Rosuvastatin Attenuates CD40L-Induced Downregulation of Extracellular Matrix Production in Human Aortic Smooth Muscle Cells via TRAF6-JNK-NF-κB Pathway

CD40L and statins exhibit pro-inflammatory and anti-inflammatory effects, respectively. They are both pleiotropic and can regulate extracellular matrix (ECM) degeneration in an atherosclerotic plaque. Statins can decrease both the CD40 expression and the resulting inflammation. However, the effects of CD40L and stains on atherosclerotic plaque ECM production and the underlying mechanisms are not well established. Moreover, prolyl-4-hydroxylase α1 (P4Hα1) is involved in collagen synthesis but its correlations with CD40L and statins are unknown. In the present study, CD40L suppressed P4Hα1 expression in human aortic smooth muscle cells (HASMCs) in a dose- and time-dependent manner, with insignificant changes in MMP2 expression and negative enzymatic activity of MMP9. CD40L increased TRAF6 expression, JNK phosphorylation, NF-κB nuclear translocation as well as DNA binding. Furthermore, silencing TRAF6, JNK or NF-κB genes abolished CD40L-induced suppression of P4Hα1. Lower NF-κB nuclear import rates were observed when JNK or TRAF6 silenced HASMCs were stimulated with CD40L compared to HASMCs with active JNK or TRAF6. Together, these results indicate that CD40L suppresses P4Hα1 expression in HASMCs by activating the TRAF6-JNK- NF-κB pathway. We also found that rosuvastatin inhibits CD40L-induced activation of the TRAF6-JNK- NF-κB pathway, thereby significantly rescuing the CD40L stimulated P4Hα1 inhibition. The results from this study will help find potential targets for stabilizing vulnerable atherosclerotic plaques.


Introduction
Atherosclerotic plaque rupture is a key event contributing to the pathogenesis of acute coronary syndrome. The risk of atherosclerotic plaque rupture depends mostly on plaque composition and vulnerability [1]. Characteristic histomorphological features of vulnerable plaques include a large lipid core and a thin fibrous cap [2]. Type 1 and type 3 collagen synthesized by human aortic smooth muscle cells (HASMCs) are the main components of the extracellular matrix (ECM) in arterial wall/plaque. A major portion of the fibrous cap is made up of these collagens that provide strength and integrity to the fibrous cap as well as maintain plaque stability [3]. Prolyl-4-hydroxylase (P4H) is a key enzyme in collagen biosynthesis, where the subunit P4Hα1 catalyzes the posttranslational processing of collagen synthesis in most cell types and tissues [4]. Matrix metalloproteinases (MMPs) are a family of enzymes that degrade ECM components of the atherosclerotic plaque, thereby inducing plaque instability. MMPs are mainly produced by macrophages and HASMCs of the atherosclerotic plaque, and are known to play a role in balancing the collagen homeostasis [5][6][7]. MMP2 and MMP9 belong to a subgroup of gelatinases that share similar proteolytic activity, and degrade denatured collagens, gelatins and various ECM components, thereby playing an important role in atherosclerotic plaque rupture [8][9][10].
The CD40 ligand (CD40L) and its receptor, CD40, belong to the tumor necrosis factor (TNF) family and tumor necrosis factor receptor (TNF-R) family, respectively [11]. It is well established that they are involved in immune regulation, inflammation and in plaque instability, and are not restricted to T-and B-lymphocytes [12][13][14][15]. CD40 and CD40L are expressed on the macrophages, endothelial cells and SMCs in the atherosclerotic plaque. [16] The soluble form of CD40L (sCD40L) is an 18-kDa protein comprising the entire TNF homologous region of CD40L. It is generated in vivo by intracellular proteolytic processing of the full length CD40L. Recombinant human soluble CD40 ligand (rhsCD40L) is a 16.3-kDa protein containing 149 amino acid residues and comprises the receptor binding TNF-like domain of CD40L. Previous studies have reported that CD40L upregulates MMPs in the atherosclerotic plaque, which eventually leads to increased collagen degradation and plaque instability [17][18][19]. The present study aims to identify the dual role of CD40L in the regulation of P4Hα1 and MMPs, 2 important proteins involved in collagen homeostasis.
TNF receptor-associated factors (TRAFs) are major signaling mediators downstream of CD40 [20] and particularly, TRAF6 has been shown to play an important role in promoting atherosclerosis [21][22]. CD40-mediated signal transduction differs significantly depending on the cell type, function, origin, differentiation stage and activation status [12,13,23,24]. Mitogen-activated protein kinase (MAPK) pathway is one of the most studied controversial signaling pathway [24]. Extracellular signal-regulated kinase (ERK), C-jun N-terminal kinase (JNK) and p38 MAPK are the 3 key pro-inflammatory MAPKs, and are involved in collagen turnover and the development of atherosclerotic lesions [25]. Nuclear factor kappa-light-chainenhancer of activated B cells (NF-κB) is activated via TRAF-mediated MAPKs and/or by TRAFs itself via CD40 signaling [24]. NF-κB can control the expression of genes that are involved in the initiation and progression of atherosclerosis [26]. The current study examines the relationship between CD40L-stimulated P4Hα1 and MMP expression, as well as the TRAF6-MAPKs-NF-κB signaling pathway in HASMCs. The results from this study will aid in finding new targets for stabilizing vulnerable atherosclerotic plaques.
HMG-CoA reductase inhibitors (statins) are widely prescribed as potent lipid lowering agents. It has been previously demonstrated that statins being pleiotropic can exert cholesterolindependent effects in atherosclerosis by reducing inflammation and stabilizing the plaque [27,28]. Previous studies have found that statins decrease CD40 expression [29,30] and the resulting inflammation [31], and also inhibit ECM degradation of the atherosclerotic plaques [32][33][34]. However, the direct effects of rosuvastatin-a statin-on CD40-induced P4Hα1 expression and collagen production are unknown. We therefore investigated whether rosuvastatin regulates ECM metabolism in CD40L-stimulated HASMCs, and if so, the possible regulatory mechanisms. 1. Cell culture and treatment HASMCs were purchased from American Type Cell Collection (Manassas, VA, USA) and cultured in Smooth Muscle Cell Medium (Cat#1101, ScienCell, Carlsbad, CA, USA) containing 2% fetal bovine serum. Cells were cultured up to the 6 th passage in a humidified incubator at 37°C, 5% CO 2 . Experiments were performed with 80%-90% confluent cells after rendering them quiescent by serum-free starvation for 24 h.
To study the effects of rosuvastatin, HASMCs were pretreated with or without 10 μM rosuvastatin calcium salt (Cat#sc-208316, Santa Cruz) for 2 h before rhsCD40L stimulation (statin was still included in the stimulation medium). This particular dose was selected because it corresponds to the highest therapeutic dose (40 mg) found in the patients' blood treated with rosuvastatin [35].

Quantitative real-time reverse-transcriptase-PCR (qRT-PCR)
Total RNA was extracted from HASMCs using TRIzol (Invitrogen, Carlsbad, CA, USA) following the manufacturer's protocol, and quantified by spectrophotometry. The cDNA was reverse transcribed in a Bio-Rad thermocycler using the RT reagent kit (Cat#RR047A, TaKaRa, Japan). Real-time PCR was then performed with a SYBR Green I Premix kit (Cat#RR820A, TaKaRa) in a 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) under the following conditions: 40 cycles at 95°C for 5 seconds and 60°C for 34 seconds. Sequence-specific primers used were as follows: The relative mRNA expression was normalized to GAPDH and analyzed using the 2 −ΔΔCT method. All the experiments were repeated at least 3 times.

Gelatin zymography
The gelatinolytic activities of MMP-2 and MMP-9 were evaluated by gelatin zymography using a gelatinase activity assay kit (Cat#30071, Genmed Scientific Inc, Shanghai, China). Equal volume of cell culture supernatants were electrophoresed in the SDS gel containing gelatin, and subjected to renaturation (1 h), digestion (20 h), staining (1 h), destaining (30 min) and termination by incubation in the corresponding reagents at room temperature. All the reagents were provided in the kit. The zymograms were photographed on a light box, and the latent or active MMP-2 and MMP-9 activity was detected as a translucent area over a dark blue background. All the experiments were repeated at least 3 times.

Electrophoretic mobility shift assay (EMSA) analysis of NF-κB
Nuclear proteins were isolated from cultured HASMCs treated with or without rhsCD40L using a NucBuster Protein Extraction Kit (Cat#71183, Merck Millipore, Darmstadt, Germany) following the manufacturer's protocol. EMSA was performed with a DIG Gel Shift Kit (Cat#03353591910, Roche, Madison, WI, USA) using the DIG-endlabeling technique to detect sequence-specific NF-κB-binding proteins. Briefly, NF-κB DNA fragment sense strand oligonucleotides (5'-AGTTGAGGGGACTTTCCCAGGC-3') were annealed with their antisense strands and labeled with digoxigenin-11-ddUTP at the 3'-ends. The labeled probes were incubated with the extracted nuclear proteins, which allowed the formation of DNA-protein complexes. The mixture was run on a 6% native polyacrylamide gel at 80 V and blotted on a positively charged nylon membrane (Cat#INYC00010, Millipore). The membrane was UVcross linked, blocked, and then incubated with anti-Digoxigenin-AP. Chemiluminescent detection was performed with CSPD and recorded on an imaging device. All the experiments were repeated at least 3 times.

Statistical analysis
Statistical analysis was performed using the SPSS software (version 20.0, SPSS, Chicago, IL, USA) for Windows. Data was presented as Mean ± SD. The differences between 2 groups were analyzed by independent-samples t test. For more than 2 groups, the differences were calculated by one-way ANOVA with LSD post hoc test. All the experiments were repeated at least 3 times. P < 0.05 was considered statistically significant. To study the dose-dependent effects of CD40L stimulation on P4Hα1 expression, HASMCs were treated with 0, 2, 5, and 10 μg/ml rhsCD40L for 8 h, following which the RNA and protein were extracted for qRT-PCR analysis and western blotting, respectively. The analysis revealed that CD40L significantly suppressed both the P4Hα1 mRNA ( Fig 1A) and protein expression ( Fig 1B). The suppression was strongest at 5 μg/ml, and the P4Hα1 protein expression corresponded with the inhibition of type 1 and 3 collagens ( Fig 1B). For the time-dependent study, HASMCs were treated with 5 μg/ml rhsCD40L for 0, 4, 8 and 12 h. The data revealed that CD40L downregulated P4Hα1 mRNA and protein expression, where P4Hα1 reached a plateau after 8 hours of stimulation (Fig 1C and 1D).

Effect of CD40L and rosuvastatin on the expression of P4Hα1 and gelatinases in HASMCs
Additional studies were performed to analyze the effect of CD40L stimulation on the expression of MMP2 and MMP9 using qRT-PCR and gelatin zymography. The results indicated that CD40L does not affect the MMP2 mRNA expression (Fig 1A and 1C) and MMP2 enzymatic activity (Fig 1E and 1F) at any of the indicated doses and time points, and it had no effects on MMP9 enzymatic activity.
1.2. Rosuvastatin attenuates the suppressive effect of CD40L on P4Hα1 expression in HASMCs. In order to investigate the effects of rosuvastatin, HASMCs were pretreated with or without 10 μM rosuvastatin for 2 h, and then stimulated with 5 μg/ml rhsCD40L for additional 8 h. The analysis revealed that pretreatment with rosuvastatin followed by CD40L stimulation significantly rescued the CD40L stimulated P4Hα1 mRNA inhibition ( Fig 1G) and protein expression with a concomitant increase in both type 1 and 3 collagen (Fig 1H). Moreover, CD40L-stimulated HASMCs expressed significantly lower P4Hα1 and collagen (both type 1 and 3) compared to control cells, which is consistent with results mentioned above. Interestingly, rosuvastatin treatment alone had no effect on the expression of P4Hα1 or collagen (Fig 1G and 1H).
2. NF-κB regulates the effect of rosuvastatin's rescue of the CD40L stimulated P4Hα1 inhibition in CD40L-stimulated HASMCs 2.1. CD40L activates NF-κB in HASMCs. Time-dependent nuclear translocation of NF-κB was demonstrated by confocal microscopy after stimulation of HASMCs with 5μg/ml rhsCD40L for 0, 15, 30, 60, and 120 min. Under normal growth conditions, most unstimulated cells showed diffused staining in the cytoplasm and a weak staining in the nucleus. After stimulation with 5 μg/ml rhsCD40L for 15 min, a considerable fraction of NF-κB accumulated in the nucleus. NF-κB is maximally concentrated in the nucleus at 30 min after rhsCD40L exposure (Fig 2A). EMSA was performed to demonstrate the time course for nuclear NF-kB mobilization. The analysis revealed that CD40L significantly upregulated the NF-κB binding activity in HASMCs compared to the control cells, where it reached a peak at 30 min ( Fig 2B). Collectively, the data indicated that 5 μg/ml CD40L is enough to strongly activate NF-κB at 30 min post-stimulation.
2.2. Rosuvastatin attenuates the suppressive effect of CD40L on P4Hα1 expression via NF-κB. To study the downstream signaling effects of rosuvastatin, HASMCs were transfected with or without 100 nM NF-κB siRNAs for 48 h, and then pretreated with 10 μM rosuvastatin for 2 h. Following this, the cells were stimulated with 5 μg/ml rhsCD40L for 8 h. The results demonstrated that the transfected HASMCs had significantly higher P4Hα1 mRNA and protein expression compared with the non-transfected HASMCs. The NF-κB gene silenced HASMCs that had been pretreated with rosuvastatin lead to a further increase in expression of P4Hα1 (Fig 2C and 2D). The effects of CD40L or/and rosuvastatin on P4Hα1 expression are consistent with the results shown in Fig 1. However, the expression of P4Hα1 was not influenced by NF-κB siRNAs transfection alone. To measure the transfection efficiency, the expression levels of NF-κB protein in the transfected group compared with the control group were used (Fig 2E).
In addition, the NF-κB nuclear import rate was significantly lower in CD40L stimulated HASMCs pretreated with rosuvastatin than in the untreated CD40L-stimulated HASMCs ( Fig  2F). Rosuvastatin itself had no significant effect on the nuclear import rate of NF-κB (S13).
3. JNK regulates the effect of rosuvastatin's rescue of the CD40L stimulated P4Hα1 inhibition in CD40L-stimulated HASMCs 3.1. CD40L induces the phosphorylation of JNK upstream of NF-κB. Western blot analysis was performed to elucidate the effect of CD40L on the phosphorylation status of several proteins. The results demonstrated that CD40L significantly induced the phosphorylation of JNK in HASMCs compared with the control cells, with the phosphorylation peaking at 15 min and then decreasing gradually ( Fig 3A). Interestingly, CD40L had no effect on the levels of phosphorylated ERK1/2 and p38 MAPK (Fig 3B and 3C).
The NF-κB nuclear import rate was significantly lower in JNK siRNAs transfected CD40Lstimulated HASMCs compared with non-transfected CD40L-stimulated HASMCs ( Fig 2F). The nuclear import rates were not influenced by ERK1/2 and p38MAPK siRNAs transfections (S13). Also, no significant changes were observed in the NF-κB nuclear import rate in HASMCs transfected with JNK, ERK1/2 or p38MAPK siRNAs alone (S13).
3.2. Rosuvastatin attenuates the suppressive effect of CD40L on P4Hα1 expression via the JNK pathway. Based on the results obtained above, we further questioned whether treatment with rosuvastatin affected P4Hα1 expression via the JNK pathway. JNK siRNA transfected HASMCs were treated as described previously. The results demonstrated that the P4Hα1 mRNA ( Fig 3D) and protein expression (Fig 3E) were significantly higher in the transfected cells stimulated with rhsCD40L compared with non-transfected CD40L stimulated HASMCs. The expression of P4Hα1 increased further when the transfected HASMCs were pretreated with rosuvastatin and then stimulated with CD40L. The effects of CD40L and/or rosuvastatin on P4Hα1 expression are consistent with the results shown in Fig 1. P4Hα1 expression was not influenced by JNK siRNAs transfection alone (Fig 3D and 3E). The expression levels of JNK, ERK1/2 and p38 MAPK protein in the transfected group compared with the control group were used to measure the transfection efficiencies (Fig 3F-3H).
Western blot analysis to quantify the levels of phosphorylated JNK protein revealed that the levels were significantly lower in HASMCs pretreated with rosuvastatin prior to stimulation with the CD40L compared with untreated CD40L stimulated HASMCs. However, rosuvastatin itself had no significant effect on the phosphorylation levels of JNK (Fig 3I).

TRAF6 regulates the effect of rosuvastatin's rescue of the CD40L stimulated P4Hα1 inhibition in CD40L-stimulated HASMCs
4.1. CD40L activates TRAF6 upstream of JNK and/or NF-κB. In order to determine the downstream signaling effects of TRAF6, HASMCs were stimulated with CD40L for 5 min, and the levels of TRAF6 mRNA and protein were detected by qRT-PCR, Western blot and immunofluorescence. The data indicated that CD40L stimulated HASMCs had significantly higher levels of TRAF6 mRNA ( Fig 4A) and protein (Fig 4B and 4C) compared with the unstimulated control cells. In addition, we questioned whether TRAF6 regulated the NF-κB nuclear import rate in CD40-stimulated HASMCs. We therefore transfected the HASMCs with 100 nM TRAF6 siRNAs for 48 h prior to stimulation with 5 μg/ml rhsCD40L for 8 h. The analysis revealed that the transfected cells had significantly lower NF-κB nuclear import rate compared with non-transfected HASMCs, both stimulated with CD40L ( Fig 2F). No significant changes were observed in the NF-κB nuclear import rate in HASMCs transfected with TRAF6 siRNAs alone (S13).
4.2. Rosuvastatin attenuates the suppressive effect of CD40L on P4Hα1 expression via TRAF6. P4Hα1 mRNA (Fig 4D) and protein expression (Fig 4E)   Interestingly, P4Hα1 expression was not influenced by TRAF6 siRNAs transfection alone ( Fig  4D and 4E). To measure the transfection efficiency, the expression levels of TRAF6 protein in the transfected group compared with the control group were used (Fig 4F).
TRAF6 expression was significantly attenuated in HASMCs pretreated with rosuvastatin before being stimulated with CD40L compared with CD40L stimulated cells alone. Rosuvastatin itself had no significant effect on the expression of TRAF6 (Fig 4A-4C).

Discussion
In the current study, we found that (1) HASMCs stimulation of with the CD40L inhibits the expression of P4Hα1 via activation of the TRAF6-JNK-NF-κB pathways, (2) Cultured HASMCs take no role in the expression changes of MMP2 and MMP9 under a P4Hα1-affected CD40L stimulation range, and (3) treatment with rosuvastatin attenuated the CD40L-mediated suppressive effect on the expression of P4Hα1 via inhibition of the CD40L-activated TRAF6-JNK-NF-κB pathways. This study therefore, contributes towards understanding the direct effects of CD40L stimulation and/or rosuvastatin on P4Hα1 expression in HASMCs, and the underlying signaling mechanisms, which may provide evidence for collagen homeostasis in atherosclerotic plaque ECM.
The collagen-rich fibrous cap that covers the lipid core, and its thickness are crucial to the stability of an atherosclerotic plaque, whereas P4Hα1 and MMPs play important roles in the production and degradation of a fibrous cap, respectively [1][2][3][4][5][6][7]. Atherosclerosis is believed to be a chronic inflammatory-immune disease, as a variety of cytokines participate in the pathogenesis of atherosclerosis. Special emphasis is placed on the role of pro-and anti-inflammatory cytokines in pathogenic and regulatory immunity [36]. Several in vitro studies have shown that TNF-α, IL-6 and TGF-β1 affect P4Hα1 expression in the ECM of an atherosclerotic plaque [37][38][39]. CD40L is a protein that is involved in the pathogenesis of atherosclerosis, and is known to actively regulate ECM metabolism [12][13][14][15][16]. Previous studies have mainly focused on CD40L-mediated ECM degradation via upregulated MMPs [17][18][19]. In the present study, we found that CD40L dose-and time-dependently reduced P4Hα1 mRNA and protein expression as well as type 1 and 3 collagen production in HASMCs. The data indicate that CD40L inhibits collagen synthesis via suppression of P4Hα1, which simultaneously increases ECM degradation and decreases its production, thereby contributing to atherosclerotic plaque rupture. Moreover, CD40L stimulation of the HASMCs did not significantly alter the expression of MMP2 and the negative enzymatic activity of MMP9, suggesting that CD40L exerts its affect independent of these matrix degrading enzymes in HASMs, at least within the range that reduced P4Hα1 expression.
Statins are cholesterol lowering drugs that have been widely prescribed in cardiovascular diseases however, over the years, several studies have reported about its pleiotropic, cholesterol-independent effects. [27,28] Additionally, it has shown that statins exhibit the cholesterol-independent effect by decreasing the inflammation associated with atherosclerosis [31,32], and also by stabilizing the plaques [33,34]. Rosuvastatin, a member of new generation of statins, has also been shown to display similar effects [40,41]. Interestingly, as with CD40L, most of the previous studies on statins mainly focused on the inhibition of matrix degradation, and therefore, the effects on matrix production remained poorly understood. To the best of our knowledge, the current study is the first to demonstrate that rosuvastatin attenuates CD40Linduced suppressive effect on P4Hα1 mRNA and protein expression, as well as the resulting collagen protein expression. However, rosuvastatin treatment without CD40L stimulation had no effect on the expression of P4Hα1 and collagen in HASMCs. These results are consistent with a previous study showing that P4Hα1 and collagen expression did not change with statin treatment alone in cultured HASMCs [42].
The CD40 signal transduction pathway differs significantly with cell type, function, origin, stage of differentiation and activation status [11,12,23,24]. The canonical TRAFs-MAPKs-NF-κB pathway mediates various proatherogenic processes in atheroma-associated cell types [11,12,23]. NF-κB controls the expression of genes directing the initiation, progression and resolution of atherosclerotic plaque, including genes for proatherogenic cytokines such as TNF-α, IL-1β, IL-6, chemokine MCP-1, adhesion molecule ICAM-1 and MMPs, where few of these factors persistently maintain NF-κB in an activated state [26,[43][44][45][46]. MAPKs play important roles in the pathogenesis of atherosclerosis, proinflammation and collagen turnover. 25 Previous studies have shown that JNK and ERK1/2 mediate the suppressive effects of TNFα and IL-6 on P4Hα1 expression in HASMCs [38,39]. TNFα shares structural homologies with CD40L, and they both belong to the TNF superfamily. Furthermore, Zhang K et al. demonstrated that oxidized-low density lipoprotein can inhibit the expression of P4Hα1 in HASMCs however, treatment with statins can abrogate this effect leading to increased P4Hα1 via p38 MAPK and ERK1/2 signaling pathway [42]. TRAF6 is a relatively independent CD40-binding protein compared with TRAF1/2/3/4/5, and CD40-TRAF6 interactions are capable of activating the canonical NF-κB pathway [23]. In the present study, we found that CD40L upregulated TRAF6 expression, JNK phosphorylation, and NF-κB nuclear translocation, as well as DNA binding, suggesting that the suppressive effect of CD40L on P4Hα1 and collagen expression in HASMCs is mediated by TRAF6, JNK and NF-κB. These finding were further confirmed by transfection studies, where the suppressive effect of CD40L on P4Hα1 expression was abolished by silencing TRAF6, JNK and NF-κB genes. Several studies have shown NF-κB to be a transcription factor for CD40, and therefore, the NF-κB nuclear import rate was measured to confirm the involvement of this pathway [24]. We found that JNK or TRAF6 siRNA transfected HASMCs when stimulated with CD40L displayed lower NF-κB nuclear import rate compared with non-transfected CD40L-stimulated HASMCs. Most importantly, we found that rosuvastatin treatment attenuated the suppressive effect of CD40L on P4Hα1 and collagen expression via inhibition of CD40L activated TRAF6-JNK-NF-κB pathways.