PAI-1 Regulates the Cytoskeleton and Intrinsic Stiffness of Vascular Smooth Muscle Cells

Abstract

Background: Plasma concentration of PAI-1 (plasminogen activator inhibitor-1) correlates with arterial stiffness. Vascular smooth muscle cells (SMCs) express PAI-1, and the intrinsic stiffness of SMCs is a major determinant of total arterial stiffness. We hypothesized that PAI-1 promotes SMC stiffness by regulating the cytoskeleton and that pharmacological inhibition of PAI-1 decreases SMC and aortic stiffness.

Methods: PAI-039, a specific inhibitor of PAI-1, and small interfering RNA were used to inhibit PAI-1 expression in cultured human SMCs. Effects of PAI-1 inhibition on SMC stiffness, F-actin (filamentous actin) content, and cytoskeleton-modulating enzymes were assessed. WT (wild-type) and PAI-1-deficient murine SMCs were used to determine PAI-039 specificity. RNA sequencing was performed to determine the effects of PAI-039 on SMC gene expression. In vivo effects of PAI-039 were assessed by aortic pulse wave velocity.

Results: PAI-039 significantly reduced intrinsic stiffness of human SMCs, which was accompanied by a significant decrease in cytoplasmic F-actin content. PAI-1 gene knockdown also decreased cytoplasmic F-actin. PAI-1 inhibition significantly increased the activity of cofilin, an F-actin depolymerase, in WT murine SMCs, but not in PAI-1-deficient SMCs. RNA-sequencing analysis suggested that PAI-039 upregulates AMPK (AMP-activated protein kinase) signaling in SMCs, which was confirmed by Western blotting. Inhibition of AMPK prevented activation of cofilin by PAI-039. In mice, PAI-039 significantly decreased aortic stiffness and tunica media F-actin content without altering the elastin or collagen content.

Conclusions: PAI-039 decreases intrinsic SMC stiffness and cytoplasmic stress fiber content. These effects are mediated by AMPK-dependent activation of cofilin. PAI-039 also decreases aortic stiffness in vivo. These findings suggest that PAI-1 is an important regulator of the SMC cytoskeleton and that pharmacological inhibition of PAI-1 has the potential to prevent and treat cardiovascular diseases involving arterial stiffening.

Keywords: RNA, small interfering; actin depolymerizing factors; collagen; gene expression; plasminogen activator inhibitor 1; sequence analysis, RNA; vascular stiffness.

Figure 1.. Pharmacologic inhibition of PAI-1 decreases vascular smooth muscle cell (SMC) stiffness.

Human coronary artery SMCs (passage number 5) were treated with vehicle control or PAI-039 (10μM) for 24 hours, after which cortical stiffness (elastic modulus) was assessed by atomic force microscopy. Results of 9–10 independent experiments for each treatment are shown with mean±SEM for each group. P value was determined with the Mann-Whitney U test.

Figure 2.. Pharmacologic inhibition of PAI-1 decreases F-actin content in vascular smooth muscle cells (SMCs).

(A) Human SMCs (passage number 5–7) were treated with PAI-039 (2.5–10μM) or vehicle control for 18 hours after which z-stack images were acquired by confocal microscopy and F/G-actin ratio was quantified. Results of 3 independent experiments for each PAI-039 concentration are shown with mean±SEM of each group. P values shown were determined by one-way repeated measures ANOVA with Geisser-Greenhouse correction and Tukey’s post hoc testing. (B) Representative images of human SMCs. Arrows indicate stress fibers. (C, D) Murine SMCs (passage number 5) were treated with PAI-039 (5μM) or vehicle control for 18 hours, after which F/G actin ratio was determined. Results of 5 independent experiments with wild-type SMCs (C) and 4 independent experiments with PAI-1-knockout SMCs (D) are shown with mean±SEM for each experimental group. P values shown were determined with the Mann-Whitney U test. (E) Representative images of murine SMCs. Scale bars represent 25 μm; F-actin is red; G-actin is green; nuclei are blue.

Figure 3.. PAI-039 promotes activation of cofilin.

Smooth muscle cells (SMCs, passage number 5–7) were treated with PAI-039 for 18 hours in serum-poor conditions (0.2% fetal bovine serum), then cell lysates were prepared and the ratio of phospho (P)-cofilin to total cofilin was quantified by SDS-PAGE and western blotting. (A) Results of 6 independent experiments with human SMCs treated with PAI-039 (10μM) or vehicle control. (B) Results of 7 independent experiments with wild-type murine SMCs treated with PAI-039 (5μM) or vehicle control. (C) Results of 7 independent experiments with PAI-1-knockout murine SMCs treated with PAI-039 (5μM) or vehicle control. In (A-C), numerical data include mean±SEM for each experimental group, a representative western blot is shown to the right of the graph, and P values shown were determined with the Mann-Whitney U test.

Figure 4.. Activation of cofilin by PAI-039 is not mediated by LIM domain kinase 1 or slingshot homolog 1.

Human smooth muscle cells (passage number 7) were treated with PAI-039 (10μM) or vehicle control for 0, 0.5, 1, or 3 hours, after which cell lysates were prepared and ratios of phosphorylated (P)-to-total protein were quantified by SDS-PAGE and western blotting for (A) cofilin, (B) LIM domain kinase 1 (LIMK) and (C) slingshot homolog 1 (SSH). All data points shown are mean±SEM of 3 independent experiments. P values shown were determined by two-way repeated measures ANOVA with Geisser-Greenhouse correction and Sidak multiple comparison post hoc testing. Representative western blots accompany each graph.

Figure 5.. PAI-039 activates cofilin through an AMPK-dependent pathway.

(A, B) Human coronary artery smooth muscle cells (SMCs, passage number 7) were serum-starved for 1 hour, then exposed to 4 experimental conditions, generated by adding dorsomorphin (4μM) or vehicle control for 30 minutes, then adding PAI-039 (10μM) or vehicle control for 60 minutes. Cell lysates were prepared, and ratios of phosphorylated (P)-to-total (A) AMPK and (B) cofilin were quantified by SDS-PAGE and western blotting with results normalized to the negative control group not treated with dorsomorphin or PAI-039. Results of 5 independent experiments, including mean±SEM of each group, are shown. (C) Representative western blots of human SMC lysates. (D) Murine SMCs isolated from wild-type or PAI-1-knockout (PAI-1-KO) mice (passage number 5) were serum starved for 1 hour, then treated for 1 hour with PAI-039 (5μM) or vehicle control. Cell lysates were prepared, and ratios of P-to-total AMPK were quantified by SDS-PAGE and western blotting with results normalized to wild-type SMCs not treated with PAI-039. Results of 4 independent experiments, including mean±SEM of each group, are shown. (E) Representative western blots of murine SMC lysates. For all graphs, P values shown were determined with the Kruskal-Wallis test with uncorrected Dunn’s test post hoc testing.

Figure 6.. PAI-1 inhibition decreases aortic stiffness and F-actin content without altering elastin or collagen content.

Age and sex-matched wild-type mice were fed a western diet containing or lacking PAI-039. (A) After 5 weeks, pulse wave velocity (PWV) in the abdominal aorta was significantly reduced in animals receiving PAI-039 (n=7 mice) vs. controls (n=7 mice). (B-H) Histologic studies. After 8.5 weeks, mice were euthanized and cross-sections from each abdominal aorta were stained for collagen and elastin content. There was no significance between groups in (B) tunica media thickness, (C) elastin content of the tunica media, or (E) collagen content of the full-thickness aortic wall. Representative images of (D) elastin (black) and (F) collagen (brick red) staining are shown. Scale bars represent 50 μm. (G) F-actin content of the tunica media was significantly lower in PAI-039-treated mice than controls. (H) Representative confocal fluorescence microscopy images of F-actin staining (red); nuclei are blue; L, lumen; scale bars represent 10 μm. All graphs include mean±SEM of each group. P values shown were determined with the Mann-Whitney U test (A and C) or the unpaired student t-test (B, E, and G).