only be useful if the signaling mechanisms responsible for inducing the specific
drug metabolizing enzymes targeted by the screening procedure have similar
structural and mechanistic properties. In other words, the relative predictive
usefulness of animal studies in screening for inducing potential of a candidate
compound will depend almost entirely upon the particular drug metabolizing
enzyme(s) whose potential induction is being investigated.
For example, members of the CYP1A family are induced by similar
mechanisms in humans and common laboratory animal species, and the
signaling molecules involved appear not to exhibit major interspecies structural
variation (Hankinson, 1995; Nebert et al., 1993; Okey et al., 1994). The number
of false negative and false positive results that might be generated by screening
candidate compounds at sufficiently high doses in animals for induction of
human CYP1A enzymes would appear to be acceptably low. In fact, animal
studies might be preferable to direct studies in humans for many of these
compounds, because of the inherent potential toxicity and genotoxicity of
many planar aromatic Ah receptor ligands.
On the contrary, using common laboratory animal models to screen
candidate compounds for their potential to induce CYP3A family members is
much more problematic. Because of the well-recognized and well-documented
interspecies differences in the structure and function of molecules mediating the
CYP3A induction signaling cascade, particularly those differences in PXR
receptor structure, such animal screening studies are likely to be poorly
predictive of induction in humans (Bertilsson et al., 1998; Blumberg et al.,
1998; Jones et al., 2000; Kliewer et al., 1998; Lehmann et al., 1998; Luo et al.,
2004; Moore et al., 2002 Zhang et al., 1999). Whereas the marked species
differences in induction of metabolic activity observed in many laboratories
(Jones et al., 2000; Kocarek et al., 1995; LeCluyse, 2001a; Lu and Li, 2001;
Moore et al., 2003) cannot be attributed to the relatively small variations in
sequence homology observed in the DNA binding domain (DBD) region of a
species-specific PXR receptor, the more extensive deviations observed in the
ligand-binding domain (LBD) region of the receptor provide a more rational
basis for the observed species differences (Goodwin et al., 2002; Kliewer and
Willson, 2002; LeCluyse, 2001a; Moore et al., 2002). Correlations between the
activation profiles of human, rat, and rabbit PXRs, and CYP3A4 induction
profiles obtained fromin vitroandin vivoexperiments for the same species,
confirm this hypothesis. Notably, mouse and rat PXRs share 97% identity in
the LBD region, while human and rhesus monkey PXRs share 95% identity in
the same region. On the contrary, sequence identity between the human LBD
region and that of the rat or mouse is 76–77% (Luo et al., 2004; Moore et al.,
2002). These structural data provide a molecular explanation for the numerous
observations that CYP3A induction profiles for many compounds vary greatly
between humans and rats or mice, yet CYP3A induction profiles for mice and
rats are very similar to each other, as are those for humans and rhesus monkey
(Jones et al., 2000; Kocarek et al., 1995; LeCluyse, 2001a; Lu and Li, 2001;
Moore et al., 2003).
ASSESSMENTS 549