2746
VOl. 68
COMMUNICATIONS T O T H E EDITOR _
_
_
~
STREPTOMYCIN. 111. 4-DESOXY-L-ERYTHROSE (THREOSE) PHENYLOSAZONE FROM STREPTOBIOSAMINE
Sir: We have found that the action of phenylhydrazine base on streptobiosamine results in the formation of a crystalline phenylosazone which has been identified by synthesis as 4-desosy-Lerythrose (threose) phenylosazone, I , HC=SS';IICaHj
CHzOAc
I I AcOCH I
1 C =XNH CeHs I HOCH I
c=o
CHs I
CHs I1
s a m i n i ~ l e ,it~ appears probable that the isolated phenylosazone is derived from a fragment of streptonoses representing carbon atoms :3 to 6 of this dicarbonyl sugar. On this premise, the asymmetric carbon atom in the phenylosazone is identical with carbon atom 3 of streptonose. Since this carbon atom has now been shown to ha\-e the I-configuration, streptonose by convention must be designated an 1-sugar. The prepzration of the phenylosazone from streptobiosamine is well reproducible, with yields of 25-300'; of the theoretical in terms of chromatographed material. From this and other considerations it appears likely that the C4-fragment was formed from streptobiosamine under the influence of phenylhydrazine base rather than as a by-product in the preparation of the disaccharide. The unusual lability of the bond between CSand C3of streptonose is undoubtedly connected with its dicarbonyl nature, since dihydrostreptob i o ~ a m i n ein , ~ which the aldehyde group of streptonose is reduced, failed to yield the osazone under similar conditions.
Streptobiosamine hydrochloride' (0.5 g.) was dissolved in water (16 ml.) and phenylhydrazine base (0.5 ml.) was added. The solution was allowed to stand a t 24' under nitrogen for seventytwo hours. 'The resulting crystalline precipitate was collected and adsorbed on alumina from benzene solution. Continued washing with benzene eluted amorphous material followed by a (3) J. Fried and 0. Wintersteiner, T H I S J O U R N A L , in press. crystalline fraction, which was recrystallized from DIVISIONOF ORGANIC CHEMISTRY benzene-hexane to a constant melting point of SQUIBBINSTITUTE FOR J. FRIED 145-146' (cor., no dec.); fine yellow prisms MEDICAL RESEARCH DORISE. WALZ XEWJERSEY 0. ~VINTERSTEINER [ C X ] * ~ D+ 1 1 3 ' (c, 0.81, pyridine); + 5 2 O after XEWBRUXSWICK, RECEIVED NOVEMBER 19, 1946 twenty-,two hours. Anal. Calcd. for C1p,H18ON4: C, 68.07; H, 6.43; N, 19.8. Found: C, 68.11; H , 6.37; N, 19.6. The ultraviolet ELECTROPHORETIC INHOMOGENEITY OF absorption spectrum (maxima a t 255 mp, E 18,000; CRYSTALLINE BETA-LACTOGLOBULIN 310 mp, e 10,200; 390 mp, E 19,700; in ethanol) was identical with t h a t of glucose phenylosazone. Sir: 1,3-Diacetyl-4-desoxy-~-erythrulose, 11, was The fact that different values for the molecular synthesized from acetyl-2-lactyl chloride via weight',? and in its content oi certain amino 3-acetyl- 1,4-bisdesoxy-1-diazo-L-erythrulose. The acids"; have been reported for cr!-stalline ,3diacetyl derivative I1 was hydrolyzed with dilute lactoglobulin suggests that the cnstalline proammonia and the hydrolysate was treated with tein may not be a single substance. Qur elecacetic acid and phenylhydrazine. The resulting trophoretic studies with this protein indicate that oil after chromatographic purification yielded the this is indeed the case. desired phenylosazone I, shown to be identical Pedersen" employecl the light absorption techwith the .compound from streptobiosamine by nique to observe the moving boundary of a 0.F-i melting point (144-145', mixed m.p. no depres- cnstalline 3-lactoglobulin solution in electrosion), rotation ([a] 25D 116' ; +SO' after twenty- phoresis and believed that he had shown that the two hours), absorption spectrum, and analysis protein is essentially homogeneous. I t is now (C, 68.26; H, 6.52; N, 19.9). generally agreed that the protein concentration Since a C-methyl group has been demonstrated in a solution should be a t least :'1 for electroin streptomycin2 and methyl dihydrostreptobio- phoretic homogeneity studies and that the Schlieren methods are more sensitive than the light (1) F. A. Kuehl, Jr., E. H. Flynn, N. G. Brink and K. Folkers, THIS J O U R X A L , 68, 2096 (1946). T h e streptobiosamine hydroabsorption procedurc. \Ye employed a 1.5'c chloride used in iiur work was prepared directly b y hydrolysis of solution of crystalline J-lactoglobulin in the
+
streptomycln trihydrochloride with 1 N H?So$a t 4.5' for fifteen hours. I t was obtained as a n amorphous reddish powder containing about 5
