Toxic isosteres

The cells could produce even more lysine if the kinase were to remain active but insensitive to a high lysine concentration. Mutants of the homoserine auxotrophs of C. glutamicum were isolated with this property by growing the organism in the presence of toxic isosteres (close structural analogues) of lysine (e.g. S-(2-aminoethyl)-L-cysteine). One way in which the cells can become resistant to the toxic isostere is to overproduce lysine. This is likely to occur in cells where the mutation alters aspartyl kinase in such a way as to make it insensitive to inhibition by lysine, while allowing it to retain its full catalytic activity. 

Substances that are isosteric equivalents of substances that are toxic or pharmacologically active may also possess these biological properties. It is also possible that biological properties may be bestowed, exacerbated, or attenuated when isosteric modifications are made. This point is illustrated by the following examples. 7-Methyl-benz[a]anthracene (36) is a known carcinogen, whereas 7-methyl-l-fluorobenz[o]... 

Medicinal chemists have used isosterism for the design of safe, effective drug substances for many years. During the development of anti-ulcer medications, for example, it was found that metiamide (38) greatly reduced acid secretion in the gastrointestinal tract by antagonizing H2-receptor sites. Its potential as auseful anti-ulcer medication was lessened by adverse effects caused by the thiourea moiety, a toxicophore (Table 4.1). This moiety is essential for H2-receptor blockade, but bestows toxicity. 

Isosteric replacement of the thiourea moiety with the cyanoguanidine moiety gave cimetidine (39), a potent H2-receptor antagonist that lacks the toxicity of 38. Cimet-idine is a widely used anti-ulcer medication because of its effectiveness in treating ulcers and relative safety. It is noteworthy that in this example, this isosteric modification selectively reduced toxicity without affecting pharmacological activity. 

As for the design of safer drug substances, isosteric substitution has been used for many years for the design of safer pesticide substances, although to a lesser extent. Isosteric replacement of carbon with silicon, for example, has resulted in a number of safe but effective pesticides [76]. An example of the use of isosteric substitution of carbon with silicon in the design of safer a pesticide is in the case of the insecticide MTI-800 (40) and its much less toxic silane isostere, 41. 

One type of isosteric substitution that should have tremendous potential in the design of safer commercial chemicals is replacement of hydrogen with fluorine. Since many bioactivation mechanisms of chemicals involve cytochrome P450-mediated hydrogen atom abstraction to yield toxic metabolites, it would seem plausible that replacement of such a hydrogen atom with fluorine would provide a safer isosteric analog, without affecting commercial efficacy. 

The data shown in Table 2 illustrate the general paucity of comparative toxicity data within an isosteric series of chemicals. In this Table a variety of toxic end-points observed for benzene and naphthalene have been compared with those of their simple heterocyclic analogues, and it is clear that it is almost impossible to derive chemical structure-biological activity relationships from the published literature for even such a simple series of compounds. Even basic estimates of mammalian toxicity such as LD50 values cannot be accurately compared due either to the absence of relevant data or the noncomparability of those available. Thus in a field where there are little comparative data on the relative toxicity to mammals of pyrrole, thiophene and furan for example, it is difficult to relate chemical structure to biological activity in historical heterocyclic poisons such as strychnine (3) and hemlock [active agent coniine (4)]. 

In theory, an isosteric/isoelectronic aromatic ring replacement within a toxic or therapeutic molecule should lead to a retention of biological activity. The fact that this is sometimes not observed may mean either that the activity is heterocycle-specific or that unforseen changes in metabolism, biological distribution and excretion, partition coefficient or stability have accompanied the molecular change. In such cases, activity may be observed in in vitro assays while not being observed in vivo (cf. the possible case of the dithiin analogue of TCDD in Table 3). 

---