Back

A 3D Endothelium-on-a-Chip Model Uncovers Cx43-Dependent Glycolytic Shift and Endothelial Barrier Dysfunction in Diabetic Kidney Disease

Lialios, P.; Kim, Y.; Trewhella, I.; Myakala, K.; Hughes, E.; Wang, X.; Levi, M.; Alimperti, S.

2025-05-23 bioengineering
10.1101/2025.05.19.654627 bioRxiv
Show abstract

Endothelial dysfunction is a key pathological feature of diabetic kidney disease (DKD), characterized by increased vascular leakiness and altered metabolic signaling. In this study, we investigated how diabetic conditions affect endothelial barrier integrity and identified the molecular mechanisms contributing to this dysfunction. Using a 3D microfluidic model that recapitulates in vivo vascular architecture and flow, we demonstrated that high glucose (HG) and high fat (HF) conditions significantly impair endothelial barrier function, as demonstrated by increased dextran permeability and loss of VE-cadherin from the cellular membrane site. Metabolomic profiling and functional assays revealed a shift toward glycolysis, marked by elevated lactate levels and upregulation of lactate dehydrogenase A (LDHA), which contributed to barrier disruption. Pharmacological inhibition of LDHA effectively restored barrier function, underscoring the pathogenic role of glycolytic reprogramming. Transcriptomic analyses of mouse and human DKD datasets further identified connexin 43 (Cx43) as a candidate mediator of this dysfunction. Cx43 expression was progressively deregulated in diabetic mouse kidneys and across multiple cell types in human DKD samples. In vitro, Cx43 overexpression in endothelial cells enhanced glycolytic flux, suppressed oxidative metabolism, disrupted VE-cadherin localization, and promoted angiogenic sprouting. Collectively, our findings establish a mechanistic link between Cx43-driven metabolic reprogramming and endothelial barrier dysfunction in DKD, highlighting Cx43 as a potential therapeutic target for preserving vascular integrity in diabetic conditions.

Matching journals

The top 5 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.