11/11 Connectome-based fingerprinting of alcohol drinking in mice: mechanisms and biomarkers

Universite De Strasbourg, Strasbourg

Investigators

Linked publications & trials

Abstract

SUMMARY In the current INIA period, we have developed multimodal brain MRI in mice, i.e. combining morphological, structural and functional data. Our data demonstrate circuitry-specific alterations of brain morphology, and reshaping of both structural and functional connectomes after 6 weeks excessive alcohol drinking. Our findings are consistent with the human literature and demonstrate our ability to characterize in-depth whole brain plasticity in a mouse model of alcohol drinking. We will build on these findings and methodologies to achieve three goals: Specific Aim 1. Longitudinal MRI biomarkers - Group level analysis: do alcohol effects on neurocircuitry depend on drinking history, and/or sex? We will continue to use the EOD-2BC model and C57BL/6 mice and incorporate female mice in longitudinal Experiment 1. We will acquire multimodal MRI images at four time points (baseline; 6 and 10 weeks drinking; 4 weeks withdrawal) and apply quantitative multiparametric analyses to investigate temporal and sex effects. Brains will be collected for histology, c-Fos analysis, metabolomics and transcriptomics (to be used in Aim 3). Specific Aim 2. Connectome/behavior individual signatures – Subject level analysis: can we classify and predict behavior and drug effects from MRI? We will move from group analysis to subject-level analysis and correlate multimodal MRI data with individual behavioral patterns using machine learning approaches. This is now feasible (GPR88 knockout mice/EOD-TBC,preliminary results). We will: Aim 2a. Identify connectome features that classify and predict alcohol drinking/withdrawal behavior. We will build a predictive multivariate model of the drinking trajectory at individual level, using longitudinal imaging data from Experiment 1. We will determine best MR-based connectome classifiers that discriminate drinking patterns and evaluate their respective causal influence on the negative affect of withdrawal. Aim 2b. Identify connectome features that predict drug treatment efficacy. We will study alcohol effects in male mice in longitudinal Experiment 2, which will include a drug treatment. We have selected a GPR-88 agonist, which reduces alcohol intake. We will expand the prediction model from Aim 2a to drug effects and include c-fos endpoints in the prediction (Kash/INIA-S). Specific Aim 3. Mechanisms: can we link connectome data with neuroimmune signaling? This aim will be synergistic with INIA partners, using notably molecular data from the same mice (see Aim 1). We will: Aim 3a. Develop hypothesis-driven analyses to identify MRI correlates for neuroadaptations in extracellular matrix (Lasek), microglia (Mangieri) and stress centers (AI, BNST, Kash). Aim 3b. Perform metabolomic profiling using HR-MAS (high resolution magic angle spinning) 1H-NMR at our site, for regions of interest based on our collaborative work. We will compare metabolic profiles of immune pathways with those obtained from mice undergoing the CIE-protocol (Kroenke-Cuzon/INIA-S). Aim 3c. Integrate transcriptomic brain mapping (Mayfield) with our MRI data to correlate altered gene expression and brain network remodeling in the EOD-TBC model. In conclusion, Specific Aim 1 will strengthen validity of our MRI-based fingerprinting approach in mice, address the yet poorly understood sex effect at whole brain level, and will be applicable to other mouse models of AUD. Specific Aim 2 will establish signatures of alcohol effects and drug treatments, with potential for diagnostic and patient stratification, confirm GPR88 as a target to treat AUD, and will be applicable to other INIA drugs. Specific Aim 3 will bridge our “mesoscale” data with molecular and cellular mapping INIA projects.