Inhibition of novel lipoprotein(a) receptor major facilitator superfamily domain containing 5 (MFSD5) reduces development of aortic valve calcification

Abstract

Abstract Funding Acknowledgements Type of funding sources: Other. Main funding source(s): Private grant from Kowa Pharmaceuticals to Brigham and Woman's Hospital Calcific aortic valve stenosis (CAVS) is the most prevent valvular heart disease in the western world increasing exponentially with age, with an 112% increase in CAVS deaths in the last three decades; however no therapeutic treatment is currently available. Recently, lipoprotein(a) [Lp(a)] has been demonstrated to be an independent and causal risk factor for CAVS, yet the understanding of its cellular uptake and catabolism is limited thus underscoring the need for further investigation. This study aimed to determine a target receptor, unique for Lp(a) on the surface of valvular interstitial cells (VICs) and ascertain the role of the receptor on the development of VIC calcification. Unbiased ligand-receptor capture mass spectrometry (TriCEPS) was used to identify target receptor, with western blotting, ELISA, qPCR, alizarin red calcium staining and immunofluorescence used to validate the targets in vitro via siRNA inhibition and overexpression. Transmission electron microscopy (TEM) was used to determine uptake of Lp(a) within excised human valves. Identification of small molecule inhibitors was assessed computationally via the L1000 dataset, with the top hit candidate validated in vitro. Genotype-phenotype studies were examined using the United Kingdom Biobank (UKBB) and the Millions Veterans Program. Linear regression was used to evaluate association between aortic stenosis and plasma Lp(a) levels, and a phenotype-wide association analysis was then performed against this generated ‘genotype’. Ligand-receptor capture mass spectrometry was used to detect novel membrane proteins with specific binding to Lp(a); MFSD5, MRC2, LDLR were identified as possible candidates. MFSD5 RNAscope demonstrated its presence in human aortic valves. Lp(a) uptake in VICs was confirmed via western blot and TEM. MFSD5 siRNA significantly reduced dil-labelled Lp(a) uptake in human VICs (p=0.003) and HEPG2 cells (p=0.0003), conversely MFSD5 overexpression increased uptake (p=0.0345, p=0.0318), whilst specificity of MFSD5 to Lp(a) alone was shown via no change in LDL uptake following MFSD5 inhibition (p=0.616, p=0.991). MFSD5 inhibition reduced RUNX2 (p=0.0124) and Osteocalcin (p<0.001) RNA expression and reduced alizarin red staining following culture in Lp(a) osteogenic media for 21 days (p<0.0033). Druggability of MFSD5 was confirmed by the L1000 database, which identified aminopurvalanol as a binding partner for MFSD5 and significantly reduced Lp(a) uptake within VICs (p=0.0091). MFSD5-loss of function within the UKBB showed no significant cardiovascular association, however 50kb +/- of the MFSD5 gene showed nominal association with hyperlipidaemia and atrial fibrillation. The current study demonstrates the novel Lp(a) receptor MFSD5 may be responsible for uptake of Lp(a) within VICs, resulting in the development of aortic valve calcification, highlighting the need for further exploration into the role of MFSD5 in aortic valve disease.