Allitol

Fig. 2. Structures of hexitols (a) allitol, (b) sorbitol (D-glucitol), (c) D-maimitol, (d) dulcitol, (e) L-iditol, and (f) D-altritol.

Very little was published on the synthesis of hexitols from unsaturated intermediates between the time of Griner s work and 1933, when it occurred to me that it might be possible to add four hydroxyl groups to divinylglycol by means of a solution of silver chlorate containing a small amount of osmic acid. In carrying out this work I had the assistance of one of my students, Joseph Wiemann. We succeeded far beyond our expectations and obtained allitol, unknown up to that time, and d,l-mannitol. 

Another method of synthesis was also used. This involved the action of chloroacetaldehyde on the Grignard reagent derived from acetylene in order to obtain the meso divinylacetylene dichlorohydrin, CH2CI—CHOH—C=C—CHOH—CH C1, from which one passed to the corresponding hexynetetrol, CH2OH—CHOH—C=C—CHOH— CHjOH. This, in turn, was reduced to the hexenetetrol, CHjOH— CHOH—CH=CH—CHOH—CH2OH, by means of Bourguel s catalyst,8 a dispersion of colloidal palladium on starch. When the hexenetetrol was hydroxylated by the use of silver chlorate and osmic acid, two hexitols, dulcitol and allitol, were obtained. 

It may be well to state at this point that the reagents used in any of the syntheses which follow are sufficiently mild to preclude the possibility of rearrangement of the hydrogen atoms on the carbon chain. Thus in the compounds which lead to allitol, those hydrogen atoms which are attached to secondary carbon atoms all lie on the same side of the carbon chain, just as they do in allitol. 

Hence, the precursors of allitol are the meso form of divinylglycol in the first of the above syntheses, and the meso form of the hexynetetrol and hexenetetrol in the second method erf synthesis. 

An amount of 16 g. of the glycol (m. p. 18°) dissolved in 250 ml. of water was oxidized with 18 g. of silver chlorate and 0.3 g. of osmic acid. The reaction mixture yielded 3 g. of allitol and no D,L-mannitol. We may therefore assign the meso configuration to the divinylglycol melting at 18°, since on hydroxylation it yielded allitol, but not D,L-mannitol. 

The mother liquor was then treated with a small amount of ether, whereupon it separated into two layers. The lower layer consisted of a black sirup, which, when treated with absolute alcohol very slowly deposited some crystals, 1 g., consisting largely of D,L-mannitol. The upper layer was treated with a large volume of ether, which caused the precipitation of a considerable quantity of a brown sirup. Crystallization began almost immediately, especially when the sirup was treated with absolute alcohol. The crystals were collected and there was obtained in this way 12 g. of crude allitol. After solution in water and precipitation with alcohol, the crystals melted at 140°, but still contained a small amount of impurity. After several recrystallizations they melted at 149°. 

Upon standing, the mother liquor deposited more allitol—some 3 g. after a period of two months. No crystalline material was obtained from the ethereal solution. We thus obtained from 100 g. of divinylglycol, 11 g. of D,L-mannitol and 15 g. of allitol. 

Allitol was characterized by its melting point, 149-150°, elementary analysis, and by the formation of a dibenzylidene derivative, which, after recrystallization from alcohol, melted at 249-250° on the Maquenne block. Later, Steiger and Reichstein7 repeated this synthesis of allitol and demonstrated the identity of the product with that obtained by the reduction of D-allose. 

The dulcitol hexaacetate mother liquor yielded a small quantity of crystalline material, m. p. 50-59°, whose nature has not been determined. Its melting point is near that of allitol hexaacetate (m. p. 61-62°).w... 

Microwave-assisted epoxide ring-openings of l,5 2,3-dianhydro-4,6-0-benzyl-idene-D-allitol with nucleobases have been reported [218], Various rapid microwave-assisted protection and deprotection methods in the area of carbohydrate chemistry are known [210], and two general review articles on microwave-assisted carbohydrate chemistry were published in 2004 [219, 220]. 

The effect of an ally lie hydroxy group was first observed in divinylglycol (1,5-hexadiene-cA-3,4-diol and 1.5-hexadiene-/raw.v-3,4-diol). It was shown that the hydroxy substitutions directed the addition of the osmium tetraoxide to syn addition, so that the cA-diol yielded allitol (all cA-hexaol) and the iraws-diol yielded mannitol42. The oxidation of the dienol 35 yielded a lactone ring 36 by cA-dihydroxylation and transesterification... 

On treatment with concentrated halogen acids, certain hexitols yield 1,6-dideoxy-l,6-dihalo compounds. The structure of the compound so obtained from galactitol, first reported by Bouchardat,404 was later verified by synthesis.405 Allitol is transformed406 into 1,4-anhydro-6-chloro-6-deoxy-DL-allitol and l,4-anhydro-5,6-dichloro-5,6-dideoxy-DL-talitol on treatment with fuming hydrochloric acid at 100°. 

Starting with the same acetal (35), but proceeding by way of the diacetate 40, instead of the dibenzoate 36, it was possible to prepare,51 in high yield, the p-toluenesulfonate 41 by sequential trityla-tion and p-toluenesulfonylation. Compound 41 was then subjected to solvolysis in the presence of sodium acetate in moist N,N-dimethyl-formamide, and the product (42) was successively acetylated and de-tritylated to give 3,4,6-tri-0-acetyl-2,5-anhydro-D-allitol (43). The... 

The stereochemistry of the hexitols affects the manner of ring fusion, as is seen in Figs. 3 and 4. In mannitol, sorbitol (gulitol) and iditol, the l,4 3,6-dianhydro rings are m-oriented, whereas in the dulcitol, allitol and altritol (talitol) series the 1,4 3,6-rings are trans-oriented. This makes the ring system in this latter series almost planar although in the case of dianhydrodulcitol and allitol there is considerable strain in the molecules. 

It is evident that syntheses of the Lespieau type can lead to a great number of polyhydroxy compounds. Wiemann has prepared compounds such as CHsCH2(CHOH)6CH2CH3, which he calls diethyl mannitol. It would seem better to designate this substance a tetra-desoxy decitol or decane hexol since its configuration is not known with certainty. Apparently this type of synthesis leads to symmetrical arrangements of hydroxyl groups, since allitol, dulcitol and D,L-mannitol are the only hexitols that were identified as products. 

When the dibromodideoxy-o-mannono-1,4-lactone (2) (Scheme 7) was treated with aqueous ammonia the 3,6-dideoxy-3,6-imino-D-allonamide (22) was formed as the only product [20]. Conversion to the ethyl ester and subsequent reduction of the ester with sodium borohydride transformed 22 into the known l,4-dideoxy-l,4-imino-L-allitol (23) [20]. 

Investigation of the inhibitory effects on glycosidases has been carried out for the hydroxylated pyrrolidines [65] and piperidines [73]. Among the pyrrolidines prepared only the l,4-dideoxy-l,4-imino-L-allitol (23) (Scheme 7) and its C-5 isomer showed any remarkable effects. They inhibited lysosomal a-manno-sidase rather than the processing a-mannosidases I and II [65], and their specificity is in accord with the structural requirements of azafuranose analogues of mannose for inhibiting mammalian a-o-mannosidases [80]. 

This elegant synthesis also permits,94 by reduction of the nitrile 89, the synthesis of l-amino-2,5-anhydro-l-deoxy-D-allitol (91),... 

Baddiley and coworkers13 further found that, under the same conditions (100° and 2 M hydrochloric acid), allitol (9) and D-altritol (D-talitol) (10) are converted into anhydrides in high yield, as shown... 

The synthesis of isomeric DL-allitol succeeded when the hydroxyl groups in cts,cis-2,4-hexadiene-l,6-diol were protected with benzoyl or mesyl groups. Epoxidation gave the appropriate derivatives of 2,3 4,5-dianhydro-DL-allitol (16%), which were then reduced with lithium aluminum hydride to DL-crt/f/iro-2,5-hexanediol. 

---