658 DIABETES 



There is also an 1-glucose series in which the members are the mirror images 

 of the above. There is a third series comprising d-galactose, d-talose, d-tagatose, 

 1-sorbose, 1-idose, 1-gulose and alpha and beta d-galtose; and a fourth series whose 

 relationship to the d-galactose series is the same as that of the 1-glucose to the 

 d-glucose series. Consideration of the d-glucose series will bring out the prin- 

 ciples common to all. Examination of the 8 formulae shows that numbers 1, 2, 

 4 and 5 have aldehyde groups (H — C =0) at the end of the chain and are hence 

 aldohexoses. Members 3 and 6 have ketone (C=0) groups, at the second car- 

 bon atom, and are therefore 2-keto-hexoses, while numbers 7 and 8 having ketone 

 groups at the third carbon atom are 3-keto-hexoses. Since each series of 8 sugars 

 has a like number of the different types there are in all 16 aldohexoses, eight 2- 

 keto- and eight 3-keto-hexoses. 



It had long been known that if a solution of any optically active sugar, such 

 as d-glucose, was alkalinized the solution gradually lost its optical activity. It 

 was later shown by Lobry de Bruyn and van Ekenstein that a solution of d- 

 glucose in very dilute alkali comes to contain a group of 4 hexoses in dynamic 

 chemical equilibrium. ^'^ Nef^' held that in such solutions there is an equili- 

 brium of at least eight hexoses as above depicted. Any one of these sugars when 

 placed in alkali reproduces the other seven, since the members of the series are 

 reciprocally convertible one into another. The same holds good for the members 

 of the d-galactose series and for the 1-galactose and 1-glucose series. But the 

 reciprocal transformation of the members of one series, such as d-glucose, into a 

 hexose of another series, such as d-galactose, occurs, if at all, only to a minute 

 degree, because such transformation involves the breaking of the hexose chain into 

 2, 3 and 4 carbon fragments with subsequent recombinations, and when this occurs 

 irreversible reactions are prone to intervene. The formation of lactic acid and 

 the saccharines are representative of these irreversible reactions. (In the body, 

 however, glucose may be converted into lactic acid in the muscles and elsewhere, 

 whereas in diabetes lactic acid can be converted easily back into glucose; in diabetes 

 galatose is convertible into glucose, etc., so that in the body the transformation 

 of hexoses of different groups one into another offers no difficulty.) 



In order to explain the effects of dilute alkali on hexoses just described, some 

 conception of labile intermediate products is a logical necessity, for when levulose 

 changes into glucose there is necessarily some intermediate phase. The nature of 

 these phases has been the subject of study by many chemists, and this study 

 involves always the question of sugar dissociation. 



Sugars are weak acids. They form salts with metals, and Cohen-^ and later 

 Michaelis and Rona^" have determined by physico-chemical methods the ioniza- 

 tion constants for glucose and other sugars. Sugars are also polyatomic alcohols, 

 and either aldehydes or ketones. Sugar chemistry reverts to the chemistry of 

 these three classes of compounds. 



A. P. Mathews'" and Michaelis have suggested that the effect of alkali on a 

 sugar such as glucose is to increase enormously the concentration of the glucose 

 anion, i. e., KOH leads to the formation of K-glucosate (see Fig. 3, p. 21), which, 

 being the combination of a powerful base with a very weak acid, has a high elec- 

 trolytic dissociation constant. These anions according to this view are sub- 

 ject to cleavages and intramolecular rearrangements. Nef also hokls that the 

 first effect of the alkali {e. g., KOH) is to form a salt, but his far more detailed 

 conception of the subsequent changes which lead to reciprocal transformations 

 of hexose sugars one into another, involves other i)rinci]>les which represent the 

 outgrowth of his earlier work on the properties of simpler aldehydes and ketones. 



These reciprocal transformations are dependent, according to Nef, upon the 

 aldehydic or ketonic character of the sugar, whereas the oxidative phenomcnaand 

 the saccharinic acid formation to be described presently, tlepend upon the alcohol 

 groups. The principles involved can best be understood if we first consider the 

 behavior of a simi)le aldehyde (acetaldehj^de) and a simple ketone (acetone). 



26Rec. trav. chim. de Pays Bas (14), 158 and 203; (15), 92; (16), 257; (19), 

 1 and 10. 



"Liebig's Annalen, 1907 (357), 294; 1910 (370), 1. 



28 Zcit. f. i)hysikal. Chem., 1901 (36), 09. 



29 Biochem. Zeit., 1912 (47), 447. 

 3" Jour. Biol. Chem., 1909 (6), 1. 



