Integrated Bioinformatics Analysis Investigating the Potential Role and Mechanism of INS in Cocaine-Induced Stroke

Cocaine abuse is associated with an increased risk of stroke, yet the underlying molecular mechanisms remain poorly understood.

An integrated framework combining network toxicology, machine learning, Mendelian randomization (MR), single-cell RNA sequencing (scRNA-seq), virtual knockout, molecular docking, and molecular dynamics (MD) simulations was applied to identify and validate key molecular links between cocaine exposure and stroke.

A total of 319 shared targets were identified, from which three core genes (TNF, INS, CDC42) were screened. MR analysis demonstrated a potential causal association between INS and stroke. scRNA-seq showed high expression of Ins2 (murine homolog of human INS) in epithelial cells, with extensive intercellular communication between epithelial and endothelial cells. Additionally, 196 genes exhibited significant changes following virtual knockout of Ins2, which were enriched in neurogenic and metabolic pathways. Molecular docking and MD simulations confirmed stable binding between cocaine and INS.

Mechanistically, cocaine-induced stroke is mediated by a mutually reinforcing pathological network forming a "mitochondrial damage-oxidative stress-inflammation-coagulation" vicious cycle, wherein cocaine crosses the blood-brain barrier to disrupt cerebral vasculature function, activate platelets, and trigger proinflammatory responses. INS, identified as a candidate gene via MR analysis (overcoming confounding biases of observational studies), is specifically highly expressed in choroid plexus epithelial cells (CPECs) and forms a choroid plexus-insulin signaling axis that regulates cerebral energy metabolism, oxidative stress, inflammation, and neurovascular homeostasis via cerebrospinal fluid. Virtual knockout of Ins2 perturbed pathways linked to energy metabolism disorder and neural repair, confirming its pivotal role in maintaining cerebral homeostasis, while stable cocaine-INS binding suggests cocaine may interfere with this protective axis.

INS may serve as a potential key regulatory factor linking cocaine exposure with stroke risk. This study provides novel mechanistic insights and a systematic analytical framework for investigating drug-induced cerebrovascular diseases.