CC2 gene alterations in invasive ... - Nature

Oncogene (1998) 16, 677 ± 679  1998 Stockton Press All rights reserved 0950 ± 9232/98 $12.00

SHORT REPORT

Low frequency of TSG101/CC2 gene alterations in invasive human breast cancers Qing Wang1,6, Keltouma Driouch2,6, SteÂphanie Courtois1, Marie HeÂleÁne ChampeÁme2, Ivan BieÁche2, Isabelle Treilleux3, Marianne Bri€od4, Ruth Rimokh1, Jean Pierre Magaud1, Patrick Curmi5, Rosette Lidereau2 and Alain Puisieux1 1

Unite d'Oncologie MoleÂculaire. Unite INSERM U453. Centre LeÂon BeÂrard, 69373 Lyon Cedex 08; 2Laboratoire d'OncogeÂneÂtique, Centre Rene Huguenin, 92210 St-Cloud; 3DeÂpartement d'Anatomie et de Cytologie Pathologiques. Centre LeÂon BeÂrard, 69373 Lyon Cedex 08; 4Laboratoire d'Anatomo-Cyto-Pathologie, Centre Rene Huguenin, F-92210 St-Cloud; 5Unite INSERM U440, 17 rue du Fer aÁ Moulin, 75005 Paris, France

Large intragenic deletions of the TSG101/CC2 gene were recently reported in seven of 15 primary metastatic breast cancers. Although the number of samples was small, this observation suggested that TSG101/CC2 alterations were a major event in breast carcinogenesis. To study the frequency of these deletions in invasive breast cancers we analysed 189 primary invasive breast tumours and 59 breast cancer metastases. We detected intragenic rearrangements in only three samples (two primary tumours and one metastasis). Northern blot analysis of 43 tumours without rearrangements failed to detect any abnormalities. Furthermore, we studied TSG101/CC2 in 11 human breast adenocarcinoma cell lines by Southern blot, RT-PCR and sequencing of the entire coding region of the gene, and detected no abnormalities. These results show that genetic alteration of TSG101/CC2 is a rare event in breast cancer. Keywords: human breast cancer; TSG101/CC2; rearrangement; mutation

The murine TSG101 gene was recently identi®ed by using a novel strategy of controlled homozygous functional knockout of allelic loci (Li and Cohen, 1996). This approach can be used to isolate unknown mammalian autosomal genes whose inactivation is associated with a de®ned phenotype. Functional knockout of TSG101 in murine ®broblasts led to cell transformation and tumorigenesis, suggesting that it was a tumour suppressor gene. A homologue of the TSG101 gene was initially identi®ed as CC2, by means of two-hybrid system methodology (Maucuer et al., 1995). It is a coiled-coil sequence interacting with stathmin, a small ubiquitous cytosolic regulatory phosphoprotein (Sobel, 1991) that is modulated during the cell cycle (Strahler et al., 1992; Larsson et al., 1995) and overexpressed in various types of cancer (Hanash et al., 1988; Brattsand et al., 1993; Friedrich et al., 1995). We thus refer to TSG101 as TSG101/CC2 in the rest of this paper. The human TSG101/CC2 gene was mapped to chromosome 11, subbands p15.1 ± p15.2 (Li et al., 1997). The 11p15 band is frequently Correspondence: A Puisieux 6 These authors have contributed equally to this work Received 4 July 1997; revised 15 September 1997; accepted 15 September 1997

altered in human solid tumours including breast cancer, stomach cancer, ovarian cancer and lung cancer (Gudmundsson et al., 1995; Winqvist et al., 1995; Ba€a et al., 1996; Tran and Newsham, 1996; Lu et al., 1997). Recently, Li et al. analysed 15 primary metastatic breast cancers and identi®ed large intragenic TSG101/CC2 deletions in seven cases (Li et al., 1997). Alterations of both TSG101/CC2 alleles were found in four of these cancers. Although the study involved a small number of tumours, it suggested that deletions of the TSG101/CC2 gene might be a frequent event in breast carcinogenesis. To assess the frequency of intragenic deletions of this gene in human breast cancers and their possible involvement in metastasis, we carried out DNA structure analysis (Southern blot or microsatellite length polymorphism analysis) in 189 primary invasive breast adenocarcinomas, 59 breast cancer metastases and 11 breast cell lines. We also analysed TSG101/CC2 RNA expression in 43 primary tumours and 11 cell lines and sequenced the entire coding region of the TSG101/CC2 gene in 11 breast cell lines. Results and Discussion In the original report by Li et al. seven of the 15 primary metastatic breast adenocarcinomas they tested showed large intragenic deletions (7 ± 16 kb) of the TSG101 gene (Li et al., 1997). Here, we attempted to con®rm this high frequency of TSG101 rearrangements in a large series of invasive human breast cancers. One hundred and eighty-nine primary breast tumours were analysed by Southern blot. Tumours were pure histological variants of invasive breast carcinomas comprising 157 ductal, 23 lobular, six mucinous, one tubular, one cribriform and one squamous cell metaplastic carcinomas. Genomic DNA was digested with three di€erent restriction endonucleases (EcoRI, BglII and TaqI) and fractionated by electrophoresis. Southern blots were probed with TSG101/CC2 cDNA. Surprisingly, only two of the 189 breast adenocarcinomas showed restriction fragment size abnormalities (Figure 1). In both cases TSG101/CC2 gene abnormalities were detected with all the restriction enzymes used. Tumour cells have a genetically determined metastatic potential. If alterations of speci®c genes are associated with the invasive process, they would

TSG101/CC2 in human breast cancer Q Wang et al

678

kb

T

1 2 3 130 142 161 247 267 552 256 510* 245 437 494 498 501 213 550 552 560

N

9.4 — 6.5 — 4.3 —

kb

T

1 2 3 130 142 161 247 267 552 256 510* 245 437 494 498 501 213 550 552 560

N

9.4 — 6.5 — 4.3 —

Figure 1 Southern blot analysis of the TSG101/CC2 gene in breast adenocarcinomas. Ten micrograms of genomic DNA from each sample were digested with EcoRI (upper panel) and BglII (lower panel) and fractionated by electrophoresis in 0.7% agarose gels. DNA was then transferred to nylon membrane ®lters with standard blotting procedures and ®lters were hybridized overnight at 658 with the denatured labeled TSG101/CC2 cDNA probe, washed in stringent conditions and autoradiographed at 7808C for an appropriate period. The human TSG101 cDNA was obtained by RT ± PCR from lymphoid cells of a normal individual by using primers (P1 and P2) and PCR conditions described elsewhere (Li et al., 1997), except that we used AmpliTaq polymerase (Perkin Elmer/Cetus). The RT ± PCR product was cloned into the pCR 2.1 vector (Invitrogen) and veri®ed by sequencing. Shown are representative autoradiographs. Asterisks indicate samples with gross alterations. The position of marker bands from DNA molecular weight marker II (Boehringer Mannheim) are indicated

probably be more frequently altered in metastases than in primary tumours. Loss of heterozygosity (LOH) on chromosome arm 11p in invasive breast cancer has been correlated with low estrogen receptor protein content and tumour size, indicating that LOH in this region may be a late event in malignant progression (Karnik et al., 1995). It is noteworthy that Li and collaborators tested only cancers that had metastasized (Li et al., 1997). To identify speci®c involvement of TSG101/CC2 in the late stages of tumour progression, we analysed 59 metastases from unrelated breast cancer patients (26 pleural e€usions, 25 solid distant metastases and eight lymph nodes) by Southern blotting after digestion with EcoRI, BglII and TaqI. Only one sample showed a restriction fragment size abnormality. These results strongly suggest that intragenic deletions of TSG101/CC2 are uncommon in both primary tumours and breast cancer metastases. We then analysed genomic DNA from primary tumours and metastases by means of microsatellite length polymorphism with the D11S921/AFM212xa11 marker, which is located in the same band as TSG101/ CC2 (Li et al., 1997). Loss of heterozygosity (LOH) was assayed by PCR ampli®cation of genomic DNA

with the following primer sequences: 5'-TGC ATT CAA CAA ATC AAC A-3', and 5'-CTT GGA CCA TTT AAT CTA AAG TAA T-3' (Dib et al., 1996). DNA was ampli®ed by using standard methods (Weber and May, 1989). After denaturation, aliquots of PCR reaction products were electrophoresed and DNA was transferred to nylon membrane ®lters. After hybridisation with labeled probe, LOH was observed in 9/35 (26%) informative primary breast tumours and 3/17 informative metastatic samples (18%). Several groups have reported that the 11p15 region is frequently a€ected by LOH, mostly at the HRas locus (Bieche and Lidereau, 1995). Our results are in keeping with these frequencies, indicating no speci®cally high frequency of LOH close to TSG101/CC2. Taken together with the observed low frequency of intragenic deletions of this gene in breast tumours, this suggested either that TSG101/CC2 was not the major target tumour suppressor gene in 11p15 or that it was inactivated by other mechanisms. TSG101/CC2 expression and the sequence of its coding region were studied in tumour cells. Total RNA was available for 43 of the primary tumours with no detectable rearrangement on Southern blot analysis. RNA was extracted from normal and tumour breast tissue (Au€ray and Rougeon, 1980) and it was fractionated by electrophoresis. Hybridisation of Northern blots were performed with TSG101/CC2 cDNA. After normalization to the ubiquitous 36B4 cDNA control (Masiakowski et al., 1982), no clearly altered TSG101/CC2 signal was observed (data not shown), suggesting that the gene was normally expressed. Next, RT ± PCR ampli®cation was performed and the entire coding region (exons 1 ± 6) of the TSG101/CC2 gene was sequenced in 11 human breast cell lines, eight of which were derived from metastases. Cancer cell lines are known to display a large number of genetic alterations, and the frequency of mutations in several tumour suppressor genes is higher in cell lines than in primary tumours (Spruck et al., 1994). For example, whereas only 20 ± 25% of invasive breast adenocarcinomas display TP53 mutations, nine of the 10 tested breast cancer cell lines expressed mutant TP53 or a homozygous deletion of this tumour suppressor gene. As a result, cell lines are often used in preliminary studies to assess the involvement of a candidate tumour suppressor gene in a speci®c tumour type. A single band of the expected size (1.1 kb) was revealed by RT ± PCR ampli®cation in all the cell lines, using primers designed to detect fulllength TSG101/CC2 cDNA (Figure 2). Complete sequence analysis of the ampli®ed products revealed no mutations. Southern blot analysis of TSG101/CC2 in the same cell lines did not reveal any abnormality (data not shown). Our results indicate that TSG101/CC2 alteration is not a frequent event in breast tumour progression, and make it unlikely that this is the major tumour suppressor gene targeted in 11p15. Although we detected rearrangements of TSG101/CC2 in three of the 248 breast cancer samples, this is nowhere near the frequency (7/15) reported by Li et al. The high frequency of TSG101/CC2 alterations reported by these authors may have resulted from the selection of a particular population of metastatic cancers or from the methodology they used. Indeed, authors based their

HBL100

Hs578T

ZR-75-30

T-47D

MDA-MB-453

MDA-MB-436

MDA-MB-361

MDA-MB-231

MDA-MB-157

BT-20

MCF7

TSG101/CC2 in human breast cancer Q Wang et al

— 1.1 kb

Figure 2 Analysis of TSG101/CC2 cDNA in human breast cancer cell lines. Breast adenocarcinoma cell lines (MCF7, BT-20, MDA-MB-157, MDA-MB-231, MDA-MB-361, MDA-MB-436, MDA-MB-453, T-47D, ZR-75-30, Hs578T) and the breast epithelial cell line HBL-100 were obtained from the American Type Culture Collection (ATCC). mRNA was extracted from cell lines with the QuickPrep Micro mRNA puri®cation kit (Pharmacia Biotech). Random hexamer-primer cDNA was synthesised using the First-Strand cDNA Synthesis kit (Pharmacia Biotech). TSG101/CC2 cDNA was then ampli®ed with primers P1 and P2 in PCR conditions described elsewhere (Li et al., 1997). A unique band (1.1 kb) was observed in all tested cell lines. The speci®c PCR products were then sequenced on an automated sequencer (ABI 373 Applied Biosystems) with the following primers: R1 (sense): 5'-GTA CAA ATA CAG AGA CCT A-3', F2 (sense): 5'-CTA TTT CGG CAT CCT ATC-3', F3 (sense): 5'-GAT AGA CCT GGA TGT CT-3', F1 (antisense): 5'TTT GCA TTA GTG CCC TCA G-3' and R2 (antisense): 5'CTG GAA CTG CCA TCG TTA-3'. Complete sequence analysis revealed no mutations

study on RT ± PCR ampli®cation and Southern blotting of PCR-ampli®ed genomic DNA. Although we did not observe any TSG101/CC2 abnormality by RT ± PCR analysis of 11 di€erent breast cancer cell lines, the use of PCR may lead to the detection of

genetic alterations present in small subpopulations of tumour cells. Advanced cancers are known to display high genomic instability (Cheng and Loeb, 1993), making it likely that a large number of genes are non speci®cally damaged. However, only alterations leading to a signi®cant growth advantage will be selected. The biological signi®cance of genetic alterations in minor populations of malignant cells in advanced cancers remains to be determined. During the reviewing process of our manuscript, two di€erent groups reported the absence of TSG101/CC2 gene alterations in human breast cancers (Steiner et al., 1997; Lee and Feinberg, 1997). Steiner et al. studied 46 primary tumours and 13 breast cancer cell lines by Southern blotting. Lee and Feinberg studied 72 tumours at the level of the cDNA by RT ± PCR and at the level of the genomic DNA by Southern hybridisation. Furthermore, they indicate that TSG101/CC2 gene maps outside regions that show the highest frequency of 11p LOH in breast cancer. Our study, performed on 189 primary tumours, 59 metastases from breast cancer and 11 cell lines, further suggests that TSG101/CC2 alteration (rearrangement or mutation) is not a major event in breast carcinogenesis. Acknowledgements We are indebted to Dr Alexandre Maucuer for providing the CC2 probe and Andre Sobel (INSERM U440) for his constant support and helpful discussions. This work was supported by the Ligue Nationale de Lutte Contre le Cancer (LNCC); the ComiteÂs ReÂgionaux des Hauts de Seine et des Yvelines; Association pour la Recherche sur le Cancer (ARC) and Institut National de la Sante et de la Recherche MeÂdicale (INSERM). RL is a research director with the Institut National de la Sante et de la Recherche MeÂdicale (INSERM).

References Au€ray C and Rougeon F. (1980). Eur J. Biochem., 107, 303 ± 314. Ba€a R, Negrini M, Mandes B, Rugge M, Ranzani GN, Hirohashi S and Croce CM. (1996). Cancer Res., 56, 268 ± 272. Bieche I and Lidereau R. (1995). Genes Chromosomes Cancer, 14, 227 ± 251. Brattsand G, Roos G, Marklund U, Ueda H, Landberg G, Nanberg E, Sideras P and Gullberg M. (1993). Leukemia, 7, 569 ± 579. Cheng KC and Loeb LA. (1993). Adv. Cancer Res., 60, 121 ± 156. Dib C, Faure S, Fizames C, Samson D, Drouot N, Vignal A, Millasseau P, Marc S, Hazan J, Seboun E, Lathrop M, Gyapay G, Morissette J and Weissenbach J. (1996). Nature, 380, 152 ± 154. Friedrich B, GroÈnberg H, LandstroÈm M, Bergh A and Gullberg M. (1995). Prostate, 27, 102 ± 109. Gudmundsson J, Barkardottir RB, Eiriksdottir G, Baldursson T, Arason A, Egilsson V and Ingvarsson S. (1995). Br. J. Cancer, 72, 696 ± 701. Hanash SM, Strahler JR, Kuick R, Chu EHY and Nichols D. (1988). J. Biol. Chem., 263, 12813 ± 12815. Karnik P, Plummer S, Casey G, Myles J, Tubbs R, Crowe J and Williams BR. (1995). Human Mol. Genet. 4, 1889 ± 1894. Larsson N, Melander H, Marklund U, Osterman O and Gullberg M. (1995). J. Biol. Chem., 270, 14175 ± 14183.

Lee MP and Feinberg AP. (1997). Cancer Res., 57, 3131 ± 3134. Li L and Cohen SN. (1996). Cell, 85, 319 ± 329. Li L, Li X, Francke U and Cohen SN. (1997). Cell, 88, 143 ± 154. Lu KH, Weitzel JN, Kodali S, Welch WR, Berkowitz RS and Mok SC. (1997). Cancer Res., 57, 387 ± 390. Masiakowski P, Breathnach R, Bloch J, Gannon F, Krust A and Chambon P. (1982). Nucl. Acid Res., 10, 7895 ± 7899. Maucuer A, Camonis JH and Sobel A. (1995). Proc. Natl. Acad. Sci. USA, 92, 3100 ± 3104. Sobel A. (1991). Trends Biochem. Sci., 16, 301 ± 305. Spruck C.H.3rd, Gonzalez-Zulueta M, Shibata A, Simoneau AR, Lin MF, Gonzales F, Tsai YC and Jones PA. (1994). Nature, 370, 183 ± 184. Steiner P, Barnes DM, Harris WH and Weinberg RA. (1997). Nature Genetics, 16, 332 ± 333. Strahler JR, Lamb BJ, Ungar DR, Fox DA and Hanash SM. (1992). Biochem. Biophys. Res. Commun., 185, 197 ± 203. Tran YK and Newsham IF. (1996). Cancer Res., 56, 2916 ± 2921. Weber JL and May PE. (1989). Am. J. Hum. Genet., 44, 388 ± 396. Winqvist R, Hampton GM, Mannermaa A, Blanco G, Alavaikko M, Kiviniemi H, Taskinen PJ, Evans GA, Wright FA, Newsham I and Cavenee WK. (1995). Cancer Res., 55, 2660 ± 2664.

679