated the interplay between NO signaling and protein phosphorylation in the cortex, a critical region of ASD pathology30, of the Shank3Δ4–22 ASD mouse model, treated or non-treated with an nNOS inhibitor, using phosphoproteomics coupled to systems biology analyses. This animal model of ASD was chosen because previous studies have shown that a complete Shank3 deletion at exons 4–22 most closely represents human mutations and recapitulates the core symptoms of this disorder31,32. Further, this is supported by our previous studies demonstrating that 7-NI reverses the behavioral deficits observed in this mouse model and restores both the SNO-proteome profile, including SNO-TSC2, and 3-nitrotyrosine levels22,33. In this study, our data suggest a possible link between NO signaling and phosphorylation-mediated signaling in the Shank3Δ4–22 ASD mouse model. We also found that ASD associated kinases and phosphatases were affected by the aberrant NO signaling, where 7-NI treatment showed a significant rescue of several phosphoproteins involved in AMPA receptor trafficking, vesicular dynamics, mTOR pathway, and MAPK signaling. Materials and methods Animal housing and tissue dissection The experimental procedures of this study (schematically presented in Fig. 1A) conformed to the guidelines of the Institutional Animal Care Committee of The Hebrew University of Jerusalem, Israel. The study protocols have been approved by the joint ethics committee (IACUC) of The Hebrew University, an Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) internationally accredited institution, and Hadassah Medical Center, Jerusalem, Israel (MD-22-16988-3). Juvenile (6–8-week-old) male Shank3Δ4–22 knockout (KO) mice (The Jackson Laboratory, Farmington, CT, USA, Strain #:03 2169) and their wild-type (WT) littermates were used in this study. The animals were kept at a room temperature of 23 °C in a 12-hour light/dark cycle and fed ad libitum with standard mouse chow and water. The pharmacological intervention was performed by daily intraperitoneal (IP) injections of the selective nNOS inhibitor 7-nitroindazole (7-NI, 80 mg/kg, Thermo Fisher Scientific, Waltham, MA, USA) for 10 consecutive days, as we have done previously22. In parallel, a group of male Shank3Δ4–22 KO mice was treated with vehicle. Thus, the following three groups of mice each one includes 3 mice were employed in this study: WT mice treated with vehicle (WT group), KO mice treated with vehicle (KO group), and KO mice treated with 7-NI (KO + 7-NI group). WT mice were not treated with 7-NI because no effect of this drug was observed in WT mice in our previous studies22. Each group included 3 biological and 1 technical replicate. On the 10th day of 7-NI injections, the mice were sacrificed, the cortex tissue was isolated from the rest of the brain tissues, as we have done previously25, and stored at -80 °C for further analysis. System biology analysis of the proteins with significantly altered phosphorylation levels in the Shank3Δ4–22 ASD mouse model (further called KO mice) compared to WT mice. (A) Graphic representation of the workflow. (B) Volcano plot of the upregulated (right) and downregulated (left) phosphorylated proteins in the KO group. Significant changes (P < 0.05) are colored in red. (C) Protein interaction network map and clustering visualization of the significantly changed phosphoproteins. (D) BP (marked in blue font) and CC (marked in red font) enrichment. (E) BP analysis of the up-regulated (Green) and the down-regulated phosphoproteins (Turquoise). (F) Enriched reactome pathways. Brain tissue homogenization and sample preparation for mass spectrometry (MS) The samples were homogenized in freshly prepared buffer (50 mM Tris, 5% SDS, and 1% phosphatase/protease inhibitor cocktail, pH 7.5) on ice using a Teflon pestle and a Jumbo Stirrer (Thermo Fisher Scientific). General chemicals were purchased from Sigma Aldrich (St. Louis, MO, USA) and Bio-Rad Laboratories (Haifa, Israel). MS analysis The LC-MS/MS and the following analysis were carried out as described previously34–36. Briefly, the proteins were cleaved with trypsin (1:50 M/M and then 1:100 M/M enzyme to protein ratio), and analyzed by liquid chromatography with tandem mass spectrometry (LC-MS/MS) using the Exploris 480 MS (Thermo Fisher Scientific). The global proteomics samples were analyzed in data-independent acquisition (DIA) mode, and the data were analyzed using DIA-NN software, v.1.9.1, for identification and quantification against mouse Uniport databases with a 1% false discovery rate (FDR) threshold. FDR, the estimated fraction of false positives in a list of peptides, was calculated with the Benjamini–Hochberg correction. Terms with FDR < 0.05 were accepted. For the phosphoproteomics, after trypsinization, the samples were enriched for phosphopeptides by TiO2, analyzed using liquid chromatography with tandem mass spectrometry (LC-MS/MS) using the Exploris 480 MS (Thermo Fisher Scientific). Later analyzed in data-dependent acquisition (DDA) mode, and the data were analyzed using MaxQuant 2.6.6.0 for identification and quantification against mouse Uniprot databases with a 1% FDR threshold. Statistical analysis of the identified phosphoproteins’ signal intensity was done by Perseus 1.6.15.0 software. Brain tissue homogenization and sample preparation for Western blot (WB) Cortical tissues (n = 6 per group) were homogenized on ice in freshly prepared Radioimmunoprecipitation Assay buffer (RIPA buffer) containing 30 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES, pH 7.4), 150 mM sodium chloride (NaCl), 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 5 mM ethylenediaminetetraacetic acid (EDTA), 1 mM sodium orthovanadate (Na₃VO₄), 50 mM sodium fluoride (NaF), 1 mM phenylmethylsulfonyl fluoride (PMSF), and protease and phosphatase inhibitor cocktails (pH 7.7), using a Teflon pestle and Jumbo Stirrer (Thermo Fisher Scientific). Homogenates were centrifuged at 17,000 × g for 30 min at 4 °C, and protein concentrations were determined using the Bicinchoninic Acid (BCA) protein assay. Western blots Equal amounts of protein were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and transferred by wet transfer onto polyvinylidene fluoride (PVDF) membranes (Bio-Rad). Membranes were blocked for 2 h at room temperature in Tris-buffered saline with Tween-20 (TBST) containing 5% dried skimmed milk. Membranes were incubated overnight at 4 °C with primary antibodies. The following antibodies were used: Anti- Synapsin 1 (CST-5297T) (1:1000), Phospho-Synapsin-1 (Ser605) (CST-88246T ) (1:1000), Anti SynGAP1 (CST-5539 S) (1:1000), Anti Phospho-MAPK/CDK Substrates(PXS*P or S*PXR/K) (CST-2325 S) (1:1000). Following TBST washes, membranes were incubated with horseradish peroxidase-conjugated secondary antibodies for 2 h at room temperature. Protein bands were detected using enhanced chemiluminescence (ECL) and visualized with the Bio-Rad ChemiDoc imaging system. Band intensities were quantified using Bio-Rad Image Lab software and normalized to total lane protein, determined by stain-free imaging of the same membrane. Immunoprecipitation assay Due to the lack of available antibody for this specific phosphorylation site of SynGAP1 on Ser788, we performed an immunoprecipitation assay (IP) of the SynGAP1 protein and immunoblotted with an antibody detecting the phospho-serine with the specific motif of PXS*P or S*PXR/K, which matches the motif of phosphorylation of Ser788 of SynGAP (S*PTK). Immunoprecipitation (IP) was performed using Dynabeads Protein G (Invitrogen). Number of cortex samples used in the IP experiment is n = 4. Briefly, proteins were extracted in RIPA lysis buffer as mentioned previously; 1000 µg of total protein were incubated with antibody-prebound Dynabeads. For bead preparation, 50 µl Dynabeads were washed in PBS with 0.01% Tween-20 and incubated with the specific antibody with dilution 1:10 (and the IgG control antibody was calculated to match the concentration) for 10 min at room temperature with rotation, following with antibody cross-linking to the beads using 20 mM dimethyl pimelimidate in 0.2 M triethanolamine, pH 8.2, and quenched with 50 mM Tris, pH 7.5. Antibody–bead complexes were then incubated with lysates for 1–2 h at 4 °C with gentle rotation, and unbound material was removed; beads were washed three times in series of detergents/Tris buffers. Bound proteins were eluted by adding 20 µl 5% SDS in 50 mM Tris, pH 7.4 and 20 µl Laemmli sample buffer with β-mercaptoethanol, followed by vigorous shaking at room temperature for 10 min and heating at 95 °C for 10 min, and the supernatant was collected for subsequent SDS–PAGE and immunoblot analysis. Bioinformatics and systems biology analysis The search tool for interacting proteins (STRING, version 12.0; http://string-db.org, accessed in 2025) was used for systems biology and protein-protein interaction network analysis of global- and phosphoproteomics. Medium reliability interactions (score > 0.4) from the neighborhood, gene fusion, co-occurrence, co-expression, experiments, databases, and text mining lists were used. For the clustering analysis, we used the K-means (obsolete) algorithm. For the Gene Ontology (GO) analysis, we used the Panther classification system 19.0 (https://pantherdb.org, accessed in 2025). For the human ASD risk gene dataset, we used the SFARI database (https://gene.sfari.org/ accessed in February 2025). We performed a system biology analysis of the significantly altered phosphorylated proteins in all three groups of mice to determine which Biological Processes (BP), Cellular Components (CC), Reactome Pathways, and Mammalian Phenotypes were enriched due to the Shank3 gene deletion and the treatment of these KO mice with 7-NI. GraphPad Prism 10.4.1 software was used to generate the schematic figures. Statistical analysis Evaluation of statistically significant differences between any two groups (e.g., KO compared to WT or KO + 7-NI compared to KO) was carried out using an unpaired two-tailed t-test. The differences were considered statistically significant when P ≤ 0.05. Evaluation of statistically significant differences between the 3 groups was carried out using One-Way ANOVA with Tukey’s post-hoc multiple comparisons test. The differences were considered statistically significant when P ≤ 0.05. Results Disturbance in NO signaling in the Shank3Δ4–22 KO mouse model was established in previous studies, and inhibition of nNOS showed therapeutic potential in both behavioral and biochemical aspects20,22,33,37. To understand the mechanistic role of NO on neuronal function, and in light of previous studies demonstrating an association between NO signaling and phosphorylation17,18, we sought to investigate this link in this mouse model of ASD. Systems analysis of cortical phosphoproteome dysregulation in the Shank3Δ4–22 mouse model Initially, we mapped the phosphoproteome to identify changes in protein phosphorylation levels in the Shank3Δ4–22 KO mouse model of ASD compared to WT mice using MS, followed by large-scale systems biology analysis to understand the biological significance of these changes. The MS analysis revealed 7099 phosphorylation sites in 2257 proteins (Fig. 1B). Among these, 341 phosphorylation sites across 282 proteins were significantly altered in KO mice compared to WT (Fig. 1B). To investigate the functional organization and interactions among these proteins, we constructed a protein–protein interaction (PPI) network, which was clustered into five distinct clusters using the k-means algorithm (k = 5), These include: ‘Postsynaptic density’, ‘Metabolic processes’, ‘System regulation’, ‘Transmission across chemical synapses’, and ‘mRNA processing’ (Fig. 1C). To gain deeper insight into the biological importance of these differentially phosphorylated proteins in the KO cortex we performed BP analysis which revealed enrichment of synaptic and neuronal development processes such as ‘Synapse organization’ (FDR = 3.38 × 10− 15), ‘Nervous system development’ (FDR = 1.68 × 10− 14), ‘Neurogenesis’ (FDR = 4.44 × 10− 10), ‘Regulation of neurotransmitter levels’ (FDR = 9.33 × 10− 10), and ‘Postsynapse organization’ (FDR = 4.14 × 10− 9) (Fig. 1D). The CC analysis revealed that these proteins were enriched in specific neuronal compartments, including ‘Synapse’ (FDR = 7.08 × 10− 52), ‘Neuron projection’ (FDR = 3.22 × 10− 37), and ‘Dendrite’ (FDR = 1.04 × 10− 33) (Fig. 1D). Further BP analysis of the significantly altered phosphorylated proteins showed that the following processes were unique to the upregulated phosphoproteins: ‘Nervous system development’ (FDR = 4.75 × 10-14), ‘Neuron differentiation’ (FDR = 2.94 × 10-8), and ‘Behavior’ (FDR = 0.00031). By contrast, ‘Cellular macromolecule localization’ (FDR = 0.0053) and ‘Trans-synaptic signaling’ (FDR = 0.033) were unique to the downregulated phosphoproteins (Fig. 1E). To define the functional pathways associated with these proteins, we analyzed the reactome pathways which included ‘Trafficking of the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors’ (FDR = 0.00018), ‘Unblocking of N-methyl-D-aspartate (NMDA) receptors, glutamate binding and activation’ (FDR = 0.0023), and ‘Activation of NMDA receptors and postsynaptic events’ (FDR = 0.0023) (Fig. 1F, For the full GO analysis see Supplementary Table 1). Systems analysis of cortical phosphorylation changes independent of protein expression in the Shank3Δ4–22 mice To further refine these findings and isolate phosphorylation-specific regulatory events, we next focused specifically on phosphoproteins showing significant changes in phosphorylation without corresponding changes in protein levels, which allows us to get a better view of phosphorylation-dependent pathway activity and regulation. Thus, we focused on 198 proteins encompassing 237 phosphorylation sites. The PPI network was divided into 8 major clusters: ‘Postsynapse organization’, ‘Unblocking of NMDA receptors, glutamate binding and activation’, ‘mRNA 3’-end processing’, ‘Microtubule-associated processes’, ‘Nucleus-associated terms’, ‘Synaptic terms’, ‘Cell junction-associated terms’, and ‘Metabolism-related terms’ (Fig. 2A). Further analysis of these proteins revealed a variety of BPs. These BPs can be grouped into three categories: Developmental, Synaptic, and mRNA metabolism (Fig. 2B-D). As ASD is a neurodevelopmental disorder, the overlap between the BP ‘Developmental process’ (FDR = 3.01 × 10− 7) and the mammalian phenotype ‘Nervous System Phenotype’ (FDR = 4.6 × 10− 16) suggests that the sixty-four observed phosphorylation changes affect proteins with established roles in neurodevelopment and neuronal function. (Fig. 2E). We examined these specific proteins and their phosphorylation sites. Among them, we found significantly increased phosphorylation levels of the protein kinase C gamma type (Prkcg) at T655, Ras/Rap GTPase-activating protein 1 (Syngap1) at S788, mitogen-activated protein kinase kinase 1 (Map2k1) at S222, and protein tyrosine kinase 2 (Ptk2) at S910, and decreased levels of the cAMP-dependent protein kinase type I-alpha regulatory subunit (Prkar1a) at S77 (Fig. 2F). We also observed that the levels of phosphorylation of most proteins were increased. Thus, 65 phosphoproteins exhibited reduced phosphorylation in the mutant mice compared to the WT ones, while 172 phosphoproteins displayed increased phosphorylation (For the full GO analysis, see Supplementary Table 2). To biochemically validate the MS results, we performed IP of SYNGAP1, which was then blotted against phospho-serine. We chose SYNGAP1 for biochemical validation as it is one of the ASD high-risk genes38. Consistent with the MS, the IP showed a similar trend of increase in the Shank3Δ4–22 mice compared to the WT mice, and a trend of reversal following the treatment (Supplementary Figure S1). System biology analysis of the proteins with significantly altered phosphorylation levels in KO compared to WT mice, without change in the protein levels. (A) Protein interaction network map and clustering visualization of the significantly changed phosphoproteins. (B) Enriched BP related to synaptic processes. (C) Enriched BP related to developmental processes. (D). Enriched BP related to mRNA metabolic processes. (E) Venn Diagram comparing the BP ‘Developmental process’ and the mammalian phenotype of the ‘Nervous system phenotype’. (F) Examples of proteins in the common area in the venn diagram presented in panel E, whose phosphorylation levels were significantly altered by the KO. (G) Table showing seven proteins, whose expression and phosphorylation levels were changed significantly in different directions. *P < 0.05 ; **P < 0.01; ***P < 0.001. Additionally, we examined proteins that had significantly altered levels of both protein phosphorylation and expression. When these changes were in the same direction, i.e., both protein phosphorylation and expression were significantly increased or decreased, the change in protein phosphorylation levels could be at least partially explained by the change in its expression. But here, we highlighted proteins whose alterations in phosphorylation and expression levels were opposite, which emphasizes the significant changes in phosphorylation independent of their expression levels. We found seven proteins of KO vs. WT mice whose changes in phosphorylation and expression levels were in opposite directions (Fig. 2G). For example, growth-associated protein 43 (Gap43) showed increased protein expression in the KO group, while the phosphorylation levels at S96 were significantly decreased. Another example is small ribosomal subunit protein eS6 (Rps6), which showed increased protein expression and decreased phosphorylation levels at S240. Systems analysis of cortical phosphoproteome alterations following 7-NI treatment in the Shank3Δ4–22 mice After establishing the baseline alterations of phosphorylation in the Shank3Δ4–22 mice, we sought to find the phosphorylation events that are regulated by NO signaling. We treated the mice with a selective inhibitor of nNOS (7-NI, at a dose of 80 mg/kg, via IP injection), and this was followed by analysis of all phosphoproteins significantly affected in the KO mice treated with 7-NI compared to the KO mice treated with vehicle. 434 proteins with 591 phosphorylation sites were significantly altered, which are shown in a Volcano plot (Fig. 3A). The network of proteins was divided into six clusters: ‘mRNA splicing, major pathway’, ‘Translation associated terms’, ‘Endocytosis and vesicular transport’, ‘Growth cone’, ‘Cytoskeleton organization associated terms’, and ‘Synapse organization and Postsynaptic specialization’ (Fig. 3B). GO analysis of the phosphoproteins altered by the 7-NI treatment showed BP similar to those of KO vs. WT, including ‘Nervous system development’ (FDR = 1.14 × 10− 22), ‘Neurogenesis’ (FDR = 1.25 × 10− 17), and ‘Synapse organization’ (FDR = 2.29 × 10− 15) (Fig. 3C). This indicates that the NO inhibition treatment affects the same processes altered by the mutation. System biology analysis of the proteins with significantly altered phosphorylation levels in KO (Shank3Δ4–22 mice treated with 7-NI (KO + 7-NI group) compared to the untreated KO mice. (A) Volcano plot of the differentially expressed cortical phosphoproteins in the KO + 7-NI mice. Significantly changed phosphoproteins (P < 0.05) are colored red. (B) Protein interaction network map and clustering visualization of the significantly changed phosphoproteins. (C) BP (marked in blue font) and CC (marked in red font) enrichment. (D) Enriched reactome pathways. (E) BP analysis of the upregulated (Green) and the down-regulated (Turquoise) phosphoproteins. CC analysis of the treated KO mice showed enrichment in synaptic compartments, such as ‘Postsynaptic specialization’ (FDR = 8.3 × 10− 36), ‘Synaptic membrane’ (FDR = 1.57 × 10− 18), and ‘Dendritic spine’ (FDR = 2.01 × 10− 16) (Fig. 3C). The mammalian phenotype ontology in these mice appeared to be enriched with ASD-related pathways, including ‘Abnormal learning/memory/conditioning’ (FDR = 6.24 × 10− 14), ‘Abnormal emotion/affect behavior’ (FDR = 1.11 × 10− 9), and ‘Abnormal anxiety-related response’ (FDR = 3.22 × 10− 9) (For the full GO analysis, see Supplementary Table 3). The Reactome pathways analysis revealed pathways related to synaptic function and NMDA receptor activation (Fig. 3D). The results of the GO analysis identified several unique BPs among upregulated phosphoproteins and unique processes to downregulated phosphoproteins. Analyzing the downregulated and upregulated phosphoproteins and their unique BPs separately is important for reflecting distinct biological mechanisms that might be direction-specific. Unique upregulated BPs include ‘mRNA processing’, ‘Positive regulation of neuron projection development’, and ‘Gene expression’. In contrast, some BPs were unique to the downregulated phosphoproteins, including ‘Behavior’, ‘Cognition’, and ‘Regulation of autophagy’ (Fig. 3E). Some enriched terms related to autophagy and apoptosis were upregulated in the KO vs. WT mice, including the BP ‘Lysosomal transport’ (FDR = 0.017) and the molecular function ‘SNARE binding’ (FDR = 0.0329), and 7-NI treatment downregulated the terms, such as the BP ‘Regulation of autophagy’ (FDR = 0.0373) and the reactome pathway ‘Macroautophagy’ (FDR = 0.0010) (Supplementary Figure S2B). Phosphorylation levels of several autophagy-related proteins were increased by the Shank3Δ4–22 mutation and reversed by 7-NI treatment. For example, this treatment inverted the elevated phosphorylation of the RB1 Inducible Coiled-Coil 1 (Rb1cc1) at S646 and the transmembrane protein 41b (Tmem41b) at S10, while the autophagy-related protein 9a (Atg9a) at S18 showed increased phosphorylation in the KO compared to the WT; it did not show a significant decrease with 7-NI treatment (Supplementary Figure S2A). Systems analysis of cortical phosphorylation-specific alterations in the 7-NI-treated Shank3Δ4–22 mice To characterize which phosphorylation-specific events are affected by NO signaling and not due to the effect of NO inhibition on the protein expression, we narrowed our data pool to 329 of these proteins with 470 phosphorylation sites that did not show protein expression alteration. These proteins were categorized into six clusters: ‘Metabolism-associated terms’, ‘mRNA splicing-major pathway’, ‘Cellular organization’, ‘Growth cone’, ‘Synaptic transmission regulation’, and ‘Postsynaptic specialization’ (Supplementary Figure S3A). Next, we performed BP analysis and grouped the enriched BPs into three categories: ‘Synaptic biological process’ (Supplementary Figure S3B), ‘Developmental biological process’ (Supplementary Figure S3C), and ‘mRNA metabolism biological process’ (Supplementary Figure S3D). The enriched BPs that appear to be caused by the mutation are similar to those seen influenced by the treatment of 7-NI, meaning the mutation and treatment affect common biological processes, For example, ‘Neurogenesis’ (FDR = 3.25 × 10− 18), ‘Nervous system development’ (FDR = 1.54 × 10− 22), ‘Regulation of mRNA metabolic process’ (FDR = 0.0025), ‘Postsynapse organization’ (FDR = 4.97 × 10− 12), and ‘Trans-synaptic signaling’ (FDR = 1.82 × 10− 7). We also compared proteins common to the two enriched terms: the BP ‘Nervous system development’ and the mammalian phenotype ‘Abnormal behavior’ (FDR = 4.84 × 10− 19). In total, 57 phosphoproteins participating in both terms were identified (Supplementary Figure S3E). Some examples of these significantly phosphorylated proteins with their phosphorylation sites are: the myristoylated alanine-rich C-kinase substrate (Marcks) at S156, Synapsin 1 (Syn1) at S605, the neural cell adhesion molecule 1 (Ncam1) at S774, and Protein Bassoon (Bsn)-S2860 (Supplementary Figure S3F). We performed biochemical validation of Synapsin 1 using western blot method, and we found that the biochemistry confirms the MS analysis (Supplementary Figure S4). For the full MS and GO analysis, see Supplementary Table 4. Systems analysis of cortical phosphorylation alterations in the Shank3Δ4–22 KO, which were rescued by 7-NI treatment Then, we searched for the phosphoproteins commonly altered in the Shank3Δ4–22 and WT mice (KO vs. WT groups) and Shank3Δ4–22 mice treated and untreated with 7-NI (KO + 7-NI vs. KO groups). We identified 115 proteins with 128 phosphorylation sites whose phosphorylation levels were significantly altered by both the mutation and the 7-NI treatment (in the KO mice) (Fig. 4A-B). The 7-NI treatment reversed most of these protein phosphorylation changes caused by the mutation, as shown in the heatmap (Fig. 4C). A chord plot analysis was conducted to show the relationships between different proteins and the biological processes in which they are involved (Fig. 4D). The BP analysis indicates that these 115 proteins are involved in the ‘Synapse organization’, ‘Behavior’, ‘Dendrite development’, ‘Regulation of AMPA receptor activity’, and ‘Regulation of neuronal synaptic plasticity’. Only 15 out of the 128 phosphorylation level alterations were not reversed by the treatment, meaning 113 phosphorylation sites on 101 proteins were reversed. In contrast, treatment exacerbated the alterations in the 15 phosphorylation sites of these proteins, further increasing the magnitude of both the upregulated and downregulated phosphorylation changes. The Reactome pathway analysis showed enriched nervous system-related pathways (For the full MS and GO analysis, see Supplementary Table 5). These proteins participate in a variety of mammalian phenotype terms that are related to ASD and nervous system abnormalities, such as ‘Abnormal anxiety-related response’ (FDR = 0.0058), ‘Abnormal emotion/affect behavior’ (FDR = 6.58 × 10− 6), and ‘Abnormal learning/memory/conditioning’ (FDR = 2.69 × 10− 9) (Supplementary Table 5). System biology analysis of the proteins with phosphorylation levels altered by both KO and 7-NI treatment and analysis of the proteins related to the ASD risk genes whose phosphorylation was altered in the KO mice and reversed with 7-NI treatment (A) Venn diagram representing 128 phosphorylation sites significantly altered by both Shank3Δ4–22 KO (341 phosphorylation sites) and the 7-NI treatment (591 phosphorylation sites) (128 phosphoryltion sites on 115 proteins - protein group A). (B) Protein-protein interaction network map of group A. (C) Heatmap analysis of phosphorylation intensity of the 115 proteins in 128 phosphorylation sites whose phosphorylation was commonly altered in all three groups of mice. (D). Chord chart of biological processes and proteins that participate in them. The width of the BP term (a segment of the circle) reflects the number of proteins participating in this process. The chord width corresponds to the log2FC of the phosphoprotein. (E) Venn diagram comparing the 101 phosphoproteins (with 113 phosphorylation sites) whose phosphorylation was reversed by the treatment, compared to the proteins encoded by the human ASD risk genes (21 proteins, with 23 phosphorylations- group B) (F) Protein interaction network map for group B categorized into 3 clusters. (G) The number of the significantly upregulated and downregulated phosphoproteins in the KO compared to WT groups of mice (group B) (H) The number of the significantly upregulated and downregulated phosphoproteins in the treated (KO + 7-NI) vs. untreated (KO) groups of mice. Next, we aimed to investigate whether the significantly rescued proteins in the Shank3Δ4–22 cortex by the 7-NI treatment overlap with the human ASD high-risk encoded genes curated by the SFARI database (https://gene.sfari.org/). This database contains ASD-associated information updated in the first quarter of 2025 and includes 1219 genes. Out of the 101 reversed proteins, 21 proteins were identified as ASD risk genes according to the SFARI database, comprising 23 phosphorylation sites whose phosphorylation levels were altered in KO mice and restored following 7-NI treatment (Fig. 4E). Clustering analysis of these 21 proteins shows involvement of ‘Presynaptic depolarization and calcium channel opening’, ‘Regulation of protein localization to synapses’, and ‘Cysteine-type deubiquitinase activity’ (Fig. 4F). These 21 ASD-related proteins participate in BPs, such as ‘Dendrite development’ (FDR = 0.00051), ‘Regulation of synaptic plasticity’ (FDR = 0.0023), and ‘Regulation of synapse structure or activity’ (FDR = 0.0329) (For the full MS and GO analysis see Supplementary Table 5). They exhibit terms of the mammalian phenotype that are relevant to ASD and CNS impairments, like ‘Abnormal excitatory postsynaptic currents’ (FDR = 0.016), ‘Impaired synaptic plasticity’ (FDR = 0.016), ‘Abnormal social/conspecific interaction behavior’ (FDR = 0.016), and ‘Abnormal behavior’ (FDR = 0.016) (Supplementary Table 5). When we examined the phosphorylation patterns of these 21 phosphoproteins, we found that most of them are uniquely phosphorylated in either the KO, WT or the 7-NI-treated group (Fig. 4G-H). The phosphoproteins of each of the comparison pairs of groups (WT vs. KO and KO + 7-NI vs. KO) were divided into upregulated and downregulated phosphoproteins (Fig. 4G-H). Most of these proteins appeared to undergo increased phosphorylation, and more than half displayed unique phosphorylation. For example, in the KO mice, the proteins were mostly phosphorylated on the sites not detected in the WT group. In the second comparison pair of groups (KO + 7-NI vs. KO), for most proteins uniquely phosphorylated in KO mice, phosphorylation levels were reduced by treatment to undetectable levels in the KO + 7-NI group (Fig. 4H). These findings indicate a role for NO signaling in driving these phosphorylation events. Systems analysis of kinases and phosphatases with altered protein expression and phosphorylation levels in the cortex of Shank3Δ4–22 mice Given the widespread occurrence of NO-associated changes in phosphorylation patterns, we next examined alterations in kinases and phosphatases in our dataset. We examined changes in kinases and phosphatases expression, as well as site-specific phosphorylation alterations in KO mice compared to the WT mice, and assessed how 7-NI treatment influenced these alterations. Using the Panther classification system, global proteomics analysis identified 28 kinases and phosphatases with significantly altered protein levels in the KO mice compared to WT mice. Of these, 24 were upregulated, and 4 were downregulated. Treatment reversed the expression of 9 upregulated kinases and phosphatases (Fig. 5A- right panel, inner circle). None of the downregulated proteins were rescued by the treatment (Fig. 5A, B). In addition, we found 23 kinases and phosphatases displaying significantly altered phosphorylation across 25 phosphorylation sites in the KO compared to the WT mice (Fig. 5A- left panel, numbers correspond to phosphorylation sites). Among these, phosphorylation levels of only 4 of these proteins appeared to be downregulated, while 21 were upregulated. Treatment with 7-NI reversed phosphorylation at 10 sites, of which only one corresponded to a downregulated site (Fig. 5A, B). The BP analysis indicated the involvement of the 28 kinases and phosphatases of KO mice, whose protein expression was significantly changed compared to their WT counterparts, in signaling and developmental processes like ‘Regulation of signaling’ (FDR = 8.22 × 10− 5), ‘Regulation of signal transduction’ (FDR = 8.53 × 10− 5), ‘Nervous system development’ (FDR = 0.0146), and ‘Neuron development’ (FDR = 0.0407) (Supplementary Table 6). CC analysis showed that these kinases and phosphatases were enriched in ‘Axon’ (FDR = 0.0168) and ‘Neuron projection’ (FDR = 0.0224) (Supplementary Table 6). We created a heatmap of the 9 proteins whose expression was reversed by the treatment according to their MS signal intensity (Fig. 5C, upper panel). These proteins included Diacylglycerol kinase gamma (Dgkg), Serine/threonine-protein phosphatase 2 A catalytic subunit alpha (Ppp2ca), and Phosphatidylinositol-3,5-bisphosphate 3-phosphatase MTMR2 (Mtmr2). Systems biology analysis of kinases and phosphatases with significantly altered protein expression and phosphorylation levels in the Shank3Δ4–22 KO vs. WT mice. (A) Venn diagram comparing the 28 kinases and phosphatases with significantly altered protein expression and proteins with reversed expression by the treatment (right panel), and a Venn diagram comparing the number of sites of kinases and phosphatases with significantly altered phosphorylation levels and proteins with phosphorylation levels reversed by the treatment (left panel). (B) Network of the proteins with changed expression levels (upper panel) and the proteins with changed phosphorylation levels (lower panel). (C) Heat map of the 9 kinases and phosphatases where protein expression was reversed by the treatment (upper panel) and Heat map of the 10 kinases and phosphatases where the phosphorylation levels were reversed by the treatment (lower panel). (D) Cellular component enrichment of the kinases and phosphatases with altered phosphorylation levels. (E) Reactome pathways enrichment of the kinases and phosphatases with altered phosphorylation levels. (F) Examples of kinases and phosphatases whose expression or phosphorylation levels (signposted with the phosphorylation sites) were altered in KO mice but reversed by the 7-NI treatment. (G) Diagram of the upstream and downstream regulators of Mink1 kinase and SynGAP1 that showed altered phosphorylation and reversal with 7-NI treatment. Then, we analyzed which kinases and phosphatases exhibited altered phosphorylation levels. A protein network was created to visualize the connections between these proteins (Fig. 5B, lower panel), and a heatmap of the 10 proteins with the phosphorylation levels reversed by the treatment (Fig. 5C, lower panel). These proteins participate in different BP, including ‘Modulation of chemical synaptic transmission’ (FDR = 5.98 × 10− 5), ‘Nervous system development’ (FDR = 0.0025), ‘Regulation of synaptic plasticity’ (FDR = 0.0264), and ‘Regulation of synaptic transmission, glutamatergic’ (FDR = 0.0264) (For the full GO analysis see Supplementary Table 6). They were enriched in neuron-related CC, such as ‘Synapse’ (FDR = 1.1 × 10− 7), ‘Postsynaptic density’ (FDR = 4.78 × 10− 6), ‘Glutamatergic synapse’ (FDR = 9.17 × 10− 6), and ‘Dendrite’ (FDR = 1.65 × 10− 5) (Fig. 5D). We also observed enrichment in the following Reactome pathways: ‘Trafficking of AMPA receptors’ (FDR = 8 × 10− 5), ‘Unblocking of NMDA receptors, glutamate binding and activation’ (FDR = 0.00068), and ‘Signal transduction’ (FDR = 0.0053) (Fig. 5E). The sole kinase that showed decreased phosphorylation levels in the KO group of mice and reversal following the treatment was misshapen-like kinase 1 (Mink1) phosphorylated at S746 (Fig. 5F). This kinase is involved in AMPA receptor trafficking and in the JNK signaling pathway, and is also known to be implicated in ASD and Alzheimer’s disease (AD)39–41. Importantly, the expression levels of its upstream effector Ppp2ca, which is shown to inhibit the activity of Mink142 were significantly altered in the KO group, but reversed by the treatment (Fig. 5F-G). Another phosphoprotein involved in the AMPA receptor trafficking is SynGAP143, whose phosphorylation on the S788 was found to be increased in the KO mice group compared to the WT and the treatment reversed the phosphorylation levels to a similar level of the WT group (Fig. 5F). One of the upstream effectors of SynGAP is Calcium/Calmodulin-Dependent Protein Kinase II Alpha (Camk2a)44, and one of its’ regulatory autophosphorylation sites (S275)45 is also seen increased in the KO mice group compared to the WT group and showed reversal in the 7-NI treatment group (Fig. 5F-G). Collectively, these results suggest that NO signaling may be associated with changes in kinase and phosphatase activity underlying the observed phosphorylation patterns. Discussion Our previous studies in two Shank3 ASD mouse models (InsG3680(+/+) and Shank3Δ4–22) and the Cntnap2−/− model revealed that excessive NO levels resulting from nNOS overactivation lead to aberrant NO-related PTMs, including S-nitrosylation (SNO), and increased nitrosative stress in the autistic brain, which together play a crucial role in ASD pathology19–22, 25. This conclusion is consistent with previous postmortem studies of ASD patients showing the accumulation of 3-nitrotyrosine29, a marker of nitrosative stress28, in the brain. Another important form of PTM is protein phosphorylation, involved in the regulation of all biological processes, both in physiological and pathological conditions 14. The phosphorylation levels directly depend on the activity of kinases and phosphatases46, which can also be influenced by other forms of PTMs, such as NO-dependent SNO, or tyrosine nitration47–52. Accordingly, this study aimed to investigate the interplay between these two forms of signaling in the autistic brain of Shank3Δ4–22 mice. To this end, we applied a selective nNOS inhibitor, 7-NI, to investigate the impact of NO signaling on the protein phosphorylation pattern in the cortex of Shank3Δ4–22 mice using phosphoproteomics and global proteomics, followed by a comprehensive systems biology analysis. The KO mice in our study displayed significant alterations in protein phosphorylation, and the pharmacological inhibition of nNOS largely reversed these abnormal changes, indicating a link between protein phosphorylation and NO levels in the brain. To the best of our knowledge, this is the first study exploring a link between excessive NO production and altered phosphorylation in the Shank3Δ4–22 model of ASD. Gene ontology analysis of the changes in the phosphoproteome of the KO mouse cortex showed enrichment of the developmental biological processes, such as ‘Nervous system development’, ‘Neurogenesis’, ‘Neuron differentiation’, as well as behavioral abnormalities like ‘Abnormal behavior’, ‘Memory and learning’, and ‘Anxiety response’, the deficits common to ASD in general and this ASD model in particular. Several studies have previously reported the involvement of impaired synaptic function and excitation/inhibition (E/I) dysbalance in ASD53–56. Pharmacological interventions aimed at reversing these abnormalities have shown the potential to attenuate ASD symptoms54, 57. We have recently found that 7-NI reverses synaptic and behavioral deficits in mouse models of ASD22. In particular, this work showed that treatment of the two ASD models, Shank3Δ4–22 and Cntnap2−/−, with 7-NI could counteract the downregulation of some important synaptic proteins, including glutamatergic NMDA receptor subunit 1, the synaptic vesicle protein synaptophysin, the glutamate decarboxylase 1, the vesicular inhibitory amino acid transporter, the postsynaptic density (PSD) scaffold protein Homer, and the membrane-associated guanylate kinase family member protein PSD95. These proteins are crucial for inhibitory and excitatory neurotransmission, synaptic plasticity, and maintenance of E/I balance. Importantly, a similar beneficial effect was observed in the human-derived induced pluripotent stem cells with the SHANK3 mutation. The reversal of the synaptic phenotype was accompanied by the rescue of ASD-like behavioral deficits in these mice treated with 7-NI22. Our analysis revealed that synaptic organization and neurodevelopmental processes are the main biological processes affected by the Shank3Δ4–22 mutation. Examination of the enrichment in the treatment (KO + 7-NI) group of mice showed that similar biological processes were affected by nNOS inhibition, linking NO signaling to synaptic development and organization in this ASD model. The different phosphoproteins that have been altered in our data can be divided into several clusters with different and distinct biological functions. The main biological cluster we found to be affected in our data is ‘AMPA receptor trafficking and activity’. The biological process ‘Regulation of AMPA receptor activity’ was considerably enriched and the Reactome pathway ‘Trafficking of AMPA receptors’ was also enriched. The AMPA receptors are ionotropic glutamatergic receptors that participate in neurodevelopment, un-silencing of the glutamatergic synapses, and excitatory neurotransmission58–60. Part of the genes encoding these receptor subunits are known as ASD risk genes. For instance, it has been found that mutations in the genes encoding AMPA receptor subunits GluA1-3 (GRIA1-3 genes, respectively) are linked to neurodevelopmental disorders, including ASD60–62. As AMPA receptor trafficking is a vulnerable pathway in ASD originating from different etiologies63, we were interested in observing which phosphoproteins are affected in the ASD brains and are influanced by NO signaling. The most direct molecular link between NO-mediated phosphorylation alterations and AMPA receptor trafficking emerged from SynGAP1, where we found increased phosphorylation of SynGAP1 at S788 in Shank3 KO mice. The 7-NI treatment reduced the phosphorylation levels of this protein to those of WT mice. SynGAP1 is a major component of the postsynaptic density in excitatory synapses64. It is a part of the NMDA receptor complex involved in AMPA receptor trafficking to the synaptic surface65. This protein also regulates synaptic plasticity and cognitive processes66. It interacts with PSD95, a major scaffolding protein in the postsynaptic density64. SYNGAP1 is one of the risk genes for ASD38. S788 phosphorylation on SynGAP1 is known to be affected by the cyclin-dependent kinase 5 (CDK5)67. Phosphorylation at this site is known to inhibit the GTPase-activating protein activity of SynGAP1. NMDAR-CamK2a-SynGAP1 signaling known to affect the recruitment and activity of AMPA receptors into the synapse65. Interestingly, Mink1 was the only kinase that showed a decreased phosphorylation (at S746) in the KO mice. This change was also attenuated by the 7-NI treatment. Mink1 is a kinase known to be localized in the postsynaptic density and to regulate neuronal morphology and surface expression of AMPA receptors68. It has been found that Mink1 prevents the depletion of surface AMPA receptors mediated by a Ras-related protein 2, Rap268. The effect of phosphorylation of this protein at S746 and the mechanism underlying its involvement in the AMPA receptor localization and function are currently unknown. It can be speculated that the reduction of Mink1 phosphorylation at this site impairs AMPA-mediated excitatory neurotransmission and synaptic plasticity. SynGAP1 and Mink1 both affect Rap2 which is a small GTPase promoting the AMPA receptor removal from the synapse surface68,69 (Fig. 5G). Another phosphoproteins related to the biological cluster of ‘AMPA receptor trafficking and activity’ are Gap43 and Prkcg. we found a significant reduction in phosphorylation levels of Gap43 at S96 in the KO group. Remarkably, the decrease in Gap43-S96 phosphorylation occurred despite the increased expression of this protein, emphasizing the significance of this alteration. Gap43 is associated with axonal growth. It is produced in the cell body of neurons during the prenatal and early postnatal stages, and it is quickly transported mainly to growth cones and axon terminals70,71. Phosphorylation of this protein at S96 is linked to increased axon growth72 and its localization in specific regions, such as the hippocampus73. Previous literature demonstrated the involvement of Gap43 in AMPA receptor endocytosis and LTD74. In ASD adults, increased levels of Gap43 are thought to be due to axonal damage and inflammation75. Previous work has shown S-nitrosylation of Gap43 in the Shank3 InsG3680(+/+) ASD mouse model, as well as in the P301S Alzheimer’s disease mouse model19,76. We assume that the increased SNO might affect the phosphorylation and protein levels of Gap43, while increased expression levels of this protein could be a compensatory reaction to its initial loss caused by the elevated NO levels in the KO mice. This AMPA trafficking defect is compounded by Prkcg overactivation, which increased phosphorylation of the Prkcg at T655 was found in the KO mice and it was reversed with the 7-NI treatment. T655 phosphorylation on Prkcg is critical for its regulation and full activation77,78. Prkcg is a Serine\Threonine protein kinase that is mainly expressed in the CNS, including the cortex, hippocampus, and amygdala79.This kinase physically binds to GluR4 subunit of AMPA receptors and regulates its function by increasing the membrane association of the GluR4 subunit80. The group of protein kinase C enzymes is implicated in the regulation of gene expression, control of cell division and differentiation, and signal transduction81. The PRKCB1 gene was previously found to be related to ASD82, and the abnormal Prkcg activity has been suggested to play a crucial role in ASD pathogenesis through impairments in β-catenin-regulated neurogenesis83. Second biological cluster we found to be affected is ‘mTOR pathway and protein synthesis’. Previous studies showed the involvement of mTOR pathway in the pathophysiology of ASD and the involvement of NO signaling in the mTOR disturbance in the Shank3Δ4–22 mouse model of ASD33. SynGAP1 also has a biological role in the mTOR activity modulation. as we mentioned previously - S788 phosphorylation on SynGAP1 is known to be affected by CDK5 and the phosphorylation at this site is known to inhibit the activity of SynGAP1. SynGAP1 is a known negative regulator of the mTOR pathway84,85 and this phosphorylation site is crucial in the SynGAP1 mediated mTOR regulation. Downstream to the mTOR signaling, the small ribosomal subunit protein Rps6 was found to be increased in its protein expression but decreased in phosphorylation levels at S240. This phosphorylation site is regulated by the mechanistic target of rapamycin (mTOR) signaling pathway via the ribosomal protein S6 kinase86–88. This signaling pathway has been found to play an important role in ASD pathogenesis33,89. Consequently, the genetic ablation of Shank3 in the ASD mouse model used in this study could cause changes in the mTOR signaling, resulting in Rps6 phosphorylation and expression changes described above- some which already have been show to occur and be NO dependent in previous study33. A third biological cluster enriched in our data is ' Ras/MAPK signaling axis’. Ras/MAPK signaling pathway is known to be implicated in the pathogenesis of ASD90. As previously thoroughly discussed, SynGAP1 is also involved in the Ras/MAPK signaling and it is one of the mechanisms SynGAP1 influences AMPA receptor levels on the surface of the synapse91. Another Phosphoprotein belonging to this cluster is MAP2K1 (Dual specificity mitogen-activated protein kinase kinase 1). Shank3Δ4–22 mutation in mice significantly increased phosphorylation at S222 of MAP2K1, which is an important site for the kinase’s activation92,93. Ras/MAPK pathways have been shown to be involved in the pathogenesis of the RASopathies, a group of disorders induced by germline mutations in several genes encoding the components of Ras/MAPK pathways94. One of these disorders is cardiofaciocutaneous (CFC) syndrome, caused by mutations of MAP2K1 and a few other genes. Importantly, individuals with CFC syndrome display a high prevalence of ASD94, which links MAP2K1 to autism. Not all changes are NO-dependent; one example is Ptk2 (also known as focal adhesion kinase, FAK)95, which showed enhanced phosphorylation at S910 in KO mice. This phosphorylation was not significantly affected by the nNOS inhibition with 7-NI. S910 phosphorylation of Ptk2, a protein related to an ASD high-risk gene PTEN96, is mediated by extracellular-regulated kinase (ERK) which is a downstream kinase in the Ras/MAPK pathway97, and is important for the regulation of cell shape and proliferation98–100. Not all changes in the data appear to be induced by the mutation itself and only the 7-NI treatment leads to them. The last biological clustering we have identified in our data is ‘Vesicular dynamics’, which is known to be implicated in neurodevelopmental disorders and in ASD specifically101. It is being effected by the 7-NI treatment with no pathological basic change. The first phosphoprotein relating to synaptic vesicle regulation is Marcks. Considerably increased phosphorylation levels of S156 were observed in Marcks in the KO mice treated with 7-NI compared to the untreated ones. The phosphorylation levels of this protein were insignificantly reduced by the Shank3 genetic ablation. S156 on Marcks has a meaningful role in the kinase’s function. Protein kinase Cδ phosphorylates Marcks on this site, which causes the translocation of phosphorylated Marcks from the membrane into the cytosol102–104. This protein takes part in the regulation of various cellular functions, such as membrane trafficking, cytoskeletal reorganization, inflammatory response, cell secretion, cell migration, etc.105. The role of Marcks in ASD pathology is unknown. However, previous studies have shown that Marcks is involved in synaptic vesicle trafficking and neurotransmitter release106, which are often disrupted in autism. Another phosphoprotein, which is important in synaptic vesicle regulation, is Synapsin1, whose phosphorylation levels were increased in the treatment group compared to the untreated KO mice. Synapsin1 was phosphorylated at S605. The Shank3 KO did not affect the phosphorylation levels of this protein compared to the WT, and the biochemical validation supports the MS results, displaying no change in phosphorylation levels on this site between KO and WT test groups, and only the treatment increased the phosphorylation. Synapsin1 is phosphorylated at S605 by calcium/calmodulin-dependent protein kinase II alpha, thereby modulating Synapsin1 activity and regulating neurotransmitter release14,107–112. This protein modification results in the dissociation of Synapsin1 from both synaptic vesicles and the actin cytoskeleton, promoting the mobilization of synaptic vesicles from the reserve pool to the readily releasable pool, facilitating neurotransmission110. Synapsin1 was previously found to undergo excessive S-nitrosylation in the Shank3 InsG3680(+/+) ASD mouse model19, which could affect its functions. Finally, Ncam1 emerged as an additional phosphoprotein of interest, which is a molecule involved in neuron-neuron adhesion, synaptic stabilization, memory, and learning113–115. This protein was previously linked to ASD and other neurodevelopmental disorders116–118. Ncam1 phosphorylation at S774 was reduced in the KO mice, and the 7-NI treatment reversed this alteration. The S774 phosphorylation site on Ncam1 is crucial for its activity119,120. We have previously reported that this protein exhibited increased SNO in the Shank3 InsG3680(+/+) mouse model of ASD19,25, providing another link between NO, protein phosphorylation, and signaling pathways. Our recent study on Shank3Δ4–22 KO mice has highlighted alterations in autophagy-related proteins in the cortex23. Changes in the phosphorylation levels and expression of several autophagy-related proteins appeared to be implicated in the Shank3Δ4–22 pathology23. Importantly, these data are consistent with the results of the current study. Increased protein and phosphorylation levels of the autophagy system proteins, including Rb1cc1 S646, Atg9a S18, Tmem41b S10, and Cltc were reversed by 7-NI treatment of the KO mice. Some BP and Reactome pathways enriched in the cortex of the KO mice were associated with autophagy. Meanwhile, the treatment downregulated BP and reactome pathways associated with autophagy. These results confirm the involvement of the autophagy-related proteins in ASD-like phenotypes. Taken together, we observed changes in phosphorylation levels of various proteins, including ASD-associated kinases and phosphatases. The nNOS inhibition resulted in significant reversal in phosphorylation of key proteins caused by the Shank3 mutation, providing a link between NO signaling and phosphorylation of different proteins in the Shank3 ASD mouse model. Thess changes affect protein stability, protein-protein interaction and function, causing impairments in CNS signaling, AMPA receptor trafficking, vesicular dynamics, mTOR pathway, and MAPK signaling.. In conclusion, our findings show an interplay between the NO signaling and the phosphorylation of various proteins involved in the regulation of neurotransmission in the brain, as many of the proteins affected are mainly neuronal and synaptic. The results of this study show that in the Shank3Δ4–22 mouse model of ASD, the aberrant NO signaling induced by the excessive NO levels leads to abnormal protein phosphorylation patterns, which mechanistically might underlie the ASD-like synaptic and behavioral phenotypes that have been previously shown to be rescued by nNOS inhibition treatment. Further investigation of the cross-talk between SNO and tyrosine nitrosylation to phosphorylation observed in the Shank3Δ4–22 mice might help unravel the biological mechanisms of this disorder and find new molecular targets for ASD treatment. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 2 Supplementary Material 3 Supplementary Material 4 Supplementary Material 5 Supplementary Material 6 Supplementary Material 7 Acknowledgements We thank the Satell Family Foundation and the Neubauer Family Foundation for their support. We thank The Smoler Proteomic Center at the Department of Biology, Technion for the LC-MS/MS analysis. Author contributions ShG: Run experiments, analyzed data, and wrote the paper.IK: Wrote ms and reviewed data.MK: Analyzed data and revised the draft.WH: Analyzed data and revised the draft.HA: Supervised and executed the project. Revised the paper. Funding This work was funded by a US Department of Defense (DoD) grant, a DFG – German Research Foundation Grant, an Israeli Science Foundation (ISF) grant, an Eagles Autism Foundation grant, a National Institute of Psychobiology in Israel (NIPI) grant, and a Berettler Centre for Research in Molecular Pharmacology and Therapeutics Grant. Data availability Data is provided within the manuscript or supplementary information files. The mass spectrometry phosphoproteomics data have been deposited with the ProteomeXchange Consortium via the PRIDE repository under the dataset identifier PXD068474. Declarations Competing interests HA is a CSO of Point6 Bio and Neuro-NOS. No funds from both companies were received for this study. All other authors do not hold any competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Supplementary Material 2 Supplementary Material 3 Supplementary Material 4 Supplementary Material 5 Supplementary Material 6 Supplementary Material 7 Data Availability Statement Data is provided within the manuscript or supplementary information files. The mass spectrometry phosphoproteomics data have been deposited with the ProteomeXchange Consortium via the PRIDE repository under the dataset identifier PXD068474.