Dopamine transmission in the anterior insula shapes the neural coding of anxiety

Abstract The anterior insula1–3 and dopamine neuromodulation4–11 both play key roles in the control of anxiety, yet how dopamine shapes anterior insula function to regulate anxiety remains unknown. Here we show that dopaminergic neurons of the ventral tegmental area preferentially target the anterior relative to the posterior insula, and that optogenetic activation of these neurons elicits dopamine signals in the anterior insula. Behaviorally, dopamine signals increased in the anterior insula during risk assessment and exploration of exposed spaces. Interestingly, neurons expressing the type-1 dopamine receptor (D1) are enriched in the anterior insula subdivision, where their optogenetic activation is anxiogenic. At the molecular level, direct D1 activation or blockade in the anterior insula bidirectionally controls anxiety, demonstrating a causal anxiogenic function of D1 in the anterior insula. Remarkably, systemic D1 activation also increased anxiety-related behaviors, together with a cellular activation of the anterior insula, and a disruption of neural coding in this region. As an example of the latter, systemic D1 activation oppositely regulated the coding reliability of exposed and protected areas, increasing the reliability of the neural code for exposed spaces. Together, our findings reveal an anterior insula D1-dependent mechanism by which dopamine can control anxiety, providing a framework for investigating dopamine dysregulation in models of anxiety disorders. Our study also introduces quantitative metrics applied to AI-based computational representations of neural activity to identify how neuromodulation reshapes neural coding of behaviors and contexts. Competing Interest Statement The authors have declared no competing interest. Footnotes Summary of revision experiments and additional analyses: 1. Cortical layer analysis of dopaminergic axonal terminals in the insular cortex, revealing preferential VTA dopaminergic innervation to the anterior relative to posterior insula. 2. Removal of valence-related experiments to improve clarity and impact of the manuscript. 3. Functional validation of dopamine release: optogenetic stimulation of VTA dopaminergic neurons elicited increased dopamine release in the anterior insula. 4. Fiber-photometry recordings and analysis of dopamine signal in the insula: 4.1. Quantification of dopamine signals as z-scored dF/F instead of percent change across all recordings. 4.2. Analysis of dopamine signal transients (frequency and amplitude) in the EPM. 4.3. Time-locked dopamine dynamics aligned to entries into EPM closed and open arms. 4.4. Behavioral impact of systemic vehicle or SKF38393 administration in the open field test, confirming the anxiogenic properties of D1 receptors across two anxiety assays. 5. EPM testing following intra-anterior insula D1 blockade (SCH23390), further supporting the anxiogenic role of D1 receptors of the anterior insula. 6. cFOS+ cell density quantification following vehicle or SKF38393 injection, revealing increased cFOS+ recruitment in the anterior insula following SKF38393 injection. 7. Bilateral optogenetic activation of anterior insula D1 neurons in the EPM is anxiogenic. 8. Expanded in vivo electrophysiology analyses: 8.1. Additional SVM decoding analyses from PCA-reduced anterior insula activity. 8.2. Comparison of classifiers (SVM vs. kNN) for CEBRA-based decoding. 8.3. CEBRA-Time analyses with behavioral overlays (entries into open and closed arms, head dips).