Page 141 - The Toxicology of Fishes

Toxicokinetics in Fishes 121

60

% OF DOSE 40

20

0
0 6 12 18 24 48
HOURS
FIGURE 3.29 Time course of phenol red accumulation in the urine (dashed line, open circles) and bile (dot-dashed line,
solid circles) of dogfish sharks following intravenous administration. Lines represent simulations generated using a physi-
ologically based toxicokinetic model, and the symbols show measured values. (Adapted from Bungay, P.M. et al., J.
Pharmacokinet. Biopharmaceut., 4, 377–388, 1976.)

Biotransformation
Currently, there is considerable interest in using in vitro metabolism information to develop PBTK
models for compounds that undergo biotransformation. Questions remain, however, concerning the
ability of in vitro systems to predict in vivo metabolism. Due to its known importance as a metabolizing
organ, most of the metabolism research conducted to date has focused on the liver. The preparations
most commonly used for in vitro liver metabolism research are microsomes, freshly isolated hepatocytes,
S9 fractions, and precision-cut liver slices. The activity of a microsomal or S9 preparation is commonly
expressed as µmoles (of product) min mg protein ; that of an isolated hepatocyte preparation is
–1
–1
–1
6
–1
generally expressed as µmoles min M cells (M = 10 ); and that of a liver slice may be expressed as
µmoles min slice , µmoles min mg tissue , or µmoles min M cells .
–1
–1
–1
–1
–1
–1
An approach for incorporating in vitro metabolism data into PBTK models was presented by Houston
and Carlile (1997a). Briefly, their approach is to: (1) estimate in vitro intrinsic clearance (CL invitro,int )
from the ratio of V max /K , determined under linear conditions with respect to time and the amount of
m
enzyme present (i.e., mg of protein or number of cells) (Rane et al., 1977); (2) extrapolate CL invitro,int to
in vivo intrinsic clearance for the whole liver (CL invivo,int ) using biologically based proportionality con-
stants; (3) incorporate CL invivo,int into an appropriate liver model that explicitly accounts for both blood-
flow limitations and chemical binding relationships; and (4) account for any nonhepatic routes of
chemical clearance. A conceptually similar approach was used by Ploemen et al. (1997) to incorporate
human in vitro data into a PBTK model for ethylene dibromide.
In a review of earlier work involving several mammalian species, Wilkinson (1987) found that hepatic
extraction ratios were generally well predicted by in vitro estimates of V max and K , determined using
m
microsomes, S9 fractions, and isolated hepatocytes. Houston and Carlile (1997b) compared CL invivo,int
estimates for 35 compounds, obtained by extrapolating data from rat microsomes, hepatocytes, and liver
slices, to observed in vivo values. Clearance rates predicted by isolated hepatocytes and liver microsomes
were highly correlated. In general, however, isolated hepatocytes provided the most accurate predictions
of in vivo clearance. CL invitro,int values measured using rat liver slices were lower than those obtained