Brain Tumor Pathol (2001) 18:83-87

9 The Japan Society of Brain Tumor Pathology 2001

Sandra R e y e s 9 D a n i e l R e m b a o 9 Julio S o t e l o

The antimalarials quinacrine and chloroquine potentiate the transplacental carcinogenic effect of ethylnitrosourea on ependymal cells

Received: May 21, 2001 / Accepted: June 28, 2001 A b s t r a c t Quinacrine and chloroquine, two widely used

antimalarials, bind strongly to deoxyribonucleic acid, thus preventing mutagenesis. We studied a possible chemoprotective effect of these substances on carcinogenesis of the nervous system induced in Wistar rats by transplacental administration of ethylnitrosourea. One experimental group consisted of rats born from mothers treated with quinacrine prior to prenatal exposure to ethylnitrosourea; a second group consisted of rats chronically treated with chloroquine after prenatal exposure to ethylnitrosourea. When compared with controls, no significant differences were observed in tumor incidence. However, early tumor growth was observed in both rats treated with quinacrine (P < 0.0004) and rats treated with chloroquine (P <'0.0Z). These differences were due mostly to rapid development of ependymomas of the spinal cord. Our results suggest that quinacrine and chloroquine do not prevent the structural alterations induced in D N A by ethylnitrosourea, which lead, in the long term, to a high incidence of neoplasms in the nervous system. Moreover, the antimalarials studied seem to promote the carcinogenic effects of ethylnitrosourea on ependymal cells.

Introduction

The changes that precede the appearance of neoplasms in animals prenatally exposed to ethylnitrosourea (ENU) are unknown. According to the somatic mutation theory, deoxyribonucleic acid (DNA) damage in embryonic cells initiates malignant transformation, which will manifest later in descendent cells. Some experimental models, such as transplacental administration of ENU, have been used to study the relationship between mutagenesis and carcinogenesis. E N U acts on the cells of the germinal matrix, apparently by alkylation of D N A bases, with the subsequent development of a malignant tumor selectively located within the central nervous system (CNS). 1-3 Mutagenesis inhibitors prevent the initiation, promotion, and progression of carcinogenesis, either by blockage of specific enzymes involved in the metabolism of xenobiotics or by antioxidant mechanisms. Chemopreventive antioxidant activities include inhibition of arachidonic acid metabolism. Most important for preventive purposes is that D N A is the main target of carcinogens, and therefore, the possible blockage of binding of a carcinogenic substance K e y w o r d s Antimalarials - Carcinogenesis 9 Chloroquine 9 to D N A might be used in studies of antimutagenesis and Ethylnitrosourea-quinacrine anticarcinogenesis. 4'5 The synthetic antimalarial compounds chloroquine and quinacrine are strong D N A chemoprotectors and phospholipase A2 inhibitors. 6-8 Complexes from these substances intercalated within the D N A prevent mutagenesis. 9 Quinacrine precludes somatic mutations in bacteria 1~ and reduces the frequency of mutations in highly mutagenic cellsJ 1-14 Additionally, quinacrine has shown anticarcinogenic activity in rat mammary gland induced by Nmethyl-N-nitrosourea, possibly through inhibition of phospholipase A2; however the chemopreventive activity S. Reyes ([~) 9J. Sotelo Neuroimmunology Unit, Instituto Nacional de Neurologfa y of quinacrine was limited by toxicity. 15 In view of these Neurocirugia, Insurgentes Sur 3877, 14269 Mexico City, Mexico antimutagenic and anticarcinogenic properties, we studied Tel. +525-606-4040; Fax +525-528-0095 the effect of chloroquine and quinacrine on ENU-induced e-mail: [email protected] tumors. To inhibit glioma induction, quinacrine, the most D. Rembao active of the antimalarials, was administered in a single Experimental Pathology Laboratory, Instituto Nacional de dose to pregnant rats before E N U administration; addiNeurologla y Neurocirugia, Mexico

84

tionally, to study the effect of antimalarials on tumor progression, chloroquine, the least toxic of its congeners, was chronically administered to the progeny of ENU-exposed rats during the latency period of tumor development.

1 ~ "~

0.8

r

0.6

~.'~

~

N,,,,,

Materials and methods Pregnant Wistar rats were randomly assigned to the control group (n = 7), the quinacrine group (n = 4), or the chloroquine group (n = 3). In animals of the quinacrine group, a single dose of quinacrine, 375mg/kg, was orally administered on the 18th day of gestation. On the 19th day, rats from all groups were intravenously injected with a single dose of ethylnitrosourea, 20mg/kg. The progeny (57 from controls, 18 from the chloroquine group, and 30 from the quinacrine group) were weaned at 6 weeks of age and examined daily for signs of neurological dysfunction. Animals born from rats of the chloroquine group were treated, starting in the 7th week of age, with chloroquine, 20mg/kg orally administered three times a week; this schedule was maintained until the end of the experiment on day 400. When neurological signs compatible with tumor growth appeared, the rat was anesthetized with ether and perfused with 10% formalin in phosphate-buffered saline. For histological examination, the brain and the spinal cord were embedded in paraffin, and 5- to 7-~m sections were stained with hematoxylin-eosin. To explain the cell origin of tumors diagnosed as ependymomas, the streptavidin-biotin peroxidase complex was used for immunohistological localization of glial fibrillary acidic protein (GFAP) and synaptophysin. All sections were deparaffinized in xylene and brought into phosphate-buffered saline (PBS) using a descending ethanol series. Endogenous peroxidase activity was blocked by treatment with 0.3% H202 in methanol for 30min, and sections were incubated at room temperature with the following antisera: normal goat serum diluted 1:10 in PBS for 20min; either anti-GFAP (Super sensitive mouse anti GFAP, BioGenex, San Ramon, CA, USA) or antisynaptophysin (Super sensitive mouse anti synaptophysin, BioGenex) for 30min; biotinylated anti-mouse immunoglobulins (HK335-5M, BioGenex) for 20min; peroxidaseconjugated streptavidin (HK330-5K, BioGenex) for 20 min. Between incubations, the sections were washed with saline solution. The peroxidase reaction was visualized with 0.05% tetrahydrochloride diaminobenzidine (Sigma, St. Louis, MO, USA) and 0.01% H202 in Tris-buffered saline. Nuclear counterstaining with hematoxylin was applied after substrate reaction. Positive control staining for GFAP or synaptophysin was performed in normal rat cerebellum. Survival, tumor location, and histological characteristics according to the classification of Zook et al. ~6 were recorded. Intragroup and intergroup statistical comparisons were performed by Kruskal-Wallis analysis of variance, followed by the Mann-Whitney U test.

o +..

-\~.

\

0.4

C

0.2 0

130

e

I

I

I

160

190

220

250

I

280

i

310

~

i

340

i

370

Days Fig. 1. Kaplan-Meier plot of survival after prenatal exposure to

ethylnitrosourea (ENU); when compared with controls (circles), mean survival time was decreased in animals treated with chloroquine (squares) (p < 0.02) or with quinacrine (triangles) (p < 0.0004)

Results Central nervous system neoplasms developed in 46 (81%) control animals after a latency period of 281 (271-309) (median with 95% confidence interval), in 13 (72%) animals administered chloroquine after a latency period of 250 (195-280) days, and in 24 (80%) animals administered quinacrine after a latency period of 232 (192-253) days. Animals treated with chloroquine or with quinacrine had a significant decrease in survival when compared with controls (P < 0.0007). Cumulative survival according to the Kaplan-Meier plot (Fig. 1) showed significant differences in animals treated with chloroquine (P < 0.02) or with quinacrine (P < 0.0004) when compared with the control group; this difference originated mainly from a high incidence of tumors in the spinal cord of animals treated either with chloroquine or with quinacrine. In the control group, 64% of tumors occurred in the brain and 36% in the spinal cord, whereas in the antimalarial groups, 54% of tumors occurred in the spinal cord and 46% in the brain, although this difference was not significant (Table 1). The number of tumors per animal varied from 1 to 3 in all groups; 53 tumors were obtained from the control group, 14 from animals treated with chloroquine, and 27 from animals treated with quinacrine. The final analysis was performed on the total number of tumors rather than on the number of experimental animals. The histopathological study showed a similar frequency in type and tumor size for all groups. However, when time of tumor development and histological type were correlated, ependymoma, particularly that located in the spinal cord, was the only tumor that developed earlier in animals treated with antimalarials than in controls (P < 0.02) (Table 1). Tumors diagnosed as ependymomas revealed a densely cellular neoplasm characterized by distinctive perivascular pseudorosettes (Fig. 2A). Malignant ependymomas showed cellular atypia and intense pleomorphism (Fig. 2B). All tumors diagnosed as ependymomas

85 Table 1. Survival data of ethylnitrosourea-induced tumors according to their CNS location and histological type Variable

Control

Chloroquine

Quinacrine

Progeny No. (%) with tumor Median latency (95% CI) No. of tumors No. (%) of spinal cord tumors No. (%) of ependymomas No. (%) of spinal cord ependymomas No. (%) of brain ependymomas Brain/cord Median latency (95% CI) for animals that died with spinal cord ependymoma Median latency (95% CI) for animals that died with spinal cord other tumors

57 46 (18) 281 (271-309) 53 20 (38) 17 (32) 10 (19) 7 (13) 62/38 291 (236-436)

18 13 (72) 250 (195-282)* 14 7 (50) 6 (43) 4 (29) 3 (21) 50/50 158 (118-208)*

30 24 (80) 232 (190-253)* 27 16 (59) 12 (44) 9 (33) 3 (11) 41/59 167 (130-219)*

252 (206-326)

260 (172-293)

263 (123-319)

*P < 0.02; CI, confidenceinterval

were negative by immunohistology to synaptophysin and reacted strongly to GFAP (Fig. 2C).

Discussion Contrary to our initial hypothesis, a potential chemoprotective effect of treatment with quinacrine or chloroquine was not observed in rats prenatally exposed to ENU, given that a decrease in incidence or an increase in time of tumor development was not obtained. Instead, chloroquine and quinacrine appear to have a discrete promoter action, since a 21% reduction in the time of tumor appearance was noted (P < 0.0007). This promoting effect appears to be particularly incisive on neoplasms derived from ependymal cells in the spinal cord, because the development of other neuroectodermal tumors was not different from that in controls (Table 1). Dutta et al. 17 found that the antimalarial drugs chloroquine and quinacrine had significant stimulant effects on the growth of mammary adenocarcinoma in rats. This stimulating effect on tumor development was not related to the intracellular effects of antimalarial drugs, such as regulation of prostaglandin E2 metabolism, riboflavin deficiency, or oxidative stress; rather, it was postulated that the immunosuppressive capacity of these agents could stimulate carcinogenesis and malignant transformation of tumors. It is conceivable that antimalarial drugs might influence the growth of transplanted tumors by reducing the overall immune function of the host. However, the promoter action of antimalarials observed in the present study does not appear to be related to a lowering of the immune surveillance, because previous studies on nitrosourea-induced tumors have not shown differences in the development of neoplasms in immunosuppressed animals. ~8 Furthermore, the immunosuppresant action of long-term administration of chloroquine is mild, and quinacrine was administered during the prenatal stage in a single dose insufficient to induce

a substantial immunosuppression in the rat or in her offspring. It has been proposed that tumor-promoting substances act by inhibition of D N A repair mechanisms. In vitro chloroquine and quinacrine have shown inhibition of polymerase 119and inhibition of the process of excision repair in xerodermas exposed'to ultraviolet light. 1~According to our results, the structural perturbations caused by antimalarials as D N A intercalating agents, rather than chemoprotectors against ENU-induced abnormalities, could have turned the D N A into a poor substrate for repair. During embryonic development, the wall of the neural tube is formed by primitive neuroepithelial cells that evolve into neurons or glial cells. The embryonic cells initially affected by transplacental administration of E N U continue to proliferate and transmit the D N A damage to their progeny, which migrate and differentiate in later stages of embryonic development, so that a tumor may develop distant from the site where the primitive cells were located at the time of exposure to ENU. It is interesting to note that ependymal cells are the only remnants of the embryonic neuroepithelium that retain their original position after the neuroblasts and glioblasts have migrated to the mantle layer2~ ependymal cells are among the first to differentiate after the histogenesis of the neural tube. 2~ The precise cell target for transplacentally induced E N U carcinogenesis is not known; precursor cells for oligodendrocytes and cells of the subependymal plate have been suggested. 22-26 We observed that antimalarial drugs hastened the development of ependymomas, and it seems possible that embryonic precursors of ependymal cells are also a main target of ENU carcinogenicity. In contrast with the neuronal origin of some tumors initially diagnosed as ependymomas, which Vaquero et al. 27 have shown to express synaptophysin, in our study all cases of ependymoma were negative for synaptophysin and had intense immunoreactivity for GFAP. The difference might be due to the time of prenatal exposure to ENU. In the study of Vaquero et al. it was on day 17, whereas in our study it was on day 19; at this late

86

state of gestation it is possible that primitive neural cells are no longer susceptible to the carcinogenic effect of ENU. The intervention of antimalarial agents in the premalignant process may have enhanced the susceptibility of ependylnal cells to the E N U alkylation target, thus provoking a higher incidence of malignant ependymoma in the lower parts of the nervous system than those usually induced by E N U in animals not treated with antimalarials. Acknowledgments We thank Guadalupe Palencia, M. Sc., and Rodolfo P6rez, M.D., for technical assistance. This work was partly supported by the National Council of Science and Technology of Mexico (CONACyT) grant No L0001-M.

References

Fig. 2. Histologic study of ENU-induced tumors. A Microscopic appearance of an ependymoma of spinal cord with distinctive perivascular pseudorosettes (hematoxylin-eosin stain, • B Malignant ependymoma with cellular atypia, intense pleomorphism, and rosette formation indicated by the a r r o w (hematoxylin-eosin stain, • C Section from an ependymoma showing intense immunoreactivity to GFAP (streptavidin-biotin-peroxidase complex, •

1. Katayama K, Ishigami N, Uetsuka K, et al (2000) Ethylnitrosourea (ENU)-induced apoptosis in the rat fetal tissues. Histol Histopathol 15:707-711 2. Nolan PM, Peters J, Vizor L, et al (2000) hnplementation of a large-scale ENU mutagenesis program: towards increasing the mouse mutant resource. Mamm Genoma 11:500-506 3. Druckrey H, Peankovic S, Preussman R (1966) Teratogenic and carcinogenic effects in the offspring after single injection of ethylnitrosourea to pregnant rats. Nature 210:1378-1379 4. De Flora S, Ramel C (1988) Mechanisms of inhibitors of mutagenesis and carcinogenesis. Classification and overview. Murat Res 202:285-306 5. Lippman SM, Benner SE, Ki HW (1993) Chemoprevention. Cancer 72:984-990 6. Blake A, Peacocke A R (1968) The interaction of aminoacridines interacting with DNA of defined base sequence. A basis for the explanation of acridine bands in chromosomes. Exp Cell Res 72: 602-608 7. Lerman LS (1962) The structure of the DNA-acridine complex. Proc Natl Acad Sci 49:94-101 8. Sotelo J (1996) Could an aminoacridine interfere with the cellular mechanisms involved in the process of human immunodeficiency virus infection? Med Hypotheses 47:43-47 9. Bach MK (1969) Reduction in the frequency of mutation to resistance to cytarabine in L1210 murine leukemic cells by treatment with quinacrine hydrochloride. Cancer Res 29:18811885 10. Thielmann HW, Popanda O, Edler L (1991) The effects of inhibitors of topoisomerase II and quinacrine on ultraviolet-lightinduced D N A incision in normal and xeroderma pigmentosum fibroblasts. J Cancer Res Clin Oncol 117:19-26 11. Sevag MG, Ashton B (1964) Evolution and prevention of drug resistance. Nature 26:1323-1326 12. Giampietri A, Fioretti MC, Goldin A, et al (1980) Drug mediated antigenic changes in murine leukemia cells: antagonistic effects of quinacrine, an antimutagenic compound. J Natl Cancer Inst 64: 297-301 13. Johnson HG, Bach MK (1969) Apparent antimutagenic activity of quinacrine hydrochloride in Detroit-98 human sternal marrow cells grown in culture. Cancer Res 29:1367-1370 14. Reyes S, Herrera L, Ostrosky P, et al (2001) Quinacrine enhances carmustine therapy in experimental glioma. Neurosurgery (in press) 15. McCormick DL (1988) Anticarcinogenic activity of quinacrine in the rat mammary gland. Carcinogenesis 9:175-178 16. Zook BC, Simmens SJ, Jones RV (2000) Evaluation of ENUinduced gliomas in rats: nomenclature, immunochemistry, and malignancy. Toxicol Pathol 28:193-201 17. Dutta P, Karmali R, Pinto JT, et al (1994) Enhanced growth of mammary adenocarcinoma in rats by chloroquine and quinacrine. Cancer Lett 30:113-119 18. Denlinger RH, Swenberg JA, Koestner A, et al (1973) Differential effect of immunosuppression on the induction of nervous system

87

19.

20. 21. 22. 23.

and bladder tumors by N-methyl N-nitrosourea. J Natl Cancer Inst 50:87-93 O'Brien RL, Olenick JG, Hahn FE (1966) Reactions of quinine, chloroquine and quinacrine with DNA and their effects on the DNA and R N A polymerase reactions. Proc Natl Acad Sci USA 55:1511-1517 Burt A M (1993) Organization and development of the nervous system. In: Textbook of Neuroanatomy. W.B. Saunders, Philadelphia, pp 3-28 Brodal P (1992) The different parts of the nervous system, ln: The central nervous system: structure and function. Oxford University Press, New York, 1992, pp 54-55 Burger PC, Shibata T, Aguzzi A (1988) Selective induction by N-nitrosoethylurea of oligodendrogliomas in fetal forebrain transplants. Cancer Res 48:2871-2875 Giordana MT, Migheli A, Mocellini C, et al (1992) Immunohistochemical observation on rat radial glia: relationship with the

24. 25. 26.

27.

origin of ethylnitrosourea-induced tumors. Acta Neuropathol 84:387-393 Lantos PL, Cox DJ (1976) The origin of experimental brain tumors: a sequential study. Experientia 32:1467-1468 Jones EL, Searle CE, Smith WT (1973) Tumours of the nervous system induced in rats by the neonatal administration of N-ethylN-nitrosourea. J Pathol 109:123-139 Schiffer D, Giordana MT, Vigliani MC, et al (1991) Relationship between glial reaction to a stab wound and tumor development after receiving transplacental ethylnitrosourea in the rat. Acta Neuropathol 83:30-38 Vaquero J, Coca S, Zurita M, et al (1992) Synaptophysin expression in "ependymal tumors" induced by ethyl-nitrosourea in rats. Am J Pathol 141:1037-1041