To evaluate how different ways of impairing Trio could affect mammalian brain development and function, we used CRISPR/Cas9 technology to generate mice bearing heterozygous Trio variant alleles in different Trio functional domains (Figure 1A and B). We chose these alleles for their discrete and measurable effects on TRIO levels or GEF activity in vitro: K1431M impairs TRIO GEF1 interaction with and subsequent activation of Rac1 in vitro up to eightfold (Figure 1—figure supplement 1A and B; Katrancha et al., 2017; Sadybekov et al., 2017), M2145T TRIO GEF2 has a reduced ability to activate RhoA as a function of protein concentration in cells (Katrancha et al., 2017), and K1918X is predicted to lead to nonsense-mediated decay and loss of Trio protein (Katrancha et al., 2017). These Trio variant alleles are found heterozygous in individuals with NDDs: +/K1431M is associated with ASD, +/K1918X is associated with SCZ, and +/M2145T is a de novo mutation found in an individual with BPD (Bircher and Koleske, 2021).

Because Trio knockout mice exhibit decreased survival (O’Brien et al., 2000; Peng et al., 2010; Zong et al., 2015), we first verified the viability of mice bearing these CRISPR/Cas9-generated Trio alleles. Mice heterozygous for any one of these three Trio alleles survived to adulthood. Mice homozygous for Trio K1431M and M2145T survived to adulthood, with genotypes from offspring of +/K1431M intercrosses observed in Mendelian ratios, but fewer than expected M2145T homozygotes obtained from +/M2145T intercrosses (Figure 1—figure supplement 1C). K1918X homozygote pups were not observed (Figure 1—figure supplement 1C), as expected for a Trio null allele (O’Brien et al., 2000; Zong et al., 2015). We focused on heterozygotes, as most rare damaging Trio variants are heterozygous in humans.

Alternative splicing of the full-length TRIO transcript generates multiple brain-specific isoforms of various sizes, whose expression level varies by brain region and developmental age: cortex-predominant Trio9S (263 kDa) and Trio9L (277 kDa), brain-wide Duet (145 kDa), and cerebellum-specific Trio8 (217 kDa; Figure 1A; McPherson et al., 2005; Portales-Casamar et al., 2006). Heterozygosity for the K1431M and M2145T alleles did not alter the levels of the predominant Trio isoforms in the brain at postnatal day 0 (P0, neonate; Figure 1C–D) or at P42 (young adult; Figure 1—figure supplement 1D–L). Trio9 protein levels were reduced by ~50% in the brains of +/K1918X mice at P0 (Figure 1C–D) and P42 (Figure 1—figure supplement 1D–L), and we did not detect the presence of residual truncated protein (expected at 217 kDa), suggesting this K1918X mutation indeed leads to nonsense-mediated decay as predicted. Meanwhile, as the K1918 site is not contained in these isoforms, the levels of Trio8 and Duet were unaffected (Figure 1—figure supplement 1D–L).

Having demonstrated that heterozygosity for the K1431M and M2145T alleles does not alter Trio protein levels in the brain, we can attribute any observed differences to the effects of these missense mutations on Trio GEF domain function rather than changes in Trio protein expression. Given the effects of these variants on TRIO GEF1/2 activities in vitro, we measured levels of active GTP-bound Rac1 and RhoA in brains of neonatal (P0) and young adult (P42) Trio variant-bearing mice (Figure 1E–H).

In concordance with the ~50% reduction in Trio levels in +/K1918X brains (Figure 1C and D), Rac1 activity was decreased in P0 +/K1918X brains (91% of WT activity, p<0.05; Figure 1E), with a trend toward decreased active RhoA levels (84% of WT, p=0.0865; Figure 1F). However, by P42, +/K1918X brains did not differ from WT in Rac1 or RhoA activity (Figure 1G and H), despite the persistent ~50% reduction in Trio9 protein level at this age (Figure 1—figure supplement 1D–F and H–J).

As K1431M decreases TRIO GEF1 catalytic activity in vitro (Figure 1—figure supplement 1A and B; Sun et al., 2021; Katrancha et al., 2017; Sadybekov et al., 2017), we anticipated Rac1 activity would also be reduced in +/K1431M mice. Instead, active Rac1 levels were significantly elevated in +/K1431M whole brain lysates compared to WT controls, showing a 111% increase at P0 (p<0.01; Figure 1E) and an even greater increase of 150% at P42 (p<0.05; Figure 1G). Despite evidence of M2145T impacting TRIO GEF2 catalytic activity in vitro (Katrancha et al., 2017), we did not detect changes in active RhoA levels at the whole brain lysate level in +/M2145T mice at either age (Figure 1F and H).

Because Trio, Rac1, and RhoA are enriched at synapses (O’Neil et al., 2021; Paskus et al., 2019; Katrancha et al., 2017; Oevel et al., 2024), we also measured GTPase activities in synaptosomes (Figure 1—figure supplement 1M–P), within the same age range as the electrophysiological recordings performed on these mice. Active Rac1 was significantly increased in +/K1431M crude synaptosomes from P42 cortex (112% of WT, p<0.01; Figure 1I), consistent with our findings in +/K1431M whole brain lysates (Figure 1E and G). We also measured decreased levels of active RhoA in P42 +/M2145T cortical synaptosomes compared to WT (73% of WT, p<0.01; Figure 1J), consistent with its reduced GEF2 activity in vitro (Katrancha et al., 2017; Figure 1F and H).

NDDs affect learning and memory, compulsivity, motor coordination, and social skills; hence, we assessed these skills using a diverse array of established behavioral tests (Katrancha et al., 2019) in P42 mice bearing Trio variants (Figure 2A).

We measured motor coordination and learning using an accelerating rotarod test. +/K1431M and +/K1918X mice of both sexes fell from an accelerating rotarod with reduced latency relative to WT littermates, while +/M2145T mice performed similarly to WT (Figure 2B). In addition, +/K1431M and +/K1918X males and +/K1431M females showed a lower rate of improvement in this skill over repeated trials (Figure 2B). No deficits in muscle strength were noted in any genotype using the Kondziela inverted screen test prior to rotarod testing.

We used an open field test (OFT) and the elevated plus maze (EPM) to assess general locomotor activity and exploratory behavior. In the OFT (Figure 2—figure supplement 1A–D), Trio variant mice did not differ significantly from WT in total distance traveled or mean speed; however, +/K1431M females spent a greater amount of time in the outer and corner zones of the OFT compared to WT (Figure 2—figure supplement 1D, right), suggesting increased anxiety-like behavior in +/K1431M female mice. WT mice prefer the protected closed arms over the open arms in the EPM (Figure 2—figure supplement 1E and F), whereas +/K1431M and +/K1918X mice of both sexes and +/M2145T females did not exhibit this preference (Figure 2—figure supplement 1F).

Deficits in social communication and social interaction are defining features of individuals diagnosed with NDDs (American Psychiatric Association, 2013). In a three-chamber social interaction test, +/K1431M mice of both sexes and +/K1918X females showed no preference for the stranger mouse over the object (Figure 2C), while +/M2145T mice of both sexes and +/K1918X males exhibited preference for the stranger mouse, similar to WT (Figure 2C). In a novel object recognition test, +/K1918X mice of both sexes and +/M2145T females failed to discriminate between the novel and familiar objects, while +/K1431M mice and +/M2145T males exhibited normal discrimination between novel and familiar objects similar to WT (Figure 2D).

Repetitive behaviors and stereotypies are often identified in individuals with NDDs (Ridley, 1994). We found a significant increase in compulsive nestled shredding in +/K1918X and +/K1431M males, with no significant changes in females (Figure 2E).

Behavioral phenotypes mostly overlapped between +/K1431M and +/K1918X mice, with differences in social interaction and memory tests. +/M2145T mice exhibited the fewest phenotypes in measured tasks. Additionally, some behavioral manifestations appeared to be selective to one sex. Together, these data indicate that these Trio alleles differentially impact behavior.

TRIO variants that reduce TRIO GEF1 activity are associated with microcephaly (Barbosa et al., 2020; Bonnet et al., 2023; Bircher et al., 2022), so we assessed head and brain size in Trio variant mice (Figure 3A–E [males], Figure 3—figure supplement 1A–E [females]). After adjusting for body weight, both the head width (Figure 3D, Figure 3—figure supplement 1D) and brain weight (Figure 3E, Figure 3—figure supplement 1E) were reduced in P42 adult +/K1431M mice of both sexes relative to WT, consistent with the microcephaly seen in patients harboring TRIO GEF1-deficient alleles. Of note, these ratios were driven by the increased body weight exhibited by P42 +/K1431M mice of both sexes compared to WT counterparts (Figure 3C, Figure 3—figure supplement 1C). +/K1918X mice of both sexes displayed a reduced brain weight-to-body weight ratio (Figure 3E, Figure 3—figure supplement 1E) but an unchanged head width-to-body weight ratio (Figure 3D, Figure 3—figure supplement 1D) relative to WT, suggesting a disproportionate reduction in brain mass in +/K1918X mice. Meanwhile, +/M2145T mice did not differ from WT in either head width or brain weight when adjusted for body weight.

We examined whether the decreased brain weight and head size observed in +/K1431M and +/K1918X mice were associated with anatomical defects. Gross histological analyses of fixed Nissl-stained brain sections in P42 male mice showed only mild morphological defects (Figure 3F–J). We observed reductions in both total cross-sectional brain area (Figure 3F–G) and cortical thickness (Figure 3H–I) in +/K1918X brains, consistent with their smaller head size and brain weight. Significant decreases in cortical layer 2/3 and layer 5 thickness were observed in +/K1918X brains (Figure 3H–J) and were in proportion to the relative decrease in cortical thickness relative to WT (Figure 3—figure supplement 1F). Meanwhile, no significant differences in brain area or cortical thickness were observed in either +/K1431M or +/M2145T mice.

Deletion of Trio has been shown to impair the migration of forebrain interneuron progenitors, resulting in fewer neurons entering the cerebral cortex and altered distribution of cortical layers (Sun et al., 2021; Wei et al., 2022). However, we observed no change in total cortical cell density or in layer-specific cell density in heterozygous Trio variant male mice, nor were numbers of cortical NeuN +neuronal cells or PV +inhibitory neurons altered relative to WT, although there were trends toward increased DAPI + cell density in +/K1918X and increased NeuN + cell density in +/M2145T motor cortex (Figure 3—figure supplement 1H–N). These data suggest that the reduced brain size of +/K1918X mice results from a loss of neuropil rather than reductions in cortical neuron number.

Altered dendritic arbor morphology and dendritic spine abnormalities are hallmarks of NDDs (Kaufmann and Moser, 2000; Huttenlocher, 1970; Huttenlocher, 1974; Huttenlocher, 1991; Purpura, 1974; Purpura, 1975; Kulkarni and Firestein, 2012). Excitatory neuron-specific ablation of one or both Trio alleles decreased dendritic arborization, increased spine density, and yielded smaller synapses in cortex area M1 Layer 5 pyramidal neurons (M1 L5 PNs; Katrancha et al., 2019).

Sholl analysis of M1 L5 PNs revealed significant reductions in both basal and apical dendritic arbor complexity and dendritic field size in P42 +/K1918X neurons, an increase in proximal basal arbor complexity in +/K1431M neurons, and a decrease in distal apical arbor complexity and apical dendritic field size in +/M2145T neurons relative to WT (Figure 3K–O, Figure 3—figure supplement 1).

Conditional Trio knockout specifically in forebrain excitatory neurons significantly increased the dendritic spine density on M1 L5 PNs, although they were smaller and appeared more immature (Katrancha et al., 2019). Notably, none of the Trio variant heterozygotes exhibited altered dendritic spine density on M1 L5 pyramidal neurons compared to WT mice on either apical or basal secondary arbors (Figure 3—figure supplement 2L and M).

Electron microscopy of cortical area M1 L5 revealed that synapse density was significantly increased in +/K1918X mice compared to WT (Figure 4A and B), possibly due to a net reduction in neuropil resulting from smaller dendritic arbors. Within synapses, postsynaptic density (PSD) length was slightly decreased in +/K1918X (by 6%) and +/M2145T (by 6.6%) mice relative to WT, but not in +/K1431M mice (Figure 4C). Cross-sectional presynaptic bouton area and spine head area were similar to WT in each Trio variant heterozygote (Figure 4D and E).

Synaptic vesicle (SV) distribution was significantly altered in +/M2145T mice relative to the other genotypes (Figure 4F and G). The densities of both docked and tethered vesicles at 15 and 50 nm from the active zone (AZ) respectively, which together can estimate the readily releasable pool (RRP) (Montesinos et al., 2015; Dong et al., 2018), were significantly increased in +/M2145T relative to the other genotypes (Figure 4F and G). Also, SV distribution at 200 nm from the AZ, thought to contribute the reserve pool, was also significantly increased in +/M2145T mice, suggesting an overall larger SV pool size. No differences in synaptic vesicle distribution were noted in +/K1431M mice, while +/K1918X showed modestly increased SVs at 200 nm from AZ relative to WT.

Loss of Trio function or disruption of TRIO GEF1 activity in slice culture decreases AMPAR levels at excitatory synapses (Katrancha et al., 2019; Tian et al., 2021; Paskus et al., 2019; Sadybekov et al., 2017), while mice bearing a GEF1-deficient Trio allele exhibited decreased gamma-aminobutyric acid receptor (GABAR)- and glycine receptor (GlyR)-mediated inhibitory miniature current (mIPSC) frequencies in the prefrontal cortex (Sun et al., 2021). To explore how Trio variants impact synaptic function, we measured both miniature excitatory currents (mEPSCs) and mIPSCs in M1 L5 PNs in each of the Trio variant heterozygotes at P35-42 consistent with other experiments.

AMPAR-mediated mEPSC amplitudes were significantly increased in +/K1431M and +/K1918X mice relative to WT littermates, with no change in their frequencies (Figure 5A–C). In contrast, AMPAR-mediated mEPSC amplitudes were unchanged in +/M2145T mice, but their frequencies were significantly increased. No significant changes in NMDAR mEPSC amplitudes were noted, while NMDAR mEPSC frequencies were decreased in +/K1431M and increased in +/M2145T mice cortex (Figure 5D–F). Notably, these findings correlated with gross alterations in the ratio of NMDAR/AMPAR-mediated evoked (e)EPSCs measured in M1 L5 PNs following stimulation in L2/3. +/K1431M and +/K1918X Trio variant heterozygotes showed decreased NMDA/AMPA ratios, indicating imbalances in NMDAR- versus AMPAR-mediated conductance (Figure 5J and K) and suggesting an increase in synaptic AMPAR signaling in both mice. Significant decreases in mIPSC frequencies were noted in +/K1431M and +/M2145T mice relative to WT mice, with no change in amplitudes (Figure 5G–I). +/K1918X mice exhibited increased mIPSC amplitude with no observed change in mIPSC frequency. Together, these data indicate that the Trio variants differentially impact excitatory and inhibitory transmission.

Finally, we tested the ability of the L5 PNs to undergo long-term potentiation (LTP) following theta-burst stimulation of L2/3 afferents (Figure 5L and M). While LTP was robustly induced and potentiated in M1 L5 PNs from WT mice, LTP induction and potentiation were deficient in slices from +/K1918X and +/K1431M mutant mice. In contrast, +/M2145T L5 PNs showed increased induction and prolonged potentiation of LTP compared to WT L5 PNs.

In addition to its postsynaptic roles, Trio localizes presynaptically and interacts with the presynaptic active zone scaffolding proteins Bassoon and Piccolo (Terry-Lorenzo et al., 2016; O’Neil et al., 2021; Paskus et al., 2019). Additionally, recent work demonstrates that Rac1 activity levels can bidirectionally affect SV probability of release (Pr) in excitatory synapses (O’Neil et al., 2021; Keine et al., 2022; Broadbelt et al., 2002). To assess the impact of Trio variants on presynaptic function, we first measured the paired-pulse ratio (PPR) to test possible changes in Pr. Synapses with a low probability of release (Pr), such as L2/3 onto L5 PNs synapses, exhibit paired-pulse facilitation (PPF), where synaptic response increases for the second of two apposed stimuli due to elevated residual Ca²⁺ promoting SV fusion. WT and +/K1918X M1 L5 PNs exhibited normal facilitation of eEPSC amplitudes that decreased with increased interstimulus interval (ISI) (Figure 6A and B). PPF was significantly enhanced in +/K1431M M1 L5 PNs at short ISIs relative to WT, suggesting a reduction in Pr at these synapses (Figure 6A and B). The +/M2145T PPR curve was complex, with significantly reduced PPF at short ISIs, yet clearly increased PPF at longer ISIs compared to WT (Figure 6A and B). The decreased PPF at initial ISIs in +/M2145T mice, together with an increase in both AMPAR and NMDAR mEPSC frequency (Figure 5A–C), demonstrates an increase in both synchronous and spontaneous glutamate Pr.

To further characterize changes in neurotransmitter release in these mice, we used high-frequency (40 Hz) stimulation (HFS) trains to quantitatively estimate glutamate Pr, RRP size, and rates of SV depletion and recovery. A plot of the normalized eEPSC responses to a HFS train stimulation again revealed facilitation upon the first 2–3 stimulations that was increased in +/K1431M and decreased in +/M2145T slices relative to WT, with no changes in +/K1918X slices (Figure 6C and D). Initial facilitation was followed by decaying eEPSC amplitudes, reflecting SV depletion under HFS. +/K1431M and +/M2145T Trio variants exhibited a slower train decay rate relative to WT during HFS with +/M2145T depleting at half the rate of WT (τ_d, +/K1431M: 3.19 s, +/M2145T: 4.79 s vs WT: 2.70 s; Figure 6D).

We used a ‘Decay’ method (Ruiz et al., 2011; Thanawala and Regehr, 2016) to estimate Pr and RRP size from HFS trains, which allows us to account for initial facilitation seen in the train eEPSCs. Glutamate Pr in the L2/3-L5 synapses was increased in +/M2145T mice, while it was decreased for +/K1431M mice (Figure 6E), consistent with the relative changes observed in PPF for these mice. RRP size was much larger in L2/3-L5 synapses of +/M2145T mice relative to WT (Figure 6F), consistent with the increased SV distribution found at 15–50 nm from the AZ in electron micrographs (Figure 4F and G). We tested the ability of Trio variant heterozygotes to recover after train depletion by pairing HFS train with a single stimulus at increasing intervals (0.01, 2, 6, 9, 12, 18 s) and calculating the fractional recovery (see Materials and methods). The recovery rate (τ_R) was significantly slower in +/K1431M L5 PNs, and they did not recover to their initial strength within 18 s, plateauing at 78% of maximal recovery compared to WT (Figure 6G). Together, Trio +/K1431M and +/M2145T mice exhibit distinct and significant alterations in short-term plasticity and synchronous glutamate release.

We used comparative proteomics from P21 cortex to identify proteins and pathways that were differentially altered by the Trio variants. We quantified a total of 7,362 proteins, finding distinct differences in the cortical proteome for each genotype (Figure 7—figure supplement 1, Figure 7—figure supplement 1—source data 1). Gene Set Enrichment Analysis (GSEA; Subramanian et al., 2005; Mootha et al., 2003) revealed alterations in distinct functions for each Trio variant (Figure 7A; Figure 7—source data 1). Of note, the only enriched gene set specific to neurons was downregulation of the synaptic vesicle pathway in +/K1431M cortex; all other gene sets were not cell-type specific.

We used SynGO (Koopmans et al., 2019) to investigate whether the Trio variants impacted synaptic functions. 1067 of the 7362 total quantified proteins were synaptic proteins listed in the SynGO gene set (Figure 7, Figure 7—source data 2). When restricted to brain-specific genes, all three Trio variant heterozygotes showed enrichment of differentially expressed proteins (DEPs) in synaptic processes (Figure 7B and C). Notably, +/M2145T upregulated DEPs and +/K1431M downregulated DEPs were significantly enriched at the presynapse (Figure 7B), but with only +/K1431M downregulated DEPs showing significant enrichment for postsynaptic receptor and synaptic vesicle cycling (Figure 7C). Meanwhile, +/K1918X showed enrichment of DEPs at the postsynapse, particularly in the postsynaptic density (Figure 7B). Together, our proteomics data point to a significant deficit in presynaptic function in both +/K1431M and +/M2145T cortex, as well as a significant effect of +/K1431M on postsynaptic function.

Given our findings from proteomic analysis and electrophysiology, we measured levels of key presynaptic regulators, including synaptophysin (Syp), syntaxin binding protein 1 (Stxbp1, also known as Munc18-1), syntaxin 1a (Stx1a) and synaptotagmin 3 (Syt3), which are crucial for synaptic vesicle (SV) tethering, docking, replenishment, and calcium-dependent replenishment, respectively. We also measured levels of presynaptic proteins in P42 cortical synaptosomes, where these proteins are enriched. Munc18-1, Syt3, and Syp levels were increased in +/M2145T synaptosomes relative to WT (Figure 7D–H). Meanwhile, Stx1a levels were significantly decreased in +/K1431M synaptosomes compared to WT, with no significant changes in +/K1918X compared to WT mice.

The elevated Rac1 activity in +/K1431M brain lysates and synaptosomes (Figure 1I) seemed at odds with previous reports that K1431M reduces TRIO GEF1 activity (Figure 1—figure supplement 1A and B; Sun et al., 2021; Katrancha et al., 2017; Sadybekov et al., 2017). We hypothesized that homeostatic compensation in +/K1431M mice may alter expression of other RhoGEFs and GAPs. Indeed, levels of the Rac1 GEF Tiam1 were increased in both +/K1431M and +/M2145T P42 cortical lysates, while VAV2 levels were increased in +/M2145T P42 lysates (Figure 7I–L). Levels of the Trio paralog Kalirin (Yan et al., 2015) were unaffected in the Trio variant mice at P42 (Figure 7J). Together, our proteomic analyses suggest that presynaptic functions are altered in +/K1431M and +/M2145T mice and may be driven by abnormal levels of crucial presynaptic regulatory proteins and changes in Rac1 and RhoA activity.

Rac1 negatively regulates synaptic vesicle replenishment and synaptic strength in excitatory synapses (O’Neil et al., 2021; Keine et al., 2022). In +/K1431M synapses, increased Rac1 activity and decreased Stx1a levels were associated with reduced synaptic strength and impaired vesicle replenishment. We tested if the acute application of the Rac1 inhibitor NSC23766 (NSC) could rescue these deficits. Treatment of +/K1431M and WT slices acutely with NSC shifted the PPF downwards in M1 L5 PNs, suggesting an increase in Pr in both cases (Figure 8A and B). Notably, at longer ISI, PPRs in NSC-treated WT slices still normalized at around 1, while PPRs in NSC-treated +/K1431M slices shifted below 1, exhibiting slight depression.

Application of NSC to WT and +/K1431M slices under HFS train stimulation decreased initial facilitation in both (Figure 8C and D), and significantly increased Pr (Figure 8E) without affecting RRP size (Figure 8F) in both, rescuing the initial Pr in +/K1431M slices (Figure 8C–F). NSC treatment also led to a faster train decay rate (τ_d) in +/K1431M slices but did not significantly change WT relative to untreated slices (+/K1431M+NSC τ_d, 2.66 s vs. initial 3.19 s; WT +NSC: 2.85 vs 2.70 s; Figure 8D).

Finally, we tested if acute Rac1 inhibition impacts the rate of recovery of the RRP following HFS stimulation. NSC treatment increased the fractional recovery rate (τ_R) in both WT and +/K1431M slices (WT +NSC: 32% faster than WT; +/K1431M+NSC: 40% faster than +/K1431M). NSC treatment of the +/K1431M variant allowed for full recovery at 18 s interval and sped up the recovery rate, but it remained significantly slower compared to WT (by 45%; Figure 8G).

Overall, we demonstrate that presynaptic Trio GEF1-dependent Rac1 signaling is crucial for maintaining synchronous glutamate Pr and SV replenishment at cortical L2/3-L5 synapses.

Individuals with mutations in TRIO present with a range of NDD-associated clinical features, including varying degrees of intellectual disability, altered head size, skeletal and facial features, and behavioral abnormalities (Ba et al., 2016; Barbosa et al., 2020; Gazdagh et al., 2023; Pengelly et al., 2016; Bircher et al., 2022; Bonnet et al., 2023; Schultz-Rogers et al., 2020). Patients with missense or truncating variants in TRIO that reduce GEF1 activity have mild developmental delay and microcephaly (Barbosa et al., 2020; Bonnet et al., 2023; Pengelly et al., 2016; Bircher et al., 2022). Similarly, we found that heterozygosity for the GEF1-deficient K1431M missense or the K1918X nonsense variants significantly reduced brain weight and/or head size compared to WT mice, along with multiple behavioral impairments. Notably, while both showed impaired motor coordination and learning, only mice bearing the ASD-associated K1431M allele exhibited social interaction deficits. In addition, we observed behavioral differences in male versus female Trio variant mice, possibly similar to human sex differences in the susceptibility to and clinical presentation of NDDs (Bölte et al., 2023).

Both +/K1431M and +/K1918X adult mice of both sexes had reduced brain-to-body weight ratios compared to WT, but these were driven by different factors. The smaller brain size in +/K1918X male mice was associated with a reduction in neuropil and reduced cortical thickness, similar to mice bearing excitatory neuron-specific ablation of one Trio allele (Katrancha et al., 2019) and paralleling the reduced gray matter volume and cortical thickness in SCZ patients (Howes et al., 2023; Dabiri et al., 2022; Harvey et al., 1993; Suddath et al., 1989; Zipursky et al., 1992). Meanwhile, +/K1431M mice of both sexes exhibited an overall increase in body weight leading to relatively decreased head width- and brain-to-body weight ratios in these mice. Adult +/K1431M male mice had no change in neuropil or cross-sectional brain area, consistent with a prior study describing normal brain size in +/K1431M at E14.5 (Sun et al., 2021). Rac1 mediates glucose-stimulated insulin secretion from pancreatic islet beta-cells (Kowluru, 2011; Kowluru, 2021; Zhou et al., 2015; Veluthakal and Thurmond, 2021; Asahara et al., 2013; Sylow et al., 2016), which may explain how chronic alterations in Rac1 activity contribute to weight changes in Trio +/K1431M and +/K1918X mice. Most studies of Trio variants have focused on neuronal effects, but expression of Trio in other tissues could explain the increased body weight in these mice, as well as the musculoskeletal abnormalities associated with TRIO variation in humans (Ba et al., 2016; Barbosa et al., 2020; Gazdagh et al., 2023; Kloth et al., 2021).

Rac1 and RhoA signaling is critical for dendrite development (Hall, 1998; Luo, 2000; Nakayama et al., 2000; Newey et al., 2005; Ba et al., 2013). We found relatively subtle effects of Rac1/RhoA-altering Trio variants on cortical L5 PN dendrites. +/K1918X L5 pyramidal neurons are smaller and less complex than WT neurons, especially in the basal compartment corresponding to L5 where EM images were obtained, consistent with the smaller brain size and reduced cortical thickness of +/K1918X mice. We posit that due to their smaller dendritic field size, L5 neurons pack more densely, contributing to the mildly increased synapse density observed in +/K1918X M1 L5 cortex. Consistent with this hypothesis, we observed a trend toward increased DAPI + cell density in M1 L5 of +/K1918X neurons.

The reductions in dendritic arborization and length in +/K1918X neurons are consistent with reports of reduced gray matter volume and dendrite alterations in individuals with SCZ versus controls (Broadbelt et al., 2002; Black et al., 2004), but were modest compared to NEX-Trio^+/fl mice lacking one copy of Trio in excitatory neurons (Katrancha et al., 2019). Reduced Trio function in other cell types, such as in inhibitory neurons or glia, or in neurons from other brain regions that project to cortical M1 L5 PNs, may ameliorate the phenotypes in excitatory neurons of +/K1918X relative to NEX-Trio^+/fl mice. In addition, changes to L5 PN dendrites in Trio variant mice appeared to be regionally selective within the arbor: +/K1431M neurons had increased arborization only in proximal basal dendrites, while +/M2145T neurons had decreased arborization only in the most distal apical dendrites. These differences may reflect the differential spatiotemporal influence of Trio in regulating Rac1 versus RhoA activity. Alternatively, given our finding for presynaptic roles for Trio, these differences may reflect differential effects of the Trio variants on both excitatory and inhibitory afferent synaptic inputs, which play critical roles in shaping the apical and basal dendrites of L5 PNs (Ramaswamy and Markram, 2015).

We show here that the K1431M variant significantly reduces TRIO GEF1 nucleotide exchange on Rac1 in vitro, consistent with previous reports (Katrancha et al., 2017; Sadybekov et al., 2017). Decreased Rac1 activity was observed in +/K1431M mice at embryonic day 14.5 in the ganglionic eminence, which is enriched for pre-migratory GABAergic interneurons (Sun et al., 2021). In contrast, we show an increase in active Rac1 levels in +/K1431M postnatal brains (at P0, P21, P42) and synaptosomes (at P42) from cortex, primarily composed of differentiated excitatory neurons. This increase aligns with observed phenotypes (e.g. reduced Pr) and reveals why reductions in brain volume, dendrites and spines, or AMPAR signaling anticipated from reduced Rac1 activity (Chen et al., 2009; Reijnders et al., 2017; Tahirovic et al., 2010; Pennucci et al., 2019; Wiens et al., 2005) were not observed in +/K1431M mice. We propose that the increased Rac1 activity we observed reflects homeostatic compensation in Rac1 regulation occurring between birth to adult ages and identify changes in Rac1-specific GEFs, for example Tiam1 and Vav2, that may contribute to this compensation. Importantly, we did not find changes in Kalirin levels in the adult brain of these Trio mutant mice, suggesting that Kalirin does not compensate for the loss of Trio GEF1 activity at this age.

We observed a significant reduction in active RhoA only in purified synaptosomes of +/M2145T brains, reflecting the synaptic compartment as a key locus of Trio function. In addition, despite +/K1918X mice having half the WT levels of Trio protein, we measured little to no change in active Rac1 and RhoA levels in +/K1918X brains. These findings are consistent with recent evidence that the spatiotemporally precise balance of Rac/Rho activity rather than absolute activity levels can mediate cytoskeletal rearrangements (Grubisha et al., 2022; Duman et al., 2015; Ba and Nadif Kasri, 2017; Pertz, 2010). Alterations in the activity of additional potential TRIO substrates, such as RhoG (Bellanger et al., 2000), Cdc42 (Blangy et al., 2000), and the neurodevelopmentally critical Rac3 (Scala et al., 2021; Hajdo-Milasinović et al., 2007; Corbetta et al., 2009), could also contribute to phenotypes in these mice.

Our results strongly complement a growing body of evidence showing that altered activation of Rho GTPases is an important mechanism affecting synaptic function in ASD (Bölte et al., 2023; Pinto et al., 2010; Zeidán-Chuliá et al., 2013; Guo et al., 2020; Carbonell et al., 2023; Zamboni et al., 2018). This highlights Rho GTPase signaling as a convergent pathway in ASD and therefore an attractive target for pharmacotherapy for these disorders.

Overexpression of a TRIO K1431M variant with reduced GEF1 activity decreased AMPAR-mediated mESPC amplitudes in rat organotypic slices (Tian et al., 2021), while Rac1 activation increased AMPAR amplitudes by promoting synaptic AMPAR clustering (Wiens et al., 2005; Hussain et al., 2015). In +/K1431M mice, we observed a decrease in the NMDA/AMPA ratio and NMDA mEPSCs frequency, with increased AMPAR mEPSCs amplitudes, with no changes in dendritic spines suggesting a possible reduction in silent synapses due to increased AMPAR incorporation. This increase in AMPAR was associated with increased active Rac1 levels measured in synaptosomes in adult +/K1431M mice. AMPAR mEPSC amplitudes were also increased in L5 PNs of +/K1918X mice, corresponding with a decrease in the NMDA/AMPAR ratio, but no changes in mEPSC frequencies, also suggesting an increased number of synaptic AMPARs. In +/K1918X cortex, Rac1 activity is reduced at P0 but increases by P21 and P42 to WT levels, suggesting that a relative increase in Rac1 during development may enhance AMPAR tone and affect LTP, especially given the precise spatiotemporal requirements of Rac/Rho activity and their effectors for LTP (Wiens et al., 2005; Hussain et al., 2015; Cui et al., 2021; Herring and Nicoll, 2016; Duman et al., 2022). The +/M2145T variant showed no changes in AMPAR or NMDAR amplitudes but had increased mEPSC frequencies in both, without changes in spine or synapse density, consistent with enhanced spontaneous neurotransmitter release.

mIPSC frequencies were decreased in +/K1431M and +/M2145T L5 PNs, while mIPSC amplitude was increased in +/K1918X slices. Sun et al. noted a similar deficit in inhibitory function in +/K1431M prefrontal cortex correlated with reduced interneuron migration to this region, including parvalbumin positive (PV+) neurons (Sun et al., 2021). In contrast, we did not observe reduced PV +interneuron numbers in the motor cortex in any Trio variant mice, suggesting that Trio variants may also impact the number of inhibitory synapses or transmission. However, we cannot fully exclude potential defects in the migration of other interneuron populations as contributors to these phenotypes.

Overall, heterozygosity for Trio variants dysregulates excitatory and inhibitory synaptic transmission in different patterns, resulting in E/I imbalance, a known driver of NDD phenotypes.

Trio-deficient excitatory neurons are unable to undergo long-term potentiation (LTP) in mouse brain slices (Katrancha et al., 2019), which is crucial for working memory in mammals. During LTP, Rac1 is suggested to be transiently activated and deactivated to regulate AMPAR (Wiens et al., 2005; Hussain et al., 2015; Cui et al., 2021; Herring and Nicoll, 2016; Duman et al., 2022). Both +/K1431M and +/K1918X L5 PNs exhibited reduced LTP induction and maintenance. The increased AMPAR resulting from elevated Rac1 activity (+/K1431M mice) or the inability of reduced levels of Trio to activate Rac1 (+/K1918X mice) may preclude LTP in these mice. +/M2145T mice showed a striking increase in the induction and maintenance of LTP, correlating with an increased glutamate Pr and SV pool size. The function of RhoA in plasticity is unknown, but the decrease in RhoA activity and increased levels in presynaptic machinery proteins measured in cortical synaptosomes may underlie the increase in LTP in these mice.

In neuroendocrine, pancreatic beta, and mast cells, Trio GEF1 activity, Rac1, and RhoA are required for regulation of exocytosis (Ferraro et al., 2007; Hong-Geller and Cerione, 2000; Lecuona et al., 2003; Momboisse et al., 2009; Pathak et al., 2012). Recent work suggests that Trio GEF1 can act through Rac1 to regulate presynaptic processes – Rac1 colocalizes with SVs in the axonal boutons to negatively regulate action potential-dependent (synchronous) glutamate Pr and SV replenishment (O’Neil et al., 2021; Keine et al., 2022; Doussau et al., 2000; Banerjee et al., 2024). Using the Rac1 inhibitor NSC, we demonstrate that elevated Rac1 activity, potentially driven by Tiam1 upregulation, drives the reduction in synchronous glutamate Pr and SV replenishment without affecting RRP in +/K1431M mice. NSC application rescues Pr and enhances SV replenishment in +/K1431M slices while also increasing both processes in WT. Decreased Stx1 levels in +/K1431M synapses, a key component of the priming machinery, may still be a limiting factor in effective priming, thus preventing full rescue of the SV recovery by NSC. Overall, NSC had a significant yet weaker effect in WT than in +/K1431M slices, likely due to the overall increased Rac1 activity in +/K1431M slices providing more of a target for NSC.

Conditional knockout of Rac1 in the Calyx of Held increased spontaneous (action potential-independent) glutamate Pr (Keine et al., 2022), aligning with the hypothesis that Rac1 activity negatively regulates neurotransmitter release by increasing the assembly of actin filaments at the AZ that impede SV fusion. +/K1431M cortical slices showed significant increases in AMPAR amplitudes, and our data are consistent with fewer silent synapses. Yet, we observed no change in AMPA mESPC frequency. We hypothesize that the Rac1-dependent decrease in Pr identified in these slices decreased AMPAR mEPSC frequency; hence, the opposing changes from these pre- and post-synaptic effects mask the expected change in AMPAR mEPSC frequency.

RhoA activity is significantly reduced in cortical synaptosomes of +/M2145T mice, and this is associated with increased RRP size. While a specific function for RhoA in regulating presynaptic release is currently unclear, levels of Munc18, Syp, and Syt3 are all increased in +/M2145T mice at P42 and can contribute to the enhanced Pr and altered SV cycling. The unusual PPR in the +/M2145T variant may result, at least in part, from altered Syt3 levels, crucial for calcium-dependent SV replenishment, as its deficiency causes similar phenotypes (Weingarten et al., 2022).

Our findings show that Trio GEF1 and GEF2-mediated Rho GTPase signaling pathways play critical and distinct roles in regulating Pr, RRP size, and SV recycling.

TRIO is a risk gene for several NDDs with different patterns of variants observed in different disorders. We show here that variants in Trio that lead to impaired Trio levels or GEF function cause both shared and distinct defects in behavior, neuroanatomy, and prominent synaptic dysfunction that may reflect variant-specific NDD clinical phenotypes. Our data also demonstrate, for the first time, the differential impact of distinct Trio lesions on glutamate Pr, RRP size, and SVs replenishment, along with alterations in presynaptic release machinery that contribute to these deficits. We demonstrate that the +/K1431M lesion in TRIO GEF1 leads to presynaptic deficits due to Rac1 dysregulation, which can be rescued by pharmacological normalization. We propose that Trio variants disrupt brain function by impairing Rho GTPase signaling, causing diverse synaptic dysfunction through combined effects on pre- and/or post-synaptic functions. Our findings support growing evidence linking synaptic dysfunction in several NDD models to Rho GTPase signaling dysregulation, identifying it as a commonly affected pathway and a potential therapeutic target.

All animal work was performed in compliance with federal guidelines and approved by the Yale Institutional Animal Care and Use Committee. All mice were maintained on a C57Bl/6 background and housed on a standard 12 hr light/dark cycle. Mice heterozygous for Trio variants K1431M (Trio^+/K1431M), K1918X (Trio^+/K1918X), or M2145T (Trio^+/M2145T) were crossed with WT (Trio^+/+) mice to produce heterozygous Trio variant and WT littermates used in all experiments. As the mice were maintained on the same genetic background, analyses of WT mice were pooled and considered one genotype, irrespective of parental Trio variant genotype. Age-matched mice of both sexes were used for behavioral experiments and for brain and body weight measurements. Only male mice were used for electrophysiological and neuroanatomical analyses to reduce potential variation in dendritic spines and activity due to the estrus cycle (Frankfurt and Luine, 2015).

Mice heterozygous for Trio variants K1431M, K1918X, or M2145T were generated via CRISPR/Cas-mediated genome editing (Yang et al., 2014; Chen et al., 2016). Potential Cas9 target guide (protospacer) sequences in the vicinity of the Trio K1431, K1918, and M2145 codons were screened using the online tool CRISPOR (Haeussler et al., 2016), and candidates were selected. Templates for sgRNA synthesis were generated by PCR from a pX330 template (Addgene), and sgRNAs were transcribed in vitro and purified (Megashortscript, MegaClear; Thermo Fisher). sgRNA/Cas9 RNPs were complexed and tested for activity by zygote electroporation, incubation of embryos to blastocyst stage, and genotype scoring of indel creation at the target sites. sgRNAs that demonstrated the highest activity were selected for creating the knock-in alleles. Guide primers for generating the template for transcription included a 5’ T7 promoter and a 3’ sgRNA scaffold sequence and were as follows (protospacer sequence capitalized):

Recombination template oligos (Integrated DNA Technologies, San Diego, CA) were designed to create the desired codon changes, with incorporation of silent mutations to destroy the PAM and prevent sgRNA recognition of the newly created alleles as well as create a new restriction site for genotyping. sgRNA/Cas9 RNP and the template oligo were electroporated into C57Bl/6 J (JAX) zygotes (Chen et al., 2016). Embryos were transferred to the oviducts of pseudopregnant CD-1 foster females using standard techniques (Nagy, 2003). Genotype screening of tissue biopsies from founder pups was performed by PCR amplification and Sanger sequencing to identify the desired base changes, followed by backcrossing to C57Bl/6 mice and sequence confirmation to establish germline transmission of the correctly targeted alleles. Recombination template oligos were as follows (silent mutations bolded and italicized; new restriction digest site underlined):

Genotypes were determined by PCR of mouse DNA using the following primer pairs:

To distinguish Trio variant and WT alleles, PCR was followed by restriction digest at 37 °C for at least 4 hours (BsrGI for K1431M, SphI for K1918X, SalI for M2145T) and subjected to agarose gel electrophoresis.

WT samples resulted in a band at 530 kb for K1431 litters, 550 kb for K1918 litters, 400 kb for M2145 litters; while Trio heterozygous variants resulted in bands at 450 and 530 kb for +/K1431M, 350 and 550 kb for +/K1918X, 350 and 400 kb for +/M2145T. All mice used in experiments were genotyped twice from DNA samples collected at two different time points, prior to weaning and again post-experiment for validation.

The K1431M point mutant was generated via site-directed mutagenesis of human WT TRIO GEF1 using the following oligos: 5’-cagcgaataacgatgtatcagctcc-3’ and 5’-ggagctgatacatcgttattcgctg-3’. Recombinant WT and K1431M human TRIO GEF1 and Rac1 proteins were purified from bacteria as previously described (Blaise et al., 2022). GEF activity was monitored by the decrease in fluorescent signal (λ_excitation = 488 nm; λ_emission = 535 nm) as GTP was exchanged for BODIPY-FL-GDP on Rac1 over 30 min, as previously described (Bircher et al., 2022; Blaise et al., 2022).

Ear-to-ear head width of P42 mice was measured with calipers in anesthetized mice. Brain weight was measured after removing the olfactory bulbs and brain stem.

Whole brain lysates from P0-P1 or P35-P42 mice were prepared as previously described (Katrancha et al., 2019; Hollingsworth et al., 1985; Scheetz et al., 2000), with minor modifications. Cerebella, hippocampi, and cortices were rapidly removed, snap-frozen in liquid nitrogen and stored at –80 °C until lysate preparation. Tissue was homogenized in ice-cold RIPA buffer (1% NP-40, 0.1% SDS, 50 mM Tris pH 8, 150 mM NaCl, 0.5% sodium deoxycholate) supplemented with protease and phosphatase inhibitors (Roche 11873580001; Roche 04906837001), then clarified by brief centrifugation; aliquots were snap frozen and stored at –80 ° C until immunoblotting. For Trio immunoblots, tissue was sonicated in homogenization buffer (1% SDS, 50 mM Tris pH 7.4, 2 mM EDTA), supplemented with protease and phosphatase inhibitors, heated at 95 ° C for 5 min, then clarified, snap frozen, and stored at –80 ° C until immunoblotting. Protein concentrations were determined with a bicinchoninic acid (BCA) assay (Pierce).

Crude synaptosomes were prepared as previously described (Hollingsworth et al., 1985; Scheetz et al., 2000). Briefly, cortices were rapidly dissected from anesthetized P39-P42 mice and immediately homogenized in cold buffer (118 mM NaCl, 4.7 mM KCl, 1.2 mM MgSO₄, 2.5 mM CaCl₂, 1.53 mM KH₂PO₄, 212.7 mM glucose) supplemented with protease and phosphatase inhibitors (Roche). Homogenates were passed through a series of nylon filters of descending pore size: 40 μm, 10 μm, and finally 5 μm. Samples were centrifuged for 15 min at 1000 x g at 4 °C; the supernatant was discarded, and the pellet was resuspended in an appropriate buffer for G-LISA measurements or immunoblotting. Mice from at least 3 litters per genotype were processed concurrently with WT littermates.

Brain lysates were separated on SDS-PAGE gels, transferred to nitrocellulose membranes, stained with Ponceau S, blocked in 5% nonfat milk in TBS-T, and incubated with primary antibodies overnight at 4 °C, then with conjugated secondary antibodies at RT for 1 hr (see Key Resources Table for list of antibodies). Images were captured by a ChemiDoc Imaging System (Bio-Rad) and quantified in ImageJ. Signal intensity was normalized to Ponceau S, then to the WT average.

Active GTP-bound GTPase levels in brain lysates were measured using G-LISA activation assay kits for Rac1 (Cytoskeleton, Inc, BK128) and RhoA (Cytoskeleton, Inc, BK124). Brains or crude synaptosome pellets were homogenized in G-LISA lysis buffer supplemented with protease and phosphatase inhibitors (Roche), normalized to the same concentration as determined by Precision Red Advanced Protein Assay, and applied to the G-LISA matrices and processed according to manufacturer’s protocols. Absorbances at 490 nm (OD₄₉₀) were background-subtracted and normalized to the WT average.

Behavioral tests were performed in both male and female mice 6–8 weeks of age (P42-P56). Mice were habituated to handling for 5 min/ day for 5 days prior to experiments, and habituated in the test facility separate from housing for at least 30 min prior to starting the task. We performed the Kondziela inverted screen test for motor strength, accelerating rotarod for motor coordination and learning, social preference test, novel object recognition task, open field test, elevated plus maze, and nestlet shredding, as previously described (Katrancha et al., 2019; Deacon, 2013; Omar et al., 2017; Rapanelli et al., 2017; Sfakianos et al., 2007). All behavioral studies were performed and analyzed by an experimenter blinded to genotype. Individual data points could be excluded from analysis if a mouse failed to properly acclimate to the testing conditions, as described in detail for specific tests below.

A five-lane rotarod treadmill (Med Associates ENV-577M 8.75’’ rod circumference) was used for accelerating analyses. Mice were acclimated to the rotating rod at 4 RPM for 10 s before beginning the test; mice that were unable to stay on the rod during this period were excluded from analyses. The rod then accelerated from 4 to 40 RPM over the course of 5 min (1.2 RPM/10 s) before leveling off at 40 RPM for 2.5 min. Five trials were performed per mouse with a rest time of at least 10 min between trials. The latency to fall was recorded per trial.

The screen apparatus consisted of a 40 cm square of wire mesh composed of 12 mm squares with 1 mm diameter wire, surrounded by a square wooden frame 3.5 cm thick. Mice were placed in the center of the screen and inverted over 2 s over a clear open box 45 cm high. During the training session on day 1, mice were habituated to hanging on the screen until a cumulative hang time of 2 min was reached; mice were excluded if they were unable to meet this requirement. During the test session on day 2, mice were placed on the screen as before and given three trials to reach a maximum inverted hang time of 8 min, with 10 min rest periods between trials. Mice that were unsuccessful at reaching the 8 min mark were excluded from the accelerating rotarod test.

Mice were placed in a large 16x16 inch square clear plexiglass SuperFlex cage and monitored using the Fusion software connected to a light beam array (Omnitech Electronics, Open Field Test, Version 4.5) for 10 min. The following measurements were analyzed based upon beam breaks: total distance traveled, ambulatory time (successive beam breaks), and vertical activity time (z axis beam breaks).

Mice were placed in a 50 cm x 50 cm plexiglass box with two wire-mesh pencil holders (open side facing downward) for 10 min for habituation. Then, the mouse was briefly removed while an inanimate object (Duplo blocks of a similar size and color to a mouse) and a male conspecific ‘‘stranger’’ mouse were placed underneath separate wire-mesh pencil holders. The test mouse was then returned to the cage for 10 min, tracked using the AnyMaze software (Stoelting Co.), and scored for entries into and time within a social target zone and an equivalent nonsocial target zone. The target zones were defined by an annulus around the wire enclosure.

During object familiarization on day 1, two identical objects were placed on the left and right sides of the testing cage. Mice explored the objects until they accumulated 30 s of tactile exploration time, defined as direct oral or nasal contact. Mice were then returned to their home cage for 48 hr. During the novel object recognition task on day 3, one familiar and one novel object were placed on either side of the testing cage, and mice explored the objects until they accumulated 30 s of tactile exploration time. Mice were excluded from analysis if they failed to explore both objects or accumulate 30 s of tactile exploration time within 6 min on either day.

The apparatus is raised 36.5 cm above the ground with a 5 cm square center platform and four 30 cm x 5 cm arms; two of the arms are enclosed by 16 cm high walls (Stoelting Co, Wood Dale, IL). Mice were tested in 3 trials that were 5 min long with at least 10 min to rest between trials. Mice were placed directly on the center platform and allowed to freely explore all arms of the maze, tracked using the AnyMaze software (Stoelting Co.), and scored for time spent in each arm.

A standard polycarbonate mouse cage (19×29 × 13 cm) was filled with unscented bedding material to a depth of 0.5 cm, the surface was leveled, and a fitted filter-top cover was placed on top. Commercially available cotton fiber nestlets (5 cm x 5 cm, 5 mm thick, 2–2.5 g each) were weighed on an analytical balance before being placed on top of the bedding. The mouse was placed into the test cage without food or water and left undisturbed for 30 min. The remaining intact nestlet was removed and allowed to dry overnight. The weight of the dried unshredded nestlet was divided by the starting weight to calculate the percentage of nestlet shredded.

Mice were transcardially perfused with heparinized PBS followed by 4% paraformaldehyde (PFA) in phosphate buffered saline (PBS). Brains were postfixed in 4% PFA at 4 °C for 24 hr, then sliced coronally at 30 μm for Nissl stain and immunohistochemistry and at 200 μm for dendritic arbor and spine analysis. Sections were processed as per application and coverslip-mounted with ProLong Diamond Antifade Mountant (P36961, Thermo Fisher). For electron microscopy, mice were perfused with 4% PFA supplemented with 2% glutaraldehyde.

Nissl staining on 30 μm brain slices was performed according to standard protocols (Paul et al., 2008). Brightfield images were acquired at 20 X on a SlideView VS200 slide scanner (Olympus). Coronal sections containing the decussation of the anterior commissure were selected for analysis and quantified using QuPath software. Measurements from three consecutive slices were averaged per mouse.

Desired sections were identified by comparison to a mouse brain atlas (Franklin and Paxinos, 2013). Three 30 μm coronal sections from the motor cortex, spaced at 120 μm intervals and containing the rostral decussation of the corpus callosum, were used for immunostaining. Mounted sections were boiled in Na-Citrate buffer (10 mM, pH 6.4) for 45 min, blocked for 1 hr in 5% normal goat serum (NGS) in 0.3% PBS-T, incubated overnight at 4 °C with primary antibodies (Key Resources Table), then with appropriate AlexaFluor-conjugated secondary antibodies for 1 hr at RT, and counterstained for 10 min with DAPI. Controls lacking primary antibody were used to confirm staining specificity. Fluorescent images were acquired at 10 X on a SlideView VS200 slide scanner (Olympus). Semi-automatic cell counts for DAPI+, NeuN+, and PV +stained cells were obtained using QuPath by an experimenter blinded to genotype.

Mice of all genotypes examined were intercrossed with mice bearing a thy1-GFP transgene (M line; Feng et al., 2000). Fluorescent images of neurons contained within 200 μm coronal sections were acquired at 20 X for arbor reconstructions (Zeiss LSM900) and at 100 x on an UltraVIEW VoX spinning disc confocal microscope for dendritic spines. Dendrites were traced and measured using the Simple Neurite Tracer Fiji plugin (SNT) in ImageJ (NIH). Spines were manually counted from max projection z-stack images on secondary apical and basal dendritic branches of L5 PNs.

Electron microscopy of synapses in M1 layer 5 was performed as previously described (Katrancha et al., 2019). Asymmetric excitatory synapses were identified as an electron-dense post-synaptic density (PSD) apposed to a presynaptic terminal containing synaptic vesicles. Measurements of synaptic features, including synapse density, PSD length, spine head area, presynaptic bouton area, and synaptic vesicle distribution, were quantified in ImageJ as previously described (Montesinos et al., 2015; Dong et al., 2018; Levy et al., 2018; González-Forero et al., 2012), from 5 to 18 fields of view (~50 μm²) per mouse by a reviewer blinded to genotype.

Acute slices were prepared from mice at P35-42 as previously described (Katrancha et al., 2019) with modifications noted here. Coronal slices of M1-M2 cortex were cut at 360 µm in ice-cold N-Methyl-D-glucamine-aCSF (NMDG-aCSF, in mM): 120 NMDG, 2.5 KCl, 7 MgSO₄, 1.25 NaH₂PO₄, 0.5 CaCl₂, 28 NaHCO₃, 13 glucose, 7 sucrose, saturated with 95% O2-5% O2 at 300–320 mOsmol/L, pH 7.4 on a vibratome (Leica). Slices were recovered in ACSF (in mM): 120 NaCl, 2 KCl, 1.2 NaH2PO4, 26 NaHCO3, 1.3 MgSO4, 1.6 CaCl2, 11 glucose, 2 ascorbic acid, 4 Na-lactic acid, 2 Na-pyruvate saturated with 95% O2-5% O2 at 300–320 mOsmol/L, pH 7.4, at 32 °C for 20 min followed by 1 hr recovery and recordings at RT. Recorded signals were acquired at a 100 kHz sampling rate and low pass filtered at 6 kHz. Cells were excluded from analysis if series resistance changed >20%.

Excitatory postsynaptic currents (EPSCs) were evoked (e) by a tungsten bipolar stimulating electrode placed in M1 L2/3 with stimulus intensities that yielded 40–50% of the maximal response. AMPAR-eEPSC amplitudes were measured at the peak of the EPSC recorded at –70 mV holding potential. NMDAR-eEPSC amplitudes were recorded from the same neurons at +40 mV and measured at 40ms after the stimulus artifact when AMPAR-eEPSCs have decayed.

Miniature EPSCs (mEPSCs) were recorded in whole-cell patch mode with 1 μM tetrodotoxin, 5 μM strychnine, and 20 μM bicuculline methiodide (BMI) in the external solution. AMPAR-mEPSCs were recorded at –70 mV with 50 μM D-2-Amino-5-phosphonovaleric acid (D-AP5) to block NMDAR. NMDAR-mEPSCs were recorded at +45 mV with 5 μM Cyanquixaline (CNQX) to block AMPAR. Gamma-aminobutyric acid receptor and glycine receptor inhibitory postsynaptic currents (GABAR/GlyR-mIPSCs) were recorded at +15 mV with D-AP5 and CNQX. Amplitude and frequency of miniature events were detected using Mini-Analysis software (Synaptosoft, Decatur, GA) and inspected visually.

Paired-pulse stimulation was applied with interstimulus intervals (ISIs) of 35, 60, 100, 200, or 300ms. The amplitude of the second eEPSC (EPSC₂), evoked by the second of the paired-pulse stimulations, was divided by eEPSC₁ amplitude to obtain a paired-pulse ratio (PPR). Long-term potentiation (LTP) of evoked EPSCs (recorded at –70 mV) was induced by theta-burst stimulation (TBS, five trains of four pulses at 100 Hz in 200ms intervals, repeated four times with an interval of 10 s) in the presence of 5 μM strychnine and 20 μM BMI.

Estimations of the readily releasable pool (RRP) size and synaptic vesicles (SVs) depletion rate were made from recordings of AMPAR-eEPSCs using high-frequency stimulation (HFS, 40 Hz) trains of 15 pulses to deplete the RRP. Trains were analyzed to estimate glutamate probability of release (Pr) and RRP for each neuron using the decay method as previously described (Ruiz et al., 2011; Thanawala and Regehr, 2016) using OriginLab 10 software. RRP recovery was approximated as fractional recovery using initial train stimulation with interstimulus intervals of 0.1, 2, 5, and 10 s, followed by a single stimulus. Fractional recovery for each ISI was calculated as (eEPSCs2-eEPSCt1ss)/(eEPSCt1-eEPSCt1ss), where eEPSCs2-amplitude of the stimulus applied after the train, eEPSCt1-first amplitude of the train, eEPSCt1ss-steady state amplitude of the train (last 5 responses). 1 mM kynurenic acid was added to prevent AMPA receptor saturation; 50 μM vD-AP5 was added to block NMDA receptors, 5 μM strychnine and 20 μM BMI to block GABA and glycine receptors.

PPR and train recordings for the rescue experiments were performed as described above with the addition of 100 μM NSC23766 into the recording solution with a 5-min incubation period to allow for efficient slice penetration.

Cortices from male P21 mice (four per genotype) were removed and immediately frozen in liquid nitrogen. Mouse brain tissues were transferred to Covaris tissueTUBE TT1 (Part No. 520001, Plug Part No. 520006), kept on dry ice, and flash frozen in liquid nitrogen prior to cryopulverization using a Covaris CP02 cryoPREP Automated Dry Pulverizer (Part No. 500001). The resulting powdered samples were transferred to 1.5 ml microcentrifuge tubes and lysed in 300 μl lysis buffer (8 M urea in 50 mM Tris pH 8.0, 150 mM NaCl, 1 mM EDTA and protease phosphatase inhibitor) on ice. Lysates were sonicated using AFA Covaris E220 (Part No. 500239). The sonicated samples were centrifuged for 30 min at 10,000 rpm. Clear supernatants were transferred to new tubes and quantified with the BCA assay (Thermo). Lysates were reduced with 10 mM TCEP and alkylated with 10 mM iodoacetamide for 30 min in the dark at room temperature. 1 mg lysates were diluted to 4 M urea with 200 mM Tris pH 8.0 and digested with Lys-C (1:50 protease:protein ratio) for 90 min at 37 °C. Samples were then diluted to less than 2 M urea with 200 mM Tris pH 8.0 and digested with trypsin at the same ratio overnight at 37 ° C. The samples were acidified with 1% formic acid and desalted using 50 mg per 1 cc Sep-Pak (Waters). A 250 μg aliquot of each sample was frozen and dried, then stored at –80 ° C until tandem mass tag (TMTpro 16, Thermo lot# WI325918) peptide labeling according to manufacturer instructions.

Desalted 16-plex TMT-labeled peptides (5 mg total) were subjected to basic reverse-phase (RP) chromatography. Peptides were separated using a 4.6 mm x 250 mm RP Zorbax 300 A Extend-C18 column (Agilent, 5 mm bead size) on an Agilent 1260 series HPLC. The 96 min gradient of solvent A (2% ACN, 5 mM ammonium formate) to solvent B (90% ACN, 5 mM ammonium formate) stayed at 0% B for 7 min, from 0% to 16% B over 6 min, from 16% to 40% over 60 min, from 40% to 44% over 4 min, from 44% to 60% over 5 min and maintained at 60% B over 14 min. Flow rate was 1 ml/min. 90 fractions were collected and concatenated down to 24 fractions. 5% of each final fraction was used for proteome analysis. The remaining 95% of each of the 25 fractions was concatenated to 13 fractions. These 13 fractions were IMAC enriched using Ni-NTA Agarose beads as described for phosphoproteome analysis (Mertins et al., 2018).

Proteome analysis was performed using an EASY-nLC 1200 UHPLC coupled to an Orbitrap Eclipse mass spectrometer (Thermo Fisher Scientific). The online peptide separation was performed using a 25 cm x 75 mm i.d. silica picofrit capillary column (New Objectives) packed with 1.9 mm ReproSil-Pur C18-AQ beads (Dr. Maisch GmbH). The 110 min method, 84 min effective gradient of solvent A (2% ACN, 0.1% FA) and solvent B (80% ACN, 0.1% FA). The gradient started at 2% B and increased to 10% B over 1 min, from 10% to 50% B over 84 min, from 50% to 72% B over 9 min, from 72% to 90% B over 1 min, stayed at 90% B over 5 min, then dropped to 60% B over 1 min and maintained at 60% B for 9 min. The flow rate was at 200 nl/min. The Eclipse mass spectrometer performed data-dependent acquisition in positive ion mode. MS1 spectra scanned a range of 350–2000 m/z at 60000 resolution, with maximum injection time of 50ms and 100% AGC target. MS2 spectra scanned at first mass of 110 m/z at 50,000 resolution. The cycle time for MS2 scan was set for 2 s for charge state of 1<x < 6, with isolation window of 0.7 m/z. The AGC target for MS2 was 60% with a maximum injection time of 105ms. An empirically determined normalized collision energy was set to 32. Dynamic exclusion was set to 20 s.

All raw files were searched using Spectrum Mill (Agilent). MS2 spectra were searched against the Uniprot Mouse Database (20171228, 47,069 mouse entries, 264 common laboratory contaminants) with a mass tolerance of 20 ppm for both the precursor and product ions. The enzyme specificity was set for Trypsin and allowed up to three missed cleavages. The fixed modification was carbamidomethylation at cysteine. TMT labeling was required at lysine, but peptide N-termini were allowed to be either labeled or unlabeled. Allowed variable modifications for whole proteome datasets were acetylation of protein N-termini, oxidized methionine, deamidation of asparagine, and pyroglutamic acid at peptide N-terminal glutamine, with a precursor MH +shift range of –18–64 Da. The false discovery rate was less than 1%. Protein identifications were discarded if the protein was only observed by a single peptide. Protein subgroups were collapsed to the proteoform with the most evidence.

Mouse UniProtIDs for identified proteins were converted to human orthologs and ranked by signed log₁₀(nominal p-value) (with sign indicating direction of fold-change from WT). GSEA 4.3.3 (Subramanian et al., 2005; Mootha et al., 2003) was used to run the GSEA Preranked tests against all gene sets in the Human MSigDB Collection C2 v2023.2. SynGO 1.2 was used to identify synaptic ontologies for differentially expressed proteins with nominal p-value <0.05.

Statistical analyses were performed using GraphPad Prism 10. Data in bar graphs are presented as mean ± SEM, with individual data points graphed when applicable. Sample size ‘n’ is annotated within the bars or in the Figure legend for each group. Distributions were tested for normality and outliers (>1.5 x the interquartile distance for behavioral data, or by Prism ROUT method at Q=5% for other metrics) were removed before proceeding with statistical tests. Specific details of statistical tests performed, adjusted with post-hoc Bonferroni test for multiple comparisons (MC) where appropriate, are noted in the Figure legends. Significance was defined by a p-value less than 0.05: ^nsp <0.1; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.