Transdifferentiation of mouse aortic smooth muscle cells to a macrophage-like state after cholesterol loading

Abstract

Mouse aortic smooth muscle cells (SMCs) were loaded for 72 h with cholesterol by using cholesterol:methyl-beta-cyclodextrin complexes, leading to approximately 2-fold and approximately 10-fold increases in the contents of total cholesterol and cholesteryl ester, respectively. Foam-cell formation was demonstrated by accumulation of intracellular, Oil Red O-stained lipid droplets. Immunostaining showed decreased protein levels of smooth muscle alpha-actin and alpha-tropomyosin and increased levels of macrophage markers CD68 and Mac-2 antigen. Quantitative real-time RT-PCR revealed that after cholesterol loading, the expression of SMC-related genes alpha-actin, alpha-tropomyosin, myosin heavy chain, and calponin H1 decreased (to 11.5 +/- 0.5%, 29.3 +/- 1.4%, 23.8 +/- 1.4%, and 3.8 +/- 0.5% of unloaded cells, respectively; P < 0.05 for all), whereas expression of macrophage-related genes CD68, Mac-2, and ABCA1 mRNA increased (to 709 +/- 84%, 330 +/- 11%, and 207 +/- 13% of unloaded cells, respectively; P < 0.05 for all), thereby demonstrating that the protein changes were regulated at the mRNA level. Furthermore, these changes were accompanied by a gain in macrophage-like function as assessed by phagocytotic activity. Expression of vascular cell adhesion molecule 1 and monocyte chemoattractant protein 1, known responders to inflammation, were not changed. In conclusion, cholesterol loading of SMC causes phenotypic changes regulated at the mRNA level that result in a transdifferentiation to a macrophage-like state. This finding suggests that not all foam cells in lesions may have a macrophage origin, despite what is indicated by immunostaining for macrophage-related markers. Furthermore, inflammatory changes in foam cells observed in vivo may not be simple consequences of cholesterol accumulation.

Fig. 7.

The effects of FC on SMC and macrophage-related gene expression. Subconfluent mouse SMCs were pretreated overnight with or without an ACAT inhibitor F-1394 (1 μM), then treated with (Chol, Chol + ACAT inhibitor) or without (Control, ACAT inhibitor alone) Chol:MβCD (10 μg/ml) in 0.2% BSA (72 h) in the presence of F-1394. Total RNA was extracted and subjected to QRT-PCR analysis. All data are averages ± SEM from independent duplicates. ^*,α-actin was not detected. †, P < 0.01 vs. Chol.

Fig. 1.

Cytotoxicity of Chol:MβCD. Subconfluent mouse SMCs were treated with increasing concentrations of Chol:MβCD in 0.2% BSA for 72 h. Cells were then incubated for 1 h with CellTiter 96 Aqueous One Solution Reagent, convertible to a colored formazan product by metabolically active cells. (A) Relative metabolic activity was determined by spectrophotometric measurement (OD = 490 nm) of the accumulation of the formazan product in the medium. (B) Total cellular protein was determined by using a Sigma kit. See Materials and Methods for details. Data are averages ± SEM from triplicate wells and repeated at least once.

Fig. 2.

Cholesterol loading leads to smooth muscle foam-cell formation. Subconfluent mouse SMCs were treated with (Chol) or without (Control) Chol:MβCD (10 μg/ml) in 0.2% BSA (72 h). Cells were fixed in 10% neutral-buffered formaldehyde and stained with Oil Red O (magnification: A, ×100; Inset, ×400) or harvested to determine cellular cholesterol (B) or to extract RNA for QRT-PCR determination of HMG-CoA reductase mRNA levels (C). The numerical data are averages ± SEM from independent duplicates.

Fig. 3.

Immunostaining of SMCs for SMC and macrophage-related proteins. Subconfluent mouse SMCs were treated with (Chol) or without (Control) Chol:MβCD (10 μg/ml) in 0.2% BSA (72 h). Cells were stained with antibodies for α-actin (red stain, ×200), α-tropomyosin (brown stain, ×200), CD68 (brown stain, ×200), and Mac-2 (brown stain, ×200).

Fig. 4.

QRT-PCR determination of SMC and macrophage-related gene expression in SMCs. Subconfluent mouse SMCs were treated with (Chol) or without (Control) Chol:MβCD (10 μg/ml) in 0.2% BSA (72 h). In C, control or cholesterol-loaded cells were treated with 30 ng/ml TNF-α for 2 h after cholesterol loading. Total RNA was extracted and subjected to QRT-PCR analysis of smooth muscle marker genes (A), macrophage marker genes (B), and macrophage-related inflammation genes (C). All data are averages ± SEM from independent duplicates.

Fig. 5.

Cholesterol loading increases the phagocytotic activity of SMCs. (A) Subconfluent mouse SMCs were treated with (Chol) or without (Control) Chol:MβCD (10 μg/ml) in 0.2% BSA (72 h), followed by incubation with 1-μm latex beads (green) for 20 h. Cells were then washed extensively, fixed, counterstained with 4′,6-diamidino-2-phenylindole (blue, for nuclei) and 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine (red, for cell membrane), and subjected to fluorescent microscopy. (Magnification, ×200.) (B) The number of cells and latex beads were also counted to obtain numerical data for phagocytotic activity. All data are averages ± SEM from independent duplicates.

Fig. 6.

The effects of cholesterol loading on SMC and macrophage-related gene expression in LLC cells (A) and NIH/3T3 fibroblasts (B). Subconfluent murine LLC cells or NIH/3T3 fibroblasts were treated with (Chol) or without (Control) Chol:MβCD (5 μg/ml) in 0.2% BSA (72 h). Total RNA was extracted and subjected to QRT-PCR analysis. All data are averages ± SEM from independent duplicates. ^*, α-actin was not detected.