The FASEB Journal • Research Communication
Regulatory polymorphism in transcription factor KLF5 at the MEF2 element alters the response to angiotensin II and is associated with human hypertension Yumiko Oishi,*,† Ichiro Manabe,*,‡,储,1 Yasushi Imai,* Kazuo Hara,§ Momoko Horikoshi,§ Katsuhito Fujiu,*,† Toshihiro Tanaka,¶ Tadanori Aizawa,# Takashi Kadowaki,§ and Ryozo Nagai*,†,‡,**,1 *Department of Cardiovascular Medicine, †Translational Systems Biology and Medicine Initiative, ‡ Global Center of Education and Research for Chemical Biology of Diseases, and §Department of Metabolic Diseases, Graduate School of Medicine, University of Tokyo, Tokyo, Japan; 储PRESTO, Japan Science and Technology Agency, Saitama, Japan; ¶Laboratory for Cardiovascular Diseases, Center for Genomic Medicine, RIKEN, Yokohama, Japan; #Department of Cardiology, Cardiovascular Institute, Tokyo, Japan; and **Translational Research Center, University of Tokyo Hospital, Tokyo, Japan Kru¨ppel-like factor 5 (KLF5) is a zincfinger-type transcription factor that mediates the tissue remodeling in cardiovascular diseases, such as atherosclerosis, restenosis, and cardiac hypertrophy. Our previous studies have shown that KLF5 is induced by angiotensin II (AII), although the precise molecular mechanism is not yet known. Here we analyzed regulatory single nucleotide polymorphisms (SNPs) within the KLF5 locus to identify clinically relevant signaling pathways linking AII and KLF5. One SNP was located at ⴚ1282 bp and was associated with an increased risk of hypertension: subjects with the A/A and A/G genotypes at ⴚ1282 were at significantly higher risk for hypertension than those with the G/G genotype. Interestingly, a reporter construct corresponding to the ⴚ1282G genotype showed much weaker responses to AII than a construct corresponding to ⴚ1282A. Electrophoretic mobility shift, chromatin immunoprecipitation, and reporter assays collectively showed that the ⴚ1282 SNP is located within a functional myocyte enhancer factor 2 (MEF2) binding site, and that the ⴚ1282G genotype disrupts the site and reduces the AII responsiveness of the promoter. Moreover, MEF2 activation via reactive oxygen species and p38 mitogen-activated protein kinase induced KLF5 expression in response to AII, and KLF5 and MEF2 were coexpressed in coronary atherosclerotic plaques. These results suggest that a novel signaling and transcription network involving MEF2A and KLF5 plays an important role in the pathogenesis of cardiovascular diseases such as hypertension.—Oishi, Y., Manabe, I., Imai, Y., Hara, K., Horikoshi, M., Fujiu, K., Tanaka, T., Aizawa, T., Kadowaki, T., Nagai, R. Regulatory polymorphism in transcription factor KLF5 at the MEF2 element alters the response to angiotensin II and is associated with human hypertension. FASEB J. 24, 1780 –1788 (2010). www.fasebj.org ABSTRACT
1780
Key Words: p38 MAP kinase 䡠 reactive oxygen species Members of the kru¨ppel-like factor (KLF) family of transcription factors are important regulators of development, cellular differentiation and growth, and the pathogenesis of various diseases, including cancer and cardiovascular disease. Recently we generated a strain of Klf5-knockout (Klf5⫹/⫺) mice, with which we were able to show that KLF5 mediates cardiovascular remodeling in vivo (1). For example, Klf5⫹/⫺ mice exhibited much less neointima formation than wild-type mice in vascular injury models and showed less cardiac hypertrophy and fibrosis in response to continuous infusion of angiotensin II (AII). KLF5 also plays an essential role in phenotypic modulation of smooth muscle cells (SMCs), which is crucially involved in neointima formation (2, 3) and controls expression of various paracrine factor genes, including PDGF-A (1, 4), which is important for tissue remodeling in the cardiovascular system. Interestingly, expression of KLF5 in coronary lesions following directional coronary atherectomy was correlated with a higher incidence of restenosis (5), suggesting a role for KLF5 in human vascular diseases. KLF5 expression is induced by AII, phorbol 12-myristate 13-acetate (PMA), and fibroblast growth factor-2 (6). Among these, AII is known to be a key factor involved in both the development and the progression of various cardiovascular diseases. The precise mechanism by which KLF5 is induced by AII remains unclear, however. 1 Correspondence: Department of Cardiovascular Medicine, Graduate School of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo, Tokyo 113-8655, Japan. E-mail: I.M.,
[email protected]; R.N.,
[email protected] doi: 10.1096/fj.09-146589
0892-6638/10/0024-1780 © FASEB
Myocyte enhancer factor 2 (MEF2) belongs to the MADS (MCM1, agamous, deficiens, SRF) family of transcription factors and plays a pivotal role in the development of various organ systems, including the cardiovascular system (7, 8). In adult tissues, MEF2 proteins act in stress response and remodeling programs and appear to be involved in mediating cell survival, apoptosis, and proliferation (7). In vascular SMCs, MEF2A has been shown to be activated by AII and to control genes such as macrophage chemoattractant protein-1 (MCP-1) and c-jun (9, 10) and has been implicated in the control of SMC differentiation marker genes (11, 12). MEF2 proteins are also known to be subjected to post-translational modification that affects their function, including phosphorylation, acetylation, and SUMOylation (13, 14). A major mediator known to modulate MEF2 is p38 mitogenactivated protein kinase (p38 MAPK) (7), which phosphorylates MEF2, thereby activating it. Within the cardiovascular system, MEF2 is expressed in rat carotid artery neointima (15), and overexpression of a dominant-negative form of MEF2A inhibited neointima formation in a rat femoral artery injury model (10). In addition, deletion of 7 aa from MEF2A reportedly leads to familial vascular disease with features of coronary artery disease/myocardial infarction (16), although subsequent case-control studies found no consistent association with myocardial infarction across different populations (17). These earlier studies suggest that MEF2 plays an important role in cardiovascular disease, but its actions remain poorly understood, due in large part to the paucity of information about its downstream target genes. Regulatory single nucleotide polymorphisms (rSNPs) are a class of SNPs located within the regulatory regions of genes that can alter recognition and binding sites for transcription factors, resulting in the loss or gain of function of the affected factors (18). Thus, rSNPs can serve as a tool with which to shed light on novel signaling pathways involved in the pathogenesis and progression of complex diseases, such as hypertension. To clarify the mechanisms by which KLF5 expression is regulated and to further investigate the clinical relevance of KLF5 in cardiovascular disease, we sought rSNPs in the KLF5 gene locus. One relatively common SNP was at ⫺1282 bp, within a MEF2-binding motif in a highly conserved region 5⬘ of the KLF5 gene. Our aim in the present study was to clarify the mechanisms by which KLF5 expression is regulated by this SNP and to further investigate the clinical relevance of KLF5 in cardiovascular disease.
MATERIALS AND METHODS Subjects The case-control group included 410 Japanese subjects (344 men and 66 women, aged 65.5⫾9.9 yr), 33–90 yr of age, who underwent coronary angiography in the Department of Cardiovascular Medicine at the University of Tokyo Hospital between October 1999 and March 2001. Written informed MEF2-KLF5 PATHWAY IN HYPERTENSION
consent was obtained from every subject after a full explanation of the study, which was approved by the Ethics Committee of the University of Tokyo Hospital. Hypertension was diagnosed according to World Health Organization criteria (19): subjects were categorized as hypertensive when they had a systolic blood pressure of ⱖ140 mmHg or a diastolic blood pressure of ⱖ90 mmHg. Subjects taking antihypertensive medication were also categorized as hypertensive. A second nonhypertensive control group was recruited from apparently healthy adult volunteers who had annual health checkups in the affiliated hospitals. Nonhypertensive controls were defined as participants not taking any antihypertensives and having a systolic blood pressure of ⬍130 mmHg and a diastolic blood pressure of ⬍85 mmHg. Genotyping All SNPs were genotyped using a MassARRAY system (Sequenom) at Hitachi Life Science (Tokyo, Japan). The primer sequences are available on request. SNPs were identified based on the sequence reported in the GenBank (KLF5: NC_000013). For the second nonhypertensive group, the SNP rs3812852 was genotyped using Taqman SNP Genotyping Assays with an ABI 7900HT (Applied Biosystems, Foster City, CA, USA) according to the manufacturer’s protocol. Reagents AII was purchased from Calbiochem (La Jolla, CA, USA). LY294002, SB203580, U0126, and diphenyleneiodonium chloride (DPI) were purchased from Sigma (St. Louis, MO, USA). Cell culture Rat aortic SMCs were isolated from the thoracic aortas of Sprague-Dawley rats (Charles River Japan, Yokohama, Japan). The cells were maintained and used for experimentation as described previously (4). Human aortic SMCs were purchased from Camblex (Rockland, ME, USA) and maintained in SmGM2 medium. Cells in passages 5–12 were used for experimentation. Plasmids, transfection, and luciferase assays The human KLF5 promoter/luciferase reporter plasmid pGL3-KLF5, which contains a KLF5 gene fragment extending from bp ⫺1920 to ⫹ 232, was constructed by cloning the SacI-XbaI fragment of pGVB-BTEB2 (6) into the pGL3-basic luciferase expression vector (Promega, Madison, WI, USA). A mutation corresponding to the ⫺1228 SNP was introduced by PCR amplification of the region [KLF5(⫺1282G)-Luc]. MEF2A cDNA was excised from pcDNA1-MEF2A (kindly provided by Dr. Eric N. Olson, University of Texas Southwestern Medical Center, Dallas, TX, USA) and inserted into pCAGMS (20) to generate the MEF2 expression vector. For transfection, rat aortic SMCs were plated at 15,000 cells/cm2 in 12-well plates. Plasmid DNA (0.75 g for each well) was transfected using Polyfect transfection reagent (Qiagen, Valencia, CA, USA) according to the manufacturer’s protocol on the day after plating. For AII stimulation, either AII (1 M) or vehicle was added to the culture medium 42 h after transfection. The cells were harvested 48 h after transfection, and luciferase activities were measured using a luciferase assay system (Promega). These activities were then normalized to the protein concentration of each cell lysate. 1781
Small interfering RNA (siRNA) siRNAs were constructed according to the manufacturer’s protocols for the Silencer siRNA construction kit (Ambion, Austin, TX, USA). For transfection, human aortic SMCs were plated at a density of 2.0 ⫻ 104/cm2 in 6-cm culture plates. Once the cells reached 90–95% confluency, 4 g of double-stranded, 21-mer siRNA was transfected using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. These cells were incubated in medium containing the siRNA-Lipofectamine 2000 complex for 4 h, after which the medium was replaced with growth medium, and the incubation was continued until the cells reached confluence (⬃12 h). The medium was then replaced with SmGM2based, serum-free medium supplemented with 2.85 mg/L insulin, 35.2 mg/L l-ascorbic acid, and 5 mg/L transferrin. Forty-eight hours after transfection, 1 M AII was added to the medium. The siRNA oligonucleotide sequences were as follows: for MEF2A, 5⬘-AGAGAAAGCCCUUCUGUAAUU-3⬘ (sense) and 5⬘-UUACAGAAGGGCUUUCUCUUU-3⬘ (antisense); for control siRNA, 5⬘-AGAGAAAGCCCGGAUGUAAUU-3⬘ (sense) and 5⬘-UUACAUCCGGGCUUUCUCUUU-3⬘ (antisense). RNA purification and real-time PCR Total RNA was purified from cells using RNeasy (Qiagen) according to the manufacturer’s instructions, after which quantitative real-time PCR was carried out using a Lightcycler system (Roche, Basel, Switzerland). The sequences of the primers for human KLF5 were described previously (21). The sequences of primers for MEF2A were 5⬘-ATGGGGCGGAAGAAAATACAA-3⬘ and 5⬘-GACTGTGACAGACATAGAGAAGTT-3⬘. Western blotting Cells were lysed in RIPA buffer and then subjected to SDS-PAGE, as described previously (20). Monoclonal antiKLF5 antibody (KM1785) (1) was used to assess KLF5 expression. An anti-MEF2 antibody that detects MEF2A (15) was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Electrophoretic mobility shift assays (EMSAs)
Chromatin immunoprecipitation (ChIP) assay ChIP assays were carried out as described previously using anti-MEF2 antibody (Santa Cruz Biotechnology) (20). Briefly, human aortic SMCs cultured in 10-cm culture dishes were treated with formaldehyde at the indicated times after AII stimulation and then prepared for immunoprecipitation. Chromatin samples were immunoprecipitated with anti-MEF2 Vol. 24
June 2010
Immunostaining Specimens of atheromatous lesions were retrieved percutaneously by directional coronary atherectomy at the Cardiovascular Institute Hospital. Written informed consent was obtained from every subject after a full explanation of the study, which was approved by the Ethics Committees of the University of Tokyo Hospital and the Cardiovascular Institute Hospital. The tissues were fixed in 20% buffered formalinmethanol mixture, embedded in paraffin, cut into 5-m-thick sections, and stained with hematoxylin-eosin and Masson’s trichrome, or immunostained for KLF5, MEF2, and SM ␣-actin as described previously (20). For KLF5 and MEF2 immunostaining, formalin-fixed, paraffin-embedded tissue sections were antigen retrieved in citrate buffer (pH 7.0). A catalyzed signal amplification system (Dako, Carpinteria, CA, USA) was then used according to the manufacturer’s protocol to visualize the labeling. Each antibody was diluted to 1:1000. Statistical analysis Comparisons of the allele frequencies and genotype distribution were made using 2 statistics with 1 and 2 degrees of freedom, respectively. Linkage disequilibrium coefficients between the gene polymorphisms were calculated from the maximum likelihood estimate. Comparisons between 2 groups were made using Student’s t test, and comparisons among multiple groups were made using ANOVA followed by a post hoc Tukey-Kramer test. Values of P ⬍ 0.05 were considered significant.
RESULTS
EMSAs were carried out as described previously (22). The sequences of the double-stranded oligonucleotides were as follows: for the MEF2 binding site within human KLF5 (⫺1295 to ⫺1271 bp), 5⬘-GTGGTATATGTAAAACTGTCTAATG-3⬘ (KLF5 ⫺1282A probe) and 5⬘-GTGGTATATGTAAGACTGTCTAATG-3⬘ (KLF5 ⫺1282G probe); for the MEF2 binding site within the mouse muscle creatine kinase (Ckm) promoter, 5⬘-GATCGCTCTAAAAATAACCCTGTCG-3⬘ (Ckm MEF2 probe). MEF2A protein was synthesized using a TNTcoupled reticulocyte lysate system (Promega). One microliter of each programmed lysate was used in the EMSAs, and the binding reactions were run on 5% polyacrylamide gels. To obtain better separation of shifted bands, gels were electrophoresed until most free probes had run off the gels.
1782
antibody and then reverse-cross-linked, purified, and subjected to PCR analysis. DNA prepared from chromatin samples served as a positive control (input). DNA prepared from chromatin samples immunoprecipitated without antibody served as a negative control. The sequences of the PCR primers were for KLF5 ⫺3000, 5⬘-CCCGCATATTCATCCGTATTTATTCCCG-3⬘ and 5⬘GTCAGGTTTCAGAAATCAATTAGTGGCC-3⬘; for KLF5 ⫺1300, 5⬘-CCATACCTTTGATAGATACACTC-3⬘ and 5⬘-CTGGCTTATTGGGAGGACGGAAC-3⬘.
KLF5 ⴚ1282 SNP associates with hypertension To identify novel transcriptional regulatory programs that control KLF5 expression in cardiovascular diseases, we sought SNPs within possible transcriptional regulatory regions. When we sequenced the KLF5 locus in the genomes of 20 Japanese individuals, including all KLF5 exons and introns, we found the single most common SNP was located at ⫺1282 bp (rs3812852) within 2000 bp 5⬘ of the transcription initiation site, in an area where cis-regulatory regions are likely concentrated. Moreover, the sequence in the vicinity of the ⫺1282 SNP was well conserved across humans, mice, and rats, suggesting that this SNP might be located within a transcriptional regulatory region. Notably, we detected a statistically significant difference in the distribution of genotypes at ⫺1282 bp (P⫽0.001) between hypertensive (HT) and normotensive (NT1) subjects (Table 1). Subjects with the A/A and A/G genotypes at that position had a significantly
The FASEB Journal 䡠 www.fasebj.org
OISHI ET AL.
TABLE 1. Genotype frequency of KLF-5 ⫺1282 SNP (rs3812852) in hypertensive and normotensive groups Genotype frequency [n(%)] Additive model Group
HT NT1 NT2
AA
AG
Dominant model GG
P
AA
AG/GG
Recessive model P
AA/AG
GG
P
274 (69.2) 117 (29.5) 5 (1.3) 274 (69.2) 122 (30.8) 391 (98.7) 5 (1.3) 54 (71.1) 16 (21.1) 6 (7.9) 0.001 54 (71.1) 22 (28.9) 0.747 70 (92.1) 6 (7.9) 0.0004 636 (73.4) 200 (23.1) 30 (3.5) 0.007 636 (73.4) 230 (26.6) 0.12 836 (96.5) 30 (3.5) 0.0266
Minor allele frequency
P
0.16 0.18 0.17
0.46 0.55
Odds ratio [95% confidence interval]: NT1, 6.7 关1.9 to 22.6兴; NT2, 2.806 关1.080 to 7.289兴.
higher risk of developing hypertension than those with the G/G genotype (P⫽0.0004). Likewise, subjects with the A/A or A/G genotype were also at significantly higher risk than those with G/G when the HT group was compared with the second normotensive control group (NT2, n⫽866, P⫽0.0266). The ⴚ1282 genotype affects AII-dependent KLF5 promoter activation Because the ⫺1282 SNP associates with hypertension and is located within a possible transcriptional regulatory region, we hypothesized that the ⫺1282 genotype might affect the transcriptional regulation of KLF5. To test that idea, we generated luciferase reporter constructs driven by the promoter spanning ⫺1920 to ⫹ 232 bp of KLF5 and containing either nucleoside A [KLF5(⫺1282A)-Luc] or G [KLF5(⫺1282G)-Luc] at ⫺1282 bp. When rat aortic SMCs were transfected with one or the other of these reporter constructs, the luciferase activities were comparable under the basal conditions (Fig. 1). However, when we treated the cells with AII, which we previously found to induce endogenous KLF5 expression (1), the reporter activity of KLF5(⫺1282A)-Luc was significantly increased, but the activity of KLF5(⫺1282G)-Luc was not, suggesting that the ⫺1282 SNP is located within the regulatory region that is important for AII-dependent KLF5 activation.
mouse muscle creatinine kinase (Cmk), which contains a defined MEF2 binding motif (Fig. 2, lane 3), and the formation of the shift complex was attenuated by the presence of an excess of cold ⫺1282A KLF5 probe, but not by cold ⫺1282G probe (Fig. 2, compare lanes 5 and 6). Thus substituting the A at ⫺1282 with G appears to disrupt the MEF2-binding element. MEF2A controls KLF5 expression in response to AII Based on the findings that the ⫺1282 SNP is located within a MEF2 binding element, and that this SNP alters AII-induced KLF5 promoter activity, we hypothesized that MEF2 might be involved in AII-dependent activation of KLF5. Human aortic SMCs express MEF2A, MEF2C, and MEF2D (data not shown), but because previous studies have shown that it is MEF2A that mediates activation of c-jun and MCP-1 in response to AII (10), we focused on
The ⴚ1282 genotype alters MEF2 binding affinity Because the ⫺1282 polymorphism altered the responsiveness of the KLF5 promoter to AII, we hypothesized that ⫺1282 bp was located within a cis-regulatory element. Consistent with that idea, we found that the sequence containing ⫺1282A (tgTAAAAATgtcta) corresponds to the MEF2 binding motif (YTA(A/T)4TAR) (14, 23). We then carried out EMSAs to test whether substituting A in the core of the motif with G (tgTAGAAATgtcta) might reduce the binding affinity of MEF2. All MEF2 family members bind to the same DNA consensus motif (14). We therefore analyzed the affinity to MEF2 using in vitro translated MEF2A, which we found to specifically bind an oligonucleotide probe corresponding to the ⫺1282A genotype, but not one corresponding to ⫺1282G (Fig. 2; compare lanes 9 and 14). MEF2A also bound an oligonucleotide probe from MEF2-KLF5 PATHWAY IN HYPERTENSION
Figure 1. KLF5 ⫺1282 genotype affects AII responsiveness. Rat aortic SMCs were transfected with either KLF5(⫺1282A)Luc or KLF5(⫺1282G)-Luc; beginning 42 h later, the cells were treated for 6 h with either AII (1 M) or vehicle. The amount of plasmid transfected was 0.75 g/well in 12-well culture plates. Luciferase activities were normalized to that in untreated cells transfected with KLF5(⫺1282A)-Luc. Data are expressed as means ⫾ sd. *P ⬍ 0.05. 1783
Figure 2. MEF2A binds to the KLF5 ⫺1282 region. EMSA analysis of MEF2 binding to the sequence between bp ⫺1295 and ⫺1271 of the human KLF5 promoter containing either ⫺1282A (KLF5 ⫺1282A) or ⫺1282G (KLF5 ⫺1282G). MEF2 element of the mouse muscle creatinine kinase (Ckm) promoter served as a positive control (Ckm MEF2). Radiolabeled probes were incubated with 1 l of either reticulocyte lysate programmed with the MEF2A expression vector or unprogrammed (unpr) reticulocyte lysates. In lanes 4 – 6, 10, 11, 15, and 16, a 100-fold excess of the indicated cold competitor was added. Free probes ran off the gel. Data are representative of 3 independent experiments. Asterisk indicates nonspecific shift bands.
MEF2A. We found that AII induced expression of both KLF5 mRNA and protein in human aortic SMCs (Fig. 3A, B), but that effect was lost when the cells were transfected with siRNA against MEF2A (Fig. 3C, D), indicating that MEF2A is required for up-regulation of KLF5 by AII. MEF2A protein was expressed under basal conditions, in
the absence of AII, and AII moderately increased the expression within 1 h (Supplemental Fig. 1). We next carried out ChIP assays to test whether MEF2 is able to bind to the ⫺1282 region of endogenous KLF5. Chromatin samples were prepared from human aortic SMCs before and after stimulation with
Figure 3. Effect of knocking down MEF2A on AII-induced KLF5 expression. A) Human aortic SMCs were treated with 1 M AII. Cells were harvested after stimulating with AII for the indicated times, and expression of KLF5 mRNA was analyzed using real-time PCR. Copy number of the transcripts was normalized to that of 18s rRNA and then further normalized to the level in untreated cells (0 h). B) Western blot analysis of KLF5 expression in SMCs treated with 1 M AII. Membrane was reprobed for -tubulin. C) Human aortic SMCs were transfected with siRNA against MEF2A or control mutant siRNA, as indicated, 48 h prior to AII stimulation. Four micrograms of siRNA was applied to each 6-cm culture dish. Levels of KLF5 and MEF2A expression after 3 h of AII stimulation were analyzed by real-time PCR. Expression level of each gene was normalized to the level of 18s and then further normalized to the levels in samples from cells that were not transfected with siRNA. D) Western blot analysis of KLF5 and MEF2A expression after 3 h of AII (1 M) stimulation. The cell lysates were electrophoresed on 2 gels and then transferred to 2 membranes, which were first used to detect KLF5 and MEF2A, after which they were reprobed for -tubulin. Data are representative of 3 independent experiments and are expressed as means ⫾ sd. *P ⬍ 0.05. Note that not all significant combinations are indicated. 1784
Vol. 24
June 2010
The FASEB Journal 䡠 www.fasebj.org
OISHI ET AL.
Figure 4. MEF2 binds to the endogenous KLF5 promoter in response to AII. Binding of MEF2 to the endogenous KLF5 promoter was analyzed by ChIP. Intact human aortic SMCs were fixed in formaldehyde after stimulation with 1 M AII for the indicated times, after which chromatin samples were subjected to ChIP analysis. A) MEF2 bound to the region containing the MEF2 element corresponding to the ⫺1282 SNP. B) No MEF2 binding was detected in the region of bp ⫺3000 in the human KLF5 promoter, which does not contain a MEF2 binding element. C) Schematic representation showing the positions of the MEF2 binding site and PCR primers. D) The ⫺1282 genotype affects AII-induced KLF5 promoter activation. Rat aortic SMCs were cotransfected with 0.375 g of either the KLF5(⫺1282A)-Luc or KLF5(⫺1282G) reporter construct plus 0.375 g of either the MEF2A expression vector or empty pCAG vector and harvested 48 h later. Luciferase activities were normalized to that in cells transfected with KLF5(⫺1282A)-Luc and empty pCAG vector. Data are expressed as means ⫾ sd. *P ⬍ 0.05.
AII, and then immunoprecipitated with a specific antibody against MEF2A. The cells were also confirmed to carry the ⫺1282AA genotype. Figure 4A shows that there was no binding of MEF2 to the KLF5 promoter before AII stimulation (0 h). After 2.5 h of AII stimulation, MEF2 binding to the ⫺1282 region was apparent, but after 5 h the MEF2 binding was no longer detectable. Notably, the binding of MEF2 to the KLF5 ⫺1282 region correlated with the level of the gene’s expression, which reached a maximum after 3 h of AII stimulation and then gradually declined (Fig. 3). No binding of MEF2 was seen in the region of ⫺3000 bp of KLF5, which does not contain a MEF2 binding motif (Fig. 4B, C).
SMCs is similarly activated. While inhibitors of p38 MAPK (SB203580) and ROS (DPI) significantly suppressed AII-induced up-regulation of KLF5 transcription, inhibitors of extracellular signal-regulated kinase (ERK) (U1026) and Akt (LY294002) did not (Fig. 5A). Likewise, the ROS and p38 MAPK inhibitors clearly reduced AII-induced expression of KLF5 (Fig. 5B),
The ⴚ1282 genotype affects MEF2-induced transactivation of KLF5 The results summarized in the last section suggest that MEF2A plays a key role in the AII-induced activation of KLF5 transcription. Furthermore, our finding that the ⫺1282G genotype disrupts MEF2A binding suggests that the ⫺1282 SNP modulates MEF2-dependent KLF5 activation. Consistent with that idea, we found that whereas the activity of KLF5(⫺1282A)-Luc was increased by overexpression of MEF2A, KLF5(⫺1282G)Luc showed a significantly weaker response to MEF2A (Fig. 4D). This strongly suggests that the insensitivity of the ⫺1282G genotype to AII reflects the lower affinity of the ⫺1282 region for MEF2A. AII-dependent induction of KLF5 is mediated by p38 MAPK and ROS Previous studies have shown that p38 MAPK activates MEF2A in rat SMCs (10). By examining the effects of various inhibitors of AII-responsive mediators (24), we asked whether AII-induced KLF5 expression in human MEF2-KLF5 PATHWAY IN HYPERTENSION
Figure 5. AII-induced KLF5 expression is mediated by ROS and p38 MAPK. A) Human aortic SMCs were incubated with the indicated kinase inhibitors (LY294002, U0126, 10 M; SB203580, 5 M) and diphenyleneiodonium chloride (DPI, 5 M) 1 h prior to AII (1 M) stimulation. Cells were harvested after 3 h of stimulation with AII, and KLF5 expression was analyzed by real-time PCR. Expression level of each gene was normalized to the level of 18s and then further normalized to the level in samples from cells not treated with AII. Data are expressed as means ⫾ sd. B) Western blot analysis of KLF5 expression in cells treated with 1 M AII and the indicated inhibitors. LY, LY294002; SB, SB203580; U, U0126. 1785
suggesting that a ROS-p38 MAPK pathway is involved in AII-induced up-regulation of KLF5.
network involving MEF2 and KLF5 is operating in SMCs in human coronary atheromatous lesions.
KLF5 and MEF2 are coexpressed in human coronary atheromatous lesions
DISCUSSION
Finally, to gain further insight into the role of MEF2 and KLF5 in vascular disease, we analyzed their expression in human coronary lesions. When we immunostained KLF5 and MEF2A in serial sections of specimens obtained from directional coronary atherectomy, we found numerous KLF5-positive cells in focal regions showing greater cellularity (Fig. 6D, F). These regions contained cells having characteristics of phenotypically modulated SMCs, including a stellate morphology, a surrounding loose light-staining extracellular matrix (25), and weaker staining for SM ␣-actin (Fig. 6B, C). Furthermore, regions containing KLF5-positive cells also contained many MEF2A-positive cells, which suggests that a transcription factor
To facilitate identification of KLF5 regulatory programs, in the present study we combined information on SNPs, sequence conservation, and clinical associations. Using this approach, along with molecular biological analyses, we have been able to show that the ⫺1282 SNP is associated with a higher prevalence of hypertension, and that it is situated within the MEF2 binding site, which is essential for AII-dependent induction of KLF5 expression. These findings clearly demonstrate the strength of this combinatorial approach to identification of disease-associated gene regulatory programs. We previously showed that Klf5 haploinsufficiency resulted in much-reduced cardiac hypertrophy and fibrosis in response to AII. Results of the present study
Figure 6. Histological analysis of human coronary atheromatous lesions. A–G) Serial sections of an atheromatous specimen obtained by directional coronary atherectomy were stained with hematoxylin/eosin (A) and Masson’s trichrome (B) and immunostained for the indicated molecules (C–G). Sections include a segment exhibiting greater cellularity and less extracellular matrix than adjacent (top right) segments. As compared to cells in the fibrous (top right) segment, cells in the high-cellularity segment exhibited much weaker staining for SM ␣-actin SMCs (C), a stellate morphology, and surrounding loose, light-staining extracellular matrix. Positive staining for KLF5 (D, F) and MEF2A (E, G) was observed in many cells within the high-cellularity segment. Note that although KLF5 staining was confined within the nucleus, MEF2A staining was seen in both the cytosol and nucleus. Original view: ⫻200 (A–E); ⫻400 (F, G). Scale bars ⫽ 50 m. H) Model for KLF5 promoter activation by AII. AII induces KLF5 transcription via MEF2A activation. The ⫺1282 SNP modulates the responsiveness to AII by altering the affinity for MEF2A. 1786
Vol. 24
June 2010
The FASEB Journal 䡠 www.fasebj.org
OISHI ET AL.
show that the response of KLF5 to AII is mediated by MEF2A in human aortic SMCs; KLF5 and MEF2 are coexpressed in coronary lesions; and the KLF5 ⫺1282 SNP, which is located within the MEF2-binding element, alters the responsiveness of KLF5 to AII and associates with hypertension. These findings further establish KLF5 as an important effector molecule of AII signaling in cardiovascular disease. Moreover, the results provide novel insights into the functional role played by MEF2A in the pathogenesis of the disease. Our finding that ROS are also required for AIIdependent activation of KLF5 suggests that KLF5 is a novel effector molecule in the ROS-p38 MAPK pathway, which can be activated by a variety of environmental and cellular cues (26). In vascular SMCs, AII is known to induce the generation of intracellular ROS (24), and Ushio-Fukai et al. (27) recently reported that AII-induced activation of p38 MAPK is mediated by ROS. For instance, DPI, an NADPH oxidase inhibitor, was shown to inhibit AII-induced phosphorylation of p38 MAPK in SMCs. In our study DPI similarly inhibited AII-induced expression of KLF5 (Fig. 5A, B). A number of studies have shown that p38 MAPK plays a pivotal role in activation of MEF2A (28). Our finding that SB203580, a p38 MAPK inhibitor, attenuates AIIinduced up-regulation of KLF5 is consistent with those earlier reports and supports a model in which AII induces KLF5 expression via a ROS-p38 MAPK-MEF2A pathway (Fig. 6H). However, our observation that basal KLF5 promoter activity was unaffected by the ⫺1282 genotype (Fig. 1) suggests that other signaling pathways may also converge on the KLF5 promoter. In fact, PMA induces KLF5 via the ERK pathway in rabbit SMCs (6), and CCAAT/enhancer binding proteins (C/EBPs) induce KLF5 in adipocytes (20). It is therefore very likely that multiple transcriptional factors are involved in responding to different environmental cues, and that different transcriptional programs control KLF5 in different cell types. Results of the present study suggest that MEF2A expressed in SMCs is an important promoter of vascular pathology. On the other hand, recent studies have shown that MEF2 expressed in endothelial cells induces KLF2, which is essential for the anti-inflammatory and antithrombotic functions of the endothelium (29). Within endothelial cells, atheroprotective shear stress stimulates induction of KLF2 via the MEK5/ERK5/ MEF2 pathway, which ultimately leads to MEF2A binding to and transactivating the KLF2 promoter (30, 31). Conversely, tumor necrosis factor ␣ inhibits KLF2 expression by inhibiting MEF2 function (32). This suggests that MEF2A responds to different signals and activates different genes in SMCs and endothelial cells. To further clarify the functional role played by MEF2A in vascular physiology and pathology, the regulation and function of MEF2A in these 2 cell types will need to be compared. We found that the major KLF5 promoter allele (⫺1282A) is associated with a higher prevalence of hypertension in Japanese individuals. Although SNPs in MEF2-KLF5 PATHWAY IN HYPERTENSION
a number of genes are known to associate with hypertension (33), only a few transcription factor genes have been shown to contain SNPs that associate with an increased risk of cardiovascular disease. To our knowledge, this the first report of a regulatory SNP in a transcription factor gene that associates with susceptibility to a cardiovascular disease. Along with our previous findings showing an association between KLF5 expression and restenosis following coronary directional atherectomy (5), the association between the ⫺1282 SNP described herein and hypertension is indicative of the clinical significance of KLF5 in cardiovascular disease. That said, blood pressure is maintained by a complex network of integrated systems including renal, neuronal, endocrine, and vascular mechanisms (34). Within each of these systems, multiple genes are thought to contribute to the specialized functions regulating blood pressure. Thus, multiple genetic and environmental factors likely contribute to the onset and progression of primary hypertension. Indeed, a recent genome-wide association study suggests that hypertension may have fewer common risk alleles of larger effect sizes than some other common complex diseases (35). The significance of the KLF5’s ⫺1282 SNP would therefore seem to warrant further evaluation using larger and more diverse populations. This study was supported in part by grants-in-aid from the Ministry of Education, Culture, Sports, Science and Technology, Japan (to R.N., I.M., and Y.O.), a research grant from the National Institute of Biomedical Innovation, Japan, and Special Coordination Funds for Promoting Science and Technology from the Japan Science and Technology Agency (to R.N.), and research grants from Japan Science and Technology Institute, NOVARTIS Foundation for the Promotion of Science, Kato Memorial Bioscience Foundation, Takeda Science Foundation, Cell Science Research Foundation, and Tokyo Biochemical Research Foundation (to I.M.). The authors declare no conflicts of interest. The authors gratefully acknowledge Noriko Yamanaka, Michiko Hayashi, Yuka Tani, Eriko Magoshi, Xiao Yingda, and Yuko Mouri for their excellent technical assistance.
REFERENCES 1.
2. 3. 4.
5.
Shindo, T., Manabe, I., Fukushima, Y., Tobe, K., Aizawa, K., Miyamoto, S., Kawai-Kowase, K., Moriyama, N., Imai, Y., Kawakami, H., Nishimatsu, H., Ishikawa, T., Suzuki, T., Morita, H., Maemura, K., Sata, M., Hirata, Y., Komukai, M., Kagechika, H., Kadowaki, T., Kurabayashi, M., and Nagai, R. (2002) Kru¨ppel-like zinc-finger transcription factor KLF5/BTEB2 is a target for angiotensin II signaling and an essential regulator of cardiovascular remodeling. Nat. Med. 8, 856 – 863 Manabe, I., and Nagai, R. (2003) Regulation of smooth muscle phenotype. Curr. Atheroscler. Rep. 5, 214 –222 Owens, G. K., Kumar, M. S., and Wamhoff, B. R. (2004) Molecular regulation of vascular smooth muscle cell differentiation in development and disease. Physiol. Rev. 84, 767– 801 Fujiu, K., Manabe, I., Ishihara, A., Oishi, Y., Iwata, H., Nishimura, G., Shindo, T., Maemura, K., Kagechika, H., Shudo, K., and Nagai, R. (2005) Synthetic retinoid Am80 suppresses smooth muscle phenotypic modulation and in-stent neointima formation by inhibiting KLF5. Circ. Res. 97, 1132–1141 Hoshino, Y., Kurabayashi, M., Kanda, T., Hasegawa, A., Sakamoto, H., Okamoto, E., Kowase, K., Watanabe, N., Manabe, I.,
1787
6.
7. 8.
9.
10.
11.
12.
13.
14. 15.
16. 17.
18. 19.
1788
Suzuki, T., Nakano, A., Takase, S., Wilcox, J. N., and Nagai, R. (2000) Regulated expression of the BTEB2 transcription factor in vascular smooth muscle cells: analysis of developmental and pathological expression profiles shows implications as a predictive factor for restenosis. Circulation 102, 2528 –2534 Kawai-Kowase, K., Kurabayashi, M., Hoshino, Y., Ohyama, Y., and Nagai, R. (1999) Transcriptional activation of the zinc finger transcription factor BTEB2 gene by Egr-1 through mitogen-activated protein kinase pathways in vascular smooth muscle cells. Circ. Res. 85, 787–795 Potthoff, M. J., and Olson, E. N. (2007) MEF2: a central regulator of diverse developmental programs. Development 134, 4131– 4140 Lin, Q., Lu, J., Yanagisawa, H., Webb, R., Lyons, G. E., Richardson, J. A., and Olson, E. N. (1998) Requirement of the MADSbox transcription factor MEF2C for vascular development. Development 125, 4565– 4574 Suzuki, E., Nishimatsu, H., Satonaka, H., Walsh, K., Goto, A., Omata, M., Fujita, T., Nagai, R., and Hirata, Y. (2002) Angiotensin II induces myocyte enhancer factor 2- and calcineurin/ nuclear factor of activated T cell-dependent transcriptional activation in vascular myocytes. Circ. Res. 90, 1004 –1011 Suzuki, E., Satonaka, H., Nishimatsu, H., Oba, S., Takeda, R., Omata, M., Fujita, T., Nagai, R., and Hirata, Y. (2004) Myocyte enhancer factor 2 mediates vascular inflammation via the p38dependent pathway. Circ. Res. 95, 42– 49 Katoh, Y., Molkentin, J. D., Dave, V., Olson, E. N., and Periasamy, M. (1998) MEF2B is a component of a smooth musclespecific complex that binds an A/T-rich element important for smooth muscle myosin heavy chain gene expression. J. Biol. Chem. 273, 1511–1518 Deaton, R. A., Su, C., Valencia, T. G., and Grant, S. R. (2005) Transforming growth factor-beta1-induced expression of smooth muscle marker genes involves activation of PKN and p38 MAPK. J. Biol. Chem. 280, 31172–31181 Gregoire, S., Tremblay, A. M., Xiao, L., Yang, Q., Ma, K., Nie, J., Mao, Z., Wu, Z., Giguere, V., and Yang, X. J. (2006) Control of MEF2 transcriptional activity by coordinated phosphorylation and sumoylation. J. Biol. Chem. 281, 4423– 4433 Black, B. L., and Olson, E. N. (1998) Transcriptional control of muscle development by myocyte enhancer factor-2 (MEF2) proteins. Annu. Rev. Cell Dev. Biol. 14, 167–196 Firulli, A. B., Miano, J. M., Bi, W., Johnson, A. D., Casscells, W., Olson, E. N., and Schwarz, J. J. (1996) Myocyte enhancer binding factor-2 expression and activity in vascular smooth muscle cells: association with the activated phenotype. Circ. Res. 78, 196 –204 Wang, L., Fan, C., Topol, S. E., Topol, E. J., and Wang, Q. (2003) Mutation of MEF2A in an inherited disorder with features of coronary artery disease. Science 302, 1578 –1581 Lieb, W., Mayer, B., Konig, I. R., Borwitzky, I., Gotz, A., Kain, S., Hengstenberg, C., Linsel-Nitschke, P., Fischer, M., Doring, A., Wichmann, H. E., Meitinger, T., Kreutz, R., Ziegler, A., Schunkert, H., and Erdmann, J. (2008) Lack of association between the MEF2A gene and myocardial infarction. Circulation 117, 185–191 Prokunina, L., and Alarcon-Riquelme, M. E. (2004) Regulatory SNPs in complex diseases: their identification and functional validation. Expert Rev. Mol. Med. 6, 1–15 World Health Organization (1999) World Health OrganizationInternational Society of Hypertension Guidelines for the Man-
Vol. 24
June 2010
20.
21.
22. 23. 24. 25.
26. 27.
28. 29. 30.
31. 32.
33. 34. 35.
agement of Hypertension. Guidelines Subcommittee. J. Hypertens. 17, 151–183 Oishi, Y., Manabe, I., Tobe, K., Tsushima, K., Shindo, T., Fujiu, K., Nishimura, G., Maemura, K., Yamauchi, T., Kubota, N., Suzuki, R., Kitamura, T., Akira, S., Kadowaki, T., and Nagai, R. (2005) Kru¨ppel-like transcription factor KLF5 is a key regulator of adipocyte differentiation. Cell. Metab. 1, 27–39 Manabe, I., and Owens, G. K. (2001) Recruitment of serum response factor and hyperacetylation of histones at smooth muscle-specific regulatory regions during differentiation of a novel P19-derived in vitro smooth muscle differentiation system. Circ. Res. 88, 1127–1134 Manabe, I., and Owens, G. K. (2001) CArG elements control smooth muscle subtype-specific expression of smooth muscle myosin in vivo. J. Clin. Invest. 107, 823– 834 Pollock, R., and Treisman, R. (1991) Human SRF-related proteins: DNA-binding properties and potential regulatory targets. Genes Dev. 5, 2327–2341 Mehta, P. K., and Griendling, K. K. (2007) Angiotensin II cell signaling: physiological and pathological effects in the cardiovascular system. Am. J. Physiol. Cell Physiol. 292, C82–C97 Kearney, M., Pieczek, A., Haley, L., Losordo, D. W., Andres, V., Schainfeld, R., Rosenfield, K., and Isner, J. M. (1997) Histopathology of in-stent restenosis in patients with peripheral artery disease. Circulation 95, 1998 –2002 Clempus, R. E., and Griendling, K. K. (2006) Reactive oxygen species signaling in vascular smooth muscle cells. Cardiovasc. Res. 71, 216 –225 Ushio-Fukai, M., Alexander, R. W., Akers, M., and Griendling, K. K. (1998) p38 Mitogen-activated protein kinase is a critical component of the redox-sensitive signaling pathways activated by angiotensin II. Role in vascular smooth muscle cell hypertrophy. J. Biol. Chem. 273, 15022–15029 Han, J., and Molkentin, J. D. (2000) Regulation of MEF2 by p38 MAPK and its implication in cardiomyocyte biology. Trends Cardiovasc. Med. 10, 19 –22 Atkins, G. B., and Jain, M. K. (2007) Role of Kru¨ppel-like transcription factors in endothelial biology. Circ. Res. 100, 1686 –1695 Parmar, K. M., Larman, H. B., Dai, G., Zhang, Y., Wang, E. T., Moorthy, S. N., Kratz, J. R., Lin, Z., Jain, M. K., Gimbrone, M. A., Jr., and Garcia-Cardena, G. (2006) Integration of flow-dependent endothelial phenotypes by Kru¨ppel-like factor 2. J. Clin. Invest. 116, 49 –58 Kumar, A., Hoffman, T. A., Dericco, J., Naqvi, A., Jain, M. K., and Irani, K. (2009) Transcriptional repression of Kru¨ppel-like factor-2 by the adaptor protein p66shc. FASEB J. 23, 4344 – 4352 Kumar, A., Lin, Z., Senbanerjee, S., and Jain, M. K. (2005) Tumor Necrosis factor alpha-mediated reduction of KLF2 is due to inhibition of MEF2 by NF-{kappa}B and histone deacetylases. Mol. Cell. Biol. 25, 5893–5903 Agarwal, A., Williams, G. H., and Fisher, N. D. (2005) Genetics of human hypertension. Trends Endocrinol. Metab. 16, 127–133 Oparil, S., Zaman, M. A., and Calhoun, D. A. (2003) Pathogenesis of hypertension. Ann. Intern. Med. 139, 761–776 Wellcome Trust Case Control Consortium (2007) Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature 447, 661– 678
The FASEB Journal 䡠 www.fasebj.org
Received for publication September 17, 2009. Accepted for publication December 17, 2009.
OISHI ET AL.
