Download distribution kinetics of salicylic acid in the dual

0090-9556/99/2703-0373–378$02.00/0
DRUG METABOLISM AND DISPOSITION
Copyright © 1999 by The American Society for Pharmacology and Experimental Therapeutics
Vol. 27, No. 3
Printed in U.S.A.
DISTRIBUTION KINETICS OF SALICYLIC ACID IN THE DUAL-PERFUSED
RAT LIVER PREPARATION
SELMA SAHIN
AND
MALCOLM ROWLAND
Hacettepe University, Faculty of Pharmacy, Ankara, Turkey (S.S.); and School of Pharmacy and Pharmaceutical Sciences, University of
Manchester, Manchester, United Kingdom (M.R.)
(Received August 19, 1997; accepted October 14, 1998)
This paper is available online at http://www.dmd.org
ABSTRACT:
area product/blood flow ratio lies between the values of 0.06 and
7.0. Our estimates (3.0 for venous output and 6.7 for arterial input)
indicate that hepatic uptake of salicylic acid is dependent on both
perfusion and permeability. The volume terms were calculated
using two different methods, standard and specific. Regardless of
the compound and method, the volume of distribution after arterial
administration was larger than that after venous administration. In
addition, a volume of distribution approximately twice that of the
total aqueous space (i.e., HA, 2.23 6 0.13 versus 1.10 6 0.07 ml/g;
PV, 1.72 6 0.16 versus 0.68 6 0.04 ml/g) implies that salicylic acid
has a significant affinity for hepatic tissue. A similar tissue-toperfusate partition coefficient associated with HA and PV input
(5.40 6 0.38 versus 6.48 6 0.56) indicates that affinity of salicylic
acid for hepatic tissue is independent of the route of input.
Hepatic elimination may be influenced by various processes, including organ perfusion, binding to blood and tissue components,
cellular activity, and membrane permeability (Rowland and Tozer,
1995). In the liver, the primary barrier is at the cellular membrane
level because the vascular endothelial membrane, with its large
fenestrae, does not offer material resistance to the movement of drug
molecules. A cell membrane rate limitation in hepatic uptake is most
likely to be seen with polar molecules.
Most of the basic physiological concepts surrounding hepatic uptake have been developed using the isolated perfused rat liver preparation, mostly receiving single perfusion via the portal vein (PV1;
Rowland and Evans, 1991). Yet, in vivo, the liver receives two blood
supplies via the hepatic artery (HA) as well as via the PV. Although
some investigators propose that these two blood supplies mix before
or within the sinusoids (Nakai et al., 1979; Watanabe et al., 1994), the
body of data supports the view that at least part of the HA flow
perfuses a specific arterial space (Field and Andrews, 1968; Ahmad et
al., 1984; Reichen, 1988; Kassissia et al., 1994; Pang et al., 1994;
Sahin and Rowland, 1998a).
Although there have been some studies comparing the hepatic
dispositional characteristics of various compounds between PV and
HA in the dual-perfused liver (Ahmad et al., 1984; Pang et al., 1994),
apparently no such studies have been reported for compounds showing membrane permeability rate-limited uptake. The aim of the current study was to address the latter issue, using salicylic acid, whose
uptake had been shown to be permeability rate-limited in the single
PV perfused rat liver (Hussein et al., 1994), as a model compound.
Materials and Methods
Salicylic acid (14C; 59 mCi/mmol) was obtained from Amersham (Amersham, UK); 14C-sucrose (0.1 mCi/ml) and 3H-water (100 mCi/ml) were obtained from ICN Biomedicals, Inc., (Costa Mesa, CA). Sodium salicylate and
all other chemicals were of analytical grade and obtained commercially.
Perfusion Procedure. The single-pass dual-perfused in situ rat liver preparation used male Sprague-Dawley rats (318 6 11 g; wet liver weight 13.5 6
0.7 g; n 5 7). Krebs-bicarbonate solution containing 5 mg/liter sodium
salicylate was used as the perfusion medium. The surgical procedure was the
same as that described previously (Sahin and Rowland, 1998b). Briefly, after
induction of anesthesia, the bile duct was cannulated and loose ligatures were
placed around the PV, ensuring exclusion of the HA. The gastroduodenal
artery and branches of the celiac artery (i.e., the left gastric and splenic arteries)
We thank the Turkish Government for a studentship (S.S.); research supported
were tied very close to their junctions. Initially, the PV was cannulated with a
by a grant from the Wellcome Trust.
16-gauge (Argyle Medicut, o.d. 1.7 3 45 mm) catheter and the perfusion
1
Abbreviations used are: HA, hepatic artery; PV, portal vein; PS, permeabilitystarted at a flow rate of 12 ml/min. The thoracic vena cava was cannulated
surface area product.
through the right atrium for the collection of the outflow perfusate. The HA
was cannulated indirectly through the celiac artery using a 18-gauge (Argyle
Send reprint requests to: Dr. Selma Sahin, Hacettepe University, Faculty of
Medicut, o.d. 1.3 3 45 mm) or 20-gauge (Argyle Medicut, o.d. 1.1 3 45 mm)
Pharmacy, 06100 Ankara, Turkey. E-mail: [email protected].
catheter and the second perfusion was started at a flow rate of 3 ml/min. All
373
Downloaded from dmd.aspetjournals.org at ASPET Journals on May 5, 2017
The hepatic distribution kinetics of salicylic acid was determined
using a single-pass dual hepatic artery (HA) and portal vein (PV)
perfused in situ rat liver preparation. Bolus doses of [14C]salicylic
acid and of reference markers ([3H]-water and [14C]-sucrose) were
injected in a random order into either the HA or PV and then, after
an appropriate interval, into the alternate vessel. The hepatic outflow profile of [14C]salicylic acid displayed a characteristic sharp
peak followed by a slower eluting tail, whereas sucrose and water
displayed unimodal outflow profiles. The biphasic outflow profile
indicates that the hepatic distribution of salicylic acid is not instantaneous but is limited by a permeability barrier. The in situ permeability surface area product for [14C]salicylic acid was 3.35 6 0.26
ml/min/g for PV and 7.45 6 1.50 ml/min/g for HA administration.
Furthermore, theory dictates that hepatic uptake is influenced by
both perfusion and permeability if effective permeability surface
374
SAHIN AND ROWLAND
C~t! z Q
f~t! 5
D
E
V C 5 Q C z MTT C
(8)
Q C 5 Q PV 1 ~1 2 f 3! z Q HA
(9)
where VC is the volume of distribution of the common space, MTTC is the MTT
after the PV injection operating in a dual perfusion mode, QC is the flow rate
to the common space, QHA and QPV are the HA and PV flow rates, respectively,
and f3 is the fraction of QHA perfusing the specific hepatic arterial space. The
fraction of HA flow to the common space (1 2 f3) was taken as 0.83 (Sahin
and Rowland, 1998a):
after the arterial injection (Sahin and Rowland, 1998a):
(1)
where C(t) is the concentration of radioactivity, Q is the total perfusate flow
(ml/s), and D is the injected dose (in dpm). The delay in the nonhepatic region
of the experimental system (i.e., PV, 2.40 6 0.03 s; HA, 2.27 6 0.06 s; n 5
6; mean 6 S.E.) was simply subtracted from the midtime of the sampling
interval. The moments of the frequency outflow against midtime profiles were
estimated by numerical integration, and then the parameters related with these
moments (e.g., VH, CV2) were calculated using the following equations:
AUC 5
In the specific method, total hepatic aqueous space is considered to consist
of a common and a specific space. The common space is perfused by both
arterial and venous flows, whereas the specific space receives a fraction of the
arterial blood (Sahin and Rowland, 1998a). Furthermore, the common and
specific spaces are parallel to each other, and solute or solvent exchange does
not take place between these spaces. The volume terms associated with each
input are given by the following equations:
after the venous injection:
MTT HA 5 f 3 z MTT sa 1 ~1 2 f 3! z MTT C
(10)
V sa 5 ~ f 3 z Q HA! z MTT sa
(11)
so that
MTT HA 5
V sa
1 ~1 2 f 3! z MTT C
Q HA
(12)
Rearrangement of eq. 12 yields
V sa 5 Q HA z @MTT HA 2 ~1 2 f 3! z MTT C#
(13)
V HA 5 V C 1 V sa
(14)
`
C~t!dt
(2)
Finally
0
MTT 5
VTT 5
E
`
E
`
t z C~t!dt
0
AUC
(3)
t 2 z C~t!dt
0
AUC
2 ~MTT! 2
(4)
where AUC is the area under the concentration versus time profile; MTT is the
mean transit time, which is the average time taken for a molecule to pass
through the organ; and VTT is variance of transit times, which is a measure of
temporal spreading or dispersion of transit times within the organ. These
equations assume that solute is not eliminated and total perfusate flow rate (Q)
is constant throughout.
The normalized variance (CV2) is given by
VTT
CV 2 5
~MTT! 2
(5)
CV2 is a dimensionless parameter and has been used as a measure of relative
dispersion of drug within the liver (Roberts et al., 1988).
The recovery (F) is defined as:
F5
AUC z Q
D
where Vsa is the volume of the specific arterial space, and VHA is the volume
of distribution after HA input.
Estimation of Permeability-Surface Area Product (PS). Estimates of the
PS have been reported for diazepam (Diaz-Garcia et al., 1992) and salicylic
acid (Hussein et al., 1994) by fitting the two-compartment dispersion model to
hepatic outflow data. This fitting procedure is complex and time-consuming in
comparison to two recently proposed model-independent approaches for the
estimation of the PS value (Wu et al., 1993; Chou, 1995). Although these two
methods evolved from the same origin (from the CV2 of the two-compartment
dispersion model defined by Yano et al. (1989) for noneliminated substances),
the one proposed by Wu et al. (1993) assumes that no metabolism or protein
binding occurs in the organ, whereas the other makes no such assumption. In
the present study, the method of Chou (1995) was adopted for the estimation
of the PS value of [14C]salicylic acid, namely:
fu B z PS 5
(6)
2
(15)
fu B z PS
Q z MTT HB
(16)
fu B z PS
Q z ~MTT H 2 MTT HB!
(17)
k 12 5
(7)
D
where fuB is the unbound fraction in the perfusate, MTTHB and CV2HB are mean
transit time and normalized variance for the extracellular marker (sucrose), and
MTTH and CV2H are the corresponding values for the test substance (salicylic
acid). The influx (k12) and efflux (k21) first-order rate constants across the
hepatocellular membrane can be evaluated from the fuBzPS value (Chou, 1995);
that is
The volume of distribution (VH) is estimated using two different methods,
namely, the standard and specific methods. The standard method is given by:
V H 5 Q z MTT
S
2Q
MTT HB
12
~CV H 2 CV 2HB!
MTT H
2
k 21 5
Downloaded from dmd.aspetjournals.org at ASPET Journals on May 5, 2017
operative procedures were completed within 20 to 30 min without interruption
of flow to the liver. Viability of the perfused liver was assessed from measurement of bile flow, perfusate recovery, hepatic arterial pressure, and from
gross appearance.
Injection Preparations. Two tracer markers (i.e., either [14C]salicylic acid
and 3H-water, or [14C]sucrose and 3H-water) were injected together as a bolus
(50 ml) in saline; doses used were [14C]salicylic acid, 0.31 mCi; 14C-sucrose,
0.16 mCi; and 3H-water, 0.74 mCi.
Experimental Procedure. All of the experiments were performed after at
least a 20-min initial stabilization period. Salicylic acid (14C) and reference
markers (14C-sucrose, tritiated water) were administered as a bolus in a random
order either into the HA or PV, and then into the alternate vessel, separated by
an appropriate time interval based on prior washout information. Immediately
after an injection, the total effluent was automatically collected at 2-s intervals
for 2 min, using a motor-driven carousel, and thereafter at increasing time
intervals, for an additional 3 to 6 min. The activities of 3H and 14C were
determined simultaneously in 200 ml of outflow perfusate, after the addition of
5 ml of scintillation fluid, with the results expressed in dpm.
Data Analysis. The frequency output [f(t), 1/s] of the injected radiolabeled
material at the midpoint time of the sampling interval was calculated using the
following equation:
HEPATIC DISTRIBUTION KINETICS OF SALICYLIC ACID
375
The apparent tissue-perfusate partition coefficient (Kp) was estimated from
the ratio of the influx and efflux rate constants (Yano et al., 1989).
Kp 5
k 12
k 21
(18)
In addition, an estimate of the intracellular unbound fraction (fuC) was
obtained using the relationship between the rate constants and volumes of
distribution (Hussein et al., 1994); it is given by
fu C 5
k 21 z V CV
k 12 z V B
(19)
where VCV is defined as the aqueous cellular volume and is calculated as the
difference between the distribution volumes of water and sucrose (VB), an
extracellular marker. This calculation assumes that salicylic acid distributes
into the total water space of the liver and that the influx and efflux PS values
are equal i.e., PSin 5 PSout (Hussein et al., 1994).
Data are presented as mean 6 S.E. and compared using a paired or unpaired
Student’s t test. A p value less than .05 was considered significant.
FIG. 1. Semilogarithmic plots of fractional rate of efflux of 14C-sucrose (■),
3
H-water (h) and [14C]salicylic acid (F) after injection into the hepatic artery
(A) and portal vein (B) of a representative dual perfused rat liver.
comparable (i.e., HA, 5.40 6 0.38; PV, 6.48 6 0.56). Furthermore,
the estimated unbound fractions of [14C]salicylic acid in the cells, fuC,
were almost identical whether injected into the HA or PV (i.e., HA,
0.36 6 0.02; PV, 0.31 6 0.03). The ratio of effective permeabilitysurface area product to blood flow (PS/Q) was 3.0 for PV injection
and 6.7 for HA injection.
Discussion
Choice of the Compound. In the present study, salicylic acid was
chosen as the model compound because it is a small, relatively polar
molecule displaying low permeability across the hepatocyte membrane (Hussein et al., 1994) and it binds to cellular constituents of
liver cells (Yoshikawa et al., 1984). Although the distribution kinetics
of salicylic acid have been investigated under various conditions,
including flow rate and concentration in the single (PV) perfused rat
liver (Hussein et al., 1994), no information is available with regard to
route of hepatic input, the subject of this study in the dual perfused rat
liver. Additionally, sucrose and water were chosen as reference markers because they represent the asymptotes with regard to hepatocyte
Downloaded from dmd.aspetjournals.org at ASPET Journals on May 5, 2017
Results
Outflow Profiles. In the presence of sodium salicylate (5 mg/l) in
the perfusate, representative frequency outflow versus midtime profiles of [14C]salicylic acid and the two reference markers (14C-sucrose
and 3H-water) after bolus administrations into the HA and PV are
depicted in Fig. 1, A and B, respectively. Regardless of the route of
entry, sucrose and water displayed unimodal outflow profiles. Irrespective of the marker, the outflow profiles after arterial administration were flatter than after PV administration (e.g., fmax: for sucrose,
HA, 0.051 6 0.003 and PV, 0.092 6 0.008 1/s; and for water, HA,
0.018 6 0.001 and PV, 0.030 6 0.002 1/s).
In contrast to the reference markers, the hepatic outflow profiles of
[14C]salicylic acid were clearly multiphasic (Fig. 2). This multiphasic
profile was characterized by a sharp peak followed by a slow eluting
flat tail. In addition, apart from a slightly diminished first peak after
HA injection (e.g., fmax, HA: 0.011 6 0.001 and PV: 0.016 6 0.002
1/s, tmax: 5.80 6 0.31 versus 7.50 6 0.75 s), arterial and venous
injections produced very similar outflow profiles.
Moment Analysis. Table 1 presents the results of moment analysis
for 14C-sucrose, 3H-water, and [14C]salicylic acid. Regardless of the
compounds, the recovery of injected material was over 90%. The
mean transit times, and hence the volumes of distribution VH, increased from sucrose to salicylic acid whether they were administered
into the HA or PV. Irrespective of the marker, the mean transit times
(s) after HA injection were longer than that after PV injection-sucrose:
21.6 6 2.3 versus 12.7 6 1.2 ( p , .05); water: 60.4 6 4.2 versus
37.0 6 2.4 ( p , .01); salicylic acid 121.1 6 3.6 versus 92.4 6 5.7
( p , .05). Also, regardless of the method used, the HA estimates for
the volume of distribution were significantly larger than those of the
PV estimates (Table 1). Although the relative spreading (CV2) of
[14C]salicylic acid was very similar whether administered into the HA
or PV, the corresponding values for sucrose and tritiated water were
larger after arterial than venous input ( p , .01).
Membrane Permeability. Values of the PS of salicylic acid (given
that fuB 5 1 in the absence of albumin) are summarized in Table 2.
The estimated PS value for salicylic acid after the HA administration
was approximately twice that of the corresponding value after PV
injection (PS, ml/min/g liver; HA: 7.45 6 1.50 and PV: 3.35 6 0.26;
p , .01). Although influx (k12) and efflux (k21) first-order rate
constants across the hepatocellular membrane were larger after arterial than venous injections (Table 2), the difference between these rate
constants was not significant. Regardless of the injection site, the
tissue-perfusate partition coefficients (KP) for [14C]salicylic acid were
376
SAHIN AND ROWLAND
permeability. Sucrose is completely excluded from the hepatocytes
(zero permeability, Alpini et al., 1986), whereas water has free access
to the hepatocytes. Accordingly, they occupy the extracellular and
total aqueous spaces, respectively.
Salicylic acid is metabolized in man and animals to salicyluric acid,
salicyl phenolic, and acyl glucuronides, gentisic acid and gentisuric
acid. In man and rat, the main metabolite is the glycine conjugate,
salicyluric acid. Available data suggest that salicyluric acid and salicyl
phenolic glucuronide can be reversibly metabolized back to salicylic
acid (Morris, 1990). Nevertheless, more recent reports support the
idea that salicylic acid is not metabolized (Hussein et al., 1994; Shetty
et al., 1994) or is minimally metabolized by the liver during single
pass (i.e., 20 min after infusion, in a recirculating mode, the percentage of metabolites in the perfusate is only about 4%; Laznicek and
Laznickova, 1994). The same was assumed to hold in the present
study. Therefore, measured total 14C-radioactivity was taken to represent [14C]salicylic acid.
Outflow Profiles. Visual comparison of the outflow profiles for
sucrose, water, and salicylic acid indicates that distribution of salicylic
acid, unlike water, is not instantaneous in the liver but is limited by a
permeability barrier. The multiphasic profile of salicylic acid has been
previously investigated by Ichikawa et al. (1992) and more recently by
Hussein et al. (1994) in the single (PV) perfused rat liver. The current
study extends this knowledge to the dual perfused liver. The similarity
in the outflow profiles of salicylic acid obtained after HA and PV
injections indicates that distribution of salicylic acid is minimally
affected by the route of input. Regardless of the route of administration, the rapidly eluting peak of [14C]salicylic acid emerged almost at
the same time as sucrose, indicating that a certain fraction of salicylic
acid, the throughput component, emerges without ever entering the
hepatocytes on transit through the liver, essentially due to poor permeability. The fraction of throughput component increases with an
increase in the flow rate (i.e., 4% for 15 ml/min and 12% for 30
ml/min; Hussein et al., 1994). In contrast, the more slowly eluting
fraction of the outflow profile, referred to as the “returning component” (Goresky, 1983) represents the fraction of [14C]salicylic acid
that has entered the hepatocytes and returned to the vascular space,
delayed by intracellular binding and limited permeability (Hussein et
al., 1994).
Moment Analysis. Preliminary experiments indicated that pres-
3~3D N 1 1! # fu B z PS/Q #
0.1~1 1 2D N!
~1 1 1.9D N! 2
(20)
For a noneliminated substance such as salicylic acid, DN can be
estimated from CV2 through the relationship DN 5 CV2/2 (Roberts et
al., 1988). This yields a value of DN of 0.42 to 0.44, given that CV2
5 0.84 – 0.88 (Table 1). Thus, for equation 20 to hold, fuB-PS/Q must
lie between 0.056 and 7.0, which is the case in the perfusate experiments being 3.0 for PV input and 6.7 for HA input. Although these
high values suggest a tendency to a perfusion rate-limitation in the
perfusate experiments in the absence of binding protein, with salicylic
acid bound to circulating albumin in plasma (fuB 5 0.11 in the
presence of 4.9% bovine serum albumin; Ichikawa et al., 1992),
fuBzPS/Q is about 0.33 and 0.74 (for PV and HA respectively),
suggesting that in vivo hepatic uptake is dependent on both perfusion
and permeability. Notwithstanding, these values for salicylic acid are
sufficiently high that distribution equilibrium between the liver and
circulating drug will be rapid (in minutes) compared with the overall
elimination of drug from the body, which takes many hours (t1/2 5
10 h after i.v. injection of 173 mg/kg salicylic acid; Hirate et al.,
Downloaded from dmd.aspetjournals.org at ASPET Journals on May 5, 2017
FIG. 2. Linear plots of fractional rate of efflux of [14C]salicylic acid after
injections into the hepatic artery (■) and portal vein (h) of a representative rat
liver operating in the dual perfusion mode.
ence of salicylate in the perfusate has no effect on the distribution
kinetics of sucrose and water. Nevertheless, this was not the case for
labeled salicylic acid; MTT was decreased and the recovery of the
administered dose was increased with the addition of salicylate to the
perfusate. With regard to [14C]salicylic acid, a longer MTT and,
hence, larger VH than for sucrose clearly indicates that salicylic acid
is not confined to the extracellular space of the liver. In addition, a VH
approximately twice that of the total aqueous space suggests that
salicylic acid has affinity for hepatic tissue. These results are in
accordance with the results of those in the in situ single (PV) perfused
rat liver preparation (Hussein et al., 1994). The VH values estimated
using the standard and specific methods support the idea that the
volume estimates are dependent upon the method used and could be
misleading. Although PV estimates were minimally affected by the
choice of the method, this effect was dramatic for HA estimates.
Membrane Permeability and Tissue Binding. The extent of distribution of a compound into an organ (i.e., liver) can be rate-limited
by either perfusion or permeability (Rowland and Tozer, 1995).
Recently an attempt has been made to correlate hepatocellular permeability with physicochemical properties (e.g., logD (n-octanol: pH
7.4 aqueous buffer; Chou et al., 1995) for variety of compounds
studied in the isolated rat liver. It was concluded that for compounds
that have logD values greater than 0, the uptake is flow limited either
if the perfusate contains no binding protein or binding of compound in
the perfusate is negligible. On the other hand, the uptake is limited by
permeability irrespective of degree of binding when the logD value of
a compound is less than 23. If the logD value lies between these
limits, as in the case of salicylic acid which has a logD value of 22.17
(Chou et al., 1995), hepatic uptake is anticipated to be governed by
both flow rate and magnitude of permeability. This expectation is
supported by considering the relative effective permeability-surface
area (fuBzPS), flow rate (Q) and the dispersion number (DN), a dimensionless parameter used as a measure of relative axial spreading of a
solute within the liver. Unlike permeability, the general similarity of
DN estimated for a variety of substances (0.2– 0.5) of differing physicochemical properties indicates that this parameter characterizes the
hepatic microvascular morphology rather than that of a drug (Rowland and Evans, 1991).
Theory dictates that hepatic uptake is influenced by both perfusion
and permeability if fuB.PS/Q lies within the following bounds (Chou
et al., 1995)
377
HEPATIC DISTRIBUTION KINETICS OF SALICYLIC ACID
TABLE 1
Mean (6S.E.) statistical moment analysis parameters for
14
C-sucrose, 3H-water and [14C]salicylic acid in the dual perfused rat liver
VH
Compound
Route of Input
CV2
MTT (s)
Standard Method
Specific Method
ml/g
Sucrose (n 5 7)
Water (n 5 14)
Salicylic acid (n 5 7)
HA
PV
HA
PV
HA
PV
21.6 6 2.3*
12.7 6 1.2
60.4 6 4.2**
37.0 6 2.4
121.1 6 3.6*
92.4 6 5.7
0.40 6 0.04*
0.23 6 0.02
1.10 6 0.07***
0.68 6 0.04
2.23 6 0.13*
1.72 6 0.16
0.27 6 0.01**
0.23 6 0.02
0.79 6 0.04***
0.67 6 0.06
1.85 6 0.14***
1.69 6 0.16
0.66 6 0.07**
0.39 6 0.01
0.85 6 0.07**
0.55 6 0.01
0.84 6 0.04
0.88 6 0.02
* p , .05; ** p , .01; *** p , .001 for the HA versus PV results.
TABLE 2
Mean (6S.E.) distributional parameters of [14C]salicylic acid in the
dual-perfused rat liver
Route of Administration
Parametera
a
HA (n 5 5)
3.35 6 0.26
0.250 6 0.025
0.039 6 0.003
6.48 6 0.56
0.31 6 0.03
7.45 6 1.50
0.390 6 0.106
0.070 6 0.017
5.40 6 0.38
0.36 6 0.02
See text for definition of symbols; * p , .01 for HA versus PV results.
1989). Also, with permeability much greater than intrinsic clearance
(PS .. CLint), at equilibrium, the unbound concentration within the
hepatocyte is expected to be essentially the same as that in perfusing
blood.
The mean PS estimate of [14C]salicylic acid after PV administration
(3.35 6 0.26 ml/min/g liver) is in good agreement with the previous
results obtained both in isolated hepatocytes in vitro (6.4 ml/min/g
liver; Ichikawa et al., 1992) and the single PV perfused in situ liver
(4.6 ml/min/g liver; Hussein et al., 1994). Although the PS value for
[14C]salicylic acid after arterial administration (7.45 6 1.50 ml/min/g
liver) was similar to that determined from in vitro hepatocytes studies
by Ichikawa et al. (1992), no such data are available in the in situ liver
preparation. In the presence of albumin, due to binding to this protein,
a substantial decrease in the PS value of salicylic acid has been
reported both in the isolated hepatocytes and in situ liver preparation
(e.g., 1.5 ml/min/g liver for isolated hepatocytes and 1.6 ml/min/g
liver for the in situ rat liver; Miyauchi et al., 1993). In addition, Hirate
et al. (1989) observed that the disposition characteristics of salicylic
acid are markedly altered in the presence of low plasma protein
concentration (i.e., analbuminemic versus control rats) due to reduction in the protein binding of salicylic acid.
Low recovery of the [14C]salicylic acid (i.e., 59% for HA and 69%
for PV) obtained in the absence of salicylic acid in the perfusate
during preliminary experiments was attributed to the binding to hepatic tissue rather than elimination of the compound. This idea was
supported by displacement of some [14C]salicylic acid after subsequent bolus injection of unlabeled material (i.e., 9 –12% of injected
dose for HA and 6% of injected dose for PV over a period of 1–2
min). Additionally, regardless of both route of input and salicylate in
the perfusate, alteration of both fractional recovery and displacement
of [14C]salicylic acid with sodium salicylate in the bolus give support
to the idea that binding to tissue is reversible and saturable. Moreover,
a Kp value of approximately 6 also suggests that salicylic acid has an
affinity to hepatic tissue whether delivered via the HA or PV. This
value of Kp is in agreement that in the in situ rat liver perfused via the
PV (about 6; Hussein et al., 1994) and also with in vivo values (e.g.,
References
Ahmad AB, Bennett PN and Rowland M (1984) Influence of route of hepatic administration on
drug availability. J Pharmacol Exp Ther 230:718 –725.
Alpini G, Garrick RA, Jones MJT, Nunes R and Tavoloni N (1986) Water and nonelectrolyte
permeability of isolated rat hepatocytes. Am J Physiol 251:C872–C882.
Chou CH (1995) Physiological modelling of hepatic elimination: A quantitative approach with
an axial dispersion model. Ph.D. thesis, University of Manchester, Manchester, England, 271
pp.
Chou CH, McLachlan AJ and Rowland M (1995) Membrane permeability and lipophilicity in the
isolated perfused rat liver: 5-Ethyl barbituric acid and other compounds. J Pharmacol Exp
Ther 275:933–940.
Diaz-Garcia JM, Evans AM and Rowland M (1992) Application of the axial dispersion model of
hepatic drug elimination to the kinetics of diazepam in the isolated perfused rat liver.
J Pharmacokinet Biopharm 20:171–193.
Field CD and Andrews WH (1968) Investigation of the hepatic arterial “space” under various
conditions of flow in the isolated perfused dog liver. Circ Res 23:611– 622.
Goresky CA (1983) Kinetic interpretation of hepatic multiple-indicator dilution studies. Am J
Physiol 245:G1–G12.
Hirate J, Kato Y, Horikoshi I, Nagase S and Ueda CT (1989) Further observations on the
disposition characteristics of salicylic acid in analbuminemic rats. Biopharm Drug Dispos
10:299 –309.
Hussein Z, McLachlan AJ and Rowland M (1994) Distribution kinetics of salicylic acid in the
isolated perfused rat liver assessed using moment analysis and the two-compartment axial
dispersion model. Pharm Res 11:1337–1345.
Ichikawa M, Tsao SC, Lin TH, Miyauchi S, Sawada Y, Iga T, Hanano M and Sugiyama Y (1992)
“Albumin-mediated transport phenomenon” observed for ligands with high membrane per-
Downloaded from dmd.aspetjournals.org at ASPET Journals on May 5, 2017
PS (ml/min/g)*
k12 (1/s)
k21 (1/s)
KP
fuC
PV (n 5 7)
6.4 after a dose of 10 mg/kg) estimated by Hussein et al. (1994), based
on the data supplied by Hirate et al. (1989). The concentrationdependent characteristics of Kp (i.e., a decrease with an increase in the
concentration of salicylate in the perfusate; Hussein et al., 1994) is
indicative of saturable tissue binding.
An absence of binding protein in the perfusate (fu 5 1) offers an
opportunity to investigate the extent of [14C]salicylic acid binding
within the hepatic tissue (i.e., 1 2 fuc). The presence of salicylate in
the perfusate results in a concentration-dependent decrease in the
hepatic tissue binding of [14C]salicylic acid (e.g. fuc values are 0.37,
0.59, and 0.93 at 0, 100, and 200 mg/l, respectively; Hussein et al.,
1994). In the present study, intracellular binding characteristics of
salicylic acid are unaffected by route of administration and the values
of fuc are in a good agreement with the literature data (Hussein et al.,
1994).
In summary, the present study demonstrates that the distribution
kinetics of salicylic acid is influenced by route of administration.
Nevertheless, its hepatic uptake process is independent of the site of
administration and governed by both flow rate and cellular permeability located at the level of hepatocytes. Also, the larger in situ PS
seen after HA input is probably attributable to the presence of the
specific arterial space and associated hepatocytes. Additionally, an
alternative method to the standard method is proposed for the estimation of volume of distribution of compounds in the dual perfused liver.
Thus, using the water content of the liver obtained by desiccation as
the reference (i.e., 0.72 6 0.01 ml/g liver; Sahin and Rowland,
1998a), the specific method for tritiated water proved superior to the
standard method, especially after arterial administration (0.79 6 0.04
versus 1.10 6 0.07, ml/g liver; Table 1).
378
SAHIN AND ROWLAND
meability: Effect of the unstirred water layer in the Disse’s space of rat liver. J Hepatol
16:38 – 49.
Kassissia I, Brault A and Huet PM (1994) Hepatic artery and portal vein vascularization of
normal and cirrhotic rat liver. Hepatology 18:1189 –1197.
Laznicek M and Laznickova A (1994) Kidney and liver contributions to salicylate metabolism in
rats. Eur J Drug Metab Pharmacokinet 19:21–26.
Miyauchi S, Sawada Y, Iga T, Hanano M and Sugiyama Y (1993) Comparison of the hepatic
uptake clearances of fifteen drugs with a wide range of membrane permeabilities in isolated
rat hepatocytes and perfused rat livers. Pharm Res 10:434 – 440.
Morris ME (1990) Pharmacokinetics and protein binding of salicylate metabolites in rats. Drug
Metab Dispos 18:809 – 811.
Nakai M, Tamura T, Kamiya A and Togawa T (1979) Control mechanisms of intrasinusoidal
flow pattern of blood. Jpn J Physiol 29:597– 608.
Pang KS, Sherman IA, Schwab AJ, Geng W, Barker III F, Dlugosz JA, Cuerrier G and Goresky
CA (1994) Role of the hepatic artery in the metabolism of phenacetin and acetaminophen: An
intravital microscopic and multiple-indicator dilution study in perfused rat liver. Hepatology
20:672– 683.
Reichen J (1988) Role of the hepatic artery in canalicular bile formation by the perfused rat liver:
A multiple indicator dilution study. J Clin Invest 81:1462–1469.
Roberts MS, Donaldson JD and Rowland M (1988) Models of hepatic elimination: Comparison
of stochastic models to describe residence time distributions and to predict the influence of
drug distribution, enzyme heterogeneity, and systemic recycling on hepatic elimination.
J Pharmacokinet Biopharm 16:41– 83.
Rowland M and Evans AM (1991) Physiological models of hepatic drug elimination, in New
Trends in Pharmacokinetics (Rescigno A and Thakur AK eds) pp 83–102, Plenum Press, New
York.
Rowland M and Tozer TN (1995) Clinical Pharmacokinetics: Concept and Applications, 3rd ed.,
The Williams & Wilkins Company, Baltimore.
Sahin S and Rowland M (1998a) Estimation of specific hepatic arterial water space. Am J Physiol
275:G228 –G236.
Sahin S and Rowland M (1998b) Development of an optimal method for the dual perfusion of
the isolated rat liver. J Pharmacol Toxicol Meth 39:35– 43.
Shetty BV, Badr M and Melethil S (1994) Evaluation of hepatic metabolism of salicylic acid in
perfused rat liver. J Pharm Sci 83:607– 608.
Watanabe Y, Puschel GP, Gardemann A and Jungermann K (1994) Presinusoidal and proximal
intrasinusoidal confluence of hepatic artery and portal vein in rat liver: Functional evidence by
orthograde and retrograde bivascular perfusion. Hepatology 19:1198 –1207.
Wu ZY, Rivory LP and Roberts MS (1993) Physiological pharmacokinetics of solutes in the
isolated perfused rat hindlimb: Characterization of the physiology with changing perfusate
flow, protein content, and temperature using statistical moment analysis. J Pharmacokinet
Biopharm 21:653– 688.
Yano Y, Yamaoka K, Aoyama Y and Tanaka H (1989) Two-compartment dispersion model for
analysis of organ perfusion system of drugs by fast inverse laplace transform (FILT).
J Pharmacokinet Biopharm 17:179 –202.
Yoshikawa T, Sugiyama Y, Sawada Y, Iga T and Hanano M (1984) Effect of pregnancy on tissue
distribution of salicylate in rats. Drug Metab Dispos 12:500 –505.
Downloaded from dmd.aspetjournals.org at ASPET Journals on May 5, 2017