Tumor suppressor Activity of KLF6 Mediated by Downregulation of the PTTG1 Oncogene

Abstract

The tumor suppressor Kruppel-like Factor 6 (KLF6) is frequently inactivated in hepatocellular carcinoma (HCC). To unearth downstream transcriptional targets of KLF6, cDNA microarray analysis of whole liver was compared between KLF6 +/+ and KLF6 +/− mice. Pituitary Tumor Transforming Gene 1 (PTTG1), an oncogene, was the most up-regulated transcript in KLF6 +/− liver. In human HCCs, KLF6 mRNA was significantly decreased, associated with increased PTTG1. In HepG2, KLF6 transcriptionally repressed PTTG1 by direct promoter interaction. Whereas KLF6 downregulation by siRNA increased HepG2 proliferation, siRNA to PTTG1 was anti-proliferative. PTTG1 downregulation represents a novel tumor suppressor pathway of KLF6.

Introduction

KLF6 is a ubiquitously expressed zinc finger transcription factor and tumor suppressor gene inactivated in several human cancers [1] [2]. To unearth potential KLF6 targets, cDNA microarray was performed comparing the transcriptional profile of liver from KLF6 (+/−) mice to livers from KLF6 (+/+) mice. The microarray analysis uncovered PTTG1 as the most highly upregulated gene in KLF6+/− livers compared with control. PTTG1, also known as securin, is a potent oncogene and mitotic checkpoint protein that inhibits sister chromatid separation [3,4].

The potential importance of both KLF6 and PTTG1 in tumorigenicity is reinforced by evidence that both play a role in HCC, through over-expression of PTTG and reduced expression of KLF6 [5,6]. The PTTG1 promoter contains GC-rich regions, which represent potential KLF6 binding sites. In the present study, we explored the potential relationship between these two molecules. Our results demonstrate direct regulation of PTTG1 by KLF6, and provide an additional mechanism whereby KLF6 displays tumor suppressor activity.

Materials and Methods

Array Analysis

Complementary DNA microarray analysis of gene expression was performed as described previously [7]. The Oncomine database (http://www.oncomine.org/) was also used to assess KLF6 expression in the prostate cancer microarray data set [7]. Primary analysis was done with the Genepix pro 4.0 software package (Molecular Devices). Cy3/Cy5 ratios were determined for KLF6, along with various other quality control parameters (e.g., intensity over local background). The normalized median of ratio values of the KLF6 and PTTG gene was log2 transformed, filtered for presence across arrays, and selected for expression levels and patterns depending on the experimental set.

Human HCC tissue samples

HCC samples from 16 patients and matched surrounding tissue (ST), were obtained, all with the approval of the Institutional Review Board (IRB) of all institutions involved, as described previously [8].

RNA Extraction

HCC RNA was extracted using Trizol reagent™ followed by column purification (Qiagen). RNA was assessed for quality by agarose electrophoresis and accepted for analysis only if the 28S/18S ratio was >0.7.

Cell line and mouse liver RNA was extracted using RNeasy™ mini kit (Qiagen) and treated with DNAse (Qiagen). A total of 1 μg of RNA was reverse transcribed per reaction using first strand complementary DNA synthesis with random primers (Promega).

Quantitative real-time PCR (QRT-PCR)

Samples were analyzed with SYBR green-based QRT-PCR and performed on an ABI PRISM 7900HT Sequence Detection System. The following primers were used: hKLF6 forward 5′ CGG ACG CAC ACA GGA GAA AA 3′ and hKLF6 reverse 5′CGG TGT GCT TTC GGA AGT G 3′; mKLF6 forward 5′ GAG TTC CTC CGT CAT TTC CA 3′ and mKLF6 reverse 5′ GTC GCC ATT ACC CTT GTC AC 3′; hPTTG1 forward 5′ GGC TTT GGG AAC TGT CAA CAG GAG ′3 and hPTTG1 reverse 5′ GGC ATC ATC TGA GGC AGG AAC AG 3′; mPTTG1 forward 5′ CTG GGC ACT GGT GTC AAG 3′ and mPTTG1 reverse 5′ GCTGTTTTGGTTGGAGGGG 3′; h/mGAPDH forward 5′ CAA TGA CCC CTT CAT TGA CC 3′ and h/mGAPDH reverse 5′ GAT CTC GCT CCT GGA AGA TG 3′. All experiments were done in triplicate and independently validated 3 times. All values were normalized to GAPDH levels.

Cell Culture

HepG2 cells were obtained from American Tissue Culture Collection (ATCC, Manassas, VA) and were cultured as described [6].

Transfection Assays

Transient transfection was performed with Fugene reagent according to the manufacturers' protocol (Roche, Indianapolis, IN)). Vectors for PCIneo-KLF6 overexpression or pSuper-KLF6 knockdown were used as previously described [9,10]. For luciferase assay, HepG2 cells cultured in 12-well plates (Corning Glass) were transfected with 1ug μg of pCIneo empty vector or pCIneo-KLF6, with the different PTTG1 promoter constructs as previously described [11]. Twenty four hours after transfection, luciferase activity was assayed in 10 μl of lysate using the dual-luciferase reporter assay system and a Dynex luminometer. Transfection efficiency was normalized to Renilla luciferase activity.

Western Blot

Cell extracts and mouse liver tissue for Western blotting were harvested in 1X lysis buffer according to standard protocols using 60 μg protein per lane. Western blotting was performed using a rabbit polyclonal antibody to KLF6, GAPDH (SC-7158, SC-25778, respectively; Santa Cruz Biotechnology Inc.), and PTTG1 (34-1500; Zymed Inc).

Chromatin Immunoprecipitation Assay

Chromatin immunoprecipitation assays were performed using a commercial kit according to the manufacturer's instructions (Active Motif – ChIP-IT™ Express. Proteins cross-linked to DNA were immunoprecipitated with 10 μg of anti-KLF6 antibody (R-173) (Santa Cruz Biotechnology), anti-Histone H3 rabbit antiserum (07-690) (Upstate Biotechnology, Inc.), or control IgG (sc-2027) (Santa Cruz Biotechnology) and protein-G-magnetic beads. Genomic sequence primers encompassing the promoter region -246 to -401 upstream of transcriptional start site were used to amplify immunoprecipitated DNA: forward, 5′- CGG CTG TTA AGA CCT GCG TGA GTG -3′; reverse, 5′- AAC GGG CTC GGA GCC ATC TAA G -3′.

Analysis of Proliferation.

Proliferation was determined by assaying [³H]thymidine incorporation as previously described[2].

Results

PTTG1 mRNA expression is increased in KLF6 (+/−) mouse livers

Livers from either KLF6 +/+ or KLF6 +/− mice previously generated by homologous recombination [12] were harvested and cDNA microarray analysis was performed to screen for genes differentially regulated with loss of a KLF6 allele. Among transcripts that were upregulated, the most significantly increased was Pituitary Tumor Transforming Gene 1 (PTTG1). To validate these results, we performed qRT-PCR on 24 additional KLF6 +/+ and +/− mouse livers. Along with the expected 50% reduction of KLF6 mRNA (Figure 1a) PTTG mRNA was increased 82% in KLF6 +/− livers (Figure 1b).

Fig. 1. KLF6 and PTTG1 mRNA levels in KLF6+/− mouse livers.

Livers were harvested from 9 WT and 9 Het mice and analyzed by QRT-PCR. (A) QRT-PCR with wildtype KLF6 primers resulted in an average 34% decrease in KLF6 transcript in the KLF6+/− mice. (B) The same samples had an average 82% increase in PTTG1 transcript when compared to wildtype control animals.

PTTG1 mRNA is over-expressed while KLF6 mRNA is decreased in HCCs compared to surrounding liver.

PTTG1 and KLF6 mRNA expression was analyzed by qRT-PCR in HCC patient samples and compared to surrounding non-malignant liver tissue. Of the 16 samples analyzed, there was an average 70% decrease in KLF6 mRNA in tumors compared to matched surrounding tissue. This decrease is consistent with our previous report [8]. Importantly, in these same samples PTTG1 was increased an average of 200% (Figure 2). Of the 16 samples, in 75% PTTG1 mRNA was increased, while the remaining 25% showed no change.

Fig. 2. KLF6 and PTTG1 mRNA levels in HCC patient samples compared with surrounding tissue.

Liver samples were analyzed from 18 patients with HCC by QRT-PCR. On average, there was a significant decrease of 70% (P < 0.0001) in levels of KLF6 transcript while PTTG1 levels were significantly up 2-fold (P < 0.001)

Reciprocal expression of KLF6 and PTTG1 in HepG2 cells

KLF6 transcriptionally regulates gene expression through binding to GC-boxes in responsive promoters. We explored KLF6 regulation of PTTG1 by transiently over-expressing KLF6 cDNA or siRNA in the HepG2cell line. A 12-fold over-expression of KLF6-pBabe in HepG2 resulted in 24% decease of PTTG1mRNA assessed by qRT-PCR (Figure 3a) and approximately 50% decrease in PTTG1 protein level (Figure 3b). Conversely, knockdown of KLF6 with siRNA resulted in 57% decrease in KLF6 mRNA, with a corresponding 3.7-fold increase in PTTG1 (Figure 4a), paralleling changes in expression of PTTG protein by Western (Figure 4b). Together these results indicate that KLF6 regulates expression of PTTG1.

Fig. 3. PTTG1 is downregulated with KLF6 overexpression.

HepG2 cells were transfected with KLF6 and RNA and protein collected after 24hrs. (A) QRTPCR on cDNA showed a 12-fold over-expression of KLF6 with a 24% decease of PTTG1 (P < 0.0005). (B) 60ug of protein from total cell extracts were loaded on 10% SDS-PAGE gel and submitted to western blot analysis for KLF6 and PTTG1. GAPDH served as the protein loading control. There was approximately 50% decrease in PTTG1 protein with several fold overexpression in KLF6. The data presented in (B) is representative of at least 3 independent experiments.

Fig. 4. PTTG1 is upregulated with KLF6 knockdown.

HepG2 cells were transfected with siKLF6 or siControl and RNA and protein collected after 24hrs. (A) QRTPCR on cDNA showed 57% decrease in KLF6 mRNA, with a corresponding 3.7-fold increase in PTTG1. (B) Protein extracts were submitted to Western Blot analysis and approximately a 50% decrease in KLF6 protein levels resulted in approximately 2-fold overexpression of PTTG1. The data presented in (B) is representative of at least 3 independent experiments.

KLF6 transcriptionally represses PTTG1 promoter activity through direct promoter interaction

Several GC-boxes in the PTTG1 promoter have been previously characterized in response to Sp1 and NF-Y transactivation, but also represent putative binding sites for KLF6. To explore this possibility, PTTG1 promoter mutation constructs (obtained as a gift from Dr. Xun Zhang) (Figure 5a) were transfected into HepG2 cells, with or without KLF6 cDNA co-transfection. Over-expression of KLF6 repressed the ‘p710’ PTTG1 promoter construct activity by 45%, while mutation of GC-box-mt2 abrogated this repression (Figure 5b). These results suggested that KLF6 directly trans-represses the promoter of PTTG1. To confirm direct binding of KLF6 to the endogenous PTTG1promoter, ChIP was performed using nuclear extracts from HepG2 cells. Binding of KLF6 was evident in the region -246 to -401 nt, which included a GC-box (Figure 6). Taken together, the data indicate that KLF6 can functionally repress the PTTG1 gene through direct binding to the PTTG1 promoter.

Fig. 5. PTTG1 promoter activity is decreased with KLF6 over-expression.

HepG2 cells were transiently transfected with either pGL3 (promoter-less vector), p710 (minimal promoter), GC-mt1, GC-mt2, GC-mt3, GC-mt4 (each representing a mutation of a GC-box in the promoter), a combination of GC-mt2 and GC-mt3, or all GC boxes mutated. Cells were also transfected with either KLF6 or PCIneo control to assess the ability of KLF6 to repress the PTTG1 promoter. (A) Schematic of PTTG1 promoter representing the GC-boxes mutated in these experiments. (B) KLF6 represses the p710 PTTG1 promoter by 30%, which was abrogated with mutation of GC-mt2.

Fig. 6. Direct interaction of KLF6 with PTTG Promoter by ChIP.

HepG2 cells were cross-linked with formaldehyde, chromatin sheared, and immunoprecipitated with KLF6 antibody, non-specific IgG, or positive control H3 antibody. The primers used span the region −401 to −246 of the PTTG1 promoter. ChIP analysis shown confirms KLF6 binding to PTTG1 promoter.

Knockdown of PTTG1 with siRNA partially abrogates the proliferation induced by KLF6 loss.

In order to establish the biological consequences PTTG1 repression by KLF6, proliferation assays were performed in HepG2 cells following KLF6 knockdown by siRNA. Decreased levels of full-length KLF6 promote cell proliferation, which has been ascribed in part to down-regulation of the cdk/cyclin inhibitor p21 [2], but upregulation of PTTG1 could represent an additional mechanism whereby KLF6 loss enhances cell growth. In order to test this hypothesis, we expressed siRNA to either KLF6, PTTG1, both KLF6 and PTTG1, or control siRNA in HepG2. DNA synthesis in cell expressing siKLF6 was increased relative to vector-expressing cells, whereas cells expressing siPTTG1 grew more slowly. However, when siPTTG1 was co-expressed with siKLF6, DNA synthesis was restored to baseline levels (Figure 7).

Fig. 7. Silencing of PTTG1 abrogates siKLF6 induced cell proliferation.

HepG2 cells were transfected with either siKLF6, siPTTG1, siControl, or a combination there of for 24, 48, and 72 hours. Cells were incubated with 1 μCi/ml [3H]thymidine for 2 hours before collection, and analyzed for radioactive thymidine incorporation. The 24hr time point is set to baseline. siPTTG1 shows a decrease in proliferation compared with siControl at both 48hr and 72hr (P < 0.05), whereas siKLF6 increases proliferation at both time points. Co-transfection of siPTTG1 and siKLF6 abrogat

Discussion

Hepatocellular carcinoma (HCC) is the fifth most frequent neoplasm worldwide with greater than 500,000 deaths annually, and the fasting rising malignancy in the US and Europe [13]. , Inactivation of several tumor suppressor genes including p53, Rb, and p16 have been documented in a proportion of HCCs [14]. More recently, reduced KLF6 expression has been observed in both HBV- and HCV-related HCCs [8].

Many cancers over-express PTTG including colon [15] and liver [4,5,16], among others. PTTG has been associated with tumor invasiveness and genetic instability[17], and can transform NIH-3T3 fibroblasts in a dose dependent manner leading to morphological changes, anchorage-independent growth in soft agar, and enhanced tumor formation in nude mice[18,19]. Recently, PTTG dysregulation has also been uncovered in HCC, with its increased expression associated with poor prognosis and increased intratumoral microvessel density [4].

Downregulation of PTTG1 by siRNA inhibits cell growth in vitro and in vivo in sarcomatoid HCC and hepatoma cell lines [5]. This tumor suppressive phenotype following PTTG1 knockdown may be partially mediated through increased p53 and apoptotic cell death [20]. Similarly, siRNA targeting of PTTG in lung tumors reduces tumor formation and decreases levels of Ki67, bFGF and CD34, which may also be relevant to the HCC phenotype[21]. In this study we show that loss of KLF6-mediated repression of PTTG1 may be an additional genetic defect in HCC. KLF6 loss is an early event in the progression of HCC, followed at later stages by a variable frequency of inactivating mutations. The loss of KLF6 and subsequent upregulation of PTTG1 could contribute to tumorigenesis through increased proliferation, angiogenesis, and possibly chromosomal instability, leading to a wide variety of mutations. Future studies focusing on the role of KLF6 in chromosomal instability mediated by PTTG1 yield further insights into this important pathway in tumorigenesis.