Benzyl ether-linked glucuronide derivative of 10-hydroxycamptothecin designed for selective camptothecin-based anticancer therapy

Article abstract of DOI:10.1021/jm701151c

A β-glucuronidase-activated prodrug approach was applied to 10-hydroxycamptothecin, a Camptotheca alkaloid with promising antitumor activity but poor water solubility. We synthesized a glucuronide prodrug of 10-hydroxycamptothecin (7) in which glucuronic acid is connected via a self-immolative 3-nitrobenzyl ether linker to the 10-OH group of 10-hydroxycamptothecin. Compound 7 was 80 times more soluble than 10-hydroxycamptothecin in aqueous solution at pH 4.0 and was stable in human plasma. Prodrug 7 was 10-to 15-fold less toxic than the parent drug to four human tumor cell lines. In the presence of β-glucuronidase, prodrug 7 could be activated to elicit similar cytotoxicity to the parent drug in tumor cells. Enzyme kinetic studies showed that Escherichia coli β-glucuronidase had a quite low K_m of 0.18 μM for compound 7 and exhibited 520 times higher catalytic efficiency for 7 than for 6 (a glucuronide prodrug of 9-aminocamptothecin). Molecular modeling studies predicted that compound 7 would have a higher binding affinity to human β-glucuronidase than compound 6. Prodrug 7 may be useful for selective cancer chemotherapy by a prodrug monotherapy (PMT) or antibody-directed enzyme prodrug therapy (ADEPT) strategy.

Full text of DOI:10.1021/jm701151c

1740
J. Med. Chem. 2008, 51, 1740–1746
Benzyl Ether-Linked Glucuronide Derivative of 10-Hydroxycamptothecin Designed for Selective
Camptothecin-Based Anticancer Therapy
Yu-Ling Leu,*,†,| Chien-Shu Chen,‡,| Yih-Jang Wu,‡ and Ji-Wang Chern*,‡
Department of Pharmacy, Chia-Nan UniVersity of Pharmacy and Science, Tainan 717, Taiwan, and School of Pharmacy, College of Medicine,
National Taiwan UniVersity, Taipei 100, Taiwan
ReceiVed September 14, 2007
A ꢀ-glucuronidase-activated prodrug approach was applied to 10-hydroxycamptothecin, a Camptotheca
alkaloid with promising antitumor activity but poor water solubility. We synthesized a glucuronide prodrug
of 10-hydroxycamptothecin (7) in which glucuronic acid is connected via a self-immolative 3-nitrobenzyl
ether linker to the 10-OH group of 10-hydroxycamptothecin. Compound 7 was 80 times more soluble than
10-hydroxycamptothecin in aqueous solution at pH 4.0 and was stable in human plasma. Prodrug 7 was 10-
to 15-fold less toxic than the parent drug to four human tumor cell lines. In the presence of ꢀ-glucuronidase,
prodrug 7 could be activated to elicit similar cytotoxicity to the parent drug in tumor cells. Enzyme kinetic
studies showed that Escherichia coli ꢀ-glucuronidase had a quite low Km of 0.18 µM for compound 7 and
exhibited 520 times higher catalytic efficiency for 7 than for 6 (a glucuronide prodrug of 9-aminocamp-
tothecin). Molecular modeling studies predicted that compound 7 would have a higher binding affinity to
human ꢀ-glucuronidase than compound 6. Prodrug 7 may be useful for selective cancer chemotherapy by
a prodrug monotherapy (PMT) or antibody-directed enzyme prodrug therapy (ADEPT) strategy.
Chart 1
Introduction
(20S)-Camptothecin (1, Chart 1), first discovered in 1966 from
Camptotheca acuminata (Nyssaceae),1 displays antitumor activ-
ity by inhibiting DNA topoisomerase I.2 Clinical use of
camptothecin, however, was limited by its extremely poor water
solubility. Efforts in developing water-soluble derivatives of
camptothecin led to the synthesis of topotecan (3)3 and
irinotecan (CPT-11, 4),4 which are derivatives of 10-hydroxy-
camptothecin (2)5,6 and have been approved for clinical use.
Irinotecan is a prodrug, which is activated by carboxylesterases
to generate the more potent metabolite, compound 5 (SN-38).7
However, carboxylesterases are ubiquitous enzymes, resulting
in a low therapeutic index of the prodrug.8 We have first
employed a ꢀ-glucuronidase-activated prodrug approach to more
tumor-selective chemotherapy as well as increased water
solubility for camptothecins.9 In normal tissues, ꢀ-glucuronidase
is localized primarily in lysosomes10 and thereby not available
for activation of glucuronide prodrugs because these prodrugs
are generally hydrophilic, thus rendering them impermeable to
cell membranes. In contrast, ꢀ-glucuronidase is found to
accumulate extracellularly at the tumor site, mainly in necrotic
areas.11 Therefore, glucuronide-based prodrugs can be used in
prodrugmonotherapy(PMTa)forselectivecancerchemotherapy.11,12
It has also been shown that the glucuronide prodrugs are useful
for application in the antibody-directed enzyme prodrug therapy
(ADEPT) strategy such that ꢀ-glucuronidase can be targeted to
tumor cells by administration of antibody-ꢀ-glucuronidase
conjugates.13 The recent development of ꢀ-glucuronidase-
activated prodrugs includes elaborations of paclitaxel14,15 and
two potent DNA repair inactivators, O6-benzylguanine and O6-
benzyl-2′-deoxyguanonsine.16
We previously designed and synthesized a glucuronide
derivative of 9-aminocamptothecin (6) as a ꢀ-glucuronidase-
cleavable prodrug in which glucuronic acid is connected via a
self-immolative carbamate linker to the 9-amino group of
9-aminocamptothecin.9 Compound 6 is water soluble, is stable
* To whom correspondence should be addressed. For Y.-L.L.: phone,
886-6-2664911, extension 200; fax, 886-6-2667318; e-mail, yulin@
mail.chna.edu.tw. For J.-W.C.: phone, 886-2-2393-9462; fax, 886-2-2393-
4221; e-mail, chern@jwc.mc.ntu.edu.tw.
†
Chia-Nan University of Pharmacy and Science.
These authors contributed equally to this work.
National Taiwan University.
|
‡
a Abbreviations: PMT, prodrug monotherapy; ADEPT, antibody-directed
enzyme prodrug therapy; PBS, phosphate-buffered saline; E. coli, Escheri-
chia coli.
10.1021/jm701151c CCC: $40.75
2008 American Chemical Society
Published on Web 03/05/2008

Glucuronide DeriVatiVe for Anticancer Therapy
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 6 1741
Chart 2. Release of 10-Hydroxycamptothecin (2) from Prodrug
7 by ꢀ-Glucuronidase Activation
in human plasma, and is a substrate for ꢀ-glucuronidase. This
prodrug is less toxic than 9-aminocamptothecin toward human
tumor cell lines and upon enzyme activation can display
cytotoxicity similar to that of the parent drug. Prodrug 6 was
also shown to display potent in vivo antitumor activity against
human tumor xenografts in nude mice.17 More studies of
prodrug 6 have been conducted to evaluate its potential clinical
utility for selective cancer therapy.18–20
In the synthesis of compound 6, the required material
9-aminocamptothecin was prepared from 10-hydroxycamptoth-
ecin via three synthetic steps. In fact, 10-hydroxycamptothecin
is a promising antitumor alkaloid that is also produced by the
roots of Camptotheca acuminata as a minor component. In
topoisomerase I inhibition assays, 10-hydroxycamptothecin was
found to be more potent than camptothecin and topotecan.21
10-Hydroxycamptothecin exhibited a broad spectrum of anti-
cancer activity against various cancer cell lines and animal tumor
models.6,22–25 Moreover, 10-hydroxycamptothecin has been
shown to be effective against multidrug-resistant cancer cell
lines, whereas other camptothecin analogues such as topotecan,
compound 5, and 9-aminocamptothecin were relatively inef-
fective.26 10-Hydroxycamptothecin has entered clinical trials,
mainly in China. To date, studies on prodrugs of 10-hydroxy-
camptothecin have been reported, including its fatty acid,27,28
poly(ethylene glycol) (PEG),29 and alkyl carbonate30 derivatives.
In light of the promising prodrug properties of 6, we wished
to apply the same approach to 10-hydroxycamptothecin. The
phenolic hydroxyl group of 10-hydroxycamptothecin was
considered to be derivatized according to the work of Toki et
al. where the peptide derivatives, containing a benzyl ether
linker, of some phenolic anticancer drugs were studied in the
protease-mediated prodrug activation strategy.31 The glucuronide
prodrug of 10-hydroxycamptothecin (7) was designed using a
self-immolative 3-nitrobenzyl spacer which was ether-linked to
the 10-OH group of 10-hydroxycamptothecin. The use of a
spacer between the drug and glucuronic acid would allow
superior enzymatic hydrolysis of prodrugs.32 In addition, for
the aromatic spacer of 7, the strong electron-withdrawing nitro
group (ortho to the glucuronyl moiety) might facilitate enzymatic
cleavage of the glycosidic bond.33 The enzymatic activation of
prodrug 7 to liberate 10-hydroxycamptothecin through a 1,6-
elimination reaction was expected as shown in Chart 2. Herein,
we describe the synthesis and biological evaluation of the
glucuronide prodrug 7.
to a relatively low electron density on the 10-OH oxygen atom
of 10-hydroxycamptothecin containing a more extended π-con-
jugated system than 6-hydroxyquinoline. We then turned to the
conversion of alcohol 10 to mesylate 11 by treatment with
methanesulfonyl chloride. In the presence of Cs2CO3, the
coupling reaction of mesylate 11 with 10-hydroxycamptothecin
could be effected in DMF at room temperature for 2 h to afford
12 in 47% yield. It should be noted that prolonged reaction time
(>2 h) would gradually lead to the formation of a byproduct.
On the basis of NMR and mass spectral data, this byproduct
was identified as the ꢀ-elimination product derived from 12 by
loss of one molecule of acetic acid at positions C4 and C5 of
the glucuronyl moiety. It has been reported in the literature that
acetyl-protected glucuronic acids can undergo this type of
elimination under basic conditions.35,36 The O-acetyl groups of
12 were deprotected with NaOMe in MeOH to yield 13. The
methyl ester 13 was treated with potassium trimethylsilanolate
(KOSiMe3) in THF, acidified with 1 N HCl, and purified by
reverse-phase column chromatography on silica gel to obtain
the target compound 7.
Chemistry
Results and Discussion
The glucuronide prodrug 7 was prepared as shown in Scheme
1. Methyl 1,2,3,4-tetra-O-acetyl-ꢀ-D-glucopyranuronate (8),
prepared as described previously,9 was brominated with TiBr4
and then reacted with 4-hydroxy-3-nitrobenzaldehyde in the
presence of Ag2O to afford 9. Reduction of the formyl group
in 9 by treatment with NaBH4 gave the alcohol 10. To prepare
compound 12, we planned to utilize the Mitsunobu reaction34
to couple alcohol 10 with 10-hydroxycamptothecin. At first a
model study on the Mitsunobu reaction was carried out using
6-hydroxyquinoline, with the structure corresponding to the AB-
ring part of 10-hydroxycamptothecin. Treatment of 10 with
6-hydroxyquinoline in DMF in the presence of diisopropyl
azodicarboxylate (DIAD) and Ph3P at room temperature for 4
days gave the coupling product in 40% yield. However, the
reaction of 10 with 10-hydroxycamptothecin did not proceed
at all under the same Mitsunobu conditions for 6 days, even at
elevated temperature (up to 80 °C). This nonreactivity of 10-
hydroxycamptothecin in the Mitsunobu reaction might be due
Camptothecins are generally poorly soluble in water and have
a feature that the lactone E-ring is susceptible to nonenzymatic
hydrolysis to form the carboxylate species at pH 7 or above.
At neutral pH, the lactone-carboxylate equilibrium is shifted
toward the hydrophilic carboxylate form.37 Consequently,
camptothecin compounds would display higher aqueous solubil-
ity at neutral pH than at acidic pH (existing in the hydrophobic
lactone form). Unfortunately, the more soluble carboxylate form
is biologically inactive. To circumvent the solubility problem,
the elaboration of glucuronide derivatives of camptothecins may
represent a potential approach based on the highly hydrophilic
nature of the glucuronyl moiety with a carboxyl and three
hydroxyl polar groups. Because of the water-solubilizing
glucuronyl moiety, the synthesized prodrug 7 was expected to
have greater aqueous solubility compared with the parent
compound. As shown in Table 1, compound 7 was 80 and 19
times more water-soluble than 10-hydroxycamptothecin at pH
4.0 and 7.0, respectively. Formulation of camptothecins at acidic

1742 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 6
Leu et al.
Scheme 1a
a Reagents and conditions: (a) TiBr4, CH2Cl2, room temp, 24 h; (b) 4-hydroxy-3-nitrobenzaldehyde, Ag2O, CH3CN, room temp, 15 h, 62%; (c) NaBH4,
silica gel, i-PrOH/CHCl3, 0 °C, 1 h, 76%; (d) methanesulfonyl chloride, TEA, CH2Cl2, 0 °C, 1 h, 97%; (e) 10-hydroxycamptothecin (2), Cs2CO3, DMF,
room temp, 2 h, 47%; (f) NaOMe, MeOH, room temp, 2 h, 46%; (g) KOSiMe3, THF, room temp, 2 h, then acidified with 1 N HCl, 12%.
Table 1. Aqueous Solubility of Compoundsa
solubility (mM)
compd
pH 4.0
pH 7.0
2
7
0.0137
1.10
0.137
2.60
a The solubility of 10-hydroxycamptothecin (2) and prodrug 7 was
determined in PBS (pH 4.0 and 7.0) at 25 °C by a single determination.
pH is important to prevent formation of the inactive carboxylate
form as a result of lactone ring opening.37 Like 9-aminocamp-
tothecin, 10-hydroxycamptothecin has poor aqueous solubility,
especially at acidic pH (10 times lower for pH 4.0 compared
with neutral pH). Thus, the 80-fold increased solubility at pH
4.0 of 7 (1.1 mM) is favorable for drug formulation. At pH
7.0, compound 7 had an aqueous solubility of 2.6 mM, which
is less than that of 6.9 However, highly polar glucuronide
prodrugs may have fast renal clearance, thus limiting their
clinical applications.38 Development of new glucuronide pro-
drugs with a less hydrophilic nature may be advantageous.
Prediction of the octanol/water partition coefficient using the
program XLOGP 2.039 gave the calculated log P values of 1.19
and 0.81 for prodrugs 7 and 6, respectively. The relatively lower
hydrophilicity of 7 may retard prodrug excretion and, hence,
result in prolonged exposure at the tumor site.
The stability of 7 was studied in 95% human plasma at 37
°C. Glucuronide 7 proved to be completely stable for at least
3 h (Figure 1), indicating its high resistance to nonspecific
enzymatic degradation in human plasma. Prior to this work,
we attempted to synthesize a carbonate-linked analogue of 7,
10-O-[4-(6-O-methyl-ꢀ-D-glucuronyloxy)benzyloxycarbonyl]-
10-hydroxycamptothecin (designated here as 10-HCG-carbon-
ate) and assess its stability. In contrast to prodrug 7, 10-HCG-
carbonate was unstable in human plasma with 70% degradation
in 10 min (Figure 1). The rapid degradation of 10-HCG-
carbonate in plasma was not likely due to chemical instability
because this compound was much more stable in PBS with
Figure 1. Stability of prodrug 7 and 10-HCG-carbonate studied in
95% human plasma: 10-HCG-carbonate, a carbonate-linked analogue
of 7 with the chemical name 10-O-[4-(6-O-methyl-ꢀ-D-glucuronyloxy-
)benzyloxycarbonyl]-10-hydroxycamptothecin.
∼70% decomposition after 11 h (data not shown). As observed
by TLC analysis, 10-HCG-carbonate was degraded in plasma
to produce the parent drug 2. This conversion presumably
resulted from direct enzymatic cleavage of the carbonate linkage
of the prodrug by esterases, which are abundantly present in
blood. Previously, Toki et al. have described protease-activated
prodrugs of the phenolic anticancer agent combretastatin A-4
with a stable benzyl ether or unstable benzyl carbonate linkage.31
In contrast, nevertheless, 10-HCG-carbonate was much more
unstable than the carbonate derivative of combretastatin A-4,
which had a half-life of 45 h in human plasma. This indicates
that a carbonate derivatization at the 10-OH group of 10-
hydroxycamptothecin is not suitable for selective activation by
ꢀ-glucuronidase.
10-Hydroxycamptothecin and prodrug 7 were evaluated for
the cytotoxicity against four human tumor cell lines (HepG2,
Colo 205, HT29, and H928) by measuring [3H]thymidine
incorporation into cellular DNA after 48 h of drug exposure.
As indicated in Table 2, prodrug 7 was 10–15 times less toxic

Glucuronide DeriVatiVe for Anticancer Therapy
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 6 1743
Table 2. Cytotoxicity of Compounds to Human Tumor Cells
IC50 (nM)a
compd
HepG2
Colo 205
HT29
H928
2
7
5.4
56.5
6.6
9.1
94.2
10.2
8.8
97.8
10.0
6.9
91.1
11.1
7 + ꢀ-glucuronidaseb
a Concentrations of compounds that inhibited incorporation of [3H]thy-
midine into cellular DNA of human hepatocellular (HepG2), colorectal (Colo
205), colorectal (HT29), or lung (H928) carcinoma cells by 50% after 48 h
are indicated. Values represent the mean of one to three experiments
performedintriplicatewithcoefficientsofvariationof<10%.b ꢀ-Glucuronidase
(5 µg/mL) was added with drugs.
Figure 4. HPLC chromatogram for the hydrolysis of 7 by ꢀ-glucu-
ronidase (0.125 µg/mL) at pH 7.0 after 20 min of incubation at 37 °C.
Table 3. Enzyme Kinetic Parameters for Prodrugs
prodrug
Km (µM)
Vmax (µM/min)
Vmax/Km
7
6
0.18
30
2.7
0.86
15
0.03
Figure 2. Cytotoxicity of compounds 2 and 7 to human colorectal
carcinoma cells (HT29). HT29 cells were exposed to compounds with
or without ꢀ-glucuronidase for 48 h before the incorporation of
[3H]thymidine into cellular DNA was measured. Results represent mean
values of triplicate determinations.
ꢀ-glucuronidase, HPLC analysis showed >50% conversion of
7 to 10-hydroxycamptothecin (2) with no spacer-drug inter-
mediate detected (Figure 4). This indicates that prodrug 7 was
susceptible to ꢀ-glucuronidase hydrolysis and that the spacer
rapidly decomposed upon enzymatic activation. To our knowl-
edge, this is the first example of a self-immolative glucuronide-
based prodrug in which the spacer was ether-linked to the drug.
Recently, a glucuronide prodrug of compound 5 was synthesized
using an aromatic spacer connected via a carbamate linkage to
the 10-OH group of 5.40 The enzyme kinetic parameters were
determined for 7 as well as 6 (Table 3). As a substrate of
ꢀ-glucuronidase, compound 7 showed a good Km of 0.18 µM
and a Vmax of 2.7 µM/min. The data showed that ꢀ-glucuronidase
exhibited 520 times higher catalytic efficiency for 7 than for 6,
as revealed by the Vmax/Km ratios (15 versus 0.029). According
to the Km values (0.18 versus 30 µM), compound 7 had an
approximately 170 times higher affinity to ꢀ-glucuronidase than
compound 6. Thus, it appears that the large difference in
catalytic efficiency between 7 and 6 can be attributed mainly
to the considerably distinct binding affinities of these two
prodrugs to the enzyme.
We were interested in understanding the interactions of
compounds 7 and 6 with ꢀ-glucuronidase and carried out
molecular docking studies to explore their probable binding
modes within the active site. To date, the only available
experimental three-dimensional structure for ꢀ-glucuronidase
is the X-ray crystal structure of human ꢀ-glucuronidase disclosed
by Jain et al.41 The crystal structure of human ꢀ-glucuronidase
was retrieved from the Protein Data Bank (PDB code 1BHG)
for the present docking studies. In our molecular modeling
process (details in Supporting Information), the known substrate
molecule p-nitrophenyl ꢀ-glucuronide42 was first docked into
the active site of ꢀ-glucuronidase using the docking program
GOLD 2.0.43 The resulting complex structure was then opti-
mized by energy minimization using the Tripos force field in
the software package SYBYL 7.1 (Tripos, Inc., St. Louis, MO).
The purpose of such initial modeling is to build a protein model
whose active site exists in the probable substrate-bound
Figure 3. Hydrolysis of prodrug 7 to 10-hydroxycamptothecin (2) by
E. coli ꢀ-glucuronidase. Prodrug 7 was incubated with 0.1 µg/mL
ꢀ-glucuronidase at pH 7.0 and 37 °C. The concentrations of 7 and 2 in
aliquots taken at the indicated times were determined by HPLC. Results
are shown as mean values of triplicate determinations.
than 10-hydroxycamptothecin. In the presence of ꢀ-glucu-
ronidase (5 µg/mL), prodrug 7 was as cytotoxic as 10-
hydroxycamptothecin itself (Figure 2, Table 2), indicating
efficient enzymatic cleavage with complete release of 10-
hydroxycamptothecin.
Enzymatic cleavage of prodrug 7 was studied using Escheri-
chia coli ꢀ-glucuronidase (0.1 µg/mL) in pH 7.0 phosphate
buffer at 37 °C. Figure 3 shows the time course of hydrolysis
of 7 on exposure to ꢀ-glucuronidase. The results reveal that
prodrug 7 was readily cleaved by ꢀ-glucuronidase to release
the active drug 2. For a 20-min incubation with 0.125 µg/mL

1744 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 6
Leu et al.
Table 4. Predicted Binding Affinities for Prodrugs 7 and 6 to
ꢀ-Glucuronidase Based on the Docking Models
predicted binding affinitya
AutoDock binding free
prodrug
energy (kcal/mol)
SCORE (pKd)
LigScore (pKi)
7
6
-10.65
-9.14
9.18
8.92
7.28
6.80
a For the scoring functions, see refs 5-7 in Supporting Information.
that compound 7 would bind to human ꢀ-glucuronidase more
tightly than 6. E. coli and human ꢀ-glucuronidases share 50%
sequence homology with highly conserved active sites.41,45
Accordingly, we believed that compound 7 could also adopt a
more favorable shape complementary to E. coli ꢀ-glucuronidase
than 6, explaining the lower Km value for 7.
In summary, we have synthesized the glucuronide prodrug 7
that showed improved water solubility and reduced cytotoxicity
relative to 10-hydroxycamptothecin. Prodrug 7 was stable in
human plasma and could efficiently liberate the free drug by
ꢀ-glucuronidase activation. Compared with prodrug 6 for
9-aminocamptothecin, prodrug 7 was much more susceptible
to ꢀ-glucuronidase hydrolysis and possessed a less hydrophilic
nature that should be considered in the development of new-
generation glucuronide prodrugs with slower clearance. The
requisite prodrug profiles of 7 suggest its potential application
for cancer prodrug monotherapy and antibody-directed enzyme
prodrug therapy of cancer.
Experimental Section
Chemistry. Melting points were obtained on an Electrothermal
apparatus and are uncorrected. 1H and 13C nuclear magnetic
resonance spectra were recorded on a Bruker DPX-200 spectrom-
eter. FAB mass spectra were recorded on a Finnigan MAT 95S
mass spectrometer. Elemental analyses for C, H, and N were carried
out on a Heraeus VarioEL-III elemental analyzer. The thin-layer
chromatographic analyses were performed using precoated silica
gel (60 F254, Merck) plates, and the spots were examined under
UV light. Column chromatography was carried out on Merck silica
gel 60 (70–230 mesh). Reverse phase column chromatography was
performed on Merck LiChroprep RP-18 (40–63 µm). 10-Hydroxy-
camptothecin was purchased from China.
Methyl 1-O-(4-Formyl-2-nitrophenyl)-2,3,4-tri-O-acetyl-ꢀ-D-
glucopyranuronate (9). A solution of methyl 1,2,3,4-tetra-O-acetyl-
ꢀ-D-glucopyranuronate (8)9 (5 g, 13.3 mmol) and TiBr4 (5.07 g,
13.8 mmol) in CH2Cl2 (50 mL) was stirred at room temperature
for 24 h. The mixture was washed with ice-cold water (50 mL)
and saturated aqueous NaHCO3 (50 mL), dried over Na2SO4, and
evaporated under reduced pressure to dryness to give 5.1 g of solid.
The solid was dissolved in CH3CN (100 mL). To this solution were
added 4-hydroxy-3-nitrobenzaldehyde (2.3 g, 13.8 mmol) and Ag2O
(3.2 g, 13.8 mmol), and the mixture was stirred at room temperature
for 15 h. The insoluble material was filtered off. The filtrate was
evaporated under reduced pressure to give a dark-brown solid,
which was washed with MeOH to give 9 (3.99 g, 62%): mp
180–182 °C; 1H NMR (200 MHz, DMSO-d6) δ 2.00 (s, 9H, CH3),
3.61 (s, 3H, OCH3), 4.77 (d, J ) 9.5 Hz, 1H, sugar-H), 5.07–5.18
(m, 2H, sugar-H), 5.45 (t, J ) 9.2 Hz, 1H, sugar-H), 5.91 (d, J )
6.0 Hz, 1H, sugar-H), 7.62 (d, J ) 8.7 Hz, 1H, ArH), 8.20 (d, J )
8.6 Hz, 1H, ArH), 8.42 (s, 1H, ArH), 9.95 (s, 1H, CHO); 13C NMR
(50 MHz, DMSO-d6) δ 21.0, 21.1, 21.2, 53.5, 69.3, 70.5, 71.3,
72.1, 98.1, 118.5, 127.1, 131.9, 135.6, 141.1, 152.9, 167.7, 169.6,
170.2, 170.4, 191.4. Anal. (C20H21NO13) C, H, N.
Figure 5. Docking models of (A) compound 7 and (B) compound 6
with human ꢀ-glucuronidase. The possible binding site of ꢀ-glucu-
ronidase is depicted as molecular surface, and the catalytic residues
Glu451 and Glu540 are colored magenta.
conformation. We chose p-nitrophenyl ꢀ-glucuronide for simu-
lation because it has smaller molecular size and is also analogous
to the spacer-glucuronyl parts of compounds 7 and 6. The
modeled substrate-bound structure of human ꢀ-glucuronidase
showed that the glycosidic bond of p-nitrophenyl ꢀ-glucuronide
was properly oriented toward the catalytic residues Glu451 and
Glu540 (Figure S1 in Supporting Information). It has been
proposed for human ꢀ-glucuronidase that in catalysis Glu451
acts as the acid/base catalyst and Glu540 as the nucleophilic
residue.44 This modeled protein structure was then used in the
prediction of a favorable binding mode of 7 or 6 to ꢀ-glucu-
ronidase by computational simulations as described above. The
predicted binding models of compounds 7 and 6 to ꢀ-glucu-
ronidase are shown in Figure 5 (see also Figures S2 and S3 in
Supporting Information). In addition, for a more reliable
comparison, the binding affinities of 7 and 6 were predicted
using three programs that have different scoring functions for
calculating the binding affinity of a ligand. According to the
calculated data (Table 4), compound 7 possesses a lower binding
free energy or a higher pKd (or pKi) value, which may allow
more favorable binding to ꢀ-glucuronidase compared with
compound 6. Analysis of the predicted binding conformations
revealed that compound 7 can adopt a linear conformation for
better fitting to the straight binding groove of ꢀ-glucuronidase.
In contrast, compound 6 has the spacer attached at the 9-position
of camptothecin, thus being unable to adopt a linear overall
molecular shape. The results from molecular modeling suggested
Methyl 1-O-(4-Hydroxymethyl-2-nitrophenyl)-2,3,4-tri-O-
acetyl-ꢀ-D-glucopyranuronate (10). A mixture of 9 (2 g, 4.14
mmol), NaBH4 (434 mg, 11.5 mmol), and silica gel (10 g) in
i-PrOH/CHCl3 (3:17) (200 mL) was stirred at 0 °C for 1 h. The
reaction was quenched with water, and the mixture was filtered to
remove silica gel. The organic layer was dried over MgSO4 and

Glucuronide DeriVatiVe for Anticancer Therapy
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 6 1745
evaporated under reduced pressure to give a residue, which was
washed with EtOH to give 10 (1.52 g, 76%): mp 167–168 °C; 1H
NMR (200 MHz, DMSO-d6) δ 1.98–2.00 (m, 9H, CH3) 3.64 (s,
3H, OCH3), 4.50 (d, J ) 5.4 Hz, 2H, CH2), 4.72 (d, J ) 9.8 Hz,
1H, sugar-H), 5.03–5.14 (m, 2H, sugar-H), 5.37–5.49 (m, 2H,
sugar-H and OH), 5.70 (d, J ) 7.8 Hz, 1H, sugar-H), 7.37 (d, J )
8.6 Hz, 1H, ArH), 7.60 (d, J ) 7.3 Hz, 1H, ArH), 7.79 (s, 1H,
ArH); 13C NMR (50 MHz, DMSO-d6) δ 21.0, 21.1, 21.2, 53.5,
62.2, 69.6, 70.8, 71.7, 71.9, 98.9, 118.6, 123.2, 132.9, 139.4, 141.1,
147.8, 167.8, 169.6, 170.2, 170.4. Anal. (C20H23NO13) C, H, N.
Methyl 1-O-(4-Methanesulfonyloxymethyl-2-nitrophenyl)-
2,3,4-tri-O-acetyl-ꢀ-D-glucopyranuronate (11). A solution of 10
(300 mg, 0.62 mmol), methanesulfonyl chloride (0.06 mL, 0.78
mmol), and triethylamine (0.1 mL, 0.71 mmol) in CH2Cl2 (15 mL)
was stirred at 0 °C for 1 h. The mixture was washed with saturated
aqueous NaHCO3, dried over MgSO4, and evaporated under reduced
pressure to dryness to give 11 (340 mg, 97%): mp 110–112 °C;
1H NMR (200 MHz, DMSO-d6) δ 2.00 (s, 9H, CH3), 3.27 (s, 3H,
SO2CH3), 3.63 (s, 3H, OCH3), 4.74 (d, J ) 9.8 Hz, 1H, sugar-H),
5.04–5.16 (m, 2H, sugar-H), 5.29 (s, 2H, CH2), 5.46 (t, J ) 9.4
Hz, 1H, sugar-H), 5.77 (d, J ) 7.7 Hz, 1H, sugar-H), 7.47 (d, J )
8.4 Hz, 1H, ArH), 7.79 (d, J ) 8.6 Hz, 1H, ArH), 8.02 (s, 1H,
ArH); FAB-MS m/z 562 (M+ - 1). Anal. (C21H25NO15S) C, H, N.
10-[4-(2,3,4-Tri-O-acetyl-6-O-methyl-ꢀ-D-glucuronyloxy)-3-ni-
trobenzyloxy]camptothecin (12). To a suspension of 10-hydroxy-
camptothecin (600 mg, 1.65 mmol) and Cs2CO3 (600 mg, 1.84
mmol) in anhydrous DMF (30 mL) was added a solution of 11 (3
g, 5.32 mmol) in anhydrous DMF (30 mL). The mixture was stirred
at room temperature for 2 h. The crude product was purified by
column chromatography on silica gel (MeOH/CHCl3 ) 2:98) to
give 12 (650 mg, 47%): mp 259–261 °C; 1H NMR (200 MHz,
DMSO-d6) δ 0.87 (t, J ) 7.1 Hz, 3H, CH3) 1.80–2.00 (m, 2H,
CH2), 2.01–2.03 (m, 9H, CH3), 3.63 (s, 3H, OCH3), 4.71 (d, J )
9.7 Hz, 1H, sugar-H), 5.08–5.18 (m, 2H, sugar-H), 5.25 (s, 2H,
CH2), 5.31 (s, 2H, CH2), 5.40–5.52 (m, 1H, sugar-H), 5.74 (d, J )
8.0 Hz, 1H, sugar-H), 6.52 (s, 1H, OH-20), 7.27 (s, 1H, ArH),
7.35–7.71 (m, 3H, ArH), 7.74–7.95 (m, 1H, ArH), 8.10 (s, 2H,
ArH), 8.53 (s, 1H, ArH); FAB-MS m/z 832 (M+). Anal.
(C40H37N3O17 ·H2O) C, H, N.
sample was injected onto a reversed phase column (LiChroCART
RP-18, 4 mm inside diameter, 250 mm length, 5 µm particle size)
using a mobile phase (1 mL/min) of 50% MeOH and 50 mM
phosphate buffer (pH 2.5). Eluted compounds were detected on a
Hitachi L-7480 fluorometer (excitation, 397 nm; emission, 482 nm).
Peak areas were analyzed with Hitachi D-7000 chromatography
data station software. Calibration curves were obtained by plotting
the peak area of standards as a function of drug concentration. The
retention times of 10-hydroxycamptothecin and prodrug 7 were 8.0
and 11.4 min, respectively.
2. Drug Solubility. Drug solubilities were determined in
phosphate buffer (100 mM, pH 7.0) or phosphate buffer (100 mM,
pH 4.0) by equilibrating an excess of solid compound in 0.25 mL
of buffer at 25 °C for 24 h. The samples were filtered through a
0.2 µm Millipore filter, diluted in HPLC mobile phase, and analyzed
by HPLC.
3. Prodrug Stability in 95% Human Serum. Prodrug stability
was determined in 95% human plasma at 37 °C for 6 h. Aliquots
were taken, neutralized with 12 mM phosphoric acid, and extracted
three times with an equal volume of ethyl acetate. The ethyl acetate
layer was evaporated under reduced pressure to dryness. This
residue was dissolved in HPLC mobile phase, filtered through a
0.2 µm Millipore filter, and analyzed by HPLC.
4. In Vitro Cytotoxicity. Exponentially growing tumor cells
at a density of 5 × 103 cells/well in RPMI medium containing
10% bovine serum, 100 U/mL penicillin, and 100 µg/mL strepto-
mycin were incubated in a 96-well microtiter plate for 72 h (37
°C, 5% CO2, humidity) with various concentrations of drug. 10-
Hydroxycamptothecin and prodrug 7 were dissolved in DMSO such
that the final concentration of DMSO in wells did not exceed 0.5%.
Control wells consisted of cells exposed to 0.5% DMSO in medium.
ꢀ-Glucuronidase, added at 1 µg/well in some experiments, was not
toxic by itself to cells. Triplicate wells were prepared for each drug
concentration and for the controls. After 48 h, cells were pulsed
for 12 h with [3H]thymidine (1 µCi/well) in complete medium.
Medium was removed, and the wells were washed once with PBS
before trypsinized cells were harvested and counted for radioactivity
in a Topcount liquid scintillation counter. The coefficient of
variation for triplicate determinations was <10%. IC50 values were
calculated from interpolation of logarithmic dose–response curves.
5. Enzyme Kinetic Analysis. A solution of various concentra-
tion of prodrug 7 was incubated with 0.125 µg/mL E. coli
ꢀ-glucuronidase at pH 7.0 for 20 min at 37 °C. The reaction was
stopped by addition of 900 µL of ice-cold MeOH to precipitate
proteins. The samples were centrifuged at 12 000 rpm for 5 min.
The supernatants were evaporated in vacuo to dryness, and the
resulting residues were dissolved in 5% H3PO4-MeOH (1:6) for
HPLC analysis. For the time-course assay of hydrolysis of 7 to the
parent compound 2, prodrug 7 was incubated with 0.1 µg/mL
ꢀ-glucuronidase at pH 7.0 and 37 °C. The concentrations of 7 and
2 in aliquots taken at the indicated times were determined by HPLC.
10-[4-(6-O-Methyl-ꢀ-D-glucuronyloxy)-3-nitrobenzyloxy]camp-
tothecin (13). A suspension of 12 (360 mg, 0.43 mmol) and NaOMe
(290 mg, 5.37 mmol) in anhydrous MeOH (160 mL) was stirred at
room temperature for 2 h. The crude product was purified by column
chromatography on silica gel (MeOH/CHCl3 ) 1:9) to give 13 (140
mg, 46%): mp 207–209 °C; 1H NMR (200 MHz, DMSO-d6) δ
0.86 (t, J ) 7.2 Hz, 3H, CH3), 1.85–1.87 (m, 2H, CH2), 3.32–3.48
(m, 3H, sugar-H), 3.64 (s, 3H, OCH3), 4.13 (d, J ) 8.9 Hz, 1H,
sugar-H), 5.26–5.53 (m, 9H, CH2, sugar-H and OH), 6.51 (s, 1H,
OH-20), 7.27 (s, 1H, ArH), 7.47–7.62 (m, 3H, ArH), 7.81 (d, J )
8.3 Hz, 1H, ArH), 8.08 (d, J ) 8.6 Hz, 2H, ArH), 8.53 (s, 1H,
ArH); FAB-MS m/z 706 (M+). Anal. (C34H31N3O14 ·CH3OH) C,
H, N.
10-[4-(ꢀ-D-Glucuronyloxy)-3-nitrobenzyloxy]camptothecin (7).
A suspension of 13 (110 mg, 0.16 mmol) and KOSiMe3 (105 mg,
0.82 mmol) in anhydrous THF (30 mL) was stirred at room
temperature for 2 h. The solvent was removed under reduced
pressure, and the resulting residue was dissolved in water and then
washed with CHCl3. The aqueous layer was acidified with 1 N
HCl and purified by reverse phase column chromatography on silica
gel (CH3CN/H2O ) 1:5) to give 7 (13 mg, 12%): mp 196–197 °C;
1H NMR (200 MHz, DMSO-d6 + D2O) δ 0.88 (t, J ) 6.6 Hz, 3H,
CH3), 1.85–1.88 (m, 2H, CH2), 3.16–3.25 (m, 3H, sugar-H), 3.72
(m, 1H, sugar-H), 5.23 (s, 2H, CH2), 5.29 (s, 2H, CH2), 5.41 (s,
2H, CH2), 5.73 (s, 1H, sugar-H), 7.28 (s, 1H, ArH), 7.54–7.81 (m,
4H, ArH), 8.04 (s, 1H, ArH), 8.08 (s, 1H, ArH), 8.52 (s, 1H, ArH);
FAB-MS m/z 692 (M+ + 1). HRMS (FAB) calcd for C33H30N3O14
Acknowledgment. The authors thank Dr. Steve R. Roffler
of Institute of Biomedical Sciences, Academia Sinica, Taipei,
Taiwan, for providing helpful experimental facilities and the
National Center for High-Performance Computing (NCHC),
Hsinchu, Taiwan, for providing computing resources. We also
thank Dr. Shin-Yu Lai and Hsiang-Ling Huang for their
technical assistance.
Supporting Information Available: Elemental analysis results
for synthesized compounds, HPLC analysis results and NMR
spectra for final compound 7, and molecular modeling details. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(M+
+ 1) 692.1728, found 692.1731. Anal. Calcd for
C33H29N3O14 ·10H2O: C, 45.47; H, 5.67; N, 4.82. Found: C, 45.34;
H, 4.82; N, 4.64.
References
Biological Tests. 1. HPLC Analysis. Drugs were analyzed by
high-pressure liquid chromatography (HPLC). Briefly, 20 µL of
(1) Wall, M. E.; Wani, M. C.; Cook, C. E.; Palmar, K. H.; McPhail, A. T.;
Sim, G. A. Plant antitumor agents I. The isolation and structure of

1746 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 6
Leu et al.
camptothecin, a novel alkaloidal leukemia and tumor inhibitor from
Acuminata. J. Am. Chem. Soc. 1966, 88, 3888–3890.
(2) Hsiang, Y.-H.; Hertzberg, R.; Hecht, S.; Liu, L. F. Camptothecin
induces protein-linked DNA breaks via mammalian DNA topoi-
somerase I. J. Biol. Chem. 1985, 260, 14873–14878.
(3) Kingsbury, W. D.; Boehm, J. C.; Jakas, D. R.; Holden, K. G.; Hecht,
S. M.; Gallagher, G.; Caranfa, M. J.; McCabe, F. L.; Faucette, L. F.;
Johnson, R. K.; Hertzberg, R. P. Synthesis of water-soluble (ami-
noalkyl)camptothecin analogues: inhibition of topoisomerase I and
antitumor activity. J. Med. Chem. 1991, 34, 98–107.
administration of natural product DNA topoisomerase I inhibitors 10-
hydroxycamptothecin and camptothecin in SCID mice bearing human
breast cancer xenografts. Int. J. Oncol. 1997, 10, 1147–1156.
(25) Zhou, J.-J.; Liu, J.; Xu, B. Relationship between lactone ring forms
of HCPT and their antitumor activities. Acta Pharmacol. Sin. 2001,
22, 827–830.
(26) Mattern, M. R.; Hofmann, G. A.; Polsky, R. M.; Funk, L. R.; McCabe,
F. L.; Johnson, R. K. In vitro and in vivo effects of clinically important
camptothecin analogues on multidrug-resistant cells. Oncol. Res. 1993,
5, 467–474.
(4) Sawada, S.; Okajima, S.; Aiyama, R.; Nokata, K.; Furuta, T.;
Yokokura, T.; Sugino, E.; Yamaguchi, K.; Miyasaka, T. Synthesis
and antitumor activity of 20(S)-camptothecin derivatives: carbamate-
linked, water-soluble derivatives of 7-ethyl-10-hydroxycamptothecin.
Chem. Pharm. Bull. 1991, 39, 1446–1450.
(5) Wani, M. C.; Wall, M. E. Plant antitumor agents. II. The structure of
two new alkaloids from Camptotheca acuminata. J. Org. Chem. 1969,
34, 1364–1367.
(27) Wang, Y.; Li, L.; Jiang, W.; Larrick, J. W. Synthesis and evaluation
of a DHA and 10-hydroxycamptothecin conjugate. Bioorg. Med. Chem.
2005, 13, 5592–5599.
(28) Takayama, H.; Watanabe, A.; Hosokawa, M.; Chiba, K.; Satoh, T.;
Aimi, N. Synthesis of a new class of camptothecin derivatives, the
long-chain fatty acid esters of 10-hydroxycamptothecin, as a potent
prodrug candidate, and their in vitro metabolic conversion by
carboxylesterases. Bioorg. Med. Chem. Lett. 1998, 8, 415–418.
(29) Greenwald, R. B.; Pendri, A.; Choe, Y. H. Acyl Polymeric Derivatives
of Aromatic Hydroxyl-Containing Compounds. U.S. Patent 6,011,042,
2000.
(6) Wani, M. C.; Ronman, P. E.; Lindley, J. T.; Wall, M. E. Plant antitumor
agents. 18. Synthesis and biological activity of camptothecin analogues.
J. Med. Chem. 1980, 23, 554–560.
(7) Kawato, Y.; Aonuma, M.; Hirota, Y. Intracellular roles of SN-38, a
metabolite of the camptothecin derivative CPT-11, in the antitumor
effect of CPT-11. Cancer Res. 1991, 51, 4187–4191.
(8) Ahmed, F.; Vyas, V.; Cornfield, A.; Goodin, S.; Ravikumar, T. S.;
Rubin, E. H.; Gupta, E. In vitro activation of irinotecan to SN-38 by
human liver and intestine. Anticancer Res. 1999, 19, 2067–2071.
(9) Leu, Y.-L.; Roffler, S. R.; Chern, J.-W. Design and synthesis of water-
soluble glucuronide derivatives of camptothecin for cancer prodrug
monotherapy and antibody-directed enzyme prodrug therapy (ADEPT).
J. Med. Chem. 1999, 42, 3623–3628.
(10) Paigen, K. Mammalian beta-glucuronidase: genetics, molecular biol-
ogy, and cell biology. Prog. Nucleic Acid Res. Mol. Biol. 1989, 37,
155–205.
(11) Bosslet, K.; Straub, R.; Blumrich, M.; Czech, J.; Gerken, M.; Sperker,
B.; Kroemer, H. K.; Gesson, J.-P.; Koch, M.; Monneret, C. Elucidation
of the mechanism enabling tumor selective prodrug monotherapy.
Cancer Res. 1998, 58, 1195–1201.
(12) Bosslet, K.; Czech, J.; Hoffmann, D. A novel one-step tumor-selective
prodrug activation system. Tumor Targeting 1995, 1, 45–50.
(13) Chen, X.; Wu, B.; Wang, P. G. Glucuronides in anti-cancer therapy.
Curr. Med. Chem.: Anti-Cancer Agents 2003, 3, 139–150.
(14) Alaoui, A. E.; Saha, N.; Schmidt, F.; Monneret, C.; Florent, J.-C. New
Taxol (paclitaxel) prodrugs designed for ADEPT and PMT strategies
in cancer chemotherapy. Bioorg. Med. Chem. 2006, 14, 5012–5019.
(15) Skwarczynski, M.; Hayashi, Y.; Kiso, Y. Paclitaxel prodrugs: toward
smarter delivery of anticancer agents. J. Med. Chem. 2006, 49, 7253–
7269.
(16) Wei, G.; Loktionova, N. A.; Pegg, A. E.; Moschel, R. C. ꢀ-Glucu-
ronidase-cleavable prodrugs of O6-benzylguanine and O6-benzyl-2′-
deoxyguanosine. J. Med. Chem. 2005, 48, 256–261.
(17) Prijovich, Z. M.; Chen, B.-M.; Leu, Y.-L.; Chern, J.-W.; Roffler, S. R.
Anti-tumour activity and toxicity of the new prodrug 9-aminocamp-
tothecin glucuronide (9ACG) in mice. Br. J. Cancer 2002, 86, 1634–
1638.
(18) Prijovich, Z. M.; Leu, Y.-L.; Roffler, S. R. Stability of the new prodrug
9-aminocamptothecin glucuronide (9ACG) in the presence of human
serum albumin. Biochem. Pharmacol. 2003, 66, 1181–1187.
(19) Prijovich, Z. M.; Leu, Y.-L.; Roffler, S. R. Effect of pH and human
serum albumin on the cytotoxicity of a glucuronide prodrug of
9-aminocamptothecin. Cancer Chemother. Pharmacol. 2007, 60, 7–
17.
(20) Chen, K.-C.; Cheng, T.-L.; Leu, Y.-L.; Prijovich, Z. M.; Chuang, C.-
H.; Chen, B.-M.; Roffler, S. R. Membrane-localized activation of
glucuronide prodrugs by ꢀ-glucuronidase enzymes. Cancer Gene Ther.
2007, 14, 187–200.
(21) Wall, M. E.; Wani, M. C.; Nicholas, A. W.; Manikumar, G.; Tele,
C.; Moore, L.; Truesdale, A.; Leitner, P.; Besterman, J. M. Plant
antitumor agents. 30. Synthesis and structure activity of novel
camptothecin analogs. J. Med. Chem. 1993, 36, 2689–2700.
(22) Ling, Y. H.; Andersson, B. S.; Nelson, J. A. DNA topoisomerase I as
a site of action for 10-hydroxycamptothecin in human promyelocytic
leukemia cells. Cancer Biochem. Biophys. 1990, 11, 23–30.
(23) Han, R. Highlight on the studies of anticancer drugs derived from
plants in China. Stem Cells 1994, 12, 53–63.
(30) He, X.; Lu, W.; Jiang, X.; Cai, J.; Zhang, X.; Ding, J. Synthesis and
biological evaluation of bis and monocarbonate prodrugs of 10-
hydroxycamptothecins. Bioorg. Med. Chem. 2004, 12, 4003–4008.
(31) Toki, B. E.; Cerveny, C. G.; Wahl, A. F.; Senter, P. D. Protease-
mediated fragmentation of p-amidobenzyl ethers: a new strategy for
the activation of anticancer prodrugs. J. Org. Chem. 2002, 67, 1866–
1872.
(32) Haisma, H. J.; van Muijen, M.; Pinedo, H. M.; Boven, E. Comparison
of two anthracycline-based prodrugs for activation by a monoclonal
antibody-ꢀ-glucuronidase conjugate in the specific treatment of
cancer. Cell Biophys. 1994, 24–25, 185–192.
(33) Florent, J. C.; Dong, X.; Gaudel, G.; Mitaku, S.; Monneret, C.; Gesson,
J. P.; Jacquesy, J. C.; Mondon, M.; Renoux, B.; Andrianomenjanahary,
S.; Michel, S.; Koch, M.; Tillequin, F.; Gerken, M.; Czech, J.; Straub,
R.; Bosslet, K. Prodrugs of anthracyclines for use in antibody-directed
enzyme prodrug therapy. J. Med. Chem. 1998, 41, 3572–3581.
(34) Mitsunobu, O. The use of diethyl azodicarboxylate and triphenylphos-
phine in synthesis and transformation of natural products. Synthesis
1981, 1–28.
(35) Duimstra, J. A.; Femia, F. J.; Meade, T. J. A gadolinium chelate for
detection of ꢀ-glucuronidase: a self-immolative approach. J. Am. Chem.
Soc. 2005, 127, 12847–12855.
(36) Tosin, M.; Murphy, P. V. Synthesis of R-glucuronic acid and amide
derivatives in the presence of a participating 2-acyl protecting group.
Org. Lett. 2002, 4, 3675–3678.
(37) Fassberg, J.; Stella, V. J. A kinetic and mechanistic study of the
hydrolysis of camptothecin and some analogues. J. Pharm. Sci. 1992,
81, 676–684.
(38) De Graaf, M.; Boven, E.; Scheeren, H. W.; Haisma, H. J.; Pinedo,
H. M. Beta-glucuronidase-mediated drug release. Curr. Pharm. Des.
2002, 8, 1391–1403.
(39) Wang, R.; Gao, Y.; Lai, L. Calculating partition coefficient by atom-
additive method. Perspect. Drug DiscoVery Des. 2000, 19, 47–66.
(40) Angenault, S.; Thirot, S.; Schmidt, F.; Monneret, C.; Pfeiffer, B.;
Renard, P. Cancer chemotherapy: a SN-38 (7-ethyl-10-hydroxycamp-
tothecin) glucuronide prodrug for treatment by a PMT (prodrug
monotherapy) strategy. Bioorg. Med. Chem. Lett. 2003, 13, 947–950.
(41) Jain, S.; Drendel, W. B.; Chen, Z. W.; Mathews, F. S.; Sly, W. S.;
Grubb, J. H. Structure of human ꢀ-glucuronidase reveals candidate
lysosomal targeting and active-site motifs. Nat. Struct. Biol. 1996, 3,
375–381.
(42) Katu, K.; Yoshida, K.; Tsukannto, H. Synthesis of p-nitrophenyl ꢀ-D-
glucopyranosiduronic acid and its utilization as a substrate for the assay
of ꢀ-glucuronidase activity. Chem. Pharm. Bull. 1960, 8, 239–242.
(43) Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R.
Development and validation of a genetic algorithm for flexible docking.
J. Mol. Biol. 1997, 267, 727–748.
(44) Islam, M. R.; Tomatsu, S.; Shah, G. N.; Grubb, J. H.; Jain, S.; Sly,
W. S. Active site residues of human ꢀ-glucuronidase. Evidence for
Glu540 as the nucleophile and Glu451 as the acid-base residue. J. Biol.
Chem. 1999, 274, 23451–23455.
(45) Geddie, M. L.; Matsumura, I. Rapid evolution of ꢀ-glucuronidase
specificity by saturation mutagenesis of an active site loop. J. Biol.
Chem. 2004, 279, 26462–26468.
(24) Zhang, R.; Cai, Q.; Lindsey, J. R.; Li, Y.; Chambless, B.; Naguib,
F. N. M. Antitumor activity and pharmacokinetics following oral
JM701151C