Detection of translocations of 10p by non-radioactive in situ hybridization of VIM gene in SV40-transformed human cell lines

Detection of Translocations of 10p by Non-Radioactive In Situ Hybridization of VIM Gene in SV40-Transformed Human Cell Lines M. Baumgartner, E. Viegas-Pequignot, F. Hoffschir, M. Ricoul, A. Bravard, and B. Dutrillaux

ABSTRACT: SV40-transformed human fibroblasts exhibit characteristic chromosome imbalances, fairly well correlated with the activity of enzymes encoded by genes located on chromosome segments either in deficiency or in excess. However, a major discrepancy existed for the express/an of vimentin gene (VIM), which was high, even though the map location of the gene (lOp) was missing in many cell lines. An in situ hybridization technique using a biotinylated probe for the human VIM was applied to detect eventual cryptic translocations, as chromosome 10p is difficult to identify. In two cell lines (WI 98 and HELl HBLT) in which a loss of copy number af 10p was assumed after karyotyping, a signal for VIM was detected in unidentified short arms of derivative chromosomes. This exemplifies that in situ hybridization is a powerful complement to classical cytogenetics to detect rearrangements in highly rearranged karyotypes from transformed or cancerous cells, These results alsa strengthen the interpretation of the correlation between karyotypic and metabolic imbalances in transformed cells.

INTRODUCTION SV40-transformed fibroblast cell lines exhibit characteristic c h r o m o s o m e imbalances, some deficiencies, such as those of c h r o m o s o m e arms 2p, 6q, and 11p, correlating with a decreased activity of enzymes e n c o d e d by genes m a p p e d on these arms [1]. The expression of the v i m e n t i n gene (VIM), m a p p e d on 10p12 [2], was found to be as high in transformed cell lines, w h i c h have frequent deficiencies of 10p, as in normal fibroblasts. This apparent contradiction was investigated by in situ h y b r i d i z a t i o n of a 2.5-kb biotinylated DNA probe for the h u m a n v i m e n t i n gene [3] to three SV40transformed fibroblast cell lines, including two previously described as deficient in chromosome 10p arm. In each of these two last cell lines, in situ h y b r i d i z a t i o n clearly i n d i c a t e d that an unidentified short arm of a marker c h r o m o s o m e was formed by w h o l e or a part of the 10p c h r o m o s o m e arm, including the v i m e n t i n gene. Non-radioactive in situ hybridization techniques a p p e a r to be a necessary c o m p l e m e n t to studies of c o m p l e x karyotypes, as is the case for transformed or malignant cells.

From InstitutCurie (M. B.,E. V.-P.,B. D.),Sectionde Biologie,Paris;and Commissariat l'EnargieAtomique (F. H., M. R., A. B., B. D.} Fontenay-aux-Roses,France. Address reprint requests to: E. Viegas-Pdquignot, Institut Curie, Section de Biologie, U B A 620 CNRS, 26 rue d'Ulm, 75231 Paris Cddex 05, France. Received October 23, 1990; accepted March I, 1991.

23 © 1991 ElsevierScience Publishing Co., Inc.

Cancer Genet Cytogenet 56:23-29 (1991}

655 Avenue of the Americas, New York, NY 10010

0165-4608/91/$03.50

24

M. Baumgartner et al.

MATERIALS AND METHODS

Cell Lines The following SV40-transformed polyploid fibroblast cell lines were investigated: JDA2 (kindly provided by Dr. M. Fellous from Institut Pasteur, Paris), HELl HBLT (kindly provided by Dr. R. Cassingena, Villejuif, France), and W198 [4]. All cell lines grew in soft agar, expressed SV40-transformed antigen, and had undergone many (above 50) generations in vitro.

Cytogenetics Metaphases were obtained according to a standard method [5]. Slides were frozen at -20°C before in situ hybridization. A characterization of chromosome imbalances observed in the three cell lines was previously published [1].

DNA Probe and In Situ Hybridization A 2.5-kb plasmid subclone of human vimentin gene (pHUVim2.5) [3, 6], encoding for the central part of the protein, was nick-translated with Biotin-ll-dUTP or Biotin-14dATP according to the BRL protocol. Chromosomes were hybridized according to the method of Viegas-Pequignot et al. [7]. The detection of hybridized sites was performed by indirect immunofluorescence using, first, an antibiotin antibody and, second, an immunofluorescein conjugate. The regional mapping was obtained by comparing the metaphase after immunofluorescence and R-banding. To obtain this banding, slides were treated by Earle's solution pH 6.5 for 12 minutes, rinsed, and stained by acridine orange (50/~g/ml) for 20 minutes. Fluorescent spots (FS) were counted as 1 when on a single chromatid from a single chromosome, for 2 when on either the 2 chromatids of a single chromosome or a single chromatid from 2 chromosomes, e t c . . . RESULTS The VIM probe of 2.5 kb has been previously located by non-radioactive in situ hybridization on 10p12 [2], a result slightly different from the proposed localization of the gene on 10p13 [8].

WI98 Cell Line The karyotype is hypotetraploid, with many rearranged chromosomes. Chromosomes 2 and 10 appear deleted (Fig. 1). After in situ hybridization, 76 metaphases exhibited at least one FS, 77 FS being distributed among 34 metaphases. These FS were very preferentially located on the short arms of chromosomes 10 (49/77 -- 64%). A second preferential hybridization site was observed on a derivative chromosome, initially identified as a del(2p): 17/77 -- 22% (Fig. 2). This result and the reassessment of chromosome banding led us to the conclusion that this chromosome was formed by a whole arm t(2;10)12ql0p~2pl0q) with loss of the der(2pl0q) after breakage in constitutive heterochromatin of both chromosomes 2 and 10. Beside normal chromosome 10 and the der(2ql0p), 11 FS were distributed at random.

HELl HBLT Cell Line Sixty-nine metaphases exhibited at least one FS, 60 FS being distributed among 36 metaphases. The large majority of FS (45/60 = 75%) were located on the short arm of the normal chromosome 10. In addition, 10 FS (14%) were localized on a derivative

10p Translocation Detection Using VIM

.

25

+.++

++ 1

~+~ 2

3

°+¢

• I

fl

4

6

7

8

9

11

12

13

14

It 10

|t 16

17

18

X

.

• ~-~ ~ 21

I~ ~ _ e ~ ~ l mar

,~-~i I~ 22

D

~t+ 10 HELl

HBLT

Figure 1 R-banded karyotype of WI 98 cell line exhibiting many rearranged chromosomes, among which are a del(2p) and a del(10p) (arrows). On the right (inset) are normal chromosomes 6 and 10 of HELl HBLT cell line, and the del(6q) (arrow), which is in fact a der(6pl0p).

chromosome, supposed to be a der(6;?)(q11;?) (Figs. 2c and 3). We concluded that this derivative chromosome resulted in fact from a whole arm t(6;10)(6p10p;6q10q), with a loss of the der(6ql0q), after breakage i n constitutive heterochromatin of both chromosomes 6 and 10 (Fig. 1, inset). Only 5 (8%) FS were observed at other chromosomal sites, with a r a n d o m distribution.

26

~

~

.~

~

~

i~

~

~

~ ~

~ ,~ '3

F i g u r e ~- In situ hybridization of the biotinylated VIM probe. Immunofluorescence detection, propidium iodide counterstaining. (a) JDA2 cell line, with normal chromosomes 10, left; R-banding, right. Same chromosomes after in situ hybridization (spots are indicated by arrows). (b) WI 98 cell line, with the der(2ql0p) after R-banding (left), after in situ hybridization of the same (center), and of another chromosome (right). (c) HELl HBLT cell line, with R-banded normal chromosome 6 (left) and der(6plOp) (center), and the same der(fplOp) after in situ hybridization (right).

~

~

~)!

;.,~

F i g u r e 3 Whole metaphase from HELl HBLT showing hybridization spots (a) on the der(6pl0p) (arrows), subsequently identified by R-banding (b).

10p Translocation Detection Using VIM

27

JDA2 Cell Line Fifty-eight metaphases exhibited at least one FS, 77 FS being distributed among 22 metaphases. Beside 10 (13%) FS randomly distributed, 67 (87%) were located on the short arm of normal chromosomes 10. It was thus concluded that, in this cell line, chromosome lop was not involved in a recurrent marker formation.

DISCUSSION The cytogenetic analysis of 7 SV40-transformed human fibroblast cell lines has indicated the presence of characteristic chromosome anomalies. These anomalies resulted from a number of events which can be summarized as follows: chromosome rearrangements preferentially affecting heterochromatic regions~ passage to endoreduplication; and chromosome losses by mitotic malsegregations, affecting both normal and derivative chromosomes [1]. These results were confirmed on a larger series, and the following chromosomes or chromosome arms were found among the most frequently deleted: 2p, 6q, 8p, 10p, 11p, and 18. These deficiencies were correlated with metabolic data, indicating that the activities of a number of enzymes encoded by genes mapped on deleted segments were low: acid phosphatase/2p) [9]: superoxide dismutase 2 (6q) [10]~ cata|ase and lactate dehydrogenase A/1 l p) [11, 12]. We confirmed a number of these correlations between low-enzyme activities and chromosome deletions, and found others that will be described elsewhere in detail [13]. However, as part of the hypothesis that deletions do not affect chromosome segments carrying actively expressed genes, a noticeable inconsistency existed for the vimentin gene. This gene is expressed in SV40-transformed human fihroblasts [6, 14], whereas it is mapped on 10p [2, 8], an arm frequently found deleted in our study. This inconsistency might have several explanations, including that chromosome 10p, of small size and fairly homogeneously stained, is difficult to identify in rearranged chromosomes. Thus, cryptic translocations might have been erroneously interpreted as deletions. We selected two cell lines with an apparent deficiency of the short arm of chromosome 10 (W198 and HELl HBLT), and one for which no quantitative differences existed between the short and the long arms IJDA2). Indeed, in the two cell lines with an apparent deficiency of chromosome 10p, in situ hybridization revealed that this chromosome arm was involved in the formation of a derivative chromosome incompletely identified. In cell line WI 98, in which the short arm of a chromosome 2 was found partially deleted, the hybridization spots for vimentin were recurrently observed on the proximal part of the deleted short arm, which was in fact composed of a whole chromosome 10p. Thus, the deletion of chromosome 2 affects its whole short arm. In cell line HELl HBLT, in which the short arm of chromosome 6 formed the long arm in of a derivative chromosome, the hybridization spots of vimentin gene were observed in the previously unidentified short arm. It is noteworthy that the short arm of chromosome 10 was found translocated with two chromosomes very frequently deleted, i.e., chromosomes 2 and 6. In both cases, the resulting derivative chromosome led to the loss of 2p and 6q. These losses were found in most other cell lines we studied but, in contrast to that of 10p, they seem to correspond to real situations because they affect large chromosome segments easy to identify. The fact that we could not identify the chromosome lOp clearly points out the limits of the cytogenetic approach in highly aneuploid and rearranged karyotypes, which frequently exist in transformed and malignant tissues. It is fairly satisfactory to observe that a certain relationship exists between the numbers of gene copies and the expression at the protein level. Indeed, disregulations

28

M. Baumgartner et al.

of gene expression may occur at m a n y stages in transformed or malignant cells, but the cases in which an overexpression is correlated with a loss of gene copies remain very rare. The single documented example seems to be p53, for which a large quantity of protein is observed in colorectal cancers [15], although the arm (17p) carrying the gene is deleted in about 80 of these cases [16]. It is likely that this overexpression is secondary to the deletion of the normal and the mutation of the r e m a i n i n g all e l e - w h i c h is highly transcribed [17]--and may contribute to transformation [18]. It is premature to consider whether this situation is exceptional or not, but our observation is compatible with the idea that the n u m b e r of chromosome arms, which may largely vary in tumor or transformed cells, is roughly proportional to the expression of the genes they carry, on the average. As a corollary, each discrepancy merits investigation, because it may indicate either cytogenetic inconsistency or a significant dysfunction of the gene considered, as in the case of p53 in colorectal cancer.

The authors are grateful to Drs. R. Cassingena and M. Fellous for providing the cell lines studied. This work was supported by Contract Cooperatif de Recherche, Institut Curie and ARC.

REFERENCES 1. Hoffschir F, Ricoul M, Dutrillaux B (19881:SV40 transformed human fibroblasts exhibit a characteristic chromosomal pattern. Cytogenet Cell Genet 49:264-268. 2. Baumgartner M, Dutrillaux B, Lemieux N, Lilienbaum A, Paulin D, Viegas-P~quignot E 11991/: Genes occupy a fixed and symmetrical position on sister chromatids. Cell 64:761-766. 3. Perreau J, Lilienbaum A, Vasseur M, Paulin D (1988]: Nucleotide sequence of the human vimentin gene and regulation of its transcription in tissues and cultured cells. Gene 62:7-16. 4. Girardi AJ, Jensen FC, Koprowski H (1965): SV-40-induced transformation of human diploid cells: crisis and recovery. J Cell Comp Physiol 65:69-84. 5. Dutrillaux B, Couturier J (1981): La pratique de l'analyse chromosomique. Masson, Paris, New York. 6. Lilienbaum A, Legagneux V, Portier MM, Dellagi K, Paulin D (1986): Vimentin gene: expression in human lymphocytes and in Burkitt's lymphoma cells. The EMBO Journal 5:2809-2814. 7. Viegas-P~quignot E, Dutrillaux B, Magdelenat H, Coppey-Moisan M (1989): Mapping of single copy DNA sequences on human chromosomes by in situ hybridization with biotinylated probes: Enhancement of detection sensitivityby intensified fluorescence digital imaging microscopy. Proc Acad Sci USA 86:582-586. 8. Human Gene Mapping 10 (1989): Cytogenet Cell Genet 51:1-4. 9. Dolbeare F, Vanderlaan M, Phares W (1980): Alkaline phosphatase and an acid arylamidase as marker enzymes for normal and transformed WI-38 cells. J Histol Cytochem 28:419-426. 10. Marlhens F, Nicole A, Sinet PM (1985): Lowered level of translatable messenger RNAs for manganese superoxide dismutase in human fibroblasts transformed by SV40. Biochem Biophys Research Commun 129:300-305. 11. Vuillaume M, Calvayrac R, BestoBelpomme M, Tarroux P, Hubert M, Decroix Y, Sarasin A (1986): Deficiency in the catalase activity of xeroderma pigmentosum cell and simian virus 40-transformed human cell extracts. Cancer Res 46:538-544. 12. Nishitani K, Namba M, Kimoto T (1981): Lactate dehydrogenase isoenzyme pattern of normal human fibroblasts and their in vitro transformed counterparts obtained by treatment with CO-60 gamma rays, SV40 or 4-nitroquinoline 1-oxide. Gann 72:300-304. 13. Bravard A, Luccioni C, Muleris M, Lefran~ois D, Dutrillaux B (1991): Relationships between UMPK and PGD activities and deletions of chromosome lp in colorectal cancers. Cancer Genet Cytogenet 56:45-56. 14. Namba M, Karai M, Kimoto T (1987): Comparison of major cytoskeletons among normal human fibroblasts, immortal human fibroblasts transformed by exposure to Coo60 gamma rays, and the latter cells made tumorigenic by treatment with Harvey murine sarcoma virus. Exp Gerontol 22:179-186.

10p Translocation Detection Using VIM

29

15. Crawford LV, Pim DC, Lamb P (1984): The cellular protein p53 in human tumours. Mol Biol Med 2:261-272. 16. Muleris M, Salmon RJ, Dutrillaux AM, Viehl P, Zafrani B, Girodet J, Dutrillaux B (1987): Characteristic chromosomal imbalances in 18 near-diploid colorectal tumors. Cancer Genet Cy~ogenet 29:289-302. 17. Baker SJ, Fearon ER, Nigro JM, Hamilton SR, Preisinger AC, Jessup JM, VanTuinen P, Ledbetter DH, Barker DF, Nakamura Y, White R, Vogelstein B (1989): Chromosome 17 deletions and p53 gene mutations in colorectal carcinomas. Science 244:217-221. 18. Michalovitz D, Halevy O, Oren M (1990): Conditional inhibition of transformation and of cell proliferation by a temperature-sensitive mutant of p53. Cell 62:671-680.