Effect of SFRP5 (Secreted Frizzled–Related Protein 5) on the WNT5A (Wingless-Type Family Member 5A)-Induced Endothelial Dysfunction and Its Relevance With Arterial Stiffness in Human Subjects

• Published in: Arteriosclerosis, thrombosis, and vascular biology Vol. 38; no. 6; pp. 1358 - 1367
• Main Authors: Cho, Yun Kyung, Kang, Yu Mi, Lee, Seung Eun, Lee, Yoo La, Seol, So Mi, Lee, Woo Je, Park, Joong-Yeol, Jung, Chang Hee
• Format: Journal Article
• Language: English
• Published: United States American Heart Association, Inc 01.06.2018
• Subjects: atherosclerosis; endothelium; nitric oxide; pulse wave velocity; phosphorylation
• DOI: 10.1161/ATVBAHA.117.310649
• ISSN: 1079-5642; 1524-4636

OBJECTIVE—SFRP5 (secreted frizzled–related protein 5) is an endogenous inhibitor of WNT5A (wingless-type family member 5a), which has been implicated in atherosclerosis. However, contradictory results have been reported about the role of SFRP5 in atherosclerosis. We aimed to investigate whether SFRP5 could restore WNT5A-induced endothelial dysfunction in vitro and ex vivo. In addition, we sought to determine whether the serum concentration of SFRP5 is associated with atherosclerosis in humans. APPROACH AND RESULTS—We measured endothelium-dependent vasorelaxation in the isolated thoracic aorta of Sprague–Dawley rats. In addition, we measured intracellular nitric oxide (NO) in human endothelial cells. The protein abundance of total and phosphorylated JNK (c-Jun N-terminal kinase), AKT (protein kinase B), and endothelial NO synthase was analyzed in human endothelial cells. Circulating SFRP5 and WNT5A levels and brachial-ankle pulse wave velocity were measured in 282 human subjects with type 2 diabetes mellitus. SFRP5 dose dependently restored Wnt5-induced impaired vasorelaxation in rat thoracic aorta by an endothelial NO synthase–dependent mechanism. SFRP5 treatment restored the WNT5A-induced reduction of NO production via endothelial NO synthase in human endothelial cells. WNT5A-induced changes in the phosphorylation of JNK, AKT, and endothelial NO synthase were ameliorated with SFRP5 administration. In humans with type 2 diabetes mellitus, the serum SFRP5 concentration positively correlated with brachial-ankle pulse wave velocity (r=0.146; P=0.024). Multivariate linear regression analysis demonstrated that the serum SFRP5 concentration was independently associated with brachial-ankle pulse wave velocity after adjustment for potential confounders [B (SE)=7.40 (3.35); P=0.028]. CONCLUSIONS—Our data suggest the possible compensatory action of SFRP5 against atherosclerosis under conditions of metabolic dysfunction.