Molecular characteristics and laminar distribution of prefrontal neurons projecting to the mesolimbic system

Distinct prefrontal projection activity and transcriptional state conversely orchestrate social competition and hierarchy

Social animals compete for limited resources, resulting in a social hierarchy. Although different neuronal subpopulations in the medial prefrontal cortex (mPFC), which has been mechanistically implicated in social dominance behavior, encode distinct social competition behaviors, their identities and associated molecular underpinnings have not yet been identified. In this study, we found that mPFC neurons projecting to the nucleus accumbens (mPFC-NAc) encode social winning behavior, whereas mPFC neurons projecting to the ventral tegmental area (mPFC-VTA) encode social losing behavior. High-throughput single-cell transcriptomic analysis and projection-specific genetic manipulation revealed that the expression level of POU domain, class 3, transcription factor 1 (Pou3f1) in mPFC-VTA neurons controls social hierarchy. Optogenetic activation of mPFC-VTA neurons increases Pou3f1 expression and lowers social rank. Together, these data demonstrate that discrete activity and gene expression in separate mPFC projections oppositely orchestrate social competition and hierarchy.

Drug-induced change in transmitter identity is a shared mechanism generating cognitive deficits

Cognitive deficits are a long-lasting consequence of drug use, yet the convergent mechanism by which classes of drugs with different pharmacological properties cause similar deficits is unclear. We find that both phencyclidine and methamphetamine, despite differing in their targets in the brain, cause the same glutamatergic neurons in the medial prefrontal cortex to gain a GABAergic phenotype and decrease their expression of the vesicular glutamate transporter. Suppressing the drug-induced gain of GABA with RNA-interference prevents the appearance of memory deficits. Stimulation of dopaminergic neurons in the ventral tegmental area is necessary and sufficient to produce this gain of GABA. Drug-induced prefrontal hyperactivity drives this change in transmitter identity. Returning prefrontal activity to baseline, chemogenetically or with clozapine, reverses the change in transmitter phenotype and rescues the associated memory deficits. The results reveal a shared and reversible mechanism that regulates the appearance of cognitive deficits upon exposure to different drugs.

The PFC-LH-VTA pathway contributes to social deficits in IRSp53-mutant mice

Dopamine (DA) neurons in the ventral tegmental area (VTA) promote social brain functions by releasing DA onto nucleus accumbens neurons, but it remains unclear how VTA neurons communicate with cortical neurons. Here, we report that the medial prefrontal cortex (mPFC)-lateral hypothalamus (LH)-VTA pathway contributes to social deficits in mice with IRSp53 deletion restricted to cortical excitatory neurons (Emx1-Cre;Irsp53fl/fl mice). LH-projecting mutant mPFC neurons display abnormally increased excitability involving decreased potassium channel gene expression, leading to excessive excitatory synaptic input to LH-GABA neurons. A circuit-specific IRSp53 deletion in LH-projecting mPFC neurons also increases neuronal excitability and induces social deficits. LH-GABA neurons with excessive mPFC excitatory synaptic input show a compensatory decrease in excitability, weakening the inhibitory LHGABA-VTAGABA pathway and subsequently over-activating VTA-GABA neurons and over-inhibiting VTA-DA neurons. Accordingly, optogenetic activation of the LHGABA-VTAGABA pathway improves social deficits in Emx1-Cre;Irsp53fl/fl mice. Therefore, the mPFC-LHGABA-VTAGABA-VTADA pathway contributes to the social deficits in Emx1-Cre;Irsp53fl/fl mice.

Self-supervised deep clustering of single-cell RNA-seq data to hierarchically detect rare cell populations

Single-cell RNA sequencing (scRNA-seq) is a widely used technique for characterizing individual cells and studying gene expression at the single-cell level. Clustering plays a vital role in grouping similar cells together for various downstream analyses. However, the high sparsity and dimensionality of large scRNA-seq data pose challenges to clustering performance. Although several deep learning-based clustering algorithms have been proposed, most existing clustering methods have limitations in capturing the precise distribution types of the data or fully utilizing the relationships between cells, leaving a considerable scope for improving the clustering performance, particularly in detecting rare cell populations from large scRNA-seq data. We introduce DeepScena, a novel single-cell hierarchical clustering tool that fully incorporates nonlinear dimension reduction, negative binomial-based convolutional autoencoder for data fitting, and a self-supervision model for cell similarity enhancement. In comprehensive evaluation using multiple large-scale scRNA-seq datasets, DeepScena consistently outperformed seven popular clustering tools in terms of accuracy. Notably, DeepScena exhibits high proficiency in identifying rare cell populations within large datasets that contain large numbers of clusters. When applied to scRNA-seq data of multiple myeloma cells, DeepScena successfully identified not only previously labeled large cell types but also subpopulations in CD14 monocytes, T cells and natural killer cells, respectively.