the standard enzyme source used in metabolic stability studies. However,
thein vitrosystem that uses liver S9 fractions, which comprises both
endoplasmic reticulum-associated and cytosolic enzymes, is also of
value, particularly for chemicals with readily conjugated functional
groups and/or susceptibility to hydrolysis. S9 fractions, although likely
to exhibit slower turnover capacities for CYP/FMO/UDP-glucurono-
syltransferase (UGT)-mediated biotransformation relative to microso-
mal fractions due to lower enzyme contents, are useful, particularly for
cofactor-independent metabolism such as that mediated by certain
hydrolyases. Therefore, S9-mediated metabolic stability is also routinely
evaluated for compounds at early discovery stages (Hewawasam et al.,
2002; Lai and Khojasteh-Bakht, 2005; Shen et al., 2003). Both CYPs and
FMOs require electron donation by the cofactor, NADPH, to elicit their
catalytic capabilities. Such a NADPH-dependency is generally consid-
ered an indicator of CYP/FMO-catalyzed reactions. In contrast to the
broad spectrum of substrates for CYPs, potential FMO substrates are
usually limited to nitrogen- and sulfur-containing nucleophiles, for
example, tertiary amines (Krueger and Williams, 2005; Ziegler, 2002).
Nonetheless, it does not appear essential to differentiate between CYP-
and FMO-mediated biotransformations in early metabolic stability
studies. FMOs tend to be thermolabile, and thus the reactions mediated
by FMOs are usually initiated by the addition of the enzymes instead of
NADPH, as is the case for CYP reactions. CYPs and FMOs both
catalyze their reactions at low to moderate rates, from an enzyme kinetic
perspective (Kruger and Williams, 2005); a 15- or 30-min incubation is
often required for these enzymes to produce quantifiable metabolites.
ii. Otherin vitroenzyme sources: cDNA-expressed recombinant proteins,
including both Phase I and Phase II enzymes, are commercially
available. While helpful for the identification of the enzymes responsible
for catalysis of specific metabolic reactions and for enzyme kinetic
characterization, recombinant enzymes should not be the primary source
for the metabolic stability studies, because their matrices are inap-
propriate surrogates for those of naturally occurring enzymes; this
mismatch could lead to skewed enzyme kinetics. On the contrary,
primary human hepatocytes, either freshly isolated or cryopreserved, are
being increasingly applied to studies of drug metabolism, including
metabolic stability. It has been shown that the assays employing these
hepatocytes are straightforward to perform, and often lead to high-
quality quantitative and qualitative results (Soars et al., 2002). Human
hepatocytes have intrinsic limitations, such as inconsistent supply,
batch-to-batch or vendor-to-vendor variations, the inability to be used
for automation/HTS, and interindividual donor variability.
These variousin vitrosystems each have value for certain applications and
provide information that might be used to predict metabolic stabilityin vivo.416 DETERMINATION OF METABOLIC RATES AND ENZYME KINETICS