• activin
• heart failure
• miR-1306-5p
• mtDNA-cn
• pulmonary hypertension
• SGLT-2 inhibitors
• telomere length

Background: Pulmonary hypertension (PH) often complicates heart failure with reduced ejection fraction (HFrEF) but can hardly be addressed therapeutically, since mechanisms involved in the pathophysiology of HFrEF-related PH (PH-HFrEF) remain unclear. Whether sodium-glucose cotransporter 2 inhibitors (SGLT2i) positively impact PH-HFrEF independent of their diuretic effect, but through what signaling pathways, is unknown.
Methods: In an observational series of seventy-four patients with HFrEF, patients were non-randomly divided into two groups receiving (Baseline T=0) either the SGLT2i dapagliflozin or empagliflozin (both at 10 mg/day) on top of conventional anti-HF therapy (gliflozin group, n=42), or a no-gliflozin anti-HF therapy (no-gliflozin group, n=32). Plasma samples from the gliflozin and the no-gliflozin groups were profiled by untargeted proteomics, microRNA, senescence and oxidative stress markers. The role of identified proteins was then validated and investigated in human plasma samples, in in vitro human cell cultures and in in vivo/ex vivo Wistar rat models.
Results: Plasma proteomics identified low protein abundance levels of activin A in the gliflozin group. Empagliflozin (EMPA) decreased activin A and increased bone morphogenic protein receptor type 2 (BMPR2) in rat hearts undergoing ischemia/reperfusion (I/R) injury and human pulmonary artery endothelial cells undergoing hypoxia. Data are available via ProteomeXchange with identifier PXD063109. Plasma from a validation cohort of patients with HFrEF associated with high echocardiographic probability of PH treated with gliflozins were confirmed to have decreased activin A levels, which was associated with better disease progression in patients at follow-up, especially in terms of improvement of World Health Organization functional class, clinical signs of right heart failure, echocardiographic probability of PH and right ventricle-pulmonary artery (RV-PA) coupling.
Conclusions: This is the first demonstration that the activin A pathway is overactivated in patients with HFrEF and high PH probability. Serum levels of activin A, as well as the expression of activin and BMPR2 in hypoxic pulmonary endothelial cells and in the heart exposed to I/R injury, are downregulated and upregulated by EMPA, respectively. SGLT2i may thus rebalance the BMP/activin/transforming growth factor (TGF)β pathway and can therefore be a valuable therapeutic addition in patients with HFrEF and associated high echocardiographic probability of PH.