Circuit-Based Biomarkers for Mood and Anxiety Disorders

Abstract

Mood and anxiety disorders are complex heterogeneous syndromes that manifest in dysfunctions across multiple brain regions, cell types, and circuits. Biomarkers using brain-wide activity patterns in humans have proven useful in distinguishing between disorder subtypes and identifying effective treatments. In order to improve biomarker identification, it is crucial to understand the basic circuitry underpinning brain-wide activity patterns. Leveraging a large repertoire of techniques, animal studies have examined roles of specific cell types and circuits in driving maladaptive behavior. Recent advances in multiregion recording techniques, data-driven analysis approaches, and machine-learning-based behavioral analysis tools can further push the boundary of animal studies and bridge the gap with human studies, to assess how brain-wide activity patterns encode and drive emotional behavior. Together, these efforts will allow identifying more precise biomarkers to enhance diagnosis and treatment.

Keywords: biomarker; circuit manipulation; data-driven analysis; multiregion recording; reverse translation.

Figure 1.. Simplified schematic of mood and anxiety networks from recent rodent studies.

Mood- and anxiety-related behavior and emotional states are mediated by local and long-range interactions across regions including medial prefrontal cortex (mPFC), orbitofrontal cortex (OFC), insular cortex (IC), ventral hippocampus (vHPC), amygdala (AMY), nucleus accumbens (NAc), bed nucleus of the stria terminalis (BNST), ventral pallidum (VP), lateral septum (LS), lateral hypothalamic area (LHA), paraventricular thalamic nucleus (PVT), intralaminar thalamus (ILT), dorsal raphe nucleus (DRN), ventral tegmental area (VTA), parabrachial nucleus (PB), and laterodorsal tegmentum (LDTg).

Figure 2.. Advances in techniques applied in rodent research for multi-region recordings or neural circuit mapping.

A. high-density extracellular electrophysiological recording (e.g., Neuropixels probes), B. multi-region calcium imaging (e.g., 1-photon, 2-photon, mesoscope, fibre photometry), C. functional magnetic resonance imaging, D. high-throughput anatomical tracing (e.g., BARseq), and E. high-resolution behavioral tracking and analysis (e.g., DeepLabCut). These approaches, either in isolation or by combined two or more of them, allow researchers to interrogate how multi-region neural activity patterns drive emotional behavior.

Figure 3.. Towards better identification and understanding of circuit-based biomarkers for mood and anxiety.

Moving beyond focal circuit interrogation and binary classifications of behavior, one can leverage recent advances in novel techniques to perform multi-region recordings, develop novel behavioral paradigms, and apply data-driven analyses to identify circuit-based biomarkers that differentiate between distinct emotional states.