- Email: [email protected]
Application of multiplex fluorescence in situ hybridization in the cytogenetic analysis of primary gastric carcinoma
Cancer Genetics and Cytogenetics 135 (2002) 23–27
Application of multiplex fluorescence in situ hybridization in the cytogenetic analysis of primary gastric carcinoma Maria I. Stamoulia, Angeliki D. Fertib,*, Anna D. Panania, John Raftakisa, Carla Consolic, Sotirios A. Raptisa, Bryan D. Youngc a
Second Department of Internal Medicine Propaedeutic, Athens University, Evangelismos Hospital, Athens, Greece b Department of Internal Medicine, Sotiria Hospital, Athens, Greece c Department of Medical Oncology, ICRF, St. Bartholomew and the Royal London School of Medicine and Dentistry, London, UK Received 29 June 2001; received in revised form 30 October 2001; accepted 31 October 2001
Abstract
The different genetic alterations observed in diffuse and intestinal types of gastric cancer suggest that these two pathological types may represent different disease entities. We present two cases of primary gastric carcinoma, a well-differentiated intestinal type adenocarcinoma and a poorly differentiated diffuse type adenocarcinoma, both studied by a 24-color multiplex fluorescence in situ hybridization technique (M-FISH). The well-differentiated intestinal type adenocarinoma exhibited fewer structural abnormalities with five noncomplex translocations, deletions of chromosomes 5q, 6q, and 17q and an i(8q). In the case of poorly differentiated diffuse carcinoma, structural abnormalities predominated and normal homologues were mostly absent. But there were also similarities between the two cases: translocations on 1p and 9p; structural abnormalities of chromosome 8 with consistent loss of 8p; structural abnormalities of 12q; partial loss of chromosome 17 and 18; and polysomy of chromosome 20. This study shows that M-FISH is valuable in identifying hidden structural abnormalities and could, therefore, be useful in the investigation of primary solid tumors. © 2002 Elsevier Science Inc. All rights reserved.
1. Introduction In gastric cancer, cytogenetic and molecular studies have identified chromosomal gains and losses, gene amplifications, and mutations in oncogenes and tumor suppressor genes. The two histological types, poorly and well-differentiated gastric cancer, have similar as well as different genetic alterations suggesting that the accumulation of genetic changes could be responsible for two different genetic pathways leading to two distinct entities of stomach cancer [1,2]. Although there is no clear agreement on the genetic changes present in this type of malignancy, the advent of multiplex fluorescence in situ hybridization (M-FISH) may clarify these genetic events in gastric cancer. Previously reported cytogenetic studies on gastric carcinoma refer to cases with simple abnormalities such as polysomy X; trisomy 8, 9, and 19; del(7q); i(8q) [3–6]; and cases with complex abnormalities involving chromosomes 1, 3, 6, 7, 11, 13, and 17 [5,7–11]. Among recent molecular cytogenetic techniques, comparative genomic hybridization (CGH) identified losses at * Corresponding author. Tel.: 30-1-770-2935; fax: 30-1-770-2935. E-mail address: [email protected] (A. Ferti).
5q14q22, 17p13, 18q11q21, and chromosome 19 and gains of chromosomes X and 20 [12–15]. We present the findings of an M-FISH analysis in primary tumor cells of two patients with gastric cancer. 2. Materials and methods Tumor specimens from two male gastric cancer patients, 68 (case 1) and 55 (case 2) years of age, were studied. Histology revealed an intestinal type adenocarcinoma with moderate differentiation in case 1 and a poorly differentiated diffuse adenocarcinoma in case 2. Tissue specimens were obtained immediately after surgery. The samples were mechanically disaggregated in small quantities of McCoy culture medium (containing 10% FCS). Colcemid was added to the cell suspension. After 2 hours of exposure to Colcemid, the cell suspension was centrifuged at 1500 rpm for 5 minutes. Hypotonic treatment with KCl (75 mM) followed. Finally, the cells were fixed using freshly prepared methanol/acetic acid (3:1). Slides were stained by a G-banding technique. Twenty-four-color M-FISH was applied on slides not previously G-banded. The SpectraVysion Assay system (Vysis, Downers Grove, IL, USA) was used. SpectraVysion consists
0165-4608/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S0165-4608(01)00 6 3 5 - 5
24
M.I. Stamouli et al. / Cancer Genetics and Cytogenetics 135 (2002) 23–27
of a 5-label, 52-probe mixture of whole chromosome paint (WCP) DNA probes that cover the 24 human chromosome analogues. The WCP probes are directly labeled with the five different fluorophores in a combinatorial labeling format to provide 24 distinct colors. The fluorophores that are directly attached to the probes include SpectrumAqua, SpectrumGreen, SpectrumGold, SpectrumRed, and SpectrumFarRed. Cot 1 DNA was included in the probe mixture to suppress sequences that are common to various chromosomes. The slides were heated at 60C for 10 minutes and were left afterwards at room temperature for about 16 hours. Then they were either used immediately or stored at 20C. The slides were denaturated in 70% formamide/2SCC at 721C for 1–3 minutes and dehydrated in a series of 1 minute EtOH washes (70%, 95%, 100%) to maintain the denatured strands in the specimen. Ten microliters of SpectraVysion probe was denatured in a 73C water bath for 5 minutes and was applied to a target area on the slide (approximately 2222 mm). Hybridization of the probe to the specimen at 37C (for 12–18 hours) followed. The nonspecifically bound probe was removed in a wash using 0.4SSC/0.3% NP-40 at 72C for 2 minutes, followed by a 2SSC/0.1% NP-40 wash at ambient temperature for 30 seconds. Then the slides were air-dried and counterstained using DAPI-III (Vysis). The hybridization of the probe with the cellular DNA site was visualized by fluorescence microscopy (PSI system). 3. Results In case 1, cells were studied both by M-FISH (5 metaphases) and G-banding (10 metaphases). In case 2, only M-FISH was successful (10 metaphases). Both cases exhibited a near triploid karyotype. 3.1. Case 1 (Figs. 1 and 3) 6472,3n.,XX,X,Y,der(1)t(1;18)(p12;?),2,4, del (5)(q?13)2,del(5)(q?),der(5)t(5;12)(p?13;q13)2, del(6) (q13),del(6)(q22),7,8, i (8)(q10),der(9)t (9;13)(p10;q10), del(10)(q?),12,16,del(17)(q?)2,18,18, der(18)t (18;18)(p?;q?),20,20,20,der(21)t(20;21)(?;?),der (22)t (8;22)(q1?;?q),23mar 3.2. Case 2 (Fig. 2) 56573n,X,Y,der(X)t(X;13)(q?;q?),der(X)t(X;11), der(1)t(1;12),der(1)t(1;17),der(1)t(1;7),der(1)t(X;1;11), der(2)t(2;5),der(3)t(3;12)2,der(3)t(X;3;4),4,der(4)t(3;4), der(4) t (3;4;15)(?;?;?),der(5)t(5;8)(q;?),der(5)t(2;5),del(5)(q?), der(5)t(5;?),6,der(6)t(6;17)(?;?),7,der(7)t(7;20)(?;?), der(8)t (8;20)(?;?),der(8)t(8;21)(q10;q10) 2, der (9)t(9;20) ( p10 ;?), del(9) (p?), 10,del(10)(?),der(10)t(8;10)(?; p1?), 11, del(11)(?),del(12)(q?)2,der(12)t(3;12)(?;q?), 13, 13 ,14,der(14)t(3;14)(?;q3?),der(14)(?),15,der(15)t(1;15; 13;20) (?; ?; ?; ?),der (16)t (8;16)(?;?), der (16) t (15;16) (? ; ?),17,17, 17, 18, 18, 19,der(19)t (17;19)(q?;?), der(20)(?),21,21,21,22,3mar
4. Discussion For several decades conventional cytogenetic banding methods have been used for karyotyping analysis. Although they have been useful in the investigation of malignancies, especially hematologic disorders, banding techniques have their limitations. In solid tumors, the detection of recurring genetic changes by karyotyping is particularly problematic because of the difficulty in routinely preparing metaphase spreads of adequate quality and quantity and the complex nature of the chromosomal aberrations. The identification of the origin of marker chromosomes, based only on banding patterns, is extremely difficult, if not impossible, even with the best banding quality. Molecular cytogenetic techniques that have emerged recently have been proven valuable in solving some of these problems. Among these techniques M-FISH, by which each of the 24 human chromosomes can be distinguished [16], has so far only been applied to a limited number of primary solid tumors. In this study, the use of M-FISH in two cases of gastric carcinoma mostly facilitated the analysis of structural abnormalities and the identification of losses and gains of chromosomes. By this technique, hidden translocations such as t(1;18) in case 1, and t(6;17) and t(X;1;11) in case 2, could be identified and complex translocations such as t(1;15;13;20) and t(3;4;15) could be analyzed. Moreover, evidence suggesting reciprocal translocations (a rare finding in solid tumors), such as t(3;4) and t(3;12), was found in one case (case 2). Between the two cases of different histologic types, there were substantial differences. The well-differentiated intestinal type adenocarinoma exhibited fewer structural abnormalities with five noncomplex translocations, deletions of chromosomes 5q, 6q, and 17q and an i(8q). In the case of poorly differentiated diffuse carcinoma, structural abnormalities predominated and normal homologues were mostly absent. These, however, are not necessarily attributable to differences due to histologic type. Still, it is remarkable that among abnormalities the following were common in both cases: translocations on 1p and 9p with loss of the short arm; deletions and translocations at different sites of 5q leading to loss of material; involvement of chromosome 8 in various structural abnormalities with consistent loss of 8p and possibly excess of 8q; structural abnormalities of 12q; partial loss of chromosomes 17 and 18; and polysomy of chromosome 20. Loss of 1p has been observed previously both by conventional [4] as well as by molecular cytogenetic techniques [13]. In our study, loss of 1p is mainly the result of translocations. The APC gene has been isolated and found to be involved in colorectal cancers. Loss of heterozygosity (LOH) for this gene has been observed in gastric cancer as well [17,18]. Loss of 5q, where APC is located, was observed in both cases of this study (with a variety of breakpoints) and has also been described by means of CGH [13,15].
Fig. 1. Multiplex fluorescence in situ hybridization (FISH) and schematic representation of a metaphase showing abnormalities from case 1. Numbers in the schematic representation indicate the chromosomes involved.
Fig. 2. Multiplex FISH and schematic representation of a metaphase showing abnormalities from case 2. Numbers in the schematic representation indicate the chromosomes involved.
M.I. Stamouli et al. / Cancer Genetics and Cytogenetics 135 (2002) 23–27 25
26
M.I. Stamouli et al. / Cancer Genetics and Cytogenetics 135 (2002) 23–27
Fig. 3. Partial karyotypes from case 1 showing abnormalities described in Results.
Abnormalities of chromosome 8 are a frequent finding in this type of malignancy [5,13,14,19,20]. Whether they result in loss or excess of chromosome 8 material is still in dispute. There is one case report where excess of 8q is the sole abnormality [6]. M-FISH results add data favoring loss of 8p and possibly the gain of 8q. Involvement of chromosome 18 has been emphasized only in molecular studies [1,13]. Loss of chromosome 18 has been observed in both of our cases. In this study chromosome 18 was also involved in structural abnormalities, such as t(1;18) and t(18;18), which would escape identification without M-FISH. Translocations on 9p and loss of material from 9q21 q24 have previously been reported [4,13]. Translocations on 9p with an apparently common breakpoint and loss of 9p material were equally observed in both cases of this study. Thus, genes on 9p could play a significant role in the evolution of gastric cancer. Gain of chromosome 20 is especially frequent in gastric cancer [12–14] and it was a common finding of this study. A combined loss of APC gene and p53 gene (located at 5q and 17p, respectively) has been suggested as a common late event in gastric carcinoma [18]. Our cases provide further supporting cytogenetic evidence for a combined loss of these gene loci. In conclusion, the results show that the M-FISH technique gives valuable information that may significantly contribute to the evaluation of genetic events in gastric carcinoma. Although the analysis of two tumors may not be sufficient to draw meaningful conclusions, these observations may serve as a starting point for the search of the rele-
vant gene alterations and gives justification for additional study of the two pathological types of gastric cancer. References [1] Kullmann F, McClelland M. Letter. N Engl J Med 1995;333:1427–8. [2] Chan AO, Luk JM, Hui WM, Lam SK. Molecular biology of gastric carcinoma: from laboratory to bedside. J Gastroenterol Hepatol 1999; 14:1150–60. [3] Ochi H, Takeuchi J, Douglass H, Sandberg AA. Trisomy X as a possible initial chromosome change in a gastric cancer. Cancer Genet Cytogenet 1984;12:57–61. [4] Ferti-Passantonopoulou AD, Panani AD, Vlachos JD, Raptis SA. Common cytogenetic findings in gastric cancer. Cancer Genet Cytogenet 1987;24:63–73. [5] Xiao S, Geng JS, Feng XL, Liu XQ, Liu QZ, Li P. Cytogenetic studies of eight primary gastric cancers. Cancer Genet Cytogenet 1992;58:79–84. [6] Panani AD, Ferti A, Malliaros S, Raptis S. Gastric cancer with an i(8q) and long survival. Cancer Genet Cytogenet 1992;58:214–15. [7] Ochi H, Douglass H, Sandberg AA. Cytogenetic studies in primary gastric cancer. Cancer Genet Cytogenet 1986;22:295–307. [8] Rodriguez E, Rao PH, Ladanui M, Altorki N, Albino AP, Kelsen DP, Jhanwar SC, Chaganti RSK. 11p13–15 is a specific region of chromosomal rearrangement in gastric and esophageal adenocarcinomas. Cancer Res 1990;50:6410–16. [9] Misawa S, Horiike S, Taniwaki M, Tsuda S, Okuda T, Kashima K, Abe T, Sugihara H, Noriki S, Fukunda M. Chromosome abnormalities of gastric cancer detected in cancerous effusions. Jpn J Cancer Res 1990;81:148–52. [10] Seruca R, Castedo S, Correia C, Gomes P, Carneiro F, Soares P, Jong B, Sobrinho-Simoes M. Cytogenetic findings in eleven gastric carcinomas. Cancer Genet Cytogenet 1993;68:42–8. [11] Panani AD, Ferti A, Malliaros S, Raptis S. Cytogenetic study of 11 gastric adenocarcinomas. Cancer Genet Cytogenet 1995;81:169–72. [12] Kokkola A, Monni O, Puolakkainen P, Larramendy ML, Victorzon M, Nordling S, Haapiainen R, Kivilaakso E, Knuutila S. 17q12–21 am-
M.I. Stamouli et al. / Cancer Genetics and Cytogenetics 135 (2002) 23–27 plification, a novel recurrent genetic change in intestinal type of gastric carcinoma. Genes Chromosomes Cancer 1997;20:38–43. [13] Sakakura C, Mori T, Sakabe T, Ariyama Y, Shinomiya T, Date K, Hagiwara A, Yamaguchi T, Takayashi T, Nakamura Y, Abe T, Inazawa J. Gains, losses and amplifications of genomic materials in primary gastric cancers analyzed by comparative genomic hybridization. Genes Chromosomes Cancer 1999;24:299–305. [14] Koo SH, Kwon KC, Shin SY, Jeon YM, Park JW, Kim SH, Noh SM. Genetic alterations in gastric cancer: comparative genomic hybridization and fluorescence in situ hybridization studies. Cancer Genet Cytogenet 2000;117:97–103. [15] Nessling M, Solinas-Toldo S, Wilgenbus KK, Borchard F, Lichter P. Mapping of chromosomal imbalances in gastric adenocarcinoma revealed amplified protooncogenes MYCN, MET, WNT2 and ERBB2. Genes Chromosomes Cancer 1998;23:307–16.
27
[16] Speicher MR, Gwyn Ballard S, Ward D. Karyotyping human chromosomes by combinatorial multi-fluor FISH. Nat Genet 1996;12:368–75. [17] McKie AB, Filipe MI, Lemoine NR. Abnormalities affecting the APC and MCC tumour suppressor gene loci on chromosome 5q occur frequently in gastric cancer but not in pancreatic cancer. Cancer 1993;55:598–603. [18] Rhyu MG, Park WS, Jung YJ, Choi SW, Meltzer SJ. Allelic deletions of MCC/APC and p53 are frequent late events in human gastric carcinogenesis. Gastroenterology 1994;106:1584–8. [19] Castedo S, Correia C, David L, Sobrinho-Simoes M. Isochromosome 8q. A recurrent change in gastric carcinoma. Cancer Genet Cytogenet 1991;54:137–8. [20] Baffa R, Santoro R, Bullrich F, Mandes B, Ishii H, Croce CM. Definition and refinement of chromosome 8p regions of loss of heterozygosity in gastric cancer. Clin Cancer Res 2000;6:1372–7.
Application of multiplex fluorescence in situ hybridization in the cytogenetic analysis of primary gastric carcinoma Maria I. Stamoulia, Angeliki D. Fertib,*, Anna D. Panania, John Raftakisa, Carla Consolic, Sotirios A. Raptisa, Bryan D. Youngc a
Second Department of Internal Medicine Propaedeutic, Athens University, Evangelismos Hospital, Athens, Greece b Department of Internal Medicine, Sotiria Hospital, Athens, Greece c Department of Medical Oncology, ICRF, St. Bartholomew and the Royal London School of Medicine and Dentistry, London, UK Received 29 June 2001; received in revised form 30 October 2001; accepted 31 October 2001
Abstract
The different genetic alterations observed in diffuse and intestinal types of gastric cancer suggest that these two pathological types may represent different disease entities. We present two cases of primary gastric carcinoma, a well-differentiated intestinal type adenocarcinoma and a poorly differentiated diffuse type adenocarcinoma, both studied by a 24-color multiplex fluorescence in situ hybridization technique (M-FISH). The well-differentiated intestinal type adenocarinoma exhibited fewer structural abnormalities with five noncomplex translocations, deletions of chromosomes 5q, 6q, and 17q and an i(8q). In the case of poorly differentiated diffuse carcinoma, structural abnormalities predominated and normal homologues were mostly absent. But there were also similarities between the two cases: translocations on 1p and 9p; structural abnormalities of chromosome 8 with consistent loss of 8p; structural abnormalities of 12q; partial loss of chromosome 17 and 18; and polysomy of chromosome 20. This study shows that M-FISH is valuable in identifying hidden structural abnormalities and could, therefore, be useful in the investigation of primary solid tumors. © 2002 Elsevier Science Inc. All rights reserved.
1. Introduction In gastric cancer, cytogenetic and molecular studies have identified chromosomal gains and losses, gene amplifications, and mutations in oncogenes and tumor suppressor genes. The two histological types, poorly and well-differentiated gastric cancer, have similar as well as different genetic alterations suggesting that the accumulation of genetic changes could be responsible for two different genetic pathways leading to two distinct entities of stomach cancer [1,2]. Although there is no clear agreement on the genetic changes present in this type of malignancy, the advent of multiplex fluorescence in situ hybridization (M-FISH) may clarify these genetic events in gastric cancer. Previously reported cytogenetic studies on gastric carcinoma refer to cases with simple abnormalities such as polysomy X; trisomy 8, 9, and 19; del(7q); i(8q) [3–6]; and cases with complex abnormalities involving chromosomes 1, 3, 6, 7, 11, 13, and 17 [5,7–11]. Among recent molecular cytogenetic techniques, comparative genomic hybridization (CGH) identified losses at * Corresponding author. Tel.: 30-1-770-2935; fax: 30-1-770-2935. E-mail address: [email protected] (A. Ferti).
5q14q22, 17p13, 18q11q21, and chromosome 19 and gains of chromosomes X and 20 [12–15]. We present the findings of an M-FISH analysis in primary tumor cells of two patients with gastric cancer. 2. Materials and methods Tumor specimens from two male gastric cancer patients, 68 (case 1) and 55 (case 2) years of age, were studied. Histology revealed an intestinal type adenocarcinoma with moderate differentiation in case 1 and a poorly differentiated diffuse adenocarcinoma in case 2. Tissue specimens were obtained immediately after surgery. The samples were mechanically disaggregated in small quantities of McCoy culture medium (containing 10% FCS). Colcemid was added to the cell suspension. After 2 hours of exposure to Colcemid, the cell suspension was centrifuged at 1500 rpm for 5 minutes. Hypotonic treatment with KCl (75 mM) followed. Finally, the cells were fixed using freshly prepared methanol/acetic acid (3:1). Slides were stained by a G-banding technique. Twenty-four-color M-FISH was applied on slides not previously G-banded. The SpectraVysion Assay system (Vysis, Downers Grove, IL, USA) was used. SpectraVysion consists
0165-4608/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S0165-4608(01)00 6 3 5 - 5
24
M.I. Stamouli et al. / Cancer Genetics and Cytogenetics 135 (2002) 23–27
of a 5-label, 52-probe mixture of whole chromosome paint (WCP) DNA probes that cover the 24 human chromosome analogues. The WCP probes are directly labeled with the five different fluorophores in a combinatorial labeling format to provide 24 distinct colors. The fluorophores that are directly attached to the probes include SpectrumAqua, SpectrumGreen, SpectrumGold, SpectrumRed, and SpectrumFarRed. Cot 1 DNA was included in the probe mixture to suppress sequences that are common to various chromosomes. The slides were heated at 60C for 10 minutes and were left afterwards at room temperature for about 16 hours. Then they were either used immediately or stored at 20C. The slides were denaturated in 70% formamide/2SCC at 721C for 1–3 minutes and dehydrated in a series of 1 minute EtOH washes (70%, 95%, 100%) to maintain the denatured strands in the specimen. Ten microliters of SpectraVysion probe was denatured in a 73C water bath for 5 minutes and was applied to a target area on the slide (approximately 2222 mm). Hybridization of the probe to the specimen at 37C (for 12–18 hours) followed. The nonspecifically bound probe was removed in a wash using 0.4SSC/0.3% NP-40 at 72C for 2 minutes, followed by a 2SSC/0.1% NP-40 wash at ambient temperature for 30 seconds. Then the slides were air-dried and counterstained using DAPI-III (Vysis). The hybridization of the probe with the cellular DNA site was visualized by fluorescence microscopy (PSI system). 3. Results In case 1, cells were studied both by M-FISH (5 metaphases) and G-banding (10 metaphases). In case 2, only M-FISH was successful (10 metaphases). Both cases exhibited a near triploid karyotype. 3.1. Case 1 (Figs. 1 and 3) 6472,3n.,XX,X,Y,der(1)t(1;18)(p12;?),2,4, del (5)(q?13)2,del(5)(q?),der(5)t(5;12)(p?13;q13)2, del(6) (q13),del(6)(q22),7,8, i (8)(q10),der(9)t (9;13)(p10;q10), del(10)(q?),12,16,del(17)(q?)2,18,18, der(18)t (18;18)(p?;q?),20,20,20,der(21)t(20;21)(?;?),der (22)t (8;22)(q1?;?q),23mar 3.2. Case 2 (Fig. 2) 56573n,X,Y,der(X)t(X;13)(q?;q?),der(X)t(X;11), der(1)t(1;12),der(1)t(1;17),der(1)t(1;7),der(1)t(X;1;11), der(2)t(2;5),der(3)t(3;12)2,der(3)t(X;3;4),4,der(4)t(3;4), der(4) t (3;4;15)(?;?;?),der(5)t(5;8)(q;?),der(5)t(2;5),del(5)(q?), der(5)t(5;?),6,der(6)t(6;17)(?;?),7,der(7)t(7;20)(?;?), der(8)t (8;20)(?;?),der(8)t(8;21)(q10;q10) 2, der (9)t(9;20) ( p10 ;?), del(9) (p?), 10,del(10)(?),der(10)t(8;10)(?; p1?), 11, del(11)(?),del(12)(q?)2,der(12)t(3;12)(?;q?), 13, 13 ,14,der(14)t(3;14)(?;q3?),der(14)(?),15,der(15)t(1;15; 13;20) (?; ?; ?; ?),der (16)t (8;16)(?;?), der (16) t (15;16) (? ; ?),17,17, 17, 18, 18, 19,der(19)t (17;19)(q?;?), der(20)(?),21,21,21,22,3mar
4. Discussion For several decades conventional cytogenetic banding methods have been used for karyotyping analysis. Although they have been useful in the investigation of malignancies, especially hematologic disorders, banding techniques have their limitations. In solid tumors, the detection of recurring genetic changes by karyotyping is particularly problematic because of the difficulty in routinely preparing metaphase spreads of adequate quality and quantity and the complex nature of the chromosomal aberrations. The identification of the origin of marker chromosomes, based only on banding patterns, is extremely difficult, if not impossible, even with the best banding quality. Molecular cytogenetic techniques that have emerged recently have been proven valuable in solving some of these problems. Among these techniques M-FISH, by which each of the 24 human chromosomes can be distinguished [16], has so far only been applied to a limited number of primary solid tumors. In this study, the use of M-FISH in two cases of gastric carcinoma mostly facilitated the analysis of structural abnormalities and the identification of losses and gains of chromosomes. By this technique, hidden translocations such as t(1;18) in case 1, and t(6;17) and t(X;1;11) in case 2, could be identified and complex translocations such as t(1;15;13;20) and t(3;4;15) could be analyzed. Moreover, evidence suggesting reciprocal translocations (a rare finding in solid tumors), such as t(3;4) and t(3;12), was found in one case (case 2). Between the two cases of different histologic types, there were substantial differences. The well-differentiated intestinal type adenocarinoma exhibited fewer structural abnormalities with five noncomplex translocations, deletions of chromosomes 5q, 6q, and 17q and an i(8q). In the case of poorly differentiated diffuse carcinoma, structural abnormalities predominated and normal homologues were mostly absent. These, however, are not necessarily attributable to differences due to histologic type. Still, it is remarkable that among abnormalities the following were common in both cases: translocations on 1p and 9p with loss of the short arm; deletions and translocations at different sites of 5q leading to loss of material; involvement of chromosome 8 in various structural abnormalities with consistent loss of 8p and possibly excess of 8q; structural abnormalities of 12q; partial loss of chromosomes 17 and 18; and polysomy of chromosome 20. Loss of 1p has been observed previously both by conventional [4] as well as by molecular cytogenetic techniques [13]. In our study, loss of 1p is mainly the result of translocations. The APC gene has been isolated and found to be involved in colorectal cancers. Loss of heterozygosity (LOH) for this gene has been observed in gastric cancer as well [17,18]. Loss of 5q, where APC is located, was observed in both cases of this study (with a variety of breakpoints) and has also been described by means of CGH [13,15].
Fig. 1. Multiplex fluorescence in situ hybridization (FISH) and schematic representation of a metaphase showing abnormalities from case 1. Numbers in the schematic representation indicate the chromosomes involved.
Fig. 2. Multiplex FISH and schematic representation of a metaphase showing abnormalities from case 2. Numbers in the schematic representation indicate the chromosomes involved.
M.I. Stamouli et al. / Cancer Genetics and Cytogenetics 135 (2002) 23–27 25
26
M.I. Stamouli et al. / Cancer Genetics and Cytogenetics 135 (2002) 23–27
Fig. 3. Partial karyotypes from case 1 showing abnormalities described in Results.
Abnormalities of chromosome 8 are a frequent finding in this type of malignancy [5,13,14,19,20]. Whether they result in loss or excess of chromosome 8 material is still in dispute. There is one case report where excess of 8q is the sole abnormality [6]. M-FISH results add data favoring loss of 8p and possibly the gain of 8q. Involvement of chromosome 18 has been emphasized only in molecular studies [1,13]. Loss of chromosome 18 has been observed in both of our cases. In this study chromosome 18 was also involved in structural abnormalities, such as t(1;18) and t(18;18), which would escape identification without M-FISH. Translocations on 9p and loss of material from 9q21 q24 have previously been reported [4,13]. Translocations on 9p with an apparently common breakpoint and loss of 9p material were equally observed in both cases of this study. Thus, genes on 9p could play a significant role in the evolution of gastric cancer. Gain of chromosome 20 is especially frequent in gastric cancer [12–14] and it was a common finding of this study. A combined loss of APC gene and p53 gene (located at 5q and 17p, respectively) has been suggested as a common late event in gastric carcinoma [18]. Our cases provide further supporting cytogenetic evidence for a combined loss of these gene loci. In conclusion, the results show that the M-FISH technique gives valuable information that may significantly contribute to the evaluation of genetic events in gastric carcinoma. Although the analysis of two tumors may not be sufficient to draw meaningful conclusions, these observations may serve as a starting point for the search of the rele-
vant gene alterations and gives justification for additional study of the two pathological types of gastric cancer. References [1] Kullmann F, McClelland M. Letter. N Engl J Med 1995;333:1427–8. [2] Chan AO, Luk JM, Hui WM, Lam SK. Molecular biology of gastric carcinoma: from laboratory to bedside. J Gastroenterol Hepatol 1999; 14:1150–60. [3] Ochi H, Takeuchi J, Douglass H, Sandberg AA. Trisomy X as a possible initial chromosome change in a gastric cancer. Cancer Genet Cytogenet 1984;12:57–61. [4] Ferti-Passantonopoulou AD, Panani AD, Vlachos JD, Raptis SA. Common cytogenetic findings in gastric cancer. Cancer Genet Cytogenet 1987;24:63–73. [5] Xiao S, Geng JS, Feng XL, Liu XQ, Liu QZ, Li P. Cytogenetic studies of eight primary gastric cancers. Cancer Genet Cytogenet 1992;58:79–84. [6] Panani AD, Ferti A, Malliaros S, Raptis S. Gastric cancer with an i(8q) and long survival. Cancer Genet Cytogenet 1992;58:214–15. [7] Ochi H, Douglass H, Sandberg AA. Cytogenetic studies in primary gastric cancer. Cancer Genet Cytogenet 1986;22:295–307. [8] Rodriguez E, Rao PH, Ladanui M, Altorki N, Albino AP, Kelsen DP, Jhanwar SC, Chaganti RSK. 11p13–15 is a specific region of chromosomal rearrangement in gastric and esophageal adenocarcinomas. Cancer Res 1990;50:6410–16. [9] Misawa S, Horiike S, Taniwaki M, Tsuda S, Okuda T, Kashima K, Abe T, Sugihara H, Noriki S, Fukunda M. Chromosome abnormalities of gastric cancer detected in cancerous effusions. Jpn J Cancer Res 1990;81:148–52. [10] Seruca R, Castedo S, Correia C, Gomes P, Carneiro F, Soares P, Jong B, Sobrinho-Simoes M. Cytogenetic findings in eleven gastric carcinomas. Cancer Genet Cytogenet 1993;68:42–8. [11] Panani AD, Ferti A, Malliaros S, Raptis S. Cytogenetic study of 11 gastric adenocarcinomas. Cancer Genet Cytogenet 1995;81:169–72. [12] Kokkola A, Monni O, Puolakkainen P, Larramendy ML, Victorzon M, Nordling S, Haapiainen R, Kivilaakso E, Knuutila S. 17q12–21 am-
M.I. Stamouli et al. / Cancer Genetics and Cytogenetics 135 (2002) 23–27 plification, a novel recurrent genetic change in intestinal type of gastric carcinoma. Genes Chromosomes Cancer 1997;20:38–43. [13] Sakakura C, Mori T, Sakabe T, Ariyama Y, Shinomiya T, Date K, Hagiwara A, Yamaguchi T, Takayashi T, Nakamura Y, Abe T, Inazawa J. Gains, losses and amplifications of genomic materials in primary gastric cancers analyzed by comparative genomic hybridization. Genes Chromosomes Cancer 1999;24:299–305. [14] Koo SH, Kwon KC, Shin SY, Jeon YM, Park JW, Kim SH, Noh SM. Genetic alterations in gastric cancer: comparative genomic hybridization and fluorescence in situ hybridization studies. Cancer Genet Cytogenet 2000;117:97–103. [15] Nessling M, Solinas-Toldo S, Wilgenbus KK, Borchard F, Lichter P. Mapping of chromosomal imbalances in gastric adenocarcinoma revealed amplified protooncogenes MYCN, MET, WNT2 and ERBB2. Genes Chromosomes Cancer 1998;23:307–16.
27
[16] Speicher MR, Gwyn Ballard S, Ward D. Karyotyping human chromosomes by combinatorial multi-fluor FISH. Nat Genet 1996;12:368–75. [17] McKie AB, Filipe MI, Lemoine NR. Abnormalities affecting the APC and MCC tumour suppressor gene loci on chromosome 5q occur frequently in gastric cancer but not in pancreatic cancer. Cancer 1993;55:598–603. [18] Rhyu MG, Park WS, Jung YJ, Choi SW, Meltzer SJ. Allelic deletions of MCC/APC and p53 are frequent late events in human gastric carcinogenesis. Gastroenterology 1994;106:1584–8. [19] Castedo S, Correia C, David L, Sobrinho-Simoes M. Isochromosome 8q. A recurrent change in gastric carcinoma. Cancer Genet Cytogenet 1991;54:137–8. [20] Baffa R, Santoro R, Bullrich F, Mandes B, Ishii H, Croce CM. Definition and refinement of chromosome 8p regions of loss of heterozygosity in gastric cancer. Clin Cancer Res 2000;6:1372–7.