Shank3 establishes AMPA receptor subunit composition at cerebellar mossy fiber-granule cell synapses and shapes regional microglia activation

9 Mutations in Shank3 are the primary genetic cause of Phelan-McDermid Syndrome (PMS), a 10 neurodevelopmental disorder frequently comorbid with autism spectrum disorder (ASD). As a key 11 scaffolding protein in the postsynaptic site, SHANK3 is critical for excitatory glutamatergic synapse 12 function by interacting with AMPARs, NMDARs, and mGluRs. While Shank3 deficiency has been 13 extensively studied in forebrain regions, its role in the cerebellum, a brain area increasingly implicated 14 in ASD pathobiology, remains comparatively underexplored. Cerebellar granule cells (CGCs) exhibit 15 high Shank3 expression. However, its role in cerebellar glutamatergic synapses is poorly understood. 16 This study aims to investigate how Shank3 loss affects mossy fiber-CGC glutamatergic synaptic 17 function. 18 Whole-cell patch clamp electrophysiological recordings from CGCs in ex vivo cerebellar brain 19 slices from adult (4-6 months old) wild type (WT) and homozygous Shank3∆ ex4-22 KO were performed to 20 record miniature, evoked, and glutamate uncaged responses. Similarly, the current-voltage (I-V) 21 relationship was analyzed with intracellular spermine and pharmacological validation of calcium-22 permeable AMPARs (CP-AMPARs) was done by IEM-1460. Immunofluorescence staining was 23 performed for microglia using IBA1 labeling. 24 We found a significant increase in m EPSC amplitude and AMPAR-mediated response to 25 glutamate uncaging, which indicates that the loss of Shank3 enhances postsynaptic AMPAR function. 26 Furthermore, the KO group showed faster AMPAR decay kinetics, inward rectification, and increased 27 sensitivity to IEM-1460, suggesting that a high proportion of CP-AMPARs with distinct biophysical 28 properties are present at the MF-CGC synapse. Furthermore, KO mice showed less ramified microglia 29 suggesting the possible presence of activated microglia in the cerebellar cortex. 30 Together, these findings highlight a critical role of Shank3 in maintaining the balance between 31 CP- and CI-AMPARs at the MF-CGC synapse, which is essential for synapse maturation and proper 32 cerebellar circuitry function. Dysregulation of this balance, with possible presence of activated microglia 33 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 3 in the cerebellum, may underscore ce rebellar-related behavioral deficits in Shank3 KO mice and may 34 suggest a potential mechanism contributing to ASD pathophysiology. 35 Key Words 36 Shank3, Autism spectrum disorder, Phelan-McDermid Syndrome, Cerebellum, AMPAR, Calcium-37 permeable AMPAR, GluA2, GluA4, Cerebellar granule cells, Microglia 38 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 4

39 Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder with a 40 heterogeneous group of symptoms typically diagnosed in early childhood (Jung & Park, 2022; H. Kim et 41 al., 2016; K. Wang et al., 2017). ASD is characterized by a group of core behavioral symptoms which 42 include impaired social communication, restricted interactions, interests or activities, and repetitive 43 behaviors. (Berkel et al., 2018; Diagnostic and Statistical Manual of Mental Disorders: DSM-5 TM, 2013; 44 Gandhi & Lee, 2021; H. Kim et al., 2016; Lord et al., 2020; K. Wang et al., 2017). Disrupted synaptic 45 structure and function has been often proposed as a converging mechanism underlying the diverse 46 phenotypic manifestations of ASD. The identification of numerous ASD-associated genes that encode 47 proteins integral to synaptic biology has led to the concepts of “synaptic hypothesis” or “synaptopathy” 48 (Bauman & Kemper, 2005; Sahin & Sur, 2015; Won et al., 2013). 49 SHANK3 is widely recognized as a prominent candidate gene strongly associated with ASD 50 (Betancur & Buxbaum, 2013; Hulbert & Jiang, 2017). Along with being the cause of Phelan-McDermid 51 Syndrome (Betancur & Buxbaum, 2013), a mutation or deletion affecting SHANK3 accounts for 52 approximately 1-2% of all ASD cases, making it one of the prominent monogenic causes of ASD with a 53 high penetrance for ASD phenotypes (Jiang & Ehlers, 2013; Soorya et al., 2013). Although classically 54 considered as a regulator of postsynaptic glutamate receptors, Shank3 plays an important role in 55 facilitating protein-protein interactions between neurotransmitter receptors, ion channels, intracellular 56 cytoskeleton components, and signal transduction pathway s (Schmeisser & Verpelli, 2016; Wan et al., 57 2021). Shank3 protein is expressed as different isoforms in the cortex, hippocampus, striatum, 58 thalamus, and cerebellum (Monteiro & Feng, 2017; Peça et al., 2011) with Shank3a, b, and e highly 59 expressed in the striatum and Shank3c and d in the cerebellum (X. Wang et al., 2014). Deletion of 60 Shank3 affects synaptic function in striatum (Peça et al., 2011; Yoo et al., 2018), hippocampus (Bey et 61 al., 2018; Bozdagi et al., 2010; Kouser et al., 2013), thalamus (B. Guo et al., 2024), and cortex (B. Guo 62 et al., 2019; Yoo et al., 2019). 63 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 5 Considerable progress has been made in explaining the neurobiological basis of synaptic 64 dysfunction, with most studies reporting alterations in excitatory glutamatergic synaptic transmission in 65 Shank3 mutant mice (Berg et al., 2018; Bozdagi et al., 2010; Jiang & Ehlers, 2013; Sala et al., 2015; 66 Wan et al., 2021). However, heterogeneity is observed depending on the specific genetic manipulation, 67 brain region, and developmental stage examined (Jung & Park, 2022; Zhang et al., 2023). Shank3∆ e4–22 68 mouse model has shown alterations in physiological processes, particularly synaptic transmission and 69 plasticity in the striatum (Bey et al., 2018; Drapeau et al., 2018; X. Wang et al., 2016). Shank3 deletion 70 is reported to increase basic cellular excitability (Bozdagi et al., 2010; X. Wang et al., 2011) and 71 decrease the frequency of s pontaneous excitatory pos tsynaptic currents (sEPSCs), as well as impair 72 long-term depression (LTD) of medium spiny neurons (MSNs) in the striatum (X. Wang et al., 2011). 73 Similarly, conditional KO of Shank3 exons 4-22 in neocortical excitatory neurons has shown an 74 increase in the NMDA/AMPA ratio in CA1 neurons of the hippocampus (Bozdagi et al., 2010). 75 Over the past two decades, many studies have highlighted the cerebellum as a key brain region 76 implicated in ASD pathogenesis (Becker & Stoodley, 2013; Fatemi et al., 2012; Hampson & Blatt, 2015; 77 Mosconi et al., 2015; Tsai, 2016; S. S. H. Wang et al., 2014). Several experiments using animal models 78 with ASD-related genes ( Tsc1, Shank2, Mecp2 ) have been conducted to investigate the role of the 79 cerebellum, primarily focusing on Purk inje cells (PCs) (Achilly et al., 2021; S. Ha et al., 2016; Kloth et 80 al., 2015, 2015; Stoodley et al., 2017; Tsai et al., 2012). These studies were critical in establishing a 81 role for ASD-linked genes specifically in PCs to shape behaviors outside of the motor domain. 82 However, gene/mRNA expression data (Furuichi et al., 2011; Lein et al., 2006; Lonsdale et al., 2013; 83 Menashe et al., 2013; Sato et al., 2008; Satterstrom et al., 2020) suggest a role for ASD-linked genes in 84 the CGCs that integrate all multisensory input entering the cerebellar cortex to then determine the firing 85 behavior of PCs. Alteration of some ASD-linked genes ( Chd8, Ib2) expressed by CGCs has shown 86 motor dysfunction as well as alterations in CGC excitability and synaptic functions (Kawamura et al., 87 2021; Soda et al., 2019). Overall, the delineation of mechanisms by which ASD-linked genes affect the 88 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 6 physiology of cerebellar cortical neurons, especially CGCs, and how such genes in the cerebellum 89 contribute to motor and non-motor processes is lacking. 90 Additionally, some studies have shown the potential involvement of glial cells (astrocytes and 91 microglia) in contributing to the neuroinflammatory profile observed in ASD (Bailey et al., 1998; Vargas 92 et al., 2005; Ahmad et al., 2017; Matta et al., 2019). Postmortem studies from individuals diagnosed 93 with ASD have frequently revealed neuroinflammation, especially within the cerebellum (Vargas et al., 94 2005). Furthermore, neuropathological investigation of postmortem brain tissue from individuals with 95 ASD has consistently demonstrated alterations in microglial characteristics (Andoh et al., 2019; Fan et 96 al., 2023; Hu et al., 2022; Hughes et al., 2023; Matta et al., 2019; Petr elli et al., 2016; Xiong et al., 97 2023, 2023). These studies reported an increased microglial number and a morphology suggestive of 98 an activated state, in multiple brain regions, including the frontal lobes and cerebellum. Despite this 99 compelling evidence of altered microglial states and increased inflammatory markers in ASD, a 100 fundamental question persists regarding whether this observed neuroinflammation and glial activation 101 represents a primary factor in ASD pathogenesis or is a secondary consequence of other underlying 102 pathological processes. 103 In our previous study, we found that the Shank3∆ e4–22 mouse demonstrated behavioral 104 impairments, mainly affecting motor function, anxiety, and repetitive behaviors, especially in adult mice 105 (Kshetri et al., 2024), which was paralleled by AMPAR-mediated response augmentation. Despite the 106 high expression of Shank3 in CGCs and its impact on AMPAR-mediated EPSC size in late 107 development, how Shank3 affects AMPAR function and the consequences of these specific changes 108 are unclear. In this study, we aimed to investigate how the loss of Shank3 alters glutamtatergic synaptic 109 function at the MF-CGC synapse and how it may affect local supporting cell morphology and CGC 110 density in the cerebellar cortex. 111

& Methods 112 Animals 113 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 7 All animal procedures were performed in accordance with the protocols approved by the 114 Institutional Animal Care and Us e Committee (IACUC) at the Souther n Illinois Univer sity School of 115 Medicine (Protocol #2023-129). All animals were group-housed, provided a normal chow diet, and 116 maintained on the reverse a 12-hour light-dark cycle to facilitate behavioral experiments during their 117 active (dark) phase. Offspring genotypes were determined by Transnetyx (Cordova, TN) using ear 118 punch or tail biopsies. Shank3Δ ex4–22 mice lack exons 4-22 of the Shank3 gene and hence lack 119 expression of all major SHANK3 protein isoforms A to F (Bozdagi et al., 2010; Drapeau et al., 2018; X. 120 Wang et al., 2014). A breeder pair of heterozygous Shank3Δ ex4–22 mice maintained on C57BL/6NJ 121 genetic background (Drapeau et al., 2018) (JAX stra in #: 032169) were acquired from the Jackson 122 Laboratory (Bar Harbor, ME) and bred in-house to generate wildtype (+/+, WT) and homozygous 123 knockout (-/-, KO) mice. All electrophysiology and histological experiments used only adult (4-6 months) 124 WT and KO mice. 125 Cerebellar slice electrophysiology 126 To prepare acute brain slices for recording from CGCs, adult (4-6 months) mice were 127 anesthetized with 3% isoflurane, and cardiac perfusion was done with artificial cerebrospinal fluid 128 (ACSF) containing 1 mM kynurenic acid. Then, the brain was rapidly removed and placed in an ice-cold 129 sucrose slicing solution. The modified form of sucrose slicing solution and recovery solution mentioned 130 in Chabrol et al., 2015 was used in this study. This solution contained the following components (in 131 mM): 2.5 KCl, 0.5 CaCl 2, 4 MgCl 2, 1.25 NaH 2PO4, 24 NaHCO 3, 25 glucose, 230 sucrose, and 1 132 kynurenic acid. The brain was then mounted on a holder and encased in agar and sliced into a 133 parasagittal section (250 μ m) using a Compresstome VF-200 (Precisionary Instruments). The 134 cerebellar slices were then transferred to a recovery solution that included the following components (in 135 mM): 85 NaCl, 2.5 KCl, 0.5 CaCl 2, 4 MgCl2, 1.25 NaH2PO4, 24 NaHCO3, 25 glucose, 75 sucrose, and 1 136 kynurenic acid maintained at 32 °C (Chabrol et al., 2015). After 30 minutes of recovery, cerebellar 137 slices were transferred to room temperature ACSF containing (in mM): 124 NaCl, 26 NaHCO 3, 1 138 NaH2PO4, 2.5 KCl, 2 MgCl2, 10 D-glucose, and 2.5 CaCl2. All solutions were saturated with 95% O2 and 139 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 8 5% CO2, had a pH of 7.3–7.4, and osmolarity of 300–310 mOsm. Slices were transferred to a custom 140 recording chamber on an upright Olympus BX51WI microscope, and CGCs in the internal granule cell 141 layer in lobules 4–5 were visualized with a 60X water-immersion objective using infrared differential 142 interference contrast. ACSF was continuously perfused into the chamber at a rate of 3–5 ml/min, 143 maintained at 32–34 °C. 144 Whole-cell voltage-clamp recordings of visually identified CGCs were made using borosilicate 145 patch pipettes (1.5 mm OD/0.86 mm ID) pulled with a P-1000 micropipette puller (Sutter Instruments) to 146 have a tip resistance of (5–8 M Ω ) when filled with CsCl-based internal solution (E Cl = 0 mV) that 147 contained (in mM): 130 CsCl, 4 NaCl, 0.5 CaCl 2, 10 HEPES, 5 EGTA, 4 Mg-ATP, 0.5 Na-GTP, and 5 148 QX314 with pH adjusted to 7.2–7.3 with CsOH and an osmolarity of 280–290 mOsm (Kaplan et al., 149 2016; Richardson & Rossi, 2017). Whole-cell patch-clamp recordings were acquired with a Multiclamp 150 700B amplifier (Molecular Devices) and sampled at 20 kHz (10 kHz low pass filter) with a Digidata 1440 151 A (Molecular Devices). Following the formation of a gigaseal (> 1G Ω ), the whole-cell configuration was 152 produced by the application of rapid negative pressure to the pipette. Whole-cell membrane properties 153 were determined by applying a 10 mV hyperpolarizing voltage step from the initial holding potential (-60 154 mV) in voltage-clamp mode. Whole cell recordings from CGCs had a series resistance of 20 ± 5 M Ω 155 and recordings with variation in series resistance of greater than 20% throughout the recording were 156 discarded. 157 Miniature excitatory postsynaptic currents (mEPSCs) 158 To isolate miniature excitatory postsynaptic currents (mEPSCs), CGCs were voltage-clamped at 159 -60 mV and the GABA A receptor antagonist, gabazine (SR95531; 10 μ M; Tocris Bioscience, catalog# 160 1262) was present in the ACSF. To isolate mEPSC s, CGCs were voltage-clamped at -70 mV with 161 gabazine and the voltage-gated sodium channel blocker tetrodotoxin (TTX; 0.5 µM; Tocris Bioscience) 162 was included in the bathing ACSF. 163 Data analysis: The acquired raw trace was filtered offline with a 2 kHz lowpass gaussian filter in 164 Clampfit 11.2 (Molecular Devices). Then, inward transient mEPSCs with a fast rise and exponential 165 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 9 decay were analyzed over a 3–5 min period with Easy Electrophysiology Software (v2.6.0) by first-pass 166 automatic threshold detection followed by manual inspection of events. All events from each CGC were 167 used to construct a cumulative distribution histogram for amplitude (1 pA bin size) or inter-event interval 168 (IEI, 100 ms bin size). Event amplitude and IEI were averaged for each cell to generate group averages 169 and for statistical comparisons between genotypes. 170 Evoked excitatory postsynaptic currents (eEPSCs) 171 A bipolar tungsten electrode (0.5 M Ω ; World Precision Instruments) was placed in the central 172 fiber bundle within the cerebellar lamina of lobe 4 & 5 to stimulate the presynaptic mossy-fiber 173 terminals. Stimulation was performed using a range of stimulus intensities from 50 to 300 µA in 50 µA 174 increments. The stimulation frequency was set at 0.2 Hz, with paired-pulse stimulation using 175 interstimulus intervals of 20 ms. To calculate the input-output curve, electrical stimulation intensities 176 ranging from 50-300 µA were used, while CGCs were held at a membrane potential of -60 mV. A 177 stimulation intensity of 100 µA was used to calculate the peak AMPA, peak NMDA, paired-pulse ratio 178 (PPR), and AMPA/NMDA ratio. During presynaptic terminal stimulation, the CGCs were initially 179 maintained at -60 mV in the presence of gabazine (10 µM) to record the AMPAR-mediated response. 180 Subsequently, the holding potential of CGCs changed to +40 mV to record the combined AMPA and 181 NMDAR-mediated response. 182 Data analysis: At least 10-15 non-failure events were selected and digitally averaged for each stimulus 183 recording. The averaged trace was then filtered using a 2 kHz Gaussian filter in Clampfit 11.2 software 184 (Molecular Devices) to obtain an average response for AMPA- and NMDAR-mediated currents. The 185 amplitudes of AMPAR and NMDAR currents were calculated by placing a 1 ms baseline window before 186 the stimulation. The AMPA/NMDA ratio was determined by calculating the ratio of the peak AMPA 187 current at -60 mV and NMDA current value after 15 ms of stimulation at +40 mV (Giza et al., 2010). 188 To calculate AMPAR-mediated decay, the peak-to-baseline decay phase of the resulting current 189 trace was fitted by the double exponential function: I = A 1e –t/ τ 1 + A 2e –t/ τ 2 and the weighted decay 190 constant was calculated using the following formula where, τ W = weighted decay constant, τ 1 = slow 191 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 10 decay time constant, τ 2 = fast decay time constant, A 1 = slow amplitude component, and A 2 = fast 192 amplitude component (Richardson et al., 2013; Schofield & Huguenard, 2007): 193 /g2028/g3050/g3404/g4666 /g2028/g2869/g1827/g2869/g3397/g2028/g2870/g1827/g2870)/(/g1827/g2869/g3397/g1827/g2870/g4667 194 Rubi-Glutamate uncaging 195 For the glutamate photo-uncaging experiment, we used a caged glutamate compound called 196 Ruthenium-bipyridine-trimethylphosphine-Glutamate (RuBi-Glutamate). This compound can be excited 197 by visible wavelengths and releases glutamate upon one- or two-photon excitation (Fino et al., 2009). 198 The CGC was maintained at a holding potential of -70 mV, and photo-evoked EPSCs were recorded in 199 the whole-cell configuration. Focal stimulation of CGCs was achieved by applying 470 nm blue LED 200 light within a 50 × 50 µm² gridded area over the cerebellar cortex. To facilitate the photo-release of 201 glutamate, the Polygon 1000 pattern Illuminator (Mightex) was used. This experiment was conducted in 202 the presence of 50 µM RuBi-Glutamate, chosen as the optimal concentration based on reliable 203 response using the stimulation protocol. Glutamate photo-release was achieved using a light 204 stimulation duration of 50 ms at a frequency of 0.05 Hz. To record the combined AMPA + NMDA 205 current evoked by photo-stimulation, the CGC was held at -70 mV, and the cerebellum slice was 206 perfused with ACSF containing RuBi-glutamate, 10 µM gabazine, and 0.5 µM tetrodotoxin (TTX; Tocris 207 Bioscience, catalog # 1069) in absence of Mg 2+. NMDA current was pharmacologically dissected by 208 subsequent supplementation with 10 µM NBQX in the bath. The NMDA current was further validated by 209 subsequent bath application of 50 µM AP5 and 50 µM 7-Chlorokynurenic acid. After the drugs were 210 introduced into the bath, at least 2-3 min were allowed for proper diffusion, and then the current 211 responses were recorded. 212 Data analysis: At least five traces with a plateau response were selected, digitally averaged, and 213 filtered using a 200 Hz Gaussian filter in Clampfit 11.2 software (Molecular Devices) to obtain an 214 average response for each current type, i.e., AMPA + NMDA, AMPA, and NMDA. The peak amplitude 215 of these currents was calculated by placing a 1 ms baseline window before the light was turned on for 216 uncaging. The AMPA trace was obtained by digitally subtracting the NMDA trace from the AMPA + 217 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 11 NMDA trace in Clampfit 11.2. The AMPA/NMDA ratio was calculated by dividing the AMPA-mediated 218 peak amplitude by the NMDA-mediated peak amplitude. The current density was calculated by dividing 219 the peak amplitude currents by the cell capacitance. 220 Inward rectification of eEPSCs 221 In the same eEPSC recording mode, a patch pipe tte filled with a CsCl- based internal solution 222 supplemented with 100 µM spermine (Tocris Bioscience, catalog# 958) was used to perform the inward 223 rectification experiment. First, CGC was held at -60 mV, and single-pulse stimulation of 100 µA at a 224 frequency of 0.2 Hz was applied in the presence of the pharmacological blockers 10 µM gabazine to 225 record AMAPR-mediated eEPSC. Then, the holding pot ential was changed from -60 to +60 mV with 226 successive increments of 20 mV, and the bath was supplemented with additional NMDAR blockers: D-227 AP5 (50 µM; Tocris Bioscience, catalog # 0106) and 7-Chlorokynurenic acid (50 µM; Tocris Bioscience, 228 catalog# 3697) to record the AMPAR peak current at positive potentials. 229 Data analysis: At least 10-15 traces were selected and digitally averaged for each stimulus recording, 230 then the averaged trace was then filtered using a 2 kHz Gaussian filter in Clampfit 11.2 software 231 (Molecular Devices) to obtain an average response for AMPAR-mediated currents. The amplitude of 232 AMPAR currents was calculated by placing a 1 ms baseline window before the stimulation. A 233 normalized current-voltage (I-V) curve was generated by normalizing the average peak AMPAR current 234 values at each holding potential to the maximum response at V h = -60 mV. The rectification index was 235 calculated by taking the ratio of the peak AMPAR current at +60 mV to the AMPAR current at -60 mV. 236 Pharmacological assessment of IEM-1460 sensitivity 237 In the same eEPSC experimental setup usi ng CsCl-based internal solution, the AMPAR-238 mediated response was recorded at -60 mV after stimulating with single-pulse stimulation of 100 µA at 239 a frequency of 0.2 Hz in the presence of gabazine (10 μ M; Tocris Bioscience, catalog# 1262) followed 240 by the supplemental application of selective calcium-permeable AMPAR blocker IEM-1460 (100 µM; 241 Tocris Bioscience, catalog# 1636). The baseline was recorded for the first 5 min in the presence of 242 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 12 gabazine, which was followed by the additional supplementation of IEM-1460 and recording for an 243 additional 15 min. 244 Data analysis: The average value of 12 traces that were recorded every minute were selected to obtain 245 the average peak AMPAR-mediated response. The amplitude of AMPAR currents was calculated by 246 placing a 1 ms baseline window before the stimulation. A normalized current-voltage (I-V) curve was 247 generated by normalizing the average peak AMPAR current values at each holding potential to the 248 maximum response at V h = -60 mV. The average responses each minute after the application of IEM-249 1460 were normalized to their respective average value of their baseline recorded for first 5 min. The 250 percentage of baseline response was calculated by calculating the average percentage of response in 251 last 3 min in IEM-1460 compared to the average baseline response. 252 Immunofluorescence 253 For all immunofluorescence studies, mice were anesthetized using isoflurane (3-5%) and then 254 transcardially perfused with 1X phosphate-buffered sa line (PBS), followed by 4% formaldehyde diluted 255 in 1X PBS. Then, brains were removed and post-fi xed for 48–72 h in 4% formaldehyde, and placed in 256 sucrose solution gradients until they sank to the bottom. Following the sucrose gradient, brain samples 257 were snap-frozen in dry ice before embedded in optimal cutting temperature (OCT) (Tissue-Tek; 258 Sakura Finetek, CA, catalog# 4583). Sagittal whole-brain slices (30 μ m thick) containing cerebellar 259 vermis were prepared on a cryostat (Leica CM 1850) at -25 °C. Subsequently, the brain slices were 260 washed in 1X PBS and then stored at -20 °C in a cryoprotecting solution until the start of 261 immunofluorescence studies. All immunofluorescence staining procedures were performed using the 262 free-floating method. 263 IBA1 staining 264 The brain slices were placed in sodium borohyd ride (1 mg/ml in PBS) for 30 min, replacing the 265 fresh solution every 10 min at room temperature. The brain slices were then permeabilized in 0.25% 266 Triton X-100 in PBS for 30 min, followed by blocking in IHC/ICC Blocking Buffer (eBiosciences) 267 containing 0.25% Triton X-100 for 1 h. Subsequently, the slices were incubated in primary antibody 268 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 13 rabbit IgG anti-IBA1 (1:500; Wako, catalog# 019-19741) overnight at 4 °C. Slices were then washed 269 with PBS and incubated in secondary antibody goat Ig G anti-rabbit Alexa Fluor 488 (1:1000; Thermo 270 Fisher Scientific, catalog# A11008). The nuclei were counterstained with Hoechst 33342 (1:2000, 271 Thermo Fisher Scientific, catalog# H3570). Finally, tissue sections were washed and transferred to 272 glass slides and mounted with Prolong Gold (Invitrogen). Imaging of microglia in the inner granule cell 273 layer of cerebellar lobe 4 &5 from the stained brain slices was performed by capturing Z-stack images 274 containing 7 optical slices at an interval of 1 µm, using a 40X oil immersion objective (numerical 275 aperture 1.3) on a confocal microscopy system (Zeiss LSM800) at a resolution of 1,024 × 1,024 pixels. 276 Data analysis: The orthogonal projections of Z-stack images were generated using ZEN Lite software (v 277 3.7 edition; Zeiss). The total surface area covered by individual microglia was determined using Imaris 278 software (v10.2). The total number of microglia was assessed by counting the number of IBA1-positive 279 microglial cell bodies in each orthogonal projection image. Then, the proportion of microglia of different 280 surface area measurements was quantified. Total 5-6 images/genotype were analyzed from 3-5 281 animals/genotype. 282 Statistical analysis 283 For analyzing electrophysiology and immunofluorescence data, the normality of data was tested 284 using Shapiro-Wilk test. An unpaired t-test was used to compare all the parameters between WT and 285 KO mice within age groups using GraphPad Prism 10.4.2. In all experiments, data values are reported 286 as mean ± standard error (SEM) with individual markers representing the value for each observation, 287 which is the cell (n) for electrophysiology assays, and the image (n) for confocal analysis. All statistical 288 comparisons with p < 0.05 were considered significant. Statistical results are provided in 289 Supplementary Table 1. 290

291 Loss of Shank3 alters postsynaptic AMPAR function at MF-CGC synapse 292 To determine the cause of augmented MF-CGC synaptic function we identified previously 293 (Kshetri et al., 2024), we investigated stochastic spontaneous neurotransmitter release and changes in 294 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 14 synaptic function at the MF-CGC synapse due to the loss of Shank3. First, quantal glutamatergic 295 miniature excitatory postsynaptic currents (mEPSCs) resulting from single vesicle release from 296 presynaptic mossy fiber terminals were recorded from CGCs at a holding potential of -70 mV in the 297 presence of 10 µM gabazine and 0.5 µM TTX ( Figure 1A ). Since TTX blocks voltage-gated sodium 298 channels, which are essential for generating action potentials and coordinated multivesicular release, 299 mEPSC recordings provide an assessment of synapt ic responses to indivi dual vesicles/single quanta 300 by eliminating the possibility of multivesicular release. 301 We also assessed tonic inhibition current in CGCs by applying 10 µM gabazine, but we did not 302 find a significant difference between WT and KO mice (t(19) = 1.179, p = 0.25). Similar to our previous 303 observation in sEPSCs (Kshetri et al., 2024), cumulative distribution of mEPSC amplitudes was shifted 304 toward higher values, indicating a change in postsynaptic AMPAR function in CGCs ( Figure 1B). The 305 average peak amplitude of mEPSCs wa s significantly larger in the Shank3 KO group compared to the 306 WT group ( Figure 1C). In contrast, there was no significant difference in the cumulative distribution or 307 average inter-event interval (IEI) between WT and Shank3 KO groups (Figure 1D, E), suggesting that 308 the probability of spontaneous presynaptic glutamate release from the presynaptic site is not affected 309 by the loss of Shank3. These findings from miniature recordings suggest that Shank3 plays a critical 310 role in maintaining postsynaptic glutamatergic receptor function at the MF-CGC synapse. 311 Deficiency of Shank3 alters the biophysical properties of AMPAR and does not affect the 312 presynaptic function 313 Next, we investigated the change in MF-CGC synapse in Shank3 KO mice by electrically 314 stimulating the presynaptic term inal and recording the evoked EPSCs (eEPSCs) from CGCs in a 315 whole-cell voltage clamp configuration at -60 mV in the presence of gabazine ( Figure 2A ). As we 316 increased the stimulus intensity from 50 to 300 µA, we observed a modest increase in the eEPSC 317 amplitude up to 200 µA stimulation in both WT and Shank3 KO, but with with high variability in the 318 amplitude between individual GCGs (Figure 2B, C). The mean amplitude of eEPSC did not significantly 319 differ between genotypes at any given stimulus strength ( Figure 2C). However, the overall distribution 320 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 15 of individual eEPSC events generated from all CGCs indicated a shift toward higher amplitude eEPSCs 321 in the Shank3 KO compared to the WT group ( Figure 2D, E ), similar to what we observed for CGC 322 sEPSCs previously (Kshetri et al., 2024). 323 After stimulating the CGCs at -60 mV to elicit the AMPA response, the holding current was 324 adjusted to +40 mV to record the composite EPSC, with AMPARs generating the fast component and 325 NMDARs producing the slower component. Then, the NMDAR-mediated currents were dissected 326 electrophysiologically by measuring the current value at 15 ms after the stimulus intensity at +40 mV 327 (Figure 3A). The average amplitudes of both AMPA- and NMDA-responses at 100 µA stimulation were 328 similar between WT and Shank3 KO (Figure 3B, C). Next, the calculated AMPA/NMDA ratio for CGCs 329 was also similar between the WT and Shank3 KO groups ( Figure 3D), which suggests that the basal 330 short-term plasticity at the MF-CGC synapse may not be affected by the loss of Shank3. 331 To determine the role of Shank3 in the presynaptic function at the MF-CGC synapse, we 332 recorded the paired-pulse ratio (PPR). Our findings showed that the PPR was similar in WT and 333 Shank3 KO groups (Figure 3E, F). The absence of change in PPR and frequency of mEPSC ( Figure 334 1D, E ) suggest that the absence of Shank3 does not alter the presynaptic function of the MF-CGC 335 synapse. Interestingly, the evoked AMPAR-mediated response showed significantly faster decay in 336 Shank3 KO mice than in WT mice ( Figure 3G, H ), indicating the possible change in biophysical 337 properties of AMPARs due to the change in subunit composition or the change in the number of 338 AMPARs at the postsynaptic site. 339 Increase in AMPAR-mediated response to uncaged glutamate in Shank3 KO mice 340 To determine the AMPAR function irrespective of their presence at the synapse, we 341 supplemented Rubi-Glutamate in the bath and made glutamate available through photolytic cleavage, 342 then recorded the inward currents from CGCs ( Figure 4A ). Glutamate uncaging allows for precise 343 spatial and temporal control of glutamate release, potentially revealing ionotropic glutamate receptor 344 properties that may not be apparent in miniature or evoked EPSC recordings. Brief exposure to blue 345 LED light in the presence of 10 µM gabazine and 0.5 µM TTX while holding a CGC at -70 mV showed a 346 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 16 combined AMPA and NMDA response ( Figure 4B). Furthermore, the NMDA and AMPA components 347 were dissected from the combined response using a pharmacological blocker (10 µM NBQX) and 348 digital subtraction (Figure 4B). We observed a significant increase in the combined AMPA- and NMDA-349 mediated response in Shank3 KO mice compared to WT ( Figure 4C). Moreover, upon analyzing the 350 individual AMPA and NMDA currents, we found a significant increase in AMPA current amplitude in the 351 Shank3 KO group (Figure 4D). In contrast, NMDA current amplitudes were similar in both the WT and 352 Shank3 KO groups ( Figure 4E ). Consequently, the Shank3 KO group showed an increased 353 AMPA/NMDA ratio compared to the WT ( Figure 4F). Next, the calculation of current densities for all 354 three currents (AMPA + NMDA, AMPA, and NMDA) demonstrated a significant increase exclusively in 355 AMPAR current density in Shank3 KO mice ( Figure 4H ). Overall, these findings suggest that the 356 upregulation of AMPAR function at the MF-CGC synapse in Shank3 KO mice may arise from either an 357 increase in the number of AMPARs at the postsynaptic site or a change in AMPAR subunit 358 composition. 359 Loss of Shank3 shows inward rectification and the presence of an increased level of CP-AMPAR 360 in CGCs 361 AMPAR kinetics are largely determined by the presence or absence of the GluA2 subunit as 362 GluA2-lacking calcium-permeable (CP) AMPARs typically exhibit faster decay kinetics compared to 363 GluA2-containing calcium-impermeable (CI) AMPARs (Anggono & Huganir, 2012; Chojnacka et al., 364 2023; Cull-Candy & Farrant, 2021; Dolgacheva et al., 2020; C. Guo & Ma, 2021; Henley & Wilkinson, 365 2016; Hollmann et al., 1991; Traynelis et al., 2010). To confirm the possibility of the presence of CP-366 AMPARs in CGCs in Shank3 KO mice, we supplemented spermine in the patch pipette and recorded 367 the eEPSC response at the same setup used for eEPSC recording ( Figure 5A). The current-voltage 368 curve (I-V curve) showed significant inward rectification at higher potentials (+20, +40, and +60 mV), 369 and a decrease in rectification index (RI) value in Shank3 KO cells than the WT cells, which indicates a 370 relative incrase in the presence of CP-AMPARs in CGCs in Shank3 KO mice (Figure 5B, C, D). 371 Next, we further determined the presence of increased CP-AMPARs by applying IEM-1460, a 372 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 17 CP-AMPAR blocker (Cull-Candy & Farrant, 2021; da Silva & Schröder, 2023; Suyama et al., 2017; 373 Twomey et al., 2018). After the application of IEM-1460, AMPAR-mediated currents decreased relative 374 to the baseline in WT neurons, suggesting the presence of CP-AMPARs in CGCs ( Figure 5E-G ). 375 These findings align with previous studies that have shown the presence of CP-AMPARs in CGCs 376 (Hack & Balázs, 1995; Savidge & Bristow, 1997). When comparing the degree of IEM-1460-induced 377 reduction in the AMPA response between WT and Shank3 KO neurons, as predicted, we observed a 378 significantly greater decrease in the AMPA response in Shank3 KO neurons compared to WT ( Figure 379 5E-G). The enhanced sensitivity to IEM-1460 in Shank3 KO further supports the conclusion that the 380 MF-CGC synapse has an elevated proportion of GluA2-lacking CP-AMPARs. 381 Since our electrophysiology data suggest an increase in the level of CP-AMPARs in the CGCs 382 of the Shank3 KO mice, which may lead to an increase in excitability, resulting in excitotoxicity and 383 neuronal loss. Therefore, to determine the possibility of neuronal loss due to the loss of Shank3, we 384 stained cerebellar slices with a neuronal marker, NeuN and counterstained with a nuclear marker, 385 Hoechst (Supplementary Fig. 2A-F). Since most of the neurons present in the cerebellar cortex are 386 CGCs, counting the NeuN-positive cells will deter mine if neuronal loss occurred due to the loss of 387 Shank3. Following the quantification, we did not find any difference in the number of NeuN-positive 388 cells when comparing the WT and Shank3 KO groups at the adult time point (4-6 months) 389 (Supplementary Fig. 2G ). This result suggests that the loss of Shank3 mainly affects synaptic 390 transmission at the MF-CGC synapse but does not lead to excitotoxicity. 391 Reduced IBA1-stained microglial surface area in Shank3 KO mice 392 We investigated changes in microglial morphology and number in Shank3 K O m i c e u s i n g 393 immunostaining with ionized calcium binding adaptor molecule 1 (IBA1) ( Figure 6A, B ). Following 394 staining, we quantified the surface area covered by IBA1-stained microglia ( Figure 6A′ , B′ ). The total 395 surface area covered by IBA1-stained microglia was significantly reduced in the Shank3 KO group 396 relative to the WT controls ( Figure 6C ). Furthermore, the distribution plot of the surface area 397 measurement showed that the majority of microglia in the Shank3 KO group covered a smaller surface 398 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 18 area relative to the WT group ( Figure 6E ), indicating a shift toward a less ramified morphology. 399 However, the number of microglia in the cerebellar cortex was similar between WT and Shank3 KO 400 mice (Figure 6D). 401 To determine the presence of astrocyte reactivity in germline Shank3 KO mice, we also 402 performed immunostaining using the astrocytic marker Glial Fibrillary Acidic Protein (GFAP) along with 403 SRY-box transcription factor 9 (SOX9), a nuclear marker for astrocytes ( Supplementary Fig. 1A-D ). 404 The percentage area covered by GFAP fluorescence was comparable between WT and Shank3 KO 405 mice (Supplementary Fig. 1E). Similarly, no significant difference was observed in GFAP fluorescence 406 intensity between the two groups ( Supplementary Fig. 1F). Additionally, staining with SOX9 revealed 407 no difference in the number of SOX9-positive astrocytes between the WT and Shank3 KO groups 408 (Supplementary Fig. 1G ). Together, these results indicate that the absence of Shank3 does not 409 induce astrocyte reactivity or alter astrocyte number in the cerebellar cortex; however, the presence of 410 less ramified microglia suggests a possible shift toward an activated microglial state. 411

412 In this study, we investigated the role of Shank3 in regulating synaptic function at the MF-CGC 413 synapse, a critical part of the cerebellar circuit that integrates sensory-motor information in the 414 cerebellar cortex. Additionally, we also studied the alteration in astrocyte reactivity and microglia 415 morphology in Shank3 KO mice. Our electrophysiological and immunofluorescence data demonstrate 416 the role of Shank3 in maintaining AMPAR function and subunit composition at the postsynaptic site 417 while indicating that the presynaptic release property remains unaffected by its absence. Additionally, 418 less ramified microglia were observed in Shank3 KO mice. This change in synaptic function, altering 419 the excitatory neurotransmission in CGCs, and altered microglia sheds light on the potential 420 mechanism of cerebellum involvement in the development of ASD-like phenotypes in Shank3 KO mice. 421 One of the main findings of our electrophysiology experiments is the enhancement of 422 postsynaptic AMPAR-mediated response in CGC from Shank3 KO mice compared to WT controls. 423 First, we observed the enhancement of amplitude but no change in the frequency of mEPSC in adult 424 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 19 Shank3Δ ex4-22 mice (Figure 1A-E). However, studies performed across various brain regions and using 425 different versions of Shank3 mutant mice have shown incons istent results regarding mEPSC 426 parameters. For instance, Shank3B KO mice showed a reduction in both frequency and amplitude of 427 mEPSC in dorsolateral striatal medium spiny neurons (MSNs) (Peça et al., 2011) and anterior cingulate 428 cortex (ACC) pyramidal neurons (B. Guo et al., 2019). However, Guo et al. reported a decrease in 429 mEPSC amplitude without a change in frequency in thalamocortical cells of Shank3B KO mice (B. Guo 430 et al., 2024). Furthermore, studies using Shank3∆ 14–16 mouse model reported an increase in frequency 431 but no change in amplitude in mPFC layer 2/3 neurons (Yoo et al., 2019), whereas both frequency and 432 amplitude were reduced in the dorsolateral striatum (Yoo et al., 2018). Additionally, the mouse model of 433 frameshift mutation (InsG3680 mutation) in Shank3 showed a significant increase in mEPSC amplitude 434 but no change in frequency in the dorsal striatum (Zhou, Kaiser, Monteiro, Zhang, Van der Goes, et al., 435 2016). Similarly, Shank3 knockdown in the anteromedial bed nucleus of the stria terminalis (BNST) 436 neurons resulted in increased mEPSC amplitude and decreased frequency (Contestabile et al., 2023). 437 These discrepancies suggest the brain-region-specific and/or mutation-specific alterations of the 438 Shank3 isoform across the brain regions. Among these, CGCs may show a unique response to the loss 439 of Shank3 compared to other brain regions. 440 Additionally, we found an increase in AMPAR response and AMPA/NMDA ratio in the uncaging 441 experiment, which further supports the idea of enhancement of AMPAR function ( Figure 4D, F), either 442 due to an increase in expression or the biophysical properties of AMPARs. Together, these increases in 443 AMPAR-response observed in CGCs in our study could arise from several potential factors. Shank3 is 444 known to be involved in the trafficking and synaptic localization of AMPARs (H. T. T. Ha et al., 2018; 445 Yang et al., 2024). Therefore, loss of Shank3 may lead to the alteration of AMPAR subunit composition, 446 potentially incorporating subunits that show higher conductance. Another possibility of an increase in 447 mEPSC amplitude could be an increased insertion of functional AMPARs at the postsynaptic 448 membrane (Chen et al., 1999; Glasgow et al., 2019), however, some evidence from other brain regions 449 suggests Shank3 deficiency may reduce AMPAR function (Xu et al., 2024; Xue et al., 2024). Beyond 450 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 20 AMPAR subunit composition, their f unction is also modulated by int egral auxiliary subunits, including 451 Transmembrane AMPAR Regulatory Proteins (TARPs), Cornichon homologs (CNIHs), Cysteine-knot 452 AMPAR Modulating Proteins (CKAMPs), and Germ cell-specific gene 1-like protein (GSG1L) (Anggono 453 & Huganir, 2012; Bissen et al., 2019; Cull-Candy & Farrant, 2021; Qneibi et al., 2024). These auxiliary 454 subunits shape AMPAR biophysical properties by dynamically tuning channel gating kinetics, single-455 channel conductance, and sensitivity to polyamine block, thereby fundamentally determining the 456 characteristics of synaptic transmission and plasticity (Coombs et al., 2023; Cull-Candy & Farrant, 457 2021; Jacobi et al., 2021; Kamalova et al., 2020; Miguez-Cabello et al., 2025; Qneibi et al., 2024; 458 Rozov et al., 2018). Disruption of Shank3 could indirectly affect the interaction between AMPARs and 459 their auxiliary subunits, potentially enhancing AMPAR function. While stable mEPSC amplitudes were 460 reported during LTP at the mossy fiber-granule cell synapse (Sola et al., 2004), a study using Islet 461 Brain-2 (IB2, a gene commonly deleted in PMS) KO found an increase in quantum size in CGCs during 462 LTP (Soda et al., 2019). Therefore, if the LTP mechanism is altered at this synapse in Shank3 KO 463 mice, it could account for the increased postsynaptic receptor function observed in our study. 464 The absence of change in mEPSC frequency in our study suggests that the presynaptic 465 neurotransmitter release machinery remains intact at the MF-CGC synapse (Figure 1D, E). This finding 466 is consistent with some studies performed in different brain regions with Shank3 mutations mentioned 467 above in the discussion (B. Guo et al., 2024; Jaramillo et al., 2016; X. Wang et al., 2011; Zhou, Kaiser, 468 Monteiro, Zhang, Van der Goes, et al., 2016) (Zhou, Kaiser, Monteiro, Zhang, Van der Goes, et al., 469 2016), suggesting that Shank3 deficiency does not universally impact the probability of spontaneous 470 presynaptic glutamate release. Conversely, a number of studies have reported either an increase (Yoo 471 et al., 2019) or decrease (B. Guo et al., 2019; Kouser et al., 2013; Peça et al., 2011; W. Wang et al., 472 2017; Yoo et al., 2018) in mEPSC frequency, indicating that the presynaptic effects of Shank3 loss may 473 depend on the brain region, the specific Shank3 isoform affected, and the developmental stage. 474 We observed a shift toward higher amplitude AMPAR-mediated eEPSCs ( Figure 2D, E) without 475 a corresponding change in AMPA/NMDA ratio in the Shank3 KO group (Figure 3D). The stability in the 476 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 21 AMPA/NMDA ratio aligns with some previous findings (Peça et al., 2011; Zhou, Kaiser, Monteiro, 477 Zhang, Van der Goes, et al., 2016) but contrasts with the others. For instance, a study showed a 478 reduction in both the eEPSC input-output curve relationship and the AMPA/NMDA ratio in ACC 479 pyramidal neurons of Shank3 InsG3680 mutant mice (B. Guo et al., 2019). Variability regarding 480 changes in the AMPA/NMDA ratio has been documented in the literature for Shank3 KO models. 481 Consistent with our results, some studies have observed no change in the AMPA/NMDA ratio (Peça et 482 al., 2011; Zhou, Kaiser, Monteiro, Zhang, Van der Goes, et al., 2016), whereas others have shown a 483 decrease in the NMDA/AMPA ratio (Jaramillo et al., 2016; Kouser et al., 2013; Speed et al., 2015). 484 Interestingly, Jaramillo et al. (2016) showed no alterations in the NMDA/AMPA ratio in the hippocampus 485 of Shank3Δ e4-9 mutant mice; however, the same study observed a significant reduction in the 486 NMDA/AMPA ratio in the striatum (Jaramillo et al., 2016), which again highlights that the effects of 487 Shank3 mutations on synaptic function can be brain region-specific. In this context, our findings 488 suggest that the cerebellum, with its distinct developmental timeline and circuit organization, may 489 exhibit unique synaptic responses to Shank3 deficiency. 490 Multiple lines of evidence from our study indicate that the loss of Shank3 may not alter the 491 probability of neurotransmitter release at the MF-CGC synapse. We found that both mEPSC frequency 492 (Figure 1D, E ) and PPR ( Figure 3F) were similar between WT and Shank3 KO, suggesting that the 493 synaptic deficits observed in our model may be predominantly postsynaptic in origin. These findings are 494 consistent with several previous studies that reported no significant changes in PPR in various Shank3-495 deficient mouse models (B. Guo et al., 2019; Jaramillo et al., 2016, 2017; Kouser et al., 2013; Moutin et 496 al., 2021; Peça et al., 2011; Speed et al., 2015; X. Wang et al., 2011; Zhou, Kaiser, Monteiro, Zhang, 497 Van der Goes, et al., 2016). However, Wang et al. (2017) noted a significant increase in PPR 498 exclusively in the D2 MSNs (striato-pallidal indirect pathway) of the Shank3B KO mice, indicating a 499 reduced presynaptic release probabilit y specifically at the glutamat ergic terminals innervating D2 500 MSNs, whereas PPR in D1 MSNs (striato-nigral direct pathway) was similar between WT and Shank3 501 KO (W. Wang et al., 2017). Although most studies collectively suggest that Shank3 loss does not 502 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 22 generally alter the presynaptic releas e probability, it is possible that Shank3 might form distinct 503 signaling complexes or express specific isoforms with unique functions in different cell types or brain 504 regions (W. Wang et al., 2017). 505 Our observation of a fast decay in AMPAR-mediated evoked EPSCs in CGCs (Figure 3G, H) of 506 Shank3 KO mice suggests a possible change in subunit composition in AMPARs, offering a novel 507 perspective on the role of Shank3 in regulating glutamatergic transmission within the cerebellum. The 508 functional diversity of AMPARs is mainly determined by their subunit composition (Anggono & Huganir, 509 2012; Chojnacka et al., 2023; Cull-Candy & Farrant, 2021; Dolgacheva et al., 2020; C. Guo & Ma, 510 2021; Henley & Wilkinson, 2016; Hollmann et al., 1991; Traynelis et al., 2010). The absence of the 511 GluA2 subunit has a profound effe ct by causing calcium perme ability, higher single-channel 512 conductance, faster kinetics, and inward rectification in its current-voltage relationship due to voltage-513 dependent block by intracellular polyamines (Burnashev et al., 1992; Cull-Candy & Farrant, 2021). In 514 addition to faster decay kinetics, we have noticed inward rectification and a comparatively high CP-515 AMPAR response in Shank3 KO, suggesting Shank3 loss may affect the homeostatic balance of 516 surface expression of CP- and CI-AMPARs in the CGCs. The change in proportion of inwardly 517 rectifying CP-AMPAR could confound the measurement of amplitude at higher potentials (Kauer & 518 Malenka, 2007), which might explain the similar AMPA/NMDA ratio in eEPSC response in both WT 519 and Shank3 KO group. 520 In the majority of neurons across various brain regions, the CP-AMPARs are rapidly replaced 521 with CI-AMPARs after P14 in rodents (Diering & Huganir, 2018). Concurrently, the expression of all 522 Shank3 isoforms significantly increases after P7, reaching a peak around four weeks postnatally (X. 523 Wang et al., 2014). A significant mechanism underlying synaptic plasticity involves activity-dependent 524 switch from CP-AMPARs to CI-AMPARs (S. J. Liu & Savtchouk, 2012; S. J. Liu & Zukin, 2007). 525 Synaptic CP-AMPARs show a unique self-regulating feature, whereby repetitive activation of these 526 receptors triggers their replacement by GluA2-containing CI-AMPARs. This phenomenon was initially 527 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 23 observed and well-characterized at the parallel fiber to stellate cell synapse in the cerebellum (S.-Q. J. 528 Liu & Cull-Candy, 2000). 529 Shank3, along with other postsynaptic proteins and zinc, has been shown to interact and assist 530 in the trafficking of GluA2-containing AMPARs at the synaptic membrane (Bariselli et al., 2016; H. T. T. 531 Ha et al., 2018; Sheng & Kim, 2000). This is mediated through the interaction of Shank3 with Glutamate 532 Receptor Interacting Proteins (GRIPs) through its SH3 domain (Bozdagi et al., 2010; Manning et al., 533 2024; Sala et al., 2015; Sheng & Kim, 2000). Furthermore, PDZ domain-containing proteins like GRIPs 534 (Dong et al., 1997, 1999; S. J. Liu & Cull-Candy, 2005) and protein interacting with C kinase (PICK1) 535 (Xia et al., 1999), interact with the C-terminal of the GluA2 subunit. This molecular bridge, formed 536 through the interactions among Shank3, PDZ-domain containing proteins (GRIPs and PICK1), and 537 GluA2, is posited to be a crucial mechanism for the appropriate synaptic localization and stabilization of 538 GluA2-containing CI-AMPARs. The molecular mechanisms underlying the switch from CP- to CI-539 AMPARs involve an increase in intracellular Ca 2+ through CP-AMPARs. Elevated intracellular Ca 2+ 540 subsequently activates protein kinase C (PKC), which in turn activates PICK1 (S. J. Liu & Savtchouk, 541 2012; S. J. Liu & Zukin, 2007). Concurrently, PKC may phosphorylate GluA3, disrupting the interaction 542 between GluA2-lacking AMPARs and GRIP, leading to the loss of these CP-AMPARs from the synapse 543 (Gardner et al., 2005; S. J. Liu & Cull-Candy, 2005). Therefore, in the case of Shank3 KO, the loss of 544 Shank3 may interrupt the AMPAR trafficking, affecting both the insertion of CI-AMPAR into the 545 membrane and the internalization of CP-AMPAR. Thus, the synaptic level of CP-AMPAR may remain 546 high, resembling that of the early postnatal stage, potentially impairing the maturation of MF-CGC 547 synapses. Additionally, the elevated intracellular Ca 2+ level could trigger intracellular signaling 548 cascades, altering intrinsic neuronal excitability and influencing cerebellar circuit output (S. J. Liu & 549 Savtchouk, 2012; S. J. Liu & Zukin, 2007). Therefore, the dysregulation of CP- and CI-AMPARs 550 dynamics at MF-CGC synapse in Shank3 KO mice may impair both synaptic transmission and neuronal 551 excitability output, ultimately contributing to cerebellar-related behavioral deficits. 552 Our observation of reduced microglial processes, suggestive of an activated state, provides 553 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 24 preliminary evidence for potential microglial involvement in the cerebellar pathology of Shank3 KO 554 mice. In contrast, a study using Shank3 heterozygous mice reported no differences in IBA1-positive 555 microglia density or cell body size in the hippocampus, mPFC, or striatum (Cope et al., 2016), 556 suggesting that gross IBA1 morphology and cell body size might not be consistently altered in Shank3 557 deficiency across all brain regions or developmental time points. Microglial morphology has been used 558 to indicate their functional state. The highly ramified cells are thought to represent a 'resting’ or 559 homeostatic phenotype, and cells exhibiting retracted processes and enlarged soma are considered 560 ‘activated’ (Davis et al., 2017; Morgan et al., 2010; Savage et al., 2019). However, recent perspectives 561 highlight heterogeneity of microglial states and emphasize that morphology analysis itself does not 562 provide insights into their functional capabilities (J. Kim et al., 2023; Paolicelli et al., 2022). Therefore, 563 our findings of IBA1 morphology in Shank3 KO microglia should be carefully interpreted. 564

565 Several studies have explored the role of Shank3 in various forebrain regions, but its function in 566 the cerebellum remains underexplored, despite its involvement in both motor and multiple non-motor 567 processes. By using various electrophysiological methods, we have found that the loss of Shank3 568 selectively disrupts the postsynaptic receptor function without impacting the presynaptic glutamate 569 release at the MF-CGC synapse. Furthermore, we have found that the loss of Shank3 results in 570 upregulation of AMPAR-mediated response and a shift towards a high level of CP-AMPARs at the MF-571 CGC synapse. These findings indicate an impairment in synapse maturation at the MF-CGC synapse, 572 potentially leading to a change in excitability and synaptic plasticity within the cerebellar cortex. 573 Moreover, the less ramified microglial processes in Shank3 KO suggested the possibility of activated 574 microglia in the cerebellar cortex. 575 Given that the present study was conducted using the germline Shank3 KO mouse model, 576 future studies using the most appropriate conditional CGC-specific Shank3 KO model may address the 577 cell-specific and developmental role of Shank3 in CGCs. Additionally, targeted pharmacological or 578 genetic interventions to restore CI-AMPAR expression may provide insights into the therapeutic 579 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 25 potential for addressing AMPAR dysfunction in ASD. 580 Together, our findings provide novel insights into the possible role of Shank3 in regulating the 581 balance between the dynamic regulation of CP- and CI-AMPARs at the MF-CGC synapse. This reveals 582 a possible novel mechanism by which the loss of Shank3 may disrupt this homeostatic balance, altering 583 the cerebellar circuitry. This study contributes significantly to our understanding of the role of Shank3 in 584 the cerebellar cortex and the neurobiological underpinnings of Shank3-associated cerebellar-related 585 behavioral deficits. 586 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 26 Declarations 587 Ethics approval - All procedures involving animals were performed in accordance with protocols 588 approved by the Institutional Animal Care and Use Committee at Southern Illinois University – School 589 of Medicine. 590 591 Consent for publication - Not applicable. 592 593 Availability of data and materials - The datasets used and/or analyzed during the current study are 594 available from the corresponding author on reasonable request with statistical analysis results included 595 with this published article’s supplementary information files. 596 597 Competing interests - The authors declare that they have no competing interests. 598 599 Funding - This work was supported by a National Institutes of Mental Health grant (R01MH129749) to 600 BDR. 601 602 Authors’ contributions – RK drafted the manuscript with BDR. RK collected and analyzed the 603 electrophysiology and immunofluorescence experiments data. BDR performed final review of data 604 analysis and prepared data figures with RK. BDR conceived of the project with RK and BDR oversaw 605 the data acquisition and generation of the final manuscript. 606 607

- We would like to thank the National Institute of Mental Health for providing 608 funding for this project. 609 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 27

610 Achilly, N. P., He, L. J., Kim, O. A., Ohmae, S., Wojaczynski, G. J., Lin, T., Sillitoe, R. V., Medina, J. F., 611 & Zoghbi, H. Y. (2021). Deleting mecp2 from the cerebellum rather than its neuronal subtypes 612 causes a delay in motor learning in mice. eLife, 10, e64833. https://doi.org/10.7554/eLife.64833 613 Ahmad, S. F., Nadeem, A., Ansari, M. A., Bakheet, S. A., Al-Ayadhi, L. Y., & Attia, S. M. (2017). 614 Upregulation of IL-9 and JAK-STAT signaling pathway in children with autism. Progress in 615 Neuro-Psychopharmacology and Biological Psychiatry, 79, 472–480. 616 https://doi.org/10.1016/j.pnpbp.2017.08.002 617 Andoh, M., Ikegaya, Y., & Koyama, R. (2019). Microglia as possible therapeutic targets for autism 618 spectrum disorders. Progress in Molecular Biology and Translational Science, 167, 223–245. 619 https://doi.org/10.1016/BS.PMBTS.2019.06.012 620 Anggono, V., & Huganir, R. L. (2012). Regulation of AMPA receptor trafficking and synaptic plasticity. 621 Current Opinion in Neurobiology, 22(3), 461–469. https://doi.org/10.1016/j.conb.2011.12.006 622 Bailey, A., Luthert, P., Dean, A., Harding, B., Janota, I., Montgomery, M., Rutter, M., & Lantos, P. 623 (1998). A clinicopathological study of autism. Brain, 121(5), 889–905. 624 https://doi.org/10.1093/brain/121.5.889 625 Bariselli, S., Tzanoulinou, S., Glangetas, C., Prévost-Solié, C., Pucci, L., Viguié, J., Bezzi, P., 626 O’Connor, E. C., Georges, F., Lüscher, C., & Bellone, C. (2016). SHANK3 controls maturation 627 of social reward circuits in the VTA. Nature Neuroscience, 19(7), 926–934. 628 https://doi.org/10.1038/nn.4319 629 Bauman, M. L., & Kemper, T. L. (2005). Neuroanatomic observations of the brain in autism: A review 630 and future directions. International Journal of Developmental Neuroscience: The Official Journal 631 of the International Society for Developmental Neuroscience, 23(2–3), 183–187. 632 https://doi.org/10.1016/j.ijdevneu.2004.09.006 633 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 28 Becker, E. B. E., & Stoodley, C. J. (2013). Autism spectrum disorder and the cerebellum. In 634 International Review of Neurobiology (Vol. 113, pp. 1–34). Academic Press Inc. 635 https://doi.org/10.1016/B978-0-12-418700-9.00001-0 636 Berg, E., Copping, N., Rivera, J. K., Pride, M., Careaga, M., Bauman, M., Berman, R., Lein, P., Harony-637 Nicolas, H., Buxbaum, J., Ellegood, J., Lerch, J., Wöhr, M., & Silverman, J. (2018). 638 Developmental social communication deficits in the Shank3 rat model of phelan‐ mcdermid 639 syndrome and autism spectrum disorder. Autism Research, 11, null. 640 https://doi.org/10.1002/aur.1925 641 Berkel, S., Eltokhi, A., Fröhlich, H., Porras-Gonzalez, D., Rafiullah, R., Sprengel, R., & Rappold, G. A. 642 (2018). Sex Hormones Regulate SHANK Expression. Frontiers in Molecular Neuroscience, 11, 643 391513. https://doi.org/10.3389/FNMOL.2018.00337/BIBTEX 644 Betancur, C., & Buxbaum, J. D. (2013). SHANK3 haploinsufficiency: A “common” but underdiagnosed 645 highly penetrant monogenic cause of autism spectrum disorders. Molecular Autism, 4(1), 1–3. 646 https://doi.org/10.1186/2040-2392-4-17/FIGURES/1 647 Bey, A. L., Wang, X., Yan, H., Kim, N., Passman, R. L., Yang, Y., Cao, X., Towers, A. J., Hulbert, S. 648 W., Duffney, L. J., Gaidis, E., Rodriguiz, R. M., Wetsel, W. C., Yin, H. H., & Jiang, Y. H. (2018). 649 Brain region-specific disruption of Shank3 in mice reveals a dissociation for cortical and striatal 650 circuits in autism-related behaviors. Translational Psychiatry, 8, 94. 651 https://doi.org/10.1038/s41398-018-0142-6 652 Bissen, D., Foss, F., & Acker-Palmer, A. (2019). AMPA receptors and their minions: Auxiliary proteins 653 in AMPA receptor trafficking. Cellular and Molecular Life Sciences, 76(11), 2133–2169. 654 https://doi.org/10.1007/s00018-019-03068-7 655 Bozdagi, O., Sakurai, T., Papapetrou, D., Wang, X., Dickstein, D. L., Takahashi, N., Kajiwara, Y., Yang, 656 M., Katz, A. M., Scattoni, M., Harris, M. J., Saxena, R., Silverman, J. L., Crawley, J. N., Zhou, 657 Q., Hof, P. R., & Buxbaum, J. D. (2010). Haploinsufficiency of the autism-associated Shank3 658 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 29 gene leads to deficits in synaptic function, social interaction, and social communication. 659 Molecular Autism, 1(1), 15. https://doi.org/10.1186/2040-2392-1-15 660 Burnashev, N., Monyer, H., Seeburg, P. H., & Sakmann, B. (1992). Divalent ion permeability of AMPA 661 receptor channels is dominated by the edited form of a single subunit. Neuron, 8(1), 189–198. 662 https://doi.org/10.1016/0896-6273(92)90120-3 663 Chabrol, F. P., Arenz, A., Wiechert, M. T., Margrie, T. W., & DiGregorio, D. A. (2015). Synaptic diversity 664 enables temporal coding of coincident multisensory inputs in single neurons. Nature 665 Neuroscience, 18(5), 718–727. https://doi.org/10.1038/nn.3974 666 Chen, L., Bao, S., Qiao, X., & Thompson, R. F. (1999). Impaired cerebellar synapse maturation in 667 waggler, a mutant mouse with a disrupted neuronal calcium channel γ subunit. Proceedings of 668 the National Academy of Sciences of the United States of America, 96(21), 12132. 669 https://doi.org/10.1073/PNAS.96.21.12132 670 Chojnacka, M., Beroun, A., Magnowska, M., Stawikowska, A., Cysewski, D., Milek, J., Dziembowska, 671 M., & Kuzniewska, B. (2023). Impaired synaptic incorporation of AMPA receptors in a mouse 672 model of fragile X syndrome. Frontiers in Molecular Neuroscience, 16, 1258615. 673 https://doi.org/10.3389/FNMOL.2023.1258615/BIBTEX 674 Contestabile, A., Casarotto, G., Musardo, S., Espinosa, P., Maltese, F., Jiang, Y. hui, Bellone, C., & 675 Glangetas, C. (2023). Shank3 deficits in the anteromedial bed nucleus of the stria terminalis 676 trigger an anxiety phenotype in mice. European Journal of Neuroscience, 57(12), 1966–1979. 677 https://doi.org/10.1111/EJN.16043 678 Coombs, I., Bats, C., Sexton, C. A., Studniarczyk, D., Cull-Candy, S. G., & Farrant, M. (2023). 679 Enhanced functional detection of synaptic calcium-permeable AMPA receptors using 680 intracellular NASPM. eLife, 12, e66765. https://doi.org/10.7554/eLife.66765 681 Cope, E. C., Briones, B. A., Brockett, A. T., Martinez, S., Vigneron, P. A., Opendak, M., Wang, S. S. H., 682 & Gould, E. (2016). Immature Neurons and Radial Glia, But Not Astrocytes or Microglia, Are 683 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 30 Altered in Adult Cntnap2 and Shank3 Mice, Models of Autism. eNeuro, 3(5). 684 https://doi.org/10.1523/ENEURO.0196-16.2016 685 Cull-Candy, S. G., & Farrant, M. (2021). Ca2+-permeable AMPA receptors and their auxiliary subunits 686 in synaptic plasticity and disease. Journal of Physiology, 599(10), 2655–2671. 687 https://doi.org/10.1113/JP279029 688 da Silva, J. A. C., & Schröder, N. (2023). The Role of Ca2+ Permeable AMPA Receptors in 689 Neurodegeneration, Neurotoxicity, and Neuroinflammation. CNS & Neurological Disorders - 690 Drug Targets, 22(5), 624–633. https://doi.org/10.2174/1871527321666220510141735 691 Davis, B. M., Salinas-Navarro, M., Cordeiro, M. F., Moons, L., & De Groef, L. (2017). Characterizing 692 microglia activation: A spatial statistics approach to maximize information extraction. Scientific 693 Reports, 7(1), 1576. https://doi.org/10.1038/s41598-017-01747-8 694 Diagnostic and statistical manual of mental disorders: DSM-5TM (5th ed.). (2013). American Psychiatric 695 Association (APA). https://doi.org/10.1176/appi.books.9780890425596 696 Diering, G. H., & Huganir, R. L. (2018). The AMPA receptor code of synaptic plasticity. Neuron, 100(2), 697 314. https://doi.org/10.1016/J.NEURON.2018.10.018 698 Dolgacheva, L. P., Tuleukhanov, S. T., & Zinchenko, V. P. (2020). Participation of Ca2+-Permeable 699 AMPA Receptors in Synaptic Plasticity. Biochemistry (Moscow) Supplement Series A: 700 Membrane and Cell Biology, 14(3), 194–204. https://doi.org/10.1134/S1990747820030046 701 Dong, H., O’Brien, R. J., Fung, E. T., Lanahan, A. A., Worley, P. F., & Huganir, R. L. (1997). GRIP: A 702 synaptic PDZ domain-containing protein that interacts with AMPA receptors. Nature, 386(6622), 703 279–284. https://doi.org/10.1038/386279a0 704 Dong, H., Zhang, P., Song, I., Petralia, R. S., Liao, D., & Huganir, R. L. (1999). Characterization of the 705 Glutamate Receptor-Interacting Proteins GRIP1 and GRIP2. The Journal of Neuroscience, 706 19(16), 6930–6941. https://doi.org/10.1523/jneurosci.19-16-06930.1999 707 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 31 Drapeau, E., Riad, M., Kajiwara, Y., & Buxbaum, J. D. (2018). Behavioral phenotyping of an improved 708 mouse model of Phelan–McDermid syndrome with a complete deletion of the Shank3 gene. 709 eNeuro, 5(3). https://doi.org/10.1523/ENEURO.0046-18.2018 710 Fan, G., Ma, J., Ma, R., Suo, M., Chen, Y., Zhang, S., Zeng, Y., & Chen, Y. (2023). Microglia Modulate 711 Neurodevelopment in Autism Spectrum Disorder and Schizophrenia. International Journal of 712 Molecular Sciences, 24(24), Article 24. https://doi.org/10.3390/ijms242417297 713 Fatemi, S. H., Aldinger, K. A., Ashwood, P., Bauman, M. L., Blaha, C. D., Blatt, G. J., Chauhan, A., 714 Chauhan, V., Dager, S. R., Dickson, P. E., Estes, A. M., Goldowitz, D., Heck, D. H., Kemper, T. 715 L., King, B. H., Martin, L. A., Millen, K. J., Mittleman, G., Mosconi, M. W., … Welsh, J. P. (2012). 716 Consensus paper: Pathological role of the cerebellum in Autism. Cerebellum, 11(3), 777–807. 717 https://doi.org/10.1007/s12311-012-0355-9 718 Fino, E., Araya, R., Peterka, D. S., Salierno, M., Etchenique, R., & Yuste, R. (2009). RuBi-Glutamate: 719 Two-photon and visible-light photoactivation of neurons and dendritic spines. Frontiers in Neural 720 Circuits, 3(MAY), 2. https://doi.org/10.3389/NEURO.04.002.2009/BIBTEX 721 Furuichi, T., Shiraishi-Yamaguchi, Y., Sato, A., Sadakata, T., Huang, J., Shinoda, Y., Hayashi, K., 722 Mishima, Y., Tomomura, M., Nishibe, H., & Yoshikawa, F. (2011). Systematizing and Cloning of 723 Genes Involved in the Cerebellar Cortex Circuit Development. Neurochemical Research, 36(7), 724 1241–1252. https://doi.org/10.1007/s11064-011-0398-1 725 Gandhi, T., & Lee, C. C. (2021). Neural Mechanisms Underlying Repetitive Behaviors in Rodent Models 726 of Autism Spectrum Disorders. Frontiers in Cellular Neuroscience, 14, 463. 727 https://doi.org/10.3389/FNCEL.2020.592710/BIBTEX 728 Gardner, S. M., Takamiya, K., Xia, J., Suh, J.-G., Johnson, R., Yu, S., & Huganir, R. L. (2005). 729 Calcium-Permeable AMPA Receptor Plasticity Is Mediated by Subunit-Specific Interactions with 730 PICK1 and NSF. Neuron, 45(6), 903–915. https://doi.org/10.1016/j.neuron.2005.02.026 731 Giza, J., Urbanski, M. J., Prestori, F., Bandyopadhyay, B., Yam, A., Friedrich, V., Kelley, K., D’Angelo, 732 E., & Goldfarb, M. (2010). Behavioral and Cerebellar Transmission Deficits in Mice Lacking the 733 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 32 Autism-Linked Gene Islet Brain-2. Journal of Neuroscience, 30(44), 14805–14816. 734 https://doi.org/10.1523/JNEUROSCI.1161-10.2010 735 Glasgow, S. D., McPhedrain, R., Madranges, J. F., Kennedy, T. E., & Ruthazer, E. S. (2019). 736 Approaches and limitations in the investigation of synaptic transmission and plasticity. Frontiers 737 in Synaptic Neuroscience, 11(JUL), 20. https://doi.org/10.3389/FNSYN.2019.00020/BIBTEX 738 Guo, B., Chen, J., Chen, Q., Ren, K., Feng, D., Mao, H., Yao, H., Yang, J., Liu, H., Liu, Y., Jia, F., Qi, 739 C., Lynn-Jones, T., Hu, H., Fu, Z., Feng, G., Wang, W., & Wu, S. (2019). Anterior cingulate 740 cortex dysfunction underlies social deficits in Shank3 mutant mice. Nature Neuroscience, 22(8), 741 1223–1234. https://doi.org/10.1038/s41593-019-0445-9 742 Guo, B., Liu, T., Choi, S., Halassa, M. M., Guo, B., Liu, T., Choi, S., Mao, H., Wang, W., Xi, K., & 743 Jones, C. (2024). Restoring thalamocortical circuit dysfunction by correcting HCN 744 channelopathy in Shank3 mutant mice. Cell Reports Medicine, 101534. 745 https://doi.org/10.1016/j.xcrm.2024.101534 746 Guo, C., & Ma, Y. Y. (2021). Calcium Permeable-AMPA Receptors and Excitotoxicity in Neurological 747 Disorders. Frontiers in Neural Circuits, 15, 82. 748 https://doi.org/10.3389/FNCIR.2021.711564/BIBTEX 749 Ha, H. T. T., Leal-Ortiz, S., Lalwani, K., Kiyonaka, S., Hamachi, I., Mysore, S. P., Montgomery, J. M., 750 Garner, C. C., Huguenard, J. R., & Kim, S. A. (2018). Shank and zinc mediate an AMPA 751 receptor subunit switch in developing neurons. Frontiers in Molecular Neuroscience, 11, 405. 752 https://doi.org/10.3389/FNMOL.2018.00405/BIBTEX 753 Ha, S., Lee, D., Cho, Y. S., Chung, C., Yoo, Y. E., Kim, J., Lee, J., Kim, W., Kim, H., Bae, Y., Tanaka-754 Yamamoto, K., & Kim, E. (2016). Cerebellar shank2 regulates excitatory synapse density, motor 755 coordination, and specific repetitive and anxiety-like behaviors. Journal of Neuroscience, 36(48), 756 12129–12143. https://doi.org/10.1523/JNEUROSCI.1849-16.2016 757 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 33 Hack, N., & Balázs, R. (1995). Properties of AMPA Receptors Expressed in Rat Cerebellar Granule 758 Cell Cultures: Ca2+ Influx Studies. Journal of Neurochemistry, 65(3), 1077–1084. 759 https://doi.org/10.1046/j.1471-4159.1995.65031077.x 760 Hampson, D. R., & Blatt, G. J. (2015). Autism spectrum disorders and neuropathology of the 761 cerebellum. Frontiers in Neuroscience, 9(NOV), 420. https://doi.org/10.3389/fnins.2015.00420 762 Henley, J. M., & Wilkinson, K. A. (2016). Synaptic AMPA receptor composition in development, 763 plasticity and disease. Nature Reviews Neuroscience, 17(6), 337–350. 764 https://doi.org/10.1038/nrn.2016.37 765 Hollmann, M., Hartley, M., & Heinemann, S. (1991). Ca2+ permeability of KA-AMPA--gated glutamate 766 receptor channels depends on subunit composition. Science (New York, N.Y.), 252(5007), 851–767 853. https://doi.org/10.1126/science.1709304 768 Hu, C., Li, H., Li, J., Luo, X., & Hao, Y. (2022). Microglia: Synaptic modulator in autism spectrum 769 disorder. Frontiers in Psychiatry, 13. https://doi.org/10.3389/fpsyt.2022.958661 770 Hughes, H. K., R.J.Moreno, & Ashwood, P. (2023). Innate immune dysfunction and neuroinflammation 771 in autism spectrum disorder (ASD). Brain, Behavior, and Immunity, 108, 245–254. 772 https://doi.org/10.1016/j.bbi.2022.12.001 773 Hulbert, S. W., & Jiang, Y. hui. (2017). Cellular and Circuitry Bases of Autism: Lessons Learned from 774 the Temporospatial Manipulation of Autism Genes in the Brain. Neuroscience Bulletin, 33(2), 775 205–218. https://doi.org/10.1007/S12264-017-0112-7/TABLES/2 776 Jacobi, E., Von Engelhardt, J., Forsythe, I., & Bé¨ıque, J.-C. B. (2021). Modulation of information 777 processing by AMPA receptor auxiliary subunits. The Journal of Physiology, 599(2), 471–483. 778 https://doi.org/10.1113/JP276698 779 Jaramillo, T. C., Speed, H. E., Xuan, Z., Reimers, J. M., Escamilla, C. O., Weaver, T. P., Liu, S., 780 Filonova, I., & Powell, C. M. (2017). Novel Shank3 mutant exhibits behaviors with face validity 781 for autism and altered striatal and hippocampal function. Autism Research, 10(1), 42–65. 782 https://doi.org/10.1002/aur.1664 783 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 34 Jaramillo, T. C., Speed, H. E., Xuan, Z., Reimers, J. M., Liu, S., & Powell, C. M. (2016). Altered Striatal 784 Synaptic Function and Abnormal Behaviour in Shank3 Exon4–9 Deletion Mouse Model of 785 Autism. Autism Research, 9(3), 350. https://doi.org/10.1002/AUR.1529 786 Jiang, Y. hui, & Ehlers, M. D. (2013). Modeling Autism by SHANK Gene Mutations in Mice. Neuron, 787 78(1), 8–27. https://doi.org/10.1016/j.neuron.2013.03.016 788 Jung, S., & Park, M. (2022). Shank postsynaptic scaffolding proteins in autism spectrum disorder: 789 Mouse models and their dysfunctions in behaviors, synapses, and molecules. Pharmacological 790 Research, 182, 106340. https://doi.org/10.1016/J.PHRS.2022.106340 791 Kamalova, A., Futai, K., Delpire, E., & Nakagawa, T. (2020). AMPA Receptor Auxiliary Subunit GSG1L 792 Suppresses Short-Term Facilitation in Corticothalamic Synapses and Determines Seizure 793 Susceptibility. Cell Reports, 32(3). https://doi.org/10.1016/j.celrep.2020.107921 794 Kaplan, J. S., Nipper, M. A., Richardson, B. D., Jensen, J., Helms, M., Finn, D. A., & Rossi, D. J. 795 (2016). Pharmacologically Counteracting a Phenotypic Difference in Cerebellar GABAA 796 Receptor Response to Alcohol Prevents Excessive Alcohol Consumption in a High Alcohol-797 Consuming Rodent Genotype. The Journal of Neuroscience, 36(35), 9019. 798 https://doi.org/10.1523/JNEUROSCI.0042-16.2016 799 Kauer, J. A., & Malenka, R. C. (2007). Synaptic plasticity and addiction. Nature Reviews Neuroscience, 800 8(11), 844–858. https://doi.org/10.1038/nrn2234 801 Kawamura, A., Katayama, Y., Kakegawa, W., Ino, D., Nishiyama, M., Yuzaki, M., & Nakayama, K. I. 802 (2021). The autism-associated protein CHD8 is required for cerebellar development and motor 803 function. Cell Reports, 35(1), 108932. https://doi.org/10.1016/j.celrep.2021.108932 804 Kim, H., Lim, C. S., & Kaang, B. K. (2016). Neuronal mechanisms and circuits underlying repetitive 805 behaviors in mouse models of autism spectrum disorder. Behavioral and Brain Functions/i1: 806 BBF, 12(1), 3. https://doi.org/10.1186/S12993-016-0087-Y 807 Kim, J., Pavlidis, P., & Ciernia, A. V. (2023). Development of a high-throughput pipeline to characterize 808 microglia morphological states at a single-cell resolution. 11(July), 1–16. 809 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 35 Kloth, A. D., Badura, A., Li, A., Cherskov, A., Connolly, S. G., Giovannucci, A., Bangash, M. A., 810 Grasselli, G., Agarikano, O. P., Piochon, C., Tsai, P. T., Geschwind, D. H., Hansel, C., Sahin, 811 M., Takumi, T., Worley, P. F., & Wang, S. S. H. (2015). Cerebellar associative sensory learning 812 defects in five mouse autism models. eLife, 4(JULY 2015). https://doi.org/10.7554/eLife.06085 813 Kouser, M., Speed, H. E., Dewey, C. M., Reimers, J. M., Widman, A. J., Gupta, N., Liu, S., Jaramillo, T. 814 C., Bangash, M., Xiao, B., Worley, P., & Powell, C. (2013). Loss of Predominant Shank3 815 Isoforms Results in Hippocampus-Dependent Impairments in Behavior and Synaptic 816 Transmission. The Journal of Neuroscience, 33, 18448–18468. 817 https://doi.org/10.1523/JNEUROSCI.3017-13.2013 818 Kshetri, R., Beavers, J. O., Hyde, R., Ewa, R., Schwertman, A., Porcayo, S., & Richardson, B. D. 819 (2024). Behavioral decline in Shank3Δ ex4–22 mice during early adulthood parallels cerebellar 820 granule cell glutamatergic synaptic changes. Molecular Autism, 15(1), 1–21. 821 https://doi.org/10.1186/S13229-024-00628-Y 822 Lein, E. S., Hawrylycz, M. J., Ao, N., Ayres, M., Bensinger, A., Bernard, A., Boe, A. F., Boguski, M. S., 823 Brockway, K. S., Byrnes, E. J., Chen, L., Chen, L., Chen, T. M., Chin, M. C., Chong, J., Crook, 824 B. E., Czaplinska, A., Dang, C. N., Datta, S., … Jones, A. R. (2006). Genome-wide atlas of 825 gene expression in the adult mouse brain. Nature, 445(7124), 168–176. 826 https://doi.org/10.1038/nature05453 827 Liu, S. J., & Cull-Candy, S. G. (2005). Subunit interaction with PICK and GRIP controls Ca2+ 828 permeability of AMPARs at cerebellar synapses. Nature Neuroscience, 8(6), 768–775. 829 https://doi.org/10.1038/nn1468 830 Liu, S. J., & Savtchouk, I. (2012). Ca2+ permeable AMPA receptors switch allegiances: Mechanisms 831 and consequences. The Journal of Physiology, 590(Pt 1), 13. 832 https://doi.org/10.1113/JPHYSIOL.2011.213926 833 Liu, S. J., & Zukin, R. S. (2007). Ca2+-permeable AMPA receptors in synaptic plasticity and neuronal 834 death. Trends in Neurosciences, 30(3), 126–134. https://doi.org/10.1016/j.tins.2007.01.006 835 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 36 Liu, S.-Q. J., & Cull-Candy, S. G. (2000). Synaptic activity at calcium-permeable AMPA receptors 836 induces a switch in receptor subtype. Nature, 405(6785), 454–458. 837 https://doi.org/10.1038/35013064 838 Lonsdale, J., Thomas, J., Salvatore, M., Phillips, R., Lo, E., Shad, S., Hasz, R., Walters, G., Garcia, F., 839 Young, N., Foster, B., Moser, M., Karasik, E., Gillard, B., Ramsey, K., Sullivan, S., Bridge, J., 840 Magazine, H., Syron, J., … Moore, H. F. (2013). The Genotype-Tissue Expression (GTEx) 841 project. Nature Genetics 2013 45:6, 45(6), 580–585. https://doi.org/10.1038/ng.2653 842 Lord, C., Brugha, T. S., Charman, T., Cusack, J., Dumas, G., Frazier, T., Jones, E. J. H., Jones, R. M., 843 Pickles, A., State, M. W., Taylor, J. L., & Veenstra-VanderWeele, J. (2020). Autism spectrum 844 disorder. Nature Reviews Disease Primers, 6(1), 1–23. https://doi.org/10.1038/s41572-019-845 0138-4 846 Manning, A., Bender, P. T. R., Boyd-Pratt, H., Mendelson, B. Z., Hruska, M., & Anderson, C. T. (2024). 847 Trans-synaptic Association of Vesicular Zinc Transporter 3 and Shank3 Supports Synapse-848 Specific Dendritic Spine Structure and Function in the Mouse Auditory Cortex. Journal of 849 Neuroscience, 44(28), e0619242024. https://doi.org/10.1523/JNEUROSCI.0619-24.2024 850 Matta, S. M., Hill-Yardin, E. L., & Crack, P. J. (2019). The influence of neuroinflammation in Autism 851 Spectrum Disorder. Brain, Behavior, and Immunity, 79, 75–90. 852 https://doi.org/10.1016/J.BBI.2019.04.037 853 Menashe, I., Grange, P., Larsen, E. C., Banerjee-Basu, S., & Mitra, P. P. (2013). Co-expression 854 Profiling of Autism Genes in the Mouse Brain. PLOS Computational Biology, 9(7), e1003128. 855 https://doi.org/10.1371/JOURNAL.PCBI.1003128 856 Miguez-Cabello, F., Wang, X., Yan, Y., Brake, N., Alexander, R. P. D., Perozzo, A. M., Khadra, A., & 857 Bowie, D. (2025). GluA2-containing AMPA receptors form a continuum of Ca2+-permeable 858 channels. Nature. https://doi.org/10.1038/s41586-025-08736-2 859 Monteiro, P., & Feng, G. (2017). SHANK proteins: Roles at the synapse and in autism spectrum 860 disorder. Nature Reviews Neuroscience, 18(3), 147–157. https://doi.org/10.1038/nrn.2016.183 861 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 37 Morgan, J. T., Chana, G., Pardo, C. A., Achim, C., Semendeferi, K., Buckwalter, J., Courchesne, E., & 862 Everall, I. P. (2010). Microglial Activation and Increased Microglial Density Observed in the 863 Dorsolateral Prefrontal Cortex in Autism. Biological Psychiatry, 68(4), 368–376. 864 https://doi.org/10.1016/j.biopsych.2010.05.024 865 Mosconi, M. W., Wang, Z., Schmitt, L. M., Tsai, P., & Sweeney, J. A. (2015). The role of cerebellar 866 circuitry alterations in the pathophysiology of autism spectrum disorders. Frontiers in 867 Neuroscience, 9, 296. https://doi.org/10.3389/FNINS.2015.00296 868 Moutin, E., Sakkaki, S., Compan, V., Bouquier, N., Giona, F., Areias, J., Goyet, E., Hemonnot-Girard, 869 A. L., Seube, V., Glasson, B., Benac, N., Chastagnier, Y., Raynaud, F., Audinat, E., Groc, L., 870 Maurice, T., Sala, C., Verpelli, C., & Perroy, J. (2021). Restoring glutamate receptosome 871 dynamics at synapses rescues autism-like deficits in Shank3-deficient mice. Molecular 872 Psychiatry, 26(12), 7596–7609. https://doi.org/10.1038/s41380-021-01230-x 873 Paolicelli, R. C., Sierra, A., Stevens, B., Tremblay, M. E., Aguzzi, A., Ajami, B., Amit, I., Audinat, E., 874 Bechmann, I., Bennett, M., Bennett, F., Bessis, A., Biber, K., Bilbo, S., Blurton-Jones, M., 875 Boddeke, E., Brites, D., Brône, B., Brown, G. C., … Wyss-Coray, T. (2022). Microglia states and 876 nomenclature: A field at its crossroads. Neuron, 110(21), 3458. 877 https://doi.org/10.1016/J.NEURON.2022.10.020 878 Peça, J., Feliciano, C., Ting, J. T., Wang, W., Wells, M. F., Venkatraman, T. N., Lascola, C. D., Fu, Z., 879 & Feng, G. (2011). Shank3 mutant mice display autistic-like behaviours and striatal dysfunction. 880 Nature, 472(7344), 437–442. https://doi.org/10.1038/nature09965 881 Petrelli, F., Pucci, L., & Bezzi, P. (2016). Astrocytes and microglia and their potential link with autism 882 spectrum disorders. Frontiers in Cellular Neuroscience, 10(FEB), 177723. 883 https://doi.org/10.3389/FNCEL.2016.00021/BIBTEX 884 Qneibi, M., Bdir, S., Bdair, M., Aldwaik, S. A., Heeh, M., Sandouka, D., & Idais, T. (2024). Exploring the 885 role of AMPA receptor auxiliary proteins in synaptic functions and diseases. FEBS Journal, 1–886 46. https://doi.org/10.1111/febs.17287 887 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 38 Richardson, B. D., Ling, L. L., Uteshev, V. V., & Caspary, D. M. (2013). Reduced GABAA Receptor-888 Mediated Tonic Inhibition in Aged Rat Auditory Thalamus. The Journal of Neuroscience, 33(3), 889 1218. https://doi.org/10.1523/JNEUROSCI.3277-12.2013 890 Richardson, B. D., & Rossi, D. J. (2017). Recreational concentrations of alcohol enhance synaptic 891 inhibition of cerebellar unipolar brush cells via pre- and postsynaptic mechanisms. Journal of 892 Neurophysiology, 118(1), 267–279. 893 https://doi.org/10.1152/JN.00963.2016/ASSET/IMAGES/LARGE/Z9K0061741660007.JPEG 894 Rozov, A., Zakharova, Y., Vazetdinova, A., & Valiullina-Rakhmatullina, F. (2018). The role of 895 polyamine-dependent facilitation of calcium permeable ampars in short-term synaptic 896 enhancement. Frontiers in Cellular Neuroscience, 12, 417438. 897 https://doi.org/10.3389/FNCEL.2018.00345/BIBTEX 898 Sahin, M., & Sur, M. (2015). Genes, circuits, and precision therapies for autism and related 899 neurodevelopmental disorders. Science, 350(6263), aab3897. 900 https://doi.org/10.1126/science.aab3897 901 Sala, C., Vicidomini, C., Bigi, I., Mossa, A., & Verpelli, C. (2015). Shank synaptic scaffold proteins: Keys 902 to understanding the pathogenesis of autism and other synaptic disorders. Journal of 903 Neurochemistry, 135(5), 849–858. https://doi.org/10.1111/jnc.13232 904 Sato, A., Sekine, Y., Saruta, C., Nishibe, H., Morita, N., Sato, Y., Sadakata, T., Shinoda, Y., Kojima, T., 905 & Furuichi, T. (2008). Cerebellar development transcriptome database (CDT-DB): Profiling of 906 spatio-temporal gene expression during the postnatal development of mouse cerebellum. 907 Neural Networks, 21(8), 1056–1069. https://doi.org/10.1016/J.NEUNET.2008.05.004 908 Satterstrom, F. K., Kosmicki, J. A., Wang, J., Breen, M. S., De Rubeis, S., An, J. Y., Peng, M., Collins, 909 R., Grove, J., Klei, L., Stevens, C., Reichert, J., Mulhern, M. S., Artomov, M., Gerges, S., 910 Sheppard, B., Xu, X., Bhaduri, A., Norman, U., … Buxbaum, J. D. (2020). Large-Scale Exome 911 Sequencing Study Implicates Both Developmental and Functional Changes in the Neurobiology 912 of Autism. Cell, 180(3), 568-584.e23. https://doi.org/10.1016/j.cell.2019.12.036 913 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 39 Savage, J. C., Carrier, M., & Tremblay, M.-È. (2019). Morphology of Microglia Across Contexts of 914 Health and Disease. In O. Garaschuk & A. Verkhratsky (Eds.), Microglia: Methods and 915 Protocols (pp. 13–26). Springer. https://doi.org/10.1007/978-1-4939-9658-2_2 916 Savidge, J. R., & Bristow, D. R. (1997). Distribution of Ca2+-permeable AMPA receptors among 917 cultured rat cerebellar granule cells: NeuroReport, 8(8), 1877–1882. 918 https://doi.org/10.1097/00001756-199705260-00017 919 Schmeisser, M., & Verpelli, C. (2016). SHANK Mutations in Intellectual Disability and Autism Spectrum 920 Disorder. https://doi.org/10.1016/B978-0-12-800109-7.00010-8 921 Schofield, C. M., & Huguenard, J. R. (2007). GABA Affinity Shapes IPSCs in Thalamic Nuclei. Journal 922 of Neuroscience, 27(30), 7954–7962. https://doi.org/10.1523/JNEUROSCI.0377-07.2007 923 Sheng, M., & Kim, E. (2000). The Shank family of scaffold proteins. Journal of Cell Science, 113(11), 924 1851–1856. https://doi.org/10.1242/jcs.113.11.1851 925 Soda, T., Mapelli, L., Locatelli, F., Botta, L., Goldfarb, M., Prestori, F., & D’Angelo, E. (2019). 926 Hyperexcitability and hyperplasticity disrupt cerebellar signal transfer in the IB2 KO mouse 927 model of autism. Journal of Neuroscience, 39(13), 2383–2397. 928 https://doi.org/10.1523/JNEUROSCI.1985-18.2019 929 Sola, E., Prestori, F., Rossi, P., Taglietti, V., & D’Angelo, E. (2004). Increased neurotransmitter release 930 during long-term potentiation at mossy fibre–granule cell synapses in rat cerebellum. The 931 Journal of Physiology, 557(3), 843–861. https://doi.org/10.1113/JPHYSIOL.2003.060285 932 Soorya, L., Kolevzon, A., Zweifach, J., Lim, T., Dobry, Y., Schwartz, L., Frank, Y., Wang, A. T., Cai, G., 933 Parkhomenko, E., Halpern, D., Grodberg, D., Angarita, B., Willner, J. P., Yang, A., Canitano, R., 934 Chaplin, W., Betancur, C., & Buxbaum, J. D. (2013). Prospective investigation of autism and 935 genotype-phenotype correlations in 22q13 deletion syndrome and SHANK3 deficiency. 936 Molecular Autism, 4(1), 1–17. https://doi.org/10.1186/2040-2392-4-18 937 Speed, H. E., Kouser, M., Xuan, Z., Reimers, J. M., Ochoa, C. F., Gupta, N., Liu, S., & Powell, C. M. 938 (2015). Autism-Associated Insertion Mutation (InsG) of Shank3 Exon 21 Causes Impaired 939 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 40 Synaptic Transmission and Behavioral Deficits. Journal of Neuroscience, 35(26), 9648–9665. 940 https://doi.org/10.1523/JNEUROSCI.3125-14.2015 941 Stoodley, C. J., D’Mello, A. M., Ellegood, J., Jakkamsetti, V., Liu, P., Nebel, M. B., Gibson, J. M., Kelly, 942 E., Meng, F., Cano, C. A., Pascual, J. M., Mostofsky, S. H., Lerch, J. P., & Tsai, P. T. (2017). 943 Altered cerebellar connectivity in autism and cerebellar-mediated rescue of autism-related 944 behaviors in mice. Nature Neuroscience 2017 20:12, 20(12), 1744–1751. 945 https://doi.org/10.1038/s41593-017-0004-1 946 Suyama, S., Ralevski, A., Liu, Z. W., Dietrich, M. O., Yada, T., Simonds, S. E., Cowley, M. A., Gao, X. 947 B., Diano, S., & Horvath, T. L. (2017). Plasticity of calcium-permeable ampa glutamate 948 receptors in proopiomelanocortin neurons. eLife, 6. https://doi.org/10.7554/ELIFE.25755 949 Traynelis, S. F., Wollmuth, L. P., McBain, C. J., Menniti, F. S., Vance, K. M., Ogden, K. K., Hansen, K. 950 B., Yuan, H., Myers, S. J., & Dingledine, R. (2010). Glutamate Receptor Ion Channels: 951 Structure, Regulation, and Function. Pharmacological Reviews, 62(3), 405–496. 952 https://doi.org/10.1124/PR.109.002451 953 Tsai, P. T. (2016). Autism and cerebellar dysfunction: Evidence from animal models. Seminars in Fetal 954 and Neonatal Medicine, 21(5), 349–355. 955 Tsai, P. T., Hull, C., Chu, Y., Greene-Colozzi, E., Sadowski, A. R., Leech, J. M., Steinberg, J., Crawley, 956 J. N., Regehr, W. G., & Sahin, M. (2012). Autistic-like behaviour and cerebellar dysfunction in 957 Purkinje cell Tsc1 mutant mice. Nature, 488(7413), 647–651. 958 https://doi.org/10.1038/nature11310 959 Twomey, E. C., Yelshanskaya, M. V., Vassilevski, A. A., & Sobolevsky, A. I. (2018). Mechanisms of 960 Channel Block in Calcium-Permeable AMPA Receptors. Neuron, 99(5), 956-968.e4. 961 https://doi.org/10.1016/j.neuron.2018.07.027 962 Vargas, D. L., Nascimbene, C., Krishnan, C., Zimmerman, A. W., & Pardo, C. A. (2005). Neuroglial 963 activation and neuroinflammation in the brain of patients with autism. Annals of Neurology, 964 57(1), 67–81. https://doi.org/10.1002/ana.20315 965 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 41 Wan, L., Liu, D., Xiao, W. B., Zhang, B. X., Yan, X. X., Luo, Z. H., & Xiao, B. (2021). Association of 966 SHANK Family with Neuropsychiatric Disorders: An Update on Genetic and Animal Model 967 Discoveries. Cellular and Molecular Neurobiology, 1–21. https://doi.org/10.1007/S10571-021-968 01054-X 969 Wang, K., Gaitsch, H., Poon, H., Cox, N. J., & Rzhetsky, A. (2017). Classification of common human 970 diseases derived from shared genetic and environmental determinants. Nature Genetics, 49(9), 971 1319–1325. https://doi.org/10.1038/ng.3931 972 Wang, S. S. H., Kloth, A. D., & Badura, A. (2014). The Cerebellum, Sensitive Periods, and Autism. 973 Neuron, 83(3), 518–532. https://doi.org/10.1016/j.neuron.2014.07.016 974 Wang, W., Li, C., Chen, Q., Van Der Goes, M. S., Hawrot, J., Yao, A. Y., Gao, X., Lu, C., Zang, Y., 975 Zhang, Q., Lyman, K., Wang, D., Guo, B., Wu, S., Gerfen, C. R., Fu, Z., & Feng, G. (2017). 976 Striatopallidal dysfunction underlies repetitive behavior in Shank3-deficient model of autism. The 977 Journal of Clinical Investigation, 127(5), 1978–1990. https://doi.org/10.1172/JCI87997 978 Wang, X., Bey, A. L., Katz, B. M., Badea, A., Kim, N., David, L. K., Duffney, L. J., Kumar, S., Mague, S. 979 D., Hulbert, S. W., Dutta, N., Hayrapetyan, V., Yu, C., Gaidis, E., Zhao, S., Ding, J.-D., Xu, Q., 980 Chung, L., Rodriguiz, R. M., … Jiang, Y. (2016). Altered mGluR5-Homer scaffolds and 981 corticostriatal connectivity in a Shank3 complete knockout model of autism. Nature 982 Communications, 7(1), 11459. https://doi.org/10.1038/ncomms11459 983 Wang, X., McCoy, P. A., Rodriguiz, R. M., Pan, Y., Je, H. S., Roberts, A. C., Kim, C. J., Berrios, J., 984 Colvin, J. S., Bousquet-Moore, D., Lorenzo, I., Wu, G., Weinberg, R. J., Ehlers, M. D., Philpot, 985 B. D., Beaudet, A. L., Wetsel, W. C., & Jiang, Y. H. (2011). Synaptic dysfunction and abnormal 986 behaviors in mice lacking major isoforms of Shank3. Human Molecular Genetics, 20(15), 3093–987 3108. https://doi.org/10.1093/hmg/ddr212 988 Wang, X., Xu, Q., Bey, A. L., Lee, Y., & Jiang, Y. H. (2014). Transcriptional and functional complexity of 989 Shank3 provides a molecular framework to understand the phenotypic heterogeneity of 990 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 42 SHANK3 causing autism and Shank3 mutant mice. Molecular Autism, 5(1), 30. 991 https://doi.org/10.1186/2040-2392-5-30 992 Won, H., Mah, W., & Kim, E. (2013). Autism spectrum disorder causes, mechanisms, and treatments: 993 Focus on neuronal synapses. Frontiers in Molecular Neuroscience, 6, 19. 994 https://doi.org/10.3389/fnmol.2013.00019 995 Xia, J., Zhang, X., Staudinger, J., & Huganir, R. L. (1999). Clustering of AMPA Receptors by the 996 Synaptic PDZ Domain–Containing Protein PICK1. Neuron, 22(1), 179–187. 997 https://doi.org/10.1016/S0896-6273(00)80689-3 998 Xiong, Y., Chen, J., & Li, Y. (2023). Microglia and astrocytes underlie neuroinflammation and synaptic 999 susceptibility in autism spectrum disorder. Frontiers in Neuroscience, 17, 1125428. 1000 https://doi.org/10.3389/FNINS.2023.1125428/BIBTEX 1001 Xu, Q. W., Larosa, A., & Wong, T. P. (2024). Roles of AMPA receptors in social behaviors. Frontiers in 1002 Synaptic Neuroscience, 16, 1405510. https://doi.org/10.3389/FNSYN.2024.1405510/BIBTEX 1003 Xue, M., Cai, Z.-L., Okoh, J., Xu, Q. W., Larosa, A., & Wong, T. P. (2024). Roles of AMPA receptors in 1004 social behaviors. Frontiers in Synaptic Neuroscience, 16, 1405510. 1005 https://doi.org/10.3389/FNSYN.2024.1405510 1006 Yang, J., Ma, G., Du, X., Xie, J., Wang, M., Wang, W., Guo, B., & Wu, S. (2024). Deciphering the Role 1007 of Shank3 in Dendritic Morphology and Synaptic Function Across Postnatal Developmental 1008 Stages in the Shank3B KO Mouse. Neuroscience Bulletin. https://doi.org/10.1007/s12264-024-1009 01330-y 1010 Yoo, T., Cho, H., Lee, J., Park, H., Yoo, Y. E., Yang, E., Kim, J. Y., Kim, H., & Kim, E. (2018). GABA 1011 neuronal deletion of Shank3 exons 14–16 in mice suppresses striatal excitatory synaptic input 1012 and induces social and locomotor abnormalities. Frontiers in Cellular Neuroscience, 12, 417328. 1013 https://doi.org/10.3389/FNCEL.2018.00341/BIBTEX 1014 Yoo, T., Cho, H., Park, H., Lee, J., & Kim, E. (2019). Shank3 Exons 14–16 Deletion in Glutamatergic 1015 Neurons Leads to Social and Repetitive Behavioral Deficits Associated With Increased Cortical 1016 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 43 Layer 2/3 Neuronal Excitability. Frontiers in Cellular Neuroscience, 13, 484477. 1017 https://doi.org/10.3389/FNCEL.2019.00458/BIBTEX 1018 Zhang, J.-W., He, D.-J., & Li, X.-J. (2023). Neuropathological insights from SHANK3 mutant animal 1019 models. Ageing and Neurodegenerative Diseases, 3(4). https://doi.org/10.20517/and.2023.18 1020 Zhou, Y., Kaiser, T., Monteiro, P., Zhang, X., Van der Goes, Marie. S., Wang, D., Barak, B., Zeng, M., 1021 Li, C., Lu, C., Wells, M., Amaya, A., Nguyen, S., Lewis, M., Sanjana, N., Zhou, Y., Zhang, M., 1022 Zhang, F., Fu, Z., & Feng, G. (2016). Mice with Shank3 Mutations Associated with ASD and 1023 Schizophrenia Display Both Shared and Distinct Defects. Neuron, 89(1), 147–162. 1024 https://doi.org/10.1016/j.neuron.2015.11.023 1025 Zhou, Y., Kaiser, T., Monteiro, P., Zhang, X., Van der Goes, Marie. S., Wang, D., Barak, B., Zeng, M., 1026 Li, C., Lu, C., Wells, M., Amaya, A., Nguyen, S., Lewis, M., Sanjana, N., Zhou, Y., Zhang, M., 1027 Zhang, F., Fu, Z., & Feng, G. (2016). Mice with Shank3 Mutations Associated with ASD and 1028 Schizophrenia Display Both Shared and Distinct Defects. Neuron, 89(1), 147–162. 1029 https://doi.org/10.1016/j.neuron.2015.11.023 1030 1031 1032 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 44 Figures and Legends 1033 Figure 1 1034 Figure 1: Increased mEPSC amplitude in CGCs of adult Shank3 KO mice. (A) Representative 1035 traces of mEPSC recorded from CGC in WT (+/+, red) and Shank3 KO (-/-, light red) mice in the 1036 presence of 10 μ M gabazine and 0.5 μ M TTX. ( B, D) Cumulative distribution histograms of mEPSC 1037 amplitudes (B) and interevent intervals ( D) for all events from WT and KO groups. ( C, E) Individual 1038 average data points from each cell (circles) and group mean ± SEM (bars) for mEPSC amplitudes (C ) 1039 and IEIs (E). WT: n = 16 cells from N = 10 mice, 1419 events; Shank3 KO: n = 18 cells from N = 10 1040 mice, 1391 events. Statistical significance was determined usi ng an unpaired t-test for mEPSC 1041 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 45 amplitude ( B) and a Mann-Whitney U test for IEI ( C). *p < 0.05 indicates a significant difference 1042 between genotypes; ns: not significant. 1043 1044 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 46 Figure 2 1045 1046 Figure 2: Increased evoked responses at higher stimulation intensities in Shank3 KO mice. ( A) 1047 Schematic diagram of cerebellar circuit showing the stimulation (mossy fiber, MF) and recording sites 1048 (cerebellar granule cell) within the cerebellar cortex. ( B) Representative traces of eEPSCs at various 1049 stimulus intensities in WT (+/+, red) and Shank3 KO (-/-, light red) mice. ( C) Quantification of the input-1050 output relationship showing mean eEPSC amplitude ± SEM (bars) at each current intensity applied to 1051 stimulate presynaptic MF terminals in WT and Shank3 KO mice. ( D) Percentage distribution of 1052 amplitude of total eEPSC individual events (excluding failure events) in 20 pA histogram bins from each 1053 CGC with the Gaussian fit of the amplitude distribution shown in the inset ( E). In C, WT: n = 40 cells 1054 from N = 14 mice; Shank3 KO: n = 42 cells from N = 20 mice. An unpaired t-test used for parametric 1055 and a Mann-Whitney test was used for non-parametric dataset for comparing responses between 1056 genotypes at each stimulus intensity. In D, WT: n = 11 cells from N = 5 mice, 382 events; Shank3 KO: n 1057 = 13 cells from N = 10 mice, 374 events. Abbreviations: ML: Molecular layer, PCL: Purkinje cell layer, 1058 GCL: Granule cell layer. 1059 1060 1061 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 47 Figure 3 1062 Figure 3: Faster decay kinetics of AMPAR-mediated eEPSC in Shank3 KO mice. ( A) 1063 Representative traces of AMPAR-mediated eEPSCs at -60 mV holding potential and NMDAR-mediated 1064 eEPSCs at +40 mV holding potential in WT and Shank3 KO mice. ( B) Average peak amplitudes of 1065 AMPAR-mediated eEPSCs. ( C) Average amplitudes of NMDAR-medi ated eEPSCs at +40 mV, 15 ms 1066 after stimulation. ( D) Ratio of average AMPAR to NMDAR amplitudes. ( E) Representative traces of 1067 paired-pulse responses (inter-stimulus interval [ISI]: 20 ms) in WT and Shank3 KO mice. ( F) 1068 Quantification of the paired-pulse ratio in both genotypes. ( G) Representative traces of normalized 1069 AMPAR-mediated eEPSC illustrating decay kinetics in both genotypes. ( H) Weighted decay tau values 1070 of AMPAR responses evoked by 100 µA stimulation in both genotypes. For panels B-D, F, and H , 1071 individual data points are shown as circles, and bars represent the mean ± SEM. WT: n = 17-23 cells 1072 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 48 from N = 10-11 mice; Shank3 KO: n = 16-24 cells from N = 13-18 mice. Statistical significance was 1073 determined using an unpaired t-test for B, D, and F , and a Mann-Whitney test for C and H. *p < 0.05 1074 indicates a significant difference between WT and KO; ns: not significant. 1075 1076 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 49 Figure 4 1077 Figure 4: Loss of Shank3 increases the total AMPAR-mediated response in CGC following 1078 glutamate uncaging. (A) Schematic diagram of glutamate uncaging experiment, illustrating brief 1079 exposure of cerebellar slice to blue LED light for photolytic cleavage of Rubi-glutamate supplied in 1080 ACSF (0 mM Mg 2+) while recording from a CGC at -70 mV. ( B) Representative current traces of 1081 combined AMPA + NMDA (red in WT, light red in Shank3 KO) current recorded upon light exposure 1082 in the presence of Rubi-glutamate, gabazine, and TTX. Subsequent addition of NBQX isolated NMDA 1083 current (gray). AMPA current trace (black) was obtained by subtracting NMDA component from the 1084 composite AMPA + NMDA current. ( C) Average peak amplitudes of the combined AMPA + NMDA 1085 response. (D) Average peak amplitudes of the AMPAR response. (E) Average peak amplitudes of the 1086 NMDA response. (F) Ratio of average AMPAR to NMDAR amplitudes. ( G) Average current density of 1087 combined AMPA + NMDA response. ( H) Average current density of the AMPA response. ( I) Average 1088 current density of the NMDA response. For panels C-I, individual data points are shown as circles, 1089 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 50 and bars represent the mean ± SEM. WT: n = 13-21 cells from N = 7-8 mice; Shank3 KO: n = 14-22 1090 cells from N = 6-8 mice. Statistical significance was determined using an unpaired t-test for the data 1091 in panels C-E, H, I and a Mann-Whitney test for the data in panels F and G . *p < 0.05 indicates a 1092 significant difference between WT and Shank3 KO; ns: not significant. 1093 1094 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 51 Figure 5 1095 Figure 5: CGCs show inward rectification and an increased proportion of CP-AMPARs in Shank3 1096 KO mice. ( A) Schematic diagram of the cerebellar circuit, illustrating the stimulation of MFs and 1097 recording a CGC in the presence of intracellular spermine. ( B) Representative current traces of 1098 AMPAR-mediated response from CGCs of WT and Shank3 KO at -60 and +60 mV. ( C) Normalized 1099 current-voltage (I-V) graph showing the eEPSC am plitude. Data points represent the mean ± SEM, 1100 normalized to the current at -60 mV. ( D) Rectification index values for CGCs from WT and Shank3 KO. 1101 (E, F ) Example ( E) and group average ( F) AMPAR-mediated EPSC responses before (ACSF) and 1102 during the IEM-1460 application, normalized to the average baseline response recorded over 5 min 1103 before drug application. ( G) Percentage of baseline response calculated from the average of the last 3 1104 min of recording in IEM-1460 from both genotypes. In panels D and G, individual data points are shown 1105 as circles, and bars represent the mean ± SEM. WT: n = 10-13 cells from N = 6-7 mice; Shank3 KO: n 1106 = 11-13 cells from N = 3-7 mice. Statistical significance was determined using an unpaired t-test for the 1107 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 52 data in panels B, D, and F. *p < 0.05 indicates a significant difference between WT and Shank3 KO. 1108 Abbreviations: ML: Molecular layer, PCL: Purkinje cell layer, GCL: Granule cell layer 1109 1110 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint 53 Figure 6 1111 Figure 6: Shank3 KO mice showed a reduced IBA1- stained fluorescence area in th1112 cerebellum. (A, B) Representative images of immunolabeled microglia with IBA1 cerebellar granu1113 cell layer of sagittal cerebellar sections from WT (+/+) and Shank3 KO (-/-) mice. ( A′ , B1114 Corresponding images showing the surface area occupied by IBA1- positive microglia (outlined 1115 yellow and magenta) from the cerebellar granule cell layer presented in panels A and B, respectivel1116 (C) Violin plot of the quantification of surface area covered by individual IBA1- stained microglia in W1117 and Shank3 KO mice. Individual data points represent the surface area of each IBA1- positiv1118 microglial cell; bars represent the median. ( D) Quantification of the total nu mber of microglia pe1119 image field (265 × 265 µm) in WT and Shank3 KO mice. Individual data points represent values fro1120 each image field, and bars indicate mean ± SEM. ( E) A histogram showing the frequency distributio1121 of the total surface area covered by IBA1-stained microglia in WT and Shank3 KO mice. A total of 51122 images were analyzed per genotype (26-29 microglia per genotype). N = 3-5 mice/genotype1123 Statistical significance was assessed using unpaired t-tests in panels C and E . * indicates p < 0.01124 ns: not significant. 1125 the nule B ′ ) d in ely. WT itive per rom tion 5-6 ype. .05; .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 1, 2025. ; https://doi.org/10.1101/2025.08.01.668222doi: bioRxiv preprint