Glycolysis Promotes Angiotensin II-Induced Aortic Remodeling Through Regulating Endothelial-to-Mesenchymal Transition...

Endothelial dysfunction plays a crucial role in aortic remodeling. Aerobic glycolysis and endothelial-to-mesenchymal transition (EndoMT) have, respectively, been suggested to contribute to endothelial dysfunction in many cardiovascular diseases. Here, we tested the hypothesis that glycolytic reprogramming is critical for EndoMT induction in aortic remodeling through an epigenetic mechanism mediated by a transcriptional corepressor CtBP1 (C-terminal binding protein 1), a sensor of glycolysis-derived NADH.

EndoMT program, aortic remodeling, and endothelial expression of the glycolytic activator PFKFB3 (6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase isoform 3) were evaluated in Ang (angiotensin) II-infused mice. Mice with endothelial-specific Pfkfb3 deficiency or CtBP1 inactivation, immunoprecipitation, chromatin immunoprecipitation, and luciferase reporter assay were employed to elucidate whether and how PFKFB3/CtBP1 epigenetically controls EndoMT.

The EndoMT program and increased endothelial PFKFB3 expression were induced in remodeled thoracic aortas. In TGF-β (transforming growth factor-β)-treated human endothelial cells, activated SMAD2/3 (SMAD Family Member 2/3) transcriptionally upregulated PFKFB3 expression. In turn, the TGF-β/SMAD signaling and EndoMT were compromised by silencing or inhibition of PFKFB3. Mechanistic studies revealed that PFKFB3-mediated glycolysis increased NADH content and activated the NADH-sensitive CtBP1. Through interaction with the transcription repressor E2F4 (E2F Transcription Factor 4), CtBP1 enhanced E2F4-mediated transcriptional repression of SMURF2 (SMAD ubiquitin regulatory factor 2), a negative regulator of TGF-β/SMAD2 signaling. Additionally, EC-specific Pfkfb3 deficiency or CtBP1 inactivation in mice led to attenuated Ang II-induced aortic remodeling.

Our results demonstrate a glycolysis-mediated positive feedback loop of the TGF-β signaling to induce EndoMT and indicate that therapeutically targeting endothelial PFKFB3 or CtBP1 activity could provide a basis for treating EndoMT-linked aortic remodeling.