June 5 , 1960
CONFORMATIONS OF METHYL IDOPYRANOSIDES
2811
in the spectrum of lactose do not disappear in hibit slight shifts in position. These are exactly the the spectrum of the ureide but are considerably changes observed. On the basis of these data, the weaker and somewhat shifted in wave length spectrum can most readily be explained as arising position. This is, upon reflection, the effect to be from an aldehydo-glucose configuration in the expected. Lactose (glucose-/3-galactoside) is a lactose ureide molecule. disaccharide containing 2 pyranose rings. Urea can Conclusions react only with the glucose portion since only that The tentative infrared band assignments for part of the molecule has a free reducing group. glucose ureide, glucose ureide urea and lactose There is little doubt from the spectrum that an ureide indicate the presence of an amide group in amide group is present in the ureide. The changes a t the longer wave lengths can be explained by the the respective molecules. In the glucose ureide fact that the bands arising from the glucopyranose and glucose ureide urea molecules there is evidence ring disappear as the reaction proceeds. The to suggest that the glucose is apparently in the galactose portion, however, still remains in the aldehydo configuration. There is some evidence ring form regardless of what happens to the glucose in the spectrum of lactose ureide to suggest that the portion and, hence, the absomtion bands associ- glucose portion of the molecule is in the aldehydo k e d with the ring v.ibrations can only be expected IOrm* Acknowledgments.-The authors express their to become weaker. But, as these bands are now resolved from the glucopyranose ring bands, their appreciation to Mrs. E. F. DuPr6 for the infrared exact wave length positions would be expected to ex- spectra.
[CONTRIBUTION
FROM THE
DEPARTMENT O F BIOCHEMISTRY AND NUTRITION, GRADUATE SCHOOL UNIVERSITY O F PITTSBURGH]
O F PUBLIC
HEALTH,
The Conformations of Methyl Idopyranosides’ BY RONALD BENTLEY RECEIVEDOCTOBER 19, 1959 Complex formation with cuprammonium solutions was studied with methyl a- and 8-D-idopyranoside (compensating complexes), methyl 2-methyl-&~-idopyranoside (levorotatory complex), methyl 3-methyl-@-~-idopyranoside(la,3acomplex) and methyl 2,3-dimethyl-@-~-idopyranoside(no complex). The oxidation of methyl P-D-idoside by chlorine was more rapid than that of the a-anomer. Possible conformations for idose derivatives are based on these observations and results already in the literature. a-Idosides are assigned a 1C conformation. The closest representation of the 8-idosides is believed to be a half-chair structure, HC3, with deviations from this ideal representation either toward the chair conformation C1 or the skew conformation 1B3. Methyl 2-methyl-P-~-idopyranosideand methyl 3-methyl-B-~-idopyranoside were prepared in the form of sirups and were characterized by means of optical rotations. The structures of these two compounds were confirmed by periodate oxidation experiments.
I n his classic studies of the conformations of pyranose rings, Reeves2 investigated the reactions of a number of idose derivatives with cuprammonium solutions. The following conformational assignments were made: 1C conformation, methyl cr-~-idoside,~ methyl 2-methyl-cr-~-idoside,methyl 4,6-benzylidene-a-~-idoside. Of the two p-idosides examined, methyl 3-methyl-P-o-idoside was assigned the C1 conformation and methyl 4,6benzylidene-/3-D-idoside, a 1C conformation. In the latter case, complex formation with the cuprammonium reagent was very poor. It was concluded that the @-idosidescould react in both chair conformations. Consideration of the “instability factors”z also suggested a C1 Ft 1C equilibrium for the p-idosides. With the realization that other stable conformations participate in the interconversion of the two chair forms,4 the description of a compound as a C1 e 1C mixture is unsatisfactory unless it is established that the theoretical intermediates in this interconversion have no stable existence. In two other instances, lyxose and (1) This work was supported in part by a Research Grant (A-725) from the National Institutes of Health, United States Public Health Service. (2) R. E. Reeves, THIS JOURNAL, 72, 1499 (1950). (3) The abbreviated form, idoside, used in this paper, refers throughout to idopyranoside structures. (4) R. E. Reeves, A n n . Rev. Biochem., 27, 15 (1958).
altrose, where C1 S 1C interconversions were originally postulated, Reeves has suggested* a stable shape in the flexible cycle of the six boat forms. The present paper describes experiments designed to provide more information about the possible conformations of idose derivatives. Experimental Methyl 4,6-benzylidene-a- and P-D-idosides were prepared from methyl a- and 0-&galactosides as described by Sorkin and R e i c h s t e h 6 Removal of the benzylidene group was accomplished by catalytic hydrogenations as follows: methyl 4,6-benzylidene-P-~-idoside (535 mg.) in glacial acetic acid (32 ml.) was shaken with 10% palladium-charcoal catalyst (450 mg.) in a hydrogen atmosphere. After about 2.5 hours no more hydrogen was taken up. The solution was filtered and evaporated t o a sirup. The sirup was taken up in water (10 ml.) and extracted three times with chloroform (15 ml. in all). The aqueous solution was treated with a small amount of Amberlite M B 2 , then Norite A. Filtration and evaporation gave methyl 8-D-idoside as a sirup, [ . : ] * l ~ -48.5’ (c 1.39, HpO). Sorkin and Reichstein6 quote the -95.0’ (c 3.799 specific rotation of this compound as [ C Z ] ~ D ~ HpO). However, in methanol) and [ a I z 0-81.l0(c3.265in in describing the conversion of this compound to D-idosan, they quote a specific rotation [CX]’~D-47’ for a 2 . 5 % solution in 2.5% sulfuric acid after 3 minutes a t room temperature. ~ (c 10 in H z ~for ) this Wiggins’later reported [ a l Z o-40.8’ compound. (5) E. Sorkin and T. Reichstein, Helo. Chim. Acta, 28, 1 (1945). (6) R . E.Reeves, THISJOURNAL, 71, 2116 (1919). (7) L. F. Wiggins, J. Chem. SOL.,1590 (1949).
2512
RONALD
BENTLEV
VOI.
s2
TABLE I COMPI.FX
F O R M A T I O N BY IDOSIDEP I N CTPKAMMONIUR.1 S O L U T I O N
[a12'431
in Hz0, c
in cupra B, c
+188.8', 1.19
+444.6', 1.21 -153.7,0.28 -919.1,1.12 - 59.0,0.33
Compound
Methyl a-D-Idoside p-D-Idoside 2-Methyl-8-D-idoside 3-h[ethyl-p-~-idoside
+103.3', 1 . 1 9 48.5,1.39 55.9,1.64 - 50.1,1.49
-
1'23?'416
-
99.7,1.39 -101.9,1.64 - 92.7,1.49
Rotational shift, deg.
.k 496" - 105 - 1670 Si0.1
SI).res.. 0.01 .lf S O i f l . Asp. in cupra h res. ohm, cm. ohm, cm.
428 418
88 7s
..
, .
465
95
Methyl a-D-idoside prepared in the same way had [ a I Z 4 ~and after cooling werc treated with 2 ml. of a freshly prepared mixture of 4 J f hydroxylamine hydrochloride and 7 N 4-103.3" (c 1.19 in H20). It was obtained as a sirup which did not crystallize even after 2 years. Sorkin and Reich- S a O H (1:l). -1fter standing for 2 minutes a t room temstein6 report [ a ] % +103.3' (c 2.716 in HzO) for the sirup perature, 2 ml. of HCl ( 1 : 3 v . / v . ) and 1 ml. of 0.74 X ferric chloride were added. The optical density was determined in and [a]% +99.8' (c 2.833 in H20) for crystals obtained the Klett-Summerson photoelectric colorimeter using a 540 after standing for many weeks. Methyl 2,3-dimethyl-fl-~-idoside was prepared from mp filter. A straight-line response was obtained over the as fol- range 0 t o 10 ptnoles. methyl 2,3-dimethyl-4,6-henzylidene-p-~-idoside~ lows. The benzylidene compound (171 mg.) in glacial acetic Results and Discussion acid ( 8 nil.) was treated with 107' palladium-charcoal catalyst and hydrogenated a t room temperature and pressure. Since in the P-series, only the methyl pyranosides When hydrogen uptake was complete ( 4 hr.) the filtered of 3-methyl-idose ant1 4.(i-benzylidene-idose had solution wiis evaporated to dryness. The sirup wits kept overnight in VUCZLO with KaOH to complete removal of the been examined,? it was desirable to extend these acetic acid. The residue was dissolved in water ( 5 ml.) and observations. Methyl 2-methyl-P-o-idoside and the solution was extracted with four portions of chloroform methyl P-D-idoside were therefore prepared and the ( 2 . 5 ml. each). Evaporation gave a slightly yellow oil; on reactions of these compounds and also those preaddition of water, a little material remained insoluble and was filtered off. The clear solution obtained on treatment viously examined by Reeves with cuprainmonium with Norite A was evaporated to yield a colorless sirup, [ a I z 4 D are shown in Table I . For methyl a-D-idoside -45.2' (c 1.86 in HzO). and methyl 3-methyl-P-~-idoside,the values for Anal. Calcd. for C9HlS06: C, 48.6; H, 8.17. Found: rotational shift and A specific resistance agreed C, 48.5; H, 8.27. well with the results of Reeves2 Slethyl p-DThe removal of the benzylidene group was also confirmed idoside resembled methyl a-midoside forming a by treatment of a dilute solution of the compound with 3 S HCl. There was no increase in the optical density at 275 complex (as indicated by the specific resistance mp which is normally observed on cleavage of the benzyli- change) which had a low rotational shift; this dene group with acid. complex could therefore have been either of the Methyl 3-methyl-p-D-idoside was prepared as described compensating or la,3a-type. l 2 Alethy1 2-methyl-Pby Reeves.' Methyl 2-methyl-P-~-idoside prepared similarly by reduction of methyl 2-methgl-4,6-benzylidene-P-~-D-idoside complexed well with cupra B, with a idoside5 was obtained as a clear glass. Optical rotations for strongly levorotatory shift ( - l , G i O o ) ; this behavior these two compounds are given in Table I . was analogous to that reported by Reeves for the Optical measurements in cupra B were carried out as dea-anomer (rotational shift - 1,572"). scribed previously.8 For resistance measurements, cupra d The possibility of complex formation between OH solution was prepared as described by Reeves and J ~ n g .It~ contained 0.75 g. of copper, 2.91 moles of ammonia and 10 groups a t positions 4 and 6 was also investigated. ml. of ethanol per liter. Measurements were made at 22" Methyl 3,3-dimethyl-P-~-idoside was prepared but with a Leeds and Northrup conductivity meter using a dip conductivity measurements established that this type cell having a cell constant of 1.075. compound did not form a complex with cupra A. Periodate Oxidation Experiments.-Methyl 2-rnethyl-PSince equatorial C1-OCH3 groups are more D-idoside (12.97 mg.) was treated with 0.01 M periodic acid solution (10 ml.) at 22'. Samples (2 ml.) were removed at rapidly oxidized by chlorine than the same axial intervals and the remaining periodic acid determined in the grouping,'O a simple test of Reeves' earlier conusual way. The consumption of periodic acid in terms of 7' clusions was possible. If methyl a-D-idoside had theory for one mole per mole of glycoside, was: 1 hr., 467,; 17 hr., 1047'. In a similar experiment, methyl 3-methyl-p- had the 1C conformation, the CI-OCH8 would D-idoside (21.81 mg.) consumed only 5.770 of the theoretical be equatorial; similarly if methyl @-midosidewere amount after 41 hr. represented by the equilibrium mixture, C l G l C , Chlorine oxidation experiments were carried out essentially as described previously.'O Experiments were performed there would be an equilibrium between axial and equatorial positions for this same group. It was with the 01- and p-anomers at the same time and the rates of flow of chlorine were adjusted t o be as close as possible. therefore to be expected that, contrary to the usual The production of aldonic acid was determined by con- experience, the a-anomer would be oxidized a t version of the aldonic acid t o lactone and subsequent assay using a modification of the hydroxylamine method described least as rapidly as and probably more rapidly than by Hestrin." The procedure was: aliquots of oxidation mix- the @-form. tures freed from chlorine by aeration, or of a standard soluShort and long term experiments on the oxidation of sodium L-idonate (kindly provided by Professor tion of methyl CY- and p-D-idosides are shown in Reichstein) were treated with one drop of dilute phenol- Table 11. The highest yield of aldonic acid was obphthalein solution and titrated with 0.1 A' NaOH to a permanent end-point; 1 ml. of 0.5 N HC1 was added and the tained with the 0-anomer after about 21 hr. and volume adjusted t o 5 ml. The solutions were placed in a thereafter remained constant a t a level of 5.25 boiling water-bath for 15 minutes to complete lactonizatioii pinoles per ml. For complete oxidation, the forma(8) K . Bentley, THISJ O U R N A L , 81, 1952 (1959). (9) R.E. Reeves and J. R. Jung, i b i d . , 71, 209 (1949). (10) R. Rentley, i b i d . , 79, 1720 (1957). (111 S He-trin. .T. B i d Chem , 180, 249 (1949).
(12) T h e complexes formed b y axial OH groups on carbon atoms separated b y a central carbon atom u-ill be reierred to as la,3a-compiexes rather than 1 e,:3e cririiplexes (see ioiitnote 2). This change is reqiiirr
