Drug Metabolism, Development Research Laboratories, Banyu
Pharmaceutical Co., Ltd., Saitama, Japan (N.T., R.I., T.K.);
Tsukuba
Research Institute, Banyu Pharmaceutical Co., Ltd., Ibaragi, Japan
(S.N., T.H.); and
Pharmaceutical Information Management, Banyu
Pharmaceutical Co., Ltd., Tokyo, Japan (H.I.)
 |
Introduction |
(6-N-formylamino-12,13-dihydro-1,11-dihydroxy-13-(
-D-glucopyranosil)5H-indolo
[2,3-a] pyrrolo
[3,4-c]carbazole-5,7(6H)-dione (NB-506)1
(Fig. 1) is derived from a novel
indolocarbazole antibiotic (BE-13793C) isolated from
Actinomyces. NB-506 demonstrated strong antitumor activity
in vitro and in experimental animal tumor systems and is currently
under development as an anticancer agent (Arakawa et al., 1995
). Recent
studies showed that NB-506 acts as a potent inhibitor of topoisomerase
I by inducing the formation of stable topoisomerase I-DNA cleavage
complexes. The compound also inhibits the activity of DNA polymerase
and RNA polymerase II (Yoshinari et al., 1995).
In phase I clinical studies, the dose of a drug is usually determined
by monitoring adverse effects and pharmacokinetics. Recently, the use
of pharmacokinetically guided dose escalation has been proposed in
phase I studies of anticancer agents to reduce the number of
dose-escalation steps before the maximum tolerated dose by monitoring
area under the curve (AUC) at each step (European Organization
for Research and Treatment of Cancer Pharmacokinetics and Metabolism
Group, 1987; Collins et al., 1990; Graham and Workman, 1992).
Therefore, it is important to evaluate the pharmacokinetics and dose
proportionality of drugs in the animals used in pharmacology and
toxicity studies.
In this study, the pharmacokinetics of NB-506 after i.v. administration
was characterized in rats and dogs. In rats, nonlinearity of the
pharmacokinetics of NB-506 was observed. To better understand the
mechanisms of nonlinear pharmacokinetics in rats, in vivo metabolites
were determined. The possible mechanism of the nonlinearity of the
pharmacokinetics of NB-506 in rats is discussed in this communication.
| |
Materials and Methods |
Materials
NB-506 and ED-501 were synthesized at Banyu Tsukuba
Research Laboratories (Tsukuba, Japan).
[14C]NB-506 was synthesized at Dai-ichi Pure
Chemical Co. Ltd. (Tokyo, Japan). Acetonitrile, methanol,
trifluoroacetic acid (TFA),
N,N,-dimethylformamide, ethyl acetate, and
dichloromethane [all high-performance liquid chromatography (HPLC)
grade] were purchased from Wako (Osaka, Japan). Propylene
glycol (PG) and polyethylene glycol 400 (PEG400, reagent grade) were
purchased from Wako. Water was purified with a Milli-Q system
(Millipore-Japan, Tokyo, Japan).
| |
Animals |
Rats (male, Sprague-Dawley, 7 weeks old) were purchased from
Charles River Breeding Laboratories, Inc. (Kanagawa, Japan). Dogs
(male, beagle, 7-9 months old) were purchased from Marshall Research
Animals, Inc., (North Rose, NY). All animals were housed under a 12-h
light/dark cycle with free access to food and water.
| |
Dosing and Sample Collection |
Pharmacokinetic Study. Rats.
According to the clinical formulation, NB-506 was dissolved in
combination with PG and PEG400. The drug solution was diluted with 2 volumes of distilled water; the diluted drug solution was administered
to rats, 2 ml/kg, via a metatarsal vein. Five groups of four rats each
received a single i.v. dose of 37.5, 120, 187.5, 250, or 375 mg/m2 NB-506. Blood samples were collected via
the tail vein into heparinized capillary tubes before dosing and then
at 5, 10, 20, and 30 min and at 1, 2, 3, 4, 6, 8, and 24 h after
drug administration. Plasma was separated immediately by centrifugation
(2 min, 4°C, 9000g) and deproteinized immediately by
mixing with 2 volumes of
N,N,-dimethylformamide/methanol (1:1) containing
an internal standard. The samples were stored at 
80°C until analysis.
Dogs.
NB-506 was dissolved in combination with PG and PEG400. According to
the dosing regimen for phase I clinical trials, the drug solutions were
diluted with 50 volumes of 5% glucose and the diluted drug solution
was administered to dogs, 10 ml/kg, via a cephalic vein in 60 min using
a portable infusion pump (CP-136; Nipro, Osaka, Japan). Three dogs were
given an i.v. infusion of NB-506 at the doses of 12, 34, and 110 mg/m2. Blood was drawn from the cephalic
vein 0.5 h before the end of infusion and at 0, 0.5, 1, 2, 4, and
6 h after the end of infusion. Plasma was separated immediately by
centrifugation (10 min, 4°C, 1800g) and stored at 80°C
until analysis.
Metabolite Isolation and Identification.
A cannula was implanted surgically into the bile duct while
the rats were under ether anesthesia. NB-506 was dissolved in combination with PG and PEG400. The drug solution was diluted with 2 volumes of distilled water, and the diluted drug solution was
administered to rats, 2 ml/kg, via a tail vein. To identify metabolites, [14C]NB-506 was administered to
rats i.v. at a dose of 187.5 mg/m2. All of the
animals received 1.85 MBq/kg of the radiolabeled compound. After
drug administration, the animals were placed in restraining cages. The
biliary output was collected at 0 to 1, 1 to 2, 2 to 4, and 4 to 6 h after dosing. For the metabolite purification, NB-506 was
administered to rats i.v. into the tail vein at a dose of 187.5 mg/m2. The biliary output was collected at 0 to
2 h after dosing.
| |
Drug Analysis |
Plasma levels of NB-506 and ED-501 after NB-506
administration were determined by HPLC as follows.
Rats.
The deproteinized plasma samples were centrifuged (5 min, 4°C,
9000g). The supernatant was filtered through a 0.5-µm
filter and 50 µl of the filtrate was injected into an HPLC system.
Dogs.
A 0.5-ml plasma sample was diluted with an equal volume of 0.1 M
glycine-HCl buffer, pH 3, containing internal standard. The diluted
sample was adsorbed onto a Bond Elut C8 column
(Varian Sample Preparation Products, Harbor City, CA), washed
with 10% methanol, and eluted with methanol. The eluate was evaporated to dryness and the residue was dissolved in 50% methanol. A 50-µl sample was injected into an HPLC system.
| |
HPLC Condition |
For the rat assays, an HPLC system (Jasco, Tokyo, Japan)
consisting of an 801-SC system controller, an 880 to 50 degasser, an
880 to 02 gradient unit, an 880-PU pump, and an 860-CO column oven was
used. The UV detector (Waters 490; Japan Millipore Limited, Tokyo,
Japan) was set at 305 nm. Integration was carried out using a
Millennium 2010J program (Japan Millipore Limited) installed in a Power
Mate 486/66i (NEC, Tokyo, Japan). A Superiorex ODS column (Shiseido,
Tokyo, Japan; 250 mm × 4.6 mm, 5 µm) was used for analysis and
a LiChrospher 100 RP-18 (5 µm; E. Merck, Darmstadt, Germany) was used
as a guard column. The mobile phase consisted of
acetonitrile/methanol/water/TFA (20:15:65:0.1, v/v/v/v) and was
delivered at a flow rate of 1 ml/min at 40°C.
For the dog assays, a Shimadzu (Tokyo, Japan) LC-10A system consisting
of a LC-10AD pump, a SPD-10A detector, an SIL-10A auto sampler, and a
CTO-10A column oven was used. A Superiorex ODS column (Shiseido; 250 mm × 4.6 mm, 5 µm) was used for analysis and an RP-18 Newguard
(Applied Biosystems, Inc., Foster City, CA; 15 mm × 3.2 mm, 7 µm) was used as a guard column. A mobile phase consisting of
acetonitrile/methanol/0.1 M sodium acetate buffer, pH 4.0 (19:15:66,
v/v/v), was delivered at a flow rate of 1 ml/min at 40°C. The eluent
was monitored at 305 nm.
| |
Identification of Metabolites in Bile |
Analysis of Radioactivity in Bile.
The bile samples were centrifuged (9000g, 4°C, 5 min) and
an aliquot of the supernatant from each sample was injected into the HPLC system. A Beckman RI-HPLC system (Beckman-Japan, Tokyo, Japan)
consisting of a module 128 pump, a 507 autosampler with a column oven
and sample cooler, a 171 radioisotope detector, a 110B solvent delivery
module, and a 406 analog interface module was used. Integration was
carried out using a System Gold (Beckman-Japan) program installed in
Prolinea 4/66 (COMPAQ-Japan, Tokyo, Japan). A Superiorex ODS column
(Shiseido; 4.6 mm × 250 mm, 5 µm) was used for the analysis and
a LiChrospher 100 RP-18 (5 µm; E.Merck) was used as a guard column.
Gradient elution was conducted with a mobile phase consisting of 1)
water/acetonitrile/TFA (81:19:0.1, v/v/v) and 2) water/acetonitrile/TFA
(40:60:0.1, v/v/v) delivered at a flow rate of 1 ml/min at 40°C. A
linear gradient was conducted as follows: 0 to 25 min, 0% B; 25 to 37 min, 0 to 25% B; 37 to 50 min, 25% B. The eluate was mixed with Ready
Flow III (Beckman-Japan) delivered at a flow rate of 1 ml/min. The
mixture was loaded continuously into the liquid scintillation flow cell
and the radioactivity was monitored.
Isolation of Metabolites.
Metabolites were extracted from bile in the following manner: The water
layers were collected after bile samples were shaken with ethyl
acetate. TFA was added to the water layer to make a final concentration
of 0.1% TFA. The acidified water layers were applied to a Bond Elut CN
(Varian) solid-phase extraction column sample cartridge that had
been prewashed with dichloromethane, methanol, and 0.1% aqueous TFA.
The cartridges were then washed with 0.1% aqueous TFA and the
metabolites of interest were eluted with 50% methanol. The eluates
were evaporated to dryness and dissolved in a small volume of 50%
methanol. Separation was achieved by HPLC on a SUPELCOSIL ABZ+Plus
column (10 × 250 mm, 5 µm; Supelco, Inc., Bellefonte, PA) with
a mobile phase consisting of 22% acetonitrile (containing 0.1% TFA).
The flow rate was 3 ml/min. Each metabolite peak was collected and
applied to a Sephadex LH-20 column that was preconditioned with water.
The column was washed with water and the metabolites were eluted with
methanol. The column eluates were evaporated to dryness for
spectroscopic characterization.
Mass Spectroscopy.
Positive and negative ion mode FAB-Mass spectra were obtained using a
JMS-SX102A (JEOL, Tokyo, Japan) mass spectrometer in m-nitrobenzyl alcohol as a matrix.
NMR Spectroscopy.
1H NMR and 13C NMR spectra
were obtained using a JNM-A500 (JEOL) NMR spectrometer. Dimethyl-d6
sulfoxide (DMSO-d6) was used as the solvent. Two-dimensional NMR
spectra were also measured.
| |
Pharmacokinetic Data Analysis |
All pharmacokinetic parameters were calculated using the
WinNonlin software (version 1.5, Scientific Consulting, Inc., NC). The
AUC0-2 h and AUC0-3 h
were calculated by the trapezoidal rule without extrapolation to
infinity. The AUC0-
was calculated with the
area from the last data point to time infinity that was estimated by
dividing the last measured plasma concentration by the terminal rate
constant. Standard moment method (Gibaldi and Perrier, 1982) was used
to calculate the following pharmacokinetic parameters: the plasma
clearance (CLp), the area under the moment curve (AUMC),
the mean residence time (MRT) and the distribution volume
(Vdss),
where Cp is the plasma concentration at time t.
Half-life was estimated from the regression line of the terminal phase
fitted to the log plasma concentration-time points by the least-squares method. Plasma volume was calculated from the allometric equation (Mordenti, 1986); Vp = 0.0429 W
0.992, where Vp
(ml) is plasma volume and W is body weight (g).
In vivo excretion clearances were calculated with the following
equation:
where CLr is the renal clearance,
Xu is the amount excreted into urine as unchanged
drug, CLh,b is the biliary excretion clearance,
and Xb is the amount excreted into bile as
unchanged drug.
| |
Scale Up of In Vitro
Vmax/Km |
To scale up the
Vmax/Km
(µl/min/mg protein) to the in vivo intrinsic clearance (ml/min/kg)
requires the knowledge of microsomal protein yield per g liver (mg/g
liver) as well as the liver weight (g/kg body weight). The scaling
factors for the calculation of the intrinsic clearance are as follows:
The microsomal protein yield is 50 mg/g. The liver weight used in
calculation was 45 g/kg for rats (Lin et al., 1996).
| |
Results |
Pharmacokinetics of NB-506 in Rats.
In rats, the plasma concentration of intact drug after single i.v.
administration of NB-506 at doses of 37.5, 120, 187.5, 250, and 375 mg/m2 decreased biexponentially with
terminal half-lives of approximately 2 h (Fig.
2). The pharmacokinetic parameters of
NB-506 are shown in Table 1. The
AUC0-2 h accounted for 90% of
AUC0-. The AUC0-
could not be determined at 37.5 mg/m2 because the
terminal phase was under the limit of quantification, therefore, the
relationship between dose and AUC was evaluated from the value of
AUC0-2 h. The AUC of NB-506 increased more than
proportionally with increasing dose (Fig.
3). The plasma clearance decreased
with increasing dose up to 187.5 mg/m2, but
remained virtually unchanged at the highest dose. The
Vdss (1-3 liters/kg) was much greater
than the plasma volume (40 ml/kg; Mordenti, 1986), indicating that
NB-506 highly distributed to tissues from plasma.
Vdss also decreased with increasing
dose up to 187.5 mg/m2. MRT values were almost
identical within this dose range. The concentration of
ED-501 peaked at initial time point (5 min) after
administration at the doses of 37.5 and 120 mg/m2. The Tmax
of ED-501 increased to 30 min with the increase of the dose
of NB-506. The Cmax of
ED-501 increased less than the dose of NB-506 increased. The
AUC ratio of ED-501 to NB-506 at the doses of 37.5, 120, 187.5, 250, and 375 mg/m2 was 31%, 28%, 22%,
24%, and 19%, respectively.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 2.
Plasma concentrations of NB-506 and
ED-501 after administration of NB-506 to rats at
doses of 37.5, 120, 187.5, 250, and 375 mg/m2.
Each value represents the mean ± S.D. (N = 4).
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 1
Pharmacokinetic parameters of NB-506 and its metabolites after i.v.
administration of NB-506 to rats (N = 4)
|
|
Pharmacokinetics of NB-506 in Dogs.
In dogs, the plasma concentrations of NB-506 and ED-501
after i.v. infusion of NB-506 at the doses of 12, 34, and 110 mg/m2 are shown in Fig.
4. NB-506 declined biexponentially with
terminal half-lives of approximately 2 h. The pharmacokinetic
parameters of NB-506 after administration in dogs are shown in Table
2. The AUC0-3 h
accounted for 90% of AUC0-. The AUC0- could not be determined at 12 mg/m2 because the terminal phase was under the
quantification limit, therefore, relationship between dose and
AUC was evaluated from the value of AUC0-3 h.
The relationship between AUC and dose was linear (Fig. 3). The
Vdss (1.6 liters/kg) was much greater
than the plasma volume (40 ml/kg; Mordenti, 1986), indicating that
NB-506 highly distributed to tissues from plasma. In contrast to rats,
ED-501 was only detected at the dose of 110 mg/m2; plasma concentration of ED-501
was very low.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 4.
Plasma concentrations of NB-506 and
ED-501 after i.v. infusion of NB-506 to dogs
at doses of 12, 34, and 110 mg/m2.
Each value represents the mean ± S.D. (N = 3).
|
|
Identification of Metabolites.
To better understand the mechanisms of nonlinear pharmacokinetics in
rats, in vivo metabolites were determined. A typical HPLC with on-line
radioisotope detector (RI-HPLC) fast atom bombardment mass (FAB-mass)
spectra chromatogram of bile sample after
[14C]NB-506 administration to rats is shown in
Fig. 5. NB-506 and ED-501,
deformyl metabolite of NB-506, were identified using the authentic
standards. Two unexpected metabolite peaks (RBM-1 and
RBM-2) were observed. These metabolites were isolated from
rat bile and identified as follows: The FAB-mass spectra of
RBM-1 showed several characteristic ion fragments at
m/z 738 [M]+ and 562 [M-176]+. From the molecular ion peak at
m/z 738.1652 [M]+ obtained using
high-resolution FAB-mass spectra, the molecular formula of
RBM-1 was estimated to be
C33H30O16N4.
Linked scan FAB-mass spectroscopic analysis suggested that
RBM-1 has two sugar moieties (Fig.
6). Table 3
lists the chemical shifts and their assignments of the NMR spectra of
RBM-1 in DMSO-d6, and also presents the results of
heteronuclear multiple bond correlation (HMBC) analysis. The phenolic
proton signal that was apparent in the spectra of the intact drug
disappeared in the spectra of RBM-1. Signals indicating a
sugar moiety were observed at 3.4 to 6.0 ppm in the spectra of
RBM-1. To determine the conjugating position of glucuronic
acid, HMBC-NMR spectra was measured. A correlation signal between the
anomeric proton (
H 5.40) and the carbon on the
position 11 (C 143.1) was observed. There were
33 signals in 13C NMR spectra, consistent with
the estimated molecular formula. These data indicate that
RBM-1 is the 11-O-glucuronide of NB-506, which is
also a synthetic standard named ED-594.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 5.
Typical RI-HPLC chromatogram of rat bile
collected 1 to 2 h after administration of
[14C]NB-506 at the dose of 187.5 mg/m2.
|
|
The FAB-mass spectra of RBM-2 showed the molecular weight of
RBM-2 was calculated to be 710 from the pseudomolecular ion
fragment that was observed at m/z 709 [M-H]
. Table 4
lists the chemical shifts and their assignments of the NMR spectra of
RBM-2 in DMSO-d6, and also presents the result of HMBC
analysis. There were no differences from the results of
RBM-1, except for the disappearance of a formyl moiety
signal and the appearance of two proton signals at 4.96 ppm. These data
indicate that RBM-2 is the 11-O-glucuronide of
ED-501, which is also a synthetic standard named
ED-595.
| |
Discussion |
The pharmacokinetics of NB-506, a novel topoisomerase I inhibitor,
was studied in rats and dogs after i.v. administration of the drug. In
rats, the AUC increased more than the increasing dose level
(37.5-187.5 mg/m2), indicating nonlinear
pharmacokinetics. Nonlinearity of the pharmacokinetics of compounds,
which has previously been reported, is mainly due to saturation in
absorption, plasma protein binding, metabolism in various tissues, and
biliary and urinary excretion processes (Ludden, 1991). In the case of
NB-506, ED-501, a metabolite of NB-506, was found in rat
plasma, but was negligible in dog and human plasma (Takenaga et al.,
1995). This suggests that NB-506 is converted to ED-501 by a
rat-specific NB-506-metabolizing enzyme. As described in a subsequent
report, NB-506 was deformylated to ED-501 in mouse and rat
plasma, but the deformylation activity of NB-506 in dog and human
plasma was negligible. In liver S9 and small intestine S9 samples from
mice and rat, the deformylation activity of NB-506 was very low. Also,
there was no activity in the liver or small intestine of dogs and
humans. The enzymatic conversion of NB-506 to ED-501 was
mouse and rat specific serine enzyme in plasma, because it was
inhibited by typical serine enzyme inhibitor but not inhibited by
SH-reagent or EDTA, with a molecular mass of 138 KDa.
To further understand the mechanisms of nonlinear pharmacokinetics, in
vivo metabolites were determined after
[14C]NB-506 administration to rats. Two unknown
metabolites (RBM-1 and RBM-2) were isolated from
rat bile and identified. From the spectrometric analysis, it is
concluded that RBM-1 is the 11-O-glucuronide of
NB-506 (ED-594), and RBM-2 is the
11-O-glucuronide of ED-501 (ED-595).
To compare the excretion and metabolic clearances, the intrinsic
clearances were calculated (Table 5). The
excretion clearances were calculated from the value of amount of drug
excreted into bile and urine taken from the reference (Ishii et al.,
1998). The metabolic clearances were calculated from the value of
Vmax/Km
(described in a subsequent report) and scaled up to in vivo
clearances. The total clearance was almost equal to the sum of
CLr (renal clearance),
CLh,b (biliary excretion clearance),
CLint, m, p (metabolic clearance in plasma), and
CLint, m, h (glucuronidation clearance in liver).
As described in a subsequent report, hepatic mixed function oxidases
involving cytochrome P-450 have no relation to NB-506 metabolism.
NB-506 was glucuronized in vitro using liver microsomes in mice, rats, and humans, but not in dogs. These observations indicated that the
elimination of NB-506 from plasma could be explained by conversion to
deformyl form and glucuronide, followed by excretion to urine and bile.
The nonlinear pharmacokinetics of NB-506 in rats is probably due to
saturation of the enzyme systems that catalyze the reaction of
deformylation and glucuronidation, because the initial plasma
concentration of NB-506 in rats after NB-506 administration at the
doses of 187.5 to 375 mg/m2 were higher than the
apparent Km value for
ED-501 formation (54 µM, 30 µg/ml) and the apparent
Km value for glucuronidation (5.8 µM, 3 µg/ml). This speculation is supported by the
Cmax of ED-501 in rats,
which increased less than the dose of NB-506 was increased, and the AUC
ratio of ED-501 to NB-506, which decreased from 31% to 19%
with the increased dose. Some carrier-mediated transport
systems, bile acid, P-glycoprotein, and canalicular multispecific
organic anion transport exist to excrete the drugs from liver to bile
(Sathirakul et al., 1994; Shimamura et al., 1994; Yamazaki et al.,
1997). The possibility that NB-506 is actively excreted into
bile by these system still remains. It would be useful to
identify the excretion mechanisms of NB-506 for determining the
possibility of nonlinear pharmacokinetics.
In contrast to rats, there was a linear relationship between the AUC
and the dose in dogs. As described in a subsequent report, the
enzymatic activities involving glucuronidation and deformylation of
NB-506 were not observed in dogs. The reason for linear
pharmacokinetics in dogs, therefore, might be due to lack of the enzyme
systems catalyzing NB-506 metabolism (deformylation and
glucuronidation) in dogs. The apparent
Km value for glucuronidation in human
liver (75 µM, 40 µg/ml) exceeded the
Cmax of NB-506 (3 µg/ml) after i.v.
administration of NB-506 at the dose of 120 mg/m2
by 1-h infusion (Takenaga et al., 1995). From these results, nonlinear
pharmacokinetics was expected for NB-506 in humans when the dose was
escalated to the maximum tolerated dose. Recently, a phase I
single-dose clinical study of NB-506 revealed a linear relationship
between dose level (37.5-180 mg/m2) and AUC
(Sasaki et al., 1995). The reason for linear pharmacokinetics of NB-506
in humans is possibly because the enzyme system catalyzing the
glucuronidation of NB-506 was not saturated at these dose levels.
In conclusion, NB-506 had a nonlinear pharmacokinetic profile in rats
and a linear profile in dogs. The metabolism of NB-506 to
ED-501 is species-dependent; it occurs only in rats. After
NB-506 administration, two unknown metabolites were isolated from rat
bile and identified as 11-O-glucuronide of NB-506 and 11-O-glucuronide of ED-501. The nonlinear
pharmacokinetics of NB-506 is probably due to saturation of these
enzyme systems that catalyze the deformylation and glucuronidation of
NB-506 in rats.
We thank Dr. Toshio Yasumori (Banyu Pharmaceutical Co., Ltd.) for
valuable comments during preparation of this manuscript.
![[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}
-D-glucopyranosil)5H-indolo
[2,3-a]pyrrolo
[3,4-c]carbazole-5,7(6H)-dione (NB-506) in
Rats and Dogs: Pharmacokinetics, Isolation, Identification, and
Quantification of Metabolites
Top
Abstract
Introduction
