CHELATION 481 



of toxicity at minimally fungistatic doses. First, copper oxine is as toxic 

 as, or even more toxic than, oxine itself (281, 313, 424). The same re- 

 lation holds for the metal complexes of other chelating agents (272, 372, 

 395, 433). Second, there is some evidence that toxicity to bacteria and 

 fungi is less in a medium low in metals than in the usual medium (5, 

 9, 14, 358). Finally, successful use of chelating agents to supply metals 

 both to fungi (Chapter 9) and to higher plants (143, 401) suggests that 

 cells in general can compete successfully for metals, with even quite 

 stable complexes. 



Metal starvation cannot, however, be dismissed completely as a factor 

 in toxicity. Chelating agents reduce the metal-dependent spore pig- 

 ment of Aspergillus niger (338); EDTA prevents spore germination of 

 Neurospora tetrasperma without apparently entering the cell (406), 

 and citrate effects on enzymatic adaptation in Azobacter spp. can at 

 least be interpreted as a depletion phenomenon (368). Inhibition of 

 sporulation by chelating agents (179, 428) suggests that a higher metal 

 requirement makes sporulation especially sensitive to metal depriva- 

 tion. 



A number of observations, covered in detail in the original literature 

 (9, 31, 32), support a plausible working hypothesis for the action of 

 oxine and its metal complexes. The hypothesis starts with the equi- 

 libria already described (Equation 5) of oxine, its charged 1:1 chelate 

 and the neutral 1:2 chelate. It is believed that the 1:2 chelate, perhaps 

 because of its higher lipid solubility, enters the cell more rapidly than 

 the 1:1 chelate and in that sense is the active species. However, within 

 the cell it seems likely that the 1:1 chelate is the more toxic, perhaps by 

 reacting directly with essential metal-binding sites or with sulfhydryl 

 groups essential to the cell. This hypothesis has the advantage of ex- 

 plaining the bimodal dose-response curve of oxine (6, 31, 281), which 

 is similar in principle to that of the dithiocarbamates (Figure 11), and 

 the complex interactions of metal and oxine concentrations (32). Still 

 to be demonstrated, however, are the roles of free metal ions and free 

 oxine, the site of the proposed toxic reaction, and the role of natural 

 chelating substances within and on the cell. Also, as with all perme- 

 ability hypotheses, the lack of direct evidence is distressing. 



8-Hydroxyquinoline-5-sulfonic acid is a chelating agent but is not 

 antibacterial (6). It fails to inhibit Monilinia fructicola but is active 

 against Macrosporium sarcinaeforme (177). It seems possible that the 

 anionic group acts by preventing access of the molecule to the site of 

 action (6); on this hypothesis, M. sarcinaeforme must be more perme- 

 able to the compound than are bacteria and M. fructicola. 



The action of the dithiocarbamates and their derivatives may be ex- 



