Cancer Letters, 50 (1990) 63-70

Elsevier Scientific Publishers Ireland Ltd. 63

Oxidation of Z-amino-9-a-D-ribofurnosylpurine-6-sulfenamide to the corresponding 6-sulfonamide facilitates changes in biologic characterization that include activity against thiopurine-refractory experimental leukemia

R.A. Finch, K.M. Vasquez, N.B. Hanna, G.R. Revankar, R.K. Robins and T.L. Avery

ICN Nucleic Acid Research Institute, 3300 Hyland Avenue, Costa Mesa, CA 92626 (U.S.A.)

(Received 15 August 1989) (Revision received 31 October 1989)

(Accepted 1 November 1989)

Summary

Preclinical investigations in vivo revealed unexpected differences in the biological char- acteristics of 2-amino-9-/3-D-ribofuranosyl- purine-6-sulfenamic!e (sulfenosine, U and 2- amino-9-/3-D-ribofuranosylpurine-6-sulfonam- ide (sulfonosine, 2): two novel but structurally related derivatives of 6thioguanosine (6TGR). Strikingly, the addition of a fully oxidized sulfur atom at the 6 position of sulfenosine produced a purine derivative (sulfonosine) that was remarkably active against experimental leuke- mia resistant to treatment with either sulfenosine or 6TGR. This slight structural modification also appeared to influence solu- bility, scheduling capability, and oral activity as well as penetration of the central nervous system (CNS) and rhe onset of cellular resis- tance. These findings underscore the dramatic changes in biologic activity that can be pro- duced by subtle modifications in molecular structure. We trust they may also contribute to the development of improved clinical therapy.

Correspondence to: T.L. Avery.

Keywords: derivatives of 6-thioguanosine; experimental leukemia; chemotherapeutic characterization; resistance; blood-brain bar- rier.

Introduction

Having been introduced by Elion, Bieber and Hitchings [Z] more than 30 years ago, 6- mercaptopurine (6MP) is one of the oldest antineoplastic agents currently in clinical use. Initially shown to be beneficial in the treatment of acute leukemia in adults and children [1,3], 6MP and its Z-amino congener, 6-thioguanine (6TG) remain valuable agents for the induction and maintenance of remissions in patients with acute myelocytic (AML) and acute lymphocytic leukemia (ALL).

Unfortunately, 6MP and 6TG have well known therapeutic limitations in addition to their established clinical utility [4,7,8, lo]. Accordingly, efforts to develop novel purine derivatives with biological advantages over these thiopurine bases have continued with major emphasis being placed on the synthesis and evaluation of purine nucleosides.

0304-3835/90/$03.50 0 1990 Elsevier Scientific Publishers Ireland Ltd.

Published and Printed in Ireland

64

The first thiopurine nucleoside to reach clini- cal trial was 6-thioguanosine, GTGR, [51. Dis- appointingly, 6TGR was cross-resistant with 6MP in man and afforded no significant therapeutic advantage. Suggestive of a slight improvement in biologic capability, however, was the determination that 6TGR was more soluble than 6TG and, when given orally, was also somewhat more active than its free base. Recently, two novel nucleoside derivatives of 6TGR (Fig. 1)) sulfenosine, 1 and sulfonosine, 2 were synthesized and targeted by routine screening in mice as potential chemotherapeu- tic agents that deserved further study [9]. We now report the results of pre-clinical investiga- tions that revealed unexpected differences in the biological characteristics of these structur- ally related nucleosides. One major difference, which suggests possibilities for improved clini- cal therapy, was the unexpected efficacy of sulfonosine in the treatment of mice with experimental leukemias refractory to sulfeno- sine or GTGR. A preliminary report of our findings was presented at the 80th Annual Meeting of the American Association for Can- cer Research in San Francisco.

Materials and methods

With noted exceptions, in vivo assessments of therapeutic efficacy with murine leukemia L1210 were performed as described previously [6]. Briefly, BDF, female mice (m 18 g) , purchased from the Jackson Labora- tory, Bar Harbor, ME were inoculated i.p. on

day 0 with 1 x lo6 L1210 cells. Treatment by i.p. bolus injection (or by catheter gavage for p.o. delivery) was commenced 24 h later, and post-inoculation life span was monitored for 60 days. The 6TGR refractory subline of L1210 designated in these studies as L1210/6TGR was obtained from the NCI. It is known to be deficient in the enzyme hypoxanthine-guanine phosphoribosyl transferase (HGPRT) and the NC1 designation, L1210/6MP,6TG indicates its resistance to treatment with 6MP or 6TG.

Drugs were solubilized immediately before use and delivered in uniform volumes of 0.01

ml/g mouse weight. This method permitted the delivery of all drugs at 10 times in mg/kg their solubility in mg/ml. Exceptions to this procedure occurred only when the limited sol- ubility of a drug required the delivery of larger volumes.

The end-points by which responses to treat- ment were gauged were the incidences of drug and leukemia related deaths and the mean post-inoculation life spans of mice that died. Temporal patterns of death and observations at necropsy examination were the major criteria for assigning death to leukemia or drug toxicity.

Studies to determine drug effects on L1210 cells in the brain deviated somewhat from the above. For this purpose, BDF, mice were injected i.p. with drug or a 0.9% solution of NaCl 1 day after i.c. inoculation with 1 x lo5 L1210 cells. Approximately 24 h later, mice were killed, the brains of 7 mice/experi- mental group were combined, homogenized, and the equivalent of half a brain was injected i.p. into each of 12 untreated mice of the same strain. Mean post-inoculation life spans were then determined and viable L1210 cells/brain were estimated using concomitantly derived inoculum response data that defined the rela- tionship between life span and inoculum size

[111.

HO 0 d OH OH 6H &i

2

Sulfonosine

Fig. 1. Chemical structures of sulfenosine and sulfon- osine .

65

Table 1. Responses of L1210 inoculated BDF, mice” to various regimens of dosing with sulfenosine.

Drugb !;chedule of Dosage Post-inoculation % T/Cd administration (mg/kg/inj) life spanc

(days1

NaCl Sulfenosine

NaCl Sulfenosine

NaCl

Sulfenosine Sulfenosine Sulfenosine Sulfenosine

Sulfenosine

NaCl Sulfenosine

Sulfenosine Sulfenosine Sulfenosine Sulfenosine

NaCl Sulfenosine Sulfenosine Sulfenosine

Sulfenosine

NaCl Sulfenosine

Sulfenosine Sulfenosine Sulfenosine Sulfenosine

Sulfenosine

qd day 1 qd day 1

-

22

-

22

6.10 + 0.50 -

11.33 + 1.53 186

qd day 1 qd day 1

6.00 & 0.45 -

10.20 + 0.45 170

qd day 1

qd day 1 bid day 1 tid day 1 qd days 1,4 qd days 1,4,7

- 6.00 & 0.67 22 11.00 + 1.00 22 10.00 f 0.00 22 10.67 f 0.58 22 10.50 + 0.58

22 10.20 + 0.84

qd day 1 - 6.00 + 0.53 qd day 1 22 10.00 + 0.00 qd day 1 13 9.33 f 1.15 qd day 1 8.1 9.33 & 0.58 qd day 1 4.8 9.33 + 0.58 qd day 1 2.9 9.67 f 0.58

qd days 1,4 qd day 1 qd days 1,4 qd days 1,4,7

qd days 1,4,7

- 6.00 + 0.00 22 11.00 + 1.00 22 9.67 + 0.58 22 10.33 f 0.58

22 10.40 f 0.89

qd days l-death qd days l-3

qd days l-6 qd days l-3 qd days l-6 qd days l-9

qcl days 1 -death

- 6.40 f 0.52 22 10.00 rt 0.71

22 8.20 f 0.45 2.2 9.80 -+ 0.45 2.2 9.80 f 0.45 2.2 9.40 + 1.52

2.2 10.20 -+ 0.45

-

183 167 178

175 170

-

167

156 156 156 161

-

183 161 172

173

-

156 (Tax)’ 153 153 147

159

‘Mice were inoculated i.p. on day 0 with 1 x 106 cells of murine leukemia L1210 and, starting 24 h later, were given bolus injections of sulfenosine using a variety of treatment schedules.

bAll solutions were delivered i.p. (0.01 ml/g mouse wt). Control mice were injected with a 0.9% solution of NaCI. Twice daily treatments (bid) were made at 1000 and 1400 h. Thrice daily treatments (tid) were made at 1000, 1200 and 1400 h.

‘Mean f 1 S.D. dTreatment responses (5 mice/treatment group; 10 mice/control group) presented as %T/C were calculated according to the equation: Mean I:‘fe span of treated mice/mean life span of control mice X 100. A T/C 3 125 is considered bio- logically significant.

‘Every mouse died from drug toxicity.

66

Results

The solubilities of sulfenosine and sulfono- sine were dissimilar. As a result, the maximum dosage of sulfenosine that could be delivered as a single bolus using our volumetric scheme of dosing (see Materials and methods) was 22 mg/kg while that of sulfonosine was 62 mg/ kg. Administered qd day 1 at those dosages, neither drug was lethally toxic for LlZlO-inoc- ulated mice but, as previously reported [9], sulfenosine (%T/C 170) produced a much greater increase in mean post-inoculation life span than did sulfonosine (%T/C 128) and was thus identified as the more potent of the two compounds.

The data in Table 1 resulted from studies in which LlZlO-inoculated mice were subjected to various regimens of dosing with sulfenosine. The main objectives of these studies were to establish sulfenosine’s maximum effectiveness against L1210 and to assess the drug’s activity as a function either of dosage or schedule of administration. As gauged by mean post-inoc- ulation life span, the data that are presented provided no indication that treatment was improved by modifications in drug dosage or delivery schedule. Thus, the repeated delivery of sulfenosine on day 1, i.e., bid or tid, was no more effective than a single treatment. Similarly, successive treatments on different days, whether consecutive or intermittent, did not produce greater increases in mean post- inoculation life span than did a single day 1 treatment. From the data presented in Fig. 2, it appears that only 1 or 2 treatments with sulfen- osine were required to generate cellular resis- tance to further treatment with the drug.

The data in Table 2 reflect the results of lim- ited dosage ranging and scheduling studies that were performed with sulfonosine. Unlike the results produced by sulfenosine, these data indicate that the therapeutic effectiveness of sulfonosine was strongly related to cumulative drug dosage. From the additional information provided by Fig. 3, it appears that mice inocu- lated with experimental leukemia cells resistant to treatment with sulfenosine also exhibited

Fig. 2. Responses of BDF, mice inoculated with untreated or previously treated L1210 cells to a single treatment with sulfenosine. Mice were inoculated i.p. on day 0 with 1 x lo6 leukemia cells and, starting 24 h

later, were injected i.p. with sulfenosine (22 mg/kg/ injection) on three different schedules of treatment. Leukemia cells that survived treatment were collected and used to inoculate other mice which were then given a

single injection of sulfenosine 24 h later. Each treatment group consisted of 5 mice. The average survival time of 4 groups of untreated control mice was 6.45 + 0.51 days.

dosage related responses to treatment with sul- fonosine. A subsequent study (Fig. 4) showed sulfonosine to be effective and sulfenosine to be ineffective in the treatment of mice inocu- lated with leukemia cells refractory to treatment with 6TGR. As indicated in materials and methods, the 6TGR resistant subline of L1210 used in this study was deficient in the phosphorylating enzyme, HGPRT.

Data reflecting the responses of L1210 inoc- ulated mice to orally administered sulfenosine and sulfonosine are presented in Table 3. These data clearly indicate the inadequacy of this route of delivery for sulfonosine even when given twice a day on multiple days. Conversely, orally administered sulfenosine produced significant increases in mean post- inoculation life span and, as was observed with parenteral administration of the drug, the increases that were produced did not appear dosage related.

In another facet of our studies, sulfenosine and sulfonosine were evaluated for their

67

Table 2. Responses of L1210 inoculated BDF, mice” to various regimens of dosing with sulfenosine.

Drugb !khedule of Dosage Post-inoculation 56 T/Cd administration (mg/kg/inj) life span’

(days)

NaCl qd day 1 - 6.00 + 0.63 -

Sulfonosine qd day 1 62 7.33 + 1.53 122

NaCl qd day 1 62 6.00 f 0.67 -

Sulfonosine qd day 1 62 7.67 f 1.15 128 Sulfonosine bid day 1 62 8.33 f 0.58 139 Sulfonosine tid day 1 62 9.33 f 0.58 156

Sulfonosine qd day 1 62 6.60 + 3.78 (Tox)’

NaCl bid days 1,4,7 - 6.20 f 0.45 -

Sulfonosine bid days 1,4,7 62 10.80 2 0.45 174

Sulfonosine bid days 1,4,7 37 9.40 f 0.55 152 Sulfonosine bid days 1,4,7 22 7.40 -+ 0.55 119

“Mice were inoculated 1.p. on day 0 with 1 x 106 cells of murine leukemia L1210 and, starting 24 h later, were given bolus injections of sulfonosine using several different treatment schedules.

bAll solutions were delivered i.p. (0.01 ml/g mouse wt). Control mice were injected with a 0.9% solution of NaCI. Multi- ple treatments on the same day were made at 1000 and 1400 h (bid); 1000, 1200 and 1400 h (tid); 1000, 1200, 1400 and 1600 h (qid).

cMean f 1 S.D. dTreatment responses (5 mice/treatment group; 10 mice/control group) presented as %T/C were calculated according

to the equation, Mean life span of treated mice/mean life span of control mice x 100. A T/C 2 125 is considered biol- ogically significant.

*Two mice died from drug toxicity.

capacity to cross t:he blood-brain barrier and enter the central nervous system (CNS). After i.c. inoculation, L’l210 cells rapidly migrate throughout the body [ 111. As a result, life span data can encompass an element of bias related

la0 r-----I

Fig. 3. Responses of BDF, mice inoculated with sulfenosine-resistant L1210 cells to treatment with sulfen- osine or sulfonosine. Treatment responses (5 mice/treat- ment group) presented as %T/C were calculated

according to the equation: Mean Life Span of Treated Mice/Mean Life Span of Control Mice x 100. A %T/C of 100, as that produced by sulfenosine, indicates no activity. Conversely, the respective values of 123, 137

and 147 that were produced when sulfonosine was given once, twice or three times, reflect increases in mean post- inoculation life span of 23, 37 and 47%. These respon- ses, which were roughly equivalent to those produced by

sulfonosine against wild type L1210, clearly indicate the sulfenosine resistant cells’ lack of cross resistance of sul- fonsine. The mean post-inoculation survival time of untreated control mice was 6.00 + 0.00 days.

68

Fig. 4. Responses of BDF, mice inoculated with 6TGR-resistant L1210 cells to treatment with 6TGR, sul- fenosine, or suifonosine. Each group consisted of 5 mice. The average survival time of untreated control mice was

6.40 f 0.55 days.

to the killing of cells outside the brain. We sought to assess the effects of treatment more directly by estimating the numbers of viable leukemia cells that remained in the brains of treated and untreated mice ~24 h after i.p. drug administration. As shown in Fig. 5, sul- fenosine penetrated the CNS and killed ~57% of L1210 cells in the brain when given once i.p. at 22 mg/kg. Under similar conditions, sulfonosine was not effective when given at a dosage of 62 mg/kg but pro- duced a cell kill of -64% when its delivery volume was increased to provide a dosage of 248 mg/kg.

Discussion

In this report, we have presented the results of pre-clinical investigations that were per- formed in vivo to provide chemotherapeutic

Table 3. Responses of L1210 inoculated BDF, mice” to oral dosing with sulfenosine and sulfenosine.

Drugb Schedule of Dosage Post-inoculation % T/Cd administration @ng/kg/inj) life span’

(days)

NaCl qd day 1 - 6.30 + 0.48 -

Sulfenosine qd day 1 22 9.80 + 0.45 153

Sulfenosine qd day 1 13 9.40 * 0.55 147

Sulfenosine bid day 1 8 9.40 -t 0.55 147

Sulfenosine bid day 1 22 9.80 f 0.45 153

NaCl bid days 1,4,7 - 6.40 + 0.55 -

Sulfonosine bid days 1,4,7 62 6.40 + 0.55 100

Sulfonosine bid days 1,4,7 37 6.60 * 0.55 103

Sulfonosine bid days 1,4,7 22 6.60 -c 0.55 103

“Mice were inoculated i.p. on day 0 with 1 X lo6 cells of murine leukemia L1210 and, starting ~24 h later, were treated with sulfenosine or sulfonosine by catheter gavage.

bAll solutions were delivered i.p. (0.01 ml/g mouse wt). Control mice were injected with a 0.9% solution of NaCl. Twice daily treatments (bid) were made at 1000 and 1400 h.

cMean -f 1 S.D. dTreatment responses (5 mice/treatment group; 10 mice/control group) presented as %T/C were calculated according

to the equation: Mean life span of treated mice/mean life span of control mice X 100. A T/C 2 125 is considered biologically significant.

J

Fig. 5. Residual populations of viable L1210 cells in

the brains of BDF, mice 24 h after i.p. treatment with sul- fenosine or sulfonosine. Control brains contained 2,083,950 leukemia cells at the time of bioassay.

characterizations of two novel nucleoside derivatives of GTGR. The nucleoside analog sulfonosine resulted from oxidation of the sul- fur atom at the C-6 position of its close conge- ner sulfenosine. Our studies indicate that this slight structural mod:ification not only increased solubility and diminished cytotoxic potency but, as shown in Table 4, essentially reversed several other biological characteristics. Sulfonosine, in sharp contrast to sulfenosine, was responsive to changes in dosage and scheduling, lacked oral activity, and was active against experimental leukemia cells refractory to treatment with 6TGR. Although the solubil- ity of sulfonosine was low relative to the drug’s

69

cytotoxic potency, it was possible to improve therapeutic performance through scheduling and also through the use of increased drug volumes.

Two clinically relevant findings in these studies were the divergent dosing characteris- tics of sulfenosine and sulfonosine and the pronounced activity of sulfonosine against experimental leukemias resistant to treatment with sulfenosine or 6TGR. The development of resistance to drugs used in standard regi- mens of therapy, e.g., resistance to 6MP when used as post-induction continuation therapy for ALL patients, is universally acknowledged as a major impediment to the effective treat- ment of cancer. The rapid onset of resistance to sulfenosine described here is indicative of that problem and the responsiveness of sulfenosine-resistant experimental leukemia to treatment with sulfonosine supports the view that drugs can be designed to overcome the problem. Chemotherapeutic agents are com- monly titrated to maximum effectiveness by delivering increasing dosages until a limiting host toxicity has been established. With 6MP however, clinical as well as preclinical trials have produced results indicating either a lack of relation between dosage and elicited response or an actual decrease in effectiveness with increasing dosage [4,7,8]. In the present studies, sulfenosine exhibited a similar limita- tion of dosage related efficacy while the thera- peutic performance of sulfonosine clearly improved with increasing dosage.

As indicated in the Materials and methods

Table 4. Divergent therapeutic characteristics of sulfenosine and sulfonosine.

Dosage Scheduling Penetration Oral Activity vs.

rar’ging capability of CNS activity L1210/6TGR

response

Sulfenosine + + Sulfonosine + + -a +

‘Sulfonosine was found I:O penetrate the CNS when an increased volume was administered to overcome its lack of potency and its limited solubility.

70

section, the 6TGR-resistant strain of L1210 used in these studies also resists treatment with 6MP and 6TG. Sulfonosine’s activity against this thiopurine resistant experimental leukemia and the drug’s dosing behavior, whereby larger dosages elicited increased responsiveness, support the possibility that this novel nucleo- side derivative of 6TGR might prove effective in the treatment of human leukemia refractory to 6MP and 6TG.

References

Burchenal, J.H., Murphy, M.L., Ellison, R.R., Sykes,

M.P., Tan, C.T.C., Leone, L.A., Kamofsky, D.A.,

Craver, L.F., Dargeon, H.W. and Rhoads, C.P. (1953) Clinical evaluation of a new antimetabolite, 6-Mercapto-

purine, in the treatment of leukemia and allied diseases.

Blood, 8,965-999.

Elion, G.B., Bieber, S. and Hitchings, G.H. (1954) The

fate of 6-mercaptopurine in mice. Ann. N.Y. Acad. Sci.,

60,297-302.

Ellison, R.R. and Burchenal, J.H. (1960) The treatment of

chronic granulocytic leukemia with the 6-substituted pur-

ines, 6-Mercaptopurine, Thioguanine, and 6-chloropurine.

Clin. Pharmacol. Ther.. 1,631-644.

He&r, S., Lennard, L., Lilleyman, J.S. and Maddocks,

J. (1982) 6-Mercaptopurtne: Apparent lack of relation

between prescribed dose and biological effect in children

with leukaemia. Br. J. Cancer, 46, 138-143.

5

6

7

8

9

10

11

K&off, I.H., Ellison, R.R. and Tan, C.T.C. (1961) Clini-

cal evaluation of thioguanosine. Cancer Res., 21, 1015-

1019.

Kuehl, M., Brixner, D.I., Broom, A.D., Avery, T.L. and

Blakley, R.L. (1988) Cytotoxicity, uptake, polyglutamate

formation, and antileukemic effects of b-deaza analogues

of methotrexate and aminoptertn in mice. Cancer Res.,

48,1481-1488.

Liliemark, J.O. Masquelier, M. and Peterson, C.O. (1988)

Paradoxical increase of formation of phosphoylated 6-

thioguanosine metabolites in leukemic cells in vitro with

decreasing dose of 6-mercaptopurtne. Proc. Am. Assoc.

Cancer Res. (Abstr. 1391) 79,350.

Matsumura, S., Hoshino, T., Wetzsaecker, M. and Deen,

D.F. (1983) Paradoxical behaviour of 6-mercaptopurine as

a cytotoxic agent: Decreasing cell kill with increasing drug

dose. Cancer Treat. Rep., 67,475-479.

Revankar, G.R., Hanna, N.B., Imamura, N., Lewis, A.F.,

Larson, S.B., Finch, R.A., Avery, T.L. and Robins, R.K.

(1989) Synthesis and in vivo antitumor activity of 2-ami-

nopurtne-6-sulfenamide, sulftnamide, and sulfonamide

and related purine ribonucleosides. J. Med. Chem., in

press.

Simone, J.V. The treatment of acute lymphoblasttc

leukemia. (1980) Br. J. Haematol., 45, l-4.

Skipper, H.E., Schabel. F.M., Wilcox, W.S., Laster,

W.R., Trader, M.W. and Thompson, S.A. (1965) Experi-

mental evaluatton of potential anticancer agents. XVIII.

Effects of therapy on viability and rate of proliferation of

leukemic cells in various anatomic sites. Cancer Chem-

other. Rep., 47.41-64.