Lack of RAC1 in macrophages protects against atherosclerosis

Abstract

The Rho GTPase RAC1 is an important regulator of cytoskeletal dynamics, but the role of macrophage-specific RAC1 has not been explored during atherogenesis. We analyzed RAC1 expression in human carotid atherosclerotic plaques using immunofluorescence and found higher macrophage RAC1 expression in advanced plaques compared with intermediate human atherosclerotic plaques. We then produced mice with Rac1-deficient macrophages by breeding conditional floxed Rac1 mice (Rac1fl/fl) with mice expressing Cre from the macrophage-specific lysosome M promoter (LC). Atherosclerosis was studied in vivo by infecting Rac1fl/fl and Rac1fl/fl/LC mice with AdPCSK9 (adenoviral vector overexpressing proprotein convertase subtilisin/kexin type 9). Rac1fl/fl/LC macrophages secreted lower levels of IL-6 and TNF-α and exhibited reduced foam cell formation and lipid uptake. The deficiency of Rac1 in macrophages reduced the size of aortic atherosclerotic plaques in AdPCSK9-infected Rac1fl/fl/LC mice. Compare with controls, intima/media ratios, the size of necrotic cores, and numbers of CD68-positive macrophages in atherosclerotic plaques were reduced in Rac1-deficient mice. Moreover, we found that RAC1 interacts with actin-binding filamin A. Macrophages expressed increased RAC1 levels in advanced human atherosclerosis. Genetic inactivation of RAC1 impaired macrophage function and reduced atherosclerosis in mice, suggesting that drugs targeting RAC1 may be useful in the treatment of atherosclerosis.

Fig 1. Increased RAC1 expression in intimal macrophages of advanced human atherosclerotic carotid plaques.

(A) Hematoxylin & Eosin-stained sections of intermediate and advanced atherosclerotic plaques in human carotid endarterectomies (upper panels). Immunofluorescence detection of RAC1 (green), and macrophages (red, MΦ) in the intermediate and advanced atherosclerotic plaques (lower panels). Nuclear staining by DAPI (blue). Arrowhead points to the internal elastic lamina bordering the intimal thickening from the medial layer. Scale bars represent 100 μm. Co-localization of RAC1 expression in MΦ in an advanced intimal thickening shown in enlarged inset. (B) Number of macrophages in intermediate and advanced atherosclerotic plaques. (C) Number of macrophages expressing RAC1 expression within the intimal thickening of intermediate and advanced atherosclerotic plaques. Mean ± SEM values (n = 9 in each). Wilcoxon signed-rank test was used. *p<0.05.

Fig 2. Bone marrow-derived macrophages deficient for RAC1 display longer cell shape.

(A) RT-PCR analysis of Rac1, CD68, and Sm22α in cultured BMMs by gel electrophoresis. 18S mRNA serves as an internal loading control. (B) Immunoblot analysis of RAC1 in Rac1^fl/fl and Rac1^fl/fl/LC BMMs. Actin was included as an internal loading control. (C) Morphological shape of Rac1^fl/fl and Rac1^fl/fl/LC BMMs as detected by actin immunofluorescence staining (left images). Inset demonstrates CD68-positivity in BMMs. Scale bars represent 10 μm. Quantification of ratios of macrophage cell elongation (right graphs). Mean ± SEM values of percentage or fold changes in at least triplicated data. Student’s t-test was used. **p<0.01.

Fig 3. Mice lacking RAC1 in macrophages develop smaller aortic atherosclerotic plaques.

(A) Representative images of Sudan IV-stained atherosclerotic aortic plaques in aortas obtained from Rac1^fl/fl/LC mice infected with AdPCSK9 (n = 12) as compared to Rac1^fl/fl control mice infected with AdPCSK9 (n = 12) (left panels). Quantification of atherosclerotic plaque size by image analysis in Rac1^fl/fl and Rac1^fl/fl/LC aortas infected with AdPCSK9 (right graphs). (B) Intima/media ratios in Rac1^fl/fl and Rac1^fl/fl/LC aortic arches from mice infected with AdPCSK9. (C) Percentage of intimal necrotic core areas in Rac1^fl/fl and Rac1^fl/fl/LC aortic arches. (D) Immunofluorescent detection of macrophages (MΦ) using anti-CD68 antibodies in Rac1^fl/fl and Rac1^fl/fl/LC aortic arches from mice infected with AdPCSK9 (left panels). Number of MΦ in Rac1^fl/fl and Rac1^fl/fl/LC aortic arches (right graphs). Scale bars represent 100 μm. Mean ± SEM values. Student’s t-test was used. *p≤0.05.

Fig 4. Secretion of inflammatory cytokines is reduced in bone marrow-derived macrophages deficient for RAC1.

(A) Secretion of IL-6 and TNF-α in primary Rac1^fl/fl or Rac1^fl/fl/LC BMMs as detected by ELISA (n = 6 in each) (B) Serum blood levels of secreted IL-6 and TNF-α in atherogenic Rac1^fl/fl or Rac1^fl/fl/LC mice (n = 6 in each). Mean ± SEM values. Student’s t-test was used. *p<0.05 and **p<0.01.

Fig 5. Deficiency of RAC1 in bone marrow-derived macrophages decreases lipid uptake.

(A) Blood levels of cholesterol and triglyceride in fast protein liquid chromatography–fractionated plasma pooled from Rac1^fl/fl mice (n = 5) that are infected with AdPCSK9 as compared to Rac1^fl/fl/LC mice (n = 5) that are infected with AdPCSK9 after a high-fat diet for 24 weeks. HDL, high-density lipoprotein; IDL, intermediate-density lipoprotein; LDL, low-density lipoprotein; and VLDL, very-low-density lipoprotein. (B) Levels of minimally-modified LDL (mmLDL) in Rac1^fl/fl and Rac1^fl/fl/LC BMMs. Student’s t-test was used. Mean ± SEM values. (C) Uptake of total LDL and oxidized LDL (OxLDL) for 8 and 24 hours in Rac1^fl/fl and Rac1^fl/fl/LC BMMs. Multiple groups were statistically evaluated by ANOVA and the Tukey-Kramer modification of Tukey’s test and comparisons between two groups were evaluated by Student’s t test. (D) Representative Immunoblots and densitometry reading of SR-B1, LXRα/β, CD36, COX2, ABCG1, and ABCA1 bands in Rac1^fl/fl and Rac1^fl/fl/LC BMMs. Mean ± SEM values of triplicated data. Student’s t-test was used. *p<0.05.

Fig 6. RAC1 interacts with FLNA in the cytoplasm and deletion of RAC1 reduces the production of cleaved C-terminal fragment of filamin A (FLNA^CT).

(A) Expression of RAC1 (green) and FLNA (red) is co-localized mainly in the cytoplasm of BMMs (yellow) as detected by immunofluorescence staining. Scale bar represents 10 μm. (B) Co-immunoprecipitation identifying FLNA^CT as an interacting partner of RAC1. Total proteins obtained from BMMs immunoprecipitated with FLNA^CT and then immunoblotted with RAC1 antibodies (upper) or vice versa (lower). IgG served as negative controls. Full-length of FLNA and FLNA^CT were indicated and actin served as internal loading control. (C) Immunoblotting of FLNA in Rac1^fl/fl and Rac1^fl/fl/LC BMMs. (D) Immunofluorescence staining of FLNA^CT in Rac1^fl/fl and Rac1^fl/fl/LC BMMs. Quantification of nuclear FLNA^CT expression in Rac1^fl/fl or Rac1^fl/fl/LC BMMs. Scale bar represents 20 μm. Mean ± SEM values of at least quadruplicated experiments. Student’s t-test was used. *p<0.05, **p<0.01.