Integrin α5β1 regulates PP2A complex assembly through PDE4D in atherosclerosis
Sanguk Yun, … , David C. Pallas, Martin A. Schwartz
Sanguk Yun, … , David C. Pallas, Martin A. Schwartz
Published August 13, 2019
Citation Information: J Clin Invest. 2019;129(11):4863-4874. https://doi.org/10.1172/JCI127692.
View: Text | PDF
Research Article Cell biology Vascular biology

Integrin α5β1 regulates PP2A complex assembly through PDE4D in atherosclerosis

Abstract

Fibronectin in the vascular wall promotes inflammatory activation of the endothelium during vascular remodeling and atherosclerosis. These effects are mediated in part by fibronectin binding to integrin α5, which recruits and activates phosphodiesterase 4D5 (PDE4D5) by inducing its dephosphorylation on an inhibitory site, S651. Active PDE then hydrolyzes antiinflammatory cAMP to facilitate inflammatory signaling. To test this model in vivo, we mutated the integrin binding site of PDE4D5 in mice. This mutation reduced endothelial inflammatory activation in atherosclerosis-prone regions of arteries and, in a hyperlipidemia model, reduced atherosclerotic plaque size while increasing markers of plaque stability. We then investigated the mechanism of PDE4D5 activation. Proteomics identified the PP2A regulatory subunit B55α as the factor recruiting PP2A to PDE4D5. The B55α-PP2A complex localized to adhesions and directly dephosphorylated PDE4D5. This interaction also, unexpectedly, stabilized the PP2A-B55α complex. The integrin-regulated, proatherosclerotic transcription factor Yap was also dephosphorylated and activated through this pathway. PDE4D5 therefore mediated matrix-specific regulation of endothelial cell phenotype via an unconventional adapter role, assembling and anchoring a multifunctional PP2A complex that has other targets. We believe these results may have widespread consequences for the control of cell function by integrins.

Authors

Sanguk Yun, Rui Hu, Melanie E. Schwaemmle, Alexander N. Scherer, Zhenwu Zhuang, Anthony J. Koleske, David C. Pallas, Martin A. Schwartz

×

Figure 3

B55α is required for PP2A-dependent PDE4D dephosphorylation.

Options: View larger image (or click on image) Download as PowerPoint
B55α is required for PP2A-dependent PDE4D dephosphorylation. (A) FN prom...
(A) FN promoted B55α binding to PDE4D5. BAECs stably expressing GFP-tagged PDE4D5-S651E were plated on dishes coated with FN or MG for 20 minutes. Lysates were immunoprecipitated with GFP-Trap. SDS-PAGE and silver staining identified a FN-specific band at approximately 55 kDa, which was identified by LC-MS (see Supplemental Figure 3) as B55α. (B) Confirmation of proteomic results. PDE4D5-S651E immune complexes prepared as in A were subjected to Western blotting to detect PP2A subunits as indicated. Results are representative of 3 independent experiments. (C) For B55α knockdown, BAECs expressing WT GFP-tagged PDE4D5 were transfected with control or B55α siRNAs and plated on FN for 6 hours. Cell lysates were Western blotted for p-PDE4D (S651) or total PDE4D (t-PDE5D) (n = 4). (D) BAECs transfected with control siRNA (siCon) or B55α siRNA were plated on FN and exposed to laminar shear (LS) for 15 minutes. Flow-dependent S635 phosphorylation of eNOS was measured by immunoblotting with a phosphorylation-specific Ab (n = 4). (E) BAECs transfected with control or B55α siRNAs were plated on FN and subjected to OSS for 2 hours. NF-κB activation was measured by Western blotting for p-p65 (S536) (n = 3). (F) BAECs transfected with control or B55α siRNA were plated on FN, treated or not with the PKI 14-22 amide inhibitor (PKI) or DMSO, and subjected to OSS for 2 hours. NF-κB activation was measured by probing for p-p65 (S536) (n = 3). *P < 0.05, by Kruskal-Wallis test with Dunn’s multiple comparisons post hoc test (C), 2-tailed Student’s t test (D), or 1-way ANOVA (E and F). Ctrl, control.