July 20, l9L7
C~LCICM ~,~-O-I~OPROPYLIDENE-L-IDOFURCROX.~TE-G-C~' 3867 [CONTRIBUTIONFROJI N A T I O N A L
BUREAUOF STANDARDS, DIVISION OF CHEMISTRY]
Reaction of Dimeric 5-Aldo-l,Z-o-isopropylidene-D-xy/o-pentofuranosewith Cyanide and the Preparation of Calcium 1,2-O-Isopropylidene-~-idofururonate-6-~2~~ Dihydrate1 BY ROBERTSCHAFFBR AND HORACE S . TSRETL RECEIVEDh f A R C I I fi. 19~57 A new method is reported for the separation of 1,2-0-isopropylidene-~-idofururonic-6-C1~ acid from the reaction product of dimeric 5-aldo-1,2-0-isopropylidene-~-~yZo-pentofuranose and C14-labeled cyanide. After separation of barium 1,2-0isopropylidene-~-glucofururonate-6-C~4 hydrate from t h e reaction mixture, crystalline calcium 1,2-0-isopropvlidene-Lidofururonate-6-C14dihydrate was obtained in 27% yield. The combined radiochemical yield of t h e ~ - i d u r o n i c - 6 - C a' ~n d ~-glucuronic-6-C acid ~ ~ epimers was 85%. The high yield shows that under the conditions used the dimeric 5-aldo-1,2-0isopropylidene-D-xylo-pentofuranoseis cleaved, and the resulting monomer gives a normal cyanohydrin reaction,
By application of the cyanohydrin synthesis t o the product went to the L-iduronate epimer. For 5-aldo-1 ,2-0-isopropylidene-~- xylo - pentofuranose, this reason a reaction product obtained from a cyanoSowden2 obtained D-gl~curonic-6-C~~ acid in about hydrin reaction buffered with sodium bisulfite was 20% yield. The cyanohydrin reaction mixture used in a search for a crystalline salt of the 1,2-O-isoturned dark with formation of by-products of un- propylidene-L-idofururonic acid. After removal of known character. Subsequently Shafizadeh and barium 1,2 - 0 - isopropylidene - D - glucofururonate Urolfrom3as well as Schaffer and Isbel14found t h a t from the reaction mixture, crystalline calcium 1,2-0by controlling the acidity of the reaction mixture isopropylidene-L-idofururonate dihydrate was prethe proportions of the epimeric acids could be al- pared. The new salt crystallizes readily and protered.j By use of a bicarbonate-carbon dioxide buf- vides a convenient means for separatioii of the fer and chroniatographic separation of the lactones, iduronate epimer in a pure state. In order to utilShafizadeh and Wolfrom obtained crystalline 1,2- ize the salt for separation of the radioactive L-idu0-isopropylidene-D-ghcurono-y-lactone in 15.5% ronate from the mother liquor of the preparation of yield and 1,2-0-isopropylidene-L-idurono-y-lactonebarium 1,2-O-isopropylidene-~-g~ucofururonate-6in 16.17, yield. From an acetate-buffered cyanohy- C14 hydrate it was necessary first t o remove the drin reaction mixture Schaffer and Isbell separated barium acetate (by decationizing and fxeze-drybarium 1,2 - 0 - isopropylidene - D - glucufururonate- ing), and the residual barium lj2-0-isopro3ylideneG-C14 monohydrate in XyGyield and a mother ~-glucofururonate-6-C" hydrate (by crystallizaliquor containing an unknown amount of the 1,2- tion). Barium in the purified mother liquor was O-isopropylidene-~-idofururonic-6-C~~ epimer. replaced b y calcium, and calcium 1,2-0-is~propyliThe continued interest in this synthesis in this dene-L-idofururonate- G - C14 dihydrate was sepaLaboratory derives from a program for the develop- rated in a radiochemical yield of 2i%.' 'The comment of methods for the preparation of position- bined yield of the D-glucuronate and L-iduronate labeled radioactive carbohydrates,'j and from in- epimers was S5Y0. The separation of the epimeric terest in the structure and properties of the crystal- acids in high yield shows t h a t under suitable conline dimer of ,?-aldo-1,2-0-isopropy~idene-~-xy~oditions the dimer is cleaved and the resulthg monopentofuranose. mer gives a normal cyanohydrin reaction. I n the present study a crystalline salt of 1,2-0By isotope dilution analysis of the product from a isopropylidene-L-idofururonic acid was sought be- bisulfite-buffered cyanohydrin reaction it was cause this would provide not only a simple means found earlier4 t h a t the D-ghcuronate ep mer had for purifying this acid-sensitive material but would been formed in 38Y0 yield; in the present study it also provide a means for its quantitative determina- was found that the L-iduronate epimer formed in tion by use of the isotope dilution technique. 3401, yield. T h e total radioactive yield of 72% for I t was shown p r e v i ~ u s l y ,by ~ isotope dilution the two epimers, considerably below the SS% yield analysis, that the D-glUCUrOnate epimer was pro- obtained in the acetate-buffered reaction, shows a duced in 5SYGyield by the cyanohydrin reaction with substantial loss of labeled cyanide in the bisulfitean acetate buffer and in 387, yield with a bisulfite buffered reaction. A greater yield of the Lbuffer. I n view of these results it seemed probable iduronate epimer (at the expense of the D-ghlcuronthat with the bisulfite buffer a larger proportion of ate) can be expected for more alkaline buffered systems, b u t the limit of alkalinity for an optimum (1) T h i s work was conducted a s p a r t of a project sponsored b y t h e Division of Research of t h e Atomic Energy Commission. yield of L-iduronate is restricted by the alkali-sen( 2 ) J. C. Sowden, THISJOCRNAL. 74, 4377 (1952). sitivity of the ~-a~do-1,2-O-isopropy~iderie-~-xy~o(3) F. Shafizadeh a n d M. L. Wolfrom, i b i d . , 77, 2568 (1955). pentofuranose. The behavior of this substance in (4) R.Schaffer a n d H. S. Isbell, J. Research Natl. Bur. Standards, 6 6 , alkaline solution is being studied and will he treated 191 (1958). ( 5 ) H. S . Isbell, U. S. P a t e n t 2,808,918(1952). in a future publication. (G) H. S. Isbell, J. V. Karabinos, H . L. F r u s h , N. B. Holt, A. Experimental Procedure Schmebel a n d T. T.Gallrowski, J. Reseavch N a t l . Barr. SLandards, 48, 163 (1952); H.S. Isbell a n d J. V. Karabinos, i b i d . , 48, 438 (1952); H.L. Frush a n d H . S. Isbell, i b i d . , SO, 133 (1953); 61, 167 (1953); 51, 307 (1953): H.S.Isbell, H. L. F r u s h a n d N. B. Holt, i b i d . , 53, 217 (1954); 63, 325 (1954); H.S. Isbell, H. L. F r u s h a n d R. Schaffer, i b i d . , 64, 201 (1955); H . S.Isbell, N. B . Holt a n d H . L. Frush, i b i d . , 6 1 , 95 (1956): H.S. Isbell a n d R . Schaffer, TEIS JOURNAL, 78, 1888 (1956).
Calcium 1,2-0-Isopropylidene-L-idofururonateDihydrate.
--An aqueous solution of 12.54 g. of 5-aldo-1,2-0-isopropyli(7) Since t h e preparation of t h e labeled epimeric cyanohydrins was made t w o years prior t o t h e isolation of t h e L-iduronate-6-Cl4, a yield greater t h a n 27% m a y have been produced originally.
denc-D-xyio-pentofuratlose, 12.65 g . of sodium pyrosulfitc and resin,8 washed through with ice-water, and collected in a flask containing 0.3 g. of barium carbonate. The partially 3.4 g. of sodium cyanide was prepared a t On, and allowed to warm to room temperature. After three days the solution neutralized effluent was freeze-dried to remove t h e acetic acid present. Addition of methanol t o an aqueous solution was heated at 90" for one hour, then treated with 7 g. of sodium carbonate, and refluxed for three hours. T h e solu- of the residue yielded 0.18 g. of crude barium 1,2-O-isopropylidene-D-glucuronate-6-C'd monohydrate. tion was cooled in a n ice-bath, and passed through a column containing 350 ml. of cation exchange resin8 at ice temperaT h e purified mother liquor was diluted with ice-water and ture. T h e efiluent, including washings, was delivered into a passed through a column of 12 nil. of ice-cold cation exchange resin8 into a flask containing 0.13 g . of calcium carflask immersed in a n ice-bath, containing 41.6 g. of barium hydroxide octahydrzte. After reinoval of excess barium bonate. Concentration of the solution under vacuum yielded calcium 1,2-//-isopropylidene-I,-idofururonate-G-~~' hydroxide by carbonation ani! Gltration, one-tenth of the dihydrate, which was separated and washed with a mixture filtrate was set aside for analysis by a n isotope dilution techof ethanol and water ( 6 : 1). Recrystallization from water nique. The remainder of the solution was concentrated and ethanol yielded 2.81 millicuries of the salt. By use of under vacuum t o a sirup t h a t yielded 7.77 g. of crude crystalnon-radioactive carrier, an additional 0.53 millicurie of this line barium 1,2-0-isopropylidene-~-giucofururonatc monohydrate. After removal of the barium salt, the mother product was obtained. T h e radiochemical yield based on the 12.5 millicuries of cyanide-C14 originally used was 27%. liquor \\-as passed through a column containing 75 ml. of cation eschange resina at ice tenipcrature. The effluent, inYield of the L-Iduronate Epimsr in a Bisulfite-buffered Cyanohydrin Reaction.-The portion of the solution from cluding v-ashitigs, was delivered into a flask containing 5 g. of calcium carbonate. T h e mixture was filtered, and t h e the preparation of non-radioactive calcium 1,2-0-isopropylifiltrate concentrated under vacuum. Calcium 1,2-0-iso- dene-L-idofururonate dihydrate kept for the isotope dilution propylidene-~-idofururoiiate-6-C'~ dihydrate separated. The analysis was diluted with a tracer consisting of 95.2 microproduct, 4.2 g., was recrystallized from wat.er with the addicuries of calcium 1,2-0-isopropylidene-~-idofururonate-6CLi dihydrate (10.6s mg.). After separation of the D-glution of methanol. -11zaZ. Calcd. for Ca C 1 8 H P ~ O 2H20: ~( C, 39.8; 11, 5.6; curonate epimer as the barium salt, the solution was passed through 15 ml. of cation exchange resin* into a flask containCa, 7.4. Found: C , 39.8; H,8.8; C ~ , 7 . 2 ; [CY];" . .~ -11.8" ing 0.75 g. of calcium carbonate. T h e excess calcium car(.c 3 , water). bonate was removed, and the calcium salt was crystallized Preparation of Calcium 1,2-O-Isopropylidene-~-idofururon-as already described. .4.fter two recrystallizations, the nirate-6-Cli Dihydrate.-The original synthesis began with dried salt had a specific radioactivity of 0.151 pc./mg. This 0.005 niole of sodium cyanide-CI4 (12.5 millicuries) and 1.11 activity corresponds to dilution of the tracer by A20 mg. of g . of 5 - a ~ d o - 1 , 2 - ~ ~ - i s o p r o p y ~ i d e n e - ~ - ~ y ~ ~ - p e n t obuff u r a n ocalcium se 1,2-0-isopropylidene-L-idofururonatedihydrate or a fered with 0.005 mole of sodium hydroxide and 0.02 mole of 34.3% yield of this substance from t h e cyanohydrin reacacetic acid. T h e mother liquor t h a t remained after separa- tion. tion of 6.75 millicuries of barium 1,2-O-isopropylidene-~Radioactivity Measurements.-Deterniiriations of C" glucofururonate-G-C" monohydrate from the cyanohydrin were made by direct count of the materials in formarnidc or synthesis previously rcported hat1 been set aside for ap- aqueous forinamide solutions.9 Samples were counted to a proximately two years bcfore the work reported here was reprobable error of 176, and the results converted to microsum cti. curies by a factor based on the C14-standnrd issued by the The mother liquor was rlissoiveti in 100 ml. of r a t e r , Sational Bureau of Standards. treated with 700 in!. of t.thani)l, m c l passed through a carThe authors express their appreciation to R. A. bon-coated filter. IJnder vacuum the filtrate was concentrated to a 50-ml. volume. T h e concentrate was passed, icePaulson and L. Tregoning for the microanalyses. cold, into a column containing 30 i d . of cation exchange (9) A. Schwebel, H. S.Isbell and J. D. hloyer, J. liesearch A'fitl. Bur. S t a n d a r d s , 53, 221 ( I Q 5 4 ) .
(8) Amberlite IR-120 H, Resinous Products Division u f R u h m a n d Haas C u . , Philadelphia, Pa.
[CONlRIEUTION FROM THE
DEPARTMENT OF
\.YASIiINGTON
25, D .
c.
CHEMISTRY OF THE O H I O STATE UNIVERSITY]
Acyl Migration in the D-Galactose Structure BY
11.L. mTOLFROM, A. TIIOMPSON' AND h'f.
INATOhIE'
RECEIVED JANUARY 17, 1957 Ii'hen 2,3,1-tri-0-acetyl-l,6-anhydro-P-~-galactose is treated successively with titanium tetrachloride and mercuric acetate, there are obtained two crystalline tetraacetates of D-galactose. The one (isomer A) is convertible t o the other (isomer B ) by traces of alkali. Both are convertible t o 8-D-galactopyranose pentaacetate on mild acetylation, Isomers A and B are resistant toward triphenylmethylation. Isomer X yields a p-toluenesulfonate which i n turn forms a methyl tri-0-acetylO-~-toluenesulfo~iyl-~-~-galactopyrat~oside, both of which derivatives are different from known isomers believed to have the p-toluenesulfonyl group in the sixth position. Both isomers A and B yield tetra-0-acetyl-3-0-methyl-~-~-galactopyrose on methylation. Positions 2, 3 or 4 are assignable for the open hydroxyl group in each of the two n-galactopyranose tetraacetates, but the present evidence disallows definite allocations.
No tetraacetate of n-galactose is known in which the terminal primary hydroxyl group is unsubstituted. An attempt to prepare such a derivative by a procedure suitable for the synthesis of the analogous D-glUCOSe structure, led t o a c r y s t a k e n-galactose tetraacetate, m.p. 138-138.5", [a]; 1 3 7 ' (chloroform), herein designated isomer A. This substance was considered t o have its position (1) Research Associate ( A T.) a n d Fellow (11,I,) of t h e Corn Industries Research Foundation. (?) A Thompson, h I I.. Wolfroni and h l . I n a t o m e , TITIS JOURNAL, 77, 3180 (1955).
six substituted by an acetate group since it was recovered unchanged on treatment with triphenylmethyl chloride and pyridine.2 Isomer A had been prepared2 by the successive action of titanium tetrachloride and mercuric acetate on 2,3,4-tri-Oacety1-1,G-anhydro-P-D-galactopyranose.A second isomer has now been isolated, by chromatographic techniques, from the same reaction mixture. The substance, m.p. 140-141.5', [ N ] D +25.5' (chloroform), designated isomer B, possesses essentially the same melting point as isomer A but the mixed
