with glucuronic acid or acyl coenzyme A are also thought to be a source of
reactive metabolites.
The study of the interaction of drugs with cellular components can be
broken down into two types of experiments, either measurement of covalent
binding after reaction with nucleophilic sites on cellular macromolecules or
studies with small molecule nucleophiles. These two types of experiments
provide complementary information and are probably both important
determinants in the overall toxicology associated with reactive intermediates.
Covalent binding studies usually require radiolabeled drug and are thus
usually carried out late in the discovery phase or in early development. A
typical experiment is done either with labeled drug in microsomes or by
administering the labeled drug to rodents. Both types of studies involve
isolation of cellular proteins through precipitation, followed by extensive
washing steps to remove noncovalently bound drug. Residual radioactivity
bound to proteins is then determined by scintillation counting. The use of this
type of data has received much recent attention as a potential predictor of
toxicity, especially of idiosyncratic toxicities, although this remains a
controversial subject.
A recent example published by Samuel et al. illustrated how^3 H-compounds
could be used in the discovery phase to reduce the extent of covalent adducts
produced by candidates from a specific chemotype (Samuel et al., 2003). These
authors found high levels of covalent binding to rat and human liver
microsomes when a lead compound was tested. The compound was then
incubated in the presence of GSH and the structure of the resultant GSH
adduct elucidated. The knowledge of the adduct structure allowed new analog
synthesis aimed at modulating metabolism of the site of reactive intermediate
formation. An analog was synthesized that reduced the amount of covalent
binding to human liver microsomes by 44-fold compared to the initial
structure. The authors propose that this approach could be generally applied
and provides a viable method of decreasing the amount of covalent binding
associated with metabolic activation during the lead optimization phase.
There are many literature examples of covalent binding studies completed
with toxicologically interesting molecules (Lecoeur et al., 1994; Masubuchi
et al., 1996; Mays et al., 1990, 1995; Naisbitt et al., 2002; Pirmohamed et al.,
1992; Singh et al., 2003; Ward et al., 1982; White et al., 1995). These studies
clearly show that there is a wide range of values for the extent of covalent
binding for different compounds. This is at least partially due to the different
experimental conditions employed for each experiment, and likely also reflects
a wide spectrum of both the efficiency of activation by microsomal enzymes
and the efficiency of trapping by cellular components displayed by these
compounds. The multifactorial nature of hepatotoxicity and blood toxicities
displayed by compounds that form reactive intermediates makes the absolute
translation of the extent of covalent binding number into a prediction of
severity of toxicity difficult to accomplish. However, many compounds that
show hepatotoxicity and blood toxicities do form reactive intermediates that
ASSESSMENT OF POTENTIAL TOXICOLOGY OF METABOLITES 247