KR Robbins1, A Bredbenner1, RA Osbaldeston2, KS Villafañe2, EE Shin2, E Merkulova1, A Clevenger1, PB Delean2, C Campos2, GC Peet2, RA Jain1,2

Visualizing active neurons and circuits in vivo is critical for investigating the neural activity that underlies behavior. While several established methodologies are available to achieve this end in larval zebrafish, they are limited by the scale of tissue visualization, temporal resolution, need to restrain larvae, and/or accessibility of necessary instruments. Here, we establish a pipeline for the visualization and quantification of spatiotemporally precise whole-brain neural activity in larval zebrafish using CaMPARI2, a genetically encoded photoconvertible calcium indicator. Using temporally specific photoconverting UV light exposures, we capture whole-brain “snapshots” of neural activity time-locked to stimuli during unrestrained larval behavior. We optimized experimental conditions for establishing sub-second neuronal activity changes across acoustically-evoked behavioral paradigms spanning minutes to hours. We then leveraged this system to pinpoint brain-wide neural activity changes during nonassociative habituation learning, observing distinct activity signatures in the subpallium, preoptic area, and habenulae that are altered through pharmacological and/or genetic disruption of habituation learning. This approach effectively complements the temporal precision achievable through post hoc activity detection methods and expands the accessibility of large-scale behavioral circuit dissection beyond highly specialized real-time volumetric imaging equipment.

1Bi-College Interdisciplinary Neuroscience Program, Haverford College, Haverford PA, USA

2Department of Biology, Haverford College, Haverford PA, USA

Funding Support: NIH R15EY031539

Peyton Presto¹, Julian Cardenas¹, Christian Bustamante¹, Brent Kisby^1,2, Guangchen Ji^1,2, Olga Ponomareva¹, Volker Neugebauer^1,2,3*, Igor Ponomarev^1,2*

Neuropathic pain is a chronic pain condition that results from damage or dysfunction in the nervous system. While mechanisms of neuropathic pain at the peripheral and spinal cord level have been extensively studied, pain mechanisms in the brain remain underexplored. The amygdala, a limbic brain region, has emerged as a critical brain area for the emotional-affective dimension of pain and pain modulation. Amygdala neuroplasticity has been associated with pain states, but exact molecular and cellular mechanisms underlying these states and the transition from acute to chronic pain are not well understood. Here, we used the spinal nerve ligation model of neuropathic pain in male rats to investigate changes in gene expression in two amygdala nuclei, basolateral (BLA) and central (CeA) at the chronic pain stage using RNA sequencing. We used an integrative approach that focuses on functional significance and cell type specificity of differentially expressed genes to nominate mechanistic targets for central regulation of chronic pain. Our integrative transcriptomic and bioinformatic analyses identified individual genes (e.g., Cxcl10, Cxcl12, Mbp, Plp1, Mag, Mog, Slc17a6, Gad1, Sst), molecular pathways (e.g., cytokine-mediated signaling pathway), biological processes (e.g., myelination, synaptic transmission), and specific cell types (e.g., oligodendrocytes, glutamatergic and GABAergic neurons) affected by chronic pain. Our results also provide evidence for hemispheric lateralization of pain processing in the amygdala. Overall, our study proposes oligodendrocyte dysfunction in the amygdala, neuroimmune signaling in the CeA, and glutamatergic neurotransmission in the BLA as mechanistic determinants of and potential therapeutic targets for the management of chronic neuropathic pain.

1. Department of Pharmacology and Neuroscience, Texas Tech University Health Sciences Center, 3601 4^th Street, Lubbock, Texas 79430, U.S.A. 2. Center of Excellence for Translational Neuroscience and Therapeutics, Texas Tech University Health Sciences Center, 3601 4^th Street, Lubbock, Texas 79430, U.S.A. 3. Garrison Institute on Aging, Texas Tech University Health Sciences Center, 3601 4^th Street, Lubbock, Texas 79430, U.S.A.

Funding Support: National Institutes of Health grants R01 NS038261 to V.N. and I.P. and R01 AA027096 to I.P.

Emma Gnatowski^1,2, Jessikah Buys^1,2, Jensen Goulette^2,3, Andrew A. George¹ and Michael F. Miles^1,2

Acute ethanol reduces anxiety in humans and animal models. Anxiety disorders increase risk for Alcohol Use Disorder (AUD) and human subjects report that stress and anxiety increase ethanol consumption. The Miles laboratory previously identified the microtubule binding protein Ninein (Nin) as a candidate gene underlying ethanol’s acute anxiolytic-like properties in BXD recombinant inbred mice. Here we report on behavioral, gene expression and GABAergic function consequences of Nin deletion in central amygdala (CeA). Deletion of Nin in CeA was done using stereotactic injections of AAV8-hSyn-GFP (control) or AAV8-hSyn-CRE-GFP (deletion) virus in Ninfl/fl mice. CeA Nin deletion increased acute ethanol anxiolysis in the light-dark box assay in male and female mice and reduced intermittent access 2-bottle choice ethanol consumption and preference x 5 weeks in female but not male mice. There were no changes in ethanol sedation (loss-of-righting reflex) or pharmacokinetics. Taste preference for quinine or saccharin were also unaffected. Bulk RNAseq analysis of stereotactic injection sites in CeA revealed striking evidence of neuroinflammatory and GABAergic gene expression alterations in Nin deletion mice. Preliminary electrophysiological studies on CeA IPSP activity measured by voltage clamp analysis showed Nin deletion altered IPSC duration, suggesting a post-synaptic site of action. Conclusions: These studies document that Nin function in CeA modulates the acute anxiolytic and consumption properties of ethanol, with the latter showing a striking sex preference. Initial mechanistic studies suggest that disruption of Nin expression in CeA produces changes in post-synaptic GABA receptor function, with coincident gene expression changes consistent with altered GABAergic neuron homeostasis and possible synaptic remodeling.

¹Dept. of Pharmacology and Toxicology, ²VCU Alcohol Research Center, and ³Dept. of Human and Molecular Genetics, Virginia Commonwealth University, Richmond, VA USA

Funding Support: NIAAA grants F31AA030727, P50AA022537, and R01AA027581.

RC Rice,¹ MN Wauhop,² GE Homanics,^3,4 SP Farris^2,5

The long noncoding RNA growth arrest specific 5 (Gas5) is differentially methylated in blood and brain of individuals with alcohol use disorder and has multiple proposed functions, including immune- and glucocorticoid signaling modulation. In mouse, we previously observed a male-specific increase in nondependent voluntary ethanol consumption following Gas5 knockdown (KD) in medial prefrontal cortex (mPFC) and persistent downregulation of mPFC Gas5 following chronic intermittent ethanol vapor (CIEV). We hypothesize mPFC Gas5 modulates CIEV-induced escalation of ethanol consumption and stress-related behaviors during ethanol abstinence. To test this, we performed mPFC-specific Gas5 KD in adult male mice (C57BL/6J background), exposed them to CIEV-2BC, then tested stress-related behaviors and peak serum CORT levels during abstinence. mPFC Gas5 KD did not alter voluntary ethanol consumption, but modulated CIEV effects on stress-related behaviors during abstinence. In the open field test, Gas5 KD in air controls mimicked the stress-like phenotype of CIEV, measured by decreased entries and time spent in the inner zone, whereas Gas5 KD attenuated this phenotype in CIEV mice. In the elevated plus maze, CIEV resulted in decreased stretch-attend postures, which was ablated in Gas5 KD CIEV mice, suggesting Gas5 KD reversed CIEV impairment on risk assessment behavior. Peak serum CORT increased in CIEV mice seven days into abstinence, with no effect of Gas5 KD. However, 21 days into abstinence, CIEV mice with Gas5 KD displayed a persistent increase in serum CORT, while their CIEV control counterparts displayed levels resembling air controls. These findings suggest Gas5 modulates stress-related phenotypes of ethanol abstinence.

¹University of Pittsburgh, Center for Neuroscience, Pittsburgh, PA, 15261, USA

²University of Pittsburgh, Department of Anesthesiology & Perioperative Medicine, Pittsburgh, PA, 15261, USA

³University of Pittsburgh, Department of Pharma cology & Chemical Biology, Pittsburgh, PA, 15261, USA

⁴University of Pittsburgh, Department of Neurobiology, Pittsburgh, PA, 15261, USA

⁵University of Pittsburgh, Department of Biomedical Informatics, Pittsburgh, PA, 15261, USA

Funding Support: NIAAA AA031168 (PI: Rice), U01 AA020889 (PI: Farris, MPI: Homanics), and R01 AA024836 (PI: Farris).

Markos Michail Chatzigiannis1,2, Hirotaka Shoji2 , Daiki Sato2,3,4 , Keizo Takao, Tsuyoshi Miyakawa2

Behavioral phenotyping across genetically modified mouse strains is extensive but lacks a coherent framework for cross strain comparison. We assembled a large scale dataset comprising more than 10,000 mice from 167 strains across 15 behavioral assays. Multifactor analysis identified two principal dimensions, locomotor activity and learning/memory, that captured the dominant components of cross strain covariance. Clustering along these axes defined six behavioral phenotypes reflecting systematic variation in activity and cognitive performance. To assess clinical relevance, each strain was assigned a disorder association score derived independently of mouse behavioral data from publicly available human gene–disease association resources. Scores were calculated for intellectual disability (ID), autism spectrum disorder (ASD), schizophrenia, and major depressive disorder. Disorder association differed across endotypes, with the strongest and most consistent enrichment observed for ID and ASD. Strains with high ID or ASD burden were concentrated in the same two profiles characterized by comparable learning impairments but opposite locomotor patterns: one predominantly hypoactive and the other hyperactive. Across disorders, specific behavioral indices showed selective correlation with disorder burden, identifying the most informative measures for distinguishing disorder relevant models. These results indicate that clinically distinct diagnostic categories share underlying behavioral structure in mouse models that is not captured by disorder titles alone. This framework enables the interpretation of large scale behavioral data and the evaluation of disorder relevance for genetically modified mice.

1. Department of Systems Medical Science, Fujita Health University Graduate School of Medicine, Kutsukake-cho, Toyoake, Japan 2. Division of Systems Medical Science, Center for Medical Science, Fujita Health University, Kutsukake-cho, Toyoake, Japan 3. Institute for Advanced Academic Research, Chiba University, Chiba, Japan 4. Graduate School of Science, Chiba University, Chiba, Japan

Naureen Hameed¹, Sergio L. Crespo Flores¹, Evan Cirone¹, Chenyue Zhao¹, Annika F. Barber^1,2,*

Central pacemaker neurons use a combination of external stimuli and neuropeptide signaling to synchronize molecular oscillations leading to circadian behaviors. The clock network structure and signaling between these pacemaker neuron groups have been well described, but how these pacemakers communicate with specific brain output regions remains poorly understood. Here, we identified how “core” clock neurons in Drosophila, the ventrolateral neurons (LNvs), signal to the proto-hypothalamic region, the pars intercerebralis (PI). Previously thought to communicate with the PI only indirectly, we provide evidence to show that LNvs functionally modulate, the PI’s insulin-producing cells (IPCs) in a time-of-day-dependent manner. This functional connectivity relies on neuropeptidergic signaling of two canonical clock neuropeptides: pigment dispersing factor (PDF) and short Neuropeptide F (sNPF). Loss of either receptor alone in PI subpopulations does not alter feeding or locomotor rhythmicity. Further, we provide insight into how these two neuropeptides may be acting together via their receptors to signal to IPCs. We identify sexually dimorphic responses of IPC response to LNv stimulation, which may be partially explained by sex differences in proximity of clock neurons to the PI. Our findings indicate that LNvs form both direct peptidergic signaling but also form indirect multisynaptic circuits with IPCs, which may model more broadly how they communicate with various other clock output regions.

¹ Waksman Institute of Microbiology, Rutgers, The State University of New Jersey, Piscataway, NJ, 08854, USA.

² Department of Molecular Biology and Biochemistry, Rutgers, The State University of New Jersey, Piscataway, NJ, 08854, USA.

* annika.barber@waksman.rutgers.edu

GE Homanics¹, SJ Sukoff Rizzo², AC Silva², PL Strick², GW Carter³, JE Park²

Mutations in the presenilin 1 (PSEN1) gene are the most common cause of familial, early-onset Alzheimer's disease (AD), yet rodent models fail to fully recapitulate human AD pathology because they do not naturally develop amyloid plaques or tau aggregates. The common marmoset (Callithrix jacchus) offers a compelling alternative. It is a small non-human primate whose brain closely resembles the human brain, and it develops spontaneous age-related amyloid and tau pathology. We used CRISPR/Cas9 gene editing to independently introduce two PSEN1 point mutations (C410Y and A426P) — the same single-nucleotide changes found in human patients — into marmoset embryos. Several genetically engineered founders were produced; however, most died prematurely. One C410Y founder survived to adulthood and sired germline offspring. Phenotypic outcomes observed in these animals will be the subject of companion talks. In addition, experiments are underway to develop marmoset models of sporadic, late-onset AD. These nonhuman primate models are unprecedented for studying the earliest molecular events that initiate AD, evaluating preventive interventions, and bridging the rodent-to-human translational gap.

Departments of ¹Anesthesiology and ²Neurobiology, University of Pittsburgh, Pittsburgh PA, 15261. ³The Jackson Laboratory, Bar Harbor, ME 04609.

Funding Support: NIA U19 AG074866 and UPMC-ITTC grant IPA 2019 No. 16.

Lauren Bailey, PhD¹, Takeshi Murai, PhD¹, Lauren Mongeau¹, Abbey Setlik¹, Tingting Zhang, PhD¹, Seung-Kwon Ha, PhD, DVM¹, Gregory W. Carter², Afonso C Silva, PhD¹ and Stacey J Sukoff Rizzo, PhD¹,

Fundamental questions remain regarding the mechanisms that initiate Alzheimer’s disease (AD), drive its progression, and link pathology to cognitive impairment. As part of our MARMO-AD consortium, we established a comprehensive testing battery sensitive to detecting age-dependent cognitive decline across the lifespan in our colony of aging marmosets, and marmosets genetically engineered with mutations in the PSEN1 gene which confers early onset AD in humans. Beginning in adolescence, marmosets are trained using touchscreens through a battery of tests that captures a spectrum of cognitive domains including spatial working memory (delayed match to position), behavioral flexibility and reversal learning (delayed non-match to position), recognition memory (trial unique delayed match to sample), attention (serial reaction time task), and episodic-like memory (paired associative learning task). Behavioral and cognitive function are aligned with longitudinal PET neuroimaging and blood-based biomarkers to track AD progression. Similar to human PSEN1 mutation carriers, plasma Aβ42:40 is significantly elevated relative to non-carrier controls Despite robust biomarker and pathological changes, PSEN1 marmosets show no significant deficits in task acquisition or performance across cognitive domains up to 4 years of age. These data are not surprising and recapitulate the disease trajectory of increased amyloid in plasma and brain years before cognitive decline. These ongoing longitudinal studies are enabling the identification of the molecular and cellular mechanisms other than amyloid that contribute to and precede cognitive decline associated with Alzheimer’s disease progression.

(1)University of Pittsburgh School of Medicine, Pittsburgh, PA, USA (2) The Jackson Laboratory, Bar Harbor, Maine, USA

This work is supported by U19AG074866 and 5T32AG021885-19.

Thais Rafael Guimarães¹, Jung Eun Park¹, Catrina Spruce², Stephanie Hachem¹, Swati Banerjee¹, Lauren K Hayrynen Schaeffer¹, Gregg E Homanics¹, Stacey J Sukoff Rizzo¹, Gregory W Carter,², Afonso C Silva¹, and Amantha Thathiah¹

Progress in preclinical Alzheimer’s disease (AD) research has been constrained by models that fail to faithfully recapitulate human aging and overt AD neuropathology. The common marmoset (Callithrix jacchus), a New World non-human primate, exhibits aging trajectories, genetic heterogeneity, and complex social behaviors closely resembling those of humans, providing a highly translational platform for age-related neurodegenerative research. Importantly, marmoset studies uniquely enable longitudinal correlation of in vitro cellular models with in vivo assessments across the marmoset lifespan. We performed an integrated ex vivo and in vitro characterization of marmoset AD and tauopathy models. Immunohistochemical analyses of postmortem brains revealed robust amyloid-β (Aβ) and tau pathology, along with the associated cellular pathology. To establish an in vitro cellular system, we adapted a well-established human direct reprogramming protocol to generate age-conserved induced neurons (iNs) from marmoset fibroblasts. Comparative, unbiased RNA-seq analyses of marmoset and human iN conversion trajectories revealed significant species-specific differences, guiding targeted optimization of the reprogramming strategy. The refined protocol achieved high-efficiency neuronal conversion, improved cell survival and maturation, and preserved AD-relevant protein expression, including amyloid precursor protein (APP)/Aβ and tau. Together, this integrated framework establishes the marmoset as a powerful translational model for AD research. This platform enables minimally invasive mechanistic studies, longitudinal analyses, high-throughput drug screening, and therapeutic discovery aimed at accelerating disease-modifying strategies for AD.

¹University of Pittsburgh School of Medicine, Pittsburgh, PA, USA

²The Jackson Laboratory, Bar Harbor, ME, USA

Gregory W. Carter

Laboratory mice and marmosets provide potential models of aging and Alzheimer’s disease with the capacity to track the initiation and progress of pathology across compressed lifespans. We have performed genetic, genomics, and proteomic analyses of plasma and postmortem tissues to map multi-modal aspects of disease development and progression in these model species. We have drawn from human genetic studies to engineer multiple mouse strains carrying candidate genetic factors for late-onset Alzheimer's disease. In the marmoset, we are combining genetic engineering of select loci with outbred genetic variability to analyze biomarkers of aging and dementia. We assembled a new telomere-to-telomere reference genome for the common marmoset (Callithrix jacchus) and performed whole-genome Illumina sequencing on over 230 marmosets. For both model systems, we have used high-coverage proteomics, Alamar NULISA, and transcriptomics to assess the disease-relevant consequences of genetic factors. We have identified a broad range of disease-associated signatures in knock-in mouse and marmoset models, including immune, metabolic, and synaptic alterations. These outcomes frequently correlate with clinically relevant biomarkers and behavioral outcomes. We have also used multi-omic signatures to understand the molecular pharmacodynamics of candidate drugs. Our findings constitute a data and model resource for identifying the appropriate animal model for understanding genetic liability in Alzheimer's and preclinical testing of targeted therapeutics.

The Jackson Laboratory, Bar Harbor, ME, USA

Leslie C. Griffith MD PhD

Neurons are among the most structurally and functionally specialized cells in the body, capable of processing, integrating, and transmitting information. Unlike many other cell types, neurons exhibit extreme morphological polarity, with distinct compartments—dendrites, soma, axon, and synaptic terminals—each requiring specifically tailored protein populations to support their localized functions. The recognition that cells are capable of locally synthesizing proteins marked a major shift in our thinking about cellular organization. Local translation enables neurons to establish functionally distinct subdomains and it allows rapid, site-specific responses to activity, supporting structural changes that are required for long-term plasticity. Localized protein synthesis and compartment-specific protein turnover allow neurons to dynamically respond to activity and environmental changes.

I will talk today about the first plasticity-related protein to be shown to be locally synthesized in neurons: Ca²⁺/calmodulin-dependent protein kinase II (CaMKII). Activity-dependent CaMKII synthesis is conserved across phyla and occurs in both pre- and postsynaptic compartments. Activity also has an additional effect on CaMKII that is equally conserved: it causes a subcellular redistribution of the protein. How the dynamic regulation of CaMKII levels and subcellular localization are related is not understood, and I will discuss recent results from my lab that address these questions. My talk will highlight how the unique biology of neurons depends on highly specialized protein landscapes—proving that in neuroscience, just like real estate, success ultimately comes down to location, location, location.

Nancy Lurie Marks Professor of Neuroscience, Brandeis University