Arabinose 2,5-anhydro

In an attempt to synthesize 2-desoxy-L-ribose, Mukherjee and Todd68 treated methyl 2,3-anhydro-jS-L-ribopyranoside (LXXII) with sodium thiomethoxide and then reduced the amorphous product. The sulfur-free sirup thus obtained was inert to the action of periodate and is therefore methyl 3-desoxy-/3-L-riboside (ayn., methyl 3-desoxy-/3-L-xyloside) (LXXIV) rather than the desired 2-desoxy derivative. The methylthio intermediate (LXXIII) was consequently a 3-methylthio-L-xylose derivative rather than a 2-methylthio-L-arabinose compound. The presence of a trace of the latter, however, is not wholly excluded since the sirupy desoxypentoside (LXXIV) gave a feeble green coloration in the Keller-Kiliani test, a reaction generally accepted as being specific for 2-desoxy-sugars. 

In the 3,4-anhydro-arabinose series, only the hydrolysis of the anomers of methyl 2-0-acetyl-3,4-anhydro-D-arabinopyranoside with aqueous acetic acid appears to have been studied 68 both compounds undergo substitution at C-3, with participation of the acetoxyl group, leading to the corresponding D-lyxose derivatives. 

In sharp contrast to the behavior observed when the three dithio-acetals aforementioned are treated with p-toluenesulfonyl chloride in pyridine, dialkyl dithioacetals of D-arabinose, treated under the same conditions, are converted into the corresponding 5-p-toluene-sulfonates, generally isolable crystalline in high yield.72 This remarkable difference has been interpreted72" on conformational grounds the D-arabinose dithioacetals are stable in the extended, planar zigzag conformation, whereas the other three examples experience some destabilization in the extended form, because of parallel 1,3-interactions.726 Furthermore, the transition state for closure of the 2,5-anhydro ring would be quite strained in the D-arabinose series, but not in the other three.72"... 

Inspection of molecular models reveals that 2,5-anhydro-arabinose and -lyxose are the only 2,5-anhydroaldopentoses wherein the simultaneous presence of a 1,4-furanoid ring is sterically possible. 

Several 2,5-anhydropentitols have been synthesized from suitably protected pentoses by standard procedures (see also, Ref. lb). Thus, ethyl 5-O-p-tolylsulfonyl-a-L-arabinofuranoside (11) was converted17 into the corresponding 2,5-anhydride (12) by treatment with metha-nolic sodium methoxide. The product was hydrolyzed, and the resulting 2,5-anhydro-L-arabinose was hydrogenated in the presence of Raney nickel to give 2,5-anhydro-L-arabinitol (13) as a syrup. 

The enantiomorph, also a syrup, was synthesized from 3,6-anhydro-4,5-O-isopropylidene-D-mannitol18 by periodate oxidation to the corresponding D-arabinose derivative, followed by reduction in the presence of Raney nickel.17 2,5-Anhydro-D-xylitol, 2,5-anhydro-D-ribitol, and 2,5-anhydro-D-lyxitol (1,4-anhydro-D-arabinitol)19,20 were... 

Desulfonylation of 2-O-p-tolylsulfonyl or 2-0-(methylsulfonyl) derivatives of D-arabinose, D-xylose, and L-fucose with aqueous barium hydroxide yields D-ribose, D-lyxose, and 6-deoxy-L-talose, respectively, by way of a 1,2-anhydro sugar similarly, 3-0-(methyl-sulfonyl)-D-fructose affords D-psicose. 

That not all of the xylans of wood are homogeneous polymers of anhydro-D-xylose units has been shown by the classical studies of O Dwyer (1923 to 1940) on hemicelluloses of American white oak. O Dwyer prepared a hemi-cellulose fraction from water-extracted, oakwood sawdust by extraction for two days with 4% aqueous sodium hydroxide solution. The polysaccharide material was obtained, after acidification, by the addition of ethanol. The product ([a]n —75° in 1 % sodium hydroxide), contained 70% of pentosan, and yielded n-xylose, n-maiinose, n-galactose, and L-arabinose on hydrolysis. 

Fehling solution. It contained anhydro-D-xylose units and 11% of it was uronic acid anhydride. D-Glucuronic acid was identified in hydrolyzates, but a test for D-galactose was negative. Hemicellulose-B (Hd —120° in 0.5% sodium hydroxide) yielded 4.8% of galactose (galactaric acid test), an undetermined amount of L-arabinose, and 64 % of uronic acid identified as galacturonic acid. ... 

Although Fischer described a lactose phenylosazone anhydride in 1887, little attention was paid to substances of this type until 1935 when Diels and Meyer reported the isolation of monoanhydrides of D-glucose phenylosazone, D-galactose phenylosazone, D-xylose phenylosazone, L-arabinose phenylosazone, lactose phenylosazone, and cellobiose phenylosazone, as well as a dianhydride of maltose phenylosazone, by boiling the corresponding osazones in alcoholic solution with a little sulfuric acid. Because of the apparent identity of the D-glucose phenylosazone anhydride with the 3,6-anhydro-D-glucose phenylosazone of Fischer and Zach, Diels and Meyer formulated these compounds as 3,6-anhydrides. 

---