Synthesis and Anti-herpetic Activity of Phosphoramidate ProTides

Article abstract of DOI:10.1002/cmdc.201300035

Among the many prodrug approaches aimed at delivering nucleoside monophosphates into cells, the phosphoramidate ProTide approach is one that has shown success, which has made it possible for some of the phosphoramidates to enter into clinical trials. Herein, we report the synthesis and antiviral activity of a series of phosphoramidate ProTides designed to bypass the thymidine kinase (TK) dependence of the parent nucleoside analogues. Phosphoramidate derivatives of (E)-5-(2-bromovinyl)-2′-deoxyuridine (BVDU) that contain L-alanine or pivaloyloxymethyl iminodiacetate (IDA-POM) exhibit anti-HSV-1 and anti-VZV activity in cell cultures, but they largely lost antiviral potency against TK-deficient virus strains. Among deazapurine nucleosides and their phosphoramidate derivatives, the 7-deazaadenine containing nucleosides and their phosphoramidate triester derivatives showed weak antiviral activity against VZV. Apparently, intracellular nucleotide delivery with these phosphoramidates is partly successful. However, none of the compound prodrugs showed superior activity to their parent drugs. Copyright

Full text of DOI:10.1002/cmdc.201300035

CHEMMEDCHEM
FULL PAPERS
DOI: 10.1002/cmdc.201300035
Synthesis and Anti-herpetic Activity of Phosphoramidate
ProTides
Munmun Maiti,[a] Leentje Persoons,[b] Graciela Andrei,[b] Robert Snoeck,[b] Jan Balzarini,[b] and
Piet Herdewijn*[a]
Dedicated to Professor Fritz Eckstein on the occasion of his 80th birthday.
Among the many prodrug approaches aimed at delivering nu-
cleoside monophosphates into cells, the phosphoramidate Pro-
Tide approach is one that has shown success, which has made
it possible for some of the phosphoramidates to enter into
clinical trials. Herein, we report the synthesis and antiviral ac-
tivity of a series of phosphoramidate ProTides designed to
bypass the thymidine kinase (TK) dependence of the parent
nucleoside analogues. Phosphoramidate derivatives of (E)-5-(2-
bromovinyl)-2’-deoxyuridine (BVDU) that contain l-alanine or
pivaloyloxymethyl iminodiacetate (IDA-POM) exhibit anti-HSV-
1 and anti-VZV activity in cell cultures, but they largely lost an-
tiviral potency against TK-deficient virus strains. Among deaza-
purine nucleosides and their phosphoramidate derivatives, the
7-deazaadenine containing nucleosides and their phosphora-
midate triester derivatives showed weak antiviral activity
against VZV. Apparently, intracellular nucleotide delivery with
these phosphoramidates is partly successful. However, none of
the compound prodrugs showed superior activity to their
parent drugs.
Introduction
Today, around 50 antiviral drugs
are on the market that have
been approved for the treat-
ment of several virus infec-
tions.[1] Of those 50, only eight
(Figure 1) are licensed to be
used as anti-herpetic drugs,
among them (E)-5-(2-bromovin-
yl)-2’-deoxyuridine (BVDU, Brivu-
din),[2] which is used against
herpes simplex virus-1 (HSV-1)
and varicella-zoster virus (VZV).
The selectivity of BVDU toward
HSV-1 and VZV depends primari-
ly on a specific sequential phos-
phorylation by the virus-en-
Figure 1. Marketed anti-herpetic drugs for clinical use.
coded
thymidine/thymidylate
kinase (TK) to form the 5’-mono-
and 5’-diphosphate derivatives.[2–4] After further phosphoryla-
tion into the 5’-triphosphate (BVDU-TP) by cellular kinases,
BVDU-TP interferes with the viral DNA polymerase as a compet-
itive/alternative substrate and thus interferes with the viral
DNA synthesis.[3]
In most cases, like BVDU, the active form of a nucleoside
drug is the 5’-O-triphosphate derivative (functioning as a DNA
chain terminator/polymerase inhibitor).[5] The formation of the
triphosphate occurs in three steps, and the first step, responsi-
ble for the formation of the nucleoside monophosphate, is
often considered as the rate-limiting step, which limits the
therapeutic potential of many modified nucleoside ana-
logues.[6] Moreover, drug-resistant virus strains are appearing
with the continuous use of drugs: for example, BVDU-resistant
HSV-1 isolates have been identified and found to be associated
[a] M. Maiti, Prof. P. Herdewijn
Laboratory of Medicinal Chemistry
Rega Institute for Medical Research, KU Leuven
Minderbroedersstraat 10, 3000 Leuven (Belgium)
E-mail: Piet.Herdewijn@rega.kuleuven.be
[b] L. Persoons, Prof. G. Andrei, Prof. R. Snoeck, Prof. J. Balzarini
Laboratory of Virology and Chemotherapy
Rega Institute for Medical Research, KU Leuven
Minderbroedersstraat 10, 3000 Leuven (Belgium)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201300035.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 10
&
1
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
with a decreased level of viral thymidine kinase activity.[7]
Therefore, direct administration of the preformed nucleoside
monophosphate (nucleotide) would be preferred in order to
bypass the rate-limiting step in the nucleoside metabolism.
However, being charged under physiological conditions, nu-
cleoside monophosphates show poor, if any, cell-membrane
permeability and are also highly susceptible to dephosphoryla-
tion by phosphatases. Therefore, several neutral and mem-
brane-permeable pronucleotides (prodrugs)[8] have been devel-
oped over the past decades. One of them is the phosphorami-
date prodrug introduced by McGuigan and co-workers in the
1990s,[9,10] which relies on intracellular enzymatic activation of
the prodrug to release the desired nucleoside monophosphate
inside the cell. The postulated mechanism of phosphoramidate
activation has been explained in several review articles.[10–12]
With this phosphoramidate triester ProTide approach, at least
four different aryloxy triester phosphoramidates (GS-9140, GS-
9131, thymectacin (NB1011), and stampidine) have entered
clinical trials.[12]
which may shed light on this hypothesis, so that the modified
nucleoside monophosphate is released intracellularly. Apart
from using the methyl ester of l-alanine (as previously synthe-
sized by Harris et al.)[13] as the amidate counterpart, we have
also synthesized the methyl and pivaloyloxymethyl (POM)
esters of iminodiacetic acid (IDA)[14] to evaluate the compara-
tive antiviral activities. As promoieties, POM[15] and isopropylox-
ymethylcarbonyl (POC)[16] have previously been introduced suc-
cessfully to increase the oral bioavailability of the parent
drugs. For example: adefovir dipivoxil[17,18] and tenofovir diso-
proxil fumarate (TDF)[19] are two FDA-approved POM- or POC-
derived antiviral prodrugs on the market, along with some
POM-derived antibiotics[18] and anticancer[20] prodrugs. The dia-
lkyl amine IDA has been selected here because we have
shown that IDA-dAMP (a phosphoramidate conjugate of IDA
and deoxyadenosine monophosphate (dAMP)) can act as
a direct substrate for viral polymerases (IDA acts as a good
leaving group) in an enzymatic assay and can incorporate the
nucleotide moiety into DNA (that is, no triphosphate formation
is required).[14] We tested the hypothesis that after the first es-
terase-mediated cleavage of the carboxyl ester group and the
subsequent release of the aryl alcohol, the phosphoramidic
acid formed could be directly recognized by the viral poly-
merase, to show an antiviral effect. Alternatively, intracellular
cleavage of the PꢀN bond by a nonenzymatic mechanism may
bypass the final step of metabolic activation in the pronucleo-
tide approach of McGuigan, Wagner and their respective co-
workers, that is, the action of the phosphoramidase enzyme to
cleave the PꢀN bond. The presence of the phosphoramidase
enzyme is considered to be cell-
Viewing this success, we were interested in synthesizing
a series of aryloxy phosphoramidate ProTide analogues of
BVDU and of several deazaadenine nucleosides (Figure 2).
BVDU is an ideal prototype to test a new nucleotide-prodrug
strategy, because HSV TKꢀ cells are available, in which BVDU is
not phosphorylated into its monophosphate (MP) form. 2’-De-
oxynucleosides with a deazaadenine base moiety (2–5) have
not been described as antiviral nucleoside analogues, which
may be due to lack of their intracellular phosphorylation.
Therefore, it is interesting to apply the ProTide approach,
type specific.[21] A pure chemical
degradation mechanism of the
prodrug could have the advant-
age of a cell-line-independent
mechanism of nucleotide deliv-
ery. Based on these successful
applications, we were interested
to pursue a prodrug approach
by combining the POM and IDA
moieties for the intracellular de-
livery of our nucleotides. Herein,
we report the synthesis of
a series of aryloxy phosphorami-
date ProTide analogues and
present their antiviral activity.
We also present the antiviral ac-
tivity of l-aspartic acid phos-
phoramidate diesters (containing
a negatively charged phosphate
group) of deazaadenine nucleo-
sides. This concept was first de-
veloped by Wagner and co-
workers.[22,23] The purpose of
choosing aspartic acid here is
that Asp-dAMP (a phosphorami-
date conjugate) has also been
Figure 2. Aryloxy phosphoramidate ProTide analogues and the corresponding nucleosides used in this study.
shown to be a direct substrate
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 10
&
2
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
for viral polymerases.[24] In addition, we have shown that the
aspartic acid phosphoramidate analogues of certain deazaade-
nine nucleosides could undergo spontaneous PꢀN bond cleav-
age to release the desired modified nucleotides.[25]
filtration step. Therefore, we have chosen the method of Lehs-
ten et al. to synthesize the series of phosphoramidate triesters
(Scheme 2) and have also successfully employed the protocol
Results and Discussion
Chemistry
The synthesis of POM-protected iminodiacetic acid hydrochlo-
ride salt 9 was carried out by starting with commercially avail-
able IDA (6), as shown in Scheme 1. After the protection of the
Scheme 2. Synthesis of aryloxy phosphoramidate ProTide analogues. Re-
agents and conditions: a) N-methylimidazole, dry CH2Cl2, ꢀ158C!RT;
b) requisite nucleoside (1/2/3/5), dry CH2Cl2, ꢀ58C!RT.
Scheme 1. Synthesis of the pivaloyloxymethyl iminodiacetate (IDA-POM) hy-
drochloride salt 9. Reagents and conditions: a) 2m NaOH, benzyl chlorofor-
mate, RT, 4 h; b) Ag2CO3, pivaloyloxymethyl iodide (11), dry DMF, RT, over-
night; c) H2, 10% Pd/C, MeOH, 5n HCl in iPrOH, RT, 4 h; d) NaI, MeCN, 308C,
overnight.
to couple the secondary amine salts to the phosphate moiety
by varying the reaction time and temperature (see the Experi-
mental Section). The phosphoramidate analogues were puri-
fied by flash column chromatography (twice) and were fully
characterized by spectroscopic and analytical techniques. The
overall yields of the reactions are in the range of 25–70%.
31P NMR investigations of the phosphoramidates displayed the
presence of two diastereomers resulting from the mixed phos-
phate stereochemistry.
imino nitrogen atom by using the carboxybenzyl (Cbz)
group,[26] the carboxylic acid groups were protected with the
POM moiety by using pivaloyloxymethyl iodide (POM-I, 11),[27]
which was prepared from pivaloyloxymethyl chloride (10) and
stored at ꢀ208C before use. As a base, both silver carbonate[28]
and cesium carbonate were tried. The yield of the reaction was
found to be better with silver carbonate. Finally, the Cbz group
was removed under a H2 atmosphere in the presence of Pd/C
and alcoholic hydrochloric acid to yield the POM ester of IDA
in the form of the hydrochloride salt 9.
All phenoxy phosphoramidate triesters (1a–c, 2a–c, 3a,b,
and 5a,b; Figure 2) were synthesized by phosphorochloridate
chemistry by using an optimized method, as described by
Lehsten et al.[29] The optimum reaction conditions (in ref. [29])
were found to be as follows: 2.5 equiv phenyl dichlorophos-
phate (PhOP(O)Cl2) with 3.5 equiv l-alanine methyl ester hydro-
chloride (1.4 equiv with respect to PhOP(O)Cl2) and 10 equiv N-
methylimidazole (NMI) in CH2Cl2. The temperature for the first
step was ꢀ10 to 08C, and for the second step was ꢀ58C to
room temperature with more than 99% purity even on multi-
gram scale. This method was first developed by McGuigan
et al.[30] with a stoichiometric amount of PhOP(O)Cl2 and an
amino acid HCl salt in the presence of triethylamine at very
low temperature (ꢀ788C). This was followed by a dry filtration
step to remove the Et3NHCl salt, and the filtrate was used in
the next step without further purification. The method opti-
mized by Lehsten et al. offers advantages in reaction condi-
tions, including the ability to perform the reaction at a temper-
ature substantially higher than ꢀ788C and omission of the first
The synthesis of the methylester-protected aspartic acid
phosphoramidate diesters 2d, 4a, and 5c (Figure 2) was per-
formed according to the method described by Wagner and co-
workers[23] by starting from the nucleoside monophosphate
and using dicyclohexylcarbodiimide (DCC)-mediated coupling.
The synthetic protocol and characterization data for com-
pounds 2, 5, 2d, and 5c have previously been communicated
in our earlier work.[25] The synthesis of compounds 3 and 4
was carried out by following previously reported methods[31]
with optimization for some of the reaction steps. The phos-
phoramidate diester 4a was synthesized by the method re-
ported in our earlier work.[25] The synthetic steps and the char-
acterization data are provided in the Supporting Information.
Antiviral activity
The modified nucleosides 1–5 and the corresponding ProTides
1a–5c have been evaluated for their potential inhibitory activi-
ty against a wide variety of DNA viruses in HEL cell cultures, in-
cluding thymidine kinase deficient HSV-1 and VZV strains. As
reference compounds, BVDU (1), Cidofovir, Acyclovir, and Gan-
ciclovir were included (Tables 1 and 2). The ProTides of BVDU
1a (EC50: (0.3ꢁ0.1) mm) and 1c (EC50: (1.1ꢁ0.4) mm) exhibited
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 10
&
3
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
potent anti-HSV-1 activity (being only three- to ten-
fold less effective than the parent BVDU compound)
but, as is also the case for the parent compound, lost
activity against thymidine kinase deficient (TKꢀ) virus
strains (Table 1). This seems to indicate a lack of ap-
preciable intracellular nucleotide delivery for the
BVDU-MP prodrugs and also supports the require-
ment for the presence of the virus-encoded thymi-
dine kinase as the activating enzyme to obtain even-
tual antiviral activity. Compound 1c showed approxi-
mately fourfold-decreased activity against HSV-
1 when compared with the activity of compound 1a,
for which the antiviral activity was published previ-
ously and correlated well with our data.[13] The de-
creased activity of compound 1c might be due to its
higher lipophilicity, which could have posed a solubili-
ty problem in the biological assay medium. Alterna-
tively, it could be that this compound is poorly pro-
cessed by intracellular enzymes to release either the
active nucleoside (BVDU, 1), or the nucleotide (BVDU-
monophosphate) because the phosphoramidase
enzyme is poorly expressed in the embryonic lung
cell line that has been used for the assays.[21] The
poor activity might also be due to an insufficient ki-
netic profile of the phosphoramidic acid conjugates
for viral polymerase for incorporation into a growing
DNA chain and action as a chain terminator.
Table 1. Anti-HSV and anti-vaccinia activities and cytotoxicity of ProTide analogues in
human embryonic lung (HEL) cell cultures.
Compound
EC50 [mm][a]
MCC [mm][b]
HSV-1 (KOS)
HSV-1 TKꢀ KOS ACVr HSV-2 (G)
VV
10
ꢂ20
>100
>4
>100
>100
>100
>20
>100
>100
>100
>20
Brivudin (1)
1a
1b
1c
2
2a
2b
2c
2d
3
3a
3b
4
4a
5
5a
0.1ꢁ0.06
0.3ꢁ0.1
>100
1.1ꢁ0.4
>100
>100
>100
>20
>100
>100
>100
>20
>100
>100
100
>100
>100
>100
2
250
>20
>100
>4
250
>20
>100
>4
>100
>100
>100
>20
>100
>100
>100
>20
>250
ꢂ100
>100
100
>100
>100
>100
100
>100
>100
>100
100
>100
>100
>100
>100
ꢂ100
>100
>250
>250
>100
>100
>100
>100
>20
>100
>100
>100
>20
>100
>100
ꢂ100
>100
>100
>100
2
>100
>100
>100
>100
69ꢁ31 >100
>100
>100
>100
>100
10
>250
>100
5b
5c
Cidofovir
Acyclovir
Ganciclovir
>100
>100
2
0.2
0.02
50
14
0.4
0.02
[a] Effective concentration required to inhibit virus-induced cytopathicity by 50%.
HSV-1: herpes simplex virus-1; HSV-2: herpes simplex virus-2; VV: vaccinia virus.
[b] Minimum cytotoxic concentration or compound concentration required to result
in a microscopically visible alteration of cell morphology.
Compound 1a (EC50: 0.21 mm) and 1c (EC50:
However, almost equal EC50 values are obtained for the BVDU
0.34 mm) showed nearly 7- to 11-fold decreased activity in HEL
cell cultures against varicella-zoster virus when compared with
the activity of the drug/parent nucleoside BVDU (1, Table 2).
ProTides bearing an amino acid l-alanine methyl ester (1a)
and a secondary amine ester pivaloyloxymethyl iminodiacetate
(1c). Apparently, the amino acid can also be replaced by a sec-
Table 2. Antiviral activity of ProTide analogues against varicella-zoster virus (VZV) and cytomegalovirus (HCMV) in human embryonic lung (HEL) cell cul-
tures.
Compound
Anti-VZV EC50 [mm][a]
Cytotoxicity [mm]
Anti-HCMV EC50 [mm][a]
AD-169 strain Davis strain
Cytotoxicity [mm]
[c]
TK+ VZV strain OKA
TKꢀ VZV strain 07-1
MCC[b]
CC50
MCC[b]
Brivudin (1)
1a
1b
1c
2
2a
2b
2c
0.03ꢁ0.007
0.21ꢁ0.02
143ꢁ8.5
>20
>20
>300
100
ꢂ100
100
>100
>100
ꢂ100
100
201ꢁ47
>100 >100
>100
100
>100
100
>100
>100
>100
100
51.7ꢁ12
>20
>100
>20
>20
>100
>20
>20
–
0.34ꢁ0.08
>100
>100
>20
20
43ꢁ4.8
>100
>100
>100
>20
–
–
–
–
>100
>100
>100
>20
>100
>100
>100
>20
>20
2d
3
3a
3b
4
4a
5
5a
5b
5c
Acyclovir
Ganciclovir
Cidofovir
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
20ꢁ8.3
>100
87ꢁ12
>100
131ꢁ27
–
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>440
–
–
–
–
–
–
–
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
–
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
–
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
–
2.4ꢁ0.8
74ꢁ7.5
68ꢁ31
37.5ꢁ2
–
–
–
>100
2.3ꢁ0.2
>440
–
–
–
–
7.5
0.8
5.2
0.8
ꢂ350
>300
–
–
[a] Effective concentration required to inhibit virus-induced cytopathicity by 50%. [b] Minimum cytotoxic concentration or compound concentration re-
quired to result in a microscopically visible alteration of cell morphology. [c] Cytotoxic concentration required to decrease cell growth by 50%.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 10
&
4
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
ondary amine moiety. Moreover, compound 1c (the IDA-POM
ester of BVDU phosphoramidate) retained moderate antiviral
activity against the TKꢀ VZV strain, whereas the parent com-
pound BVDU (1) showed no activity at all. This result indicates
that some degree of kinase bypass may be achieved and that,
to some extent, BVDU-monophosphate is being delivered in-
tracellularly or the phosphoramidic acid formed might have
been recognized as a direct substrate by the viral enzyme to
exert some antiviral activity. However, the apparent cytostatic
activity being tenfold higher for compound 1c than for BVDU
could also be related to the observed (marginal) antiviral activi-
ty of 1c against the VZV TKꢀ strain.
Stability in human serum
In order to investigate the stability of these phosphoramidate
analogues in biological conditions, the stabilities of two repre-
sentative ProTides, 1a and 1b, were investigated at 378C in
the presence of human serum by using 31P NMR spectroscopy.
The ProTide was dissolved in a mixture of [D6]DMSO and D2O,
and then human serum was added. The reaction was moni-
tored over a period of 14 h, as depicted in Figure 3.
From Tables 1 and 2, it is evident that compound 1c (the
IDA-POM ester of BVDU phosphoramidate) has significant anti-
viral activity against TK+ HSV-1 and VZV, whereas compound
1b (the IDA-Me ester of BVDU phosphoramidate) is completely
inactive. This result suggests that, between the methyl and
POM esters of iminodiacetic acid, the POM ester has a much
greater potency than the methyl ester, probably due to its
higher lipophilicity assisting permeation through the lipid-rich
cell membrane. This suggests that the POM ester might be
a potential protecting group to consider for the future design
of phosphoramidate prodrugs. Although there are some con-
cerns for toxicity, due to liberation of formaldehyde and pivalic
acid (which interrupts carnitine homeostasis in humans) during
POM-derived-prodrug metabolism, it is suggested that normal
formaldehyde input from the diet and the environment ex-
ceeds the exposure from formaldehyde-releasing prodrugs,
and simultaneous carnitine supplementation may be adminis-
tered with pivalate-generating prodrugs in case of long-term
treatment.[18,32]
The antiviral activities of deazaadenine nucleoside deriva-
tives have been reviewed by Cristalli and co-workers,[33] and
some of these compounds showed antiviral activity. For exam-
ple, 7-deaza-2’-methyladenosine and its derivatives were re-
ported as highly potent agents against the hepatitis C
virus.[33,34] We were therefore interested to evaluate the antivi-
ral activity of some deazaadenine nucleosides and the corre-
sponding l-alanine methyl and IDA-POM containing phosphor-
amidate derivatives. Among all of the deazaadenine nucleo-
sides and phosphoramidate derivatives evaluated, only the 7-
deaza-nucleoside 5 (EC50: 2.4 mm) showed significant activity
against the VZV TK+ strain, although its potency against the
TKꢀ VZV strain was eightfold lower. The triester phosphorami-
date analogues 5a and 5b (EC50: 37–68 mm) exhibited very
weak anti-VZV activity; the activities are nearly 15- to 28-fold
lower than the corresponding nucleoside 5.
Figure 3. Stack plot of 31P NMR spectra illustrating the stability of ProTides
A) 1a and B) 1b in human serum at various time intervals; control: no
serum added.
The spectra show that both of the ProTides were reasonably
stable under these conditions. In fact, more than 90% of the
parent compound was still present after 14 h of incubation at
378C. Compound 1a was partly hydrolyzed into the proposed
metabolite 15 (Figure 4) because a single peak appeared at
d=7.07 ppm (Figure 3A). For compound 1b, a peak appeared
at d=5.61 ppm (Figure 3B), which might correspond to the
metabolite 16 (Figure 4). The proposed metabolites were not
characterized further, and the prediction is based on previously
published literature, which reported similar 31P chemical shifts
The antiviral activities of some phosphoramidate diesters
bearing l-aspartyl methyl esters (2d, 4a, and 5c) were also
evaluated, but none of these compounds showed antiviral ac-
tivity against any of the tested viruses. No antiviral activity was
observed against HSV-2, vaccinia virus (VV), or human cytome-
galovirus (HCMV) for any of the compounds in this study.
Figure 4. Tentative proposals for the metabolites formed during the serum
stability studies with ProTides 1a and 1b.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 10
&
5
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
that were proved to be characteristic 31P peaks due to the for-
mation of similar metabolites for this type of molecule in
human serum.[35] The two small peaks at dꢃ1.77 and
ꢀ0.32 ppm are derived from human serum, as confirmed with
blank serum. Therefore, from the data, it could be suggested
that this type of phosphoramidate derivative could be reasona-
bly stable in the extracellular medium to allow intact uptake
by the target cells.
hydrous magnesium sulfate and concentrated under vacuum to
give iodomethyl pivalate (11) as a yellow liquid (85%): 1H NMR
(300 MHz, CDCl3): d=5.89 (s, 2H, CH2), 1.16 ppm (s, 9H, C(CH3)3);
13C NMR (75 MHz, CDCl3): d=176.5, 39.1, 31.7, 26.8 ppm.
N-Benzyloxycarbonyl iminodiacetic acid (7): Benzyl chloroformate
(18 mL, 0.13 mol) was added dropwise to a solution of iminodiace-
tic acid, 6 (15 g, 0.1 mol) in NaOH (2m, 600 mL) at 08C. The mixture
was allowed to stir at room temperature for 4 h. The mixture was
then acidified to pH 2 with HCl (2m) and extracted three times
with diethyl ether. The combined diethyl ether layers were dried
over anhydrous MgSO4, and the solvent was evaporated under re-
duced pressure to obtain the crude product as an oil. The crude
product was then purified on a silica gel column eluted with ethyl
acetate/hexane (6:4) to obtain the pure product (7) as a colorless
Conclusions
A series of phosphoramidate ProTide analogues of BVDU and
of deazaadenine nucleosides were synthesized and investigat-
ed for their antiviral activity and cytotoxicity against DNA virus-
es. The BVDU aryloxy phosphoramidates of l-alanine methyl
ester and pivaloyloxymethyl iminodiacetate ester showed sig-
nificant activity against HSV-1 and VZV, although relatively less
activity than the corresponding parent drug BVDU. The amino
acid l-alanine methyl ester and secondary amine IDA-POM
ester were found to be equally potent against varicella-zoster
virus. Some success was achieved in delivering the nucleoside
monophosphate intracellularly for phosphoramidate com-
pounds bearing the IDA-POM moiety (for example, 1c) against
VZV to bypass the thymidine kinase pathway, which indicates
the POM moiety is a potential carboxyl protecting group in
the design of prodrugs. The application of the phosphorami-
date ProTide technology was only partly successful because
none of these compounds showed a significant increase in an-
tiviral activity against the TK-deficient virus strains.
1
oil (70%): H NMR (300 MHz, CDCl3): d=9.88 (brs, 2H, OH), 7.31–
7.28 (m, 1H, aromatic H), 5.15 (s, 2H, CH2), 4.18 (s, 2H, CH2),
4.13 ppm (s, 2H, CH2); 13C NMR (75 MHz, CDCl3): d=174.6, 173.9,
156.4, 135.9, 128.9, 128.7, 128.2, 68.8, 50.2, 50.1 ppm; HRMS (ESIꢀ):
calcd for C12H12NO6 [MꢀH]ꢀ: 266.0670; found: 266.0674.
Bis(pivaloyloxymethyl)-N-benzyloxycarbonyl iminodiacetate (8):
Silver carbonate (32.5 g, 118 mmol) and pivaloyloxymethyl iodide
(11; 35.4 mL, 0.24 mol) were added to a solution of 7 (6.3 g,
23.6 mmol) in dry N,N-dimethylformamide (DMF; 120 mL). The
color of the reaction mixture changed from greenish to brown in
15 min and was allowed to stir at room temperature overnight.
The solid particles were removed by filtration and washed with
EtOAc. The filtrate was concentrated under reduced pressure and
purified by flash column chromatography eluted with EtOAc/
hexane (1:9 to 2:8) to obtain 8 as a colorless oil (87%): Rf =0.2
1
(EtOAc/hexane, 1:9); H NMR (300 MHz, CDCl3): d=7.34 (m, 5H, ar-
omatic H), 5.79 (s, 2H, CH2), 5.73 (s, 2H, CH2), 5.14 (s, 2H, CH2), 4.21
(s, 2H, CH2), 4.15 (s, 2H, CH2), 1.21 (s, 9H, C(CH3)3), 1.17 ppm (s, 9H,
C(CH3)3); 13C NMR (75 MHz, CDCl3): d=177.15, 177.07, 168.4, 168.3,
155.9, 136.0, 128.7, 128.4, 127.9, 80.1, 80.0, 68.3, 49.3, 49.1, 38.9,
38.8, 27.0, 26.9 ppm; HRMS (ESI+): calcd for C24H34NO10 [M+H]+:
496.2177; found: 496.2171.
Experimental Section
General: NMR spectra were recorded on a Bruker Avance II
300 MHz spectrometer with 5 mm broad band probe and on
a 500 MHz spectrometer, which is equipped with a TXI-HCP Z gra-
dient probe. The spectra were processed with Bruker Topspin 2.1
software. Chemical shifts (d) were expressed in parts per million
Bis(pivaloyloxymethyl)iminodiacetate hydrochloride salt (9):
10% Pd/C (2 g, 8.3 mmol) was added to a solution of 8 (10.2 g,
20.6 mmol) in methanol (500 mL). HCl (5n in isopropanol, 4 mL)
was added to the resulting mixture, which was then stirred under
H2 at room temperature for 4 h. The catalyst was filtered over
Celite and washed with methanol. The filtrate was evaporated to
dryness and triturated with diethyl ether to obtain pure compound
9 as a white solid (98%): Rf =0.14 (EtOAc/hexane, 3:7); 1H NMR
(300 MHz, CDCl3): d=5.86 (s, 4H, CH2), 4.16 (s, 4H, CH2), 1.22 ppm
(s, 18H, C(CH3)3); 13C NMR (75 MHz, CDCl3): d=177.2, 167.8, 80.4,
48.2, 39.0, 27.1 ppm; HRMS (ESI+): calcd for C16H28NO8 [M+H]+:
362.1809; found: 362.1808.
1
(ppm). The H and 13C NMR chemical shifts were referenced relative
to a tetramethylsilane peak (d=0.00 ppm). All signals in the
proton and carbon 1D NMR spectra were assigned by using 2D
COSY, 1H–13C (natural abundance) HSQC, and 1H–13C HMBC NMR
spectra. Mass spectra were acquired on a quadrupole orthogonal
acceleration time-of-flight mass spectrometer (Synapt G2 HDMS,
Waters, Milford, MA, USA). Samples were infused at 3 mLminꢀ1, and
spectra were obtained in positive (or negative) ionization mode
with a resolution of 15000 (FWHM) by using leucine enkephalin as
a lock mass. Chemicals of analytical and synthetic grade were ob-
tained from commercial sources and were used as received. Flash
silica column chromatography was performed on 60 A, 0.035–
0.070 mm silica gel (Acros Organics). Human serum was purchased
from Sigma Aldrich.
General procedure for synthesis of phosphoramidates: Step 1: A
solution/suspension of the appropriate amine/amino acid ester hy-
drochloride salt (3.5 equiv) in anhydrous CH2Cl2 was prepared and
cooled to ꢀ158C. Dichlorophenylphosphate (2.5 equiv) was added
slowly. After 10 min, a solution of N-methylimidazole (10 equiv) in
dry CH2Cl2 was added dropwise. The mixture was left to stir for
1 h. (In the cases of secondary amines, the mixtures were allowed
to reach room temperature slowly and left for 10–12 h.) Step 2: In
a separate flask, a suspension of the appropriate nucleoside
(1 equiv) in anhydrous CH2Cl2 was cooled to ꢀ58C. With stirring,
the solution prepared above was added slowly over a period of
1 h, with the temperature kept near ꢀ58C. The cooling bath was
removed, and the reaction was left to stir at room temperature
Pivaloyloxymethyl iodide (11): Sodium iodide (45 g, 0.3 mol) was
added to a solution of pivaloyloxymethyl chloride (10; 25 g,
0.17 mol) in acetonitrile (50 mL) and the suspension was allowed
to stir at 308C overnight under an argon atmosphere. A mixture of
CH2Cl2 (250 mL) and water (250 mL) was added. The mixture was
allowed to stir for 10 min, then separated into two layers. The
aqueous layer was discarded and the organic layer was washed
with sodium thiosulfate (2%). The organic layer was dried over an-
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 10
&
6
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
until TLC indicated a reasonable amount of product formation. The
reaction mixture was then evaporated to dryness under reduced
pressure, and the residue was purified by column chromatography
eluted with CH2Cl2/MeOH in different proportions (generally, 2–5%
methanol in CH2Cl2). The yields of the reactions were in the range
of 25–70%. The product structures were confirmed by NMR spec-
parent t, 1H, J1’,2’a =J1’,2’b =6.5 Hz, H1’), 6.48 (d, 1H, J1,2 =5.4 Hz, H1),
5.34 (s, 2H, NH2), 4.75–4.51 (m, 2H, H3’, NH), 4.44–4.43 (m, 2H, H5’,
H5’’), 4.31–4.30 (m, 1H, H4’), 4.08–4.02 (m, 1H, H-a-ala), 3.72, 3.70
(2s, 3H, OCH3), 2.81–2.53 (m, 2H, H2’a, H2’b), 1.39, 1.35 ppm (d,
3H, CH3-ala); 13C NMR (125 MHz, CDCl3): d=173.7, 173.6 (CO-ala),
150.3, 150.2 (phenyl C), 146.5 (C4), 145.6 (C6), 145.1 (C2), 138.3,
138.1 (C8), 129.3 (phenyl C), 124.6 (phenyl C), 123.4 (C5), 119.8
(2phenyl C), 103.0 (C1), 84.7, 85.6 (C4’), 83.6, 83.5 (C1’), 70.6 (C3’),
66.2, 65.9 (C5’), 52.1 (OCH3), 49.9 (C-a-ala), 39.7 (C2’), 20.2,
20.1 ppm (CH3-ala); 31P NMR (202 MHz, CDCl3): d=3.10, 2.90 ppm;
HRMS (ESI+): calcd for C21H27N5O7P [M+H]+: 492.1642; found:
492.1651.
1
troscopy and HRMS, as described below. H, 13C and 31P NMR spec-
tra are provided in the Supporting Information.
Brivudin-5’-[phenyl(methoxy-l-alaninyl)]phosphate (1a): Yield:
60%; Rf =0.47 (CH2Cl2/MeOH, 9.5:0.5); 1H NMR (500 MHz, CDCl3):
d=7.63, 7.60 (2s, 1H, H6), 7.41, 7.38 (2d, 1H, J=13.7 Hz, CH=
CHBr), 7.33–7.14 (m, 5H, OPh), 6.71, 6.67 (2d, 1H, J=13.7 Hz, CH=
CHBr), 6.27, 6.22 (two sets of apparent t, 1H, J1’,2’a =J1’,2’b =6.3 Hz,
H1’), 4.52, 4.47 (2m, 1H, H3’),4.43–4.19 (m, 3H, H5’, H5’’, NH), 4.17–
4.09 (m, 1H, H4’), 4.06–3.95 (m, 1H, H-a-ala), 3.70, 3.69 (2s, 3H,
OCH3), 2.44–2.38 (m, 1H, H2’a), 2.06–2.02 (m, 1H, H2’b), 1.37,
1.35 ppm (d, 3H, CH3-ala); 13C NMR (125 MHz, CDCl3): d=174.3,
174.2 (CO-ala), 161.9, 161.8 (C4), 150.4, 150.3 (phenyl C), 149.6 (C2),
137.8, 137.7 (C6), 130.0 (phenyl C), 128.7 (CH=CHBr),125.5 (phenyl
C), 120.3–120.2 (phenyl C), 111.7, 111.6 (C5), 110.0 (CH=CHBr), 85.8,
85.4 (C1’), 85.3 (C4’), 70.9, 70.5 (C3’), 66.2, 65.9 (C5’), 52.9, 52.8
(OCH3), 50.5, 50.4 (C-a-ala), 40.6, 40.4 (C2’), 20.9, 20.8 ppm (CH3-
ala); 31P NMR (202 MHz, CDCl3): d=3.47, 3.17 ppm; HRMS (ESI+):
calcd for C21H26BrN3O9P [M+H]+: 574.0585; found: 574.0579.
2’-Deoxy-1-deazaadenosine-5’-[phenylbis(methoxyiminodiace-
tyl)] phosphate (2b): Yield: 62%; Rf =0.55 (CH2Cl2/MeOH, 9.0:1.0);
1H NMR (300 MHz, CDCl3): d=8.15, 8.00 (2s, 1H, H8), 7.97 (d, 1H,
J
J
2,1 =5.5 Hz, H2), 7.28–7.08 (m, 5H, OPh), 6.54 (apparent t, 1H,
1’,2’a =J1’,2’b =6.6 Hz, H1’), 6.41 (d, 1H, J1, 2 =5.5 Hz, H1), 5.15 (s, 2H,
NH2), 4.75–4.55 (m, 1H, H3’), 4.44–4.32 (m, 2H, H5’, H5’’), 4.26–4.13
(m, 1H, H4’), 4.09–3.86 (m, 4H, CH2-IDA), 3.63, 3.61 (2s, 3H, OCH3),
2.78–2.38 ppm (m, 2H, H2’a, H2’b); 13C NMR (75 MHz, CDCl3): d=
169.9, 169.8 (CO-IDA), 150.3, 150.2 (phenyl C), 146.6 (C4), 145.5
(C6), 145.1 (C2), 138.2, 138.0 (C8), 129.7, 129.4, 128.9 (phenyl C),
124.8 (phenyl C), 123.5 (C5), 119.9, 119.8, 119.7 (phenyl C), 102.9
(C1), 84.6–84.4 (C4’), 83.4, 83.2 (C1’), 70.8, 70.7 (C3’), 66.3, 65.8 (C5’),
51.8, 51.7 (OCH3), 47.6, 47.5 (CH2-IDA), 39.7 ppm (C2’); 31P NMR
(202 MHz, CDCl3): d=4.47, 4.37 ppm; HRMS (ESI+): calcd for
C23H29N5O9P [M+H]+: 550.1697; found: 550.1704.
Brivudin-5’-[phenylbis(methoxyiminodiacetyl)]phosphate (1b):
Yield: 60%; Rf =0.46 (CH2Cl2/MeOH, 9.5:0.5); 1H NMR (500 MHz,
CDCl3): d=7.65, 7.59 (2s, 1H, H6), 7.41, 7.39 (2d, 1H, J=13.6 Hz,
CH=CHBr), 7.32–7.13 (m, 5H, OPh), 6.72, 6.67 (2d, 1H, J=13.6 Hz,
2’-Deoxy-1-deazaadenosine-5’-[phenylbis(pivaloyloxymethylimi-
nodiacetyl)] phosphate (2c): Yield: 70%; Rf =0.6 (CH2Cl2/MeOH,
CH=CHBr), 6.28, 6.25 (two sets of apparent t, 1H, J1’,2’a =J1’,2’b
=
1
6.5 Hz, H1’), 4.51–4.48 (m, 1H, H3’), 4.43–4.42 (m, 2H, H5’, H5’’),
4.11–3.87 (m, 5H, H4’, CH2-IDA), 3.66, 3.65 (2s, 6H, OCH3), 2.45–
2.36 (m, 1H, H2’a), 2.07–2.01 ppm (m, 1H, H2’b); 13C NMR
(125 MHz, CDCl3): d=170.6, 170.4 (CO-IDA), 162.0, 161.9 (C4),
150.6, 150.5, 150.4 (phenyl C), 149.7 (C2), 138.0, 137.9 (C6), 130.3–
130.09 (phenyl C), 128.9, 128.8 (CH=CHBr), 125.7 (phenyl C), 120.4–
120.3 (phenyl C), 111.8, 111.7 (C5), 110.2, 110.1 (CH=CHBr), 85.5,
85.4, 85.3, 85.2 (C1’ and C4’), 70.7, 70.6 (C3’), 66.5, 66.4 (C5’), 52.7,
52.6 (OCH3), 48.4 (CH2-IDA), 40.7, 40.6 ppm (C2’); 31P NMR
(202 MHz, CDCl3): d=4.81, 4.78 ppm; HRMS (ESI+): calcd for
C23H28BrN3O11P [M+H]+: 632.0640; found: 632.0646.
9.0:1.0); H NMR (500 MHz, CDCl3): d=8.09 (d, 1H, J2, 1 =5.6 Hz, H2),
8.06 (s, 1H, H8), 7.41–7.22 (m, 5H, OPh), 6.62 (apparent t, 1H,
J
1’,2’a =J1’,2’b =6.6 Hz, H1’), 6.53 (d, 1H, J1, 2 =5.6 Hz, H1), 5.85–5.81
(m, 4H, CH2-POM), 5.14 (s, 2H, NH2), 4.75–4.72 (m, 1H, H3’), 4.52–
4.45 (m, 2H, H5’, H5’’), 4.32–4.26 (m, 1H, H4’), 4.19–4.06 (m, 4H,
CH2-IDA), 2.91–2.52 (m, 2H, H2’a, H2’b), 1.28 ppm (s, 18H,
tBu);13C NMR (125 MHz, CDCl3): d=176.7, 176.6 (CO-POM), 168.3,
168.2 (CO-IDA), 150.1, 150.0 (phenyl C), 146.5 (C4), 145.5 (C6), 145.1
(C2), 138.3, 138.0 (C8), 129.4, 128.9, 124.9 (phenyl C), 123.6 (C5),
119.8 (2phenyl C), 102.9 (C1), 84.5, 84.3 (C4’), 83.4 (C1’), 79.6 (CH2-
POM), 70.9, 70.7 (C3’), 66.2, 65.9 (C5’), 47.6, 47.5 (CH2-IDA), 39.4,
39.3 (C2’), 38.4 (tBu quaternary C), 26.4 ppm (tBu); 31P NMR
(202 MHz, CDCl3): d=4.19, 4.07 ppm; HRMS (ESI+): calcd for
C33H45N5O13P [M+H]+: 750.2746; found: 750.2747.
Brivudin-5’-[phenylbis(pivaloyloxymethyliminodiacetyl)] phos-
phate (1c): Yield: 40%; Rf =0.55 (CH2Cl2/MeOH, 9.5:0.5); 1H NMR
(500 MHz, CDCl3): d=9.51 (s, 1H, NH), 7.60, 7.58 (2s, 1H, H6), 7.42,
7.39 (2d, 1H, J=13.6 Hz, CH=CHBr), 7.34–7.16 (m, 5H, OPh), 6.70,
6.68 (2d, 1H, J=13.6 Hz, CH=CHBr), 6.26, 6.23 (two sets of appar-
ent t, 1H, J1’,2’a =J1’,2’b =6.3 Hz, H1’), 5.78–5.70 (m, 4H, CH2-POM),
4.50–4.40 (m, 3H, H3’, H5’, H5’’), 4.16–3.94 (m, 5H, H4’, CH2-IDA),
2.46–2.38 (m, 1H, H2’a), 2.08–2.01 (m, 1H, H2’b), 1.20, 1.19 ppm
(2s, 18H, tBu-POM); 13C NMR (125 MHz, CDCl3): d=177.4 (2CO-
POM), 168.9, 168.8 (CO-IDA), 161.7 (C4), 150.4, 150.3 (3phenyl C),
149.5 (C2), 137.9, 137.8 (C6),130.2 (phenyl C), 128.8, 128.7 (CH=
CHBr), 125.9 (phenyl C), 120.4–120.3 (4phenyl C), 111.8, 111.7 (C5),
110.3, 110.2 (CH=CHBr), 85.5 (C1’), 85.2, 85.1 (C4’), 80.4 (CH2-POM),
70.9, 70.4 (C3’), 66.4, 66.3 (C5’), 48.4, 48.3 (CH2-IDA), 40.7, 40.5 (C2’),
39.1 (tBu quaternary C), 27.1 ppm (tBu); 31P NMR (202 MHz, CDCl3):
d=4.55, 4.53 ppm; HRMS (ESI+): calcd for C33H44BrN3O15P [M+H]+
: 832.1688; found: 832.1680.
2’-Deoxy-6-chloro-3-deazaadenosine-5’-[phenyl(methoxy-l-ala-
ninyl)] phosphate (3a): Yield: 50%; Rf =0.51 (CH2Cl2/MeOH,
1
9.5:0.5); H NMR (500 MHz, CDCl3): d=8.30, 8.21 (2s, 1H, H8), 8.15
(d, 1H, J2,3 =5.7 Hz, H2), 7.40, 7.38 (2d, 1H, J3,2 =5.7 Hz, H3), 7.27–
7.06 (m, 5H, OPh), 6.26 (apparent t, 1H, H1’), 4.68–4.62 (m, 1H,
H3’), 4.46–4.20 (m, 4H, NH, H4’, H5’, H5’’), 3.93–3.86 (m, 1H, H-a-
ala), 3.62, 3.60 (2s, 3H, OCH3), 2.53–2.35 (m, 2H, H2’a, H2’b), 1.28,
1.22 ppm (2d, 3H, CH3-ala); 13C NMR (125 MHz, CDCl3): d=174.2–
174.1 (CO-ala), 150.6–150.5 (phenyl C), 143.1, 143.0 (C6), 142.8,
142.7 (C8), 142.0, 141.9 (C2), 139.4, 139.3 (C4), 138.2 (C5), 130.0
(phenyl C), 125.5, 125.4 (phenyl C), 120.2, 120.1 (phenyl C), 106.5,
106.4 (C3), 86.0–85.7 (C1’, C4’), 70.9, 70.8 (C3’), 66.4–65.9 (C5’), 52.8,
52.7 (OCH3), 50.4 (C-a-ala), 40.6, 40.5 (C2’), 20.9–20.8 ppm (CH3-ala);
31P NMR (202 MHz, CDCl3): d=3.23, 2.95 ppm; HRMS (ESI-): calcd
for C21H23ClN4O7P [MꢀH]ꢀ: 509.0998; found: 509.0999.
2’-Deoxy-1-deazaadenosine-5’-[phenyl(methoxy-l-alaninyl)]
phosphate (2a): Yield: 65%; Rf =0.36 (CH2Cl2/MeOH, 9.0:1.0);
1H NMR (500 MHz, CDCl3): d=8.17, 8.13 (2s, 1H, H8), 8.04 (d, 1H,
2’-Deoxy-6-chloro-3-deazaadenosine-5’-[phenylbis(pivaloyloxy-
methyliminodiacetyl)] phosphate (3b): Yield: 40%; Rf =0.64
1
J
2,1 =5.4 Hz, H2), 7.38–7.16 (m, 5H, OPh), 6.64, 6.61 (two sets of ap-
(CH2Cl2/MeOH, 9.5:0.5); H NMR (500 MHz, CDCl3): d=8.31 (s, 0.5H,
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 10
&
7
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
H8 of one diastereomer), 8.19–8.18 (2d, 1H, H2), 8.16 (s, 0.5H, H8
of one diastereomer), 7.43–7.39 (2d, 1H, H3), 7.30–7.11 (m, 5H,
OPh), 6.29–6.26 (two sets of apparent t, 1H, H1’), 5.74–5.69 (m, 4H,
CH2-POM), 4.70–4.62 (m, 1H, H3’), 4.38–4.26 (m, 2H, H5’, H5’’),
4.27–4.19 (m, 1H, H4’), 4.10–3.88 (m, 4H, CH2-IDA), 2.62–2.35 (m,
2H, H2’a, H2’b), 1.18 ppm (s, 18H, tBu); 13C NMR (125 MHz, CDCl3):
d=177.3, 177.2 (CO-POM), 168.8, 168.7 (CO-IDA), 150.5, 150.4
(phenyl C), 143.2, 143.1 (C6), 142.6, 142.5 (C8), 142.0, 141.9 (C2),
139.5, 139.4 (C4), 138.4, 138.3 (C5), 130.1, 130.0 (phenyl C), 125.6,
125.5 (phenyl C), 120.2, 120.1 (phenyl C), 106.4, 106.3 (C3), 85.8–
85.6 (C1’, C4’), 80.3 (CH2-POM), 71.1, 70.9 (C3’), 66.6, 66.3 (C5’), 48.2
(CH2-IDA), 40.5, 40.4 (C2’), 38.9 (tBu quaternary C), 27.0 ppm
(tBu);31P NMR (202 MHz, CDCl3): d=4.46, 4.32 ppm; HRMS (ESI+):
calcd for C33H43ClN4O13P [M+H]+: 769.2247; found: 769.2258.
value or effective compound concentration (expressed in micromo-
lar concentrations) required for decreasing virus-induced cytopa-
thogenicity or viral plaque formation by 50%.
Cytostatic/toxicity assays: Cytotoxicity measurements were based
on the inhibition of cell growth. HEL cells were seeded at a rate of
5ꢁ103 cells per well into 96-well microtiter plates and allowed to
proliferate for 24 h. Medium containing different concentrations of
the test compounds was then added. After three days of incuba-
tion at 378C, the cell number was determined with a Coulter coun-
ter. The cytostatic concentration (expressed in micromolar concen-
trations) was calculated as the CC50 value or the compound con-
centration required for decreasing cell proliferation by 50% relative
to the number of cells in the untreated controls. CC50 values were
estimated from graphic plots of the number of cells (percentage of
control) as a function of the concentration of the test compounds.
Alternatively, cytotoxicity of the test compounds was expressed as
the minimum cytotoxic concentration (MCC) or the compound
concentration that caused a microscopically detectable alteration
of cell morphology.
2’-Deoxy-7-deazaadenosine-5’-[phenyl(methoxy-l-alaninyl)]
phosphate (5a): Yield: 60%; Rf =0.35 (CH2Cl2/MeOH, 9.5:0.5);
1H NMR (500 MHz, CDCl3): d=8.25 (s, 1H, H2), 7.29–7.10 (m, 5H,
OPh), 7.05–7.00 (2d, 1H, J8,7 =3.7 Hz, H8), 6.69–6.65 (two sets of
apparent t, overlapped, H1’), 6.35–6.33 (2d, overlapped, 1H, J7,8
=
3.7 Hz, H7), 5.60 (s, 2H, NH2), 4.62–4.45 (m, 2H, H3’, NH), 4.30–4.26
(m, 2H, H5’, H5’’), 4.14–4.10 (m, 1H, H4’), 4.01–3.96 (m, 1H, H-a-
ala), 3.64, 3.62 (2s, 3H, OCH3), 2.49–2.36 (m, 2H, H2’a, H2’b), 1.31,
1.28 ppm (2d, 3H, CH3-ala); 13C NMR (125 MHz, CDCl3): d=173.8–
173.7 (CO-ala), 156.5 (C6), 151.5, 151.4 (C2), 150.3, 150.2 (phenyl C),
150.0, 149.9 (C4), 129.4 (phenyl C), 124.7 (phenyl C), 121.3, 121.2
(C8), 119.8–119.7 (phenyl C), 103.3, 103.2 (C5), 98.9 (C7), 83.9, 83.8
(C4’), 82.8, 82.7 (C1’), 70.9, 70.8 (C3’), 66.2, 65.9 (C5’), 52.2, 52.1
(OCH3), 49.9, 49.8 (C-a-ala), 39.6 (C2’), 20.4, 20.3 ppm (CH3-ala);
31P NMR (202 MHz, CDCl3): d=3.26, 2.99 ppm; HRMS (ESI+): calcd
for C21H27N5O7P [M+H]+: 492.1642; found: 492.1644.
Stability assay in human serum: The experiment was carried out by
dissolving ProTide 1a and 1b (5–6 mg) in [D6]DMSO (0.06 mL) and
D2O (0.18 mL). After recording the 31P NMR spectra at 378C as
a control, human serum (0.35 mL) was added to the sample. Next,
a series of 31P NMR spectra were recorded at 378C over a period of
14 h. The 31P NMR data were processed and analyzed with the
Bruker Topspin 2.1 program.
Acknowledgements
2’-Deoxy-7-deazaadenosine-5’-[phenylbis(pivaloyloxymethylimi-
nodiacetyl)] phosphate (5b): Yield: 25%; Rf =0.26 (CH2Cl2/MeOH,
9.65:0.35); H NMR (500 MHz, CDCl3): d=8.26 (s, 1H, H2), 7.30–7.16
This work was financed by FWO and KU Leuven grants GOA and
IDO. We are indebted to Prof. Jef Rozenski (KU Leuven) for provid-
ing HRMS data. Mass spectrometry was made possible by the
support of the Hercules Foundation of the Flemish Government
(grant 20100225-7). We thank Luc Baudemprez for NMR technical
assistance, Frieda De Meyer, Lies Van den Heurck, Steven Car-
mans, and Anita Camps for assistance with the antiviral assays,
and Chantal Biernaux for excellent editorial help.
1
(m, 5.5H, OPh, H8 of one diastereomer), 7.02 (d, 0.5H, J8,7 =3.3 Hz,
H8 of one diastereomer), 6.70–6.68 (two sets of apparent t, over-
lapped, H1’), 6.39–6.38 (2d, 1H, H7), 5.73–5.63 (m, 6H, CH2-POM,
NH2), 4.65–4.58 (m, 1H, H3’), 4.35–4.33 (m, 2H, H5’, H5’’), 4.12–3.92
(m, 5H, H4’, CH2-IDA), 2.59–2.40 (m, 2H, H2’a, H2’b), 1.25, 1.18 ppm
(2s, 18H, tBu);13C NMR (125 MHz, CDCl3): d=177.3, 177.2 (CO-
POM), 168.8, 168.7 (CO-IDA), 156.9 (C6), 151.7, 151.6 (C2), 150.7–
150.5 (phenyl C, C4), 129.9 (phenyl C), 125.4 (phenyl C), 121.9–
121.6 (C8), 120.2 (phenyl C), 103.7, 103.6 (C5), 99.4–99.2 (C7), 84.3–
84.2 (C4’), 83.2–82.9 (C1’), 80.1, 80.0 (CH2-POM), 71.2–71.1 (C3’),
66.8–66.3 (C5’), 48.0–47.9 (CH2-IDA), 40.1, 39.9 (C2’), 38.9 (tBu qua-
ternary C), 27.5, 26.9 ppm (tBu); 31P NMR (121 MHz, CDCl3): d=4.08,
3.94 ppm; HRMS (ESI+): calcd for C33H45N5O13P [M+H]+: 750.2746;
found: 750.2729.
Keywords: antiviral agents · drug delivery · nucleosides ·
phosphoramidates · protecting groups
[1] a) E. De Clercq, Adv. Virus Res. 2009, 73, 1–53; b) G. Antonelli, O. Turri-
ziani, Int. J. Antimicrob. Agents 2012, 40, 95–102.
[2] E. De Clercq, Med. Res. Rev. 2005, 25, 1–20.
[3] H. S. Allaudeen, J. W. Kozarich, J. R. Bertino, E. De Clercq, Proc. Natl.
Acad. Sci. USA 1981, 78, 2698–2702.
[4] J. A. Fyfe, Mol. Pharmacol. 1982, 21, 432–437.
Antiviral assays: The compounds were evaluated against the fol-
lowing viruses: HSV-1 strain KOS, thymidine kinase deficient (TKꢀ)
HSV-1 KOS strain resistant to ACV (ACVr), HSV-2 strain G, vaccinia
virus Lederle strain, varicella-zoster virus (VZV) strain Oka, TKꢀ VZV
strain 07-1, and human cytomegalovirus (HCMV) strains AD-169
and Davis. The antiviral assays were based on inhibition of virus-in-
duced cytopathicity or plaque formation in human embryonic lung
(HEL) fibroblasts. Confluent cell cultures in microtiter 96-well plates
were inoculated with 100 CCID50 of virus (with 1 CCID50 being the
virus dose required to infect 50% of the cell cultures) or with 20
plaque-forming units (PFU; VZV) in the presence of varying con-
centrations of the test compounds. Viral cytopathicity or plaque
formation (VZV) was recorded as soon as it reached completion in
the control virus-infected cell cultures that were not treated with
the test compounds. Antiviral activity was expressed as the EC50
[5] a) J. Balzarini, Pharm. World Sci. 1994, 16, 113–126; b) C. R. Wagner, S. L.
Chang, G. W. Griesgraber, H. Song, E. J. McIntee, C. L. Zimmerman, Nu-
cleosides Nucleotides 1999, 18, 913–919; c) S. Chang, G. W. Griesgraber,
P. J. Southern, C. R. Wagner, J. Med. Chem. 2001, 44, 223–231.
[6] J. Balzarini, H. Egberink, K. Hartmann, D. Cahard, T. Vahlenkamp, H.
Thormar, E. De Clercq, C. McGuigan, Mol. Pharmacol. 1996, 50, 1207–
1213.
[7] B. A. J. Wilber, J. J. Docherty, J. Gen. Virol. 1994, 75, 1743–1747.
[8] C. R. Wagner, V. V. Iyer, E. J. McIntee, Med. Res. Rev. 2000, 20, 417–451.
[9] K. G. Devine, C. McGuigan, T. J. O’Connor, S. R. Nicholls, D. Kinchington,
AIDS 1990, 4, 371–373.
[10] D. Cahard, C. McGuigan, J. Balzarini, Mini Rev. Med. Chem. 2004, 4, 371–
381.
[11] D. P. Drontle, C. R. Wagner, Mini Rev. Med. Chem. 2004, 4, 409–419.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 10
&
8
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
[12] Y. Mehellou, J. Balzarini, C. McGuigan, ChemMedChem 2009, 4, 1779–
1791.
[23] T. W. Abraham, T. I. Kalman, E. J. McIntee, C. R. Wagner, J. Med. Chem.
1996, 39, 4569–4575.
[13] S. A. Harris, C. McGuigan, G. Andrei, R. Snoeck, E. De Clercq, J. Balzarini,
Antiviral Chem. Chemother. 2001, 12, 293–300.
[24] O. Adelfinskaya, P. Herdewijn, Angew. Chem. 2007, 119, 4434–4436;
Angew. Chem. Int. Ed. 2007, 46, 4356–4358.
[14] A. Giraut, X. P. Song, M. Froeyen, P. Marliere, P. Herdewijn, Nucleic Acids
Res. 2010, 38, 2541–2550.
[15] D. Farquhar, D. N. Srivastva, N. J. Kattesch, P. P. Saunders, J. Pharm. Sci.
1983, 72, 324–325.
[25] M. Maiti, S. Michielssens, N. Dyubankova, M. Maiti, E. Lescrinier, A. Ceu-
lemans, P. Herdewijn, Chem. Eur. J. 2012, 18, 857–868.
[26] A. C. Benniston, P. Gunning, R. D. Peacock, J. Org. Chem. 2005, 70, 115–
123.
[16] a) L. Naesens, N. Bischofberger, P. Augustijns, P. Annaert, G. van den
Mooter, M. N. Arimilli, C. U. Kim, E. De Clercq, Antimicrob. Agents Chemo-
ther. 1998, 42, 1568–1573; b) H. B. Fung, E. A. Stone, F. J. Piacenti, Clin.
Ther. 2002, 24, 1515–1548.
[17] a) J. E. Starrett, Jr., D. R. Tortolani, M. J. Hitchcock, J. C. Martin, M. M.
Mansuri, Antiviral Res. 1992, 19, 267–273; b) L. Naesens, J. Balzarini, N.
Bischofberger, E. De Clercq, Antimicrob. Agents Chemother. 1996, 40,
22–28; c) P. Annaert, R. Kinget, L. Naesens, E. De Clercq, P. Augustijns,
Pharm. Res. 1997, 14, 492–496; d) P. Barditch-Crovo, J. Toole, C. W. Hen-
drix, K. C. Cundy, D. Ebeling, H. S. Jaffe, P. S. Lietman, J. Infect. Dis. 1997,
176, 406–413.
[18] J. Rautio, H. Kumpulainen, T. Heimbach, R. Oliyai, D. Oh, T. Jarvinen, J.
Savolainen, Nat. Rev. Drug Discovery 2008, 7, 255–270.
[19] a) J. Balzarini, A. Holy, J. Jindrich, L. Naesens, R. Snoeck, D. Schols, E. De
Clercq, Antimicrob. Agents Chemother. 1993, 37, 332–338; b) E. De
Clercq, Antiviral Res. 2007, 75, 1–13.
[20] a) L. L. Siu, D. D. Von Hoff, A. Rephaeli, E. Izbicka, C. Cerna, L. Gomez,
E. K. Rowinsky, S. G. Eckhardt, Invest. New Drugs 1998, 16, 113–119; b) T.
Reid, F. Valone, W. Lipera, D. Irwin, W. Paroly, R. Natale, S. Sreedharan, H.
Keer, B. Lum, F. Scappaticci, A. Bhatnagar, Lung Cancer 2004, 45, 381–
386.
[27] B. P. Bandgar, R. J. Sarangdhar, S. Viswakarma, F. A. Ahamed, J. Med.
Chem. 2011, 54, 1191–1201.
[28] Y. Feng, J. K. Coward, J. Med. Chem. 2006, 49, 770–788.
[29] D. M. Lehsten, D. N. Baehr, T. J. Lobl, A. R. Vaino, Org. Process Res. Dev.
2002, 6, 819–822.
[30] a) C. McGuigan, R. N. Pathirana, N. Mahmood, K. G. Devine, A. J. Hay, An-
tiviral Res. 1992, 17, 311–321; b) C. McGuigan, R. N. Pathirana, J. Balzari-
ni, E. De Clercq, J. Med. Chem. 1993, 36, 1048–1052.
[31] a) Y. Mizuno, T. Itoh, K. Saito, Chem. Pharm. Bull. 1964, 12, 866–872;
b) F. Seela, H. Rosemeyer, S. Fischer, Helv. Chim. Acta 1990, 73, 1602–
1611; c) C. Crey-Desbiolles, J. Lhomme, P. Dumy, M. Kotera, J. Am. Chem.
Soc. 2004, 126, 9532–9533.
[32] E. P. Brass, Pharmacol. Rev. 2002, 54, 589–598.
[33] S. Vittori, D. Dal Ben, C. Lambertucci, G. Marucci, R. Volpini, G. Cristalli,
Curr. Med. Chem. 2006, 13, 3529–3552.
[34] M. J. Sofia, W. S. Chang, P. A. Furman, R. T. Mosley, B. S. Ross, J. Med.
Chem. 2012, 55, 2481–2531.
[35] a) M. Derudas, D. Carta, A. Brancale, C. Vanpouille, A. Lisco, L. Margolis,
J. Balzarini, C. McGuigan, J. Med. Chem. 2009, 52, 5520–5530; b) C.
McGuigan, P. Murziani, M. Slusarczyk, B. Gonczy, J. Vande Voorde, S.
Liekens, J. Balzarini, J. Med. Chem. 2011, 54, 7247–7258.
[21] N. Korsisaari, D. J. Rossi, K. Luukko, K. Huebner, M. Henkemeyer, T. P.
Makela, Mol. Cell. Biol. 2003, 23, 3929–3935.
Received: January 24, 2013
Revised: March 15, 2013
[22] a) T. W. Abraham, C. R. Wagner, Nucleosides Nucleotides 1994, 13, 1891–
1903; b) E. J. McIntee, R. P. Remmel, R. F. Schinazi, T. W. Abraham, C. R.
Wagner, J. Med. Chem. 1997, 40, 3323–3331.
Published online on && &&, 0000
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 10
&
9
&
ÞÞ
These are not the final page numbers!

FULL PAPERS
M. Maiti, L. Persoons, G. Andrei,
Key to membrane permeability: The
introduction of the pivaloyloxymethyl
(POM) moiety as a carboxyl protecting
group has shown promise in the design
of antiviral phosphoramidate prodrugs
for the intracellular delivery of nucleo-
side monophosphates. Antiviral data
also support the substitution of a pri-
mary amine with a secondary amine in
the intracellular delivery of ProTide ana-
logues.
R. Snoeck, J. Balzarini, P. Herdewijn*
&& – &&
Synthesis and Anti-herpetic Activity of
Phosphoramidate ProTides
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 10
&10
&
ÞÞ
These are not the final page numbers!