Animal model tissues and biofluids are indispensable yet underused assets in preclinical research. We argue that standardized, shareable biobanks could reduce duplication, improve technical reproducibility, strengthen biomarker translation, lower animal use, and create scalable infrastructures for academia and industry.

Psychotic symptoms, including hallucinations, delusions, and disorganized thought, occur across psychiatric and neurological disorders, yet their underlying mechanisms remain poorly integrated across scales. Here, we propose a transdiagnostic framework in which disturbances in dopamine, glutamate, GABA, serotonin, and acetylcholine alter neuronal gain and synaptic efficacy, disrupting the precision weighting of sensory evidence, prediction errors, and internally generated representations. These local disturbances propagate through cortico-striato-thalamo-cortical circuits and large-scale functional networks, producing abnormal integration-segregation dynamics and unstable attractor states. This multilevel model links molecular and cellular perturbations to circuit dysfunction, network instability, and symptom expression while allowing distinct biological pathways to converge on shared computational vulnerabilities. Emerging neuroimaging, dynamical systems, and artificial intelligence approaches may enable individualized tests of these mechanisms and support more precise, circuit-informed interventions for psychotic symptoms.

The lateral hypothalamic area (LHA) contains γ-aminobutyric acid (GABAergic) and glutamatergic neurons that converge on midbrain dopamine circuits, exerting opposing control over feeding and reinforcement. How these neurons behave during consummatory behavior and how they shape dopamine release across the striatum remains unresolved. Here, we show that LHA GABAergic and glutamatergic neurons independently scale their activity as mice consume rewarding and aversive solutions and that their balance sets a spatially organized dopamine landscape along the striatum's anterior-posterior axis: anterior regions track solution value and recent history, posterior regions encode sensorimotor features, and release at each site is controlled locally rather than propagated across subregions. This landscape tracks LHA activity, depends on it causally, and reinforces licking initiation. These results position the LHA as a central regulator of a striatum-wide dopamine landscape during consumption, providing a circuit-level framework for how the hypothalamus shapes motivated behavior.

The persistence or loss of memory depends on the stability of synaptic connections. Microglia regulate this stability in part by phagocytic elimination of synapses, while the molecular mechanism that sets their intrinsic phagocytic capacity at homeostasis remains unknown. Here, we show that TMEM119, a microglia-specific transmembrane protein, functions as a constitutive cell-intrinsic brake on microglial surveillance dynamics and phagocytosis. Mechanistically, TMEM119 sequesters tankyrase, preventing axin degradation, thereby limiting the pool of free adenomatous polyposis coli protein available to nucleate actin assembly and drive phagocytosis. In adult mice, acute ablation of TMEM119 from microglia exaggerates synapse elimination and destabilizes established fear memories. These effects are recapitulated by a membrane-permeable peptide that enters microglia in the brain after intravenous delivery to competitively disrupt TMEM119-tankyrase association. Our results identify the TMEM119-tankyrase-axin axis as a pharmacologically accessible rheostat of microglial phagocytosis and establish a new framework for memory manipulation.

Profound autism spectrum disorder (ASD) is frequently attributable to single-gene mutations, with SCN2A, encoding the voltage-gated sodium channel NaV1.2, among the most penetrant. Cortico-striatal circuitry is a key node implicated in ASD, yet how SCN2A deficiency alters human neural circuits remains unclear. Using a human cortico-striatal assembloid model, we show that autism-causing heterozygous SCN2A protein-truncating variants (PTVs) impair long-range cortical axonal projections, reduce striatal spine density, and attenuate excitatory cortico-striatal synaptic transmission. Genotype-defined assembloids reveal contributions from both impaired cortical projections and intrinsic striatal vulnerability. Paradoxically, despite these connectivity deficits, SCN2A-deficient neurons exhibit increased intrinsic excitability, and assembloids show elevated spontaneous network firing, suggesting a potentially distinctive feature of human neural models. Notably, canine adenovirus type 2-mediated delivery of human SCN2A rescued cellular and circuit deficits. Collectively, our study unveils human circuit dysfunction caused by SCN2A deficiency and supports the therapeutic potential of targeted gene replacement for SCN2A-related ASD.

Recent evidence in a 16p11.2 deletion mouse model of autism spectrum disorder (ASD) revealed brain endothelial abnormalities postnatally, but the endothelial alterations eliciting these changes remained unknown. Using 14-day-old and adult 16p11.2-deficient and wild-type male mice, we now show that the 16p11.2 deletion causes a bioenergetic failure selectively in endothelial cells (ECs) with reduced intracellular ATP levels. Intra- or extracellular ATP supplementation rescued the function of 16p11.2-deficient ECs in vitro via P2 purinergic receptor activation, specifically P2Y2 receptors. Activating P2Y2 receptors restored cerebrovascular reactivity in 16p11.2-deficient arterioles ex vivo, activity-dependent cerebral blood flow in vivo, and rescued 16p11.2 deletion-associated mouse behaviors. Taken together, this study demonstrates that metabolic reprogramming of brain ECs via purinergic receptor engagement represents a promising therapeutic avenue for ASD.

Integrating high-resolution spatial transcriptomics with metabolomics is essential for linking cellular identity to metabolic state, yet same-section, single-cell alignment is limited by incompatible substrates, chemistries, and imaging. We introduce OpenFISH, an open-source, low-cost, modular spatial transcriptomics platform and a matrix-assisted laser desorption ionization mass spectrometry imaging (MALDI-MSI)-compatible workflow. Through optimized tissue handling and a guided registration pipeline, OpenFISH enables cell-type-aware co-mapping of transcripts and metabolites. We demonstrate its utility by revealing cell-type-specific metabolic heterogeneity in the central nervous system and compartment-level metabolic zonation in the hippocampus. Exploratory 5xFAD experiments identify concordant cell-type-associated metabolic changes, including prominent phospholipid changes in microglia, metabolic shifts in excitatory neurons masked in aggregate analyses, and elevated lysophosphatidylethanolamine-related features in oligodendrocytes. OpenFISH also independently quantifies transposable-element activation after immune challenge and identifies candidate striatal D1-neuron patterning changes in Reeler mutants. This work provides an open, low-cost, and customizable framework for same-section spatial multi-omics.

Non-oscillatory, aperiodic activity, is an emerging signal of interest whose physiological and functional origins are just beginning to be probed for their potential link to systems-level physiology and cognition. But there are several pernicious, practical, and statistical hazards to be aware of when deploying spectral parameterization tools for disentangling oscillations and aperiodic activity over short, noisy time windows. In this primer, we provide a set of recommendations and guidelines for time-resolved and event-related spectral parameterization to help researchers meet these challenges. Specifically, we provide recommendations on how to optimize and assess model fits, how to report analysis decisions to maximize reproducibility, and what to do when spectral parameterization fails to identify the presence of oscillations. These recommendations will help guide the study of the distinct dynamics of periodic and aperiodic neural activity in cognition, behavior, and disease at fine timescales.

The 17q21.31 inversion locus is among the strongest genetic modifiers of tau-related neurodegenerative disease, yet its structure and function across ancestries remain incompletely understood. By integrating long-read sequencing, transcriptomics from postmortem human brain tissue, induced pluripotent stem cell (iPSC)-derived neurons, and astrocytes, we identified H1_A, a previously unrecognized non-inverted H1 sub-haplotype enriched in individuals of African ancestry (AA). H1_A carries rs8070723, a variant widely used to define the protective H2 haplotype, revealing the limitations of current MAPT haplotype assignment in non-European populations. Across brain tissue and neural cell types, ancestry and haplotype influenced MAPT expression, tau isoform composition, extracellular matrix pathways, immune signaling, and oxidative stress responses. Functional studies further demonstrated effects on astrocyte migration and glutathione regulation. These findings uncover ancestry-dependent regulatory programs at 17q21.31 and provide a framework for more accurate interpretation of MAPT-associated risk in neurodegenerative disease.

Meningeal lymphatic vessels drain brain fluid and waste into the cervical lymph nodes, but how removing these nodes during surgery affects brain health is unknown. Retrospective analysis of cancer patients who underwent neck lymphadenectomy without chemo-radiation therapy found that up to 25% developed mild-to-severe cognitive impairment, and neuroimaging showed accelerated temporal horn enlargement after bilateral surgery. To determine mechanistically how cervical lymphadenectomy contributes to cognitive impairment, mouse and rat models were utilized to phenocopy abrupt disruption of lymphatic flow to the cervical and submandibular nodes. Impaired lymphatic drainage induced a cerebrospinal fluid signature of metabolic stress and inflammation. In the hippocampus, redox stress drove oxidative posttranslational protein modifications and microaggregation with biochemical, molecular, and structural features distinct from amyloid aggregates in Alzheimer's models. Such changes were absent when lymphatic drainage was gradually reduced. The data link cognitive impairment following acute neck lymphadenectomy with the biochemical consequences of impaired metabolic waste clearance.

Chemokine receptor CXCR3 mediates T cell recruitment into inflamed tissues, but its role in tauopathies is unclear. Here, we show that hippocampal injection of interferon-γ (IFNγ) acutely triggers CXCL10 upregulation and brain parenchymal T cell infiltration. Genetic deletion or antibody-mediated blockade of CXCR3 prevented IFNγ-induced T cell infiltration. In a mouse model of tauopathy and neurodegeneration, chronic systemic anti-CXCR3 treatment markedly reduced parenchymal CD4[+] and CD8[+] T cell accumulation, attenuated neurodegeneration, and improved aspects of cognition. CXCR3 blockade decreased microglial MHC-II expression without broadly suppressing classical disease-associated inflammatory phenotypes. Single-cell RNA sequencing and flow cytometry further revealed a reduction in the proportion of activated CD4[+] T cell populations and elevated CD8[+] T cell terminal exhaustion in the brain. These findings identify CXCR3-dependent chemotaxis as a critical signaling pathway for subtypes of T cells linked to tau-mediated neurodegeneration and highlight CXCR3 blockade as a potential disease-modifying therapeutic strategy for tauopathies.

The dorsal horn is remarkably diverse, yet how this cellular complexity enables discrimination of sensory modalities remains a fundamental question. Neurons expressing the gastrin-releasing peptide receptor (Grpr[+]) have long been considered dedicated to itch, yet broad activation also evokes pain-related behavior. Whether Grpr[+] neurons are required for pain and how itch- and pain-related functions are organized within this population remain unresolved. Here, we show that Grpr[+] neurons comprise functionally distinct subpopulations defined by tachykinin-1 (Tac1) and correspond to species-conserved transcriptomic subtypes. Convergent loss- and gain-of-function approaches show that the Tac1- subpopulation is required for chemical itch, whereas a bombesin-insensitive, Tac1-enriched population is necessary for mechanical hypersensitivity across diverse injury states. Computationally identified enhancers provide subtype-enriched genetic access, confirming their roles in itch, mechanical hypersensitivity, and sustained pain. The findings resolve functional diversity within the Grpr[+] population, including identification of a convergent node for mechanical hypersensitivity, and provide subtype-enriched tools for further investigation.

Artificial neural networks trained with gradient descent (GD) are widely used to model brain learning, but GD allows synapses to switch between excitatory and inhibitory and produces weight distributions unlike those observed experimentally. We show that an alternative rule, exponentiated gradient (EG), avoids these problems because its multiplicative, exponential weight updates mathematically prohibit synapses from changing sign and preferentially scale already-large weights, naturally generating log-normal distributions. This mechanism also concentrates task-relevant computation onto a smaller subset of strong connections, which we show makes EG-trained recurrent neural networks more robust to synaptic pruning and better able to learn from sparsely relevant inputs than GD-trained networks, while matching GD's performance on standard cognitive tasks. These findings tie a normative optimization principle directly to observed features of synaptic organization in the brain.

Central post-stroke pain (CPSP) represents a failure of homeostatic recovery, yet how acute injury transitions to chronic pain remains elusive. While glial reactivity is a hallmark of stroke, the effectors translating this activation into sustained neuronal hyperexcitability are undefined. Here, we identify microglia-derived secreted phosphoprotein 1 (SPP1) as a key chronification switch for CPSP-like behaviors in mice. Hemorrhage-induced ATP release activates P2X7 receptors, triggering robust SPP1 expression in microglia. This engages a self-amplifying, autocrine integrin β1 (ITGB1) signaling loop that perpetuates a chemokine-rich, pro-nociceptive milieu. Optogenetic activation of this microglial state is sufficient to induce mechanical allodynia and thalamic hyperexcitability without injury. Genetic and pharmacological dissections reveal a critical temporal dissociation: while SPP1 is induced acutely, its signaling via ITGB1 is specifically required to maintain chronic pain. These findings define the SPP1-ITGB1 axis as a fundamental driver of maladaptive sensory plasticity and a promising therapeutic target for refractory central pain.