AG was purchased from Hanpoong Pharm and Foods Company (Jeonju, Korea). AG roots were ground and subsequently extracted with 30% ethanol. The freeze-dried mixture was stored at −80°C. Human umbilical vein endothelial cells (HUVECs) were gifted by Kwang Seok Kim at Korea Institute of Radiological and Medical Science (Seoul, Korea). Human brain microvascular pericytes were purchased from ScienCell Research Laboratories, Inc. (1200; Carlsbad, CA, USA). HUVECs were cultured in endothelial medium supplemented with 5% fetal bovine serum (FBS) and 1% endothelial cell growth supplement, microvascular pericytes were cultured in pericyte medium supplemented with 5% FBS, 1% pericyte growth supplement, and 1% penicillin/streptomycin solution. FBS, growth supplements and media were purchased from ScienCell Research Laboratories, Inc. (Carlsbad, CA, USA). Cells were cultured at humidified incubator (5% CO₂; 95% relative humidity) at 37°C.

To investigate ET-1-mediated RhoA activation, pericytes were incubated with 100 nM/l ET-1 (Sigma-Aldrich, Merck KGaA, Darmstadt, Germany). To measure the inhibitory effect of AG on cold-induced RhoA activation, HUVECs and pericytes were pretreated with AG (100 or 200 µg/ml) for 30 min, followed by a warm (37°C) or cold (25°C) incubation for 30 min. Cells were lysed for protein extraction using ice-cold radioimmunoprecipitation assay buffer containing 50 mM Tris-HCl, (pH 7.5), 150 mM NaCl, 1% triton X-100, 2 mM EDTA, 0.1% SDS and 1% sodium deoxycholate (R2002; Biosesang Co., Ltd., Seoul, Korea). The extracted proteins were mixed with sample buffer (EBA-1052; Daejeon, Korea) and boiled at 100°C for 10 min. Protein amount was analyzed by using Bio-Rad protein assay kit (500–0006; Bio-Rad GmbH, Munchen, Germany). Equal amount of proteins (10 µg) were separated by 8–12% SDS-PAGE and transferred onto a nitrocellulose blotting membrane. The blots were probed with the following primary antibodies: Mouse-anti-active RhoA monoclonal (26904; 1:500; NewEast Biosciences, Malvern, PA, USA), rabbit-anti-phospho-focal adhesion kinase (FAK) monoclonal (8556; 1:500) and rabbit-anti-proto-oncogene tyrosine-protein kinase Src (SRC) polyclonal (2101; 1:1,000; both from Cell Signaling Technology, Inc., Danvers, MA, USA), mouse-anti-phospho-extracellular signal-related kinase (ERK) monoclonal (sc7383; 1:1,000) and mouse-anti-β-actin monoclonal (sc73615; 1:1,000; both from Santa Cruz Biotechnology, Inc., Dallas, TX, USA). After blocking the membrane with 2% skim milk for non-phosphoform of proteins or 2% Bovine serum albumin (BSA) in TBST for phosphoform of proteins, primary antibodies were incubated with the membrane overnight at 4°C on a shaker. Secondary antibodies for mouse (7076; 1:1,000–3,000) and rabbit (074; 1:1,000–3,000) were purchased from Cell Signaling Technology (Danvers, MA, USA) and incubated with the membrane for 1 h at room temperature on shaker. Antibody-conjugated membranes were incubated with ECL reagent (DG-WP250; DoGen, Seoul, Korea).

ET-1 production in HUVEC-conditioned medium was evaluated using an endothelin-1 ELISA kit, according to the manufacturer's protocol (ADI-900-020A; Enzo Life Sciences, Inc., Farmingdale, NY, USA).

The cold-induced formation of stress fibers and focal adhesion complexes was assessed in HUVECs. Cells were treated as aforementioned, and fixed with 4% paraformaldehyde (Junsei Chemical Co., Ltd., Tokyo, Japan) for 30 min and permeabilized with 0.1% Triton X-100 (T8787; Sigma-Aldrich, Merck KGaA, Darmstadt, Germany) for 15 min. Cells were stained with rhodamine-phalloidin (R415; Thermo Fisher Scientific, Inc., Waltham, MA, USA). The cells were visualized with Olympus FV10i self-contained confocal laser system (Olympus America Inc., PA, USA).

The differences of means between the groups were analyzed one-way analysis of variance. P<0.05 was considered to indicate a statistically significant difference.

Pericytes contribute to microvascular contraction; therefore, the inhibitory effect of AG on the activity of RhoA in cold- or ET-1-exposed pericytes was examined by western blotting with active RhoA-GTP monoclonal antibody. AG treatment inhibited cold-induced RhoA activation (Fig. 1A), and ET-1-induced RhoA activation (Fig. 1B).

The impact of AG treatment on cold-induced RhoA activation was investigated in HUVECs. AG treatment decreased cold-induced RhoA activation (Fig. 2). Therefore, AG treatment demonstrated an ability to inhibit cold-induced RhoA activation in both pericytes and HUVECs.

ET-1 is produced by ECs and serves a key role in vasoconstriction under cold conditions (7–9). Therefore, the inhibitory effect of AG treatment on ET-1 production in cold-exposed ECs was examined. HUVECs exposed to cold for 30 min increased ET-1 production by approximately 2-fold compared with the cells under normal conditions (P<0.05), however, AG pretreatment did not impact on cold-induced ET-1 production (Fig. 3).

Cold-mediated RhoA activation induces the phosphorylation of FAK, which stimulates the formation of stress fibers and focal adhesion complexes (4). Cold exposure induced the formation of stress fibers and focal adhesion complexes, whereas AG treatment limited these cold-mediated responses (Fig. 4A). Furthermore, FAK phosphorylation stimulates the phosphorylation of SRC and ERK, and western blot analysis demonstrated that the cold-induced phosphorylation of FAK, SRC and ERK was inhibited by AG treatment (Fig. 4B). Expression of unphosphorylated FAK, SRC and ERK was not changed (data not shown).