MECHANISM OF THE DIABETOGENIC ACTION 415 



ably chelates Zn++ and other ions much better than alloxan (page 378) 

 (Resnik and Cecil, 1956). Unfortunately, there has been no thorough in- 

 vestigation of the effects of alloxan on the metalloenzymes so that we do 

 not know if alloxan can complex with a protein-bound metal ion. 



(F) Reaction with ^-cell membranes. The rapid and irreversible cell damage 

 seen with alloxan might be more readily interpreted as an effect on the 

 /?-cell membranes whereby their permeability or integrity is altered. Dixon 

 etal. (1960 b) observed that alloxan causes an early separation of the /?-cells 

 from each other and the surrounding capillaries, and suggested that al- 

 loxan might react with a layer of protein on the surface of the cells. Wat- 

 kins et al. (1961) studied the permeability of toadfish islet cells to mannitol- 

 C^^ as a measure of the integrity of the plasma membrane, since mannitol 

 normally does not penetrate rapidly. Alloxan at 0.25 mM was found to 

 double the rate of mannitol penetration. Alloxanate is inactive. The per- 

 meability of kidney, heart, brain, gill, muscle, and liver cells is not altered 

 by alloxan. These data are thus consistent with the idea that alloxan 

 selectively alters the /?-cell membrane. The membrane component with 

 which alloxan acts is, of course, unknown, but it might be involved in the 

 release of insulin since the /5-cells are specialized for this function. 



{G) Release of trypsin from the exocrine tissue. Ligation of the pancreatic 

 duct leads to atrophy of the pancreatic parenchyma without affecting the 

 islets, and simultaneously there is a reduction in the susceptibility to 

 alloxan (page 408) (Walpole and Innes, 1946). This might imply that the 

 acinar tissue is involved in the diabetogenic action, perhaps by causing 

 the release of trypsin which damages the /5-cells. However, trypsin injec- 

 tions do not disturb the /5-cells. Urinary diastase excretion is essentially 

 abolished by alloxan and this might also be interpreted as an effect on the 

 acinar tissue (Bernhard et al., 1947). Ethionine administration for long 

 periods produces resistance to alloxan and causes a fibrosclerosis of the 

 exocrine tissue (Gambassi and Del Gatto, 1956). Pancreatic duct ligation 

 leading to parenchymal atrophy does not prevent the development of 

 diabetes from alloxan in rats, but prevents it in rabbits although the 

 /5-cells are damaged, so that it was concluded that trypsin release is not 

 an important factor (Ries and Allegretti, 1957). It is difficult to understand 

 how the /5-cells could be eliminated in rabbits without the appearance 

 of the diabetic state. One notes also that injury to the /5-cells is not ob- 

 served in vitro, even at high concentrations of alloxan (Becker et al., 1962). 

 The significance of these observations cannot be evaluated at this time. 



Progress in this field must await more detailed investigations of the 

 metabolic and functional peculiarities of the /?-cells. This has been started 

 in the work of Humbel and Renold (1963) on the glucose and amino acid 

 metabolism of cultured toadfish islet tissue, wherein they showed that 



