Intracellular trapping of cyclosal-pronucleotides: Modification of prodrugs with amino acid esters

Article abstract of DOI:10.1021/jm800815b

A new class of d4TMP-cycloSal-pronucleotides bearing enzymatically cleavable amino acid esters is reported. These compounds are designed to trap the pronucleotide inside the cell by a fast conversion of a nonpolar ester group into a charged carboxylate. This should prevent efficient diffusion equilibrium across the cell membrane to the extracellular environment, leading to an intracellular accumulation of the compounds. This initial conversion is followed by a slow release of the nucleoside monophosphate (i.e., d4TMP). The concept is proven by hydrolysis studies in phosphate buffer, cell extracts, and human serum. These investigations revealed a high sensitivity of some amino acid ester modifications to conversion by cellular extracts, resulting in the fast release of a charged intermediate, whereas no cleavage of the modification is found in phosphate buffer. In addition, antiviral activities against HIV are presented.

Full text of DOI:10.1021/jm800815b

6592
J. Med. Chem. 2008, 51, 6592–6598
Intracellular Trapping of cycloSal-Pronucleotides: Modification of Prodrugs with Amino Acid
Esters‡
Henning J. Jessen,† Jan Balzarini,§ and Chris Meier*,†
Chemistry Department, Organic Chemistry, Faculty of Sciences, UniVersity of Hamburg, Martin-Luther-King-Platz 6, 20146 Hamburg,
Germany, and Rega Institute for Medical Research, Katholieke UniVersiteit LeuVen, Minderbroedersstraat 10, 3000 LeuVen, Belgium
ReceiVed July 7, 2008
A new class of d4TMP-cycloSal-pronucleotides bearing enzymatically cleavable amino acid esters is reported.
These compounds are designed to trap the pronucleotide inside the cell by a fast conversion of a nonpolar
ester group into a charged carboxylate. This should prevent efficient diffusion equilibrium across the cell
membrane to the extracellular environment, leading to an intracellular accumulation of the compounds.
This initial conversion is followed by a slow release of the nucleoside monophosphate (i.e., d4TMP). The
concept is proven by hydrolysis studies in phosphate buffer, cell extracts, and human serum. These
investigations revealed a high sensitivity of some amino acid ester modifications to conversion by cellular
extracts, resulting in the fast release of a charged intermediate, whereas no cleavage of the modification is
found in phosphate buffer. In addition, antiviral activities against HIV are presented.
Introduction
intracellular concentrations of the pronucleotide, which then
presumably also leads to high concentrations of the released
nucleotide. In order to efficiently trap the lipophilic prodrug,
we designed the second generation of cycloSal-pronucleotides
(“lock-in” cycloSal triesters).14,15 These compounds are
bearing a (carboxy)esterase-cleavable ester (trigger) attached
to the aromatic ring. To avoid a considerable reduction of
the chemical stability due to the electron-withdrawing effect
of the ester moieties, a C2-spacer separates the trigger from
the aromatic system.
Nucleoside analogues are commonly used as antiviral or
antitumor agents.1 The antiviral effect of nucleoside analogues,
such as 2′,3′-dideoxy-2′,3′-didehydrothymidine (d4T, 1), de-
pends on their conversion into the ultimately bioactive tri-
phosphates via mono- and diphosphate formation by cellular
kinases. In the case of d4T, the metabolism-limiting step is the
first phosphorylation catalyzed by the salvage pathway enzyme
thymidine kinase (TK).2,3 As a consequence, attempts to directly
deliver d4T-monophosphate into intact cells would be warranted.
Different lipophilic nucleotide-releasing systems (pronucle-
otides) were designed to achieve a passive membrane transport
of the nucleotide.4 Among these, cycloSal-pronucleotides were
successfully used in the intracellular delivery of e.g. d4TMP
(TK bypass).5 This delivery depends on a chemically triggered
cascade reaction.6 The cycloSala approach has also been applied
to other nucleoside analogues, resulting in improved antiviral
potency of the compounds.7-9Moreover, it was shown that this
technology converts nonactive nucleoside analogues like 2′-
ribofluoro-2′,3′-dideoxyadenosine10 or 5-(E)-bromovinyl-2′-deox-
yuridine (BVDU)11into potent antivirals or could expand the
antiviral spectrum for other nucleoside analogues such as
BVDU.11 Likewise, cycloSal-acyclovirMP compounds entirely
retain their antiviral activity against acyclovir-resistant HSV-1
strains.12 The latter compounds also proved active against pox
viruses, while the parent compound is nonactive.13
Two types of ester-bearing cycloSal triesters have been
developed: the cycloSal-d4TMP acid ester and the cycloSal-
d4TMP alcohol ester. As simple esters were not cleaved by
esterases, more elaborate acylal systems were used to release
the corresponding carboxylates. This approach was based on
the acetoxymethyl (AM) and pivaloyloxymethyl (POM)
modification. Hydrolysis studies in phosphate buffer (PBS,
pH 7.3) and in T-lymphocyte CEM cell extracts showed that
an effective intracellular trapping should be possible if highly
polar cycloSal-d4TMP acids such as 4 (Figure 1) are released
from cycloSal acid esters such as 3 (Figure 1). Chemical
hydrolysis of 4 via the cascade reaction finally led to an
intracellular d4TMP delivery.14,15 The trapping concept was
further proven by enrichment studies with intrinsically
fluorescent cycloSal-pronucleotides using a U-tube setup.16
However, the AM modification suffered from a low stability
in RPMI/FCS incubation media, resulting in modest antiviral
data. Although the POM modification proved to be more
stable under these conditions, the release of a potentially
cytotoxic pivaloic acid represents the main drawback.5
As a result of the the lipophilic character of cycloSal-
phosphate triesters such as 2 (Figure 1) and the chemically
triggered delivery mechanism, a drug concentration equilib-
rium across the cell membrane is formed (Figure 2). However,
for a strong antiviral effect it is necessary to achieve high
In this paper we disclose the synthesis, properties, and
antiviral evaluation of a new series of compounds, based on
the modification of cycloSal acid 4 with different amino acid
esters. The possible cleavage mechanism is summarized in
Figure 3. All metabolites display higher polarity compared
to the parent pronucleotide, in principle preventing diffusion
from the cytoplasm once they are formed. In this context we
assume that a high stability of the pronucleotides in phosphate
buffer or human serum (step a, Figure 2) but a low stability
in cell extracts (step d, Figure 2) is favorable to induce an
‡
This work was presented at the 21st International Conference on
Antiviral Research (ICAR), Montreal, Canada, April 13-17, 2008.
* To whom correspondence should be addressed. Phone: +49-40-42838-
4324. Fax: +49-40-42838-2495. E-mail: chris.meier@chemie.uni-ham-
burg.de.
†
University of Hamburg.
Katholieke Universiteit Leuven.
§
a Abbreviations: AM, acetoxymethyl; POM, pivaloyloxymethyl; cy-
cloSal:, cyclosaligenyl; DCC, dicyclohexylcarbodiimide; HOBt, hydroxy-
benzotriazole.
10.1021/jm800815b CCC: $40.75
2008 American Chemical Society
Published on Web 10/01/2008

Trapping of cycloSal-Pronucleotides
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 20 6593
Figure 1. Structures of target compounds 5, other cycloSal-d4TMPs 2-4, and d4T 1.
Scheme 1. Synthesis of 5-Propionyl-cycloSal-d4TMP Acid 4a
Figure 2. “Lock-in” concept.
a Reagents and conditions. Method A: N,N-DMF-dineopentylacetale,
tBuOH, toluene, reflux. Method B: phenylboronic acid, propionic acid,
paraformaldehyde, toluene, reflux. Method C: H2O2, 0°C. Method D: PCl3,
pyridine, Et2O, -20 °C. Method E: d4T 1, DIPEA, CH3CN, tBuOOH, -20
°C to room temp. Method F: TFA, CH2Cl2.
possible modifications to adjust stability and polarity and the
possibility to circumvent the release of pivaloic acid and
formaldehyde.
Chemistry
The key intermediate 5-propionyl-cycloSal-d4TMP acid 4 was
synthesized using a previously published route (Scheme 1).15
Briefly, commercially available 3-(4-hydroxyphenyl)propanoic
acid 6 was converted into the corresponding tert-butyl ester 7
and subsequently transformed to salicylic alcohol 8 by an ortho-
selective hydroxymethylation.17 Salicylic alcohol 8 was treated
with PCl3 and pyridine in Et2O. Pyridinium chloride was then
Figure 3. Possible cleavage mechanisms of amino acid ester func-
tionalized cycloSal-d4TMPs.
accumulation of the “lock-in” pronucleotides. The main
advantages of the amino acid ester modification compared
to, for example, POM modification are the flexibility of

6594 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 20
Scheme 2. Synthesis of Target Compounds 5a
Jessen et al.
the C2-spacer separates the electron withdrawing amide bond
effectively from the aromatic system, resulting in a slightly
increased stability compared to prototype compound 2 (t1/2
)
4.4 h). Additionally, the chemical stability seems to be roughly
independent of the amino acid and the type of ester, supporting
the assumption that the chemical hydrolysis does only take place
at the phosphate moiety. L-Alanine compound 5g bearing a
nonesterified amino acid differs from this behavior. It displays
a 2-fold increased stability of 12 h. This increase was also found
for 5-propionyl-cycloSal-d4TMP acid 4. We assume that the
negatively charged carboxylate repulsively interacts with hy-
droxide ions, thus enhancing stability of the compounds. The
separated diastereomers of triester 5f illustrate the strong effect
of the configuration at phosphorus on the chemical hydrolysis
properties. The diastereomer 5f fast is hydrolyzed twice as fast
as 5f slow. This is in line with previous experiments.5
In summary, d4T monophosphate was released from all
compounds within a reasonable period of time. We did not find
any interaction of the amino acid ester modification with the
chemical hydrolysis mechanism of cycloSal triesters. Despite
the increased average half-life of compound 5g, this derivative
also selectively released d4TMP.
a Reagents and conditions. Method A: DCC, HOBt, DIPEA, salts of
amino acid esters, DMF. Method B: TFA, CH2Cl2.
removed by filtration, yielding saligenyl chlorophosphite 9 in
excellent purity. This compound was used to prepare 5-tBu-
propionyl-cycloSal-d4TMP 10 in 78% yield using the standard
procedure.15 The tert-butyl ester was subsequently cleaved with
TFA to give key intermediate 4 in a 95% yield as a 0.9:1 mixture
of diastereomers.
The following coupling of 5-propionyl-cycloSal-d4TMP acid
4 with different amino acid ester salts was achieved in DMF
using DCC, HOBt, and DIPEA (Scheme 2). Generally, this
synthesis resulted in excellent yields except for D-alanine methyl
ester 5c. L-Alanine tert-butyl ester 5e was converted to 5-L-
Ala-propionyl-cycloSal-d4TMP 5g by cleavage of the ester
group with TFA. With acids 4 and 5g, reference compounds
for HPLC analysis were available to investigate the different
mechanisms of hydrolysis (Figure 3). Further, both compounds
were used as references in antiviral evaluation.
Hydrolysis in Cell Extracts and Human Serum. The
cleavage of the different ester groups was investigated in
hydrolysis studies using T-lymphocyte CEM cell extracts (CE).
We were able to show that at least for L-alanine bearing
compounds 5d-f the hydrolysis in PBS buffer proceeds via
path 5 (Figure 3) to yield d4TMP directly, whereas in cellular
extracts the amino acid ester is cleaved by an esterase (path 1,
Figure 3). We did not observe a cleavage of the amide bond by
peptidases that may be present in the serum (paths 2 and 3,
dashed lines). The stabilities of the compounds are summarized
in Table 1. A very fast enzymatic cleavage of the trigger moiety
was found for esters of L-alanine and L-proline. Additionally,
the cleavage was found to be faster for benzyl esters compared
to methyl esters. This led to the desired increase of polarity
(e.g., 5f log P ) 3.37 vs 5g log P ) 1.56; Table 1). There was
no dependence of the cleavage rate on the configuration at
phosphorus, which can be taken from the results obtained for
5f fast and 5f slow (t1/2 ) 0.5 h each). These results clearly
show that the rate of cleavage strongly depends on different
factors, which are the amino acid, the stereochemistry of the
amino acid (D-AlaOMe 5c is not cleaved by esterases, whereas
L-AlaOMe 5d was cleaved rapidly), and the type of ester (the
tBu-ester of 5e is not cleaved enzymatically). Such structure-
activity relationship has also been observed for the arylphos-
phoramidate ester prodrugs.18 If there is no enzymatic cleavage
of the trigger, the measured stabilities resemble those obtained
for pure chemical hydrolysis, as this is now the main pathway
taking place. The most distinct difference in chemical stability
and enzymatic stability was found for 5f slow. We measured
an almost 18-fold reduction of stability in cell extracts compared
to phosphate buffer. This complies with our intention to have a
fast enzymatic hydrolysis of, for example, 5f slow preceding
the slow chemical hydrolysis of the trapped compound 5g to
give d4T monophosphate.
As aforementioned, cycloSal triesters are diastereomeric entities.
To investigate the effect of the configuration at phosphorus, we
enriched both diastereoisomers of 5f up to a ratio of 9:1 (judged
from 31P NMR) by preparative RP-HPLC. Because we failed to
assign the absolute stereochemistry up to now, these compounds
will be named as 5f fast and 5f slow, which refer to the mobility
properties on the reversed-phase silica gel chromatography column.
Compounds bearing L-proline caused a mixture of diastereomers
and rotamers, which was proven by coalescence NMR experiments.
1
All triesters 5 were characterized by means of H, 13C, and
31P NMR spectroscopy as well as high resolution mass
spectrometry.
Results and Discussion
Chemical Stability. Stability studies of cycloSal-triesters 5
were conducted in aqueous 25 mM phosphate buffer (PBS pH
7.3). The hydrolysis products were detected by analytical RP-
HPLC. Additionally, hydrolysis studies of compounds 5 in a
1:1 mixture of DMSO-d6 and PBS (pH 7.3) followed by 31P
NMR spectroscopy revealed d4TMP and d4T diphosphate
(d4TDP) as the sole phosphorus containing products resulting
from chemical hydrolysis of triesters 5 by either hydroxide (to
give d4TMP) or phosphate (to give d4TDP, Figure 4). Though
the detection of small amounts of d4TDP is more reliable with
NMR techniques, it can also be found in chromatograms if
appropriate methods are used. The HPLC-determined half-lives
are summarized in Table 1. They relate to the slow chemically
induced cleavage of the phosphorus triester and not to the
cleavage of the trigger moiety. The average half-life of all
diastereomeric cycloSal-triesters 5 is about 6 h. This shows that
Further investigations in 5% and 50% human serum diluted
with phosphate buffered saline (pH 6.8) were carried out with
compounds 5f fast and 5f slow. In 5% human serum there was
no effect on the stability, and the half-lives measured slightly
increased compared to those obtained at pH 7.3. However, in
50% human serum we observed a striking difference for both
diastereomers. Compound 5f fast remained stable under these
conditions with hydrolysis half-lives of 4.5 h in PBS (pH 7.3)

Trapping of cycloSal-Pronucleotides
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 20 6595
Figure 4. Chemical hydrolysis of 5i followed by 31P NMR spectroscopy {solvent, DMSO-d6/50 mM PBS buffer (pH 7.3), 1:1 (v/v)}. Spectra were
recorded at t0 (top) and at the end of the hydrolysis (bottom), revealing the release of 85% d4TMP and 15% d4TDP. H3PO4 was used as external
reference.
Table 1
EC50 (µM)c
t1/2 (h)b
CEM/Oh
CEM/TK- i
HIV-2
a
substituent
log Pcalcd
PBSe
CEf
serumg
HIV-1
HIV-2
CC50(µM)d
5a
5b
5c
5d
GlyOMe-pr
GlyOBn-pr
D-AlaOMe-pr
L-AlaOMe-pr
L-AlaOtBu-pr
L-AlaOBn-pr
L-AlaOBn-pr
L-AlaOBn-pr
L-Ala-pr
L-ProOMe-pr
L-ProOBn-pr
5-POM-pr
5-AM-pr
H
1.05
2.83
1.59
1.59
2.42
3.37
3.37
3.37
1.56l
1.82
3.60
4.04
2.18
0.28
2.10l
-0.48
6.5
6.2
5.0
6.4
5.7
4.5
8.8
6.4
12
4.5
4.8
6.2
1.1
6.8
0.5
0.5
ndk
7
1.1
0.4
0.38
0.25
4.0
ndk
naj
ndk
ndk
ndk
ndk
ndk
4.4
1.3
ndk
ndk
ndk
ndk
ndk
ndk
ndk
ndk
naj
2.0 ( 0.0
1.6 ( 0.6
0.47 ( 0.12
0.4 ( 0.07
0.7 ( 0.0
4.0 ( 0.0
0.4 ( 0.0
4.2 ( 0.5
2.5 ( 0.7
3.8 ( 0.0
1.0 ( 0.5
1.0 ( 0.3
0.3 ( 0.1
4.1 ( 0.8
2.8 ( 1.1
21 ( 1.4
220
75
61
152
53
48
61
23
77
182
46
0.31 ( 0.08
0.4 ( 0.07
0.8 ( 0.3
5e
0.3 ( 0.07
0.3 ( 0.0
0.4 ( 0.1
5f fast
5f slow
5f mix
5g
5h
5i
3a
3b
2
4
0.3 ( 0.1
0.3 ( 0.0
0.4 ( 0.3
0.3 ( 0.1
0.4 ( 0.0
0.5 ( 0.2
0.4 ( 0.0
7.9
6.1
5.6
4.3
4.4
13
1.3 ( 0.0
1.6 ( 0.6
0.7 ( 0.42
0.23 ( 0.04
0.20 ( 0.11
0.10 ( 0.02
0.14 ( 0.10
0.48 ( 0.45
1.1 ( 0.44
0.33 ( 0.11
0.53 ( 0.39
0.13 ( 0.04
0.80 ( 0.20
0.63 ( 0.21
0.7 ( 0.08
25 ( 19.1
0.90 ( 0.28
50 ( 30
24
81
31
74
C(O)OH-pr
1
naj
47.5 ( 26.3
234
a Calculated partition coefficients (log P) calculated using log P function implemented in ChemDraw 6.0. b Hydrolysis half-lives. c Antiviral activity in
T-lymphocytes: 50% effective concentration (shown values are the mean of two to three independent experiments). d Cytotoxicity: 50% cytotoxic concentration.
e 25 mM phosphate buffer (pH 7.3). f CEM cell extracts (pH 6.9). g 50% human serum, pH 6.8. h Wild-type CEM cells. i Thymidine kinase-deficient CEM
cells. j na: not applicable. k nd: not determined. l Calculated for protonated acid.
and 4.4 h in human serum. In contrast, the half-life of compound
5f slow was reduced from 8.8 h in PBS (pH 7.3) to 1.3 h in
50% human serum. Thus, in human serum, stereochemistry at
phosphorus is important for the enzymatic stability, which was
not the case in cell extracts. We have not yet figured out the
reason for this behavior, but first investigations may point toward
an interaction of the triester with butyrylcholine esterase (BChE).
Previous experiments have shown that always one diastereomer
of different cycloSal triesters inhibited BChE by irreversible
binding to the active site of the enzyme. To circumvent this
problem, bulky substituents can be placed in the 3-position of
the aromatic moiety.19 However, the reduced half-life of
compound 5f slow may also originate from unspecific protein
binding.
proved to be equipotent against HIV-1 and HIV-2 in wild-type
CEM/0 cells as 1. All triesters 5 are markedly more active
against HIV-2 in CEM/TK- cells compared to the parent
nucleoside 1. The worst TK--activity was found for the
L-proline benzyl ester 5i, which may be due to its poor solubility
in water. In contrast, a full retention of activity in TK--cells
was found for 5-GlyOBn-Pr-cycloSal-d4TMP 5b and 5-L-
AlaOBn-Pr-cycloSal-d4TMP 5f, indicating that a benzyl ester
is a favorable modification. However, we have no explanation
of why the mixture of compounds 5f is 3-fold more active in
TK-deficient cells compared to the separated diastereomers. We
cannot exclude that this is within the experimental error. On
the other hand, the L-alanine modified triester 5g showed a slight
loss of TK- activity, which may result from its hampered
diffusion across the cell membrane. Otherwise, 5f should not
have higher TK- activity compared to 5g, as the first one is
rapidly converted into the latter in intact cells. Thus, if
compound 5g is released inside the cell, e.g., from compound
Antiviral Evaluation. All newly prepared triesters were
evaluated for their anti-HIV activity in wild-type CEM/0 and
mutant thymidine kinase-deficient CEM/TK- cells (Table 1).
As reference compound, d4T 1 was used. cycloSal-d4TMPs 5

6596 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 20
Jessen et al.
combustion of the compounds or varying amounts of water but
were found to be pure by rigorous HPLC analysis. Diethyl ether
was dried over sodium/benzophenone and distilled under nitrogen.
THF was dried over potassium/benzophenone and distilled under
nitrogen. Pyridine, CH2Cl2, and CH3CN were distilled from calcium
hydride under nitrogen. N,N-Diisopropylethylamine and triethy-
lamine were distilled from sodium prior to use. The solvents for
HPLC were obtained from Merck (CH3CN, HPLC grade).
Compounds 4 and 7-10 were synthesized as reported in ref 14,
and analytical data were according to those published.
5f, an intracellular accumulation seems to be possible. The
increased cytotoxicity of some of the triesters 5 may reflect the
higher amount of bioactive nucleoside phosphates formed in
the cells interacting with the cellular processes compared to the
situation starting the metabolism from the parent nucleoside 1.
Generally, we did not find cytotoxic effects of salicylic alcohol
and masks derived from it.
Conclusion
General Procedure for the Preparation of Amino Acid
Ester Modified cycloSal-d4TMPs 5a-i. The reactions were carried
out under a nitrogen atmosphere. Amino acid ester salts (2.0 equiv)
were suspended in dry DMF and solubilized by addition of dry
DIPEA (Hu¨nigs base, 2.0 equiv). 5-Propionyl-cycloSal-d4TMP acid
(4, 1.0 equiv), HOBt (1.1 equiv), and DCC (1.3 equiv) were added
successively. The solution was stirred 16 h at room temperature.
The solvent was removed in vacuo. The residue was diluted with
ethyl acetate and washed twice with water. The organic layer was
dried and concentrated (Na2SO4), and the crude products were
purified by preparative TLC (Chromatotron) and lyophilized from
CH3CN/H2O, 1:1 (v/v).
A new synthetic route to amino acid ester modified cycloSal-
pronucleotides has been developed. It has been shown that some
of the compounds are rapidly converted in cell extracts to
compounds with increased polarity. Further, the cleavage
products release d4T monophosphate in a subsequent chemical
hydrolysis. Therefore, if 5-L-AlaOBn-Pr-cycloSal-d4TMP 5f is
delivered into cells, a fast hydrolysis to 5-L-Ala-Pr-cycloSal-
d4TMP 5g should enrich the prodrug and eventually lead to
higher concentrations of d4TMP. The chemical stabilities of
compounds 5 were independent of the amino acid ester attached
to the C2-spacer, but enzymatic stabilities strongly depended
on both the amino acid and ester. In general, modifications with
L-alanine benzyl esters seem to have the most promising
properties, both concerning their hydrolysis and antiviral profile,
which is similar to results obtained for phosphoramidate
prodrugs, although the structure is considerably different.18 As
a result, the modification of cycloSal-pronucleotides with amino
acid esters can be used as a trigger that is chemically inert under
physiological conditions but is cleaved in cellular extracts by
enzymes. This led to an increase of polarity (5f log P ) 3.37
vs 5g log P ) 1.56), which may hamper diffusion of the
compound into the extracellular matrix. Further, with this
concept, polarity and enzymatic stability can be adjusted.
The novel 5-L-AlaOBn-propionyl-cycloSal mask has already
been applied to other nucleoside analogues, and results of the
antiviral evaluation of the compounds will be published
elsewhere. To further prove the possible enrichment of lock-in
modified compounds, we are currently conducting fluorescence
microscopy studies. Finally, we are running our recently
developed stereoselective synthesis to obtain compounds 5 as
pure isomers.
5-GlyOMe-propionyl-cycloSal-d4TMP (5a). Quantities were
as follows: 5-propionyl-cycloSal-d4TMP acid (4, 30 mg, 65 µmol),
DIPEA (22 µL, 130 µmol), glycine methyl ester ·HCl (16 mg, 130
µmol), HOBt (11 mg, 72 µmol), DCC (18 mg, 85 µmol), dissolved
in 1 mL dry DMF. Purification was done by preparative TLC
(Chromatotron; CH2Cl2/MeOH, gradient elution 0-10%). Yield:
29 mg (54 µmol, 83%) of a diastereomeric mixture (ratio 0.8:1.0)
1
as a colorless foam. H NMR (400 MHz, DMSO-d6): δ ) 11.34
(2 × s, 1H, NHhet), 8.33 (t, 1H, J ) 5.8 Hz, NH-11), 7.30-7.24
(m, 1H, H-4), 7.20, 7.19 (2 × q, 1H, J ) 1.2 Hz, H-6het), 7.15-7.08
(m, 2H, H-6, H-3), 6.82-6.76 (m, 1H, H-1′), 6.42, 6.36 (2 × ddd,
J ) 1.7 Hz, J ) 1.7 Hz, J ) 6.0 Hz, H-3′), 6.05-5.98 (m, 1H,
H-2′), 5.53-5.30 (m, 2H, H-7), 5.00-4.90 (m, 1H, H-4′), 4.40-4.21
(m, 3H, H-12, H-5′), 3.82 (2 × d, 2H, J ) 5.8 Hz, H-12), 3.61 (s,
3H, OMe), 2.90-2.77 (m, 2H, H-8), 2.47-2.36 (m, 2H, H-9), 1.66,
1.59 (2 × d, 3H, J ) 0.9 Hz, Mehet) ppm. 31P NMR (162 MHz, 1H
decoupled, DMSO-d6): δ ) -8.71, -8.77 ppm.
5-GlyOBn-propionyl-cycloSal-d4TMP (5b). Quantities were as
follows: 5-propionyl-cycloSal-d4TMP acid (4, 50.0 mg, 108 µmol),
DIPEA (38 µL, 0.22 mmol), glycine benzyl ester ·p-TsOH (73 mg,
0.22 mmol), HOBt (18 mg, 0.12 mmol), DCC (30 mg, 0.14 mmol),
dissolved in 1 mL dry DMF. Purification was done by flash
chromatography (CH2Cl2/MeOH, 15:1 v/v). Yield: 51 mg (84 µmol,
78%) of a diastereomeric mixture (ratio 0.9:1.0) as a colorless foam.
1H NMR (400 MHz, DMSO-d6): δ ) 11.34, 11.33 (2 × s, 1H,
NHhet), 8.38 (2 × t, 1H, J ) 5.8 Hz, NH-11), 7.40-7.30 (m, 5H,
Experimental Section
NMR spectra were recorded with a Bruker AMX 400 or a Bruker
1
DRX 500 Fourier transform spectrometer. All H and 13C NMR
H
H
aryl), 7.28-7.23 (m, 1H, H-4), 7.20, 7.19 (2 × q, 1H, J ) 1.2 Hz,
het-6), 7.14-7.06 (m, 2H, H-6, H-3), 6.81-6.78 (m, 1H, H-1′),
chemical shifts (δ) are quoted in parts per million (ppm) downfield
from tetramethylsilane and calibrated on solvent signals. The 31P
NMR chemical shifts (proton decoupled) are quoted in ppm using
H3PO4 as the external reference. The spectra were recorded at room
temperature. Mass spectra were obtained with a VG Analytical VG/
70-250 F [FAB (double focusing), matrix m-nitrobenzyl alcohol]
spectrometer. ESI mass spectra were recorded with a VG Analytical
Finnigan ThermoQuest MAT 95 XL spectrometer. For thin layer
chromatography (TLC) Merck precoated 60 F254 plates with a 0.2
mm layer of silica gel were used; sugar-containing compounds were
visualized with sugar spray reagent (0.5 mL of 4-methoxybenzal-
dehyde, 9 mL of EtOH, 0.5 mL of concentrated sulfuric acid, and
0.1 mL of glacial acetic acid). All preparative TLC experiments
were performed on a Chromatotron (Harrison Research, model
7924T) using glass plates coated with 1, 2, or 4 mm layers of Merck
60 PF254 silica gel containing a fluorescent indicator. For column
chromatography, Merck silica gel 60, 230-400 mesh, was used.
Analytical HPLC was performed on a Merck-Hitachi HPLC system
(D-7000) equipped with a LiChroCART 125-3 column containing
reversed phase silica gel Lichrospher 100 RP 18 (5 µM; Merck,
Darmstadt, Germany). The lyophilized products 5 did not give
useful microanalytical data most probably because of incomplete
6.41, 6.35 (2 × ddd, 1H, J ) 1.7 Hz, J ) 1.7 Hz, J ) 6.0 Hz,
H-3′), 6.05-5.97 (m, 1H, H-2′), 5.51-5.30 (m, 2H, H-7), 5.12 (s,
2H, H-14), 4.98-4.92 (m, 1H, H-4′), 4.38-4.22 (m, 2H, H-5′),
3.89, 3.88 (2 × d, 2H, J ) 5.9 Hz, H-12), 2.89-2.75 (m, 2H, H-8),
2.47-2.40 (m, 2H, H-9), 1.66, 1.58 (2 × d, 3H, J ) 1.1 Hz, Mehet
)
ppm. 31P NMR (162 MHz, 1H decoupled, DMSO-d6): δ ) -8.70,
-8.77 ppm.
5-D-AlaOMe-propionyl-cycloSal-d4TMP (5c). Quantities were
as follows: 5-propionyl-cycloSal-d4TMP acid (4, 40 mg, 86 µmol),
DIPEA (30 µL, 0.17 mmol), D-alanine methyl ester·HCl (25 mg,
0.18 mmol), HOBt (15 mg, 0.11 mmol), DCC (24 mg, 0.11 mmol),
dissolved in 1 mL dry DMF. Purification was done by preparative
TLC (Chromatotron; EtOAc/MeOH, gradient elution 0-2%; second
purification with pure CH2Cl2). Yield: 17 mg (31 µmol, 36%) of a
diastereomeric mixture (ratio 0.9:1.0) as a colorless foam. 1H NMR
(400 MHz, DMSO-d6): δ ) 11.35, 11.34 (2 × s, 1H, NHhet), 8.27
(d, 1H, J ) 6.5 Hz, NH-11), 7.23-7.15 (m, 2H, Hhet-6, H-4),
7.11-7.09 (m, 1H, H-6), 7.03, 7.00 (2 × d, 1H, J ) 8.4 Hz, H-3),
6.82-6.75 (m, 1H, H-1′), 6.42, 6.35 (2 × ddd, J ) 1.6 Hz, J )
1.6 Hz, J ) 6.1 Hz, H-3′), 6.04-5.98 (m, 1H, H-2′), 5.50-5.30
(m, 2H, H-7), 5.00-4.90 (m, 1H, H-4′), 4.35-4.20 (m, 3H, H-12,

Trapping of cycloSal-Pronucleotides
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 20 6597
H-5′), 3.60 (s, 3H, OMe), 2.78 (t, 2H, J ) 7.2 Hz, H-8), 2.39 (t,
2H, J ) 7.5 Hz, H-9), 1.68, 1.62 (2 × d, 3H, J ) 0.9 Hz, Mehet),
1.23, 1.22 (2 × d, 3H, J ) 7.3 Hz, Me) ppm. 31P NMR (162 MHz,
1H decoupled, DMSO-d6): δ ) -9.25, -9.28 ppm.
5-L-Ala-propionyl-cycloSal-d4TMP (5g). 5-L-AlaOtBu-propio-
nyl-cycloSal-d4TMP (5e, 28 mg, 47 µmol) was dissolved in 10
mL of CH2Cl2, and an amount of 2 mL of TFA was added. After
2.5 h at room temperature, the solvent was removed in vacuo. The
residue was purified by preparative TLC (Chromatotron; CH2Cl2
[+0.5% HOAc]/MeOH, gradient elution 1-10%) and lyophilized
from CH3CN/H2O, 1:1 (v/v). Yield: 21 mg (39 µmol, 83%) of a
diastereomeric mixture (ratio 0.9:1.0) as a colorless foam. 1H NMR
(400 MHz, DMSO-d6): δ ) 12.39 (s, 1H, COOH), 11.35, 11.34 (2
× s, 1H, NHhet), 8.14 (d, 1H, J ) 7.2 Hz, NH-11), 7.24-7.15 (m,
2H, Hhet-6, H-4), 7.12-7.07 (m, 1H, H-6), 7.02, 7.00 (2 × d, 1H,
J ) 8.4 Hz, H-3), 6.81-6.77 (m, 1H, H-1′), 6.41, 6.35 (2 × ddd,
J ) 1.6 Hz, J ) 1.6 Hz, J ) 6.0 Hz, H-3′), 6.03-5.98 (m, 1H,
H-2′), 5.48-5.30 (m, 2H, H-7), 4.99-4.90 (m, 1H, H-4′), 4.36-4.14
(m, 3H, H-12, H-5′), 2.82-2.70 (m, 2H, H-8), 2.38 (t, 2H, J ) 7.6
Hz, H-9), 1.68, 1.62 (2 × d, 3H, J ) 1.0 Hz, Mehet), 1.23, 1.22 (2
× d, 3H, J ) 7.3 Hz, Me) ppm. 31P NMR (162 MHz, 1H decoupled,
DMSO-d6): δ ) -9.27, -9.30 ppm.
5-L-ProOMe-propionyl-cycloSal-d4TMP (5h). Quantities were
as follows: 5-propionyl-cycloSal-d4TMP acid (4, 50.0 mg, 108
µmol), DIPEA (38 µL, 0.22 mmol), L-proline methyl ester·HCl
(37 mg, 0.22 mmol), HOBt (18 mg, 0.12 mmol), DCC (30 mg,
0.14 mmol), dissolved in 1 mL dry DMF. First purification was
done by flash chromatography (CH2Cl2/MeOH, 9:1 v/v), then by
preparative TLC (Chromatotron; CH2Cl2/MeOH, gradient elution
0-5%). Yield: 51 mg (89 µmol, 82%) of a diastereomeric and
rotameric mixture as a colorless foam. 1H NMR (400 MHz, DMSO-
d6): δ ) 11.34 (s, 1H, NHhet), 7.35-7.25 (m, 1H, H-4), 7.22, 7.19
(2 × q, 1H, J ) 1.2 Hz, H-6het), 7.16-7.07 (m, 2H, H-6, H-3),
6.82-6.78 (m, 1H, H-1′), 6.42, 6.35 (2 × ddd, J ) 1.6 Hz, J )
1.6 Hz, J ) 6.0 Hz, H-3′), 6.06-5.98 (m, 1H, H-2′), 5.54-5.30
(m, 2H, H-7), 5.00-4.90 (m, 1H, H-4′), 4.60-4.20 (m, 3H, H-12,
H-5′), 3.67, 3.54, 3.61, 3.60 (4 × s, 3H, OMe), 3.55-3.37 (m, 2H,
H-11a), 2.81, 2.80 (2 × t, 2H, J ) 7.5 Hz, H-8), 2.55 (t, 2H, J )
7.6 Hz, H-9), 2.34-2.08 (m, 1H, H-11c), 2.06-1.70 (m, 3H, H-11b,
H-11c), 1.65, 1.62, 1.60, 1.59 (4 × d, 3H, J ) 0.8 Hz, Mehet) ppm.
31P NMR (162 MHz, 1H decoupled, DMSO-d6): δ ) -8.47, -8.75
ppm.
5-L-AlaOMe-propionyl-cycloSal-d4TMP (5d). Quantities were
as follows: 5-propionyl-cycloSal-d4TMP acid (4, 50.0 mg, 108
µmol), DIPEA (38 µL, 0.22 mmol), L-alanine methyl ester ·HCl
(30 mg, 0.22 mmol), HOBt (18 mg, 0.12 mmol), DCC (30 mg,
0.14 mmol), dissolved in 1 mL dry DMF. Purification was done
by flash chromatography (CH2Cl2/MeOH, 9:1 v/v). Yield: 56.0 mg
(102 µmol, 94%) of a diastereomeric mixture (ratio 0.9:1.0) as a
colorless foam. 1H NMR (400 MHz, DMSO-d6): δ ) 11.34, 11.33
(2 × s, 1H, NHhet), 8.31, 8.30 (2 × d, 1H, J ) 6.9 Hz, NH-11),
7.28-7.22 (m, 1H, H-4), 7.20, 7.19 (2 × q, 1H, J ) 1.2 Hz, H-6het),
7.15-7.08 (m, 2H, H-6, H-3), 6.82-6.76 (m, 1H, H-1′), 6.42, 6.35
(2 × ddd, J ) 1.8 Hz, J ) 1.8 Hz, J ) 6.0 Hz, H-3′), 6.05-5.99
(m, 1H, H-2′), 5.52-5.30 (m, 2H, H-7), 5.00-4.90 (m, 1H, H-4′),
4.40-4.20 (m, 3H, H-12, H-5′), 3.61, 3.60 (2 × s, 3H, OMe),
2.87-2.75 (m, 2H, H-8), 2.47-2.35 (m, 2H, H-9), 1.66, 1.59 (2 ×
d, 3H, J ) 1.1 Hz, Mehet), 1.25, 1.23 (2 × d, 3H, J ) 7.3 Hz, Me)
ppm. 31P NMR (162 MHz, 1H decoupled, DMSO-d6): δ ) -8.71,
-8.74 ppm.
5-L-AlaOtBu-propionyl-cycloSal-d4TMP (5e). Quantities were
as follows: 5-propionyl-cycloSal-d4TMP acid (4, 60.0 mg, 129
µmol), DIPEA (45 µL, 0.26 mmol), L-alanine tert-butyl ester ·HCl
(47 mg, 0.26 mmol), HOBt (22 mg, 0.14 mmol), DCC (36 mg,
0.17 mmol), dissolved in 1 mL dry DMF. Purification was done
by preparative TLC (Chromatotron; CH2Cl2/MeOH, gradient elution
0-2%). Yield: 53 mg (90 µmol, 70%) of a diastereomeric mixture
1
(ratio 0.9:1.0) as a colorless foam. H NMR (400 MHz, DMSO-
d6): δ ) 11.35, 11.34 (s, 1H, NHhet), 8.14 (d, 1H, J ) 7.0 Hz,
NH-11), 7.24-7.15 (m, 2H, Hhet-6, H-4), 7.12-7.07 (m, 1H, H-6),
7.02, 7.00 (2 × d, 1H, J ) 8.3 Hz, H-3), 6.82-6.75 (m, 1H, H-1′),
6.41, 6.34 (2 × ddd, J ) 1.6 Hz, J ) 1.6 Hz, J ) 6.0 Hz, H-3′),
6.03-5.98 (m, 1H, H-2′), 5.48-5.30 (m, 2H, H-7), 4.99-4.90 (m,
1H, H-4′), 4.36-4.20 (m, 2H, H-12, H-5′), 4.09 (m, 1H, H-12),
2.82-2.70 (m, 2H, H-8), 2.38 (t, 2H, J ) 7.5 Hz, H-9), 1.68, 1.62
(2 × s, 3H, Mehet), 1.38 (s, 9H, t-Bu), 1.23, 1.22 (2 × d, 3H, J )
5-L-ProOBn-propionyl-cycloSal-d4TMP (5i). Quantities were
as follows: 5-propionyl-cycloSal-d4TMP acid (4, 40 mg, 86 µmol),
DIPEA (30 µL, 0.17 mmol), L-proline benzyl ester ·HCl (42 mg,
0.17 mmol), HOBt (15 mg, 0.11 mmol), DCC (24 mg, 0.11 mmol),
dissolved in 2 mL dry DMF. Purification was done by preparative
TLC (Chromatotron; EtOAc/MeOH, gradient elution 0-2%). Yield:
46 mg (71 µmol, 83%) of a diastereomeric and rotameric mixture
1
7.1 Hz, Me) ppm. 31P NMR (162 MHz, H decoupled, DMSO-
d6): δ ) -9.15, -9.52 ppm.
5-L-AlaOBn-propionyl-cycloSal-d4TMP (5f fast and 5f slow).
Quantities were as follows: 5-propionyl-cycloSal-d4TMP acid (4,
30.0 mg, 65 µmol), DIPEA (22 µL, 0.13 mmol), L-alanine benzyl
ester·p-TsOH (46 mg, 0.13 mmol), HOBt (11 mg, 72 µmol), DCC
(18 mg, 85 µmol), dissolved in 1 mL dry DMF. Purification was
done by preparative TLC (Chromatotron; EtOAc/MeOH, gradient
elution 0-5%). Yield: 36 mg (58 µmol, 88%) of a diastereomeric
mixture (ratio 0.9:1.0) as a colorless foam. The diastereomers 5f fast
and 5f slow were enriched by means of preparative RP-HPLC
(CH3CN/H2O, 1:4) up to a ratio of 9:1 for each compound (as
1
as a colorless foam. H NMR (400 MHz, DMSO-d6): δ ) 11.35,
11.34 (2 × s, 1H, NHhet), 7.40-7.30 (m, 5H, Haryl), 7.27-7.19 (m,
1H, H-4), 7.19, 7.16 (2 × q, 1H, J ) 1.2 Hz, H-6het), 7.15-7.10
(m, 1H, H-6), 7.05-6.95 (m, 1H, H-3), 6.82-6.76 (m, 1H, H-1′),
6.41, 6.35 (2 × ddd, J ) 1.8 Hz, J ) 1.8 Hz, J ) 6.0 Hz, H-3′),
6.05-5.98 (m, 1H, H-2′), 5.50-5.27 (m, 2H, H-7), 5.11 (s, 2H,
H-14), 4.99-4.91 (m, 1H, H-4′), 4.40-4.20 (m, 3H, H-12, H-5′),
3.55-3.37 (m, 2H, H-11a), 2.78 (t, 2H, J ) 7.5 Hz, H-8), 2.57 (t,
2H, J ) 7.6 Hz, H-9), 2.25-2.10 (m, 1H, H-11c1), 1.94-1.78 (m,
1
judged from 31P NMR). 5f fast: H NMR (400 MHz, MeOD): δ
) 7.40-7.27 (m, 6H, Hhet-6, Haryl), 7.20 (d, 1H, J ) 8.4 Hz, H-4),
7.06 (s, 1H, H-6), 6.98 (d, 1H, J ) 8.4 Hz, H-3), 6.90 (ddd, 1H, J
) 1.6 Hz, J ) 1.6 Hz, J ) 3.5 Hz, H-1′), 6.41 (ddd, 1H, J ) 1.4
Hz, J ) 1.4 Hz, J ) 6.0 Hz, H-3′), 6.01-5.96 (m, 1H, H-2′), 5.37
(dd, 2H, J ) 9.4 Hz, J ) 14.0 Hz, H-7), 5.15 (s, 2H, H-14),
5.02-4.96 (m, 1H, H-4′), 4.45-4.37 (q, 1H, J ) 7.3 Hz, H-12),
4.40-4.30 (m, 2H, H-5′), 2.95-2.80 (m, 2H, H-8), 2.50 (t, 2H, J
) 7.4 Hz, H-9), 1.77 (s, 3H, Mehet), 1.33 (d, 3H, J ) 7.3 Hz, Me)
ppm. 31P NMR (162 MHz, 1H decoupled, MeOD): δ ) -8.23 ppm.
5f slow: 1H NMR (400 MHz, MeOD): δ ) 7.36-7.29 (m, 6H,
3H, H-11b, H-11c2), 1.68, 1.61 (2 × d, 3H, J ) 1.1 Hz, Mehet
)
ppm. 31P NMR (162 MHz, 1H decoupled, DMSO-d6): δ ) -9.23,
-9.29 ppm.
Hydrolysis Studies of cycloSal Phosphate Triesters. Hydrolysis
studies of cycloSal nucleotides (phosphate buffer, pH 7.3 and pH
7.6) by HPLC analysis have been described before.20 Studies in
cell extracts and human serum (obtained from pooled blood samples
from the university hospital in Hamburg) were performed as
reported in ref 15 with different incubation times but without using
acetic acid to stop the reaction. Method 1 parameters were as
follows: 0-25 min, H2O/CH3CN gradient (5-100%); 25-35 min,
H2O/CH3CN (5%); flow rate 0.5 mL/min. Method 2 parameters
were as method 1 parameters except TBAH ion buffer (0.55 mM)
was used instead of water.
Hhet-6, Haryl), 7.18 (d, 1H, J ) 8.4 Hz, H-4), 7.05 (d, 1H, J ) 1.7
Hz, H-6), 6.94 (d, 1H, J ) 8.4 Hz, H-3), 6.87 (ddd, 1H, J ) 1.7
Hz, J ) 1.7 Hz, J ) 3.4 Hz, H-1′), 6.35 (ddd, 1H, J ) 1.7 Hz, J
) 1.7 Hz, J ) 6.0 Hz, H-3′), 6.00-5.96 (m, 1H, H-2′), 5.40-5.34
(m, 2H, H-7), 5.15 (s, 2H, H-14), 5.02-4.96 (m, 1H, H-4′),
4.45-4.30 (m, 3H, H-12, H-5′), 2.95-2.80 (m, 2H, H-8), 2.50 (t,
2H, J ) 7.5 Hz, H-9), 1.77 (d, 3H, J ) 1.1 Hz, Mehet), 1.34 (d,
31P NMR Hydrolysis Studies of cycloSal Phosphate Tri-
mesters. 31P NMR hydrolysis studies of cycloSal nucleotides 5
and 6 were carried out as described before.20
1
3H, J ) 7.3 Hz, Me) ppm. 31P NMR (162 MHz, H decoupled,
MeOD): δ ) -8.55 ppm.

6598 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 20
Jessen et al.
ery of d4TMP. Mol. Pharmacol. 2000, 58, 928–935. (b) Balzarini, J.;
Naesens, L.; Aquaro, S.; Knispel, T.; Perno, C.-F.; De Clercq, E.;
Meier, C. Intracellular Metabolism of cycloSaligenyl-3′-azido-2′,3′-
dideoxythymidine Monophosphate, a Prodrug of 3′-Azido-2′,3′-
dideoxythymidine (Zidovudine). Mol. Pharmacol. 1999, 56, 1354–
1361.
Antiretroviral Evaluation. Human immunodeficiency virus type
1 (HIV-1) was originally obtained from a persistently HIV-infected
H9 cell line, as described previously, and was kindly provided by
Dr. R. C. Gallo (then at the National Institutes of Health, Bethesda,
MD). Virus stocks were prepared from the supernatants of HIV-
1-infected MT-4 cells. HIV-2 (strain ROD) was kindly provided
by Dr. L. Montagnier (then at the Pasteur Institute, Paris, France),
and virus stocks were prepared from the supernatants of HIV-2-
infected MT-4 cells. CEM cells were obtained from the American
Tissue Culture Collection (Rockville, MD). CEM cells were infected
with HIV as previously described. Briefly, 4 × 105 CEM cells/mL
were infected with HIV-1 or HIV-2 at ∼100 CCID50 (50% cell
culture infective dose) per mL of cell suspension. The thymidine
kinase-deficient CEM cell cultures were also infected with HIV-2.
Then 100 µL of the infected cell suspensions was transferred into
96-well microtiter plate wells and mixed with 100 µL of the
appropriate dilutions of the test compounds. After 4-5 days, giant
cell formation was recorded microscopically in the HIV-infected
cell cultures.
(7) Meier, C.; Lorey, M.; De Clercq, E.; Balzarini, J. cycloSal-2′,3′-
dideoxy-2′,3′-didehydrothymidine Monophosphate (cycloSal-d4TMP):
Synthesis and Antiviral Evaluation of a New d4TMP Delivery System.
J. Med. Chem. 1998, 41, 1417–1427.
(8) Meier, C.; Habel, L.; Haller-Meier, F.; Lomp, A.; Herderich, M.;
Klo¨cking, R.; Meerbach, A.; Wutzler, P. Chemistry and Anti-Herpes
Simplex Virus Type-1 Evaluation of cycloSal-Nucleotides of Acyclic
Nucleoside Analogs. AntiViral Chem. Chemother. 1998, 9, 389–402.
(9) Meier, C.; Knispel, T.; De Clercq, E.; Balzarini, J. cycloSal-
Pronucleotides of 2′,3′-Dideoxyadenosine and 2′,3′-Dideoxy-2′,3′-
didehydroadenosine: Synthesis and Antiviral Evaluation of a Highly
Efficient Nucleotide Delivery System. J. Med. Chem. 1999, 42, 1604–
1614.
(10) Meier, C.; Knispel, T.; Marquez, V. E.; De Clercq, E.; Balzarini, J.
cycloSal-Pronucleotides of 2′-Fluoro-ara- and 2′-Fluoro-ribo-2′,3′-
dideoxyadenosine as a Strategy To Bypass a Metabolic Blockade.
J. Med. Chem. 1999, 42, 1615–1624.
(11) Meier, C.; Lomp, A.; Meerbach, A.; Wutzler, P. cycloSal-BVDUMP
Pronucleotides: How To Convert an Antiviral-Inactive Nucleoside
Analogue into a Bioactive Compound against EBV. J. Med. Chem.
2002, 45, 5157–5172.
(12) Balzarini, J.; Haller-Meier, F.; De Clercq, E.; Meier, C. Antiviral
Activity of cycloSaligenyl Prodrugs of Acyclovir, Carbovir and
Abacavir. AntiViral Chem. Chemother. 2001, 12, 301–306.
(13) Sauerbrei, A.; Meier, C.; Meerbach, A.; Schiel, M.; Helbig, P.; Wutzler,
P. In Vitro Activity of cycloSal-Nucleoside Monophosphates and
Polyhydroxycarboxylates against Orthopoxviruses. AntiViral Res. 2005,
67, 147–154.
(14) Meier, C.; Ruppel, M. F. H.; Vukadinoviæ, D.; Balzarini, J. Second
Generation of cycloSal-Pronucleotides with Esterase-Cleavable Sites:
The “Lock-In”-Concept. Nucleosides, Nucleotides Nucleic Acids 2004,
23, 89–115.
(15) Meier, C.; Ducho, C.; Jessen, H.; Vukadinoviæ-Tenter, D.; Balzarini,
J. Second-Generation cycloSal-d4TMP Pronucleotides Bearing Es-
terase-Cleavable Sites. The “Trapping” Concept. Eur. J. Org. Chem.
2006, 197–206.
Acknowledgment. We are grateful to the Deutsche Fors-
chungsgemeinschaft (DFG), the University of Hamburg, the
“Geconcerteerde Onderzoeksacties (GOA No. 019/05), and the
Fonds voor Wetenschappelijk Onderzoek (FWO)sVlaanderen
for financial support. The authors also thank Leen Ingels and
Ria Van Berwaer for excellent technical assistance.
Supporting Information Available: 13C NMR and IR spec-
troscopic data, Rf values, and analytical HPLC data of new cycloSal
compounds. This material is available free of charge via the Internet
at http://pubs.acs.org.
References
(1) De Clercq, E. Strategies in the Design of Antiviral Drugs. Nat. ReV.
Drug DiscoVery 2002, 1, 13–25.
(2) Balzarini, J. Metabolism and Mechanism of Antiretroviral Action of
Purine and Pyrimidine Derivatives. Pharm. World Sci. 1994, 16, 113–
126.
(3) Herdewijn, P.; Balzarini, J.; De Clercq, E. 2′,3′-Dideoxynucleoside
Analogues as Anti-HIV Agents. In AdVances in AntiViral Drug Design;
De Clercq, E., Ed.; JAI Press Inc.: Greenwich, London, 1993; Vol. 1,
pp 233-318.
(16) Jessen, H.; Fendrich, W.; Meier, C. Synthesis and Properties of
Fluorescent cycloSal-Nucleotides Based on the Pyrimidine Nucleoside
m5K. Eur. J. Org. Chem. 2006, 92, 4–931.
(17) Nagata, W.; Okada, K.; Aoki, T. Ortho-Specific R-Hydroxyalkylation
of Phenols with Aldehydes. An Efficient Synthesis of Saligenol
Derivatives. Synthesis 1979, 365–368.
(4) (a) Meier, C. Pro-Nucleotides. Recent Advances in the Design of
Efficient Tools for the Delivery of Biologically Active Nucleoside
Monophosphates. Synlett 1998, 233–242. (b) Wagner, C. R.; Iyer,
V. V.; McIntee, E. J. Pronucleotides: Towards the in Vivo Delivery
of Antiviral and Anticancer Nucleotides. Med. Res. ReV. 2000, 20,
417–451.
(18) (a) McGuigan, C.; Thiery, J. C.; Daverioa, F.; Jiang, W. G.; Davies,
G.; Mason, M. Anti-Cancer ProTides: Tuning the Activity of BVDU
Phosphoramidates Related to Thymectacin. Bioorg. Med. Chem. 2005,
13 (9), 3219–3227. (b) Cahard, D.; McGuigan, C.; Balzarini, J.
Aryloxy Phosphoramidate Triesters as Pro-Tides. Mini-ReV. Med.
Chem. 2004, 4 (4), 371–381.
(5) (a) Meier, C. cycloSal-Pronucleotides. Design of Chemical Trojan
Horses. Mini-ReV. Med. Chem. 2002, 2, 219–234. (b) Meier, C.
cycloSal Phosphates as Chemical Trojan Horses for Intracellular
Nucleotide and Glycosylmonophosphate Delivery. Chemistry Meets
Biology. Eur. J. Org. Chem. 2006, 108, 1–1102. (c) Meier, C. cycloSal-
Pronucleotides. Design of the Concept, Chemistry, and Antiviral
Activity. In AdVances in AntiViral Drug Design; De Clercq, E., Ed.;
Elsevir: Amsterdam, 2004; Vol. 4, pp 147-213.
(19) Meier, C.; Ducho, C.; Go¨rbig, U.; Esnouf, R.; Balzarini, J. Interaction
of cycloSal-Pronucleotides with Cholinesterases from Different Ori-
gins. A Structure-Activity Relationship. J. Med. Chem. 2004, 47,
2839–2852.
(20) Ducho, C.; Balzarini, J.; Naesens, L.; De Clercq, E.; Meier, C. Aryl-
Substituted and Benzo-Annulated cycloSal-Derivatives of 2′,3′-Dideoxy-
2′,3′-didehydrothymidine Monophosphate. Correlation of Structure,
Hydrolysis Properties and Anti-HIV Activity. AntiViral Chem. Chemoth-
er. 2002, 13, 129–141.
(6) (a) Balzarini, J.; Aquaro, S.; Knispel, T.; Rampazzo, C.; Bianchi, V.;
Perno, C.-F.; De Clercq, E.; Meier, C. Cyclosaligenyl-2′,3′-didehydro-
2′,3′-dideoxythymidine Monophosphate: Efficient Intracellular Deliv-
JM800815B