386 4. ALLOXAN 



and it may be noted that ninhydrin is nondiabetogenic. Lazarow (1954 a) 

 has given a full discussion of the structural requirements for diabetogenic 

 activity, but at present it is impossible to formulate a specific grouping 

 which will produce diabetes. Certain data indicate the importance of the 

 — NH— CO— NH— CO— moiety and its enolized tautomer — N=C(OH)— 

 — N=C(OH) — but a number of compounds have this grouping and are 

 inactive. Little attention seems to have been paid to the possibility that 

 the quinonoid structure of alloxan is the reactive form. It may be noted 

 that none of the inactive compounds can assume such a structure. 



Effects of Alloxan on Tissue Thiols 



A relationship between the diabetogenic action of alloxan and its reac- 

 tion with glutathione or other thiols has been postulated, particularly by 

 Lazarow (1954 a). There is no doubt that alloxan injected intravenously 

 brings about a profound decrease in blood glutathione. Leech and Bailey 

 (1945) showed that the injection of 200 mg/kg of alloxan into rabbits 

 reduces the blood glutathione 50-100%, the maximal effect occurring at 

 1-2 min, and from the parallel disappearance of alloxan and glutathione 

 they concluded that reaction between them took place. This was confirmed 

 by Briickmann and Wertheimer (1947) in the rat. However, reduction of 

 blood glutathione cannot be the mechanism of the effect on the /5-cells. 

 For one thing, the glutathione is mainly in the erythrocytes and its loss 

 there would not so rapidly affect other tissues, and also several substances 

 lower blood glutathione and do not cause diabetes, examples being ninhy- 

 drin (Briickmann and Wertheimer, 1947), iodoacetate, and iodoacetamide 

 (Hultquist, 1958). The effects on blood glutathione merely demonstrate the 

 ability of alloxan to react rapidly with glutathione. One might expect that 

 this is a simple oxidation of glutathione, but Bhattacharya et al. (1956) 

 showed that alloxan lowers only GSH, the GSSG level remaining unaffected, 

 so that there is a loss of total glutathione. This points to a combination of 

 alloxan with glutathione, possibly to form Complex 305 or a related sub- 

 stance. 



Of more importance are the effects of alloxan on tissue thiols and par- 

 ticularly the changes possibly occurring in the /?-cells. DeCaro and Rovida 

 (1937) reported that alloxan produces moderate falls in liver and intestinal 

 glutathione within 10 min. There is a decrease in liver and kidney glutathione 

 during severe diabetes resulting from alloxan, but these results were ob- 

 tained at 48 hr and, as Houssay et al. (1947) pointed out, tissue thiols fall 

 after pancreatectomy. Thus only the early changes are of significance 

 for our purpose. Brada (1951) determined the glutathione level in various 

 tissues of the rat following injection of alloxan (see accompanying tabula- 

 tion) and believed the drops in adrenals, pancreas, and liver to be significant, 

 although it seems likely that the renal effect is also real. Except for the adre- 



