Myeloid p53 regulates macrophage polarization and venous thrombus resolution by inflammatory vascular remodeling in mice

Abstract

Deep venous thrombosis (DVT) remains a common and serious cardiovascular problem with both fatal and long-term consequences. The consequences of DVT include the development of postthrombotic syndrome in 25% to 60% of DVT patients. Despite the clinical importance of venous thrombus resolution, the cellular and molecular mediators involved are poorly understood, and currently there is no molecular therapy to accelerate this process. Several lines of evidence suggest that a complex and interrelated array of molecular signaling processes are involved in the inflammatory vascular remodeling associated with the resolution of DVT. Here, we have identified a role for the tumor suppressor gene p53 in regulating venous thrombus resolution. Using the stasis model of venous thrombosis and resolution in mice, we found that genetic deficiency of p53 or pharmacologic inhibition by pifithrin impairs thrombus resolution and is associated with increased fibrosis and altered expression of matrix metalloproteinase-2. The effect of p53 loss was mediated by cells of the myeloid lineage, resulting in enhanced polarization of the cytokine milieu toward an M1-like phenotype. Furthermore, augmentation of p53 activity using the pharmacological agonist of p53, quinacrine, accelerates venous thrombus resolution in a p53-dependent manner, even after establishment of thrombosis. Together, these studies define mechanisms by which p53 regulates thrombus resolution by increasing inflammatory vascular remodeling of venous thrombi in vivo, and the potential therapeutic application of a p53 agonist as a treatment to accelerate this process in patients with DVT.

Figure 1.

p53 deficiency enhances venous thrombus resolution. Thrombus weights of Tp53^−/− mice compared with Tp53^+/+ mice at (A) day 4 (n = 6-7 per group), (B) day 8 (n = 5-6 per group), and (C) day 12 (n = 8-9 per group) after vena cava ligation. Comparison of thrombus weights of C57BL/6 mice treated with vehicle or PFT at day 4 (D) and day 12 (E) after vena cava ligation (n = 5-6 per group).

Figure 2.

Characterization of venous thrombi from Tp53^+/+and Tp53^−/−mice at 4 days after vena cava ligation. Immunohistochemical analysis of intrathrombotic neutrophil accumulation was performed using anti-Lys6G antibodies (A) and macrophage accumulation was analyzed using anti-F4/80 antibodies (B) in venous thrombus samples from Tp53^+/+ and Tp53^−/− mice. (C) Collagen content in resolving Tp53^+/+ and Tp53^−/− thrombi at day 4 was determined in histological sections after Picrosirius Red staining; original magnification ×200. Scale bar, 200 μm. Representative results from 4 to 5 independent animals are shown. The numbers of Lys6G⁺ cells (neutrophils) (D) and F4/80⁺ cells (macrophages) (E) were determined as described in “Methods.” All values represent the mean ± standard error of the mean (SEM) (n = 4 Tp53^+/+ and 5 Tp53^−/−).

Figure 3.

Effect of p53 deficiency on collagen remodeling. (A) Collagen content in resolving Tp53^+/+ and Tp53^−/− thrombi at day 12 was determined in histological sections after Picrosirius Red staining; original magnification ×100. Scale bar, 400 μm. Representative images from 5 independent mice from each genotype are shown. Collagen content was quantified as a measure of fibrosis as described in “Methods.” All values represent the mean ± SEM (n = 5 Tp53^+/+ and 5 Tp53^−/−). *P < .05, Tp53^+/+ vs Tp53^−/−. (B) Representative gel images of intrathrombotic MMP-2 and MMP-9 activities in venous thrombus samples from Tp53^+/+ and Tp53^−/− mice as measured by gelatin gel zymography (n = 5 and 4 for Tp53^+/+ and Tp53^−/−, respectively). Gel images were subjected to semiquantitative analysis as described in “Methods.” All values represent the mean ± SEM (n = 4-5 per group). *P < .03, Tp53^+/+ vs Tp53^−/−. (C) Intrathrombotic mRNA expression of MMP-2 and MMP-9 in Tp53^+/+ and Tp53^−/− mice at day 8 was determined by quantitative PCR (qPCR). All values represent the mean ± SEM (n = 4 per group). *P < .03, Tp53^+/+ vs Tp53^−/−.

Figure 4.

Effect of absence of p53 on intrathrombotic cytokine expression. (A) Intrathrombotic expression of IL-6 mRNA at day 8 and day 12 after vena cava ligation was determined by qPCR. Values represent the mean ± SEM (n = 3 Tp53^+/+ and 4 Tp53^−/−). *P < .05, Tp53^+/+ vs Tp53^−/−. (B) Intrathrombotic expression of IL-6 protein at day 12 was determined by enzyme-linked immunosorbent assay (ELISA). All values represent the mean ± SEM (n = 4 Tp53^+/+ and 5 Tp53^−/−). **P < .02, Tp53^+/+ vs Tp53^−/−. (C) Enumeration of intrathrombotic macrophages at day 12 after vena cava ligation by immunohistochemical staining using anti-F4/80 antibody; original magnification ×200. Scale bar, 200 μm. Representative images from 4 independent animals are shown. (D) The number of macrophages was quantified as described in “Methods.” All values represent the mean ± SEM (n = 4 Tp53^+/+ and 4 Tp53^−/−). (E) Intrathrombotic gene expression of macrophage polarization markers at 8 days (A) and 12 days (B) after vena cava ligation in Tp53^+/+ and Tp53^−/− animals were determined by qPCR. All values represent the mean ± SEM (n = 3 Tp53^+/+ and 4 Tp53^−/−). *P < .05, Tp53^+/+ vs Tp53^−/−. hpf, high-powered field; iNOS, inducible NO synthase.

Figure 5.

Myeloid cell-specific p53 deficiency impairs venous thrombus resolution. (A) Thrombus weights of Mp53^+/+ and Mp53^−/− mice 12 days after vena cava ligation (n = 7 Tp53^+/+ and 6 Tp53^−/−). *P < .05, Mp53^+/+ vs Mp53^−/− mice. (B) Intrathrombotic F4/80 mRNA expression at 12 days after vena cava ligation in Mp53^+/+ and Mp53^−/− mice, determined by qPCR. All values represent the mean ± SEM (n = 4 Tp53^+/+ and 4 Tp53^−/−). (C) Intrathrombotic mRNA expression of macrophage polarization markers at 12 days after vena cava ligation in Mp53^+/+ and Mp53^−/− mice, as determined by qPCR. All values represent the mean ± SEM (n = 4 Tp53^+/+ and 4 Tp53^−/−). *P < .05, Mp53^+/+ vs Mp53^−/−. (D) Histochemical analysis of intrathrombotic collagen content by Picrosirius Red staining at day 12 after vena cava ligation; original magnification ×100. Representative images from 4 to 5 independent mice of each genotype are shown. The intrathrombus collagen area was quantitated as a measure of fibrosis as described in “Methods.” All values represent the mean ± SEM (n = 5 Tp53^+/+ and 4 Tp53^−/− n= 4-5). *P < .05, p53 Mp53^+/+ vs Mp53^−/−.

Figure 6.

Quinacrine enhances venous thrombus resolution through myeloid p53-dependent and MMP-2-independent manners. (A) Thrombus weights of vehicle- and quinacrine-treated CD1 mice at (A) day 4 (n = 6-7 per group) and (B) day 12 (n = 5-6 per group), after vena cava ligation. (C) Thrombus weights of Mp53^−/− mice at 12 days after vena cava ligation are unaffected after pretreatment with quinacrine (n = 6) or PFT (n = 5) compared with vehicle (n = 5). (D) Representative gel images of intrathrombotic MMP-2 and MMP-9 activities measured by gelatin gel zymography in day 8 venous thrombus samples from CD1 mice pretreated with vehicle (n = 4) or quinacrine (n = 4). Gel images were subjected to semiquantitative analysis as described in “Methods.” All values represent the mean ± SEM. (E) Thrombus weights of Mmp2^−/− mice at 12 days after vena cava ligation following pretreatment with vehicle or quinacrine (n = 5 vehicle and 5 quinacrine).

Figure 7.

Quinacrine treatment alters intrathrombotic macrophage polarization and fibrosis. (A) Intrathrombotic mRNA expression of macrophage polarization markers at day 12 after vena cava ligation following treatment with vehicle or quinacrine, as determined by qPCR. All values represent the mean ± SEM (n = 4 vehicle and 4 quinacrine). *P < .05, vehicle vs quinacrine. (B) Intrathrombotic expression of IL-6 mRNA at day 12 after vena cava ligation following treatment with vehicle or quinacrine by qPCR. Values represent the mean ± SEM (n = 4 vehicle and 4 quinacrine). *P < .05, vehicle vs quinacrine. (C) Intrathrombotic expression of IL-6 protein at day 12 was determined by ELISA. All values represent the mean ± SEM (n = 4 vehicle and 4 quinacrine). **P < .05, vehicle and vs quinacrine. (D) Histochemical analysis of intrathrombotic collagen content by Picrosirius Red staining after treatment with vehicle or quinacrine at day 12 after vena cava ligation; original magnification ×100. Representative images from 4 to 5 independent mice of each group. The intrathrombus collagen area was quantified as a measure of fibrosis as described in “Methods.” All values represent the mean ± SEM (n = 4 vehicle and 5 quinacrine). *P < .05, vehicle vs quinacrine treated. (E) Effect of quinacrine treatment on established thrombi. CD-1 mice were treated with vehicle or quinacrine from 3 days post–vena cava ligation, and thrombus weights measured at 12 days after vena cava ligation (n = 8 vehicle and 9 quinacrine). *P < .05, vehicle vs quinacrine.