Despite the traditional view of the brain as an immunological privileged organ, recent discoveries have revealed that a continuous crosstalk between microglia, the brain residential immune cells, and neurons is required for the maintenance of brain homeostasis and for the sculpting of neuronal connections during development. Indeed, defects in this bidirectional communication have been described in brain diseases, where altered microglia function, synaptic activity and plasticity may produce profound changes in nervous circuits and associated functions.

The Triggering receptor expressed on myeloid cells 2 (Trem2) is a myeloid cell-specific gene expressed in brain microglia, with variants that are associated with neurodegenerative diseases, including Alzheimer’s disease. During development, Trem2 is essential for microglia-mediated synaptic refinement, but whether it also contributes in shaping early neuronal development remains unclear. Here, we demonstrate that Trem2 plays a key role in controlling the bioenergetic profile of pyramidal neurons during development. In the absence of Trem2, developing neurons in the hippocampal CA1 but not CA3 region, display compromised energetic metabolism, accompanied by a transcriptional rearrangement that include a pervasive alteration of metabolic and synaptic signatures, ultimately leading to a delay in the maturation of CA1 pyramidal neurons. Such early derangement is mantained at later developmental windows, leading to synaptic and circuitry alterations. Altogether, these results unveil a novel role of Trem2 in controlling neuronal development by regulating the metabolic fitness of neurons in a region-specific manner.

Macroautophagy, a conserved mechanism delivering cellular constituents to the lysosome, has emerged as a pivotal player of brain homeostasis and synaptic function. We and others demonstrated that long-term synaptic depression requires the on-demand autophagic turnover of synaptic proteins. Moreover, in our recent work we immunopurified autophagic vesicles from the mouse brain to characterize the constitutive brain autophagic degradome, revealing the synaptic and organellar substrates of autophagy in brain cells. Following up on this work, here we show that brain autophagic vesicles additionally contain RNA-binding proteins and diverse RNA species in their lumen. We provide a detailed profiling of these RNA cargoes, enriched in small-RNA species. Moreover, we identify a group of proteins as putative selective autophagy receptors (SARs) for RNA sequestration within brain autophagic vesicles. Finally, we explore the role of autophagy in regulating the turnover and/or trafficking of these RNAs, focusing on neurons. Our findings reveal a previously unknown interplay between autophagy and gene expression regulation in the brain, redefining and enriching our understanding of how autophagy regulates brain homeostasis.

Rabphilin 3A (Rph3A) is a synaptic protein crucial for regulating exo- and endocytosis processes at presynaptic terminals and dendritic spines. The protein promotes the stabilization of the NMDA-type glutamate receptors (NMDARs) at the cell surface, forming a ternary complex with the PSD-95 and GluN2A subunit of the receptor. This complex is needed for long-term potentiation induction and NMDAR-dependent hippocampal behaviors, such as spatial learning. Rph3A overexpression increases dendritic spine density and GluN2A-containing NMDARs at synaptic membranes. Overall, these results suggest that aberrant Rph3A could be also involved in pathological alterations of NMDAR.

By trio-based exome sequencing, GeneMatcher and screening of 100,000 Genomes Project data, we identified six heterozygous variants in RPH3A gene. Four cases had a neurodevelopmental disorder (NDD) with untreatable epileptic seizures [p.(Gln73His)dn; p.(Thr450Ser)dn; p.(Arg209Lys); p.(Gln508His)], and two cases [p.(Asn618Ser)dn; p.(Arg235Ser)] showed high functioning autism spectrum disorder (ASD). Using primary hippocampal neuronal cultures, we demonstrated that p.(Arg209Lys), p.(Thr450Ser), p.(Gln508His) and p.(Asn618Ser) reduce PSD-95/GluN2A co-localization. Rph3A variants p.(Arg209Lys), p.(Thr450Ser) and p.(Gln508His) also increased the surface levels of GluN2A. These results indicate that RPH3A mutants induce GluN2A-containing NMDARs to localize prevalently at the extrasynaptic sites, suggesting that their stimulation can promote a detrimental effect. In addition, RPH3A mutants induce defects in NMDAR activity at synaptic sites, affecting postsynaptic calcium and glutamate influx at dendritic level. In conclusion, we provide evidence that missense gain-of-function variants in RPH3A increase GluN2A-containing NMDARs at extrasynaptic sites and alter synaptic function leading to a clinically variable neurodevelopmental presentation, ranging from untreatable epilepsy to ASD.

Alzheimer's disease (AD) manifests with extracellular amyloid-β plaques and intracellular tau-containing neurofibrillary tangles in the brain. While these pathologies synergistically affect tau spread, they exert antagonistic effects on population activity in hippocampal and cortical circuits. Emerging evidence suggests that familial AD (fAD) mutations and soluble amyloid-β contribute to neural circuit hyperexcitability, leading to augmented neuronal firing rate, hypersynchrony, and epileptiform discharges, particularly during low-arousal brain states in presymptomatic fAD mouse models. In contrast, mutations in human tau are associated with a reduction in spontaneous neuronal activity in transgenic mouse models of tauopathy.

We hypothesize a link between the pathophysiology of neuronal activity in cortico-hippocampal circuits and the dysfunctional homeostatic regulation of mean firing rate (MFR) at the neuronal population level. Here, we investigated how fAD and tau-associated mutations influence neuronal stability amidst ongoing perturbations. Employing long-term extracellular recordings of spikes in cultured hippocampal networks grown on micro-electrode arrays, we observed stable MFRs around a set-point value during unperturbed conditions. Chronic inhibition by baclofen induced a transient drop in MFR, normalized within a day in its presence. Notably, fAD APP/PS1 mutations or the P301L-variant human tau (hTau_P301L) mutation, linked to frontotemporal dementia, did not alter the MFR homeostatic response to chronic inactivity. In contrast, the hTau-A152T mutation, associated with AD and other tauopathies, impaired homeostatic MFR recovery to inactivity. The MFR homeostatic response to inactivity remained resilient to bi-directional changes in tau expression levels.

In conclusion, our findings highlight distinct effects of fAD and hTau mutations on the homeostatic regulation of MFR in hippocampal circuits, potentially elucidating the antagonistic effects of amyloid-β and tau pathologies on neuronal activity in cortico-hippocampal circuits during early disease stages, preceding memory decline.

Stress can lead to synapse dysfunction and contribute to the development of depression. While most individuals are resilient to stress exposure, it can cause pathologies in vulnerable individuals. Women are more frequently and more severely affected than men by stress-related psychiatric disorders like depression. The underlying biological mechanisms causing this sex bias remain unknown. However, a potential role of sex hormones and their receptors has manifested over the past years. I want to understand which biological factors determine susceptibility to stress, how these differ between sexes, and investigate the potential role of sex hormones and receptors in stress vulnerability. To examine this, a 2-week social instability stress (SIS) paradigm is applied to mice during adolescence – a particularly vulnerable period for developing affective disorders. Evaluation of behavioral and physiological phenotypes, using both classical tests and home-cage phenotyping, is used to distinguish between stress-vulnerable and resilient mice. Preliminary results show that although both sexes have a similar percentage of stress susceptible mice, there is a clear sex difference in stress response. Alterations in the expression of sex hormone receptors is determined using immunohistochemistry and subsequent brain mapping in vulnerable and resilient male and female mice. I will also examine how stress impacts the synaptic proteome in sex-specific ways, using proteomics and phosphoproteomics of isolated synaptoneurosomes from multiple brain regions. This study will provide a deeper understanding of sex differences in psychiatric conditions and enable development of more targeted therapies and precise gender-based psychiatric treatments in the future.

Olfactory sensory neurons (OSNs) detect odours at a wide range of intensities. In insects, volatile compounds are perceived by odorant receptors (ORs), which are made up of an odour-specific protein (OrX) and the ubiquitous odorant co-receptor Orco. In principle, ORs tune the sensitivity of odour detection, with some OSNs exhibiting exceptionally high sensitivity. To test whether additional mechanisms underlie odour-specific neuronal processing, we investigated synapses between OSNs and projection neurons in the antennal lobe, the first relay station of the olfactory pathway. Here, we studied the molecular structure and plasticity of the presynaptic active zone (AZ), the specialized site of neurotransmitter release. We focused on a highly sensitive OSN type that expresses the receptor Or56a and exclusively detects geosmin, an odorant signaling ecologically harmful stimuli. Using confocal microscopy, our results uncover a differential arrangement of the AZ proteins Bruchpilot (Brp) and Unc13A at Or56a and conventional OSNs. Interestingly, our investigations also show that Or56a-OSNs display a limited capacity to undergo homeostatic plasticity in response to a genetic reduction of presynaptic release probability. We hypothesise that this difference to conventional OSNs reflects the basal tuning of geosmin-sensing neurons towards maximum levels of performance.

SynGAP1-related non-syndromic intellectual disability, more commonly known as SynGAP Syndrome, is a rare genetic, neurodevelopmental disorder (NDD) caused by mutations in the SYNGAP1 gene. SynGAP, a majorly expressed brain protein is a RasGAP. The overarching aim is to contribute to our understanding of the multifaceted role of SynGAP in neuropsychiatric disorders and it’s connection with the Ras pathway. Our goals include establishing cell-based assays to study the activity of SynGAP protein variants and identifying novel Ras targets to understand their binding affinity. We have utilized functional imaging experiments to explore HRas-dependent effects on targets like sodium channels and assess their regulation in the context of the Ras-ERK pathway; and successfully activated the Ras-ERK pathway for the assay. The ultimate objective is to uncover molecular pathways involving SynGAP1, contributing to the development of targeted therapeutic interventions for individuals affected by SynGAP syndrome, thus improving their quality of life.

Presynaptic forms of plasticity occur throughout the nervous system and play an important role in learning and memory but the underlying molecular mechanisms are insufficiently understood. Here we show that the small GTPase Rab3 is a key mediator of cyclic AMP (cAMP)-induced presynaptic plasticity in Drosophila. Pharmacological and optogenetic cAMP production triggered concentration-dependent alterations of synaptic transmission, including potentiation and depression of evoked neurotransmitter release, as well as strongly facilitated spontaneous release. These changes correlated with a nanoscopic rearrangement of the active zone protein Unc13A and required Rab3. To link these results to animal behaviour, we turned to the established role of cAMP signalling in memory formation and demonstrate that Rab3 is necessary for olfactory learning. As Rab3 is dispensable for basal synaptic transmission, these findings uncover a molecular pathway specifically recruited for tuning neuronal communication and adaptive behaviour.

Fragile-X-Syndrome, the most common inherited form of intellectual disability, is caused by transcriptional silencing of the Fmr1 gene, that encodes for fragile-X-messenger ribonucleoprotein (FMRP). FMRP is an RNA-binding protein involved in regulating many synaptic proteins. In fact, Fmr1y/- mice present impaired synaptic plasticity, but the mechanisms underlying such deficits are largely unknown; the aim of this work is to bridge this gap. Acute hippocampal slices were prepared from WT and Fmr1y/- mice and long-term potentiation (LTP) was induced with five theta-bursts, a protocol known to induce the release of endogenous BDNF. We observed an impairment in LTP of CA1 synapses of Fmr1y/- mice. Blockade of BDNF-TrkB signalling further impaired LTP in slices from Fmr1y/- mice, while blocking NMDA receptors (NMDAR) was without effect. Fmr1 downregulation may impair LTP by affecting BDNF-mediated control of synaptic NMDAR. This was investigated using primary cultures of hippocampal neurons transfected with a shRNA to knock down (KD) Fmr1 expression, and analysing BDNF-induced upregulation of synaptic surface NMDAR by immunocytochemistry. Fmr1 KD had no effect on synaptic GluN2A- and GluN2B-containing NMDAR under resting conditions, but abolished BDNF-induced upregulation of synaptic NMDAR. Furthermore, Fmr1 KD impaired BDNF-induced dendritic accumulation of Pyk2, a kinase regulator of NMDAR synaptic stability. Finally, single particle tracking by quantum dots in neurons after Fmr1 KD showed a decrease in mobility of GluN2A-containing NMDAR when compared to control, while GluN2B-containing NMDAR become so after BDNF treatment. Our data show an impairment in BDNF-induced synaptic regulation of NMDAR after Fmr1 KD which may account for the deficits in LTP.

(Supported by MSCA [ITN-#813986] and FCT)

Antisense oligonucleotides (ASO) are powerful tools to alter gene expression and ASO’s are even under clinical use and clinical trials to treat human diseases. Whether ASO’s targeting protein-coding genes embedded with non-coding RNA’s affect non-coding RNA expression or function is relatively unknown. While studying how glial cells regulate axonal morphogenesis in larval zebrafish, we made a serendipitous observation that ASO targeting a splice site of one of our candidate protein-coding genes led to defects in axonal morphogenesis. We observed ASO-induced intron retention events and increased gene expression. The splice site-targeted ASO-induced phenotype was not rescued by candidate gene translation-blocking ASO’s, thus potentially ruling out the role of truncated protein. However, the phenotype was rescuable by the knockdown of embedded non-coding RNA. We believe our results highlight a blind spot in ASO-based research and call for a careful evaluation of results from ASO studies when targeting protein-coding genes embedded with non-coding RNAs.

Neurotransmitter release at the neuronal synapse is a fundamental process for signal transduction between neurons. The SNARE proteins play a crucial role here by eliciting the fusion of the synaptic vesicle membrane with the presynaptic plasma membrane. A fusion pore will open, releasing neurotransmitters into the synaptic cleft.

In their monomeric pre-fusion state, the SNARE proteins are intrinsically disordered. They do not exhibit a well-defined structure and show high internal flexibility. The SNARE proteins are membrane-anchored. However, the mode of interaction between the SNARE proteins and the lipid membrane is not well understood. We employ Nuclear Magnetic Resonance (NMR) spectroscopy to obtain novel structural and dynamic insights into SNARE proteins in their monomeric pre-fusion form at an atomic resolution.

At the conference, we will present recently published [1,2] and unpublished insights into the structural dynamics of the SNARE proteins synaptobrevin-2 and SNAP25a at the lipid membrane interface.

Literature:

[1] Lakomek N.A., Yavuz H., Jahn R., Perez-Lara A., Structural dynamics and transient lipid binding of synaptobrevin-2 tune SNARE assembly and membrane fusion, Proc. Natl. Acad. Sci. U.S.A. 2019, 16(18), 8699-8708.

[2] Stief T., Gremer L., Pribicevic S., Espinueva D.F., Vormann K., Biehl R., Jahn R.., Pérez-Lara Á., Lakomek N.A., J. Mol. Biol. 2023 Mar 30;435(10):168069

The ionotropic NMDA receptors (NMDARs) greatly contribute to the excitatory glutamatergic drive in the central nervous system. Their function is crucial for neuronal physiology and synaptic plasticity. In addition, their dysfunctions have been associated with the etiology of several major neurological and psychiatric diseases. Therefore, understanding how these receptors are regulated in healthy and diseased brains has been a major challenge for many laboratories over the past decades.

Recent developments in the super-resolution microscopy field have revealed that beyond the diffraction-limited view, NMDARs are organized into nanometer-sized clusters, termed nanodomains. Using super-resolution imaging and single nanoparticle tracking in rat hippocampal neurons it was possible to unveil the nanoscale topography of native GluN2A- and GluN2B-NMDAR within the synapse and along the dendritic arbor and the contribution of the different nanoarchitecture to synaptic plasticity (Kellermayer*, Ferreira*, Dupuis* et al., 2018; Ferreira et al., 2020).

Our recent studies reveal that the dynamic nature of GluN2B-containing NMDARs (GluN2B-NMDAR) makes them key players in the NMDAR-dependent regulation of synaptic adaptation. We have shown that GluN2B-NMDARs are important molecular hubs for the synaptic localization of cell-adhesion proteins (CAMs), such as Neurexins and the ephrin type-A receptor 4 (EphA4). We proposed that the interplay of NMDAR with CAMs, possibly through their extracellular domains, may be instrumental for the receptor function and organization, to the same extent as the well-known intracellular NMDAR partners. Understanding how the trans-synaptic signaling may influence NMDAR nanoarchitecture will shed new light on glutamatergic synapse function and regulation.

Perineuronal nets (PNNs) are specialised lattice-like networks of extracellular matrix (ECM) proteins that ensheath the cell bodies and proximal dendrites of inhibitory neurones. They create a neuroprotective barrier against excitotoxicity and oxidative stress, maintain homeostatic plasticity and promote synaptogenesis. Their importance is highlighted by PNN dysregulation in psychiatric disease; mutations in genes for PNN proteins have been found in patients with bipolar disorder (BD) and schizophrenia (SZ). Furthermore, many studies report a loss of PNNs in post mortem tissue and animal models of numerous psychiatric diseases.

We have discovered a novel role for the presynaptic scaffolding protein Piccolo in astrocyte secretion. Specifically, secretion of ECM components such as brevican and neurocan. Mutations in the PCLO gene have also been found in BD and SZ patients. Another unifying feature of these disorders is loss of astrocyte function and/or astrocyte cell loss.

This ongoing study investigates the secretion of PNN components from astrocytes in a Piccolo mutant rat model (Pclogt/gt). Preliminary data reveals a loss of the PNN proteoglycan neurocan around cortical Pclowt/wt and Pclogt/gt neurones when co-cultured with Pclogt/gt but not Pclowt/wt astrocytes, and an accumulation of neurocan inside Pclogt/gt astrocytes. Western blot analysis of cell culture supernatant vs. astrocyte cell lysate are being performed and consequences for network excitability will be investigated using calcium imaging and immunocytochemistry.

DNM1 epileptic encephalopathy is a specific form of developmental epileptic encephalopathy, characterised by severe to profound intellectual disability, hypotonia and epilepsy. All individuals identified with this disorder have de novo heterozygous missense mutations in the DNM1 gene, which encodes the large GTPase Dynamin-1. Importantly, almost all individuals with DNM1 mutations have intractable epilepsy, making the identification of novel therapeutic interventions an urgent unmet challenge.

Correct neurotransmission and circuit function requires the efficient fusion, retrieval and refilling of neurotransmitter-containing synaptic vesicles (SVs) at the presynapse. Dynamin-1 is essential for SV endocytosis, and undergoing a conformational change on GTP hydrolysis to provide force for the final stages on vesicle fission. It has a modular structure with an N-terminal GTPase domain, followed by domains essential for self-assembly (middle and GTPase effector domains), membrane lipid binding (pleckstrin homology, PH) and protein interactions (C-terminal proline-rich domain). To determine how mutations in the middle domain of DNM1 gene could translate into epilepsy, we reproduced a series of human mutations in the middle domain of Dynamin-1. Using live-imaging and biochemistry approaches, we characterised, at both the cell and molecular level, the effect of these mutations on Dynamin-1 GTPase activity, its interactions, subcellular localisation and at specific stages of the SV life cycle. Importantly, the phenotype can be corrected at the cell level by a drug which accelerates endocytosis.

These studies revealed a series of disease signatures, which will provide a framework for future translational studies and importantly reveal that synaptic vesicle recycling may be a viable therapeutic target for monogenic intractable epilepsies.

Autoantibodies directed against the N-methyl-D-aspartate receptor (NMDAR-Ab) are pathogenic immunoglobulins detected in patients suffering from NMDAR encephalitis. NMDAR-Ab alter the receptor membrane trafficking, synaptic transmission and neuronal network properties, leading to patients’ neurological and psychiatric symptoms. Patients often have very little neuronal damage but rapid and massive (treatment-responsive) brain dysfunctions related to unknown early mechanism of NMDAR-Ab. Our understanding of the early molecular cascade underpinning these synaptic dysfunctions remains in fact surprisingly fragmented. Here, we used a combination of single molecule-based imaging of membrane proteins to unveil the spatio-temporal action of NMDAR-Ab onto live hippocampal neurons. We first demonstrate that NMDAR-Ab primarily affect extrasynaptic -and not synaptic- NMDAR. In the first minutes, NMDAR-Ab increase extrasynaptic NMDAR membrane dynamics, de-clustering its surface interactome. NMDAR-Ab also rapidly reshuffle all membrane proteins located at the extrasynaptic compartment. Consistent with this alteration of multiple proteins, NMDAR-Ab effects were not mediated through the sole interaction between NMDAR and EphB2 receptor. At the long-term, NMDAR-Ab reduce NMDAR synaptic pool by slowing down receptor membrane dynamics in a cross-linking independent manners. Remarkably, exposing only extrasynaptic NMDAR to NMDAR-Ab was sufficient to produce their full-blown and long-term pathogenic effect. Collectively, we demonstrate that NMDAR-Ab first impair extrasynaptic proteins, and then the synaptic ones. These data shed thus new, and unsuspected, lights on the mode of action of NMDAR-Ab and likely to our understanding of (extra)synaptopathies.

Mammalian synapses facilitate high-frequency sustained neurotransmission through the fusion of synaptic vesicles (SVs) with the presynaptic membrane, followed by rapid recycling of SV machinery, including SNARE components, SV proteins, and membrane lipids over milliseconds to minutes. Visualizing these SV exo-endocytosis events is challenging due to their speed and nanoscale localization. To address this, we identified candidate SV proteins with abundant representation on SVs for precise labelling, aiming to enhance contrast. Conventional methods like over-expression, antibody-based labelling, and small molecule tags can compromise protein function and targeting, leading to misinterpretation of SV cycling mechanisms. Therefore, we employed the CRISPR-Cas9 system (TKIT method) for endogenous labelling of candidate proteins. We developed a library of potential target sequences, tested their efficiency in cell lines, and selected optimal guide RNAs (gRNAs). Utilizing two gRNAs and a Donor for Knock-In, we achieved high labelling precision of SV fusion events. Our objective is to detect changes in intraluminal SV pH upon fusion and recycling to investigate fusion and compensatory endocytosis mechanisms. To achieve this, we knocked-in super ecliptic pHluorin (SEP) into the luminal domain of candidate SV proteins. Following successful incorporation of the fluorescent tag, we assessed primary neuron fluorescence with field stimulation. Following this we detected and evaluated SV fusion and recycling. This approach demonstrates an effective methodology for labelling, visualizing, and studying SV protein recycling at the presynapse.

Dopamine is an essential brain neuromodulator involved in reward and motor control, altered in Parkinson’s disease and addiction. Dopaminergic neurons projects to most brain areas, with particularly enriched innervation in the striatum. Dopamine contained in vesicles is released by axons and binds to G protein coupled receptors to modulate neuronal activity of target neurons. However, the basic features of dopamine release sites, e.g. their location relative to other neuronal processes, or the organisation of synaptic vesicles within them, are still largely unknown. Here, we have used cryo-correlative light electron microscopy (cryo-CLEM) and cryo-electron tomography to analyse the ultrastructure of synaptosomes isolated from the striatum of adult mice. We identified dopaminergic (DA) synaptosomes, labelled via AAV-Flex-mNeonGreen injected in the VTA o DAT-Cre mice, and compared it with glutamatergic (Glu) synaptosomes isolated from VGluT1-Venus KI mice. We show that DA synaptosomes have ~10 times fewer vesicles than Glu ones, and that these vesicles are bigger and less round. We also analyzed other features of these vesicles, such as tethers, which will reveal the molecular parameters of dopamine release. Moreover, we found previously that ~30% of DA synaptosomes contact cortico-striatal, VGLUT1 positive synapses, forming a hub structure (Paget-Blanc et al. 2022 DOI 10.1038/s41467-022-30776-9). We have used a double labelling strategy to identify DA hub synapses and characterize the adhesion site between DA and Glu synapses with cryo-CLEM. These findings will help to understand the basic mechanisms of dopamine release, its modulation and its alterations in pathologies in the basal ganglia.

The actin binding protein (ABP) Drebrin (Dbn) regulates cytoskeletal functions during neuronal development and is thought to contribute to structural and functional synaptic changes associated with aging and Alzheimer's disease (AD). Interestingly, decreased Dbn protein levels have been reported to be associated with mild cognitive impairment and AD. We previously identified Dbn phosphorylation at S647 to increase protein stability and stress resilience at the spines. Long-term depression (LTD), defined as a long-lasting weakening of a synapse, has been suggested to be altered in age-related memory deficits. Our ongoing work shows that induced chemical LTD (cLTD) in cortical neurons rapidly decreases Dbn protein levels and alters phosphorylation of a number of amino acid residues. We aim to characterize posttranslational modifications observed during cLTD, and use Dbn as an entry point to characterize the actin interactome in synaptosomes isolated from aged and AD model mouse brains. This work will provide mechanistic and physiological insights into the functional role of the actin cytoskeleton to provide neuronal protection upon stress in neurodegeneration and with the progression of aging.

Post-translational modifications of proteins by Ubiquitin-like modifiers (Ubls) regulate a wide range of cellular processes. Indeed, increasing evidence links Ubls to neuronal development, but their role at synapses remains enigmatic. To address this issue, we examined the role of three essential Ubls, SUMO2, NEDD8, and UFM1, in synaptic transmission. Using autaptic hippocampal cultures from corresponding conditional knock-out mouse lines, we combined imaging approaches and electrophysiological recordings to systematically characterize the morphology and synaptic properties of excitatory glutamatergic neurons lacking SUMO2, NEDD8, or UFM1. SUMO2 deficient neurons showed reduced dendrite complexity and synapse numbers, along with a decrease in evoked and spontaneous synaptic transmission, and consequently reduced synaptic short-term depression. In contrast, NEDD8-deficient neurons showed increased vesicular release probability and consequently increased synaptic short-term depression, with largely unaltered morphology. Finally, UFM1-deficient neurons showed severely reduced dendrite complexity and synapse numbers, along with strong decreases in evoked and spontaneous synaptic transmission, while synaptic short-term depression was not changed. Altogether, our data indicate that SUMO2, NEDD8, and UFM1 each have specific ‘synaptic signatures’ and regulate the development and function of neurons via largely distinct mechanisms.

A Disintegrin And Metalloproteinase 10 (ADAM10) plays a pivotal role in shaping neuronal networks by orchestrating the activity of numerous membrane proteins through the shedding of their extracellular domains. Despite its significance in the brain, the specific cellular localization of ADAM10 remains not well understood due to a lack of appropriate tools. Here, using a specific ADAM10 antibody suitable for immunostainings, we discover that ADAM10 is localized to presynapses and especially enriched at presynaptic vesicles of mossy fiber (MF)-CA3 synapses in the hippocampus. These synapses undergo pronounced frequency facilitation of neurotransmitter release, a process that play critical roles in information transfer and neural computation. We demonstrate, that in conditional ADAM10 knockout mice the ability of MF synapses to undergo this type of synaptic plasticity is greatly reduced. The loss of facilitation depends on the cytosolic domain of ADAM10 and association with the calcium sensor synaptotagmin 7 rather than its proteolytic activity. Our findings unveil a new pathway contributing to the regulation of synaptic vesicle exocytosis.

Synaptic vesicle priming, and regulation of spontaneous and Ca2+-triggered release are core release functions that depend on both the synaptic SNARE-complex (sc) and the Calcium sensor synaptotagmin 1 (syt). NMR and X-ray crystallography experiments showed that the sc and syt bind via a so-called primary interface (Zhou et al, 2015). We performed site-directed mutagenesis on the sc proteins syntaxin-1 and SNAP-25 as well as on syt to interfere with this interface and performed electrophysiological recordings with rescue experiments in cultured murine neurons to test the relevance of the primary interface for vesicle priming, fusion clamping and Ca2+-triggered release. We find that individual single point mutants fall into three distinct categories that 1. mimic general loss of syt function; 2. exclusively impair vesicle priming and clamping, but not Ca2+-triggering; and 3. affect Ca2+-triggered release but not vesicle priming and fusion clamping. Our results generally support the notion that sc-syt interaction via the primary interface is relevant for all three release functions. Moreover, our results lead us also to postulate that the sc-syt interface is a critical switch for the transition from prefusion-clamped to calcium-activated states of the release apparatus, where sc-syt stay in contact during the transition but change their relative spatial orientation to trigger fast Ca2+-mediated release.

Endocytic adaptors ensure the correct sorting of membrane proteins. Impairments in endocytic proteins can cause mislocalizations of membrane proteins which in turn can lead to pathologies ranging from epilepsy to Alzheimer's disease. This emphasizes the critical role of accurate sorting in brain function.

EPS15 and EPS15R are closely related endocytic adaptor proteins implicated in brain function. Loss of their single homolog in C.elegans and D.melanogaster causes defects in synaptic vesicle endocytosis and synapse formation. In mammals, constitutive EPS15R KO mice are perinatally lethal, whereas EPS15 KO mice are unaffected, due to partial redundancy between Eps15 and Eps15R. The EPS15R KO mice also exhibit reduction in synaptic vesicles and increase in endosomal like vacuoles. To study neuronal functions of EPS15/15R in mice we generated neuron-specific DKO mice. Currently, we are characterizing the phenotypic effects of EPS15/15R DKO mice.

Our preliminary analyses of the brain-specific DKO mice demonstrate the importance of Eps15/Eps15R for mammalian brain function and organismal survival. DKO animals display a decreased postnatal weight gain and increased mortality with about 25% of mice dying before 2 months of age. In addition, DKO mice show hindlimb clasping and suffer from seizures. Moreover, first observations of home cage behavior have revealed specific behavioral alterations in DKO mice such as decreased digging and decreased occupation with the supplied paper towel. We also investigated the possible involvement of EPS15R in the endocytosis of the AMPA-type glutamate receptor GluA1, as it was previously suggested that ubiquitinated GluA1 is a cargo for EPS15. Indeed, we observed a significant surface accumulation of GluA1 in absence of Eps15/Eps15R in primary neurons via pHluorin assay. We could also successfully co-precipitate GluA1 with EPS15R from synaptosomal membrane fractions. This suggests that Eps15 and Eps15R act redundantly in GluA1 internalization and therefore might be relevant for neurotransmission and synaptic plasticity.

The precise mechanisms responsible for regulation of activity set points are still largely unknown. While NMDA receptors (NMDARs) are widely recognized for their role in Hebbian-like synaptic plasticity, their contribution to homeostatic regulation of neuronal activity has remained controversial. Utilizing long-term multi-electrode array recordings in hippocampal networks ex vivo, we discovered that sustained inhibition of NMDARs by structurally distinct blockers, including ketamine, effectively reduces the population mean firing rate (MFR) set point in hippocampal networks while preserving the homeostatic response to other perturbations. The reduction in MFR set point is mediated by a decrease in excitation/inhibition (E/I) ratio and the augmentation of fast-spiking interneurons’ intrinsic excitability. This mechanism requires eEF2K and BDNF. Notably, NMDAR-eEF2K signaling within inhibitory neurons alone is sufficient to stabilize MFR set point by disabling homeostatic compensation. In behaving mice, continuous long-term NMDAR blockade in CA1 stably suppresses MFR across sleep-wake states without inducing a compensatory response. These results highlight NMDARs as a network-wide MFR set-point modulator and extend NMDAR function beyond its canonical role in synaptic plasticity, raising the possibility that some NMDAR-dependent behavioral effects are mediated by homeostatic regulation of activity set points.

AMPA-type glutamate receptors (AMPARs) mediate fast excitatory neurotransmission, and their surface levels play a crucial role in determining synaptic strength during learning. However, these levels are often altered in neurological disorders, including Alzheimer's disease (AD). We previously found that CALM, a highly validated genetic risk factor for Alzheimer’s disease, regulates synaptic plasticity by selectively facilitating the endocytosis of ubiquitinated GluA1-homomeric Ca2+-permeable (CP) AMPARs. This mechanism relies on ubiquitin recognition, but is independent of clathrin. Here, using a combination of amyloid beta (Aβ) aggregation assays, mouse genetics, electrophysiological recordings and proteomics, we report that CALM directly bind Aβ42 in vitro thereby preventing its aggregation. Moreover, we show that neuronal loss of CALM leads to increased levels of Aβ species, accelerates Aβ pathology and impairs neurotransmission in an AD mouse model. Finally, TMT-labelling mass spectrometry elucidates the proteomic changes at synaptic resolution, shedding light on the role of CALM during AD progression.

Ketamine produces rapid and sustained antidepressant effects after brief exposure to a single dose. Counterintuitively, while ketamine acts primarily as a blocker of postsynaptic N-methyl-D-aspartate receptors (NMDARs), increased signalling at glutamatergic synapses has been reported. Due to technical limitations, however, it remains unclear whether ketamine directly increases presynaptic glutamate release or acts via postsynaptic or network-level mechanisms. To address this knowledge gap, we used presynaptic capacitance measurements to directly monitor glutamate release in a cerebellar synapse. Ketamine increased glutamate release within minutes and this effect persisted >30 minutes after washout. MK-801, another NMDAR blocker, had no effect on glutamate release. Mechanistically, we show that the ketamine-mediated enhancement of presynaptic release results from an increase in both calcium influx and the number of release-ready vesicles. Our data uncover a rapid effect of ketamine on key presynaptic properties of central glutamatergic synapses, which has important implications for the development of antidepressant drugs.

Efficient neuronal communication relies on the activity-dependent exo- and endocytosis of synaptic vesicles at the active zone. It is established that for each vesicle that fuses at the synapse equivalent membrane complement is internalized, however how this close coupling of exo- and endocytosis is achieved in molecular terms is still not fully understood. BAR proteins sense and stabilize membrane curvature and often act as hubs connecting various machineries (such as cytoskeleton or dynamin) facilitating membrane internalization. With the use of human neurons, cell lines, and model membranes we demonstrate that activity-dependent presynaptic membrane tension fluctuations could act as link between exo- and endocytosis and identify a tension change-driven conformational switch in a member of the early endocytic machinery that activates formation of endocytic invagination as well as recruitment of the downstream effectors such as dynamin.

Within a neuronal network, the strength of individual synaptic connections affects network dynamics. The ability of synapses to release multiple quanta of neurotransmitter (multivesicular release; MVR) in response to a single action potential can increase the range of synaptic strength (Rudolph et al., 2015). At certain glutamatergic synapses, protein kinase A (PKA) signaling pathways have been implicated as a mechanism to control the extent of MVR, through putative regulation of the pool of readily-releasable synaptic vesicles (Vaden et al., 2019). Is this mechanism more general across cell types or limited to particular synapses? Moreover, does PKA signaling activation result in similar effects on the physiological properties of synapses releasing different types of neurotransmitters, such as glutamate and GABA? In this project we examine the effects of activating PKA signaling on synaptic transmission in cultured excitatory glutamatergic and inhibitory GABAergic neurons in primary culture from mouse hippocampus. Using electrophysiology and pharmacology, we find that activating PKA signaling with forskolin treatment increases the extent of MVR in glutamatergic synapses, with minimal effects on release probability. Conversely, in GABAergic neurons, we observe a clear effect of forskolin treatment on release probability, and a more ambiguous effect on MVR. We observed a similar trend on release probability effects by PKA activation on synaptic transmission in glutamatergic and GABAergic human neurons induced from pluripotent stem cells to their respective murine primary neuron counterpart. Taken together, our results will provide insight into mechanisms of synaptic diversity between cell types and the degree to which these mechanisms are conserved between species.

The brain is composed of billions of neurons, which connect to each other using electrical or chemical signals. To sustain this complex system, the brain consumes an intensive amount of oxygen and glucose to fulfil the needs in ATP. As a result, neurons produce a robust amount of reactive oxygen species (ROS). Uncontrolled production and insufficient elimination of ROS from the antioxidant system can be detrimental and damage essential macromolecules such as enzymes and structural proteins leading to brain dysfunction. The Ataxia-Telangiectasia Mutated (ATM) kinase is mutated in the disease ataxia-telangiectasia, a rare autosomal disorder characterized by neuronal degeneration, cancer, immune deficiency, increased sensitivity to ionizing radiation, growth retardation and premature aging. Aside from the classical nuclear function of ATM in sensing DNA double strand breaks, ATM also exploits cytosolic signaling pathways by responding, for example, to oxidative stress. In fact, several mechanisms point towards a protective effect of ATM against ROS-induced damage of the brain. One of the hallmarks of ATM is the maintenance of proteostasis by coordinating both the ubiquitin proteasome system and the lysosome/autophagy pathway. Recently, we also showed that ATM can phosphorylate the actin binding protein Drebrin to improve the stress resilience in dendritic spines, the post-synaptic site of glutamatergic synapses suggesting that ATM control of the actin cytoskeleton may also protect against ROS induced synapse dysfunction. Further understanding how dysregulation of ATM may lead to anomalies in the physiological properties of the synapse will open new avenues on therapeutic strategies for patients with Ataxia-Telengiectasia or other neurodegenerative diseases.

With the prospect of a better translation compared to mouse-derived systems, hIPSC-derived neuronal networks are a promising model system for overcoming the valley of death in cognitive drug discovery in the neurodegenerative disease field. However, this requires a network capable of synapse strengthening and weakening, which is supported by human astrocytes and microglia: A fully human model of a minimal brain circuit. So far, the field has been able to show that human induced neurons cultured for a sufficiently long time express pre-synaptic markers and show activity in multi-electrode recordings. But convincing post-synaptic immunostainings are rare, as are functional assays investigating pre- and post-synaptic function and synaptic maturation. In our hIPSC-derived neuronal cell culture, we validated two functional assays, calcium and FM1-43 dye imaging. While calcium imaging can give insights into cell activity and pre- and post-synaptic function, FM1-43 dye, a fluorescent stryryl dye, inserts into the membrane and is taken up through synaptic vesicle recycling. The dye release kinetics follow an exponential decay indicating mature synaptic vesicle recycling. These functional pipelines can be used for the determination of synaptic maturity, and in the future for the evaluation of compounds that could treat cognitive diseases.

Recent genomic studies propose that schizophrenia genetic risk converges on the structure and function of the synapse. To improve therapeutics and prevention strategies, understanding the context where genetic risk converges on vulnerable biological systems is crucial. Genes implicated in schizophrenia risk have known roles in synaptic processes such as synaptogenesis, neurite outgrowth and mature synaptic transmission and plasticity. These synaptic events rely on rapid protein production driven by local translation of gene transcripts in context-specific synapses. However, no prior study has explored the relative contribution to risk from different synaptic subtypes, or from locally vs somatically translated genes.

Using published single-synapse transcriptomics and proteomics datasets in conjunction with the latest case-control genotype data, we examined enrichment for common genetic variants and rare de novo coding variants associated with schizophrenia in localised transcripts and proteins across a range of synapse subtypes, cell types and brain regions.

We observed an enrichment of common and rare schizophrenia associations in inhibitory and excitatory synapses of the hippocampus and cortex. In both hippocampal and cortical excitatory and inhibitory neurons, we observed an enrichment of schizophrenia associations in genes encoding localised synaptic transcripts compared to locally translated genes without synaptic function and genes translated somatically. The associated gene sets were related to synaptic structure, glutamatergic and GABAergic signalling pathways.

Our results suggest that schizophrenia risk variants specifically impact locally translated genes with synaptic function, particularly in inhibitory and excitatory synapses of the hippocampus and cortex. In turn, our findings support the involvement of early synaptic connectivity and maturation during development and mature synaptic transmission and plasticity in schizophrenia pathogenesis. This warrants future research focussing on the investigation of activity-dependent synaptic genes under the control of local translation, particularly at the isoform level.

Understanding the intricate interplay between dendritic, cytoskeletal organization, and synaptic compartmentalization is crucial in deciphering neuronal function. While much is known about the role of actin filaments and microtubules in excitatory neurons, their organization in inhibitory GABAergic neurons remains relatively understudied. This research delves into the distinct cytoskeletal organization within aspiny GABAergic Parvalbumin-positive neurons, particularly focusing on their dendritic microtubule arrangement and its interaction with postsynaptic sites. Unlike excitatory neurons where actin patches typically localize to dendritic spines, Parvalbumin neurons exhibit a prominent mesh of cortical actin outlining the dendritic membrane. Moreover, postsynaptic densities of excitatory synapses in Parvalbumin neurons are directly located on the dendritic shaft, in close proximity to microtubules. Investigating these features under physiological conditions and in a mouse model of Autism spectrum disorder, we uncover insights into the intricate synaptic-cytoskeletal interactions. Notably, a global knockout of a synaptic protein implicated in Autism spectrum disorder results in an Autism-related phenotype, mirroring outcomes achieved by specific knockout in GABAergic or Parvalbumin-expressing neurons. This suggests that Autism dysfunction may predominantly stem from within interneuronal populations. Our findings illuminate the distinctive cytoskeletal organization in GABAergic neurons and propose a role for the ASD-associated mutation in altering synaptic-cytoskeletal interactions, potentially disrupting molecular trafficking pathways.

Sumoylation is a post-translational modification essential to the modulation of several neuronal functions, including neurotransmitter release and synaptic plasticity. Altered sumoylation has been associated with neurological disorders. Here, we demonstrated for the first time that Oligophrenin-1 (OPHN1) is a SUMO target. OPHN1 is a Rho-GAP protein highly expressed in neurons. In humans, all Ophn1 mutations cause the loss of function of OPHN1 leading to syndromic intellectual disability (ID). In neurons, OPHN1 regulates dendritic spine density and architecture, actin dynamics and AMPA receptor (AMPAR) trafficking.

By combining molecular biology with live imaging and super resolution microscopy we addressed the role of sumoylation in controlling OPHN1 function in hippocampal neurons. Excitingly, we demonstrated that sumoylation controls the activation state of OPHN1 by promoting a conformational change that induces the release of the autoinhibitory interaction between the BAR and the GAP domain of OPHN1. By this mechanism, sumoylation tunes the actin dynamics, which in turn impacts dendritic spine density and architecture. Furthermore, we demonstrated that a novel ID-related missense mutation G412D compromises OPHN1 sumoylation by altering the proper conformation of the protein and impairing its physiological regulation. This rises thrilling hypothesis that compromised sumoylation participates to synaptic dysfunctions associated to the ID phenotype by altering the autoinhibitory mechanism that controls OPHN1 function. Finally, we demonstrated that repairing the proper OPHN1 3D structure in the G412D mutant restore OPHN1 sumoylation and its SUMO-dependent conformational changes.

Altogether our results, beside identifying a new SUMO target at the synapse, add a novel level of complexity to the regulation of OPHN1 function in neurons further confirming the emerging role of sumoylation as key regulator of synaptic function.

Seizures arise from disruption in mechanisms that control neuronal excitability and the M-current is one such mechanism. This low-threshold potassium current modulates neuronal excitability and suppresses repetitive firing. M-channels assemble as tetramers of Kv7 subunits and mutations in the KCNQ2 gene, encoding Kv7.2, are linked to epilepsy and intellectual disabilities. Despite this common epileptogenic mechanism, understanding M-channel regulation remains limited. We identified type 1 TARPs, well-known Transmembrane AMPA-receptors Regulatory proteins, as new interactors of the Kv7.2 subunit of M-channels, characterized the Kv7.2-TARPs interaction and assessed its functional relevance for neuronal excitability. Through co-immunoprecipitation and proximity-ligation assays, we demonstrated that type 1 TARPs and Kv7.2 interact in neurons, with this interaction increasing upon neuronal activation. Co-expression of TARPs (γ2, γ3 and γ4) with Kv7.2 enhanced the channel’s surface expression and potentiated Kv7.2-mediated currents. Conversely, silencing TARP-γ2 in cortical neurons reduced the M-current, suggesting that endogenous TARP-γ2 is necessary for normal M-channel function. TARP-γ2 depletion also impaired the nano-structure organization of Kv7.2 channels. Notably, an intellectual disability-associated variant of TARP-γ2 failed to potentiate M-currents. In a knock-in mouse harbouring this variant, the hippocampal M-currents were diminished, as determined by a 50% reduction of the medium after burst-hyperpolarization and a decreased response to retigabine, an M-channel activator. Moreover, these defects led to increased susceptibility to pentylenetetrazol-induced seizures, indicating that disruption of the regulation of M-channels by TARP-γ2 is epileptogenic.

Collectively, this work provides groundbreaking evidence of a synaptic protein directly involved in neuronal intrinsic excitability regulation, with important implications for epilepsy.

Pools of proteins in solution are known to undergo liquid-liquid phase separation (LLPS) by interacting via multivalent, low affinity binding sites, such as intrinsically ordered regions or proline rich motifs. Recently, LLPS has been shown for a mixture mimicking the synaptic vesicle (SV) pool in neurons [1]. In the study they mixed synapsin1, α-synuclein and SVs and showed that synapsin can create a condensate which is able to recruit the other two constitutents. Importantly, the stability and structure of the resulting condensates is dependent on the relative concentration of the different proteins and SVs. This raises the question to which extent a neuron can control SV release via changing the concentration of the involved proteins or varying specific interactions. To obtain a detailed understanding of the underlying physical mechanism involved in LLPS of the SV cluster, we use the computational approach of patchy particles [2]. Patchy particles are spheres wíth smaller interaction sites on top. This allows us to include directional interactions which contain structural information. Thereby, we preserve crucial molecular information, while the coarse-grained character enables simulations at sizes sufficiently large to study LLPS.

With our model, we link microscopic and molecular information inside a condensate to macroscopic porperties of the cluster. We are specifically interested to which extend competition for binding sites and valency of the proteins affects the phase behaviour.

[1] C. Hoffmann et al., Journal of Molecular Biology, 166961, 2021

[2] Jorge R. Espinosa et al., PNAS, Vol. 117, Pages 13238-13247, 2020

The murine cortical layer 6b (L6b) is considered to be a remnant of the subplate, which contains some of the earliest generated cortical neurons that are involved in the establishment of thalamocortical and intracortical circuits during development. In the murine sensorimotor cortex, Drd1a-Cre+ neurons in L6b project to higher-order thalamic nuclei and are the only cortical projection neurons responsive to the wake-promoting neuropeptide orexin, suggesting that L6-Drd1a neurons play an important role in sensory processing, brain state modulation, attention, and cognition (Hoerder-Suabedissen et al., 2018; Zolnik et al., 2023). In the medial prefrontal cortex (mPFC), whether these neurons play an analogous role in gating cortical arousal remains poorly understood. Thus, we investigated the effects of ‘silencing’ L6-Drd1a neurons in a conditional Snap25 KO mouse model (Drd1a-Cre+/-;Snap25fl/fl) on the dopaminergic modulation of network activity in mPFC acute brain slices using planar high-density multielectrode arrays (MaxWell Biosystems) and patch clamp recordings. In WT slices, we found a significant increase in delta activity and evoked network spiking in the infralimbic mPFC after D1 receptor agonist SKF-81297 administration, which was eliminated upon chronic silencing of L6-Drd1a neurons. Application of the D2 receptor agonist quinpirole had no effect in WT slices, but chronically silencing L6-Drd1a neurons unmasked a quinpirole-induced increase in delta activity and evoked network spiking in the infralimbic mPFC, suggesting that L6-Drd1a neurons mediate an inhibitory effect of D2 receptor activation and/or a compensatory shift in D2 receptor function. We followed up these findings with whole-cell electrophysiological characterization of these L6-Drd1a neurons. Our results provide novel insight into the dopaminergic modulation of L6-Drd1a neurons in the mPFC that may contribute to their orexinergic response in mediating attention and arousal, with relevant therapeutic implications for stress-related neurodevelopmental disorders.

Modifications of nucleotides in mRNA expands the repertoire of post-transcriptional regulation and adds a novel dimension due the potential for dynamic and epigenetic control. Dynamic methylation in mRNA occurs at cap adjacent nucleotides on the ribose and on adenosine if the first nucleotide, and internally most frequently on adenosine (m6A), cytidine (m5C) and on the ribose.

Cap methyl transferases (CMTrs) methylate cap-adjacent nucleotides of animal, protist and viral mRNAs at the 2` position of the ribose (cOMe). Animals generally have two CMTrs, while trypanosomes have three and many viruses encode one in their genome. Even though cOMe is the most prominent mRNA modification, their function remains largely unknown.

Here, we used a Drosophila model to generate a double knock-out of both CMTrs. Intruiguingly, these flies are viable and CMTrs reduntantly add cOMe to the first cap-adjacent nucleotide. Consistent with prominent neuronal expression, they have a reward learning defect that can be rescued by conditional expression in mushroom body neurons before training. Among CMTr targets are cell adhesion and signaling molecules. Many are relevant for learning, and are also targets of Fragile X Metal Retardation Protein (FMRP). Like FMRP, cOMe is required for localization of untranslated mRNAs to synapses and enhances binding of the cap binding complex in the nucleus. Hence, our study reveals a mechanism to co-transcriptionally prime mRNAs by cOMe for localized protein synthesis at synapses.

Haussmann et al., 2022 Nat Comm 13, 1209. Dix et al., 2022 RNA 28, 1377. Anreiter et al., 2023, BioEssays 45, 2200198.

Synaptic vesicles (SVs) play a key role in interneuronal communication. The molecular composition of SVs was thoroughly investigated in the previous studies, mainly by proteomic analysis and fluorescent microscopy. A model of an average SV was shown to contain proteins responsible for NT uptake, vesicle docking, and fusion with the plasma membrane (Takamori S. et al, Cell, 2006), however, understanding the differences between individual SVs is lacking. We used cryo-electron tomography (cryo-ET) to analyze individual SVs, isolated from mouse brain tissue and inside synapses of cultured neurons, and to visualize their morphological and molecular details at molecular resolution. We quantitatively characterize several classes of small proteins on the surface of SVs and long proteinaceous densities within SVs. We identified large V-ATPases and its additional partner that we propose to be Synaptophysin. Interestingly, V-ATPases were randomly distributed on the surface of SVs irrespective of vesicle sizes. We demonstrated a soccer-ball symmetry of clathrin coats on a subpopulation of the isolated SVs and SVs inside cells. Finally, we provide descriptions of clathrin baskets without SV inside, their mini-coat type of symmetry, and their preferred localization in statistically significant proximity to the cell membrane. Our analysis advances understanding of the diversity of SVs and their molecular architecture.

Calcium imaging takes advantage of naturally occurring changes in internal calcium concentrations following action potentials. Modern genetically encoded sensors have the necessary sensitivity to also report sub-threshold calcium influx through calcium-permeable glutamatergic NMDA receptors at individual synapses. By imaging neuronal cultures transduced with synapsin promoter-driven GCaMP6f AAV in the absence of magnesium and presence of TTX, we can detect dendritic spine calcium transients from single presynaptic vesicle release events via NMDA receptor-dependent influx of calcium. We have created an automated analysis pipeline to characterize the effects of compounds on synaptic calcium transients. This pipeline can be used to screen compounds for effects on synapse strengthening or weakening by reporting changes in presynaptic vesicle release (frequency at individual synapses), post-synaptic receptor number (amplitude, decay time) and total number of active synapses. We established the dynamic range of synaptic activity using the NMDA receptor antagonist APV, and the phorbol ester PDBu which interacts with Munc-13 in presynaptic boutons leading to an increase in release probability of synaptic vesicles. Using selective blockers, we then tested the contribution of AMPA receptors, L-type calcium channels and internal calcium sources, to synaptic calcium transients. As a proof of principle of the pipeline, we tested patient-derived NMDA receptor autoantibodies found in patients suffering from anti-NMDA receptor encephalitis. We found that these NMDA receptor autoantibodies decreased the number of spontaneously active synapses observed by calcium imaging and decreased the frequency and decay time of synaptic calcium transients to varying degrees. We are now additionally testing the effects of the psychedelic compounds ketamine and LSD, and memantine: an NMDA receptor antagonist prescribed to mitigate and slow the symptoms of advanced Alzheimer’s disease and dementia. Our goal is to establish a pipeline for drug development to screen for compounds that strengthen or weaken synapses and may treat cognitive diseases.

Neurons comprise only less than half of the nervous system, yet some of all those other cells have been largely ignored. Microglia are the tissue resident macrophages of the brain. These cells play a key role in shaping neuronal network formation during development and subsequently help to maintain a stable and healthy brain. After birth our brain contains overabundant neuronal connections and some of these must be refined and eliminated. Microglia scavenge the unnecessary connections via synaptic pruning and shape the neuronal network. One of the paradigms explaining synaptic pruning suggests that neuron-microglia interactions may be activity-dependent. However, it remains unclear how microglia are involved in synaptic vesicle release and trafficking under normal conditions, and whether there is direct or indirect interaction with SNARE complex proteins.

To evaluate how neuronal activity changes the behaviour and dynamic of microglia, we used mouse model and we selectively removed parts of the Snap25 gene, and it produced a non-functioning protein in a subset of cortical layer 5 projection neurons and some hippocampal dentate gyrus neurons and thus also eradicated their ability to release synaptic vesicles in a regulated manner from around the time of birth in a chronic fashion.

We observed that abolition of Snap25 expression in selected subset of layer 5 neurons and dentate gyrus reduced microglial activity and affected morphological properties. Moreover, chronic synaptic ‘silencing’ altered synaptic maturation resulting in increased synaptic density and increased engulfment of presynaptic compartments. In addition, abolition of Snap25 affected neuron-microglia communication by disrupting normal chemokine and phagocytic signalling. Lastly, we observed sex-specific changes in neuron-microglia communications, and this indicated that male mice were more sensitive to abolished Snap25-mediated release of synaptic vesicles during brain development.

The cerebral cortex is the brain region responsible for crucial functions that make us humans unique. Cortical areas are specified prenatally by a complex interplay of intrinsic genetic programs and extrinsic patterning cues mainly provided by the early thalamocortical afferent (TCA). These axons reach the immature cortical areas during mid-fetal gestation and establish transient synaptic circuits with the early-born mature neurons of the transient subplate zone (SP). However, the role of these early thalamocortical connections in shaping the development and specification of the cerebral cortices has not been investigated in the human brain due to technical limitations and the limited availability of human fetal brain samples.

In this study, we traced the early TCA in human post-abortion mid-fetal brains and showed that they do not only project to the SP neurons but also send collaterals to cortical germinal compartments in humans. Here, they establish close anatomical connections with the outer radial glial cells (oRGC), a specific population of cortical progenitors found in gyrencephalic mammals. TCA can release important neurotrophins in the cortex in an activity-dependent manner upon the establishment of synaptic circuits with the SP. Amongst these, VGF (non-acronymic), that promoted further functional activity in SP neurons alongside neurogenic differentiation over proliferation of cortical progenitors in in vitro models of human cortical development.

Furthermore, thalamic Vgf expression appeared to have shifted over evolution from sensorial nuclei to the associative mediodorsal nucleus (MD) in human. The MD directly innervates the prefrontal cortex (PFC), which shows a developed layer 4 only in primates. Given VGF's layer-specific effects on promoting the generation and maturation of layer 4 neurons in rodents, our findings suggest a role for VGF in this unique feature of the human PFC.

This study sheds light on the interplay between early synaptic circuits and evolution of the the human cortex.

Synapses rely on the dynamic nature of their organelles and proteome to promote and support functional and structural modifications underlying synaptic plasticity. The endoplasmic reticulum (ER) plays a pivotal role in these processes as it is responsible for calcium homeostasis, translocation and secretion of proteins, and mitochondrial dynamics, among other functions, rearranging its functional according to cellular demands (1). In dendrites, ER has mainly a tubular morphology that may expand into dendritic spines, determining their responsiveness to synaptic activity (2, 3). However, limitations on the spatiotemporal resolution of the used methodologies have constrained the understanding on how synaptic transmission affects ER’s nanoscale organization and dynamics within dendritic spines. Hence, we started by characterizing the nanoscale organization of ER in dendritic spines of mature hippocampal neurons using STED microscopy. Spine ER can present a variety of structural morphologies, ranging from simple tubules to complex structures with multiple subdomains that cover most of spine area.

To understand if and how ER structural diversity correlates with synaptic transmission, we took advantage of our recently published event-triggered STED (etSTED) nanoscopy approach. EtSTED is a sample-adaptative method that enables rapid and localized STED time-lapse imaging at the site of detected subcellular events (such as calcium spikes triggered by synaptic activity) (4). Besides resolving the ER structural organisation, the versatility of this technique also gave us the opportunity to study how the ER dynamics is affected by synaptic activity and synaptic strength in this subcellular compartment.

References:

(1) N. L. Chanaday, E. T. Kavalali, Current Opinion in Neurobiology. 73, 102538 (2022).

(2) N. Holbro, Å. Grunditz, T. G. Oertner, Proc. Natl. Acad. Sci. U.S.A. 106, 15055–15060 (2009).

(3) A. Perez-Alvarez et al., Nat Commun. 11, 5083 (2020).

(4) J. Alvelid et al., Nat Methods. 19, 1268–1275 (2022).

Subplate neurons are among the earliest generated neurons in the cerebral cortex that orchestrate the development of the intra and extracortical connections. A large proportion of these neurons will undergo cellular death throughout postnatal development. The remaining cells will form the mature mouse brain’s thin and compact cortical layer 6b (L6b). The functional aspects of L6b have received little attention compared to the other layers. Recent studies have argued that the extensive connections and cortico-thalamic-cortical loops of L6b neurons might enable sensory processing to extract salient features of sensory stimuli. This layer is characterized by being the only cortical layer containing projection neurons that respond directly to the polypeptide Orexin, a neuropeptide known to be involved in cortical arousal and emotive behaviour. Nevertheless, the exact mechanisms underlying these responses are not fully understood. We hypothesize that cortical circuits activated by these orexin-sensitive L6b neurons (Drd1a-Cre+) are not only responsible for detecting salient features of sensory stimuli but are also involved in the regulation of emotional states. Here, we show the Drd1a-Cre+ neurons are selectively sensitive to orexin and gate the activation of the prefrontal network. We also found that inactivating these neurons completely hinders the orexinergic activation of the prefrontal cortex, and demonstrates the anxiolytic effects of chronically silencing this population (Drd1a-Cre+/+:Snap25fl/fl) across the cortical mantle on several anxiety-related behavioural paradigms. This suggests that the orexin-responsive cortical layer 6b neuronal population is a potential target for the manipulation of anxiety-like behaviour.

Neurohypophysis (posterior pituitary) is a major neuroendocrine interface in the brain through which water homeostasis is maintained. Neurohypophysis majorly consists of glial pituicytes, neuropeptides oxytocin- and vasopressin-loaded loaded synapses, and permeable capillaries. We recently identified that pituicyte-derived secreted factor can regulate neurohypophyseal neurovascular morphogenesis. However, the role of other secreted factors expressed in neurohypophysis in neurovascular morphogenesis is unknown. Towards this goal, we have been employing pharmacological and genetic perturbations to explore the roles of candidate molecules that could regulate neurohypophyseal synapse morphogenesis. Our studies of the glial pituicytes are expected to reveal novel players in the development of a key neuroendocrine interface conserved in vertebrates.

TIMM50 is a vital subunit of the TIM23 complex, involved in mitochondrial protein import. Defects in TIMM50 reportedly lead to severe neurological phenotypes characterized by mitochondrial epileptic encephalopathy, Leigh syndrome, and intellectual disability.

In this research, we characterized the biochemical and molecular effects of TIMM50 mutation in human patient fibroblasts and noted significant decrease in the protein level of TIMM50, as well as the core components of TIM23 complex - TIMM17A/B and TIMM23. Upon knockdown of TIMM50 in mice neurons, we observed a similar protein expression behavior of the TIM23 core components, suggesting a promising mimicry model for understanding TIMM50 disease in mammalian neurons.

Strikingly, our analysis indicates that TIMM50 and TIM23 complex deficiency does not affect the steady state level of most mitochondrial matrix and inner membrane proteins in patient fibroblasts as well as mice neurons. However, it does lead to a significant and specific decrease of many oxidative phosphorylation and mitochondrial ribosome complex subunits. This led to declined respiration rates in both cases, decreased mitochondrial membrane potential in fibroblasts and neuronal cell bodies, reduced ATP levels and defective mitochondrial trafficking in neuronal processes, possibly contributing to developmental defects in patients of TIMM50 disease. Finally, increased neuronal electrical activity correlated with reduced KCNJ10 and KCNA2 potassium channel levels, likely underlying patients’ epileptic phenotype.

Local translation is a fundamental process in synaptic plasticity and neuronal homeostasis. Translation of most mRNAs is regulated at the initiation step in a cap-dependent manner by initiation factors, including eukaryotic translation initiation factor 4E (eIF4E). Regulation of eIF4E has been shown to be critical for synaptic plasticity. eIF4E is exclusively phosphorylated by MAPK interacting kinases 1 and 2 (MNK1 and MNK2). MNK1 and MNK2 are non-essential kinases mainly activated by the p38 and ERK/MAPK pathways. In the brain, the MNKs have been implicated in many aspects of adaptive behavior, including learning and social behaviors. However, the individual regulatory mechanisms of MNK1 and MNK2 are poorly understood. In particular, the role of MNK2 in the nervous system has not previously been investigated.

Here, we have characterized the individual roles of MNK1 and MNK2 in the nervous system. We show that MNK1 and MNK2 are expressed in overlapping neuronal populations throughout cortex and hippocampus. To investigate molecular differences, we used a combination of ex vivo brain slices and subcellular fractionation of brain tissue in knockout mice lacking each kinase. In addition, we performed proteomic analysis on isolated synaptoneurosomes. This data shows that MNK1 regulates the expression of proteins related to ribosome biogenesis, whereas MNK2 regulates proteins linked to the pre-and post-synaptic function. Furthermore, the behavioral phenotypes are distinctly different; MNK1KO mice show reduced social interaction and memory, whereas MNK2KO mice show increased social interaction and object recognition. Taken together, these results suggest that MNK1 and MNK2 have distinct functional roles in the nervous system.

Phospholipid-phosphatase related protein 3 (PLPPR3, previously known as Plasticity Related Gene 2 or PRG2) belongs to a family of transmembrane proteins, highly expressed in neuronal development, which regulate critical growth processes in neurons. Prior work established crucial functions of PLPPR3 in axon guidance, filopodia formation and axon branching. However, little is known regarding the signaling events regulating PLPPR3 function. With the use of mass spectrometry 26 high-confidence phosphorylation sites were identified in the intracellular domain of PLPPR3. Biochemical characterization established one of these – S351 – as a bona fide phosphorylation site of PKA. Experiments in neuronal cell lines suggest that phosphorylation of S351 does not regulate filopodia formation. Instead, it regulates binding to BASP1, a signaling molecule previously implicated in axonal growth and regeneration. Interestingly, both PLPPR3 intracellular domain and BASP1 enrich in presynapses in primary neurons. We propose that the presynaptic PLPPR3-BASP1 complex may function as novel signaling integrator at neuronal synapses.

Parkinson's disease (PD) is characterized by the preferential degeneration of dopaminergic neurons in the substantia nigra, yet the precise molecular mechanisms driving this selective vulnerability remain elusive. Emerging evidence implicates excessive dopamine oxidation as a common pathological phenotype across various genetic and idiopathic forms of PD. Concurrently, mutations in PD-associated genes underscore the critical role of mitochondrial function in disease pathogenesis, potentially initiating a cascade of mitochondrial oxidant stress culminating in elevated dopamine oxidation.

Given the ATP-dependent packaging of dopamine into synaptic vesicles by vesicular monoamine transporter 2 (VMAT2), disrupted mitochondrial function may lead consequently to cytosolic dopamine accumulation and subsequent oxidation into neurotoxic quinones. To elucidate this complex relationship, I employed CRISPR-edited DJ-1 knockout human induced pluripotent stem cells differentiated into dopaminergic neurons as a model system. Through my investigations, I aim to uncover the mechanistic underpinnings linking synaptic mitochondrial dysfunction, ATP depletion, and dopamine oxidation in the context of Parkinson's disease (PD).

My investigations revealed a depolarized mitochondrial state accompanied by decreased ATP levels in mutant neurons, suggesting a correlation between mitochondrial dysfunction and cytosolic energy depletion. Furthermore, I observed diminished VMAT2 activity alongside abnormalities in vesicle dynamics, implying dysregulated vesicle formation and vesicular dopamine packaging in PD neurons, potentially contributing to the observed excess of oxidized dopamine.

In summary, these findings underscore the significance of ATP and dopamine metabolism in dopaminergic neuronal vulnerability in PD, shedding light on potential therapeutic targets aimed at restoring mitochondrial function and mitigating dopamine oxidation to impede disease progression, particularly within the intricate dynamics of the synapse.

Autism is a developmental condition affecting 1-2% of people in the world, with high heterogeneity and high heritability. This high heterogeneity leads to a spectrum of neurobiological differences, making diagnosis and treatment approaches difficult to develop. The symptoms vary from alteration in social communication and behavior to stereotyped behavior. The high heterogeneity is not exclusive to the variety of symptoms but also the causative factors: there are hundreds of genes identified to be associated with autism. Many of these genes converge onto shared pathways, including genes associated with transcriptional and translational regulation, and synapse formation and function. Several proteins at the synapse are linked to autism, including those involved in activity-dependent synaptic protein synthesis.

To address heterogeneity in autism one approach is to focus on the behavioral phenotypes and investigate the underlying biological mechanism. Here, to better assess heterogeneity in a mouse model of autism, we use a multidimensional phenotyping approach of the Shank3KO mice line, and we use this detailed behavioral information to examine how molecular phenotypes correlate with behavioral phenotypes. In particular, we use open-source toolkits, including DeepLabCut, SimBA, and DeepOF, to perform deep phenotyping of individual animals across multiple behavioral domains. Using this approach, we identified three clusters of mice with distinct behavioral profiles. To identify molecular mechanisms linked to behavioral phenotype we use Data Independent Acquisition (DIA) proteomics of isolated synaptoneurosomes from multiple brain regions. We want to investigate if clusters of mice with similar behavioral profiles also share similar molecular profiles, independent of genotype or sex.

Touch sensation is crucial for interacting with our physical world. The perception of touch occurs when physical forces are transformed to electrical stimuli by the opening of mechanically-activated ion channels in sensory nerves. Here we identify a new mechanically-activated ion channel, Elkin1, that, when ablated in mice, leads to a profound behavioural touch insensitivity, such that Elkin1-/- mice respond only 50% of the time to light brushing of cotton swab compared to ~90% response rate of wildtype mice. The neuronal basis of this touch insensitivity in Elkin1-/- mice was caused by a loss of mechanically-activated currents (MA-currents) in around 50% of sensory neurons that are activated by light touch (low threshold mechanoreceptors, LTMRs). Reintroduction of Elkin1 into sensory neurons from Elkin1-/- mice acutely restored MA-currents. Small-interfering RNA-mediated knockdown of Elkin1 from human sensory neurons reduced indentation induced MA-currents, therefore suggesting that Elkin1 is a core component of touch transduction in mice and potentially in humans. Additionally, we find that Elkin1 is present in many nociceptive sensory neurons which detect potentially damaging and painful mechanical force. Higher forces are required to elicit responses from the sensory neuron cell bodies of Elkin1-/- mice. Furthermore, Elkin1-/- mice show decreased sustained firing of nociceptor terminals of only mechanically activated C-fibres, but not polymodal C-fibres. Indentation of cell bodies of polymodal C-fibres (marked by IB4) showed that these nociceptor subtype has inherently less MA-currents. We further utilized an ultra-low input proteomic approach to reveal subtype specific proteome of IB4 positive, non-peptidergic sensory neurons and TrkA positive, peptidergic sensory neurons. This phenotype specific proteomic approach has the potential to reveal protein candidates important in mechanosensation of nociceptors.

The extracellular matrix (ECM) is a dense meshwork in which neurons and glial cells are embedded in the central nervous system (CNS).1 ECM components such as lecticans co-aggregate to form highly ordered structures around neurons and synapses. ADAMTS-5 is a zinc metalloproteinase that primarily cleaves lecticans (especially aggrecan).1 Several studies showed an upregulation of ADAMTS-5 in neurodegenerative diseases such as Alzheimer’s disease. Overexpression of ADAMTS-5 leads to remodelling and disruption of ECM architecture.2 Thus, ADAMTS-5 is a promising target for therapeutic intervention. Pharmaceutical companies have investigated the potential ADAMTS-5 inhibitors for osteoarthritis, but none of them successfully completed all phases of clinical trials. Currently, there are no therapies specifically targeted at ADAMTS-5. In this study, we used drug repurposing, as it offers notably cost-effective solutions for identifying new treatments for diseases. We conducted in-silico structure-based virtual screening, followed by in-silico ADME-Tox (chemical absorption, distribution, metabolism, excretion and toxicity) profiling to identify novel ADAMTS-5 inhibitors in FDA-approved list. The catalytic domain of ADAMTS-5 was obtained from the Protein Data Bank and then subjected to molecular docking against set of FDA-approved drugs sourced from the ZINC database. For in-vitro experiments, we modified existing fluorogenic substrate-based method of ADAMTS-5 proteolytic activity detection to be compatible with TaqMan chemistry and high-throughput screening. Proposed method allows detection in real-time PCR machine as well is in standard fluorimeter. Our finding shows that in-silico screening protocol makes valuable predictions and designed in-vitro method allows to check these predictions in a really high-throughput manner, being compatible with a large variety of common laboratory devices. These findings are promising and pave the way for further exploration of targeted therapies for ADAMTS-5 inhibition.

References:

1. Y. Mohamedi at al., Biomolecules, 2020, 10, 1–17.

2. M. S. Gurses et al., Aging Dis., 2016, 7, 479–490.

Recent developments in cryo-electron tomography combined with cryo-CLEM have led to numerous insights into biological fine-structures on the molecular level. Usually achieved by sequential labeling of proteins of interest (POIs) with fluorescent dyes and heavy metals for light and electron microscopy, respectively. This project highlights a complete solution to address extracellular and/or cytosolic POIs by labeling them with combined fluorophore/heavy metal conjugates, an approach that is presented as cryoCLASSICAL (for cryoCLEM-compatible Live And Specific Solution for IntraCellular Au Labelling).

In brief, the SNAP-tag (20 kDa) labeling system is selectively addressed with a new generation of compounds that comprise 1) a cell impermeable or penetrating agent to target POIs in living cells, 2) a dye with high brightness resistant to vitrification, and 3) a 1.4 nm gold nanoparticle (AuNP). This approach is circumventing the use of antibodies in fixed cells, yielding specificity in the most native context and is characterized by a total small size of ~25 kDa, smaller than the size of green fluorescent protein. Being a minimally perturbative system, cryoCLASSICAL allows visualizing POIs independent of their size, shape, and copy number in cryo tomograms, and tagging them for subsequent sub-tomogram averaging. This technology bears the potential to become a mainstay in cryo-CLEM and in cellulo cryo-ET.

Protein structures determine their specific function and the working mechanisms on the molecular level. X-ray crystallography and cryo-EM allow atomic level insights into proteins with static conformations. Actual dynamics can only be addressed by solution state techniques, and an extreme example for dynamic protein systems are intrinsically disordered proteins. Among these techniques, single molecule fluorescence spectroscopy allows real-time observation of protein dynamics in solution. Specific long-range distances between residues can be measured using single molecule Förster Resonance Energy Transfer (smFRET). The choice of the label—ideally bright, photostable, soluble and small—and its covalent attachment to the protein are critical. In addition, the background in the green spectral regime requests the use of red-emitting dyes, especially for samples with high background, such as the interior of the cell or liquid-liquid phase separating conditions. Unfortunately, red-emitting dyes are often less bright and more sticky. We describe deuterated and sulfonated fluorophores in the red and NIR spectrum with improved quantum yields, reduced bleaching rates and higher solubility. Their improved characteristics warrant undiscovered potential to replace state-of-the-art fluorophores used in the field smFRET.

The endo-lysosomal system is a complex network of membrane-bound organelles responsible for intracellular trafficking of molecules and maintaining proteostasis through degradation. ESCRT (Endosomal Sorting Complexes Required for Transport) proteins are essential components of the endo-lysosomal pathway involved in sorting membrane cargos and key players in the formation of multivesicular bodies. Despite their widespread role in regulating membrane protein levels through degradation by the endo-lysosomal pathway in non-neuronal cells, their role in neurons and synaptic proteostasis remains poorly understood. This study leverages the ESCRT-0 protein Hrs, known for its role in recognizing and clustering ubiquitinated proteins for degradation, as a tool to unravel the mechanisms of synaptic protein degradation in neurons. By utilizing RNAi to selectively deplete Hrs in rat hippocampal neurons, we integrate molecular biology techniques and advanced microscopy to investigate the role of the ESCRT-dependent pathway in synaptic proteostasis. Additionally, we employ a novel metabolic protein labeling approach known as THRONCAT, along with Stable Isotope Labeling by Amino Acids in Cell Culture (SILAC) coupled with Mass Spectrometry, to elucidate the synaptic protein turnover and the substrate specificity of lysosomal and autophagosomal pathways in synapses.

Connector Enhancer of Kinase Suppressor of Ras-2 (CNKSR2), also known as CNK2 or membrane-associated guanylate kinase-interacting protein-1 (MAGUIN), is a multidomain scaffold protein that is predominantly expressed in the brain, specifically localized at the postsynaptic density (PSD) of excitatory synapses. Our overarching aim is to understand the functional role of the CNK2 scaffold molecule at the synapses. We validated novel interaction partners of CNK2 through co-immunoprecipitation and discovered an interaction between CNK2 and other synaptic scaffold molecules localized at the postsynaptic density.

Further, based on our preliminary unpublished data, we know that the disease-associated truncation mutation, CNKR712*, which is truncated at the C-terminal region, is associated with altered spine morphology. We utilized this mutant in a comparative mass spectrometry (MS)-based approach to map disease-mediated changes in protein-protein interactions to study the potential of the C-terminal region to regulate the scaffolding function of the protein, possibly through the modulation of specific interactions with other proteins. We identified a differential interaction between six different isoforms of the 14-3-3 family with the disease-associated mutant. In the future, we will explore this differential interaction in depth to contribute to the understanding of how the alterations in CNK2 might cause developmental defects.

Neurons largely lack Golgi apparatuses in their processes, yet the cellular machineries facilitating the secretion of locally translated proteins remain elusive. In non-neuronal cells, recent discoveries implicate autophagic vesicles – typically degradative organelles – in the trafficking of cargoes to the plasma membrane (PM). However, whether secretory autophagy occurs within neurons under basal conditions remains unknown. Here, we report for the first time that secretory autophagic vesicles (sAVs), marked by the co-expression of LC3 and the vesicular SNARE SEC22B, are found in neuronal processes. By combining their immunopurification from brain with proteomic profiling experiments, we reveal an enrichment for soluble and integral proteins in sAVs, including many proteins key for synaptic function. Combined with cell biology studies performed in vitro and in vivo, we confirm sAVs as key contributors to Type-III Golgi-independent secretion within neurons. Taken together, our results highlight a novel, non-canonical role for autophagosomes in shaping the synaptic surfaceome, revealing an additional layer of autophagic activity in the governance of neuronal proteostasis and function.

Hippocampal CA1 pyramidal neurons receive excitatory inputs from both ipsi- and contralateral CA3 and CA2 regions. While synaptic plasticity at the intrahippocampal Schaffer collateral synapses has been extensively described, synaptic plasticity mechanisms at interhippocampal

commissural fiber synapses are less well understood. In this project, we investigate cyclic adenosine monophosphate (cAMP)-dependent plasticity at commissural synapses in the CA1 region. We performed in vivo stereotactic injections of AAVs to express ChrimsonR unilaterally in CA3/CA2 glutamatergic neurons of one hippocampus. ChrimsonR, a red-light sensitive Channelrhodopsin, allows us to trigger transmitter release with short orange light flashes

selectively at ipsi- or contralateral synapses in acute brain slices. We then induced cAMP production either pharmacologically by application of forskolin, or optically by blue light-stimulation of the light-activated adenylyl cyclase bPAC, which we coexpressed with ChrimsonR to achieve specific presynaptic manipulation of cAMP. Both the pharmacological stimulation of endogenous adenylyl cyclases with forskolin and the activation of bPAC enhanced synaptic

transmission at commissural fiber synapses. Therefore, we propose that commissural synapses in CA1 express a presynaptic, cyclic adenosine monophosphate (cAMP)-dependent form of plasticity. It remains to be investigated from which specific cell population or network these cAMP-sensitive synapses originate, and further research is needed to uncover the intricate mechanisms and functions underlying this phenomenon.

Introduction: The endoplasmic reticulum (ER), as a major intracellular Ca2+ store, contributes to the amplification and propagation of Ca2+ signals through the channels 'inositol trisphosphate receptor' (IP3R) and 'ryanodine receptor' (RyRs). Both are expressed as three isoforms (1-3) in mammals, being IP3R1/RyR2/RyR3 the most functionally relevant and widely expressed isoforms in the rodent hippocampal neurons. Dendritic spines, as the main postsynaptic site of excitatory synapses, can rely on the ER’s amplification of locally generated Ca2+ transients to propagate synaptic activity. Interestingly, RyR3 is enriched in dendritic spines of hippocampal neurons, while RyR2 is not present, and IP3R1’s presence is debated. Different refilling mechanisms have been also associated to RyRs- and IP3Rs- Ca2+ reservoirs, suggesting a specialization of these channels in their function and subcellular localization. Methods: Here, we use 8-fold Expansion Microscopy and Airyscan fluorescence imaging to resolve the ER of mice hippocampal neurons, and we studied IP3R1/RyR2/RyR3 distribution. We immunostained cultures for IP3R1/RyR2/RyR3 along with the synaptic marker PSD95, the vesicle priming protein Munc13-1, and the membrane probe mCLING. Results: This methodology allowed us to resolve individual ER tubules/cisterns in the soma of neurons, with IP3R1/RyR2/RyR3 punctae signals adhered to them. Cluster size measurements showed differences between the three channels. Furthermore, we corroborated the presence of IP3R1 in dendritic spines, along with a strong signal distributed with a pattern that resembled nanocolumns proteins. Colocalization analysis of this IP3R1 with PSD95 and Munc13-1 confirmed a pre-synaptic localization. RyR3’s enrichment in dendritic spines showed no specific preference on its subcellular localization. Discussion: These results contribute to the characterization of this complex and fined-tuned ER network that acts as an intracellular Ca2+ store. More research is needed to further understand how this specialized distribution of ER-resident Ca2+ channels may be altered in diseases that have a dysregulation of Ca2+ homeostasis.

Neurons communicate with each other at synaptic contact sites, where a pool of docked neurotransmitter-filled synaptic vesicles (SVs) fuse in response to an action potential at specialized presynaptic active zones (AZ). Our work focuses on the hippocampal mossy fibre (MF) synapse, formed between dentate gyrus granule cells and CA3 pyramidal cells. The complexity of this synapse manifests in its large presynaptic bouton’s size, which contains many individual AZs, a very low initial synaptic strength, and a particularly pronounced short-term facilitation (STF)1, with several mechanisms postulated to contribute to MF STF in recent years2. However, ultrastructural information linking these activity-dependent processes to spatially defined SV subpopulations is required to fully interpret short-term plasticity processes. Light stimulation-coupled high-pressure freezing (“flash-and-freeze”) combined with electron microscopy (EM) are powerful methods for probing ultrastructure-function relationships at synapses by revealing the structural organization of synapses at subcellular resolution, both at rest and during defined activity states3,4.

Our previous studies using these methodologies showed that hippocampal MF synapses harbour two distinct morphological vesicle pools in the proximity of AZs at rest (docked and tethered SVs5) and that long, high-frequency stimulation protocols designed to induce steady-state depression caused a partial depletion of docked SVs3. How these two SV pools are recruited during different activity patterns and how they contribute to shaping MF STF is currently not understood. The goal of the present study is to analyse the dynamic organisation and remodelling of AZ-proximal SV pools during STF at the hippocampal MF synapse in intact circuits using shorter stimulation trains, and to dissect the underlying molecular mechanisms. Our experimental approach includes a deletion mutant of the Ca2+ sensor Synaptotagmin-7, recently identified as a major contributor to MF STF6, to isolate aspects of activity-dependent SV pool remodelling most relevant to the facilitated functional state.

Brain-derived neurotrophic factor (BDNF) plays a critical role in postnatal development by modulating the architecture of specific neuronal populations and brain areas. However, the precise molecular program controlling this differential responsiveness to BDNF is still unclear. In the present study, we describe that this program is governed by the restricted expression of the mitogen- and stress-activated protein kinase-1 (MSK1) in GABAergic neurons. Also, we show that while Msk1 expression declines in cortical interneurons along early postnatal development, its expression in striatal neurons increases until adulthood. Utilizing a novel MSK1 loss-of-function mouse model, we reveal its essential role in postnatal growth of the striatum, as it interacts with and modulates the BDNF-dependent phosphorylation of the methyl-CpG binding protein-2 (MeCP2). Furthermore, these mutant mice exhibit an altered transcription pattern of genes involved in the control of the dopamine and GABAergic signalling pathways. Consequently, MSK1 knockout mice behaviour is markedly altered, showing social dysfunction, altered anxiety- and depressive-like responses unequally manifested in males and females. These results elucidate how disruptions in the BDNF/MSK1 pathway impact GABAergic neurite outgrowth and contribute to behaviours reminiscent of schizophrenia in humans.

Neuronal function is intricately tied to the dynamic regulation of the synaptic local proteome through post-transcriptional modifications, trafficking and local synthesis in both excitatory and inhibitory synapses. In turn a proper control of the excitatory versus inhibitory connection balance (E/I balance) in neural networks is essential to cognitive functions. Indeed, dysregulation the E/I balance has been reported in a large number of neuropathologies such as epilepsy, autism spectrum disorders and neurodegenerative disorders.

In Alzheimer’s disease (AD), synaptic dysfunction is concomitant to or shortly follows an excessive accumulation of amyloid-β (Aβ), generated from amyloid precursor protein (APP) -tau pathology being observed only subsequently. Here, we present evidence that APP accumulates preferentially at excitatory presynapses in physiological conditions in cultured and acute brain tissue. Thus, we propose that alterations in APP processing/Aβ generation at excitatory presynapses is pivotal in the early stages of AD.

In our experimental paradigm, we pharmacologically interfere with APP processing in primary neuronal cultures to examine the impact of APP metabolites accumulation on presynaptic function and E/I balance. We use a large range of techniques such as metabolic labeling, live-cell microscopy, immunochemical analysis, and ELISA assays. Our data reveal distinct alterations in neuronal network activity driven by specific E/I imbalance induced by different APP metabolites. We observed detrimental effects of specific metabolites on local protein synthesis within axons and presynapse, affecting bouton stability and function. These observations also support disruptions in axoplasmic trafficking of sub-cellular organelles, compromising overall neuron function. Taken together, we highlight an APP metabolite-specific effect on E/I presynaptic (dys)regulation, neuronal (dys)homeostasis and network (im)balance, observed in early stages of AD pathogenesis.

Munc13s are presynaptic active-zone proteins that are necessary for the maintenance of a pool of primed and fusion-competent synaptic vesicles (SVs) at CNS synapses. The SV priming process is rate-limiting for replenishing the readily-releasable SV pool following synaptic activity, thereby influencing the synapses' ability to reliably transmit presynaptic action potential firing across various frequencies and durations. SV priming is regulated via multiple intracellular signaling pathways, many of which converge onto Munc13s. To study the mechanism of how the C1 domain of Munc13-1 modulates short term plasticity and SV priming at a CNS synapse in situ, we made use of a mouse line carrying a C1 domain point mutation in the dominant Munc13 paralog Munc13-1 (histidine replaced with lysine at AA position 567; H567K). Homozygous Munc13-1H567K/H567K mice cause perinatal lethality in mice, while heterozygous Munc13-1wt/- mice show normal synaptic function. The use of a brainstem-specific conditional Munc13-1 knock-out mouse line, thus, enabled us to compare synapses carrying either a single Munc13-1 H567K mutant allele (Munc13-1H567K/-) or a single Munc13-1 wt allele (Munc13-1wt/-) and to characterize functional consequences at the calyx of Held synapse using patch clamp techniques.

Cells employ various pathways, including the mammalian target of rapamycin (mTOR) signaling axis, to monitor nutrient availability and adapt to metabolic stress, such as amino acid (AA) deprivation. While adaptations in non-neuronal cells involve autophagy activation via mTOR complex 1 (mTORC1) inhibition, understanding adaptation in neurons upon starvation remains limited. Neurons, with their complex morphology, encounter spatial constraints in responding to metabolic stress. Notably, AA sensing and energy regulation may vary between neuronal compartments. However, the impact of metabolic stress on neuronal physiology in these compartments and the role of autophagy in this process remain elusive.

Here, we delve into the molecular and cellular mechanisms underlying neuronal susceptibility to AA withdrawal and investigate how synapses and neuronal soma respond to metabolic stress under conditions of autophagy deficiency, using wild-type and ATG5 knockout cortical neurons. Contrary to non-neuronal cells, AA withdrawal results in limited mTORC1 inhibition, as monitored by phosphorylation of its substrate S6. This phenotype is accompanied by the selective accumulation of damaged mitochondria at synapses and rapid neuronal cell death, which can be rescued by inhibiting mTORC1 activity with rapamycin and/or additional serum withdrawal. Additionally, we find that AA withdrawal does not induce synaptic autophagy, instead, autophagy flux at synapses is stalled. Interestingly, loss of autophagy does not exacerbate AA withdrawal-induced cell death but surprisingly rescues ATP rundown, likely via metabolic rewiring to β-oxidation. This metabolic rewiring has consequences for neuronal physiology, as activity-dependent Ca2+ signaling differs across synapses and soma upon autophagy deficiency, with AA withdrawal inducing a decrease in neuronal activity in synapses but not in soma in WT neurons—a phenotype reversed by autophagy inhibition.

Taken together, these findings shed light on the intricate interplay between mTOR signaling, autophagy, and neuronal metabolism in response to metabolic stress, highlighting the importance of compartment-specific responses in neuronal survival and function.

Major depressive disorder is a leading cause of disability worldwide that affects about 5% of adults. Depression is believed to be the consequence of abnormal activity in key structures supporting mood and reward, namely cortico-meso-limbic structures. Current treatments, that is serotonin-based pharmacotherapies and psychotherapies, have a delayed onset of action and still up to one-third of patients are resistant. Thus, the recent discovery that a subanesthetic dose of ketamine (KET), a non-competitive N-methyl-D-aspartate glutamate receptor (NMDAR) antagonist, induce a rapid-acting and sustained antidepressant effect has risen new hopes for the treatment of depression. Although intensely investigated, the mechanisms through which ketamine acts on NMDAR signaling within the cortico-meso-limbic network to produce its antidepressant effect are still unclear. Based on previous results, we hypothesized that ketamine elicits changes in NMDAR synaptic distribution allowing rearrangements in the functional connectivity of cortico-meso-limbic structures, thereby alleviating the symptoms of depression.

Synaptotagmin-1 (Syt1) plays a crucial role not only in triggering fast neurotransmitter release but also in the docking and priming process of synaptic vesicles (SVs) at the presynaptic active zone. Another isoform, Synaptotagmin-7 (Syt7) shares some functions with Syt1, but the specific mechanisms behind Syt7’s role are not well understood. To explore the potential redundant functions of Syt1 and Syt7 in SV docking and priming, we performed a variety of experiments. This included high-pressure freezing and electron microscopy on hippocampal cultures, and electrophysiological recordings on excitatory hippocampal autaptic neurons from Syt1/7 double knockout and Syt7 knockout mice. Our findings revealed that, similar to Syt1, Syt7 supports the docking of SVs. Additionally, both Syt1 and Syt7 contribute to the priming of synaptic vesicles and the suppression of spontaneous release in a redundant manner, consistent with previous research. To investigate the structural function of Syt7’s C2 domains in priming/docking of SVs and in exocytosis we performed lentiviral overexpression experiments of Syt7 mutants carrying charge-neutralizing and membrane-binding mutations located at the top-loops of its C2A and C2B domains. We could show that Syt7 distinctly requires these domains to exert those functions. The results indicate that SV priming and clamping of spontaneous release are lost upon mutating the putative Ca2+-binding site of the C2A but not the C2B domain. Intriguingly, our investigation of Ca2+-triggered release uncovered that Syt7 acts as negative regulator of release probability, with this regulatory function being mediated specifically through the C2A domain of the protein, providing a mechanism of its proposed role as regulator of presynaptic short-term plasticity. These findings contribute to a better understanding of the distinct functions of Syt7 compared to Syt1 and its regulatory role in neurotransmitter release.

The functional capabilities of our brain, such as memory capacity and processing speed, progressively decline over the course of a lifetime. Concurrently, age becomes the main risk factor for our increased susceptibility to dementia and neurodegenerative diseases, with Alzheimer’s disease being the most common. Thousands of studies conducted over the past decades have focused on neurons, since cognitive decline during aging was thought to be associated with neuronal loss. Nonetheless, more recent and advanced stereological studies suggest that neuronal degeneration is minimal. Rather, subtle and region-specific changes in protein abundance and protein-protein interactions (PPIs) that ultimately alter network dynamics are observed as we age. Here, we present a suite of novel tools and methodologies based on crosslinking mass-spectrometry (XL-MS) combined with biochemical and computational approaches to reveal the age-related changes in the synaptic interactome. We aim to discover novel possible targets that might be used as therapeutic approaches to attenuate age-related neuronal dysfunction and cognitive decline in the elderly.

The early stages of dementia involve synapse loss and Tau phosphorylation. Hibernating animals, such as hamsters, exhibit similar patterns; when they enter torpor, they experience increased tau phosphorylation and synapse loss (>50%). However, while synaptic loss is progressive in dementia, it is completely reversible in hibernating animals. The underlying mechanisms of this phenomenon remain elusive.

This project aims to unravel the pathways responsible for synapse loss and the subsequent recovery in hibernating hamsters. As proof-of-concept, we have purified and cross-linked mouse-derived synaptosomes. We propose a quantitative cross-linking mass spectrometry (XL-MS) pipeline to investigate altered protein interactions and abundance during hibernation cycles in hamster with the final goal to find novel potential therapeutic targets to alleviate synapse loss in human tauopathies.

Glutamate, the main excitatory amino acid neurotransmitter plays a crucial role in neuronal physiology. Glutamatergic neurotansmission disturbance can result from primary de novo mutations of GRIN genes, encoding for the N-methyl-D-Aspartate receptor (NMDAR) subunits. These rare autosomic dominant conditions cause GRIN-related disorders (GRD, also called Grinpathies), a group of severe developmental encephalopathies. GRD display a clinical spectrum including intellectual disability, hypotonia, ASD traits, motor impairment, epilepsy, and gastro-intestinal distress, in a gene- and residue-dependent manners. Accordingly, as for other channelopathies, the functional annotation of GRIN de novo variants is critical i) to understand GRD pathophysiology, ii) to evaluate potential therapeutic strategies and iii) to define personalised therapeutic approaches. In order to address these issues we have created a multi-angled GRIN cluster initiative, merging computational, experimental, translational, and clinical neuroscience approaches. Bioinformatic analysis was used to build-up a comprehensive and specific GRIN variants database compiling genetic, structural, functional and clinical annotations. This database allowed to define a superimposition structural algorithm drastically increasing GRIN variants annotations with a high predictive likelihood ultimately accelerating GRIN variants functional annotations. Further, an experimental pipeline has been developed for the annotation of GRIN-orphan variants and their functional stratification. Finally, we evaluated and experimentally demonstrated the potential therapeutic benefit of nutraceutical interventions for the rescue of LoF GRIN variants, both in preclinical cellular and animal models, in proof-of-concept GRD cases and currently in the first reported GRD clinical trial. Beyond GRD personalised therapies, our findings open the avenue for for future treatments of genetic and/or environmental conditions perturbing the glutamatergic synapse.

GRIN-related neurodevelopmental disorders (GRDs) are a group of rare genetic diseases associated with intellectual disability, epilepsy, movement disorders, development delay, autism spectrum disorder, and schizophrenia. The primary cause of GRD is the presence of de novo GRIN mutations that result in the presence of dysfunctional GluN subunits of the N-methyl D-Aspartate receptor (NMDAR), which plays a pivotal role in neuronal development, synaptic plasticity, and neuron survival(3). GRIN variants identification and reporting are exponentially growing and there is an urgent need to classify these variants as disease-causing or neutral and further to functionally stratify them as Loss-of-function (LoF), Gain-of-function (GoF), or Complex. The stratification is essential to define a strategic therapy (i.e. L-serine for LoF and memantine for GoF). However, their clinical, genetic, and functional annotations remain highly fragmented or there is no previous information on the variant, representing a bottleneck in GRD patient stratification. To shorten the gap between GRIN variant identification and stratification we have constructed GRIN Database, a platform containing all available clinical, genetic, and functional information on all previously reported GRIN variants. Furthermore, we have developed and experimentally validated GRIN algorithm, a structural algorithm that allows to computationally predict the pathogenicity and functional annotations of unannotated GRIN variants, resulting in the duplication of pathogenic GRIN variants assignment, reduction by 30% of GRIN variants with uncertain significance, and increase by 70% of functionally annotated GRIN variants. GRIN algorithm was implemented into GRIN variants Database (http://lmc.uab.es/grindb), providing together a computational resource and tool that accelerates GRIN missense variants stratification, contributing to clinical therapeutic decisions for this neurodevelopmental disorder.

Mossy fiber boutons are highly plastic synapses in the middle of the hippocampal circuitry and therefore play an important role in learning and memory formation. They are able to react to a wide range of input frequencies with different forms of presynaptic plasticity. These include short- and long-term plasticity, which are thought to underlie distinct forms of learning. Despite the relevance, the precise mechanisms of presynaptic plasticity at this synapse remain to be elucidated.

Synapsins are presynaptic proteins that cluster synaptic vesicles. Therefore, they play a crucial role in neurotransmission and plasticity throughout the nervous system. Three synapsin genes exist, but synapsin III expression is usually downregulated early in development. However, in hippocampal mossy fiber boutons synapsin III expression persists in adulthood. Previous research on synapsins in mossy fiber boutons focused on synapsin I and II. Thus, a complete picture regarding the role of synapsin in mossy fiber plasticity is still missing.

Here, we investigated presynaptic plasticity at hippocampal mossy fiber boutons in a mouse model lacking all synapsin isoforms. In a combined approach, we investigated both synaptic function and ultrastructure, using electrophysiological field recordings and transmission electron microscopy. We found decreased short-term plasticity - i.e. decreased facilitation and post-tetanic potentiation - but increased long-term potentiation in male synapsin triple knockout mice. At the ultrastructural level, we observed a higher density of active zones in mossy fiber boutons from knockout animals. When chemically potentiated with forskolin, the SynTKO boutons also showed an increase in the density of docked synaptic vesicles. Our results indicate that all synapsin isoforms are required for the fine regulation of short- and long-term presynaptic plasticity at the hippocampal mossy fiber synapse.

Synaptic transmission, the process of information transfer between neurons, occurs at synapses and is controlled by the coordinated activity of different protein complexes. Unraveling the mechanisms of synaptic transmission and the functional differences between synapse subtypes, therefore, requires knowledge of protein composition and organization. I use proximity labeling, an unbiased, mass-spectrometric-based approach, to elucidate the proteomic composition and organization of molecular complexes at presynaptic subcompartments. Our protocol produces highly specific readouts of presynaptic cytosolic and membranous compartments. I propose that monitoring the molecular composition and organization of the presynapse will help in understanding synaptic transmission mechanisms and how synaptic function and dysfunction are defined.

Munc13-1 is a highly-conserved and essential protein for synaptic transmission, acting as a regulator of synaptic strength and plasticity. A pivotal role for Munc13-1 in human neurological disorders is emerging. Patients carrying variants in the protein-coding sequence of the UNC13A gene show a neurodevelopmental disorder characterized by profound developmental delay, intellectual disability and dyskinesia/intention tremor. In addition, deep intronic single nucleotide polymorphisms in non-coding regions of the UNC13A gene have been repeatedly identified as strong risk factors for amyotrophic lateral sclerosis and frontotemporal dementia (ALS and FTD). Recent findings highlight a mechanism whereby in neurons with ALS/FTD pathology, Munc13-1 levels are gradually reduced due to mis-splicing events, but how lower UNC13 levels eventually contribute to ALS/FTD pathology remains unknown. Here, we study the consequences of gradual Munc13-1 removal from hippocampal mouse neurons for neurotransmission and neuronal survival, using a novel conditional knock-out mouse line and human motor neurons. These efforts will aid in determining how Munc13-1 levels control different properties of synaptic transmission and in¬ the development of approaches to stabilize UNC13A mRNA expression levels, that are currently sought for as a therapeutic approach for ALS/FTD.

Neurons transmit chemical signals through specialized structures called synapses, which comprise presynaptic and postsynaptic compartments. A functional presynapse requires an extensive machinery for neurotransmission and the majority of its components (e.g. synaptic vesicles, active zone proteins) need to be synthesized in the neuronal soma. Due to the large distance between the soma and its presynapses, the timely and stoichiometrical delivery of the presynaptic protein machinery is a unique challenge.

In our previous work we analyzed actively transported axonal Synaptophysin-positive vesicles in iPS-derived human neurons by first recruiting them to mitochondria. Using correlative light and electron microscopy (CLEM), we found that the majority of these vesicles were devoid of a dense core and had a size of at least 50-100 nm distinct from synaptic vesicles (Rizalar et al., (2023)).

The two potential drawbacks of this approach were the overexpression of Synaptophysin and the capture of Synaptophysin-containing vesicles moving anterogradely and retrogradely. To selectively analyze anterogradely moving precursor vesicles carrying synaptic vesicle proteins, we labeled endogenous Synaptotagmin1 with HaloTag by using Crispr/Cas9-based genome editing in induced pluripotent human stem cells (iPSCs). The versatility of the HaloTag system will allow us to selectively target newly synthesized Synaptotagmin1 prior to its incorporation into mature synaptic vesicles. By combining a pulse-chase approach in induced human neuron KI-lines with CLEM we aim to specifically identify precursor vesicles by correlative light and electron microscopy (live CLEM). Additionally, the Halo-Tag will enable the biochemical isolation of precursor vesicles by utilizing affinity ligands for subsequent mass spectrometry.

We predict the combination of endogenous tagging of newly synthesized synaptic vesicle proteins (i.e. Synaptotagmin1-Halo) in genome-engineered iPS with live CLEM and the investigation of the precursor vesicle proteome to yield new insights into the mechanisms of presynapse assembly in excitatory human neurons.