Investigating the role of load in failing human hearts

Left ventricular assist devices (LVADs) have proven to be a highly effective treatment strategy for end-stage heart failure with reduced ejection fraction (HFrEF). Notably, there are discernible physiological differences between the apex and base of the left ventricle (LV). We propose that these differences are load dependent and play a crucial role in LV remodelling during LVAD support. The aim of our study is to investigate the effects of unloading on the structure and function of cardiomyocytes, as well as the extracellular matrix (ECM), in both rat and human, focusing on the apical and basal regions of the LV. Methods: a rat model of partial mechanical unloading (PMU) through heterotopic abdominal heart and lung transplantation was used to examine the impact on t-tubules and their associated structural proteins (Junctophillin2, Caveolin3). We evaluated the function of cardiomyocytes in the apical and basal regions by stimulating β2 adrenergic receptors and measuring single-cell contraction and FRET (Förster resonance energy transfer). Furthermore, we assessed t-tubule structure and ECM remodelling in heart failure patients who underwent heart transplantation (HTx), comparing those with and without LVAD support. Additionally, we investigated ECM remodelling in the right ventricle (RV) after LVAD implantation. Results: in control rat basal cardiomyocytes have a denser (Apex 29.26±1.095 vs. Base 38.71±0.78 n=21 p<0.05) and more organized t-tubules network (Power of Regularity (PoR) Apex 1.12e+8±9.14e+6 vs. BCM 1.84e+8±1.62e+7 n=21 p<0.05) with higher organisation of structural proteins Junctophillin-2 and Cav3. After 8 weeks of PMU these differences were equalised (Apex 9.01e+7±1.05e+7 vs. Base 9.52e+7±1.07e+7 n=21,NS). b2AR response is more tightly controlled in basal cardiomyocytes (Contraction fold of increase Apex 0.52±0.19 vs. Base 0.11±0.08 n=12 p<0.05.) This difference is equalized after unloading (Contraction Apex 0.40±0.14 vs. Base 0.48±0.15 n=10, NS). This correlates to the equilibration of cAMP responses in PKA-RII domains 6 observed with FRET. Human basal cardiomyocytes have a denser and more organised t-tubules network in control and HFrEF (Apex 9.9x105 ± 8.6x105 vs Base 1.5x106 ± 7.6x105, p=0.02). In patients with LVAD regional differences were equalised (PoR 1.2x106 ± 9.7x105, p=0.18) and ttubules diameter was wider. Myocardial fibrosis increases post LVAD and is associated with increased tissue biomarkers (miRNA-27b-5p p=0.04; miRNA-497 p=0.03) and its pattern is region specific, occurring more in the base (Apex 13.1 ± 7.5% vs base 36.5 ± 15.9%, p= 0.009). This process is not drive by hypoxia. However, HIF-1 a is reduced in patient who had rehospitalization for HF at follow up. Late RV failure is a common complication post LVAD. It is associated with tissue hypoxia (HIF1-a RV p=0.02) and increased Collagen3 expression (p=0.0014). In conclusion LV unloading equalises regional structural differences in rats and humans. ECM remodelling post LVAD is region specific. RV failure is secondary to chronic volume overload. Regional differences across the LV are load dependent and are necessary for cardiac structure and function. Novel adjuvant therapies might help in achieving optimised LV remodelling during LVAD support.