摘要
Background: Infarct volume is associated with prognosis after ischemic stroke, yet molecular programs underlying differences in infarct severity remain poorly characterized. This study aims to identify cellular interactions and candidate transcriptional regulators associated with stroke severity.
Methods: We reanalyzed rat single-nucleus RNA sequencing data (GSE250245) from sham rats and rats with moderate or severe permanent middle cerebral artery occlusion. CellChat identified severity-associated communication patterns. The selected cell population was examined through subpopulation analysis, metabolic gene-set scoring, scFEA flux inference, and lactylation-related gene-set scoring. Severity-associated differentially expressed genes were integrated with hdWGCNA and MEGENA networks, followed by functional enrichment. KLF6 activity was evaluated with SCENIC and DoRothEA. Virtual perturbations were examined with CellOracle, scTenifoldKnk, and scTenifoldNet. Klf6 expression was assessed in two independent mouse astrocyte datasets.
Results: Astrocyte-to-neuron cell-cell communication showed the greatest increase with severity. The Astro1 astrocyte subset showed higher predicted pyruvate-to-lactate flux and lactate balance, while lactylation-related gene-set scores increased across severity groups. Key co-expression modules associated with stroke severity and lactylation were primarily enriched for cell adhesion, neuronal death, MAPK signaling, and histone modification, among other terms. Integrating differential expression and network results identified KLF6 as a hub transcription factor, and its activity also increased with severity. CellOracle simulations suggested shifts from the severe-associated Astro1 state toward Astro2/3 after virtual Klf6 knockout and shifts toward Astro1 after simulated overexpression. Virtual Klf6 knockout predicted prominent network perturbations involving neuroactive ligand–receptor interactions, cell adhesion, hypoxia, and inflammatory responses, whereas oxidative phosphorylation showed comparatively weaker perturbation in the model. Both external datasets supported increased astrocytic Klf6 expression after stroke.
Conclusions: Infarct severity was associated with changes in astrocyte communication and metabolic transcriptional programs, and KLF6 was identified as a candidate transcription factor. Further experimental studies are needed to determine whether and how KLF6 contributes to these changes.
