Mll5 haploinsufficiency attenuates microglial phagocytosis through dysregulated TREM2-SGK3-GSK3β signaling in autism

Mll5 haploinsufficiency attenuates microglial phagocytosis through dysregulated TREM2-SGK3-GSK3β signaling in autism | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Mll5 haploinsufficiency attenuates microglial phagocytosis through dysregulated TREM2-SGK3-GSK3β signaling in autism Shumin Gao, Qingxiu Lin, Xiaotong Liu, Meixiang Jia, Anyi Zhang, and 12 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6665030/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder characterized by persistent deficits in social communication and repetitive behaviors. Recent studies indicated that heterozygous mutations in the mixed lineage leukemia 5 ( MLL5 ) gene are implicated in the ASD susceptibility and associated with neurodevelopmental abnormalities. However, the detailed mechanisms remain unclear. Here, we demonstrate that Mll5 haploinsufficiency in mice impairs microglial phagocytosis, drives neuronal hyperexcitability, and recapitulates core ASD-like behaviors. We also show that Mll5 acts as an epigenetic regulator, modulating microglial phagocytosis via the TREM2-SGK3-GSK3β signaling axis, which is associated with deficient glucose metabolism. Furthermore, ASD individual-derived microglia exhibit parallel reductions in MLL5 expression and phagocytic function. By targeting this pathway, lithium chloride, a GSK3β inhibitor, rescues both microglial phagocytosis deficits and behavioral abnormalities in Mll5 haploinsufficiency mice. Our findings highlight MLL5 ’s critical role in ASD and its potential as a therapeutic target. Autism spectrum disorder Mll5 microglia phagocytosis Sgk3 Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Autism spectrum disorders ( ASD ) is a neurodevelopment disorder characterized by impairment in social communication and interaction, along with restricted and repetitive behavioral patterns 1 . Individuals with ASD vary greatly in cognitive development and often comorbid with other neuropsychiatric disorders, making its clinical manifestations highly heterogeneous 2 . Although twin and family studies provide evidence that ASD is predominantly heritable, the exact underlying genetic determinants are largely unknown 3 , 4 . Recent large scale genomic studies have identified over hundreds of ASD susceptible genes, many of which are involved in chromatin modification and synaptic connections 5 , 6 . Mixed lineage leukemia 5 ( MLL5 ), also known as lysine (K)-specific methyltransferase 2E ( KMT2E ), is one of the genes associated with ASD in both genome-wide association study and whole exome sequencing studies 5 , 7 , 8 . Probands with MLL5 mutation bare either heterozygous variants or microdeletions, leading to haploinsufficiency of the MLL5 function. KMT2E -related neurodevelopmental disorder is a genetically defined condition caused by pathogenic variants in the KMT2E gene, manifesting core features of global developmental delay, mild-to-moderate intellectual disability, and central hypotonia 9 – 11 . This disorder is further associated with a heterogeneous clinical spectrum encompassing autism spectrum disorder (ASD), structural brain anomalies (microcephaly, macrocephaly), epilepsy, and subtle dysmorphic facial features 11 . Nevertheless, the exact neurobiological function of MLL5 in the developing central nervous system ( CNS ) is still unclear. MLL5 is suggested to act as a transcriptional regulator by affecting histone H3 lysine 4 (H3K4) methylation through direct or indirect mechanisms, exerting a diverse range of biological functions 12 , 13 . Previous study has been primarily investigated using cancer cell lines, which showed that MLL5 knockdown inhibited cell cycle progression and disrupted genomic stability. In vivo experiments have demonstrated that MLL5 was essential for normal hematopoiesis and spermatogenesis, possibly through its function in epigenetic modulation and reactive oxygen species production 14 . In the CNS, evidence have suggested that Mll5 was involved in regulating neuronal activity, visual function, and glucose metabolism 15 , 16 . Notably, Mll5 +/− mice exhibited social impairment and disturbed electroretinogram responses 15 , 16 . Additionally, the deficiency of Mll5 has been linked to deficits in immune surveillance in peripheral tissues 17 , 18 , suggesting its potential role in regulating microglial function within the CNS. Microglia are tissue-resident macrophages derived from myeloid progenitors generated in the yolk sac and migrate into the brain parenchyma prior to blood-brain barrier closure 19 , 20 . Besides its role in inflammatory response, recent studies have highlighted the participation of microglia in synaptic pruning and extracellular matrix remodeling during brain development, which are crucial for the synaptic maturation and neuronal circuit formation 21 – 23 . Dysfunction of microglia has been associated with several neurodevelopmental disorders including ASD 24 , 25 . Despite the known roles of MLL5 in regulating cellular processes, its specific contributions to microglial function within the CNS remains incompletely understood. In this study, we investigated the role of Mll5 in regulating microglia function during critical postnatal neurodevelopmental windows. We demonstrated that Mll5 +/− mice exhibit impaired hippocampal microglial phagocytosis and ASD-like behavioral phenotypes. Mll5 haploinsufficiency in microglia impaired the clearance of synapses and disrupted neuronal excitability. Mechanistically, Mll5 haploinsufficiency attenuated microglial phagocytic capacity by dysregulating metabolic pathways mediated by the TREM2-SGK3-GSK3β axis. Notably, microglia derived from individuals with ASD exhibited reduced MLL5 expression and diminished phagocytic activity, implicating MLL5 dysfunction in ASD-associated microglial pathophysiology. Pharmacological intervention with lithium chloride (LiCl) restored microglial phagocytic function and ameliorated ASD-like behaviors in Mll5 +/− mice. Collectively, our findings establish MLL5 as a critical regulator of microglial function during postnatal neurodevelopment, with direct relevance to ASD pathogenesis. Results Mll5 +/− mice display behavioral abnormalities Mutations in Mll5 have been implicated in a range of behavioral abnormalities associated with neurodevelopmental disorders 11 , 15 . Here, we determined the behavioral performances resulting from Mll5 haploinsufficiency in mice (Fig. 1 A and Supplementary Fig. 1A). Behaviorally, male Mll5 +/− pups (postnatal days (P) 9) that were separated from their mothers emitted ultrasonic vocalizations (USVs), which were recorded and analyzed to assess social communication deficits. Compared to their male wild type (WT) counterparts, male Mll5 +/− pups exhibited fewer calls, shorter total call durations, and a reduced mean call duration (Fig. 1 B). However, female Mll5 +/− pups did not exhibit abnormal USVs (Supplementary Fig. 1B). These results suggest atypical postnatal development of mother-pup communication in male Mll5 +/− pups. We used the three-chamber social test to assess social interaction behavior. To evaluate sociability, we compared the time that the experimental mouse spent interacting with either a stranger mouse chamber or an empty wired cup chamber. As expected, adult WT mice displayed normal sociability, as reflected by their preferential interaction with the stranger mouse. In contrast, Mll5 +/− mice showed no preference for the stranger mouse chamber over the empty cup chamber, indicating impaired sociability (Fig. 1 C and Supplementary Fig. 1C). Next, in a social novelty test that assesses social novelty recognition, Mll5 +/− mice spent less time in stranger 2 chamber (Fig. 1 C and Supplementary Fig. 1D), indicating impaired social novelty recognition. Mll5 +/− mice also exhibited higher levels of self-grooming (Fig. 1 D and Supplementary Fig. 1E) but buried a significantly lower number of marbles in the marble burying task (Fig. 1 E and Supplementary Fig. 1F). We found that Mll5 +/− mice produced low-quality nests compared to the WT mice in nest building test (Fig. 1 F and Supplementary Fig. 1G). Male Mll5 +/− mice showed increased immobility in forced swimming test (Fig. 1 G) and tail suspension test (Fig. 1 H), however no significant changes were observed in female Mll5 +/− mice compared to the control mice (Supplementary Fig. 1H, 1I), suggesting Mll5 +/− mice displayed depression-like behavior in a sex-specific manner. To evaluate recognition memory, novel object recognition tests were performed. Mll5 +/− mice did not show any preferences in exploring two identical objects compared to the WT mice (Fig. 1 I and Supplementary Fig. 1J). Additionally, the Morris water maze test was used to investigate the spatial learning and memory. Our results indicated that both WT and Mll5 +/− mice spent a similar amount of time in the platform quadrant and had a comparable number of platform crossings during the test (Fig. 1 J). Moreover, there was no significant difference in swimming speed between the two groups (Supplementary Fig. 1K). These results suggested that deficiency of Mll5 in mice resulted in autistic-like behaviors, including abnormal vocal communication, social deficits and stereotyped behaviors, accompanied depressive-like behaviors. Mll5 +/− mice exhibit overall normal brain morphology but reduced microglial activity during postnatal brain development MLL5 mRNA expression was detected throughout all brain regions during development and adulthood, peaking around the perinatal period and gradually decreasing after birth (Supplementary Fig. 2A, B). MLL5 was expressed in all cell types of the adult brain, with a higher expression in glia cells compared to neuronal cells (Supplementary Fig. 2C, D). Previous study has shown that full deletion of Mll5 ( Mll5 −/− ) led to significant postnatal growth retardation 18 , while we found Mll5 haploinsufficiency mice had comparable brain/body weight ratio and brain/body length ratio throughout development (Supplementary Fig. 2E, F). Adult Mll5 +/− brain appeared normal morphology and expected thickness using Nissl staining (Supplementary Fig. 3A). No obvious cortical lamination abnormality was observed when stained for CTIP2 + and CUX1 + neurons (Supplementary Fig. 3B, C). Consistently, Mll5 +/− embryonic brain had normal numbers of PAX6 + and KI67 + cells at embryonic day (E)14.5, indicating that neural progenitor cell proliferation was not affected by Mll5 deficiency (Supplementary Fig. 3D). We focused on the hippocampus in juvenile (P17-21) mice, a developmental window coinciding with active synaptic refinement. Microglial density was quantified through immunostaining for the microglia marker ionized calcium-binding adapter molecule 1 (IBA1) in brain sections from Mll5 +/− and WT littermates. No significant differences in microglial density were observed across subregions of the hippocampus (Fig. 2 A, B and Supplementary Fig. 4A) or in other brain regions, including medial prefrontal cortex (mPFC) (Supplementary Fig. 4A, 4B) and paraventricular nucleus of the hypothalamus (PVN) (Supplementary Fig. 4A, B). Consistent with these observations, fluorescence-activated cell sorting (FACS) of the CD11B + CD45 low hippocampal microglia revealed comparable proportions across genotypes (Supplementary Fig. 5A). Morphological analysis identified a hyper-ramified phenotype specifically in the Mll5 +/− hippocampi, characterized by an increased number of branches and junctions per cell (Fig. 2 C, D), indicative of reduced microglial activation compared to WT. Consistently, we found decreased CD68-immunolabeled microglial phagosomes per IBA1 + volume in the Mll5 +/− mice (Fig. 2 E). Furthermore, FACS analysis of the CD11B + CD45 low microglia indicated lower mean fluorescence intensity (MFI) of CD40 (Figure S5B), while expression levels of major histocompatibility complex class II (MHCII) and CD11B were comparable between Mll5 +/− mice and its littermates (Supplementary Fig. 5B). Although MLL5 was reported to suppress innate immune response in peripheral macrophages, we did not observe significant elevated inflammatory status in the Mll5 +/ − brain. The mRNA levels of Tnf , Ccl2 and Tgfb1 , as well as the levels of TNFα, IL-6 and IL-1β, were not differentially expressed between Mll5 +/− mice and the WT controls (Supplementary Fig. 5C, D). Similarly, we observed no differences in the number of microglia in the subregions of the hippocampus in adult Mll5 +/ − mice (Supplementary Fig. 4E). However, microglia from adult Mll5 +/ − mice exhibited persistent morphological alterations, including increased branching complexity (Fig. 2 C, D) and reduced percentage of CD68/IBA1 volume (Fig. 2 J), indicating lasting effects of Mll5 haploinsufficiency. Collectively, these findings demonstrate that Mll5 haploinsufficiency impairs microglial activity selectively within the hippocampus without inducing a concomitant inflammatory response. Mll5 +/− mice are impaired in synapse elimination The reduced microglial activity in Mll5 +/− mice implicated impaired synaptic development and disrupted synapse elimination. To investigate this, we performed Golgi-Cox staining to assess dendritic spine density in hippocampal neurons. Golgi-Cox staining revealed a significant increase in dendritic spine density specifically in the dentate gyrus (DG) region of Mll5 +/− mice (Fig. 3 A). No significant differences in NEUN + cell number were observed (Fig. 3 B and Supplementary Fig. 6A). Immunofluorescence staining demonstrated protein levels of VGLUT1 and HOMER1 in juvenile Mll5 +/− mice (Fig. 3 C and Supplementary Fig. 6B). In adult Mll5 +/− mice, VGLUT1 and HOMER1 protein levels remained elevated (Fig. 3 D and Supplementary Fig. 6C). We next analyzed synaptic markers via western blot using hippocampal homogenates from juvenile Mll5 +/− mice. Mll5 +/− mice exhibited elevated of the immature synapse marker SAP102 compared with littermate controls (Fig. 3 E). Levels of MAP2, GLUR1, GLUR2, GRIN2B, PSD95 and VGLUT1 were similar across groups (Fig. 3 E and Supplementary Fig. 6D), indicating that altered synaptic sites were independent of changes in neuronal number in Mll5 +/− mice. Furthermore, adult Mll5 +/− mice exhibited increased expression of synaptic proteins, including PSD95, SYNAPIN1, and VGLUT1 (Fig. 3 F), whereas SNAP25 and GEPHYPIN protein levels showed no significant difference change (Supplementary Fig. 6E). These results revealed that Mll5 haploinsufficiency impaired synaptic development and disrupted synapse elimination of the hippocampus. Mll5 is required for microglia phagocytosis in vitro We first examined whether synaptic numbers and dendritic complexity differed between primary Mll5 +/− and WT neurons. Immunofluorescence and Sholl analyses revealed comparable synaptic numbers and dendritic complexity across all groups (Supplementary Fig. 7A, 7B). Consistently, we did not find an obvious difference in spontaneous excitatory postsynaptic currents (sEPSCs) or spontaneous inhibitory postsynaptic currents (sIPSCs) frequency and amplitude between Mll5 +/− and WT neurons (Supplementary Fig. 7C), indicating that Mll5 haploinsufficiency did not directly influence neuronal activity. We therefore hypothesized that the synaptic deficits in Mll5 +/− mice arose specifically from microglial dysfunction. To address this hypothesis, we performed primary culture of microglia from P1-3 Mll5 +/− and WT brains. The purity of the cultured microglia was identified by immunostaining for IBA1 (Supplementary Fig. 8A). Mll5 haploinsufficiency had no effect on microglia viability and migration (Supplementary Fig. 8B, C). Mll5 +/− microglia phagocytosed significantly fewer pHrodo-conjugated synaptosomes, as indicated by decreased proportion of cells exhibited red fluorescence (Fig. 4 A). However, the efficiency of phagocytosing zymosan was similar across groups (Supplementary Fig. 8D), suggesting a mechanism that is target- or receptor-specific. Furthermore, we generated a stable Mll5 knockdown cell line (shRNA- MLL5 ) in the human derived microglial cell line HMC3. Notably, MLL5 knockdown reduced phagocytosis of pHrodo-conjugated synaptosomes compared with scramble controls (Supplementary Fig. 8F). Neuron-microglia co-culture experiments demonstrated that primary neurons co-cultured with Mll5 +/− microglia exhibited increased synaptic density, recapitulating in vivo observations in mouse brains (Fig. 4 B and Supplementary Fig. 8F). To determine whether microglial phagocytosis is impaired in ASD, we generated induced pluripotent stem cells (iPSCs) from five individuals with ASD and five age- and sex-matched healthy controls, followed by microglial differentiation (Fig. 4 C). iPSC and microglial identity were confirmed by immunostaining for the stem cell markers TRA-1-60 and OCT4, and the microglial markers PU.1 and CD11B (Fig. 4 D). Microglial cell numbers did not differ significantly between ASD and control groups (Fig. 4 E). ASD-derived microglia exhibited reduced MLL5 mRNA expression compared with healthy controls (Fig. 4 F). The morphology of microglia is dependent on their activation state, specifically activated or dividing microglia are amoeba-like, while quiescent microglia are more branched-like. Following exposure to pHrodo-labeled synaptosomes, ASD-derived microglia displayed an increased proportion of round and ramified morphologies and a fewer intermediate state (Fig. 4 J, H). ASD-derived microglia exhibited significantly impaired phagocytosis of pHrodo-labeled synaptosomes (Fig. 4 I). Overall, these data suggested that microglial Mll5 is essential for synapse pruning. Mll5 haploinsufficiency impaired TREM2-SGK3-GSK3β signaling pathway To elucidate the mechanisms underlying Mll5 -mediated regulation of microglial phagocytosis, we conducted RNA sequencing (RNA-seq) on FACS-sorted P19 hippocampal microglia to assess genome-wide transcriptional alterations resulting from Mll5 deficiency (Fig. 5 A). We identified 23 differentially expressed genes (DEGs) comparing Mll5 +/− and WT mice, with 9 genes upregulated and 14 genes downregulated in the Mll5 +/− microglia (Fig. 5 B and Supplementary Table 1). Five of these DEGs are annotated in the SFARI database, implicating a potential association between Mll5 and ASD (Supplementary Fig. 9B). To further confirm the decreased phagocytic activity in the Mll5 +/− microglia, we compared the expression of DEGs between Tau-GFP + and Tau-GFP- microglia isolated from tau-GFP mice, in which neurons express a GFP-fusion protein and in which only microglia that have phagocytosed neuronal materials are labeled with GFP, suggesting a more potent phagocytic microglia 26 . Notably, GFP- microglia exhibited lower expression of the DEGs gene signature, suggesting decreased phagocytic activity (Supplementary Fig. 9A). Serum/glucocorticoid regulated kinase 3 ( Sgk3 ) emerged as the most significantly downregulated gene in the Mll5 +/− microglia. SGK3, a member of the AGC kinases family, shares overlapping substrate specificity with the serine-threonine kinase AKT (also known as protein kinase B), which functions as a downstream effector of phosphoinositide 3-kinase (PI3K) 27 . SGK3 regulates glycogen synthase kinase 3β (GSK3β) activity via its serine/threonine kinase function 28 . Triggering receptor expressed on myeloid cells 2 (TREM2) plays a pivotal role in regulating microglial functions by modulating their phagocytosis, energetic metabolism, migration, and proliferation, thereby sculpting their functional profile within the brain 29 , 30 . TREM2 are upstream activators of PI3K and deficiency in TREM2 disrupted phagocytosis and mitochondrial metabolism, at least in part through GSK3β activity 31 , 32 . We found that SGK3 protein and the ratio of P-GSK3β/GSK3β were significantly decreased in Mll5 +/− primary microglia, as well as in shRNA- MLL5 HMC3 cells (Fig. 5 C and Supplementary Fig. 9C). Moreover, genes related to oxidoreductase activity, such as Mical1 , Aldh1l2 , were also found to be significantly changed in our analysis (Fig. 5 B and Supplementary Table 1). We next investigated whether decreased SGK3 signaling in Mll5 +/− was associated with disturbance in TREM2 signaling. We found that TREM2 expression was decreased both in Mll5 +/− primary microglia and shRNA- MLL5 HMC3 cells (Fig. 5 C and Supplementary Fig. 9D), whereas the ratio of P-mTOR/mTOR and P-AKT/AKT remained unchanged (Fig. 5 C and Supplementary Fig. 8C). Therefore, we asked whether anabolic and energetic metabolism were disturbed in Mll5 +/− mice. We analyzed the metabolic states of microglia using metabolic flux assays. Our extracellular acidification rate (ECAR) results revealed that the glycolytic pathway was significantly reduced in Mll5 +/− primary microglia, including glycolysis, and non-glycolytic acidification (Fig. 5 D and Supplementary Fig. 9E). Mll5 +/− primary microglia showed impaired mitochondrial basal respiration, maximal respiration, proton leak, and ATP production in oxygen consumption rate (OCR) measurements (Fig. 5 E and Supplementary Fig. 9F-G). Hexokinase 2 (HK2), the rate-limiting glycolytic enzyme, was also decreased in Mll5 +/− primary microglia (Supplementary Fig. 9H). However, MLL5 did not increase mitochondrial membrane potential, as measured by TMRM fluorescence intensity (Supplementary Fig. 9I). MLL5 was reported to have histone lysine methyltransferase activity that could function as an epigenetic regulator 16 , 33 . We asked whether the decreased expression of Sgk3 and Trem2 in Mll5 +/− mice was associated with changes in their epigenetic modifications. We used chromatin-immunoprecipitation quantitative PCR (ChIP-qPCR) assay to investigate the histone H3 methylation levels at the promoter regions of TREM2 , SGK3 and MICAL1 genes and found that H3K4me3 levels were significantly decreased in shRNA- MLL5 HMC3 cells compared to the-scramble knockdown cells (Fig. 5 F and Supplementary Fig. 9J). Consistently, compared with shRNA-Scramble HMC3 cells, shRNA- MLL5 HMC3 cells showed a decrease in the expression level of Trem2 and Sgk3 (Supplementary Fig. 9D, C). Interestingly, we did not find the overall levels of H3K4me, H3K4me3, H3K9me3, and H3K27me3 showed significant difference between WT and Mll5 +/− hippocampus (Supplementary Fig. 9K). Ultimately, treatment of BV2 cells with an SGK3 degrader resulted in a suppression of phagocytosis in these cells (Fig. 5 G). Concurrently, a decrease in the P-GSK3β/GSK3β ratio was observed (Fig. 5 H). Taken together, our data suggest that MLL5 could promote microglial phagocytosis via the TREM2-SGK3-GSK3β signaling pathway, which associated with changes in anabolic and energetic metabolism. LiCl treatment partially rescued autism-like behaviors and microglial phagocytosis in Mll5 +/− mice Our mechanism-driven screening approach identified GSK3β hyperactivity as a therapeutic target for the Mll5 +/− mice. Based on clinical efficacy profiles and phenotypic relevance, we employed a drug repurposing strategy and identified 35 candidate GSK3β inhibitors (Supplementary Table 4). Among these candidates, Indirubin and Tideglusib are currently clinically approved or under active investigation. SB216763 exhibits high lipophilicity and moderate polarity, properties that enhance blood-brain barrier penetration, a critical advantage for neurological therapeutics 34 . LiCl, a well-characterized GSK3β inhibitor, has shown therapeutic potential in an ASD model 35 . Therefore, we investigated whether these four inhibitors could rescue the disrupted phagocytic function in shRNA- MLL5 HMC3 cells. Notably, all four compounds restored phagocytic activity in shRNA- MLL5 HMC3 cells (Supplementary Fig. 10A-D). At concentrations sufficient to rescue phagocytosis, only LiCl significantly inhibited GSK3β activity (Supplementary Fig. 10E). Furthermore, LiCl has a well-documented safety profile 36 . Based on these findings, we selected LiCl for in vivo therapeutic evaluation. To determine whether pharmacological GSK3β inhibition ameliorated behavioral deficits associated with Mll5 haploinsufficiency, we administered LiCl (45 mg/kg/d, i.p.) or vehicle (Veh) control to Mll5 +/− mice for 4 weeks prior to behavioral assessment (Fig. 6 A). LiCl treatment did not induce significant alterations in body weight (Supplementary Fig. 11A). Notably, LiCl administration ameliorated marble-burying and self-grooming behaviors, and enhanced social novelty preference, while baseline sociability remained unaffected (Fig. 6 B-D). To investigate whether LiCl treatment alters behavioral changes through modulation of microglial function, we assessed its effects on microglial number, morphology, and phagocytic activity. Our findings demonstrated that LiCl treatment significantly increased microglial cell density in the DG region of Mll5 +/− mice (Fig. 6 E and Supplementary Fig. 11B), while reducing branching complexity and junctional density (Fig. 6 E). No changes in the number of microglia were observed in the CA1 and CA3 regions of Mll5 +/− mice upon LiCl treatment (Supplementary Fig. 11C). Furthermore, the volume of the phagocytosis-related molecule CD68 in microglial cells was elevated in Mll5 +/− mice following LiCl treatment (Fig. 6 F). Concomitantly, expression levels of excitatory synapse-related molecules, VGLUT1 and HOMER1, were reduced in Mll5 +/− mice after LiCl administration (Supplementary Fig. 11D, E). To further assess synaptic plasticity, we performed electrophysiological analysis of miniature excitatory and inhibitory postsynaptic currents (mEPSCs and mIPSCs) in DG region. Adult Mll5 +/− mice exhibited elevated mEPSCs frequency, with no alterations in mIPSCs frequency and amplitudes (Fig. 6 J and Supplementary Fig. 11F). LiCl treatment restored mEPSCs frequency in Mll5 +/− mice (Fig. 6 J and Supplementary Fig. 11F). In summary, these results suggest that LiCl ameliorates behavioral abnormalities in Mll5 +/− mice by modulating microglial function. Discussion Our study provides novel insights into the role of MLL5 in postnatal brain development, particularly in regulating microglial activity and the process of synaptic refinement. Consistent with previous findings 15 , our study revealed a critical role for Mll5 in the autism-related behaviors, including USVs issues, sociability deficits, repetitive behaviors and depression-like behaviors. Notably, Mll5 +/− mice exhibited reduced microglial activity during postnatal period, which was accompanied by impairments in synapse maturation and neurotransmission. Furthermore, we demonstrated that the suppressed microglial phagocytosis in Mll5 +/− mice were associated with reduced metabolism via the TREM2-SGK3-GSK3β pathway. Additionally, we found that microglia derived from individuals with ASD exhibited reduced phagocytosis and decreased MLL5 mRNA expression. Pharmacological intervention with LiCl partially rescued ASD-like behavioral phenotypes and restored microglial phagocytic activity. These results contribute to the evidence linking the role of Mll5 gene in the pathophysiology of ASD. Microglia regulate diverse neurodevelopmental processes, including immune surveillance, phagocytosis, synaptic pruning, and the extracellular matrix clearing 37 . Studies demonstrate that microglia dynamically interact with synapses via their motile processes, enabling real-time monitoring and modulation of synaptic activity 38 . This interaction allows microglia to precisely identify synapses requiring either elimination or strengthening. In addition to direct synaptic contact, microglia secrete cytokines and growth factors that regulate synaptic plasticity, thereby optimizing neural circuit refinement 22 , 39 , 40 . Seminal studies demonstrate that microglia execute synaptic pruning via a synapse recognition-phagocytosis-clearance mechanism 41 . Beyond the complement system, phagocytic “eat-me” (e.g., TREM2) and “don’t-eat-me” (e.g., CD47/SIRPα, CD22) signals further regulate microglial phagocytic specificity 38 . Through these mechanisms, microglia fine-tune synaptic density and neural circuit architecture, thereby establishing functional neural connectivity. Previous study have shown that MLL5 suppresses antiviral innate immune responses 18 , and Mll5 −/− mice exhibit heightened susceptibility to bacterial infections 17 . We therefore hypothesized that Mll5 gene is critical for microglial immune surveillance in the CNS. Surprisingly, Mll5 +/− mice showed no evidence of enhanced neuroinflammation. Instead, we found that Mll5 haploinsufficiency induced morphological abnormalities in hippocampal microglia, and reduced lysosomal phagocytic markers, such as CD68 at P17-21, a time window that is active for microglia-dependent synaptic pruning. Notably, these deficits persisted into adulthood, suggesting that developmental disruptions may exacerbate neurological phenotypes. Consistent with Fabia et al’ study that observed TREM2 mediated synaptic refinement by microglia during the early stages of brain development 42 . TREM2 modulates microglial phagocytosis, morphology, and motility, thereby shaping neurodevelopment and disease progression 43 . We observed decreased Trem2 expression levels in Mll5 +/− microglia and this reduction was epigenetically modified by Mll5 ’s histone methyltransferases activity. This finding establishes an epigenetic link between Mll5 and TREM2, implicating Mll5 in microglia-dependent synaptic pruning and neurodevelopment, thereby advancing our understanding of its CNS roles. Synaptic dysfunction is implicated in numerous neurological and psychiatric disorders, including ASD 44 . The hippocampus is involved in cognitive functions such as social interaction, spatial cognition, and memory, and plays a role in the pathogenesis of ASD 45 . Our data demonstrate that MLL5 haploinsufficiency elevates dendritic spine density in the hippocampal DG, accompanied by increased frequency of mEPSCs. These findings are consistent with recent studies that report synaptic hyperconnectivity associated with ASD. For instance, Fmr1 knockout mice exhibit increased mEPSC frequency and dendritic spine density in CA1 and DG hippocampal neurons 46 – 48 . However, it seems that the spine density and morphology varied across different animal models with ASD. In the valproic acid (VPA)-induced ASD mouse or rat model, an increase in dendritic spines 49 , 50 .Notably, postmortem studies of ASD individuals have similarly reported increased dendritic spine density in frontal and temporal cortices 51 . Conversely, decreased dendritic spine and reduced excitatory synaptic transmission were reported in the anterior cingulate cortex in Shank3 knockout mice 52 . Knockdown of Cntnap2 in layer 2/3 pyramidal neurons of the PFC reduced excitatory and inhibitory synaptic transmission, presumably by decreasing the number of functional synapses 53 . Furthermore, disruption of signaling between microglia and neurons leads to an excess of immature synaptic connections, which is thought to be the result of impaired phagocytosis of synapses by microglia 54 . These observations together with our findings in the Mll5 +/− mice, suggest that dysfunctional synapse in ASD may be dependent on a region and genetic specific manner. Epigenetic regulation is crucial for numerous fundamental cellular functions, and a growing body of research indicates that epigenetics significantly influences ASD 55 . MLL5 is an epigenetic regulator that is involved in histone modification, particularly by affecting H3K4 methylation 33 . Our transcriptome analysis in Mll5 +/− microglia showed that among the differentially expressed genes, several of which were also implicated in SFARI database, suggesting a role of Mll5 in regulating the expression of autism-related genes. Consistently, we found that H3K4me3 levels were significantly decreased at the promoter regions of Sgk3, Trem2 and Mical1 . However, different from previous studies 16 , we failed to find an overall decreased H3K4me3 level in the hippocampus. We next focused on Sgk3 , which was the most significantly differentially genes between Mll5 +/− and controls. Sgk3 shares similar substrate specificity with AKT and can act as a downstream mediator of PI3K 27 . Microglia was reported to phagocytose amyloid beta in part through TREM2-AKT mediated signaling 29 , and this process was compromised if GSK3β signaling was impaired 29 , 56 . We thus hypothesized that Mll5 may modulate microglial phagocytosis through the TREM2-SGK3-GSK3β signaling pathway. Accordingly, we found that TREM2, SGK3 and the ratio of P-GSK3β/GSK3β were significantly decreased in Mll5 +/− microglia. Yet, the ratio of P-mTOR/mTOR and P-AKT/AKT was not affected by Mll5 deficiency. Oxidative phosphorylation and glycolysis are the main metabolic pathways of energy production in cells 57 . Previous studies have shown that glycolysis and energetic metabolism may be involved in effective microglial phagocytosis 58 , 59 . We found that Mll5 +/− primary microglia exhibited reduced glycolysis and ATP production. Moreover, we observed that levels of HK2, a key rate-limiting enzyme in glycolysis specifically expressed in microglia 60 , were decreased when MLL5 was knockdown, suggesting a role of Mll5 in regulating cellular metabolism at least in part through the TREM2-SGK3-GSK3β signaling pathway. Li, as a mood stabilizer, is widely used in the treatment of bipolar disorder and has demonstrated efficacy as a GSK3β inhibitor in ameliorating psychiatric disorders 61 . Recent studies have explored its potential for ASD therapy 62 . In our study, microglia with Mll5 haploinsufficiency exhibited significantly elevated GSK3β activity, prompting us to evaluate LiCl’s potential to ameliorate ASD-like behaviors in mice. LiCl administration partially improved core behavioral deficits, including increased marble-burying behavior, reduced self-grooming duration, and enhanced social novelty preference in Mll5 +/− mice. These behavioral improvements align with prior reports demonstrating Li’s efficacy in rescuing ASD-like behaviors in Shank3 - and Fmr1 - deficient mouse models 62 , 63 , underscoring its broad therapeutic relevance across genetically distinct ASD subtypes. In a VPA-induced rat model of ASD, Li ameliorated social cognitive impairments, enhanced social memory, and reduced anxiety levels. Furthermore, Li treatment reduced pro-inflammatory markers and increased neuroprotective histone modifications in the hippocampus, suggesting modulation of neuroinflammatory pathways implicated in ASD 64 . Li regulates synaptic plasticity, particularly through GSK3β inhibition, and modulates microglial function. It reduces AMPA receptor-mediated mEPSCs amplitude while altering postsynaptic homeostatic plasticity, suggesting regulation through AMPA receptor activity 65 . Li increases glial cell numbers in the DG 66 , suppresses microglial activation 67 , and elevates C3 expression in microglia 68 , suggesting enhanced phagocytic function. Our findings indicate that the frequency of mEPSCs in Mll5 +/− mice decreased following treatment with LiCl. Additionally, we observed morphological changes in microglia, along with an enhanced phagocytic capacity of these cells after LiCl treatment. These alterations may contribute to the observed improvement in autism-like behaviors associated with LiCl administration. These alterations may underlie the behavioral improvements associated with lithium, indicating its therapeutic potential for ASD via GSK3β and microglial modulation. In conclusion, our study expands our understanding of Mll5 ’s role in the postnatal neurodevelopment and highlights its potential as a therapeutic target for ASD. Methods Ethical approval The study received ethical approval from the Ethical Review Board of Peking University Health Science Center and Peking University Sixth Hospital. Mice The generation and genotyping of Mll5 +/- mice were previously described 18 . The genotyping of WT and Mll5 +/- micewere performed by PCR analysis using genomic DNA isolated from the tail tips using the primers listed in Table S3. All animals were kept on a 12 h:12 h reversed light/dark cycle, and all experiments were performed in the animals’ dark phase. No more than 5 animals per cage. The mice had ad libitum access to food and water.Behavioral experiments were performed using 8-12-week-old mice. Behavioral test was done blind to genotype of each mouse with age-matched littermates of mice. Patient selection and clinical assessment Five male individuals diagnosed with ASD according to the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) criteria and five age- and gender-matched healthy controls were recruited from the Child and Adolescent Psychiatric Outpatient Department of Peking University Sixth Hospital. The healthy controls were screened based on the absence of history of mental health and substance use disorders. Those with a history of psychological and psychiatric disorders were excluded (Table S2). The research aim, procedure, benefit and risk were explained to the parents of the participating subjects, and the written informed consent were signed by them. LiCl administration Adult Mll5 +/- mice received daily intraperitoneal (i.p.) injections of LiCl (45 mg/kg/day; Thermo Fisher Scientific, MA, USA) or vehicle for 28 days. Western blot analysis Tissues and cells were homogenized in cell lysis buffer (Meilunbio, Dalian, China), which was supplemented with EDTA-free protease inhibitor cocktail (Thermo Fisher Scientific) and protein phosphatase inhibitor (Solarbio, Beijing, China). The sample was centrifuged for 10 min at 13,000 g at 4°C followed by protein concentration quantification using the Pierce BCA Protein Assay Kit (Thermo Fisher). Samples were run on a NuPAGE Bis-Tris protein gel (Thermo Fisher) and transferred to PVDF membrane (Merck KGaA, Darmstadt, Germany), followed by blocking with fast block buffer (Epizyme, Shanghai, China) for 20 min at room temperature (RT). Primary antibody was incubated overnight at 4°C followed by secondary antibody incubation for one hour at RT. The blot was developed with Immobilon Western Chemiluminescent HRP Substrate (Merck). Quantitative real-time PCR (qPCR) Total RNA was isolated using TRIzol reagent (Thermo Fisher) and then subjected to reverse transcription with a HiScript III 1st Strand cDNA Synthesis Kit (Vazyme, Nanjing, China). The resulting cDNA was used for qPCR analysis with AceQ Universal SYBR qPCR Master Mix (Vazyme), and specific primers in a QuantStudio 5 Real-Time PCR System (Thermo Fisher). qPCR primer sequences were presented in supplementary table (Table S3). Gene expression levels were calculated using the 2 −ΔΔCt analysis method, and the samples were normalized to GAPDH. Immunofluorescence staining Under deep anesthesia, mice were transcardially perfused with 4% paraformaldehyde followed by brain dissection. After postfixation, brains were dehydrated in 30% sucrose, and then embedded in O.C.T. for cryopreservation. The samples were cut coronally at 50-μm sections on cryostat (Leica, Wetzlar, Germany). Sections were washed in 1× PBS and incubated with 3% BSA and 0.3% Triton X-100 in PBS for 1 h at RT to minimize nonspecific binding. Then, the sections were incubated overnight at 4 °C in primary antibody solution followed by washing and incubating with a species-specific secondary antibody for 1 h at RT. The sections were further washed with 1× PBS and incubated with DAPI at RT for 10 min, and washed twice with PBS before mounting. The samples were observed under a laser confocal scanning microscope (Leica). Enzyme-linked immunosorbent assay (ELISA) The concentrations of TNFα, IL-6, and IL-1β in the mouse hippocampus were measured using the specific ELISA kits (ABclonal, Wuhan, China) and the assays were conducted in strict accordance with the protocols supplied. The levels of TNFα, IL-6, and IL-1β in each sample were calculated according to the standard curve generated by serial dilution of the recombinant mouse TNFα, IL-6, and IL-1β. Slice electrophysiology Mice were deeply anesthetized with isoflurane inhalation and decapitated. Transverse hippocampal slices (250 µm thick) were prepared with a vibratome (Leica VT1200S) in an ice-cold cutting solution containing the following (in mM): 80 NaCl, 26 NaHCO 3 , 3.0 KCl, 1.0 NaH 2 PO 4 , 1.3 MgCl 2 , 1.0 CaCl 2 , 20 D-glucose, and 75 sucrose, saturated with 95% O2 and 5% CO 2 from P16-P21 male littermates. Slices were recovered for 30 min at 33 °C, and then at RT until recordings in artificial cerebrospinal fluid (ACSF) containing (in mM): 124 NaCl, 26 NaHCO 3 , 3.0 KCl, 1.0 NaH 2 PO 4 , 1.3 MgCl 2 , 2 CaCl 2 , 20 D-glucose. Slices were perfused with oxygenated ACSF at 33°C at 2.0 mL/min. Signals were amplified by a Multiclamp 700B amplifier (Axon Instruments/Molecular Devices, San Jose, CA, USA) and digitized using Digidata 1550A (Molecular Devices). Recordings were made from hippocampal DG pyramidal neurons. For miniature excitatory post-synaptic currents (mEPSCs) recordings, 1 mM TTX were added to the ACSF. Pipettes contained (in mM): 135 CsGluconate, 1 EGTA, 10 HEPES, 2 MgCl 2 , 4 MgATP, and 0.3 Tris-GTP, (pH 7.4). Primary cortical neuron culture Primary neurons were isolated from E17 mice. Briefly, cortices were isolated and dissociated with 0.25% trypsin-EDTA. The cells were plated onto poly-D-lysine-coated coverslips and first cultured in DMEM medium containing 10% fetal bovine serum (FBS), 1% GlutaMax and 1% penicillin/streptomycin for 4 h. Then he medium was switched to Neurobasal plus medium, which contained 1% GlutaMax, 2% B27 plus supplement and 1% penicillin/streptomycin. All experiments were performed with cultures that were 13-15 days in vitro (DIV). Primary microglia culture The cerebral cortices of P1-3 mice were isolated and treated with 0.25% trypsin-EDTA to digest the tissues. The single cells were seeded in T12.5 cm 2 flasks coated with poly-L-lysine and cultured in DMEM/F12 medium containing 10% FBS and 1% penicillin/streptomycin. Medium change was conducted one day after plating and then after 3 days. At 12 days, flasks were mechanically shaking to collect microglia. Microglia were plated on 96-well plates in DMEM/F12 containing 10% FBS and 1% penicillin/streptomycin to continue the downstream experiments. IN microglia-neuron co-culture experiments, WT or Mll5 +/- microglia were added to neurons (13-14 DIV) at microglia to neuron ratio of 1.5:1 for 24 h. Cell culture recording Whole cell voltage clamp recording of cultured neurons from WT and Mll5 +/- mice were performed at 13-15 days in vitro . During recordings cell were bathed in a standard external solution containing (in mM): 125 NaCl, 5 KCl, 1.2 MgSO 4 , 1.2 KH 2 PO 4 , 2 CaCl 2 , 6 glucose and 25 HEPES-NaOH, pH 7.4. Recording pipettes (resistances of 3-5 MΩ) were filled with a standard intracellular solution containing (in mM):135 CsGluconate, 1 EGTA, 10 HEPES, 2 MgCl2, 4 MgATP, and 0.3 Tris-GTP, (pH 7.4). Golgi-Cox staining Golgi-Cox staining was performed using a FD Rapid Golgistain Kit (FD NeuroTechnologies, Columbia, MD, USA). Briefly, freshly dissected brains were immersed in solution A and B for 2 weeks at RT and then transferred into solution C for 72 h at RT in the dark. 150 µm slices were prepared and the sections placed on a gelatin-coated glass slide and dried naturally at RT for 2 days in the dark. Slices were placed in a mixture of solution D: E: distilled water at 1:1:2 ratio for 10 min followed by serial dehydration in 50%, 75%, 95% and 100% ethanol. Clear in xylene, three times for 4 minutes each time, and the cover slides. Images of dendritic spines were acquired using a 100× objective. Spines that started from 50 µm distance of the apical dendrite were counted within a 20 µm segment. The density of the spines was analyzed from an individual blind to genotype of neurons. For frequency distributions, spines were separated into thin, mushroom and stubby spines based on their shape. Nissl staining Brains were perfused as described above, and consecutive 50 μm-thick coronal sections were mounted on slides, rehydrated, stained with tar purple dye for 30 min (Solarbio, Beijing, China). After being rinsed with distilled water and 95% alcohol, the sections were treated with conventional dehydration, transparency and tablet sealing. Images were observed and photographed using an optical microscope. Induced pluripotent stem cells (iPSCs) reprogramming iPSCs were generated from five male individuals diagnosed with ASD (ASD 1, ASD 2, ASD 3, ASD 4, ASD 5) according to the DSM-5 and five male control participants (Control 1, Control 2, Control 3, Control 4, Control 5) according to the published protocol 69 .In brief,Epstein-Barr virus immortalized B-lymphocytes were reprogrammed using the Yamanaka Episomal vector, consisting of three sets of episomal plasmids expressing reprograming factors: pCXLE-hOCT3/4-shp53, pCXLE-hSK (SOX2, KLF4) and pCXLE-hUL (L-MYC, LIN28). Two to three iPSCs colonies were collected from each individual and Standard G-banding karyotype and immunostaining of OCT4 and TRA-1-60 marker were performed to check iPSC quality. Differentiation to human iPSCs-derived microglia Differentiation to human iPSC-derived microglia was as previously described 70 .In brief,generation of CD43+ primitive hematopoietic progenitor cells (HPCs) from iPSCs were conducted according to STEMdiff™ Hematopoietic Kit (Stem Cell Technologies), then harvested HPCs were induced to microglia. On day 0, HPCs were seeded onto Matrigel-coated plates, culturing in microglia medium (DMEM/F12, 2× insulin-transferrin-selenite, 2× B27, 0.5× N2, 1× glutamax, 1× non-essential amino acids, 400 μM monothioglycerol, 5 μg/mL insulin), supplemented with 100 ng/mL IL-34 (PeproTech), 50 ng/mL TGFβ1 (Novoprotein), and 25 ng/mL M-CSF (Novoprotein). The medium was changed every other day. On day 25, microglial medium was supplemented with additional 100 ng/mL CD200 (Novoprotein) and 100 ng/mL CX3CL1 (Novoprotein). On day 28, cells were collected for functional assays. Synaptosomes purification and engulfment assay in vitro Synaptosomes of cortex were extracted using Syn-PER™ Synaptic Protein Extraction Reagent (Thermo Fisher). Cortex was homogenized with Dounce grinder with 10 slow strokes followed by centrifugation at 1200g for 10 min at 4 °C. After discarding the pellet, the supernatant was centrifuged at 15,000g for 20 min at 4 °C. The supernatant part was the cytosolic fraction and the pellet containing the synaptosomes was dissolved with the extraction reagent. Protein quantification was performed using Pierce BCA Protein Assay Kit (Thermo Fisher). Resulting synaptosomes were collected and labeled with the pH-sensitive fluorescent dye pHrodo (Thermo Fisher) as per manufacturer’s instructions. Cells were seed into 96-well plate overnight, and pHrodo labeled synaptosomes were added to the cell cultures. Phagocytosis was determined as the red object counts per image over time. Flow Cytometry Mice aged P17-P21 were sacrificed via decapitation and the hippocampi were dissected in sterile, ice-cold Hank’s balanced salt solution (HBSS). Single cell suspensions were prepared using the brain dissociation kit (Miltenyi Biotec, Bergisch Gladbach, Germany) and the single cell suspension dissociator (Shenzhen Ruiwode Life Technology Co., Ltd., Shenzhen, China). Brain cell suspensions were then resuspended in 30% Percoll followed by centrifugation at 900g for 20 min without brake. The interface was removed and the deposition was washed with HBSS. Red blood cells were lysed using blood cell lysis buffer. Cell suspensions were then spun down at 300g for 10 min and pellets were resuspended in flow cytometry buffer, labeled with a combination of the following conjugated antibodies: CD11b-PE, CD45-PeCy7, MHC II-APC, and CD40-FITC. FACS analysis of the microglial profile was performed by gating CD11b + CD45 low cells on a FACS Aria SORP (BD Biosciences). Appropriate antibody IgG isotype controls were used for all staining. Data were analyzed with FlowJo Software (TreeStar). Seahorse XFe96 Analyzer The oxygen consumption rate (OCR) and the extracellular acidification rate (ECAR) were measured using a Seahorse XFe96 Analyzer (Agilent, California, CA, USA) with Seahorse XF Cell Mito Stress Test Kit (Agilent) and Seahorse XF Glycolysis Stress Test Kit (Agilent), respectively, following the manufacturer’s instructions. Briefly, microglia were seeded in 96-well Seahorse assay plates at a concentration of 2 × 10 4 cells/well, and cultured overnight for attachment. At indicated time points, microglia were washed and analyzed in XF Running buffer A (XF assay medium, 10 mM glucose, 1 mM pyruvate sodium, 2 mM l-Glutamine) for OCR measurements or XF Running buffer B (XF assay medium, 2 mM l-Glutamine) for ECAR measurements. Measurements were obtained in real-time under basal conditions (no drug treatment) and with the sequential treatment with different drugs: 1 μM oligomycin, 1 μM FCCP, 1 μM rotenone plus 5 μM antimycin A for the Mito-Stress assay, or 10 mM glucose, 1 μM oligomycin and 50 mM 2-DG for the Glycolysis stress assay. Ultrasonic vocalizations of isolated mice pups On the testing day, dams and pups were acclimated to the experimental room for ≥30 minutes in a temperature-controlled environment (25°C). Prior to testing, individual pups were randomly separated from dams and littermates and placed into a sound-attenuated chamber for a 1-minute acclimation period. Vocalizations were recorded for 5 minutes using Avisoft Recorder USGH software (Avisoft Bioacoustics) paired with an ultrasonic microphone. Raw audio files were imported into Avisoft SASLab Pro (Avisoft Bioacoustics) and filtered to eliminate amplitudes under 30 kHz before further processing. All the call traces were extracted blindly using an automatic detection function with a proper threshold and manually checked. Call numbers, total duration, and mean duration were analyzed from the labeled call traces. Three-chamber social test The three-chamber apparatus (40 cm × 60 cm × 23 cm) was constructed with three chambers (40 cm × 20 cm × 23 cm). Each side compartments contained an empty wire cage. The test is divided into two phases. (A) Habituation: the mice were allowed to explore the center chamber for 5 min. (B) Sociability and social novelty test: Following habituation, Stranger Mouse 1 (age- and gender-matched) was placed in one wire mesh cage, while the opposite cage remained empty. The empty cage functioned as a non-social control object. The subject mouse was placed in the central chamber and given 10 minutes to explore the entire apparatus. Subsequently, Stranger Mouse 2 (novel mouse) was introduced into the previously empty cage, and the subject mouse was allowed to explore the apparatus again for 10 min. Total time spent in the chamber and in close interaction with the cages was recorded to evaluate the social function and social memory of the mice. Marble burying Briefly, the clean mouse cage was prepared by placing 12 glass marbles (14 mm diameter) evenly on 5 cm deep bedding material. The experimental animal was left undisturbed for 30 min in the test cage in an isolated place. A marble was considered buried when 2/3 of the marble was covered in bedding material. Self-grooming test This test was conducted in a new mouse cage with a thin layer bedding reducing neophobia. The cumulative time that the test mouse spent grooming itself was measured for 30 min following 10 min period of acclimatization to the test cage. Novel object recognition test Both object recognition tests were performed in a square white plastic box (40 cm × 27 cm × 18cm). Mice were allowed to explore two identical objects (A) in the testing arena for 3 min (Trial 1-Training Trial). Subsequently, the mice were habituated to the arena for 15 min. The last stage, the mice were returned to the arena, with one of the familiar objects (A) now replaced by a novel object (B) for 3 min (Trial 2-Testing trial). The performance was evaluated by novel object preference index= B exploration time/ (A recognition time + B exploration time). Tail suspension test The tail-suspension test was performed using a specially manufactured tail-suspension box (55 × 60 cm) divided into four 15 cm compartments. In each compartment, a small plastic bar hangs from the base. The tail of each mouse was taped to this bar and animals were suspended for 6 min. The immobility time was determined. Forced swimming test Mice were gently placed into 5 L of beaker filled with 3.5 L 25°C water for 6 min. The immobility time of each mouse spent during the test was counted online by two independent observers blind to the animal treatments. An increase in immobile time was an indication of depression. Nest building test Nest building test is an innate behavior that assesses sensorimotor function in mice. The mice were transferred to a new home cage with clean bedding and new nest-building material, a 6 × 6 cm square of compressed cotton. After 36 h, the nest was scored on a 1-5 rating scale, with 1 meaning the nestlet more than 90% intact; 2, the nestlet was partially torn but 50-90% intact; 3, the nestlet was torn with less than 50% remaining intact, but no obvious nest with shreds spread around the cage; 4, a flat nest without raised walls, more than 90% of the nestlet was shredded; 5, the nest was built perfectly with walls higher than mouse height and no unshredded pieces. Morris water maze test The Morris water maze test was conducted to assess spatial learning and memory. Mice were trained to locate a submerged escape platform positioned in the northeastern quadrant of a circular pool (diameter: 110 cm), which was filled with opacified water maintained at 25°C. During the acquisition phase, mice performed 4 trials per day from random starting position for 5 consecutive days with a maximum trial duration of 60 seconds. Spatial memory was assessed by removing the platform and conducting 60-second probe trials. The number of entries into the quadrants and other relevant parameters were analyzed using Morris maze video tracking software (Zhongshi Di Chuang, Beijing, China). RNA-seq. Total RNA was extracted from the sorted microglia of P17-21 mice hippocampus using a TRIzol reagent (Thermo Fisher). mRNA was reverse transcribed and amplified using the SMART-seq method. Complementary DNA libraries were constructed using the TruePrep DNA Library Prep Kit V2 for Illumina (Vazyme), and sequenced with a HiSeq 2500 system (Illumina). Differentially expressed genes were identified using the edgeR package (3.18.1) in R (3.4.1). Differentially expressed genes were required to have an adjusted p -value < 0.1. Chromatin immunoprecipitation (ChIP) qPCR HMC3 cells were fixed with 1% formaldehyde for 10 min at RT to cross-link proteins to DNA. This reaction was subsequently stopped by adding 0.125 M glycine for 5 min. Chromatin is extracted and sheared by sonication. Appropriate antibodies were mixed and incubated overnight at 4 °C with gentle shaking. Agarose protein A/G beads (Santa Cruz Biotechnology) were used to precipitate the antibody-protein-DNA complexes. Following washing of the beads, the crosslinking was reversed. The DNA captured by the immunoprecipitation was subjected to qPCR analysis. In parallel, an input control sample, which consisted of sheared chromatin collected prior to the addition of antibodies and equivalent to 10% of the immunoprecipitation sample volume, was processed to standardize the pull-down efficiencies. Statistical analysis GraphPad Prism 8.0.1 was used for statistical analyses. Statistical significance was estimated with Student's t test when variances were equal or with Mann-Whitney U test when variances were not equal. Differences between multiple groups were assessed with one-way or two-way ANOVA. Declarations Data availability Gene expression profiling data have been deposited at GEO database (GEO: GSE280750) and are publicly available as of the date of publication. Source data are provided with this paper. Acknowledgements We thank Professor Lih Wen Deng for sharing MLL5 antibody with us, and we thank all the participating subjects. This work was funded by National Natural Science Foundation of China (82130040, 82288101, 82171530, 81801344), the Youth Program of National Natural Science Foundation of China (82009Y3510), Nursing Science Research Fund of Peking University (TYZH2023002), and The Fundamental Research Funds for the Central Universities (71006Y2557). Author contributions Y.B.W. conceived and designed the experiments. S.G., Q.L., X.L., Z.F., L.H., and N.Q. performed the experiments and analyzed the data. Y.B.W. was responsible for validation. M.J., A.Z, X.X.L., H.Z., X.Z., Z.J., Y.Z., L.L., S.J., J.J.L. and Y.B.W. provided resources. S.G. wrote the manuscript. Y.B.W., J.J.L, J.S. and L.L. revised the manuscript. Y.B.W. and J.J.L supervised the project. Competing interests The authors declare no competing interests. Materials & Correspondence Correspondence and requests for materials should be addressed to Shi Jie, Jia Jia Liu or Ya Bin Wei. References Lord C et al (2006) Autism from 2 to 9 years of age. Arch Gen Psychiatry 63:694–701. 10.1001/archpsyc.63.6.694 Lord C et al (2020) Autism spectrum disorder. Nat Rev Dis Primers 6:5. 10.1038/s41572-019-0138-4 Bailey A et al (1995) Autism as a strongly genetic disorder: evidence from a British twin study. Psychol Med 25:63–77. 10.1017/s0033291700028099 Sandin S et al (2014) The familial risk of autism. JAMA 311:1770–1777. 10.1001/jama.2014.4144 Satterstrom FK et al (2020) Large-Scale Exome Sequencing Study Implicates Both Developmental and Functional Changes in the Neurobiology of Autism. Cell 180, 568–584 e523. 10.1016/j.cell.2019.12.036 Zhou X et al (2022) Integrating de novo and inherited variants in 42,607 autism cases identifies mutations in new moderate-risk genes. Nat Genet 54:1305–1319. 10.1038/s41588-022-01148-2 Grove J et al (2019) Identification of common genetic risk variants for autism spectrum disorder. Nat Genet 51:431–444. 10.1038/s41588-019-0344-8 Dong S et al (2014) De novo insertions and deletions of predominantly paternal origin are associated with autism spectrum disorder. Cell Rep 9:16–23. 10.1016/j.celrep.2014.08.068 Sharawat IK, Panda PK, Dawman L (2021) Clinical Characteristics and Genotype-Phenotype Correlation in Children with KMT2E Gene-Related Neurodevelopmental Disorders: Report of Two New Cases and Review of Published Literature. Neuropediatrics 52:98–104. 10.1055/s-0040-1715629 Abreu NJ et al (2022) Novel truncating variant in KMT2E associated with cerebellar hypoplasia and velopharyngeal dysfunction. Clin Case Rep 10:e05277. 10.1002/ccr3.5277 O'Donnell-Luria AH et al (2019) Heterozygous Variants in KMT2E Cause a Spectrum of Neurodevelopmental Disorders and Epilepsy. Am J Hum Genet 104:1210–1222. 10.1016/j.ajhg.2019.03.021 Ali M et al (2013) Molecular basis for chromatin binding and regulation of MLL5. Proc Natl Acad Sci U S A 110:11296–11301. 10.1073/pnas.1310156110 Lemak A et al (2013) Solution NMR structure and histone binding of the PHD domain of human MLL5. PLoS ONE 8:e77020. 10.1371/journal.pone.0077020 Tasdogan A et al (2016) DNA Damage-Induced HSPC Malfunction Depends on ROS Accumulation Downstream of IFN-1 Signaling and Bid Mobilization. Cell Stem Cell 19:752–767. 10.1016/j.stem.2016.08.007 Li YJ et al (2023) KMT2E Haploinsufficiency Manifests Autism-Like Behaviors and Amygdala Neuronal Development Dysfunction in Mice. Mol Neurobiol 60:1609–1625. 10.1007/s12035-022-03167-w Zhang X et al (2022) MLL5 is involved in retinal photoreceptor maturation through facilitating CRX-mediated photoreceptor gene transactivation. iScience 25:104058. 10.1016/j.isci.2022.104058 Heuser M et al (2009) Loss of MLL5 results in pleiotropic hematopoietic defects, reduced neutrophil immune function, and extreme sensitivity to DNA demethylation. Blood 113:1432–1443. 10.1182/blood-2008-06-162263 Zhou P et al (2018) MLL5 suppresses antiviral innate immune response by facilitating STUB1-mediated RIG-I degradation. Nat Commun 9:1243. 10.1038/s41467-018-03563-8 Swinnen N et al (2013) Complex invasion pattern of the cerebral cortex bymicroglial cells during development of the mouse embryo. Glia 61:150–163. 10.1002/glia.22421 Ginhoux F et al (2010) Fate mapping analysis reveals that adult microglia derive from primitive macrophages. Science 330:841–845. 10.1126/science.1194637 Paolicelli RC et al (2011) Synaptic pruning by microglia is necessary for normal brain development. Science 333:1456–1458. 10.1126/science.1202529 Schafer DP et al (2012) Microglia sculpt postnatal neural circuits in an activity and complement-dependent manner. Neuron 74:691–705. 10.1016/j.neuron.2012.03.026 Nguyen PT et al (2020) Microglial Remodeling of the Extracellular Matrix Promotes Synapse Plasticity. Cell 182, 388–403 e315. 10.1016/j.cell.2020.05.050 Lukens JR, Eyo UB (2022) Microglia and Neurodevelopmental Disorders. Annu Rev Neurosci 45:425–445. 10.1146/annurev-neuro-110920-023056 Wamsley B et al (2024) Molecular cascades and cell type-specific signatures in ASD revealed by single-cell genomics. Science 384:eadh2602. 10.1126/science.adh2602 He D et al (2022) Disruption of the IL-33-ST2-AKT signaling axis impairs neurodevelopment by inhibiting microglial metabolic adaptation and phagocytic function. Immunity 55:159–173e159. 10.1016/j.immuni.2021.12.001 Scortegagna M et al (2015) PDK1 and SGK3 Contribute to the Growth of BRAF-Mutant Melanomas and Are Potential Therapeutic Targets. Cancer Res 75:1399–1412. 10.1158/0008-5472.Can-14-2785 Liu F, Wu X, Jiang X, Qian Y, Gao J (2018) Prolonged inhibition of class I PI3K promotes liver cancer stem cell expansion by augmenting SGK3/GSK-3β/β-catenin signalling. J Exp Clin Cancer Res 37:122. 10.1186/s13046-018-0801-8 Wang S et al (2022) TREM2 drives microglia response to amyloid-β via SYK-dependent and -independent pathways. Cell 185:4153–4169e4119. 10.1016/j.cell.2022.09.033 Liu Z et al (2020) TMEM59 interacts with TREM2 and modulates TREM2-dependent microglial activities. Cell Death Dis 11:678. 10.1038/s41419-020-02874-3 Keren-Shaul H et al (2017) A Unique Microglia Type Associated with Restricting Development of Alzheimer's Disease. Cell 169:1276–1290e1217. 10.1016/j.cell.2017.05.018 Wang Y et al (2015) TREM2 lipid sensing sustains the microglial response in an Alzheimer's disease model. Cell 160:1061–1071. 10.1016/j.cell.2015.01.049 Zhang X, Novera W, Zhang Y, Deng LW (2017) MLL5 (KMT2E): structure, function, and clinical relevance. Cell Mol Life Sci 74:2333–2344. 10.1007/s00018-017-2470-8 Xiao H, Deng M, Yang B, Tang J, Hu Z (2017) Role of glycogen synthase kinase 3 in ischemia-induced blood-brain barrier disruption in aged female rats. J Neurochem 142:194–203. 10.1111/jnc.14051 Hu W et al (2023) Lithium Cholesterol Sulfate: A Novel and Potential Drug for Treating Alzheimer's Disease and Autism Spectrum Disorder. CNS Neurol Disord Drug Targets 22:1250–1258. 10.2174/1871527321666220825114236 Szklarska D, Rzymski P (2019) Is Lithium a Micronutrient? From Biological Activity and Epidemiological Observation to Food Fortification. Biol Trace Elem Res 189:18–27. 10.1007/s12011-018-1455-2 Sierra A, Paolicelli RC, Kettenmann H (2019) Cien Años de Microglía: Milestones in a Century of Microglial Research. Trends Neurosci 42:778–792. 10.1016/j.tins.2019.09.004 Pereira-Iglesias M et al (2025) Microglia as hunters or gatherers of brain synapses. Nat Neurosci 28:15–23. 10.1038/s41593-024-01818-w Bollinger JL, Burns B, C. M., Wellman CL (2016) Differential effects of stress on microglial cell activation in male and female medial prefrontal cortex. Brain Behav Immun 52:88–97. 10.1016/j.bbi.2015.10.003 Chen Y et al (2019) miR-124/VAMP3 is a novel therapeutic target for mitigation of surgical trauma-induced microglial activation. Signal Transduct Target Ther 4. 10.1038/s41392-019-0061-x Ding X et al (2030) Loss of microglial SIRPα promotes synaptic pruning in preclinical models of neurodegeneration. Nat Commun 12. 10.1038/s41467-021-22301-1 (2021) Filipello F et al (2018) The Microglial Innate Immune Receptor TREM2 Is Required for Synapse Elimination and Normal Brain Connectivity. Immunity 48:979–991e978. 10.1016/j.immuni.2018.04.016 Suárez-Calvet M et al (2016) Early changes in CSF sTREM2 in dominantly inherited Alzheimer's disease occur after amyloid deposition and neuronal injury. Sci Transl Med 8:369ra178. 10.1126/scitranslmed.aag1767 Sun XD et al (2016) Lrp4 in astrocytes modulates glutamatergic transmission. Nat Neurosci 19:1010–1018. 10.1038/nn.4326 Banker SM, Gu X, Schiller D, Foss-Feig JH (2021) Hippocampal contributions to social and cognitive deficits in autism spectrum disorder. Trends Neurosci 44:793–807. 10.1016/j.tins.2021.08.005 Jawaid S et al (2018) Alterations in CA1 hippocampal synapses in a mouse model of fragile X syndrome. Glia 66:789–800. 10.1002/glia.23284 Grossman AW et al (2010) Developmental characteristics of dendritic spines in the dentate gyrus of Fmr1 knockout mice. Brain Res 1355:221–227. 10.1016/j.brainres.2010.07.090 Zhang L et al (2024) Anterior piriform cortex dysfunction underlies autism spectrum disorders-related olfactory deficits in Fmr1 conditional deletion mice. Neuropsychopharmacology. 10.1038/s41386-024-02027-6 Cheng Y et al (2022) Degraded cortical temporal processing in the valproic acid-induced rat model of autism. Neuropharmacology 209:109000. 10.1016/j.neuropharm.2022.109000 Yang EJ, Ahn S, Lee K, Mahmood U, Kim HS (2016) Early Behavioral Abnormalities and Perinatal Alterations of PTEN/AKT Pathway in Valproic Acid Autism Model Mice. PLoS ONE 11:e0153298. 10.1371/journal.pone.0153298 Hutsler JJ, Zhang H (2010) Increased dendritic spine densities on cortical projection neurons in autism spectrum disorders. Brain Res 1309:83–94. 10.1016/j.brainres.2009.09.120 Guo B et al (2019) Anterior cingulate cortex dysfunction underlies social deficits in Shank3 mutant mice. Nat Neurosci 22:1223–1234. 10.1038/s41593-019-0445-9 Sacai H et al (2020) Autism spectrum disorder-like behavior caused by reduced excitatory synaptic transmission in pyramidal neurons of mouse prefrontal cortex. Nat Commun 11:5140. 10.1038/s41467-020-18861-3 Weinhard L et al (2018) Microglia remodel synapses by presynaptic trogocytosis and spine head filopodia induction. Nat Commun 9:1228. 10.1038/s41467-018-03566-5 Yan Y et al (2022) ASH1L haploinsufficiency results in autistic-like phenotypes in mice and links Eph receptor gene to autism spectrum disorder. Neuron 110:1156–1172e1159. 10.1016/j.neuron.2021.12.035 Ennerfelt H et al (2022) SYK coordinates neuroprotective microglial responses in neurodegenerative disease. Cell 185:4135–4152e4122. 10.1016/j.cell.2022.09.030 Mascaraque-Checa M et al (2022) Metformin overcomes metabolic reprogramming-induced resistance of skin squamous cell carcinoma to photodynamic therapy. Mol Metab 60:101496. 10.1016/j.molmet.2022.101496 Ulland TK, Colonna M (2018) TREM2 - a key player in microglial biology and Alzheimer disease. Nat Rev Neurol 14:667–675. 10.1038/s41582-018-0072-1 Ulland TK et al (2017) TREM2 Maintains Microglial Metabolic Fitness in Alzheimer's Disease. Cell 170:649–663. .e613 Hu Y et al (2022) Dual roles of hexokinase 2 in shaping microglial function by gating glycolytic flux and mitochondrial activity. Nat Metab 4:1756–1774. 10.1038/s42255-022-00707-5 Chien T et al (2018) GSK3β negatively regulates TRAX, a scaffold protein implicated in mental disorders, for NHEJ-mediated DNA repair in neurons. Mol Psychiatry 23:2375–2390. 10.1038/s41380-017-0007-z Ioannidis V et al (2024) Disrupted extracellular matrix and cell cycle genes in autism-associated Shank3 deficiency are targeted by lithium. Mol Psychiatry 29:704–717. 10.1038/s41380-023-02362-y Liu Z, Smith CB (2014) Lithium: a promising treatment for fragile X syndrome. ACS Chem Neurosci 5:477–483. 10.1021/cn500077p Wang J et al (2023) Effects of different doses of lithium on the central nervous system in the rat valproic acid model of autism. Chem Biol Interact 370:110314. 10.1016/j.cbi.2022.110314 Gideons ES, Lin PY, Mahgoub M, Kavalali ET, Monteggia LM (2017) Chronic lithium treatment elicits its antimanic effects via BDNF-TrkB dependent synaptic downscaling. Elife 6. 10.7554/eLife.25480 Rajkowska G et al (2016) Differential effect of lithium on cell number in the hippocampus and prefrontal cortex in adult mice: a stereological study. Bipolar Disord 18:41–51. 10.1111/bdi.12364 Dong H et al (2014) Lithium ameliorates lipopolysaccharide-induced microglial activation via inhibition of toll-like receptor 4 expression by activating the PI3K/Akt/FoxO1 pathway. J Neuroinflammation 11:140. 10.1186/s12974-014-0140-4 Yu Z et al (2015) Therapeutic concentration of lithium stimulates complement C3 production in dendritic cells and microglia via GSK-3 inhibition. Glia 63:257–270. 10.1002/glia.22749 Stern S et al (2018) Neurons derived from patients with bipolar disorder divide into intrinsically different sub-populations of neurons, predicting the patients' responsiveness to lithium. Mol Psychiatry 23:1453–1465. 10.1038/mp.2016.260 McQuade A et al (2018) Development and validation of a simplified method to generate human microglia from pluripotent stem cells. Mol neurodegeneration 13:67. 10.1186/s13024-018-0297-x Additional Declarations The authors declare no competing interests. Supplementary Files SupplementaryinformationGaoetal.docx Supplementary information Supplementary Fig. 1-11 and Table S1-S3 SupplementaryTable4.xlsx Supplementary Table 4. Candidate GSK3β inhibitors Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6665030","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":456595802,"identity":"6d87f0ac-9939-4576-a75a-ee2702440391","order_by":0,"name":"Shumin Gao","email":"","orcid":"","institution":"Peking University","correspondingAuthor":false,"prefix":"","firstName":"Shumin","middleName":"","lastName":"Gao","suffix":""},{"id":456595803,"identity":"a86f02b2-eab8-4552-84d3-f517cfd16a06","order_by":1,"name":"Qingxiu Lin","email":"","orcid":"","institution":"Peking University","correspondingAuthor":false,"prefix":"","firstName":"Qingxiu","middleName":"","lastName":"Lin","suffix":""},{"id":456595804,"identity":"dd6272ee-41d6-482b-aac5-029e9005f384","order_by":2,"name":"Xiaotong Liu","email":"","orcid":"","institution":"Peking University","correspondingAuthor":false,"prefix":"","firstName":"Xiaotong","middleName":"","lastName":"Liu","suffix":""},{"id":456595805,"identity":"0f04ae79-651e-42ca-b6d8-9bbdea917de7","order_by":3,"name":"Meixiang Jia","email":"","orcid":"","institution":"Peking University Sixth Hospital","correspondingAuthor":false,"prefix":"","firstName":"Meixiang","middleName":"","lastName":"Jia","suffix":""},{"id":456602235,"identity":"857cc0f8-27b0-430f-bc9c-e6e4647ec013","order_by":4,"name":"Anyi Zhang","email":"","orcid":"","institution":"Beijing Children's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Anyi","middleName":"","lastName":"Zhang","suffix":""},{"id":456602236,"identity":"fd721368-a971-4a2a-8155-01d02113fa59","order_by":5,"name":"Zhendong Feng","email":"","orcid":"","institution":"Peking University","correspondingAuthor":false,"prefix":"","firstName":"Zhendong","middleName":"","lastName":"Feng","suffix":""},{"id":456602237,"identity":"51d8049e-72b4-490f-9b1a-e368e9a220a6","order_by":6,"name":"Lei Han","email":"","orcid":"","institution":"Peking University","correspondingAuthor":false,"prefix":"","firstName":"Lei","middleName":"","lastName":"Han","suffix":""},{"id":456602238,"identity":"4d3906bc-f460-48fa-84c7-4be1099f987b","order_by":7,"name":"Nianzhuang Qiu","email":"","orcid":"","institution":"Peking University","correspondingAuthor":false,"prefix":"","firstName":"Nianzhuang","middleName":"","lastName":"Qiu","suffix":""},{"id":456602239,"identity":"aa68751c-abf3-4bdb-932e-e7325e8c0ef0","order_by":8,"name":"Huajie Zhai","email":"","orcid":"","institution":"Liaocheng Dongchangfu District Maternal and Child Health Hospital","correspondingAuthor":false,"prefix":"","firstName":"Huajie","middleName":"","lastName":"Zhai","suffix":""},{"id":456602240,"identity":"7cc30649-7cdb-44ee-919a-9fa25b8fcf3f","order_by":9,"name":"Xiaoyan Zhang","email":"","orcid":"","institution":"Liaocheng Dongchangfu District Maternal and Child Health Hospital","correspondingAuthor":false,"prefix":"","firstName":"Xiaoyan","middleName":"","lastName":"Zhang","suffix":""},{"id":456602241,"identity":"7df24875-0f43-477e-82ed-e2ea70335298","order_by":10,"name":"Jing Zhang","email":"","orcid":"","institution":"Beijing Haidian Maternal \u0026 Child Health Hospital","correspondingAuthor":false,"prefix":"","firstName":"Jing","middleName":"","lastName":"Zhang","suffix":""},{"id":456602242,"identity":"6f897d49-fcde-483a-9484-9fc4dfa88538","order_by":11,"name":"Xiaodan Ding","email":"","orcid":"","institution":"Shanghai Jiao Tong University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Xiaodan","middleName":"","lastName":"Ding","suffix":""},{"id":456602243,"identity":"eb06dd60-1361-4da8-a893-8f4dc4a9eaa9","order_by":12,"name":"Yan Zhang","email":"","orcid":"","institution":"Shanghai Jiao Tong University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Yan","middleName":"","lastName":"Zhang","suffix":""},{"id":456602244,"identity":"3fafc5fc-7658-4a9a-adee-29483b1c2f52","order_by":13,"name":"Lin Lu","email":"","orcid":"","institution":"Peking University Sixth Hospital","correspondingAuthor":false,"prefix":"","firstName":"Lin","middleName":"","lastName":"Lu","suffix":""},{"id":456602245,"identity":"95ffb95a-8ebe-43d9-9946-efef078d36bf","order_by":14,"name":"Jie Shi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzklEQVRIiWNgGAWjYDACCcYGgwQGmwTGBhCPjXgtaSRpAZOHEyA8YrTIz25uKHhQcT6PedoZA4YPZYcZ+Gc34NdicOcg0GFnbhczzs4xYJxx7jCDxJ0DBLRIJDYYJLbdTmwEamHmbTsMFEkg4LAZIC3/zkG0/CVGC8MNkJaGAxAtjMRoMQBpSTiWDPRLWsHBnnPpPBI3CDos/Znhjxq7PMPZyRsf/CizluOfQchhwLgwAJGGDQwMB4A0D0H1QMD8AGwdMUpHwSgYBaNgZAIAhVFHhCl2fGMAAAAASUVORK5CYII=","orcid":"","institution":"Peking University","correspondingAuthor":true,"prefix":"","firstName":"Jie","middleName":"","lastName":"Shi","suffix":""},{"id":456602246,"identity":"ad0deb9e-a10a-403a-b082-7e4f77a3fc98","order_by":15,"name":"Jia Jia Liu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCUlEQVRIiWNgGAWjYBACPmYGBoMEBjYeEEeCoeIAM1iYB48WNlQtZ4jRgsyRYGw7wEBYCzuPQcGDGj4Z/gb2h7d5591h152RwPjgbRuDvDlOh/EYGCQcY+OROMBjbM277Rmz2Y0EZsO5bQyGOxvwaWED+uUAD5s077bDIC1ARhtDgsEBfFr+sfHIH2B/Js07B6yF/TdBLYltbDxABWbSvA0QW5jxa2ErMEjsY+MxBPrFcs4xoF/OPGyWnHNOwnADDi38/Ie3Gf74dsxe7gD7wxtvau4kmx1PPvjhTZmNPC5bQBYZMDAcY2CQfwDmJTMwMDYwgKIVD2AGqq2B8+zwKR0Fo2AUjIKRCQDeMU/BXFfrxwAAAABJRU5ErkJggg==","orcid":"","institution":"Peking University","correspondingAuthor":true,"prefix":"","firstName":"Jia","middleName":"Jia","lastName":"Liu","suffix":""},{"id":456602247,"identity":"3ccecb55-e7dc-4b7e-8ab3-d8664db5a363","order_by":16,"name":"Ya Bin Wei","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5UlEQVRIie3PMQrCMBSA4SeFuITOLYXWI0S6epgUwSmKOIibghAXD6C38AiRgF1aXSMuuqtLD6A1OjilugnmX96j7UdSAJvtJ8MgHgPV50D0A/ExwdkX5JnHXksVIWm+ln2QoetfxGDPIXQVrRV9E8l6VC5Axijo0bjLIfYVdYKFiQhGJIZDwgNGHiRZKYocbCK7syZj7meajKuJep5CkYc1oaSK+OpKJSa3Jsflv7Ct11xmp2lgIu6uKws86kTRLN/EbNiK3LS9LkykIfT19I7K4ZWzNjEAgOjttXM0fmqz2Wx/2x0klksTvB9CjgAAAABJRU5ErkJggg==","orcid":"","institution":"Peking University","correspondingAuthor":true,"prefix":"","firstName":"Ya","middleName":"Bin","lastName":"Wei","suffix":""}],"badges":[],"createdAt":"2025-05-14 14:15:46","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-6665030/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6665030/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":82799695,"identity":"81e25032-54b1-4623-8f34-1a7de462560f","added_by":"auto","created_at":"2025-05-15 11:01:45","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":136072,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMale \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eMll5\u003c/strong\u003e\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u003cstrong\u003e+/-\u003c/strong\u003e\u003c/em\u003e\u003c/sup\u003e\u003cstrong\u003e mice exhibited impairments in repetitive and social behaviors\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Timeline for behavioral testing (schematic created with BioRender) of male mice.\u003c/p\u003e\n\u003cp\u003e(B) Statistical analysis showed that the call number, total duration, and mean duration of \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003epups were decreased at P9. Data represent mean ± SEM.\u003c/p\u003e\n\u003cp\u003e(C) Using the three-chamber paradigm, the sociability and social novelty tests were performed in WT and \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice.\u003c/p\u003e\n\u003cp\u003e(D-F) \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice spent significantly more time in the self-grooming test (D), buried less marbles (E) and had lower nest building scores (F).\u003c/p\u003e\n\u003cp\u003e(G-H) The immobility time in the forced swimming test (G) and the tail suspension test (H) were significantly longer in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice.\u003c/p\u003e\n\u003cp\u003e(I) The discrimination index obtained in the novel object recognition task by WT and \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice.\u003c/p\u003e\n\u003cp\u003e(J) Quantification of the number of platform crossings and the percentage time in target quadrant in the Morris water maze test.\u003c/p\u003e\n\u003cp\u003ens: not significant, *\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05, **\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01, ***\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001. Data represent mean ± SD, \u003cem\u003en \u003c/em\u003e= 8-12 mice per group. P, postnatal day. SEM, standard error of the mean. WT, wild type. SD, standard deviation.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6665030/v1/6df98df7ec041dd1df640d0b.jpg"},{"id":82799696,"identity":"a01edda9-5399-4475-8d3b-611a1d021184","added_by":"auto","created_at":"2025-05-15 11:01:45","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":160821,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eReduced microglial activity in \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eMll5\u003c/strong\u003e\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u003cstrong\u003e+/-\u003c/strong\u003e\u003c/em\u003e\u003c/sup\u003e\u003cstrong\u003e mice\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Representative images of IBA1+ microglia in the hippocampus and the DG region of juvenile WT and \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice. Scale bars, 200 μm (hippocampus) and 100 μm (DG region).\u003c/p\u003e\n\u003cp\u003e(B) IBA1+ microglial density in the DG, CA1, and CA3 regions of juvenile WT and\u003cem\u003e Mll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice showed no significant differences.\u003c/p\u003e\n\u003cp\u003e(C) Quantitative analysis demonstrated elevated microglial branching complexity in the DG region of juvenile and adult \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e \u003c/em\u003emice relative to WT littermates.\u003c/p\u003e\n\u003cp\u003e(D) Quantitative analysis revealed increased junctions per microglia in the DG region of juvenile and adult \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice compared to WT.\u003c/p\u003e\n\u003cp\u003e(E) 3D reconstruction of CD68+ phagocytic structures in the DG region of juvenile WT and \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice. Relative quantification of CD68+ structures per IBA1 volume was significantly reduced in juvenile \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice. Scale bar, 10 μm.\u003c/p\u003e\n\u003cp\u003e(F) 3D reconstruction of CD68+ structures in the DG region of adult WT and \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice. Relative quantification of CD68+ structures per IBA1 volume indicated a significant decrease in adult \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice. \u003cem\u003en\u003c/em\u003e= 3 mice per group. Scale bar, 10 μm.\u003c/p\u003e\n\u003cp\u003ens, not significant; *\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05; **\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01; ***\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001. Data represent mean ± SD, \u003cem\u003en \u003c/em\u003e= 3-5 mice per group. WT, wild type. DG, dentate gyrus. cornu ammonis 1, CA1. mPFC, medial prefrontal cortex. PVN, paraventricular nucleus of hypothalamus. SD, standard deviation.\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6665030/v1/2f7546704f8944ec870523a9.jpg"},{"id":82799697,"identity":"3777bdb6-a3d7-42ba-87d1-1e8c25b9ef68","added_by":"auto","created_at":"2025-05-15 11:01:45","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":155672,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eImpaired synaptic development in \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eMll5\u003c/strong\u003e\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u003cstrong\u003e+/-\u003c/strong\u003e\u003c/em\u003e\u003c/sup\u003e\u003cstrong\u003e mice\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Golgi-Cox staining in the hippocampal DG of juvenile WT and \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice. Quantification revealed increased spine density in the DG of juvenile \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice, with no changes in the CA1 or CA3 regions. \u003cem\u003en \u003c/em\u003e= 4-6 mice per group. Scale bar, 10 μm.\u003c/p\u003e\n\u003cp\u003e(B), Representative images and quantification of NEUN+ neuronal density in the DG, CA1, and CA3 regions of juvenile mice. Scale bar, 20 μm.\u003c/p\u003e\n\u003cp\u003e(C) Representative VGLUT1 immunofluorescence in the DG and relative quantification showing significantly increased VGLUT1 area in juvenile \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice versus WT. Scale bar, 10 μm. \u003cem\u003en\u003c/em\u003e= 3 mice per group.\u003c/p\u003e\n\u003cp\u003e(D) VGLUT1 immunofluorescence in the DG of adult mice demonstrated persistently increased VGLUT area \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice. Scale bar, 10 μm. \u003cem\u003en\u003c/em\u003e= 3 mice per group.\u003c/p\u003e\n\u003cp\u003e(E) Quantitative western blot analysis of synaptic proteins (MAP2, PSD95, SAP102, and VGLUT1) in juvenile hippocampal lysates. \u003cem\u003en \u003c/em\u003e= 6-7 mice per group.\u003c/p\u003e\n\u003cp\u003e(F) Western blot analysis of synaptic markers (MAP2, PSD95, SYNAPSIN I, and VGLUT1) in adult hippocampal lysates., \u003cem\u003en \u003c/em\u003e= 5-7 mice per group.\u003c/p\u003e\n\u003cp\u003ens, not significant; *\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05; **\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01; ***\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001. Data represent mean ± SD. WT, wild type. DG, dentate gyrus. CA1, cornu ammonis 1, CA3, cornu ammonis 3.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6665030/v1/03a5327909d0cdf0221c8734.jpg"},{"id":82800412,"identity":"d29f88f5-c0f4-4b77-b997-404469e0ae45","added_by":"auto","created_at":"2025-05-15 11:09:45","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":172315,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMicroglial \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eMll5\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e haploinsufficiency impaired phagocytosis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Representative images and quantification of pHrodo-conjugated synaptosome phagocytosis by WT and \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e primary microglia (µglia). Phagocytic activity was significantly reduced in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e µglia compared to WT. \u003cem\u003en \u003c/em\u003e= 2-3 mice per group. Scale bars, 100µm.\u003c/p\u003e\n\u003cp\u003e(B) Representative immunostaining of synaptic protein VGLUT1 was performed in primary neuron (PrN) derived from WT mice and cocultured with WT or \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e μglia. The puncta of synaptic protein VGLUT1 was increased in PrN cocultured with \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e μglia, compared to PrN cocultured with WT μglia. \u003cem\u003en \u003c/em\u003e= 3 mice per group. Scale bars, 10 µm, 5 µm.\u003c/p\u003e\n\u003cp\u003e(C) Schematic diagram of the induction protocol for microglia.\u003c/p\u003e\n\u003cp\u003e(D) Immunofluorescence images showing staining for iPSCs and microglia. Scale bar, 100 μm.\u003c/p\u003e\n\u003cp\u003e(E) ASD and control derived microglia showed similar percentage of CD11B\u003csup\u003e+\u003c/sup\u003ePU.1\u003csup\u003e+\u003c/sup\u003e double immunofluorescence cells.\u003cem\u003e n\u003c/em\u003e= 5 per group.\u003c/p\u003e\n\u003cp\u003e(F)\u003cem\u003e MLL5\u003c/em\u003e mRNA expression was significantly reduced in ASD individuals derived microglia versus controls.\u003cem\u003e n\u003c/em\u003e= 5 per group.\u003c/p\u003e\n\u003cp\u003e(G-H) Morphological classification of microglia (ramified, intermediate, round) (G). ASD-derived microglia exhibited increased round/ramified-like morphologies and reduced intermediate states post-synaptosome challenge (H). \u003cem\u003en\u003c/em\u003e= 5 per group.\u003c/p\u003e\n\u003cp\u003e(I) Representative images of microglial phagocytosis of pHrodo-conjugated synaptosomes. Compared to the control, the phagocytosis ability of ASD derived microglia was significantly decreased. \u003cem\u003en\u003c/em\u003e= 5 per group. Scale bar, 20 μm.\u003c/p\u003e\n\u003cp\u003ens, not significant; *\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05; ***\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001; ***\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001. Data represent mean ± SD. WT, wild type. SD, standard deviation. iPSCs, induced pluripotent stem cells.\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6665030/v1/793d89269c0af9b53ec87fe8.jpg"},{"id":82800414,"identity":"cbe387e2-42d8-46a3-bf50-880b973e5ffb","added_by":"auto","created_at":"2025-05-15 11:09:45","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":115471,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eMll5\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e haploinsufficiency impaired TREM2-SGK3-GSK3β signaling pathway\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Schematic diagram of RNA-seq of microglia sorted from hippocampus.\u003c/p\u003e\n\u003cp\u003e(B) Volcano plot analysis of differentially expressed genes in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e versus WT microglia. Orange: significantly upregulated genes; blue: significantly downregulated genes.\u003cem\u003e n \u003c/em\u003e= 3 mice per group.\u003c/p\u003e\n\u003cp\u003e(C) Western blot analysis of P-GSK3β, GSK3β, P-mTOR, mTOR, SGK3 and TREM2 in primary microglia. Data represent mean ± SD, \u003cem\u003en \u003c/em\u003e= 5 mice per group.\u003c/p\u003e\n\u003cp\u003e(D) ECAR of primary microglia at various time points followed by injections of glucose, oligomycin and 2-DG were obtained. Glycolysis was presented as bar graphs. Data represent mean ± SEM and mean ± SD, n = 8 per group.\u003c/p\u003e\n\u003cp\u003e(E) OCR of primary microglia with consecutive injections of oligomycin, FCCP, and Rotenone/antimycin A. Basal respiration was presented as bar graphs. Data represent mean ± SEM, n = 8 per group.\u003c/p\u003e\n\u003cp\u003e(F) ChIP-qPCR assay showed that H3K4me3 association with the \u003cem\u003eTREM2\u003c/em\u003e and\u003cem\u003e SGK3\u003c/em\u003e promoter were decreased in shRNA-\u003cem\u003eMLL5\u003c/em\u003e HMC3 cells. Data represent mean ± SD, \u003cem\u003en \u003c/em\u003e= 3 per group.\u003c/p\u003e\n\u003cp\u003e(G) In comparison to the control group, BV2 cells treated with the SGK3 degrader exhibited a markedly reduced capacity for phagocytosis. Data represent mean ± SD, \u003cem\u003en \u003c/em\u003e= 5 per group. Scar bar, 100 µm.\u003c/p\u003e\n\u003cp\u003e(H) Western blot analysis demonstrated a significant reduction in both the P-GSK3β/GSK3β ratio and SGK3 expression levels in BV2 cells following exposure to the SGK3 degrader, whereas the p-mTOR/mTOR ratio remained unchanged. Data represent mean ± SD. n= 3 per group.\u003c/p\u003e\n\u003cp\u003ens, not significant; *\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05; **p \u0026lt; 0.01; ***p \u0026lt; 0.001. ECAR, extracellular acidification rate. 2-DG, 2-deoxy-glucose. OCR, oxygen consumption rate. FCCP, carbonyl cyanide-p-trifluoromethoxyphenylhydrazone. ChIP-qPCR, Chromatin immunoprecipitation (ChIP)-qPCR. SD, standard deviation. SEM, standard error of the mean.\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6665030/v1/48785c055f3625d587e21ef7.jpg"},{"id":82800413,"identity":"6697b7e9-06b2-448d-a9e1-821b59987ac5","added_by":"auto","created_at":"2025-05-15 11:09:45","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":115874,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLiCl partially restored autism-like behaviors and microglial function in \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eMll5\u003c/strong\u003e\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u003cstrong\u003e+/-\u003c/strong\u003e\u003c/em\u003e\u003c/sup\u003e\u003cstrong\u003e mice\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Schematic timeline of LiCl administration and behavioral tests (created with BioRender).\u003c/p\u003e\n\u003cp\u003e(B) After LiCl treatment, the marble-burying behavior of the mice was rescued in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice. \u003cem\u003en \u003c/em\u003e= 8-10 mice per group.\u003c/p\u003e\n\u003cp\u003e(C) LiCl significantly reduced self-grooming duration in\u003cem\u003e Mll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice versus vehicle (Veh)-treated controls. \u003cem\u003en \u003c/em\u003e= 8-11 mice per group.\u003c/p\u003e\n\u003cp\u003e(D) LiCl treatment did not alter sociability in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice during the three-chamber social interaction test. LiCl normalized social novelty preference in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice.\u003cem\u003e n \u003c/em\u003e= 8-11 mice per group.\u003c/p\u003e\n\u003cp\u003e(E) Representative images of IBA1 in the hippocampal region. Statistical analysis of microglial morphology showed that after LiCl treatment, the branches and junctions of microglia were reduced in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice.\u003cem\u003e n \u003c/em\u003e= 3 mice per group. Scale bar, 100 μm.\u003c/p\u003e\n\u003cp\u003e(F) 3D reconstruction of CD68+ phagocytic structures in the DG region. LiCl increased CD68+ structures per IBA1 volume was significantly increased in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice. \u003cem\u003en \u003c/em\u003e= 3 mice per group. Scale bar, 10 μm.\u003c/p\u003e\n\u003cp\u003e(G) Representative traces of mEPSCs recordings in DG hippocampal acute slices LiCl decreased mEPSC frequency without altering amplitude in\u003cem\u003e Mll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/-\u003c/em\u003e\u003c/sup\u003e mice. Data represent mean ± SEM, \u003cem\u003en \u003c/em\u003e= 3 mice per group. Scale bars, 10 pA, 1 sec.\u003c/p\u003e\n\u003cp\u003ens, not significant; *\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05; **\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01; ***\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001. Data represent mean ± SD. SD, standard deviation. mEPSCs, miniature excitatory postsynaptic currents. SEM, standard error of the mean.\u003c/p\u003e","description":"","filename":"6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6665030/v1/50bf709db132daac26fc3bee.jpg"},{"id":82802068,"identity":"97f4db1a-2163-453e-bba9-9e453883cf06","added_by":"auto","created_at":"2025-05-15 11:33:46","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2378131,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6665030/v1/6db7f6ff-997d-4261-ac83-6cbeebd5fb86.pdf"},{"id":82799717,"identity":"a7ccd519-4155-473a-9552-26340a4275b6","added_by":"auto","created_at":"2025-05-15 11:01:45","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":8159729,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary information\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSupplementary Fig. 1-11 and Table S1-S3\u003c/p\u003e","description":"","filename":"SupplementaryinformationGaoetal.docx","url":"https://assets-eu.researchsquare.com/files/rs-6665030/v1/d595c94150c18caa3bd055b3.docx"},{"id":82800411,"identity":"de25199d-e336-4e61-9da4-905a8ee04790","added_by":"auto","created_at":"2025-05-15 11:09:45","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":49577,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Table 4. Candidate GSK3β inhibitors\u003c/p\u003e","description":"","filename":"SupplementaryTable4.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-6665030/v1/c32d9bc2a931b88edb630cee.xlsx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eMll5\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e haploinsufficiency attenuates microglial phagocytosis through dysregulated TREM2-SGK3-GSK3β signaling in autism\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAutism spectrum disorders (\u003cb\u003eASD\u003c/b\u003e) is a neurodevelopment disorder characterized by impairment in social communication and interaction, along with restricted and repetitive behavioral patterns\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. Individuals with ASD vary greatly in cognitive development and often comorbid with other neuropsychiatric disorders, making its clinical manifestations highly heterogeneous\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Although twin and family studies provide evidence that ASD is predominantly heritable, the exact underlying genetic determinants are largely unknown\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. Recent large scale genomic studies have identified over hundreds of ASD susceptible genes, many of which are involved in chromatin modification and synaptic connections \u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Mixed lineage leukemia 5 (\u003cb\u003eMLL5\u003c/b\u003e), also known as lysine (K)-specific methyltransferase 2E (\u003cb\u003eKMT2E\u003c/b\u003e), is one of the genes associated with ASD in both genome-wide association study and whole exome sequencing studies\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. Probands with \u003cem\u003eMLL5\u003c/em\u003e mutation bare either heterozygous variants or microdeletions, leading to haploinsufficiency of the \u003cem\u003eMLL5\u003c/em\u003e function. \u003cem\u003eKMT2E\u003c/em\u003e-related neurodevelopmental disorder is a genetically defined condition caused by pathogenic variants in the \u003cem\u003eKMT2E\u003c/em\u003e gene, manifesting core features of global developmental delay, mild-to-moderate intellectual disability, and central hypotonia\u003csup\u003e\u003cspan additionalcitationids=\"CR10\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. This disorder is further associated with a heterogeneous clinical spectrum encompassing autism spectrum disorder (ASD), structural brain anomalies (microcephaly, macrocephaly), epilepsy, and subtle dysmorphic facial features\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. Nevertheless, the exact neurobiological function of \u003cem\u003eMLL5\u003c/em\u003e in the developing central nervous system (\u003cb\u003eCNS\u003c/b\u003e) is still unclear.\u003c/p\u003e \u003cp\u003e \u003cem\u003eMLL5\u003c/em\u003e is suggested to act as a transcriptional regulator by affecting histone H3 lysine 4 (H3K4) methylation through direct or indirect mechanisms, exerting a diverse range of biological functions\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. Previous study has been primarily investigated using cancer cell lines, which showed that \u003cem\u003eMLL5\u003c/em\u003e knockdown inhibited cell cycle progression and disrupted genomic stability. \u003cem\u003eIn vivo\u003c/em\u003e experiments have demonstrated that \u003cem\u003eMLL5\u003c/em\u003e was essential for normal hematopoiesis and spermatogenesis, possibly through its function in epigenetic modulation and reactive oxygen species production\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. In the CNS, evidence have suggested that \u003cem\u003eMll5\u003c/em\u003e was involved in regulating neuronal activity, visual function, and glucose metabolism\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. Notably, \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice exhibited social impairment and disturbed electroretinogram responses\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. Additionally, the deficiency of \u003cem\u003eMll5\u003c/em\u003e has been linked to deficits in immune surveillance in peripheral tissues\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e, suggesting its potential role in regulating microglial function within the CNS. Microglia are tissue-resident macrophages derived from myeloid progenitors generated in the yolk sac and migrate into the brain parenchyma prior to blood-brain barrier closure\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. Besides its role in inflammatory response, recent studies have highlighted the participation of microglia in synaptic pruning and extracellular matrix remodeling during brain development, which are crucial for the synaptic maturation and neuronal circuit formation\u003csup\u003e\u003cspan additionalcitationids=\"CR22\" citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. Dysfunction of microglia has been associated with several neurodevelopmental disorders including ASD\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. Despite the known roles of \u003cem\u003eMLL5\u003c/em\u003e in regulating cellular processes, its specific contributions to microglial function within the CNS remains incompletely understood.\u003c/p\u003e \u003cp\u003eIn this study, we investigated the role of \u003cem\u003eMll5\u003c/em\u003e in regulating microglia function during critical postnatal neurodevelopmental windows. We demonstrated that \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice exhibit impaired hippocampal microglial phagocytosis and ASD-like behavioral phenotypes. \u003cem\u003eMll5\u003c/em\u003e haploinsufficiency in microglia impaired the clearance of synapses and disrupted neuronal excitability. Mechanistically, \u003cem\u003eMll5\u003c/em\u003e haploinsufficiency attenuated microglial phagocytic capacity by dysregulating metabolic pathways mediated by the TREM2-SGK3-GSK3β axis. Notably, microglia derived from individuals with ASD exhibited reduced \u003cem\u003eMLL5\u003c/em\u003e expression and diminished phagocytic activity, implicating \u003cem\u003eMLL5\u003c/em\u003e dysfunction in ASD-associated microglial pathophysiology. Pharmacological intervention with lithium chloride (LiCl) restored microglial phagocytic function and ameliorated ASD-like behaviors in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice. Collectively, our findings establish \u003cem\u003eMLL5\u003c/em\u003e as a critical regulator of microglial function during postnatal neurodevelopment, with direct relevance to ASD pathogenesis.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e \u003cb\u003eMll5\u003c/b\u003e \u003csup\u003e \u003cb\u003e+/\u0026minus;\u003c/b\u003e \u003c/sup\u003e \u003cb\u003emice display behavioral abnormalities\u003c/b\u003e\u003c/p\u003e \u003cp\u003eMutations in \u003cem\u003eMll5\u003c/em\u003e have been implicated in a range of behavioral abnormalities associated with neurodevelopmental disorders\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. Here, we determined the behavioral performances resulting from \u003cem\u003eMll5\u003c/em\u003e haploinsufficiency in mice (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA and Supplementary Fig.\u0026nbsp;1A). Behaviorally, male \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e pups (postnatal days (P) 9) that were separated from their mothers emitted ultrasonic vocalizations (USVs), which were recorded and analyzed to assess social communication deficits. Compared to their male wild type (WT) counterparts, male \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e pups exhibited fewer calls, shorter total call durations, and a reduced mean call duration (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). However, female \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e pups did not exhibit abnormal USVs (Supplementary Fig.\u0026nbsp;1B). These results suggest atypical postnatal development of mother-pup communication in male \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e pups. We used the three-chamber social test to assess social interaction behavior. To evaluate sociability, we compared the time that the experimental mouse spent interacting with either a stranger mouse chamber or an empty wired cup chamber. As expected, adult WT mice displayed normal sociability, as reflected by their preferential interaction with the stranger mouse. In contrast, \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice showed no preference for the stranger mouse chamber over the empty cup chamber, indicating impaired sociability (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC and Supplementary Fig.\u0026nbsp;1C). Next, in a social novelty test that assesses social novelty recognition, \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice spent less time in stranger 2 chamber (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC and Supplementary Fig.\u0026nbsp;1D), indicating impaired social novelty recognition. \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice also exhibited higher levels of self-grooming (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD and Supplementary Fig.\u0026nbsp;1E) but buried a significantly lower number of marbles in the marble burying task (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE and Supplementary Fig.\u0026nbsp;1F). We found that \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice produced low-quality nests compared to the WT mice in nest building test (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eF and Supplementary Fig.\u0026nbsp;1G). Male \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice showed increased immobility in forced swimming test (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eG) and tail suspension test (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eH), however no significant changes were observed in female \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice compared to the control mice (Supplementary Fig.\u0026nbsp;1H, 1I), suggesting \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice displayed depression-like behavior in a sex-specific manner. To evaluate recognition memory, novel object recognition tests were performed. \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice did not show any preferences in exploring two identical objects compared to the WT mice (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eI and Supplementary Fig.\u0026nbsp;1J). Additionally, the Morris water maze test was used to investigate the spatial learning and memory. Our results indicated that both WT and \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice spent a similar amount of time in the platform quadrant and had a comparable number of platform crossings during the test (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eJ). Moreover, there was no significant difference in swimming speed between the two groups (Supplementary Fig.\u0026nbsp;1K). These results suggested that deficiency of \u003cem\u003eMll5\u003c/em\u003e in mice resulted in autistic-like behaviors, including abnormal vocal communication, social deficits and stereotyped behaviors, accompanied depressive-like behaviors.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eMll5\u003c/b\u003e \u003csup\u003e \u003cb\u003e+/\u0026minus;\u003c/b\u003e \u003c/sup\u003e \u003cb\u003emice exhibit overall normal brain morphology but reduced microglial activity during postnatal brain development\u003c/b\u003e\u003c/p\u003e \u003cp\u003e \u003cem\u003eMLL5\u003c/em\u003e mRNA expression was detected throughout all brain regions during development and adulthood, peaking around the perinatal period and gradually decreasing after birth (Supplementary Fig.\u0026nbsp;2A, B). \u003cem\u003eMLL5\u003c/em\u003e was expressed in all cell types of the adult brain, with a higher expression in glia cells compared to neuronal cells (Supplementary Fig.\u0026nbsp;2C, D). Previous study has shown that full deletion of \u003cem\u003eMll5\u003c/em\u003e (\u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u0026minus;/\u0026minus;\u003c/em\u003e\u003c/sup\u003e) led to significant postnatal growth retardation\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e, while we found \u003cem\u003eMll5\u003c/em\u003e haploinsufficiency mice had comparable brain/body weight ratio and brain/body length ratio throughout development (Supplementary Fig.\u0026nbsp;2E, F). Adult \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e brain appeared normal morphology and expected thickness using Nissl staining (Supplementary Fig.\u0026nbsp;3A). No obvious cortical lamination abnormality was observed when stained for CTIP2\u0026thinsp;+\u0026thinsp;and CUX1\u0026thinsp;+\u0026thinsp;neurons (Supplementary Fig.\u0026nbsp;3B, C). Consistently, \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e embryonic brain had normal numbers of PAX6\u0026thinsp;+\u0026thinsp;and KI67\u0026thinsp;+\u0026thinsp;cells at embryonic day (E)14.5, indicating that neural progenitor cell proliferation was not affected by \u003cem\u003eMll5\u003c/em\u003e deficiency (Supplementary Fig.\u0026nbsp;3D).\u003c/p\u003e \u003cp\u003eWe focused on the hippocampus in juvenile (P17-21) mice, a developmental window coinciding with active synaptic refinement. Microglial density was quantified through immunostaining for the microglia marker ionized calcium-binding adapter molecule 1 (IBA1) in brain sections from \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e and WT littermates. No significant differences in microglial density were observed across subregions of the hippocampus (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, B and Supplementary Fig.\u0026nbsp;4A) or in other brain regions, including medial prefrontal cortex (mPFC) (Supplementary Fig.\u0026nbsp;4A, 4B) and paraventricular nucleus of the hypothalamus (PVN) (Supplementary Fig.\u0026nbsp;4A, B). Consistent with these observations, fluorescence-activated cell sorting (FACS) of the CD11B\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003elow\u003c/sup\u003e hippocampal microglia revealed comparable proportions across genotypes (Supplementary Fig.\u0026nbsp;5A). Morphological analysis identified a hyper-ramified phenotype specifically in the \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e hippocampi, characterized by an increased number of branches and junctions per cell (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC, D), indicative of reduced microglial activation compared to WT. Consistently, we found decreased CD68-immunolabeled microglial phagosomes per IBA1\u0026thinsp;+\u0026thinsp;volume in the \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE). Furthermore, FACS analysis of the CD11B\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003elow\u003c/sup\u003e microglia indicated lower mean fluorescence intensity (MFI) of CD40 (Figure S5B), while expression levels of major histocompatibility complex class II (MHCII) and CD11B were comparable between \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice and its littermates (Supplementary Fig.\u0026nbsp;5B). Although MLL5 was reported to suppress innate immune response in peripheral macrophages, we did not observe significant elevated inflammatory status in the \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u003c/em\u003e\u0026minus;\u003c/sup\u003e brain. The mRNA levels of \u003cem\u003eTnf\u003c/em\u003e, \u003cem\u003eCcl2\u003c/em\u003e and \u003cem\u003eTgfb1\u003c/em\u003e, as well as the levels of TNFα, IL-6 and IL-1β, were not differentially expressed between \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e mice and the WT controls (Supplementary Fig.\u0026nbsp;5C, D). Similarly, we observed no differences in the number of microglia in the subregions of the hippocampus in adult \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u003c/em\u003e\u0026minus;\u003c/sup\u003e mice (Supplementary Fig.\u0026nbsp;4E). However, microglia from adult \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u003c/em\u003e\u0026minus;\u003c/sup\u003e mice exhibited persistent morphological alterations, including increased branching complexity (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC, D) and reduced percentage of CD68/IBA1 volume (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eJ), indicating lasting effects of \u003cem\u003eMll5\u003c/em\u003e haploinsufficiency. Collectively, these findings demonstrate that \u003cem\u003eMll5\u003c/em\u003e haploinsufficiency impairs microglial activity selectively within the hippocampus without inducing a concomitant inflammatory response.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eMll5\u003c/b\u003e \u003csup\u003e \u003cb\u003e+/\u0026minus;\u003c/b\u003e \u003c/sup\u003e \u003cb\u003emice are impaired in synapse elimination\u003c/b\u003e\u003c/p\u003e \u003cp\u003eThe reduced microglial activity in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e mice implicated impaired synaptic development and disrupted synapse elimination. To investigate this, we performed Golgi-Cox staining to assess dendritic spine density in hippocampal neurons. Golgi-Cox staining revealed a significant increase in dendritic spine density specifically in the dentate gyrus (DG) region of \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e mice (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). No significant differences in NEUN\u0026thinsp;+\u0026thinsp;cell number were observed (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB and Supplementary Fig.\u0026nbsp;6A). Immunofluorescence staining demonstrated protein levels of VGLUT1 and HOMER1 in juvenile \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e mice (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC and Supplementary Fig.\u0026nbsp;6B). In adult \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e mice, VGLUT1 and HOMER1 protein levels remained elevated (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD and Supplementary Fig.\u0026nbsp;6C). We next analyzed synaptic markers via western blot using hippocampal homogenates from juvenile \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e mice. \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e mice exhibited elevated of the immature synapse marker SAP102 compared with littermate controls (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). Levels of MAP2, GLUR1, GLUR2, GRIN2B, PSD95 and VGLUT1 were similar across groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE and Supplementary Fig.\u0026nbsp;6D), indicating that altered synaptic sites were independent of changes in neuronal number in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e mice. Furthermore, adult \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e mice exhibited increased expression of synaptic proteins, including PSD95, SYNAPIN1, and VGLUT1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF), whereas SNAP25 and GEPHYPIN protein levels showed no significant difference change (Supplementary Fig.\u0026nbsp;6E). These results revealed that \u003cem\u003eMll5\u003c/em\u003e haploinsufficiency impaired synaptic development and disrupted synapse elimination of the hippocampus.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eMll5\u003c/b\u003e \u003cb\u003eis required for microglia phagocytosis\u003c/b\u003e \u003cb\u003ein vitro\u003c/b\u003e\u003c/p\u003e \u003cp\u003eWe first examined whether synaptic numbers and dendritic complexity differed between primary \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e and WT neurons. Immunofluorescence and Sholl analyses revealed comparable synaptic numbers and dendritic complexity across all groups (Supplementary Fig.\u0026nbsp;7A, 7B). Consistently, we did not find an obvious difference in spontaneous excitatory postsynaptic currents (sEPSCs) or spontaneous inhibitory postsynaptic currents (sIPSCs) frequency and amplitude between \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e and WT neurons (Supplementary Fig.\u0026nbsp;7C), indicating that \u003cem\u003eMll5\u003c/em\u003e haploinsufficiency did not directly influence neuronal activity. We therefore hypothesized that the synaptic deficits in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice arose specifically from microglial dysfunction. To address this hypothesis, we performed primary culture of microglia from P1-3 \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e and WT brains. The purity of the cultured microglia was identified by immunostaining for IBA1 (Supplementary Fig.\u0026nbsp;8A). \u003cem\u003eMll5\u003c/em\u003e haploinsufficiency had no effect on microglia viability and migration (Supplementary Fig.\u0026nbsp;8B, C). \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e microglia phagocytosed significantly fewer pHrodo-conjugated synaptosomes, as indicated by decreased proportion of cells exhibited red fluorescence (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). However, the efficiency of phagocytosing zymosan was similar across groups (Supplementary Fig.\u0026nbsp;8D), suggesting a mechanism that is target- or receptor-specific. Furthermore, we generated a stable \u003cem\u003eMll5\u003c/em\u003e knockdown cell line (shRNA-\u003cem\u003eMLL5\u003c/em\u003e) in the human derived microglial cell line HMC3. Notably, \u003cem\u003eMLL5\u003c/em\u003e knockdown reduced phagocytosis of pHrodo-conjugated synaptosomes compared with scramble controls (Supplementary Fig.\u0026nbsp;8F). Neuron-microglia co-culture experiments demonstrated that primary neurons co-cultured with \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e microglia exhibited increased synaptic density, recapitulating \u003cem\u003ein vivo\u003c/em\u003e observations in mouse brains (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB \u003cb\u003eand\u003c/b\u003e Supplementary Fig.\u0026nbsp;8F). To determine whether microglial phagocytosis is impaired in ASD, we generated induced pluripotent stem cells (iPSCs) from five individuals with ASD and five age- and sex-matched healthy controls, followed by microglial differentiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). iPSC and microglial identity were confirmed by immunostaining for the stem cell markers TRA-1-60 and OCT4, and the microglial markers PU.1 and CD11B (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD). Microglial cell numbers did not differ significantly between ASD and control groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE). ASD-derived microglia exhibited reduced \u003cem\u003eMLL5\u003c/em\u003e mRNA expression compared with healthy controls (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF). The morphology of microglia is dependent on their activation state, specifically activated or dividing microglia are amoeba-like, while quiescent microglia are more branched-like. Following exposure to pHrodo-labeled synaptosomes, ASD-derived microglia displayed an increased proportion of round and ramified morphologies and a fewer intermediate state (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eJ, H). ASD-derived microglia exhibited significantly impaired phagocytosis of pHrodo-labeled synaptosomes (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eI). Overall, these data suggested that microglial \u003cem\u003eMll5\u003c/em\u003e is essential for synapse pruning.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eMll5\u003c/b\u003e \u003cb\u003ehaploinsufficiency impaired TREM2-SGK3-GSK3β signaling pathway\u003c/b\u003e\u003c/p\u003e \u003cp\u003eTo elucidate the mechanisms underlying \u003cem\u003eMll5\u003c/em\u003e-mediated regulation of microglial phagocytosis, we conducted RNA sequencing (RNA-seq) on FACS-sorted P19 hippocampal microglia to assess genome-wide transcriptional alterations resulting from \u003cem\u003eMll5\u003c/em\u003e deficiency (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). We identified 23 differentially expressed genes (DEGs) comparing \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e and WT mice, with 9 genes upregulated and 14 genes downregulated in the \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e microglia (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB and Supplementary Table\u0026nbsp;1). Five of these DEGs are annotated in the SFARI database, implicating a potential association between \u003cem\u003eMll5\u003c/em\u003e and ASD (Supplementary Fig.\u0026nbsp;9B). To further confirm the decreased phagocytic activity in the \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e microglia, we compared the expression of DEGs between Tau-GFP\u0026thinsp;+\u0026thinsp;and Tau-GFP- microglia isolated from tau-GFP mice, in which neurons express a GFP-fusion protein and in which only microglia that have phagocytosed neuronal materials are labeled with GFP, suggesting a more potent phagocytic microglia\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. Notably, GFP- microglia exhibited lower expression of the DEGs gene signature, suggesting decreased phagocytic activity (Supplementary Fig.\u0026nbsp;9A). Serum/glucocorticoid regulated kinase 3 (\u003cem\u003eSgk3\u003c/em\u003e) emerged as the most significantly downregulated gene in the \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e microglia. SGK3, a member of the AGC kinases family, shares overlapping substrate specificity with the serine-threonine kinase AKT (also known as protein kinase B), which functions as a downstream effector of phosphoinositide 3-kinase (PI3K)\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. SGK3 regulates glycogen synthase kinase 3β (GSK3β) activity via its serine/threonine kinase function\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. Triggering receptor expressed on myeloid cells 2 (TREM2) plays a pivotal role in regulating microglial functions by modulating their phagocytosis, energetic metabolism, migration, and proliferation, thereby sculpting their functional profile within the brain\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e,\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. TREM2 are upstream activators of PI3K and deficiency in TREM2 disrupted phagocytosis and mitochondrial metabolism, at least in part through GSK3β activity\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. We found that SGK3 protein and the ratio of P-GSK3β/GSK3β were significantly decreased in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e primary microglia, as well as in shRNA-\u003cem\u003eMLL5\u003c/em\u003e HMC3 cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC and Supplementary Fig.\u0026nbsp;9C). Moreover, genes related to oxidoreductase activity, such as \u003cem\u003eMical1\u003c/em\u003e, \u003cem\u003eAldh1l2\u003c/em\u003e, were also found to be significantly changed in our analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB and Supplementary Table\u0026nbsp;1). We next investigated whether decreased SGK3 signaling in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e was associated with disturbance in TREM2 signaling. We found that TREM2 expression was decreased both in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e primary microglia and shRNA-\u003cem\u003eMLL5\u003c/em\u003e HMC3 cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC and Supplementary Fig.\u0026nbsp;9D), whereas the ratio of P-mTOR/mTOR and P-AKT/AKT remained unchanged (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC and Supplementary Fig.\u0026nbsp;8C). Therefore, we asked whether anabolic and energetic metabolism were disturbed in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e mice. We analyzed the metabolic states of microglia using metabolic flux assays. Our extracellular acidification rate (ECAR) results revealed that the glycolytic pathway was significantly reduced in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e primary microglia, including glycolysis, and non-glycolytic acidification (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD and Supplementary Fig.\u0026nbsp;9E). \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e primary microglia showed impaired mitochondrial basal respiration, maximal respiration, proton leak, and ATP production in oxygen consumption rate (OCR) measurements (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE and Supplementary Fig.\u0026nbsp;9F-G). Hexokinase 2 (HK2), the rate-limiting glycolytic enzyme, was also decreased in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e primary microglia (Supplementary Fig.\u0026nbsp;9H). However, \u003cem\u003eMLL5\u003c/em\u003e did not increase mitochondrial membrane potential, as measured by TMRM fluorescence intensity (Supplementary Fig.\u0026nbsp;9I).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003eMLL5\u003c/em\u003e was reported to have histone lysine methyltransferase activity that could function as an epigenetic regulator\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. We asked whether the decreased expression of \u003cem\u003eSgk3\u003c/em\u003e and \u003cem\u003eTrem2\u003c/em\u003e in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice was associated with changes in their epigenetic modifications. We used chromatin-immunoprecipitation quantitative PCR (ChIP-qPCR) assay to investigate the histone H3 methylation levels at the promoter regions of \u003cem\u003eTREM2\u003c/em\u003e, \u003cem\u003eSGK3\u003c/em\u003e and \u003cem\u003eMICAL1\u003c/em\u003e genes and found that H3K4me3 levels were significantly decreased in shRNA-\u003cem\u003eMLL5\u003c/em\u003e HMC3 cells compared to the-scramble knockdown cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eF and Supplementary Fig.\u0026nbsp;9J). Consistently, compared with shRNA-Scramble HMC3 cells, shRNA-\u003cem\u003eMLL5\u003c/em\u003e HMC3 cells showed a decrease in the expression level of \u003cem\u003eTrem2\u003c/em\u003e and \u003cem\u003eSgk3\u003c/em\u003e (Supplementary Fig.\u0026nbsp;9D, C). Interestingly, we did not find the overall levels of H3K4me, H3K4me3, H3K9me3, and H3K27me3 showed significant difference between WT and \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e hippocampus (Supplementary Fig.\u0026nbsp;9K). Ultimately, treatment of BV2 cells with an SGK3 degrader resulted in a suppression of phagocytosis in these cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eG). Concurrently, a decrease in the P-GSK3β/GSK3β ratio was observed (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eH). Taken together, our data suggest that \u003cem\u003eMLL5\u003c/em\u003e could promote microglial phagocytosis via the TREM2-SGK3-GSK3β signaling pathway, which associated with changes in anabolic and energetic metabolism.\u003c/p\u003e \u003cp\u003e \u003cb\u003eLiCl treatment partially rescued autism-like behaviors and microglial phagocytosis in\u003c/b\u003e \u003cb\u003eMll5\u003c/b\u003e\u003csup\u003e\u003cb\u003e+/\u0026minus;\u003c/b\u003e\u003c/sup\u003e \u003cb\u003emice\u003c/b\u003e\u003c/p\u003e \u003cp\u003eOur mechanism-driven screening approach identified GSK3β hyperactivity as a therapeutic target for the \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice. Based on clinical efficacy profiles and phenotypic relevance, we employed a drug repurposing strategy and identified 35 candidate GSK3β inhibitors (Supplementary Table\u0026nbsp;4). Among these candidates, Indirubin and Tideglusib are currently clinically approved or under active investigation. SB216763 exhibits high lipophilicity and moderate polarity, properties that enhance blood-brain barrier penetration, a critical advantage for neurological therapeutics\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. LiCl, a well-characterized GSK3β inhibitor, has shown therapeutic potential in an ASD model\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. Therefore, we investigated whether these four inhibitors could rescue the disrupted phagocytic function in shRNA-\u003cem\u003eMLL5\u003c/em\u003e HMC3 cells. Notably, all four compounds restored phagocytic activity in shRNA-\u003cem\u003eMLL5\u003c/em\u003e HMC3 cells (Supplementary Fig.\u0026nbsp;10A-D). At concentrations sufficient to rescue phagocytosis, only LiCl significantly inhibited GSK3β activity (Supplementary Fig.\u0026nbsp;10E). Furthermore, LiCl has a well-documented safety profile\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. Based on these findings, we selected LiCl for \u003cem\u003ein vivo\u003c/em\u003e therapeutic evaluation. To determine whether pharmacological GSK3β inhibition ameliorated behavioral deficits associated with \u003cem\u003eMll5\u003c/em\u003e haploinsufficiency, we administered LiCl (45 mg/kg/d, i.p.) or vehicle (Veh) control to \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice for 4 weeks prior to behavioral assessment (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). LiCl treatment did not induce significant alterations in body weight (Supplementary Fig.\u0026nbsp;11A). Notably, LiCl administration ameliorated marble-burying and self-grooming behaviors, and enhanced social novelty preference, while baseline sociability remained unaffected (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB-D). To investigate whether LiCl treatment alters behavioral changes through modulation of microglial function, we assessed its effects on microglial number, morphology, and phagocytic activity. Our findings demonstrated that LiCl treatment significantly increased microglial cell density in the DG region of \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE and Supplementary Fig.\u0026nbsp;11B), while reducing branching complexity and junctional density (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE). No changes in the number of microglia were observed in the CA1 and CA3 regions of \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice upon LiCl treatment (Supplementary Fig.\u0026nbsp;11C). Furthermore, the volume of the phagocytosis-related molecule CD68 in microglial cells was elevated in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice following LiCl treatment (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eF). Concomitantly, expression levels of excitatory synapse-related molecules, VGLUT1 and HOMER1, were reduced in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice after LiCl administration (Supplementary Fig.\u0026nbsp;11D, E). To further assess synaptic plasticity, we performed electrophysiological analysis of miniature excitatory and inhibitory postsynaptic currents (mEPSCs and mIPSCs) in DG region. Adult \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice exhibited elevated mEPSCs frequency, with no alterations in mIPSCs frequency and amplitudes (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eJ and Supplementary Fig.\u0026nbsp;11F). LiCl treatment restored mEPSCs frequency in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eJ and Supplementary Fig.\u0026nbsp;11F). In summary, these results suggest that LiCl ameliorates behavioral abnormalities in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice by modulating microglial function.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eOur study provides novel insights into the role of \u003cem\u003eMLL5\u003c/em\u003e in postnatal brain development, particularly in regulating microglial activity and the process of synaptic refinement. Consistent with previous findings\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e, our study revealed a critical role for \u003cem\u003eMll5\u003c/em\u003e in the autism-related behaviors, including USVs issues, sociability deficits, repetitive behaviors and depression-like behaviors. Notably, \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice exhibited reduced microglial activity during postnatal period, which was accompanied by impairments in synapse maturation and neurotransmission. Furthermore, we demonstrated that the suppressed microglial phagocytosis in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice were associated with reduced metabolism via the TREM2-SGK3-GSK3β pathway. Additionally, we found that microglia derived from individuals with ASD exhibited reduced phagocytosis and decreased \u003cem\u003eMLL5\u003c/em\u003e mRNA expression. Pharmacological intervention with LiCl partially rescued ASD-like behavioral phenotypes and restored microglial phagocytic activity. These results contribute to the evidence linking the role of \u003cem\u003eMll5\u003c/em\u003e gene in the pathophysiology of ASD.\u003c/p\u003e \u003cp\u003eMicroglia regulate diverse neurodevelopmental processes, including immune surveillance, phagocytosis, synaptic pruning, and the extracellular matrix clearing\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. Studies demonstrate that microglia dynamically interact with synapses via their motile processes, enabling real-time monitoring and modulation of synaptic activity\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. This interaction allows microglia to precisely identify synapses requiring either elimination or strengthening. In addition to direct synaptic contact, microglia secrete cytokines and growth factors that regulate synaptic plasticity, thereby optimizing neural circuit refinement\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e,\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. Seminal studies demonstrate that microglia execute synaptic pruning via a synapse recognition-phagocytosis-clearance mechanism\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. Beyond the complement system, phagocytic \u0026ldquo;eat-me\u0026rdquo; (e.g., TREM2) and \u0026ldquo;don\u0026rsquo;t-eat-me\u0026rdquo; (e.g., CD47/SIRPα, CD22) signals further regulate microglial phagocytic specificity\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. Through these mechanisms, microglia fine-tune synaptic density and neural circuit architecture, thereby establishing functional neural connectivity. Previous study have shown that MLL5 suppresses antiviral innate immune responses\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e, and \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u0026minus;/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice exhibit heightened susceptibility to bacterial infections\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. We therefore hypothesized that \u003cem\u003eMll5\u003c/em\u003e gene is critical for microglial immune surveillance in the CNS. Surprisingly, \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice showed no evidence of enhanced neuroinflammation. Instead, we found that \u003cem\u003eMll5\u003c/em\u003e haploinsufficiency induced morphological abnormalities in hippocampal microglia, and reduced lysosomal phagocytic markers, such as CD68 at P17-21, a time window that is active for microglia-dependent synaptic pruning. Notably, these deficits persisted into adulthood, suggesting that developmental disruptions may exacerbate neurological phenotypes. Consistent with Fabia et al\u0026rsquo; study that observed TREM2 mediated synaptic refinement by microglia during the early stages of brain development\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. TREM2 modulates microglial phagocytosis, morphology, and motility, thereby shaping neurodevelopment and disease progression\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. We observed decreased \u003cem\u003eTrem2\u003c/em\u003e expression levels in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e microglia and this reduction was epigenetically modified by \u003cem\u003eMll5\u003c/em\u003e\u0026rsquo;s histone methyltransferases activity. This finding establishes an epigenetic link between \u003cem\u003eMll5\u003c/em\u003e and TREM2, implicating \u003cem\u003eMll5\u003c/em\u003e in microglia-dependent synaptic pruning and neurodevelopment, thereby advancing our understanding of its CNS roles.\u003c/p\u003e \u003cp\u003eSynaptic dysfunction is implicated in numerous neurological and psychiatric disorders, including ASD\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. The hippocampus is involved in cognitive functions such as social interaction, spatial cognition, and memory, and plays a role in the pathogenesis of ASD\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. Our data demonstrate that MLL5 haploinsufficiency elevates dendritic spine density in the hippocampal DG, accompanied by increased frequency of mEPSCs. These findings are consistent with recent studies that report synaptic hyperconnectivity associated with ASD. For instance, \u003cem\u003eFmr1\u003c/em\u003e knockout mice exhibit increased mEPSC frequency and dendritic spine density in CA1 and DG hippocampal neurons\u003csup\u003e\u003cspan additionalcitationids=\"CR47\" citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. However, it seems that the spine density and morphology varied across different animal models with ASD. In the valproic acid (VPA)-induced ASD mouse or rat model, an increase in dendritic spines\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e,\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e.Notably, postmortem studies of ASD individuals have similarly reported increased dendritic spine density in frontal and temporal cortices\u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e. Conversely, decreased dendritic spine and reduced excitatory synaptic transmission were reported in the anterior cingulate cortex in \u003cem\u003eShank3\u003c/em\u003e knockout mice\u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e. Knockdown of \u003cem\u003eCntnap2\u003c/em\u003e in layer 2/3 pyramidal neurons of the PFC reduced excitatory and inhibitory synaptic transmission, presumably by decreasing the number of functional synapses\u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e. Furthermore, disruption of signaling between microglia and neurons leads to an excess of immature synaptic connections, which is thought to be the result of impaired phagocytosis of synapses by microglia\u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e. These observations together with our findings in the \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice, suggest that dysfunctional synapse in ASD may be dependent on a region and genetic specific manner.\u003c/p\u003e \u003cp\u003eEpigenetic regulation is crucial for numerous fundamental cellular functions, and a growing body of research indicates that epigenetics significantly influences ASD\u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e. \u003cem\u003eMLL5\u003c/em\u003e is an epigenetic regulator that is involved in histone modification, particularly by affecting H3K4 methylation\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. Our transcriptome analysis in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e microglia showed that among the differentially expressed genes, several of which were also implicated in SFARI database, suggesting a role of \u003cem\u003eMll5\u003c/em\u003e in regulating the expression of autism-related genes. Consistently, we found that H3K4me3 levels were significantly decreased at the promoter regions of \u003cem\u003eSgk3, Trem2 and Mical1\u003c/em\u003e. However, different from previous studies\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e, we failed to find an overall decreased H3K4me3 level in the hippocampus. We next focused on \u003cem\u003eSgk3\u003c/em\u003e, which was the most significantly differentially genes between \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e and controls. \u003cem\u003eSgk3\u003c/em\u003e shares similar substrate specificity with AKT and can act as a downstream mediator of PI3K\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. Microglia was reported to phagocytose amyloid beta in part through TREM2-AKT mediated signaling\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e, and this process was compromised if GSK3β signaling was impaired\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e,\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e. We thus hypothesized that \u003cem\u003eMll5\u003c/em\u003e may modulate microglial phagocytosis through the TREM2-SGK3-GSK3β signaling pathway. Accordingly, we found that TREM2, SGK3 and the ratio of P-GSK3β/GSK3β were significantly decreased in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e microglia. Yet, the ratio of P-mTOR/mTOR and P-AKT/AKT was not affected by \u003cem\u003eMll5\u003c/em\u003e deficiency. Oxidative phosphorylation and glycolysis are the main metabolic pathways of energy production in cells\u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e. Previous studies have shown that glycolysis and energetic metabolism may be involved in effective microglial phagocytosis \u003csup\u003e\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e,\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e. We found that \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e primary microglia exhibited reduced glycolysis and ATP production. Moreover, we observed that levels of HK2, a key rate-limiting enzyme in glycolysis specifically expressed in microglia\u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e, were decreased when \u003cem\u003eMLL5\u003c/em\u003e was knockdown, suggesting a role of \u003cem\u003eMll5\u003c/em\u003e in regulating cellular metabolism at least in part through the TREM2-SGK3-GSK3β signaling pathway.\u003c/p\u003e \u003cp\u003eLi, as a mood stabilizer, is widely used in the treatment of bipolar disorder and has demonstrated efficacy as a GSK3β inhibitor in ameliorating psychiatric disorders \u003csup\u003e\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e. Recent studies have explored its potential for ASD therapy\u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e. In our study, microglia with \u003cem\u003eMll5\u003c/em\u003e haploinsufficiency exhibited significantly elevated GSK3β activity, prompting us to evaluate LiCl\u0026rsquo;s potential to ameliorate ASD-like behaviors in mice. LiCl administration partially improved core behavioral deficits, including increased marble-burying behavior, reduced self-grooming duration, and enhanced social novelty preference in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice. These behavioral improvements align with prior reports demonstrating Li\u0026rsquo;s efficacy in rescuing ASD-like behaviors in \u003cem\u003eShank3\u003c/em\u003e- and \u003cem\u003eFmr1\u003c/em\u003e- deficient mouse models\u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e,\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e, underscoring its broad therapeutic relevance across genetically distinct ASD subtypes. In a VPA-induced rat model of ASD, Li ameliorated social cognitive impairments, enhanced social memory, and reduced anxiety levels. Furthermore, Li treatment reduced pro-inflammatory markers and increased neuroprotective histone modifications in the hippocampus, suggesting modulation of neuroinflammatory pathways implicated in ASD\u003csup\u003e\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e\u003c/sup\u003e. Li regulates synaptic plasticity, particularly through GSK3β inhibition, and modulates microglial function. It reduces AMPA receptor-mediated mEPSCs amplitude while altering postsynaptic homeostatic plasticity, suggesting regulation through AMPA receptor activity\u003csup\u003e\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e\u003c/sup\u003e. Li increases glial cell numbers in the DG\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e, suppresses microglial activation\u003csup\u003e\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e, and elevates C3 expression in microglia\u003csup\u003e\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e, suggesting enhanced phagocytic function. Our findings indicate that the frequency of mEPSCs in \u003cem\u003eMll5\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice decreased following treatment with LiCl. Additionally, we observed morphological changes in microglia, along with an enhanced phagocytic capacity of these cells after LiCl treatment. These alterations may contribute to the observed improvement in autism-like behaviors associated with LiCl administration. These alterations may underlie the behavioral improvements associated with lithium, indicating its therapeutic potential for ASD via GSK3β and microglial modulation.\u003c/p\u003e \u003cp\u003eIn conclusion, our study expands our understanding of \u003cem\u003eMll5\u003c/em\u003e\u0026rsquo;s role in the postnatal neurodevelopment and highlights its potential as a therapeutic target for ASD.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eEthical approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study received ethical approval from the Ethical Review Board of Peking University Health Science Center and Peking University Sixth Hospital.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMice\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe generation and genotyping of \u003cem\u003eMll5\u003csup\u003e+/-\u003c/sup\u003e\u003c/em\u003e mice were previously described\u003csup\u003e18\u003c/sup\u003e. The genotyping of WT and \u003cem\u003eMll5\u003csup\u003e+/-\u003c/sup\u003e\u003c/em\u003e micewere performed by PCR analysis using genomic DNA isolated from the tail tips using the primers listed in Table S3. All animals were kept on a 12 h:12 h reversed light/dark cycle, and all experiments were performed in the animals’ dark phase. No more than 5 animals per cage. The mice had ad libitum access to food and water.Behavioral experiments were performed using 8-12-week-old mice. Behavioral test was done blind to genotype of each mouse with age-matched littermates of mice.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePatient selection and clinical assessment\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFive male individuals diagnosed with ASD according to the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) criteria and five age- and gender-matched healthy controls were recruited from the Child and Adolescent Psychiatric Outpatient Department of Peking University Sixth Hospital. The healthy controls were screened based on the absence of\u0026nbsp;history of mental health and substance use disorders. Those with a history of psychological and psychiatric disorders were excluded (Table S2). The research aim, procedure, benefit and risk were explained to the parents of the participating subjects, and the written informed consent were signed by them.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLiCl administration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAdult \u003cem\u003eMll5\u003csup\u003e+/-\u003c/sup\u003e\u0026nbsp;\u003c/em\u003emice received daily intraperitoneal (i.p.) injections of LiCl (45 mg/kg/day; Thermo Fisher Scientific, MA, USA) or vehicle for 28 days.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWestern blot analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTissues and cells were homogenized in cell lysis buffer (Meilunbio, Dalian, China), which was supplemented with EDTA-free protease inhibitor cocktail (Thermo Fisher Scientific) and protein phosphatase inhibitor (Solarbio, Beijing, China). The sample was centrifuged for 10 min at 13,000 g at 4°C followed by protein concentration quantification using the Pierce BCA Protein Assay Kit (Thermo Fisher). Samples were run on a NuPAGE Bis-Tris protein gel (Thermo Fisher) and transferred to PVDF membrane (Merck KGaA, Darmstadt, Germany), followed by blocking with fast block buffer (Epizyme, Shanghai, China) for 20 min at room temperature (RT). Primary antibody was incubated overnight at 4°C followed by secondary antibody incubation for one hour at RT. The blot was developed with Immobilon Western Chemiluminescent HRP Substrate (Merck).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eQuantitative real-time PCR (qPCR)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTotal RNA was isolated using TRIzol reagent (Thermo Fisher) and then subjected to reverse transcription with a HiScript III 1st Strand cDNA Synthesis Kit (Vazyme, Nanjing, China). The resulting cDNA was used for qPCR analysis with AceQ Universal SYBR qPCR Master Mix (Vazyme), and specific primers in a QuantStudio 5 Real-Time PCR System (Thermo Fisher). qPCR primer sequences were presented in supplementary table (Table S3). Gene expression levels were calculated using the 2\u003csup\u003e−ΔΔCt\u003c/sup\u003e analysis method, and the samples were normalized to GAPDH.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImmunofluorescence staining\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUnder deep anesthesia, mice were transcardially perfused with 4% paraformaldehyde followed by brain dissection. After postfixation, brains were dehydrated in 30% sucrose, and then embedded in O.C.T. for cryopreservation. The samples were cut coronally at 50-μm sections on cryostat (Leica, Wetzlar, Germany). Sections were washed in 1× PBS and incubated with 3% BSA and 0.3% Triton X-100 in PBS for 1 h at RT to minimize nonspecific binding. Then, the sections were incubated overnight at 4 °C in primary antibody solution followed by washing and incubating with a species-specific secondary antibody for 1 h at RT. The sections were further washed with 1× PBS and incubated with DAPI at RT for 10 min, and washed twice with PBS before mounting. The samples were observed under a laser confocal scanning microscope (Leica).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEnzyme-linked immunosorbent assay\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;(ELISA)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe concentrations of TNFα, IL-6, and IL-1β in the mouse hippocampus were measured using the specific ELISA kits (ABclonal, Wuhan, China) and the assays were conducted in strict accordance with the protocols supplied. The levels of TNFα, IL-6, and IL-1β in each sample were calculated according to the standard curve generated by serial dilution of the recombinant mouse TNFα, IL-6, and IL-1β.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSlice electrophysiology\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMice were deeply anesthetized with isoflurane inhalation and decapitated. Transverse hippocampal slices (250 µm thick) were prepared with a vibratome (Leica VT1200S) in an ice-cold cutting solution containing the following (in mM): 80 NaCl, 26 NaHCO\u003csub\u003e3\u003c/sub\u003e, 3.0 KCl, 1.0 NaH\u003csub\u003e2\u003c/sub\u003ePO\u003csub\u003e4\u003c/sub\u003e, 1.3 MgCl\u003csub\u003e2\u003c/sub\u003e, 1.0 CaCl\u003csub\u003e2\u003c/sub\u003e, 20 D-glucose, and 75 sucrose, saturated with 95% O2 and 5% CO\u003csub\u003e2\u0026nbsp;\u003c/sub\u003efrom P16-P21 male littermates. Slices were recovered for 30 min at 33 °C, and then at RT until recordings in artificial cerebrospinal fluid (ACSF) containing (in mM): 124 NaCl, 26 NaHCO\u003csub\u003e3\u003c/sub\u003e, 3.0 KCl, 1.0 NaH\u003csub\u003e2\u003c/sub\u003ePO\u003csub\u003e4\u003c/sub\u003e, 1.3 MgCl\u003csub\u003e2\u003c/sub\u003e, 2 CaCl\u003csub\u003e2\u003c/sub\u003e, 20 D-glucose. Slices were perfused with oxygenated ACSF at 33°C at 2.0 mL/min. Signals were amplified by a Multiclamp 700B amplifier (Axon Instruments/Molecular Devices, San Jose, CA, USA) and digitized using Digidata 1550A (Molecular Devices). Recordings were made from hippocampal DG pyramidal neurons. For miniature excitatory post-synaptic currents (mEPSCs) recordings, 1 mM TTX were added to the ACSF. Pipettes contained (in mM): 135 CsGluconate, 1 EGTA, 10 HEPES, 2 MgCl\u003csub\u003e2\u003c/sub\u003e, 4 MgATP, and 0.3 Tris-GTP, (pH 7.4).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePrimary cortical neuron culture\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePrimary neurons were isolated from E17 mice. Briefly, cortices were isolated and dissociated with 0.25% trypsin-EDTA. The cells were plated onto poly-D-lysine-coated coverslips and first cultured in DMEM medium containing 10% fetal bovine serum (FBS), 1% GlutaMax and 1% penicillin/streptomycin for 4 h. Then he medium was switched to Neurobasal plus medium, which contained 1% GlutaMax, 2% B27 plus supplement and 1% penicillin/streptomycin. All experiments were performed with cultures that were 13-15 days in vitro (DIV).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePrimary microglia culture\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe cerebral cortices of P1-3 mice were isolated and treated with 0.25% trypsin-EDTA to digest the tissues. The single cells were seeded in T12.5 cm\u003csup\u003e2\u003c/sup\u003e flasks coated with poly-L-lysine and cultured in DMEM/F12 medium containing 10% FBS and 1% penicillin/streptomycin. Medium change was conducted one day after plating and then after 3 days. At 12 days, flasks were mechanically shaking to collect microglia. Microglia were plated on 96-well plates in DMEM/F12 containing 10% FBS and 1% penicillin/streptomycin to continue the downstream experiments. IN microglia-neuron co-culture experiments, WT or \u003cem\u003eMll5\u003csup\u003e+/-\u003c/sup\u003e\u0026nbsp;\u003c/em\u003emicroglia were added to neurons (13-14 DIV) at microglia to neuron ratio of 1.5:1 for 24 h.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCell culture recording\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWhole cell voltage clamp recording of cultured neurons from WT and \u003cem\u003eMll5\u003csup\u003e+/-\u003c/sup\u003e\u0026nbsp;\u003c/em\u003emice were performed at 13-15 days \u003cem\u003ein vitro\u003c/em\u003e. During recordings cell were bathed in a standard external solution containing (in mM): 125 NaCl, 5 KCl, 1.2 MgSO\u003csub\u003e4\u003c/sub\u003e, 1.2 KH\u003csub\u003e2\u003c/sub\u003ePO\u003csub\u003e4\u003c/sub\u003e, 2 CaCl\u003csub\u003e2\u003c/sub\u003e, 6 glucose and 25 HEPES-NaOH, pH 7.4. Recording pipettes (resistances of 3-5 MΩ) were filled with a standard intracellular solution containing (in mM):135 CsGluconate, 1 EGTA, 10 HEPES, 2 MgCl2, 4 MgATP, and 0.3 Tris-GTP, (pH 7.4).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGolgi-Cox staining\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGolgi-Cox staining was performed using a FD Rapid Golgistain Kit (FD NeuroTechnologies, Columbia, MD, USA). Briefly, freshly dissected brains were immersed in solution A and B for 2 weeks at RT and then transferred into solution C for 72 h at RT in the dark. 150 µm slices were prepared and the sections placed on a gelatin-coated glass slide and dried naturally at RT for 2 days in the dark. Slices were placed in a mixture of solution D: E: distilled water at 1:1:2 ratio for 10 min followed by serial dehydration in 50%, 75%, 95% and 100% ethanol. Clear in xylene, three times for 4 minutes each time, and the cover slides. Images of dendritic spines were acquired using a 100× objective. Spines that started from 50 µm distance of the apical dendrite were counted within a 20 µm segment. The density of the spines was analyzed from an individual blind to genotype of neurons. For frequency distributions, spines were separated into thin, mushroom and stubby spines based on their shape.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNissl staining\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBrains were perfused as described above, and consecutive 50 μm-thick coronal sections were mounted on slides, rehydrated, stained with tar purple dye for 30 min (Solarbio, Beijing, China). After being rinsed with distilled water and 95% alcohol, the sections were treated with conventional dehydration, transparency and tablet sealing. Images were observed and photographed using an optical microscope.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInduced pluripotent stem cells (iPSCs) reprogramming\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eiPSCs were generated from five male individuals diagnosed with ASD (ASD 1, ASD 2, ASD 3, ASD 4, ASD 5) according to the DSM-5 and five male control participants (Control 1, Control 2, Control 3, Control 4, Control 5) according to the published protocol\u003csup\u003e69\u003c/sup\u003e.In brief,Epstein-Barr virus immortalized B-lymphocytes were reprogrammed using the Yamanaka Episomal vector, consisting of three sets of episomal plasmids expressing reprograming factors: pCXLE-hOCT3/4-shp53, pCXLE-hSK (SOX2, KLF4) and pCXLE-hUL (L-MYC, LIN28). Two to three iPSCs colonies were collected from each individual and Standard G-banding karyotype and immunostaining of OCT4 and TRA-1-60 marker were performed to check iPSC quality.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDifferentiation to human iPSCs-derived microglia\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDifferentiation to human iPSC-derived microglia was as previously described\u003csup\u003e70\u003c/sup\u003e.In brief,generation of CD43+ primitive hematopoietic progenitor cells (HPCs) from iPSCs were conducted according to STEMdiff™ Hematopoietic Kit (Stem Cell Technologies), then harvested HPCs were induced to microglia. On day 0, HPCs were seeded onto Matrigel-coated plates, culturing in microglia medium (DMEM/F12, 2× insulin-transferrin-selenite, 2× B27, 0.5× N2, 1× glutamax, 1× non-essential amino acids, 400 μM monothioglycerol, 5 μg/mL insulin), supplemented with 100 ng/mL IL-34 (PeproTech), 50 ng/mL TGFβ1 (Novoprotein), and 25 ng/mL M-CSF (Novoprotein). The medium was changed every other day. On day 25, microglial medium was supplemented with additional 100 ng/mL CD200 (Novoprotein) and 100 ng/mL CX3CL1 (Novoprotein). On day 28, cells were collected for functional assays.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSynaptosomes purification and engulfment assay \u003cem\u003ein vitro\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSynaptosomes of cortex were extracted using Syn-PER™ Synaptic Protein Extraction Reagent (Thermo Fisher). Cortex was homogenized with Dounce grinder with 10 slow strokes followed by centrifugation at 1200g for 10 min at 4 °C. After discarding the pellet, the supernatant was centrifuged at 15,000g for 20 min at 4 °C. The supernatant part was the cytosolic fraction and the pellet containing the synaptosomes was dissolved with the extraction reagent. Protein quantification was performed using Pierce BCA Protein Assay Kit (Thermo Fisher). Resulting synaptosomes were collected and labeled with the pH-sensitive fluorescent dye pHrodo (Thermo Fisher) as per manufacturer’s instructions. Cells were seed into 96-well plate overnight, and pHrodo labeled synaptosomes were added to the cell cultures. Phagocytosis was determined as the red object counts per image over time.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFlow Cytometry\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMice aged P17-P21 were sacrificed via decapitation and the hippocampi were dissected in sterile, ice-cold Hank’s balanced salt solution (HBSS). Single cell suspensions were prepared using the brain dissociation kit (Miltenyi Biotec, Bergisch Gladbach, Germany) and the single cell suspension dissociator (Shenzhen Ruiwode Life Technology Co., Ltd., Shenzhen, China). Brain cell suspensions were then resuspended in 30% Percoll followed by centrifugation at 900g for 20 min without brake. The interface was removed and the deposition was washed with HBSS. Red blood cells were lysed using blood cell lysis buffer. Cell suspensions were then spun down at 300g for 10 min and pellets were resuspended in flow cytometry buffer, labeled with a combination of the following conjugated antibodies: CD11b-PE, CD45-PeCy7, MHC II-APC, and CD40-FITC. FACS analysis of the microglial profile was performed by gating CD11b\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003elow\u003c/sup\u003e cells on a FACS Aria SORP (BD Biosciences). Appropriate antibody IgG isotype controls were used for all staining. Data were analyzed with FlowJo Software (TreeStar).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSeahorse XFe96 Analyzer\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe oxygen consumption rate (OCR) and the extracellular acidification rate (ECAR) were measured using a Seahorse XFe96 Analyzer (Agilent, California, CA, USA) with Seahorse XF Cell Mito Stress Test Kit (Agilent) and Seahorse XF Glycolysis Stress Test Kit (Agilent), respectively, following the manufacturer’s instructions. Briefly, microglia were seeded in 96-well Seahorse assay plates at a concentration of 2 × 10\u003csup\u003e4\u003c/sup\u003e cells/well, and cultured overnight for attachment. At indicated time points, microglia were washed and analyzed in XF Running buffer A (XF assay medium, 10 mM glucose, 1 mM pyruvate sodium, 2 mM l-Glutamine) for OCR measurements or XF Running buffer B (XF assay medium, 2 mM l-Glutamine) for ECAR measurements.\u0026nbsp;Measurements were obtained in real-time under basal conditions (no drug treatment) and with the sequential treatment with different drugs: 1 μM oligomycin, 1 μM FCCP, 1 μM rotenone plus 5 μM antimycin A for the Mito-Stress assay, or 10 mM glucose, 1 μM oligomycin and 50 mM 2-DG for the Glycolysis stress assay.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eUltrasonic vocalizations of isolated mice pups\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOn the testing day, dams and pups were acclimated to the experimental room for ≥30 minutes in a temperature-controlled environment (25°C). Prior to testing, individual pups were randomly separated from dams and littermates and placed into a sound-attenuated chamber for a 1-minute acclimation period. Vocalizations were recorded for 5 minutes using Avisoft Recorder USGH software (Avisoft Bioacoustics) paired with an ultrasonic microphone. Raw audio files were imported into Avisoft SASLab Pro (Avisoft Bioacoustics) and filtered to eliminate amplitudes under 30 kHz before further processing. All the call traces were extracted blindly using an automatic detection function with a proper threshold and manually checked. Call numbers, total duration, and mean duration were analyzed from the labeled call traces.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThree-chamber social test\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe three-chamber apparatus (40 cm × 60 cm × 23 cm) was constructed with three chambers (40 cm × 20 cm × 23 cm). Each side compartments contained an empty wire cage. The test is divided into two phases. (A) Habituation: the mice were allowed to explore the center chamber for 5 min. (B) Sociability and social novelty test: Following habituation, Stranger Mouse 1 (age- and gender-matched) was placed in one wire mesh cage, while the opposite cage remained empty. The empty cage functioned as a non-social control object. The subject mouse was placed in the central chamber and given 10 minutes to explore the entire apparatus. Subsequently, Stranger Mouse 2 (novel mouse) was introduced into the previously empty cage, and the subject mouse was allowed to explore the apparatus again for 10 min. Total time spent in the chamber and in close interaction with the cages was recorded to evaluate the social function and social memory of the mice.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMarble burying\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBriefly, the clean mouse cage was prepared by placing 12 glass marbles (14 mm diameter) evenly on 5 cm deep bedding material. The experimental animal was left undisturbed for 30 min in the test cage in an isolated place. A marble was considered buried when 2/3 of the marble was covered in bedding material.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSelf-grooming test\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis test was conducted in a new mouse cage with a thin layer bedding reducing neophobia. The cumulative time that the test mouse spent grooming itself was measured for 30 min following 10 min period of acclimatization to the test cage.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNovel object recognition test\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBoth object recognition tests were performed in a square white plastic box (40 cm × 27 cm × 18cm). Mice were allowed to explore two identical objects (A) in the testing arena for 3 min (Trial 1-Training Trial). Subsequently, the mice were habituated to the arena for 15 min. The last stage, the mice were returned to the arena, with one of the familiar objects (A) now replaced by a novel object (B) for 3 min (Trial 2-Testing trial). The performance was evaluated by novel object preference index= B exploration time/ (A recognition time + B exploration time).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTail suspension test\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe tail-suspension test was performed using a specially manufactured tail-suspension box (55 × 60 cm) divided into four 15 cm compartments. In each compartment, a small plastic bar hangs from the base. The tail of each mouse was taped to this bar and animals were suspended for 6 min. The immobility time was determined.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eForced swimming test\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMice were gently placed into 5 L of beaker filled with 3.5 L 25°C water for 6 min. The immobility time of each mouse spent during the test was counted online by two independent observers blind to the animal treatments. An increase in immobile time was an indication of depression.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNest building test\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNest building test is an innate behavior that assesses sensorimotor function in mice. The mice were transferred to a new home cage with clean bedding and new nest-building material, a 6 × 6 cm square of compressed cotton. After 36 h, the nest was scored on a 1-5 rating scale, with 1 meaning the nestlet more than 90% intact; 2, the nestlet was partially torn but 50-90% intact; 3, the nestlet was torn with less than 50% remaining intact, but no obvious nest with shreds spread around the cage; 4, a flat nest without raised walls, more than 90% of the nestlet was shredded; 5, the nest was built perfectly with walls higher than mouse height and no unshredded pieces.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMorris water maze test\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe Morris water maze test was conducted to assess spatial learning and memory. Mice were trained to locate a submerged escape platform positioned in the northeastern quadrant of a circular pool (diameter: 110 cm), which was filled with opacified water maintained at 25°C. During the acquisition phase, mice performed 4 trials per day from random starting position for 5 consecutive days with a maximum trial duration of 60 seconds. Spatial memory was assessed by removing the platform and conducting 60-second probe trials. The number of entries into the quadrants and other relevant parameters were analyzed using Morris maze video tracking software (Zhongshi Di Chuang, Beijing, China).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRNA-seq.\u003c/strong\u003e Total RNA was extracted from the sorted microglia of P17-21 mice hippocampus using a TRIzol reagent (Thermo Fisher). mRNA was reverse transcribed and amplified using the SMART-seq method. Complementary DNA libraries were constructed using the TruePrep DNA Library Prep Kit V2 for Illumina (Vazyme), and sequenced with a HiSeq 2500 system (Illumina). Differentially expressed genes were identified using the edgeR package (3.18.1) in R (3.4.1). Differentially expressed genes were required to have an adjusted \u003cem\u003ep\u003c/em\u003e -value \u0026lt; 0.1.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eChromatin immunoprecipitation (ChIP) qPCR\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHMC3 cells were fixed with 1% formaldehyde for 10 min at RT to cross-link proteins to DNA. This reaction was subsequently stopped by adding 0.125 M glycine for 5 min. Chromatin is extracted and sheared by sonication. Appropriate antibodies were mixed and incubated overnight at 4 °C with gentle shaking. Agarose protein A/G beads (Santa Cruz Biotechnology) were used to precipitate the antibody-protein-DNA complexes. Following washing of the beads, the crosslinking was reversed. The DNA captured by the immunoprecipitation was subjected to qPCR analysis. In parallel, an input control sample, which consisted of sheared chromatin collected prior to the addition of antibodies and equivalent to 10% of the immunoprecipitation sample volume, was processed to standardize the pull-down efficiencies.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGraphPad Prism 8.0.1 was used for statistical analyses. Statistical significance was estimated with Student's t test when variances were equal or with Mann-Whitney U test when variances were not equal. Differences between multiple groups were assessed with one-way or two-way ANOVA.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGene expression profiling data have been deposited at GEO database (GEO: GSE280750) and are publicly available as of the date of publication. Source data are provided with this paper.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank Professor Lih Wen Deng for sharing MLL5 antibody with us, and we thank all the participating subjects. This work was funded by National Natural Science Foundation of China (82130040, 82288101, 82171530, 81801344), the Youth Program of National Natural Science Foundation of China (82009Y3510), Nursing Science Research Fund of Peking University (TYZH2023002), and The Fundamental Research Funds for the Central Universities (71006Y2557).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eY.B.W. conceived and designed the experiments. S.G., Q.L., X.L., Z.F., L.H., and N.Q. performed the experiments and analyzed the data. Y.B.W. was responsible for validation. M.J., A.Z, X.X.L., H.Z., X.Z., Z.J., Y.Z., L.L., S.J., J.J.L. and Y.B.W. provided resources. S.G. wrote the manuscript. Y.B.W., J.J.L, J.S. and L.L. revised the manuscript. Y.B.W. and J.J.L supervised the project.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMaterials \u0026amp; Correspondence\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCorrespondence and requests for materials should be addressed to Shi Jie, Jia Jia Liu or Ya Bin Wei.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eLord C et al (2006) Autism from 2 to 9 years of age. Arch Gen Psychiatry 63:694\u0026ndash;701. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/archpsyc.63.6.694\u003c/span\u003e\u003cspan address=\"10.1001/archpsyc.63.6.694\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLord C et al (2020) Autism spectrum disorder. Nat Rev Dis Primers 6:5. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41572-019-0138-4\u003c/span\u003e\u003cspan address=\"10.1038/s41572-019-0138-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBailey A et al (1995) Autism as a strongly genetic disorder: evidence from a British twin study. Psychol Med 25:63\u0026ndash;77. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1017/s0033291700028099\u003c/span\u003e\u003cspan address=\"10.1017/s0033291700028099\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSandin S et al (2014) The familial risk of autism. JAMA 311:1770\u0026ndash;1777. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jama.2014.4144\u003c/span\u003e\u003cspan address=\"10.1001/jama.2014.4144\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSatterstrom FK et al (2020) Large-Scale Exome Sequencing Study Implicates Both Developmental and Functional Changes in the Neurobiology of Autism. \u003cem\u003eCell\u003c/em\u003e 180, 568\u0026ndash;584 e523. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.cell.2019.12.036\u003c/span\u003e\u003cspan address=\"10.1016/j.cell.2019.12.036\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou X et al (2022) Integrating de novo and inherited variants in 42,607 autism cases identifies mutations in new moderate-risk genes. Nat Genet 54:1305\u0026ndash;1319. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41588-022-01148-2\u003c/span\u003e\u003cspan address=\"10.1038/s41588-022-01148-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrove J et al (2019) Identification of common genetic risk variants for autism spectrum disorder. Nat Genet 51:431\u0026ndash;444. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41588-019-0344-8\u003c/span\u003e\u003cspan address=\"10.1038/s41588-019-0344-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDong S et al (2014) De novo insertions and deletions of predominantly paternal origin are associated with autism spectrum disorder. Cell Rep 9:16\u0026ndash;23. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.celrep.2014.08.068\u003c/span\u003e\u003cspan address=\"10.1016/j.celrep.2014.08.068\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSharawat IK, Panda PK, Dawman L (2021) Clinical Characteristics and Genotype-Phenotype Correlation in Children with KMT2E Gene-Related Neurodevelopmental Disorders: Report of Two New Cases and Review of Published Literature. Neuropediatrics 52:98\u0026ndash;104. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1055/s-0040-1715629\u003c/span\u003e\u003cspan address=\"10.1055/s-0040-1715629\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbreu NJ et al (2022) Novel truncating variant in KMT2E associated with cerebellar hypoplasia and velopharyngeal dysfunction. Clin Case Rep 10:e05277. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/ccr3.5277\u003c/span\u003e\u003cspan address=\"10.1002/ccr3.5277\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eO'Donnell-Luria AH et al (2019) Heterozygous Variants in KMT2E Cause a Spectrum of Neurodevelopmental Disorders and Epilepsy. Am J Hum Genet 104:1210\u0026ndash;1222. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ajhg.2019.03.021\u003c/span\u003e\u003cspan address=\"10.1016/j.ajhg.2019.03.021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAli M et al (2013) Molecular basis for chromatin binding and regulation of MLL5. Proc Natl Acad Sci U S A 110:11296\u0026ndash;11301. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1073/pnas.1310156110\u003c/span\u003e\u003cspan address=\"10.1073/pnas.1310156110\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLemak A et al (2013) Solution NMR structure and histone binding of the PHD domain of human MLL5. PLoS ONE 8:e77020. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pone.0077020\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0077020\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTasdogan A et al (2016) DNA Damage-Induced HSPC Malfunction Depends on ROS Accumulation Downstream of IFN-1 Signaling and Bid Mobilization. Cell Stem Cell 19:752\u0026ndash;767. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.stem.2016.08.007\u003c/span\u003e\u003cspan address=\"10.1016/j.stem.2016.08.007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi YJ et al (2023) KMT2E Haploinsufficiency Manifests Autism-Like Behaviors and Amygdala Neuronal Development Dysfunction in Mice. Mol Neurobiol 60:1609\u0026ndash;1625. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12035-022-03167-w\u003c/span\u003e\u003cspan address=\"10.1007/s12035-022-03167-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang X et al (2022) MLL5 is involved in retinal photoreceptor maturation through facilitating CRX-mediated photoreceptor gene transactivation. iScience 25:104058. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.isci.2022.104058\u003c/span\u003e\u003cspan address=\"10.1016/j.isci.2022.104058\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHeuser M et al (2009) Loss of MLL5 results in pleiotropic hematopoietic defects, reduced neutrophil immune function, and extreme sensitivity to DNA demethylation. Blood 113:1432\u0026ndash;1443. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/blood-2008-06-162263\u003c/span\u003e\u003cspan address=\"10.1182/blood-2008-06-162263\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou P et al (2018) MLL5 suppresses antiviral innate immune response by facilitating STUB1-mediated RIG-I degradation. Nat Commun 9:1243. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41467-018-03563-8\u003c/span\u003e\u003cspan address=\"10.1038/s41467-018-03563-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSwinnen N et al (2013) Complex invasion pattern of the cerebral cortex bymicroglial cells during development of the mouse embryo. Glia 61:150\u0026ndash;163. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/glia.22421\u003c/span\u003e\u003cspan address=\"10.1002/glia.22421\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGinhoux F et al (2010) Fate mapping analysis reveals that adult microglia derive from primitive macrophages. Science 330:841\u0026ndash;845. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1126/science.1194637\u003c/span\u003e\u003cspan address=\"10.1126/science.1194637\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePaolicelli RC et al (2011) Synaptic pruning by microglia is necessary for normal brain development. Science 333:1456\u0026ndash;1458. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1126/science.1202529\u003c/span\u003e\u003cspan address=\"10.1126/science.1202529\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchafer DP et al (2012) Microglia sculpt postnatal neural circuits in an activity and complement-dependent manner. Neuron 74:691\u0026ndash;705. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.neuron.2012.03.026\u003c/span\u003e\u003cspan address=\"10.1016/j.neuron.2012.03.026\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNguyen PT et al (2020) Microglial Remodeling of the Extracellular Matrix Promotes Synapse Plasticity. \u003cem\u003eCell\u003c/em\u003e 182, 388\u0026ndash;403 e315. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.cell.2020.05.050\u003c/span\u003e\u003cspan address=\"10.1016/j.cell.2020.05.050\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLukens JR, Eyo UB (2022) Microglia and Neurodevelopmental Disorders. Annu Rev Neurosci 45:425\u0026ndash;445. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1146/annurev-neuro-110920-023056\u003c/span\u003e\u003cspan address=\"10.1146/annurev-neuro-110920-023056\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWamsley B et al (2024) Molecular cascades and cell type-specific signatures in ASD revealed by single-cell genomics. Science 384:eadh2602. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1126/science.adh2602\u003c/span\u003e\u003cspan address=\"10.1126/science.adh2602\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHe D et al (2022) Disruption of the IL-33-ST2-AKT signaling axis impairs neurodevelopment by inhibiting microglial metabolic adaptation and phagocytic function. Immunity 55:159\u0026ndash;173e159. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.immuni.2021.12.001\u003c/span\u003e\u003cspan address=\"10.1016/j.immuni.2021.12.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eScortegagna M et al (2015) PDK1 and SGK3 Contribute to the Growth of BRAF-Mutant Melanomas and Are Potential Therapeutic Targets. Cancer Res 75:1399\u0026ndash;1412. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1158/0008-5472.Can-14-2785\u003c/span\u003e\u003cspan address=\"10.1158/0008-5472.Can-14-2785\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu F, Wu X, Jiang X, Qian Y, Gao J (2018) Prolonged inhibition of class I PI3K promotes liver cancer stem cell expansion by augmenting SGK3/GSK-3β/β-catenin signalling. J Exp Clin Cancer Res 37:122. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s13046-018-0801-8\u003c/span\u003e\u003cspan address=\"10.1186/s13046-018-0801-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang S et al (2022) TREM2 drives microglia response to amyloid-β via SYK-dependent and -independent pathways. Cell 185:4153\u0026ndash;4169e4119. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.cell.2022.09.033\u003c/span\u003e\u003cspan address=\"10.1016/j.cell.2022.09.033\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu Z et al (2020) TMEM59 interacts with TREM2 and modulates TREM2-dependent microglial activities. Cell Death Dis 11:678. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41419-020-02874-3\u003c/span\u003e\u003cspan address=\"10.1038/s41419-020-02874-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKeren-Shaul H et al (2017) A Unique Microglia Type Associated with Restricting Development of Alzheimer's Disease. Cell 169:1276\u0026ndash;1290e1217. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.cell.2017.05.018\u003c/span\u003e\u003cspan address=\"10.1016/j.cell.2017.05.018\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang Y et al (2015) TREM2 lipid sensing sustains the microglial response in an Alzheimer's disease model. Cell 160:1061\u0026ndash;1071. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.cell.2015.01.049\u003c/span\u003e\u003cspan address=\"10.1016/j.cell.2015.01.049\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang X, Novera W, Zhang Y, Deng LW (2017) MLL5 (KMT2E): structure, function, and clinical relevance. Cell Mol Life Sci 74:2333\u0026ndash;2344. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00018-017-2470-8\u003c/span\u003e\u003cspan address=\"10.1007/s00018-017-2470-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXiao H, Deng M, Yang B, Tang J, Hu Z (2017) Role of glycogen synthase kinase 3 in ischemia-induced blood-brain barrier disruption in aged female rats. J Neurochem 142:194\u0026ndash;203. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/jnc.14051\u003c/span\u003e\u003cspan address=\"10.1111/jnc.14051\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu W et al (2023) Lithium Cholesterol Sulfate: A Novel and Potential Drug for Treating Alzheimer's Disease and Autism Spectrum Disorder. CNS Neurol Disord Drug Targets 22:1250\u0026ndash;1258. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2174/1871527321666220825114236\u003c/span\u003e\u003cspan address=\"10.2174/1871527321666220825114236\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSzklarska D, Rzymski P (2019) Is Lithium a Micronutrient? From Biological Activity and Epidemiological Observation to Food Fortification. Biol Trace Elem Res 189:18\u0026ndash;27. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12011-018-1455-2\u003c/span\u003e\u003cspan address=\"10.1007/s12011-018-1455-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSierra A, Paolicelli RC, Kettenmann H (2019) Cien A\u0026ntilde;os de Microgl\u0026iacute;a: Milestones in a Century of Microglial Research. Trends Neurosci 42:778\u0026ndash;792. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.tins.2019.09.004\u003c/span\u003e\u003cspan address=\"10.1016/j.tins.2019.09.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePereira-Iglesias M et al (2025) Microglia as hunters or gatherers of brain synapses. Nat Neurosci 28:15\u0026ndash;23. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41593-024-01818-w\u003c/span\u003e\u003cspan address=\"10.1038/s41593-024-01818-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBollinger JL, Burns B, C. M., Wellman CL (2016) Differential effects of stress on microglial cell activation in male and female medial prefrontal cortex. Brain Behav Immun 52:88\u0026ndash;97. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.bbi.2015.10.003\u003c/span\u003e\u003cspan address=\"10.1016/j.bbi.2015.10.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen Y et al (2019) miR-124/VAMP3 is a novel therapeutic target for mitigation of surgical trauma-induced microglial activation. Signal Transduct Target Ther 4. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41392-019-0061-x\u003c/span\u003e\u003cspan address=\"10.1038/s41392-019-0061-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDing X et al (2030) Loss of microglial SIRPα promotes synaptic pruning in preclinical models of neurodegeneration. \u003cem\u003eNat Commun\u003c/em\u003e 12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41467-021-22301-1\u003c/span\u003e\u003cspan address=\"10.1038/s41467-021-22301-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFilipello F et al (2018) The Microglial Innate Immune Receptor TREM2 Is Required for Synapse Elimination and Normal Brain Connectivity. Immunity 48:979\u0026ndash;991e978. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.immuni.2018.04.016\u003c/span\u003e\u003cspan address=\"10.1016/j.immuni.2018.04.016\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSu\u0026aacute;rez-Calvet M et al (2016) Early changes in CSF sTREM2 in dominantly inherited Alzheimer's disease occur after amyloid deposition and neuronal injury. Sci Transl Med 8:369ra178. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1126/scitranslmed.aag1767\u003c/span\u003e\u003cspan address=\"10.1126/scitranslmed.aag1767\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun XD et al (2016) Lrp4 in astrocytes modulates glutamatergic transmission. Nat Neurosci 19:1010\u0026ndash;1018. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nn.4326\u003c/span\u003e\u003cspan address=\"10.1038/nn.4326\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBanker SM, Gu X, Schiller D, Foss-Feig JH (2021) Hippocampal contributions to social and cognitive deficits in autism spectrum disorder. Trends Neurosci 44:793\u0026ndash;807. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.tins.2021.08.005\u003c/span\u003e\u003cspan address=\"10.1016/j.tins.2021.08.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJawaid S et al (2018) Alterations in CA1 hippocampal synapses in a mouse model of fragile X syndrome. Glia 66:789\u0026ndash;800. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/glia.23284\u003c/span\u003e\u003cspan address=\"10.1002/glia.23284\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrossman AW et al (2010) Developmental characteristics of dendritic spines in the dentate gyrus of Fmr1 knockout mice. Brain Res 1355:221\u0026ndash;227. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.brainres.2010.07.090\u003c/span\u003e\u003cspan address=\"10.1016/j.brainres.2010.07.090\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang L et al (2024) Anterior piriform cortex dysfunction underlies autism spectrum disorders-related olfactory deficits in Fmr1 conditional deletion mice. Neuropsychopharmacology. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41386-024-02027-6\u003c/span\u003e\u003cspan address=\"10.1038/s41386-024-02027-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCheng Y et al (2022) Degraded cortical temporal processing in the valproic acid-induced rat model of autism. Neuropharmacology 209:109000. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.neuropharm.2022.109000\u003c/span\u003e\u003cspan address=\"10.1016/j.neuropharm.2022.109000\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang EJ, Ahn S, Lee K, Mahmood U, Kim HS (2016) Early Behavioral Abnormalities and Perinatal Alterations of PTEN/AKT Pathway in Valproic Acid Autism Model Mice. PLoS ONE 11:e0153298. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pone.0153298\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0153298\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHutsler JJ, Zhang H (2010) Increased dendritic spine densities on cortical projection neurons in autism spectrum disorders. Brain Res 1309:83\u0026ndash;94. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.brainres.2009.09.120\u003c/span\u003e\u003cspan address=\"10.1016/j.brainres.2009.09.120\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo B et al (2019) Anterior cingulate cortex dysfunction underlies social deficits in Shank3 mutant mice. Nat Neurosci 22:1223\u0026ndash;1234. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41593-019-0445-9\u003c/span\u003e\u003cspan address=\"10.1038/s41593-019-0445-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSacai H et al (2020) Autism spectrum disorder-like behavior caused by reduced excitatory synaptic transmission in pyramidal neurons of mouse prefrontal cortex. Nat Commun 11:5140. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41467-020-18861-3\u003c/span\u003e\u003cspan address=\"10.1038/s41467-020-18861-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeinhard L et al (2018) Microglia remodel synapses by presynaptic trogocytosis and spine head filopodia induction. Nat Commun 9:1228. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41467-018-03566-5\u003c/span\u003e\u003cspan address=\"10.1038/s41467-018-03566-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYan Y et al (2022) ASH1L haploinsufficiency results in autistic-like phenotypes in mice and links Eph receptor gene to autism spectrum disorder. Neuron 110:1156\u0026ndash;1172e1159. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.neuron.2021.12.035\u003c/span\u003e\u003cspan address=\"10.1016/j.neuron.2021.12.035\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEnnerfelt H et al (2022) SYK coordinates neuroprotective microglial responses in neurodegenerative disease. Cell 185:4135\u0026ndash;4152e4122. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.cell.2022.09.030\u003c/span\u003e\u003cspan address=\"10.1016/j.cell.2022.09.030\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMascaraque-Checa M et al (2022) Metformin overcomes metabolic reprogramming-induced resistance of skin squamous cell carcinoma to photodynamic therapy. Mol Metab 60:101496. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.molmet.2022.101496\u003c/span\u003e\u003cspan address=\"10.1016/j.molmet.2022.101496\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUlland TK, Colonna M (2018) TREM2 - a key player in microglial biology and Alzheimer disease. Nat Rev Neurol 14:667\u0026ndash;675. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41582-018-0072-1\u003c/span\u003e\u003cspan address=\"10.1038/s41582-018-0072-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUlland TK et al (2017) TREM2 Maintains Microglial Metabolic Fitness in Alzheimer's Disease. Cell 170:649\u0026ndash;663. .e613\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu Y et al (2022) Dual roles of hexokinase 2 in shaping microglial function by gating glycolytic flux and mitochondrial activity. Nat Metab 4:1756\u0026ndash;1774. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s42255-022-00707-5\u003c/span\u003e\u003cspan address=\"10.1038/s42255-022-00707-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChien T et al (2018) GSK3β negatively regulates TRAX, a scaffold protein implicated in mental disorders, for NHEJ-mediated DNA repair in neurons. Mol Psychiatry 23:2375\u0026ndash;2390. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41380-017-0007-z\u003c/span\u003e\u003cspan address=\"10.1038/s41380-017-0007-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIoannidis V et al (2024) Disrupted extracellular matrix and cell cycle genes in autism-associated Shank3 deficiency are targeted by lithium. Mol Psychiatry 29:704\u0026ndash;717. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41380-023-02362-y\u003c/span\u003e\u003cspan address=\"10.1038/s41380-023-02362-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu Z, Smith CB (2014) Lithium: a promising treatment for fragile X syndrome. ACS Chem Neurosci 5:477\u0026ndash;483. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1021/cn500077p\u003c/span\u003e\u003cspan address=\"10.1021/cn500077p\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang J et al (2023) Effects of different doses of lithium on the central nervous system in the rat valproic acid model of autism. Chem Biol Interact 370:110314. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.cbi.2022.110314\u003c/span\u003e\u003cspan address=\"10.1016/j.cbi.2022.110314\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGideons ES, Lin PY, Mahgoub M, Kavalali ET, Monteggia LM (2017) Chronic lithium treatment elicits its antimanic effects via BDNF-TrkB dependent synaptic downscaling. Elife 6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.7554/eLife.25480\u003c/span\u003e\u003cspan address=\"10.7554/eLife.25480\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRajkowska G et al (2016) Differential effect of lithium on cell number in the hippocampus and prefrontal cortex in adult mice: a stereological study. Bipolar Disord 18:41\u0026ndash;51. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/bdi.12364\u003c/span\u003e\u003cspan address=\"10.1111/bdi.12364\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDong H et al (2014) Lithium ameliorates lipopolysaccharide-induced microglial activation via inhibition of toll-like receptor 4 expression by activating the PI3K/Akt/FoxO1 pathway. J Neuroinflammation 11:140. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12974-014-0140-4\u003c/span\u003e\u003cspan address=\"10.1186/s12974-014-0140-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu Z et al (2015) Therapeutic concentration of lithium stimulates complement C3 production in dendritic cells and microglia via GSK-3 inhibition. Glia 63:257\u0026ndash;270. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/glia.22749\u003c/span\u003e\u003cspan address=\"10.1002/glia.22749\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStern S et al (2018) Neurons derived from patients with bipolar disorder divide into intrinsically different sub-populations of neurons, predicting the patients' responsiveness to lithium. Mol Psychiatry 23:1453\u0026ndash;1465. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/mp.2016.260\u003c/span\u003e\u003cspan address=\"10.1038/mp.2016.260\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcQuade A et al (2018) Development and validation of a simplified method to generate human microglia from pluripotent stem cells. Mol neurodegeneration 13:67. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s13024-018-0297-x\u003c/span\u003e\u003cspan address=\"10.1186/s13024-018-0297-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Peking University","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Autism spectrum disorder, Mll5, microglia, phagocytosis, Sgk3","lastPublishedDoi":"10.21203/rs.3.rs-6665030/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6665030/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAutism spectrum disorder (ASD) is a complex neurodevelopmental disorder characterized by persistent deficits in social communication and repetitive behaviors. Recent studies indicated that heterozygous mutations in the mixed lineage leukemia 5 (\u003cem\u003eMLL5\u003c/em\u003e) gene are implicated in the ASD susceptibility and associated with neurodevelopmental abnormalities. However, the detailed mechanisms remain unclear. Here, we demonstrate that \u003cem\u003eMll5\u003c/em\u003e haploinsufficiency in mice impairs microglial phagocytosis, drives neuronal hyperexcitability, and recapitulates core ASD-like behaviors. We also show that \u003cem\u003eMll5\u003c/em\u003e acts as an epigenetic regulator, modulating microglial phagocytosis via the TREM2-SGK3-GSK3β signaling axis, which is associated with deficient glucose metabolism. Furthermore, ASD individual-derived microglia exhibit parallel reductions in \u003cem\u003eMLL5\u003c/em\u003e expression and phagocytic function. By targeting this pathway, lithium chloride, a GSK3β inhibitor, rescues both microglial phagocytosis deficits and behavioral abnormalities in \u003cem\u003eMll5\u003c/em\u003e haploinsufficiency mice. Our findings highlight \u003cem\u003eMLL5\u003c/em\u003e\u0026rsquo;s critical role in ASD and its potential as a therapeutic target.\u003c/p\u003e","manuscriptTitle":"Mll5 haploinsufficiency attenuates microglial phagocytosis through dysregulated TREM2-SGK3-GSK3β signaling in autism","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-15 11:01:40","doi":"10.21203/rs.3.rs-6665030/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"d5d5d438-99ee-4d71-9d1b-39b3399d1046","owner":[],"postedDate":"May 15th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-05-15T11:01:40+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-15 11:01:40","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6665030","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6665030","identity":"rs-6665030","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}