- Email: [email protected]
Baseline and mitomycin-c-induced sister chromatid exchanges in a melanoma and a colon tumor cell line
Baseline and Mitomycin-C-induced Sister Chromatid Exchanges in a Melanoma and a Colon Tumor Cell Line Shende Li,* William W. Au, R. L. Schmoyer, Jr., and T. C. Hsu
Baseline and mitomycin C (MMC)-induced sister chromatid exchanges (SCEs) in two human tumor cell lines (a colon tumor and a melanoma) and in a normal fibroblast cell line were analyzed and compared. The tumor cells showed numerical and structural chromosomal abnormalities. Their baseline SCE rate was slightly, but not significantly, higher than that of the control. Each tumor cell line showed a dose-dependent increase in SCE frequency above the spontaneous level in its own specific manner. The response in the melanoma cell was consistently below that of the control, but only the response to the highest dose of MMC (10 _9 M) was significantly lower than that of the control. The response of the colon tumor cells varied with respect to that of the control. Thus, it appears that koryotypic instability in tumor cells is not necessarily associated with elevated baseline or induced SCE/chromosome rates. In addition, within each cell line dose group, the SCE frequency was proportional to the number of chromosomes. Thus, the SCE/chromosome is a better expression of genetic damage than SCE/metaphase in analyses involving heteraploid cells.
ABSTRACT:
INTRODUCTION In the field of genetic toxicology, SCE is a sensitive indicator of genetic damage i n d u c e d by various physical and chemical agents [1-3]. Whether this response to external insult is a quantitative indicator of the damage is still u n k n o w n [4]. The relationship of SCE to hereditarily determined genetic damage (genetic instability) has been extensively investigated. The p h e n o m e n o n of genetic instability is characteristic in "chromosome instability syndromes" [5]. A m o n g those syndromes, Bloom's syndrome is characterized by an increase in baseline chromosome breakage and SCE [6]. However, increase i n baseline chromosome breakage for two other types of patients (ataxia telangiectasia and Fanconi anemia) are not correlated with an increase in SCE (reviewed by Arlett and L e h m a n n [7]). Neoplastic cells offer another system for analysis of the relationship between SCE and chromosome instability. Such studies have been conducted by several inFrom the Department of Cell Biology, The University of Texas M.D. Anderson Hospital and Tumor Institute at Houston, Houston (S.L., T.C.H.), and the Biology Division, Oak Ridge National Laboratory (W.W.A)and the Nuclear Division,Union Carbide Corporation,Oak Ridge, Tennessee(R.L.S). *Dr. Li is on leave from Ritan Hospital and Cancer Institute, Chinese Academy of Medical Sciences, Beijing,People's Republicof China. Address requests for reprints to Dr. T.C. Hsu, Department of Cell Biology, The University of Texas M.D. Anderson Hospital and Tumor Institute, Houston TX 77030. Received August 14, 1981; accepted October 5, 1981.
243 Elsevier SciencePublishingCo., Inc., 1982 52 VanderbiltAve., New York, NY 1 0 0 1 7
Cancer Geneticsand Cytogenetics6, 243-246 (1982) 0165-4608/82/070243-0652.75
244
s. Li et al. vestigators (reviewed by Shirashi and Sandberg [8]), but most of these investigations were l i m i t e d to analyses involving leukemic cells. We report here a s t u d y comparing the baseline and M M C - i n d u c e d SCE in two tumors (a colon t u m o r and a melanoma) and a normal cell line.
MATERIALS AND METHODS
Three h u m a n cell lines were used in this study. Two were t u m o r cell lines: a colon t u m o r (SW 620) and a m e l a n o m a (SW 1687). They were established by A. Leibovitz at the Scott and White Clinic, Temple, Texas. Line SW 620 was established on November 20, 1973 from a metastatic lesion of a colon carcinoma. We used it at the 83rd passage onwards. Its history and growth characteristics have been reported [9]. The m e l a n o m a line was established on July 15, 1977 from a metastatic lesion in the thigh of the patient. We used it at the 20th passage onwards. The cells of the m e l a n o m a line continue to synthesize m e l a n i n in vitro. A normal d i p l o i d h u m a n fibroblast cell line (BAS) used as control was p r o v i d e d by Dr. Lee Kronenberg of Lee Biomolecular Research Laboratory, San Diego, California. The tumor cell lines were m a i n t a i n e d in Leibovitz L-15 m e d i u m s u p p l e m e n t e d with 10% fetal calf serum; whereas, the BAS line was m a i n t a i n e d in McCoy's 5a m e d i u m w i t h 10% fetal calf serum.
Figure 1 A differentially stained metaphase from the melanoma cell line treated with 10 -~ M MMC showing 72 chromosomes with 22 SCE.
7 I
245
Baseline a n d M i t o m y c i n - C - i n d u c e d SCEs
Confluent cultures were trypsinized and 5 × 10 ~ cells were seeded into T-25 flasks in 10 ml of m e d i u m . Duplicate flasks were treated with various concentrations of MMC (10 -~, 10 -1°, 10-11M) and b r o m o d e o x y u r i d i n e (BrdU) (20 ~g/ml). The BAS cultures were havested after a 40 hr treatment. However, w h e n tumor cell cultures were harvested at 48 and 72 hr, very few cells had proceeded through two cell cycles as indicated by the absence of differential staining of the sister chromatids. Therefore, after 72 hr the culture m e d i u m was replaced with one c o n t a i n i n g BrdU alone. Cultures were harvested 24 hr later. With this protocol, sufficient n u m b e r s of cells were i n mitosis and the majority of metaphases showed differential staining. A n example of a differentially stained metaphase from the m e l a n o m a cell line treated with 10 .9 M MMC is s h o w n in Figure 1. Cultures were harvested after 4 hr by colcemid (0.04 ~g/ml) treatment. Cells were trypsinized, treated with 0.7% sodium citrate, and fixed three times with Carnoys' fixative. Air-dried cytological preparations were treated with the FPG techniques of Perry and Wolff to show differential staining of sister chromatids [10]. For each cell line and each dose, 20 metaphases were scored for SCE frequency and for chromosome number. T h e n the entire procedure was repeated; thus, the study actually consisted of two i n d e p e n d e n t experiments. RESULTS Table 1 summarizes the results of the study. Analysis of variance of the two indep e n d e n t l y conducted experiments showed that experiments and all interactions involving experiments have a negligible effect (p = 0.48 for n u m b e r of chromosomes and p = 0.34 for SCE/chromosome). Within the limits of experimental error the results of the two experiments are identical, and we have c o m b i n e d the results of the two experiments in Table 1. Analysis of the distribution of chromosomes showed that all three cell lines were significantly different i n terms of chromosome n u m b e r (p -< 0.0001). This difference was considered w h e n comparing the dose respons~ of the different cell lines. As s h o w n in Table 1, the means of the baseline SCE/chromosome for the two tumor
Table 1
Chromosome distribution and MMC-induced sister chromatid exchanges in three cell lines
Number of chromosomes/cell Cell line
Dose (M)
BAS (control)
0 10 -'1 10 -l° 10 -9 0 10 -ll 10 -1° 10 -9 0 10 -11 10 -~° 10 -9
Colon tumor
Melanoma
Mode (min-max) 46 46 46 46 50 50 50 50 67 64 65 65
(41-46) (40-46) (41-46) (42-46) (34-52) (35-53) (38-55) (38-53) (50-75) (50-72) (49-72) (50--72)
SCE/chromosome
SCE/chromosome control mean
Mean
(SE)
Mean
(SE)
Mean
45.4 45.8 45.5 45.7 49.1 49.9 48.9 48.6 64.3 60.7 63.1 63.9
(0.2) (0.2) (0.2) (0.2) (0.5) (0.3) (0.5) (0.4) (1.0) (0.7) (0.9) (0.8)
0.077 0.089 0.132 0.202 0.093 0.111 0.111 0.263 0.094 0.103 0.146 0.166
(0.005) (0.004) (0.005) (0.008) (0.006) (0.008) (0.007) (0.018) (0.004) (0.005) (0.009) (0.007)
-1.14 1.71 2.60 -1.19 1.19 2.84 -1.10 1.55 1.75
(SE) (0.05) (0.07) (0.10) (0.09) (0.08) (0.19) (0.06) (0.10) (0.07)
246
s. Li et al.
cell lines were higher than that of the control. However, there was no significant difference among t h e m (p -- 0.28). The sensitivity of the e x p e r i m e n t was such that we could be 90% certain of detecting differences as small as 0.037 in SCE/chromosome for the different treatment groups. All cell types s h o w e d a d o s e - d e p e n d e n t increase in SCE/chromosome. The dose of MMC was the d o m i n a n t factor in determ i n i n g this SCE response (p -~ 0.0001), although there was also a significant d o s e cell line interaction (p -~ 0.0001). The SCE/chromosome of each dose was n o r m a l i z e d by its own untreated control and was plotted against the dose as s h o w n in Figure 2. A l t h o u g h the baseline SCE rate for the m e l a n o m a cells was higher than that of the BAS cells, the n o r m a l i z e d response of the treated m e l a n o m a cells was consistently b e l o w that of the control culture. The d o s e - c e l l line interaction is most clearly illustrated by the behavior of the colon tumor cells, w h i c h showed a threshold effect in response to the damage i n d u c e d by MMC at 10 11 and 10 -1° M. The n o r m a l i z e d SCE frequency of the colon tumor i n d u c e d by 10 -1° M MMC was significantly different from those of the other two cultures. The m e l a n o m a line was significantly different from the other two at 10 -9 M. (This conclusion is based on analysis by the Duncan m u l t i p l e range test.) With regard to the appropriateness of the end point, SCE/chromosome, we considered the following models: SCE = (constant d e p e n d i n g on dose and cell line) × (number of chromosomes) x (random error), SCE = (constant d e p e n d i n g on dose, cell line, and n u m b e r of chromosomes) x (random error).
(1) (2)
Notice that m o d e l (1) is a special case of m o d e l (2). R a n d o m error represents factors other than dose, cell line, or n u m b e r of chromosomes. (Log r a n d o m error was assumed to have normal distribution, and the two models were c o m p a r e d by the Figure 2 Mitomycin-C-induced changes in SCE/chromosome relative to background. Vertical bars represent 95% confidence interval for the true response at specified cell lines and dose. For simplicity, these have been included only for those cell lines differing significantly from the other two at the specified dose. 3.0 • Colon
Tumor
it
,,' .. B A S • • .° o°'°
E
°°~?
°.°°o°°., •• °o° i °°°°°°°°°°"j7It ~
¢-
0
02.0 E E 0
0 t,tO LU
*
/
°°00°
C O
°°°° t ~ ° ' ° ° " J
0 o3
Melanoma
¢
1.0
I 10 -11
L
10 -10 Mitomycin
10 . 9
C Concentration(M)
Baseline and M i t o m y c i n - C - i n d u c e d SCEs
247
method of analysis of variance.) We found that model (2) does not provide a significantly better fit to the data than model (1) (p = 0.96). The sensitivity of this analysis is such that differences on the order of 17% in the expected SCE frequencies under models (1) and (2) would be detected with 90% certainty. Thus, for a given cell line-dose group, within the limits of experimental error SCE is equal to a constant times the number of chromosomes; hence, SCE/chromosome is constant. This is certainly not true for SCE/metaphase. We think, therefore, that SCE/chromosome is a better indicator of cytogenetic damage than SCE/metaphase. DISCUSSION Studies on the frequency of SCE in tumor cells have been limited mostly to analyses of baseline SCE frequencies in myelo- and lymphoproliferative diseases [8]. In general, it was concluded that leukemic cells were not characterized by an inherently high incidence of SCE, regardless of the type of leukemia. Some of the exceptions could be accounted for by prior drug treatment. On the other hand, increased SCE frequency was usually observed in cells from patients receiving chemotherapy and continued for a short time after the cessation of chemotherapy. Similar analyses on cells derived from solid tumors were rarely performed. The colon tumor and the melanoma cell cultures used in this study are similar in terms of genetic instability to most heteroploid cell lines established from solid tumors. Both showed an abnormal distribution of chromosomes (Table 1). Chromosomal rearrangements have been observed in the colon tumor [9] and in the melanoma (S. Pathek, unpublished observation). In addition, we observed an elevated rate of baseline chromosome breakage in these two tumor cell lines when the rates were compared with that of the control (0.42 lesions/metaphase for the colon tumor, 0.18 for melanoma, and 0.04 for BAS culture). The control culture, BAS, was karyotypically normal (L. Kronenberg, unpublished observation). Our data showed that the baseline SCE frequency of the tumor cells was higher than that of the control cells although they were not significantly different from one another. However, significantly higher baseline SCE frequencies were observed by Chen in five heteroploid cell lines and sublines from melanomas [11]. Shirashi et al. observed significant increases in MMC-induced SCE in leukemia and lymphoma cell lines as compared with normal lymphocytes [12]. Conceivably, induction of genetic damage by chemicals can be influenced by a number of factors, e.g., penetration, metabolism of the chemical, and repair of induced damage. Indeed, variable responses were observed in this study. Although the cell lines showed dose-dependent increases in SCE, each had its own characteristic and reproducible response. The response of the melanoma cells was consistently lower than that of the control. The colon tumor cells showed a threshold effect not seen in the control, while at high doses of MMC the SCE response was as high as that of the control. Such lack of consistent elevation of baseline and induced SCE frequencies in the tumor cells indicated that karyotypic instability is not necessarily correlated with SCE/chromosome rates. The cell line-specific response cannot be explained simply by the long cell cycle time available for repair of damage in the tumor cells although the role of cell cycle in repair has been observed [13,14]. A better understanding of such a phenomenon might be achieved by using more dose level of MMC, by using other inducing agents, and by using other assaying techniques. W.W. Au's research is supported by the Office of Health and Environmental Research, U. S. Department of Energy, under contract W-7405-eng-26 with the Union Carbide Corporation.
248
S. Li et al. This is a joint research project of The University of Texas System Cancer Center M.D. Anderson Hospital and Tumor Institute and the John S. Dunn Research Foundation of Houston, Texas. We would like to thank Mr. Dindyal Ramkissoon for his expert technical assistance.
REFERENCES 1. Latt SA (1975): Sister chromatid exchanges, indices of human chromosome damage and repair: Detection by fluorescence and induction by mitomycin C. Proc Natl Acad Sci USA 71, 3162-3166. 2. Perry P, Evans HJ (1975): Cytological detection of mutagen-carcinogen exposure by sister chromatid exchange. Nature 258, 121-125. 3. Carrano AV, Thompson LH, Lindl PA, Minkler JL (1978): Sister chromatid exchange as an indicator of mutagenesis. Nature 271,551-553. 4. Wolff S (1978): Relation between DNA repair, chromosome aberrations, and sister chromatid exchanges. In : DNA Repair Mechanism, EC Friedberg, CF Fox, eds. Academic Press, New York, pp. 751-760. 5. German J (1972): Genes which increase chromosomal instability in somatic cells and predispose to Cancer. Prog Med Genet 8, 61-101. 6. Chaganti RSK, Schonberg S, German J (1974): A manyfold increase in sister chromatid exchanges in Bloom's syndrome lymphocytes. Proc Natl Acad Sci USA 71, 4508-4512. 7. Arlett CF, Lehmann AR (1978): Human disorders showing increased sensitivity to the induction of genetic damage. Ann Rev Genet 12, 95-115. 8. Shirashi Y, Sandberg AA (1980): Sister chromatid exchange in human chromosomes, including observations in neoplasia. Cancer Genet Cytogenet 1,363-380. 9. Leibovitz A, Stinson JC, McCombs WB, McCoy CE, Mazurk C, Mabry ND (1976): Classification of human colorectal adenocarcinoma cell lines. Cancer Res 36, 4562-4569. 10. Perry P, Wolff S (1974): New Giemsa method for the differential staining of sister chromatids. Nature 251,250-258. 11. Chen TR (1981}: Frequencies of sister chromatid exchanges in heteroploid cell lines of human melanoma origin. J Natl Cancer Inst 66, 273-277. 12. Shirashi Y, Minowada J, Sandberg AA (1979): Differential response of sister chromatid exchange and chromosome aberrations to mitomycin C of normal and abnormal human lymphocytic cell lines. Oncolo~ 36, 76-83. 13. Fornace AJ, Nagasawa H, Little JB (1980): Relationship of DNA repair to chromosome aberrations, sister chromatid exchanges and survival during liquid-holdings recovery in x-irradiated mammalian cells. Mutat Res 70, 323-336. 14. Preston RJ (1980): DNA repair and chromosome aberration: The effect of cytosine arabinoside on the frequency of chromosome aberrations induced by radiation and chemicals. Teratog Carcinog Mutagen 1, 147-159.
Baseline and mitomycin C (MMC)-induced sister chromatid exchanges (SCEs) in two human tumor cell lines (a colon tumor and a melanoma) and in a normal fibroblast cell line were analyzed and compared. The tumor cells showed numerical and structural chromosomal abnormalities. Their baseline SCE rate was slightly, but not significantly, higher than that of the control. Each tumor cell line showed a dose-dependent increase in SCE frequency above the spontaneous level in its own specific manner. The response in the melanoma cell was consistently below that of the control, but only the response to the highest dose of MMC (10 _9 M) was significantly lower than that of the control. The response of the colon tumor cells varied with respect to that of the control. Thus, it appears that koryotypic instability in tumor cells is not necessarily associated with elevated baseline or induced SCE/chromosome rates. In addition, within each cell line dose group, the SCE frequency was proportional to the number of chromosomes. Thus, the SCE/chromosome is a better expression of genetic damage than SCE/metaphase in analyses involving heteraploid cells.
ABSTRACT:
INTRODUCTION In the field of genetic toxicology, SCE is a sensitive indicator of genetic damage i n d u c e d by various physical and chemical agents [1-3]. Whether this response to external insult is a quantitative indicator of the damage is still u n k n o w n [4]. The relationship of SCE to hereditarily determined genetic damage (genetic instability) has been extensively investigated. The p h e n o m e n o n of genetic instability is characteristic in "chromosome instability syndromes" [5]. A m o n g those syndromes, Bloom's syndrome is characterized by an increase in baseline chromosome breakage and SCE [6]. However, increase i n baseline chromosome breakage for two other types of patients (ataxia telangiectasia and Fanconi anemia) are not correlated with an increase in SCE (reviewed by Arlett and L e h m a n n [7]). Neoplastic cells offer another system for analysis of the relationship between SCE and chromosome instability. Such studies have been conducted by several inFrom the Department of Cell Biology, The University of Texas M.D. Anderson Hospital and Tumor Institute at Houston, Houston (S.L., T.C.H.), and the Biology Division, Oak Ridge National Laboratory (W.W.A)and the Nuclear Division,Union Carbide Corporation,Oak Ridge, Tennessee(R.L.S). *Dr. Li is on leave from Ritan Hospital and Cancer Institute, Chinese Academy of Medical Sciences, Beijing,People's Republicof China. Address requests for reprints to Dr. T.C. Hsu, Department of Cell Biology, The University of Texas M.D. Anderson Hospital and Tumor Institute, Houston TX 77030. Received August 14, 1981; accepted October 5, 1981.
243 Elsevier SciencePublishingCo., Inc., 1982 52 VanderbiltAve., New York, NY 1 0 0 1 7
Cancer Geneticsand Cytogenetics6, 243-246 (1982) 0165-4608/82/070243-0652.75
244
s. Li et al. vestigators (reviewed by Shirashi and Sandberg [8]), but most of these investigations were l i m i t e d to analyses involving leukemic cells. We report here a s t u d y comparing the baseline and M M C - i n d u c e d SCE in two tumors (a colon t u m o r and a melanoma) and a normal cell line.
MATERIALS AND METHODS
Three h u m a n cell lines were used in this study. Two were t u m o r cell lines: a colon t u m o r (SW 620) and a m e l a n o m a (SW 1687). They were established by A. Leibovitz at the Scott and White Clinic, Temple, Texas. Line SW 620 was established on November 20, 1973 from a metastatic lesion of a colon carcinoma. We used it at the 83rd passage onwards. Its history and growth characteristics have been reported [9]. The m e l a n o m a line was established on July 15, 1977 from a metastatic lesion in the thigh of the patient. We used it at the 20th passage onwards. The cells of the m e l a n o m a line continue to synthesize m e l a n i n in vitro. A normal d i p l o i d h u m a n fibroblast cell line (BAS) used as control was p r o v i d e d by Dr. Lee Kronenberg of Lee Biomolecular Research Laboratory, San Diego, California. The tumor cell lines were m a i n t a i n e d in Leibovitz L-15 m e d i u m s u p p l e m e n t e d with 10% fetal calf serum; whereas, the BAS line was m a i n t a i n e d in McCoy's 5a m e d i u m w i t h 10% fetal calf serum.
Figure 1 A differentially stained metaphase from the melanoma cell line treated with 10 -~ M MMC showing 72 chromosomes with 22 SCE.
7 I
245
Baseline a n d M i t o m y c i n - C - i n d u c e d SCEs
Confluent cultures were trypsinized and 5 × 10 ~ cells were seeded into T-25 flasks in 10 ml of m e d i u m . Duplicate flasks were treated with various concentrations of MMC (10 -~, 10 -1°, 10-11M) and b r o m o d e o x y u r i d i n e (BrdU) (20 ~g/ml). The BAS cultures were havested after a 40 hr treatment. However, w h e n tumor cell cultures were harvested at 48 and 72 hr, very few cells had proceeded through two cell cycles as indicated by the absence of differential staining of the sister chromatids. Therefore, after 72 hr the culture m e d i u m was replaced with one c o n t a i n i n g BrdU alone. Cultures were harvested 24 hr later. With this protocol, sufficient n u m b e r s of cells were i n mitosis and the majority of metaphases showed differential staining. A n example of a differentially stained metaphase from the m e l a n o m a cell line treated with 10 .9 M MMC is s h o w n in Figure 1. Cultures were harvested after 4 hr by colcemid (0.04 ~g/ml) treatment. Cells were trypsinized, treated with 0.7% sodium citrate, and fixed three times with Carnoys' fixative. Air-dried cytological preparations were treated with the FPG techniques of Perry and Wolff to show differential staining of sister chromatids [10]. For each cell line and each dose, 20 metaphases were scored for SCE frequency and for chromosome number. T h e n the entire procedure was repeated; thus, the study actually consisted of two i n d e p e n d e n t experiments. RESULTS Table 1 summarizes the results of the study. Analysis of variance of the two indep e n d e n t l y conducted experiments showed that experiments and all interactions involving experiments have a negligible effect (p = 0.48 for n u m b e r of chromosomes and p = 0.34 for SCE/chromosome). Within the limits of experimental error the results of the two experiments are identical, and we have c o m b i n e d the results of the two experiments in Table 1. Analysis of the distribution of chromosomes showed that all three cell lines were significantly different i n terms of chromosome n u m b e r (p -< 0.0001). This difference was considered w h e n comparing the dose respons~ of the different cell lines. As s h o w n in Table 1, the means of the baseline SCE/chromosome for the two tumor
Table 1
Chromosome distribution and MMC-induced sister chromatid exchanges in three cell lines
Number of chromosomes/cell Cell line
Dose (M)
BAS (control)
0 10 -'1 10 -l° 10 -9 0 10 -ll 10 -1° 10 -9 0 10 -11 10 -~° 10 -9
Colon tumor
Melanoma
Mode (min-max) 46 46 46 46 50 50 50 50 67 64 65 65
(41-46) (40-46) (41-46) (42-46) (34-52) (35-53) (38-55) (38-53) (50-75) (50-72) (49-72) (50--72)
SCE/chromosome
SCE/chromosome control mean
Mean
(SE)
Mean
(SE)
Mean
45.4 45.8 45.5 45.7 49.1 49.9 48.9 48.6 64.3 60.7 63.1 63.9
(0.2) (0.2) (0.2) (0.2) (0.5) (0.3) (0.5) (0.4) (1.0) (0.7) (0.9) (0.8)
0.077 0.089 0.132 0.202 0.093 0.111 0.111 0.263 0.094 0.103 0.146 0.166
(0.005) (0.004) (0.005) (0.008) (0.006) (0.008) (0.007) (0.018) (0.004) (0.005) (0.009) (0.007)
-1.14 1.71 2.60 -1.19 1.19 2.84 -1.10 1.55 1.75
(SE) (0.05) (0.07) (0.10) (0.09) (0.08) (0.19) (0.06) (0.10) (0.07)
246
s. Li et al.
cell lines were higher than that of the control. However, there was no significant difference among t h e m (p -- 0.28). The sensitivity of the e x p e r i m e n t was such that we could be 90% certain of detecting differences as small as 0.037 in SCE/chromosome for the different treatment groups. All cell types s h o w e d a d o s e - d e p e n d e n t increase in SCE/chromosome. The dose of MMC was the d o m i n a n t factor in determ i n i n g this SCE response (p -~ 0.0001), although there was also a significant d o s e cell line interaction (p -~ 0.0001). The SCE/chromosome of each dose was n o r m a l i z e d by its own untreated control and was plotted against the dose as s h o w n in Figure 2. A l t h o u g h the baseline SCE rate for the m e l a n o m a cells was higher than that of the BAS cells, the n o r m a l i z e d response of the treated m e l a n o m a cells was consistently b e l o w that of the control culture. The d o s e - c e l l line interaction is most clearly illustrated by the behavior of the colon tumor cells, w h i c h showed a threshold effect in response to the damage i n d u c e d by MMC at 10 11 and 10 -1° M. The n o r m a l i z e d SCE frequency of the colon tumor i n d u c e d by 10 -1° M MMC was significantly different from those of the other two cultures. The m e l a n o m a line was significantly different from the other two at 10 -9 M. (This conclusion is based on analysis by the Duncan m u l t i p l e range test.) With regard to the appropriateness of the end point, SCE/chromosome, we considered the following models: SCE = (constant d e p e n d i n g on dose and cell line) × (number of chromosomes) x (random error), SCE = (constant d e p e n d i n g on dose, cell line, and n u m b e r of chromosomes) x (random error).
(1) (2)
Notice that m o d e l (1) is a special case of m o d e l (2). R a n d o m error represents factors other than dose, cell line, or n u m b e r of chromosomes. (Log r a n d o m error was assumed to have normal distribution, and the two models were c o m p a r e d by the Figure 2 Mitomycin-C-induced changes in SCE/chromosome relative to background. Vertical bars represent 95% confidence interval for the true response at specified cell lines and dose. For simplicity, these have been included only for those cell lines differing significantly from the other two at the specified dose. 3.0 • Colon
Tumor
it
,,' .. B A S • • .° o°'°
E
°°~?
°.°°o°°., •• °o° i °°°°°°°°°°"j7It ~
¢-
0
02.0 E E 0
0 t,tO LU
*
/
°°00°
C O
°°°° t ~ ° ' ° ° " J
0 o3
Melanoma
¢
1.0
I 10 -11
L
10 -10 Mitomycin
10 . 9
C Concentration(M)
Baseline and M i t o m y c i n - C - i n d u c e d SCEs
247
method of analysis of variance.) We found that model (2) does not provide a significantly better fit to the data than model (1) (p = 0.96). The sensitivity of this analysis is such that differences on the order of 17% in the expected SCE frequencies under models (1) and (2) would be detected with 90% certainty. Thus, for a given cell line-dose group, within the limits of experimental error SCE is equal to a constant times the number of chromosomes; hence, SCE/chromosome is constant. This is certainly not true for SCE/metaphase. We think, therefore, that SCE/chromosome is a better indicator of cytogenetic damage than SCE/metaphase. DISCUSSION Studies on the frequency of SCE in tumor cells have been limited mostly to analyses of baseline SCE frequencies in myelo- and lymphoproliferative diseases [8]. In general, it was concluded that leukemic cells were not characterized by an inherently high incidence of SCE, regardless of the type of leukemia. Some of the exceptions could be accounted for by prior drug treatment. On the other hand, increased SCE frequency was usually observed in cells from patients receiving chemotherapy and continued for a short time after the cessation of chemotherapy. Similar analyses on cells derived from solid tumors were rarely performed. The colon tumor and the melanoma cell cultures used in this study are similar in terms of genetic instability to most heteroploid cell lines established from solid tumors. Both showed an abnormal distribution of chromosomes (Table 1). Chromosomal rearrangements have been observed in the colon tumor [9] and in the melanoma (S. Pathek, unpublished observation). In addition, we observed an elevated rate of baseline chromosome breakage in these two tumor cell lines when the rates were compared with that of the control (0.42 lesions/metaphase for the colon tumor, 0.18 for melanoma, and 0.04 for BAS culture). The control culture, BAS, was karyotypically normal (L. Kronenberg, unpublished observation). Our data showed that the baseline SCE frequency of the tumor cells was higher than that of the control cells although they were not significantly different from one another. However, significantly higher baseline SCE frequencies were observed by Chen in five heteroploid cell lines and sublines from melanomas [11]. Shirashi et al. observed significant increases in MMC-induced SCE in leukemia and lymphoma cell lines as compared with normal lymphocytes [12]. Conceivably, induction of genetic damage by chemicals can be influenced by a number of factors, e.g., penetration, metabolism of the chemical, and repair of induced damage. Indeed, variable responses were observed in this study. Although the cell lines showed dose-dependent increases in SCE, each had its own characteristic and reproducible response. The response of the melanoma cells was consistently lower than that of the control. The colon tumor cells showed a threshold effect not seen in the control, while at high doses of MMC the SCE response was as high as that of the control. Such lack of consistent elevation of baseline and induced SCE frequencies in the tumor cells indicated that karyotypic instability is not necessarily correlated with SCE/chromosome rates. The cell line-specific response cannot be explained simply by the long cell cycle time available for repair of damage in the tumor cells although the role of cell cycle in repair has been observed [13,14]. A better understanding of such a phenomenon might be achieved by using more dose level of MMC, by using other inducing agents, and by using other assaying techniques. W.W. Au's research is supported by the Office of Health and Environmental Research, U. S. Department of Energy, under contract W-7405-eng-26 with the Union Carbide Corporation.
248
S. Li et al. This is a joint research project of The University of Texas System Cancer Center M.D. Anderson Hospital and Tumor Institute and the John S. Dunn Research Foundation of Houston, Texas. We would like to thank Mr. Dindyal Ramkissoon for his expert technical assistance.
REFERENCES 1. Latt SA (1975): Sister chromatid exchanges, indices of human chromosome damage and repair: Detection by fluorescence and induction by mitomycin C. Proc Natl Acad Sci USA 71, 3162-3166. 2. Perry P, Evans HJ (1975): Cytological detection of mutagen-carcinogen exposure by sister chromatid exchange. Nature 258, 121-125. 3. Carrano AV, Thompson LH, Lindl PA, Minkler JL (1978): Sister chromatid exchange as an indicator of mutagenesis. Nature 271,551-553. 4. Wolff S (1978): Relation between DNA repair, chromosome aberrations, and sister chromatid exchanges. In : DNA Repair Mechanism, EC Friedberg, CF Fox, eds. Academic Press, New York, pp. 751-760. 5. German J (1972): Genes which increase chromosomal instability in somatic cells and predispose to Cancer. Prog Med Genet 8, 61-101. 6. Chaganti RSK, Schonberg S, German J (1974): A manyfold increase in sister chromatid exchanges in Bloom's syndrome lymphocytes. Proc Natl Acad Sci USA 71, 4508-4512. 7. Arlett CF, Lehmann AR (1978): Human disorders showing increased sensitivity to the induction of genetic damage. Ann Rev Genet 12, 95-115. 8. Shirashi Y, Sandberg AA (1980): Sister chromatid exchange in human chromosomes, including observations in neoplasia. Cancer Genet Cytogenet 1,363-380. 9. Leibovitz A, Stinson JC, McCombs WB, McCoy CE, Mazurk C, Mabry ND (1976): Classification of human colorectal adenocarcinoma cell lines. Cancer Res 36, 4562-4569. 10. Perry P, Wolff S (1974): New Giemsa method for the differential staining of sister chromatids. Nature 251,250-258. 11. Chen TR (1981}: Frequencies of sister chromatid exchanges in heteroploid cell lines of human melanoma origin. J Natl Cancer Inst 66, 273-277. 12. Shirashi Y, Minowada J, Sandberg AA (1979): Differential response of sister chromatid exchange and chromosome aberrations to mitomycin C of normal and abnormal human lymphocytic cell lines. Oncolo~ 36, 76-83. 13. Fornace AJ, Nagasawa H, Little JB (1980): Relationship of DNA repair to chromosome aberrations, sister chromatid exchanges and survival during liquid-holdings recovery in x-irradiated mammalian cells. Mutat Res 70, 323-336. 14. Preston RJ (1980): DNA repair and chromosome aberration: The effect of cytosine arabinoside on the frequency of chromosome aberrations induced by radiation and chemicals. Teratog Carcinog Mutagen 1, 147-159.