Application of ProTide technology to gemcitabine: A successful approach to overcome the key cancer resistance mechanisms leads to a new agent (NUC-1031) in clinical development

Article abstract of DOI:10.1021/jm401853a

Gemcitabine is a nucleoside analogue commonly used in cancer therapy but with limited efficacy due to a high susceptibility to cancer cell resistance. The addition of a phosphoramidate motif to the gemcitabine can protect it against many of the key cancer resistance mechanisms. We have synthesized a series of gemcitabine phosphoramidate prodrugs and screened for cytostatic activity in a range of different tumor cell lines. Among the synthesized compounds, one in particular (NUC-1031, 6f) was shown to be potent in vitro. Importantly, compared with gemcitabine, 6f activation was significantly less dependent on deoxycytidine kinase and on nucleoside transporters, and it was resistant to cytidine deaminase-mediated degradation. Moreover, 6f showed a significant reduction in tumor volumes in vivo in pancreatic cancer xenografts. The ProTide 6f is now in clinical development with encouraging efficacy signals in a Phase I/II study, which strongly supports the ProTide approach to generate promising new anticancer agents.

Full text of DOI:10.1021/jm401853a

Article
pubs.acs.org/jmc
Application of ProTide Technology to Gemcitabine: A Successful
Approach to Overcome the Key Cancer Resistance Mechanisms
Leads to a New Agent (NUC-1031) in Clinical Development
Magdalena Slusarczyk,† Monica Huerta Lopez,† Jan Balzarini,‡ Malcolm Mason,§ Wen G. Jiang,∥
Sarah Blagden,⊥ Emely Thompson,# Essam Ghazaly,# and Christopher McGuigan*,†
†Cardiff School of Pharmacy & Pharmaceutical Sciences, Cardiff University, King Edward VII Avenue, Cardiff CF10 3NB, United
Kingdom
‡Rega Institute for Medical Research, Katholieke Universiteit Leuven, Minderbroedersstraat 10, Leuven B-3000, Belgium
§Section of Oncology & Palliative Medicine, Cardiff University School of Medicine, Velindre Hospital, Cardiff CF14 2TL, United
Kingdom
∥Metastasis Research Group, University Department of Surgery, Cardiff University, School of Medicine, Heath Park, Cardiff CF14
4XN, United Kingdom
⊥Department of Oncology, Imperial College, Hammersmith Campus, Du Cane Road, London W12 0HS, United Kingdom
#Centre for Haemato-Oncology, Barts Cancer Institute, Queen Mary University of London, Charterhouse Square, London, EC1M
6BQ, United Kingdom
S
* Supporting Information
ABSTRACT: Gemcitabine is a nucleoside analogue com-
monly used in cancer therapy but with limited efficacy due to a
high susceptibility to cancer cell resistance. The addition of a
phosphoramidate motif to the gemcitabine can protect it
against many of the key cancer resistance mechanisms. We
have synthesized a series of gemcitabine phosphoramidate
prodrugs and screened for cytostatic activity in a range of
different tumor cell lines. Among the synthesized compounds,
one in particular (NUC-1031, 6f) was shown to be potent in
vitro. Importantly, compared with gemcitabine, 6f activation was significantly less dependent on deoxycytidine kinase and on
nucleoside transporters, and it was resistant to cytidine deaminase-mediated degradation. Moreover, 6f showed a significant
reduction in tumor volumes in vivo in pancreatic cancer xenografts. The ProTide 6f is now in clinical development with
encouraging efficacy signals in a Phase I/II study, which strongly supports the ProTide approach to generate promising new
anticancer agents.
cytidine during DNA replication, which leads to cell arrest.5
Failure to repair the DNA subsequently triggers apoptosis and
INTRODUCTION
■
Nucleoside analogues and nucleobases constitute a major class
blocks tumor growth. In addition, several self-potentiating
of chemotherapeutic agents and are widely used for the
mechanisms have been reported, such as inhibition of DNA
treatment of patients with cancer. This group of agents, known
synthesis by dFdCTP and of ribonucleotide reductase (RNR)
as antimetabolites, includes a variety of pyrimidine and purine
by dFdCDP.6 Notably all of these mechanisms require
nucleoside derivatives with cytotoxic activity in both hemato-
logical and solid tumors. Gemcitabine1 (2′,2′- difluoro-2′-
phosphorylation of gemcitabine at least to the monophosphate
level.
deoxycytidine) is a pyrimidine nucleoside analogue, shown to
be active against several solid tumor types.2 Following FDA
Both innate and acquired resistance to nucleoside analogues
(such as gemcitabine, 5-fluorouracil, cytarabine, and fludar-
approval in 1996, gemcitabine has become the standard of care
for the treatment of pancreatic cancer.3 More recently, the
abine) is a common problem in the treatment of cancer and is
regarded as a key driver of poor patient survival outcomes.
compound has also gained approval for treating non-small cell
lung, ovarian, bladder, and breast cancer.4
Gemcitabine, like all nucleoside analogues, faces numerous
inherent and acquired cancer resistance mechanisms that
At a molecular level, the mechanisms underpinning the
dramatically limit its effectiveness. These include (i) poor
anticancer effect of gemcitabine rely on the sequential
conversion of the compound into the monophosphate
(dFdCMP), diphosphate (dFdCDP), and then triphosphate
(dFdCTP) forms. Gemcitabine triphosphate replaces deoxy-
Received: December 2, 2013
Published: January 28, 2014
© 2014 American Chemical Society
1531
dx.doi.org/10.1021/jm401853a | J. Med. Chem. 2014, 57, 1531−1542

Journal of Medicinal Chemistry
Article
conversion of the drug into the active dFdCDP and dFdCTP
forms; (ii) rapid degradation into toxic byproducts; and (iii)
limited uptake by cancer cells. These effects are due to the
following: (i) down-regulation of the key initial phosphorylat-
ing enzyme deoxycytidine kinase (dCK)7 required to convert
gemcitabine into the monophosphate form, considered the key
rate-limiting step in its activation process;8 (ii) expression of
the key deactivating enzyme cytidine deaminase;9 and (iii)
deficiency of nucleoside transporter10 proteins. In addition,
increased expression and/or activity of cytidine deaminase
(CDA) increases the degradation of gemcitabine into the toxic
metabolite 2′,2′-difluoro-2′-deoxyuridine (dFdU).9 Because of
these processes, single agent gemcitabine has limited activity in
cancer treatment, with response rates of less than 10% in
pancreatic cancer.11
cornerstone of pancreatic cancer treatment, gemcitabine, and
the screening process that has identified a novel lead candidate,
now in clinical development.
The potency of the compounds varies with the individual
components (aryl, ester, and amino acid) of the phosphor-
amidate moiety, and therefore we examined a panel of
gemcitabine ProTides. Thus, herein we report for the first
time in the open literature the design, synthesis, antitumor
evaluation, and metabolic studies of a panel of L-alanine-based
gemcitabine phosphoramidates. NUC-1031 (6f, first patented
by us in 2005 as CPF-3126) emerged as the most promising
compound. The compound is currently making progress
through the clinical development stage and has already
shown very encouraging activity against a range of solid tumors
and a favorable safety profile in a phase I study.27
Various strategies to improve the treatment efficiency of
gemcitabine have been developed, including the chemical
modifications and functionalization of its 4-(N) and 5′-sites
(reviewed in Moysan et al.).12 In 2009, Bender et al. reported
an orally active prodrug of gemcitabine, LY2334737 which is
significantly less prone to degradation by CDA due to a
valproic acid linkage at the 4-(N)-position.13 Based on in vivo
data in the HCT-116 human colon xenograft, LY2334737 has
been further developed and advanced into phase I clinical
studies, either alone or in combination therapy with other
agents.14 As an alternative approach, CP-4126 was developed15
to enhance the cellular uptake of gemcitabine, by attaching a
fatty acid chain to the ribose 5′-position, but this failed to show
superiority over gemcitabine in clinical trials.16 While these
strategies may offer theoretical benefits, they only address one
or two of the three known resistance mechanisms associated
with gemcitabine. Notably, these approaches do not overcome
the inability of the resistant cancer cell to activate gemcitabine
through phosphorylation. This obligatory first-phosphorylation
step is likely to limit the therapeutic effectiveness of these
prodrugs, as is established for the parent nucleoside analogue.
Therefore, a different approach, utilizing a preactivated
monophosphate form of these nucleoside analogues, has
emerged as a therapeutic approach.17
CHEMISTRY
■
The synthesis of gemcitabine ProTides is largely based on
phosphorochloridate chemistry and requires the preparation of
two components: a regioselectively protected gemcitabine on
the 3′−OH group and an aryl aminoacyl phosphorochloridate.
First, gemcitabine (1) was selectively protected on the 3′−
OH group28 in order to favor the 5′-regioselectivity of the
phosphorylation with the aryl aminoacyl phosphorochloridates
(Scheme 1). Thus, gemcitabine was allowed to react with tert-
a
Scheme 1. Selective Protection of Gemcitabine (1)
a
Reagents and conditions: (a) DBOC, Na2CO3, 1,4-dioxane/H2O
(4:1 v/v), rt, 48 h.
A major strategy currently applied in the modulation of
several antiviral nucleoside analogues is the ProTide technology
developed in 1995 by our laboratories.18 This method is
designed to overcome the key resistance mechanisms associated
with nucleoside analogues in general. Bearing one phosphate
group protected with biolabile groups (termed the phosphor-
amidate motif), the ProTides are preactivated, thereby
bypassing the initial rate-limiting step of kinase phosphorylation
into the monophosphate form. Importantly, the cellular uptake
of these compounds is independent of nucleoside transporters,
and their resistance to deactivation by deaminases and other
enzymes is superior to nucleoside analogues.19 The addition of
the phosphoramidate motif will also change the pharmacoki-
netic/dynamic properties, which may be favorable for therapy.
Overcoming cancer resistance mechanisms should lead to
significant clinical benefits.
butyl dicarbonate in 1,4-dioxane and water (4:1 v/v) at room
temperature for 48 h to give the corresponding intermediate
(4) with a 75% yield after column chromatography.
In parallel, the aryl aminoacyl phosphorochloridates (3),
which are key reagents in the ProTide methodology, are
prepared using two reagents: an esterified amino acid and an
aryloxy phosphorodichloridate (Scheme 2).29 The first reagent,
an amino acid ester, if not commercially available, can be
prepared by esterification of the appropriate amino acid via
standard esterification methods.30 The second reagent is an aryl
(phenyl or 1-naphthyl) phosphorodichloridate; the phenyl
phosphorodichloridate is commercially available, whereas the 1-
naphthyl analogue (2) can be obtained by reacting 1-naphthol
with phosphorus oxychloride and triethylamine following
published procedures.31
In our laboratory we have extensively investigated the
phosphoramidate ProTide approach and successfully applied
this technology to many active antiviral agents such as d4T,20
abacavir,21 and various anti-hepatitis C virus agents.22 Similarly,
we generated ProTides based on several anticancer nucleosides
such as cladribine,23 clofarabine,24 and recently 5-fluoro-2′-
deoxyuridine (FUDR),25 with very encouraging results. Here
we report the application of the ProTide technology to the
The aryl aminoacyl phosphorochloridates (3) were formed
by reaction of phenyl or 1-naphthyl phosphorodichloridates
with appropriate amino acid esters in the presence of
triethylamine, as outlined in Scheme 2. Chiral amino acid
esters generated a pair of diastereoisomers at the phosphate
center, in about 1:1 ratio, visible by two close peaks in the 31P
NMR spectra. In most cases, the aryl aminoacyl phosphoro-
chloridate (3) syntheses were completed within two hours as
1532
dx.doi.org/10.1021/jm401853a | J. Med. Chem. 2014, 57, 1531−1542

Journal of Medicinal Chemistry
Article
a
Scheme 2. Synthesis of Aryl Aminoacyl Phosphorochloridates (3a−l)
a
Reagents and conditions: (a) POCl3, Et3N, dry Et2O, −78 °C for 1 h, then rt for 1 h; (b) (2) or phenyl phosphorodichloridate, Et3N, dry DCM,
−78 °C, 1−3 h.
monitored by 31P NMR. Because of their limited stability, these
products (3) were then used as crude materials, or after rapid
silica gel chromatography, in the ProTide syntheses.
Next, each aminoacyl phosphorochloridate (3) was reacted
with the 3′-protected gemcitabine (4) in the presence of tert-
butylmagnesium chloride (1.0 M solution in THF) as shown in
Scheme 3. Overnight reactions at ambient temperature
ies32 and prior work suggested that the L-alanine phosphor-
amidates were the most potent gemcitabine ProTides.26 Thus,
we further explored this ProTide family bearing the L-alanine
residue by altering the aryl and ester moieties. This led us to
prepare both series of phenyl and 1-naphthyl L-alanine
gemcitabine ProTides with branched (isopropyl, neopentyl),
linear (pentyl, hexyl), cyclic (cyclopentyl, cyclohexyl), and
benzyl ester motifs.
Scheme 3. Preparation of Gemcitabine ProTides (5a−l, 6a−
a
BIOLOGICAL ACTIVITY IN VITRO
l) under Coupling Conditions
■
Nucleoside analogues face numerous inherent and acquired
cancer resistance mechanisms that can dramatically limit their
effectiveness. ProTides are specifically designed to overcome
these key cancer resistance pathways and thereby achieve a
superior antineoplastic effect. Over the years, our laboratory has
created a large library of over 3500 ProTides and built
considerable know-how as to structure/activity relationships
through multiple rounds of pharmacomodulation and func-
tional tests. By applying the ProTide chemistry to gemcitabine,
more than 80 different compounds were generated with various
natural amino acids (data not published). Early chemical and
biological screens during the development of this ProTide
family led to the conclusion that L-alanine-based molecules best
enhanced functional activity. This is in keeping with previous
experience. Several L-alanine phosphoramidate compounds
were synthesized, with numerous variations on the aryl and
ester groups, and these were screened in vitro for cytotoxic
activity against a range of cancer cell lines. Of these, six
compounds showed activity comparable to or better than
gemcitabine and the results are summarized in Table 1.
ProTides showed a broad spectrum of activity levels around the
gemcitabine IC50 reference value, which was used as a positive
control. Notably, within the series of compounds tested,
ProTides 6f, 6i, and 6k demonstrated better (2- to 4-fold
increase) potency than gemcitabine in pancreatic (MiaPaCa-2,
BxPC-3) cell lines. Because gemcitabine is the cornerstone of
pancreatic cancer treatment, these data were particularly
encouraging.
a
Reagents and conditions: (a) (ArO)(R2OC(O)CR1NH)P(O)Cl
(3a−l), tert-BuMgCl, THF, rt, 16 h; (b) TFA/DCM (1:1 v/v), 0
°C, 2−4 h.
generated crude materials that were purified by column
chromatography to provide good yields of 3′-Boc-protected
phosphoramidates 5a−l. Finally, the end products 6a−l were
obtained through acidic deprotection, typically 2−4 h reaction
at 0 °C in TFA/DCM (1:1 v/v), followed by basic workup and
column purification on silica gel. As a result of the phosphate
chirality, ProTides 6a−l were isolated as diastereoisomeric
To exert their anticancer activity, the ProTides should be
metabolized to release the monophosphate form, which will
then generate the active compounds, gemcitabine di- and
triphosphate. The proposed intracellular activation route of the
ProTides has been described33 for other ProTide families, such
as the FUDR ProTides, and is depicted in Scheme 4. The
mixtures, evidenced by 31P NMR, HPLC (two peaks), H
1
NMR, and 13C NMR (with splitting of many nucleoside
signals). Extensive structure/activity relationship (SAR) stud-
1533
dx.doi.org/10.1021/jm401853a | J. Med. Chem. 2014, 57, 1531−1542

Journal of Medicinal Chemistry
Article
Table 1. IC50 (μM) Values of Selected Prodrug Derivatives against Four Cancer Cell Lines
a
IC50 (μM)
entry
amino acid
ester
iPrO
aryl
Ph
L1210
>5
CEM
>5
MP-2
>5
BxPC-3
6a
6b
6c
L-Ala
L-Ala
L-Ala
L-Ala
L-Ala
L-Ala
>5
Pentyl
Hexyl
CH2tBu
cHexyl
Bn
Ph
0.024 0.009
0.016 0.005
0.39 0.04
0.6 0.06
0.08 0.011
0.081 0.008
0.44 0.04
0.42 0.17
0.1 0.03
N.D.
N.D.
Ph
N.D.
N.D.
6d
Ph
1.41 0.57
N.D.
0.36 0.13
N.D.
6e
Ph
6f
Ph
0.035 0.02
0.16 0.03
0.022 0.005
0.18 0.03
0.32 0.03
0.035 0.03
0.03 0.01
0.013 0.02
0.44 0.06
N.D.
0.15 0.04
N.D.
b
6g
DMG
Pentyl
Hexyl
cPentyl
cHexyl
Bn
1-Naph
1-Naph
1-Naph
1-Naph
1-Naph
1-Naph
0.13 0.01
0.075 0.005
0.11 0.00
0.14 0.02
0.10 0.03
0.073 0.004
0.086 0.003
6h
L-Ala
L-Ala
L-Ala
L-Ala
L-Ala
N.D.
N.D.
6i
0.46 0.24
N.D.
0.15 0.05
N.D.
6j
6k
0.57 0.22
N.D.
0.24 0.05
N.D.
61
Pentyl
dFdC (1)
1.04 0.71
0.67 0.34
a
b
IC50 (50% Inhibitory Concentration) is the compound concentration required to inhibit tumor cell proliferation by 50%. DMG is dimethylglycine.
Data shown are the means of at least two independent experiments.
Scheme 4. Proposed Activation Pathway of Gemcitabine ProTide 6f
hypothesized metabolic route would begin with the hydrolysis
of the ProTide ester (i), mediated by a carboxyesterase-type
enzyme to form the intermediate (A). This step is followed by a
spontaneous cyclization (ii) displacing the aryl moiety via an
internal nucleophilic attack of the carboxylate residue on the
phosphorus center to yield (B). In a third step (iii), the
unstable cyclic mixed anhydride would be hydrolyzed to release
the intermediate (C). The final step (iv) would then involve a
phosphoramidase-type enzyme, which would cleave off the
amino acid to generate the corresponding monophosphate (D).
To support the proposed activation pathway ProTide 6f was
incubated with carboxypeptidase Y. This assay34 was designed
to verify whether an enzymatic cleavage of the ester motif
would be sufficient to trigger the first steps of the activation
route and generate the intermediate (C). Compound 6f was
dissolved in acetone-d6 in the presence of Trizma buffer (pH
7.6) and treated with carboxypeptidase Y to monitor the
metabolic conversion of the ProTide by 31P NMR spectroscopy
over time (Figure 1). Two peaks at δ 3.64 and 3.37 ppm in the
blank spectrum recorded at 25 °C correspond to the
diastereoisomers of the parent compound 6f. Within 10 min
of incubation with carboxypeptidase Y, a single peak at δ 4.35
ppm started appearing, consistent with the chemical shift of the
metabolite (A) lacking the ester moiety. A further processing of
the intermediate (A) led to the formation of a third peak δ 6.80
ppm after 10 min, which is consistent with the single chemical
shift expected for the achiral intermediate (C). The peak
increased over time and persisted, suggesting the accumulation
of a stable product. Therefore, the spectra recorded during the
enzymatic reaction suggest that within 12 h of assay 6f was
almost fully converted into the metabolite (C) via the
intermediate (A) and strongly support the proposed pathway
as the actual ProTide activation route.
The optimal metabolism of these ProTides depends on their
chemical stability. Upon administration, the circulating
ProTides are exposed to challenging conditions, such as
catalytic enzymes in the liver and should resist long enough
to enter tumor cells before being degraded. Conversely,
ProTides unaffected by enzymatic digestion might not be
efficiently converted into active compounds inside the cancer
cells. To further select the lead candidates with suitable
chemical stability, six ProTides were incubated with human
hepatocytes and liver microsomes. The half-lives and the
percentage of ProTide remaining after one hour were measured
1534
dx.doi.org/10.1021/jm401853a | J. Med. Chem. 2014, 57, 1531−1542

Journal of Medicinal Chemistry
Article
Figure 1. 31P NMR spectra of ProTide 6f over time after carboxypeptidase Y digestion showing the signals of metabolites.
and are summarized in Table 2. In the human hepatocyte
extracts the half-lives of the six compounds varied from 79 min
Table 2. Metabolic Stability of ProTides Incubated with
Human Hepatocytes and Liver Microsomes
human hepatocytes half-life
(min)
liver microsomes % remain after
l h
entry
6a
6d
6e
6f
219
1065
N.D.
139
93
64
27
2
18
0
6j
6k
79
1
for 6k up to >1000 min for 6d. The selection criteria included
identifying a compound with a mid-range half-life in human
hepatocytes. Importantly, compounds 6f and 6j met this
criterion. Next, the stability of ProTides was further
investigated using liver microsomes. After one hour incubation,
18% of 6f remained detectable, while 6j had entirely been
processed. These results, together with the cytotoxic data
shown earlier, pointed to ProTide 6f (L-Ala-OBn phenyl) as the
lead candidate for further analyses.
Next, the stability of the lead compound, ProTide 6f in the
presence of human serum was investigated using in situ 31P
NMR (Figure 2). The ProTide was dissolved in deuterated
solvent and scanned at 37 °C at regular intervals over 13 h to
monitor the formation of metabolites. The resolution of the
recordings was improved using the Lorentz-Gauss deconvolu-
tion method. The spectra initially displayed a single peak due to
human serum at δ 2.42 ppm and a broad peak at δ 4.15 ppm,
which can be assigned to the ProTide 6f. After 5 h 30 min
incubation, a third signal appeared at δ 7.27 ppm consistent
with the chemical shift expected for metabolite C (Scheme 4).
The reaction mixture contained 59% of the original compound
6f after 13 h together with 41% of the putative intermediate C.
This suggests that in the presence of human serum the ProTide
Figure 2. Deconvoluted 31P NMR spectra of ProTide 6f over time
after incubation with human serum showing the signals of metabolites.
6f was converted to the metabolite C with an estimated half-life
of over 14 h. Also, these results indicate that the stability of the
compound should be suitable for clinical studies.
The ProTides are specifically designed to overcome the key
drug resistance mechanisms associated with nucleoside
analogue treatments against cancer. At the cellular level,
gemcitabine resistance, in particular, is dependent on three
parameters: (i) the down-regulation of deoxycytidine kinase,
necessary for the activation into the phosphorylated moiety;
(ii) the reduced expression of nucleoside transporters, in
particular, hENT1 required for uptake by cancer cells; and (iii)
the up-regulation of catalytic enzymes especially cytidine
deaminase that degrades gemcitabine. Thus, the lead ProTide
6f was evaluated for its ability to bypass these three key
resistance mechanisms. Because the ProTide is synthesized as a
masked monophosphate, it is expected to bypass the dCK-
dependent phosphorylation event for its activation, which is the
rate-limiting step to generate the di- and triphosphate active
1535
dx.doi.org/10.1021/jm401853a | J. Med. Chem. 2014, 57, 1531−1542

Journal of Medicinal Chemistry
Article
Figure 3. ProTide 6f overcomes the key cancer resistance mechanisms associated with gemcitabine in vitro. (A) Effect of deoxycytidine-mediated
dCK inhibition on gemcitabine (GEM) and 6f cytotoxicity in RT112 cells. (B) Effect of dipyridamole-mediated nucleoside transport inhibition on
gemcitabine (GEM) and 6f cytotoxicity in PANC1 cells. (C,D) Absorbance spectra of gemcitabine and ProTide 6f exposed to cytidine deaminase
over 30 min.
moieties. To test this hypothesis, the cytotoxicity of 6f and
gemcitabine on RT112 cancer cells was measured in the
presence of a substrate for the dCK enzyme, 2′-deoxycytidine
(Figure 3A). This compound was shown to compete with
gemcitabine for dCK activity,7b,9b which mimics a down-
regulation of the enzyme. In the presence of deoxycytidine, the
gemcitabine concentration required to block cell proliferation,
EC50, increased 75-fold from 1.4 nM to 104.9 nM, as
determined in the colorimetric MTT assay measuring the
number of viable cells in culture. By contrast, 6f was more
cytotoxic than gemcitabine (i.e., 6f: EC50 lower than
gemcitabine EC50) and its activity was not significantly affected
by deoxycytidine (6f: EC50 = 0.7nM in the presence of
deoxycytidine as opposed to 0.2 nM without deoxycytidine).
This result indicates that 6f can bypass the dCK-mediated
monophosphorylation step to block tumor cell growth.
Next, we explored whether the ProTide could overcome the
requirement for nucleoside analogue transporters to enter the
cancer cells. In a similar assay, the growth inhibition effect of
gemcitabine and 6f was studied on PANC1 pancreatic cancer
cells in the presence of an inhibitor of nucleoside transport,
dipyridamole (Figure 3B). This inhibitor blocks hENT1
transporter activity to resemble cancers where the transporters
responsible for nucleoside uptake are down-regulated. When
gemcitabine was incubated with dipyridamole, its cytotoxic
activity was markedly reduced, with the EC50 significantly
increasing from 611 μM to >2000 μM. In contrast, the
ProTide, which had a baseline activity ∼3-fold superior to
gemcitabine (190 μM compared to 611 μM for gemcitabine),
maintained a strong cytotoxic level in the presence of the
inhibitor (EC50 = 162 μM). These data suggest that 6f is not
reliant on the nucleoside transporter hENT1 to enter the cells
and exert its anticancer effect.
Cytidine deaminase found in the serum and inside the cells
can break down gemcitabine and thereby limit its activity.
Expression of this enzyme is thought to play a key role in the
resistance to the drug.35 Unprotected hydroxyl groups at
positions 3′ and 5′ of the nucleoside analogue were shown to
be necessary for the cleavage to occur.36 In addition, the amino
acid ester prodrugs of gemcitabine have been reported
previously to be resistant to deamination by cytidine deaminase
in vitro.37
Because 6f bears a phosphoramidate moiety on the 5′-carbon
of the ribose it was predicted to resist deamination. To assess
their resistance to enzymatic breakdown, the ProTide and
gemcitabine were incubated with cytidine deaminase and
monitored by UV absorption spectroscopy at wavelengths
between 220 and 350 nm (Figure 3C). The absorbance of
gemcitabine shifted from 267 to 257 nm within the first 2 min
of the UV assay, suggesting breakdown of gemcitabine and
conversion to the corresponding dFdU toxic metabolite. The
spectrum of 6f remained unchanged (Abs 267 nm) at 30 min
and demonstrates that the ProTide is resistant to deaminase
breakdown. Because 6f is less susceptible to degradation, it is
anticipated that it should release less harmful deaminated
catabolite, dFdU, and have a more favorable safety profile in
patients. Together, the cytotoxicity and absorbance data
support a model where the ProTide 6f is able to overcome
the three key resistance mechanisms associated gemcitabine.
Gemcitabine is a prodrug which requires further processing
into the di- and triphosphate active moieties (dFdCDP and
dFdCTP, respectively) to block cancer cell growth. The
triphosphate form is considered the main driver of the
anticancer effect, directly interfering with DNA replication,
while the diphosphate derivative is thought to further
potentiate the activity.5 To test whether 6f cytotoxic activity
1536
dx.doi.org/10.1021/jm401853a | J. Med. Chem. 2014, 57, 1531−1542

Journal of Medicinal Chemistry
Article
Figure 4. ProTide 6f generates high intracellular levels of the active moiety bypassing key drug resistance mechanisms in vitro. (A) Intracellular levels
of dFdCTP (gemcitabine triphosphate) in pancreatic cell lines treated with gemcitabine (red and pink) or 6f (dark and light blue) in the presence
(pink and light blue) or absence (red and dark blue) of the kinase dCK inhibitor, 2T2D. (B) Intracellular levels of dFdCTP in pancreatic cell lines
treated with gemcitabine (red and pink) or 6f (dark and light blue) in the presence (pink and light blue) or absence (red and dark blue) of the
hENT1 (nucleoside transporter) inhibitor, NBTI.
correlates with high levels of intracellular dFdCTP, the ProTide
was tested in a cell growth assay for 24 h before measuring the
intracellular dFdCTP levels (Figure 4). The cells were again
exposed to different inhibitors to recreate the drug resistance
conditions associated with nucleoside analogue treatment (i.e.,
dCK kinase and hENT1 transporter down-regulation). As
expected, when the pancreatic cancer cells (BxPC-3 and
MiaPaCa-2) were exposed to gemcitabine, the dFdCTP levels
were significantly decreased upon 2T2D and NBTI treatments
(Figure 4A and B; compare RED and PINK for both cell lines).
In contrast, the ProTide (in dark and light blue) maintained
steady levels of the triphosphate derivative even in the presence
of the inhibitors. These results demonstrate the capacity of 6f
to bypass the requirement for dCK-mediated activation and
hENT1-dependent uptake to generate high concentrations of
intracellular dFdCTP. Together, these analyses along with the
cytotoxicity results strongly suggest that the ProTide 6f is
efficiently metabolized into the active anticancer moiety,
overcoming the key cancer resistance mechanisms associated
with gemcitabine and blocking tumor cell growth.
The data gathered in vitro prompted us to examine whether
the ProTide 6f would also have a strong anticancer activity in
vivo against well-established mouse models with human tumor
xenografts. First, because the ProTide had previously been
shown to undergo rapid degradation in rodents due to blood
circulating esterases, we examined 6f stability in rat serum and
compared it with dogs as a nonrodent specie reference (Figure
5A). As expected, when the ProTide was incubated with rat
serum and monitored by 31P NMR over time, the spectra
revealed the complete conversion of 6f to the achiral metabolite
C after 6.5 min, whereas the compound was still stable after 14
h in dog serum (>80% 6f; 20% metabolite C). The estimated
half-life of ProTide 6f in rat serum was <5 min but exceeded
100 h in dog serum. These analyses of ProTide 6f and
metabolite C confirmed the prior findings that rodent models
are likely to reduce the advantages seen from the ProTides over
the parent nucleoside analogue.
with 6f was less than 4%, and similar to the gemcitabine treated
mice (Figure 5B). After the treatment was stopped, the mean
body weight of both groups increased, indicating that the
weight loss may have been due to the compound’s toxicity.
When examining 6f and gemcitabine anticancer activities, both
compounds proved effective, with 6f significantly reducing
tumor volume compared to control (vehicle) (Figure 5C).
Importantly, 6f (0.076 mmol/kg) achieved significantly greater
reduction in tumor volume than gemcitabine on Day 7 after the
first administration of the compounds. Moreover, tumor
growth was inhibited following the first administration of the
ProTide and continued to decrease until the dosing was
stopped. These results were further confirmed with a second
human pancreatic cancer cell line known to be resistant to
gemcitabine, BxPC-3 where 6f achieved a significant reduction
in tumor growth compared to control unlike gemcitabine
(Figure 5E). Interestingly, over the treatment period the mean
body weight change was less in the mice treated with the
ProTide than with gemcitabine, suggesting that 6f was better
tolerated than the nucleoside analogue (Figure 5D). Overall,
even in rodents, where the ProTide is very unstable, the
reported in vivo tolerability and efficacy of the lead compound
6f were at least equivalent to or better than gemcitabine. These
results are particularly encouraging because the ProTides are
more stable in other species (i.e., no circulating rodent
esterases), including humans, and this is expected to translate
into superior clinical efficacy.
CONCLUSIONS
■
Tumor cells have numerous inherent and acquired resistance
mechanisms that can render conventional chemotherapy with
nucleoside analogues less effective. Here we describe the
successful application of the ProTide technology to generate
new anticancer agents to overcome the key cancer resistance
mechanisms associated with nucleoside analogues, and in
particular, gemcitabine. The compounds synthesized through
phosphoramidate chemistry demonstrated superior in vitro and
in vivo efficacy compared to gemcitabine, even in experimental
conditions designed to limit gemcitabine activity. By adding a
specific phosphoramidate moiety onto the gemcitabine back-
bone, a series of novel and potent antimetabolite agents were
created with superior efficacy and an improved safety profile,
which were able to bypass the key cancer resistance
To determine the effect of 6f on tumor growth, nude mouse
xenograft models of MiaPaCa-2 human pancreatic cancers were
administrated with equimolar doses of gemcitabine, 6f, and
vehicle. The body weight was also recorded to assess the
animal’s tolerance of the compound. Over the course of the
study, the observed change in body weight for the mice treated
1537
dx.doi.org/10.1021/jm401853a | J. Med. Chem. 2014, 57, 1531−1542

Journal of Medicinal Chemistry
Article
Gemcitabine is a prodrug with limited clinical utility due to
its reliance on deoxycytidine kinase expression for activation;7
requirement on nucleoside transporters for cellular entry;10 and
susceptibility to degradation by deactivating enzymes such as
cytidine deaminase.9 In vitro, the ProTides showed better
cytotoxic activity than gemcitabine against a range of tumor cell
lines. Importantly, the selected lead compound, 6f, generated
much higher intracellular levels of the active moiety,
gemcitabine triphosphate, in experimental conditions known
to be associated with gemcitabine resistance. In vivo, the
ProTide demonstrated a significant reduction in tumor size
against pancreatic xenograft models compared with the
gemcitabine treated group, and less adverse effects on body
weight, indicating a better safety profile. Together these data
strongly suggest that the ProTides are not reliant on kinases or
nucleoside transporters to exert their activity inside tumor cells
and remain stable in the presence of deaminases.
The ProTide NUC-1031 (6f) is currently advancing through
phase I/II clinical studies and has already generated strong
pharmacokinetic data that confirm significantly higher intra-
cellular levels of gemcitabine triphosphate, together with
promising early efficacy signals and a favorable safety profile.27
The phosphoramidate chemistry is potentially a great source of
new and very effective anticancer agents, bringing a
considerable array of advanced treatments specifically designed
to overcome cancer resistance mechanisms that will benefit a
greater proportion of patients.
EXPERIMENTAL SECTION
■
Cytostatic Activity Assays. Murine leukemia L1210 and its dCK-
deficient congener L1210/dCK− and human lymphocyte CEM and its
dCK-deficient congener CEM/dCK− were seeded at −50 to 75 × 103
cells/200 μL in the wells of a 96-well microtiter plate that contained a
variety of drug concentrations. After 48 h (L1210) or 72 h (CEM) the
cell number was determined by counting with a Coulter Counter. The
IC50 (50% inhibitory concentration) was determined as the drug
concentration required to inhibit tumor cell proliferation by 50%.
Metabolic Stability (Cryopreserved Hepatocytes, Human)
Assay. The assay was contracted and performed by Cerep (Seattle,
WA, USA Laboratories, 15318 N. E 95th Street, Redmond, WA,
98052, USA) according to the published procedure (Cerep ref.
1432).38 Pooled cryopreserved hepatocytes were thawed, washed, and
resuspended in Krebs-Heinslet buffer (pH 7.3). The reaction was
initiated by adding the test compound (1 μM final concentration) into
cell suspension and incubated in a final volume of 100 μL on a flat-
bottom 96-well plate for 0 and 60 min, respectively, at 37 °C/5% CO2.
The reaction was stopped by adding 100 μL of acetonitrile into the
incubation mixture. Samples were then mixed gently and briefly on a
plate shaker, transferred completely to a 0.8 mL V-bottom 96-well
plate, and centrifuged at 2550 × g for 15 min at room temperature.
Each supernatant (150 μL) was transferred to a clean cluster tube,
followed by HPLC-MS/MS analysis on a Thermo Electron
triplequadrupole system.
Metabolic Stability (Liver Microsomes, Human) Assay. The
stability assay was contracted and carried out by Cerep (Seattle, WA,
USA Laboratories, 15318 N. E 95th Street, Redmond, WA, 98052,
USA) according to the published procedure (Cerep ref. 0607).39
Pooled liver microsomes are preincubated with NADPH-regenerating
system (1 mM NADP, 5 mM G6P, and 1 U/mL G6PDHase) in
phosphate buffer (pH 7.4) containing 3 mM MgCl2 and 1 mM EDTA
in a 2 mL-block 96-well plate for 10 min in a 37 °C shaking water bath.
The reaction was initiated by adding the test compound (final
concentration 0.5 μM) and incubated in a final volume of 700 μL in
the 37 °C shaking water-bath. At 0, 15, 30, 45, and 60 min after
reaction initiation, 100 μL of the incubation mixture is transferred to
100 μL of acetonitrile/methanol (50/50, v/v) in a 0.8 mL V-bottom
96-well plate. Samples are then mixed on a plate shaker for 5 min and
Figure 5. (A) Metabolic stability of 6f incubated with rat and dog
serum. (B) Mean body weight change (%) of MiaPaCa-2 bearing nude
mice following administration of vehicle (n = 8), gemcitabine (n = 8),
and 6f (0.19 mmol/kg, 2×/week; n = 7). (C) Antitumor effect of
gemcitabine and 6f against MiaPaCa-2 pancreatic xenograft tumors in
nude mice. (D) Mean body weight change (%) of BxPC-3 bearing
nude mice following administration of vehicle (n = 8), gemcitabine
(n = 8), and 6f (0.19 mmol/kg, 2×/week; n = 7). (E) Antitumor effect
of gemcitabine and 6f against BxPC-3 pancreatic xenograft tumors in
nude mice.
mechanisms that limit gemcitabine activity. One ProTide in
particular, 6f, showed exceptional activity and was selected for
clinical studies in patients with advanced cancers.
1538
dx.doi.org/10.1021/jm401853a | J. Med. Chem. 2014, 57, 1531−1542

Journal of Medicinal Chemistry
Article
centrifuged at 2550 × g for 15 min at room temperature. Each
supernatant is transferred to a clean cluster tube, followed by HPLC-
MS/MS analysis on a Thermo Electron triplequadrupole system.
MTS Cell Viability Assay. The assay was contracted and carried
out by WuXi AppTec (Shanghai) Co., Ltd. The corresponding cell
lines MiaPaCa-2 (MP2) and BxPC-3 were seeded at range of density
0.5 to 100 × 103 cells/well in the wells of 96-well plate the day before
incubation, then incubated for 72 h with the different concentrations
of compound tested. Then 50 μL of MTS was added and cells were
incubated for 4 h at 37 °C. The data were read and collected by
Spectra Max 340 Absorbance Microplate Reader.
Maximum Tolerated Dose (MTD) Assay. MTD was contracted
and performed by WuXi AppTec (Shanghai) Co., Ltd. Balb/c nude
mouse (Slac Laboraory Animal Co., Shanghai, China): female, six to
eight week old were intraperitoneally given compounds 6f (i.p 0.228
mmol/kg, 132.3 mg/kg, 2×/WK) and 6j (i.p 0.228 mmol/kg, 141.9
mg/kg, 2×/WK) or vehicle for 2 weeks. Compounds were dissolved in
40% Captisol solution. (40% Captisol was prepared by dissolving 20
mg of Captisol with pure water, and made the final volume 50 mL.
The solvent was filtered with 0.22 μm filter). Mice were monitored
daily for body weight change and clinical symptoms for 2 weeks. The
results were expressed as means SEM.
cytidine deaminase, His-tagged expressed in E. coli, in 20 mM Tris-
HCl buffer pH 8.0, containing 1 mM DTT, 2 mM EDTA, 100 mM
NaCl, 40% glycerol, concentration 0.5 mg/mL as determinated by
Bradford assay) was purchased from Creative BioMart. All reactions
were carried out under an argon atmosphere. Reactions were
monitored with analytical TLC on Silica Gel 60-F254 precoated
aluminum plates and visualized under UV (254 nm) and/or with 31P
NMR spectra. Column chromatography was performed on silica gel
(35−70 μM). Proton (1H), carbon (13C), phosphorus (31P), and
fluorine (19F) NMR spectra were recorded on a Bruker Avance 500
spectrometer at 25 °C. Spectra were autocalibrated to the deuterated
solvent peak and all 13C NMR and 31P NMR were proton-decoupled.
The purity of final compounds was verified to be >95% by HPLC
analysis using Varian Polaris C18-A (10 μM) as an analytic column
with a gradient elution of H2O/MeOH from 100/0 to 0/100 in 40
min (method 1), and with a gradient elution of H2O/CH3CN from
100/0 to 0/100 in 35 min (method 2). The HPLC analysis was
conducted by Varian Prostar (LC Workstation-Varian prostar 335 LC
detector). Low and high resolution mass spectra were performed as a
service by Cardiff University, using electrospray spectrometry (ESMS).
CDA assays were performed with Varian Cary 50 Bio UV/vis
spectrophotometer. UPLC-MS/MS Analysis: dFdCTP was resolved
using an ultraperformance liquid chromatography system (Accela
UPLC, Thermo Scientific, UK) equipped with Ascentis Express 2.7
μm C18 4.6 × 100 mm HPLC column (Sigma Aldrich, UK). Mobile
phase gradient was employed, comprising buffer A (10 mM
ammonium acetate) = 98% and buffer B (acetonitrile) = 2% at 0
min. Buffer A from 98% to 30% over 6 min and from 30% to 98% over
2 min, all at a flow rate of 700 μL/min. Eluting compounds of interest
were detected using a triple stage quadrupole Vantage mass
spectrometry system (Thermo Scientific, UK) equipped with heated
electrospray (HESI) ion source. The ion source conditions for mixed
ion mode (positive and negative) HESI analysis were as follows: ion
spray voltage 3500 V, vaporizer temperature 350 °C, collision gas
(argon) 1.5 mTorr, sheath and auxiliary gases (nitrogen) were 60 and
20 arbitrary units, respectively. The optimum transitional daughter
ions mass and collision energy of dFdCTP was m/z p 502.0 → 159.0
(collision energy 37 V).
In Vivo Xenograft Models of MiaPaCa-2 and BxPC-3 Bearing
Mice Assay. Assay was contracted and performed by WuXi AppTec
(Shanghai) Co., Ltd.: BxPC-3 cell line, MiaPaCa-2 cell line (ATCC,
USA); RPMI 1640 medium (Invitrogen, USA); FBS (Invitrogen,
Australia); Balb/c nude mouse (Slac Laboratory Animal Co., Shanghai,
China): female, six to eight week old, 20
2 g; Captisol (CyDex
Pharmaceuticlas Inc., USA). BxPC-3 and MiaPaCa-2 cell line cultures
were performed according to the cell culture protocols of WuXi
AppTec Shanghai) Co., Ltd. BxPC-3 and MiaPaCa-2 bearing mice
were treated with test compounds or control for 3 weeks. Mice
continued to grow for one week without the above treatment. The
results were expressed as means
SEM. Data was analyzed using
Student’s t test. Significant differences were considered to exist for
those probabilities below 5% (p < 0.05).
Carboxypeptidase Y (EC 3.4.16.1) Assay. The experiment was
carried out by dissolving ProTide 6f (3.0 mg) in acetone-d6 (0.15 mL)
followed by addition of 0.30 mL of Trizma buffer (pH 7.6). After
recording the control 31P NMR at 25 °C, a previously defrosted
carboxypeptidase Y (0.1 mg dissolved in 0.15 mL of Trizma) was
added to the sample, which was then immediately submitted to the 31P
NMR experiments (at 25 °C). The spectra were recorded every 7 min
over 14 h. 31P NMR recorded data were processed and analyzed with
the Bruker Topspin 2.1 program.
Stability Assay in Serum. The experiment was carried out by
dissolving ProTide 6f (5.0 mg) in DMSO (0.050 mL) and D2O (0.15
mL). After recording the control 31P NMR at 37 °C, a previously
defrosted human, rat, or dog serum (0.30 mL) was added to the
sample, which was next submitted to the 31P NMR experiments at 37
°C. The spectra were recorded every 30 min over 13 h. 31P NMR
recorded data were processed and analyzed with the Bruker Topspin
2.1 program. The Lorentz−Gauss deconvolution method was further
used to improve a visualization of results.
Cytidine Deaminase (CDA) Assay. CDA assays were performed
using modified procedure.40 Stock solution of gemcitabine and
ProTide 6f were dissolved in the assay buffer (50 mM Tris-HCl
buffer pH 7.5) to a final concentration of 100 μM. Cytidine deaminase
(100 μg in 200 μL 20 mM Tris-HCl buffer pH 8.0, containing 1 mM
DTT, 2 mM EDTA, 100 mM NaCl, 40% glycerol) was aliquoted for 4
portions and stored at −20 °C prior to usage. 50 μL aliquot of cytidine
deaminase was added to 150 μL of assay buffer. For each assay, 1 mL
of 100 μM of gemcitabine and 6f was placed in a 1 cm UV cuvette at
25 °C, and UV spectrum was recorded over the range 220 to 350 nm.
A portion of enzyme solution (200 μL) was added, and spectra were
recorded in 1 min intervals for 30 min.
General Method for the Preparation of Phosphorochlor-
idates (3a−l). Anhydrous triethylamine (2.0 mol equiv) was added
dropwise at −78 °C to a stirred solution of the appropriate aryl
dichlorophosphate (1.0 mol equiv) and an appropriate amino acid
ester (1.0 mol equiv) in anhydrous DCM under argon atmosphere.
Following the addition, the reaction mixture was allowed to slowly
warm to room temperature and stirred for 1−2 h. Formation of a
desired compound was monitored by 31P NMR. After the reaction was
completed, the solvent was evaporated under reduced pressure and the
resulting residue was redissolved in anhydrous Et2O and filtered. The
filtrate was reduced to dryness to give a crude product as an oil, which
was in some cases used without further purification in the next step.
Most of aryl phosphorochloridates, in particular, those obtained from
the amino acid tosylate salt were purified by flash column
chromatography using EtOAc/Hexane (7:3) as an eluent.
Phenyl(isopropoxy-L-alaninyl) Phosphorochloridate (3a).
Yellowish oil; yield 80% (2.94 g); mixture of diastereoisomers; 31P
1
NMR (202 MHz, CDCl3): δP 8.08, 7.71 (int. 1.00:1.00). H NMR
(500 MHz, CDCl3): δH 7.41 − 7.36 (m, 2H, ArH), 7.31 − 7.27 (m,
3H, ArH), 5.21 − 5.01 (m, 1H, OCH(CH3)2), 4.21 − 4.06 (m, 1H,
CHCH3), 1.51 (d, J = 6.9 Hz, 3H, CHCH3), 1.26 − 1.19 (m, 6H,
OCH(CH3)2).
3′-O-(tert-Butoxycarbonyl)-gemcitabine (4). Prepared accord-
ing to the published procedure28 and obtained as a white solid. Yield,
75% (1.03 g). 19F NMR (470 MHz, MeOD): δF − 115.74, − 115.75
1
(d, J = 243 Hz, F), − 119.21 (broad signal, F). H NMR (500 MHz,
MeOD): δH 7.68 (d, 1H, J = 7.5, H-6), 6.18 (t, 1H, J = 8.5, H-1′), 5.84
(d, 1H, J = 7.5, H-5, 5.13 (q, 1H, J = 6.0, H-3′), 4.18 − 4.15 (m, 1H,
H-4′), 3.82 (dd, 1H, J = 13.0, J = 3.0, H-5′a), 3.68 (dd, 1H, J = 13.0, J =
3.0, H-5′b), 1.39 (s, 9H, C(CH3)3).
Chemistry. General. Anhydrous solvents were obtained from
Aldrich and used without further purification. Amino acid esters were
purchased from Carbosynth; Carboxypeptidase Y, human serum, and
buffers from Sigma-Aldrich. Cytidine deaminase (recombinant human
General Method for the Preparation of Gemcitabine
ProTides (5a−l). To a solution of 3′-O-(tert-butoxycarbonyl)-
1539
dx.doi.org/10.1021/jm401853a | J. Med. Chem. 2014, 57, 1531−1542

Journal of Medicinal Chemistry
Article
gemcitabine 4 (1.0 equiv) in dry THF (10 mL), tert-BuMgCl (1.0 M
solution in THF, 1.2 equiv) was added in one portion followed by
addition of appropriate phosphorochloridate (2.0 equiv) dissolved in
anhydrous THF (3 mL). The reaction mixture was stirred for 16 to 18
h and then evaporated in vacuo to give a crude residue that was
purified by column chromatography on silica gel, eluting with CH2Cl2-
MeOH as a gradient (0% to 5% MeOH) to afford the products 5a−l
as white solid.
General Method for the Preparation of Gemcitabine
ProTides (6a−l). The 3′-O-(tert-butoxycarbonyloxy)-gemcitabine
phosphates (5a−l) were treated with a mixture of TFA/DCM (1:1)
(6 mL) at 0 °C. The resulting reaction mixture was stirred at 0 °C for
2−4 h. After the reaction was completed the solvents were evaporated
and the residue was treated with saturated NaHCO3, and extracted
with EtOAc. The organic layers were combined, dried (MgSO4),
filtered, reduced to dryness, and purified on silica gel with gradient of
eluent MeOH/DCM (2% to 8% MeOH).
ACKNOWLEDGMENTS
■
We thank Helen Murphy and Lindsey Leven for secretarial
assistance and NuCana BioMed Ltd and the KU Leuven (GOA
10/014) for financial support. We thank Sofiya Abramchuk and
Lizette van Berckelaer for excellent laboratory assistance.
ABBREVIATIONS
■
dFdC, 2′,2′- difluoro-2′-deoxycytidine; dCyd, deoxycytidine;
d4T, stavudine; dCK, deoxycytidine kinase; L1210, wild-type
murine leukemia; CEM, human T-lymphocyte CEM; FUDR,
2′-deoxy-5-fluorouridine; NBTI, S-(4-nitrobenzyl)-6-thioino-
sine; hENT1, human equilibrative nucleoside transporter-1;
TLC, thin layer chromatography; HPLC, high performance
liquid chromatography
2′-Deoxy-2′,2′-difluoro-3′-O-(tert-butoxycarbonyloxy)-D-cy-
tidine-5′-O-[phenyl (isopropoxy-L-alaninyl)]phosphate (5a).
Obtained from 3′-O-(tert-butoxycarbonyl)-gemcitabine 4 (0.25 g,
0.69 mmol), the phosphorochloridate 3a (0.42 g, 1.37 mmol) and tert-
BuMgCl (0.83 mL, 0.83 mmol) as a white solid. Yield 83% (0.36 g);
mixture of diastereoisomers (45%, 55%). 31P NMR (202 MHz,
REFERENCES
■
(1) Hertel, L. W.; Kroin, J. S.; Misner, J. W.; Tustin, J. M. Synthesis
of 2-deoxy-2,2-difluoro-D-ribose and 2′-deoxy-2′,2′-difluoro-D-ribo-
furanosyl nucleosides. J. Org. Chem. 1988, 53, 2406−2409.
(2) (a) Hertel, L. W.; Boder, G. B.; Kroin, J. S.; Rinzel, S. M.; Poore,
G. A.; Todd, G. C.; Grindey, G. B. Evaluation of the antitumor activity
of gemcitabine (2′,2′-difluoro-2′-deoxycytidine). Cancer Res. 1990, 50,
4417−4422. (b) Braakhuis, B. J.; Ruiz van Haperen, V. W.; Welters, M.
J.; Peters, G. J. Schedule-dependent therapeutic efficacy of the
combination of gemcitabine and cisplatin in head and neck cancer
xenografts. Eur. J. Cancer 1995, 31A, 2335−2340. (c) Merriman, R. L.;
Hertel, L. W.; Schultz, R. M.; Houghton, P. J.; Houghton, J. A.;
Rutherford, P. G.; Tanzer, L. R.; Boder, G. B.; Grindey, G. B.
Comparison of the antitumor activity of gemcitabine and ara-C in a
panel of human breast, colon, lung and pancreatic xenograft models.
Invest. New Drugs 1996, 14, 243−247. (d) Manegold, C. Gemcitabine
(Gemzar) in non-small cell lung cancer. Expert Rev. Anticancer Ther.
2004, 4, 345−360. (e) Heinemann, V. Gemcitabine in metastatic
breast cancer. Expert Rev. Anticancer Ther. 2005, 5, 429−443.
(3) Eckel, F.; Schneider, G.; Schmid, R. M. Pancreatic cancer: a
review of recent advances. Expert Opin. Invest. Drugs 2006, 15, 1395−
1410.
1
MeOD): δP 3.86, 3.79. H NMR (500 MHz, MeOD): δH 7.56, 7.48
(2d, J = 7.50 Hz, 1H, H-6), 7.41 − 7.39 (m, 2H, ArH), 7.35 − 7.19 (m,
3H, ArH), 6.30 − 6.28 (m, 1H, H-1′), 5.89, 5.85 (2d, J = 7.50 Hz, 1H,
H-5), 5.28 − 5.22 (m, 1H, H-3′), 5.04 − 4.98 (m, 1H, OCH(CH3)2),
4.54 − 4.50 (m, 1H, H-5′a), 4.46 − 4.39 (m, 2H, H-5′b, H-4′), 4.01 −
3.98 (m, 1H, CHCH3), 1.51, 1.50 (2s, 9H, C(CH3)3), 1.38 (d, J = 6.30
Hz, 3H, CHCH3).
2′-Deoxy-2′,2′-difluoro-D-cytidine-5′-O-[phenyl(isopropoxy-
L-alaninyl)] phosphate (6a). Obtained from 2′-deoxy-2′,2′-difluoro-
3′-O-(tert-butoxycarbonyloxy)-D-cytidine-5′-O-[phenyl(isopropoxy-L-
alaninyl)]phosphate (5a) (0.36 g, 0.57 mmol) as a white solid. Yield
50% (0.15 g). (ES+) m/z: found: (M + Na+) 555.15.
C21H27F2N4O8NaP required: (M+) 532.43. Mixture of diaster-
eoisomers (45%, 55%). 31P NMR (202 MHz, MeOD): δP 3.80,
3.73. 19F NMR (470 MHz, MeOD): δF − 118.0 (d, J = 240 Hz), −
1
118.2 (d, J = 240 Hz, F), − 123.43 (− 123.74) (broad signal, F). H
NMR (500 MHz, MeOD): δH 7.60, 7.55 (2d, J = 7.72 Hz, 1H, H-6),
7.41 − 7.36 (m, 2H, ArH), 7.29 − 7.22 (m, 3H, ArH), 6.30 − 6.25 (m,
1H, H-1′), 5.94, 5.89 (2d, J = 7.72 Hz, 1H, H-5), 5.03 − 4.97 (m, 1H,
OCH(CH3)2), 4.56 − 4.35 (m, 2H, H-5′), 4.28 − 4.21 (m, 1H, H-3′),
4.14 − 4.09 (m, 1H, H-4′), 3.97 − 3.91 (m, 1H, CHCH3), 1.37 − 1.32
(m, 3H, CHCH3), 1.25 − 1.22 (m, 6H, OCH(CH3)2). 13C NMR (125
MHz, MeOD): δC 174.4 (CO, ester), 167.62 (C-NH2), 157.80
(CO base), 152.18, 152.12 (C-Ar), 142.5, 142.41 (CH-base), 130.88
(4) Toschi, L.; Finocchiaro, G.; Bartolini, S.; Gioia, V.; Cappuzzo, F.
Role of gemcitabine in cancer therapy. Future Oncol. 2005, 1, 7−17.
(5) (a) Huang, P.; Chubb, S.; Hertel, L. W.; Grindey, G. B.; Plunkett,
W. Action of 2′,2′-difluorodeoxycytidine on DNA synthesis. Cancer
Res. 1991, 51, 6110−6117. (b) Ruiz van Haperen, V. W.; Veerman, G.;
Vermorken, J. B.; Peters, G. J. 2′,2′-Difluorodeoxycytidine (gemcita-
bine) incorporation into RNA and DNA of tumor cell lines. Biochem.
Pharmacol. 1993, 46, 762−766.
(6) Heinemann, V.; Xu, Y. Z.; Chubb, S.; Sen, A.; Hertel, L. W.;
Grindey, G. B.; Plunkett, W. Cellular elimination of 2′,2′-
difluorodeoxycytidine-5′-triphosphate: a mechanism of self-potentia-
tion. Cancer Res. 1992, 52, 533−539.
(7) (a) Ruiz van Haperen, V. W.; Veerman, G.; Eriksson, S.; Boven,
E.; Stegmann, A. P.; Hermsen, M.; Vermorken, J. B.; Pinedo, H. M.;
Peters, G. J. Development and molecular characterization of a 2′,2′-
difluorodeoxycytidine-resistant variant of the human ovarian carcino-
ma cell line A2780. Cancer Res. 1994, 54, 4138−4143. (b) Kroep, J. R.;
Loves, W. J.; van der Wilt, C. L.; Alvarez, E.; Talianidis, I.; Boven, E.;
Braakhuis, B. J. M.; van Groeningen, C. J.; Pinedo, H. M.; Peters, G. J.
Pretreatment deoxycytidine kinase levels predict in vivo gemcitabine
sensitivity. Mol. Cancer Ther. 2002, 1, 371−376. (c) Galmarini, C. M.;
Clarke, M. L.; Jordheim, L.; Santos, C. L.; Cros, E.; Mackey, J. R.;
Dumontet, C. Resistance to gemcitabine in a human follicular
lymphoma cell line is due to partial deletion of the deoxycytidine
kinase gene. BMC Pharmacol. 2004, 4, 8.
(8) (a) Heinemann, V.; Hertel, L. W.; Grindey, G. B.; Plunkett, W.
Comparison of the cellular pharmacokinetics and toxicity of 2′,2′-
difluorodeoxycytidine and 1-β-D-arabinofuranosylcytosine. Cancer Res.
1988, 48, 4024−4031. (b) Gandhi, V.; Plunkett, W. Modulatory
activity of 2′,2′-difluorodeoxycytidine on the phosphorylation and
3
1
(d, JC−P = 3.75 Hz, CH−Ar), 126.33 (CH−Ar), 122.46 (d, JC−F
=
256 Hz, CF2), 121.42, 121.40 (2d, 3JC−P = 2.5 Hz, CH−Ar), 96.8 (CH-
base), 85.6 (broad signal, C-1′), 80.40 (C-4′), 71.62, 71.20 (2d, 2JC−F
=
37.5 Hz, C-3′), 70.24 (OCH(CH3)2), 65.71, 65.66 (2d, 2JC−P = 5.0 Hz,
C-5′), 51.89, 51.75 (CHCH3), 21.99, 21.91, 21.82 (OCH(CH3)2),
16.47 (CHCH3). Reverse HPLC, eluting with H2O/MeOH from 100/
0 to 0/100 in 40 min, showed two peaks of the diastereoisomers with
tR 22.73 min and tR 23.53 min (44%, 54%).
ASSOCIATED CONTENT
■
S
* Supporting Information
Preparative methods, spectroscopic and analytical data for
target compounds. This material is available free of charge via
the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*Phone/fax: +44 2920874537. E-mail: mcguigan@cardiff.ac.uk.
Notes
The authors declare no competing financial interest.
1540
dx.doi.org/10.1021/jm401853a | J. Med. Chem. 2014, 57, 1531−1542

Journal of Medicinal Chemistry
Article
cytotoxicity of arabinosyl nucleosides. Cancer Res. 1990, 50, 3675−
3680.
action of masked alaninyl d4T-MP derivatives. Proc. Natl. Acad. Sci.
U.S.A. 1996, 93, 7295−7299.
(9) (a) Beumer, J. H.; Eiseman, J. L.; Parise, R. A.; Joseph, E.; Covey,
J. M.; Egorin, M. J. Modulation of gemcitabine (2′,2′-difluoro-2′-
deoxycytidine) pharmacokinetics, metabolism, and bioavailability in
mice by 3,4,5,6-tetrahydrouridine. Clin. Cancer Res. 2008, 14, 3529−
3535. (b) Mini, E.; Nobili, S.; Caciagli, B.; Landini, I.; Mazzei, T.
Cellular pharmacology of gemcitabine. Ann Oncol. 2006, 17 (Suppl 5),
v 7−12.
(10) (a) Mackey, J. R.; Mani, R. S.; Selner, M.; Mowles, D.; Young, J.
D.; Belt, J. A.; Crawford, C. R.; Cass, C. E. Functional nucleoside
transporters are required for gemcitabine influx and manifestation of
toxicity in cancer cell lines. Cancer Res. 1998, 58, 4349−4357.
(b) Gourdeau, H.; Clarke, M. L.; Ouellet, F.; Mowles, D.; Selner, M.;
(21) McGuigan, C.; Harris, S. A.; Daluge, S. M.; Gudmundsson, K.
S.; McLean, E. W.; Burnette, T. C.; Marr, H.; Hazen, R.; Condreay, L.
D.; Johnson, L.; De Clercq, E.; Balzarini, J. Application of
phosphoramidate pronucleotide technology to abacavir leads to a
significant enhancement of antiviral potency. J. Med. Chem. 2005, 48,
3504−3515.
(22) Vernachio, J. H.; Bleiman, B.; Bryant, K. D.; Chamberlain, S.;
Hunley, D.; Hutchins, J.; Ames, B.; Gorovits, E.; Ganguly, B.; Hall, A.;
Kolykhalov, A.; Liu, Y.; Muhammad, J.; Raja, N.; Walters, C. R.; Wang,
J.; Williams, K.; Patti, J. M.; Henson, G.; Madela, K.; Aljarah, M.;
Gilles, A.; McGuigan, C. INX-08189, a phosphoramidate prodrug of 6-
O-methyl-2′-C-methyl guanosine, is a potent inhibitor of hepatitis C
virus replication with excellent pharmacokinetic and pharmacodynamic
properties. Antimicrob. Agents Chemother. 2011, 55, 1843−1851.
(23) Walsby, E.; Congiatu, C.; Schwappach, A.; Walsh, V.; Burnett,
A. K.; McGuigan, C.; Mills, K. I. Nucleoside analogues of cladribine
produce enhanced responses in cell lines. Blood 2005, 106, 941A−
942A (Abstract 3369)..
̀
Richard, A.; Lee, N.; Mackey, J. R.; Young, J. D.; Jolivet, J.; Lafreniere,
R. G.; Cass, C. E. Mechanisms of uptake and resistance to
troxacitabine, a novel deoxycytidine nucleoside analogue, in human
leukemic and solid tumor cell lines. Cancer Res. 2001, 61, 7217−7224.
(11) (a) www.GEMZAR.com. Locally advanced or metastatic
pancreatic cancer. (b) Liau, S.-S.; Whang, E. HMGA1 is a molecular
determinant of chemoresistance to gemcitabine in pancreatic
adenocarcinoma. Clin. Cancer Res. 2008, 14, 1470−1477.
(24) McGuigan, C.; Congiatu, C.; Mills, K. Phosphoramidate
derivatives of nucleoside compounds for use in the treatment of
cancer. EP1866324A1, WO 2006100439 A1.
(12) Moysan, E.; Bastiat, G.; Benoit, J. P. Gemcitabine versus
modified gemcitabine: a review of several promising chemical
modifications. Mol. Pharmaceutics 2013, 10, 430−444.
(13) Bender, D. M.; Bao, J.; Dantzig, A. H.; Diseroad, W. D.; Law, K.
L.; Magnus, N. A.; Peterson, J. A.; Perkins, E. J.; Pu, Y. J.; Reutzel-
Edens, S. M.; Remick, D. M.; Starling, J. J.; Stephenson, G. A.; Vaid, R.
K.; Zhang, D.; McCarthy, J. R. Synthesis, crystallization, and biological
evaluation of an orally active prodrug of gemcitabine. J. Med. Chem.
2009, 52, 6958−6961.
(14) Koolen, S. L. W.; Witteveen, P. O.; Jansen, R. S.; Langenberg,
M. H. G.; Kronemeijer, R. H.; Nol, A.; Garcia-Ribas, I.; Callies, S.;
Benhadji, K. A.; Slapak, C. A.; Beijnen, J. H.; Voest, E. E.; Schellens, J.
H. M. Phase I study of oral gemcitabine prodrug (LY2334737) alone
and in combination with erlotinib in patients with advanced solid
tumors. Clin. Cancer Res. 2011, 17, 6071−6082.
(15) (a) Bergman, A. M.; Adema, A. D.; Balzarini, J.; Bruheim, S.;
Fichtner, I.; Noordhuis, P.; Fodstad, O.; Myhren, F.; Sandvold, M. L.;
Hendriks, H. R.; Peters, G. J. Antiproliferative activity, mechanism of
action and oral antitumor activity of CP-4126, a fatty acid derivative of
gemcitabine, in in vitro and in vivo tumor models. Invest. New Drugs
2011, 29, 456−466. (b) Adema, A. D.; Smid, K.; Losekoot, N.;
Honeywell, R. J.; Verheul, H. M.; Myhren, F.; Sandvold, M. L.; Peters,
G. J. Metabolism and accumulation of the lipophilic deoxynucleoside
analogs elacytarabine and CP-4126. Invest. New Drugs 2012, 30, 1908−
1916.
(25) (a) Vande Voorde, J.; Liekens, S.; McGuigan, C.; Murziani, P.
G. S; Slusarczyk, M.; Balzarini, J. The cytostatic activity of NUC-3073,
a phosphoramidate prodrug of 5-fluoro-2′-deoxyuridine, is independ-
ent of activation by thymidine kinase and insensitive to degradation by
phosphorolytic enzymes. Biochem. Pharmacol. 2011, 82, 441−452.
(b) McGuigan, C.; Murziani, P. G. S.; Slusarczyk, M.; Gonczy, B.;
Vande Voorde, J.; Liekens, S.; Balzarini, J. Phosphoramidate ProTides
of the anticancer agent FUDR successfully deliver the preformed
bioactive monophosphate in cells and confer advantage over the
parent nucleoside. J. Med. Chem. 2011, 54, 7247−7258.
(26) (a) McGuigan, C. Preparation of amino acid-containing
nucleotide phosphoramidates as antitumor agents. PCT. Int. Appl.
2005 WO 2005012327 A2 20050210. (b) McGuigan, C. Nucleotide
phosphoramidates as anticancer agents. US 7951787 B2, 2005.
(27) Ghazaly E. A; Simon J.; Gribben, J. G.; Mohammad, T.;
Oluwadunni E.; Stavraka, Ch.; Hopkins, T.; Gabra, H.; Harpreet, W.;
Habib, N. A.; Leonard, R. C. F.; McGuigan, C.; Slusarczyk, M.;
Blagden, S. P. ProGem1: Phase 1 first-in-human study of the novel
nucleotide analogue NUC-1031 in adult patients with advanced solid
tumors. 2013 ASCO Annual Meeting J. Clin. Oncol. 2013, 31 (suppl;
Abstract 2576).
(28) Guo, Z.; Gallo, J. M. Selective protection of 2′,2′-
difluorodeoxycytidine (Gemcitabine). J. Org. Chem. 1999, 64, 8319−
8322.
(16) Ahmed, T. Clavis and Clovis collaborate on anticancer drug CP-
4126. Pharma Deals Review 2009, 12, 122. In 2012, Clavis Pharma and
Clovis Oncology suspended further development of CP-4126 after
demonstrating no difference in overall survival between CP-4126 and
gemcitabine in hENT1-low patients.
(29) McGuigan, C.; Pathirana, R. N.; Balzarini, J.; De Clercq, E.
Intracellular delivery of bioactive AZT nucleotides by aryl phosphate
derivatives of AZT. J. Med. Chem. 1993, 36, 1048−1052.
(30) Greene, T. W.; Wuts, P. G. M. Protective groups, in Organic
Chemistry, 2nd ed.; Wiley: New York, 1991.
(17) (a) Bazzanini, R.; Manfredini, S.; Durini, E.; Groschel, B.;
Cinatl, J.; Balzarini, J.; De Clercq, E.; Imbach, J. L.; Peirgaud, C.;
́
(31) Curley, D.; McGuigan, C.; Devine, K. G.; O′Connor, T. J.;
Jeffries, D. J.; Kinchington, D. Synthesis and anti-HIV evaluation of
some phosphoramidate derivatives of AZT: studies on the effect of
chain elongation on biological activity. Antiviral Res. 1990, 14, 345−
356.
(32) McGuigan, C.; Devine, K. G.; O′Connor, T. J.; Kinchington, D.
Synthesis and anti-HIV activity of some haloalkyl phosphoramidate
derivatives of 3′-azido-3′-deoxythymidine (AZT): potent activity of
the trichloroethyl methoxyalaninyl compound. Antiviral Res. 1991, 15,
255−263.
(33) Saboulard, D.; Naesens, L.; Cahard, D.; Salgado, A.; Pathirana,
R.; Velazquez, S.; McGuigan, C.; De Clercq, E.; Balzarini, J.
Characterization of the activation pathway of phosphoramidate triester
prodrugs of stavudine and zidovudine. Mol. Pharmacol. 1999, 56, 693−
704.
̈
Gosselin, G. Prodrugs of Ara-CMP and Ara-AMP with a S-acyl-2-
thioethyl (SATE) biolabile phosphate protecting group: synthesis and
biological evaluation. Nucleoside Nucleotides 1999, 18, 971−972.
(b) Galmarini, C. M.; Warren, G.; Senanayake, M. T.; Vinogradov,
S. V. Efficient overcoming of drug resistance to anticancer nucleoside
analogs by nanodelivery of active phosphorylated drugs. Int. J. Pharm.
2010, 395, 281−289.
(18) Cahard, D.; McGuigan, C.; Balzarini, J. Aryloxy phosphor-
amidate triesters as protides. Mini-Rev. Med. Chem. 2004, 4, 371−381.
(19) McGuigan, C.; Shackleton, J. M.; Tollerfield, S. M.; Riley, P. A.
Synthesis and evaluation of some novel phosphate and phosphinate
derivatives of araA. Studies on the mechanism of action of phosphate
triesters. Nucleic Acids Res. 1989, 17, 10171−10177.
(20) Balzarini, J.; Karlsson, A.; Aquaro, S.; Perno, C. F.; Cahard, D.;
Naesens, L.; De Clercq, E.; McGuigan, C. Mechanism of anti-HIV
(34) Birkus, G.; Wang, R.; Liu, X.; Kutty, N.; MacArthur, H.; Cihlar,
T.; Gibbs, C.; Swaminathan, S.; Lee, W.; McDermott, M. Cathepsin A
1541
dx.doi.org/10.1021/jm401853a | J. Med. Chem. 2014, 57, 1531−1542

Journal of Medicinal Chemistry
Article
is the major hydrolase catalyzing the intracellular hydrolysis of the
antiretroviral nucleotide phosphonoamidate prodrugs GS-7340 and
GS-9131. Antimicrob. Agents Chemother. 2007, 51, 543−550.
(35) Bengala, C.; Guarneri, V.; Giovannetti, E.; Lencioni, M.;
Fontana, E.; Mey, V.; Fontana, A.; Boggi, U.; Del Chiaro, M.; Danesi,
R.; Ricci, S.; Mosca, F.; Del Tacca, M.; Conte, P. F. Prolonged fixed
dose rate infusion of gemcitabine with autologous haemopoietic
support in advanced pancreatic adenocarcinoma. Brit. J. Cancer. 2005,
93, 35−40.
(36) Camiener, G. W. Studies of the enzymatic deamination of
cytosine arabinoside. 3. Substrate requirements and inhibitors of the
deaminase of human liver. Biochem. Pharmacol. 1967, 16, 1691−1702.
(37) Song, X.; Lorenzi, P. L.; Landowski, C. P.; Vig, B. S.; Hilfinger, J.
M.; Amidon, G. L. Amino acid ester prodrugs of the anticancer agents
gemcitabine: synthesis, bioconversion, metabolic bioevasion, and
hPEPT1-mediated transport. Mol. Pharm. 2005, 2, 157−167.
(38) Li, A. P.; Lu, C.; Brent, J. A.; Pham, C.; Fackett, A.; Ruegg, C. E.;
Silber, P. M. Cryopreserved human hepatocytes: characterization of
drug-metabolizing enzyme activities and applications in higher
throughput screening assays for hepatotoxicity, metabolic stability,
and drug-drug interaction potential. Chem. Biol. Interact. 1999, 121,
17−35.
(39) Kuhnz, W.; Gieschen, H. Predicting the oral bioavailability of
19-nortestosterone progestins in vivo from their metabolic stability in
human liver microsomal preparations in vitro. Drug Metab. Dispos.
1998, 26, 1120−1127.
(40) (a) Murakami, E.; Niu, C.; Bao, H.; Micolochick Steuer, H. M.;
Whitaker, T.; Nachman, T.; Sofia, M. A.; Wang, P.; Otto, M. J.;
Furman, P. A. The mechanism of action of ß-D-2′-deoxy-2′-fluoro-2′-
C-methylcytidine involves a second metabolic pathway leading to ß-D-
2′-deoxy-2′-fluoro-2′-C-methyluridine 5′-triphosphate, a potent inhib-
itor of the hepatitis C virus RNA-dependent RNA polymerase.
Antimicrob. Agents Chemeother. 2008, 52, 458−464. (b) CDA Activity
Assay, Material Safety Datasheets available at www.acris-antibodies.
com.
1542
dx.doi.org/10.1021/jm401853a | J. Med. Chem. 2014, 57, 1531−1542