Synthesis, in vitro antiviral evaluation, and stability studies of bis(s-acyl-2-thioethyl) ester derivatives of 9-[2- (phosphonomethoxy)ethyl]adenine (PMEA) as potential PMEA prodrugs with improved oral bioavailability

Article abstract of DOI:10.1021/jm960289o

A new series of hitherto unknown 9-[2-(phosphonomethoxy)ethyl]adenine (PMEA) phosphonodiester derivatives incorporating carboxyesterase-labile S- acyl-2-thioethyl (SATE) moieties as transient phosphonate-protecting groups was prepared in an attempt to increase the oral bioavailability of the antiviral agent PMEA. We report here a direct comparison of the in vitro anti-HIV and anti-HSV activities as well as the in vitro stability between the bis(SATE) derivatives and the already known PMEA prodrugs, namely, bis[(pivaloyloxy)methyl (POM)]and bis[dithiodiethy] (DTE)]PMEA. All of the compounds tested showed an enhanced in vitro antiviral activity compared to the parent PMEA. The bis(POM)- and bis(tBu-SATE)PMEA derivatives were the most effective. However, striking differences between these two compounds were found during the stability studies. In particular the bis(tBu-SATE)PMEA was found to be more stable than bis(POM)PMEA in human gastric juice and human serum, suggesting it could be considered as a promising candidate for further in vivo development.

Full text of DOI:10.1021/jm960289o

4958
J . Med. Chem. 1996, 39, 4958-4965
Syn th esis, in Vitr o An tivir a l Eva lu a tion , a n d Sta bility Stu d ies of
Bis(S-a cyl-2-th ioeth yl) Ester Der iva tives of
9-[2-(P h osp h on om eth oxy)eth yl]a d en in e (P MEA) a s P oten tia l P MEA P r od r u gs
w ith Im p r oved Or a l Bioa va ila bility†
Samira Benzaria,‡,∇ He´le`ne Pe´licano,‡ Richard J ohnson,‡ Georges Maury,‡ J ean-Louis Imbach,‡
Anne-Marie Aubertin,§ Georges Obert,§,| and Gilles Gosselin*,‡
Laboratoire de Chimie Bioorganique, case courrier 008, UMR CNRS-USTL 5625, Universite´ Montpellier II, Sciences et
Techniques du Languedoc, Place Euge`ne Bataillon, 34095 Montpellier Ce´dex 5, France, and Institut de Virologie de la Faculte´
de Me´decine de Strasbourg, Unite´ INSERM 74, 3 Rue Koeberle´, 67000 Strasbourg, France
Received April 17, 1996X
A new series of hitherto unknown 9-[2-(phosphonomethoxy)ethyl]adenine (PMEA) phospho-
nodiester derivatives incorporating carboxyesterase-labile S-acyl-2-thioethyl (SATE) moieties
as transient phosphonate-protecting groups was prepared in an attempt to increase the oral
bioavailability of the antiviral agent PMEA. We report here a direct comparison of the in
vitro anti-HIV and anti-HSV activities as well as the in vitro stability between the bis(SATE)
derivatives and the already known PMEA prodrugs, namely, bis[(pivaloyloxy)methyl (POM)]-
and bis[dithiodiethyl (DTE)]PMEA. All of the compounds tested showed an enhanced in vitro
antiviral activity compared to the parent PMEA. The bis(POM)- and bis(tBu-SATE)PMEA
derivatives were the most effective. However, striking differences between these two compounds
were found during the stability studies. In particular the bis(tBu-SATE)PMEA was found to
be more stable than bis(POM)PMEA in human gastric juice and human serum, suggesting it
could be considered as a promising candidate for further in vivo development.
In tr od u ction
9-[2-(Phosphonomethoxy)ethyl]adenine [PMEA (1);
Figure 1] has demonstrated broad spectrum antiviral
activity against human immunodeficiency virus (HIV)
and other retroviruses.1,2 PMEA is also active against
various DNA viruses, including hepatitis B virus, herpes
simplex virus (HSV), cytomegalovirus, and Epstein-Barr
virus.1,2 In addition to in vitro activity, PMEA has
demonstrated in vivo efficacy when administrated in-
travenously, intraperitoneally, or intramuscularly.1,2
Thus, PMEA is of interest both as a potential antiret-
roviral drug for HIV infections and for the treatment of
some of the opportunistic infections associated with
AIDS. It has undergone phase I/II clinical trials where
it exhibited activity against HIV in vivo.3 However, the
potential therapeutic use of PMEA could be limited by
its poor oral bioavailability, which has been reported
to be <1% in monkeys4 and 7.8-11% in rats.5,6 The
poor oral bioavailability is due to the phosphonate
negative charges that are present in PMEA at physi-
ological pH. Therefore, the concept of temporarily
F igu r e 1. Structure of PMEA and its phosphonodiester
derivatives studied.
masking these charges with neutral substituents to
form more lipophilic derivatives capable of crossing the
gastrointestinal wall and reverting back to the parent
PMEA in plasma was attempted.
Compared to prodrugs of nucleoside monophos-
phates,7-9 relatively few examples of PMEA derivatives
have so far been reported in the literature. One PMEA
prodrug has been synthesized by linking a synthetic
polymer bearing mannosylated residues to PMEA.10
More recently, several derivatives, including hydrogeno-
phosphinate,11 mono- or bis(phosphonoamidate),6 and
mono- or bis(phosphonoester)6,12-16 functionalities, have
been prepared as potential prodrugs of PMEA. The best
results were obtained when PMEA was esterified with
two transient, enzyme-labile phosphonate-protecting
groups, which were later removed by a specific enzy-
matic system. For instance, we have previously re-
ported16 that the bis[dithiodiethyl (DTE)]PMEA (2;
† This work is taken, in part, from the Ph.D. Dissertation of S.
Benzaria, Universite´ de Montpellier II, J une 23, 1995. It has been
presented in preliminary form at the Eleventh International Round
Table on Nucleosides, Nucleotides, and their Biological Applications,
Sept. 7-11, 1994, Leuven, Belgium. For the proceedings, see: Ben-
zaria, S.; Gosselin, G.; Pe´licano, H.; Maury, G.; Aubertin, A.-M.; Obert,
G.; Kirn, A.; Imbach, J .-L. New Prodrugs of 9-(2-phosphonomethoxy-
ethyl)adenine [PMEA]: synthesis and stability studies. Nucleosides
Nucleotides 1995, 14, 563-565.
* Author for correspondence. Tel: (33) 4-67-14-38-55. Fax: (33) 4-67-
04-20-29. E-mail: gosselin@univ-montp2.fr.
‡ Universite´ Montpellier II.
§ Unite INSERM 74.
∇ Present address: Laboratory of Medicinal Chemistry, Develop-
mental Therapeutics Program, Division of Cancer Treatment, National
Cancer Institute, National Institutes of Health, Bethesda, MD 20892.
| Deceased on J une 20, 1995.
X Abstract published in Advance ACS Abstracts, November 1, 1996.
S0022-2623(96)00289-0 CCC: $12.00 © 1996 American Chemical Society

SATE Derivatives as Potential PMEA Prodrugs
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25 4959
Sch em e 1. Synthesis of Bis(POM)PMEA (3)
Sch em e 2. Synthesis of Bis(SATE)PMEA Prodrugs
4a -d and Their Corresponding Monoesters 9a -d a
Figure 1), whose activation is mediated by reductases,
showed an increase in the in vitro anti-HIV activity of
PMEA. More importantly, (acyloxy)alkyl groups have
also been studied as carboxyesterase-mediated bio-
reversible groups,6,17 and the bis[(pivaloyloxy)methyl
(POM)]PMEA (3; Figure 1) was examined as an efficient
PMEA prodrug.6,15 Antiretroviral activity and phar-
macokinetics of orally administrated bis(POM)PMEA in
mice have been reported,18-21 and recent studies have
demonstrated that this compound improved PMEA oral
bioavailability 2-fold in rats6 and 5-fold in monkeys,22
despite its low aqueous solubility and stability.22,23
Phase I/II clinical trials are currently ongoing to evalu-
ate its efficacy following oral administration in AIDS
patients.24,25
In our laboratory, we have independently developed
the carboxyesterase-labile S-acylthioethyl (SATE) group
for the transient protection of nucleoside monophos-
phates.26-30 Based on the favorable results obtained
with the SATE protection of various drugs, in terms of
in vitro anti-HIV activity and preliminary stability, we
decided to prepare and study a series of PMEA bis-
(SATE) derivatives (4a -d ; Figure 1). We report herein
the synthesis of these new PMEA prodrugs, as well as
their in vitro anti-HIV and anti-HSV activities, and
their chemical and enzymatic stabilities compared to the
two known PMEA prodrugs, bis(DTE)- and bis(POM)-
PMEA.
a(i) 5, MSNT, pyridine; (ii) CH3CO2H/H2O/MeOH. a , R ) Me;
b, R ) iPr; c, R ) tBu; d , R ) Ph.
Ta ble 1. Antiviral Activity of the Phosphonodiester
Derivatives 2-4 Compared to That of PMEA (1) in Two Cell
Lines Infected with HIV-1a
MT-4
CEM-SS
b
c
b
c
compd
EC50
CC50
EC50
CC50
1
2
>10
>10
50
0.42
2
>10
67
6.9
3
0.08
1.1
0.51
0.65
0.41
0.74
7.6
3.8
1.5
2.4
0.04
0.1
0.09
0.03
0.05
4.4
23
13
5.8
3.9
4a
4b
4c
4d
a
All data represent average values for at least three separate
experiments. The variation of these results under standard
Resu lts a n d Discu ssion
b
operating procedures is below (10%. EC50, 50% effective con-
Ch em istr y. PMEA (1) was prepared according to
Holy’s procedure31 with some modifications32-35 that
improved the overall yield and facilitated the execution
of some steps, which were performed more rapidly, or
sometimes without any preliminary purification. The
synthesis of bis(DTE)PMEA (2) was effected as previ-
ously reported16 and involved the condensation of a
hydroxylated derivative with a base-protected PMEA.
Concerning the synthesis of the bis(POM)PMEA (3), we
did not use the published procedure,6,15 which consisted
of reacting PMEA with chloromethyl pivaloate in the
presence of the hindered base N,N′-dicyclohexylmor-
pholinecarboxamidine, to allow solubilization in DMF.
We opted to overcome the solubility problem by using
the N6-monomethoxytritylated derivative (5)16 of PMEA
(Scheme 1), which is soluble in organic solvents. Thus,
reaction of 516 with iodomethyl pivaloate,36 followed by
treatment with acid, gave a 18% yield of the bis(POM)-
PMEA (3).
centration (in µM) or concentration required to inhibit the replica-
tion of HIV-1 by 50%. c CC50, 50% cytotoxic concentration (in µM)
or concentration required to reduce the viability of uninfected cells
by 50%.
ing protected phosphonodiesters 7a -d in yields of 58-
86% (Scheme 2). The monoesters 8b-d were also
isolated as byproducts. Treatment of 7a -d with acetic
acid provided the target PMEA prodrugs 4a -d as oils
in yields of 75-82% after purification by silica gel
column chromatography. It is noteworthy that deriva-
tives 4a -c could be crystallized, unlike 4d , bis(DTE)-
PMEA (2), and bis(POM) PMEA (3). Furthermore,
authentic samples of the corresponding mono(SATE)
derivatives (9a -d ) of PMEA were required as standards
to identify the decomposition products from the stability
studies in different media. The monoesters 9b-d were
prepared by acid treatment of 8b-d and purified on
Dowex resin (acetate form).31 The mono(Me-SATE)-
PMEA (9a ) was obtained as a byproduct (3%) during
the detritylation of 7a .
For the synthesis of the title compounds 4a -d , we
chose an approach similar to that developed in the case
of bis(DTE)PMEA. The hydroxythioester precursors
6a -d , which were prepared by reacting 2-iodoethanol
with the corresponding thio acid,26 were condensed with
5 in pyridine in the presence of 1-mesitylene-2-sulfonyl-
3-nitro-1,2,4-triazole (MSNT) to afford the correspond-
An tivir a l Activity. PMEA (1) and the phosphono-
diester derivatives 2, 3, and 4a -d were evaluated for
their inhibitory effects on the replication of HIV-1 in
two cell culture systems (Table 1). Under the assay
conditions, all the tested prodrugs enhanced the in vitro

4960 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25
Benzaria et al.
Ta ble 2. Measured Partition Factors (between Octanol and
Water) of the Phosphonodiesters 2-4 Compared to That of
PMEA (1)
and it should be lipophilic enough to cross the gas-
trointestinal wall. Therefore, to test the usefulness of
2 and the newly synthesized bis(SATE) derivatives 4a -
d , it was necessary to study their chemical and enzy-
matic stabilities. The bis(POM) ester 3 was chosen as
a reference compound for this evaluation because it is
the most effective PMEA oral prodrug reported to
date.6,22
1
2
3
4a
4b
4c
4d
log Pa -4.11
0.21
2.48
0.32
2.31
3.41
3.93
((0.03) ((0.01) ((0.01) ((0.05) ((0.04) ((0.01) ((0.01)
aThe determination of partition coefficients was carried out by
HPLC by using a microscale method adapted from that described
previously for various nucleoside analogues (Ford, H., J r.; Merski,
C. L.; Kelley, J . A. A rapid microscale method for the determi-
nation of partition coefficients by HPLC. J . Liquid Chromatogr.
1991, 14, 3365-3386).
In order to measure the relative chemical and enzy-
matic stabilities of the PMEA prodrugs, the decomposi-
tion pathways and kinetic data for compounds 2, 3, and
4a -d (initial concentration 5 × 10-5 M) were studied
at 37 °C (i) in water, (ii) in pH 7.2 buffer, (iii) in RPMI
1640 containing 10% heat-inactivated fetal calf serum
(culture medium), (iv) in RPMI 1640 alone, (v) in pH 2
buffer, (vi) in human gastric juice, and (vii) in human
serum. These various media were thought to be valid
in vitro models for the different types of degradation
that may affect the prodrugs during oral administration.
Crude aliquots of incubates were directly analyzed by
using the recently described,26 on-line HPLC cleaning
method. The use of a reverse-phase (RP) precolumn
(Guard-Pak, δ-pak C18, Waters) in this technique
allowed the elimination of proteins before the sample
reached the RP analytical column. During this cleaning
step, we used an ion-pairing reagent (tetrabutylammo-
nium sulfate) in order to avoid the coelimination of the
polar PMEA and monoester derivatives produced after
degradation of the prodrugs in the various media.
anti-HIV activity of PMEA. The lower EC50 values for
the neutral phosphonodiester derivatives 2-4, com-
pared to that of the parent molecule 1, can be attributed
to an increase in cellular uptake followed by intracel-
lular release of PMEA. The prodrugs appear to be
lipophilic enough to cross the cellular membrane through
simple diffusion, unlike the PMEA, which seems to
require an endocytosis-like process,37 or the ATP mem-
brane receptor1 for intracellular transport. The bis-
(POM)PMEA (3) and the bis(tBu-SATE)PMEA (4c) were
found to be the most potent antiviral agents. The DTE
derivative 2 showed limited efficacy, presumably due
to its lower lipophilicity compared to the other diesters,
as reflected by their measured log P values (Table 2).
We can tentatively conclude from the results pre-
sented in Tables 1 and 2 that lipophilicity is not the
only factor that influences the antiviral activity of these
enzyme-labile prodrugs. For instance, compound 4d is
more lipophilic than 3, but it is not more potent. The
enzymatic system involved in prodrug activation and
the prodrugs’ relative stabilities are also likely to be
crucial factors.
We observed that all the prodrugs tested in water and
in pH 7.2 buffer had the same decomposition pathway,
giving rise to the corresponding monoesters. For 4a -d
the decomposition products were unambiguously identi-
fied by HPLC coinjection of the corresponding mono-
(SATE) derivatives 9a -d . Due to the lack of authentic
samples in the case of 2 and 3, we made an extrapola-
tion based on the retention times and UV spectra of the
compounds formed. Table 4 shows that the bis(SATE)
prodrugs 4a -d were the most stable toward chemical
hydrolysis at neutral (pH 7.2) or slightly acidic (milliQ
water, pH 5.5) pH.
PMEA (1) and its phosphonodiester derivatives 2-4
were also evaluated against HSV-1 and -2 in two cell
lines (Table 3). Again, compounds 3 and 4c proved to
be the most potent, with EC50 values of 0.91 and 0.97
µM, respectively, in HSV-2-infected MRC-5 cells, com-
pared to 33.5 µM for PMEA. The inhibiting capacity of
the derivatives followed the order 3 > 4c ∼ 4b > 4d >
4a > 2 ∼ 1.
The comparable decomposition rates obtained for
4a -d in water and at pH 7.2 suggest that the variation
of the SATE chain does not influence the reactivity of
this protecting group under the conditions of the assay.
This result is in accordance with a hydrolysis mecha-
nism involving nucleophilic attack at the R carbon of
the phosphorous atom38 furthest from the acyl thioester
moiety (Scheme 3).
It should be noted that every prodrug-related gain in
antiviral potency was in all cases accompanied by a
proportional enhancement of cytotoxicity, as reflected
by the CC50 values (Tables 1 and 3). Although it is
possible that these toxic effects are due to the release
of the phosphonate-protecting groups, we do not believe
this to be the case. Indeed, we have recently shown that
use of SATE groups as transient phosphate protections
of 5’-mononucleotides does not induce any additive
toxicity compared to the parent nucleoside (manuscript
in preparation). A more likely explanation would be
that the derivatization of PMEA increases not only its
intracellular concentration but the consequent intrac-
ellular concentration of the phosphorylated anabolites,
which are the active and cytotoxic forms of the drug.1
Therefore, in the case of PMEA, it appears that the
prodrug approach is not well-suited to improve the
antiviral selectivity index (ratio CC50/EC50). Neverthe-
less, this strategy may prove useful to deliver the PMEA
into plasma after oral administration, as already dem-
onstrated for the bis(POM)PMEA.6,22
For the study in RPMI 1640, which represents a free-
enzyme system but a nucleophile-enriched medium, only
2, 3, and 4a were selected for testing. The bis(POM)-
PMEA (3) was the least stable under these conditions
(Table 5). It proved to be more sensitive to chemical
hydrolysis than the bis(DTE) derivative 2 (half-life of 5
h for 3, compared to >24 h for 2), which was not
apparent in simple media, such as water and pH 7.2
buffer. The corresponding monoesters were detected as
single decomposition products for the three compounds
tested.
In culture medium, we additionally observed PMEA
formation (as ascertained by coinjection with authentic
sample), presumably following the decomposition path-
way: phosphonodiester derivative f phosphonomo-
noester f PMEA. We have recently published26 the
Sta bility Stu d ies. For a designed oral prodrug to
be effective, it must be resistant enough to any hydroly-
sis that might occur before it reaches the bloodstream,

SATE Derivatives as Potential PMEA Prodrugs
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25 4961
Ta ble 3. Anti-HSV-1 and -2 Activity of the Phosphonodiester Derivatives 2-4 Compared to That of PMEA (1) in Two Infected Cell
Linesa
Vero
MRC-5
b
b
EC50
EC50
c
c
compd
HSV-1
HSV-2
CC50
>100
>100
HSV-1
HSV-2
CC50
>100
>100
1
2
3
4a
4b
4c
4d
38.5 ( 6.9
22.3 ( 7.4
0.87 ( 0.62
14.2 ( 2.5
1.35 ( 0.45
1.07 ( 0.24
2.56 ( 0.89
32.7 ( 10.5
27.4 ( 8.5
1.15 ( 0.39
10.2 ( 3.1
1.14 ( 0.39
1.12 ( 0.39
2.89 ( 0.97
29.7 ( 7.6
32.3 ( 8.3
1.23 ( 0.73
12.6 ( 4.6
1.69 ( 0.37
1.43 ( 0.28
3.48 ( 1.23
33.5 ( 9.6
26.5 ( 6.8
0.91 ( 0.32
8.9 ( 2.1
1.20 ( 0.54
0.97 ( 0.38
4.75 ( 2.1
22.56 ( 8.4
126 ( 38
23 ( 11.4
36.8 ( 6.3
12.7 ( 1.8
15.7 ( 5.1
148 ( 27
30.2 ( 9.6
48.5 ( 8.7
18.7 ( 3.4
a
All data represent average values for at least three different experiments. b,c See the corresponding footnotes in Table 1.
Ta ble 4. Calculated Rate of Decomposition in Water (Milli-Q,
pH 5.5) and at pH 7.2 (Ammonium Acetate Buffer, 0.02 M) for
the Phosphonodiester Derivatives 2-4
Ta ble 6. Calculated Rate of Hydrolysis at pH 2 (Glycine/HCl
Buffer) in Human Gastric J uice and Human Serum for the
Phosphonodiester Derivatives 2-4
compd
water
pH 7.2
compd
pH 2
human gastric juicea
human serumb
2
3
4a
4b
4c
4d
t1/2 3.9 days
t1/2 3.7 days
13.2%a
13.9%b
32.7%c
4%c
2
3
4a
4b
4c
4d
t1/2 4.4 days
t1/2 3.2 days
20.8%c
t1/2 4.8 days
t1/2 2.4 days
t1/2 5.3 days
25.7%d
t1/2 <5 min
t1/2 <5 min
t1/2 <5 min
t1/2 11 min
t1/2 4 h
8.9%a
1.6%c
3%c
12.8%c
14.2%a
7.8%c
11.1%d
10.2%a
1.7%c
9%c
5%d
t1/2 1.5 h
a
b
a
Percent of decomposition after 9 days of incubation. Percent
The human gastric juice (pH 1.3) [obtained from Dr J .-C.
Cuber (Institut National de la Sante´ et de la Recherche Me´dicale,
U-45, Lyon, France)] was centrifuged for 15 min at 3000 rpm at 4
°C. The corresponding compounds were dissolved in the filtered
gastric juice up to a 50 µM concentration. After incubation at 37
°C, aliquots were periodically removed and directly analyzed by
of decomposition after 26 h of incubation. c Percent of decomposi-
tion after 24 h of incubation.
Sch em e 3. Proposed Hydrolysis Mechanism of PMEA
Prodrugs 4a -d in Water (Milli-Q, pH 5.5) and at pH 7.2
(Ammonium Buffer, 0.02 M)
b
HPLC. Human serum from healthy volunteers was a gift from
the Centre Regional de Transfusion Sanguine (Montpellier, France).
c Percent of decomposition after 7 days of incubation. Percent of
d
decomposition after 5 days of incubation.
group, and (ii) phosphodiesterase activity, which leads
to the elimination of the second SATE protecting group
(we also report in Table 5 the half-life of the PMEA
monoester derivatives 9a -d ). It is noteworthy that
except for 4a , all the SATE prodrugs proved to be more
stable than bis(POM)PMEA (3), for which the measured
half-life of 5 h compares favorably with the previously
reported value.22 This result suggests that the SATE
thioester might be more stable toward carboxyesterase
activity than a simple ester. The relative stability of 2
is quite surprising (half-life 1.6 days) and might be
attributed to the lack of reductases in the medium or
to the poor affinity of 2 toward the reductases. As Table
5 clearly points out, the bis(tBu-SATE)PMEA (4c)
emerged as the most stable prodrug in media such as
culture medium, which represents an appropriate model
for neutral conditions where chemical and enzymatic
hydrolysis can occur.
For their acid stability, the PMEA prodrugs were
evaluated not only in the usual pH 2 test buffer but also
in human gastric juice (pH 1.3). The values reported
in Table 6 show a good correlation for these two media,
for which we observed the same decomposition pathway,
consisting of the formation of the corresponding mo-
noester derivative as the sole product for all the pro-
drugs tested. In each case, the SATE prodrugs 4a -d
proved to be more acid resistant than the already known
PMEA prodrugs 2 and 3. This last result leads us to
suppose that 4a -d could be more stable than 2 and 3
in the gastric environment following oral administra-
tion, which would increase their absorption. However,
as reflected by their half-lives in human serum (Table
6), the two SATE prodrugs 4a ,b, as well as 2 and 3, are
likely to be hydrolyzed immediately to PMEA once in
Ta ble 5. Calculated Half-Lives of the Phosphonodiester
Derivatives 2-4 and the Phosphonomonoesters 9a -d in RPMI
1640 and in Culture Medium
t1/2
compd
RPMI 1640
culture medium
2
3
>24 h
5 h
1.6 days
5 h
4a
4b
4c
4d
9a
9b
9c
9d
>24 h
NDa
ND
ND
ND
ND
ND
ND
3.7 h
8.3 h
3.4 days
25.7 h
19.2 h
1.2 days
9.6 days
2.7 days
a
ND, not determined.
proposed mechanisms for the decomposition of SATE
prodrugs and the release of the parent mononucleotide
in such culture media. It should be noted that we
confirmed hereby the formation of the hypothesized
intermediates corresponding to the mono(SATE) deriva-
tives. The half-life of 4a in culture medium was shorter
than in RPMI alone (Table 5). This suggests that in
the heat-inactivated serum added to RPMI there re-
mains (i) carboxyesterase activity, which leads to a
faster elimination of the first enzyme-labile SATE

4962 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25
Benzaria et al.
9-{2-[O,O′-Bis[(pivaloyloxy)m ethyl]phosphonom eth oxy]-
eth yl}a d en in e (3). Iodomethyl pivaloate36 (966.75 mg, 3.99
mmol) was added to a solution of 516 (544 mg, 0.78 mmol) in
anhydrous pyridine (10 mL) and stirred at room temperature
for 36 h. The reaction mixture was neutralized with an
aqueous solution of 1 M triethylammonium bicarbonate buffer
(pH 7.5, 8 mL), evaporated under reduced pressure, and
coevaporated with toluene and methanol. The crude residue
was treated with a (8:1:1, v/v/v) mixture of acetic acid, water,
and methanol (25 mL) and stirred at room temperature for
12 h. The residue obtained after evaporation was dissolved
in chloroform (50 mL), and the organic layer was washed with
water, dried over sodium sulfate, filtered, and evaporated
under reduced pressure. Column chromatography of the
residue on silica gel with a stepwise gradient of methanol (0-
3%) in dichloromethane afforded the title compound 3 as an
oil (67.5 mg, 18%): 1H NMR (DMSO-d6) δ 8.14 and 8.09 (2s,
1H and 1H, 2-H and 8-H), 7.33 (s, 2H, NH2), 5.54 (d, 4H, J )
12.7 Hz, 2 × -O-CH2-), 4.32 (t, 2H, J ) 5.1 Hz, -CH2-N), 3.95
(d, 2H, J ) 7.8 Hz, -CH2-P-), 3.88 (t, 2H, CH2-CH2-N), 1.12 (s,
18H, 2 × (CH3)3C-); 31P NMR (DMSO-d6) δ 22.2; FAB MS (>0,
NBA) m/ e 502 (M + H)+; FAB MS (<0, G/T) m/ e 386 (M -
Piv-O-CH2)-, 272 (M - 2Piv-O-CH2 + H)-; HRMS 502.2000
(M + H), calcd for C20H33N5O8P 502.2067; UV (ethanol) λmax
260 nm (ꢀ 12 000), λmin 228 nm (ꢀ 3700).
Gen er a l P r oced u r e for th e P r ep a r a tion of th e P h os-
p h on od iester Der iva tives 7a -d a n d th e P h osp h on om o-
n oester s 8b-d . 1-Mesitylene-2-sulfonyl-3-nitro-1,2,4-triazole
(MSNT; 3 equiv) was added to a solution of 516 and the
appropriate hydroxythioester 826 in anhydrous pyridine (24
mL/mmol of 5). After stirring at room temperature for 12 h,
the reaction mixture was neutralized with an aqueous solution
of 1 M triethylammonium hydrogenocarbonate buffer (pH 7.5,
2 times the number of moles of MSNT) and extracted with
chloroform and water. The organic layer was dried over
sodium sulfate, filtered, and evaporated to dryness under
reduced pressure.
circulation. On the other hand, the two other SATE
prodrugs (4c,d ) showed greater stability in human
serum (half-life, respectively, of 4 and 1.5 h), which
indicated that they could show in vivo bioavailability
and biodistribution different from (and probably better
than) the parent PMEA.
Con clu sion
The present results demonstrate that the bis(SATE)-
PMEA prodrugs 4a -d are chemically and enzymatically
more stable than the already known bis(DTE)- and bis-
(POM)PMEA, 2 and 3, respectively. This greater stabil-
ity in neutral and acidic conditions, as well as in serum,
indicates that the bis(SATE) derivatives might enhance
PMEA oral bioavailability. Moreover, the crystalline
form of 4a -c might facilitate the formulation of oral
forms of administration. Among the SATE series, the
tBu-SATE derivative 4c emerged as the most promising
compound, combining an antiviral potency similar to
that of the currently developed bis(POM)PMEA (3) with
a markedly greater chemical and enzymatic stability.
In vivo studies with 4c are presently ongoing to deter-
mine its antiretroviral efficacy after oral administration
vis-a`-vis the bis(POM) derivative 3 and the parent
compound PMEA (1).
Exp er im en ta l Section
Ch em ica l Syn th esis. Evaporation of solvents was carried
out on a rotary evaporator under reduced pressure. Melting
points were determined with a Gallenkamp MFB-595-010-M
apparatus and are uncorrected. 1H NMR spectra were run at
ambient temperature in (CD3)2SO (DMSO-d6) or DMSO-d6
+
D2O with a Bruker AC 250 spectrometer. Chemical shifts are
given in δ values, (CD3)(CD2H)SO being set at δH 2.49 as a
reference. Deuterium exchange and decoupling experiments
were performed in order to confirm proton assignments. 31P
NMR spectra were recorded at ambient temperature on a
Bruker AC 250 spectrometer with proton decoupling. Chemi-
cal shifts are reported relative to external H3PO4. FAB mass
spectra were recorded in the positive-ion or negative-ion mode
on a J EOL DX 300 mass spectrometer operating with a J MA-
DA 5000 mass data system. Xe atoms were used for the gun
at 3 kV with a total discharge current of 20 mA. The matrix
used was 3-nitrobenzyl alcohol (NBA) or a mixture (50:50, v/v)
of glycerol and thioglycerol (G/T). High-resolution mass
spectra (HRMS) were obtained by using FAB+. UV spectra
were recorded on an Uvikon 810 (Kontron) spectrometer.
Elemental analyses were performed by the Service de Mi-
croanalyses du CNRS, Division de Vernaison (France), and the
results were within (0.4% of the theoretical values. TLC was
performed on precoated aluminum sheets of silica gel 60 F254
(Merck, article 5554), visualization of products being ac-
complished by UV absorbance followed by charring with 10%
ethanolic sulfuric acid with heating; phosphorous-containing
compounds were detected by spraying with Hanes molybdate
reagent.39 Column chromatography was carried out on silica
gel 60 (Merck, article 9385) at atmospheric pressure. High-
performance liquid chromatography (HPLC) studies were
carried out on a Waters Assoc. unit equipped with a model
600E multisolvent delivery system, a model 600E system
controller, a model U6K sample injector, a 486 tunable
absorbance detector, and a base line 810 data workstation. The
column was a reverse-phase analytical column (Hypersil, C18,
100 × 4.6 mm, 3 µm) protected by a prefilter and a precolumn
(Guard Pak, C18). The compound to be analyzed was eluted
using the same system as indicated below in the Stability and
Decomposition Studies section. The different retention times
are also indicated in this section.
N6-(4-Mon om eth oxytr ityl)-9-{2-[O,O′-bis[(S-acetylth io)-
eth yl]p h osp h on om eth oxy]eth yl}a d en in e (7a ). Silica gel
chromatography [eluent, stepwise gradient of methanol (0-
2%) in dichloromethane] gave 7a as a solid (59%) after
lyophilization in dioxane: 1H NMR (DMSO-d6) δ 8.15 and 7.90
(2s, 1H and 1H, 2-H and 8-H), 7.3-6.8 (m, 15H, NH, 14H
aromatic), 4.32 (2H, t, J ) 4.7 Hz, CH2N), 4.0-3.8 (m, 8H, 2
× S-CH2-CH2-O, CH2-P, CH2-CH2-N), 3.70 (s, 3H, -OCH3), 3.01
(t, 4H, J ) 6.4 Hz, 2 × S-CH2-CH2-O), 2.30 (s, 6H, 2 × CH3-
C(O)-); 31P NMR (DMSO-d6) δ 22.5; FAB MS (>0, G/T) m/ e
750 (M + H)+.
N 6 -(4-M o n o m e t h o x y t r i t y l)-9-{2-[O ,O ′-b i s [(S -i s o -
bu tyr ylth io)eth yl]ph osph on om eth oxy]eth yl}aden in e (7b)
a n d N6-(4-Mon om eth oxytr ityl)-9-{2-[O-m on o[(S-isobu ty-
r ylt h io)et h yl]p h osp h on om et h oxy]et h yl}a d en in e (8b ).
Silica gel chromatography [eluent, stepwise gradient of metha-
nol (0-100%) in dichloromethane] gave 7b as a solid (86%)
after lyophilization in dioxane and the more polar 8b as a
powder (8%) after lyophilization in a solution of dioxane and
water. 7b: 1H NMR (DMSO-d6) δ 8.15 and 7.90 (2s, 1H and
1H, 2-H and 8-H), 7.3-6.8 (m, 15H, NH, 14H aromatic), 4.32
(t, 2H, J ) 4.8 Hz, CH2N), 4.0-3.9 (m, 8H, 2 × S-CH2-CH2-O,
CH2-P, CH2-CH2-N), 3.70 (s, 3H, -OCH3), 3.01 (t, 4H, J ) 6.4
Hz, 2 × S-CH2-CH2-O), 2.8-2.7 (m, 2H, 2 × (CH3)2CH-), 1.08
(d, 12H, J ) 6.9 Hz, 2 × (CH3)2CH); 31P NMR (DMSO-d6) δ
22.45; FAB MS (>0, G/T) m/ e 806 (M + H)+; FAB MS(<0,
G/T) m/ e 804 (M - H)-, 674 (M - iPr-C(O)-S-CH2-CH2)-, 604
(M - iPr-C(O)-S-CH2-CH2 - iPr-C(O) + H)-.
8b: 1H NMR (DMSO-d6) δ 8.28 and 7.88 (2s, 1H and 1H,
2-H and 8-H), 7.3-6.8 (m, 15H, NH, 14H aromatic), 4.27 (t,
2H, J ) 4.7 Hz, CH2N), 3.8-3.6 (m, 9H, S-CH2-CH2-O, CH2-
P, CH2-CH2-N, -OCH3), 2.86 (t, 2H, J ) 6.7 Hz, S-CH2-CH2-
O), 2.8-2.7 (m, 1H, (CH3)2CH-), 1.04 (d, 6H, J ) 6.8 Hz,
(CH3)2CH); 31P NMR (DMSO-d6) δ 12.5; FAB MS (<0, G/T)
m/ e 674 (M - H)-, 544 (M - iPr-C(O)-S-CH2-CH2)-.
N 6-(4-Mo n o m e t h o x y t r it y l)-9-{2-[O ,O ′-b is [(S -p iv a l-
oylth io)eth yl]ph osph on om eth oxy]eth yl}aden in e (7c) an d
N6-(4-Mon om eth oxytr ityl)-9-{2-[O-m on o[(S-p iva loylth io)-
eth yl]p h osp h on om eth oxy]eth yl}a d en in e (8c). Silica gel
The test compounds were found to be pure by rigorous
HPLC analysis, high-field multinuclear NMR spectroscopy,
and high-resolution mass spectroscopy.

SATE Derivatives as Potential PMEA Prodrugs
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25 4963
chromatography [eluent, stepwise gradient of methanol (0-
100%) in dichloromethane] gave 7c as a solid (58%) after
lyophilization in dioxane and 8c as a 3:97 mixture of the acidic
and triethylammonium forms (10%) after lyophilization in a
solution of dioxane and water. 7c: 1H NMR (DMSO-d6) δ 8.15
and 7.89 (2s, 1H and 1H, 2-H and 8-H), 7.3-6.8 (m, 15H, NH,
14H aromatic), 4.32 (t, 2H, J ) 4.8 Hz, CH2N), 4.0-3.8 (m,
8H, 2 × S-CH2-CH2-O, CH2-P, CH2-CH2-N), 3.70 (s, 3H,
-OCH3), 2.99 (t, 4H, J ) 6.4 Hz, 2 × S-CH2-CH2-O), 1.14 (s,
18H, 2 × (CH3)3C-C(O)-); 31P NMR (DMSO-d6) δ 22.5; FAB
MS (>0, G/T), m/ e 834 (M + H)+; FAB MS (<0, G/T), m/ e
832 (M - H)-, 688 (M - Piv-S-CH2-CH2)-, 604 (M - Piv-S-
CH2-CH2 - Piv + H)-.
1H and 1H, 2-H and 8-H), 7.19 (s, 2H, NH2), 4.31 (t, 2H, J )
5.0 Hz, CH2N), 4.0-3.9 (m, 8H, 2 × S-CH2-CH2-O, CH2-P, CH2-
CH2-N), 3.03 (t, 4H, J ) 6.4 Hz, 2 × S-CH2-CH2-O), 2.8-2.7
(m, 2H, 2 × (CH3)2CH-), 1.09 (d, 12H, 2 × (CH3)2CH-); 31P NMR
(DMSO-d6) δ 22.55; FAB MS (>0, G/T) m/ e 534 (M + H)+,
136 (BH2)+; FAB MS (<0, NBA) m/ e 402 (M - iPr-C(O)-S-
CH2-CH2)-, 272 (M - 2iPr-C(O)-S-CH2-CH2 + H)-; HRMS
534.1595 (M + H), calcd for C20H33N5O6PS2 534.1610; UV
(ethanol) λmax 260 (ꢀ 15 000), 231 nm (ꢀ 11 000), λmin 240 (ꢀ
10 400), 223 nm (ꢀ 10 100). Anal. (C20H32N5O6PS2) C, H, N,
P, S.
9-{2-[O ,O ′-B i s [(S -p i v a lo y lt h i o )e t h y l]p h o s p h o n o -
m eth oxy]eth yl}a d en in e (4c). Silica gel column chromatog-
raphy [stepwise gradient of methanol (0-6%) in dichlo-
romethane] gave 4c (76%) which was crystallized from toluene
(60%): mp 66 °C; 1H NMR (DMSO-d6) δ 8.11 and 8.06 (2s, 1H
and 1H, 2-H and 8-H), 7.17 (s, 2H, NH2), 4.31 (t, 2H, J ) 5.0
Hz, CH2N), 4.0-3.9 (m, 8H, 2 × S-CH2-CH2-O, CH2-P, CH2-
CH2-N), 3.01 (t, 4H, J ) 6.4 Hz, 2 × S-CH2-CH2-O), 1.16 (s,
18H, 2 × (CH3)3C-C(O)-); 31P NMR (DMSO-d6) δ 22.6; FAB
MS (>0, NBA) m/ e 562 (M + H)+; FAB MS (<0, NBA) m/ e
416 (M - Piv-S-CH2-CH2)-, 272 (M - 2Piv-S-CH2-CH2 + H)-;
HRMS 562.1876 (M + H), calcd for C22H37N5O6PS2 562.1923;
UV (ethanol) λmax 260 (ꢀ 16 100), 231 nm (ꢀ 11 700), λmin 242
(ꢀ 11 100), 223 nm (ꢀ 10 400). Anal. (C22H36N5O6PS2) C, H,
N, P, S.
9-{2-[O,O′-Bis[(S-benzoylthio)ethyl]phosphonomethoxy]-
et h yl}a d en in e (4d ). Silica gel column chromatography
[stepwise gradient of methanol (0-6%) in dichloromethane]
gave 4d (75%) as an oil: 1H NMR (DMSO-d6) δ 8.11 and 8.07
(2s, 1H and 1H, 2-H and 8-H), 7.9-7.5 (m, 10H, 2 Ph), 7.18 (s,
2H, NH2), 4.30 (t, 2H, J ) 5.0 Hz, CH2N), 4.10 (q, 4H, 2 ×
S-CH2-CH2-O), 3.94 (d, 2H, J ) 8.2 Hz, CH2-P), 3.88 (t, 2H, J
) 5.0 Hz, CH2-CH2-N), 3.26 (t, 4H, J ) 6.3 Hz, 2 × S-CH2-
CH2-O); 31P NMR (DMSO-d6) δ 22.7; FAB MS (>0, G/T) m/ e
602 (M + H)+; FAB MS (<0, G/T) m/ e 436 (M - Bz-S-CH2-
CH2)-, 272 (M - 2Bz-S-CH2-CH2 + H)-; HRMS 602.1316 (M
+ H), calcd for C26H29N5O6PS2 602.1297; UV (ethanol) λmax 260
(ꢀ 30 900), 244 nm (ꢀ 27 800), λmin 250 (ꢀ 26 500), 223 nm (ꢀ
12 600).
Gen er a l P r oced u r e for th e P r ep a r a tion of th e P h os-
p h on om on oester s 9b-d . The corresponding 8 was treated
with a (8:1:1, v/v/v) mixture of acetic acid, water, and methanol
(70 mL/mol of 8). After 12 h of stirring at room temperature,
the reaction mixture was evaporated under reduced pressure
and extracted with water and chloroform. The aqueous layer
was evaporated under reduced pressure.
9-{2-[O,O′-Mon o[(S-isob u t yr ylt h io)et h yl]p h osp h on o-
m eth oxy]eth yl}a d en in e (9b). Dowex 1 × 2 resin (acetate
form) chromatography [eluent: linear gradient of acetic acid
(0-2 M)] gave 9d (46%) as a solid after lyophilization in
water: 1H NMR (DMSO-d6) δ 8.20 and 8.18 (2s, 1H and 1H,
2-H and 8-H), 7.70 (s, 2H, NH2), 4.33 (t, 2H, J ) 5.0 Hz, CH2N),
3.9-3.7 (m, 6H, S-CH2-CH2-O, CH2-P, CH2-CH2-N), 2.98 (t, 2H,
J ) 6.5 Hz, S-CH2-CH2-O), 2.8-2.7 (m, 1H, (CH3)2CH-), 1.09
(d, 6H, J ) 6.9 Hz, (CH3)2CH-); 31P NMR (DMSO-d6) δ 19.1;
FAB (<0, NBA) m/ e 805 (2M - H)-, 402 (M - H)-, 272 (M -
iPr-C(O)-S-CH2-CH2)-.
9-{2-[O,O′-Mon o[(S -p iva loylt h io)e t h yl]p h osp h on o-
m eth oxy]eth yl}a d en in e (9c). Dowex 1 × 2 resin (acetate
form) chromatography [eluent: linear gradient of acetic acid
(0-2 M)] gave 9c (54%) as a solid after lyophilization in
water: 1H NMR (DMSO-d6) δ 8.20 and 8.18 (2s, 1H and 1H,
2-H and 8-H), 7.72 (s, 2H, NH2), 4.34 (t, 2H, J ) 5.1 Hz, CH2N),
3.9-3.7 (m, 6H, S-CH2-CH2-O, CH2-P, CH2-CH2-N), 2.97 (t, 2H,
J ) 6.6 Hz, S-CH2-CH2-O), 1.15 (s, 9H, (CH3)3C-C(O)-); 31P
NMR (DMSO-d6) δ 19.1; FAB MS (<0, G/T) m/ e 833 (2M -
H)-, 416 (M - H)-, 272 (M - Piv-S-CH2-CH2)-.
8c: 1H NMR (DMSO-d6) δ 8.28 and 7.89 (2s, 1H and 1H,
2-H and 8-H), 7.3-6.8 (m, 15H, NH, 14H aromatic), 4.28 (t,
2H, J ) 4.6 Hz, CH2N), 3.8 (m, 6H, S-CH2-CH2-O, CH2-P, CH2-
CH2-N), 3.70 (s, 3H, -OCH3), 2.85 (t, 2H, J ) 6.7 Hz, S-CH2-
CH2-O), 1.11 (s, 9H, (CH3)3C-C(O)-); 31P NMR (DMSO-d6) δ
12.4; FAB MS (<0, G/T), m/ e 688 (M - H)-, 544 (M - Piv-S-
CH2-CH2)-.
N 6-(4-Mo n o m e t h o x y t r it y l)-9-{2-[O ,O ′-b is [(S -b e n z-
oylth io)eth yl]ph osph on om eth oxy]eth yl}aden in e (7d) an d
N6-(4-Mon om eth oxytr ityl)-9-{2-[O-m on o[(S-ben zoylth io)-
eth yl]p h osp h on om eth oxy]eth yl}a d en in e (8d ). Silica gel
chromatography [eluent, stepwise gradient of methanol (0-
100%) in dichloromethane] gave 7d as a solid (58%) after
lyophilization in dioxane and 8d as a 3:97 mixture of the acidic
and triethylammonium forms (10%) after lyophilization in a
solution of dioxane and water. 7d : 1H NMR (DMSO-d6) δ 8.2-
6.8 (m, 27H, 2-H, 8-H, NH, 24H aromatic), 4.31 (t, 2H, J )
4.8 Hz, CH2N), 4.08 (q, 4H, 2 × S-CH2-CH2-O), 3.94 (d, 2H, J
) 8.3 Hz, CH2-P), 3.87 (t, 2H, CH2-CH2-N), 3.68 (s, 3H, -OCH3),
3.23 (t, 4H, J ) 6.3 Hz, 2 × S-CH2-CH2-O); 31P NMR (DMSO-
d6) δ 22.6; FAB MS (>0, G/T), m/ e 874 (M + H)+; FAB MS
(<0, G/T) m/ e 872 (M - H)-, 708 (M - Bz-S-CH2-CH2)-.
8d : 1H NMR (DMSO-d6) δ 8.3-6.8 (m, 22H, 2-H, 8-H, NH,
19H aromatic), 4.27 (t, 2H, J ) 4.4 Hz, CH2N), 3.8-3.7 (m,
9H, S-CH2-CH2-O, CH2-P, CH2-CH2-N, -OCH3), 3.08 (t, 2H, J
) 6.4 Hz, S-CH2-CH2-O); 31P NMR (DMSO-d6) δ 12.5; FAB MS
(<0, G/T) m/ e 708 (M - H)-, 544 (M - Bz-S-CH2-CH2)-.
Gen er a l P r oced u r e for th e P r ep a r a tion of th e P h os-
p h on od iester Der iva tives 4a -d . The corresponding 7 was
treated with a (8:1:1, v/v/v) mixture of acetic acid, water, and
methanol (75 mL/mol of 7). After 12 h of stirring at room
temperature, the reaction mixture was evaporated under
reduced pressure.
9-{2-[O,O′-Bis[(S-acetylth io)eth yl]ph osph on om eth oxy]-
eth yl}a d en in e (4a ) a n d 9-{2-[O,O′-Mon o[(S-a cetylth io)-
eth yl]p h osp h on om eth oxy]eth yl}a d en in e (9a ). Silica gel
column chromatography [stepwise gradient of methanol (0-
6%) in dichloromethane] gave 4a (89%) which was crystallized
from toluene (67%) and 9a (3%) after evaporation of the
appropriate fractions and purification on Dowex 1 × 2 resin
(acetate form) [eluent: linear gradient of acetic acid (0.5-2
1
M)]. 4a : mp 68-69 °C; H NMR (DMSO-d6) δ 8.11 and 8.06
(2s, 1H and 1H, 2-H and 8-H), 7.19 (s, 2H, NH2), 4.31 (t, 2H,
J ) 5.0 Hz, CH2N), 4.0-3.9 (m, 8H, 2 × S-CH2-CH2-O, CH2-P,
CH2-CH2-N), 3.03 (t, 4H, J ) 6.4 Hz, 2 × S-CH2-CH2-O), 2.8-
2.7 (m, 2H, 2 × (CH3)2CH-), 1.09 (d, 12H, J ) 6.9 Hz, 2 ×
(CH3)2CH-); 31P NMR (DMSO-d6) δ 22.55; FAB MS (>0, G/T)
m/ e 570 (M + G + H)+, 478 (M + H)+; FAB MS (<0, G/T) m/ e
374 (M - Ac-S-CH2-CH2)-; HRMS 478.0852 (M + H), calcd
for C16H25N5O6PS2 478.0984; UV (ethanol) λmax 260 (ꢀ 14 100),
230 nm (ꢀ 10 400), λmin 240 (ꢀ 9200), 223 nm (ꢀ 9800). Anal.
(C16H24N5O6PS2) C, H, N, P, S.
9a : 1H NMR (D2O) δ 8.46 and 8.44 (2s, 1H and 1H, 2-H
and 8-H), 4.57 (t, 2H, J ) 4.9 Hz, CH2N), 4.00 (t, 2H, CH2-
CH2-N), 3.8-3.7 (m, 4H, S-CH2-CH2-O, CH2-P), 2.83 (t, 2H, J
) 6.5 Hz, S-CH2-CH2-O), 2.34 (s, 3H, CH3-C(O)-); 31P NMR
(D2O) δ 17.8; FAB MS (<0, G/T) m/ e 749 (2M - H)-, 374 (M
- H)-, 272 (M - AcSCH2CH2 + H)-.
9-{2-[O,O′-Bis[(S -isob u t yr ylt h io)e t h yl]p h osp h on o-
m eth oxy]eth yl}a d en in e (4b). Silica gel column chromatog-
raphy [stepwise gradient of methanol (0-6%) in dichlo-
romethane] gave 4b (81%) which was crystallized from CCl4
(75%): mp 83-84 °C; 1H NMR (DMSO-d6) δ 8.11 and 8.06 (2s,
9-{2-[O,O′-Mon o[(S -b e n zoylt h io)e t h yl]p h osp h on o-
m eth oxy]eth yl}a d en in e (9d ). Dowex 1 × 2 resin (acetate
form) chromatography [eluent: linear gradient of acetic acid
(0-2 M)] gave 9d (49%) as a solid after lyophilization in a
mixture of water and dioxane (9:1, v/v): 1H NMR (D2O +
DMSO-d6) δ 8.57 and 8.38 (2s, 1H and 1H, 2-H and 8-H), 8.0-
7.7 (m, 5H, Ph), 4.62 (t, 2H, J ) 5.0 Hz, CH2N), 4.1-3.8 (m,
6H, S-CH2-CH2-O, CH2-P, CH2-CH2-N), 3.22 (t, 2H, J ) 6.3

4964 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25
Benzaria et al.
Hz, S-CH2-CH2-O); 31P NMR (D2O + DMSO-d6) δ 16.8; FAB
MS (<0, G/T) m/ e 873 (2M - H)-, 436 (M - H)-.
Ack n ow led gm en t. These investigations were sup-
ported by grants from the CNRS, Agence Nationale de
Recherches sur le SIDA (ANRS, France), and Associa-
tion pour la Recherche sur le Cancer (ARC, France). Two
of us are particularly grateful to ANRS (S.B.) and ARC
(H.P.) for a fellowship. We warmly thank S. Schmidt
and G. Albrecht for excellent assistance in performing
the anti-HIV-1 assays. We are indebted to Dr. Victor
Marquez (NIH, Bethesda, MD) for critical reading of the
manuscript. The assistance of G. S. Hanrahan in typing
this manuscript is also greatly appreciated.
Biologica l Meth od s. An ti-HIV Assa ys on Cell Cu ltu r e.
The origin of the viruses and the techniques used for measur-
ing inhibition of virus multiplication have previously been
described.16 Briefly, in MT-4 cells, the determination of the
antiviral activity of the pronucleotides was based on a reduc-
tion of HIV-1-IIIB-induced cytopathogenicity, the metabolic
activity of the cells being measured by the property of
mitochondrial dehydrogenases to reduce 3-(4,5-dimethyl-
thiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) into for-
mazan.16,40,41 For CEM cells, the production of virus HIV-LAI
was measured by quantification of the reverse transcriptase
activity associated with the virus particle released in the
culture supernatant.16,41 Cells, MT-4 and CEM-SS, were
respectively incubated with 50 or 100 TCID50 of viruses during
30 min; after virus adsorption, unbound particles were elimi-
nated by two washes, and cells were cultured in the presence
of different concentrations of test compounds for 5 days before
virus production determination. The 50% effective concentra-
tion (EC50) was derived from the computer-generated median
effect plot of the dose-effect data.42 In parallel experiments,
cytoxicity of the test compounds was measured after an
incubation of 5 days in their presence using the colorimetric
MTT test.43 The 50% cytotoxic concentration (CC50) is the
concentration at which OD540 was reduced by one-half and was
calculated using the program mentioned above.
An ti-HSV Assa ys on Cell Cu ltu r e. The anti-HSV assays
were performed by a method adapted from that described by
De Clercq.44,45 The herpes simplex virus type 1 [HSV-1, strain
F (ATCC VR-733)] and the herpes simplex virus type 2 [HSV-
2, strain G (ATCC VR-734)] were propagated on Vero cells and
human embryonic fibroblasts (cell line MRC-5).
All assays were carried out on confluent cells in 96-well
microtiter plates. The cells were grown in Eagle’s minimum
essential medium (MEM) supplemented with 5% fetal bovine
serum (FBS). They were infected with 100 CCID50 of virus
for 1 h at 37 °C and immediately thereafter exposed to various
concentrations of the test compounds (from 1 mM to 1 µM).
The viral cytopathic effect (CPE) was recorded daily. Antiviral
activity was expressed as EC50 (50% effective concentration),
that is, the concentration of compound required to reduce the
viral CPE by 50% when it had reached completion in the
control virus-infected cell cultures.
Drug cytotoxicity was determined using the quantitative
colorimetric MTT assay.43 In this assay, different dilutions
of the test compounds were added to 96-well plates containing
about 3 × 103 cells (approximately 30% confluent). When cells
in the control wells (without drug) reached confluence, 0.05
mL of MTT (5 mg/mL) was added to each well. After a 3 h
incubation at 37 °C, acidified 2-propanol was added to each
well. The optical density in each well of the plate was
determined using a 96-well plate photometer at 540 nm.
Toxicity was considered to occur if the monolayer remained
less than 50% confluent.
Sta bility a n d Decom p osition Stu d ies. We used the
previously described HPLC procedure.26 Briefly, the cleaning
precolumn was a Guard-Pak insert (Delta-Pak C18 100 Å) in
a Guard-Pak holder (Waters, Saint Quentin, France). The
analytical column used was a Hypersil C18, 3 µm, 120 Å, 4.6
× 100 mm (Shandon, Eragny, France). The elution system
was prepared as follows: stock solution S, 1.0 M, pH 6.0, from
ammonium acetate and acetic acid pro analysis (Merck,
Darmstadt, Germany); eluent A 2.5 mM buffered tetrabutyl-
ammonium sulfate [one bottle of PIC A reagent (Waters) in
2.0 L of water]; eluent B (v/v), S 10, water 90; eluent C (v/v/v),
S 10, acetonitrile 50, water 40. The crude sample (80 µL,
initial concentration of 2, 3, or 4a -c of 5 × 10-5 M) was
injected into the precolumn and eluted with eluent A during
5 min. Then, the switching valve for connecting the precolumn
to the column was activated, and a linear gradient from eluent
B to eluent C programmed over 30 min was used. The
retention times (min) were 1, 13.2; 2, 22.1; 3, 32.1; 4a , 23.9;
4b, 33.3; 4c, 37; 4d , 36.3; 9a , 16.8; 9b, 20.7; 9c, 22.8; 9d , 23.1;
mono(DTE)PMEA, 16.6; mono(POM)PMEA, 20.5.
Refer en ces
(1) For a review, see: Naesens, L.; Balzarini, J .; De Clercq, E.
Therapeutic potential of PMEA as an antiviral drug. Med. Virol.
1994, 4, 147-159.
(2) De Clercq, E. Broad-spectrum anti-DNA virus and anti-retro-
virus activity of phosphonylmethoxyalkylpurines and -pyrim-
idines. Biochem. Pharmacol. 1991, 42, 963-972.
(3) Walker, R. E.; Vogel, S. E.; J affe, H. S.; Polis, M. A.; Kovacs, J .
A.; Faloon, J .; Davey, R. T.; Ebeling, D.; Cundy, K.; Paar, D.;
Markowitz, N.; Masur, H.; Lane, H. C. A phase I/II study of
PMEA in HIV infected patients. Abstracts of Papers; 1st National
Conference on Human Retroviruses and Related Infections,
Washington, DC, 1993; Abstract 522.
(4) Balzarini, J .; Naesens, L.; Slachmuylders, J .; Niphuis, H.;
Rosenberg, I.; Holy, A.; Schellekens, H.; De Clercq, E. 9-(2-
Phosphonylmethoxyethyl)adenine (PMEA) effectively inhibits
retrovirus replication in vitro and simian immunodeficiency
virus infection in rhesus monkeys. AIDS 1991, 5, 21-28.
(5) Bronson, J . J .; Ghazzouli, I.; Hitchcock, M. J . M.; Russel, J . W.;
Klunk, L. J .; Kern, E. R.; Martin, J . C. In vivo anti-retrovirus
and anti-cytomegalovirus activity of 9-((2-phosphonylmethoxy)-
ethyl)adenine (PMEA). Abstracts of Papers; 5th International
Conference on AIDS, Montreal, Canada, 1989; Abstract M.C.P.
74.
(6) Starrett, J . E., J r.; Tortolani, D. R.; Russel, J .; Hitchock, M. J .
M.; Whiterock, V.; Martin, J . C.; Mansuri, M. M. Synthesis, oral
bioavailability determination, and in vitro evaluation of prodrugs
of the antiviral agent 9-[2-(phosphonomethoxy)ethyl]adenine
(PMEA). J . Med. Chem. 1994, 37, 1857-1864.
(7) Alexander, P.; Holy, A. Prodrugs of analogs of nucleic acid
components. Collect. Czech. Chem. Commun. 1994, 59, 2127-
2165.
(8) Pe´rigaud, C.; Girardet, J .-L.; Gosselin, G.; Imbach, J .-L. Com-
ments on nucleotide delivery forms. In Advances in antiviral
drug design; De Clercq, E., Ed.; J AI Press Inc.: Greenwich, CT,
1996; Vol. 2, pp 147-172.
(9) J ones, R. J .; Bischofberger, N. Minireview: nucleotides prodrugs.
Antiviral Res. 1995, 27, 1-17.
(10) Midoux, P.; Negre, E.; Roche, A. C.; Mayer, R.; Monsigny, M.;
Balzarini, J .; De Clercq, E.; Mayer, E.; Ghaffar, A.; Gangemi, J .
D. Drug targeting: anti-HSV-1 activity of mannosylated polymer-
bond 9-(2-phosphonylmethoxyethyl)adenine. Biochem. Biophys.
Res. Commun. 1990, 167, 1044-1049.
(11) Alexander, P.; Holy, A.; Masojidkova, M. Synthesis of 9-(2-
phosphinomethoxyethyl)adenine and related compounds. Collect.
Czech. Chem. Commun. 1994, 59, 1870-1878.
(12) Alexander, P.; Holy, A.; Masojidkova, M. Preparation of 9-(2-
phosphonomethoxyethyl)adenine esters as potential prodrugs.
Collect. Czech. Chem. Commun. 1994, 59, 1853-1869.
(13) Glazier, A. Potent topical anti-herpes activity of a lipophilic
phosphorus prodrug for the antiviral agent PMEA. Antiviral Res.
1995, 26, A306.
(14) Glazier, A.; Buckheit, R.; Yanachkova, M.; Yanachkov, I.; Wright,
G. E. Lipophilic phosphorus prodrugs for the antiviral agent
PMEA. Antiviral Res. 1994, 23 (Suppl. 1), 65.
(15) Starrett, J . E., J r.; Tortolani, D. R.; Hitchock, M. J . M.; Martin,
J . C.; Mansuri, M. M. Synthesis and in vitro evaluation of a
phosphonate prodrug: bis(pivaloyloxymethyl)9-(2-phosphonyl-
methoxyethyl)adenine. Antiviral Res. 1992, 19, 267-273.
(16) Puech, F.; Gosselin, G.; Lefebvre, I.; Pompon, A.; Aubertin, A.
M.; Kirn, A.; Imbach, J .-L. Intracellular delivery of nucleoside
monophosphates through a reductase-mediated activation pro-
cess. Antiviral Res. 1993, 22, 155-174.
(17) Srinivas, R. V.; Robbins, B. L.; Connely, M. C.; Gong, Y. F.;
Bischofberger, N.; Fridland, A. Pivaloyloxymethyl esters of
acyclic nucleoside phosphonates. Int. Antiviral News 1994, 2,
53-55.
(18) Naesens, L.; Neyts, J .; Balzarini, J .; Bischofberger, N.; De Clercq,
E. Antiviral efficacy in mice of oral bis(POM)-PMEA, the bis-
(pivaloyloxymethyl) ester prodrug of 9-(2-phosphonylmethoxy-
ethyl)adenine. Antiviral Res. 1995, 26, A276.

SATE Derivatives as Potential PMEA Prodrugs
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25 4965
(19) Naesens, L.; Neyts, J .; Balzarini, J .; Bischofberger, N.; De Clercq,
E. Pharmacokinetics in mice of bis(POM)-PMEA, the bis-
(pivaloyloxymethyl) ester prodrug of 9-(2-phosphonylmethoxy-
ethyl)adenine. Antiviral Res. 1995, 26, A277.
(20) Naesens, L.; Neyts, J .; Balzarini, J .; Bischofberger, N.; De Clercq,
E. In vivo antiretroviral efficacy of oral bis(POM)-PMEA, the
bis(pivaloyloxymethyl)prodrug of 9-(2-phosphonylmethoxyethyl)-
adenine (PMEA). Nucleosides Nucleotides 1995, 14, 767-770.
(21) Naesens, L.; Balzarini, J .; Bischofberger, N.; De Clercq, E.
Antiretroviral activity and pharmacokinetics in mice of oral bis-
(pivaloyloxymethyl)-9-(2-phosphonylmethoxyethyl)adenine, the
bis(pivaloyloxymethyl) ester prodrug of 9-(2-phosphonylmeth-
oxyethyl)adenine. Antimicrob. Agents Chemother. 1996, 40, 22-
34.
(22) Srinivas, R. V.; Robbins, B. L.; Connely, M. C.; Gong, Y. F.;
Bischofberger, N.; Fridland, A. Metabolism and in vitro anti-
retroviral activities of bis(pivaloyloxymethyl)prodrugs of acyclic
nucleoside phosphonates. Antimicrob. Agents Chemother. 1993,
37, 2247-2250.
(23) Cundy, K. C.; Fishback, J . A.; Shaw, J .-P.; Lee, M. L.; Socke, K.
F.; Yisor, G.C.; Lee, W. A. Oral bioavailability of the antiretro-
viral agent 9-(2-phosphonylmethoxyethyl)adenine (PMEA) from
three formulations of the prodrug bis(pivaloyloxymethyl)-PMEA
in fasted male cynomolgus monkeys. Pharm. Res. 1994, 11, 839-
843.
(31) Holy, A.; Rosenberg, I. Synthesis of 9-[2-(phosphonylmethoxy)-
ethyl]adenine and related compounds. Collect. Czech. Chem.
Commun. 1987, 52, 2801-2809.
(32) Bronson, J . J .; Ghazzouli, I.; Hitchcock, M. J . M.; Webb, R. R.,
II; Martin, J . C. Synthesis and antiviral activity of the nucleotide
analogue (S)-1-[3-hydroxy-2(phosphonylmethoxy)propyl]cytosine.
J . Med. Chem. 1989, 32, 1457-1463.
(33) Bronson, J . J .; Kim, C. U.; Ghazzouli, I.; Hitchcock, M. J . M.;
Kern, E. R.; Martin, J . C. Synthesis and antiviral activity of
phosphonylmethoxyethyl derivatives of purine and pyrimidine
bases. In Nucleotides analogues as antiviral agents; Martin, J .
C., Ed.; American Chemical Society: Washington, DC; 1989; Vol.
401, pp 72-87.
(34) Bailey, W. F.; Rivera, A. D. Acetolysis of cyclic acetals
:
regioselective acylative cleavage of cyclic formals. J . Org. Chem.
1984, 49, 4958-4964.
(35) Foye, W. O.; Kauffman, J . M.; Kim, Y. H. Synthesis of 4-alkyl-
4H-benzo[e]- and 4H-pyrido[2,3-e]-1,2,4-triazin-3-one 1-oxides for
potential anticancer activity. J . Heterocycl. Chem. 1982, 19, 497-
501.
(36) Miyake, A.; Yamaoka, M. Method for producing 1-iodoalkyl
acylates. Eur. Patent Appl. E.P. 143 601, J une 5, 1985; Chem.
Abstr. 1986, 104, 5589z.
(37) Palu, G.; Stefanelli, S.; Rassu, M.; Parolin, C.; Balzarini, J .; De
Clercq, E. Cellular uptake of phosphonylmethoxyalkylpurine
derivatives. Antiviral Res. 1991, 16, 115-119.
(24) GS840, a PMEA prodrug, enters trials for HIV-infection. Anti-
viral Agents Bull. 1994, 7, 232-233.
(25) Barditch-Crovo, P. A.; Toole, J .; Burgee, H.; Wachsman, M.;
Ebeling, D.; Cundy, K. C.; J affe, H. S.; Lietman, P. S. A
randomized, double-blind, placebo-controlled phase I/II evalu-
ation of 9-[-2-(bispivaloyloxy-methyl) phosphonylmethoxyethyl]
adenine (bis-POM PMEA), an orally bioavailable prodrug of the
anti-HIV nucleotide, PMEA. Antiviral Res. 1995, 26, A229.
(26) Lefebvre, I.; Pe´rigaud, C.; Pompon, A.; Aubertin, A.-M.; Girardet,
J .-L.; Kirn, A.; Gosselin, G.; Imbach, J .-L. Mononucleoside
phosphotriester derivatives with S-acyl-2-thioethyl bioreversible
phosphate-protecting groups: intracellular delivery of 3′-azido-
2′,3′-dideoxythymidine 5′-monophosphate. J . Med. Chem. 1995,
38, 3941-3950.
(27) Pe´rigaud, C.; Gosselin, G.; Lefebvre, I.; Girardet, J . L.; Benzaria,
S.; Barber, I.; Imbach, J . L. Rational design for cytosolic delivery
of nucleoside monophosphates: “SATE” and “DTE” as enzyme-
labile transient phosphate protecting groups. Bioorg. Med. Chem.
Lett. 1993, 3, 2521-2526.
(28) Pe´rigaud, C.; Aubertin, A. M.; Benzaria, S.; Pe´licano, H.;
Girardet, J .-L.; Maury, G.; Gosselin, G.; Kirn, A.; Imbach, J .-L.
Equal inhibition of the replication of human immunodeficiency
virus in human T-cell culture by ddA bis(SATE)phosphotriester
and 3′-azido-2′,3′-dideoxythymidine. Biochem. Pharmacol. 1994,
48, 11-14.
(29) Benzaria, S.; Girardet, J .-L.; Pe´rigaud, C.; Aubertin, A. M.;
Pe´licano, H.; Maury, G.; Gosselin, G.; Kirn, A.; Imbach, J .-L.
The SATE pronucleotide derivative of ddA : a more potent HIV
inhibitor than AZT. Nucleic Acids Symp. Ser. 1994, 31, 129-
130.
(38) Blackburn, G. M.; Gait, M. J . Nucleic acids in chemistry and
biology; IRL Press: Oxford, New York, Tokyo, 1990; pp 71-135.
(39) Hanes, C. S.; Isherwood, F. A. Separation of the phosphoric
esters on the filter paper chromatogram. Nature 1949, 154,
1107-1112.
(40) Pauwels, R.; Balzarini, J .; Baba, M.; Snoeck, R.; Schols, D.;
Herdewijn, P.; Desmyter, G.; De Clercq, R. Rapid and automated
tetrazolium-based colorimetric assay for the detection of anti-
HIV compounds. J . Virol. Methods 1988, 20, 309-321.
(41) Moog, C.; Wick, A.; Le Bur, P.; Kirn, A.; Aubertin, A.-M. Bicyclic
imidazo derivatives, a new class of highly selective inhibitors
for the human immunodeficiency virus type 1. Antiviral Res.
1994, 24, 274-288.
(42) Chou, J .; Chou, T.-C. Dose-effect analysis with microcomput-
ers: quantitation of ED50, LD50, synergism, antagonism, low-
dose risk, receptor binding and enzyme kinetics. Computer
software for Apple II series and IBM-PC and instruction manual;
Elsevier-Biosoft, Elsevier Science Publishers: Cambridge, U.K.,
1985; pp 19-28.
(43) Mosmann, T. Rapid colorimetric assay for cellular growth and
survival: application to proliferation and cytotoxic assays. J .
Immunol. Methods 1983, 65, 55-63.
(44) De Clercq, R.; Holy, A.; Rosenberg, I.; Sakuma, T.; Balzarini,
J .; Maudgal, P. L. A novel selective broad-spectrum anti-DNA
virus agent. Nature 1986, 323, 464-467.
(45) De Clercq, E.; Descamps, J .; Verhelst, J .; Walker, R. T.; J ones,
A. S.; Torrence, P. F.; Shugar, D. Comparative efficacy of
antiherpes drugs against different strains of herpes simplex
virus. J . Infect. Dis. 1980, 141, 563-574.
(30) Girardet, J .-L.; Pe´rigaud, C.; Aubertin, A.-M.; Gosselin, G.; Kirn,
A.; Imbach, J .-L. Increase of the anti-HIV activity of D4T in
human T-cell culture by the use of the SATE pronucleotide
approach. Bioorg. Med. Chem. Lett. 1995, 5, 2981-2984.
J M960289O