746 6. ARSENICALS 



be partly explained by the effect of the As = group on the pK„ of the COOH 

 group. It is noteworthy that the presence of these acidic groups does not 

 alter the acidic properties of the As=0 group. The interesting series, 

 wherein the length of the chain bearing a terminal COOH group is increased, 

 shows an unexpected rise in trypanocidal activity in the butyric and va- 

 leric derivatives, without simultaneous changes in the toxicity, a fact 

 taken advantage of in developing compounds (e. g. butarsen) useful in 

 trypanosomiasis. If the differences here are a matter of penetration, the 

 permeability properties of these membranes are indeed unique. A further 

 complication is the possibility that the acidic derivatives are not active 

 themselves but are changed into active substances in the cells. The evidence 

 for this is as follows (Eagle and Doak, 1951): (1) the rate of excretion 

 of the acidic compounds is rapid at first but then levels off suddenly, 

 indicating a change to some other type of substance; and (2) death of the 

 animals is delayed relative to other derivatives. The low parasiticidal activ- 

 ity may be due to the fact that the protozoa cannot bring about this 

 change, whereas the host tissues can. Both of these facts may be explained 

 in other ways, however, since a slow penetration into cells with subsequent 

 tight binding would account for the rapid initial excretion (as long as free 

 arsenical remains) and the delayed lethal action (one might expect that 

 the more readily a derivative penetrates, the faster would be the toxic 

 or lethal action). Such results also argue against the simple assumption 

 that the negative charge is the major factor in determining the activity 

 here. It is true that a rough correlation may be made between activity 

 and the pK^ of the substituted group, but there are exceptions; e. g., the 

 3-NO2-4-COOH derivative is more acidic and the activity is greater (see 

 page 777 for the effects of pH). It is interesting also that the effect of adding 

 an NH2 group adjacent to the COOH group largely cancels the effect of 

 the acidic group. This might be explained on the basis of the amphoteric 

 nature of the disubstituted derivative and an over-aU neutral charge at 

 physiological pH, or of the reduction in acidity of the COOH group induced 

 by the vicinal NH3+ group; however, this will not explain the decreased 

 toxicity so readily. 



The polysubstituted derivatives present some interesting and unexpected 

 behavior. No general additivity rules can be formulated; i. e., from the ef- 

 fects of the single substituents it is impossible to predict accurately the 

 activity of the disubstituted derivative. Furthermore, it is not just a matter 

 of the groups themselves but of their relative positions on the ring, as may 

 be seen in the case of the aminophenols and aminobenzoates. Even where 

 only one type of group is involved the effect cannot be predicted, as in the 

 3,4-diOH derivative, which is even more toxic than phenylarsenoxide, 

 despite the fact that both the 3-OH and 4-OH derivatives are much less so. 

 It is true that certain rough rules may be formulated, e. g., the addition of 



