RNA interference targeting STIM1 suppresses vascular smooth muscle cell proliferation and neointima formation in the rat

Abstract

Our objective was to study the expression and function of stromal interaction molecule 1 (STIM1), an endoplasmic reticulum protein recently identified as the calcium sensor that regulated Ca(2+)-released activated channels in T cells. STIM1 was found to be upregulated in serum-induced proliferating human coronary artery smooth muscle cells (hCASMCs) as well as in the neointima of injured rat carotid arteries. Growth factors-induced proliferation was significantly lower in hCASMC transfected with STIM1 siRNA than in those transfected with scrambled siRNA (increase relative to 0.1% S: 116 +/- 12% and 184 +/- 16%, respectively, P < 0.01). To assess the role of STIM1 in preventing vascular smooth muscle cells (VSMCs) proliferation in vivo, we infected balloon-injured rat carotid arteries with an adenoviral vector expressing a short hairpin (sh) RNA against rat STIM1 mRNA (Ad-shSTIM1). Intima/media ratios reflecting the degree of restenosis were significantly lower in Ad-shSTIM1- infected arteries than in Ad-shLuciferase-infected arteries (0.34 +/- 0.02 vs. 0.92 +/- 0.11, P < 0.006). Finally, we demonstrated that silencing STIM1 prevents activation of the transcription factor NFAT (nuclear factor of activated T cell). In conclusion, STIM1 appears as a major regulator of in vitro and in vivo VSMC proliferation, representing a novel and original pharmacological target for prominent vascular proliferative diseases.

Figure 1

STIM1 is expressed in vascular smooth muscle cells (VSMCs). (a) Detection of STIM1 by immunofluorescence in a rat balloon-injured carotid artery. STIM1 is labeled in red, with a-STIM1; the green fluorescence represents autofluorescence of elastin and identifies the media. m: media, ni: neointima (b) Western blots of total extracts from human coronary artery (CA), human and rat VSMC and of Jurkat T cells hybridized with anti-GOK/Stim. (c) Confocal imaging of hCASMC labeled with anti-STIM1 and anti-SERCA2 (IID8).

Figure 2

STIM1 is upregulated in proliferative VSMC. (a) Relative STIM1 mRNA levels normalized to RPL32 mRNA in quiescent (0.1%S) and proliferative (5% S) hCASMC. (b) Western blot showing expression of STIM1, calcineurin (PP2B), and cyclin D1 in quiescent (0.1%S) and proliferative (5% S) hCASMC. (c) STIM1 (gray bars) and cyclin D1 (black bars) protein levels normalized to PP2B level in quiescent and proliferative hCASMC. ^**P < 0.01; ^***P < 0.001.

Figure 3

STIM1 knockdown inhibited hCASMC proliferation in vitro. (a) Western blot showing the disappearance of STIM1 and the reduction of cyclin D1 expression 72 hours after transfection with STIM1 siRNA compared to the negative control (scrambled) siRNA. (b) Proliferation (measured by BrDU incorporation) of hCASMC in presence of 5% supplement mix or (c) 50 nmol/l PDGF-BB in control cells or cells transfected with STIM1 or scrambled siRNA for 72 hours, or treated with 5 µmol/l cyclosporin A (CsA) for 24 hours. ^*P < 0.05; ^**P < 0.01.

Figure 4

Adenoviral vector expressing specific STIM1 shRNA prevented in vivo neointima formation in rat injured carotid artery. (a) Sequence of STIM1 shRNA. (b) Representative hematoxylin-eosin sections of noninjured (NI) and injured carotid arteries infected with Ad-sh Luciferase or Ad-sh STIM1, 14 days after surgery. (c) Average intima/media thickness ratios for the above three groups (^***P < 0.001 compared with Ad-shLuc). M indicates media; ni, neointima; and ad, adventitia (n = 5 for noninjured carotid, n = 4 for Ad-shLuc and n = 6 for Ad-shSTIM1). (d) PCR analysis of DNA extracted from the vessels in the respective conditions. (e) Immunofluorescence analysis of expression of STIM1 in carotid arteries 14 days after injury.

Figure 5

STIM1 silencing inhibited TRP single-channel activity. Representative single-channel activity recordings obtained from cell-attached membranes on hCASMC transfected with either scrambled siRNA (upper trace) or either STIM1 siRNA (lower trace) at a patch membrane potential maintained to −80 mV. Extracts a and b are time-scale expansion of the channel activity from the above trace.

Figure 6

RNA interference-induced STIM1 silencing prevented NFAT nuclear translocation and activity. (a,b) Measurement of NFAT activity using a NFAT-driven luciferase construct in control cells as in a or cells treated with thapsigargin (1 µmol/l for 6 hours) as in b (P < 0.001). (c) Localization of NFAT protein by immunofluorescence. (d) Measurement of the relative MCIP mRNA level normalized to RPL32 mRNA. Cyclosporin A (CsA) is used as a negative control (^*P < 0.05).