Oncogene (2007) 26, 4428–4434 & 2007 Nature Publishing Group All rights reserved 0950-9232/07 $30.00 www.nature.com/onc SHORT COMMUNICATION In vivo regulation of p21 by the Kruppel-like factor 6 tumor-suppressor gene in mouse liver and human hepatocellular carcinoma G Narla1,2,3,4,7, S Kremer-Tal1,3,7, N Matsumoto1,3, X Zhao1,3, S Yao5, K Kelley6, M Tarocchi1,3 and SL Friedman1,3 1Department of Medicine, The Mount Sinai School of Medicine, New York, NY, USA; 2Department of Human Genetics, The Mount Sinai School of Medicine, New York, NY, USA; 3Division of Liver Diseases, The Mount Sinai School of Medicine, New York, NY, USA; 4Division of Hematology/Oncology, The Mount Sinai School of Medicine, New York, NY, USA; 5Department of Oncological Sciences, The Mount Sinai School of Medicine, New York, NY, USA and 6Brookdale Department of Molecular, Cell and Developmental Biology, The Mount Sinai School of Medicine, New York, NY, USA Kruppel-like factor (KLF) 6 is a tumor-suppressor gene Kruppel-like factor 6 (KLF6) belongs to the Kruppel- functionally inactivated by loss of heterozygosity, somatic like family of transcription factors, which play roles in mutation and/or alternative splicing that generates a the regulationof diverse cellular processes including dominant-negative splice form, KLF6-SV1. Wild-type development, differentiation, proliferation and apopto- KLF6 (wtKLF6) expression is decreased in many human sis (Bieker, 2001). Functional inactivation of the KLF6 malignancies, which correlates with reduced patient gene occurs through several mechanisms, including loss survival. Additionally, loss of the KLF6 locus in the of heterozygosity (LOH), somatic mutationand/or absence of somatic mutation in the remaining allele occurs increased alternative splicing that yields a dominant- in a number of human cancers, raising the possibility that negative splice isoform, KLF6-SV1. KLF6 dysregula- haploinsufficiency of the KLF6 gene alone contributes tion has been demonstrated in a number of human to cellular growth dysregulation and tumorigenesis. Our cancers including prostate (Narla et al., 2001; Chen earlier studies identified the cyclin-dependent kinase et al., 2003), colorectal (Reeves et al., 2004), non-small- inhibitor p21 as a transcriptional target of the KLF6 cell lung (Ito et al., 2004), gastric (Cho et al., 2005), gene in cultured cells, but not in vivo. To address this nasopharyngeal (Chen et al., 2002), hepatocellular issue, we have generated two genetic mouse models to (Kremer-Tal et al., 2004) and ovarian carcinomas define the in vivo role of KLF6 in regulating cell (DiFeo et al., 2006b) as well as glioma (Jeng et al., proliferation and p21 expression. Transgenic overexpres- 2003). Furthermore, decreased KLF6 mRNA expression sion of KLF6 in the liver resulted in a runted phenotype is associated with reduced patient survival in prostate with decreased body and liver size, with evidence of (Singh et al., 2002; Glinsky et al., 2004) and lung cancers decreased hepatocyte proliferation, increased p21 and (Kettunen et al., 2004). Interestingly, reconstitution of reduced proliferating cell nuclear antigen expression. In KLF6 decreases cell proliferationandreverts tumor- contrast, mice with targeted deletion of one KLF6 allele igenicity in glioblastoma cell lines (Kimmelman et al., (KLF6 þ /À) display increased liver mass with reduced 2004). p21 expression, compared to wild type littermates. Depending on the cell type and context, KLF6’s Moreover, in primary hepatocellular carcinoma samples, growth-suppressive properties have beenassociated with there is a significant correlation between wtKLF6 and p21 key pathways disrupted in human cancer, including p53- mRNA expression. Combined, these data suggest that independent upregulation of p21 (Narla et al., 2001), haploinsufficiency of the KLF6 gene may regulate cellular reduced interaction of cyclin D1 with CDK4 (Benzeno proliferation in vivo through decreased transcriptional et al., 2004), induction of apoptosis (Ito et al., 2004) and activation of the cyclin-dependent kinase inhibitor p21. inhibition of c-jun (Slavin et al., 2004). Recently, a single Oncogene (2007) 26, 4428–4434; doi:10.1038/sj.onc.1210223; nucleotide polymorphism in the KLF6 gene has been published online 5 February 2007 associated with increased prostate cancer risk (Narla et al., 2005). Keywords: KLF6; Kruppel-like factor; tumor-suppressor Overall, the majority of the published data to date gene; p21; haploinsufficiency report frequent LOH of the KLF6 gene locus in primary hepatocellular carcinoma (HCC) patient-derived sam- ples. KLF6 LOH was reported in36% of 14 informative Correspondence: Dr S Friedman, Department of Medicine, Box 1123, HCC patient samples and somatic mutations were Mount Sinai School of Medicine, 1425 Madison Avenue, Room 11-70, detected inthree patientsamples (Wang et al., 2004). New York, NY 10029, USA. These findings complement our original report describ- E-mail: [email protected] ing frequent loss and somatic mutation in primary HCC 7These two authors contributed equally to the work. Received 19 June 2006; revised 6 November 2006; accepted 7 November tumor samples (Kremer-Tal et al., 2004). Ina separate 2006; published online 5 February 2007 study, somatic mutations were identified in 8.7% of In vivo regulation of p21 by KLF6 G Narla et al 4429 patient samples (Pan et al., 2006). Inaddition,LOH was Table 1 Features of WT and KLF6 TG mice reported in6.8% of tumors with nomutationor Wt (n ¼ 6) TG (n ¼ 6) promoter methylationina Koreancohort of HCCs (Song et al., 2006). Onthe other hand,two reports have Albumin(g/dl) 2.7 70.1 2.171.0 either failed to identify KLF6 mutations in HCC Total protein(g/dl) 5.2 70.1 3.870.5 ALT (U/l) 25762174 samples (Boyault et al., 2005) or did not find a decrease AST (U/l) 62736576 in KLF6 mRNA expressioninHCCs (Wang et al., Weight (g) 1170.7 871 2004). However, methodologic differences may account for these discrepant results (Narla et al., 2003). We and Abbreviations: KLF6, Kruppel-like factor 6; TG, transgenic; WT, others have demonstrated significant downregulation of wild-type. Clinical parameters, including serum analysis and body wtKLF6 expressionby both quantitativereal-time weight for TG mice (six) and WT littermates (six) are provided. Serum proteinandalbuminlevels are reduced inthe TG compared with WT. polymerase chain reaction (qRT–PCR) analysis and In addition, there was no evidence of inflammation in either of the microarray studies inboth HBV- andHCV-derived groups to account for these changes as there is no difference in ALT HCC patient samples (Lee et al., 2004; Kremer-Tal and AST levels. Body weight is reduced in the TG mice compared with et al., 2006). Although the frequency of somatic WT (Po0.01). mutationinthe KLF6 gene is quite variable, the bulk of evidence from published studies supports a role for the KLF6 tumor-suppressor gene in the development proliferating cell nuclear antigen (PCNA) in hepatocytes and progression of HCC, through either KLF6 loss and/ of 4-week-old TG pups was markedly diminished, or somatic mutationanddecreased wtKLF6 expression. consistent with reduced hepatocyte proliferation (Figure Interestingly, there are a number of tumor types in 1c and d). There was no increase in cellular apoptosis as which loss of one KLF6 allele occurs inthe absenceof assessed by TdT-mediated dUTP nick end labeling somatic mutation in the remaining allele, including (TUNEL) (data not shown). Correlating with the glioblastoma, ovarian, gastric, and head and neck decreased liver mass, TG mice yielded B50% fewer squamous cell cancers. This finding raises the possibility hepatocytes than their non-TG littermates, following that haploinsufficiency of the KLF6 gene alone might cell isolation using standard methods – this difference contribute to increased cellular proliferation and tumor was a result of decreased cellular proliferation, as development in vivo. To explore this possibility, and to assessed by 3H thymidine incorporation and PCNA investigate the in vivo biologic activity of KLF6, we staining. No differences in apoptosis or cellular viability generated transgenic (TG) mice with hepatocyte-specific were noted between TG and WT-derived hepatocytes, as overexpressionof KLF6 by usinga well-validated TG measured by TUNEL staining and fluorescence-acti- construct (Wu et al., 1996) inwhich the humanKLF6 vated cell sorting analysis. Of note, the altered weight, cDNA was cloned downstream of the transthyretin histology and KLF6 expression were confined only to (TTR) promoter. Three independent lines of mice, the liver and there were no differences in these features TTR1-KLF6, TTR4-KLF6 and TTR9-KLF6, were inanyother tissues. generated with modest (Btwo- to threefold) but Because anantiproliferative effect of KLF6 was reproducible expressionof KLF6. Expressionof the apparent in the hepatocytes of TG mice, and as we transgene is confined specifically to hepatocytes with had previously established that KLF6 transactivates p21 variable expressioninthe choroid plexus of the brainat independent of p53 (Narla et al., 2001), we examined the high transgene copy number (Wu et al., 1996) (data not expression of p21, an inhibitor of several cyclin- shown). TG mice were analysed at 6 weeks of age as dependent kinases and a key regulator of the G1/S this is the period of rapid murine liver growth and transition (el-Deiry et al., 1993). By Westernblot, there differentiation (Walthall et al., 2005), and studies of was a threefold increase in KLF6 and a 10-fold