Biotransformation of a novel antimitotic agent, I-387, by mouse, rat, dog, monkey, and human liver microsomes and in vivo pharmacokinetics in mice

Article abstract of DOI:10.1124/dmd.110.036673

3-(1H-Indol-2-yl)phenyl)(3,4,5-trimethoxyphenyl)methanone (I-387) is a novel indole compound with antitubulin action and potent antitumor activity in various preclinical models. I-387 avoids drug resistance mediated by P-glycoprotein and showed less neurotoxicity than vinca alkaloids during in vivo studies. We examined the pharmacokinetics and metabolism of I-387 in mice as a component of our preclinical development of this compound and continued interest in structure-activity relationships for antitubulin agents. After a 1 mg/kg intravenous dose, noncompartmental pharmacokinetic analysis in plasma showed that clearance (CL), volume of distribution at steady state (Vd_ss), and terminal half-life (t_1/2) of I-387 were 27 ml per min/kg, 5.3 l/kg, and 7 h, respectively. In the in vitro metabolic stability study, half-lives of I-387 were between 10 and 54 min by mouse, rat, dog, monkey, and human liver microsomes in the presence of NADPH, demonstrating interspecies variability. I-387 was most stable in rat liver microsomes and degraded quickly in monkey liver microsomes. Liquid chromatography-tandem mass spectrometry was used to identify phase I metabolites. Hydroxylation, reduction of a ketone group, and O-demethylation were the major metabolites formed by the liver microsomes of the five species. The carbonyl group of I-387 was reduced and identified as the most labile site in human liver microsomes. The results of these drug metabolism and pharmacokinetic studies provide the foundation for future structural modification of this pharmacophore to improve stability of drugs with potent anticancer effects in cancer patients. Copyright

Full text of DOI:10.1124/dmd.110.036673

Supplemental material to this article can be found at:
http://dmd.aspetjournals.org/content/suppl/2011/01/13/dmd.110.036673.DC1.html
0090-9556/11/3904-636–643$20.00
DRUG METABOLISM AND DISPOSITION
Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics
DMD 39:636–643, 2011
Vol. 39, No. 4
36673/3676457
Printed in U.S.A.
Biotransformation of a Novel Antimitotic Agent, I-387, by Mouse,
Rat, Dog, Monkey, and Human Liver Microsomes and In Vivo
□
S
Pharmacokinetics in Mice
Sunjoo Ahn, Jeffrey D. Kearbey, Chien-Ming Li, Charles B. Duke III, Duane D. Miller,
and James T. Dalton
Division of Pharmaceutics, College of Pharmacy, the Ohio State University, Columbus, Ohio (S.A., C.-M.L., J.T.D.); Department
of Pharmaceutical Sciences, College of Pharmacy, University of Tennessee Health Science Center, Memphis, Tennessee
(C.B.D., D.D.M.); and Preclinical Research and Development, GTx, Inc., Memphis, Tennessee (S.A., J.D.K., C.M.L.,
D.D.M., J.T.D.)
Received October 8, 2010; accepted January 12, 2011
ABSTRACT:
3-(1H-Indol-2-yl)phenyl)(3,4,5-trimethoxyphenyl)methanone (I-387) 10 and 54 min by mouse, rat, dog, monkey, and human liver mi-
is a novel indole compound with antitubulin action and potent crosomes in the presence of NADPH, demonstrating interspecies
antitumor activity in various preclinical models. I-387 avoids drug variability. I-387 was most stable in rat liver microsomes and de-
resistance mediated by P-glycoprotein and showed less neurotox- graded quickly in monkey liver microsomes. Liquid chromatogra-
icity than vinca alkaloids during in vivo studies. We examined the phy-tandem mass spectrometry was used to identify phase I me-
pharmacokinetics and metabolism of I-387 in mice as a component tabolites. Hydroxylation, reduction of
a ketone group, and
of our preclinical development of this compound and continued O-demethylation were the major metabolites formed by the liver
interest in structure-activity relationships for antitubulin agents. microsomes of the five species. The carbonyl group of I-387 was
After a 1 mg/kg intravenous dose, noncompartmental pharmaco- reduced and identified as the most labile site in human liver mi-
kinetic analysis in plasma showed that clearance (CL), volume of crosomes. The results of these drug metabolism and pharmacoki-
distribution at steady state (Vdss), and terminal half-life (t1/2) of netic studies provide the foundation for future structural modifica-
I-387 were 27 ml per min/kg, 5.3 l/kg, and 7 h, respectively. In the
tion of this pharmacophore to improve stability of drugs with
in vitro metabolic stability study, half-lives of I-387 were between potent anticancer effects in cancer patients.
Introduction
Because it is necessary to achieve and maintain effective con-
centrations of drugs at their sites of actions, pharmacokinetic
studies to characterize absorption, distribution, metabolism, and
elimination are important in optimization of therapeutic dose and
schedule. Although in vivo studies provide definitive data regard-
ing the disposition of a drug in an animal, in vitro studies of drug
metabolism provide the opportunity for efficient and rapid char-
acterization of the metabolic fate of a compound in numerous
species without the need for synthesis of large quantities of drugs
or use of animals. Drug biotransformation is divided into two types
3-(1H-Indol-2-yl)phenyl)(3,4,5-trimethoxyphenyl)methanone (I-387),
is a new antitubulin agent with potent in vivo and in vitro antitumor
activity in various preclinical cancer models (Fig. 1A) (Ahn et al., 2010).
I-387 destabilizes microtubules by binding to the colchicine-binding
domain. In addition, I-387 is not a substrate for ATP-binding cassette
transporters and maintained potent tumor growth inhibition in xenograft
models using cell lines overexpressing P-glycoprotein. This investiga-
tional agent is particularly attractive for further development because no
significant neurotoxicity was observed during in vivo and in vitro neu-
rotoxicity studies (Ahn et al., 2010). To this end, we sought to understand of reactions: phase I (hydrolysis, oxidation, and reduction) and
the absorption, distribution, metabolism, and excretion of I-387 as a
means to evaluate its toxicity and potency and to optimize the I-387
pharmacophore for better exposure and stability in vivo.
phase II (conjugation). The cytochrome P450 enzyme superfamily
plays a dominant role in phase I metabolism, and phase II conju-
gation is caused by UDP-glucuronosyltransferase, N-acetyl transferase,
glutathione transferase, sulfotransferase, and so on (Schlichting et al.,
2000; Sheweita, 2000). Although the isolated perfused liver could
give an excellent representation of the in vivo situation, current
guidelines for human drug development allow the use of in vitro
systems for metabolism studies, including microsomes, Super-
somes, cytosol, the S9 fraction, cell lines, transgenic cell lines, and
primary hepatocytes (FDA Drug Metabolism/Drug Interaction
Sunjoo Ahn, Jeffrey D. Kearbey, Chien-Ming Li, Duane D. Miller, and James T.
Dalton are employees of GTx, Inc. and hold stock options in the company.
Article, publication date, and citation information can be found at
http://dmd.aspetjournals.org.
doi:10.1124/dmd.110.036673.
□S The online version of this article (available at http://dmd.aspetjournals.org)
contains supplemental material.
ABBREVIATIONS: I-387, 3-(1H-Indol-2-yl)phenyl)(3,4,5-trimethoxyphenyl)methanone; UDPGA, UDP-glucuronic acid; LC, liquid chromatography;
MS/MS, mass spectrometry; amu, atomic mass units; EIC, extracted ion chromatogram.
636

METABOLISM AND PHARMACOKINETICS OF I-387
637
FIG. 1. Synthesis scheme. A, synthesis of I-387. Reagents and conditions:
a, 3,4,5-trimethoxyphenyllithium, tetrahydrofuran, Ϫ78°C; b, pyridinium
dichromate, CH2Cl2; c, aqueous NaOH, EtOH, reflux. For better under-
standing putative structures of I-387 metabolites, we will refer to sections
denoted in the figure as the A, B, and C rings. B, the structure of indole-15
(internal standard).
Pharsight Corporation, Mountain View, CA). The area under the plasma
concentration-time curve from time 0 to infinity (AUCinf) was calculated by
the trapezoidal rule with extrapolation to time infinity. The terminal t1/2 was
calculated as ln2/␭z, where ␭z was the first-order rate constant associated with
the terminal (log-linear) portion of the curve. The plasma CL was calculated as
dose/AUCinf. The maximum plasma concentration (Cmax) and the time when it
occurred (Tmax) were obtained by visual inspection of the plasma concentra-
tion-time curve. The apparent Vdss was calculated by CL ϫ MRTinf, where
Studies in the Drug Development Process: Studies In Vitro, http://
www.fda.gov) (Brandon et al., 2003).
In this study, we examined the pharmacokinetics of I-387 in mice
after intravenous, intraperitoneal, and oral dosing in mice. The in vitro
metabolic fate of I-387 was investigated in mouse, rat, dog, monkey,
and human liver microsomes. Our studies provide the first evidence
that intravenous and intraperitoneal administration of I-387 in mice
can achieve systemic drug exposure sufficient for an in vivo antitumor
effect (Ahn et al., 2010). In addition, we identified a metabolically
labile site in the I-387 pharmacophore that can be modified to design
more stable drug candidates.
MRT is the mean residence time extrapolated to infinity calculated as AUMCinf
/
AUCinf, where AUMCinf is the area under the first moment curve extrapolated to
infinity. The bioavailability (F) was calculated as (AUCnon-i.v. ϫ dosei.v.)/(AUCi.v. ϫ
dosenon-i.v.).
In Vitro Metabolism Studies. For both phase I and phase I and II metab-
olism studies, animal microsomes was used as described previously (Li et al.,
2010). In brief, the incubation mixture consisted of 1 mg/ml liver microsomal
proteins, 3 mM NADPH, and 0.5 ␮M test compound in 65 mM potassium
phosphate buffer (pH 7.4). The concentration of 0.5 ␮M was selected for these
studies on the basis of preliminary experiments showing that the Km value for
metabolism was Ͼ10 ␮M. For the phase I and II metabolism, 5 mM UDP-
glucuronic acid (UDPGA), 5 mM D-saccharolactone, 50 ␮g/ml alamethicin,
and 3 mM magnesium chloride were added to the phase I metabolism study
mixture. The concentration of methanol (used for dissolving the substrate) was
1% (v/v). The total volume of the incubation was 200 ␮l, and the reaction
mixtures were incubated at 37°C. For metabolite identification, the reaction
mixture was incubated for 2 h. To generate stability curves for I-387 different
incubations were stopped at 10, 20, 30, 60, and 90 min for analysis of I-387
remaining. All reactions were stopped by the addition of 200 ␮l of ice-cold
acetonitrile containing an internal standard for quantification. The samples
were then centrifuged at 8000g for 5 min, and supernatant was analyzed by
LC-MS/MS. The mean peak area ratio of I-387 to internal standard observed
at time 0 was 100%. Replicate studies were conducted on at least two separate
occasions for each species.
LC-MS/MS Analyses. The LC-MS/MS analysis was performed on a 4000
Q TRAP triple quadrupole/linear ion trap mass spectrometer (Applied Biosci-
ences, Foster City, CA) with a nanospray interface. The nanospray temperature
was set as 500°C, curtain gas at 30 psi, ion spray energy at 5500 V, nebulizer
gas (gas 1) at 30, and nano gas (gas 2) at 40 psi. Nitrogen was used as the
collision gas. The collision energy was 45 eV. The protonated molecules were
investigated. Aliquots of samples were injected into the high-performance
liquid chromatography system (model 1100 series ChemStation; Agilent Tech-
nologies, Santa Clara, CA). The gradient mode was used to achieve separation
of the analytes using mixtures of mobile phase A (95% water and 5%
acetonitrile with 0.1% formic acid) and mobile phase B (95% acetonitrile and
5% water with 0.1% formic acid) at a flow rate of 300 ␮l/min.
Materials and Methods
Chemicals, Microsomes, and Animals. I-387and a putative metabolite of
I-387 were synthesized using methods similar to these described previously for
indole-15 (Ahn et al., 2011). In this study, indole-15 was used as an internal
standard. Intermediate 1 was treated with 3,4,5-trimethoxyphenyllithium to
afford alcohol 2 (Fig. 1A). Alcohol 2 was then oxidized with pyridinium
dichromate in dichloromethane at room temperature to form ketone 3, which
was deprotected with aqueous sodium hydroxide to afford I-387 (Fig. 1A). The
putative metabolite, alcohol 4, was also synthesized by deprotection of com-
pound 2 with reflux under basic conditions (Fig. 1A). Complete details of
synthesis are forthcoming in a separate publication.
Pooled liver microsomes from mice, rats, dogs, monkeys, and humans were
purchased from XenoTech, LLC (Lenexa, KS). The NADPH-regenerating
system was purchased from BD (Franklin Lakes, NJ). All other chemicals were
purchased from Sigma-Aldrich (St. Louis, MO). Male ICR mice (20–25 g)
were purchased from Harlan Biosciences (Indianapolis, IN). All animal pro-
tocols were approved by the animal care and use committee at The Ohio State
University or the University of Tennessee Health Science Center.
Pharmacokinetic Studies in Mice. Male ICR mice (5–6 weeks, 20–25 g)
were used. For I-387, three doses (1, 5, and 15 mg/kg) were administered via
the intravenous, intraperitoneal, and oral routes. Dosing vehicles were com-
posed of 10% dimethyl sulfoxide in PEG300. Intravenous doses were admin-
istered via the tail vein. Oral doses were administered by gavage. At each time
point, three to four mice were euthanized by isoflurane (Baxter Healthcare,
Deerfield, IL), and blood samples (up to 600 ␮l each) were taken from the
posterior vena cava. Plasma samples were stored at Ϫ20°C before analysis.
Sample Preparation for LC-MS/MS Analysis of Mouse Pharmacoki-
netics. Plasma proteins were precipitated by the addition of acetonitrile (150
␮l, containing the internal standard) to 100 ␮l of mouse plasma. Samples were
vortexed and then centrifuged at 8000g for 10 min. The supernatant was
transferred to a clean vial for injection into the mass spectrometer for analysis.
Pharmacokinetic Data Analysis. Mean plasma concentration-time data
were analyzed using noncompartmental methods (WinNonlin version 5.2.1;
Samples for pharmacokinetic studies were separated with a Halo 2.1 ϫ 50
mm C18 column (Advanced Materials Technology, Wilmington, DE) within a

638
AHN ET AL.
runtime of 8 min. Mobile phase A was used at 80% from 0 to 1 min followed
by a linearly programmed gradient to 80% of mobile phase B within 1.5 min,
and 80% of mobile phase B was maintained for 2 min before a quick ramp to
20% mobile phase B for 30 s. Then 20% of mobile phase B and 80% of mobile
phase A were continued for another 3 min toward the end of analysis. Multiple
reaction monitoring mode scanning of m/z 388.2 3 192 was used to obtain the
most sensitive signals for I-387.
Metabolite separation was performed on an Alltima HP 2.1 ϫ 100 mm C18
column (Alltech Associates, Deerfield, IL). Total run time was 15 min. The
mobile phase comprised 100% A for the first 1 min and increased to 100% B
in a linear gradient over 7 min. Then 100% of mobile phase B was maintained
for 2 min, and it quickly returned to 0% for 30 s; 100% mobile phase A was
continued for 4.5 more min until the end of analysis. Q1 full (data were not
shown), precursor ion, neutral loss, and enhanced product ion scan modes were
used to identify metabolites of I-387.
TABLE 2
Half life of I-387 in mouse, rat, dog, monkey, and human liver microsomes
n ϭ 2 to 3.
t1/2
Phase I
Phase I and II
min
Mouse
Rat
Dog
Monkey
Human
37 Ϯ 3.5
42 Ϯ 1.3
29 Ϯ 4.4
15 Ϯ 2.3
24 Ϯ 2.7
39 Ϯ 6.2
54 Ϯ 4.3
33 Ϯ 1.2
11 Ϯ 6.9
30 Ϯ 3.4
Fig. 3). A drug administered by intraperitoneal injection is subject
to the same potential hepatic first-pass metabolism as oral admin-
istration. However, the intraperitoneal bioavailability (65–95%) of
I-387 greatly exceeded that of orally administered drug (8–24%),
suggesting that the oral bioavailability of I-387 was primarily
limited by its poor stability in gastric or intestinal fluids and/or gut
wall metabolism or its incomplete absorption because of poor
permeability and/or low aqueous solubility. The approximate ter-
minal t1/2 (mean t1/2 ϭ 7 h) after intravenous administration was
similar to that observed after intraperitoneal and oral doses (mean t1/2 ϭ
7–8 h) (Table 1).
Metabolic Stability of 387 in Mouse, Rat, Dog, Monkey, and
Human Liver Microsomes. Metabolic stability was investigated by
incubating I-387 with liver microsomes from mouse, rat, dog,
monkey, and human. To evaluate the metabolism by phase II
conjugation, UDPGA was introduced to the microsomal prepara-
tion. Half-life was determined with a two-parameter single expo-
nential decay curves using SigmaPlot (version 10.0; Systat Soft-
ware, Inc., Chicago, IL). Metabolic stability of I-387 in the
presence and absence of UDPGA was not significantly different as
evidenced by similar half-lives during incubation with phase I and
phase I and II systems (Table 2). This result suggests that the
primary metabolic pathway of I-387 predominantly involves phase
I enzymes. I-387 was most stable in rat liver microsomes with the
longest t1/2 (41.9 min) and degraded quickly in monkey liver
microsomes with the shortest t1/2 (15 min) in phase I metabolism.
The phase I metabolic stability of I-387 in human liver microsomes
(23.6 min) appears to be more similar to those of mouse and dog
with half-lives of 36.7 and 29.4 min, respectively, than those of the
other two species.
Results
Pharmacokinetics of I-387 in Mice. Doses of 1, 5, and 15 mg/kg
were administered via the intravenous, intraperitoneal, and oral
route to elucidate the pharmacokinetic parameters of I-387 in ICR
mice. Plasma concentrations of I-387 in mice declined in a biex-
ponential manner after intravenous injection, with a terminal t1/2 of
4 to 9 h. The Vdss of I-387 was 3.8 to 9.8 l/kg after intravenous
administration indicating high tissue binding. The high tissue
binding of I-387 is probably a result of the hydrophobic nature of
this compound. The estimated logP and clogP values of I-387 were
4.15 and 4.77, respectively (ChemDraw Ultra version 11.0; Cam-
bridgeSoft Corporation, Cambridge, MA). CL slightly decreased
with increasing dose for the 1, 5, and 15 mg/kg with values of 27,
21, and 19 ml per min/kg, respectively (Table 1; Supplemental Fig.
1), which is 40 to 60% of hepatic plasma flow (Davies and Morris,
1993). MRTinf increased from 3 and 3.3 h at the 1 and 5 mg/kg
doses, respectively, to 8.8 h after the 15 mg/kg dose after intrave-
nous administration (Table 1). After intraperitoneal administration,
I-387 was absorbed quickly with Tmax values of 5 to 10 min.
Systemic exposure after intraperitoneal doses was 65 to 95% of the
analogous intravenous dose (Table 1; Supplemental Fig. 2). Plasma
concentrations of I-387 peaked at approximately 30 min after oral
administration with absolute bioavailability of 24, 13, and 8% after
doses of 1, 5, and 15 mg/kg, respectively (Table 1; Supplemental
TABLE 1
Pharmacokinetic parameters of I-387 in male ICR mice after intravenous,
intraperitoneal, and oral administration
n ϭ 3/dose group.
Identification of Metabolites in Human Liver Microsomal In-
cubation. To identify metabolites in human liver microsomes, I-387
was incubated with human liver microsomes in the presence of
NADPH for 2 h. Samples were analyzed by LC-MS/MS. Total ion
current chromatograms of the acetonitrile extracted I-387 in precursor
ion and neutral loss scan modes are shown in Fig. 2. The results are
described using the terms of A, B, and C rings of I-387 as shown in
Fig. 1. The product ion spectrum of I-387 (m/z 388.2) results in two
prominent fragment ions of m/z 192 and 220 (Fig. 3A). Thus, the
precursor ion scan of m/z 192 and the neutral loss scan of m/z 167
were used to identify metabolites of I-387 and its product fragments.
The precursor ion scan of m/z 192 was used to identify modifications
in the carbonyl group and/or C ring, whereas metabolites containing
metabolic modifications in rings A or B or the carbonyl could be
identified by the neutral loss of 167 m/z.
Route
Unit
1 mg/kg
5 mg/kg
15 mg/kg
Intravenous
t1/2
MRTinf
Cmax
AUCinf
CL
Vss
Intraperitoneal
t1/2
MRTinf
Cmax
Tmax
AUCinf
F
h
h
7
4.3
3
2.86
237
22
8.9
8.8
7.53
805
19
3.3
0.55
37
27
5.3
␮g/ml
min ϫ ␮g/ml
ml per min/kg
l/kg
3.8
9.8
h
h
8
6.4
0.25
5
24
65
4.9
5.5
1.44
5
225
95
8.3
7
4.29
10
755
94
␮g/ml
min
min ϫ ␮g/ml
%
Oral
t1/2
MRTinf
Cmax
Tmax
AUCinf
F
h
h
8.7
9
0.03
30
8.9
24
7.7
8.9
0.10
30
31
13
7.9
8.9
0.26
30
64
8
In addition to unchanged I-387, three metabolites (M1, M2, and
M3) were detected after a 2-h incubation in human liver microsomes
(Fig. 2). In the precursor scan mode, M1 showed the [M ϩ H]ϩ ion
at m/z 374.7, which was 13.5 amu lower than that of the parent
␮g/ml
min
min ϫ ␮g/ml
%

METABOLISM AND PHARMACOKINETICS OF I-387
639
FIG. 2. Chromatograms of I-387 incubated with human liver microsomes in the precursor ion of m/z 192 and neutral loss of 167 m/z scan modes. Chromatograms of I-387
samples in the precursor ion scan mode (A and B) and in the neutral loss scan mode (C and D) are shown.
compound (388.2 amu), suggesting the loss of a CH3 group from the also found. The LC-MS/MS fragmentation pattern of M5, M6, and
C ring. M2 observed in neutral loss as well as precursor scan modes M7 yielded the characteristic fragment ion at m/z 208, indicating
showed the [M ϩ H]ϩ ion at m/z 390.6, which was 2.4 amu higher that demethylation and hydroxylation occurred.
than that of the parent compound. These results suggest that the
The amount of each metabolite was obtained from peak areas in
carbonyl group was reduced. M3 showed the [M ϩ H]ϩ ion at m/z the EIC (Fig. 4). The metabolism of I-387 in monkey and human
404.7, which was 16.5 amu higher than that of the parent compound liver microsomes showed similar patterns, with reduction of the
in the neutral loss scan mode, indicating the addition of an oxygen carbonyl group and hydroxylation. In dog liver microsomes, hy-
atom to the A or B ring of parent compound. The fragment ion masses droxylation was the primary metabolic pathway. Dog liver micro-
of M1, M2, and M3 obtained in the product ion spectrum clearly somes generate three different hydroxylated metabolites (M3, M8,
indicated O-demethylation in the C ring, reduction of the carbonyl and M9). M3 and/or M8 were also observed in the microsomes of
group between rings B and C, and hydroxylation in the A or B ring other species. O-Demethylation was observed to a lesser degree in
(Fig. 3). The structure of M2 was verified using a synthetic reference dog, monkey, and human microsomes. Incubation of mouse and rat
standard.
liver microsomes with I-387 produced the reduced carbonyl, O-
Metabolites of I-387 in Mouse, Rat, Dog, Monkey, and Hu- demethylated, and hydroxylated metabolites in similar proportions.
man Liver Microsomes. To compare its metabolic fate in different Monkey liver microsomes produced patterns of metabolism of
species, I-387 was incubated for 2 h with mouse, rat, dog, and I-387 similar to those of human liver microsomes in vitro. Figure
monkey liver microsomes under the same conditions applied to the 5 depicts the proposed metabolic pathways of I-387 in the liver
human liver microsomes. The total ion current chromatogram of microsomes of the five species.
the linear ion trap product ion scan for m/z 375, 390.5, and 404.5
for each species is shown in Fig. 4. Nine metabolites (M1–M9)
Discussion
were observed in mouse, rat, dog, and monkey liver microsomes.
Microtubules have been considered to be one of the major and
Table 3 shows the retention time, [M ϩ H]ϩ, and major fragment best targets for cancer chemotherapy for decades. Despite the
ions of I-387 and its metabolites in the liver microsomal incuba- clinical successes of several antitubulin agents including vinca
tions of the five species. In the EIC of m/z 375 and 404.5, alkaloids and taxane analogs, toxicity and drug resistance have
additional metabolites (M4, M8, and M9) were observed, with limited the effectiveness of antimitotic agents in clinical use (Du-
different retention times than M1 and M3. M4 yielded the same montet and Sikic, 1999; Verrills and Kavallaris, 2005; Malik and
fragment ion at m/z 192 as M1. Both M8 and M9 generated the Stillman, 2008). We identified a novel compound, I-387, that
characteristic fragment ion at m/z 208 as did M3. In the EIC of m/z inhibits the in vitro growth of a number of human cancer cell lines
390.5, not only was the metabolite with the reduced carbonyl group with nanomolar IC50 values. I-387 inhibits tubulin polymerization
(M3) detected but additional metabolites (M5, M6, and M7) were via the colchicine binding site and circumvents P-glycoprotein-

640
AHN ET AL.
FIG. 3. Total ion current chromatograms of extracted ions of I-387 parent and metabolites in human liver microsomes using product ion scan mode and their
fragmentation patterns; I-387 parent with m/z 388 (A), the metabolite with m/z 375 (B), the metabolite with m/z 390 (C), and the metabolite with m/z 404.5 (D).
MW, molecular weight.
mediated multidrug resistance. An in vivo tumor xenograft study in mice. The relatively long half-life of I-387 after intravenous,
using immunodeficient mice provided compelling evidence of the intraperitoneal, and oral administration provides the opportunity
ability of I-387 to inhibit tumor growth. Furthermore, in vitro and for less frequent dosing. The bioavailability of I-387 was Ͼ65 and
in vivo studies showed that I-387 was less neurotoxic than vin- Ͻ24% after intraperitoneal and oral administration, respectively.
blastine and vincristine, tubulin destabilizers with known neuro- Therefore, intraperitoneal injection achieved sufficient exposure to
toxicity at equieffective doses.
I-387 compared with intravenous injection, without the technical
For the pharmacokinetic studies of I-387 in ICR mice, three complications or potential morbidity associated with repetitive
doses (1, 5, and 15 mg/kg) of I-387 were administered via the intravenous or oral administration. In the recent xenograft study,
intravenous, intraperitoneal, and oral route. Noncompartmental 10 mg/kg I-387 was administered via intraperitoneal injection and
pharmacokinetic analysis after intravenous injection of I-387 induced 68 and 76% tumor growth inhibition (Ahn et al., 2010).
showed that it has a large volume of distribution and low clearance The pharmacokinetic study showed that plasma concentrations of

METABOLISM AND PHARMACOKINETICS OF I-387
641
FIG. 4. Total ion current chromatograms of extracted ions of I-387 metabolites in mouse (A), rat (B), dog (C), monkey (D), and human (E) liver microsomes using product
ion scan mode. The amount of each metabolite was analyzed on the basis of its fragmentation pattern and is shown in right panel: O-demethylation (f), reduction of a ketone
group (p), and hydroxylation in A or B ring (ꢀ).

642
AHN ET AL.
In vitro metabolic stability in liver microsomes of five different
TABLE 3
species was examined in the presence of NADPH. I-387 was
metabolized mainly via phase I pathways in mouse, rat, dog,
monkey, and human liver microsomes. Because I-387 showed
stability in mouse and dog liver microsomes similar to that in
human liver microsomes, further efficacy studies in mice and
pharmacokinetic studies in dogs might be helpful to predict the
effects of I-387 in humans compared with those in the other two
species. In vitro metabolite identification studies showed that
carbonyl reduction, O-demethylation, and/or hydroxylation are
probably the predominant metabolic pathways for I-387. The po-
sitions of hydroxylation and/or demethylation varied between spe-
cies. Quantitative differences in metabolism were also observed in
the liver microsomes of different species. As a result, carbonyl
reduction was the most important metabolic pathway in humans,
suggesting that further modification in the carbonyl linkage could
improve metabolic stability of this structural pharmacophore in
humans.
Summary of the metabolite profile generated after incubation of I-387 with liver
microsomes of mouse, rat, dog, monkey, and human
ϩ
Compound
Retention Time
͓M ϩ H͔
Fragment Ions
min
I-387
11.3–11.4
388
375
192
192
Demethylation in C ring
M1
M4
10.6
10.8
Reduction of ketone
M2
10.5
390.5
390.5
154, 194, 222
208
Demethylation in C ring and
hydroxylation in A/B ring
M5
9.5
9.7
M6
M7
11.4
Hydroxylation in A/B ring
M8
M3
M9
10
10.2–10.3
10.6
404.5
208
In conclusion, a promising antitubulin agent, I-387, was slowly
cleared, moderately distributed, and moderately metabolized in mice.
The metabolic fate of I-387 varied among species. The metabolic
identification study indicates that further structural optimization of
I-387 in the carbonyl linkage could improve metabolite stability, and
the inclusion of heterocyclic rings instead of the phenyl B-ring linkage
may improve aqueous solubility and oral bioavailability. Our study
suggests that further structural optimization of I-387 will lead to the
development of new orally available antitubulin agents for the treat-
ment of drug-resistant cancer.
I-387 (15 mg/kg i.p.) were maintained at Ͼ71 ng/ml for 30 h,
suggesting that systemic drug concentrations approaching 50 ng/ml
are necessary to achieve an in vivo antitumor effect. The low to
moderate oral bioavailability of I-387 suggests that additional
structural modifications may present a plausible approach to iden-
tifying an antitubulin agent that could be administered orally. We
reported recently the synthesis and promising in vitro and in vivo
anticancer activity of a series of 2-aryl-4-benzoyl-imidazoles with
greatly improved aqueous solubility (Chen et al., 2010) as an
advance toward this goal.
FIG. 5. The proposed metabolic schemes of I-387 in the mouse, rat,
dog, monkey, and human liver microsomes.

METABOLISM AND PHARMACOKINETICS OF I-387
643
Davies B and Morris T (1993) Physiological parameters in laboratory animals and humans.
Pharm Res 10:1093–1095.
Authorship Contributions
Participated in research design: Ahn, Li, Miller, and Dalton.
Conducted experiments: Ahn.
Contributed new reagents or analytic tools: Ahn and Duke.
Wrote or contributed to the writing of the manuscript: Ahn, Kearbey, and
Dalton.
Dumontet C and Sikic BI (1999) Mechanisms of action of and resistance to antitubulin agents:
microtubule dynamics, drug transport, and cell death. J Clin Oncol 17:1061–1070.
Li CM, Lu Y, Narayanan R, Miller DD, and Dalton JT (2010) Drug metabolism and pharma-
cokinetics of 4-substituted methoxybenzoyl-aryl-thiazoles. Drug Metab Dispos 38:2032–
2039.
Malik B and Stillman M (2008) Chemotherapy-induced peripheral neuropathy. Curr Neurol
Neurosci Rep 8:56–65.
References
Schlichting I, Berendzen J, Chu K, Stock AM, Maves SA, Benson DE, Sweet RM, Ringe D,
Petsko GA, and Sligar SG (2000) The catalytic pathway of cytochrome p450cam at atomic
resolution. Science 287:1615–1622.
Sheweita SA (2000) Drug-metabolizing enzymes: mechanisms and functions. Curr Drug Metab
1:107–132.
Ahn S, Duke CB 3rd, Barrett CM, Hwang DJ, Li CM, Miller DD, and Dalton JT (2010) I-387,
a novel antimitotic indole, displays a potent in vitro and in vivo antitumor activity with less
neurotoxicity. Mol Cancer Ther 9:2859–2868.
Ahn S, Hwang DJ, Barrett CM, Yang J, Duke CB 3rd, Miller DD, and Dalton JT (2011) A novel
bis-indole destabilizes microtubules and displays potent in vitro and in vivo antitumor activity
in prostate cancer. Cancer Chemother Pharmacol 67:293–304.
Verrills NM and Kavallaris M (2005) Improving the targeting of tubulin-binding agents: lessons
from drug resistance studies. Curr Pharm Des 11:1719–1733.
Brandon EF, Raap CD, Meijerman I, Beijnen JH, and Schellens JH (2003) An update on in vitro
test methods in human hepatic drug biotransformation research: pros and cons. Toxicol Appl
Pharmacol 189:233–246.
Chen J, Wang Z, Li CM, Lu Y, Vaddady PK, Meibohm B, Dalton JT, Miller DD, and Li W
(2010) Discovery of novel 2-aryl-4-benzoyl-imidazoles targeting the colchicines binding site
in tubulin as potential anticancer agents. J Med Chem 53:7414–7427.
Address correspondence to: Dr. James T. Dalton, 3 N. Dunlap St., Memphis,
TN 38163. E-mail: jdalton@gtxinc.com