Faculty of Pharmaceutical Sciences (H.W., S.K., E.K., F.S.A.,
K.W.R.) and British Columbia Research Institute for Children's and
Women's Health, Department of Obstetrics and Gynecology, Faculty of
Medicine (D.W.R.), The University of British Columbia, Vancouver,
British Columbia, Canada
Dose-dependent pharmacokinetics and metabolism of valproic acid
(VPA) were studied in newborn and adult sheep to assess age-related differences in plasma protein binding and metabolic elimination. Newborn lambs received either a 10- (n = 8), 50- (n = 5), 100- (n = 4), or
250-mg/kg (n = 4) VPA i.v. bolus.
Individual adult sheep (n = 5) received all four
doses in a random order with an appropriate washout period between
experiments. Unbound or metabolic clearance of VPA was significantly
higher in adult sheep at the two lower doses when compared with lambs,
and similar to the lambs at the two higher doses. Plasma protein
binding was nonlinear at all doses. Estimates of binding capacity
(Bmax1) at the saturable site were higher in
adults (91.8 µg/ml) when compared with lambs (44.9 µg/ml), whereas
the opposite trend was observed for binding affinity
[Kd1 = 9.6 µg/ml (adult) versus 3.2 µg/ml (lambs)]. Characterization of developmental differences in
overall VPA metabolic elimination involved fitting of unbound VPA
plasma concentration data to a two-compartment model with
Michaelis-Menten elimination. This resulted in similar in vivo
estimates of apparent Vmax [445.0 µg/min/kg (adult) versus 429.9 µg/min/kg (lambs)]. However,
apparent Km estimates appeared to be higher
in lambs [30.0 µg/ml (adult) versus 69.6 µg/ml (lambs)]. Similar
findings were obtained from in vivo estimates of
Vmax and Km for
VPA glucuronidation obtained from VPA-glucuronide metabolite urinary
excretion data. Thus, it appears that age-related differences in
metabolic clearance may be related to differences in the apparent in
vivo Km as opposed to
Vmax of VPA glucuronidation.
 |
Introduction |
Valproic
acid (2-propylpentanoic acid; VPA1) is a broad
spectrum anticonvulsant with a unique branched-chain fatty acid
structure (Davis et al., 1994
). Despite cases of VPA-induced
idiosyncratic hepatotoxicity, VPA remains the drug of choice for
treating seizures of various etiologies in neonates, infants, and
children due to its broad spectrum of activity and minimal cognitive
side effects (Sarisjulis and Olivier, 1999). Studies examining
VPA pharmacokinetics during the immediate newborn period in humans have
reported extended elimination half-lives in newborns when compared with
the adult (Levy and Shen, 1995). Similar findings have been observed in studies with other species, such as rats (Haberer and Pollack, 1994)
and guinea pigs (Yu et al., 1985, 1987). Valproic acid is mainly
eliminated by hepatic metabolism and exhibits a low hepatic extraction
ratio. Thus, any developmental changes in plasma protein binding and/or
metabolic capacity will influence VPA systemic clearance (Levy and
Shen, 1995). The limited available data in human neonates indicate a
low unbound VPA clearance in this population, suggesting low intrinsic
metabolic clearance due to immature activity of the enzymes responsible
for VPA elimination (Levy and Shen, 1995). In vivo developmental
studies in rats (Haberer and Pollack, 1994) and guinea pigs (Yu et al.,
1985, 1987) suggest that differences in both plasma protein binding and
VPA metabolism contribute to age-related alterations in VPA
disposition. Similarly in sheep, we observed a lower metabolic
clearance and a higher unbound fraction in 10-day-old lambs when
compared with adult sheep (Wong et al., 2000). The lower metabolic
clearance in 10-day-old lambs observed in this study was attributed to
age-related differences in VPA-glucuronidation (Wong et al., 2000).
Thus, the purpose of this study is to examine in more detail the role
of plasma protein binding and metabolic elimination in determining VPA
total body clearance in 10-day-old lambs and adult sheep. This was
accomplished by conducting a dose-ranging experiment in these two age
groups. Dose-dependent changes in VPA metabolism were also examined.
Sheep were chosen for these studies due to their similarity to humans
in terms of the main metabolic pathways (i.e., glucuronidation,
-oxidation, and P-450-catalyzed pathways) and VPA metabolites
[i.e., VPA-glucuronide, 3-keto VPA (2-n-propyl-3-oxopentanoic acid)] previously observed in
sheep (Kumar et al., 2000a). Furthermore, the use of chronically
catheterized lambs and adult sheep overcomes limitations in the
available sampling volume of biological fluids associated with smaller
animal models allowing for more detailed studies.
| |
Materials and Methods |
Animals and Surgical Preparation.
All studies were approved by the University of British Columbia Animal
Care Committee, and the procedures performed on sheep conformed to the
guidelines of the Canadian Council on Animal Care.
Adult sheep.
Five nonpregnant Dorset Suffolk cross-bred ewes, with a body weight of
61.9 ± 7.3 kg (mean ± S.D.), were surgically prepared at
least 3 days before experimentation. Polyvinyl catheters (Dow Corning,
Midland, MI) were implanted in a femoral artery and vein (catheter i.d.
1.02 mm and o.d. 2.16 mm) as described by Kumar et al. (1999). On the
morning of the experiment, a Foley bladder catheter was inserted via
the urethra of the ewe and attached to a sterile polyvinyl bag for
cumulative urine collection.
Newborn lambs.
A total of 21 Dorset Suffolk cross-bred lambs were used in this study.
Lambs were divided into a 10-mg/kg group (n = 8), a 50-mg/kg group (n = 5), a 100-mg/kg group
(n = 4), and a 250-mg/kg group (n = 4).
All lambs were surgically prepared at least 3 days before the
experiment under isoflurane (1%) anesthesia. Briefly, polyvinyl
catheters (Dow Corning) were implanted in a carotid artery, a jugular
vein, and the urinary bladder as described by Wong et al. (2000). On
the day of the experiment, the lambs were moved to monitoring pens
adjacent to and in full view of their mothers. The urinary bladder
catheter was allowed to drain by gravity into a sterile reservoir.
While in the holding pens, lambs were fed Deluxe Lamb Milk Replacer
(Canadian Nurs-ette Distributor Ltd., Canrose, AB, Canada) and had free
access to hay, grain, and water.
Experimental Protocols.
Adult sheep
Experiments involved administration of an i.v. bolus of VPA (sodium
valproate, Sigma Chemical Co., St. Louis, MO) equivalent to 10, 50, 100, or 250 mg of VPA/kg of body weight (mean ewe body weight = 61.9 ± 7.3 kg) followed by a 5-day washout period, after which
the next dose was administered. This continued until each ewe received
one of each dose (i.e., a total of four experiments were performed on
each animal). Doses were administered over 1 min via the femoral vein
in a randomized order. Serial blood samples (~3 ml) were collected
for adult sheep from the femoral artery at 5, 15, 30, 45, 60 min, and
2, 4, 6, 9, 12, 15, 24, 36, 48, 60, and 72 h following drug
administration. For the 10- and 50-mg/kg doses, the experiments
continued for only 36 and 48 h, respectively. Cumulative urine was
also collected for the full duration of the experiment. The adequacy of
the 5-day washout period was confirmed by the absence of any detectable
VPA or its metabolites in the last collected plasma and urine samples
of each experiment.
Newborn lambs.
Experiments in newborn lambs were initiated at approximately 10 days
following birth. Average ages of the newborn lamb groups on the first
day of the experiment were 10.9 ± 1.4 days (10-mg/kg group),
10.4 ± 0.5 days (50-mg/kg group), 10.3 ± 1.0 days
(100-mg/kg group), and 10.5 ± 1.0 days (250-mg/kg group). Mean
lamb body weights were 5.8 ± 1.4 kg (10-mg/kg group), 6.0 ± 1.0 kg (50-mg/kg group), 7.1 ± 1.6 kg (100-mg/kg group), and
6.8 ± 2.1 kg (250-mg/kg group). Lambs were administered an i.v.
bolus dose of VPA equivalent to 10, 50, 100, or 250 mg of VPA/kg of
body weight depending on which group the lamb had been assigned to.
Drug administration in the lambs was via the jugular vein over 1 min.
Serial blood samples (~2 ml) were collected from the carotid artery
at 5, 15, 30, 45, 60 min, and 2, 4, 6, 9, 12, 24, 36, 48, 60, and
72 h following drug administration. Cumulative urine samples were
also collected for the full duration of the experiment. The only
exceptions were for four of the eight lambs in the 10-mg/kg group where
urine collection was incomplete due to catheter failure; these were excluded from data analysis.
All doses were prepared in sterile water for injection and were
sterilized by filtering through a 0.22-µm nylon syringe filter (MSI,
Westboro, MA) into a capped empty sterile injection vial. All blood
samples collected were placed into heparinized Vacutainer tubes (Becton
Dickinson, Rutherford, NJ) and centrifuged at 2000g for 10 min. The plasma supernatant was removed and placed into clean
borosilicate test tubes with polytetrafluoroethylene-lined caps. Plasma
and urine samples were stored frozen at 
20°C until the time of analysis.
Note: Some of the data from the lamb and adult 10-mg/kg VPA i.v. bolus
experiments have been presented in an earlier manuscript (Wong et al.,
2000).
Determination of VPA Plasma Protein Binding.
Unbound plasma concentrations of VPA were determined ex vivo in all
adult sheep and postnatal lamb plasma samples by an ultrafiltration procedure at sheep body temperature (39°C). The procedure involved centrifuging at 1000g for 30 min using Centrifree
micropartition devices (Amicon Inc., Danvers, MA). Plasma samples for
the determination of unbound VPA concentrations were stored in separate
aliquots to avoid repetitive thawing that could result in lipolysis and release of free fatty acids and hence competitive displacement of bound
VPA from plasma binding sites (Haberer and Pollack, 1994).
Determination of Protein Concentrations in Adult and Lamb Plasma.
Concentrations of total protein in adult sheep and 10-day-old lamb
plasma were determined using a hand-held refractometer (model 5433, Fisher Scientific Instruments, Tustin, CA) as described by Kwan (1989).
Drug and Metabolite Assay.
Concentrations of VPA and its metabolites in all biological fluids and
plasma ultrafiltrate were measured simultaneously using an established
gas chromatographic-mass spectrometric analytical method (Yu et
al., 1995). The variability and bias of all analytes measured using
this analytical method was determined to be <15% in earlier assay
validation studies (Yu et al., 1995). VPA and metabolite calibration
and quality control standards as well as control (blank) biological
fluid samples were run with each batch of study samples. Concentrations
of the VPA-glucuronide metabolite in both adult and lamb urine were
measured using a base hydrolysis procedure described as follows. Urine
samples were adjusted to pH 12.5, incubated at 60°C for 1 h, and
the total VPA (unconjugated + conjugated) was quantified by the above
gas chromatographic-mass spectrometric analytical method. The
concentration of the VPA-glucuronide metabolite was estimated as the
difference between total and unconjugated (unhydrolyzed) VPA
concentrations. This described procedure was preferred over hydrolysis
with -glucuronidase because VPA-glucuronide has been shown to
rearrange to at least six -glucuronidase-resistant structural
isomers via migration of the acyl moiety away from the C-1 position and
subsequent ring opening, mutarotation, and lactone formation (Dickinson
et al., 1984). These rearrangements are pH-, temperature-, and storage
time-dependent (Dickinson et al., 1984). Hydrolysis with alkali,
however, is capable of measuring total VPA-glucuronide despite these
possible rearrangements (Dickinson et al., 1984).
Pharmacokinetic Analyses.
Ex vivo protein binding data was analyzed by first calculating the
bound VPA concentrations from the difference between the corresponding
experimentally determined total and unbound concentrations. Rosenthal
plots (bound/unbound concentration versus bound concentration) were
constructed for identification of the multiplicity of binding sites.
Bound versus unbound concentrations were then fitted using the
nonlinear least-squares regression program ADAPT II (D'Argenio and
Schumitzky, 1997) to a two-site binding model (a high-affinity saturable and a low-affinity linear site) according to the following equation:
|
(1)
|
where Cb and
Cu are the corresponding bound and unbound
concentrations, and Bmax1 and
Bmax2 are the maximal binding capacities of
the first and second binding site, respectively.
Kd1 and Kd2 are
the equilibrium dissociation constants of VPA at the first and second
binding site, respectively.
Adult sheep estimates of plasma protein binding parameters were
obtained for individual animals and are presented as mean ± S.D.
This was possible since a wide range of plasma VPA concentrations was
achieved in each animal following the four dose-ranging experiments. For lambs, only a single estimate of the binding parameters could be
estimated since each lamb only received one VPA dose, and therefore plasma concentration data from individual animals were insufficient to
provide for individual estimates. Thus, VPA plasma concentration data
for lambs were pooled together to generate estimates, and consequently
plasma protein binding parameters are presented for lambs as an
estimate followed by its coefficient of variation (CV) in parentheses.
In vivo estimates of apparent Michaelis-Menten parameters
(Vmax, Km) for
overall VPA elimination were obtained through simultaneous fitting of
unbound concentration-time data from the 50-, 100-, and 250-mg/kg
experiments of an individual adult animal (more details as to why only
the three higher doses were modeled will be provided under
Results). A two-compartment model with Michaelis-Menten elimination provided the best "fit" of the data from all adult animals. Briefly, the first step involved generating microconstant estimates characterizing the movement of drug between the central and
peripheral compartments for individual adult animals. This was
accomplished by modeling of the unbound concentration-time data from
their respective 10-mg/kg experiments to a standard two-compartment
model. The resulting microconstant estimates were fixed for subsequent
modeling involving simultaneous fitting of unbound concentration-time
data from multiple experiments (i.e., 50-, 100-, and 250-mg/kg
experiments). Model selection was based upon lower Akaike's
Information Criterion (AIC) and Schwarz Criterion values generated when
using a two-compartment model with Michaelis-Menten elimination as
opposed to a simpler one-compartment model with similar elimination
characteristics (Wagner, 1993; Bourne, 1995). In contrast to adult
sheep, individual neonatal lambs received only a single dose of VPA;
therefore, individual animal estimates of apparent
Vmax and Km
could not be obtained. Instead, pooled plasma profiles (i.e., unbound
plasma VPA concentrations at each time point were averaged) were
constructed for each dose, and the resulting profiles were modeled as
described for the adult above. Similar to adult sheep, a
two-compartment model with nonlinear elimination provided the best fit
for the pooled lamb data. CVs for all estimates of apparent
Vmax and Km
were <7% and <20%, respectively. Adult estimates for both
parameters are presented as a mean ± S.D. The pooled estimate of
Vmax and Km
obtained for neonatal lambs is presented as the parameter estimate
followed by its respective CV in parentheses. As for plasma protein
binding, all data was modeled using ADAPT II (D'Argenio and
Schumitzky, 1997).
In vivo estimates of apparent Vmax and
Km for VPA glucuronidation were determined
by constructing plots of urinary excretion rate of the glucuronide
metabolite (V) versus the unbound VPA concentration at the
midpoint of the urine collection interval (C 
). Data was pooled from all
experiments in adult sheep and in neonatal lambs resulting in one plot
for adult sheep and one plot for lambs. Both plots were fit to a
standard Michaelis-Menten equation as follows:
|
(2)
|
where V and C are as
defined above, Vmax is the maximal
formation rate of VPA-glucuronide, and Km is the Michaelis-Menten constant (Gibaldi and Perrier, 1982). Data were
fit to eq. 2 using ADAPT II (D'Argenio and Schumitzky, 1997). CVs
generated for Vmax and
Km were <8% and <20%, respectively. The
pooled estimate of apparent Vmax and
Km obtained for both adult sheep and
neonatal lambs is presented as the parameter estimate followed by its
respective CV in parentheses.
Due to the nonlinear/saturable nature of VPA plasma protein binding,
the parameters fp (area weighted unbound
fraction of the drug) and Vdss' (steady state
volume of distribution parameter corrected for the effects of saturable
protein binding) were also calculated as follows:
|
(3)
|
|
(4)
|
For drugs exhibiting saturable protein binding, the
Vdss when calculated using the
"model-independent" approach (Gibaldi and Perrier, 1982)
overestimates the "true" Vdss and is
concentration-dependent (McNamara et al., 1983). Similarly,
Vd
is constant only for a particular
fp value and can be used to relate
steady-state plasma concentrations to the amount of the drug in the
body if steady-state unbound fraction of the drug is equal to
fp (McNamara et al., 1983). Thus, both
Vdss and Vd are poor
indicators of drug distribution. Instead, the
Vdss' parameter is more reflective of shifts in
drug mass into or out of the vascular space (i.e., information
traditionally provided by the Vdss parameter) (McNamara et al., 1983). As with Vd above, this
Vdss' parameter is also constant only for a
particular fp or a steady-state plasma
unbound fraction equivalent to fp.
All other pharmacokinetic parameters were calculated by standard
methods as described in Gibaldi and Perrier (1982).
Statistical Analysis.
All data are reported as mean ± S.D. Pharmacokinetic parameters
were compared using either a t test for comparison between two groups or an analysis of variance followed by a Fischer's least
significant difference multiple comparison test for multiple group
comparisons. The significance level was p < 0.05 in
all cases.
| |
Results |
Dose-Dependent Pharmacokinetics of VPA in Newborn Lambs and Adult
Sheep.
Figure 1, A to D show mean
semilogarithmic plots of VPA (unbound and total) concentration versus
time for adult sheep following i.v. administration of a 10-, 50-, 100-, and 250-mg/kg VPA bolus. Figure 2, A to D
show similar semilogarithmic plots of pooled VPA (unbound and total)
concentration versus time data from newborn lambs. In a previous study,
we observed that plasma protein binding was nonlinear following the
administration of a 10-mg/kg i.v. bolus of VPA (Wong et al., 2000). In
cases where binding is nonlinear, a parameter that can be used to
assess overall changes in unbound fraction is the area weighted unbound
fraction (fp). Table
1 presents dose-dependent changes in
fp, AUC
(AUC of
unbound VPA), and Vdss' for newborn lambs
and adult sheep. As expected, fp increased
with increasing dose for both age groups. At the three higher doses,
fp values for 10-day-old lambs were similar
to their corresponding adult fp estimates.
This contrasts to what is observed at the lowest dose.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 1.
Mean VPA [total ( ) and unbound ( )]
plasma concentration versus time profiles for adult sheep (n = 5)
following i.v. bolus administration of a 10- (A), 50- (B), 100- (C),
and 250-mg/kg (D) dose of VPA.
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 2.
Mean VPA [total () and unbound ()]
plasma concentration versus time profiles for newborn lambs (10 days
old) following i.v. bolus administration of a 10- (A; n = 8), 50- (B; n = 5), 100- (C; n = 4), and 250-mg/kg (D; n = 4)
dose of VPA.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 1
Dose-dependent changes in fp,
AUC , and Vdss' in adult sheep and
newborn lambs
Data is presented as the mean ± S.D.
|
|
Dose-dependent changes in AUC were examined as
opposed to AUC0-
(AUC of total drug) since
changes in AUC are independent of changes in
plasma protein binding. Increases in dose were associated with expected
increases in AUC for both age groups (Table 1).
AUC was significantly higher for 10-day-old
lambs at the two lower doses; however, for the two higher doses the
adult and lamb estimates were similar. For adult sheep, the increase in
AUC with increasing dose is nonlinear,
suggestive of Michaelis-Menten elimination. In newborn lambs,
AUC increased linearly with dose until the
250-mg/kg dose, where its increase was nonlinear. Additional evidence
of nonlinear elimination can be observed in the unbound VPA plasma
profiles in Figs. 1 and 2 that displayed an increasing convex curvature
with increasing dose. Although the plasma profiles for total VPA
concentrations also showed signs of convexity, this could be a result
of saturable protein binding rather than nonlinear elimination.
For drugs exhibiting saturable protein binding, the volume terms
Vdss and Vd do not provide any
information on possible shifts of drug into or out of the vascular
space. However, this information is provided by the
Vdss'; therefore, this term has been presented in
Table 1 to examine changes in drug distribution with dose. The
Vdss' for both age groups increased with an
increase in dose, suggesting movement of drug out of the vascular space
with increasing doses. In addition, for the three higher doses,
Vdss' was significantly larger in lambs in
comparison with adult sheep.
Figure 3A depicts changes in VPA total
body clearance (CLtb) with dose for newborn lambs
and adult sheep. In both lambs and adult sheep,
CLtb increases with dose to a maximum at the
100-mg/kg dose before decreasing at the highest dose. Figure 3B
illustrates changes in VPA unbound clearance (CL
). For adult sheep, the CL decreases significantly with
increasing dose, consistent with metabolic saturation (Table 2). For neonatal lambs,
CL decreases only following administration of the
highest dose (Table 2). Adult CL is significantly
higher than CL for newborn lambs at the two lower
doses. However, at the two higher doses, CL is
similar between the two age groups (Fig. 3B).
View larger version (9K):
[in this window]
[in a new window]
|
Fig. 3.
Changes in total (A) and unbound (B) drug
clearances in adult sheep (n = 5) and 10-day-old lambs (10 mg/kg,
n = 8; 50 mg/kg, n = 5; 100 mg/kg, n = 4; 250 mg/kg,
n = 4) for different doses of VPA.
*, significant difference between the adult and lamb estimates
(p < 0.05).
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 2
Changes in total and unbound VPA clearance in newborn lamb (10 days
old) and adult sheep with increasing dose
Data is presented as the mean ± S.D.
|
|
Ex Vivo Determination of Plasma Protein Binding Parameters.
Rosenthal plots for a representative adult sheep and pooled lamb data
are presented in Fig. 4, A and B. Plots
for both adult sheep and newborn lambs were biphasic in nature. The
initial steep declining portion of the biphasic Rosenthal plots
suggests the presence of a high-affinity saturable binding site,
whereas the relatively flat portion of the plots suggests the presence
of a low-affinity linear (nonsaturable) site. Fitting the bound versus unbound concentration data to a two-site binding model with a saturable
and a nonsaturable binding site (eq. 1) resulted in statistically
better fits (lower AIC and Schwarz Criterion, smaller CV values for
fitted parameters) when compared with a one-site binding model.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 4.
Rosenthal plot (A) and plots of the
relationship between bound drug (Cb) versus unbound drug
(Cu) in plasma (C) for a representative
adult sheep (E6208) and 10-day-old lambs (B and D).
A and C are generated from plasma data pooled from all experiments (10, 50, 100, and 250 mg/kg) performed in E6208. B and D are similar plots
generated from plasma data pooled from all 10-day-old lamb experiments
(n = 21). C and D, a model-predicted line obtained
from a fit of the data to a two-site binding model is depicted.
|
|
A scatter plot of bound versus unbound VPA plasma concentration data
for an individual adult sheep (i.e., plasma data is pooled from the
10-, 50-, 100-, and 250-mg/kg experiments from a single animal) is
presented in Fig. 4C. A similar plot was constructed for plasma data
from all the newborn lamb experiments and is presented in Fig. 4D. The
model-predicted lines based upon fitting to eq. 1 are also depicted,
indicating a good fit of the data to the two-site binding model.
Estimates of binding parameters for adult sheep and newborn lambs are
presented in Table 3. From the data, it
appears that binding capacity at the saturable binding site (Bmax1) is higher in adult sheep. In
contrast, the higher Kd1 estimate in adult
sheep suggests that binding affinity at the saturable site is lower
when compared with lambs.
Total Protein Concentrations in Adult Sheep and Lamb Plasma.
The total protein concentration in plasma from adult sheep
(n = 5) and 10-day-old lambs (n = 18)
was 73.8 ± 6.4 and 58.0 ± 4.6 mg/ml, respectively. Three
lambs were excluded from total protein determination due to the lack of
availability of remaining plasma following drug and metabolite
analysis. Total protein concentration was significantly higher in adult
sheep in comparison to 10-day-old lambs (unpaired t test,
p < 0.05).
Dose-Dependent Changes in Urinary Excretion of VPA and Its
Metabolites.
Tables 4 and
5 present dose-dependent changes in
the percentage of the total VPA dose recovered in urine as unchanged
VPA and its metabolites in adult sheep and newborn lambs. Aside from the group of lambs receiving the 10-mg/kg dose, almost the entire dose
(>90%) was recovered as either unchanged VPA or one of its metabolites (Tables 4 and 5). The majority of the dose in both age
groups and at all doses was recovered as VPA-glucuronide. In adult
sheep, 68.3 to 77.9% of the dose appeared in urine as the glucuronide
metabolite, and there was no significant change in the recovery of this
metabolite with dose (Table 4). In neonatal lambs, recovery of
VPA-glucuronide in urine increased significantly from 29.2% at the
lowest dose to 65.7 to 69.6% at the higher doses (Table 5). Recovery
of the administered dose in urine as the unchanged drug appeared to
follow no pattern with increasing dose. In adults, recovery of VPA in
urine was significantly less at the 50- and 100-mg/kg doses than at the
10-mg/kg dose (Table 4). No such changes occurred with increasing dose
in neonatal lambs (Table 5).
View this table:
[in this window]
[in a new window]
|
TABLE 4
Recovery of VPA and its metabolites in urine following i.v. bolus
administration of a 10-, 50-, 100-, or 250-mg/kg dose of VPA in adult
nonpregnant sheep (n = 5)
Data is presented as the mean ± S.D.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 5
Recovery of VPA and its metabolites in urine following i.v. bolus
administration of a 10- (n = 4), 50- (n = 5), 100- (n = 4), or 250- (n = 4) mg/kg dose of VPA in 10-day-old lambs
Data is presented as the mean ± S.D.
|
|
The major -oxidation metabolite recovered in urine in both lambs and
adult sheep was 3-keto VPA. For both age groups, the recovery of this
metabolite in urine decreased significantly with increasing dose,
suggestive of metabolic saturation of the -oxidation pathway (Tables
4 and 5). At the 50- and 100-mg/kg doses, the percentage of the dose
recovered as this metabolite in neonatal lambs was significantly higher
than observed in adult sheep. The renal excretion of the other
-oxidation metabolites, (E)-2-ene and
(E)-3-ene VPA, were limited in both adult sheep and lambs accounting for no more than 0.08% of the dose. Similar to 3-keto VPA,
the recovery of 3-OH VPA decreased with increasing dose in neonates
(Table 5). In adults, the mass balance of this metabolite appeared to
follow no trend (Table 4).
The prominent metabolite formed by cytochrome P-450 pathways that was
recovered in urine in both age groups appeared to be 4-OH VPA. In both
newborn lambs and adult sheep, the urinary recovery of 4-OH VPA
significantly increased with increasing dose, going as high as ~11%
of the dose in lambs (Table 5) and ~7% of the dose in adults (Table
4). Similarly, in both age groups, 2-PGA increased significantly with
increasing dose. All other P-450 metabolites (i.e., 5-OH, 4-ene and
4-keto VPA, and 2-PSA) accounted for less than 1% of the VPA dose at
all dosing levels. Similar to (E)-2-ene and
(E)-3-ene VPA, the recovery of 4-ene VPA in urine appeared
to be particularly low, accounting for no more than ~0.05% of the
dose in any of the adult sheep or lamb groups.
In Vivo Estimation of Vmax and
Km of Overall VPA Elimination.
Unbound concentration versus time profiles from the 50-, 100-, and
250-mg/kg bolus experiments were modeled simultaneously to obtain in
vivo estimates of apparent Vmax and
Km values of overall VPA elimination. The
10-mg/kg experiments from both age groups were excluded from this
modeling exercise due to our inability to recover the majority of the
10-mg/kg VPA dose administered to neonatal lambs (i.e., only ~50%
could be recovered). For the three larger doses we were able to recover
the majority of the dose as known VPA metabolites derived from hepatic
metabolism (i.e., >80% of the administered dose) in both age groups.
Thus, the apparent Vmax and
Km values estimated from the modeling of these doses are hybrid constants largely reflective of overall metabolic elimination. The unbound concentration versus time data from
both age groups were modeled using a two-compartment model with
Michaelis-Menten elimination based upon a better fit (lower AIC and
Schwarz Criterion, smaller CVs for fitted parameters) with this model
when compared with fitting to a one-compartment model with nonlinear
elimination. Unbound concentration versus time profiles from adult
sheep and neonatal lambs are presented in Fig.
5 along with their model-predicted plasma
profile lines. Adult animals were individually modeled; however, for
lambs the mean unbound plasma concentration versus time profiles at
each dose were modeled. Estimates of apparent
Vmax and Km
values obtained from modeling are presented in Table
6. From the estimates, it appears that
Vmax is similar between the two age groups.
However, Km appears to be higher for lambs.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 5.
Unbound VPA plasma concentration versus time
profiles from a representative adult sheep (E6208) (A, B, and C) and
mean unbound VPA plasma concentration versus time profiles from
10-day-old lambs (n = 8; D, E, and F) receiving a 50-mg/kg bolus
(n = 5; A and D), a 100-mg/kg bolus (n = 4; B and E), and a
250-mg/kg bolus (n = 4; C and F).
In all cases, a model-predicted line obtained from fit of the data to a
two-compartment model with Michaelis-Menten elimination is depicted.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 6
In vivo estimation of Vmax and Km of overall VPA
elimination in adult sheep and 10-day-old lambs via modeling of unbound
VPA concentrations in plasma
|
|
In Vivo Estimation of Apparent Vmax and
Km of VPA Glucuronidation.
A condition for estimating the apparent
Vmax and Km
values of VPA glucuronidation using urine data is that the appearance of VPA-glucuronide in urine is formation rate-limited as opposed to
elimination rate-limited (Gibaldi and Perrier, 1982). In previous investigations with VPA using pregnant sheep, the glucuronide metabolite did not appear to accumulate substantially in plasma consistent with formation rate-limited urinary excretion (S. Kumar, unpublished results). Metabolites demonstrating formation rate-limited urinary excretion exhibit plasma profiles that decline in parallel to
the plasma profile of parent compound (Houston, 1986). Thus, apparent
plasma half-lives determined from the terminal slopes should be similar
for both the parent compound and its metabolite. Due to the lack of
plasma following drug (unbound and total) and metabolite analysis, we
could not determine VPA-glucuronide levels in plasma. As an
alternative, a plot of urinary excretion rate versus
tmid (i.e., time at the midpoint of the
urine collection interval) was constructed (Gibaldi and Perrier, 1982).
The apparent plasma half-life of the metabolite can be obtained from
the terminal slopes of the semilogarithmic form of these plots (Gibaldi
and Perrier, 1982). Since elimination rate-limited urinary excretion of
the metabolite is likely to occur at the highest dose of VPA, we
constructed semilogarithmic plots of excretion rate versus tmid for the 250-mg/kg experiments (plots
not shown). VPA-glucuronide half-lives of 4.70 ± 1.03 h
(adult sheep) and 3.88 ± 0.62 h (10-day-old lambs) obtained
from the terminal slopes of these plots were not significantly
different from unbound VPA half-lives determined from plasma data
(adult, 4.86 ± 2.50 h; 10-day-old lamb, 3.28 ± 0.34 h) (unpaired t test, p < 0.05).
Thus, our assumption of formation rate-limited urinary excretion of the
glucuronide metabolite appears to be reasonable.
Plots of V versus C for adult
sheep and neonatal lambs were fitted to eq. 2 and are presented in Fig. 6. The resulting estimates of
Vmax and Km
values of VPA-glucuronidation are presented in Table
7. As with
Vmax estimates of overall VPA elimination,
the Vmax estimates of VPA-glucuronidation
were similar between the two age groups. However,
Km estimates appeared to be higher for
neonatal lambs. In fact, Km estimates of
VPA glucuronidation for both adult and neonatal sheep (Table 7) are
very similar to Km estimates of overall VPA
elimination (Table 6).
View larger version (8K):
[in this window]
[in a new window]
|
Fig. 6.
Relationship between the rate of urinary
excretion of VPA-glucuronide (V) versus the unbound VPA plasma
concentration at the midpoint of the urine collection interval
(C) for adult sheep (A; n = 5) and 10-day-old
lambs (B; n = 17).
In both cases, a model-predicted line obtained from fit of data to a
standard Michaelis-Menten equation is depicted.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 7
In vivo estimation of apparent Vmax and Km of VPA
glucuronidation in adult sheep and 10-day-old lambs via modeling of
urine data
|
|
| |
Discussion |
Dose-Dependent Pharmacokinetics in Adult Sheep and Neonatal Lambs.
The pharmacokinetics of VPA are unique, exhibiting both
saturable/nonlinear plasma protein binding and
saturable/capacity-limited metabolism at clinically relevant plasma
concentrations. These characteristics are largely due to the high
therapeutic doses of VPA in comparison with other drugs. Since VPA
exhibits a low hepatic extraction and is mainly eliminated via hepatic
metabolism (Levy and Shen, 1995), according to the well stirred model
(Wilkinson and Shand, 1975) the CLtb of VPA can
be described as
|
(5)
|
where fu is the unbound fraction and
CLint is the hepatic intrinsic clearance. Thus,
saturation of plasma protein binding and metabolism influence VPA
CLtb in opposite directions. Specifically, saturation of plasma protein binding increases drug free fraction, resulting in an increase in CLtb. In
contrast, metabolic saturation acts to decrease
CLtb by decreasing intrinsic clearance.
Significant alterations in plasma protein binding occurred in our
dose-ranging experiments as evidenced by the observed changes in
fp (a measure of overall drug free
fraction) (Table 1). In addition, the nonlinear increases in
AUC in both adult sheep and neonatal lambs,
especially at the highest dose, provide evidence of metabolic
saturation. Thus, dose-dependent alterations in VPA
CLtb will be the result of relative changes in
plasma protein binding and metabolism that occur with increasing dose.
For both adult sheep and neonatal lambs, we observed significant
increases in VPA CLtb up to the 100-mg/kg dose
(Table 2). At the highest dose (250-mg/kg), CLtb
for both age groups fell to a level similar to that observed at the
10-mg/kg dose (Table 2). The observed increases in
CLtb are a result of the observed increases in
overall drug free fraction (i.e., fp) with
increasing dose (Table 1). The influence of metabolic saturation on VPA CLtb is only evident at the 250-mg/kg dose
for both age groups. This is also the dose at which we observed the
most substantial nonlinear increases in AUC (i.e., a 2.5-fold increase in dose resulted in an ~6- and an
~4.5-fold increase in AUC in adult sheep and
neonatal lambs, respectively). In previous studies examining the
dose-dependent pharmacokinetics of VPA in adult guinea pigs (Yu et al.,
1987, 1993), rats (Liu et al., 1990; Liu and Pollack, 1993), and humans
(Bowdle et al., 1980; Anderson et al., 1992; Gómez Bellver et
al., 1993), CLtb either increased with dose, decreased with dose, or exhibited no changes with dose. This
discrepancy in alterations in CLtb with dose is
largely related to differences in the range of doses used in these
studies. Species and individual differences in VPA plasma protein
binding and metabolic capacity may also play a role.
Unlike VPA CLtb, CL appeared
to follow different trends in adult and 10-day-old sheep. In adults,
CL decreased significantly with increasing dose.
Similar decreases in CL were observed in
dose-ranging studies conducted in adult guinea pigs (Yu et al., 1987,
1993) and humans (Bowdle et al., 1980; Anderson et al., 1992;
Gómez Bellver et al., 1993). The observed decreases in
CL in these studies are indicative of metabolic
saturation as the unbound clearance of VPA approximates hepatic
intrinsic clearance (see eq. 5). Surprisingly, CL in
neonatal lambs did not decrease significantly until the highest dose
(Table 2). In contrast, a decrease in CL was observed with increasing dose in both 3-day-old and 21-day-old guinea
pigs (Yu et al., 1987). However, in these experiments two of the three
doses administered (i.e., 20, 200, and 600 mg/kg) were either similar
or substantially larger than the highest dose (i.e., 250 mg/kg) used in
our studies. The use of similar doses in lambs would likely result in a
similar trend in CL.
As expected, Vdss' increased with increasing dose
in both adult sheep and neonatal lambs (Table 1). The modest increase
in Vdss' is consistent with the VPA's low tissue
binding (Davis et al., 1994). Interestingly, the
Vdss' in lambs is significantly larger than in
adults at the 50-, 100-, and 250-mg/kg doses. The larger
Vdss' observed in lambs may be related to the
larger total body water content in the young (Moreselli et al., 1980).
A similar phenomenon is observed in human neonates who exhibit higher
volumes of distribution (i.e., 0.28-0.43 l/kg) in comparison with
adults (i.e., 0.13-0.20 l/kg) (Levy and Shen, 1995).
Age-Related Differences in VPA Plasma Protein Binding.
Plasma protein binding exhibits nonlinear characteristics in adult and
neonatal lambs following the administration of a clinical dose (i.e.,
10 mg/kg) of VPA (Wong et al., 2000). One of our goals of the
dose-ranging study was to saturate plasma protein binding to such an
extent that we would have a wide enough range of bound (Cb) and unbound
(Cu) drug concentrations required for
appropriate characterization of plasma protein binding parameters.
Examination of our Rosenthal (Fig. 4, A-B) and our
Cb versus Cu
plots (Fig. 4, C-D) provide obvious evidence that a two-site binding
model is required to describe VPA plasma protein binding. This was not surprising since a similar model was used previously to characterize VPA plasma protein binding in pregnant sheep (Kumar et al., 2000b). The presence of a high-affinity saturable binding site and a
low-affinity nonsaturable site has also been demonstrated in rats
(Semmes and Shen, 1990; Haberer and Pollack, 1994; Slattum et al.,
1996), guinea pigs (Yu and Shen, 1992), and humans (Riva et al., 1984; Scheyer et al., 1990). Our estimates of binding parameters for adult
sheep (Table 3) appear to be reasonably similar to estimates obtained
from humans (i.e., Bmax1 ~ 169 µg/ml
and Kd1 ~ 6-13 µg/ml; Riva et al.,
1984; Scheyer et al., 1990). The Kd1
(9.6 ± 5.9 µg/ml) estimate obtained for adult sheep is
consistent with observable saturation of VPA plasma protein binding at
therapeutic concentrations (i.e., 50-100 µg/ml).
A comparison of VPA binding parameters obtained from previous
experiments in pregnant sheep and current estimates in nonpregnant sheep revealed a subtle difference in binding capacity. Kumar et al.
(2000b) estimated a Bmax1 of 62.8 µg/ml
in pregnant sheep as opposed to Bmax1
estimates of 91.8 ± 24.3 µg/ml in nonpregnant animals. A
reduction of albumin (the primary protein involved in VPA plasma
protein binding) concentrations that occurs with pregnancy has been
suggested as one of mechanisms involved in reduced VPA binding in
pregnant women (Nau et al., 1984; Riva et al., 1984; Nau and Krauer,
1986). A similar phenomenon may occur in pregnant sheep and result in
the lower observed Bmax1 estimate.
Kd1 estimates were similar in pregnant and
nonpregnant sheep (7.6 and 9.6 µg/ml, respectively).
Estimates of the linear component (i.e.,
Bmax2/Kd2) of
VPA plasma protein binding for adult sheep and neonatal lambs were similar (Table 3). However, both the binding capacity
(Bmax1) and affinity
(Kd1) of the high-affinity saturable
binding site of the two age groups appeared to be different.
Bmax1 estimates from adult sheep were
two-fold higher than the corresponding estimate in 10-day-old lambs
(44.9 µg/ml). The lower binding capacity may play a role in the
higher fp previously observed in 10-day-old lambs when compared with adult sheep (Wong et al., 2000). Similar apparent reductions in plasma protein binding have been observed in
neonates for several anticonvulsants (including VPA), antibiotics, and
analgesics (Kearns and Reed, 1989; Reed and Besunder, 1989; Levy and
Shen, 1995). Causes for differences in binding include lower plasma
levels of total protein, albumin, and
1-glycoprotein in neonates compared with
adults (Kearns and Reed, 1989; Reed and Besunder, 1989). Furthermore,
neonates exhibit an increased level of unconjugated bilirubin and free
fatty acids, which can act to displace albumin-bound drugs from their
binding sites (Kearns and Reed, 1989; Reed and Besunder, 1989). Since
Bmax1 is a measure of binding capacity, the
presence of binding inhibitors (i.e., unconjugated bilirubin and free
fatty acids) would have no effect on this parameter. Thus, it is likely
that the lower Bmax1 observed in neonatal
lambs is a result of lower total protein and albumin concentrations.
Our measurements of total protein in adult and neonatal sheep plasma
support this explanation, since we observed significantly lower
concentrations of total protein in neonatal plasma in comparison with
the adult. Similar increases in binding capacity of VPA with age were
observed in developmental studies in rats (Haberer and Pollack, 1994)
and guinea pigs (Yu et al., 1985).
Surprisingly, the binding affinity of VPA appeared to be higher in
neonatal lambs (Table 3). As mentioned, in the young there is generally
a higher concentration of substances such as unconjugated bilirubin and
free fatty acids, which can act as inhibitors of VPA binding (Kearns
and Reed, 1989; Reed and Besunder, 1989). The presence of such
inhibitors influences binding by decreasing binding affinity (i.e.,
increase in Kd1). In addition, the
lingering presence of fetal albumin in neonates usually results in a
decrease in binding affinity (Moreselli, 1976). Consistent with the
expected situation, binding affinity increases with age in rats
(Haberer and Pollack, 1994). An explanation for the unexpected results observed in sheep may be the presence of a rumen in these animals. In
newborn ruminants, blood glucose and fatty acid concentrations are
similar to monogastric animals. As the animal ages and the rumen
develops, glucose levels in the blood fall to half of what is observed
in nonruminants. In contrast, volatile fatty acid concentration in
blood increases substantially due to their production in the rumen and
subsequent absorption (Annison and Lewis, 1959). Higher concentrations
of fatty acids in adult sheep plasma may act as binding inhibitors,
resulting in the observed reduction in VPA binding affinity (i.e.,
increase in Kd1) in adult sheep.
Dose-Dependent Changes in VPA Metabolism.
The interpretation of VPA metabolite mass balance data from the
dose-dependent studies is complex, as the contributions to overall
elimination of the different metabolic pathways are not entirely
independent of each other. At high concentrations, saturation of
primary pathways of elimination results in the "shunting" of the
administered dose to normally minor routes of elimination. Our mass
balance data are consistent with metabolic saturation of -oxidation.
For both age groups, we observed an overall decrease in the percentage
of the administered dose recovered as -oxidation metabolites [i.e.,
(E)-2-ene, (E)-3-ene, 3-OH, and 3-keto VPA], with 3-keto VPA being the primary metabolite at all doses (Tables 4 and
5). A similar phenomenon has been previously observed in dose-ranging
studies in humans (Granneman et al., 1984; Dickinson et al., 1989). In
contrast, the contribution of 
-oxidation (5-OH VPA and 2-PGA)
and -1-oxidation (4-OH and 4-keto VPA, and 2-PSA) to VPA elimination
increased with increasing dose (Tables 4 and 5). Of the metabolites
derived from P-450 metabolism, the increase in the percentage of the
dose recovered as the -1-oxidation metabolite, 4-OH VPA, was the
greatest, suggesting increased formation of P-450 metabolites. As
mentioned, dose dependence has not been previously observed for -
and -1-oxidation in humans; however, the doses used in these
experiments were substantially lower than the doses used in our sheep
studies. The urinary excretion of 4-ene VPA was minor for both adult
and neonatal sheep and appeared only at the higher doses.
By far the main pathway of VPA elimination is glucuronidation. In adult
sheep, no significant change was observed in the recovery of VPA with
increasing dose. In fact, VPA-glucuronide accounted for 68.3 ± 6.5 to 78.7 ± 5.6% of the administered VPA at all doses (Table
4). A different situation was observed in lambs. At the 10-mg/kg dose,
29.2% of the administered VPA was recovered as the glucuronide
metabolite. This percentage significantly increased to 65.7 to 69.6%
at the three higher doses (Table 5). Previous dose-ranging experiments
in adult rats (Dickinson et al., 1979) and humans (Granneman et al.,
1984; Dickinson et al., 1989; Anderson et al., 1992) showed a pattern
similar to neonatal lambs wherein VPA-glucuronidation appeared to
increase with dose. The observed differences in changes in
VPA-glucuronidation with dose in adult rat and humans in comparison
with adult sheep could be a result of species differences in
VPA-glucuronidation. Species differences in the relative contribution
of other routes of elimination (i.e., especially -oxidation) may
also play a role since mass balance data for each metabolic pathway is
not independent of other routes of elimination.
In Vivo Estimation of Apparent Vmax and
Km of Overall VPA Elimination and VPA
Glucuronidation.
The mass balance data presented in Tables 4 and 5 clearly show that VPA
glucuronidation is the main pathway responsible for VPA elimination in
both adult sheep and neonatal lambs. In fact at the 50-, 100- and
250-mg/kg doses, the majority of the administered dose was recovered as
this metabolite in both neonatal lambs (~65-70%) and adult sheep
(~70-80%). Thus, the estimated apparent in vivo
Vmax and Km for
overall VPA metabolic elimination is largely reflective of
VPA-glucuronidation. This would explain the similar information
obtained from the Vmax and
Km estimates using plasma data and urine
data. Estimates from both methods suggest that
Vmax is similar in both age groups (Tables
6 and 7). The higher estimates of Vmax
obtained from plasma data (445.0 µg/min/kg for adult sheep and 429.9 µg/min/kg for lambs) in comparison with the urine data (288.5 µg/min/kg for adult sheep and 326.5 µg/min/kg for lambs) may be a
consequence of the fact that VPA-glucuronidation does not account for
elimination of the entire dose. The presence of other routes of
elimination will also influence the parameter estimates obtained from
plasma data. However, urine data directly monitor the production of the
specific metabolite of interest. Apparent
Km estimates from the two methods were
surprisingly similar with both, suggesting that
Km is higher in lambs. A similar phenomenon was observed in a study examining acetaminophen glucuronidation in
adult and fetal sheep microsomes (Wang et al., 1986). In this study,
estimates of Km obtained from microsomes
were higher for the fetus than for the adult. This difference in
Km was attributed to either a different
form of UDP-glucuronosyltransferase being present in fetal microsomes
or age-related differences in enzyme function resulting from
differences in the immediate environment of the
UDP-glucuronosyltransferases (Wang et al., 1986). It is possible that
the higher Km estimate observed for VPA
glucuronidation in neonatal lambs is a result of similar causes since
the neonatal lamb is an early phase in the transition from the fetal to
the adult situation.
The Michaelis-Menten parameter Vmax is
defined as the maximum enzyme capacity and is related to the total
concentration of enzyme. The other Michaelis-Menten parameter,
Km, is the Michaelis-Menten constant and
has an inverse relationship to enzyme affinity (Rowland and Tozer,
1980). Therefore, the apparent Vmax and
Km values from our dose-ranging studies
suggest that metabolic capacity is similar in the two age groups of
interest. However, since Km was estimated in vivo, it is not possible to determine whether a difference in
Km is related to developmental changes in
enzyme affinity and/or enzyme function. A higher apparent
Km in 10-day-old lambs explains the
relative changes in CL with increasing dose in both
age groups. Earlier we stated that VPA CL approximates intrinsic clearance. Intrinsic clearance is related to
Vmax and Km by
the following equation:
|
(6)
|
where CLint is the intrinsic clearance and
Cu is the unbound drug concentration
(Rowland and Tozer, 1980; Gibaldi and Perrier, 1982). At lower drug
concentrations, the denominator approximates Km and thus CLint 
Vmax/Km. Since
Km is smaller in adult sheep, VPA
CL estimates should be higher for adults than for
lambs in this first scenario. At higher concentrations, the denominator
of eq. 6 approximates Cu, and
CLint Vmax/Cu. Therefore,
as concentrations increase, the CL in adult and
neonatal sheep should become increasingly similar since
Vmax is similar for both age groups. Our
CL data follow this exact pattern, with adult
CL values being significantly higher at the two
lower doses (Fig. 3). By the 100-mg/kg dose, the adult
CL estimates have decreased to the point where they
are no longer different than the lamb values. The CL estimates from both age groups remain similar at the highest dose (Fig.
3).
Differences observed in the recovery of VPA-glucuronide in adult
(73.8% of the dose) and neonatal lambs (29.2% of the dose) following
the 10-mg/kg VPA dose are also consistent with a higher Km in neonatal lambs. VPA metabolism is
essentially a competition between various metabolic pathways.
Therefore, a decrease in the ability of one pathway to eliminate VPA
will result in the elimination of the compound by one of many alternate
metabolic routes. Thus, the higher apparent
Km of VPA glucuronidation in lambs would
allow a larger portion dose to be eliminated by other processes than would occur in adult sheep. With increases in dose, elimination pathways tend to saturate, which would allow a larger portion of VPA to
be glucuronidated. This theory is consistent with our data, as VPA
glucuronidation in neonatal lambs increases substantially with an
increase in dose. In fact, at the 50-mg/kg dose, almost the entire dose
is recovered (92.3 ± 7.3%), with the majority being
VPA-glucuronide (65.7 ± 8.5% of the dose). We have yet to recover ~50% of the dose in 10-day-old lambs following the
administration of the lowest dose (i.e., 10 mg/kg). Thus, the presence
of a high-affinity, low-capacity VPA elimination process would explain
our results. The nature of this process remains to be identified.
Vmax and Km
values of VPA glucuronidation have been previously estimated in adult
guinea pigs using both in vitro (i.e., 5% liver homogenate and
microsomes) and in vivo methodologies (Yu et al., 1993; Yu and Shen,
1996). In these studies, in vitro Vmax estimates of ~260 and ~170 µg/min/kg were obtained using liver homogenate and microsomes, respectively. A similar apparent
Vmax estimate of ~220 µg/min/kg was
obtained in vivo in dose-ranging drug infusion experiments.
Km estimates obtained for guinea pigs from
these studies were similar [~45 µg/ml (5% liver homogenate), ~23 µg/ml (microsomes), and ~22 µg/ml (in vivo)] (Yu et al.,
1993; Yu and Shen, 1996). These adult guinea pig estimates are
comparable with our estimates obtained for adult sheep using urine data
(Vmax = 288.5 µg/min/kg and
Km = 30.0 µg/ml; Table 7).
Km estimates from both adult sheep and
guinea pigs are within the clinical range of unbound drug (Yu, 1984)
suggesting that VPA glucuronidation is nonlinear at therapeutic doses
in these two species.
Estimation of in vivo apparent Vmax and
Km of overall VPA elimination has also been
assessed using plasma data from developing rats (Haberer and Pollack,
1994). The Vmax (~302 µg/min/kg) and Km (~100 µg/ml) for 10-day-old rats in
this study are comparable with our own estimates obtained from lamb
plasma data (Vmax = 429.9 µg/min/kg and
Km = 69.6 µg/ml; Table 6). However,
Vmax estimates in slightly older rats
(i.e., 20 days and 60 days) were substantially larger [i.e., 975 µg/min/kg (20-day-old) and 4460 µg/min/kg (60-day-old)] than even
our adult estimates. It must be mentioned that these estimates were
obtained following single bolus dose experiments. Reliable estimation
of in vivo Vmax and Km requires dose-ranging experiments at
three to four dose levels with at least some doses wherein saturation
kinetics is evident (Metzler and Tong, 1981; Gibaldi and Perrier,
1982). Although not nearly as reliable, estimates can still be obtained
from single dose studies exhibiting some degree of saturation kinetics.
However, signs of saturation kinetics (i.e., nonlinear terminal slope
of semilogarithmic plasma concentration versus time plot) were only present in the plasma profiles of the 5- and 10-day-old rats in these
studies. Thus, it is difficult to assess the validity of the estimates
obtained from the older rats (i.e., 20- and 60-day-old) and make
comparisons with our own adult data.
In summary, developmental differences exist in both plasma protein
binding and metabolism of valproic acid. VPA plasma protein binding
capacity was higher in adult sheep, whereas binding affinity appeared
to be lower. The unexpected lower binding affinity in adult sheep may
be attributed to higher volatile fatty acid levels in plasma of adult
ruminants in comparison with lambs and monogastric animals. Differences
in apparent Km rather than the metabolic capacity of the glucuronidation pathway appeared to be primarily responsible for the differences in CL previously observed between adult and neonatal sheep (Wong et al., 2000). Since
mass balance data from an earlier study (Wong et al., 2000) suggest
that rapid changes in VPA glucuronidation occur during the time period
between 10 days and 2 months of age, future dose-ranging studies in 1- and 2-month-old lambs would provide further insight on developmental
changes in VPA-glucuronidation.
These studies were supported by funding from the Medical
Research Council of Canada. H.W. was the recipient of a Pharmaceutical Manufacturers Association of Canada/-Health Research Foundation and
Medical Research Council Graduate Scholarship in Pharmacy. S.K.
was the recipient of a University of British Columbia Graduate Fellowship. D.W.R. is the recipient of an investigatorship award from
the British Columbia Children's Hospital Foundation.
Abbreviations used are:
VPA, valproic
acid;
AIC, Akaike's information criterion;
AUC0-, area
under the curve of arterial plasma concentration-time profile;
AUC, AUC0- of unbound drug;
Cb, protein bound drug concentration;
CLint, intrinsic clearance;
CLtb, total body
clearance of the total drug;
CL, total body
clearance based upon unbound drug concentrations;
Cu, unbound drug concentration;
C, unbound plasma concentration at
the midpoint of the urine collection interval;
CV, coefficient of
variation;
fp, area weighted unbound
fraction of the drug;
fu, unbound fraction;
tmid, time at the midpoint of the urine
collection interval;
Vdss, steady-state volume of
distribution;
Vd, steady-state volume of
distribution based upon unbound drug concentrations;
Vdss', steady-state volume of distribution corrected for the effects of
saturable protein binding;
2-ene VPA, 2-n-propyl-2-pentenoic acid;
3-ene VPA, 2-n-propyl-3-pentenoic acid;
4-ene VPA, 2-n-propyl-4-pentenoic acid;
3-keto VPA, 2-n-propyl-3-oxopentanoic acid;
4-keto VPA, 2-n-propyl-4-oxopentanoic acid;
3-, 4-, and 5-OH VPA,
3-, 4-, and 5-hydroxy VPA, respectively;
2-PSA, 2-propylsuccinic acid;
2-PGA, 2-propylglutaric acid.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
2-valproic acid to rat plasma proteins.
Pharm Res
7:
461-467[Medline].
