[CANCER RESEARCH 54, 1319-1323. March 1, 1994] Lowered Amounts of the Tissue-specific Transcription Factor LFB1 (HNF1) Correlate with Decreased Levels of Glutathione 5-Transferase a Messenger RNA in Human Renal Cell Carcinoma1 Annette Clairmont, Thomas Ebert, Heike Weber, Christiane Zoidl, Peter Eickelmann, Wolfgang A. Schulz, Helmut Sies, and Gerhart U. Ryffel2 institut fur Zellbiologie (Tumorforschung), Universitätsklinikum Essen, D-45122 Essen ¡A. C., H. W., C. Z., G. U. R.], and Urologische Klinik ¡T.E./ and institut für Physiologische Chemie I, ¡P. E., W. A. S., H. S./, Universität Düsseldorf,D-40225 Düsseldorf, Germany ABSTRACT Human renal cell carcinoma is characterized by the loss of differentia tion markers such as glutathione S-transferase a (GST-a). In this paper we show that the promoter of a GST-a gene contains a functional binding site for the cell-specific transcription factor LFB1 (HNF1). To investigate the potential role of LFB1 in the down-regulation of GST-a expression, we have compared the amount and the binding activity of the LFB1 protein between normal kidney and tumor tissue. By Western analysis and gel retardation assay using a monoclonal antibody specific for LFB1 we show that in 11 of 14 carcinomas the amount of LFB1 is clearly reduced com pared to the corresponding normal tissue and that in all 14 renal carci nomas LFB1 binding activity is diminished. As in the same samples the abundance of GST-a mRNA is lower than in the normal tissue, we pos tulate that the loss of LFB1 binding activity might be responsible for the decreased expression of the GST-a gene in renal cell carcinoma. INTRODUCTION LFB1 (also referred to as HNF1) has been identified as a transcrip tion factor recognizing a promoter element present in several genes specifically expressed in liver cells (1-3). Using the cloned comple mentary DNAas well as antibodies to determine the tissue distribution of LFB1, it has been established that LFB1 is not restricted to the liver but is also found in other tissues such as the intestine, stomach, and kidney (4-8). Thus far the potential target genes of LFB1 are not known in these nonhepatic tissues but it is generally assumed that LFB1 is involved in establishing the differentiated phenotype in these tissues. In the kidney in situ hybridization has shown that LFB1 mRNA is not uniformly distributed but is restricted to the proximal and distal tubules (9). Most renal cell carcinoma in humans is considered to be derived from the proximal tubule of the kidney. This assumption is based on electron microscopic (10) as well as immunological (11, 12) findings. On the other hand the analysis of the expression of specific genes in renal cell carcinoma revealed lower levels of differentiation markers such as mRNAs encoding pro-EGF3 (13), HER2/neu (14), and GST-a (15). All these changes have been interpreted as the consequence of a dedifferentiation program occurring during the development and pro gression of renal cancer. As the sequence analysis of the promoter of the human GST-a gene suggested the presence of a potential LFB1- binding site (16), we wondered whether the level of LFB1 is dimin ished in renal cell carcinoma. To address this question we have com pared in this paper the presence of LFB1 between normal kidney and renal carcinoma using a monoclonal antibody specific for LFB1 (7). Received 9/9/93; accepted 1/5/94. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ' This work was supported by the Deutsche Forschungsgemeinschaft (SFB 354 to G. U. R.) as well as by the Doktor-Robert-Pfleger Stiftung, Bamberg, and National Foundation for Cancer Research, Bethesda (to H. S.). 2 To whom requests for reprints should be addressed. 3 The abbreviations used are: EOF, epidermal growth factor; GST-a, glutathione 5-transferase a. MATERIALS AND METHODS Tissue Samples. Malignant as well as autologous normal kidney tissues were dissected from radical nephrectomy specimens immediately following removal of the kidney. Tissue cubes were transferred to liquid nitrogen, shock- frozen, and finally stored at -80°C.All tumors were histologically classified as renal cell carcinomas of similar tumor stage (pT2-3) (17) and tumor grade (2-3) (18). Preparation of Whole Cell Extract and Gel Retardation Assays. To prepare whole cell extracts from frozen tissue the material was minced, ho mogenized in 5 volumes of extraction buffer [20 ITIM4-(2-hydroxyethyl)-l- piperazineethanesulfonic acid (pH 7.8), 450 min sodium chloride, 0.2 mM EDTA, 0.5 mM dithiothreitol, 25% glycerine, and 10 mM disodium molybdate with 0.5 mM benzamidine, 25 fig/ml pepstatin A, and 25 fig/ml leupeptin], three times shock-frozen, and cleared by centrifugation for 15 min at 2°Cand 50,000 rpm. Rat liver nuclear extract was prepared essentially as described (19). Gel retardation assays were carried out as described previously (19) and the conditions for upshift experiments by the monoclonal antibody RADI are published (7). Cell Culture, Transient Transfection, Luciferase-Assay. SK-RC-47 cells (20) were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum, penicillin/streptomycin (100 units/ml), and 2.5 mM glutamine and transfected with lipofectin according to the manufacturer's recommendation (Life Technologies, Inc.). Cells (4 X 105/3.5 cm dish) plated 20 h prior to transfection were transfected using 5 j¿lipofectin, 1 ^g reporter DNA, and 400 ng DNA of the expression vector in 1 ml OptiMEM. T47D cells cultured in RPMI 1640 supplemented with 10% fetal calf serum, penicillin/streptomycin (100 units/ml), 1 mM glutamine, and bovine insulin (0.6 jj.g/ml) were transfected with the reporter constructs (1 ^ig) and the expression vector using DEAE-dextran (0.5 mg/ml) essentially as described (21). After 24 h 10-50 /j,g protein extract of SK-RC-47 or T47D cells were used for the luciferase assay (Luciferase-Assay-System; Promega). Western Blot Analysis. Aliquots of extracts were separated on an sodium dedecyl sulfate-polyacrylamide gel and electrotransferred to a nitrocellulose membrane. The blot was incubated overnight with the monoclonal antibody RADI specific for LFB1. A peroxidase-conjugated rabbit antibody to mouse immunoglobulin G was used as second antibody to visualize LFB1 by the enhanced chemiluminescence system (Amersham). The integrity of the pro teins was verified by staining of control gels. RNA Analysis. RNA extraction, preparation of polyadenylated RNA, and Northern blot analysis were as described (15). RESULTS The GST-a Promoter Contains a Weak LFB1 Binding Site. To analyze whether the human CST-A2 gene is regulated by the tissue- specific transcription factor LFB1 we transfected the luciferase re porter gene GST-luc containing the promoter sequence from -745 to -1 of the GST-A2 gene (16) into the human renal cancer cell line SK-RC-47 (20) that based on gel retardation assays and Western blots lacks any LFB1. Clearly cotransfection of the expression vector RV/ CMV (22) encoding the chimeric LFB1 with the acidic activation domain of VP16 resulted in about a 3-fold increase in the luciferase activity (Fig. 1). In control experiments with luciferase reporters con- 1319 Research. on November 18, 2015. © 1994 American Association for Cancer cancerres.aacrjournals.org Downloaded from