- Email: [email protected]
Analysis of MYC and Chromosome 8 Copy Number Changes in Gastrointestinal Cancers by Dual-Color Fluorescence In Situ Hybridization
Analysis of MYC and Chromosome 8 Copy Number Changes in Gastrointestinal Cancers by Dual-Color Fluorescence In Situ Hybridization Yasuo Takahashi, Kaori Shintaku, Yukimoto Ishii, Satoshi Asai, Koichi Ishikawa, and Masashi Fujii
ABSTRACT: We performed dual (two-color) fluorescence in situ hybridization (FISH) by using direct fluorescent labeling probes for C-MYC and chromosome 8 in six gastrointestinal (three stomach and three colon) cancers. There are several reports of increased C-MYC copy numbers in solid tumors. To date, however, genetic rearrangements including those of the C-MYC gene have not actually been detected by FISH. Metaphase FISH demonstrated the C-MYC gene on other chromosomes (i.e., in addition to chromosome 8) in one gastric cancer. Somatic mutations such as chromosome translocation or insertion including the C-MYC gene are assumed to have occurred in this case. Our results suggest that genetic rearrangements, in addition to an increased C-MYC copy number, may be a mechanism of MYC oncogene activation in solid tumors. © Elsevier Science Inc., 1998
INTRODUCTION Fluorescence in situ hybridization (FISH) with the use of probes targeted to specific genes, which has recently gained popularity, allows the detection of genetic rearrangements and increases in copy number, participating in oncogene activation, at the individual cell level [1, 2]. In most FISH studies of solid tumors, changes in the copy number of specific genes or chromosomes were evaluated in interphase cell nuclei [3, 4]. However, genetic rearrangements cannot be demonstrated by interphase FISH. There have been reports of increased C-MYC copy numbers in various solid tumors [5, 6]. To date, however, genetic rearrangements including those involving the C-MYC gene have not actually been detected by FISH. Herein, we performed dual (two-color) FISH by using direct fluorescent labeling probes for C-MYC and the chromosome 8 centromere to study DNA rearrangements and copy numbers of MYC and chromosome 8 in gastrointestinal (three stomach and three colon) cancers. Metaphase FISH demonstrated the C-MYC gene on other chromo-
From the Department of Pharmacology (Y. T., S. A., K. I.) and the Third Department of Surgery (Y. T., Y. I., M. F.), Nihon University School of Medicine, Tokyo, Japan; and the Research and Development Center, Fujisawa Pharmaceutical Company (K. S.), Osaka, Japan. Address reprint requests to: Dr. Yasuo Takahashi, Department of Pharmacology, Nihon University School of Medicine, 30 Oyaguchi-Kami Machi, Itabashi, Tokyo 173, Japan. Received December 2, 1997; accepted April 22, 1998. Cancer Genet Cytogenet 107:61–64 (1998) Elsevier Science Inc., 1998 655 Avenue of the Americas, New York, NY 10010
somes; that is, in addition to chromosome 8. Somatic mutations such as chromosome translocation or insertion including the C-MYC gene were thus assumed to have occurred in one gastric cancer. Our results suggest that genetic rearrangements, in addition to an increased C-MYC copy number, may be a mechanism of MYC oncogene activation in solid tumors. METHODS Sample Collection Tissue specimens for FISH were obtained from fresh surgically resected primary tumors of patients hospitalized at Nihon University School of Medicine, Tokyo. All samples were obtained before the administration of chemotherapy or radiation. A part of each specimen was used for routine histopathological examinations. Peripheral blood lymphocytes obtained from a healthy adult were used as a negative control. A Burkitt lymphoma cell line, Daudi, which has a reciprocal chromosomal translocation involving the IgH and C-MYC loci, was used as a positive control for C-MYC rearrangement [7]. Slide Preparation Tissue for cytogenetic analysis was minced into 1–2 mm3 fragments and then incubated at 37⬚C for 30 minutes in RPMI-1640 medium with 0.2% collagenase and 10% fetal bovine serum. A single-cell suspension was made by passing cells through a 100-mesh sieve. Cells were centrifuged and washed in phosphate buffered saline, then resus-
0165-4608/98/$19.00 PII S0165-4608(98)00089-2
62
Y. Takahashi et al.
pended in 0.075 M KCl at room temperature for 20 minutes, centrifuged and subsequently fixed in methanol:glacial acetic acid (3:1), and chilled at ⫺20⬚C. Slides were prepared according to standard techniques. Probes and Hybridization A directly labeled probe for the chromosome 8 centromere (Spectrum Green CEP-8; Vysis, Framingham, MA) and the directly labeled MYC probe (Spectrum Orange LSI c-myc; Vysis), as well as reagents necessary for hybridization, were purchased. Dual-color hybridizations were performed by using a supplemental probe kit (Vysis) in accord with the manufacturer’s instructions. The nuclei were counterstained with DAPI and then examined with a fluorescence microscope, a cooled CCD camera (SenSys 1400; Photometrics), and a computerized image-analyzing system (QUIPS XL; Vysis). Only signals in nonoverlapping, apparently intact, nuclei (n ⫽ 100) were counted. RESULTS In peripheral blood lymphocytes and gastric tissues obtained from normal areas of gastrectomy specimens (cases 1 and 2), serving as normal controls, 93–100% of nuclei had two signals each for the C-MYC and chromosome 8 probes (Table 1). In two of the three resected gastric tumors (cases 1 and 3), 60% or more of nuclei showed a C-MYC signal gain (see Table 1). Case 3 also had a chromosome 8 signal gain; the modal signal count for chromosome 8 was similar to that for C-MYC. The other, interestingly, had normal chromosome 8 signal counts despite
55% of nuclei showing five signals for C-MYC. Both tumors with MYC amplification had features typical of poorly differentiated adenocarcinoma histologically, whereas the one without MYC amplification (case 2) was a mucinous adenocarcinoma. In two of three malignant colon tumors (Table 1, cases 4 and 5), most nuclei showed a C-MYC signal gain. Case 4 also had a signal gain for chromosome 8, but the degree did not correspond to that for C-MYC. In case 5, chromosome 8 aneusomy was consistent with numbers of C-MYC signals and the proportion of nuclei involved. Both cases with MYC amplification (cases 4 and 5) had morphologically ulcerative advanced cancers, whereas case 6, without MYC amplification, had an early polyplimited cancer. To obtain more information about the tumor in case 1, in which 55% of nuclei showed five signals for C-MYC despite only two signals for chromosome 8, we employed metaphase FISH analysis. First, we tested the Daudi cell line as a positive control (Fig. 1). Metaphase FISH in case 1 detected C-MYC signals on five chromosomes, two which had chromosome 8 signals. The C-MYC signals were thus clearly detectable on other chromosomes, besides chromosome 8. DISCUSSION Oncogenes are activated by several mechanisms including copy number increases, DNA rearrangements, and point mutations. FISH, with the use of probes targeted to specific genes, allows the detection of genetic rearrangements and copy number increases triggering oncogene activation
Table 1 Distribution of FISH signals Percent of nuclei with Number of signals Case
Origin
Pathology
1
Stomach
Adenocarcinoma poorly diff. N
2
Stomach
Mucinous adenocarcinoma N
3
Stomach
4
Colon
Adenocarcinoma poorly diff Adenocarcinoma poorly diff. M
5
Colon
6
Colon
Control
PBL
Adenocarcinoma mod. diff. Adenocarcinoma well diff.
Probe
1
2
3
4
5
⬎6
MYC Chr 8 MYC Chr 8 MYC Chr 8 MYC Chr 8 MYC Chr 8 MYC Chr 8 MYC Chr 8 MYC Chr 8 MYC Chr 8 MYC Chr 8
0 1 1 0 0 0 6 3 3 3 4 2 0 0 3 3 4 2 0 0
17 95 99 100 98 97 93 96 34 35 48 52 12 16 17 17 93 96 97 98
2 3 0 0 2 1 1 1 59 62 4 38 24 51 9 13 2 1 1 0
14 1 0 0 0 1 0 0 3 0 22 2 50 27 55 52 1 1 2 2
55 0 0 0 0 1 0 0 0 0 14 2 8 4 11 10 0 0 0 0
11 0 0 0 0 0 0 0 1 0 8 4 6 2 5 5 0 0 0 0
Boldface numbers indicate modal signal count/nucleus. Abbreviations: PBL, peripheral blood lymphocytes; mod, moderately; diff., differentiated; N, normal tissue; M, liver metastasis; Chr 8, chromosome 8.
63
MYC and Chromosome 8 in Gastrointestinal Cancers
Figure 1 Photomicrographs of MYC and chromosome 8 hybridizations. Orange signals are hybridization sites for the C-MYC probe. Green signals are hybridization sites for the chromosome 8 centromeric probe. (a) Normal gastric mucosa from case 1 showing normal hybridization patterns, two signals each for chromosome 8 and C-MYC per nucleus. (b) Gastric carcinoma from case 1. There are five signals for MYC and two for chromosome 8 per nucleus. (c) Metaphase of a gastric carcinoma cell from case 1. (d) Metaphase cell from Daudi cell (positive control) showing MYC gene rearrangement.
[2]. However, FISH results are often unsatisfactory in solid tumors compared with blood or cell lines, because the abundant connective tissue in solid tumors hampers separation into single cells and produces excessive background debris, making the preparation of clear specimens difficult. Thus, most FISH studies of solid tumors have as-
sessed only changes in the copy numbers of specific genes and chromosomes in interphase cell nuclei [3, 4, 6]. However, because genetic rearrangements cannot be detected, FISH cannot be used to full advantage, employing only interphase analysis. We have achieved reliable hybridization, in both interphase and metaphase cell nuclei from
64 surgically resected solid tumors, by employing recently developed direct labeling probes targeted to DNA sequences with a cosmid size of tens of kilobases or more and a computerized image analyzer. We avoided changes in the cell population harboring the original mutation due to selective pressure by using primary samples rather than subcultures of cell lines. We are thus confident that the amplification and rearrangement of the C-MYC gene observed in this study were present in the original cancer cells, rather than being subsequently induced artifacts. Using this technique, we studied rearrangements and changes in the copy number of the C-MYC gene in stomach and colon cancers. Because chromosome aberrations involving chromosome 8 have been reported in various solid tumors, such as stomach and lung cancers [3, 4], numerical abnormalities involving chromosome 8, in which C-MYC is located, have been suggested as a mechanism accounting for the increased MYC copy number. We, therefore, performed dual (two-color) FISH by using direct fluorescent labeled probes for C-MYC and the chromosome 8 centromere and compared the copy numbers obtained. In peripheral blood lymphocytes and gastric tissues obtained from normal areas of gastrectomy specimens (cases 1 and 2; Table 1), 93– 100% of nuclei had two signals for the C-MYC and two for the chromosome 8 probe (Table 1). However, among the six gastrointestinal cancers, increased MYC copy numbers were observed in four—that is, two stomach and two colon cancers. Of these four cases, three (cases 3, 4, and 5; Table 1) showed increases in chromosome 8 signals. Four MYC gene copies would theoretically be present in nuclei in the process of DNA synthesis before mitosis. However, the number of centromeric signals is reportedly unchanged throughout the cell cycle [8]. Therefore, the increase in chromosome 8 centromeric signals suggests an abnormally increased chromosome 8 copy number in these cases and that MYC amplification was secondary to this copy number increase. Similarly, the possibility that most MYC gene duplication is secondary to polysomy of chromosome 8 has also been suggested in breast cancer [6], supporting our results. Only one stomach cancer case (case 1; Table 1) had a normal chromosome 8 copy number and an increased MYC copy number. We performed metaphase FISH to examine MYC rearrangement as well as C-MYC amplification as possible oncogenic mechanisms. Fortunately, hybridization of metaphase cells was successful in case 1. In malignant cells from this case, C-MYC signals were detected at five loci; that is, two on chromosome 8 and one each on three other chromosomes (Fig. 1c). In addition, these results were consistent with the interphase FISH results; that is, cells with nuclei with two signals for chromosome 8 and five for C-MYC constituted a major cell population (case 1; Table 1; Fig. 1b). The C-MYC gene on chromosome 8 has been suggested to move to other chro-
Y. Takahashi et al. mosomes either by insertion or by chromosome translocation. Both events have been reported in numerous hematopoietic malignancies, including leukemias and Burkitt lymphoma [7, 9, 10], but neither has been observed in solid tumors even by FISH analysis. The possibility that such an event occurred in case 1, thereby leading to MYC gene amplification as detected by interphase analysis in previous studies, cannot be ruled out. Therefore, more cases, particularly with nuclei in which the MYC and chromosome 8 copy numbers are discordant on interphase analysis, must be studied by metaphase FISH to confirm these observations. The chromosome to which C-MYC moved could not be determined in this karyotype study. Furthermore, it was not possible to distinguish between insertion and translocation, because we used a probe for the centromere of chromosome 8. We will evaluate these issues in our next study.
REFERENCES 1. Chen Z, Morgan R, Berger CS, Sandberg AA (1992): Application of fluorescence in situ hybridization in hematological disorders. Cancer Genet Cytogenet 63:62–69. 2. Pinkel D, Landegent J, Collins C, Fuscoe J, Segraves R, Lucas J, Gray J (1988): Fluorescence in situ hybridization with human chromosome-specific libraries: detection of trisomy 21 and translocations of chromosome 4. Proc Natl Acad Sci USA 85:9138–9142. 3. Atkin NB, Baker MC (1991): Numerical chromosome changes in 165 malignant tumors: evidence for a nonrandom distribution of normal chromosomes. Cancer Genet Cytogenet 52:113– 121. 4. Cajulis RS, Frias HD (1993): Detection of numerical chromosomal abnormalities in malignant cells in fine needle aspirates by fluorescence in situ hybridization of interphase cell nuclei with chromosome-specific probes. Acta Cytol 37:391– 396. 5. Field JK, Spandidos DA (1990): The role of ras and myc oncogenes in human solid tumours and their relevance in diagnosis and prognosis (review). Anticancer Res 10:1–22. 6. Visscher DW, Wallis T, Awussah S, Mohamed A, Crissman JD (1997): Evaluation of MYC and chromosome 8 copy number in breast carcinoma by interphase cytogenetics. Genes Chromosom Cancer 18:1–7. 7. Haluska FG, Tsujimoto Y, Croce CM (1987): The t(8;14) chromosome translocation of the Burkitt lymphoma cell line Daudi occurred during immunoglobulin gene rearrangement and involved the heavy chain diversity region. Proc Natl Acad Sci USA 84:6835–6839. 8. Hopman AH, Voorter CE, Ramaekers FC (1994): Detection of genomic changes in cancer by in situ hybridization. Mol Biol Rep 19:31–44. 9. Sandberg AA (1990): The Chromosomes in Human Cancer and Leukemia. 2, Elsevier Science Publishing Co., New York, pp. 738–739. 10. Erikson J, Finger L, Sun L, Ar RA, Nishikura K, Minowada J, Finan J, Emanuel BS, Nowell PC, Croce CM (1986): Deregulation of c-myc by translocation of the alpha-locus of the T-cell receptor in T-cell leukemias. Science 232: 884–886.
ABSTRACT: We performed dual (two-color) fluorescence in situ hybridization (FISH) by using direct fluorescent labeling probes for C-MYC and chromosome 8 in six gastrointestinal (three stomach and three colon) cancers. There are several reports of increased C-MYC copy numbers in solid tumors. To date, however, genetic rearrangements including those of the C-MYC gene have not actually been detected by FISH. Metaphase FISH demonstrated the C-MYC gene on other chromosomes (i.e., in addition to chromosome 8) in one gastric cancer. Somatic mutations such as chromosome translocation or insertion including the C-MYC gene are assumed to have occurred in this case. Our results suggest that genetic rearrangements, in addition to an increased C-MYC copy number, may be a mechanism of MYC oncogene activation in solid tumors. © Elsevier Science Inc., 1998
INTRODUCTION Fluorescence in situ hybridization (FISH) with the use of probes targeted to specific genes, which has recently gained popularity, allows the detection of genetic rearrangements and increases in copy number, participating in oncogene activation, at the individual cell level [1, 2]. In most FISH studies of solid tumors, changes in the copy number of specific genes or chromosomes were evaluated in interphase cell nuclei [3, 4]. However, genetic rearrangements cannot be demonstrated by interphase FISH. There have been reports of increased C-MYC copy numbers in various solid tumors [5, 6]. To date, however, genetic rearrangements including those involving the C-MYC gene have not actually been detected by FISH. Herein, we performed dual (two-color) FISH by using direct fluorescent labeling probes for C-MYC and the chromosome 8 centromere to study DNA rearrangements and copy numbers of MYC and chromosome 8 in gastrointestinal (three stomach and three colon) cancers. Metaphase FISH demonstrated the C-MYC gene on other chromo-
From the Department of Pharmacology (Y. T., S. A., K. I.) and the Third Department of Surgery (Y. T., Y. I., M. F.), Nihon University School of Medicine, Tokyo, Japan; and the Research and Development Center, Fujisawa Pharmaceutical Company (K. S.), Osaka, Japan. Address reprint requests to: Dr. Yasuo Takahashi, Department of Pharmacology, Nihon University School of Medicine, 30 Oyaguchi-Kami Machi, Itabashi, Tokyo 173, Japan. Received December 2, 1997; accepted April 22, 1998. Cancer Genet Cytogenet 107:61–64 (1998) Elsevier Science Inc., 1998 655 Avenue of the Americas, New York, NY 10010
somes; that is, in addition to chromosome 8. Somatic mutations such as chromosome translocation or insertion including the C-MYC gene were thus assumed to have occurred in one gastric cancer. Our results suggest that genetic rearrangements, in addition to an increased C-MYC copy number, may be a mechanism of MYC oncogene activation in solid tumors. METHODS Sample Collection Tissue specimens for FISH were obtained from fresh surgically resected primary tumors of patients hospitalized at Nihon University School of Medicine, Tokyo. All samples were obtained before the administration of chemotherapy or radiation. A part of each specimen was used for routine histopathological examinations. Peripheral blood lymphocytes obtained from a healthy adult were used as a negative control. A Burkitt lymphoma cell line, Daudi, which has a reciprocal chromosomal translocation involving the IgH and C-MYC loci, was used as a positive control for C-MYC rearrangement [7]. Slide Preparation Tissue for cytogenetic analysis was minced into 1–2 mm3 fragments and then incubated at 37⬚C for 30 minutes in RPMI-1640 medium with 0.2% collagenase and 10% fetal bovine serum. A single-cell suspension was made by passing cells through a 100-mesh sieve. Cells were centrifuged and washed in phosphate buffered saline, then resus-
0165-4608/98/$19.00 PII S0165-4608(98)00089-2
62
Y. Takahashi et al.
pended in 0.075 M KCl at room temperature for 20 minutes, centrifuged and subsequently fixed in methanol:glacial acetic acid (3:1), and chilled at ⫺20⬚C. Slides were prepared according to standard techniques. Probes and Hybridization A directly labeled probe for the chromosome 8 centromere (Spectrum Green CEP-8; Vysis, Framingham, MA) and the directly labeled MYC probe (Spectrum Orange LSI c-myc; Vysis), as well as reagents necessary for hybridization, were purchased. Dual-color hybridizations were performed by using a supplemental probe kit (Vysis) in accord with the manufacturer’s instructions. The nuclei were counterstained with DAPI and then examined with a fluorescence microscope, a cooled CCD camera (SenSys 1400; Photometrics), and a computerized image-analyzing system (QUIPS XL; Vysis). Only signals in nonoverlapping, apparently intact, nuclei (n ⫽ 100) were counted. RESULTS In peripheral blood lymphocytes and gastric tissues obtained from normal areas of gastrectomy specimens (cases 1 and 2), serving as normal controls, 93–100% of nuclei had two signals each for the C-MYC and chromosome 8 probes (Table 1). In two of the three resected gastric tumors (cases 1 and 3), 60% or more of nuclei showed a C-MYC signal gain (see Table 1). Case 3 also had a chromosome 8 signal gain; the modal signal count for chromosome 8 was similar to that for C-MYC. The other, interestingly, had normal chromosome 8 signal counts despite
55% of nuclei showing five signals for C-MYC. Both tumors with MYC amplification had features typical of poorly differentiated adenocarcinoma histologically, whereas the one without MYC amplification (case 2) was a mucinous adenocarcinoma. In two of three malignant colon tumors (Table 1, cases 4 and 5), most nuclei showed a C-MYC signal gain. Case 4 also had a signal gain for chromosome 8, but the degree did not correspond to that for C-MYC. In case 5, chromosome 8 aneusomy was consistent with numbers of C-MYC signals and the proportion of nuclei involved. Both cases with MYC amplification (cases 4 and 5) had morphologically ulcerative advanced cancers, whereas case 6, without MYC amplification, had an early polyplimited cancer. To obtain more information about the tumor in case 1, in which 55% of nuclei showed five signals for C-MYC despite only two signals for chromosome 8, we employed metaphase FISH analysis. First, we tested the Daudi cell line as a positive control (Fig. 1). Metaphase FISH in case 1 detected C-MYC signals on five chromosomes, two which had chromosome 8 signals. The C-MYC signals were thus clearly detectable on other chromosomes, besides chromosome 8. DISCUSSION Oncogenes are activated by several mechanisms including copy number increases, DNA rearrangements, and point mutations. FISH, with the use of probes targeted to specific genes, allows the detection of genetic rearrangements and copy number increases triggering oncogene activation
Table 1 Distribution of FISH signals Percent of nuclei with Number of signals Case
Origin
Pathology
1
Stomach
Adenocarcinoma poorly diff. N
2
Stomach
Mucinous adenocarcinoma N
3
Stomach
4
Colon
Adenocarcinoma poorly diff Adenocarcinoma poorly diff. M
5
Colon
6
Colon
Control
PBL
Adenocarcinoma mod. diff. Adenocarcinoma well diff.
Probe
1
2
3
4
5
⬎6
MYC Chr 8 MYC Chr 8 MYC Chr 8 MYC Chr 8 MYC Chr 8 MYC Chr 8 MYC Chr 8 MYC Chr 8 MYC Chr 8 MYC Chr 8
0 1 1 0 0 0 6 3 3 3 4 2 0 0 3 3 4 2 0 0
17 95 99 100 98 97 93 96 34 35 48 52 12 16 17 17 93 96 97 98
2 3 0 0 2 1 1 1 59 62 4 38 24 51 9 13 2 1 1 0
14 1 0 0 0 1 0 0 3 0 22 2 50 27 55 52 1 1 2 2
55 0 0 0 0 1 0 0 0 0 14 2 8 4 11 10 0 0 0 0
11 0 0 0 0 0 0 0 1 0 8 4 6 2 5 5 0 0 0 0
Boldface numbers indicate modal signal count/nucleus. Abbreviations: PBL, peripheral blood lymphocytes; mod, moderately; diff., differentiated; N, normal tissue; M, liver metastasis; Chr 8, chromosome 8.
63
MYC and Chromosome 8 in Gastrointestinal Cancers
Figure 1 Photomicrographs of MYC and chromosome 8 hybridizations. Orange signals are hybridization sites for the C-MYC probe. Green signals are hybridization sites for the chromosome 8 centromeric probe. (a) Normal gastric mucosa from case 1 showing normal hybridization patterns, two signals each for chromosome 8 and C-MYC per nucleus. (b) Gastric carcinoma from case 1. There are five signals for MYC and two for chromosome 8 per nucleus. (c) Metaphase of a gastric carcinoma cell from case 1. (d) Metaphase cell from Daudi cell (positive control) showing MYC gene rearrangement.
[2]. However, FISH results are often unsatisfactory in solid tumors compared with blood or cell lines, because the abundant connective tissue in solid tumors hampers separation into single cells and produces excessive background debris, making the preparation of clear specimens difficult. Thus, most FISH studies of solid tumors have as-
sessed only changes in the copy numbers of specific genes and chromosomes in interphase cell nuclei [3, 4, 6]. However, because genetic rearrangements cannot be detected, FISH cannot be used to full advantage, employing only interphase analysis. We have achieved reliable hybridization, in both interphase and metaphase cell nuclei from
64 surgically resected solid tumors, by employing recently developed direct labeling probes targeted to DNA sequences with a cosmid size of tens of kilobases or more and a computerized image analyzer. We avoided changes in the cell population harboring the original mutation due to selective pressure by using primary samples rather than subcultures of cell lines. We are thus confident that the amplification and rearrangement of the C-MYC gene observed in this study were present in the original cancer cells, rather than being subsequently induced artifacts. Using this technique, we studied rearrangements and changes in the copy number of the C-MYC gene in stomach and colon cancers. Because chromosome aberrations involving chromosome 8 have been reported in various solid tumors, such as stomach and lung cancers [3, 4], numerical abnormalities involving chromosome 8, in which C-MYC is located, have been suggested as a mechanism accounting for the increased MYC copy number. We, therefore, performed dual (two-color) FISH by using direct fluorescent labeled probes for C-MYC and the chromosome 8 centromere and compared the copy numbers obtained. In peripheral blood lymphocytes and gastric tissues obtained from normal areas of gastrectomy specimens (cases 1 and 2; Table 1), 93– 100% of nuclei had two signals for the C-MYC and two for the chromosome 8 probe (Table 1). However, among the six gastrointestinal cancers, increased MYC copy numbers were observed in four—that is, two stomach and two colon cancers. Of these four cases, three (cases 3, 4, and 5; Table 1) showed increases in chromosome 8 signals. Four MYC gene copies would theoretically be present in nuclei in the process of DNA synthesis before mitosis. However, the number of centromeric signals is reportedly unchanged throughout the cell cycle [8]. Therefore, the increase in chromosome 8 centromeric signals suggests an abnormally increased chromosome 8 copy number in these cases and that MYC amplification was secondary to this copy number increase. Similarly, the possibility that most MYC gene duplication is secondary to polysomy of chromosome 8 has also been suggested in breast cancer [6], supporting our results. Only one stomach cancer case (case 1; Table 1) had a normal chromosome 8 copy number and an increased MYC copy number. We performed metaphase FISH to examine MYC rearrangement as well as C-MYC amplification as possible oncogenic mechanisms. Fortunately, hybridization of metaphase cells was successful in case 1. In malignant cells from this case, C-MYC signals were detected at five loci; that is, two on chromosome 8 and one each on three other chromosomes (Fig. 1c). In addition, these results were consistent with the interphase FISH results; that is, cells with nuclei with two signals for chromosome 8 and five for C-MYC constituted a major cell population (case 1; Table 1; Fig. 1b). The C-MYC gene on chromosome 8 has been suggested to move to other chro-
Y. Takahashi et al. mosomes either by insertion or by chromosome translocation. Both events have been reported in numerous hematopoietic malignancies, including leukemias and Burkitt lymphoma [7, 9, 10], but neither has been observed in solid tumors even by FISH analysis. The possibility that such an event occurred in case 1, thereby leading to MYC gene amplification as detected by interphase analysis in previous studies, cannot be ruled out. Therefore, more cases, particularly with nuclei in which the MYC and chromosome 8 copy numbers are discordant on interphase analysis, must be studied by metaphase FISH to confirm these observations. The chromosome to which C-MYC moved could not be determined in this karyotype study. Furthermore, it was not possible to distinguish between insertion and translocation, because we used a probe for the centromere of chromosome 8. We will evaluate these issues in our next study.
REFERENCES 1. Chen Z, Morgan R, Berger CS, Sandberg AA (1992): Application of fluorescence in situ hybridization in hematological disorders. Cancer Genet Cytogenet 63:62–69. 2. Pinkel D, Landegent J, Collins C, Fuscoe J, Segraves R, Lucas J, Gray J (1988): Fluorescence in situ hybridization with human chromosome-specific libraries: detection of trisomy 21 and translocations of chromosome 4. Proc Natl Acad Sci USA 85:9138–9142. 3. Atkin NB, Baker MC (1991): Numerical chromosome changes in 165 malignant tumors: evidence for a nonrandom distribution of normal chromosomes. Cancer Genet Cytogenet 52:113– 121. 4. Cajulis RS, Frias HD (1993): Detection of numerical chromosomal abnormalities in malignant cells in fine needle aspirates by fluorescence in situ hybridization of interphase cell nuclei with chromosome-specific probes. Acta Cytol 37:391– 396. 5. Field JK, Spandidos DA (1990): The role of ras and myc oncogenes in human solid tumours and their relevance in diagnosis and prognosis (review). Anticancer Res 10:1–22. 6. Visscher DW, Wallis T, Awussah S, Mohamed A, Crissman JD (1997): Evaluation of MYC and chromosome 8 copy number in breast carcinoma by interphase cytogenetics. Genes Chromosom Cancer 18:1–7. 7. Haluska FG, Tsujimoto Y, Croce CM (1987): The t(8;14) chromosome translocation of the Burkitt lymphoma cell line Daudi occurred during immunoglobulin gene rearrangement and involved the heavy chain diversity region. Proc Natl Acad Sci USA 84:6835–6839. 8. Hopman AH, Voorter CE, Ramaekers FC (1994): Detection of genomic changes in cancer by in situ hybridization. Mol Biol Rep 19:31–44. 9. Sandberg AA (1990): The Chromosomes in Human Cancer and Leukemia. 2, Elsevier Science Publishing Co., New York, pp. 738–739. 10. Erikson J, Finger L, Sun L, Ar RA, Nishikura K, Minowada J, Finan J, Emanuel BS, Nowell PC, Croce CM (1986): Deregulation of c-myc by translocation of the alpha-locus of the T-cell receptor in T-cell leukemias. Science 232: 884–886.