**1. Introduction**

Atherosclerosis is considered a chronic inflammatory process and involves a complex pathophysiological effect, including endothelial dysfunction, low-density lipoprotein (LDL) oxidation, foam cell formation, and vascular smooth muscle cell (VSMC) proliferation and migration at different stages of this disease [1,2]. Elevated plasma LDL concentration contributes to the initiation of atherosclerosis [3]. Oxidized LDL triggers endothelial cells to release chemokines in contribution to recruitment of monocytes, resulting in the transformation of the lipid-laden macrophages into foam cells [3]. In the lesion progression, these activated macrophages still secrete proinflammatory

cytokines, especially tumor necrosis factor-alpha (TNFα), which enhances VSMC migration and proliferation [1,3]. Subsequently, VSMC transforms and proliferates into foam cells, and thus the accumulation of foam cells leading to fatty streaks results in the formation of atherosclerotic plaques [2]. Thus, inhibition of abnormal VSMC migration and proliferation is an attractive strategy for clinical therapy of atherosclerosis and restenosis after percutaneous coronary interventions.

VSMC is normally quiescent, but upon vascular injury, it transforms into a more synthetic phenotype with progressively increasing capacity for activation, proliferation, and migration [1,4]. In the atherosclerotic process, VSMC migrates from the media to the intima, forms the neointima progressively with abundant levels of extracellular matrix (ECM) proteins, and then eventually leads to plaque formation [2]. Identification of key proteins involved in the process, such as matrix metalloproteinases (MMPs), is vital for understanding atherosclerosis and devising new therapies. MMPs are a subfamily of the metzincin superfamily of endogenous proteinases that break down components of ECM. Among them, the gelatinases (MMP-2 and MMP-9) degrade e fficiently native collagen types IV and laminin, and promote a VSMC migratory phenotype [5]. Moreover, the gene expression of MMPs is majorly regulated by the transcriptional factors, such as activator protein-1 (AP-1) or nuclear factor-kappaB (NF-κB) through the serine/threonine protein kinase PKB (also known as Akt) or extracellular signal-regulated kinase (ERK) pathways, or by the MMP protein activators or inhibitors. One review study concluded that oxidative stress could enhance MMP activity and expression [6], and recent studies further indicate that MMP-mediated ECM remodeling is modulated by reactive oxygen species (ROS) [7]. Hence, MMPs and their regulatory signaling have been considered as promising targets for anti-atherosclerotic agents [8].

In arterial media, VSMC is at low proliferative indices (<0.05%) and remains in the G0/G1 phase of the cell cycle [4]. However, VSMC re-enters into the cell cycle from the quiescent state to proliferate under the stimulation of several cytokines in pathological processes, which plays an important role in the development of atherosclerosis [1]. VSMC begins to divide in response to cytokines, exits the G1 phase, and then enters the S phase. During the G1/S transition, cyclin D1/cyclin-dependent kinase (cdk) 4 and cyclin E/cdk2 complexes are required. The complexes participate in the hyperphosphorylation of retinoblastoma (Rb) tumor suppressor, leading phosphorylated Rb (p-Rb) to release E2F transcription factor, allowing the cells to progress into S phase [9]. The kinase activities of these cyclin/cdk complexes are regulated by cdk inhibitors (cki), including p16, p21, and p27. The gatekeeper of the mammalian cell cycle, p53, plays a key role in controlling G0/G1 arrest through its downstream factor, such as p21 [10].

Previous studies have reported that *Hibiscus*leaf, an edible part of*H. sabdari*ff*a* Linne (*Malvaceae*) [11], possesses hypoglycemic [12], hypolipidemic [13,14], and antioxidant [13,15] e ffects, as demonstrated by various experimental models (Table S1). For the standardization of each extract, our studies also indicated that (–)-epicatechin gallate (ECG; 16.5 ± 5.6%) was identified to be present in the highest level in *Hibiscus* leaf polyphenols (HLPs), followed by ellagic acid (EA; 10.31 ± 3.43%) and catechin (Cat; 7.4 ± 2.6%), and traces of only quercetin (Que; 0.8 ± 0.4%) and ferulic acid (FA; 0.7 ± 0.3%) were detected (Table S2) [14]. In this regard, the aqueous and methanol extracts of *H. sabdari*ff*a* leaves showed anti-atherogenic e ffects in hyperlipidemia animals induced by cholesterol [11,12], and inhibited foam cell formation, as well as protected endothelial cells from injury in vitro [11,16]. Our recent studies also revealed that *Hibiscus* leaf aqueous extract, due to its high content in polyphenols, has apoptotic and anti-migratory e ffects on prostate cancer cells [17,18]. However, little information is available on the isolation and characterization of a polyphenolic extract of *H. sabdari*ff*a* leaves. In the present study, HLP was partially characterized by biochemical and spectroscopic assays, and evaluated for the ability to inhibit TNFα-stimulated VSMC dysfunction.

Many studies have indicated that plant-derived polyphenols have various pharmacological and biological e ffects, such as antioxidant, anti-inflammatory, anti-hyperlipidemia, anti-diabetes, anti-atherogenic, and anti-tumor abilities [19]. Furthermore, although the protective e ffects of HLPs on endothelial cells and macrophages have been demonstrated previously, the in vivo function and

the molecular target of HLPs on VSMC remain to be elucidated in cardiovascular microenvironment. Using a model of VSMC exposed to TNFα and the well-established atherosclerotic rabbit experiment, to our knowledge, this is the first report revealing the TNFα-antagonist potential of HLPs in vitro and in vivo.
