trioses might be formed, glyceraldose and glyceroketose; thus CH 2 (OH)+CH 2 (OH) 2 ->CH 2 (OH).CH(OH).CH(OH) 2 +OH 2 CH(OH) 2 ^CH 2 (OH).C(OH) 2 .CH 2 .OH+OH 2
Both compounds are known: they are easily obtained by oxi- dising glycerol, CH 2 (OH)-CH(OH).CH 2 (OH). In solution, in presence of a trace of alkali, the one is rapidly converted into the other: consequently a solution made from either is a mixture of the two in equilibrium; but as the molecule of gly- ceraldose is asymmetric, this compound is present in two forms of opposite optical activity:
CH 2 .OH CH 2 .OH
HC.OH CH(OH) 2
HO.CH
CH(OH) 2
Fructose and sorbose, the two ketohexoses obtained in labor- atory operations, in the manner described, whether from formal- dehydrol, glycollic aldehydrol or the complex triose mixture re- ferred to, are constituted as represented by the formulae
H OH OH H OH H
CH 2 (OH).CO.C. C. C.CH 2 (OH)CH 2 (OH).CO. C. C. C. CH 2 (OH) OH H H OH H OH
Fructose Sorbose
The formation of the two isomerides is accounted for, and is indeed to be expected, on the assumption that the condensation is effected equally between glyceroketose and each of the two oppositely active forms of glyceraldose, which would necessarily be present in equal proportions: for the same reason each isomer would be produced in its two forms of opposite optical activity. It is a remarkable fact, therefore, that whereas fructose is of universal occurrence in plants, sorbose is very rarely met with: this is one of the many indications that in plants the course of synthesis is narrowly directed.
It is conceivable that if six molecules of formaldehydrol were brought into position side by side and condensation took place throughout the series, all the possible hexoses might arise, through the fortuitous arrangement of the molecules in the many possible ways. The force of the argument is lessened by the probability that affinities would come into play which would determine arrangements in particular ways; probably the number would be less than is conceivable but yet greater than it actually happens to be. This conclusion, however, but serves to confirm the argu- ment used above as to the actual course of the process: that it is essentially a live stage process. Perhaps in nature, at least, pentoses may be formed directly, but to judge from laboratory experience it is equally if not more probable that the hexoses are the only primary products and that other simple carbohydrates are derived from them: in other words that the hexoses are both primary products and reserve materials.
The preferential formation and the superior stability of the hexose system is to be referred to certain peculiarities of structure which are probably innate in the component elements. It has long been held that the aldehydic sugars are not true aldehydes and the ketonic not true ketones; they are too inert in behaviour to pass as such. The true aldehydrols and kethydrols, if present at all, enjoy but a fleeting existence in solution; their place is taken by closed chain derivation. Thus:
CH 2 .OH CH 2 .OH
HO.HC CH.OH
.HO.HC CH.CH(OH). CH 2 (OH) \/ O
As the " terminal " group concerned in this change carries two hydroxyls either of which may be active, and the group be- comes asymmetric in the course of the change a new asymmet- ric system being created in the molecules, it is to be foreseen that two isomeric compounds will arise in this way. As a matter of fact " glucose " is known to be an equilibrated mixture of an a and b form, differing in optical activity and other ways, which
CH.OH CH.OH
HC.OH ->
CH.OH c CH.OH HO.CH.OH
-C.OH CH.OH 5 CH.OH
. CH.OH
can be separated. If either be redissolved in ordinary water it soon passes over into the other until equilibrium is again reached. If hydrogen chloride be added to a solution of glucose in methylic alcohol, after a time two methyl-glucosides can be separated, an a and a ft form: HO.HC - CH.OH HO.C CH(OH)
H
\
\
\
CH 3
\
\
CH.CH(OH).CH 2 (OH) C CH.CH(OH).CH 2 (OH)
CH 3 O O HO
These are neutral, very stable compounds as compared with the parent glucoses. They are the prototypes of a large class of glucosides met with in plants and may be hydrolysed by enzymes which attack these latter. Hence it is possible to classify gluco- sides generally, in so far as they can be hydrolysed by enzymes which hydrolyse either a or /3 methylglucoside and can thus be correlated with either the a or the /3 form of glucose. The en- zyme maltase or a-glucase, present in yeasts, is used in charac- terising a-glucosides, the j3-glucase in almond emulsin in charac- terising /3-glucosides.
All sugars of the aldose and ketose types behave as described. The fructose sugars exist as condensed stable systems similar to that of glucose and, therefore, should persist, if formed when for- maldehydrol undergoes change and is converted ultimately into hexose sugars. Their non-production gives further weight to the argument that these latter are formed from the trioses.
Recently a third or -y form of methylglucoside has been found in the mixture obtained by the interaction of glucose and meth- ylic alcohol. This new glucoside is very different from the a and /3 forms; it is easily hydrolysed and easily oxidized by perman- ganate and very active in other ways. Probably it is a condensed system of the ethylenic oxide type:
CH 2 CH.OCH,
\ /\
O O-C
/ (CH.OH) 3
CH 2 CH 2 .OH
Ethylenic oxide and glucoside
The discovery is of primary importance, as it has led to the dis- covery of a similar form of fructoside and has given the clue to the nature of cane sugar, long remarkable on account of the ex- treme ease with which, in comparison with other sugars, it is hydrolysed by acids and by the special enzyme invertase. The formula suggested for cane sugar is:
i - O - ,
Glucose section CH 2 (OH). CH(OH). CH.(CH.OH) 2 CH
I O
Fructose " CH 2 (OH). CH(OH). CH(OH).CH.C.CH 2 (OH)
\S O
The difficulty in accepting this interpretation is that sugar is shown as either an a or a /3-glucoside and that it is hydrolised only by a specific enzyme, not by either a or /3-glucose. There can be little doubt that the fructose element is present in the y form; if the glucose were also in the y form the peculiar be- haviour of cane sugar towards acids would be even better explained.
Although, in the laboratory, the sugars obtained from formal- dehydrol are the two ketoses, fructose and sorbose, in the plant glucose plays the preponderating part to a remarkable extent. Only three of the sixteen possible hexaldoses and two ketoses, glucose, mannose, galactose, fructose and perbose, are met with as such or in combination in plants. Three of these are reversibly interrelated glucose, mannose and fructose. If a solution of any one of the three be made alkaline and kept, gradually the other two make their appearance. A natural process is at work which seems to assure even the rapid passage of any one of the three into the others. It has been shown that, during fermentation with the aid of the juice expressed from yeast, an enzyme phos- phetose is active, which, in presence of phosphate gives rise to a diphosphoric glucoside, C6Hi O4(PO4P 2 )2 , the result is the same whether mannose, glucose or fructose be taken, but when
