VOL. 12 (1953) EFFECTS OF INSULIN ON MELANOMA 339 



equivalent to the average room temperature) had a mean body temperature of 37.5° C 

 (range 36.50 to 38.40). 



Response of brain in heat-treated mice 



Since brain shoes of normal mice uniformly failed to respond to insulin (by elevation 

 of Q^^ ) it was of interest to determine whether brain tissues from heat-treated mice 

 bearing insulin-responsive melanomas would show an effect. In two experiments, in- 

 volving mice held for ca. 19 hours at 35 °C, insulin (4 units crystalline zinc-insulin per ml) 

 slightly inhibited (^co of brain slices (4 and 17% respectively) ; whereas tumor slices from 

 the same animals showed marked stimulation (46 and 45% respectively). A third 

 experiment involving a melanoma-bearing mouse maintained at room temperature gave 

 similar results except that the percentage stimulation in the tumor was not so large 

 (brain —8%, tumor 18%). 



The influence of inorganic zinc 



Although crystalline insulin contains zinc [ca. 0.7%) the metal can be largely 

 removed by dialysis and other means. However, since there are considerable stores of 

 zinc in the body it is difficult, from in vivo experiments, to assess the requirement for zinc 

 in insulin function. Nevertheless, because of the presence of zinc in freshly isolated 

 insulin, the capacity of insulin crystals to combine with large quantities of zinc^^, and 

 the relation of zinc content to the duration of insulin action following the injection of 

 such crystals, it has been suggested that this element may have a special in vivo relation 

 to action of the hormone^^. 



The present results secured with the S-91 melanoma also suggest that zinc plays a 

 special role in insulin action, and that the relative concentrations of the two substances 

 are critical. Thus in a number of experiments with the S-91 melanoma, employing low- 

 zinc insulin, crystalline zinc-insulin, and inorganic zinc it was found that the tumors fell 

 into a graded pattern of response. This is best illustrated by comparing the two extremes 

 in the response pattern (Table IV). In experiments A and B (Table IV), each involving 

 slices of different individual tumors, the addition of inorganic zinc (as ZuSO^-yHgO) 

 at concentration of i.i y* to 8.0 y per ml (as zinc ions) did not cause significant changes 

 in the (2c6 • However, both low-zinc insulin, and crystalline zinc-insulin increased the 

 Q^o^- When i.i y of inorganic zinc were added per 4 units of the low-zinc insulin, the 

 (Jcoj was considerably higher than with insuhn alone (Table IV, expts A and B) . On the 

 other hand when 8.0 y of inorganic zinc were added per 4 units of low-zinc insulin, the 

 stimulating effect of the insulin was largely abolished (Table IV) . That crystalline zinc- 

 insulin was not always equivalent to low-zinc insulin plus added zinc is indicated by the 

 data in Table IV, experiment B. 



In contrast to tumors of the type just described, those that lay at the other extreme 

 of the response pattern (Table IV, expts C and D) showed inhibition of (^co^ when either 

 concentration of inorganic zinc alone was added. Furthermore, in the same tumors 

 crystalhne zinc-insulin was markedly less stimulatory than low-zinc insulin. The addition 

 of even the lowest concentration of inorganic zinc (i.i y/ml) to low-zinc insulin practically 

 abolished the insulin-induced stimulation of acid formation of these zinc-inhibited 

 tumors. It might be concluded from the data (Table IV, C and D) that insulin counteracted 



I.I y of zinc is approximately equivalent to the zinc content of 4 units of crystalline zinc- 

 insulin. 



References p. 346. 



