REACTIONS OF THE ARSENICALS 613 



why the ring structure is more stable than the linear one for the thioarsinites 

 has not been completely answered. In fact, whether such increased sta- 

 bility occurs has been questioned, since there is no resonance which can 

 account for it (Eagle and Doak, 1951). Metal chelates possess special sta- 

 bility in part, because of entropy factors and a similar explanation may 

 be involved here. Also, the steric factor must play a role in the case of 

 most monothiols attaching to a single arsenic atom in pairs; in addition, 

 electrostatic repulsion could contribute if ionized groups occur on the thiol. 



Reactions with Proteins 



There has been very little quantitative work on the binding of arseni- 

 cals to different types of protein. Indirect evidence indicates that arseni- 

 cals often react with proteins, but the mechanism of the binding is un- 

 known. For example, the fact that seralbumin decreases the ability of 

 oxophenarsine to inactivate trypanosomes shows that a complex is formed 

 (Toyoshima, 1952). When radioactive arsenite is injected into rats, the 

 relatively large amounts in the erythrocytes appear to be bound to the 

 globin portion of hemoglobin, with little attached to the stroma (Hunter 

 and Kip, 1941; Hunter et al., 1942). However, in other species much less 

 occurs in the blood and little seems to be bound to either hemoglobin or 

 serum proteins. If the liver proteins are fractionated on the basis of solu- 

 bility, the arsenite is found not to be uniformly distributed between the 

 various fractions (Lowry, et al., 1942). When rabbit blood is incubated 

 in vitro with oxophenarsine, more arsenical is bound to hemoglobin than 

 to the serum proteins or stroma; however, relatively more is bound to 

 urease and the binding is about 3-fold tighter, suggesting the importance 

 of SH groups (Fink and Wright, 1948 b). The protein SH groups are lowered 

 about 80% in renal cortex, medulla, and the distal tubules following incu- 

 bation of kidney sections with rather high concentrations of arsenite, the 

 effect being as marked as with HgClg or iodoacetate (Cafruny et al., 1955 a). 

 These types of result, important as they are in various ways, do not provide 

 much information on how the arsenicals react with proteins. 



What evidence is there that arsenicals can react with protein SH groups? 

 Rosenthal (1932 b) incubated various proteins with 1 mM oxophenarsine, 

 and determined how much of the arsenical appeared in a dialysate (see 

 accompanying tabulation). None of the purer proteins in their native state 

 bound appreciable arsenical, but did so upon denaturation, which was 

 interpreted as due to the release of SH groups. It was also shown that the 

 nitroprusside reaction for SH groups declines and disappears when oval- 

 bumin is incubated with oxophenarsine. This is not due to oxidation be- 

 cause no oxygen is taken up and cyanide does not restore the SH groups. 

 Rosenthal calculated that 1 g crystalline ovalbumin binds 12.72 mg 

 oxophenarsine; this is about 2.4 moles of arsenical per mole of protein, 



