- Email: [email protected]
Detection of i(17q) chromosome by fluorescent in situ hybridization (FISH) with interphase nuclei in medulloblastoma
Detection of i(17q) Chromosome by Fluorescent In Situ Hybridization (FISH) with Interphase Nuclei in Medulloblastoma A. M. Vagner-Capodano, H. Zattara-Cannoni, D. Gambarelli, J. C. Gentet, L. Genitori, G. Lena, N. Graziani, C. Raybaud, M. Choux, and 17. Grisoli
ABSTRACT: Medulloblastomas are the most frequent primitive neurectodermal tumors in children. An isochromosome for the long arm of 17, i(17q), is found in 30% of medulloblastomas. For some authors, this abnormality is observed in cases with a shorter survival time. In our cylogenetic studies of 30 medulloblastomas, we observed i(17q) in only three cases, a monosomy 17 in two cases, a monosomy 22 in four cases, nonspecific numerical or structural abnormalities in five cases, and normal karyotypes in 12 cases. We compared the results of karyotypic analysis after culture and FISH with a chromosome 17 alpha satellite DNA probe on interphase nuclei in five cases of medulloblastoma. In one case, i(17q) was only observed in four cells in karyotypic analysis, in three cases a normal karyatype was found, and in one case karyotypic analysis was impossible. In all of these cases, i(17q) was observed in a great number of nuclei by FISH on interphase nuclei. Our study shows that the FISH on interphase nuclei permitted us to observe i(17q) in the cases where it was not or could not be completely detected by karyotypic analysis. The association of these two techniques is required to detect i(17q), an abnormality whose prognosis value in medulloblastomas is now recognized.
INTRODUCTION Primitive neurectodermal tumors (PNETs) of the central nervous system are the most malignant tumors in children. Medulloblastomas are the most frequent of PNETs [1] and are located in the posterior fossa. They account for 25% of childhood brain tumors. The long-term prognosis for many of these children remains negative. Efforts are being made to identify new prognostic factors; cytogenetics and molecular aspects of these tumors can be prognostic factors. Cytogenetics studies show that in 30-60% of medulloblastomas an i(17q) is the most constant chromosomal abnormality to be observed [2, 3]. Other less frequent changes are gains of chromosome 1, deletions of lp or lq, 6q, 11, and 16(t [2-4], unbalanced translocations of 20(t13 and t(8;11} (qll; p11) [4, 5], and double-minute chromosomes (dmin) [6]. For some authors, the presence of i(17q) is often associated with a shorter survival time [4, Z 8]. So it is very important to find
From the Cytogenetic Oncology Laboratory, Neurapathology Laboratory, Departments of Pediatric Oncology, Pediatric Neurosurgery, and Neurosurgery, H6pitol Timone, Marseille, France. Address reprint requests to: Professor A. M. Vagner-Capodano, H6pitol d'Adultes de la Timone, Bd. Jean Mou|in, 13385 Marseille, Cedex 5, France. Received November 1Z 1993; accepted January 21, 1994.
this abnormality in these tumors. However, culturing medulloblastomas is often difficult because in vitro these tumors have a slow proliferation activity and a low mitotic index. For this reason, we thought it would be particularly interesting to apply a fluorescence in situ hybridization technique (FISH) on interphase nuclei to show i(17q) in the cases where this abnormality was not observed in culture, especially in the cases in which only normal diploid cells were observed, when culture was unsuccessful, or when tumoral material was insufficient for culture. We then compared the results of karyotypic analysis after culture with FISH data on interphase nuclei obtained by imprints on slides of fresh tumor tissue. MATERIALS AND METHODS Thirty-one medulloblastoma tumors were collected over a period of 3 years from 29 children and two adults. Histologic sections from the tumors were considered as medulloblastomas using the traditional World Health Organization Classification [1]. Tumor biopsies were obtained from primary tumors.
Cytogenetic Technique Sterile tumor tissue (in 30 cases) was obtained directly from the operating room and transported immediately to the lab-
1 © 1994 Elsevier Science Inc. 655 Avenue of the Americas. New York, NY 10010
Cancer Genet Cytogenet 78:1-6 {1994) 0165-4608/94/$07.00
2 oratory in sterile RPMI medium. For culture the tissue was minced with disposable scalpels in a small volume of medium and then the fragments were dissociated enzymatically using RPMI medium containing 0.02 % collagenase type II. and incubated at 37°C for approximatively I hour. Cell suspensions were centrifuged and the pellets were seeded into 25-cm 2 tissue culture Falcon flasks with medium containing RPMI and antibiotics supplemented with 10% fetal calf serum. These culture flasks were incubated at 37°C in a 5% CO 2 atmosphere. Cultures were observed daily through a binocular magnifying glass, and when cultures entered exponential growth (5-12 days), chromosomal analysis was performed by a routine cytogenetic technique. R-banding technique [9] was used for chromosomal identification. In one case the tumor material was not sterile when it arrived at the laboratory and culture was impossible. In this case, only imprints of fresh tumor on slides were performed. Fish Procedure In five medulloblastomas, FISH was performed on interphase nuclei that were either obtained by smears on slides of cell pellets provided from cultures, in two cases, or by imprints on slides of flesh tumor, in three cases. FISH was realized with a biotinylated chromosome 17 alpha satellite DNA (D17Z17 [2.7 Kb, EcoRl fragment, Oncor), which was denatured at 76°C for 5 minutes. Slides were incubated in ribonuclease (RNase 7 (100 ~g/ml7 for 1 hour at 37°C. Afterwards, the slides were dehydrated in a graded ethanol series. A pepsin treatment (10 I~g/ml) was performed, followed by postfixation in paraformaldehyde. Denaturation of chromosomal DNA was done in 70% formamide, 2 × SSC, at 76°C for 3 minutes. Then the slides were dehydrated at 4°C. Hybridization buffer consisted of 0.4 ng/l~l probe, 60% formamide, 10% dextran sulfate and salmon sperm DNA (500 ~g/ml). Hybridization was performed in a humid chamber at 37°C for 15-18 hours. After this time, slides were washed once in a shaker far 20 minutes at 37°C in 60% formamide, followed by two washings in 2 x SSC at 43°C. Immunologic detection was performed according to the method of Benkhalifa et al. [10], with modifications. Slides were incubated with 5 ~g/m] avidin-FlTC (Vector Laboratories) with 15% nonfat dry milk, 4 x SSC, 0.2% Tween 20. Hybridization signals were amplified by alternating incubation with 0.5 I~g/ml of biotinylated anti-avidin goat antibody (Vector Laboratories) and avidin-FITC for 45 minutes at 37°C. After each incubation, slides were washed three times in 4 x SSC, 0.2% Tween 20 at 37°C for 5 minutes each. Counterstaining of interphase nuclei was done with propidium iodide (5 I~g/ml). Preparations were mounted in antifading agent and observed under a fluorescent microscope (filter combination, Axiophot ZEISS). On the other hand, FISH was performed on interphase nuclei of a lymphocyte culture provided by a normal control subject, under the same conditions as previously described. The number of fluorescent signals was counted in 500 to 800 cells according to the cases. The percentage of nuclei with one fluorescent spot, two fluorescent spots, or more than two spots was determined in the five medulloblastomas and in the normal control subject.
A . M . Vagner-Capodano et al. RESULTS The results of karyotypic analysis are shown in Table 1. We have observed only five cases with an abnormality concerning chromosome 17: either an i(17q) (cases 2, 8, 22) [Fig. 1), or a monosomy 17 (cases 1, 6). Other abnormalities described in medulloblastomas, such as dmin, were observed in one case (case 207. Nonspecffic chromosomal abnormalities were observed, such as isolated monosomy 22 in four cases (case Z 9, 11, and 13). In five cases (cases 1, 6, 8, 18, and 21) nonspecific structural or numerical changes were observed. In addition, normal karyotypes were found in 12 cases; in four cases the culture was unsuccessful. We thought it could be interesting to find i(17q) by FISH on interphase nuclei in all cases where this abnormality had not been observed by karyotypic analysis. Thus, in five medulloblastomas we compared the results of karyotypic analysis and those of FISH on interphase nuclei (Table 2). In case 22, we observed i(17q) by karyotypic analysis after culture in only four cells; only 30 metaphases could be analyzed. FISH, on the other hand, permitted us to study 800 interphase nuclei. By FISH an interphase nuclei we saw the presence of three bright fluorescent spots in some nuclei, one isolated spot and a duplicated fluorescence spot (Fig. 2). The isolated spot represents a normal chromosome 17. The centromeric region is indicated by one bright fluorescent spot. The two spots slightly apart from each other represent the centromeric region of i(17q). It has been well demonstrated that isochromosomes look like dicentric chromosomes [11]. In this patient i(17q) was found in 20% of the cells on smears of cell pellets, and in 18% of the ceils on imprints. In three cases [cases 5, 12, and 20) karyotypic analysis did not reveal the i(17q). Among these cases, two had normal karyotypes, one had a hyperdiploid clone and a normal clone [case 20). In all of these cases, we observed i(1;'q) in a great number of nuclei by FISH (Table 27. In case 20, we observed hyperdiploid nuclei with four fluorescent spots: two isolated spots which represent two normal chromosomes 17 and two spots apart from each other which represent an i(1Rt) (Fig. 3). In culture, we did not observe i(17a3 in the hyperdiploid clone. In the last case (case 31) i(17q) was found in 30% of the nuclei obtained by FISH procedure on imprints on slides of fresh tumor. But in this case a karyotype could not be performed because the tumor tissue was not sterile, and culture was not possible. FISH procedure performed in a normal lymphocyte culture showed that approximately 90% of the nuclei had two fluorescent spots. Approximately 10% of the nuclei had more or less than two fluorescent spots. When three spots were present they were set wide apart.
DISCUSSION Our cytogenetic results agree with those reported in the literature [3, 4, 12, 13]. Isochromosome 17q is the most frequent change observed in medulloblastomas. Other less frequent numerical or structural abnormalities are also observed, as well as diploid cells [4, 13]. Furthermore, we have observed four cases of medulloblastoma presenting a monosomy 22. This abnormality has been reported only in one medullo-
Detection of i(17)q with FISH in Medulloblastoma Table 1 Patient
Results of karyotypic analysis in 30 medulloblastomas Age (yr)
Histologic type
1
11
Medulloblastoma
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
42 2 5 4 3 15 12 3 11 5 12 6 1 1 3 10 10 4 35 10 8 11 6 1 4 2 2 6 8
Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medullomyoblastoma Desmoplastic medulloblastoma Medulloblastoma Desmoplastic medulloblastoma Medulloblastoma Desmoplastic medulloblastoma Medulloblastoma Desmoplastic medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma
Karyotype 46,XX,t(1;8){q25;q22},add(3)(q29),del(9){p12),dup(12)(q12q23) - 17, ÷ 2215]/46,XX[15] 49-52,XY, + 1,i(17}(q10)[cp10] 46,XX[16] 44-50,XY, + 21[cp15] 46,XY[25] 46,XY,add(10){q26), - 17, + mar[16] 45,XX, - 22115] 45,X, - X,i(17)(q10)[8]/46,XX[10] 46,XY, - 22110]/46,XY[5] No mitosis 43-45,XX, - 22[cp10] 46,XY[30] 45,XX, - 22141/46,XX[10] No mitosis 46,XY[18] 46,XY[15] No mitosis 45,XX,add(9)(p24),- 10115] 46,XY[201 46-51,XY, + 1, - 18,3dmin[cp10]/46,XY[51 49,XX, + 7, + 11, + 1214]/46,XX[10] 46,XX,i(17)(q10)[4]/46,XX[26] 46,XX[20] 46,XX[15] 48-54,XX, + 1, + 11[cp15] No mitosis 46,XX[15] 46,XY[18] 46,XY[20] 46,XY[161
blastoma [2]. But it is interesting to note that in all our cases the monosomy 22 was the single abnormality observed in these tumors that presented the same histologic aspect, i.e., desmoplastic medulloblastoma. Loss of 22 has been demonstrated in other brain tumors: in m e n i n g i o m a [14], acoustic neurinoma [15], neurofibromatosis [16], and in some gliomas [17]. Recently, James et al. [18] showed loss of genetic information from chromosome 22 in some medulloblastomas. These findings suggest that one or several different recessive tumor genes on chromosome 22 may play a critical role in CNS tumors [19]. Isochromosome 17q seems to be the only chromosomal abnormality with a prognosis value in these tumors. Isochromosome 17q is not an abnormality specific to medulloblastomas, although it is not observed in other brain tumors. This abnormality has been observed in some hematologic malignancies, such as chronic myelocytic leukemia [20] and acute lymphoblastic leukemia [21]. In chronic myelocytic leukemia, i(17q) appears with the blastic crisis of this disease [22]. In medulloblastomas, the presence of i(17q) might also be linked to tumoral progression and represent a poor prognosis factor. The studies of MacDonald et al. [8] with restriction fragment-length p o l y m o r p h i s m (RFLP) analysis of DNA in 31 tumors have shown that the clinically staged "good risk" patients with medulloblastoma have a signifi-
cantly shorter time to tumor progression and a significantly worse survival time if 17p deletions are present in their tumors. Clinically staged "poor risk" patients have a significantly longer survival time if 17p is present. In our study, by FISH using a chromosome 17-specific DNA probe with interphase nuclei, we have interpreted the presence of two bright fluorescent spots apart from each other as the two centromeric regions of i(17q). Indeed, it is generally assumed that i(17q) arises by transverse misdivision of the centromere [11] and the presence of two centromeric regions in i(17q) has been reported in several cases [23-25]. Nakagawa et al. [26] reported similar aspects of i(17q) by FISH using a chromosome 17-specific alpha satellite probe with interphase nuclei of chronic myelocytic leukemia. Some authors [26-30] showed that the FISH procedure with chromosome-specific probes on interphase nuclei might overcome the difficulty of analyzing a large number of metaphases in tumor cultures that have a low mitotic index. Indeed, in our cytogenetic study of 30 medulloblastomas, we observed i(17q) in only three cases. The comparative studies of results in karyotypic analysis and FISH procedure in five cases have shown that FISH on interphase nuclei allowed us to observe i(17q) in a greater number of cells than in cultures, thus permitting a quantitative study on a large number of interphase nuclei. In addition, the FISH proce-
~allg
'
'
ill 1
2
3
4
5
Sl ,J!lt :)!It J 'A 6
7
8
9
iitt ill'It 10
I1
Ott
12
X
It0
13
~l!~i
14
allt 16
15
IItU* n~ 17
18
8~
ma
19
20
21
t
O
22
Figure 1
Table 2
:It
Karyotype of a cell from a medulloblastoma (case 8): 45,X,- X,i(17)(qlO) (arrows).
C o m p a r a t i v e r e s u l t s of k a r y o t y p i c a n a l y s i s a n d FISH p r o c e d u r e in 5 m e d u l l o b l a s t o m a s Number of cells with i(17q) in metaphases after culture
Percentage of interphase nuclei with i(17q) by FISH procedure 14% 600 nuclei counted) ° 14% 600 nuclei counted) a 17% '600 nuclei counted) b
22
0 (25 mitosis analyzed: normal) 0 (30 mitosis analyzed: normal) 0 (10 hyperdiploid mitosis, 5 normal mitosis) 4 (30 mitosis analyzed)
31 Control lymphocyte culture
No culture 0 (30 mitosis analyzed: normal)
Patient 5 12 20
° FISH performed on smears of cell pellets. b FISH performed on imprints.
20% 18% 30% 0%
400 400 500 800
nuclei nuclei nuclei nuclei
counted) a counted) b counted) b counted] ~
Detection of i(17)q w i t h FISH in M e d u l l o b l a s t o m a
5
F i g u r e 2 FISH with D17Z1 probe in interphase nuclei (case 22). Three fluorescent spots are observed. One isolated spot indicates a normal chromosome 17; the duplicated spot indicates the presence of i(17q) (arrow).
Figure 3 FISH with D17Z1 probe in hyperdiploid interphase nuclei (case 20). We observed two fluorescent spots, which represent two normal chromosomes 17, and one duplicated fluorescent spot, which indicates the presence of one i(17q) (arrow).
d u r e detected this a b n o r m a l i t y in the case w h e r e o n l y normal d i p l o i d cells w e r e o b s e r v e d in culture. FISH on interphase nuclei also allowed us to detect i(17q) in the case w h e r e the c u l t u r e was i m p o s s i b l e d u e to insufficient or nonsterile t u m o r a l material. These results show that karyotypic analysis and FISH proc e d u r e are two c o m p l e m e n t a r y and necessary techniques for the identification of i(17q), w h o s e detection is crucial because this a b n o r m a l i t y is c o n s i d e r e d to be an i m p o r t a n t prognostic factor in m e d u l l o b l a s t o m a s .
8. McDonald JD, Daneshvar L, Willert JR, Metzger A, Mack E, Edwards MSB, Cogen PH (1993): Clinical and molecular genetic analysis of medulloblastoma. 5th International Symposium on Pediatric Neuro-Oncology: Medulloblastoma. Marseilles, June 12-15, 1993. 9. Dutrillaux B, Lejeune J (1971): Sur une nouvelle technique d'analyse du caryotype humain (a new technique for human karyotype analysis). CR Acad Sci 272:2638-2640. 10. Benkhalifa M, Arnold N, Le Corvaisier B, Geneix A, Malet P (1991): Human sex determination by in situ hybridization using nonradioactive probes. Genet Sel Evol 23:70S-72S. 11. Testa JR, Cohen BC (1986): Dicentric chromosome 17 in patients with leukemia. Cancer Genet Cytogenet 23:47-52. 12. Biegel JA, Rorke LB, Packer RJ, Sutton LN, Schut L, Bonner K, Emanuel BS (1989): lsochromosome 1RI in primitive neurectodermal tumors of the central nervous system. Genes Chrom Cancer 1:139-147. 13. Rasheed BKA, Bigner SH (1991): Genetic alterations in gliomas and medulloblastomas. Cancer Metastasis Rev 10:289-299. 14. Zang KD (1982): Cytological and cytogenetical studies on human meningiomas. Cancer Genet Cytogenet 6:249-274. 15. Seizinger BR, Martuza RL, Gusella JF (1986): Loss of genes on chromosomes 22 in tumorigenesis of human acoustic neuroma. Nature 322:644-647. 16. Rouleau GA, Wertelecki W, Haines JL, Hobbs WJ, Trofatter ]A, Seizinger BR, Martuza RL, Superneau DW, Conneally PM, Gusella JF (1987): Genetic linkage of bilateral acoustic neurofibromatosis to a DNA marker on chromosome 22. Nature 329:246-248. 17. Yamada K, Kondo T, Yoshiaka M, Oami H (1980): Cytogenetic studies in twenty human brain tumors: association of no. 22 chromosome abnormalities with tumors of the brain. Nature 2:293-307. 18. James CD, Carlbom E, Mikkelsen T, Ridderheim PA. Cavenee WK, Collins VP (1990): Loss of genetic information in central nervous system tumors common to children and young adults. Genes Chrom Cancer 2:94-102. 19. Rey JA, Bello JB, de Campos JM, Vaquero l, Kusak ME, Sarasa JL, Pestana A (1993): Abnormalities of chromosome 22 in human brain tumors determined by combined cytogenetic and molecular genetic approaches. Cancer Genet Cytogenet 66:1-10.
This work was supported by the "Association pour la Recherche Contre le Cancer." We thank Dr. Gouzien and Mrs. Y. Breusa, S. Durocher, D. Intagliata, and A. Ruoppolo for technical assistance.
REFERENCES 1. Rorke LB, Gilles FH, Davis RL, Becker LE (1985): Revision of the World Health Organization classification of brain tumors for childhood brain tumors. Cancer 56:1869-1886. 2. Bigner SH, Mark J, Friedman HS, Biegel JA, Bigner DD (1988): Structural chromosomal abnormalities in human medulloblastoma. Cancer Genet Cytogenet 30:91-101. 3. Griffin CA, Hawkins AL, Packer RJ, Rorke LB, Emanuel BS (1988): Chromosomal abnormalities in pediatric brain tumors. Cancer Res 48:175-180. 4. Bigner SH, Vogelstein B (1990): Cytogenetics and molecular genetics of malignant gliomas and medulloblastomas. Brain Pathol 1:12-18. 5. Callen DF, Cirocco L, Moore L (1989): A der(11) t(8;11) in two medulloblastomas: a possible nonrandom cytogenetic abnormality. Cancer Genet Cytogenet 38:255-260. 6. Cox D, Yuncken C, Spriggs AL (1985): Minute chromatin bodies in malignant tumors of childhood. Lancet 11:55-58. 7. Cogen P, Daneshvar L, Metzger AK, Edwards MSB (1990): Deletion mapping of the medulloblastoma locus on chromosome 17p. Genomics 8:279-285.
6 20. Misawa S, Taniwaki M, Abe T, Takino T (1987): Monocentric and dicentric structure of an isochromosome for the long arm of chromosome 17, i(17q), in blast crisis of chronic myelogenous leukemia. Acta Haematol Jpn 50:902-905. 21. Pui CH, Raimondi SC, Williams DL (1988}: lsochromosome 17q in childhood acute lymphoblastic leukemia: an adverse cytogenetic feature in association with hyperdiploidy? Leukemia 2:222-225. 22. Sandberg AA (1980): The cytogenetics of chronic myelocytic leukemia (CML): chronic phase and blastic crisis. Cancer Genet Cytogenet 1:217-228. 23. Atkin NB, Amin S, Brito-Babapulle V (1981}: Three or four copies of a dicentric 17q isochromosome in acute myeloproliferative disorder. Cancer Genet Cytogenet 3:75-80. 24. Wang-Peng J, Lee E, Knutsen T, Solanki D (1981}: Dicentric isochromosome for the long arm of chromosome 17, dic i(17q), in a patient with chronic myelogenous leukemia (CML). Cancer Genet Cytogenet 3:233-236. 25. Borgstrom GH, Vuopio T, de la Chapelle A (1982}: Abnormalities of chromosome 17 in myeloproliferative disorder. Cancer Genet Cytogenet 5:123-135. 26. Nakagawa H, Inazawa J, Misawa S, Tanaka S, Takashima T,
A. M, V a g n e r - C a p o d a n o et al.
Taniwaka M, Abe T, Kashima K (1992): Detection of an i(17q) chromosome by fluorescent in situ hybridization with a chromosome 17 alpha satellite DNA probe. Cancer Genet Cytogenet 62:140-143. 27. Nederlof PM, van der Flier S, Raap AK, Tanke HJ, van der Ploeg M, Kornips F, Geraedts JPM (1989): Detection of chromosome aberrations in interphase tumor nuclei by non radioactive in situ hybridization. Cancer Genet Cytogenet 42:87-98. 28. Cromer T, Langedent JE, Bruckner A, School HP, Shardin M, Hager HD, Devilee P, Pearson PL, van der Ploeg M (1986): Detection of chromosome aberrations in the human interphase nucleus by visualization of specific target DNAs with radioactive and nonradioactive in situ hybridization techniques: diagnosis of trisomy 18 with probe. Hum Genet 74:346-352. 29. Dekken H, Pizzolo JG, Kelsen DE Melamed MR (1990): Targeted cytogenetic analysis of gastric tumors by in situ hybridization with a set of chromosome specific DNA probes. Cancer 66: 491-497. 30. Viegas-Pequignot E, Jeanpierre M, Dutrillaux AM, GerbaultSeureau M, Muleris M, Dutrillaux B (1989): Detection of lq polysomy in interphase nuclei of human solid tumors with a hiotinylated probe. Hum Genet 81:311-314.
ABSTRACT: Medulloblastomas are the most frequent primitive neurectodermal tumors in children. An isochromosome for the long arm of 17, i(17q), is found in 30% of medulloblastomas. For some authors, this abnormality is observed in cases with a shorter survival time. In our cylogenetic studies of 30 medulloblastomas, we observed i(17q) in only three cases, a monosomy 17 in two cases, a monosomy 22 in four cases, nonspecific numerical or structural abnormalities in five cases, and normal karyotypes in 12 cases. We compared the results of karyotypic analysis after culture and FISH with a chromosome 17 alpha satellite DNA probe on interphase nuclei in five cases of medulloblastoma. In one case, i(17q) was only observed in four cells in karyotypic analysis, in three cases a normal karyatype was found, and in one case karyotypic analysis was impossible. In all of these cases, i(17q) was observed in a great number of nuclei by FISH on interphase nuclei. Our study shows that the FISH on interphase nuclei permitted us to observe i(17q) in the cases where it was not or could not be completely detected by karyotypic analysis. The association of these two techniques is required to detect i(17q), an abnormality whose prognosis value in medulloblastomas is now recognized.
INTRODUCTION Primitive neurectodermal tumors (PNETs) of the central nervous system are the most malignant tumors in children. Medulloblastomas are the most frequent of PNETs [1] and are located in the posterior fossa. They account for 25% of childhood brain tumors. The long-term prognosis for many of these children remains negative. Efforts are being made to identify new prognostic factors; cytogenetics and molecular aspects of these tumors can be prognostic factors. Cytogenetics studies show that in 30-60% of medulloblastomas an i(17q) is the most constant chromosomal abnormality to be observed [2, 3]. Other less frequent changes are gains of chromosome 1, deletions of lp or lq, 6q, 11, and 16(t [2-4], unbalanced translocations of 20(t13 and t(8;11} (qll; p11) [4, 5], and double-minute chromosomes (dmin) [6]. For some authors, the presence of i(17q) is often associated with a shorter survival time [4, Z 8]. So it is very important to find
From the Cytogenetic Oncology Laboratory, Neurapathology Laboratory, Departments of Pediatric Oncology, Pediatric Neurosurgery, and Neurosurgery, H6pitol Timone, Marseille, France. Address reprint requests to: Professor A. M. Vagner-Capodano, H6pitol d'Adultes de la Timone, Bd. Jean Mou|in, 13385 Marseille, Cedex 5, France. Received November 1Z 1993; accepted January 21, 1994.
this abnormality in these tumors. However, culturing medulloblastomas is often difficult because in vitro these tumors have a slow proliferation activity and a low mitotic index. For this reason, we thought it would be particularly interesting to apply a fluorescence in situ hybridization technique (FISH) on interphase nuclei to show i(17q) in the cases where this abnormality was not observed in culture, especially in the cases in which only normal diploid cells were observed, when culture was unsuccessful, or when tumoral material was insufficient for culture. We then compared the results of karyotypic analysis after culture with FISH data on interphase nuclei obtained by imprints on slides of fresh tumor tissue. MATERIALS AND METHODS Thirty-one medulloblastoma tumors were collected over a period of 3 years from 29 children and two adults. Histologic sections from the tumors were considered as medulloblastomas using the traditional World Health Organization Classification [1]. Tumor biopsies were obtained from primary tumors.
Cytogenetic Technique Sterile tumor tissue (in 30 cases) was obtained directly from the operating room and transported immediately to the lab-
1 © 1994 Elsevier Science Inc. 655 Avenue of the Americas. New York, NY 10010
Cancer Genet Cytogenet 78:1-6 {1994) 0165-4608/94/$07.00
2 oratory in sterile RPMI medium. For culture the tissue was minced with disposable scalpels in a small volume of medium and then the fragments were dissociated enzymatically using RPMI medium containing 0.02 % collagenase type II. and incubated at 37°C for approximatively I hour. Cell suspensions were centrifuged and the pellets were seeded into 25-cm 2 tissue culture Falcon flasks with medium containing RPMI and antibiotics supplemented with 10% fetal calf serum. These culture flasks were incubated at 37°C in a 5% CO 2 atmosphere. Cultures were observed daily through a binocular magnifying glass, and when cultures entered exponential growth (5-12 days), chromosomal analysis was performed by a routine cytogenetic technique. R-banding technique [9] was used for chromosomal identification. In one case the tumor material was not sterile when it arrived at the laboratory and culture was impossible. In this case, only imprints of fresh tumor on slides were performed. Fish Procedure In five medulloblastomas, FISH was performed on interphase nuclei that were either obtained by smears on slides of cell pellets provided from cultures, in two cases, or by imprints on slides of flesh tumor, in three cases. FISH was realized with a biotinylated chromosome 17 alpha satellite DNA (D17Z17 [2.7 Kb, EcoRl fragment, Oncor), which was denatured at 76°C for 5 minutes. Slides were incubated in ribonuclease (RNase 7 (100 ~g/ml7 for 1 hour at 37°C. Afterwards, the slides were dehydrated in a graded ethanol series. A pepsin treatment (10 I~g/ml) was performed, followed by postfixation in paraformaldehyde. Denaturation of chromosomal DNA was done in 70% formamide, 2 × SSC, at 76°C for 3 minutes. Then the slides were dehydrated at 4°C. Hybridization buffer consisted of 0.4 ng/l~l probe, 60% formamide, 10% dextran sulfate and salmon sperm DNA (500 ~g/ml). Hybridization was performed in a humid chamber at 37°C for 15-18 hours. After this time, slides were washed once in a shaker far 20 minutes at 37°C in 60% formamide, followed by two washings in 2 x SSC at 43°C. Immunologic detection was performed according to the method of Benkhalifa et al. [10], with modifications. Slides were incubated with 5 ~g/m] avidin-FlTC (Vector Laboratories) with 15% nonfat dry milk, 4 x SSC, 0.2% Tween 20. Hybridization signals were amplified by alternating incubation with 0.5 I~g/ml of biotinylated anti-avidin goat antibody (Vector Laboratories) and avidin-FITC for 45 minutes at 37°C. After each incubation, slides were washed three times in 4 x SSC, 0.2% Tween 20 at 37°C for 5 minutes each. Counterstaining of interphase nuclei was done with propidium iodide (5 I~g/ml). Preparations were mounted in antifading agent and observed under a fluorescent microscope (filter combination, Axiophot ZEISS). On the other hand, FISH was performed on interphase nuclei of a lymphocyte culture provided by a normal control subject, under the same conditions as previously described. The number of fluorescent signals was counted in 500 to 800 cells according to the cases. The percentage of nuclei with one fluorescent spot, two fluorescent spots, or more than two spots was determined in the five medulloblastomas and in the normal control subject.
A . M . Vagner-Capodano et al. RESULTS The results of karyotypic analysis are shown in Table 1. We have observed only five cases with an abnormality concerning chromosome 17: either an i(17q) (cases 2, 8, 22) [Fig. 1), or a monosomy 17 (cases 1, 6). Other abnormalities described in medulloblastomas, such as dmin, were observed in one case (case 207. Nonspecffic chromosomal abnormalities were observed, such as isolated monosomy 22 in four cases (case Z 9, 11, and 13). In five cases (cases 1, 6, 8, 18, and 21) nonspecific structural or numerical changes were observed. In addition, normal karyotypes were found in 12 cases; in four cases the culture was unsuccessful. We thought it could be interesting to find i(17q) by FISH on interphase nuclei in all cases where this abnormality had not been observed by karyotypic analysis. Thus, in five medulloblastomas we compared the results of karyotypic analysis and those of FISH on interphase nuclei (Table 2). In case 22, we observed i(17q) by karyotypic analysis after culture in only four cells; only 30 metaphases could be analyzed. FISH, on the other hand, permitted us to study 800 interphase nuclei. By FISH an interphase nuclei we saw the presence of three bright fluorescent spots in some nuclei, one isolated spot and a duplicated fluorescence spot (Fig. 2). The isolated spot represents a normal chromosome 17. The centromeric region is indicated by one bright fluorescent spot. The two spots slightly apart from each other represent the centromeric region of i(17q). It has been well demonstrated that isochromosomes look like dicentric chromosomes [11]. In this patient i(17q) was found in 20% of the cells on smears of cell pellets, and in 18% of the ceils on imprints. In three cases [cases 5, 12, and 20) karyotypic analysis did not reveal the i(17q). Among these cases, two had normal karyotypes, one had a hyperdiploid clone and a normal clone [case 20). In all of these cases, we observed i(1;'q) in a great number of nuclei by FISH (Table 27. In case 20, we observed hyperdiploid nuclei with four fluorescent spots: two isolated spots which represent two normal chromosomes 17 and two spots apart from each other which represent an i(1Rt) (Fig. 3). In culture, we did not observe i(17a3 in the hyperdiploid clone. In the last case (case 31) i(17q) was found in 30% of the nuclei obtained by FISH procedure on imprints on slides of fresh tumor. But in this case a karyotype could not be performed because the tumor tissue was not sterile, and culture was not possible. FISH procedure performed in a normal lymphocyte culture showed that approximately 90% of the nuclei had two fluorescent spots. Approximately 10% of the nuclei had more or less than two fluorescent spots. When three spots were present they were set wide apart.
DISCUSSION Our cytogenetic results agree with those reported in the literature [3, 4, 12, 13]. Isochromosome 17q is the most frequent change observed in medulloblastomas. Other less frequent numerical or structural abnormalities are also observed, as well as diploid cells [4, 13]. Furthermore, we have observed four cases of medulloblastoma presenting a monosomy 22. This abnormality has been reported only in one medullo-
Detection of i(17)q with FISH in Medulloblastoma Table 1 Patient
Results of karyotypic analysis in 30 medulloblastomas Age (yr)
Histologic type
1
11
Medulloblastoma
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
42 2 5 4 3 15 12 3 11 5 12 6 1 1 3 10 10 4 35 10 8 11 6 1 4 2 2 6 8
Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medullomyoblastoma Desmoplastic medulloblastoma Medulloblastoma Desmoplastic medulloblastoma Medulloblastoma Desmoplastic medulloblastoma Medulloblastoma Desmoplastic medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma Medulloblastoma
Karyotype 46,XX,t(1;8){q25;q22},add(3)(q29),del(9){p12),dup(12)(q12q23) - 17, ÷ 2215]/46,XX[15] 49-52,XY, + 1,i(17}(q10)[cp10] 46,XX[16] 44-50,XY, + 21[cp15] 46,XY[25] 46,XY,add(10){q26), - 17, + mar[16] 45,XX, - 22115] 45,X, - X,i(17)(q10)[8]/46,XX[10] 46,XY, - 22110]/46,XY[5] No mitosis 43-45,XX, - 22[cp10] 46,XY[30] 45,XX, - 22141/46,XX[10] No mitosis 46,XY[18] 46,XY[15] No mitosis 45,XX,add(9)(p24),- 10115] 46,XY[201 46-51,XY, + 1, - 18,3dmin[cp10]/46,XY[51 49,XX, + 7, + 11, + 1214]/46,XX[10] 46,XX,i(17)(q10)[4]/46,XX[26] 46,XX[20] 46,XX[15] 48-54,XX, + 1, + 11[cp15] No mitosis 46,XX[15] 46,XY[18] 46,XY[20] 46,XY[161
blastoma [2]. But it is interesting to note that in all our cases the monosomy 22 was the single abnormality observed in these tumors that presented the same histologic aspect, i.e., desmoplastic medulloblastoma. Loss of 22 has been demonstrated in other brain tumors: in m e n i n g i o m a [14], acoustic neurinoma [15], neurofibromatosis [16], and in some gliomas [17]. Recently, James et al. [18] showed loss of genetic information from chromosome 22 in some medulloblastomas. These findings suggest that one or several different recessive tumor genes on chromosome 22 may play a critical role in CNS tumors [19]. Isochromosome 17q seems to be the only chromosomal abnormality with a prognosis value in these tumors. Isochromosome 17q is not an abnormality specific to medulloblastomas, although it is not observed in other brain tumors. This abnormality has been observed in some hematologic malignancies, such as chronic myelocytic leukemia [20] and acute lymphoblastic leukemia [21]. In chronic myelocytic leukemia, i(17q) appears with the blastic crisis of this disease [22]. In medulloblastomas, the presence of i(17q) might also be linked to tumoral progression and represent a poor prognosis factor. The studies of MacDonald et al. [8] with restriction fragment-length p o l y m o r p h i s m (RFLP) analysis of DNA in 31 tumors have shown that the clinically staged "good risk" patients with medulloblastoma have a signifi-
cantly shorter time to tumor progression and a significantly worse survival time if 17p deletions are present in their tumors. Clinically staged "poor risk" patients have a significantly longer survival time if 17p is present. In our study, by FISH using a chromosome 17-specific DNA probe with interphase nuclei, we have interpreted the presence of two bright fluorescent spots apart from each other as the two centromeric regions of i(17q). Indeed, it is generally assumed that i(17q) arises by transverse misdivision of the centromere [11] and the presence of two centromeric regions in i(17q) has been reported in several cases [23-25]. Nakagawa et al. [26] reported similar aspects of i(17q) by FISH using a chromosome 17-specific alpha satellite probe with interphase nuclei of chronic myelocytic leukemia. Some authors [26-30] showed that the FISH procedure with chromosome-specific probes on interphase nuclei might overcome the difficulty of analyzing a large number of metaphases in tumor cultures that have a low mitotic index. Indeed, in our cytogenetic study of 30 medulloblastomas, we observed i(17q) in only three cases. The comparative studies of results in karyotypic analysis and FISH procedure in five cases have shown that FISH on interphase nuclei allowed us to observe i(17q) in a greater number of cells than in cultures, thus permitting a quantitative study on a large number of interphase nuclei. In addition, the FISH proce-
~allg
'
'
ill 1
2
3
4
5
Sl ,J!lt :)!It J 'A 6
7
8
9
iitt ill'It 10
I1
Ott
12
X
It0
13
~l!~i
14
allt 16
15
IItU* n~ 17
18
8~
ma
19
20
21
t
O
22
Figure 1
Table 2
:It
Karyotype of a cell from a medulloblastoma (case 8): 45,X,- X,i(17)(qlO) (arrows).
C o m p a r a t i v e r e s u l t s of k a r y o t y p i c a n a l y s i s a n d FISH p r o c e d u r e in 5 m e d u l l o b l a s t o m a s Number of cells with i(17q) in metaphases after culture
Percentage of interphase nuclei with i(17q) by FISH procedure 14% 600 nuclei counted) ° 14% 600 nuclei counted) a 17% '600 nuclei counted) b
22
0 (25 mitosis analyzed: normal) 0 (30 mitosis analyzed: normal) 0 (10 hyperdiploid mitosis, 5 normal mitosis) 4 (30 mitosis analyzed)
31 Control lymphocyte culture
No culture 0 (30 mitosis analyzed: normal)
Patient 5 12 20
° FISH performed on smears of cell pellets. b FISH performed on imprints.
20% 18% 30% 0%
400 400 500 800
nuclei nuclei nuclei nuclei
counted) a counted) b counted) b counted] ~
Detection of i(17)q w i t h FISH in M e d u l l o b l a s t o m a
5
F i g u r e 2 FISH with D17Z1 probe in interphase nuclei (case 22). Three fluorescent spots are observed. One isolated spot indicates a normal chromosome 17; the duplicated spot indicates the presence of i(17q) (arrow).
Figure 3 FISH with D17Z1 probe in hyperdiploid interphase nuclei (case 20). We observed two fluorescent spots, which represent two normal chromosomes 17, and one duplicated fluorescent spot, which indicates the presence of one i(17q) (arrow).
d u r e detected this a b n o r m a l i t y in the case w h e r e o n l y normal d i p l o i d cells w e r e o b s e r v e d in culture. FISH on interphase nuclei also allowed us to detect i(17q) in the case w h e r e the c u l t u r e was i m p o s s i b l e d u e to insufficient or nonsterile t u m o r a l material. These results show that karyotypic analysis and FISH proc e d u r e are two c o m p l e m e n t a r y and necessary techniques for the identification of i(17q), w h o s e detection is crucial because this a b n o r m a l i t y is c o n s i d e r e d to be an i m p o r t a n t prognostic factor in m e d u l l o b l a s t o m a s .
8. McDonald JD, Daneshvar L, Willert JR, Metzger A, Mack E, Edwards MSB, Cogen PH (1993): Clinical and molecular genetic analysis of medulloblastoma. 5th International Symposium on Pediatric Neuro-Oncology: Medulloblastoma. Marseilles, June 12-15, 1993. 9. Dutrillaux B, Lejeune J (1971): Sur une nouvelle technique d'analyse du caryotype humain (a new technique for human karyotype analysis). CR Acad Sci 272:2638-2640. 10. Benkhalifa M, Arnold N, Le Corvaisier B, Geneix A, Malet P (1991): Human sex determination by in situ hybridization using nonradioactive probes. Genet Sel Evol 23:70S-72S. 11. Testa JR, Cohen BC (1986): Dicentric chromosome 17 in patients with leukemia. Cancer Genet Cytogenet 23:47-52. 12. Biegel JA, Rorke LB, Packer RJ, Sutton LN, Schut L, Bonner K, Emanuel BS (1989): lsochromosome 1RI in primitive neurectodermal tumors of the central nervous system. Genes Chrom Cancer 1:139-147. 13. Rasheed BKA, Bigner SH (1991): Genetic alterations in gliomas and medulloblastomas. Cancer Metastasis Rev 10:289-299. 14. Zang KD (1982): Cytological and cytogenetical studies on human meningiomas. Cancer Genet Cytogenet 6:249-274. 15. Seizinger BR, Martuza RL, Gusella JF (1986): Loss of genes on chromosomes 22 in tumorigenesis of human acoustic neuroma. Nature 322:644-647. 16. Rouleau GA, Wertelecki W, Haines JL, Hobbs WJ, Trofatter ]A, Seizinger BR, Martuza RL, Superneau DW, Conneally PM, Gusella JF (1987): Genetic linkage of bilateral acoustic neurofibromatosis to a DNA marker on chromosome 22. Nature 329:246-248. 17. Yamada K, Kondo T, Yoshiaka M, Oami H (1980): Cytogenetic studies in twenty human brain tumors: association of no. 22 chromosome abnormalities with tumors of the brain. Nature 2:293-307. 18. James CD, Carlbom E, Mikkelsen T, Ridderheim PA. Cavenee WK, Collins VP (1990): Loss of genetic information in central nervous system tumors common to children and young adults. Genes Chrom Cancer 2:94-102. 19. Rey JA, Bello JB, de Campos JM, Vaquero l, Kusak ME, Sarasa JL, Pestana A (1993): Abnormalities of chromosome 22 in human brain tumors determined by combined cytogenetic and molecular genetic approaches. Cancer Genet Cytogenet 66:1-10.
This work was supported by the "Association pour la Recherche Contre le Cancer." We thank Dr. Gouzien and Mrs. Y. Breusa, S. Durocher, D. Intagliata, and A. Ruoppolo for technical assistance.
REFERENCES 1. Rorke LB, Gilles FH, Davis RL, Becker LE (1985): Revision of the World Health Organization classification of brain tumors for childhood brain tumors. Cancer 56:1869-1886. 2. Bigner SH, Mark J, Friedman HS, Biegel JA, Bigner DD (1988): Structural chromosomal abnormalities in human medulloblastoma. Cancer Genet Cytogenet 30:91-101. 3. Griffin CA, Hawkins AL, Packer RJ, Rorke LB, Emanuel BS (1988): Chromosomal abnormalities in pediatric brain tumors. Cancer Res 48:175-180. 4. Bigner SH, Vogelstein B (1990): Cytogenetics and molecular genetics of malignant gliomas and medulloblastomas. Brain Pathol 1:12-18. 5. Callen DF, Cirocco L, Moore L (1989): A der(11) t(8;11) in two medulloblastomas: a possible nonrandom cytogenetic abnormality. Cancer Genet Cytogenet 38:255-260. 6. Cox D, Yuncken C, Spriggs AL (1985): Minute chromatin bodies in malignant tumors of childhood. Lancet 11:55-58. 7. Cogen P, Daneshvar L, Metzger AK, Edwards MSB (1990): Deletion mapping of the medulloblastoma locus on chromosome 17p. Genomics 8:279-285.
6 20. Misawa S, Taniwaki M, Abe T, Takino T (1987): Monocentric and dicentric structure of an isochromosome for the long arm of chromosome 17, i(17q), in blast crisis of chronic myelogenous leukemia. Acta Haematol Jpn 50:902-905. 21. Pui CH, Raimondi SC, Williams DL (1988}: lsochromosome 17q in childhood acute lymphoblastic leukemia: an adverse cytogenetic feature in association with hyperdiploidy? Leukemia 2:222-225. 22. Sandberg AA (1980): The cytogenetics of chronic myelocytic leukemia (CML): chronic phase and blastic crisis. Cancer Genet Cytogenet 1:217-228. 23. Atkin NB, Amin S, Brito-Babapulle V (1981}: Three or four copies of a dicentric 17q isochromosome in acute myeloproliferative disorder. Cancer Genet Cytogenet 3:75-80. 24. Wang-Peng J, Lee E, Knutsen T, Solanki D (1981}: Dicentric isochromosome for the long arm of chromosome 17, dic i(17q), in a patient with chronic myelogenous leukemia (CML). Cancer Genet Cytogenet 3:233-236. 25. Borgstrom GH, Vuopio T, de la Chapelle A (1982}: Abnormalities of chromosome 17 in myeloproliferative disorder. Cancer Genet Cytogenet 5:123-135. 26. Nakagawa H, Inazawa J, Misawa S, Tanaka S, Takashima T,
A. M, V a g n e r - C a p o d a n o et al.
Taniwaka M, Abe T, Kashima K (1992): Detection of an i(17q) chromosome by fluorescent in situ hybridization with a chromosome 17 alpha satellite DNA probe. Cancer Genet Cytogenet 62:140-143. 27. Nederlof PM, van der Flier S, Raap AK, Tanke HJ, van der Ploeg M, Kornips F, Geraedts JPM (1989): Detection of chromosome aberrations in interphase tumor nuclei by non radioactive in situ hybridization. Cancer Genet Cytogenet 42:87-98. 28. Cromer T, Langedent JE, Bruckner A, School HP, Shardin M, Hager HD, Devilee P, Pearson PL, van der Ploeg M (1986): Detection of chromosome aberrations in the human interphase nucleus by visualization of specific target DNAs with radioactive and nonradioactive in situ hybridization techniques: diagnosis of trisomy 18 with probe. Hum Genet 74:346-352. 29. Dekken H, Pizzolo JG, Kelsen DE Melamed MR (1990): Targeted cytogenetic analysis of gastric tumors by in situ hybridization with a set of chromosome specific DNA probes. Cancer 66: 491-497. 30. Viegas-Pequignot E, Jeanpierre M, Dutrillaux AM, GerbaultSeureau M, Muleris M, Dutrillaux B (1989): Detection of lq polysomy in interphase nuclei of human solid tumors with a hiotinylated probe. Hum Genet 81:311-314.