cyclophosphamide induced in vivo sister chromatid ...

2 downloads 0 Views 412KB Size Report
The inducibility of sister chromatid exchanges (SCEs) by cycIophosphamide (CP) in bone marrow ceIls .... sification of twisted chromatids as SCEs. The analysis ...
Can. J. Genet. Cytol. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/04/13 For personal use only.

CYCLOPHOSPHAMIDE INDUCED IN VZVO SISTER CHROMATID EXCHANGES (SCE) IN MUS MUSCULUS. I: STRAIN DIFFERENCES AND EMPIRICAL ASSOCIATION WITH RELATIVE CHROMOSOME SIZE DOROTHY L. REIMERAND SHIVAM. SINGH Dc'j?urtrnetlt o,f'Zoolog~..Uni\*ersit\ of Western Onturio, Lontion, Ontario. Cunuda N6A 5B7 The inducibility of sister chromatid exchanges (SCEs) by cycIophosphamide (CP) in bone marrow ceIls was evaluated in vivo in the three genetic strains of mice (C3H/s, C57BL/6J, and Balbic). Female mice (10 to 12 wks old, mean = 22.9g, SD = 3.2g) were administered with nine hourly injections of 214.19 mgikg 5-Bromo-2' deoxyuridine (BrdU) folIowed by 0 , 0.048, 0.449, 4.585 or 46.93 mgikg CP and 4 mgikg colcemid. SCEs were evaluated following differential staining procedures of Perry and Wolff (1974). The base-line SCEs were similar in all strains with about ten SCEsicell. Increasing CP concentrations yielded an increased level of SCEs. Most cells showed extensive damage in CP doses exceeding 4.55 mgikg. No SCE evaluation was possible beyond this concentration. Strain differences were evident at every dose of CP, and Balbi c was the least susceptible strain to SCE induction. F, hybrids involving C3His P and Balbic d showed SCE values closer to Balbic. Data on the association between chromosome length and frequency of SCEs are provided. They empirically establish a positive correlation (r = 0.90) between the two features. Most induced SCEs were interstitially located rather than terminally positioned on the chromosome. On a CvaluC it1 \,i\,o la capacitc d'induction d'echangcs de chromatides soeurs (SCEs) par Ic cyclophosphamide (CP) dans Ies ccllules de moelle osseuse chez trois lignces genctiques de souris (C3H:s. C57B1~6J.et Balb!c). On a administre 1des souris femelles (rige: 10 1 I? semaines. poids moyen: 7 2 . 9 ~DS: . 3.29) neuf injections consCcutives 1 une hcure d'intervalle d'une dose de 211.19 mpkg de 5-Bronio-2'd6oxyuridine (BrdU) huivie\ de doses de 0. 0.048. 0.149, 1.585 ou 16.93 n1g:kg de CP et de 1mg kg dc dimCcolcine (colcemid).On a fait Ie dCcompte des SCEs aprt:s application de la m6thodc dc coloration diffkrcntielle de Perry et Wolff (1971). On a obtenu des taux de baxc de SCE\ xemblablcs chez touteh les lignecj. c'est-1-dire approximativement dix SCEs cellule. Des concentrations plus fortes de C'P ont entrain6 une augmentation du nombre deh SCEs. La rnaJoritC des cellulej ont manifest6 d'irnportants dommages i dex doses de CP depassant 1.55 mg.kg. Au deli de ce xeuil il est devenu imposhible de faire Ic dCcompte des SCEs. Leh lignies se sont comportCes de fagon diffirente 1 chaque dose de CP. et Balb,c s'est avcree la lignce la rnoins xuaceptible h I'induction de SCEs. Les hybrides F, issus de C3H;s 2 et de Balbi c s' ont fourni des taux de SCE se rapprochant de ceux de Balb c. On fournit des donnCes sur la correlation entre la longueur des chromoson~eset la friquence des SCEs. Ces donnees ktablissent ernpiriquernent une correlation positive ( r = 0.90) entre les deux parametres. La plupart des SCEs induits Ctaient localist!s dans des interstices plut6t qu'en position terminale sur le chromosome. [Traduit par le journal]

Introduction The n~olecularbasis and biological sigificance of sister chromatid exchanges ( S C E s ) are unknown or at best circumstantial. SCE formation is coupled to DNA synthesis (Kato, 1 9 8 0 ) . It presumably represents the interchange of DNA replication products at apparently homologous loci. and involves breakage and reunion at the level of the DNA duplex. In spite of these uncertainties, studies on SCE have provided an understanding towards the basic structure of eukaryotic chromosomes. SCE analysis is generally regarded to be among the most sensitive methods for detecting chromosome damage by certain agents. It is also used as an indicator of cellular response to genetic damage induced by a wide variety of mutagens and carcinogens. Furthermore, it is widely used to differentiate between chromosome fragility diseases (Latt et a l . , 1 9 7 9 ) . Recently, Latt et ul. ( 1 9 8 0 ) and Kato ( 1 9 7 7 ) , among others, reviewed different aspects of SCE studies, most involving in Manuscript received November 20. 198 1 .

Can. J. Genet. Cytol. 24: 521-528, 1982.

Can. J. Genet. Cytol. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/04/13 For personal use only.

522

D. L. REIMER AND S. M. SINGH

~jitrowork. They emphasized the importance of in vivo studies in evaluating the individual variation in SCE formation. Furthermore, in vivo studies are useful in evaluating the role of "metabolic activation" and tissue specificity in causing SCEs. Kato (1977) suggested that spontaneous SCEs may be subject to genetic control and vary from individual to individual. In this context, Galloway and Evans (1975) reported SCE frequencies ranging from 0.2 1 to 0.48 per chromosome in 12 normal individuals. In addition, -it has been speculated that the number of SCEs is related to the genome size (Kato, 1977), relative DNA content of the chromosome (Carrano and Wolff, 1975), and roughly proportional to the length of the metaphase chromosome (Dutrillaux, et al., 1974). Although the relationship between SCEs and the length of the metaphase chromosomes is realized, there are exceptions to this rule. For example, the smaller chromosomes of groups, E , F, and G in humans exhibit fewer exchanges than would be expected based on their metaphase length (Chaganti et a l . , 1974; Latt, 1974; Galloway and Evans, 1975). These exceptions are most likely due to the preferential occurrence of exchanges in specific chromosome regions. An understanding of these exceptional regions, in chromosomes of species used in testing for mutagenesis, is desirable for appropriate interpretation of the data. Mrrs rnusc.rrlris, with 20 pairs of telocentric chromosomes (Davisson and Roderick, 1978). is used in extensive cytological experimentation. Misclassification of SCEs resulting from centroineric twisting is avoided with the use of these chromosomes. Furthermore. well established strains are available for appropriate genetic analysis in this species. In this report, we present data on in 1-i\w-induced SCEs on individual chromosomes of three genetic strains of mice. 5-Bromo-2'deoxyuridine (BrdU) in conjunction with different doses of cyclophosphan~ide(CP). a known mutagen (Hill. 1975). is used to assess differences between strains and provide statistical correlation between frequency of SCEs and relative length of metaphase chromosomes. These results are discussed in the light of possible attributions of genetic determinants in the formation of SCEs. Materials and Methods In t~ivoSCE studies involve BrdU incorporation for one cell cycle: however, sister chromatid differentiation requires two rounds of DNA replication. BrdU is provided either as an implant tablet with continuous release in the system (Allen et (11.. 1977). or by hourly i.p. injections of a suitable dose to permit good sister chromatid differentiation (Allen and Latt, 1976). We administered nine i.p. injections of 2 14.19 2 25.52 mglkg BrdU at hourly intervals to 10 to 12 wk old (22.9 2 3.2 g) female mice of genetic strains C3Hls. C57BLI6J. and Balb!c. Fifteen minutes following the last BrdU injection. at hour 9 . they were given a single 0 . 4 ml i.p. injection of cyclophosphamide dissolved in H 2 0 to provide a concentration of 0. 0.0483 2 0.008. 0.449 2 0.053. 4.58 2 0.52. or 46.93 t 5.87 mglkg C P . After 12 h. at hour 21, each individual was given a 0 . 4 ml i.p. dose of colcemid (-4 mglkg body weight). BrudU (Sigma), C P (Sigma). and Colcemid (Gibco) were of analytical grade. and solutions were used within 2 wks of their preparation. During this period. they were stored in light proof containers at 4°C. Animals were sacrificed 2 h following colcemid administration. at hour 23. by cervical dislocation. Both femurs were removed in the dark. and bone marrow cells were collected by forcing 37°C PBS (pH = 7.0) through the marrow canals using a 26 gauge 12.7 mni (0.5 inch) needle. The cells were treated with 0.075 M KC1 for 25 min at 37°C. centrifuged (1200 rpm. 10 min) and resuspended in fixative ( 3 methanol : 1 acetic acid). Air dried slides were prepared by dropping the cell suspension onto clean glass slides and were stored in a light-tight box until further processing. The differential staining was performed following Perry and Wolff (1974) with the exception that the slides were immersed in PBS (pH = 7.0) and exposed to ultraviolet light for 60 min prior to incubation in 2 x SSC at 65'C ( 2 min) and 10 min staining in 2% Giemsa (Fisher Scientific Co.). Slides were evaluated for differential staining of sister chromatids and SCEs. using a light microscope. Approximately 20 well-spread metaphase cells were photographed. They were projected on a screen where the length of each chromosome and its SCEs were recorded. The location of each SCE on the chromosome was also characterized as interstitial or terminal.

IN VIVO SCE-INDUCTION IN MURINE STRAINS

523

Can. J. Genet. Cytol. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/04/13 For personal use only.

Analysis of variance was used to evaluate the strain differences, treatment ( C P dose) effects, and the effect of relative chromosome length on the frequency of SCEs. Regression lines were established to evaluate the relationship between relative chromosome length and induced SCE level.

Results and Discussion The method yielded excellent sister chromatid differentiation in metaphase chromosomes from bone marrow cells. The telocentric mouse chromosomes avoided misclassification of twisted chromatids as SCEs. The analysis of variance showed that the frequency of SCEs varied significantly among strains (F,,.,,,, = 18.6'") (*Denotes significance beyond 0.01 probability) which were differently distributed on chromosomes = 12.9*). Furthermore, increasing' doses of C P yielded elevated frequencies of (F,,,,,,,, SCEs (Ft3.215) = 13.5%)in three genetic strains. Figure 1 shows the SCE frequencyicell t SE in the three strains (C3Hls, C57BLl6J, and Balblc) of mice in response to different doses of cyclophosphamide. Here, the baseline SCEs are around lOicell and not significantly different among strains. However, increased SCEs, in response to different doses of CP, produced significant strain difference~. The Balblc has consistently lower values of SCEsicell at every treatment which are significantly different from C3HIs with relatively higher values as shown in the insert of this figure. Correlation coefficient for the linear regression of SCEs induced by C P for the three strains was 0.97. The only other results using C P and involving these strain differences were reported by Dragani et a / . ( 1981). who had similar results. Our values of baseline SCEsIcell, however, are relatively higher ( - 1Olcell) as compared to about 3lcell.

DOSE OF CP ( m g / k g )

Fig. 1: Mean SCESIcell induced by cyclophosphamide (CP) in three strains of mice. Insert shows C P dose response in these strains and hybrid involving C3Hls P and Balblc d .

Can. J. Genet. Cytol. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/04/13 For personal use only.

5 24

D. L. REIMER AND S. M . SINGH

reported by these workers. This may reflect on the methodological differences in the administration of BrdU in these two reports. It should be pointed out, however, that Dragani et al. (1981) used BrdU tablet implants while we i.p. injected BrdU for nine hourly intervals to label the chromatids. Also, it is realized that there is a minimum spontaneous level of SCEs in a given cell cycle, and they are also induced by BrdU (Lambert et al., 1976; Stetka and Carrano. 1977). Higher base-line values of SCEsIcell in this report, therefore, may also reflect on a differential dose of BrdU to obtain consistent sister chromatid differentiation. At a higher dose of CP (46.93 mglkg), most cells from all three strains showed extensive genetic damage, and only a few cells were available where SCE frequencies could be counted. In one such cell of C3H/s, up to 80 SCEs were counted. At a CP dose over 450 mglkg, no good quality metaphases were observed. Most cells showed numerous micronuclei, a reflection of extensive damage. This pattern was similar in the three strains. F, hybrids (C x A) between C3HIs Q and Balblc c? were also treated with 0 , 0.048, 0.449, or 4.585 mglkg doses of CP and produced SCEIcell values of 8.15 t 0.55, 14.0 t 2.59, 16.5 -+ 2.1 1, and 22.75 t 1.95, respectively. These values are more like values 1.39, and 22.25 t 2.27) and obtained for Balblc (8.8 -+ 1.61, 1 1.5 -+ 1.74, 12.15 differ significantly from C3HIs (Fig. 1). Similar genotype dependent differential response of Mitomycin C and BCNU ( 1,3-bis [2-chloroethyll- 1 -nitrosurea) in producing SCEs was also reported by Kram and Schneider (1978) and Biegel et al. (1980) in mice. Cyclophosphamide is an alkylating agent, which is mutagenic and carcinogenic (Mohn and Ellenberger, 1976) and requires metabolic activation by liver microsomes (Hill, 1975). Differential genotypic response as observed here may be due to a differential activation of CP by individual genotypes andlor differential susceptibility of genotypes to damage caused by CP and other such agents, eg. cysteine (Rosin and Stich, 1978) which is capable of trapping free radicals. However. these results probably deal more with the chetnistry of the inducing agent than with alterations in cellular response to the damage induced (Latt et al.. 1980). Similar responses may be expected if enzymes like superoxide dismutase (Nordenson, 1977) and glutathione peroxidase (La1 et a l . , 1980) can protect cells from chromosome damage. Further in l i l ~ ostudies are needed to evaluate the role of genotypes in metabolic activation of mutagens and genotype dependent differential cellular response to the damage induced.

*

Distribiitiotz qf ltzd~icedSCES 0 1 2 the Chromosomes The physical properties of the mouse chromosomes allow accurate evaluation of SCEs. However. the identification of individual chromosomes in this report is based primarily on relative chromosome length, and ambiguity may result. As mentioned earlier, we measured the relative length of each chromosome and recorded the number and location (terminal or interstitial) of its SCEs. The data permit the evaluation of the relationship between chromosome size and the frequency of SCE. Figure 2 shows such a relationship for the three strains at the base-line (A) and induced SCE following 4.585 mglkg CP treatment (B). The correlation coefficients for regression lines A and B of the three strains are significant (range = 0.80 to 0.90). They give the statistical dependence of the probability of SCE on chromosome size. A significant difference between the slopes of regression lines A and B in all three strains suggests a relatively greater increase of SCEs on larger chron~osomesas a result of cyclophosphamide treatment. However, when the points of the B line (induced SCE) were corrected for chromosome length (i.e. for chromosome i, ci = (llrelative length for chromosome i ) x mean SCE i), and regressed against chromosome length, line C was produced as shown in Fig. 2 for the three strains. Line C is parallel to A (the base-line) for the two strains, C3HIs and Balblc. This suggests that like the base-line SCEs, the induced SCEs are also dependent on the chromosome size. The difference, a, between lines A and C, therefore, reflects on the effect of 4.585 mglkg

Can. J. Genet. Cytol. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/04/13 For personal use only.

IN V I V O SCE-INDUCTION IN MURINE STRAINS

525

Fig. 3: Relationship involving induced SCESlchromosome and relative length of the chromosome at metaphase. in three ~nurinestrains. Line A (solid circles) represent 213.19 mg/kg!h BrdU. Line B (open circles) represents 213.19 mglkglh BrdU accompanied by 3.585 mglkg cyclophosphamide. Line C shows induced SCEs/ chromosome when line B is corrected for chromosome length (see text for details). a is the induced SCEsl chromosome for the three strains.

cyclophosphamide in causing SCES. The C line for C57BLl6J is almost parallel to A. but its u value is approximately 34% higher than the other strains. Although a chromosome with near 0.68 relative size of C57BLl6J has high SCE frequency as compared to other

Can. J. Genet. Cytol. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/04/13 For personal use only.

D. L. REIMER AND S. M. SINGH

.

Fig. 3. Distribution of SCEs on individual chromosomes in three strains of mice. *Chromosome identity is based only on relative length. absence of SCE. two SCEs three or more SCEs. one SCE,

chromosomes. the increase in SCE frequency in this strain. in general. is distributed over all chromosomes. Figure 3 shows the distribution of number of SCEs (0.1,2,3, or more) induced by 4.585 mgikg C P on individual chromosomes, along with their relative lengths. These data

Can. J. Genet. Cytol. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/04/13 For personal use only.

IN VIVO SCE-INDUCTION IN MURINE STRAINS

527

are based on 19 cells for each strain. The regression line of number of SCEs and relative chromosome length is given in Fig. 2 (line B). It shows that in general most of the multiple SCEs are restricted to larger chromosomes, and most of the chromosomes with no SCEs belong to smaller groups. Caution should be used when interpreting these data, since chromosome identification is based solely on relative length. The high frequencies of SCEs observed on a larger chromosome could come from two possible sources. On a given chromosome, many cells possess at least a single SCE, or when SCE is present, it exists in greater numbers. Chromosome 11, for example in C57BL/6J, has about 50% of cells with one SCE, and chromosomes 9 to 11 of C3HIs have up to 25% of cells with two or more SCEs, all producing approximately the same frequency of SCEs on these chromosomes. Also Balblc chromosomes with low SCEs have, in general, a higher proportion of chromosomes with no SCEs. Thus, strain differences for the frequency of total SCEs in response to cyclophosphamide are explained using differential general susceptibility of all chromosomes under different genetic backgrounds, rather than involvement of only a few chromosomes. Although most reports have implied a relationship between SCEs and chromosome size, to our knowledge no statistical evidence has ever been reported in the literature. This report, therefore,provides the first empirical evidence for such an association in mice. For in vivo treatment with 4.585 mglkg cyclophosphamide over different strains, the correlation between induced SCEs and chromosome length is 0.9 1 with regression equation Y = - 1 . 4 ~+ 39.9, where Y is the frequency of SCEs and times the relative size of the chromosome. Most induced SCEs were found to be interstitially located (x' = 13.6, P = 0.99) in all three strains. This phenomenon may be accounted for by the fact that internal positions posses a larger portion of the chromosome than do terminal positions, making the former more available to additional SCEs. These relationships are useful when strains of mice are used to evaluate in \'i~'ogenetic damage by a given mutagen.

Acknowledgments This study was financed by a Natural Sciences Engineering Research Council grant to Dr. Singh. We are thankful to Dr. R. Green for statistical advice and help during the data analysis. References Allen. J . W. and Latt. S. A. 1976. I n rsir30BrdU-33258 Hoechst analysis of DNA replication kinetics and sister chromatid exchange formation in mouse somatic and meiotic cells. Chromosoma, 58: 325-340. Allen. J . W . . Shuler. C. F.. Mendes. R. W. and Latt. S. A. 1977. A simplified technique for in tire analysis of sister chromatid exchanges using 5-bromodeoxyuridine tablets. Cytogenet. Cell Genet. 18: 231-237. Biegel. J . A , . Boggs. S . S. and Connors. M. K. 1980. Comparison of BCNU-induced SCE in bone-marrow cells of AKRIJ and BDF, mice. Mutat. Res. 79: 87-90. Carrano. A. V. and Wolff, S . 1975. Distribution of sister chromatid exchanges in euchromatin and heterochromatin of the Indian muntjac. Chromosoma, 53: 36 1-369. Chaganti, R. S . K.. Schonberg. S . and German. J. 1974. A many fold increase in sister chromatid exchanges in Bloom's syndrome lymphocytes. Proc. Natl. Acad. Sci. U.S.A. 71: 4508-45 12. Davisson. M. T . and Roderick. T. H. 1978. Status of the linkage map of the mouse. Cytogenet. Cell Genet. 22: 552-557. Dragani, T . A , . Zunino. A. and So7zi. G. 1981. Differences in sister chromatid exchange (SCE)-induction in r,iro by cyclophosphamide in murine strains. Carcinogenesis, 2: 219-222. Dutrillaux. B.. Fosse. A. M.. Prieur. M. and Lejeune. J . 1974. Analyses de kchanges de chromatides dans le cellules somatiques humaines. Chromosoma. 48: 327-340. Galloway, S . M . and Evans, H. J. 1975. Sister chromatid exchange in human chromosomes from normal individuals and patients with ataxia telangiectasia. Cytogenet. Cell Genet. 15: 17-29. Hill, D. H. 1975. A review of cyclophosphamide. Thomas Press. Springfield, Ill., U.S.A. Kato. H. 1977. Spontaneous and induced sister chromatid exchanges as revealed by BUdR-labeling method. Int. Rev. Cytol. 49: 55-97. Kato, H. 1980. Evidence that the replication point is the site of sister chromatid exchange. Cancer Genet. Cytogenet. 2: 69-77.

Can. J. Genet. Cytol. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/04/13 For personal use only.

528

D. L. REIMER AND S. M. SINGH

Kram, D. and Schneider, E. L. 1978. Reduced frequencies of mitomycin-C induced sister chromatid exchanges in AKR mice. Hum. Genet. 41: 45-51. Lal, A. K . , Ansari. N. H.. Awasthi, Y. C . , Snyder, L. M . , Fortier. N. L. and Srivastava. S. K. 1980. Defense of mouse red blood cells against oxidative damage by phenylhydrazine. glutathione peroxidase and catalase deficiency. J . Lab. Clin. Med. 95: 536-551. Lambert. B., Hansson, K., Lindsten, J . . Sten, M. and Werelius, B. 1976. Bromodeoxyuridine-induced sister chromatid exchanges in human lymphocytes. Hereditas. 83: 163-174. Latt, S . A. 1974. Localization of sister chromatid exchanges in human chromosomes. Science. 185: 74-76. Latt, S . A., Schreck, R. R., Loveday, K. S . and Shuler. C. F. 1979. In vitro and in vivo sister chromatid exchange. Pharm. Rev. 30: 501 -533. Latt. S . A , , Schreck, R. R.. Loveday. K. S . , Dougherty. C. P. and Shuler. C. F. 1980. Sister chromatid exchanges. I n Advances in human genetics. Vol 10. Edited by H. Harris and K. Hirschorn. Plenum Press, New York. pp. 267-331. Mohn, G . R . and Ellenberger, J. 1976. Genetic effects of cyclophosphamide, ifosfamine. and trofosfamide. Mutat. Res. 32: 33 1-360. Nordenson, I. 1977. Effect of superoxide dismutase and catalase on spontaneously occurring chromosome breaks in patients with Fanconi's anemia. Hereditas, 86: 147-150. Perry, P. E. and Wolff. S. 1974. New Giemsa method for differential staining of sister chromatids. Nature (London), 261: 156- 158. Rosin. M. P. and Stich, H. F. 1978. The inhibitory effect of cysteine on the mutagenic activities of several carcinogens. Mutat. Res. 54: 73-81. Stetka. D. and Carrano, A. V. 1977. 'The interaction of Hoechst 33258 and BrdU substituted DNA in the formation of sister chromatid exchanges. Chromosoma. 63: 2 1-31.