July 1965 Acknowledgment.-The
METABOLITE OF DICHLOROMETHOTREXATE authors are indebted to Dr.
R. R. Mohan, Nr. R. S. Pianotti, and hlr. B. S. Schwartz, Department of r\ Iicrobiology of the Warner-
513
Lambert Research Institute, research affiliate of the Warner-Chilcott Laboratories, for carrying out the inhibition studies.
The Metabolite of 3',5'-Dich1oro-4-amino-4-deoxy-N10-methy1pteroy1g1utamic Acid (Dichloromethotrexate) TI LI LOO .4SD
RICHARD
H.
Lh.kXSOS
Laborator!/ of Chemical Pharmacology, 'Yational Cancer Institute, Sational Institutes of Health, Bethesrla, dfar!jland Received February 1 , 1965 I)irlilorometliotrexatr ( I X l l I ) is partly metabolized by man and three rodent species. This metabolite is also obtained by incubating U C J I with rat liver homogenate. Based on degradation experiments, the metntmlite is shown to be 3',~'-dichloro-4-amino-4-deoxJ.-7-hydrolcg-N1o-methylpteroylglutsmic acid.
I n studies of the physiological dispositionla-c of dichloromethotrexateza-c (abbreviated D C J I in this paper), a pteroylglutamic acid antagonist of clinical interest6 in the treatment of leukemia, the presence of a metabolite of D C l I was recognized in the urine of animals that had received the drug by different routes. Employing C136-labeled DCM, it was observed that almost all the radioactivity was recoverable from the urine and feces of man,'C dog, and three rodent species'b treated with this drug. In addition to unaltered DCAI, one and only one metabolite, amounting to almost one-third of t'he dose of DC3I administered, was det'ected. The dog, however, excreted DCM unchanged. ,4n identical metabolite was isolated from the in vit7.o incubation of DCRI with rat liver homogenate.' As in the case of t'he in vivo experiments, dog liver homogenate was ineffective in metabolizing DCM. Because of the minute quantity of pure D C l I metabolite available for study and the inherent difficult'ies involved in pteridine chemist,ry, classical techniques are not always useful in structural elucidation.8 To (1) (a) V. T. Oliverio and T. L. Loo, Proc. A m . Assoc. Cancer Res., 3 , 140 (1960); (b) V. T. Oliverio and J. D . Davidson, J . P h a m m " . Ezpll. Therap.. 131, 76 (1962); (c) J. D . Davidson and V. T.Oliverio, Clin. Pharmacol. Ezptl. Therap., in press. (2) (a) Dichloromethotrexate is the 3',5'-dichloro derivative of methotrexate. Methotrexate or amethopterin are generic names for 4-amino-4-deoxyWO-methylpteroylglutamic acid or N-(p- ( [ (2,4-diamino-6-pteridinyl)methyljmethylamino) benzoy1)glutamic acid. (b) The D C M was supplied by Lederle Laboratories Division of American Cyanamid Co. through the courtesy of Cancer Chemotherapy National Service Center of the Sational Cancer Institute. T h e drug used in the rabbit experiments was estimated t o be over 90% pure b y chromatography.a For incubation with r a t liver homogenate, DC.M was purified b y column chromatography on DEAE cellulose.~ (c) T h e synthesis of methotrexate was reported in ref. 4, and t h a t of D C M in ref. 5 , a n d also b y R. Aneier a n d W. V . Curran. J . Am. Chem. SOC., 81, 2814 (1959). (3) V. T. Oliverio, Anal. Chem., 83, 263 (1961). (4) D . R. Seeger, D. B. Cosulich, J. M. Smith, Jr., and M. E. Hultquist IJ. A m . Chem. Soc., 11, 1753 (1949)l reported absorption maximum a t 369 mp i n alkali. (5) D. B. Cosulich, D . R. Seeger, M. J. Fahrenbach, B. Roth, J. H . Mowat. J. M. Smith, Jr., a n d M. E. Hultquist, ibid., 13, 2554 (1951). (6) E. Frei, 111. C. L. Spurr, C. 0. Brindley, 0. Selawry, J. F. Holland, D. P. Rall. L. R. Wasserman, B. Hoogstraten. B. I. Shnider. 0. R. McIntyre, L. B. M a t t h e w . Jr., and S. P. Miller, Clin. Pharmacol. Ezptl. Therap., 6, 160 (1965). (7) T. L. Loo and R . H. Adamson, Biochem. Pharmacol., 11, 170 (1962). I n this paper t h e structure proposed for t h e metabolite 3',5'-dichloro-ihydroxy-N'0-methylpteroylglutamic acid or 4-deamino-4,7-dihydroxy-DC11.I was later shown to be erroneous. See text. (8) For an excellent discussion on this point, see A. Albert, Quart. Rev. (London), 6, 197 (1952).
establish identity most of the following criteria were relied upon: elementary analysis if possible, ultraviolet absorption spectros~opy,~ electrophoretic niobility,'O and position of elution from a diethylaniinoethylcellulose c01umn.~ Taken alone, none of these criteria would be sufficient to warrant a conclusive structural proof. The fact that there is only one metabolite which is not identical with any simple pteridine related to DCRI excludes the possibility that it is a cleavage product of DCM. Also, the metabolite is distinctly different from 3',5'-dichloro-X"o-methylpteroylglutaniic acid6 in spectra and electrophoretic mobility and hence cannot result from D C N by simple hydrolysis of the 4-amino group. Analytical data of DCAI and its magnesium salt can best be expressed by the respective empirical formulas, C20H20ClzS806.H20and C2oH1&12hfgS806. 5HzO." I n other words, tentatively, the metabolite seems to differ from its parent compound only in an oxygen atom. Existing knowledge of drug metabolism suggests that the extra oxygen atom is located in either the pteridine or the benzene ring. A comparison of the ultraviolet absorption spectrum of DCRI in dilute alkali with that of its metabolite discloses that the absorption maximum at 3T0 mp for DCRI has been displaced 26 mp to 344 for the nietabolite. Such a hypsochromic shift is strongly reminiscent of a similar shift of 26 mp in dilute alkali from 364 for 2,4-diamino-6-methylpteridine* to 338 m p for 2,4diamino-T-hydro~y-6-methylpteridine.'~ As a whole, the general pattern of change in the spectra in both acid and alkali matches closely for these two pairs of compounds. Nevertheless, in view of the corresponding (9) Infrared spectroscopy, on the other hand, proves to be of little application to our problem owing t o overlap of hydrogen-bonded OH and N H stretching frequencies. (10) See J. D. Davidson, J. Natl. Cancer Inst., 29, 792 (1962), for a description of method and equipment. (11) I n pteridine chemistry the interpretation of analytical d a t a must proceed with extreme caution.8 I n the present case, ultimate analysis provides no basis for discrimination between the proposed empirical formula CzoHzoClzNsOe.HzO and CzoHzoClzN80~. 2Hz0 ( D C M dihydrate) or CmH18ClzMgNsO6.5HzO and CmH~sClzMgNsOs.6Hz0 (hexahydrated magnesium salt of D C M ) . Determination of water content by the Karl Fisher reagent gave inconclusive results in our hands probably because of addition of water across double bonds in some of these pteridines.8 (12) G. B. Elion, G. H. Hitchings, and P. B. Russell [ J . A m . Chem. Soc.. 1 2 , 78 (1950)l reported absorption maximum in alkali a t 340 mp.
Experimentali6 Isolation of DCM Metabolite from Rabbit Vrine.-l )CSI ~ t s :tdiiiinistered intraperitoneally to six rahhits of 2-3 kg. Eac.li anirnal received 2 ml. ( J f i~ solution of 50 nig. of I>C1L/nil. of dilute SaOH of pH 8.5. T-rine excreted b y the animals during the 4 hr. after injection was oollec+ed and filtered. It \vas d j u s t e d to pH 8 with NH&H aiid applied direct1)- to ii I)E.-11
