Liver-targeted prodrugs of 2′-C-methyladenosine for therapy of hepatitis C virus infection

Article abstract of DOI:10.1021/jm0701021

2′-C-Methyladenosine exhibits impressive inhibitory activity in the cell-based hepatitis C virus (HCV) subgenomic replicon assay, by virtue of intracellular conversion to the corresponding nucleoside triphosphate (NTP) and inhibition of NS5B RNA-dependent

Full text of DOI:10.1021/jm0701021

J. Med. Chem. 2007, 50, 3891-3896
3891
Liver-Targeted Prodrugs of 2′-C-Methyladenosine for Therapy of Hepatitis C Virus Infection
Scott J. Hecker,*,† K. Raja Reddy,† Paul D. van Poelje,† Zhili Sun,† Wenjian Huang,† Vaibhav Varkhedkar,† M. Venkat Reddy,†
James M. Fujitaki,† David B. Olsen,‡ Kenneth A. Koeplinger,‡ Serge H. Boyer,† David L. Linemeyer,† Malcolm MacCoss,§ and
Mark D. Erion†
Departments of Biochemistry and Medicinal Chemistry, Metabasis Therapeutics, Inc., 11119 North Torrey Pines Road,
La Jolla, California 92037, Departments of AntiViral Research and Drug Metabolism, Merck Research Laboratories,
West Point, PennsylVania 19486, and Department of Chemistry, Merck Research Laboratories, Rahway, New Jersey 07065
ReceiVed January 25, 2007
2′-C-Methyladenosine exhibits impressive inhibitory activity in the cell-based hepatitis C virus (HCV)
subgenomic replicon assay, by virtue of intracellular conversion to the corresponding nucleoside triphosphate
(NTP) and inhibition of NS5B RNA-dependent RNA polymerase (RdRp). However, rapid degradation by
adenosine deaminase (ADA) limits its overall therapeutic potential. To reduce ADA-mediated deamination,
we prepared cyclic 1-aryl-1,3-propanyl prodrugs of the corresponding nucleoside monophosphate (NMP),
anticipating cytochrome P450 3A-mediated oxidative cleavage to the NMP in hepatocytes. Lead compounds
identified in a primary rat hepatocyte screen were shown to result in liver levels of NTP predictive of
efficacy after intravenous dosing to rats. The oral bioavailability of the initial lead was below 5%; therefore,
additional analogues were synthesized and screened for liver NTP levels after oral administration to rats.
Addition of a 2′,3′-carbonate prodrug moiety proved to be a successful strategy, and the 1-(4-pyridyl)-1,3-
propanyl prodrug containing a 2′,3′-carbonate moiety displayed oral bioavailability of 39%.
Introduction
Hepatitis C virus (HCV)a infection is a serious worldwide
health problem that according to the World Health Organization
affects 2% of the world population.1 While the early stages of
the disease are typically asymptomatic, the majority of HCV
infections progress to chronic infection, which is associated with
an increased risk of cirrhosis, hepatocellular carcinoma, and liver
failure. At present, the only therapy available for patients with
Figure 1. Conversion of 2′-C-methyladenosine (1) to 2′-C-methyl-
inosine (2) by ADA.
chronic HCV infection is a combination of ribavirin and
interferon-based therapies, which leads to a sustained virologic
response in only about half the patients treated.2
Efforts to discover more effective drugs to treat HCV-infected
patients have focused on several possible targets, including the
NS5B RNA-dependent RNA polymerase (RdRp).3,4 A number
of nucleoside analogues are reported to inhibit RdRp (as the
corresponding triphosphate) in enzyme and cell-based assays,
and particularly nucleoside analogues containing the 2′-C-
methylribose sugar.5 Within this group, 2′-C-methyladenosine
(1, Figure 1) has shown impressive activity in in vitro systems
such as the cell-based HCV subgenomic replicon assay (EC50
) 0.3 µM).6 However, pharmacokinetic studies have shown very
rapid clearance of 1 upon intravenous dosing in rats, and further
studies to investigate the reason revealed that 1 is an efficient
substrate for adenosine deaminase (ADA).6-8
Figure 2. Mechanism of activation of cyclic 1-aryl-1,3-propanyl
(HepDirect) prodrugs.
phate form (NMP) specifically to cells expressing CYP 3A such
as hepatocytes (Figure 2).9,10 Oxidation of the prodrug moiety
by CYP 3A yields a hemiketal, which spontaneously isomerizes
to a ring-opened form and undergoes â-elimination to yield the
NMP. This delivery approach bypasses the initial nucleoside
phosphorylation step known to be rate-limiting in the conversion
of many nucleosides to their active NTP form. By selectively
enhancing NTP generation in the liver, prodrugs of this type
may offer an improved therapeutic index, particularly if the
metabolic fate of the NTP involves a path other than conversion
back to nucleoside and release from the liver into the blood-
stream.
The challenge of circumventing metabolic degradation and
achieving delivery of the intact nucleoside to the liver is
particularly well-suited to a prodrug approach previously
reported from these laboratories, which delivers the monophos-
* To whom correspondence should be addressed: phone 858-622-3910;
fax 858-622-5583; e-mail hecker@mbasis.com.
† Metabasis Therapeutics, Inc.
‡ Departments of Antiviral Research and Drug Metabolism, Merck
Research Laboratories.
§ Department of Chemistry, Merck Research Laboratories.
a Abbreviations: HCV, hepatitis C virus; RdRp, RNA-dependent RNA
polymerase; ADA, adenosine deaminase; NTP, nucleoside triphosphate;
NMP, nucleoside monophosphate; DMSO, dimethyl sulfoxide; LC-MS/
MS, liquid chromatography with tandem mass spectrometry.
It has been reported that ADA requires a free 5′-OH group
for processing of substrates,11 so it would be expected that
1-aryl-1,3-propanyl 5′-monophosphate prodrugs would avoid
metabolic degradation by this enzyme. Armed with this
10.1021/jm0701021 CCC: $37.00 © 2007 American Chemical Society
Published on Web 07/18/2007

3892 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 16
Hecker et al.
Figure 3. Synthesis of 1-aryl-1,3-propanyl monophosphate prodrugs of 1.
Table 1. Yield and Purity Data for Compounds 6a-s, 9a-c, 9f, 9g, 9i,
and 9p
rationale, a program was initiated whose objective was to
discover a prodrug of 2′-C-methyladenosine that would achieve
therapeutic concentrations of nucleoside triphosphate (NTP) in
the liver upon oral administration.
yield (%)
yield (%)
compd
3 f 6
6 f 9
HPLC area (%)
MS
6a
6b
6c
6d
6e
6f
6g
6h
6i
35
25
30
8
94.9
95.4
93.8
96.0
96.8
93.4
93.4
95.1
93.9
93.3
94.0
94.8
93.8
90.1
99.5
94.4
97.4
95.4
94.6
96.0
95.0
94.3
97.8
92.2
98.0
95.6
512.0
546.0
546.0
558.0
556.0
496.0
514.2
514.0
514.0
530.0
573.8
503.1
503.0
479.0
479.0
479.0
557.0
509.1
493.1
538.0
572.0
572.0
522.0
540.2
540.0
505.4
Results and Discussion
Synthesis of Prodrugs. The synthesis of prodrugs of 2′-C-
methyladenosine was carried out as depicted in Figure 3. The
nucleoside was protected as its 2′,3′-isopropylidene ketal 3 and
was then phosphorylated at 5′ with trans-p-nitrophenylphosphate
reagents 4a-s.9,12 Removal of the protecting group from
compounds 5a-s afforded the desired cis-prodrugs 6a-s. Yields
(unoptimized) for the two-step conversion of 3 to 6a-s were
generally 25-60%, although in a few cases lower yields were
obtained. Previous work had shown that an electron-withdrawing
group on the phenyl ring is important for sufficient chemical
stability, so only analogues of this type were prepared.9 In most
cases, the phosphorylating reagents 4a-s were prepared in
racemic form, and therefore the prodrugs 6a-s are a mixture
of two cis-diastereomers; a single cis-diastereomer was prepared
in the cases of 3-chlorophenyl prodrug 6a and 4-pyridyl prodrug
6p, since the corresponding phosphorylation reagents 4a and
4p were available as single enantiomers from other projects.10,13
Yield and purity data, as well as results of mass spectrometric
(MS) analysis confirming the structures, are shown in Table 1.
Activation in Rat Hepatocytes. Activation of prodrugs 6a-s
to the corresponding NTPs, which requires prodrug cleavage
as well as two phosphorylation events catalyzed by cellular
nucleotide kinases, was evaluated in rat hepatocytes (Table 2).
Many prodrugs were efficiently cleaved and phosphorylated,
affording levels of NTP comparable to those derived from the
unmodified nucleoside 1. The highest activation levels were seen
with halo-substituted phenyl prodrugs, with the highest level
observed with dichlorophenyl prodrug 6c. Pyridyl prodrugs
6n-s were also activated but gave lower levels of NTP. The
beneficial effect of halogen substitution was noted in the pyridyl
series as well, as seen with 5-bromo-3-pyridyl derivative 6q.
Evaluation of Liver NTP Levels in Rat. On the basis of
the intracellular concentration of NTP in the replicon assay at
the EC90, we sought to achieve a concentration of NTP in the
liver of 30 nmol/g.14 Levels of NTP generated in the liver
following administration of 1 and its prodrug 6a were evaluated
in the rat (Figure 4); doses of prodrugs were normalized to
nucleoside equivalents (ne), that is, the quantity of nucleoside
contained. Oral dosing of 1 resulted in negligible levels of the
corresponding NTP in liver, whereas intravenous dosing af-
forded modest exposure at the level of 10 nmol/g. Importantly,
8
20
61
39
25
41
30
35
66
7
30
48
40
12
40
6j
6k
6l
6m
6n
6o
6p
6q
6r
6s
9a
9b
9c
9f
72
76
45
45
95
80
79
9g
9i
9p
intravenous administration of prodrug 6a achieved liver levels
of NTP of up to 50 nmol/g, with sustained exposure above our
target of 30 nmol/g for 12 h.
Next, prodrug 6a was evaluated orally at a dose 5-fold higher
than used intravenously (iv). Liver levels of NTP were below
10 nmol/g; the oral bioavailability of 6a based on the AUC of
NTP in liver following oral versus intravenous dosing was less
than 5%.
In response to the observation of low oral bioavailability,
selected prodrugs were screened in the rat for liver levels of
NTP at the 3 h time point following an oral dose of 50 mg/kg.
None of the prodrugs came close to achieving target levels of
NTP in liver, as shown in Table 4. The low liver NTP levels
for these compounds are likely due to limited oral absorption,
resulting from their high molecular weight and multitude of polar
functional groups.
Optimization of Oral Bioavailability. Encouraged by the
results of iv dosing of 6a, we embarked on a program to further
functionalize this compound in order to achieve better oral
absorption. One approach used to improve the oral bioavail-
ability of nucleoside drugs is attachment of an amino acid

LiVer-Targeted Prodrugs of 2′-C-Methyladenosine
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 16 3893
Figure 4. Rat liver levels of 2-C-methyladenosine triphosphate after
dosing with nucleoside 1 and prodrug 6a.
Table 2. Activation of Prodrugs 6a-s in Rat Hepatocytes (25 µM, 2 h)
compd
Ar
[NTP], nmol/g
1
209
389
396
585
122
99
328
323
139
250
328
72
137
41
93
73
47
6a
6b
6c
6d
6e
6f
6g
6h
6i
6j
6k
6l
6m
6n
6o
6p
6q
6r
6s
(S)-3-Cl-phenyl
3,4-Cl2-phenyl
3,5-Cl2-phenyl
2-Br-phenyl
3-Br-phenyl
3-F-phenyl
2,3-F2-phenyl
2,5-F2-phenyl
3,5-F2-phenyl
4-Cl-3-F-phenyl
2-Br-5-F-phenyl
3-CN-phenyl
4-CN-phenyl
2-pyridyl
Figure 5. Initial prodrugs to improve oral bioavailability.
3-pyridyl
(S)-4-pyridyl
5-Br-3-pyridyl
2-OMe-4-pyridyl
5-Me-3-pyridyl
194
93
73
Figure 6. Synthesis of carbonate prodrugs.
Table 3. Activation of Prodrugs 9a-c, 9f, 9g, 9i, and 9p in Rat
Hepatocytes (25 µM, 2 h)
acyl groups to the amine and hydroxy functionalities of the
nucleoside. Carbamate derivatives of the amino group of
cytosine-containing nucleosides have been shown to be ef-
ficiently cleaved in vivo.16 To this end, n-pentyl carbamate 8
(Figure 5) was synthesized; like valine ester 7, it showed good
conversion to the NTP in rat hepatocytes but failed to offer a
corresponding increase in liver levels upon oral administration
to rats.
As we initiated efforts to modify the hydroxy groups of the
sugar moiety, we conceived the use of a 2′,3′-carbonate prodrug,
which would remove two H-bond donors with only a minor
increase in molecular weight. We were gratified to find that
2′,3′-carbonate 9a afforded a 10-fold increase in rat liver NTP
levels upon oral dosing, and efforts were initiated to explore
the generality of application of this functional group to 1-aryl-
1,3-propanyl monophosphate prodrugs of 1.
2′,3′-Carbonate Derivatives of 6a-p. The 2′,3′-carbonate
derivatives 9a-p were prepared by treatment of prodrugs 6a-p
with carbonyl diimidazole in N,N-dimethylformamide (DMF)
(Figure 6); yield and purity data are provided in Table 1. The
activation of these prodrugs to the NTP in rat hepatocytes is
shown in Table 3. In general, the addition of the 2′,3′-carbonate
group does not hamper formation of the NTP. However,
reduction in activation was observed with 3,5-dichlorophenyl
prodrugs 9b and 9c, while improvement in the level of activation
was seen with 4-pyridyl prodrug 9p.
compd
Ar
[NTP], nmol/g
9a
9b
9c
9f
9g
9i
(S)-3-Cl-phenyl
3,4-Cl2-phenyl
3,5-Cl2-phenyl
3-F-phenyl
2,3-F2-phenyl
3,5-F2-phenyl
(S)-4-pyridyl
296
97
49
313
280
113
185
9p
Table 4. Rat Liver NTP Levels at 3 h Following Oral Dosing of
Prodrugs at 50 mg/kg
HepDirect alone
HepDirect 2,3-carbonate
Ar
compd [NTP], nmol/g compd
[NTP], nmol/g
(S)-3-Cl-phenyl
3,4-Cl2-phenyl
3,5-Cl2-phenyl
3-F-phenyl
2,3-F2-phenyl
3,5-F2-phenyl
(S)-4-pyridyl
6a
6b
6c
6f
6g
6i
2.9
2.0
2.4
4.1
8.2
4.1
3.6
9a
9b
9c
9f
9g
9i
29.2
11.8
5.0
22.9
14.7
11.2
28.3
6p
9p
residue, such as valine, in order to facilitate uptake by amino
acid transporters.15 Thus, the 3′-valine ester 7 (Figure 5) was
prepared; it afforded reasonable activation to NTP in hepatocytes
but did not lead to an increase in liver NTP levels (relative to
6a) following oral administration.
A second approach for improving oral absorption surrounded
reducing the number of hydrogen-bond donors by appending
Application of the 2′,3′-carbonate in all cases resulted in a
marked improvement in oral bioavailability, although the degree

3894 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 16
Hecker et al.
prodrug, the 2′,3′-carbonate, affords a dramatic increase in oral
bioavailability. Dual prodrug 9p displays properties warranting
its further study as a potential new agent for treating HCV.
Experimental Section
General Information. Glassware for moisture-sensitive reactions
was flame-dried and cooled to room temperature in a desiccator,
and all reactions were carried out under an atmosphere of nitrogen.
Anhydrous solvents were purchased from Aldrich or Acros. Thin-
layer chromatography was performed on EM Science silica gel 60
F254 plates and visualized with a UV lamp (254 nm). Column
chromatography was performed on 230-400 mesh EM Science
silica gel 60. 1H and 13C NMR were obtained on a Varian Mercury-
300 operating at 300 and 75 MHz, respectively. 1H and 13C NMR
spectra were recorded in δ units with the solvent’s chemical shift
as the reference line. C, H, N microanalyses were performed by
Robertson Microlit Laboratories, Inc., Madison, NJ, or NuMega
Resonance Labs, Inc. San Diego, CA. Purity was assessed on a
Hitachi HPLC instrument (L-6000 pump, L-4000 UV detector) with
a YMC-ODS-AQ column (250 × 4.6 mm); gradient elution was
employed, with 0.05% trifluoroacetic acid/water and methanol at
a flow rate of 1.0 mL/min (detector wavelength 280 nm). Mass
spectral data were acquired on a PE Sciex API-2000 LC/MS
instrument, operating in the positive ion mode. Centrifugations were
performed in an Eppendorf microcentrifuge at 14 000 rpm in a cold
room. All protocols involving animal experimentation were re-
viewed and approved by the Metabasis Therapeutics IACUC
(Institution Animal Care and Use Committee) and follow the
guidelines established by the NRC Guide for the Care and Use of
Laboratory Animals.
Figure 7. Liver levels of 2′-C-Me-ATP after dosing of rats with
prodrug 9p.
Figure 8. Stability of compound 9p in rat plasma at 37 °C.
cis-5-O-[4-(S)-(3-Chlorophenyl)-2-oxo-1,3,2-dioxaphosphori-
nan-2-yl]-2′-C-methyladenosine, Trifluoroacetic Acid Salt
(6a): (A) Phosphorylation. To a solution of 2′-C-methyladenosine-
2′,3′-acetonide20 (3, 1.04 g, 3.24 mmol) in DMF (20 mL), being
stirred under nitrogen at 0 °C, was added a 1 M solution of
t-butylmagnesium chloride in tetrahydrofuran (THF) (5.50 mL, 5.50
mmol). After 20 min, trans-4-[(S)-3-chlorophenyl]-2-(4-nitro-
phenoxy)-2-oxo-1,3,2-dioxaphosphorinane9 (4a, 1.41 g, 3.87 mmol)
was added. The mixture was allowed to warm to room temperature
and was stirred for 3 h. The solvent was removed with a rotary
evaporator, and the residue was subjected to column chromatog-
raphy on silica gel (5-8% methanol in dichloromethane), affording
1.00 g (1.82 mmol, 56%) of compound 5a.
of enhancement varied (Table 4). The highest liver levels of
NTP were observed with 3-chlorophenyl prodrug 9a and
4-pyridyl prodrug 9p. Further profiling of these two compounds
showed that 9p had 3-fold higher aqueous solubility (0.45 vs
0.16 mg/mL in pH 5.1 citrate buffer), and therefore 9p was
selected for a more extensive pharmacokinetic evaluation.
Liver levels of NTP following intravenous (10 mg/kg) and
oral (50 mg/kg) dosing of dual prodrug 9p in rats are shown in
Figure 7. In this experiment, absorption appears to have been
delayed relative to the observations in the earlier screening assay,
as the NTP level at the 3 h time point is only ∼5 nmol/g versus
28.3 nmol/g in the screening assay. This difference cannot be
attributed to formulation, since the same vehicle (PEG-400) was
used in both experiments. The maximum liver NTP concentra-
tion was achieved at a surprisingly late time point of 16 h,17 an
observation that may be attributable in part to the fact that food
was returned to the fasted rats a few hours after dosing. Most
importantly, the target concentration of 30 nmol/g is achieved
during the period of 8-16 h. The oral bioavailability of 9p,
calculated by comparing AUCs of NTP in liver following oral
and intravenous dosing, is 39%.
Compound 9p shows adequate chemical stability, with a half-
life of 13 h in pH 7 buffer at 25 °C. In rat plasma at 37 °C, 9p
is efficiently converted to des-carbonate 6p (Figure 8), indicating
that cleavage is primarily enzyme-mediated. Compound 6p
undergoes no further degradation under these conditions.
Although we have not performed studies to identify the enzyme
responsible for hydrolysis of the 2′,3′-carbonate moiety, the
enzyme paraoxonase has been implicated in the hydrolysis of
cyclic carbonates.18,19
(B) Deprotection. Compound 5a (1.00 g, 1.82 mmol) was added
to 70% aqueous trifluoroacetic acid at 0 °C. After being stirred
overnight at 0 °C, the mixture was allowed to warm to room
temperature and was stirred for 4 h. The volatiles were removed
with a rotary evaporator, and the residue was partitioned between
20% methanol in ethyl acetate and saturated aqueous sodium
bicarbonate solution. The aqueous layer was extracted three times
with 20% methanol in ethyl acetate, and the combined organic phase
was washed with brine, dried over sodium sulfate, and filtered; the
solvent was removed with a rotary evaporator. The crude product
was subjected to column chromatography on silica gel (5-10%
methanol in dichloromethane), affording 703 mg (62%) of com-
1
pound 6a as an off-white solid. H NMR (DMSO-d6) δ 8.24 (s,
1H), 8.16 (s, 1H), 7.45 (s, 2H), 7.3-7.4 (m, 4H), 6.01 (s, 1H),
5.73 (m, 1H), 5.55 (d, J ) 10 Hz, 1H), 5.40 (s, 1H), 4.4-4.6 (m,
4H), 4.1-4.25 (m, 2H), 2.19 (m, 2H), 0.82 (s, 3H). Anal. (following
repurification to remove residual TFA; C20H23ClN5O7P) C, H, N.
cis-5′-O-[4-(S)-(Pyridin-4-yl)-2-oxo-1,3,2-dioxaphosphorinan-
2-yl]-2′-C-methyladenosine (6p): (A) Phosphorylation. To a
solution of 2′-C-methyladenosine-2′,3′-acetonide (3, 171 mg, 0.530
mmol) in DMF (5 mL), being stirred under nitrogen, was added a
1 M solution of t-butylmagnesium chloride in THF (1.00 mL, 1.00
mmol). After 30 min, a solution of trans-4-[(S)-pyridin-4-yl]-2-(4-
nitrophenoxy)-2-oxo-1,3,2-dioxaphosphorinane9 (4p, 220 mg, 0.65
mmol) was added in one portion. After being stirred at room
temperature overnight, the mixture was concentrated with a rotary
evaporator, and the residue was subjected to column chromatog-
Conclusions
Application of the prodrug approach described herein suc-
cessfully circumvents the rapid metabolism of 2′-C-methylad-
enosine, achieving target concentrations of the triphosphate in
the liver following intravenous dosing. The use of an additional

LiVer-Targeted Prodrugs of 2′-C-Methyladenosine
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 16 3895
raphy on silica gel (5-10% methanol in dichloromethane), affording
168 mg (0.323 mmol, 61%) of compound 5p.
was reconstituted with 100 µL of mobile phase and then analyzed
for nucleotides by an LC-MS/MS method as described below.
The reconstituted extracts in mobile phase A (20 mM N,N-
dimethylhexylamine and 10 mM propionic acid in 20% methanol)
were analyzed by LC-MS/MS (Applied Biosystems, API 4000)
equipped with an Agilent 1100 binary pump and a LEAP injector.
Ten microliters of sample was injected onto an Xterra MS C18
column (3.5 um, 2.1 × 50 mm, Waters Corp.) with a SecurityGuard
C18 guard column (5 µm, 4.0 × 3.0 mm, Phenomenex) and eluted
with a gradient of mobile phases A and B (20 mM N,N-
dimethylhexylamine and 10 mM propionic acid in 80% methanol)
at a flow rate of 0.3 mL/min (0 min, 0% B, 0-1 min, 0-50% B;
1-3 min, 50-100% B, 3-6 min, 100% B; 6-6.1 min, 100-0%
B; 6.1-9 min, 0% B). NTP was detected by using MS/MS mode
(M-/78.8). The quantitative analysis of liver NTP was calculated
on the basis of a calibration curve generated with an authentic
standard of 2′-C-methyladenosine triphosphate (0.01, 0.03, 0.1, 0.3,
1.0, 3, 10, and 30 µM).
Evaluation of Stability in Blood and Plasma. 2′-C-Methyl-
adenosine and prodrugs thereof were separately incubated in
heparinized whole rat blood or plasma at 37 °C. Aliquots of the
blood and plasma samples were removed periodically and extracted
with perchloric acid and 0.2% formic acid in acetonitrile, respec-
tively, and then centrifuged (22000g, 20 min, 4 °C). The acidic
supernatants were neutralized with potassium carbonate and
centrifuged again (22000g, 20 min, 4 °C). The neutralized
supernatants and acetonitrile extracts were then analyzed for the
major metabolite of 2′-C-methyladenosine, that is, 2′-C-methyl-
inosine, as described below.
(B) Deprotection. Compound 5p (168 mg, 0.323 mmol) was
added to 70% aqueous trifluoroacetic acid at 0 °C. After being
stirred overnight at 0 °C, the mixture was concentrated with a rotary
evaporator, and the residue was subjected twice to column
chromatography on silica gel (5-12.5% methanol in dichlo-
romethane containing 1% ammonium hydroxide, then 8-10%
methanol in acetonitrile containing 1% ammonium hydroxide),
affording 120 mg (78%) of compound 6p as an off-white solid. 1H
NMR (DMSO-d6) δ 8.55 (d, J ) 4 Hz, 2H), 8.25 (s, 1H), 8.20 (s,
1H), 7.36 (m, 4H), 6.02 (s, 1H), 5.74 (m, 1H), 5.51 (d, J ) 10 Hz,
1H), 5.40 (s, 1H), 4.4-4.6 (m, 4H), 4.1-4.3 (m, 2H), 2.0-2.3 (m,
2H), 0.83 (s, 3H). Anal. (C19H23N6O7P‚2.5H2O) C, H, N.
cis-5′-O-[4-(S)-(3-Chlorophenyl)-2-oxo-1,3,2-dioxaphosphori-
nan-2-yl]-2′-C-â-methyladenosine-2′,3′-carbonate (9a). To a solu-
tion of compound 6a (0.80 g, 1.56 mmol) in THF (15 mL), being
stirred at room temperature, was added 1,1′-carbonyldiimidazole
(0.70 g, 4.32 mmol). After 3 h, the solvent was removed with a
rotary evaporator, and the residue was subjected to column
chromatography on silica gel (5-6% methanol in dichloromethane),
1
affording 0.60 g (72%) of compound 9a as an off-white solid. H
NMR (DMSO-d6) δ 8.20 (s, 1H), 8.16 (s, 1H), 7.3-7.5 (m, 6H),
6.59 (s, 1H), 5.72 (m, 1H), 5.08, (d, J ) 4 Hz, 1H), 4.35-4.65 (m,
5H), 2.1-2.3 (m, 2H), 1.25 (s, 3H). Anal. (C21H21N5O8ClP‚1.0H2O)
C, H, N.
cis-5′-O-[4-(S)-(Pyridin-4-yl)-2-oxo-1,3,2-dioxaphosphorinan-
2-yl]-2′-C-methyladenosine-2′,3′-carbonate (9p). To a solution of
compound 6p (600 mg, 1.25 mmol) in DMF (10 mL), being stirred
at 0 °C, was added 1,1′-carbonyldiimidazole (407 mg, 2.51 mmol).
The mixture was allowed to warm to room temperature and was
stirred for 3 h. The solvent was removed with a rotary evaporator,
and the residue was subjected to column chromatography on silica
gel (0-10% methanol in dichloromethane), affording 513 mg (79%)
of compound 9p as an off-white solid. 1H NMR (DMSO-d6) δ 8.57
(d, J ) 4 Hz, 2H), 8.22 (s, 1H), 8.16 (s, 1H), 7.44 (s, 2H), 7.37 (d,
J ) 4 Hz, 2H), 6.59 (s, 1H), 5.76 (m, 1H), 5.08 (d, J ) 4 Hz, 1H),
4.35-4.65 (m, 5H), 2.1-2.3 (m, 2H), 1.25 (s, 3H). Anal.
(C20H21N6O8P‚1.2H2O) C, H, N.
The plasma extracts were analyzed by HPLC on an Agilent 1100
instrument. Analysis (50 µL) was performed on an Agilent Zorbax
SB-Aq column (4.6 × 150 mm) eluted with a gradient consisting
of a mixture of buffer A (20 mM potassium phosphate, pH 6.2)
and buffer B (acetonitrile) (0-10 min, 0-10% buffer B; 10-20
min, 10-80%, 20-21 min, 80-0%; 21-30 min, 0%) and UV
absorbance monitoring at 265 nm. The flow rate was 1.5 mL/min
and the column temperature was set at 40 °C. Concentrations of
metabolite were determined from calibration curves prepared by
spiking known amounts of standards to plasma and processing as
before. The LOQ of 2′-C-methylinosine was 1 µM.
NTP Generation in Rat Hepatocytes. Hepatocytes were
prepared from fed Sprague-Dawley rats (250-300 g) according
to the procedure of Berry and Friend21 as modified by Groen et
al.22 Hepatocytes (20 mg/mL wet weight, >85% trypan blue
viability) were incubated at 37 °C in 2 mL of Krebs-bicarbonate
buffer containing 20 mM glucose and 1 mg/mL bovine serum
albumin (BSA) for 2 h in the presence of 1-250 µM nucleoside
or prodrug (from 25 mM stock solutions in DMSO). Following
the incubation, a 1600 µL aliquot of the cell suspension was
centrifuged and 300 µL of acetonitrile was added to the pellet, which
was vortexed and sonicated until the pellet broke down. Then 200
µL of water was added to make a 60% acetonitrile solution. After
10 min of centrifugation at 14 000 rpm, the resulting supernatant
was transferred to a new vial and evaporated to near dryness in a
Savant SpeedVac Plus at room temperature. The dried residue was
reconstituted with 200 µL of water and the mixture was centrifuged
for 10 min at 14 000 rpm. A mixture of 35 µL of supernatant and
35 µL of mobile phase A (20 mM N,N-dimethylhexylamine and
10 mM propionic acid in 20% methanol) was analyzed by LC-
MS/MS (Applied Biosystems, API 4000) equipped with an Agilent
1100 binary pump and a LEAP injector. NTP was detected by using
MS/MS mode (M-/78.8) and quantified on the basis of comparison
to a standard of 2′-C-methyladenosine-5′-triphosphate.
Supporting Information Available: Results from elemental
analysis and NMR spectra. This material is available free of charge
via the Internet at http://pubs.acs.org.
References
(1) Shepard, C. W.; Finelli, L.; Alter, M. J. Global epidemiology of
hepatitis C virus infection. Lancet Infect. Dis. 2005, 5, 558-567.
(2) Hughes, C. A.; Shafran, S. D. Chronic hepatitis C virus manage-
ment: 2000-2005 update. Ann. Pharmacother. 2006, 40, 74-82.
(3) Beaulieu, P. L.; Tsantrizos, Y. S. Inhibitors of the HCV NS5B
polymerase: new hope for the treatment of hepatitis C infections.
Curr. Opin. InVest. Drugs 2004, 5, 838-850.
(4) Wu, J. Z.; Yao, N.; Walker, M.; Hong, Z. Recent advances in
discovery and development of promising therapeutics against hepatitis
C virus NS5B RNA-dependent RNA polymerase. Mini ReV. Med.
Chem. 2005, 5, 1103-1112.
(5) Carroll, S. S.; Olsen, D. B. Nucleoside analog inhibitors of hepatitis
C virus replication. Infect. Disord. Drug Targets 2006, 6, 17-29.
(6) Carroll, S. S.; Tomassini, J. E.; Bosserman, M.; Getty, K.; Stahlhut,
M. W.; Eldrup, A. B.; Bhat, B.; Hall, D.; Simcoe, A. L.; LaFemina,
R.; Rutkowski, C. A.; Wolanski, B.; Yang, Z.; Migliaccio, G.; De
Francesco, R.; Kuo, L. C.; MacCoss, M.; Olsen, D. B. Inhibition of
hepatitis C virus RNA replication by 2′-modified nucleoside analogs.
J. Biol. Chem. 2003, 278, 11979-11984.
(7) Eldrup, A. B.; Allerson, C. R.; Bennett, C. F.; Bera, S.; Bhat, B.;
Bhat, N.; Bosserman, M. R.; Brooks, J.; Burlein, C.; Carroll, S. S.;
Cook, P. D.; Getty, K. L.; MacCoss, M.; McMasters, D. R.; Olsen,
D. B.; Prakash, T. P.; Prhavc, M.; Song, Q.; Tomassini, J. E.; Xia,
J. Structure-activity relationship of purine ribonucleosides for inhibi-
tion of hepatitis C virus RNA-dependent RNA polymerase. J. Med.
Chem. 2004, 47, 2283-2295.
Evaluation of Liver NTP Levels in Rat. Nucleoside analogues
and their prodrugs were administered to Sprague-Dawley rats by
oral gavage. At 2 or 3 h following drug administration, liver samples
(∼1 g) were collected, snap-frozen, and homogenized in 3 volumes
of ice-cold 70% methanol containing 20 mM EDTA/EGTA.
Following centrifugation to clarify the homogenate, aliquots of the
supernatants (100 µL) were evaporated to dryness on a Savant
Speed-Vac Plus (1 h, room temperature). The resulting dried residue
(8) Eldrup, A. B.; Prhavc, M.; Brooks, J.; Bhat, B.; Prakash, T. P.; Song,
Q.; Bera, S.; Bhat, N.; Dande, P.; Cook, P. D.; Bennett, C. F.; Carroll,
S. S.; Ball, R. G.; Bosserman, M.; Burlein, C.; Colwell, L. F.; Fay,
J. F.; Flores, O. A.; Getty, K.; LaFemina, R. L.; Leone, J.; MacCoss,

3896 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 16
Hecker et al.
M.; McMasters, D. R.; Tomassini, J. E.; Von,Langen, D.; Wolanski,
B.; Olsen, D. B. Structure-activity relationship of heterobase-modified
2′-C-methyl ribonucleosides as inhibitors of hepatitis C virus RNA
replication. J. Med. Chem. 2004, 47, 5284-5297.
(16) Shimma, N.; Umeda, I.; Arasaki, M.; Murasaki, C.; Masubuchi, K.;
Kohchi, Y.; Miwa, M.; Ura, M.; Sawada, N.; Tahara, H.; Kuruma,
I.; Horii, I.; Ishitsuka, H. The design and synthesis of a new tumor-
selective fluoropyrimidine carbamate, capecitabine. Bioorg. Med.
Chem. 2000, 8, 1697-1706.
(17) In reponse to this unusual finding, a second study was run in which
levels were assessed at 16 and 24 h, with similar results.
(18) Billecke, S.; Draganov, D.; Counsell, R.; Stetson, P.; Watson, C.;
Hsu, C.; La Du, B. N. Human serum paraoxonase (PON1) isozymes
Q and R hydrolyze lactones and cyclic carbonate esters. Drug Metab.
Dispos. 2000, 28, 1335-1342.
(19) Biggadike, K.; Angell, R. M.; Burgess, C. M.; Farrell, R. M.;
Hancock, A. P.; Harker, A. J.; Irving, W. R.; Ioannou, C.; Procopiou,
P. A.; Shaw, R. E.; Solanke, Y. E.; Singh, O. M.; Snowden, M. A.;
Stubbs, R. J.; Walton, S.; Weston, H. E. Selective plasma hydrolysis
of glucocorticoid γ-lactones and cyclic carbonates by the enzyme
paraoxonase: an ideal plasma inactivation mechanism. J. Med. Chem.
2000, 43, 19-21.
(20) Jenkins, S. R.; Arison, B.; Walton, E. Branched-chain sugar nucleo-
side. IV. 2′-Methyladenosine. J. Org. Chem. 1968, 33, 2490-2494.
(21) Berry, M. N.; Friend, D. S. High-yield preparation of isolated rat
liver parenchymal cells. A biochemical and fine structural study. J.
Cell Biol. 1969, 43, 506-520.
(22) Groen, A. K.; Sips, H. J.; Vervoorn, R. C.; Tager, J. M. Intracellular
compartmentation and control of alanine metabolism in rat liver
parenchymal cells. Eur. J. Biochem. 1982, 122, 87-93.
(9) Erion, M. D.; Reddy, K. R.; Boyer, S. H.; Matelich, M. C.; Gomez-
Galeno, J.; Lemus, R. H.; Ugarkar, B. G.; Colby, T. J.; Schanzer, J.;
Van Poelje, P. D. Design, synthesis, and characterization of a series
of cytochrome P(450) 3A-activated prodrugs (HepDirect prodrugs)
useful for targeting phosph(on)ate-based drugs to the liver. J. Am.
Chem. Soc. 2004, 126, 5154-5163.
(10) Erion, M. D.; van Poelje, P. D.; Mackenna, D. A.; Colby, T. J.;
Montag, A. C.; Fujitaki, J. M.; Linemeyer, D. L.; Bullough, D. A.
Liver-targeted drug delivery using HepDirect prodrugs. J. Pharmacol.
Exp. Ther. 2005, 312, 554-560.
(11) Ikehara, M.; Fukui, T. Nucleosides and nucleotides. LVIII. Deami-
nation of adenosine analogs with calf intestine adenosine deaminase.
Biochim. Biophys. Acta, Gen. Subj. 1974, 338, 512-519.
(12) Reddy, K. R.; Boyer, S. H.; Erion, M. D. Stereoselective synthesis
of nucleoside monophosphate HepDirect prodrugs. Tetrahedron Lett.
2005, 46, 4321-4324.
(13) In this study as in others, the corresponding cis-diastereomer prodrugs
derived from the (R)-diol precursors to 4a and 4p were efficiently
activated in rat hepatocytes, although to a lesser extent (up to 2-fold)
than those derived from the (S)-diol precursors (data not shown).
(14) Unpublished data.
(15) Balimane, P. V.; Sinko, P. J. Involvement of multiple transporters
in the oral absorption of nucleoside analogues. AdV. Drug DeliVery
ReV. 1999, 39, 183-209.
JM0701021