Heterozygous inactivation of the Nf1 gene in myeloid cells enhances neointima formation via a rosuvastatin-sensitive cellular pathway

Abstract

Mutations in the NF1 tumor suppressor gene cause Neurofibromatosis type 1 (NF1). Neurofibromin, the protein product of NF1, functions as a negative regulator of Ras activity. Some NF1 patients develop cardiovascular disease, which represents an underrecognized disease complication and contributes to excess morbidity and mortality. Specifically, NF1 patients develop arterial occlusion resulting in tissue ischemia and sudden death. Murine studies demonstrate that heterozygous inactivation of Nf1 (Nf1(+/-)) in bone marrow cells enhances neointima formation following arterial injury. Macrophages infiltrate Nf1(+/-) neointimas, and NF1 patients have increased circulating inflammatory monocytes in their peripheral blood. Therefore, we tested the hypothesis that heterozygous inactivation of Nf1 in myeloid cells is sufficient for neointima formation. Specific ablation of a single copy of the Nf1 gene in myeloid cells alone mobilizes a discrete pro-inflammatory murine monocyte population via a cell autonomous and gene-dosage dependent mechanism. Furthermore, lineage-restricted heterozygous inactivation of Nf1 in myeloid cells is sufficient to reproduce the enhanced neointima formation observed in Nf1(+/-) mice when compared with wild-type controls, and homozygous inactivation of Nf1 in myeloid cells amplified the degree of arterial stenosis after arterial injury. Treatment of Nf1(+/-) mice with rosuvastatin, a stain with anti-inflammatory properties, significantly reduced neointima formation when compared with control. These studies identify neurofibromin-deficient myeloid cells as critical cellular effectors of Nf1(+/-) neointima formation and propose a potential therapeutic for NF1 cardiovascular disease.

Figure 1.

Ly6C^hi monocytes are mobilized in a cell-autonomous manner. (A) Representative gating strategy: within the CD115-positive population (red box), monocytes are distinguished by positive staining for Ly6C (x-axis) and negative staining for CD11c (y-axis). Pro-inflammatory Ly6C^hi monocytes are identified within the green box for representative WT, Nf1^fl/+;LysMcre and Nf1^fl/fl;LysMcre mice. (B) Absolute number of Ly6C^hi monocytes in peripheral blood of WT, Nf1^+/–, Nf1^fl/+;LysMcre and Nf1^fl/fl;LysMcre. Data represent mean percentage ± SEM, n = 10. *P < 0.02 for WT versus Nf1^+/– and Nf1^fl/+;LysMcre. **P < 0.0001 for WT, Nf1^+/– and Nf1^fl/+;LysMcre versus Nf1^fl/fl;LysMcre. Analysis by one-way ANOVA with Tukey's post-hoc test. There was no statistical difference between Nf1^+/− and Nf1^fl/+;LysMcre mice.

Figure 2.

Heterozygous inactivation of Nf1 in myeloid cells is sufficient for enhanced neointima formation. (A) Representative Van Gieson-stained cross sections of uninjured and injured carotid arteries from WT, Nf1^+/– and Nf1^fl/+;LysMcre mice. Black arrows indicate neointima boundaries. Black boxes identify area of injured artery that is magnified in the far-right panels. Scale bars: 100 μm. (B and C) Quantification of neointima area (B) and I/M ratio (C) of injured carotid arteries from WT, Nf1^+/– and Nf1^fl/+;LysMcre mice. Data represent the mean neointima area of three arterial cross sections (400, 800 and 1200 μm proximal to the ligation) ± SEM, n = 10–12. *P < 0.01 for WT uninjured versus WT injured, Nf1^+/− uninjured versus Nf1^+/− injured and Nf1^fl/+;LysMcre uninjured versus Nf1^fl/+;LysMcre injured. **P < 0.01 for WT injured versus Nf1^+/– injured and Nf1^fl/+;LysMcre injured. Analysis by one-way ANOVA. There was no statistical difference between Nf1^+/− and Nf1^fl/+;LysMcre injured. (D) Representative photomicrograph of injured carotid artery cross section from Nf1^fl/+;LysMcre mice stained with anti-Mac3 antibody (brown) and counterstained with hematoxylin (blue). Black arrows indicate neointima boundaries. Black arrowheads represent positive Mac3 staining. Black box identifies area of injured artery that is magnified in the far-right panel. Scale bars: 100 μm.

Figure 3.

Macrophages and neutrophils are recruited to the evolving neointima by 7 days post-injury. Representative photomicrographs from injured carotid arteries from WT (top panels) and Nf1^+/−(bottom panels) mice demonstrating macrophage (left column) and neutrophil (right column) infiltration into the neointima 7 days after arterial ligation. Black arrows indicate neointima boundaries. Black arrowheads represent positive macrophage (anti-Mac3) and neutrophil (anti-NIMP-R14) staining. Black box identifies area of injured artery that is magnified below. Scale bars: 100 μm. For each carotid artery, Mac-3-positive and NIMP-R14-positive neointimal cells were counted separately in five distinct 40 × images at 400, 800 and 1200 μm proximal to the bifurcation, and average cells per HPF were calculated (n = 5 per group). There was no difference in number of macrophages or neutrophils per HPF in the intimal layer between WT and Nf1^+/− mice.

Figure 4.

Arterial occlusion and stenosis is regulated by neurofibromin. (A) Representative Van Gieson-stained cross section of uninjured and injured carotid arteries from Nf1^fl/fl;LysMcre mice. Black arrows indicate neointima boundaries. Scale bar: 100 μm. (B) Quantification of lumen stenosis of injured carotid arteries from WT, Nf1^+/−, Nf1^fl/+;LysMcre and Nf1^fl/fl;LysMcre arteries. Data represent the mean percentage of lumen stenosis of three arterial cross sections (400, 800 and 1200 μm distal to the ligation) ± SEM, n = 10–12. *P < 0.01 for WT versus Nf1^+/− and Nf1^fl/+;LysMcre. **P < 0.01 for WT, Nf1^+/−, Nf1^fl/+;LysMcre mice versus Nf1^fl/fl;LysMcre. Analysis by one-way ANOVA. There was no statistical difference between Nf1^+/− and Nf1^fl/+;LysMcre.

Figure 5.

Rosuvastatin reduces proliferation, migration and adhesion in Nf1^+/− macrophages. (A) WT (white bars) and Nf1^+/– (black bars) macrophage proliferation in response to stimulation with M-CSF (20 ng/ml) in the presence of rosuvastatin, where indicated. Data represent thymidine incorporation reported as mean counts per minute (cpm) ± SEM, n = 6. *P < 0.001 for WT versus Nf1^+/– macrophages stimulated with M-CSF. **P < 0.001 for WT macrophages without growth factor versus WT macrophages incubated with M-CSF and Nf1^+/– macrophages without growth factor versus Nf1^+/– macrophages incubated with M-CSF. #P < 0.01 for Nf1^+/– macrophages stimulated with M-CSF versus Nf1^+/– macrophages stimulated with M-CSF in the presence of 10 and 20 μm rosuvastatin. Analysis was performed by one-way ANOVA. (B) WT (white bars) and Nf1^+/– (black bars) macrophage migration in response to stimulation with M-CSF (100 ng/ml) in the presence of rosuvastatin, where indicated. Data represent average number of migrated cells per HPF ± SEM, n = 6. *P < 0.01 for WT macrophages versus Nf1^+/– macrophages in the absence and presence of M-CSF. **P < 0.001 for WT macrophages without M-CSF versus WT macrophages incubated with M-CSF and Nf1^+/– macrophages without M-CSF versus Nf1^+/– macrophages incubated with M-CSF. ^#P < 0.001 for WT macrophages stimulated with M-CSF versus WT macrophages stimulated with M-CSF in the presence of 10 μm rosuvastatin and Nf1^+/– macrophages stimulated with M-CSF versus Nf1^+/– macrophages stimulated with M-CSF in the presence of 10 μm rosuvastatin. Analysis by one-way ANOVA. (C) WT (white bars) and Nf1^+/– (black bars) macrophage adhesion to fibronectin. Data represent mean optical density ± SEM, n = 4. *P < 0.01 for WT versus Nf1^+/– macrophages; **P < 0.01 for Nf1^+/– macrophages versus Nf1^+/– macrophages incubated with 10μm rosuvastatin at 15 and 30 min. Analysis by one-way ANOVA.

Figure 6.

Nf1^+/− VSMCs have increased response to TNFα that is abrogated by incubation with rosuvastatin. (A) WT (white bars) and Nf1^+/– (black bars) VSMC proliferation in response to stimulation with TNFα. Data represent thymidine incorporation reported as mean cpm ± SEM, n = 5. *P < 0.001 for WT versus Nf1^+/– VSMCs stimulated with indicated concentrations of TNFα. **P < 0.01 for WT and Nf1^+/– VSMCs versus WT and Nf1^+/– VSMCs stimulated with indicated concentrations of TNFα. Analysis by one-way ANOVA with Tukey's post-hoc test. (B) WT (white bars) and Nf1^+/– (black bars) VSMC proliferation in response to stimulation with TNFα in the presence of rosuvastatin, where indicated. Data represent thymidine incorporation reported as mean cpm ± SEM, n = 6. *P < 0.001 for WT VSMC stimulated with TNFα versus Nf1^+/– VSMCs stimulated with TNFα. **P < 0.001 for WT and Nf1^+/– VSMCs versus WT and Nf1^+/– VSMCs stimulated with TNFα. ^#P < 0.001 for Nf1^+/– VSMCs stimulated with TNFα versus Nf1^+/– VSMCs stimulated with TNFα in the presence of 5 and 10 μm rosuvastatin. Analysis by one-way ANOVA.

Figure 7.

Rosuvastatin reduces neointima formation in Nf1^+/− mice. (A) Representative Van Gieson-stained cross sections of uninjured and injured carotid arteries from rosuvastatin-treated WT and Nf1^+/– mice. Representative photomicrographs of H&E-stained carotid arteries from WT (top panels) and Nf1^+/− (bottom panels) mice 28 days, following no injury (left panels), injury and PBS treatment (middle panels) or injury and rosuvastatin treatment (right panels). Black arrows indicate neointima boundaries. Black boxes identify area of injured artery that is magnified below. Scale bars: 100 μm. (B and C) Quantification of neointima area (B) and I/M ratio (C) of injured carotid arteries from PBS and rosuvastatin-treated WT and Nf1^+/– mice. Data represent the mean neointima area of three arterial cross sections (400, 800 and 1200 μm distal to the ligation) ± SEM, n = 10–12. *P < 0.01 for WT uninjured versus WT injured with PBS treatment and Nf1^+/− uninjured versus Nf1^+/− injured with PBS treatment. **P < 0.01 for WT injured with PBS treatment versus Nf1^+/– injured with PBS treatment. ^#P < 0.01 for Nf1^+/− injured with PBS treatment versus Nf1^+/− injured with rosuvastatin treatment. Analysis by one-way ANOVA.