Rosiglitzone Suppresses Angiotensin II-Induced Production of KLF5 and Cell Proliferation in Rat Vascular Smooth Muscle Cells

Krüppel-like factor (KLF) 5, which initiates vascular smooth muscle cell (VSMC) proliferation, also participates in Angiotensin (Ang) II-induced vascular remodeling. The protective effect of rosiglitazone on vascular remodeling may be due to their impact on VSMC proliferation. However, the underlying mechanisms involved remain unclear. This study was designed to investigate whether the antiproliferation effects of rosiglitazone are mediated by regulating Ang II/KLF5 response. We found that, in aortas of Ang II-infused rats, vascular remodeling and KLF5 expression were markedly increased, and its target gene cyclin D1 was overexpressed. Co-treatment with rosiglitazone diminished these changes. In growth-arrested VSMCs, PPAR-γ agonists (rosiglitazone and 15d-PGJ2) dose-dependently inhibited Ang II-induced cell proliferation and expression of KLF5 and cyclin D1. Moreover, these effects were attenuated by the PPAR-γ antagonists GW9662, bisphenol A diglycidyl ether and PPAR-γ specific siRNA. Furthermore, rosiglitazone inhibited Ang II-induced phosphorylation of protein kinase C (PKC) ζ and extracellular signal-regulated kinase (ERK) 1/2 and activation of early growth response protein (Egr). In conclusion, in Ang II-stimulated VSMCs, rosiglitazone might have an antiproliferative effect through mechanisms that include reducing KLF5 expression, and a crosstalk between PPAR-γ and PKCζ/ERK1/2/Egr may be involved in. These findings not only provide a previously unrecognized mechanism by which PPAR-γ agonists inhibit VSMC proliferation, but also document a novel evidence for the beneficial vascular effect of PPAR-γ activation.


Introduction
Vascular remodeling is closely involved in the progression of atherosclerosis and restenosis, and is also present in hypertension-and diabetes-induced vascular complications [1]. and DharmaFECT 2 transfection reagent (T-2002-02) were obtained from Dharmacon (Lafayette, CO, USA). Cell Cycle Phase Determination Kit was from Cayman Chemical (USA). NE-PER Nuclear and Cytoplasmic Extraction kit was from Pierce (IL, USA), TRIzol kit and Su-perScript III Platinum SYBR-Green Two-Step qRT-PCR kit were provided by Invitrogen Corp. (Carlsbad, CA,USA). DNA-free kit was from Ambion (Austin TX, USA). Agarose gels were from Spanish Biochemicals (Pronadisa, Madrid, Spain). Reagents for the enhanced chemiluminescence were from Pierce Corp. (Rockford, IL, USA).

In vivo experiments
The study was approved by the Institutional Animal Research and Ethics Committee of Xi'an Jiaotong University (SCXK2007-001) and was conducted in accordance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85-23, revised 1996). Male Sprague-Dawley rats (200-220 g) were divided into 4 groups for treatment (n = 8): control, Ang II, Ang II+rosiglitazone and rosiglitazone alone. Rats were anesthetized with methoxyflurane, and then osmotic minipumps (model 2004, Durect Corp., Cupertino, CA) were inserted subcutaneously to deliver Ang II (150 ng/kg/min) or the same volume of 0.9% saline for controls for 28 days. Rosiglitazone was administrated (5 mg/kg/d in drinking water) for 29 days, starting the day before Ang II infusion. Systolic blood pressure was measured by the tail-cuff method. The animals were killed by injecting an excess amount of pentobarbital. One portion of the aorta was dissected and cleaned of fat, then frozen in liquid nitrogen for RNA extraction. Another portion was fixed in 4% formaldehyde solution and embedded in paraffin for immunohistochemical analysis.

Vascular morphology
Cross-sections of thoracic aorta segments collected at the time of sacrifice were paraffin-embedded and stained with alizarin blue. Morphometric assessment was performed using the Qwin 550 quantitative image analysis system (Leica, German). Equally spaced (every 45°) measurements of lumen diameter (4 measurements) and wall thickness (8 measurements) were made. The averaged wall thickness was divided by the averaged lumen diameter to calculate a final wall/lumen ratio.

Immunohistochemistry
Primary antibodies against KLF5 (1:400) and cyclin D1 (1:250) were added to paraffinembedded sections of rat thoracic aortas and incubated overnight at 4°C. Biotinylated and affinity-purified IgG (Zymed, USA) was used as a secondary antibody and incubated for 1 hr at 37°C. A streptavidin-enzyme conjugate was sequentially added for 20 min and incubated with substrate 3', 3' -diaminobenzidine (DAB), followed by haematoxylin nuclear counterstaining. Negative controls were without the primary antibody. Quantification results detected by Leica QWin550CW Image Acquireing & Analysis System (Leica, German) are presented as gray scale levels.

Cell culture
VSMCs were obtained from thoracic aortas of Sprague-Dawley rats as previously described [25] and cultured in DMEM containing 10% FBS, penicillin (100 U/ml) and streptomycin (100 U/ml). VSMCs between passages 3 to 7, confirmed positive (99%) for smooth muscle αactin immunostaining, were used in the experiments. For subsequent experiments, cells at 80% confluence in culture wells were growth-arrested by serum starvation for 48 hrs.

Cell proliferation assay
Cell proliferation was determined using the cell counting kit-8 (CCK-8, Dojindo, JAPAN) according to the manufacturer's protocol. Briefly, 2500 cells/well were plated in a 96-well plate. After treatment, 10 μl of CCK-8 solution was added to each well and incubated for 3 hrs. The cell viability in each well was determined by reading the optical density at 450nm.

Small-interfering RNA
The VSMCs (5×10 6 ) were seeded into six-well plates and were grown until 60-80% confluent. The cells were transiently transfected with 150 pM of PPAR-γ siRNA, KLF5 siRNA, PKCz siRNA or NC siRNA and using DharmaFECT 2 transfection reagent according to the manufacturer's instructions. After 48 hrs, PPAR-γ mRNA, KLF5 mRNA and PKCz mRNA levels were detected by quantitative real time-PCR respectively. Knock down efficiency of PPAR-γ, KLF5 and PKCz were showed in S1

Real-time RT-PCR
Total RNA was extracted by use of TRIzol reagent, and DNA was removed by use of the DNAfree kit (Ambion, Austin TX, USA). Real-time qRT-PCR with SYBR involved use of the Super-Script III Platinum Two-Step qRT-PCR Kit on an ABI PRISM 7000 sequence detection PCR system (Applied Biosystems, Foster City, CA, USA) according to the manufacturers' instructions. Primers were designed with use of Beacon Designer v4.0 (Premier Biosoft, USA; Table 1). Results are expressed as fold difference in gene expression to that of GAPDH by the 2 -ΔΔCT method [26]. To validate our real-time PCR protocol, gene-specific standard curves for each gene and GAPDH were generated from serial 10-time dilutions of the cDNA. Expression slopes of each gene were similar to that of GAPDH. A melting-curve analysis was also performed to check for the absence of primer dimers.

Gene
Primer sequence Accession No of bands were scanned and quantified with the Gel Doc 2000 system (Bio-Rad). Results are expressed as fold increase as compared with that of the control.

Immunofluorescene
Cells were fixed with 4% paraformaldehyde, permeablized with 0.3% Triton-X 100, and incubated with primary antibody at 4°C overnight, and then subjected to Cy3 labeled secondary antibody for 1 hr at room temperature. Visualization was performed by using Zeiss LSM 510 laser scanning confocal microscope.

DNA-binding assay
Nuclear proteins were extracted by use of the NE-PER kit. DNA-binding activity of PPAR-γ and early growth response (Egr) transcription factor was detected by an ELISA-based method with PPAR-γ (Cayman Chemical, USA) and Egr (Genlantis, USA) transcription assay kits according to the manufacturers' instructions. Briefly, 10 μg nuclear protein was added to the 96-well plate pre-coated with PPAR or Egr specific double-strand DNA containing the sequence for peroxisome proliferators-response element (PPRE) or Egr response element, and then incubated over night at 4°C. Bound PPAR-γ or Egr was detected by the specific PPAR-γ or Egr antibody. A horseradish peroxidase-conjugated secondary antibody was then added for colorimetric reading.

Statistical analysis
Results are expressed as mean ± s.e.m. Statistical significance between groups was assessed by one-way ANOVA, followed by post-hoc Duncan multiple comparisons, with use of SPSS v11.5 (SPSS Inc., Chicago, IL). A P <0.05 was considered statistically significant.

Systolic blood pressure, body weight and vascular morphology changes in Ang II-infused rats
After the 28-day treatment, body weights were similar in all groups ( Table 2). Ang II infusion induced a substantial increase in systolic blood pressure, (p<0.05 versus Control), which was attenuated by rosiglitazone (p<0.05 versus Ang II and control). The

Rosiglitazone attenuated KLF5 and cyclin D1 expression in Ang IIinfused rats
In rats infused with Ang II for 28 days, KLF5 mRNA expression was induced in the aortas, and this change was significantly inhibited by rosiglitazone treatment (Fig 1A). To define the cellular localization and protein expression of KLF5 after the appropriate treatment, we examined the expression of KLF5 protein by immunohistochemistry within the aortas (Fig 1B). Robust KLF5 expression was observed in the aortic VSMCs of Ang II-treated animals, which was significantly decreased by rosiglitazone treatment. Rosiglitazone alone had no effect on KLF5 expression. We then evaluated whether rosiglitazone diminished VSMC proliferation in parallel with KLF5 downregulation by measuring cyclin D1 expression in the aorta. Infusion of Ang II increased the expression of cyclin D1 within the thoracic aorta of Ang II-infused rats (Fig 1A and  1B). Rosiglitazone significantly inhibited this Ang II-mediated effect. Expression and activation of PPAR-γ in the thoracic aorta were identified with real-time RT-PCR and DNA binding assay. As show in (Fig 1C). The increase of PPAR-γ mRNA expression and activation was observed after rosiglitazone treatment.

PPAR-γ agonists inhibited Ang II-induced KLF5 and cyclin D1 expression in VSMCs
Results from real-time reverse transcription-PCR analysis showed that both rosiglitazone and 15d-PGJ2 substantially reduced the increased levels of KLF5 mRNA in response to Ang II in a dose-dependent manner ( Fig 3A) (P<0.05 versus Ang II). Notably, rosiglitazone and 15-d-PGJ2 had no significant effect on the basal expression of KLF5 in VSMCs (P>0.05 versus control). Similar results were found by using immunofluorescene (Fig 3B) and western blot analysis for detecting of total ( Fig 3C) and nuclear (S3 Fig) KLF5 protein in VSMCs. In addition, as assessed by real-time RT-PCR, rosiglitazone and 15d-PGJ2 also dose dependently inhibited Ang II-induced cyclin D1 mRNA expression, which was well known as a target gene of KLF5 ( Fig 3D).

Relationship between the effect of rosiglitazone on Ang II-induced cell proliferation and KLF5 in VSMCs
As mentioned above, rosiglitazone is able to inhibit Ang II-induced cell proliferation and KLF5 expression in VSMCs in vivo and in vitro. To confirm whether rosiglitazone suppresses Ang II-induced proliferation responses via KLF5, VSMCs were pretreated with KLF5 siRNA for 48 hrs prior to the addition of rosiglitazone (5 μM) for 1 hr, and subsequently stimulated with Ang II (0.1 μM) for 24 hr. As shown in Fig 4, When compared with the control, stimulating the cells with Ang II induced cell proliferation (P<0.05 versus control), whereas the KLF5 siRNA and rosiglitazone partially reversed this effect in VSMCs (P<0.05 versus Ang II). Moreover, rosiglitazone did not further reverse the effect of Ang II induced cell proliferation when

Suppression of KLF5 expression is mediated by PPAR-γ activation
As shown in Fig 5A, Ang II treatment for 24 hrs elicited a trend to an induction of PPARγ mRNA expression, although it did not achieve statistical significance (P>0.05 versus control). Moreover, Ang II significantly decreased PPAR-γ DNA binding activity as compared with the control (P<0.05; Fig 5B). Exposure to rosiglitazone and 15d-PGJ2 for 24 hrs resulted in a significant increase in the DNA-binding activity to PPRE. Moreover, pretreatment with rosiglitazone or 15d-PGJ2 substantially increased PPAR-γ DNA-binding activity in comparison with Ang II treatment (P<0.05 versus Ang II).
As shown in Fig 5C and 5D, western blot and real time RT-PCR analysis showed that rosiglitazone, 15d-PGJ2 and pioglitazone were all able to suppress Ang II induced KLF5 mRNA and protein expression. In contrast, Pretreated cells with PPAR-γ antagonists (GW9662 and BADGE) or PPAR-γ specific siRNA diminished these effects (Fig 5C and 5D). These results clearly indicated that the effect of PPAR-γ agonists on Ang II-induced KLF5 expression is mediated at least in part by PPAR-γ activation.

Effect of rosiglitazone on ANG II receptors and signaling in regulating KLF5 expression
To clarify whether blockade of the Ang II receptor signaling was involved in the inhibitory effect of rosiglitazone on Ang II-induced KLF5 expression in VSMCs, VSMCs were subjected to losartan (1 μM), PD123319 (10 μM) for 30min, followed by treatment of rosiglitazone (5 μM) for further 1 hr, and finally stimulated with Ang II (0.1 μM) for 24 hrs. As seen in Fig 6, AT 1 antagonist losartan inhibited Ang II-induced KLF5 expression in VSMCs, whereas the AT2 antagonist PD123319 had no effect (data not shown). Although rosiglitazone alone also inhibited Ang II-induced KLF5 expression in VSMCs, it did not have any additional effects to AT 1  blockades, suggesting that the inhibitory effects of PPAR-γ agonist on Ang II stimulatory events might be downstream of the AT 1 receptor.
We then explored the regulation of AT 1 R/PKC signaling in the inhibitory effect of rosiglitazone on Ang II-induced KLF5 expression in VSMCs. As shown in Fig 7A and 7B, phosphorylation of both PKCε and PKCz was induced by Ang II (0.1 μM) treatment for 30 min, whereas pretreatment of cells with rosiglitazone (5 μM) abrogated only PKCz phosphorylation (by more than 80%). Moreover, as seen in Fig 7, PKCz specific siRNA significantly diminished Ang II induced KLF5 expression (P<0.05 vs. Ang II) and rosiglitazone treatment potentiated the inhibitory effect of the PKCz RNAi on KLF5 expression. Taken together, these results indicate that the inhibitory effect of PPAR-γ agonist on Ang II-induced KLF5 expression in VSMCs is at least part from Ang II/ PKCz pathway.
Finally, we observed whether the blockade of ERK1/2 was required for the inhibitory effect of rosiglitazone on Ang II-induced KLF5 expression (Fig 8). VSMCs were pretreated with the specific ERK1/2 inhibitor PD98059 (1 μM) for 30min before the addition of rosiglitazone (5 μM) for 1 hr, and then stimulated with Ang II (0.1 μM), As illustrated in Fig 8A, Erk1/2 inhibitor PD98059 significantly attenuated Ang II induced KLF5 expression, whereas, rosiglitazone did not have any additional effects on ERK1/2 blockade, which suggests that Ang II/ ERK1/2 signaling may be involved in the inhibitory effect of PPAR-γ agonist on Ang IIinduced KLF5 expression. Furthermore, as shown in Fig 8B, ERK1/2 phosphorylation induced by Ang II (0.1 μM) treatment for 30 min was significant impaired by rosiglitazone. These results suggest that Ang II/ERK1/2 signaling might also contribute to the inhibitory effect of rosiglitazone on Ang II-induced KLF5 expression in VSMCs.

Inhibitory effect of rosiglitazone on EGR transcription activity
To evaluate whether KLF5 expression inhibition by rosiglitazone is mediated by the downregulation of Egr activation, we tested the effect of rosilgitazone on Egr DNA-binding activity. As shown in Fig 9, rosiglitazone significantly abolished Ang II-induced Egr DNA-binding activity (P<0.05 versus Ang II).

Discussion
PPAR-γ activation by its agonists such as TZDs, has been demonstrated previously to inhibit VSMC proliferation and prevent vascular remodeling in hypertension, restenosis, and atherosclerosis in both early clinical trials and animal experiments [21][22][23][24]27]. Understanding the molecular mechanisms responsible for their beneficial efficacy in cardiovascular disease provides   an important basis for the future development of PPAR-γ agonist in the treatment of vascular diseases. A considerable body of evidence indicates the pivotal role of KLF5 in cell proliferation, and inhibition of KLF5 expression has been demonstrated recently to inhibit neointima formation, atherosclerosis and hypertension related vascular remodeling [11,28]. In the present study, we investigated the regulation of KLF5 expression by the PPAR-γ agonist such as rosiglitazone, demonstrated that rosiglitazone-induced activation of PPAR-γ suppresses Ang II-induced KLF5 expression, likely by interfering with the Ang II/PKCz/ERK1/2 pathway, and subsequently downregulated Egr transcriptional function (Fig 10). These findings demonstrate a previously unrecognized mechanism by which PPAR-γ agonists inhibit VSMC proliferation and document a novel evidence for the beneficial vascular effect of PPAR-γ activation.
In VSMCs, as the major target for Ang II in vascular remodeling, KLF5 is critically involved in the pathogenesis of cell proliferation and migration [11] and therefore may have a fundamentally significant contribution to the pathophysiological relationship between vascular remodeling and disorders [7]. Furthermore, our recent reports [10] and those of other researchers have shown that KLF5 in VSMCs is also induced by Ang II and plays a crucial role in Ang II-induced VSMC proliferation [9,12,29]. Consistent with the previous report, we observed a significant increase of cell proliferation and KLF5 induction by Ang II in VSMCs both in vivo and in vitro. Importantly, the present study provides the first evidence that the PPAR-γ agonist suppresses Ang II-induced KLF5 expression in VSMCs. Moreover, rosiglitazone did not further inhibit Ang II-induced cell proliferation in VSMCs when pretreatment of cells with KLF5 siRNA, thus suggesting that activation of PPAR-γ may intervene in the impact of Ang II on the KLF5 to inhibit VSMC proliferation.
Although TZDs have been reported to exert their effects through PPAR-γ-dependent andindependent mechanisms, the observed inhibition of Ang II-induced KLF5 expression we found is likely mediated by the activation of PPAR-γ, as demonstrated by several lines of evidence. Firstly, we confirmed previous data that Ang II decreased DNA-binding activity of PPAR-γ to PPRE, whereas pretreating cells with rosiglitazone and 15d-PGJ2 significantly upregulated PPAR-γ expression and the DNA-binding activity. Secondly, both a different synthetic TZD PPAR-γ agonist and a natural agonist, 15d-PGJ2, mimicked the effect of the PPAR-γ activation on KLF5 expression with equivalent inhibitory potency. Finally, we tested both the specific pharmacological inhibitors of PPAR-γ (BADGE and GW966) and the PPAR-γ specific siRNA on KLF5 suppression by PPAR-γ agonist. The findings that PPAR-γ inhibitions by BADGE, GW9662 and RNAi overcame KLF5 suppression by PPAR-γ agonist demonstrate an essential involvement of PPAR-γ in this effect.
Ang II acts through its binding to a specific AT 1 receptor that regulates KLF5 expression. Previous studies demonstrated that PPAR-γ activation might transcriptionally regulate AT 1 expression [30]. We showed that rosiglitazone cotreatment had little effect on AT 1 receptor mRNA expression in Ang II-challenged VSMCs, which indicates that this mechanism may not be involved in the inhibitory effect of PPAR-γ agonists on Ang II-induced KLF5 expression. Our finding is in accordance with those of Takeda [31] and Sugawara [32], who demonstrated that 24-hr treatment with rosiglitazone had the little effect on AT 1 expression in VSMCs as compared with untreated cells.
Several studies have demonstrated distinct roles of PKC isoforms in regulating cell proliferation. PKCα and -δ activation may inhibit cell proliferation [33,34]; however, PKCz and -ε were reported to induce VSMC proliferation in response to many simulators, including Ang II [35,36]. In light of previous research [9,29], Ang II-induced KLF5 expression in VSMCs is PKC dependent, so we explored the possibility whether PPAR-γ could inhibit Ang II-induced KLF5 induction by suppressing PKC activity. As indicated by their phosphorylation, both PKCz and -ε were indeed activated by Ang II, but rosilgitazone inhibited only the PKCz activation. Moreover, siRNA PKCz significantly downregulated Ang II induced KLF5 expression in VSMC. Therefore, the decrease in Ang II-induced KLF5 expression and cell proliferation by PPAR-γ agonists we observed is, at least in part, by interfering with the switch of PKCz activation in the Ang II signaling pathway, and other PKC-independent pathway may also be involved. The interaction between the PPAR-γ and PKC pathway was also demonstrated in several other studies. PPAR-γ activated by TZDs inhibited glucose-induced activation of diacylglycerol-PKC signaling pathway in endothelial cells [37]. In PDGF-stimulated VSMCs, rosiglitazone and troglitazone significantly inhibited PKCδ activation [38]. Indeed a recent study suggested that rosiglitazone may negatively modulate high-glucose-induced signaling through PKCz in mesangial cells [39]. ERK1/2 is downstream of the PKCz signal pathway in regulating VSMC proliferation in response to Ang II [36] and is implicated in vascular diseases [40]. Our study demonstrated that PPAR-γ agonist markedly diminished Ang II-induced ERK1/2 phosphorylation. Blockade of ERK1/2 by PD98059 significantly diminished Ang II induced KLF5 expression. Moreover, rosiglitazone did not have any additional effects on KLF5 expression when ERK1/2 was blocked, which suggests that ERK1/2 signaling, at least in part, may be involved in the inhibitory effect of PPAR-γ agonist on Ang II-induced KLF5 expression in VSMCs. In accordance with our study, previous studies have demonstrated that PPAR-γ activation may inhibit Ang II induced ERK1/2 activation in endothelial cells [41,42] and VSMCs [43,44]. Recently, Kim et al. has showed the similar result that rosiglitazone might attenuate VSMCs proliferation at least by inhibit ERK1/2 activation and other cell signaling such as, mTOR-p70S6K and -4EBP1 systems [45]. It has been demonstrated that Egr is present in the promoter of KLF5 genes [43], and PPARγ activation may inhibit Egr activation and expression [46,47]. In this study we demonstrated that rosiglitazone inhibited Ang II-induced Egr DNA binding activity, which indicated that PPAR-γ agonist might regulate KLF5 expression by inhibiting Egr transcriptional activation. Taken together, our results indicated that AT1/PKCz/ERK1/2/Egr signal pathways might implicate in the suppression of KLF5 expression and cell proliferation in VSMCs in response to rosilgitzone.
In summary, our findings provide evidence for the beneficial effects of the PPAR-γ agonist to counter-regulate KLF5 expression induced by Ang II. More importantly, anti-proliferation of PPAR-γ activation, by interfering with the Ang II /KLF5 signaling pathway, works in concert to protect against vascular remodeling.