J.Med. Chem.
1990,33,833-838
833
Synthesis and Immunosuppressive Activity of Some Side-Chain Variants of Mycophenolic Acidt Peter H. Nelson,***Elsie Eugui,l Ching C. Wang," and Anthony C. Allison8 Syntex Research, 3401 Hillview Ave., Palo Alto, California 94304, and Department of Pharmaceutical Chemistry, University of California, San Francisco, California 94143. Received June 12,1989 The syntheses and immunosuppressive bioassays of 12 side-chain variants of mycophenolic acid (la)are described. The compounds were made either from mycophenolic acid itself or from 5-(chloromethyl)-l,3-dihydro-4-hydroxy6-methoxy-7-methyl-3-oxoisobenzofuran (3), a versatile intermediate for the synthesis of diverse side-chain variants. Replacement of the methylated E double bond of the natural product with a triple bond, a 2 double bond, a saturated bond, or a sulfur atom, with overall chain lengths equal to or greater than that of mycophenolic acid, produced compounds devoid of significant activity. Replacement of the side-chain double bond with difluoro, dibromo, or unsubstituted cyclopropane rings also removed most activity. Replacement of the double bond with an allenic linkage yielded a compound with about one-fifth of the immunosuppressive activity of mycophenolic acid. Some possible causes for the unusual specificity of structure and activity are discussed.
Mycophenolic acid (la) is produced by the fermentation of a number of penicillium species.' The compound was first described in 18962and the structure was determined in 1952.3 The compound has been shown to have antifungal: antibacterial: antiviral: and immunosuppressive properties,6 and it is also effective in several in vitro and in vivo tumor model^.^^^ Mycophenolic acid has been examined in clinical trials against a variety of tumors with little success8and also in prolonged trials against psoriasis, in which the compound was effe~tive.~ Over a considerable period of time, several research groups have attempted to obtain better therapeutic agents based on mycophenolic acid by means of chemicallo or microbiologicall' modifications or by latentiation.12 Although mycophenolic acid itself continues to be used in psoriasis on a compassionate basis,9eno derivative has yet reached the market. Mycophenolic acid is a potent inhibitor of inosine monophosphate dehydrogenase (IMPD) and guanosine monophosphate synthetase, with Ki values in the lo-@M range,7aand it is therefore an inhibitor of guanosine nucleotide biosynthesis. The in vitro properties can be reversed by the addition of exogenous guanine,7aand it is likely that other biological properties are also due to the inhibition of guanosine nucleotide synthesis. Mycophenolic acid is also an inhibitor of soybean lipoxygenase (ICso55 pm),13and it has been hypothesized that this effect may contribute to efficacy in psoriasis, though the antiproliferative effect of the inhibition of purine synthesis seems likely to be more relevant. The high potency of IMPD inhibition by mycophenolic acid is not reflected in low clinical doses (doses in excess of 1 g/day are required in psoriasisgbJ4)presumably because of rapid clearance of the drug.15 In the course of our search for selective nontoxic immunosuppressive agents for use in autoimmune disease, we have examined the effects of inhibitors of purine biosynthesis on lymphocyte responses to mitogenic stimulation. This approach was suggested by the knowledge that lymphocyte responses are reduced in adenosine deaminase deficiency,16but are essentially normal in Lesch-Nyhan syndrome (inherited deficiency of hypoxanthine-guanine phosphoribosyl tran~ferase),'~ observations that suggest that lymphocytes are highly dependent on de novo purine biosynthesis. On the basis of the foregoing and of the observed reduction in intracellular guanosine monophosphate levels 'Contribution No. 770 from the Institute of Organic Chemistry. 1 Syntex Institute of Organic Chemistry. 8 Syntex Institute of Biological Sciences. 11 University of California, San Francisco, CA.
in a variety of cell types treated with mycophenolic acid or other IMPD inhibitors,18J9we examined the effect of mycophenolic acid on mitogen-induced lymphocyte proliferation. The compound was found to be a potent supOxford, A. E.; Raistrick, H. Biochem. J. 1932,26,1441,1902; 1933,27,1176,1473. Gosio, B. Riuista Zgiene Sanita Pubblica Ann. 1896,7, 825, 869,961. Birkinshaw, J. H.; Raistrick, H.; Ross, D. J. Biochem. J. 1952, 50, 630. Florey, H. W.; Gilliver, K.; Jennings, M. A.; Sanders, A. G. Lancet 1946,46.Gilliver, K. Ann. Bot. (London) 1946,10,271. (a) Williams, R. H.; Lively, D. H.; De Long, D. C.; Cline, M. J.; Poore, G. A.; Larsen, S. H. J . Antibiot. 1968,21,463. (b) Sweeney,M. J.; Cline, J. C.; Williams, R. H. Proc. Am. Assoc. Cancer Res. 1969,10,91. Mitsui, A.; Suzuki, S. J. Antibiot. 1969,22,358. (a) Carter, S. B.; Franklin, T. J.; Jones, D. F.; Leonard, B. J.; Mills, S.D.; Turner, R. W.; Turner, W. B. Nature 1969,223, 848. (b) Williams, R. H.; Boeck, L. D.; Cline, J. C.; De Long, D. C.; Gerzon, K.; Gordee, R. S.; Gorman, M.; Holmes, R. E.; Larson, S.H.; Lively, D. H.; Matthews, T. R.; Nelson, J. D.; Poore, G. A,; Stark, W. M.; Sweeney,M. J. Antimicrob. Agents Chemother. 1968,229. (c) Sweeney, M. J.; Gerzon, K.; Harris, P. N.; Holmes, R. E.; Poore, G. A.; Williams, R. H. Cancer Res. 1972,32,1795. Knudtzon, S.; Nissen, N. I. Cancer Chemother. Rep. 1972,56, 221. (a) Jones, E. L.; Epinette, W. W.; Hackney, V. C.; Menendez, L.; Frost, P. J. Invest. Dermatol. 1973,60,246.(b) Marinari, R.; Fleischmajer, R.; Schragger, A. H.; Rosenthal, A. L. Arch. Dermatol. 1977,113,930.(c) Jones, E. L.; Frost, P.; Epinette, W. W.; Gomez, E. In Psoriasis, Proc. 2nd Int. Symp.; Farba, E. M., Cox, E. J., Eds.; Yorke Medical Books, New York, 1977; p 440. (d) Spatz, S.; Rudicka, A,; McDonald, C. J. Br. J . Dermatol. 1978,98,429. (e) Epinette, W. W.; Parker, C. M.; Linn Jones, E.; Griest, M. C. J . Am. Acad. Dermatol. 1987,17, 962. Jones, D.F.; Mills, S.D. J. Med. Chem. 1971,14,305. Jones, D. F.;Moore, R. H.; Crawley, G. C. J. Chem. SOC. 1970, 1725. Suzuki, S.; Takaku, S.; Mori, T. J . Antibiot. 1976,29, 275. Sircar, J. C.;Schwender, C. F. Prostaglandins Leukotrienes Med. 1983,11, 373. Linn Jones. E.: EDinette. W. W.: Hacknev. V. C.: Menendez. L.; Frost, P.J.'I;uest. Dermatol. 1975,65; 537. . (15) (a) Lintrup, J.;Hyltoft-Peterson, P.; Knudtzon, S.; Nissen, N. I. Cancer Chemother. Rep. 1972,32, 1803. (16) Giblett, E.L.; Anderson, J. E.; Cohen, F.; Pollara, B.; Meuwissen, H. J. Lancet 1972,ii, 1067. (17) Allison, A. C.; Hovi, T.; Watts, R. W. E.; Webster, A. D. B. Lancet 1975,ii, 1179. (18) Cohen, M. F.;Maybaum, J.; Sadee, W. J. Biol. Chem. 1981, 256,8713. (19) (a) Lowe, J. K.; Brox, L.; Henderson, J. F. Cancer Res. 1977, 37,736. (b) Malinoski, F.;Stollar, V. Virology 1981,110,281.
0022-2623/90/1833-0833$02.50/00 1990 American Chemical Society
Nelson et al.
834 Journal of Medicinal Chemistry, 1990, Vol. 33, No. 2 Table I. In Vitro Bioassays inhibn of lymphocyte responses to mitogens: ICm, Irm inhibn of IMPD: compd (RS) PHA' PWM" SPA" ICm, w 0.06 0.02 0.06 0.02 1 2.6 2.8 10 3.1 2a 3.1 4.4 > 10 >10 2b 1.1 4.3 3.8 1.7 2c 10.9 >10 >lo >10 4a 22.7 >10 >10 >IO 4b 2.6 7.5 4.8 2.4 4c 2.5 >10 6.0 6.0 4d 1.8 4.8 4.8 4.8 4e NTb >10 >10 >10 4f 1.2 0.15 0.35 0.35 4g 0.94 4.3 1.6 4h 2.5 4i > 10 > 10 >10 1.8 PHA, phytohaemmagglutinin; PWM, pokeweed mitogen; SPA, staphylococcus protein A. Not tested.
The remaining analogues were prepared from the (chloromethy1)phenol 3,24by means of displacement reactions with 3-mercaptopropionic acid, 4-mercaptobutyric acid, and the bis(brom0magnesium) salts of pent-Cynoic and hex-5-ynoic acids. These reactions proceeded rapidly and in good yields, presumably because of elimination of hydrogen chloride from 3a to give a highly reactive oquinone methide. The procedure is a facile route to side-chain variants of mycophenolic acid and, if it was justified by biological results, could be used to prepare a wide range of analogues by means of nucleophilic displacement of the chloride. The acetylenic products of the latter reactions were partially or completely hydrogenated to produce the 2 olefins and the saturated side-chain analogues.
~~
OR2
0
CH3
pressor of the responses of both T and B lymphocytes to mitogens, at concentrations of M or less. Since mycophenolic acid is relatively innocuous in experimental animals (the LDw in rats and mice is 452 and 1917 mg/kg per day, re~pectively~~), we have attempted to synthesize analogues more effective than the parent compound. Previous work, in which analogues of mycophenolic acid were tested for antimitoticloor antitumoF activity, had shown that biological activity was lost if significant modifications were made to the structure. In view of the fact that hydrogenation of the side-chain double bond was known to cause marked loss of antimitotic potency,'O we synthesized a group of compounds that retained r-electron character in the appropriate positions on the side chain relative to the nucleus and the terminal carboxyl group.
Chemistry The cyclopropane analogues 2a-c were prepared from appropriately protected derivatives of mycophenolic acid. Attempts to add difluorocarbene, generated by thermolysis of sodium chlorodifluoroacetate, to mycophenolic acid itself, apparently gave no reaction20 so a protected substrate was examined. The corresponding reaction of methyl 0-methylmycophenolate" (lb) gave a good yield of the adduct 2d, and selective demethylation using boron trichloride,21followed by alkaline hydrolysis of the ester, then afforded 2a. The dibromocyclopropane 2b was also prepared from a doubly protected mycophenolic acid; in this case both the hydroxyl groups were protected as MEM ethers. Dibromocarbene was generated under phasetransfer conditions from bromoform and aqueous sodium hydroxide. The allenic derivative 4g was prepared from the dibromocyclopropane with use of the CrCl,/LAH reagent of Hiyama et a1.22 Since yields were low in the initial preparation of the dibromocyclopropane adduct, different protecting groups were used when larger quantities were required (see the Experimental Section). The unsubstituted cyclopropane 2c was prepared by means of a modified Simmon-Smith reactionB on mycophenolic acid itself. ~~
~~~~
(20) It was later found that la and 2a have the same R, in the solvent system used, so some 2a may have been formed. (21) Dean, D. M.; Goochild, J.; Houghton, L. E.; Martin, J. A.; Morton, R. B.; Parton, B.; Price, A. W.; Somviechien, N. Tetrahedron Lett. 1966, 4153. (22) Hiyama, T.; Okude, Y.; Kimura, K.; Noyaki, H. Bull. Chem. SOC.J p n . 1982, 55, 561. (23) Harrison, I. T.; Rawson, R. .J. J Org. Chem. 1970, 35. 2057.
la: R ' = R ' = H b: R' = R2 = CH3 C: R' = R2 = CH20(CH2)20CH3 d: R' = CH20(CH2)20CH3; R2 = CH3
2a: b: c: d: e:
R' = R' = H; x = F R' = R2 = H; X = Br R1=R2=X=H
R ' = R2=CH3; X = F R' = H; R2 = CH3; X = F
f: R' = CH20(CH2)20CH3; R2 = CH3; X = Br
3
4a: R' = R' = H; x = s; n = 2 b: R ' = R 2 = H ; X = S ; n = 3 C:
R ' = R ~ = H ;x = c E c ; ~ = ~
d: R'=R2=H; X = C z C ; n = 3 e: R ' = R 2 = H ; X = ( Z ) - C H = C H ; n = 2 f: R ' = R ~ = H ;x = ( z ) - c H = c H ; ~ = ~ 9: R' = R2 = H; X = CH=C=C(CH3); n= 2 h: R' = R2 = H; X = (CHp)?;n = 2 i: R' = R2 =H; X = (CHp)?;n = 3 j: R ' = H ; R 2 = C H 3 ; X = C H = C = C ( C H 3 ) ; n = 2
Discussion All the analogues were assayed for their ability to inhibit mitogen-induced human lymphocyte proliferation. Eleven of the compounds were also tested for inhibition of lymphoid-derived IMPD. The results are shown in Table I. As stated earlier, mycophenolic acid is a potent inhibitor of the blastic transformation, with ICm values in the M range. It exhibits similar activity against different lymphocyte subsets; thus responses to PHA, which stimulates T cells,25and PWM and SPA, which respectively (24) Birch, A. J.; Wright, J. J. Austral. J . Chem. 1969, 22, 2635. (25) Weksler, M. E.; Kuntz, M. M. Immunology 1976, 31, 273.
Journal of Medicinal Chemistry, 1990, Vol. 33, No. 2 835
Side-Chain Variants of Mycophenolic Acid
Table 11. Effect of la and 4g on in Vivo Antibody Response to SRBC in Mice treatment,” daw 0 to 4
no. of animals
PO dose, mg/kg per day
vehicle la 4g 4g
6 6 6 6
25 25 40
inhibn
PPM‘
inhibn
WBC/spleen x 108
69 18 15
751 264d 630 583
65 16 22
158 137 155 169
%
PFC/spleenb Experiment 1 117416 36 027d 96 583 99 333
%
Experiment 2 7 84 142 544 153 46 308d 44 142 45 333d la 6 25 -33 657 -20 172 112416 6 50 4g 80 114d 79 154 17 OOOd 4g 6 100 “All animals received 1 X 108 SRBC ip on day 0. bFiguresare means of the total number of PFC/spleen per group (quadruplicatesamples from each animal). ‘Plaques per million (PPM) are the total number of plaques in the spleen divided by 10” times the total number of nucleated cells. d p
