COMMUNICATIONS TO THE EDITOR
5006
nitrophenylhydrazone (m.p. 119.5-120.5’) was prepared from each acetal. Mixed melting points with an authentic sample7showed no depression. Apparently this reduction is of general applicability. The orthoformic, orthoacetic, and orthovaleric ethyl esters, prepared from the nitriles, were converted to the corresponding acetals in good yields. An article giving detailed results of our experiments will be submitted in the near future. (7) Cf C D Hurd and L L Gershbein, ’Tms
JOURNAL,
69, 2728
(1047) CHEMISTRY DEPARrMEN r ~JNIVERSITY O F ROCHESTrR ROCHESIFR, N Y .
CART, J. CLAWS 1, h f O K G E N I I i k t r , [K TlscrrvPrj Aircirsi fi, 1951
1~01.7;:
Chemical and physical evidence indicate that 11, 111 and IV are representatives of a new class uf tctrahydroptcroylglutamic acid derivatives, in which an imidazoline or imidazolidine ring has been formed linking the N5- and NlO-positions by a single carbon bridge derived from the formyl group. Such ring compounds may be intermediates involved in the synthesis of leucovorin.‘ DONNAB. COSULICH BARBARA ROTH CALCOCHEMICAL DIVISION AMERICAN CYANAMID COMPANY JAMES M. SMITH, JR. Bor;xn BROOK, N. J. MARTIXE. HULTQUIST ROBERT P. PARKER RRCEIVEDAUGUST13, 1951
JOHN
~
..
.
. .
CATION EXCHANGE EQUILIBRIA
Sir: ACID TRANSFORMATION PRODUCTS OF LEUCOVORIN
In a recent valuable paper on cation exchange equilibrium, Lowen and associated authors’ disSir : cuss a paper by Krishnamoorthy, Davis, and Communications from these laboratories1 have Overstreet.2 Apparently two misconceptions are described the synthesis of a formyltetrahydro- involved in this discussion. (a) The authors pteroylglutamic acid, leucovorin (I), with activity state “It should be mentioned, however, that the for Leuconostoc citrovorum 8081.2 From our chemi- arbitrary substitution of equivalent fraction for cal studies on the pure crystalline vitamin, we mole fraction in the resin phase, in the equilibrium have concluded that the probable structure of expression given by these authors (which may leucovorin is 5-formyl-5,6,7,8-tetrahydropteroyl- possibly be justified on the basis of the statistical glutamic acid.3 During this investigation we have nature of the exchange of ions among resin sites), isolated and characterized several acid transforma- results in decreased apparent hysteresis and more tion products of I. nearly constant values of K,.” At @H 1.3 or below, leucovorin yielded isoleucoIt should be pointed out that the terms used in vorin chloride (11), repeatedly crystallized by the equation derived by Davis for the paper in solution in 12 N hydrochloric acid and dilution to Science represent moles (or, statistically, ions) 2 N ; boat-shaped crystals; nL, 1.508 f 0.004; and not equivalents (or, statistically, individual ns, 1.84 f 0.01; m.p. 250-251” (dec.). Anal. charges). This situation is more clearly repreCalcd. for CzoHzzN706C1: c , 48.8; H, 4.51; N, sented in a detailed paper by Davisa 19.9; C1, 7.21; CHO, 5.90a4 Found: C, 49.4; (b) The work reported in THIS JOURNAL inII, 4.73; N, 20.2; Cl, 7.32; CHO, 5.74 (corrected volved basic cation-hydrogen ion exchanges. Both for 6% water, Karl Fischer titration). Crystalliza- in the Science paper cited and in another paper tion of I1 a t pH 2 yielded anhydroleucovorin-A by Davis4 it is shown that the exchange “constant” (111), hair-like needles from 0.01 N hydrochloric is probably not invariant for exchanges involving acid; parallel extinction: ns, 1.63; n L , 1.87; hydrogen ions. The statistical theory applies oblique extinction (30’): ns, 1.48; nL, > 1.90; only to completely dissociated ions, so that the ionm.p. (anhydrous form) 250-257’ (dec.). Anal. exchanger bond strength is not a function of the Calcd. for CzoHzlN706,4Hz0: c, 45.5; H, 5.42; concentration (at least to a first approximation). N, 18.6; CH0,4 5.50; HzO,, 13.7. Found: C, I should like to comment that probably all 45.4; H, 5.44; N, 18.6; CHO, 5.68; HzO, 13.4. successful attempts to formulate theoretical ion I1 or I11 a t p H 4 (hot) gave anhydroleucovorin-B exchange equations represent only approximately (IV), tablets; ns, >1.90; nL, 1.57; anhydrous adequate working hypotheses, which may have form, non-melting below 330’. Anal. Calcd. considerable heuristic value in relation to physical for C z ~ z i N ~ 6 ~ 1 / 2 H zC, o : 51.7; H I 4.78; N, intuition. It is doubtful that any such equations 21.1; CH0,4 6.25; HzO, 1.94. Found: C, 51.7; are completely adequate. We may except, of H, 4.68; N,. 21.1; CHO, 6.22; HzO, 1.6 (by Karl course, those expressed in terms of activities, but, Fischer titration). I n 0.1 N hydrochloric acid as Lowen and his group point out, this result is 11, I11 and IV exhibit a maximum at 356 mp unavoidable in view of the circular dependency of ( T = 35 f. lY0, 10 mg./liter); in anaerobic sodium K a and the a values. hydroxide solution each is converted to leucovorin DIVISION OF SOILS in high yield. UNIVERSITY OF CALIFORNIA LANNESE. DAVIS (1) J. A. Brockman, Jr., e: al., THIS JOURNAL, 73, 4325 (1850); B. Roth, c1 a!., ibdd., in preparation. (2) H. E. Sauberlich and C. A. Baumann, J . Biol. Chcm., 176, 165 (1948). (3) W. Shive, ct ai., proposed the structure 5-formyl-5,6,7,8-tetrahydropteroylglutamic acid for “folinic acid-SF”; 119th meeting, American Chemical Society, Boston, Mass., April, 1951. (4) Formic acid is liberated by the drastic acid hydrolysis in the usual analysis for the formyl group.
DAVIS,CALIFORNIA RECEIVED AUGUST17, 1951
( I ) W.K.Lowen, R. W. Stoenner. W. J . Argersinger, Jr., A. W. Davidson and D . N. Hume, THIS JOURNAL, 78,2665 (1951). (2) C. Krishnamoorthy, L. E. Davis and R . Overstreet, Science, 108, 439 (1948). (3) L. E. Davia, J . Colloid Scicrcc. 6,71 (1949). (4) I. fr. Davis. i b i d . , 6 , 107 (1950).
