THE TOXICITY

OF ADENOSINE ANALOGUES BO v1s iN V1TRO ELIZABETH A. KERR

AGAINST

BABE,%4

and ANNETTE M. GERO*

School of Biochemistry, University of New South Wales, P. 0. Box 1, Kensington, New South Wales 2033, Australia (Received 12 cocci

I 99 1; accepted 7 June 1991)

E. A. and GEROA. M. 1991. The toxicity of adenosine analogues against Babes& bovis in vitro. Inrernationnl Journalfor Parasitology 21: 747-7.51. The toxicities of 20 analogues of deoxyadenosine or adenosine were tested in vitro against the intraerythrocytic parasite Babe&z bovis. XC,, values (the con~ntration of compound required to reduce cell survival to 37%) were determined for each compound. Tubercidin (7-deaza-adenosine), Z-bromo-adenosine, 8-bromo-3-ribosyl adenine and 6-phenylaminodeoxyadenosine were shown to be the most toxic towards B. bovis. Comparison of the toxicity results for these compounds in B. bovis with those in human melanoma cell lines indicated a differential toxicity, in that many of the compounds were toxic towards II. bovis but were relatively non-toxic towards human melanoma cell lines and vice versa. These results suggest that the mechanism of toxicity of the deoxyadenosine and adenosine analogues, whose normal metabolism involves transport, metabolism and incorporation into nucleic acids, may vary sjgni~cantly between B. bovis and mammalian cells, allowing such drugs to be considered for parasite chemotherapy. Abstract-KERR

INDEX KEY WORDS: Anti-babesial agents; Eabesiu bovis; adenosine analogues; purine nucleosides; sangivamycin; tubercidin.

THE prevention and control of babesiosis in Australian cattle are mainly dependent on chemotherapeutic and immunological techniques (Kuttler, 1988; Ristic & Montenegro-James, 1988; Mahoney & Wright, 1976). However, research for more effective antibabesial compounds is still being actively pursued. A logical approach to the control of this disease is to develop agents which will specifically inhibit the growth and reproduction of Babesia bovis in its host cell, the bovine erythrocyte. The ideal treatment would specifically target the Babes&infected cell and leave the normal host tissues unaffected, thus resulting in few side effects. One way to achieve this is to exploit the presence of parasite-specific metabolic pathways which are only present in the infected cell, or to utiiize the presence of transport systems in infected cells which are different from those present in normal cells. Like many other parasitic protozoa (Sherman, 1984; Hammond & Gutteridge, 1984), Babesia species are incapable of purine biosynthesis de now, and are thus dependent upon the salvage of purine bases and nucleosides from the extracellular medium for the synthesis of their nucleotides and nucleic acids (Irvin & Young, 1977, 1979; Irvin, Young & Purnell, 1978). Normal bovine erythrocytes are different from other mammalian cells in that they lack a functional nucleoside transporter so that they cannot salvage

nucleosides (Young & Jarvis, 1983; Gero, 1989). However, Babesia has been shown to insert its own nucleoside transport systems into the host erythrocytic membrane, allowing the salvage of nucleosides (Conrad, 1986; Gero, 1989), and to introduce new metabolic pathways into these erythrocytes (Hassan, Phillips & Coombs, 1987; Matias, Nott, Bagnara, O’Sullivan & Gero, 1990) In an attempt to exploit these differences between the host tissues and the parasite, analogues of purine nucleosides have been investigated for their potential antibabesial ability. In particular, deoxyadenosine analogues were considered to be of interest as potential antibabesial compounds because of their recognized antiviral and antitumour activities (Parsons, Bowman & Blakley, 1986). The structurefunction relationships of these compounds against Babesia in culture were compared with the published results of the toxicities of these compounds in human melanoma cell lines (Parsons et al., 1986). B. bovis (Samford and Ka isolates) were obtained from Dr LG. Wright (CSIRO, Division of Tropical Animal Production, Long Pocket Laboratories, Indooroopilly, Queensland), and were maintained in culture in vitro by a modified version of the techniques of Levy and Ristic (1980). Stock, asynchronous cultures contained a 10% (v/v) suspension of erythrocytes (O.l-2% parasitized cells) in RPM1 1640 medium supplemented with 30 FM-HEPES-KOH (pH 7.2) and 40% bovine serum. Cultures were maintained at 37°C in a gas mixture of 5% COZ, 5%0, and 90% N,. 747

E. A.

748 TABLE I-IC,,

KERR

and A. M.

GERO

v.~ues FORADENOS~YE ANALOGUES AGAINSTB. bovis

Adenosine

HUMANMELANOMA CELLLINES

2-Bromo-deoxyadenosine

Sangivdmycin

Compound

fc,, (W) B. bovis

Adenosine 2’-Deoxyadenosine Tubercidin (Fdeaza-adenosine) Sangivamycin (7-deaza-amido-adenosine) S-Aza-2-amino-deoxyadenosine 8-Aza-7-deaza-deoxyadenosine 8Bromo-adenosine 8-Bromo-deoxyadenosine 8-Bromo-3-ribosyi adenine X-Bromo-3-(2’deoxyribosyl)adenine 8-Trifluoromethyl-3-(2’-deoxyribosyl)adenine 2-bromo-adenosine 2-Bromo-deoxyadenosine 2-Chloro-deoxyadenosine 2-Fluoro-deoxyadenosine 2-Methylthio-deoxyadenosine 2-Hydroxy-deoxyadenosine 2-Amino-deoxyadenosine 2-Trifluoromethyl-deoxyadenosine 6-Phenyian~ino-deoxyadenosin~ 6-Methylamino-deoxyadenosine h-Dimethylamino-deoxyadenosine

AND

> 100 > 100 0.13 (0.07:) > 100(60$) > 100 > 100 60 45 I2 > 100 > 100 7.5 60 >I00 > 100 20 >I00 >I00 >iOO 15+ z-100 >lOO

Melanoma? 450 250 0.023 0.37 0.30 > 20(0527) >20 >20 >20 >20 0.18 0.022 0.1 I

>20 >20 >20(134) >20 >20 >20 > 20(Y)

Average IC,, values for each test compound against the Samford or Ka (*) isolates of B. hovi.c. Values are the mean of triplicate wells of each dose used over a range of five or more concentrations. The IC,? value is the ~on~ent~dtion at which the compound reduces cell survival to 37%. The toxicities of the same adenosine analogues in human tumour cells are included for comparison. + Determined by Parsons er (II. (1986) and Parsons & Hayward (1986) for the MM96 human melanoma. unless otherwise indicated. flC,,, values (concentration of compound causing death of 50% ofcells). $ IC,,, value determined by Smith, Lummis & Grady (1959) in KB strain of human carcinoma cells. cIC,,, values determined in Sarcoma I80 cells (Glazer and Hartman, 1981) or CCRF-CEM cells (Huang, Hatfield, Roetker.

Montgomery & Blakley. I981--in brackets).

Parasitic growth for each experiment performed in vitro was monitored by measuring the incorporation of [G-~Hlhypoxanthine into parasitic nucleic acids (Desjardins, Canfield. Haynes & Chulay, 1979). Microculture plates were prepared according to the technique of Nott, O’Sullivan, Gero & Bagnara (1990) using asynchronous parasitized cultures of B. hovis (10% haematocrit, 1% parasitized cells). Each plate contained a range of concentrations (up to 100 p) of

each compound being tested. After incubation for 24 h, [G-‘Hlhypoxanthine (3.7 kBq) was added to each well, and the plates were incubated for a further 18~20 h. The cells were harvested on Whatman GFjC glass hbre filters under vacuum to collect nucleic acids. Filters were counted for radioactivity in 0.4% w;v PPO, 0.01% wiv POPOP in toluene in a Packard TriCarb Liquid Scintillation Spectrometer. The IC,, reduces which of drug values ~concentration

Research Note

149

n B. BOVIS q MELANOMA

\‘.‘\ \‘\‘\ \‘\‘\ \‘\‘\ .‘\‘. \‘\‘\ \‘\‘\ .‘.I. \‘.‘\ \‘\‘\ \‘,‘\ .‘\‘. \‘\‘\ \‘\‘\ \‘\‘\ \‘\‘\ \‘\‘\ \‘\‘\ \‘\‘\ \‘\‘\ \‘\‘\ \‘\‘. \‘\I. \‘\‘\ \‘\‘\ .‘.‘\ \‘.I\ \‘\‘\ \‘\‘\

Tub

2BrA

6PdA

PMSdA

6BrdA

COMPOUND

\\\

_I PBrdA

~

PCldA

\‘\‘\ \‘\‘\ \‘\‘\ \‘\‘\ .I.‘. \‘\‘\ \‘\‘\ \‘\‘\ \‘.‘\ \‘\‘\ \‘\‘\ \‘.‘\ \‘\‘\ \‘\‘\ X’S’\ \‘\‘\ \‘\‘\ \‘.‘\ \‘.‘\ \‘\‘\ \‘\‘\ \‘\‘. \‘\‘\ \‘,‘\ .‘\‘\ \‘\‘\

6azadA

2FdA

Fm. 1.Comparative bar graphs of the relative toxicities of several adenosine analogues in B. bovis and human carcinoma or melanoma cells. The Toxicity Index is the inverse of the XC,, value, as tabulated in Table 1, except for tubercidin and sangivamycin which are IC,, values. Abbreviations are as follows: Tub, tubercidin; Sang, sangivamycin; 2BrA, 2-bromoadenosine; 6PdA. 6-phenylamino-deoxyadenosine; 2MSdA, 2-methylthio-deoxyadenosine; 8BrdA, 8-bromo-deoxyadenosine; 2BrdA. 2-bromo-deoxyadenosine; 2CIdA. 2-chloro-deoxyadenosine; 8azadA, 8-aza-2-amino-deoxyadenosine; 2FA, 2fluoro-deoxyadenosine. [‘Hlhypoxanthine incorporation to 37% of the control) were determined for each drug tested. The results were confirmed by microscopic analysis of Giemsa-stained blood smears. All synthetic deoxyadenosine and adenosine analogues were synthesized by Dr R. L. Blakley (St Judes’ Children’s Hospital, Memphis, TN, U.S.A.) and supplied by Dr P. Parsons (Queensland Institute of Medical Research, Brisbane, Australia), except for 8-bromo-adenosine, 8-bromo-3-ribosyl-adenine and sangivamycin (7-deaza-7-amido-adenosine), which were supplied by the National Cancer Institute (Bethesda, MD, U.S.A.). Tubercidin (7-deazaadenosine) was purchased from Sigma Chemical Co. (St Louis, MO, U.S.A.). Stock solutions of the test compounds were prepared in single strength RPM1 1640 medium. The results of drug screening in vitro have been summarized in Table 1. Tubercidin was the most toxic compound towards B. bovis with an IC,, value of 0.13 PM. However, the three other compounds with altered aza groups in the purine ring, sangivamycin, 8-aza-7deaza-deoxyadenosine and 8-aza-2-amino-deoxyadenosine, were not toxic at concentrations of up to 100 PM.

The substituted adenosine analogue, 2-bromoadenosine, demonstrated toxicity towards B. bovis in vitro, being approximately eight-fold more toxic than 8-bromo-adenosine. Substitution of the ribose of 2bromo-adenosine with deoxyribose (2-bromo-deoxyadenosine) decreased the toxicity. Conversely, the 8bromo-deoxyadenosine analogue showed slightly greater potency when compared with the corresponding 8-bromo-adenosine. When the sugar moiety was linked via the Ni of the purine ring (i.e. 8bromo-3-ribosyl-adenine), the toxicity compared with 8-bromo-adenosine was increased almost five-fold. However, in the case of 8-bromo-deoxyadenosine, attachment of the sugar via the Ni position (i.e. 8bromo-3-(2’-deoxyribosyl)adenine) rendered the compound non-toxic towards Babesiu. Another N3linked sugar compound, 8-trifluoromethyl-3-(2’deoxyribosyl)adenine, was also not toxic towards B. bovis at 100 PM. These data may indicate a complex mode of action of these analogues in this organism. One further compound, 6-phenylamino-deoxyadenosine, was potent, with an IC,, value of 15 PM. 2Chloro-, 2-fluoro, 2-hydroxyand 2-amino-deoxyadenosine were not toxic at the concentrations tested (Up

t0

100

PM).

750

E. A. KERR and A. M. GERO

of these compounds towards B. with the effect on melanoma cell lines, it was clear that there was a differential activity, in that those derivatives which were the most toxic towards B. bovis, most notably 2-bromo-adenosine, were not significantly toxic towards melanoma cells. Conversely, some compounds, such as the halogen0 analogues, which were not active against B. bovis were very active against melanoma cells. A diagrammatic summary of the salient features of these results is shown in Fig. 1. The different toxicities suggest that these analogues act markedly differently in B. bovis from the way they do in human melanoma cells. The toxicity in vitro of halogeno-deoxyadenosine compounds has been widely reported in several human cell types. 2-Chloro- and 2-fluoro-deoxyadenosine exhibit cytoxic effects in nanomolar concentrations against malignant lymphoblastoid cell lines, and are toxic to normal human T-lymphocytes in higher concentrations with longer exposure times (Carson, Wasson, Kaye, Ulhnan, Martin, Robins&Montgomery, 1980; Carson, Wasson, Taetle & Yu, 1983). The 2fluoro-, 2-chloro-, 2-bromo- and 8-aza-2-amino-deoxyadenosine derivatives demonstrate toxicity towards human melanoma cell lines at concentrations less than 1,UM(Parsons et al., 1986). Yet these same agents were not toxic to B. bovis except for 2-bromo-deoxyadenosine which had an IC,, value of 60 PM. The toxicity of 2-bromo-deoxyadenosine in tumour cells is associated with inhibition of DNA synthesis and DNA fragmentation. In human cell lines, halogenated deoxyadenosine derivatives are not substrates for adenosine deaminase. In lymphocytes, these nucleosides are phosphorylated by deoxycytidine kinase, and are incorporated into DNA (Carson et al., 1980). Inhibition of DNA synthesis may be due to inhibition of ribonucleotide reductase by the accumulated analogue nucleoside triphosphates or by its incorporation into DNA (Hentosh, Koob & Blakley, 1990). The fact that these compounds have little toxicity in B. bovis suggests that there are significant differences in either the transport of these compounds into B. bovisinfected cells and/or the metabolism of such compounds in B. bovis compared with human cell lines. Thus, the greater toxicity of the adenosine analogues in B. bovis compared to deoxy analogues might suggest the lack of suitable kinases in B. bovis for the metabolism of deoxy compounds, or a resistance to S-triphosphate inhibition by the babesial ribonucleotide reductase. Toxicity of the adenosine nucleosides is most likely due to direct inference with the pathways of nucleotide metabolism and is the subject of further investigation. Such analogues of natural metabolites and of cytotoxic agents can be useful aids in defining the features of a compound that are important for toxicity. In this case, the effects produced by modification to the fundamental nucleoside structure may aid in the further design of synthetic analogues with greater babesicidal activity. When the potency

bovis was compared

Acknowledgements-We wish to thank Dr R. L. Blakley, St Judes’ Children’s Hospital, Memphis, TN, U.S.A. and Dr Peter Parsons, Queensland Institute of Medical Research, Brisbane, Australia for the adenosine analogues, Dr Ian Wright. CSIRO Division of Tropical Animal Production. Long Pocket Laboratories, Indodroopilly, Queensland, for supplies of B. hovis-infected blood and Dr Aldo Bagnara and Sue Nott for helpful advice. This work was supported by grants from the -National Health and Medical -Research Council of Australia and the UNDPiWorld Bank/WHO Special Programme for Research and Training in Tropical Diseases. REFERENCES CARSOND. A., WA~SSOND. B., KAYE J., ULLMAN B., MARTIN D. W., JR, ROBINS R. K. & MONTGOMERYJ. A. 1980. Deoxycytidine kinase-mediated toxicity of deoxyadenosine analogs toward malignant human lymphoblasts in vilro and toward murine Ll210 leukaemia in viva. Proceedings of the National Academy of Sciences of the United States of America 77: 6865-6869. CARSON D. A., WA~SSON D. B., TAETLE R. & Yu A. 1983. Specific toxicity of 2-chlorodeoxyadenosine toward resting and proliferating human lymphocytes. Blood 62: 737-743: CONRAD P. A. 1986. Uptake of tritiated nucleic acid precursors by Babesia bovis in vitro. International Journal f?w Parasitology 16:263-268. DESJARDINSR. E., CANPIELDC. J., HAYNESJ. D. & CHULAYJ. D. 1979. Quantitative assessment of anti-malarial activity by a semiautomated microdilution technique. Antimicrobial Agents and Chemotherapy 16: 710-718. GERO A. M. 1989. Induction of nucleoside transport sites into the host cell membrane of Babesia bovis-infected erythrocytes. Molecular and Biochemical Parasitology 35: 269-276. GLAZER R. I. & HARTMAN K. D. 1981. Cytokinetic and biochemical effects of sangivamycin in human colon carcinoma cells in culture. Molecular Pharmacology 20: 657-66 1. HAMMOND D. J. & GUTTER~DGEW. E. 1984. Purine and pyrimidine metabolism in the Trypanosomatidae. Molecular and Biochemical Parasitologv 13: 243-261. HASSAN H. F., PHILLIPSR. S. & CO&BS G. H. 1987. Purinemetabolising enzymes in Babesia divergens. Parasitology Research 73: 121L125. HENTOSH P., Koos R. & BLAKLEYR. L. 1990. Incorporation of 2-halogeno-2’-deoxyadenosine 5-triphosphates into DNA during replication by human polymerasks alpha and beta. Journal of Biolopical Chemistrv 265: 40334040. HUANG M.-C., HATFIELD K., ROE*& A. W., MONTGOMERY J. A. & BLAKLEYR. L. 1981. Analogs of 2’-deoxyadenosine: facile enzymatic preparation and growth inhibitory effects on human cell lines. Biochemical Pharmacology 30: 2663-2671. IRVIN A. D. & YOUNG E. R. 1977. Possible in vitro test for screening drugs for activity against Babesia and other blood protozoa. Nature (London) 269: 407409. IRVINA. D., YOUNG E. R. & PURNELLR. E. 1978. The uptake of tritiated nucleic acid precursors by Babesia species of cattle and mice. international Journal for Parasitology 8: 19-24. IRVIN A. D. & YOUNG E. R. 1979. Further studies on the uptake of tritiated nucleic acid precursors by Bahesia species of cattle and mice. International Journal for Parasitology

KUTTLER K.

9: 109-l 14.

L. 1988. Chemotherapy of babesiosis. In: Babesiosis of Domestic Animals and Man (Edited by RIS~[~ M.), pp. 227-244. CRC Press, Boca Raton, FL.

Research LEVY M. G. & RIS-~CM. 1980. Babesia bovis: continuous cultivation in a microaerophilous stationary phase culture. Science 207: 1218-1220. MAHONEY D. F. & WRIGHT I. G. 1976. Babesia argeniina: immunization of cattle with a killed antigen against infection with a heterogeneous strain. Veterinary Parasitology 2: 273-28 1. MAT~ASC., Norm S. E., BAGNARAA. S., O’SULLIVAN W.J. & GERO A. M. 1990. Purine salvage and metabolism in Babesia bovis. Parasitology Research 76: 207-213. NO~TS. E., O’SULLIVANW. J., GERO A. M. & BAGNARAA. S. 1990. Routine screening for potential babesicides using cultures of Babesia bovis. International Journal for Parasifolony 20: 797-802. PARSON;~. G., BOWMAN E. P. W. & BIAKLEYR. L. 1986. Selective toxicity of deoxyadenosine analogues in human melanoma cell lines. Biochemical Pharmacology 35: 40254029.

Note

751

PARSONSP. G. &HAYWARD I. P. 1986. Human melanoma cells sensitive to deoxyadenosine and deoxyinosine. Biochemical Pharmacology 35: 655-660. RISTIC M. & MONTENEGRO-JAMESS. 1988. Immunization against Babesia. In: Babesiosis of Domestic Animals & Man (Edited by RISTIC M.), pp. 163-189. CRC Press, Boca Raton, FL. SHERMAN I. W 1984. Metabolism. In: Handbook of Experimental Pharmacology, Vol. 68 (Edited by PETERS W. and RICHARDS W. H. G.), pp. 31-80. Springer, Berlin. SMITHC. G., LUMMISW. L. &GRADY J. E. 1959. An improved tissue culture assay. II. Cytotoxicity studies with antibiotics, chemicals and solvents. Cancer Research 19: 847-852. YOUNGJ. D. & JARVIS S. M. 1983. Nucleoside transport in animal cells. Bioscience Report 3: 309-322.