J.Med. Chem. 1989,32, 852-856
852
with c = 29000 at 300 nm. The results are expressed in Table
v.
Acknowledgment. This work was supported by grants Refrom the Cancer Research Campaign and Baker for determining the search Council. We thank mass spectra. J. Hawkes and F. Gall of King's College kindly provided NMR spectra under the auspices of the University of London Intercollegiate Research Service. Donna Nichols expertly typed the manuscript. We thank
Dr. D. A. Matthews for permission to reproduce the drawing in Chart 11. Registry No. 2,2941-78-8; 3, 19181-53-4; 4, 106585-52-8; 5, 106585-53-9; 6a, 13726-52-8; 6b,2378-95-2; 6c, 70280-71-6; 6d, 76858-71-4; 6e,76858-72-5; 7a,106585-62-0;7b,106585-59-5;7c, 106585-60-8;7 4 106585-61-9;7e,106585-54-0; 8a, 106585-70-0; 8b,106585-67-5;&, 106585-133-6;8 4 106585-69-7;&, 106585-65-3; TS, 9031-61-2; DHFR, 9002-03-3; HCONH,, 75-12-7; tBuCOOCH2C1, 18997-19-8; MeI, 74-88-4; EtI, 75-03-6.
Comparison of the Biological Effects of Selected 5,8-Dideazafolate Analogues with Their 2-Desamino Counterparts' John B. Hynes,*pt Shirish A. Patil,? Robert L. Hagan,t Aimee Cole,* William Kohler,i and James H. Freisheimt Department of Pharmaceutical Sciences, Medical University of South Carolina, Charleston, South Carolina 29425, and Department of Biochemistry, Medical College of Ohio, Toledo, Ohio 43699. Received August 12, 1988
Three new 5,8-dideaza analogues of folic acid devoid of an amino group at position 2 have been prepared by using acid, 2b, 2synthetic routes patterned after earlier methodologies. They were 2-desamino-5,8-dideaaisofolic desamino-lO-thia-5,8-dideazafolicacid, 2c,and 2-desamino-lO-oxa-5,8-dideazafolic acid, 2d. These compounds were found to be 4-6-fold more cytoxic toward L1210 leukemia cells than their 2-NHzcounterparts and to be poor inhibitom of mammalian thymidylate synthase. However, they were only 1.5-3-fold less inhibitory toward dihydrofolate reductase than the analogous compounds containing a 2-NHz group. The known thymidylate synthase inhibitors 2desamino-l0-propargyl-5,&dideazafolicacid and lO-propargyl-5,8-dideazafolicacid were included in this study for purposes of comparison. Efforts to discover a folate analogue having potent inhibitory activity toward mammalian thymidylate synthase, TS, culminated in the development of lO-propargyl-5,8dideazafolic acid, la, which became referred to as CB3717.2 On the basis of its high degree of efficacy against L1210 leukemia in mice, CB3717 was introduced into clinical trials.3 However, renal toxicity, apparently resulting from poor aqueous solubility, caused the cessation of human ~ t u d i e s . ~Subsequently, it was found that the removal of the 2-amino group of la and related compounds resulted in improved solubility and cytotoxicity. For example, 2-desamino-l0-propargyl-5,8-dideazafolic acid, 2a, was found to be approximately 10-fold more cytotoxic toward L1210 leukemia cells in culture but 8-fold less inhibitory toward TS from L1210 leukemia cells than la.6p6 0
COOH
l a : Y = CH2NCH2CECH) b: Y = NHCH2 c: Y = CHZS d: Y=CH2O
The folate analogue 5,8-dideazaisofolicacid, l b (IAHQ), has demonstrated activity against the colon tumor 38 in mice, the CX-1 human colon tumor implant in the nude m o w , a human osteogenic sarcoma xenograft in hamsters, as well as a variety of established human colon cell lines in culture.'-'l However, high doses of IAHQ were required for activity in animal models, apparently due to its slow rate of influx into target cells as recently demonstrated with use of the human colon adenocarcinoma cell line HCT-8." Medical University of South Carolina.
* Medical College of Ohio.
In spite of the fact that IAHQ has greatly reduced affinity for TS as compared to CB 3717, it appeared to exert its antitumor action by inhibition of TS, since its cytotoxicity toward HCT-8 cells was largely reversed by thymidine.7 I t was of interest, therefore, to prepare the 2desamino analogue of IAHQ, 2b, in an effort to achieve enhanced antitumor efficacy. Also, prepared were 2desamino-lO-thia-5,8-dideazafolicacid, 2c, and 2desamino-lO-oxa-5,8-dideazafolic acid, 2d, in order to test the generality of cytotoxicity enhancement caused by removal of the 2-NH2 group. The parent compound of the former, IC, was shown to be a reasonably effective inhibitor This paper has been presented in part. See: Proc. Am. Assn. Cancer Res. 1988,29, 281. Jones, T. R.; Calvert, A. H.; Jackman, A. L.; Brown, S. J.; Jones, M.; Harrap, K. R. Eur. J. Cancer 1981, 17, 11. Calvert, A. H.; Alison, D. L.; Harland, S. J.; Robinson, B. A.; Jackman, A. L.; Jones, T. R.; Newell, D. R.; Siddik, Z. H.; Wiltahaw, E.; McElwain, T. J.; Smith, 1. E.; Harrap, K. R. J. Clin. Oncol. 1986, 4, 1245. Jackman, A. L.; Jones, T. R.; Calvert, A. H. In Experimental and Clinical Progress in Cancer Chemotherapy; Muggia, F. M., Ed.; Martinus Nijhoffi Boston, 1985; pp 155-210. Jones, T. R.; Thornton, T. J.; Flinn, A.; Jackman, A. L.; Newell, D. R.; Calvert, A. H. J. Med. Chem. 1989, 32, 847. Jackman, A. L.; Newell, D. R.; Taylor, G. A.; O'Connor, B.; Hughes, L. R.; Calvert, A. H. Proc. Am. Assn. Cancer Res. 1987, 28, 271. Fernandes, D. J.; Bertino,J. R.; Hynes, J. B. Cancer Res. 1983, 43, 1117. Hynes, J. B.; Smith, A. B.; Gale, G. R. Cancer Chemother. Pharmacol. 1986, 18, 231. Tsang, K.-Y.; Hynes, J. B.; Fudenberg, H. H. Chemotherapy 1982,28, 276. Hynes, J. B.; Yang, Y. C. S.; McGill, J. E.; Harmon, S. J.; Washtein, W. L. J. Med. Chem. 1984, 27, 232. McGuire, J. J.; Sobrero, A. F.; Hynes, J. B.; Bertino, J. R. Cancer Res. 1987, 47, 5975. Sobrero, A. F.; McGuire, J. J.; Bertino, J. R. Biochem. Pharmacol. 1988, 37, 997.
0022-2623/89/1832-0852$01.50/00 1989 American Chemical Society
Journal of Medicinal Chemistry, 1989, Vol. 32, No. 4 853
5,8-Dideazafolate Analogues 0
OH
COOH
Scheme I
2a:
Y = CH2NCH2CECH) b: Y = NHCH2 C : Y = CH2S d : Y = CH20
of mammalian TS,13 while lO-oxa-5,8-dideazafolic acid, Id, was found to be the most effective inhibitor of avian 5aminoimidazole-4-carboxamideribonucleotide transformylase (AICAR Tfase) of a series of 5,8-dideazafolates tested.14
H
OH
Chemistry The preparation of 2-desamino-5,8-dideazaisofolic acid (2b) was facilitated by the observation that the reaction of 5-nitroisatoic anhydride (6-nitro-2H-3,l-benzoxazine2,4(1H)-dione) (3) with formamide gave 4-hydroxy-6nitroquinazoline (4) in excellent yield (Scheme I). Catalytic hydrogenation of 4 using the procedure described earlier afforded amine 5.15 Subsequent reductive condensation with di-tert-butyl N-(4-formylbenzoyl)-~glutamate (6)'O gave di-tert-butyl ester 7 in modest yield. Treatment of 7 with anhydrous trifluoroacetic acid then afforded the target compound 2b. The access to 2-desamino analogues modified at position 10 was facilitated by the recent synthesis of 6-(bromomethyl) - 3,4-di hydro-4- oxo- 3- [ (pivaloyloxy)me t hyl] quinazoline 8.5 The sodium salt of diethyl (4-mercaptobenzoyl)-L-glutamate (9) was generated according to the literature procedure16and then reacted with 5 to yield the protected diethyl ester 10 (Scheme 11). Subsequent treatment with base removed the ester as well as the (pivaloy1oxy)methyl groups to afford 2-desamino-10-thia5,8-dideazafolic acid (2c) in respectable overall yield. On the basis of earlier experience with the preparation of lO-oxa-5,8-dideazafolic acid (la),17 a stepwise approach was employed for the preparation of its 2-desamino counterpart 2d as shown in Scheme 111. The 6-bromomethyl compound 8 reacted with methyl 4-hydroxybenzoate in the presence of cesium bicarbonate to yield the fully protected intermediate 11,which in the presence of dilute sodium hydroxide gave 2-desamino-lO-oxa-5,8dideazapteroic acid (12) in good yield. Coupling of this key intermediate to di-tert-butyl L-glutamate using diethyl phosphorocyanidate afforded the di-tert-butyl intermediate 13. Upon treatment with trifluoroacetic acid, 13 was acid converted into 2-desamino-lO-oxa-5,8-dideazafolic
(2d). Biological Evaluation The three new target compounds 2b-d together with 2a were evaluated as inhibitors of human WIL2/DHFR18 and L1210 leukemia TS.19 The results obtained are presented (13) Hvnes. J. B.: Patil. S. A.: TomaiiE, A,: Kumar, A.: Pathak, A.: Tan, K.; Xianqiang, L.; 'Ratnam, M.;'Delcamp, T. J.; Freisheim, J. H. J. Med. Chem. 1988,31,449. Mueller, W. T.; Smith, G. K.; Benkovic, S. J.; Hynes, J. B. Biochem. Pharmacol. 1988, 37, 449. Baker, B. R.; Schaub, R. E.; Joseph, J. P.; McEvoy, F. J.; Williams, J. H. J. Org. Chem. 1952, 17, 141. Kim, Y. H.; Gaumont, Y.; Kisliuk, R. L.; Mautner, H. G. J. Med. Chem. 1975,18,776. Oatis, J. E., Jr.; Hynes, J. B. J. Med. Chem. 1977, 20, 1393. Delcamp, T. J.; Susten, S. S.; Blankenship, D. T.; Freisheim, J. H. Biochemistry 1983, 22, 633. Jackman, A. L.; Alison, D. L.; Calvert, A. H.; Harrap, K. R. Cancer Res. 1986, 46, 2810.
4
3
5
0
?H
COO-t-Eu
CFaCOOH
I
(CH&,COO-
2b
t - BU
7
Scheme I1
10
2c Table I. Comparison of the Biochemical and Antitumor Effects of Selected 5,8-Dideazafolates versus Their 2-Desamino Counterparts DHFR: TS: L1210/S: L1210/R81: no. Z,,"pM ZSO,Ll'bpM IC,,bpM IC,,bpM 116 0.014' 1.4 la 0.91' 75 0.13 0.27 2a 7.0 3.2 >200 1.3' lb 0.11' >200 25 0.53 2b 0.32 >200 0.22c 7.6 IC 0.33' 1.3 >200 13.8 2c 0.65 39 12' 1.2 Id 0.17c 0.28 >200 49 2d 0.26 0.0044 186 MTX 0.0043 20' Limits of variability
