Back

Disrupted ErbB4 splicing with region-specific severity across the cortical visuospatial working memory network in schizophrenia

Chung, Y.; Bazmi, H. H.; Lewis, D. A.; Chung, D. W.

2025-04-25 psychiatry and clinical psychology
10.1101/2025.04.23.25326229 medRxiv
Show abstract

BackgroundVisuospatial working memory (vsWM) depends on gamma oscillations generated in multiple areas of a cortical network, including dorsolateral prefrontal (DLPFC), posterior parietal (PPC), visual association (V2), and primary visual (V1) cortices. Gamma oscillations require parvalbumin-expressing interneurons (PVI), and deficient gamma power in the vsWM network of schizophrenia (SZ) is thought to arise from lower PVI activity. We previously proposed that PVI activity, and activity-dependent PV levels, are regulated by shifts in splicing of erb-b2 receptor tyrosine kinase 4 (ErbB4) transcripts between major variants that enhance PVI activity and inactive minor variants. Here, we investigated the region-specific pattern of this splicing shift across the vsWM network and its alterations in SZ. MethodsLevels of ErbB4 splice variants and PV mRNA were quantified from 16 pairs of unaffected comparison (UC) and SZ subjects across DLPFC, PPC, V2, and V1. ResultsIn UC, the major ErbB4 variants showed progressive enrichment relative to minor variants from the rostral to caudal regions. In SZ, this gradient was disrupted by abnormal shifts toward minor variants, with the magnitude of shifts greater in caudal regions. These major-to-minor splicing shifts correlated with lower PV mRNA levels across all regions in SZ. ConclusionGreater enrichment of major variants in caudal areas suggests spatially regulated molecular mechanism supporting region-specific levels of PVI activity across the vsWM network. In SZ, the region-specific magnitude of splicing shift to minor variants may drive reduced PVI activity throughout the vsWM network, leading to a network-wide dysfunction that contributes to cognitive impairments.

Published in Cerebral Cortex (predicted rank #15) · training set

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.