1650
J. K. DANKAND J. W. GATES,J K . [COMNUKICATIOK S O . 1854 FROM
RESEARCH LABORATORIES,
Vol. 7'3 EASYMAN KOUAK CO.]
The Reaction of Djenkolic Acid and Aqueous Calcium Hydroxide. The Isolation and Identification of Racemic Thiazolidine-4-carboxylic Acid BY J. K. DANNAND J. W. GATES,JR. RECEIVED OCTOBER 10, 1956 Oiie of the materials produced by the treatment of djenkolic acid at room temperature with aqueous calcium hydroxide has been identified as thiazolidine-4-carboxylic acid. The isolation, purification and comparison of this material with the synthetic compound have been described. The identity of this material has been further confirmed by comparison of its iufrared and ultraviolet absorption spectra 1% ith those of the synthetic material.
Introduction this was the ethyl ester of thiazolidine-4-carboxylic Djenkolic acid (I) CH2(SCH2CH(NH~)COOH)zacid (VII) and not the diethyl ester of 2-methyl(Chart I) is a naturally occurring sulfur-containing thiazolidine-2,4-dicarboxylic acid (IX) as had been amino acid which was first detected in the urine of obtained from cystine treated with aqueous calnatives of Java who included the djenkol bean in cium h y d r ~ x i d e . ~The analysis also indicated that their diet. It was separated from djenkol beans the material was not the diethyl ester of djenkolic in 1935 by van Veen and Hyman,' who extracted acid (VIII). The ester VI1 was prepared from synthe djenkol beans with dilute barium hydroxide a t thetic thiazolidine-4-carboxylic acid (VI), which, in 30' and isolated the acid from the reaction mixture. turn, had been prepared from L-cysteine (V), and The acid was found to be stable to concentrated formaldehyde by the method of Ratner and Clarke.s It was found that the hydrochlorideg of the synhydrochloric acid a t loo', but hydrolysis with concentrated sulfuric acid gave cystine and formalde- thetic ester VI1 melted some 20" higher than the hyde. Saturated barium hydroxide gave no sul- hydrochloride of the ester I V obtained from djenfites or sulfates with djenkolic acid. A dibenzoyl kolic acid treated with aqueous calcium hydroxide. derivative and a dihydantoin were prepared. The Ultraviolet absorption spectra of the two coinsynthesis of djenkolic acid was carried out by du pounds were identical but showed no characteristic Vigneaud and Patterson, who treated cystine in absorption. Infrared absorption spectra also inliquid ammonia with methylene chloride. Arm- dicated that the two compounds VI1 and IV were strong and du Vigneaud3 later synthesized djen- apparently identical; only the spectra (Fig. 1) of kolic acid from 1 mole of formaldehyde and 2 moles the synthetic compound VI1 is given since the two of cysteine in strong hydrochloric acid solution. spectra are nearly indistinguishable. Infrared Middlebrook and Phillips4 obtained djenkolic acid absorption spectra were obtained on a Baird doubleby adding formaldehyde to a wool cystine sub- beam spectrophotometer using rock-salt optics. fraction. Evidence that djenkolic acid occurs in a The samples were prepared as pressed plates in free state was obtained by van Veen and Latua- potassium bromide. Since a racemized thiazolidine was isolated from ~ 3 1 1 ,who ~ separated djenkolic acid from djenkol beans by a process of dialysis, extr?ction and cystine which had been treated with aqueous calchromatographic absorption, while Xgren and cium hydroxide,' it seemed quite plausible to exEklund6 obtained djenkolic acid from a hydroly- pect the same phenomena in the case of djenkolic zate of a peptide in calf plasma. They also found acid. This proved to be the case, as the synthetic ester hydrochloride VI1 was found to have a traces in muscle tissue. in alcohol In the present work, djenkolic acid was treated specific rotation [ a ] ~ * ~of. 4 -79.0' with aqueous calcium hydroxide a t room tempera- (c 0.01 g./cc.), while the ester hydrochloride IV ture for 2 months in the same manner as was used obtained from djenkolic acid was found to be opin earlier work in these Laboratories for the similar tically inactive. tThen the synthetic ester hydrotreatment of cystine.' The calcium salt I1 was chloride V I 1 was treated with a catalytic amount obtained by drying down the filtered reaction mix- of potassium hydroxide in ethanol by the procedure ture a t room temperature. Esterification of this of McKenzie and Wren, lo the low-melting estcr salt with absolute ethanol and dry hydrogen chlo- hydrochloride IV was obtained. When untreated ride, followed by neutralization with potassium djenkolic acid (I), was esterified in the same inancarbonate and distillation in vacuo, gave the liquid ner as was 11, a solid ester hydrochloride VI11 was ester 111in 30% yield, which, upon treatment with obtained which was not similar to either I V or VII. dry hydrogen chloride, gave the solid ester hydroExperimental chloride IV. Elemental analysis indicated that ( 1 ) A . G. van Veen a n d A. J. H y m a n , Geneesk. Tijdschr. N e d w l I n d i e , 7 3 , 991 (1933); Rec. tuav. chim., 64, 493 (1935). (2) V. d u Vigneaud a n d W. I. Patterson, J . B i d . C h e w , 114, 533 (l!)HIi), (3) A I . 11. Armstrong a n d V. d u Vigneaud, i h i d . , 114, :373 (1947). (4) W. 13. Middlebrook a n d H. P. Phillips, Riockenz. J . , 41, 218 (1947).
( 5 ) A . G . v a n Veen and H. E , Latuasan, Chroizica Naturae, 105, 288 (1949); C. A , , 44, 4548 (1950). i G ) G. Agren a n d S . I!klund, A r l o C h e m . S r a m . , 6 , 1129 (1952). (7) J . J
