PDZK1 prevents neointima formation via suppression of breakpoint cluster region kinase in vascular smooth muscle

Abstract

Scavenger receptor class B, type I (SR-BI) and its adaptor protein PDZK1 mediate responses to HDL cholesterol in endothelium. Whether the receptor-adaptor protein tandem serves functions in other vascular cell types is unknown. The current work determined the roles of SR-BI and PDZK1 in vascular smooth muscle (VSM). To evaluate possible VSM functions of SR-BI and PDZK1 in vivo, neointima formation was assessed 21 days post-ligation in the carotid arteries of wild-type, SR-BI-/- or PDZK1-/- mice. Whereas neointima development was negligible in wild-type and SR-BI-/-, there was marked neointima formation in PDZK1-/- mice. PDZK1 expression was demonstrated in primary mouse VSM cells, and compared to wild-type cells, PDZK1-/- VSM displayed exaggerated proliferation and migration in response to platelet derived growth factor (PDGF). Tandem affinity purification-mass spectrometry revealed that PDZK1 interacts with breakpoint cluster region kinase (Bcr), which contains a C-terminal PDZ binding sequence and is known to enhance responses to PDGF in VSM. PDZK1 interaction with Bcr in VSM was demonstrated by pull-down and by coimmunoprecipitation, and the augmented proliferative response to PDGF in PDZK1-/- VSM was abrogated by Bcr depletion. Furthermore, compared with wild-type Bcr overexpression, the introduction of a Bcr mutant incapable of PDZK1 binding into VSM cells yielded an exaggerated proliferative response to PDGF. Thus, PDZK1 has novel SR-BI-independent function in VSM that affords protection from neointima formation, and this involves PDZK1 suppression of VSM cell proliferation via an inhibitory interaction with Bcr.

Fig 1. PDZK1 deletion causes exaggerated carotid artery neointima formation.

A, B. Male SR-BI^+/+ and SR-BI^-/- mice (10–12 weeks of age) underwent unilateral left carotid artery ligation for the evaluation of neointima formation. Arteries were harvested 21d later, and representative images of sections from SR-BI^+/+ and SR-BI^-/- mice stained with hematoxylin and eosin are shown. The scale bar indicates 100 μm. C, D. Neointima formation was also evaluated in male PDZK1^+/+ and PDZK1^-/- mice. Neointima is indicated by arrow. E. Summary data for intima-to-media (IM) ratio for SR-BI^+/+ versus SR-BI^-/-. F. Summary data for IM ratio for PDZK1^+/+ versus PDZK1^-/-. In E and F, values are mean±SEM, n = 7–9, *p<0.05 vs. PDZK1^+/+.

Fig 2. VSM cells populate the neointima that develops in PDZK1-/- mice.

Male wild-type PDZK1^+/+ (A, C) and PDZK1^-/- mice (B, D) underwent unilateral left carotid artery ligation for the evaluation of neointima formation. Arteries were harvested 21d later, and immunofluorescent staining was performed with anti-αSMA monoclonal antibody to detect smooth muscle cells (A and B, green fluorescent staining for αSMA and blue for DAPI), and immunohistochemistry was done with anti-F4/80-polyclonal antibody to detect macrophages (C and D). Positive F4/80 signal is seen in rare luminal macrophages in C. The scale bar indicates 100 μm.

Fig 3. VSM cells from PDZK1-/- mice display enhanced PDGF-stimulated cell proliferation and migration.

A. Aortic VSM cells were isolated from PDZK1^+/+ or PDZK1^-/- mice and immmunfluorescent staining was performed with anti-α-SMA monoclonal antibody (red). The scale bar indicates 100 μm. B. PDZK1 mRNA was quantified in VSM cells and liver from PDZK1^+/+ and PDZK1^-/- mice. Results are expressed relative to cyclophilin. C. PDZK1 protein expression was evaluated by immunoblotting using anti-PDZK1 polyclonal antibody. Equal protein loading was confirmed by immunoblotting using anti-GAPDH antibody. D. Cell growth under quiescent conditions was assessed by quantification of ³H-thymidine incorporation. E.Cell proliferation was evaluated in VSM incubated in the presence of 0–20 ng/ml PDGF-BB for 24h. F. Cell migration was assessed in VSM incubated in the presence of 0–30 ng/ml PDGF-BB for 4h. In B and D-F, values are mean±SEM, n = 3, *p<0.05 vs. no PDGF, †p<0.05 vs. PDZK1^+/+.

Fig 4. Bcr is expressed in VSM cells and interacts with PDZK1.

A. PDZK1^+/+ or PDZK1^-/- VSM cells were transfected with control dsRNA (Con) or dsRNA targeting Bcr, and Bcr expression was evaluated by immunoblot analysis 24h later. B. Wild-type VSM cells were infected with adenovirus expressing protein G sequence-containing Tag alone (lane 1) or Tag-PDZK1 (lane 2), and 60h later the cells were lysed and Tag or Tag-PDZK1 were precipitated using IgG sepharose. Tag and Tag-PDZK1 were detected by HRP-linked secondary antibody binding to the Protein G sequence in the Tag (left panel). The pull-down of Bcr was detected by immunoblotting of samples from Tag-PDZK1 infected cells (right panel). C. Endogenous Bcr was immunoprecipitated from PDZK1^+/+ or PDZK^-/- VSM cells by anti-Bcr antibody, and the presence of Bcr and PDZK1 in the immunoprecipitates was detected by immunoblotting. D. Endogenous PDZK1 was immunoprecipitated by anti-PDZK1 antibody, and Bcr and PDZK1 in the immunoprecipitates was detected by immunoblotting.

Fig 5. Bcr interaction with PDZK1 blunts VSM cell proliferation.

A. VSM cells from PDZK1+/+ or PDZK1-/- mice were transfected with control dsRNA (Con siRNA) or dsRNA targeting Bcr (Bcr siRNA), and cell proliferation in response to PDGF-BB (10 ng/ml) was assessed by quantification of 3H-thymidine incorporation. Effective knockdown of Bcr is shown in Fig 4A. PDGF-BB-stimulated proliferation is expressed relative to proliferation in control, PBS-treated cells. Values are mean±SEM, n = 8, *p<0.05 vs. PDZK1^+/+, †p<0.05 vs. Con siRNA. B. VSM cells were infected with adenovirus expressing vector alone (sham), wild-type Bcr or Bcr-V1271A, and relative Bcr expression was evaluated by immunoblot analysis. C. In cells described in B, proliferation was assessed in control PBS-treated cells (open bar) and in response to PDGF-BB (10 ng/ml, closed bar). PDGF-BB-stimulated proliferation is expressed relative to proliferation in control-treated cells. Values are mean±SEM, n = 3, *p<0.05 vs. no PDGF, †p<0.05 vs. wild-type Bcr.