Valence, a hedonic value assigned to sensory stimuli that determines subsequent behaviors, is a fundamental organizing principle of affective states and emotional behaviors. Valence influences physiological processes across multiple organ systems — effects that may be mediated in part by the serotonergic (5-HT) system, which itself has widespread distribution throughout the body. The brain serotonergic system projects to the amygdala and its extended structure, the bed nucleus of the stria terminalis (BNST), regions heavily implicated in valence encoding. Despite their anatomical and functional connectivity, the amygdala and BNST express distinct 5-HT receptor subtypes, receive heterogeneous 5-HT projections, and process valence differently. The current study examines the significance of 5-HT projections in the computation of valence in the CeA and BNST. Here, multi-site in vivo electrophysiology was combined with anterograde chemogenetic modulation of 5-HT terminals in either the BNST or the CeA, while recording from both regions during valence presentation tasks where mice were exposed to the odorants of saline (neutral), opposite-sex urine (positive), and trimethylthiazoline (TMT, negative valence) in a counterbalanced manner. Neural responses were then compared across the neutral, appetitive, and aversive stimuli. Our results showed dorsal raphe 5-HT activation in the BNST supported valence coding in the CeA and BNST during stimulus presentation. During the post-stimulus phase, this modulation engaged only CeA-supported valence coding. Conversely, the inhibition of 5-HT terminals in the BNST supported local valence coding. Dorsal raphe 5-HT activation in the CeA enhanced valence coding in both the CeA and BNST only after stimulus offset. However, its suppression in the CeA enhanced valence coding in both the CeA and BNST during and after stimulus presentation. Taken together, our results showed that dorsal raphe serotonergic modulation creates a phasic response pattern in the CeA/BNST circuit by moderating response timing relative to stimulus valence.
Glioblastoma (GBM), a high-grade glioma, is the most common primary brain tumor in adults, with a dismal prognosis and a very high recurrence rate. Major challenges to effective treatment arise from the tumor’s locally aggressive, diffusely infiltrative nature and extensive intratumoral heterogeneity. Recent advances in cancer neuroscience have uncovered bidirectional neuron-glioma interactions including the formation of functional neuron-glioma synapses (NGS). Neuronal activity drives tumor growth and its synaptic integration into neural circuits; tumor-induced circuit remodeling reciprocally influences neural activity. In parallel, emerging evidence implicates the immunological synapse- a dynamic interface between T cells and antigen-presenting cells analogous to its neural counterpart- as a key driver of GBM pathogenesis and progression. The tumor microenvironment (TME) in GBM is characterized by local immunosuppression, immune evasion, and resistance to immunotherapy. Notably, neural and immune synapses share fundamental architectural and functional principles–spatial confinement of signaling, receptor clustering and adhesion-mediated stabilization, directional information transfer, and use-dependent plasticity–and an expanding catalog of molecular hubs operates across both interfaces. In this context, glioblastoma can be reconceptualized as a disease of coupled neural and immune synaptic networks cooperating to drive tumor progression. Therapeutic strategies targeting these hubs may offer advantages over current approaches by simultaneously disrupting tumor-promoting neuronal input and enhancing antitumor immunity, providing a “double hit” with the potential to dismantle the network architecture sustaining tumor growth and resilience.
Appropriate transmission of sensory afferent input from the peripheral nervous system to spinal locomotor circuits is crucial for efficient and adaptive locomotion. This is partly mediated by spinal interneurons (INs) that regulate afferent input relayed to locomotor networks. After a spinal cord injury (SCI), sensory afferent feedback becomes critical for motor recovery, however these pathways undergo plasticity. Previously, we observed a loss in the inhibitory sensory afferent pathway to locomotor rhythm generating Shox2 INs after SCI. Importantly, the inhibitory INs within sensory afferent pathways to Shox2 INs have yet to be defined. Thus, the goal of the present study was to identify inhibitory INs interposed between sensory afferents and Shox2 INs in the uninjured mouse. Using electrophysiology and pharmacology to test for connectivity, we demonstrate that lumbar inhibitory INs within sensory afferent pathways to Shox2 INs are glycinergic (GlyT2) INs in the medial deep dorsal horn. Afferent stimulation evoked currents in medial deep dorsal GlyT2 INs that were mostly from excitatory, putative monosynaptic connections. Consistent with the electrophysiology, parvalbumin+ afferent fibers densely innervated the region containing the medial deep dorsal GlyT2 INs. Additionally, the medial deep dorsal GlyT2 INs expressed multiple previously described deep dorsal horn molecular markers. These results identify an inhibitory sensory afferent pathway to Shox2 INs comprised of a molecularly diverse population of GlyT2 INs that may be important for the regulation of low-threshold afferent input to the spinal locomotor network.
Interoceptive signals reflecting internal bodily states influence emotional processing. The paraventricular thalamus (PVT) constitutes a key node within the interoceptive network, monitoring circulating humoral factors in the ventricles, and is thus considered to link interoceptive signals to emotion-related behaviors. Here, we demonstrate that the vagus nerve (VN), a principal afferent pathway for interoceptive signaling, also affects PVT neuronal activity in mice. Left vagus nerve stimulation (VNS) induced pronounced c-Fos expression in PVT neurons and elicited transient increases and decreases in neuronal spike rates across all PVT subregions. These VNS-responsive neuronal populations were distinct from those exhibiting transient-to-sustained delayed responses following lipopolysaccharide injection after vagotomy, which selectively activates VN-independent pathways, including humoral factors. VNS-responsive PVT neurons projected to limbic structures and modulated anxiety-like behavior. Taken together, these findings demonstrate that PVT neuronal populations integrate and relay interoceptive signals across multiple pathways and timescales, thereby playing a pivotal role in linking vagus nerve-related bodily states to emotional responses.
Sterile alpha and TIR motif-containing protein 1 (SARM1) is an enzyme that cleaves nicotinamide adenine dinucleotide (NAD+) and plays a role in disrupting neural circuits through axon degeneration and cell death. SARM1 is activated by changes in the nicotinamide mononucleotide (NMN)/NAD+ ratio and by various post-translational modifications, but its complete regulatory mechanism remains poorly understood. Here, we report that tripartite motif-containing protein 32 (TRIM32) activates SARM1 through specific ubiquitination triggered by anticancer drug treatment. TRIM32 promotes the attachment of Lys27-linked ubiquitin chains to Lys173 and Lys375 within the ARM domain, which is an autoinhibitory region of SARM1. This ubiquitination by TRIM32 increases SARM1’s NAD+-cleaving catalytic activity and enhances its ability to induce neurite degeneration and cell death. A mutant form of TRIM32 lacking the enzymatically active RING domain fails to promote SARM1 ubiquitination, and reducing TRIM32 levels decreases SARM1 ubiquitination. Additionally, ubiquitin-specific peptidase 13 (USP13), a known negative regulator of SARM1, can deubiquitinate SARM1. These findings suggest that TRIM32 is a key regulator of axon degeneration and cell death through its ubiquitination of SARM1.
