Transcriptional and translational downregulation of H ... - Nature

Oncogene (1998) 17, 1305 ± 1312  1998 Stockton Press All rights reserved 0950 ± 9232/98 $12.00 http://www.stockton-press.co.uk/onc

Transcriptional and translational downregulation of H-REV107, a class II tumour suppressor gene located on human chromosome 11q11-12 Knut Husmann1, Christine Sers1,2, Ellen Fietze2, Antoaneta Mincheva3, Peter Lichter3 and Reinhold SchaÈfer1,2 1

Division of Cancer Research, Department of Pathology, University of ZuÈrich, Schmelzbergstr. 12, CH-8091 ZuÈrich, Switzerland; Institute of Pathology, University Hospital ChariteÂ, Humboldt-University Berlin, Schumannstr. 20/21, D-10117 Berlin, Germany; 3 German Cancer Research Center, Im Neuenheimer Feld 280, D-69120 Heidelberg, Germany 2

The H-rev107 tumour suppressor was isolated as a gene speci®cally expressed in rat ®broblasts resistant toward malignant transformation by the activated HRAS gene (Sers et al., 1997; Hajnal et al., 1994). Here we describe the human homologue of the rat H-rev107 gene. The predicted rat and human proteins are highly conserved exhibiting an overall amino acid identity of 83%. The HREV107-1 gene is ubiquitously expressed with the exception of haematopoetic cells and tissues. In contrast, H-REV107-1 mRNA was found only in eight of 27 cell lines derived from mammary carcinoma, lung carcinoma, gastric carcinoma, kidney carcinoma, melanoma, neuroblastoma and other tumours. The H-REV107-1 protein was not detectable in any of these tumour cells. Loss of H-REV107-1 expression was not restricted to cultured human tumour cell lines, but also found in primary squamous cell carcinomas. Gross structural aberrations of the H-REV107-1 gene were absent in tumorigenic cell lines. Thus, the block to H-REV107-1 expression is achieved both at the level of transcription and translation. By ¯uorescence in situ hybridisation the human HREV107-1 gene was localised to chromosome 11q11-12. Keywords: expression genetics; cellular transformation; RAS; oncogenic signalling; transcriptional target

Introduction A limited number of tumour suppressor genes relevant for both hereditary and sporadic tumours were isolated until today. The predominant strategy for suppressor gene identi®cation has been positional cloning, guided by linkage analysis in families of patients predisposed for certain types of cancers, by cytogenetically detectable chromosomal lesions and by deletion analysis at the level of DNA. Usually, the causal role of candidate tumour suppressor genes, located in the vicinity of the speci®c chromosome deletion in the tumour, is veri®ed by a high prevalence of mutations that inactivate normal gene function. Numerous allelic losses found in human cancers serve as the hallmark of additional tumour suppressor genes, which have not yet been identi®ed at the molecular level (Sager, 1997; Stanbridge, 1991; Lasko et al., 1991). Two genes, Krev-1

Correspondence: R SchaÈfer KH and CS contributed equally to this paper Received 7 November 1997; revised 21 April 1998; accepted 21 April 1998

(Kitayama et al., 1989) and rsp-1 (Cutler et al., 1992), were isolated due to their ability of suppressing the transformed phenotype in cells transformed by oncogenes rather than by positional cloning. Moreover, class II tumour suppressor genes were identi®ed on the basis of distinct expression dierences between closely related normal and tumorigenic mammary cell lines. Their expression was blocked in the tumorigenic cells and the reconstitution of normal gene function was shown to suppress certain neoplastic properties (Zou et al., 1994; Sager et al., 1993; Lee et al., 1991). The tumour suppressor genes E-CAD (Schipper et al., 1991), VHL (Herman et al., 1994), p16/CDKN2/MTS1 (Merlo et al., 1995) and BRCA1 (Thompson et al., 1995) were shown to be inactivated either by an expression block or by mutation. These ®ndings support the notion that the downregulation of suppressor gene activity is of equal importance for the loss of growth constraints in tumours as is mutational inactivation. Consequently, tumour suppressor identi®cation by positional cloning based on chromosomal deletions will fail to detect critical genes inactivated by an expression block at the level of RNA or protein. For identifying candidate tumour-suppressing genes, we have used a set of closely related rat cell lines consisting of normal cells, HRAS-transformed tumorigenic cells and of phenotypic revertants (Griegel et al., 1986; SchaÈfer et al., 1988). RAS oncogenes frequently contribute to human neoplasia (Bos, 1989) and permanently activated RAS signalling inhibits the transcription of genes controlling growth-limiting functions (SchaÈfer, 1994). By subtractive cDNA cloning we recovered several structurally unrelated genes, designated H-rev, expressed in non-tumorigenic rat 208F ®broblasts, but not expressed in the FE-8 cell line, an HRAS transformed derivative thereof. The repressed H-rev gene activity was restored in a phenotypic revertant cell line derived from FE-8 cells (Kiess et al., 1995; Hajnal et al., 1993, 1994). At least two of the H-rev genes identi®ed so far are capable of suppressing the transformed phenotype. The H-rev142 gene (rrg-1/lox) encodes lysyl oxidase, a secreted enzyme involved in extracellular matrix stabilization (Kenyon et al., 1991; Contente et al., 1990; Hajnal et al., 1993). The expression of the lox gene can be enhanced by transforming growth factor-b (Boak et al., 1994; Shibanuma et al., 1993) and by the antioncogenic transcription factor IRF-1 (interferonregulatory factor-1) (Tan et al., 1996). H-rev107 was isolated as a novel gene associated with the resistance of cells toward neoplastic transformation by HRAS oncogenes and with density-dependent growth arrest

Downregulation of H-REV107 gene K Husmann et al

1306

(Hajnal et al., 1994). The H-rev107 gene is frequently downregulated in rat tumorigenic cell lines and in experimental tumours mediated by targeted expression of a HRAS transgene. Moreover, forced expression of H-rev107 cDNA controlled by heterologous promoters in de®cient rodent cell lines transformed by HRAS resulted in growth inhibition in vivo and in vitro (Sers et al., 1997). Interestingly, H-rev107 expression can also be up-regulated by interferon-g (Kuchinke et al., 1995). This suggests that one important function of the H-rev107 and lox/H-rev142 genes is to mediate the anti-proliferative eects of cytokines and negative growth factors. In view of the use of cytokines in cancer therapy (Gutterman, 1994), it is desirable to study the role of H-rev genes in human tumorigenesis. Here we describe the sequence, chromosomal localization, and expression pattern of the human H-REV107-1 gene. We also isolated a related gene, H-REV107-2. The H-REV107-1 gene is frequently downregulated at the level of transcription or translation in tumorigenic cell lines and in primary tumours.

Results Isolation, sequence analysis and expression of human H-REV107-1 cDNA Four clones isolated from a human liver cDNA-library with the rat H-rev107 cDNA as a probe were sequenced. The longest cDNA was 1064 bp in size and contained the complete open reading frame. Alignment of the human and the rat H-REV107 cDNA revealed an overall sequence identify of 82%. Translation of the human H-REV107-1 cDNA predicted a 162 amino acid protein, two amino acids longer than the rat protein and containing a conserved hydrophobic a-helix at the C-terminus (Figure 1). Comparison of rat and human amino acid sequences showed 83% identity. Protein pattern and motif

searches of the H-REV107-1 translation product identi®ed two potential protein kinase C phosphorylation sites (amino acids 55 and 71). The ®rst site is conserved between human and rat. The H-REV107-1 protein was detected on Western blots in total protein extracts prepared from COS-7 cells transiently transfected with the recombinant human H-REV107-1 cDNA expression vectors pCDNA3+871 and pCDNA3+971. The H-rev107 antiserum raised against the rat protein (Hajnal et al., 1994) detected a protein band of approximately 17 kD (Figure 4b). Expression of H-REV107-1 in normal tissues and tumour cell lines A 1.2 kb H-REV107-1 transcript was present in RNA from all tissues analysed with the exception of thymus and peripheral blood leukocytes. In addition, we observed a related 6.5 kb transcript in some tissues (Figure 2a). Next we investigated the expression of the human gene in tumorigenic cell lines derived from dierent tissues. The 1.2 kb H-REV107-1 mRNA was detected only in eight of 27 human tumour cell lines, while the related transcript of 6.5 kb was undetectable (Figure 3a). To rule out the possibility that the lack of H-REV107-1 expression is restricted to cell cultures, we analysed RNA from human primary tumours by RT ± PCR using H-REV107-1- speci®c primers. As shown in Figure 3b, the speci®c H-REV107-1 PCR product was not detectable in squamous cell carcinomas, while the corresponding histologically normal tissue was positive. To rule out gross structural alterations of the HREV107-1 gene in tumour cells, we compared DNA from ®ve tumour cell lines expressing H-REV107-1 mRNA with DNA from four negative cell lines by Southern blot analysis. Dierences between the two sets of cell lines were not found (not shown), suggesting that the H-REV107 gene is intact in these cells, although small deletions will remain undetected by this method.

Figure 1 Amino acid alignment of the predicted translation products of human and rat H-REV107-1 genes and the human HREV107-2 gene. Two potential protein kinase C phosphorylation sites (amino acids 55 and 71, respectively) are indicated in bold letters. The membrane-associated a-helix (amino acids 136 ± 155) is underlined. Accession numbers of nucleotide sequences in EMBL database: X76453 (rat H-rev107), X92814 (human H-REV107-1)


1307

Figure 2 Normal tissue Northern blots showing H-REV107-1 mRNA and related transcripts. 2 mg of poly(A)+-RNA are loaded per lane. The following probes were used: (a) H-REV107-1 cDNA; the major 1.2 kb and the minor 6.5 kb H-REV107-1 transcripts are indicated. (b) H-REV107-2 cDNA; the 1 kb H-REV107-2 mRNA is indicated. Size markers are shown on the right of the ®gure

The H-REV107 protein was undetectable in the HREV107-1 mRNA-positive cell lines by Western blot and immuno¯uorescence analysis (not shown). Therefore we analysed the eect of forced human HREV107-1 expression in the human liver carcinoma cell line SK-HEP1 which exhibits low levels of endogenous H-REV107-1 mRNA but no protein. Cells were stably transfected with the human HREV107-1 cDNA expression vectors pCDNA3+871 and pCDNA3+971, the rat H-rev107 cDNA expression vector pCDNA107, and the control vector. Transfectants were selected in medium containing G418. Representative clones were analysed by Northern and Western blotting (Figure 4a and b). While most of the transfected SK-HEP-1 cells expressed elevated levels of the exogenous H-REV107-1 mRNA, only three of 30 transfectants contained the protein. One of the rare positive clones (8-1/6; Figure 4b) exhibited a protein of aberrant size which is likely to be non-functional. The two other clones contained only weak H-REV107 expression (Figure 4b) indicating heterogeneous intraclonal expression as was previously shown in ®broblasts and hepatocellular carcinomas cell lines transfected with the rat H-rev107 cDNA. The fraction of positive cells diminished rapidly on continuous culture of transfectants (Sers et al., 1997).

One of the four H-REV107 cDNA clones harboured an insertion of 95 bp resulting in a frame shift mutation that eliminated the hydrophobic C-terminal domain of the encoded protein. This cDNA cloned into the recombinant vector pCDNA3+371 can be eciently translated in SK-HEP cells (Figure 4b). These results con®rm an earlier observation that the permanent proliferation of transfected cells is not compatible with the presence of the wild-type HREV107 protein. Chromosomal localization of H-REV107-1 The rat H-rev107 gene was mapped to chromosome 1 (Szpirer et al., 1996), which corresponds to human chromosome 11. To identify the precise position of the human H-REV107-1 gene in the human genome, we performed ¯uorescence in situ hybridization (FISH) analysis using the two genomic H-REV107-1 clones as described in Materials and methods. Forty of 49 (81%) metaphase spreads hybridized with the probe 6-1 exhibited ¯uorescent signals in band q11-12 on both chromosome 11 homologues (Figure 5). Thirty of 36 (83%) metaphase spreads hybridized with 6-2 DNA exhibited an identical hybridization pattern. The ¯uorescence signals were highly speci®c, as more than 80% of the target sequences were labelled with each of


1308

Figure 3 (a) Northern blot analysis of H-REV107-1 mRNA in human tumour cell lines. Ten mg of total RNA were loaded per lane: (1) SK-Mel 23, (2) SK-Mel 28, (3) Wei Mel (malignant melanomas), (4) SK-N-SH, (5) SK-N-MC (neuroblastomas), (6) BT-20, (7) MCF-7, (8) SK-Br-3 (mammary adenocarcinomas), (9) Kato, (10) RF-1 (gastric adenocarcinomas), (11) RF-48 (metastatic gastric adenocarcinoma), (12) NCI-H441 (papillary lung adenocarcinoma), (13) NCI-H596 (adenosquamous lung carcinoma), (14) NCI-H582 (small cell lung carcinoma), (15) LN 215, (16) LN 235, (17) LN 428 (gliomas), (18) Hep G2 (hepatocellular carcinomas), (19) C4 I, (20) C4 II, (21) C33 A (cervical carcinomas), (22) Caki I, (23) Caki II (clear cell carcinomas), (24) A498 (kidney carcinoma), (25) A431 (epidermoid carcinoma), (26) SW 579 (thyroid carcinoma), (27) HT 1080 (®brosarcoma). 28S and 18S ribosomal RNA as size markers are indicated, the H-REV107-1 speci®c signal is marked by an arrow. (b) RT ± PCR analysis of HREV107-1 mRNA in primary squamous cell carcinomas. The H-REV107-1-speci®c PCR product is 564 base pairs in size. Upper panel: MCF-7 cell line (positive control), SKBR-3 cell line (negative control), N, normal squamous epithelium, T1, T2, squamous cell carcinomas. Note that the normal epithelium was derived from the same patient as tumour T1. Lower panel: control PCR, equal amounts of 205 bp PCR products were obtained in tissue samples using GAPDH-speci®c primers

the genomic clones and no additional ¯uorescence signal was found in other regions of the human genome. H-REV-107 is a member of a gene family We identi®ed several expressed sequence tags (EST) containing partial H-REV107-1 sequences and four related ESTs by comparing the H-REV107-1 cDNA with the GenEMBL data bases using the FASTA program. Two identical cDNA clones, TS 53 996 and TS 71 932, referred to as H-REV107-2 therafter, were obtained from the Washington University-Merck EST Project and sequenced. The longest clone has an insert of 720 bp carrying a complete open reading frame, as well as 5' and 3' untranslated sequences. Alignment of human H-REV107-1 and H-REV107-2 cDNA revealed an identity of 65%, con®ned to the coding region (Figure 1). The human H-REV107-2 cDNA encodes a protein of 216 amino acids, 54 amino acids longer than

the human H-REV107-1 protein. Comparison of the predicted amino acid sequences revealed an overall identity of 51%. The 50 amino terminal residues of both proteins exhibit an identity of 72% (Figure 1). The product of the H-REV107-2 gene is a 1.0 kb mRNA. In contrast to H-REV107-1, H-REV107-2 mRNA is abundant in leukocytes and thymus, but undetectable in brain, testis and pancreas (Figure 2b). We conclude that H-REV107-1 is a member of a small gene family. The expression pattern of H-REV107-2 in tumour cells has not yet been analysed. Discussion The H-REV107-1 gene is ubiquitously expressed in normal tissues, but is consistently downregulated in tumorigenic human cell lines derived from melanoma, neuroblastoma, adenocarcinoma of the breast, lung, and stomach, carcinoma of the cervix, thyroid, and


1309

Figure 4 (a) Northern blot analysis of H-REV107-1 expression in total RNA prepared from stably transfected SK-HEP-1 cells. Lanes 1 ± 14, individual clones isolated after transfection with the H-REV107-1 expression vector pCDNA3+871; lane 7, non-transfected SK-HEP-1 cells; lane P, pool of pCDNA3+871 transfectants; lanes V1, V2, control pCDNA3 transfectants. 28S and 18S rRNAs as size markers are indicated, the H-REV107-1 speci®c mRNA is indicated. Equal amounts of RNA (10 mg) were loaded on each lane. (b) Western blot analysis of H-REV107-1 protein in stable SK-HEP1 transfectants. Recombinant expression vectors used for transfection are given in brackets: Clones 8-1/5, 8-1/6, 8-1/4 and 8-1/2 (pCDNA3+871); clones 900/0, 900/7, 900/2 and 900/0 (pcDNA107+); clones 3-1/8, 3-1/7 and 3-1/3 (pCDNA3+371); clones T53/7 and T53/1 (H-REV107-2 cDNA cloned into pCDNA3). Cos/8-1, COS7 cells transiently transfected with pCDNA3+871 (control). Molecular weight markers are indicated. Extracts from clones T53/7 to cos/8-1 were on the same gel, extracts from clones 3-1/8 to 3-1/3 were on a separate gel. 75 mg of protein were loaded on each lane

kidney, and ®brosarcoma. The loss of H-REV107-1 expression was not restricted to cultured cells, but was also found in primary tumours. In addition, we have shown that the proliferation of the human SK-Hep-1 cell line is not compatible with H-REV107-1 protein expression, because cells exhibiting the endogenous or exogenous H-REV107 protein could not be maintained in culture. This con®rms the anti-proliferative activity of the rat H-rev107 protein observed in rodent mesenchymal cells and in a hepatocellular cell line transformed by activated HRAS genes (Sers et al., 1997). Therefore, we conclude that H-REV107-1 is a suppressor for multiple cancers such as the MTS1 gene described earlier (Kamb et al., 1994). The H-REV107-1 gene was assigned to band q11q12 of human chromosome 11. This chromosome is known to carry several tumour suppressor genes including the WT1 gene at 11p13 (Gessler et al., 1990; Call et al., 1990) and a yet unknown gene at 11p15 (Henry et al., 1989) involved in nephroblastoma. The presence of additional suppressor genes was indicated by chromosome transfer studies (Gioeli et al., 1997; Negrini et al., 1994; Loh et al., 1992; Weissman et al., 1987) and by allelic deletions at q11-q12 in oesophageal and lung cancer (Iizuka et al., 1995; Whang Peng et al., 1990). Thus, the H-REV107-1 gene is a candidate suppressor gene involved in oesophageal and lung cancer. Our ®nding supports

the notion that cloning strategies based on speci®c RNA expression patterns in well-de®ned tumour cells as compared to normal cells provide an equally useful and rapid alternative to the classical approach for tumour suppressor gene isolation (Sager, 1997). The gross structure of the H-REV107-1 gene was intact in human tumour cell lines which were negative for the encoded mRNA and/or protein. This indicates that the H-REV107-1 gene is inactivated by a block to its expression, rather than by mutation and represents a class II tumour suppressor gene (Sager, 1997; Lee et al., 1991). Class II tumour suppressor function can be abolished by transcriptional silencing through methylation of CpG-islands in the 5'-regulatory region as shown for the VHL and MTS1 genes (Herman et al., 1994; Merlo et al., 1995) or by the loss of transcriptional activation as shown for the maspin gene (Zhang et al., 1997). A well studied example of post-transcriptional inactivation aects the tumour suppressor TP53. Cytoplasmic sequestration of the p53 protein was shown to abolish the growthconstraining function in breast carcinomas and undierentiated neuroblastomas (Moll et al., 1992). The loss of H-rev107 expression occurs both at the level of RNA and protein (Sers et al., 1997; this paper). The majority of human tumour cell lines and a limited set of primary tumours analysed lack H-REV107-1 mRNA suggesting transcriptional silencing. Some cell


1310

Figure 5 Assignment of the human H-REV107-1 gene to chromosome 11q11-12 by ¯uorescence in situ hybridization. Left panel: Total metaphase spread hybridized with the genomic H-REV107-1 probe 6-1. Right panel: Sections of metaphase spreads showing chromosome 11 only

lines expressed H-REV107-1 mRNA, but no protein indicating that H-REV107-1 function can also be impaired at the post-transcriptional level. H-REV1071 mRNA was eciently translated in transiently transfected COS cells, however, the protein was not detectable in stable SK-HEP-1 transfectants in spite of elevated levels of H-REV107-1 mRNA. This suggests that the lack of signi®cant H-REV107 protein levels in proliferating cells is governed by alternative mechanisms aecting the transcription of the gene, the translation of mRNA or protein stability. The product of the H-rev107 gene is an approx. 17 ± 18 kDa protein without sequence similarity to proteins present in the databases suggesting a novel transformation-suppressive function. The ®rst hints for inhibition of growth and neoplastic transformation came from the observation that the H-rev107 protein accumulated in the membrane fraction of density-arrested ®broblasts (Hajnal et al., 1994) and that three dierent cell lines resistant toward HRAS-induced transformation (SchaÈfer et al., 1988; Katz and Carter, 1986; Franza et al., 1986) contained elevated levels of mRNA (Hajnal et al., 1994). The C-terminal hydrophobic domain responsible for membrane-association of the H-rev107 protein was required for full anti-proliferative activity in HRAS-transformed cells (Sers et al., 1997). Since forced expression of H-rev107 cDNA does not aect the colony-forming ability of nontransformed fibroblasts (Sers et al., 1997), the protein may act as a negative regulator of oncogenic Ras signalling, for instance by binding to and inhibiting a downstream eector molecule.

The role of H-rev107 as a negative growth regulator is supported by an independent line of evidence related to interferon signalling. H-rev107 is upregulated in cells treated with interferon-g (Kuchinke et al., 1995). Interferon signalling pathways mediate anti-tumour eects (Lengyel, 1993) and are involved in the resistance toward RAS transformation (Tanaka et al., 1994). Three other interferon-responsive genes, IRF-1 (Harada et al., 1993), PKR (Meurs et al., 1993) and Hrev142/lox (Contente et al., 1990) have been shown to possess tumour suppressor activity. In addition, the anti-oncogenic capacity of interferon regulatory factor1 may be mediated by the products of PKR (Kirchho et al., 1995) and H-rev142/lox (Tan et al., 1996) genes. H-rev142/lox and H-rev107 genes are characterized by an identical expression pattern in normal cells, HRAStransformed cells and in phenotypic revertants (Oberhuber et al., 1995; Hajnal et al., 1993, 1994). In view of the transformation-suppressing ability of both genes (Sers et al., 1997; Contente et al., 1990) and their modulation of expression by interferons (Oberhuber et al., 1995; Contente et al., 1990), it is tempting to speculate that H-rev107 genes contribute to the antitumour eects of interferons. Our results provide an important link between oncogene activation and H-rev tumour suppressor inactivation. The RAS signalling pathway is permanently activated in many cancers, either by mutations in one of the RAS genes themselves (Bos, 1989), by mutation of Ras eectors (Basu et al., 1992), or by activation of upstream components such as membrane tyrosine kinase receptors (Janes et al., 1994). Identifica-


tion of the complete set of H-rev genes will help to understand the multifaceted features of cancer cells carrying mutant RAS genes or containing elevated RAS ± GTP levels in the absence of RAS mutations. Materials and methods Cell culture and cell lines Human tumour cell lines were cultured in RPMI medium containing 10% fetal calf serum (FCS), 100 U/ml penicillin and 100 mg/ml streptomycin. SK-Mel 23, Wei Mel, LN 215, LN 235 and LN 428 cell lines were a kind gift of Dr Judith Johnson (Institute for Immunology, MuÈ nchen), all other cell lines were obtained from the American Type Culture Collection. COS-7 cells were grown in Dulbecco's modi®ed Eagle's medium (DMEM) containing 10% FCS, 100 U/ml penicillin and 100 mg/ml streptomycin. Isolation of human H-REV107-1 cDNA and genomic DNA A human liver cDNA library (3.66105 clones; Clontech, no. HL1188x) was hybridized with a 406 bp (a32P-dCTP (Amersham) labelled PCR fragment (pos. 98 ± 503) of the rat H-Rev107 cDNA. Puri®ed positive clones were converted into pDR-2 plasmid by in vivo excision according to the protocol of the manufacturer. Two HREV107-1 related cDNA clones were obtained from the Washington University-Merck EST Project (Genome Sequencing Center, Washington University School of Medicine, St. Louis). Sequencing was done with an automated sequencing system (Applied Biosystem). For the isolation of genomic sequences of human H-REV107-1, 3.66105 clones of a human leukocyte EMBL-3 libary (Clontech, no. HL1006d) were screened with the same fragment used for the cDNA libary. Two independent clones (6-1 and 6-2) were isolated and speci®city was con®rmed by Southern blotting and partial sequencing. H-REV107-1 expression plasmids BamHI/XbaI inserts from cDNA clones pDR-2/3-1, pDR2/8-1 and pDR-2/9-1 were subcloned into the BamHI/XbaI site of pCDNA3 (Invitrogen, Leek). Plasmid pCDNA3+81 harbours 718 bp including the complete coding sequence and 62 bp of the 5'-untranslated region. pCDNA3+9-1 harbours 672 bp including the complete coding sequence and 16 bp of the 5'-untranslated region. pCDNA3+3-1 carries a 95 bp insertion that results in a frame shift eliminating the hydrophobic C-terminal domain of the translation product. pCDNA107+ harbours the rat Hrev107 cDNA (Sers et al., 1997). Northern blot analysis Total cellular RNA of human tumor cell lines was prepared as described (Chomzcynski and Sacchi, 1987). For Northern blotting, total RNA prepared from cell lines (10 mg per lane) were hybridized with an H-REV107-1 cDNA probe 32P-labelled by random priming. Normal human tissue RNA blots were obtained from Clontech (No. 7759-1 and 7760-1) and hybridized with a 384 bp PCR-fragment (pos. 24 ± 407) of human H-REV107-1 or with a 339 bp PCR-fragment (pos. 382 ± 720) of human HREV107-2 cDNA (Clone TS 53992). Hybridization was carried out in 50% (v/v) deionized formamide, 10% dextrane sulfate, 1 M sodium chloride, 56 Denhardt's solution, 1% sodium dodecylsulfate (SDS), 50 mM Tris/ HCl pH 7.5 and 200 mg/ml denatured salmon sperm DNA at 428C for 18 h. Blots were washed with 0.26SSC/0.1% SDS at 658C and exposed to X-ray ®lms.

Human tumour samples and RT ± PCR analysis Samples from normal and tumorigenic human tissues were snap frozen in liquid nitrogen. Total RNA was prepared as described (Chomzcynski and Sacchi, 1987). Two mg total RNA were reverse transcribed using AMV reverse transcriptase (Boehringer, Mannheim). 2 ml of each reaction were used for PCR with either GAPDH speci®c primers GAPDH for (5'-GACCTTCATTGACCTCAAC3') and GAPDHrev (5'-CCAGCCTTCTCCATGGGG-3') at 568C or H-REV107-1 speci®c primers A (5'-CTACGCAGCGAAATCGAGCC-3') and G (5'-GTCATCGCGACAGACAGTC-3') at 548C using 33 cycles. Reactions with individual tumour RNAs were repeated at least twice and analysed on 1% agarose gels. Fluorescence in situ hybridization (FISH) For chromosomal mapping, complete undigested phage DNA of the genomic clones 6-1 and 6-2 were labelled with digoxigenin-11-dUTP by nick-translation and hybridized to human metaphase chromosomes as described (Lichter et al., 1990). Brie¯y, 80 ng of digoxigenin-labelled probe was combined with 3 mg of human Cot 1-DNA and 7 mg salmon sperm DNA in 10 ml hybridization mix and hybridized to human lymphocyte metaphase chromosomes from a normal male 46, XY. Hybridized probes were detected via rhodamine and chromosomes were counterstained with DAPI (4,6-diamino-2-phenylindole-dihydrochloride). DAPI and rhodamine ¯uorescence were recorded separately by using a cooled CCD camera (Photometrics), carefully aligned and electronically overlayed. DNA transfection and Western blot analysis Thirty mg of recombinant vector DNA were transfected into 106 COS cells by calcium-phosphate precipitation (Wigler et al., 1978). After 40 h, cells were washed twice with phosphate buered saline (pH 7.4, PBS), lysed in 300 ml 46SDS sample buer (0.25 M Tris/HCl pH 6.8, 20% (v/v) b-mercaptoethanol, 40% (v/v) glycerol, 9.2% (w/v) SDS, and 0.015% (w/v) bromophenol blue), boiled for 10 min and stored at 7208C. Total protein concentration was quanti®ed as described (Schaner and Weissmann, 1973). Equal amounts of protein were fractionated on 14% SDS polyacrylamide gels (Laemmli, 1970), transferred to nitrocellulose membranes and blocked overnight at 48C in TBS (25 mM Tris/HCl, pH 7.4, 150 mM NaCl) containing 4% (w/v) low fat milk powder. Membranes were incubated for 5 h at room temperature with an anity-puri®ed anti-rat H-rev107 antiserum (Hajnal et al., 1994), diluted 1 : 750 in blocking solution. After washing with TBST (TBS plus 0.05% Tween 20), the membrane was incubated for 1 h at room temperature with a goat-anti-rabbit alkaline phosphatase-conjugated antibody (Promega) diluted 1 : 5000 in blocking solution. HREV107-1 protein was detected by 330 mg/ml nitro blue tetrazolium (NBT) and 165 mg/ml 5-bromo-4-chloro-3indoyl phosphate (BCIP) in AP-buer (100 mM Tris/HCl, 100 mM NaCl, 5 mM MgCl2, pH 9.5).

Acknowledgements The authors thank Jacqueline Gheyselinck for technical assistance and Christa Scholz for photographic work. Our work was supported by Schweizerischer Nationalfonds (grant no. 31-37525 to RS), Schweizerische Krebsliga, Krebsliga des Kantons ZuÈrich, and by the Humboldt University Medical School (research grants to CS, EF and RS).

1311


1312

References Basu TN, Gutmann DH, Fletcher JA, Glover TW, Collins FS and Downward J. (1992). Nature, 356, 713 ± 715. Boak AM, Roy R, Berk J, Taylor L, Polgar P, Goldstein RH and Kagan HM. (1994). Am. J. Respir. Cell. Mol. Biol., 11, 751 ± 755. Bos JL. (1989). Cancer Res., 49, 4682 ± 4689. Call KM, Glaser T, Ito CY, Buckler AJ, Pelletier J, Haber DA, Rose EA, Kral A, Yeger H, Lewis WH, Jones C and Housman DE. (1990). Cell, 60, 509 ± 520. Chomzcynski P and Sacchi N. (1987). Anal. Biochem., 162, 156 ± 159. Contente S, Kenyon K, Rimoldi D and Friedman RM. (1990). Science, 249, 796 ± 798. Cutler ML, Bassin RH, Zanoni L and Talbot N. (1992). Mol. Cell. Biol., 12, 3750 ± 3756. Franza BRJ, Maruyama K, Garrels JI and Ruley HE. (1986). Cell, 44, 409 ± 418. Gessler M, Poustka A, Cavenee W, Neve RL, Orkin SH and Bruns GAP. (1990). Nature, 343, 774 ± 778. Gioeli D, Conway K and Weissman BE. (1997). Cancer. Res., 57, 1157 ± 1165. Griegel S, Traub O, Willecke K and SchaÈfer R. (1986). Int. J. Cancer, 38, 697 ± 705. Gutterman JU. (1994). Proc. Natl. Acad. Sci. USA, 91, 1198 ± 1205. Hajnal A, Klemenz R and SchaÈfer R. (1993). Cancer Res., 53, 4670 ± 4675. Hajnal A, Klemenz R and SchaÈfer R. (1994). Oncogene, 9, 479 ± 490. Harada H, Kitagawa M, Tanaka N, Yamamoto H, Harada K, Ishihara M and Taniguchi T. (1993). Science, 259, 971 ± 974. Henry I, Grandjouan S, Couillin P, Barichard F, HuerreJeanpierre C, Glaser T, Philip T, Lenoir G, Chaussain JL and Junien C. (1989). Proc. Natl. Acad. Sci. USA, 86, 3247 ± 3251. Herman JG, Latif F, Weng YK, Lerman MI, Zbar B, Liu S, Samid D, Duan DS, Gnarra JR, Linehan WM and Baylin SB. (1994). Proc. Natl. Acad. Sci. USA, 91, 9700 ± 9704. Iizuka M, Sugiyama Y, Shiraishi M, Jones C and Sekiya T. (1995). Genes Chromosomes Cancer, 13, 40 ± 46. Janes PW, Daly RJ, Defazio A and Sutherland RL. (1994). Oncogene, 9, 3601 ± 3608. Kamb A, Gruis NA, Weaver-Feldhaus J, Liu Q, Harshman K, Tavtigian SV, Stockert E, Day RSI, Johnson BE and Skolnick MH. (1994). Science, 264, 436 ± 440. Katz E and Carter BJ. (1986). J. Natl. Cancer Inst., 77, 909 ± 914. Kenyon K, Contente S, Trackman PC, Tang J, Kagan HM and Friedman RM. (1991). Science, 253, 802. Kiess M, Scharm B, Aguzzi A, Hajnal A, Klemenz R, Schwarte-Waldho I and SchaÈfer R. (1995). Oncogene, 10, 61 ± 68. Kirchho S, Koromilas AE, Schaper F, Grasho M, Sonenberg N and Hauser H. (1995). Oncogene, 11, 439 ± 445. Kitayama H, Sugimoto Y, Matsuzaki T, Ikawa Y and Noda M. (1989). Cell, 56, 77 ± 84. Kuchinke W, Hart RP and Jonakait GM. (1995). Neuroimmunomodulation, 2, 347 ± 355.

Laemmli UK. (1970). Nature, 227, 680 ± 685. Lasko D, Cavenee W and NordenskjoÈld M. (1991). Ann. Rev. Genet., 25, 281 ± 314. Lee SW, Tomasetto C and Sager R. (1991). Proc. Natl. Acad. Sci. USA, 88, 2825 ± 2829. Lengyel P. (1993). Proc. Natl. Acad. Sci. USA, 90, 5893 ± 5895. Lichter P, Tang CC, Call K, Hermanson G, Evans GA, Housman D and Ward DC. (1990). Science, 247, 64 ± 69. Loh WE, Scrable HJ, Livanos E, Arboleda MJ, Cavenee WK, Oshimura M and Weissman BE. (1992). Proc. Natl. Acad. Sci. USA, 89, 1755 ± 1759. Merlo A, Herman JG, Mao L, Lee DJ, Gabrielson E, Burger PC, Baylin SB and Sidransky D. (1995). Nature Med., 1, 686 ± 692. Meurs EF, Galabru J, Barber GN, Katze MG and Hovanessian AG. (1993). Proc. Natl. Acad. Sci. USA, 90, 232 ± 236. Moll UM, Riou G and Levine AJ. (1992). Proc. Natl. Acad. Sci. USA, 89, 7262 ± 7266. Negrini M, Sabbioni S, Possati L, Rattan S, Corallini A, Barbantibrodano G and Croce CM. (1994). Cancer Res., 54, 1331 ± 1336. Oberhuber H, Seliger B and SchaÈfer R. (1995). Mol. Carcin., 12, 198 ± 204. Sager R, Anisowicz A, Neveu M, Liang P and Sotiropoulou G. (1993). FASEB J., 7, 964 ± 970. Sager R. (1997). Proc. Natl. Acad. Sci. USA, 94, 952 ± 955. Schaner W and Weissmann C. (1973). Anal. Biochem., 56, 502 ± 514. SchaÈfer R, Iyer J, Iten E and Nirkko AC. (1988). Proc. Natl. Acad. Sci. USA, 85, 1590 ± 1594. SchaÈfer R. (1994). Rev. Phys. Biochem. Pharm., 124, 29 ± 92. Schipper JH, Frixen UH, Behrens J, Unger A, Jahnke K and Birchmeier W. (1991). Cancer Res., 51, 6328 ± 6237. Sers C, Emenegger U, Husmann K, Bucher K, Andres A-C and SchaÈfer R. (1997). J. Cell. Biol., 136, 935 ± 944. Shibanuma M, Mashimo J, Mita A, Kuroki T and Nose K. (1993). Eur. J. Biochem., 217, 13 ± 19. Stanbridge EJ. (1991). Annu. Rev. Genet., 24, 615 ± 657. Szpirer C, Szpirer J, Riviere M, Hajnal A, Kiess M, Scharm B and SchaÈfer R. (1996). Mammal. Genome, 7, 701 ± 703. Tan RS-P, Taniguchi T and Harada H. (1996). Cancer Res., 56, 2417 ± 2421. Tanaka N, Ishihara M, Kitagawa M, Harada H, Kimura T, Matsuyama T, Lamphier MS, Aizawa S, Mak TW and Taniguchi T. (1994). Cell, 77, 829 ± 839. Thompson ME, Jensen RA, Obermiller PS, Page DL and Holt JT. (1995). Nature Genet., 9, 444 ± 450. Weissman BE, Saxon PJ, Pasquale SR, Jones GR, Geiser AG and Stanbridge EJ. (1987). Science, 236, 175 ± 180. Whang Peng J, Banks Schlegel SP and Lee EC. (1990). Cancer Genet. Cytogenet., 45, 101 ± 120. Wigler M, Pellicer A, Silverstein S and Axel R. (1978). Cell, 14, 725 ± 731. Zhang M, Maass N, Magit D and Sager R. (1997). Cell Growth Di., 8, 179 ± 186. Zou Z, Anisowicz A, Hendrix MJC, Thor A, Neveu M, Sheng S, Ra®di K, Seftor E and Sager R. (1994). Science, 263, 526 ± 529.