Inhibition of bone morphogenetic proteins protects against atherosclerosis and vascular calcification

Abstract

Rationale: The bone morphogenetic proteins (BMPs), a family of morphogens, have been implicated as mediators of calcification and inflammation in the vascular wall.

Objective: To investigate the effect of altered expression of matrix Gla protein (MGP), an inhibitor of BMP, on vascular disease.

Methods and results: We used MGP transgenic or MGP-deficient mice bred to apolipoprotein E mice, a model of atherosclerosis. MGP overexpression reduced vascular BMP activity, atherosclerotic lesion size, intimal and medial calcification, and inflammation. It also reduced expression of the activin-like kinase receptor 1 and the vascular endothelial growth factor, part of a BMP-activated pathway that regulates angiogenesis and may enhance lesion formation and calcification. Conversely, MGP deficiency increased BMP activity, which may explain the diffuse calcification of vascular medial cells in MGP deficient aortas and the increase in expression of activin-like kinase receptor 1 and vascular endothelial growth factor. Unexpectedly, atherosclerotic lesion formation was decreased in MGP-deficient mice, which may be explained by a dramatic reduction in expression of endothelial adhesion molecules limiting monocyte infiltration of the artery wall.

Conclusions: Our results indicate that BMP signaling is a key regulator of vascular disease, requiring careful control to maintain normal vascular homeostasis.

Figure 1. Increased MGP expression limits the development of atherosclerotic lesions

Apoe^-/-and MGP^tg/wt;Apoe^-/-mice were maintained on Western diet for 16 weeks. (A-D) Mean lesion area (A), representative Oil Red O stained aortic sinus section (B), aortic lesion coverage in en face specimens (C), and representative Sudan IV stained aortas (D) of male and female Apoe^-/-and MGP^tg/wt;Apoe^-/-mice. Statistically significant differences were determined by the Mann-Whitney rank sum test.

Figure 2. Increased MGP expression limits lesion calcification and medial (non-lesion) calcification

Apoe^-/-and MGP^tg/wt;Apoe^-/-mice were maintained on Western diet for 16 weeks. (A) Total calcium in aortic tissues (n=4) as compared to chow-fed mice. (B) Schematic representation of analysis of lesion calcification. (C) Calcification rate in Apoe^-/-and MGP^tg/wt;Apoe^-/-mice. (D) Mean lesion calcification in Apoe^-/- and MGP^tg/wt;Apoe^-/-mice that exhibited calcification (male and female mice were analyzed together). (E) Representative Oil Red O stained aortic sinus sections with calcification from male and female Apoe^-/-and MGP^tg/wt;Apoe^-/-mice. (F) Medial calcification in Apoe^-/-and MGP^tg/wt;Apoe^-/-mice. Medial calcification was analyzed as schematically shown in insert. Representative section of Oil Red O stained medial calcification. Asterisks indicate statistically significant differences *<0.05, **<0.01, ***<0.001, Tukey's test. In panel D, statistically significant differences were determined by the Mann-Whitney rank sum test.

Figure 3. Fat-feeding enhances and increased MGP expression limits aortic BMP-signaling

Apoe^-/-and MGP^tg/wt;Apoe^-/-mice fed chow (CH) or Western (HF) diet were examined. (A) Aortic expression of BMP-2, BMP-4 and MGP, as determined by real-time PCR. (B) Serum levels of BMP-4, as determined by ELISA. (C) Aortic levels of BMP-4, as determined by immunoblotting with densitometry. (D) MGP expression and localization, as determined by immunohistochemistry, in aortic segments without atherosclerotic lesions. (E) Control experiment using MGP-/- aorta and the same MGP-antibodies as in panel D. (F) BMP-4 and MGP expression, as determined be immunofluorescence, in lesions of in Apoe^-/-and MGP^tg/wt;Apoe^-/-mice fed a Western diet. (G) BMP activity in total aorta, as determined by pSMAD1/5/8 immunoblotting with densitometry. (H) BMP activity in atherosclerotic lesions from Apoe^-/-and MGP^tg/wt;Apoe^-/-mice fed a Western diet, as determined by immunofluorescence. Asterisks indicate statistically significant differences compared to control (Apoe^-/- mice on chow). **<0.01, ***<0.001, Tukey's test.

Figure 4. Increased MGP expression attenuates aortic expression of ALK1 and VEGF

(A-C) Aortic expression of ALK2, ALK1 and VEGF in Apoe^-/-and MGP^tg/wt;Apoe^-/-mice fed chow (CH) or Western (HF) diet, as determined by real-time PCR (A), immunoblotting with densitometry (B), and immunohistochemistry (C). (D) Expression of ALK1 and VEGF colocalize with expression of vWF in atherosclerotic lesions from Apoe^-/-and MGP^tg/wt;Apoe^-/-mice fed a Western diet. Asterisks indicate statistically significant differences compared to control (Apoe^-/- mice on chow). **<0.01, ***<0.001, Tukey's test.

Figure 5. Increased MGP expression suppresses inflammatory activity

(A-D) Aortic expression of CD68, IL-6, HSP70, ICAM-1, VCAM-1 and E-selectin in Apoe^-/-and MGP^tg/wt;Apoe^-/-mice fed chow (CH) or Western (HF) diet, as determined by real-time PCR (A), immunoblotting with densitometry (B, C), and immunofluorescence (D). (E) Serum levels of IL-6 and HSP70, as determined by immunoblotting with densitometry. Asterisks indicate statistically significant differences. **<0.01, ***<0.001, Tukey's test.

Figure 6. Reduced lesion formation in extensively calcified arteries from in MGP^-/-;Apoe^-/- mice

Apoe^-/-, MGP^+/-;Apoe^-/- and MGP^-/-;Apoe^-/-mice were maintained on chow for 18 weeks. Male and female mice were analyzed together. (A) Mean lesion area. (B) Representative Oil Red O stained aortic sinus section from male and female Apoe^-/-and MGP^-/-;Apoe^-/-mice. (C) Total calcium in aortic tissues (n=4). (D) Calcification rate. (E) Mean lesion calcification. Asterisks indicate statistically significant differences. ***<0.001, Tukey's test. In panel A and D, statistically significant differences were determined by the Mann-Whitney rank sum test.

Figure 7. Increased aortic BMP-signaling and expression of ALK1 and VEGF in MGP^-/-;Apoe^-/- mice

(A,B) BMP activity in aortas of Apoe^-/- and MGP^-/-;Apoe^-/- mice as determined by immunoblotting with densitometry for pSMAD1/5/8 (A), and in lesions as determined by immunofluorescence (B) Structural changes in aortas of MGP^-/-;Apoe^-/- mice are visible in the H&E-stained sections (B, top). (C-E) Expression of ALK2, ALK1, VEGF, BMP-2 and -4 in aortas from Apoe^-/- and MGP^-/-;Apoe^-/- mice as determined by real-time PCR (C) and immunoblotting with densitometry (D), and in lesions as determined by immunofluorescence (E). Asterisks indicate statistically significant differences. **<0.01, ***<0.001, Tukey's test.

Figure 8. Increased aortic BMP-signaling but reduced expression of inflammatory proteins in MGP^-/-;Apoe^-/- mice

(A-D) Expression of CD68, IL-6, HSP70, ICAM-1, VCAM-1 and E-selectin in aortas of Apoe^-/- and MGP^-/-;Apoe^-/- mice as determined by real-time PCR (A), immunoblotting with densitometry for pSMAD1/5/8 (B,C), and in lesions as determined by immunofluorescence (D). Asterisks indicate statistically significant differences. ***<0.001, Tukey's test.