- Email: [email protected]
Flow cytometric DNA analysis and chromosomal aberrations in malignant glioblastomas
Cancer Letters 138 (1999) 101±106
Flow cytometric DNA analysis and chromosomal aberrations in malignant glioblastomas Volker Ehemann a,*, Birgit Hashemi b, Adelheid Lange a, Herwart F. Otto a a
Institute of Pathology, University of Heidelberg, Im Neuenheimer Feld 220-22, 69120Heidelberg, Germany b Department of Neurosurgery, University of Heidelberg, Heidelberg, Germany Received 8 June 1998; accepted 4 December 1998
Abstract In this study we combined ¯ow cytometry with ¯uorescence in situ hybridization to detect numerical aberrations in chromosomes. Fifty-nine human malignant gliomas were examined by ¯ow cytometry for DNA-content and cell cycle analysis and for numerical aberrations of chromosome 1 by in situ hybridization using a chromosome speci®c centromere probe. Of the gliomas analysed, 42% were diploid and 58% showed aneuploid tumour cell populations. The DNA index was heterogeneous ranging from 1.0 to 2.3. The S-phase analysis showed proliferation activity from a very low range of 0.7% up to 17.0%. In general, diploid gliomas exhibited a lower S-phase activity than aneuploid gliomas. Of the aneuploid gliomas, 15% showed a peridiploid pattern with a DNA index mean of 1.1. In these peridiploid tumours a trisomy of chromosome 1 could be detected by ¯uorescence in situ hybridization (FISH). The frequency of trisomic chromosome 1 in malignant gliomas re¯ects a very slight increase in DNA index from diploid to peridiploid (DNA index 1.1). Comparison of chromosome numbers and DNA content gave good correlation. Also important, the results re¯ects the cell cycle, speci®cally the extent of S-phase activity. In general, cell proliferation of diploid and peridiploid gliomas is much less than in higher aneuploid gliomas. The analysis of DNA content may thus yield results with respect to the biological behaviour of tumours in general. q 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Glioblastoma; Cytogenetics; Flow cytometry; DNA index; Cell cycle; FISH
1. Introduction It is well established that human malignant gliomas, in common with other solid tumours, may have an abnormal cellular DNA content which can be expressed as DNA aneuploidy. Tumour heterogeneity may be de®ned as a variation in the genotype and/or phenotype within a tumour [1±4,7,17]. DNA index and cell cycle analysis give information about the grade of ploidy and the growth rate characteristics * Corresponding author. Tel.: 1 49-6221-562626; fax: 1 496221-565251.
of tumours [4,7,12]. Fluorescence in situ hybridization (FISH) using chromosome-speci®c probes [5] allows the detection of numerical chromosome aberrations in the interphase nucleus. We studied malignant gliomas using ¯ow cytometry in native tissue and primary cell cultures. The results from ¯ow cytometry were related to cytogenetic characteristics of the tumours [4,9,10]. Peridiploid tumours indicate only a slight modi®cation in the DNA index. The increase of a DNA index to 1.1 represents a change in cellular DNA content of 2±3%. The trisomic chromosome 1 re¯ects the increase of the whole DNA content in this range. Therefore, high resolution ¯ow cytometry is
0304-3835/99/$ - see front matter q 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(98)00383-8
102
V. Ehemann et al. / Cancer Letters 138 (1999) 101±106
important in detecting such small changes in DNA [8,14] and the implicated genomic alterations. Nuclear DNA content roughly corresponds to chromosomal numbers [4,16]. Of the gliomas analysed, 42% were diploid and 58% showed aneuploid tumour cell populations. The DNA index was heterogeneous ranging from 1.0 to 2.3. The S-phase analysis showed proliferation activity from a very low range of 0.7% up to 17.0%. In general, diploid gliomas exhibited a lower S-phase activity than aneuploid gliomas. Of the aneuploid gliomas, 15% showed a peridiploid pattern with a DNA index mean of 1.1. Remarkably, in these peridiploid tumours a trisomy of chromosome 1 could be detected by FISH [7]. We conclude that in malignant gliomas there is a close correlation between DNA index, proliferative activity and cytogenetic tumour characteristics. Therefore, DNA ¯ow cytometry is an additional objective parameter in the clinical evaluation of gliomas and other malignancies.
2. Materials and methods 2.1. Tissue preparation and cell culture Fifty-nine native and sterile surgical specimens of sampled gliomas were subdivided from the same area for cell culture as well as for ¯ow cytometric analysis. Primary human glioma cells were routinely grown in RPMI-1640 medium (Eurobio, Germany) supplemented with 10% heat inactivated foetal calf serum (Eurobio, Germany), 5 mM l-glutamine, 50 IU/ml penicillin, and 25 mg/ml streptomycin (Eurobio, Germany). The cells were cultivated at 378C in a humidi®ed atmosphere of 95% air and 5% CO2. Experiments were performed with cells in exponential growth, hypotonic KCl (75 mM) for 15 min at 378C, trypsinated and ®xed in acetic acid/methanol (1:3, v/ v) after treatment with colcemide. Chromosomal spreads on glass slides were performed and rested for a minimum of 8 days before the hybridization with the pUC 1.77 DNA probe (1.77 kb) cloned in a pUC 9 vector (kindly provided by Prof. Dr. T. Cremer, Institute of Human Genetics and Anthropology, University of Heidelberg) major binding site; the
centromeric region of human chromosome 1 was started 2.2. In situ hybridization on slides The procedure for FISH was a modi®ed procedure originally described by Pinkel et al. and Cremer et al. [5,15]. In brief, cells were ®xed with methanol acetic acid, spotted on clean microscope slides and placed in ethanol for 1 h. The DNA probe, pUC 1.77, was labelled by a random primer method with a biotinylated-dUTP (ENZO-Kit). A hybridization mixture consisting of 1 mg/ml sonicated herring sperm DNA, 10% dextran sulfate, and 2 £ SSC (1 £ SSC is 0.15 M NaCl; 0.015 M sodium citrate) in 50% formamide was used. A 100-ng quantity of the biotinlabelled chromosome-speci®c probe, pUC 1.77, and the target DNA were denatured at 728C for 12 min and allowed to hybridize at 378C for 16 h. Hybridization was visualized with avidin±FITC (ENZO-KIT) counterstaining the nuclei with propidium iodide (Sigma Unterschleiûheim, Germany) at 0.2 mg/ml. An Ortholux ¯uorescence microscope (Leitz Stuttgart/ Germany), equipped with a 100 W mercury arc lamp and a 63£ oil immersion objective lens, was used to screen the slides. Photoprints were made using a 400 ASA Kodak dia®lm. 2.3. Flow cytometric analyses Flow cytometric analyses were performed using a PAS II ¯ow cytometer (Partec, MuÈnster/Germany) equipped with a mercury vapour lamp 100 W and a ®lter combination for 2,4-diamidino-2-phenylindole (DAPI) stained single cells. After harvesting, cells with trypsin/EDTA from primary cell cultures, as well as native sampled tissues, were treated with 2.1% citric acid/0.5% Tween 20 according to the method of Otto [14] with slight modi®cations. A phosphate buffer (7.2 g Na2HPO4 £ 2H2O in 100 ml distilled H2O) pH 8.0 containing DAPI for staining the cell suspension was used. Each histogram, depicting the DNA index and cell cycle, represents 30 000±90 000 cells. We used the Multicycle program (Phoenix Flow Systems, San Diego, CA) for histogram analysis. Human lymphocyte nuclei from healthy donors were used as an internal standard for determination of the diploid DNA index. The mean CV of diploid lymphocytes was 0.8±1.0.
V. Ehemann et al. / Cancer Letters 138 (1999) 101±106
Fig. 1. Calculated DNA indices and standard deviations for diploid, peridiploid and aneuploid cell populations in malignant gliomas.
3. Results We analysed 59 malignant human gliomas in native tissue as well as in primary cell cultures. Flow cytometric analyses showed 25 (42%) diploid gliomas and 34 (58%) aneuploid stemlines in malignant gliomas. The DNA index was heterogeneous ranging from 1.0 up to 2.3. Five of the aneuploid gliomas had a promi-
103
nent peridiploid stemline with a DNA index mean of 1.1. The general DNA index range in all aneuploid gliomas was 1.9 (Fig. 1). Only by careful preparation and measurement for cellular DNA is it possible to analyse smooth DNA-index variations in a peridiploid range of 1.04 to 1.1 with a coef®cient of variance (CV) of 1.3 to 1.9. DNA indices of the S-phase fraction for diploid gliomas ranged from 1.5 to 2.8% in peridiploid gliomas compared to a signi®cantly higher level of 5.9% in aneuploid gliomas (Fig. 2). In diploid and peridiploid gliomas the average Sphase fraction shows very distinct variations of 0.8 and 0.5%, respectively. In aneuploid gliomas the average was 3.8%. The ¯ow cytometric analyses of ®ve peridiploid gliomas (DNA index 1.1) shows in all cases a trisomy of chromosome 1, detected with a biotinylated centromere speci®c DNA probe for chromosome 1 (Fig. 3). These probes give a single signal for each centromere. The number of ¯uorescent hybridization signals within each nucleus was counted and, by convention, is referred to as the chromosome number. The number of trisomic interphase cells with three FISH dots ranged from 55 up to 80%, depending on whether the cell fractions were diploid (red marked fraction) or peridiploid (blue marked fraction) in each probe (Fig. 4). This was determined by counting the numbers of trisomic positive interphase cells with three dots in contrast to cells with the normal two dots. Two hundred cells were counted per slide. In the same way we also analysed ®ve diploid (DNA index 1.0) and ten aneuploid gliomas (DNA index . 1.3). We could not detect any trisomic chromosome 1 cells in the diploid cells; however, in three of the ten higher aneuploid gliomas a trisomic chromosome 1 was found. 4. Discussion
Fig. 2. Calculated S-phases and standard deviations for diploid, peridiploid, and aneuploid cell populations in malignant gliomas.
The aim of this study was to understand the biological effects of speci®c genetic events that occur within tumour subpopulations. Flow cytometry and in situ hybridization are powerful methods to study both numerical chromosome aberrations [6], as well as tumour cell heterogeneity [4] in malignancies. The combination of using ¯ow cytometry to detect very distinct DNA indices
104
V. Ehemann et al. / Cancer Letters 138 (1999) 101±106
Fig. 3. Fluorescence in situ hybridization on isolated glioma cells, determined as peridiploid with a biotinylated repetitive probe for centromere region chromosome 1. Hybridized probes are visualized with FITC-conjugated avidin, counterstaining with propidium iodide (PI). Metaphase and interphase cells with trisomy of chromosome 1.
and the in situ hybridization to de®ne and quantify the source of these sometimes small changes in the DNA content is the most important aspect of this work [7]. In situ hybridization in combination with ¯ow cytometry analyses was described in only a few cases of breast tumour cells and human gliomas [2,4]. The increase of a large chromosome, for example to a trisomy of chromosome 1, changes the whole cellular DNA content by only 2±3%. Only by high resolution ¯ow cytometry, careful preparation of the tissue and/or cells with appropriate staining, and the modality of measurement is it possible to observe such slight alterations in the DNA content, ranging from 1.04 to 1.1 [2,8]. In this paper we have shown that such slight alterations of the DNA index are most likely related to chromosomal aberrations. Interestingly, in 5 poorly diploid (DNA index 1.0) and in 7 out of 10 higher aneuploid gliomas (DNA index . 1.3) a trisomy of chromosome 1 was not detected. Of course other chromosomes or combinations of chromosomal aberrations could be present [4,16,17]. A trisomy in malignant gliomas of chromo-
some 7 and a monosomy of chromosome 10 is described by Hecht et al. [9]. Most aberrations in gliomas are found in chromosome 1 as deletions from the chromosome arm 1p [3]. Shapiro and Shapiro [17] found a loss of chromosome 1 to be most frequent in cultured glioma cell lines. Such numerical aberrations were also found and described for other cancers [1,2,10]. The DNA index is an important factor for measuring chromosomal aberrations [11±13,15,16]. The S-phase fraction of gliomas also exhibited signi®cant properties. Diploid gliomas showed a moderate S-phase (2.8%) fraction, in contrast to peridiploid gliomas which had a 1.5% S-phase fraction and presented a lower growth fraction. A signi®cantly higher S-phase of 5.9% was detected in higher aneuploid gliomas. Cell proliferation in peridiploid gliomas is reduced compared to higher aneuploid gliomas [12]. The occurrence of peridiploid cells in malignant gliomas which present a trisome chromosome 1 is, in our opinion, the most interesting result of the present study. In summary it could be demonstrated that ¯ow cytometry repre-
V. Ehemann et al. / Cancer Letters 138 (1999) 101±106
105
Fig. 4. (A±C) DNA histograms from human gliomas with peridiploid tumour cell populations. Red presents the diploid cell population, blue presents the peridiploid (aneuploid) cell population. A, DNA index 1.05. B, DNA index 1.07. C, DNA index 1.04.
sents an important supplement to diagnostic techniques, which in combination with FISH gives more detailed results regarding tumour biology and tumour characteristics.
References [1] N.B. Atkin, Chromosome 1 aberrations in cancer, Cancer Genet. Cytogenet. 21 (1985) 279±285. [2] M. Balazs, K. Matsumura, D. Moore, D. Pinkel, J.W. Gray,
106
[3]
[4] [5]
[6]
[7]
[8] [9]
V. Ehemann et al. / Cancer Letters 138 (1999) 101±106 F.M. Waldman, Karyotypic heterogeneity and its relation to labeling index in interphase breast tumor cells, Cytometry 20 (1995) 62±73. M.J. Bello, P.E. Leone, P. Nebreda, J.M. de Campos, M.E. Kusak, J. Vaquero, J.L. Sarasa, P. Garcia-Miguel, A. Queizan, J.L. Hernandes-Moneo, Allelic status of chromosome 1 in neoplasms of the nervous system, Cancer Genet. Cytogenet. 83 (1995) 160±164. S.W. Coons, P.C. Johnson, J.R. Shapiro, Cytogenetic and ¯ow cytometric DNA analysis of regional heterogeneity in a low grade human glioma, Cancer Res. 55 (1995) 1569±1577. T. Cremer, P. Lichter, J. Borden, D. Ward, L. Manuelidis, Detection of chromosome aberrations in metaphase and interphase tumor cells by in situ hybridization using chromosome speci®c library probes, Hum. Genet. 80 (1988) 235± 246. P. Deville, R.F. Thierry, T. Kievits, R. Kolluri, A.H. Hopman, H.F. Willard, P.L. Pearson, C.F. Cornelisse, Detection of chromosome aneuploidy in interphase nuclei from human primary breast tumors using chromosome speci®c repetitive DNA probes, Cancer Res. 51 (1991) 1020±1025. V. Ehemann, B. Eifert, K. MuÈnkel, A. Lange, H.F. Otto, Wertigkeit der Flow Cytometrie bei maligner Gliomen, 9. Heidelberger Zytometrie Symposium Abstract Band 1996 p.18, 1996. W. GoÈhde, M. Us-Krasovec, A. Pogacnik, Die Bedeutung der Messgenauigkeit bei der DNS-Zytophotometrie, Verh. Dtsch. Ges. Zyt. 19 (1995) 156±170. B.K. Hecht, C. Turc Carel, M. Chatel, P. Grellier, J. Gioanni, R. Attias, P. Gaudra, F. Hecht, Cytogenetics in malignant
[10]
[11]
[12]
[13]
[14] [15] [16]
[17]
gliomas: The autosomes with reference to rearrangements, Cancer Genet. Cytogenet. 84 (1995) 1±8. A.H.N. Hopman, O. Moesker, A.W.G.B. Smeets, R.P.E. Pauwels, G.P. Vooijs, F.C.S. Ramaekers, Numerical chromosome 1, 7, 9, and 11 aberrations in bladder cancer detected by in situ hybridization, Cancer Res. 51 (1991) 644±651. O. Jimenez, A. Timms, P. Quirke, J.E. McLaughlin, Prognosis in malignant glioma: A retrospective study of biopsy specimens by ¯ow cytometry, Neuropathol. Appl. Neurobiol. 15 (1989) 331±338. P. Mathew, T. Look, X. Luo, R. Ashmun, M. Nash, A. Gajjar, A. Walter, L. Kun, R.L. Heideman, DNA index of glial tumors in children: Correlation with tumor grade and prognosis, Cancer 78(4) (1996) 881±886. P.M. Nederlof, A.K. van der Raap, H.J. Tanke, M. van der Ploeg, F. Kornips, J.P.M. Geraedts, Detection of chromosome aberrations in interphase tumor nuclei by nonradioactive in situ hybridization, Cancer Genet. Cytogenet. 42 (1989) 87±98. F.J. Otto, High resolution analysis of nuclear DNA employing the ¯uorochrome DAP, Methods Cell Biol. 41 (1994) 211± 217. D. Pinkel, T. Straume, J.W. Gray, Cytogenetic analysis using quantitative, high-sensitivity ¯uorescence hybridization, Proc. Natl. Acad. Sci. USA 83 (1986) 2934±2938. J. Schlegel, G. Stumm, H.D. Mennel, J. RuÈschoff, Chromosome numbers and DNA-content in intracerebrally transplanted experimental gliomas, Exp. Pathol. 41 (1991) 135± 145. J.R. Shapiro, W.R. Shapiro, Clonal tumor cell heterogeneity, Prog. Exp. Tumor Res. 27 (1984) 49±66.
Flow cytometric DNA analysis and chromosomal aberrations in malignant glioblastomas Volker Ehemann a,*, Birgit Hashemi b, Adelheid Lange a, Herwart F. Otto a a
Institute of Pathology, University of Heidelberg, Im Neuenheimer Feld 220-22, 69120Heidelberg, Germany b Department of Neurosurgery, University of Heidelberg, Heidelberg, Germany Received 8 June 1998; accepted 4 December 1998
Abstract In this study we combined ¯ow cytometry with ¯uorescence in situ hybridization to detect numerical aberrations in chromosomes. Fifty-nine human malignant gliomas were examined by ¯ow cytometry for DNA-content and cell cycle analysis and for numerical aberrations of chromosome 1 by in situ hybridization using a chromosome speci®c centromere probe. Of the gliomas analysed, 42% were diploid and 58% showed aneuploid tumour cell populations. The DNA index was heterogeneous ranging from 1.0 to 2.3. The S-phase analysis showed proliferation activity from a very low range of 0.7% up to 17.0%. In general, diploid gliomas exhibited a lower S-phase activity than aneuploid gliomas. Of the aneuploid gliomas, 15% showed a peridiploid pattern with a DNA index mean of 1.1. In these peridiploid tumours a trisomy of chromosome 1 could be detected by ¯uorescence in situ hybridization (FISH). The frequency of trisomic chromosome 1 in malignant gliomas re¯ects a very slight increase in DNA index from diploid to peridiploid (DNA index 1.1). Comparison of chromosome numbers and DNA content gave good correlation. Also important, the results re¯ects the cell cycle, speci®cally the extent of S-phase activity. In general, cell proliferation of diploid and peridiploid gliomas is much less than in higher aneuploid gliomas. The analysis of DNA content may thus yield results with respect to the biological behaviour of tumours in general. q 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Glioblastoma; Cytogenetics; Flow cytometry; DNA index; Cell cycle; FISH
1. Introduction It is well established that human malignant gliomas, in common with other solid tumours, may have an abnormal cellular DNA content which can be expressed as DNA aneuploidy. Tumour heterogeneity may be de®ned as a variation in the genotype and/or phenotype within a tumour [1±4,7,17]. DNA index and cell cycle analysis give information about the grade of ploidy and the growth rate characteristics * Corresponding author. Tel.: 1 49-6221-562626; fax: 1 496221-565251.
of tumours [4,7,12]. Fluorescence in situ hybridization (FISH) using chromosome-speci®c probes [5] allows the detection of numerical chromosome aberrations in the interphase nucleus. We studied malignant gliomas using ¯ow cytometry in native tissue and primary cell cultures. The results from ¯ow cytometry were related to cytogenetic characteristics of the tumours [4,9,10]. Peridiploid tumours indicate only a slight modi®cation in the DNA index. The increase of a DNA index to 1.1 represents a change in cellular DNA content of 2±3%. The trisomic chromosome 1 re¯ects the increase of the whole DNA content in this range. Therefore, high resolution ¯ow cytometry is
0304-3835/99/$ - see front matter q 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(98)00383-8
102
V. Ehemann et al. / Cancer Letters 138 (1999) 101±106
important in detecting such small changes in DNA [8,14] and the implicated genomic alterations. Nuclear DNA content roughly corresponds to chromosomal numbers [4,16]. Of the gliomas analysed, 42% were diploid and 58% showed aneuploid tumour cell populations. The DNA index was heterogeneous ranging from 1.0 to 2.3. The S-phase analysis showed proliferation activity from a very low range of 0.7% up to 17.0%. In general, diploid gliomas exhibited a lower S-phase activity than aneuploid gliomas. Of the aneuploid gliomas, 15% showed a peridiploid pattern with a DNA index mean of 1.1. Remarkably, in these peridiploid tumours a trisomy of chromosome 1 could be detected by FISH [7]. We conclude that in malignant gliomas there is a close correlation between DNA index, proliferative activity and cytogenetic tumour characteristics. Therefore, DNA ¯ow cytometry is an additional objective parameter in the clinical evaluation of gliomas and other malignancies.
2. Materials and methods 2.1. Tissue preparation and cell culture Fifty-nine native and sterile surgical specimens of sampled gliomas were subdivided from the same area for cell culture as well as for ¯ow cytometric analysis. Primary human glioma cells were routinely grown in RPMI-1640 medium (Eurobio, Germany) supplemented with 10% heat inactivated foetal calf serum (Eurobio, Germany), 5 mM l-glutamine, 50 IU/ml penicillin, and 25 mg/ml streptomycin (Eurobio, Germany). The cells were cultivated at 378C in a humidi®ed atmosphere of 95% air and 5% CO2. Experiments were performed with cells in exponential growth, hypotonic KCl (75 mM) for 15 min at 378C, trypsinated and ®xed in acetic acid/methanol (1:3, v/ v) after treatment with colcemide. Chromosomal spreads on glass slides were performed and rested for a minimum of 8 days before the hybridization with the pUC 1.77 DNA probe (1.77 kb) cloned in a pUC 9 vector (kindly provided by Prof. Dr. T. Cremer, Institute of Human Genetics and Anthropology, University of Heidelberg) major binding site; the
centromeric region of human chromosome 1 was started 2.2. In situ hybridization on slides The procedure for FISH was a modi®ed procedure originally described by Pinkel et al. and Cremer et al. [5,15]. In brief, cells were ®xed with methanol acetic acid, spotted on clean microscope slides and placed in ethanol for 1 h. The DNA probe, pUC 1.77, was labelled by a random primer method with a biotinylated-dUTP (ENZO-Kit). A hybridization mixture consisting of 1 mg/ml sonicated herring sperm DNA, 10% dextran sulfate, and 2 £ SSC (1 £ SSC is 0.15 M NaCl; 0.015 M sodium citrate) in 50% formamide was used. A 100-ng quantity of the biotinlabelled chromosome-speci®c probe, pUC 1.77, and the target DNA were denatured at 728C for 12 min and allowed to hybridize at 378C for 16 h. Hybridization was visualized with avidin±FITC (ENZO-KIT) counterstaining the nuclei with propidium iodide (Sigma Unterschleiûheim, Germany) at 0.2 mg/ml. An Ortholux ¯uorescence microscope (Leitz Stuttgart/ Germany), equipped with a 100 W mercury arc lamp and a 63£ oil immersion objective lens, was used to screen the slides. Photoprints were made using a 400 ASA Kodak dia®lm. 2.3. Flow cytometric analyses Flow cytometric analyses were performed using a PAS II ¯ow cytometer (Partec, MuÈnster/Germany) equipped with a mercury vapour lamp 100 W and a ®lter combination for 2,4-diamidino-2-phenylindole (DAPI) stained single cells. After harvesting, cells with trypsin/EDTA from primary cell cultures, as well as native sampled tissues, were treated with 2.1% citric acid/0.5% Tween 20 according to the method of Otto [14] with slight modi®cations. A phosphate buffer (7.2 g Na2HPO4 £ 2H2O in 100 ml distilled H2O) pH 8.0 containing DAPI for staining the cell suspension was used. Each histogram, depicting the DNA index and cell cycle, represents 30 000±90 000 cells. We used the Multicycle program (Phoenix Flow Systems, San Diego, CA) for histogram analysis. Human lymphocyte nuclei from healthy donors were used as an internal standard for determination of the diploid DNA index. The mean CV of diploid lymphocytes was 0.8±1.0.
V. Ehemann et al. / Cancer Letters 138 (1999) 101±106
Fig. 1. Calculated DNA indices and standard deviations for diploid, peridiploid and aneuploid cell populations in malignant gliomas.
3. Results We analysed 59 malignant human gliomas in native tissue as well as in primary cell cultures. Flow cytometric analyses showed 25 (42%) diploid gliomas and 34 (58%) aneuploid stemlines in malignant gliomas. The DNA index was heterogeneous ranging from 1.0 up to 2.3. Five of the aneuploid gliomas had a promi-
103
nent peridiploid stemline with a DNA index mean of 1.1. The general DNA index range in all aneuploid gliomas was 1.9 (Fig. 1). Only by careful preparation and measurement for cellular DNA is it possible to analyse smooth DNA-index variations in a peridiploid range of 1.04 to 1.1 with a coef®cient of variance (CV) of 1.3 to 1.9. DNA indices of the S-phase fraction for diploid gliomas ranged from 1.5 to 2.8% in peridiploid gliomas compared to a signi®cantly higher level of 5.9% in aneuploid gliomas (Fig. 2). In diploid and peridiploid gliomas the average Sphase fraction shows very distinct variations of 0.8 and 0.5%, respectively. In aneuploid gliomas the average was 3.8%. The ¯ow cytometric analyses of ®ve peridiploid gliomas (DNA index 1.1) shows in all cases a trisomy of chromosome 1, detected with a biotinylated centromere speci®c DNA probe for chromosome 1 (Fig. 3). These probes give a single signal for each centromere. The number of ¯uorescent hybridization signals within each nucleus was counted and, by convention, is referred to as the chromosome number. The number of trisomic interphase cells with three FISH dots ranged from 55 up to 80%, depending on whether the cell fractions were diploid (red marked fraction) or peridiploid (blue marked fraction) in each probe (Fig. 4). This was determined by counting the numbers of trisomic positive interphase cells with three dots in contrast to cells with the normal two dots. Two hundred cells were counted per slide. In the same way we also analysed ®ve diploid (DNA index 1.0) and ten aneuploid gliomas (DNA index . 1.3). We could not detect any trisomic chromosome 1 cells in the diploid cells; however, in three of the ten higher aneuploid gliomas a trisomic chromosome 1 was found. 4. Discussion
Fig. 2. Calculated S-phases and standard deviations for diploid, peridiploid, and aneuploid cell populations in malignant gliomas.
The aim of this study was to understand the biological effects of speci®c genetic events that occur within tumour subpopulations. Flow cytometry and in situ hybridization are powerful methods to study both numerical chromosome aberrations [6], as well as tumour cell heterogeneity [4] in malignancies. The combination of using ¯ow cytometry to detect very distinct DNA indices
104
V. Ehemann et al. / Cancer Letters 138 (1999) 101±106
Fig. 3. Fluorescence in situ hybridization on isolated glioma cells, determined as peridiploid with a biotinylated repetitive probe for centromere region chromosome 1. Hybridized probes are visualized with FITC-conjugated avidin, counterstaining with propidium iodide (PI). Metaphase and interphase cells with trisomy of chromosome 1.
and the in situ hybridization to de®ne and quantify the source of these sometimes small changes in the DNA content is the most important aspect of this work [7]. In situ hybridization in combination with ¯ow cytometry analyses was described in only a few cases of breast tumour cells and human gliomas [2,4]. The increase of a large chromosome, for example to a trisomy of chromosome 1, changes the whole cellular DNA content by only 2±3%. Only by high resolution ¯ow cytometry, careful preparation of the tissue and/or cells with appropriate staining, and the modality of measurement is it possible to observe such slight alterations in the DNA content, ranging from 1.04 to 1.1 [2,8]. In this paper we have shown that such slight alterations of the DNA index are most likely related to chromosomal aberrations. Interestingly, in 5 poorly diploid (DNA index 1.0) and in 7 out of 10 higher aneuploid gliomas (DNA index . 1.3) a trisomy of chromosome 1 was not detected. Of course other chromosomes or combinations of chromosomal aberrations could be present [4,16,17]. A trisomy in malignant gliomas of chromo-
some 7 and a monosomy of chromosome 10 is described by Hecht et al. [9]. Most aberrations in gliomas are found in chromosome 1 as deletions from the chromosome arm 1p [3]. Shapiro and Shapiro [17] found a loss of chromosome 1 to be most frequent in cultured glioma cell lines. Such numerical aberrations were also found and described for other cancers [1,2,10]. The DNA index is an important factor for measuring chromosomal aberrations [11±13,15,16]. The S-phase fraction of gliomas also exhibited signi®cant properties. Diploid gliomas showed a moderate S-phase (2.8%) fraction, in contrast to peridiploid gliomas which had a 1.5% S-phase fraction and presented a lower growth fraction. A signi®cantly higher S-phase of 5.9% was detected in higher aneuploid gliomas. Cell proliferation in peridiploid gliomas is reduced compared to higher aneuploid gliomas [12]. The occurrence of peridiploid cells in malignant gliomas which present a trisome chromosome 1 is, in our opinion, the most interesting result of the present study. In summary it could be demonstrated that ¯ow cytometry repre-
V. Ehemann et al. / Cancer Letters 138 (1999) 101±106
105
Fig. 4. (A±C) DNA histograms from human gliomas with peridiploid tumour cell populations. Red presents the diploid cell population, blue presents the peridiploid (aneuploid) cell population. A, DNA index 1.05. B, DNA index 1.07. C, DNA index 1.04.
sents an important supplement to diagnostic techniques, which in combination with FISH gives more detailed results regarding tumour biology and tumour characteristics.
References [1] N.B. Atkin, Chromosome 1 aberrations in cancer, Cancer Genet. Cytogenet. 21 (1985) 279±285. [2] M. Balazs, K. Matsumura, D. Moore, D. Pinkel, J.W. Gray,
106
[3]
[4] [5]
[6]
[7]
[8] [9]
V. Ehemann et al. / Cancer Letters 138 (1999) 101±106 F.M. Waldman, Karyotypic heterogeneity and its relation to labeling index in interphase breast tumor cells, Cytometry 20 (1995) 62±73. M.J. Bello, P.E. Leone, P. Nebreda, J.M. de Campos, M.E. Kusak, J. Vaquero, J.L. Sarasa, P. Garcia-Miguel, A. Queizan, J.L. Hernandes-Moneo, Allelic status of chromosome 1 in neoplasms of the nervous system, Cancer Genet. Cytogenet. 83 (1995) 160±164. S.W. Coons, P.C. Johnson, J.R. Shapiro, Cytogenetic and ¯ow cytometric DNA analysis of regional heterogeneity in a low grade human glioma, Cancer Res. 55 (1995) 1569±1577. T. Cremer, P. Lichter, J. Borden, D. Ward, L. Manuelidis, Detection of chromosome aberrations in metaphase and interphase tumor cells by in situ hybridization using chromosome speci®c library probes, Hum. Genet. 80 (1988) 235± 246. P. Deville, R.F. Thierry, T. Kievits, R. Kolluri, A.H. Hopman, H.F. Willard, P.L. Pearson, C.F. Cornelisse, Detection of chromosome aneuploidy in interphase nuclei from human primary breast tumors using chromosome speci®c repetitive DNA probes, Cancer Res. 51 (1991) 1020±1025. V. Ehemann, B. Eifert, K. MuÈnkel, A. Lange, H.F. Otto, Wertigkeit der Flow Cytometrie bei maligner Gliomen, 9. Heidelberger Zytometrie Symposium Abstract Band 1996 p.18, 1996. W. GoÈhde, M. Us-Krasovec, A. Pogacnik, Die Bedeutung der Messgenauigkeit bei der DNS-Zytophotometrie, Verh. Dtsch. Ges. Zyt. 19 (1995) 156±170. B.K. Hecht, C. Turc Carel, M. Chatel, P. Grellier, J. Gioanni, R. Attias, P. Gaudra, F. Hecht, Cytogenetics in malignant
[10]
[11]
[12]
[13]
[14] [15] [16]
[17]
gliomas: The autosomes with reference to rearrangements, Cancer Genet. Cytogenet. 84 (1995) 1±8. A.H.N. Hopman, O. Moesker, A.W.G.B. Smeets, R.P.E. Pauwels, G.P. Vooijs, F.C.S. Ramaekers, Numerical chromosome 1, 7, 9, and 11 aberrations in bladder cancer detected by in situ hybridization, Cancer Res. 51 (1991) 644±651. O. Jimenez, A. Timms, P. Quirke, J.E. McLaughlin, Prognosis in malignant glioma: A retrospective study of biopsy specimens by ¯ow cytometry, Neuropathol. Appl. Neurobiol. 15 (1989) 331±338. P. Mathew, T. Look, X. Luo, R. Ashmun, M. Nash, A. Gajjar, A. Walter, L. Kun, R.L. Heideman, DNA index of glial tumors in children: Correlation with tumor grade and prognosis, Cancer 78(4) (1996) 881±886. P.M. Nederlof, A.K. van der Raap, H.J. Tanke, M. van der Ploeg, F. Kornips, J.P.M. Geraedts, Detection of chromosome aberrations in interphase tumor nuclei by nonradioactive in situ hybridization, Cancer Genet. Cytogenet. 42 (1989) 87±98. F.J. Otto, High resolution analysis of nuclear DNA employing the ¯uorochrome DAP, Methods Cell Biol. 41 (1994) 211± 217. D. Pinkel, T. Straume, J.W. Gray, Cytogenetic analysis using quantitative, high-sensitivity ¯uorescence hybridization, Proc. Natl. Acad. Sci. USA 83 (1986) 2934±2938. J. Schlegel, G. Stumm, H.D. Mennel, J. RuÈschoff, Chromosome numbers and DNA-content in intracerebrally transplanted experimental gliomas, Exp. Pathol. 41 (1991) 135± 145. J.R. Shapiro, W.R. Shapiro, Clonal tumor cell heterogeneity, Prog. Exp. Tumor Res. 27 (1984) 49±66.