Department of Reproductive Medicine, First Affiliated Hospital, Xi'an Jiaotong University of Medical School, Xi'an, People's Republic of China
Department of Biochemistry and Molecular Biology, Xi'an Jiaotong University of Medical School, Xi'an, People's Republic of China
Division of Cancer Stem Cell Research, Key Laboratory of Environment and Genes Related to Diseases, Ministry of Education, Xi'an Jiaotong University of Medical School, Xi'an, People's Republic of China
Department of Reproductive Medicine, First Affiliated Hospital, Xi'an Jiaotong University of Medical School, Xi'an, 710061, People's Republic of China
The Krüppel-like factor 4 (KLF4) protein, a zinc finger transcription factor that is highly expressed in epithelial tissues such as the gut and skin, has been implicated in both tumor suppression and progression. However, the role of KLF4 in human cervical carcinoma is still unclear.
The expression of KLF4 in cervical carcinoma tissues and cervical cancer cell lines was examined with immunohistochemistry and Western blot assay. The effects of KLF4 silencing and overexpression on the cell proliferation, cell viability, and tumor formation of cervical cancer cells were investigated.
KLF4 protein expression showed a pattern of gradual decrease from normal cervix to cervical carcinoma in situ and then to invasive cervical carcinomas (P < .05). Overexpression of KLF4 in SiHa and C33A cells resulted in significantly inhibited cell growth and significantly attenuated tumor formation. Consistently, KLF4 silencing in HeLa cells significantly promoted cell growth and tumor formation. Furthermore, KLF4 overexpression caused cell cycle arrest at the G1/S transition, along with the up-regulated expression of p27Kip1 protein. Promoter analysis revealed that KLF4 transactivated the expression of p27Kip1 through the specific motif that is between the nucleotides of −435 and −60 in its promoter. The results from chromatin immunoprecipitation assays demonstrated the physical interaction between KLF4 protein and this specific motif in p27Kip1 promoter.
Infection with high-risk human papillomavirus (HR-HPV), predominantly the types 16 and 18, can cause cervical cancer.1, 2 As in almost all human cancers, additional host cell alterations in addition to the HR-HPV infection are required for the formation and development of cervical cancers.3 The etiology and pathogenesis of cervical carcinoma remain unclear, although heterogeneous genetic alterations have been reported to be responsible.4, 5 The most common genetic alterations include mutations and loss of heterozygosity of tumor suppressors, such as TP53, CDK2A, and PTEN.6 Even in the absence of mutations, tumor suppressor genes (eg, CDH1, DAPK, HIC1, and PCDH10) may still be inactivated by epigenetic alterations, such as promoter hypermethylation in cervical carcinoma.7-10 Therefore, continued identification of signature genetic and epigenetic alterations in cervical carcinoma should provide a conceptual framework for future analyses of this complex disease and future development of strategies for early detection and effective treatment.
The gut-enriched Krüppel-like factor 4 (KLF4) is highly expressed in the gastrointestinal tract and other epithelial tissues, and it has been reported to regulate cell proliferation, apoptosis, differentiation,11-14 and the maintenance of telomerase activity.15 KLF4's expression can be increased by serum deprivation, contact inhibition, and DNA damage.16, 17 In many types of tumors, the expression of KLF4 is significantly down-regulated.18-22 Consistently, the overexpression of KLF4 induces cell growth arrest in colon cancer cells and apoptosis in bladder, gastric cancer cells, and leukemia and reduces the tumorigenicity of colonic and gastric cancer cells in vivo.19, 23, 24 Consistently, it has been shown that KLF4 undergoes promoter methylation and loss of heterozygosity in gastrointestinal cancer21, 23 and medulloblastoma.19 All these findings suggest that KLF4 functions as a tumor suppressor. This notion is further supported by the finding that specific ablation of KLF4 in mouse gastric epithelium caused premalignant changes, including polypoid lesions.25 However, KLF4 has also been proposed to be an oncogene in breast cancer,26-28 and persistent expression of KLF4 can serve as a marker for the progression and poor prognosis of head and neck squamous cell carcinoma.29
In the present study, we performed detailed analyses on the role of KLF4 in cervical carcinoma. We demonstrated that KLF4 was down-regulated in the development and progression of cervical carcinoma. The overexpression of exogenous KLF4 protein inhibited the cell growth and tumor formation by activating the cell cycle suppressor p27Kip1, which supports the hypothesis that KLF4 is a tumor suppressor in cervical carcinoma.
MATERIALS AND METHODS
Cell Lines and Culture Conditions
Human cervical carcinoma cell lines HeLa, SiHa, and C33A and human teratoma cell line Tera-1 were all purchased from the American Type Culture Collection (Manassas, Va). HeLa, SiHa, and C33A cells were cultured in Dulbecco Modified Eagle Medium (Sigma-Aldrich, St Louis, Mo) supplemented with 10% fetal bovine serum (FBS; Invitrogen, Carlsbad, Calif) at 37°C in an atmosphere with 5% CO2. Tera-1 cells were cultured in McCoy 5A medium (Sigma-Aldrich) with 15% FBS at 37°C in an atmosphere with 5% CO2.
Human Tissue Specimens and Immunohistochemistry
A total of 43 normal cervixes, 28 cervical carcinomas in situ, and 64 cervical carcinoma tissues were obtained from the First Affiliated Hospital of Xi'an Jiaotong University between January 2008 and December 2009. Ten normal cervix and 18 invasive squamous carcinoma of the cervix (SCC) fresh tissues were collected from the First Affiliated Hospital of Xi'an Jiaotong University for Western blot analysis.
Four-millimeter sections of formalin-fixed and paraffin-embedded tissue specimens were prepared. A standard immunostaining procedure was performed using a rabbit polyclonal antibody against human KLF4 (1:400 dilution; sc-20691; Santa Cruz Biotechnology, Santa Cruz, Calif). A positive reaction was defined as a reddish-brown precipitate being observed in the nuclei. The KLF4 staining was classified into 3 categories: negative, weak, and strong expression, based on the percentage of positive cells and the staining intensity.30 The percentage of positive cells was divided into 5 ranks of scores: <10% (0), 10% to 25% (1), 25% to 50% (2), 50% to 75% (3), and >75% (4). The intensity of staining was divided into 4 ranks of scores: no staining (0), light brown (1), brown (2), and dark brown (3). The positivity of KLF4 staining was determined by the following formula: immunohistochemistry (IHC) score = percentage score × intensity score. The overall score of ≤3 was defined as negative, >3 but ≤6 as weak positive, and >6 as strong positive.
For the expression of KLF4 in cells, similar immunocytochemistry was performed, after the cells were seeded on cover slips for 48 hours, fixed with 4% paraformaldehyde for 20 minutes, and permeabilized with 0.2% Triton X-100 for 20 minutes at room temperature.
Cell Growth and Cell Viability Assays
Cells (5 × 104) were seeded in triplicate in 2-mL media in 6-well plates. The cells were trypsinized and then counted every day for 1 week using a hemocytometer. Cell growth curves were generated to assess the cell proliferation. Cell viability was assessed using 3-(4,5-dimethylthiazole-yl)-2,5-diphenyl tetrazolium bromide (Sigma-Aldrich) dye according to standard protocol. The number of viable cells was determined by measuring the absorbance at 490 nm.
Flow Cytometry Analysis
A cell cycle analysis was performed using fluorescence-activated cell sorting (FACS; Becton Dickinson, Franklin Lakes, NJ) according to the manufacturer's protocol. The cells were harvested and fixed in 70% ethanol overnight at 4°C. Thirty minutes before FACS analysis, the cells were treated with RNaseA and then stained with propidium iodide (Sigma-Aldrich). Cell cycle distribution was analyzed with the FACSCalibur flow cytometer using ModFit LT software.
Western Blot Assays
Western blot analyses were performed as previously described17 with the lysates from fresh tissues and cells. The rabbit polyclonal antibody against human KLF4, p27Kip1 (1:500 dilution; sc-528; Santa Cruz Biotechnology) and mouse monoclonal antibody against human β-actin (1:500 dilution; sc-47778; Santa Cruz Biotechnology) were incubated with the membranes at 4°C overnight, followed by a secondary incubation using horseradish peroxidase-conjugated antirabbit or antimouse immunoglobulin G (IgG; Thermo Fisher Scientific, New York, NY). Proteins were briefly incubated with an enhanced chemiluminescence reagent (Millipore, Billerica, Mass) and then visualized on x-ray films. The KLF4 Western blots results were normalized to those of β-actin blotting for quantification.
Vector Construction and Transfection
Full-length KLF4 cDNA was amplified using the following primers: forward, 5′-ggc gaa ttc gac atg gct gtc agc gac-3′ and reverse, 5′-gcg gga tcc tta aaa atg cct ctt cat gtg-3′. The KLF4 DNA fragment was subsequently cloned into the EcoRI and BamHI sites of the internal ribosome entry site vector pIRES2-AcGFP (Clontech, Mountain View, Calif) to generate the pIRES2-AcGFP-KLF4 recombinant plasmid.
To generate plasmids that express KLF4-specific short hairpin RNA (shRNA), 2 sequences of KLF4 were used as follows: gat cca ttc gcc cgc tca gat gaa ttc aag aga ttc atc tga gcg ggc gaa ttt a and gat ccc cta cac aaa gag ttc cca ttc aag aga tgg gaa ctc ttt gtg tag gtt a. A scrambled shRNA sequence was used as a negative control. Double-stranded DNA oligonucleotides were cloned into pSilencer4.1 neo vector according to the manufacturers' protocol (Ambion, Austin, Tex).
All transfection experiments were done using Lipofectamine 2000 reagent (Invitrogen) according to the manufacturer's instructions.
Analysis of p27Kip1 Promoter Activity
The p27Kip1 promoter constructs were originally obtained from Dr. Toshiyuki Sakai31 and Dr. Keping Xie.22 Plasmids containing firefly luciferase reporters were cotransfected into tumor cells in triplicate using Lipofectamine 2000, with the thymidine kinase promoter-Renilla luciferase reporter plasmid (pRL-TK) as an internal control. The activity of both firefly and Renilla luciferase reporters was determined 48 hours after transfection using the Dual Luciferase Assay kit (Promega, Madison, Wis). Specific promoter activity was presented as the relative ratio of firefly luciferase activity to Renilla luciferase activity. The specific promoter activity was presented as the change in the experimental group versus in the control group.
Quantitative Chromatin Immunoprecipitation
Chromatin immunoprecipitation (ChIP) assays were performed as previously described.22 For quantitative ChIP analysis, regions of interest were amplified from precipitated samples by real-time polymerase chain reaction (PCR). Each sample was assayed in triplicate, and the amount of precipitated DNA was calculated as the percentage of the input sample.32 The primers used in quantitative ChIP assays are listed in Table 1.
Table 1. Sequences of the Oligonucleotides for Quantitative ChIP
Cells in the exponential growth phase were harvested for inoculation. Tumor cells (1 × 105 or 1 × 104) were injected into the subcutis on the dorsum of 4- to 6-week-old female BALB/c-nude mice (6 mice per group). The tumor volume (V) was determined by the length (a) and width (b) as V = ab2/2.33 The experimental protocols were evaluated and approved by the Animal Care and Use Committee of the Medical School of Xi'an Jiaotong University.
Total RNA from cervical carcinoma cells was extracted using the TRIzol Reagent (Invitrogen). Total cDNA was used as a template for PCR amplification of KLF4 with β-actin as internal control. Real-time quantitative PCR was performed in triplicate for each primer set and in each cell sample, using an iQ5 multicolor real-time PCR Detection System (Bio-Rad, Hercules, Calif). The protocol for real-time PCR was 1 cycle of 95°C for 30 seconds, 40 cycles of 95°C for 5 seconds, 60°C for 30 seconds, and then a dissociation stage. The cycle threshold value was determined as the point at which the fluorescence exceeded a preset limit determined by the instrument's software.
Statistical analysis was performed with SPSS 16.0 software (SPSS Inc., Chicago, Ill). The 2-tailed chi-square test was used to determine the significance of the differences between the covariates. For 2-group analyses, the Student t test was used to determine the statistical significance. To examine the relationship between 2 quantitative variables, Pearson linear regression analysis was performed. In all tests, P < .05 was defined as statistically significant.
KLF4 Expression in Normal Cervixes and in Different Cervical Lesions
KLF4 expression has been shown to be both down-regulated21, 23, 24, 34 and up-regulated35, 36 in human tumors. To determine the expression of KLF4 protein in human cervical carcinoma, we did immunohistochemistry with paraffin-embedded normal cervix, cervical carcinoma in situ, and cervical carcinoma tissues. KLF4 was localized in the nucleus of all positive cells with different levels in various cervical tissues, including normal cervix (Fig. 1A, a1-a3), cervical carcinoma in situ (Fig. 1A, b1-b3), and cervical invasive carcinomas (Fig. 1A, c1-c3). In normal cervical epithelia, the basal undifferentiated cells, the only cells with proliferative ability, had almost undetectable KLF4 staining, whereas the differentiated epithelial cells without proliferative ability had a relatively high level of KLF4 expression (Fig. 1A, a1). These results suggest that KLF4 is not required in the cell proliferation of normal cervix.
To determine whether KLF4 is involved in the development of cervical carcinoma, we compared the KLF4 IHC scores among normal cervix, carcinoma in situ, and cervical carcinoma tissue specimens. The number of specimens with negative or weak positive staining was gradually increased from normal cervix to cervical carcinoma in situ and then to cervical invasive carcinoma. However, the number of specimens with strong positive staining was decreased from 83.72% in normal cervixes to 53.57% in cervical carcinoma in situ and ultimately to 6.25% in cervical invasive carcinoma (Table 2 and Fig. 1B). The average IHC score of KLF4 staining was 9.30 ± 2.85 in normal cervix (n = 43), 6.29 ± 3.28 in carcinoma in situ (n = 28), and 2.45 ± 2.94 in invasive carcinoma (n = 64; P < .05 by 2-tailed t test; Fig. 1C). Furthermore, the expression of KLF4 was detected by Western blot assays in 10 normal cervices and 18 invasive cervical carcinomas (Fig. 1D), with an average relative KLF4 expression of 0.32 ± 0.08 in the 18 invasive cervical carcinomas but 2.05 ± 0.40 in the 10 normal cervices (Fig. 1E). The normal cervix tissues had 6.67× the KLF4 expression as the invasive cervical carcinoma tissues did (P < .01). Both Western blot and IHC semiquantitative analyses consistently support the notion that the loss of function of KLF4 as a tumor suppressor may stimulate the development of cervical cancer.
Table 2. KLF4 Expression Levels in Different Tissue Specimens
Negative, No. (%)
Weak, No. (%)
Strong, No. (%)
Abbreviation: KLF4, Krüppel-like factor 4.
Pearson 2-tailed chi-square test was done with the SPSS software program to determine the statistical significance of the level of expression of KLF4 in different tissue specimens.
We also evaluated the expression of KLF4 in cervical cancer cell lines by immunocytochemistry, real-time PCR, and Western blotting (Fig. 2A and B). We detected a high level of KLF4 expression in HeLa cells but a low level in SiHa cells. In C33A cells, KLF4 expression was not detectable by either method.
To determine whether KLF4 affects the proliferation of these cervical cancer cell lines, we overexpressed exogenous KLF4 in SiHa and C33A cells by stable gene transfection (Fig. 2C) and knocked down the expression of KLF4 in HeLa cells by small interfering RNA (Fig. 2F). The SiHa and C33A cells overexpressing KLF4 (SiHa-KLF4, C33A-KLF4) showed significantly lower proliferation ability than the control cells (SiHa-GFP, C33A-GFP), as measured by both cell growth curve assay (P < .005; Fig. 2D) and cell viability assay (P < .001; Fig. 2E). Meanwhile, the KLF4 knockdown HeLa cells (HeLa-shKLF4-1 and -2) had much lower proliferation ability than the control cells (HeLa-shControl), as measured by both cell growth curve assay (P < .05; Fig. 2G) and cell viability assay (P < .05; Fig. 2H). These results suggest that KLF4 inhibits the growth of cervical cancer cells, despite of the difference in the expression levels of endogenous KLF4.
KLF4 Suppresses the Tumor Formation of Cervical Carcinoma Cells
To investigate whether KLF4 also affects tumor formation ability, xenograft assays were performed with nude mice, and the development and growth of solid tumors were monitored. The palpable tumor formation of SiHa-GFP cells and SiHa-KLF4 cells occurred at the same time after inoculation. However, tumor development with SiHa-KLF4 cells was significantly slower, with both amounts of cells used for inoculation (1 × 105 or 1 × 104 cells; P < .01; Fig. 3A and B). Similar data were obtained from C33A-GFP and C33A-KLF4 cells (data not shown). However, the tumor formation of HeLa-shKLF4 cells occurred earlier and progressed much faster than that of HeLa-shControl cells when 1 × 105 cells were inoculated (P < .01; Fig. 3C). When fewer cells (1 × 104) were inoculated, the HeLa-shKLF4 cells formed small tumors, and the HeLa-shControl cells did not form any tumors (Fig. 3D).
We then examined the expression of KLF4 and Ki67 proteins by immunohistochemistry in all xenograft tumor tissues formed by the SiHa-KLF4, SiHa-GFP, HeLa-shKLF4, and HeLa-shControl cells (Fig. 3E). Tumor tissues formed by SiHa-KLF4 cells expressed much more KLF4 but much less Ki67 than those formed by SiHa-GFP cells. Meanwhile, tumor tissues formed by HeLa-shKLF4 cells expressed much less KLF4 but much more Ki67 than those formed by HeLa-shControl cells. These results demonstrate that KLF4 suppresses the tumor formation and development of cervical cancer, and this suppression may be caused by the inhibited cell proliferation.
KLF4 Blocks G1/S Phase Cell Cycle Transition of Cervical Cancer Cells
It has been reported that KLF4 inhibits the proliferation of the colon cancer cell line by blocking the G1/S phase transition of the cell cycle.37 We tested whether it also occurs in cervical cancer cells with FACS. As shown in Figure 4A and B, the percentage of SiHa-KLF4 cells in G0/G1 phase increased significantly to 66.98%, together with a decrease in the percentage of S phase cells to 22.56% (Fig. 4A and B). The ratio of G1 phase to S phase SiHa-KLF4 cells (66.98%/22.56% = 2.96) was much higher than that of SiHa-GFP cells (46.87%/34.80% = 1.34), suggesting that KLF4 overexpression induced cell cycle arrest in G1/S phase transition. A similar effect was observed in C33A-KLF4 cells (Fig. 4C and D). Consistently, knockdown of KLF4 in HeLa cells (HeLa-shKLF4) resulted in a decrease of G1/S ratio (54.86%/44.44% = 1.23), compared with that of the HeLa-shControl cells (67.94%/27.86% = 2.43; Fig. 4E and F), suggesting that knockdown of KLF4 promoted the G1/S transition in HeLa cells.
KLF4 Positively Regulates p27Kip1 Expression
We analyzed the expression levels of a number of cell cycle-related genes. Real-time PCR revealed that the cyclin-dependent kinase inhibitor p27Kip1 had the most significant up-regulation in SiHa-KLF4 cells (Fig. 5A). Furthermore, the p27Kip1 levels were increased in KLF4 overexpression cells but decreased in KLF4 knockdown cells, as revealed by both Western blot analysis and immunocytochemistry or immunohistochemistry (Fig. 5B and C). The expression levels of p27Kip1 were also positively correlated with those of KLF4 in the xenograft tumor tissues (Fig. 5D). These results suggest that KLF4 inhibits cell growth by up-regulating the expression of p27Kip1.
KLF4 Binds to the Special Region of p27Kip1 Promoter In Vivo
It has been found that p27Kip1 promoter contains 3 KLF4-binding motifs (Fig. 5E), and that KLF4 transactivates p27Kip1 expression in human pancreatic cancer.22 To identify the relationship between KLF4 and p27Kip1 protein in cervical carcinoma, we transfected the p27Kip1 promoter-luciferase constructs into SiHa-KLF4 and SiHa-GFP cells. The p27 SacII construct with the shortest fragment containing all 3 KLF4 motifs in p27Kip1 promoter showed the strongest luciferase signals in SiHa-KLF4 cells, compared with the SiHa-GFP cells (Fig. 5F), suggesting that KLF4 transactivates p27kip1 expression through the specific KLF4-binding motifs in p27kip1 promoter in cervical carcinoma.
Furthermore, we used quantitative ChIP assays to identify the specific region of p27Kip1 promoter that KLF4 binds to. Three pairs of primers were designed to amplify the specific regions (p1, p2, and p3; Fig. 5E), and 1 of these pairs was used to amplify a 150-bp fragment in the p27Kip1 3′-untranslated region as a control. From the immunoprecipitate by the KLF4 antibody, p1, p2, and p3 were amplified >4 to 5× more from SiHa-KLF4 cells than from SiHa-GFP cells (Fig. 5G). However, no significant differences were observed in the control IgG immunoprecipitate for all primers (Fig. 5H), demonstrating the specificity of KLF4's binding to the p27Kip1 promoter in cervical carcinoma cells.
In the present study, we investigated the expression and potential role of KLF4 in the development and progression of cervical carcinoma. KLF4 protein was down-regulated during the development of cervical carcinoma from normal cervix in situ, and during the progression of cervical carcinoma in situ to invasive cervical carcinoma, as revealed by IHC semiquantitative and Western blot analysis (Fig. 1). KLF4 also inhibited the proliferation of cervical carcinoma cell lines SiHa, C33A, and HeLa in vitro by blocking G1/S cell cycle transition (Fig. 2 and 4). Xenograft assays with nude mice revealed that KLF4 may suppress the formation of cervical tumors by inhibiting cell growth (Fig. 3). Furthermore, KLF4 protein transactivated the expression of p27Kip1 protein, by which the cell growth and tumor formation were inhibited. Collectively, these results indicate that KLF4 is a potential tumor suppressor in cervical carcinoma.
Cervical carcinoma arises in women who present a persistent infection with HR-HPV with a multistage process of carcinogenesis.38 A progression of intraepithelial lesions from slight to marked, and furthermore to invasive cancer, has been postulated.39 For example, CINIII, also called the premalignant phase of cervical carcinogenesis or carcinoma in situ, which is a precursor for lesion detection in screening programs, can progress to invasive cancer. In our study, the expression of KLF4 was gradually decreased from 83.72% in normal cervixes to 53.57% in cervical carcinoma in situ and ultimately to 6.25% in cervical invasive carcinoma. The average KLF4 staining score was 9.30 ± 2.85 in normal cervix (n = 43), 6.29 ± 3.28 in carcinoma in situ (n = 28), and 2.45 ± 2.94 in invasive carcinoma (n = 64), respectively. These results demonstrate that KLF4 functions in the cervical cancer progression.
KLF4 has been implicated in the regulation of the differentiation of pancreatic ductal cells40, 41 and colorectal epithelium cells.12 In normal cervical epithelia, the basal undifferentiated cells with proliferative ability showed very low-level or even undetectable KLF4 expression, although the differentiated epithelial cells had very strong KLF4 staining (Fig. 1A, a1-a3). This indicates that KLF4 has a more important role in cell differentiation in normal cervix. KLF4 was highly expressed in the basal cells of normal cervix, but gradually decreased in cervical carcinoma in situ, as well as in invasive cervical carcinomas. These results suggest that KLF4 may function as a tumor suppressor gene in the development (normal cervix vs carcinoma in situ; P < .05) and progression (carcinoma in situ vs cervical carcinoma; P < .05) of cervical carcinoma (Fig. 1). Whether KLF4 inactivation in cervical carcinoma is triggered by genetic or epigenetic mechanisms is being investigated in another related project of ours.
Although KLF4 is highly expressed in human cervical carcinoma HeLa cells, it is almost completely inactivated in SiHa and C33A cells. Exogenously expressed KLF4 in SiHa (SiHa-KLF4) suppressed the tumorigenicity of SiHa cells, and knockdown of KLF4 in HeLa enhanced the tumorigenicity of HeLa cells, indicating that KLF4 works as a tumor suppressor gene in both SiHa and HeLa cells. All of these results support the notion that KLF4 is a tumor suppressor in the pathogenesis of cervical carcinoma.
Several lines of evidence suggest that KLF4 is a negative regulator of cell proliferation and an inducer of epithelial terminal differentiation.42 KLF4 inhibits cell proliferation through up-regulating the inhibitors of proliferation, while down-regulating the genes that promote proliferation.18 In this study, we also found that KLF4 inhibited cell growth by blocking the G1/S transition in 3 different cervical cancer cell lines, SiHa, C33A, and HeLa. Xenograft assays with nude mice revealed that KLF4 may suppress the tumor formation of cervical cancer cells by inhibiting cell growth. Furthermore, we found that KLF4 up-regulated the expression of p27Kip1 and, to a lesser extent, the expression p21CIP1, but down-regulated the expression of cyclinD1 (data not show). Previous studies have reported that KLF4's regulation of p21CIP1 expression was through a specific Sp1-like cis-element in the proximal promoter of p21CIP1.16, 20, 25, 43 Interestingly, the same element is also required for p53 to activate p21CIP1 expression, although p53 does not bind to this element.44 Similarly, KLF4 negatively regulates cyclin D1 expression through its binding to the Sp1 motif in the proximal region of cyclinD1 promoter.45 In this study, promoter analysis revealed that KLF4 activated p27Kip1 expression by directly binding to the proximal CACCC consensus sequence and the CG-rich regions of the p27Kip1 promoter in cervical carcinoma cells (Fig. 5E-H). These results are consistent with the findings of Xie's group on KLF4 in human pancreatic cancer cells.22 p27Kip1 is a universal cyclin-dependent kinase inhibitor that directly inhibits the activity of cyclin-CDK complexes, resulting in cell cycle arrest at G0/G1 phase.46 Therefore, the results from our study revealed that the cell growth retardation induced by KLF4 in cervical carcinoma is probably through activating the cell cycle inhibitor p27Kip1.
In this report, we provide evidence to support the hypothesis that KLF4 is a tumor suppressor in cervical carcinoma. We also confirmed the positive regulation of p27Kip1 by KLF4 in cell proliferation, especially the G1/S transition. Our study points in the direction of restoring the expression of KLF4 and/or down-regulating the expression of p27Kip1 as a new method to treat certain types of cervical cancers.
We thank Dr. Keping Xie and Dr. Daoyan Wei (The University of Texas MD Anderson Cancer Center, Houston, Texas) and Dr. Toshiyuki Sakai (Kyoto Prefectural University of Medicine, Kyoto, Japan) for providing the p27Kip promoter constructs.
This work was supported by a general grant (No. 30571951) and a special scientific Distinguished Young Scientists Fund (No. 30725043) to P.-S.Z. from the National Natural Science Foundation of China.