Lipase-mediated stereoselective hydrolysis of stampidine and other phosphoramidate derivatives of stavudine

Article abstract of DOI:10.1016/j.bmc.2004.03.068

Enzymatic hydrolysis of stampidine and other aryl phosphate derivatives of stavudine were investigated using the Candida Antarctica Type B lipase. Modeling studies and comparison of the hydrolysis rate constants revealed a chiral preference of the lipase

Full text of DOI:10.1016/j.bmc.2004.03.068

Bioorganic & Medicinal Chemistry 12 (2004) 3371–3381
Lipase-mediated stereoselective hydrolysis of stampidine and other
phosphoramidate derivatives of stavudine
T. K. Venkatachalam, P. Samuel, G. Li, S. Qazi, C. Mao, S. Pendergrass and F. M. Uckun*
Department of Chemistry, Pharmaceutical Sciences, Bioinformatics, Structural Biology, and Virology, Parker Hughes Institute,
2699 Patton Road, St. Paul, MN 55113, USA
Received 18 June 2003; accepted 10 March 2004
Available online 6 May 2004
Abstract—Enzymatic hydrolysis of stampidine and other aryl phosphate derivatives of stavudine were investigated using the
Candida Antarctica Type B lipase. Modeling studies and comparison of the hydrolysis rate constants revealed a chiral preference of
the lipase active site for the putative S-stereoisomer. The in vitro anti-HIV activity of these compounds correlated with their
susceptibility to lipase- (but not esterase-) mediated hydrolysis. We propose that stampidine undergoes rapid enzymatic hydrolysis in
the presence of lipase according to the following biochemical pathway: During the first step, hydrolysis of the ester group results in
the formation of carboxylic acid. Subsequent step involves an intramolecular cyclization at the phosphorous center with simulta-
neous elimination of the phenoxy group to form a cyclic intermediate. In the presence of water, this intermediate is converted into
the active metabolite Ala-d4T-MP. We postulate that the lipase hydrolyzes the methyl ester group of the L-alanine side chain to
form the cyclic intermediate in a stereoselective fashion. This hypothesis was supported by experimental data showing that chlo-
roethyl substituted derivatives of stampidine, which possess a chloroethyl linker unit instead of a methyl ester side chain, were
resistant to lipase-mediated hydrolysis, which excludes the possibility of a direct hydrolysis of stampidine at the phosphorous center.
Thus, our model implies that the lipase-mediated formation of the cyclic intermediate is a key step in metabolism of stampidine and
relies on the initial configuration of the stereoisomers.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
resistant clinical HIV-1 isolate in severe combined
immunodeficient (SCID) mice reconstituted with
peripheral blood (PBL) mononuclear cells from sero-
negative human donors.4 In the feline immunodeficiency
virus (FIV)-infected domestic cat model for AIDS,
orally administered stampidine showed a dose-dependent
anti-retroviral effect in chronically FIV-infected cats.5
Stampidine therapy was not associated with any clinical
or laboratory evidence of toxicity at dose levels as high
as 500 mg kgꢀ1 or at cumulative dose levels as high as
8.4 g kgꢀ1. Stampidine exhibited favorable pharmaco-
kinetic behavior in mice, rats, dogs, and cats following
oral administration.6;7 The documented in vitro potency
of stampidine against primary clinical HIV-1 isolates
with genotypic and/or phenotypic NRTI- or NNRTI-
resistance as well as nonB envelope subtypes together
with its in vivo anti-retroviral activity in HIV-infected
Hu-PBL SCID mice and FIV-infected cats warrants its
further development as a new anti-HIV drug.
Stampidine is a novel phosphoramidate derivative of
stavudine, which is being developed as a potential anti-
HIV agent.1–5 Stampidine was 100-times more active
than stavudine and twice as active as zidovudine against
nine clinical HIV-1 isolates of nonB envelope subtypes
(A, C, F, and G) originating from South America, Asia,
and subSaharan Africa.1 Stampidine was effective
against 20 genotypically and phenotypically nucleoside
analog reverse transcriptase inhibitor (NRTI)-resistant
and six nonnucleoside inhibitor (NNRTI)-resistant
HIV-1 isolates at subnanomolar to low nanomolar
concentrations.1 Stampidine was active against HIV-1
isolates with five thymidine analog mutations at subn-
anomolar concentrations. Orally or intraperitoneally
administered stampidine exhibited significant and dose
dependent in vivo anti-HIV activity against an NRTI-
The generation of the active metabolite of stampidine
was originally thought to require the esterase-mediated
hydrolysis of the carbomethoxy group associated with
Keywords: Lipase; Stampidine; Anti-HIV; Stereochemistry; Modeling.
* Corresponding author. Tel.: +1-651-796-5450; fax: +1-651-796-5493;
e-mail: fatih_uckun@ih.org
0968-0896/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.03.068

3372
T. K. Venkatachalam et al. / Bioorg. Med. Chem. 12 (2004) 3371–3381
the alanine side chain of stampidine.8–13 Recent reviews
by Wagner et al.14;15 and Meier16–18 illustrates the pro-
drug methodology for the in vivo delivery of various
nucleotides. In addition, Wagner and co-workers19 have
reported the pharmacokinetics of amino acid phosphor-
amidate monoesters of zidavudine in rats and came to
the conclusion that these compounds showed greater
tissue distribution and improved pharmacokinetic
properties of over zidavudine. We hypothesized that in
various tissue microenvironments the metabolism of
stampidine may occur through the action of hydrolytic
enzymes other than esterases as well. The purpose of the
present study was to evaluate the potential role of lip-
ases in the activation of stampidine. Our experimental
results provide evidence that stampidine as well as other
halogen-substituted phosphoramidate derivatives of
stavudine can be metabolized by lipase-mediated
hydrolysis. Modeling studies and comparison of the
hydrolysis rate constants revealed a chiral preference of
the lipase active site for the putative S-stereoisomer of
these compounds. Our model indicates that the lipase-
mediated formation of the cyclic intermediate is a key
step in metabolism of stampidine and relies on the initial
configuration of the stereoisomers. To our knowledge
this study is the first to provide experimental evidence
for a lipase-mediated metabolism of a nucleoside phos-
phoramidate prodrug. A stereoselective hydrolytic acti-
vation of such compounds by a lipase has never been
reported or even considered. In our earlier publication7;8
the metabolism of these phosphoramidate derivatives
with cellular enzymes such as esterases have been stud-
ied. The results show that apart from esterase’s other
enzymes may be involved in the metabolic pathway of
these phosphoramidate derivatives. To further under-
stand the metabolic pathway, we chose candida B lipase
as a model enzyme in the present study.
mediated hydrolysis but not esterase-mediated hydro-
lysis. These findings prompted us to further elucidate the
lipase-mediated hydrolysis of stampidine and other 4-
halogen substituted phosphoramidate derivatives of
stavudine.
2.2. Stereoisomers of stampidine
Due to the stereochemistry of its phosphorous chiral
center, stampidine exists as a mixture of two possible
diastereomers (Fig. 2). The individual stereoisomers
were separated by preparative HPLC due to their dis-
tinct retention times of 20.0 and 21.2 min, respectively,
lyophilized, and examined by IR, polarimetry, HPLC,
NMR, and melting point analysis. No significant dif-
ferences in optical rotation were observed ()30ꢁ vs
1
)27.5ꢁ) by polarimetric analysis and their individual H
NMR spectra were identical. Repetitive HPLC and 31P
NMR analyses during a 36-h kinetics experiment con-
firmed that the separated diastereomers are stable and
do not undergo interconversion. The physicochemical
constants of the purified diastereomers are described in
the experimental section. The in vitro anti-HIV potency
of the purified diastereomers (IC50: 0.001 lM for both)
was identical to the in vitro anti-HIV potency of the
diastereomeric mixture of stampidine (IC50: 0.001 lM).
Since there appears to be no difference in the anti-iviral
activity of the two distereoisomers, the observed modest
differences in substrate specificity (4–5-fold) implies that
the lipase discrimination is of little consequence to the
biological activity. However, our goal was to establish
that Ala-Stav-MP is the key intermediate responsible for
its biological activity since this derivative lacks chirality
at the phosphorus center and hence both the isomers are
expected to show equipotency. We have also established
in our earlier studies7;8 that indeed AlaSTVMP is much
more active than STVMP using authentic samples. In
addition, we point out that the observed discrimination
by lipase occurs at the phosphorus chiral center after
initial hydrolysis at the ester group present in the mol-
ecule to form the unstable cyclic intermediate.
2. Results and discussion
2.1. Lipase-mediated hydrolysis of stampidine
It has been postulated that phosphoramidate derivatives
of stavudine are activated after an enzymatic hydrolysis
mediated by esterase’s.20–22 No other enzymes have yet
been proposed as alternative mediators of the hydrolysis
of this promising new class of anti-HIV compounds.
Since lipases have been shown to be capable of hydro-
lyzing other chiral compounds with a phosphorous
center,23–28 we sought to determine if the 4-Br substi-
tuted phosphoramidate derivative stampidine could be
hydrolyzed by a lipase as well. The structurally well-
characterized lipase derived from Candida Antarctica
Type B was selected for the experiments. The porcine
liver esterase was included as a control for comparison.
Remarkably, lipase-mediated hydrolysis of stampidine
was much more efficient than its esterase-mediated
hydrolysis (Fig. 1). Furthermore, the unsubstituted
compound 1 with 2-fold lower anti-HIV activity showed
a 3-fold faster hydrolysis in the presence of the esterase,
but 3-fold slower hydrolysis in the presence of the lipase.
Thus, the in vitro anti-HIV activity of these two com-
pounds correlated with their susceptibility to lipase-
2.3. Structural model of lipase-mediated hydrolysis of
stampidine
The crystal structure of C. Antarctica lipase is
resolved.29 Several homologous lipases complexed with
various substrate analogs have also been crystallized
and their active sites have been extensively explored. The
structural basis of the hydrolytic function of the lipase
has been deciphered and linked to the spatial arrange-
ment and concerted action of the active site residues30–44
We obtained the published coordinates from PDB (pdb
access code: 1 tcc, 1 tcb, 1 tca)29 and modeled stampidine
and similar stavudine derivatives into the active site
using our previously published docking procedures with
appropriate modifications.45 Since these phosphoram-
idate derivatives of stavudine have multiple rotatable
bonds, which can yield several conformations, we first
generated up to 255 conformers for each compound and
docked them individually into the lipase active site. The

T. K. Venkatachalam et al. / Bioorg. Med. Chem. 12 (2004) 3371–3381
3373
Figure 1. Lipase and esterase mediated hydrolysis of stampidine at various time intervals.
majority of these conformers were excluded during the
earliest stages of the docking process because of too
many steric collisions with the lipase protein before their
carboxylester oxygen atoms would descend toward the
protein core at the catalytic center. The final model was
determined by sorting and evaluating the energy cost
and LUDI interaction score for all pairs.46–51 The mol-
ecules of both R- and S-stereoisomers of each com-
pound were modeled and compared in order to
determine, which enantiomer was better suited for the
active site of the lipase. The final model for the inter-
actions of the S-stereoisomer of stampidine and lipase is
shown in Figure 3A. The phenoxy ring is buried in a
tight binding pocket and the stavudine group half buried

3374
T. K. Venkatachalam et al. / Bioorg. Med. Chem. 12 (2004) 3371–3381
O
Me
COOMe
Me
COOMe
HN
O
N
HN
P O
HN
P
O
N
HN
O
O
Br
O
O
O
O
O
H
O
H
Br
'S' Configuration
"R" Configuration
Figure 2. Structures of two individual isomers of stampidine.
in another tight pocket, neither of which is fit to the
binding site of the other group. When the S-configura-
tion is changed to the corresponding R-configuration,
the phenoxy ring is switched in position with the
stavudine group. Because of significant geometrical
differences none of the two groups would remotely fit
into the binding site of the other favorably. Because of a
major steric ‘clash’, neither the phenoxy ring nor the
stavudine group could be easily accommodated in the
binding position for the P@O group. Therefore, the
interactions of the R-stereoisomer with the lipase active
site are energetically less favorable than those of the S-
stereoisomer. This model predicts a faster and more
efficient hydrolysis of the S-stereoisomer due to the tight
initial binding between this stereoisomer and the lipase
active site. The phenoxy ring is fit well in a binding
pocket that consists of a protein backbone near Thr40
stacking against with its ring plane and several hydro-
phobic residues on the sideline including Ile285, Ala281,
Ile189, and Leu278 (Fig. 3B). A 2,6-dimethoxy substi-
tution to the phenoxy ring was predicted to significantly
impair the binding and to slow down the hydrolysis.
There is room for a 4-position substitution that would
bring a close contact with the backbone of the 278 res-
idue and thus is favored for halogen substitutions, as
suggested by the experiments that 4-position halogens
are good for hydrolysis. Interestingly, comparing this
lipase active site model with a homologous model of the
esterase active site, we observed a larger spatial opening
near the entrance of the active site in the esterase than
that in the lipase. The larger opening corresponds to
roomier binding sites for both stavudine and the phen-
oxy group. The esterase thus appears to be more for-
giving in accommodating a different structure or both
enantiomers and thus less discriminative.
Figure 3. (A) The model of the S-stereoisomer of the stampidine
molecule in the active site of the C. Antarctic lipase, a solid surface
representation based on the crystal structure of the protein and the
computation of energetically favored conformations for the ligand by a
previously published docking procedure.48–52 (B) The model of the S-
stereoisomer of the stampidine molecule in the active site of the C.
Antarctic lipase, based on the crystal structure of the protein and the
computation of energetically favored conformations for the ligand by a
previously published docking procedure.48–52 The close-up view of the
phenoxy ring interacting with the surrounding protein residues
(labeled). Prepared with InsightII.52
2.4. Stereoselectivity of lipase-mediated hydrolysis of
stampidine
We next set out to determine if stampidine and five other
phosphoramidate derivatives of stavudine are hydro-
lyzed by C. Antarctica lipase in a stereoselective manner
as predicted from the modeling and docking studies
detailed above. The chemical reactions of lipase with
these compounds were monitored using HPLC and the
rate of lipase-mediated hydrolysis at room temperature
was calculated for the two isomers of each compound
using a first order rate equation. The rate constants are
given in Table 1. The selectivity index indicates the rel-
ative rates of hydrolysis for the two isomers. Impor-
tantly, the stereoisomers of the four substituted
compounds, including stampidine, corresponding to
peak #1 in their HPLC chromatograms underwent
hydrolysis much faster than their enantiomers corre-
sponding to peak #2. Figure 4 shows representative

T. K. Venkatachalam et al. / Bioorg. Med. Chem. 12 (2004) 3371–3381
3375
Table 1. Enzyme mediated hydrolysis rate constants of various arylphosphoramidate derivatives and their anti-viral activity
O
HN
X
O
P
NH
O
O
N
O
O
H
MeOOC
Lipase
H
Me
Compd
X
Esterase
2
IC50 (nM)
HTLVIIIB RTMDR
1
2
SI
1
SI
1
2 (Stampidine)
H
4Br
0.9
3.2
4.9
1.7
1.6
0.05
1.3
0.8
0.9
0.9
0.6
0.05
0.7
4.0
5.4
1.9
2.7
1.0
0.8
0.3
1.8
1.5
2.4
0.12
0.8
0.2
1.4
1.2
2.6
0.09
1.0
1.5
1.3
1.3
0.9
1.3
2.0 0.6
1.0 0.0
1.0 0.3
1.0 0.0
4.0 2.0
6.0 3.0
534 308
21 12
62 43
3
4
5
6
4Cl
4F
218 12.6
1014 586
1104 637
4OMe
2,6-OMe
1 and 2 represents rate constants/hour for each of the isomers. SI denotes the selectivity index.
HPLC chromatogram profiles for a fluoro substituted
phosphoramidate derivative treated with lipase for var-
ious time intervals showing the preferential hydrolysis of
the isomer corresponding to peak #1, concomitant with
the emergence of the peaks corresponding to the
metabolites d4T and Ala-d4T-MP. The mechanism
through which such metabolite formed has been con-
firmed by using LC/mass, HPLC profiles, and indepen-
dent authentic compound synthesis for this purpose.7;8
lipase-mediated hydrolytic activation of stampidine and
compound 3 may play an important role for their anti-
HIV activity. In contrast, we observed no such associ-
ation for the rate constants of the esterase-mediated
hydrolysis. Our lead compound stampidine was a par-
ticularly poor substrate for esterase-mediated hydroly-
sis. These results call in question previous assertions that
esterase-mediated hydrolysis of phosphoramidate
derivatives of stavudine plays an essential role in their
activation. In agreement with the predictions of the
modeling studies, esterase-mediated hydrolysis showed
less stereoselectivity (Table 1).
Similar results were obtained from the 31P NMR anal-
ysis of the lipase-treated stampidine diastereomer mix-
ture. The signal corresponding to diastereomer #1 (but
not the signal corresponding to diastereomer #2) was
drastically reduced after lipase treatment consistent with
a stereoselective hydrolysis.
2.5. Putative pathway of lipase-mediated hydrolysis of
stampidine
Based on the results of the modeling studies, the isomers
corresponding to peak #1 undergoing much faster
hydrolysis was tentatively designated the ‘S-stereo-
isomers’. Table 2 indicates the characteristic HPLC and
31P NMR peaks corresponding to the individual ste-
reoisomers. For each compound the two rate constants
for the hydrolysis of the stereoisomers were log trans-
formed and correlated with Hammett sigma values.
Both rate constants showed a positive association with
the Hammett sigma values (S-stereoisomer: R-
square ¼ 0.63, p ¼ 0:006; R-stereoisomer: R-square ¼
0.71, p ¼ 0:002. The rate constants for the S-stereoi-
somers of stampidine (3.2 hꢀ1) and the 4-Cl substituted
compound 3 (4.9 hꢀ1) were higher than those of the
unsubstituted compound 1 (0.9 hꢀ1) or the 4-OMe
substituted compound 4 (1.6 hꢀ1). These two com-
pounds also exhibited more potent anti-HIV activity
against the NRTI-sensitive HIV-1 strain HTLVIIIB as
well as the NRTI-resistant HIV-1 strain RT MDR than
compounds 1 and 4. As predicted from the modeling
studies, compound 5 with 2,6Di-OMe substitution was a
poor substrate for the lipase and exhibited 6-fold higher
IC50 values than stampidine or compound 3.
We hypothesize that stampidine undergoes rapid enzy-
matic hydrolysis in the presence of lipase according to
the following putative pathway (Fig. 5): In the first step,
compound B is generated by lipase-mediated hydrolysis
of the methylester side chain of stampidine. The sub-
sequent step involves an intramolecular cyclization step
involving the phosphorus center with simultaneous
elimination of the phenoxy group to form the cyclic
intermediate (D). In the presence of water, this inter-
mediate is converted into the active metabolite (E). We
believe that the lipase hydrolyzes the methyl ester group
of the L-alanine side chain to form the cyclic interme-
diate D in a stereoselective fashion. This notion is sup-
ported by experimental data showing that chloroethyl
substituted derivatives (Scheme 1) of stampidine, which
possess a chloroethyl linker unit instead of a methyl
ester side chain, were resistant to lipase-mediated
hydrolysis, which excludes the possibility of a direct
hydrolysis of stampidine at the phosphorous center
(Table 3). Although alanine methyl amide analog would
have been an alternative, however it would also hydro-
lyze the compound similar to the stampidine hydrolysis.
The object of the present study was to totally eliminate
any hydrolyzable group in the structure of the molecule
in order to establish whether the enzyme attacks on the
phosphorus center only. To further substantiate our
The observed association of the higher rate constants
with better IC50 values prompts the hypothesis that

3376
T. K. Venkatachalam et al. / Bioorg. Med. Chem. 12 (2004) 3371–3381
Figure 4. HPLC profiles of 4-fluoro phenylphosphoramidate derivative with lipase at various time intervals at room temperature.
claim, we recently prepared the ‘
amidate derivatives and compared their hydrolysis rate
with enzymes, and found that the isomer underwent
10-fold faster hydrolysis as compared to the ‘ ’ isomer
Br ( ): 3.94; Br ( ): 0.40; F ( ): 2.60; F( ): 0.27 implying
that the hydrolysis occurs at the ester side chain and not
at the phosphorus center of the molecule. We indicate
that there should be no apparent difference in the elec-
D
’ isomer of phosphor-
derivative and hence the slower rate observed under
identical experimental conditions proves beyond doubt
that there is no attack on the phosphorus center
(Venkatachalam et al., manuscript under preparation).
Thus, our model implies that the lipase-mediated for-
mation of the cyclic intermediate is a key step in
metabolism of stampidine and relies on the initial con-
figuration of the stereoisomers. The S-stereoisomer
enters preferentially into the active site of the lipase and
undergoes rapid hydrolysis to form the cyclic interme-
L
D
L
D
L
D
tronic charge or steric hindrance between ‘
alanine methylester substituted phosphoramdiate
L’ and ‘D’

T. K. Venkatachalam et al. / Bioorg. Med. Chem. 12 (2004) 3371–3381
3377
Table 2. Proposed configuration, HPLC retention time, and 31P NMR chemical shifts observed for each of the isomers of phosphoramidate
derivatives of stavudine using enzymatic hydrolysis profiles and molecular modeling results
O
HN
X
O
P
NH
O
O
N
O
O
H
MeOOC
H
Me
Compd
X
Peak #1a (Rt) (ppm)
Peak #2a (Rt) (ppm)
1
2 (Stampidine)
H
4-Br
R(4.47) (2.66)
S(7.86) (3.11)
S(7.04) (3.57)
S(5.01) (3.80)
S(4.48) (3.82)
R(44.3) (4.28)
S(4.65) (3.20)
R(8.37) (2.56)
R(7.45) (2.82)
R(5.22) (3.22)
R(4.75) (3.11)
S(48.3) (4.97)
3
4
5
6
4-Cl
4-F
4-OMe
2,6-OMe
a Peaks are numbered according to the elution time in HPLC. Rt refers their respective retention times and ppm refers to 31P NMR chemical shifts
relative to phosphoric acid as standard at 0 ppm.
O
O
HN
HN
O
O P
NH
O
O
N
Enzymes
CH3OH
O
O P
NH
O
O
N
X
O
X
O
H
MeOOC
H
Me
HO
H
Me
A
O
O
B
O
N
HN
HN
O
O P
O
N
O
O
O
-O P
O
O
NH
NH
O
O
X
O-
H
Me
HO
H
Me
D
O
C
E
X-C6H4OH
Enzymes involved : Lipase, Esterase,
Figure 5. New Pathway for the lipase-mediated hydrolysis of phosphoramidate derivatives of stavudine based on various enzymes involved.
O
HN
Cl
Cl
Cl
Cl
O
N
O
P
O
O
O P O
N
a
b
Cl
N
R
O P N
R
O
Cl
Cl
Cl
Cl
R=H, CH3, OCH3, F, Br
Scheme 1. Synthetic scheme for chloroethyl substituted phosphoramidate derivatives of stavudine. (a) Phenol, triethylamine, 1,2,4-triazole; (b)
stavudine.
diate. At subsequent steps of the metabolism, the chi-
rality of the cyclic intermediate at its phosphorous
center becomes irrelevant due to the fact that the active
metabolite of stampidine, which is the final product of
this pathway, is not chiral.
pounds with phosphorous chiral centers and proposed a
general reaction scheme for their selectivity. Using
molecular modeling and X-ray diffraction studies, they
showed that the ‘R’ isomer of methyl hexyl phosphonate
fits into the enzyme pocket better than the ‘S’ isomer
and was more sensitive to the enzymatic action of the
lipase. In their studies, the target compound had a hal-
ogen atom directly attached to the phosphorous center;
In a previous study, Cygler et al.25 demonstrated that
lipases can distinguish between enantiomers of com-

3378
T. K. Venkatachalam et al. / Bioorg. Med. Chem. 12 (2004) 3371–3381
Table 3. Area under curve (HPLC) observed for the chloroethyl substituted aryl phosphoramidate derivatives of stavudine with lipase and esterase at
various intervals of time
O
HN
O
O
N
X
O
P
O
O
N
H
Cl
Cl
X
0 min
#2
40 min
120 min
#2
Comments
#1
#1
#2
#1
OMe
H
Br
996
387
615
640
968
388
571
599
985
382
627
635
949
378
589
599
986
385
ND
635
956
376
ND
591
No hydrolysis
No hydrolysis
No hydrolysis
No hydrolysis
Me
Lipase was used for the above studies under identical conditions used for alanine substituted phosphoramidate derivatives of stavudine. #1 and #2
indicates each of the isomers peak observed in HPLC chromatogram.
therefore, one could intuitively imagine hydrolysis
occurring at this site for chiral recognition by the
enzyme. However in our phosphoramidate derivatives,
there were two chiral centers, one at the carbon center
and the other at the phosphorus center. The chiral rec-
ognition of the lipase occurs at the phosphorus center
due to the cyclic intermediate formation and there is no
direct reaction of the enzyme towards the phosphorus
center.
Our results indicate that the lipase-mediated formation
of the cyclic intermediate is a key step in metabolism of
stampidine and relies on the initial configuration of the
stereoisomers. We postulate that the lipase hydrolyzes
the methyl ester group of the L-alanine side chain to
form the cyclic intermediate in a stereoselective fashion.
4. Material and methods
All chemicals were purchased from Aldrich (Milwaukee,
WI) or Sigma and were used without further purifica-
tion. Unless otherwise noted, each reaction vessel was
secured with a rubber septum, and the reaction was
performed under nitrogen atmosphere. 1H and 13C
NMR was obtained on a Varian Mercury 300 instru-
ment at ambient temperature in DMSO-d6 Chemical
shifts are reported as ꢁd values in parts per million
downfield from tetramethylsilane (d ¼ 0:0 ppm) as an
internal standard or from the residual dimethylsulfoxide
2.6. Possible lipase activation of the prodrug
Based on the results obtained, we hypothesize that along
with esterase, lipase may be also involved in the enzy-
matic conversion of these phosphoramdiate derivatives
of stavudine to the metabolite. The enzymatic conver-
sion of Ala-d4TMP is the first step in the process and in
the subsequent step it is converted into d4T. However,
the interesting aspect is that the above conversion is
normally very slows step and is typically rate limiting. In
our experiments, it is evident that this conversion is
rapid (see Fig. 5) and there was no evidence of d4TMP
formation. This is consistent with enzymatic cleavage of
the 50–O–P bond. If this were the case, it would explain
the apparent therapeutic advantage of lipase activation.
The above discussion is further substantiated by our
recent findings that proteases were also found to
hydrolyze these phosphoramdiate derivatives (unpub-
lished results).
1
signal (d ¼ 2:49 ppm for H NMR or d ¼ 39:7 ppm for
13
C NMR). Splitting patterns are designated as follows:
s, singlet; d, doublet; t, triplet; m, multiplet; br, broad
peak. FT-IR spectra were recorded on a Nicolet Protege
460 spectrometer. Mass spectra were performed on a
Hewlett Packard MALDI-TOF spectrometer (Model
G2025A LD-TOF). Melting points were determined
using a Melt John’s apparatus and are uncorrected.
HPLCs were performed using a Hewlett Packard 1100
series instrument consisting of an automatic sampler, an
electronic degasser, a thermostatic control unit, and a
diode array detector in conjunction with a Chemstation
software assembly. The column used was an analytical
RP–18 Lichrospher column 5 mm (4.6 · 150 lm) and
eluent was acetonitrile/water mixture. The flow rate was
maintained at 1.0 mL minꢀ1 and the detection wave-
length was set at 275 nm. The column was maintained at
room temperature throughout the analysis. Column
chromatography was performed using silica gel obtained
from the Baker Company. The solvents used for elution
varied depending on the compound and included either
one or a combination of the following: ethyl acetate,
3. Conclusion
Stampidine and other halogen-substituted phospho-
ramidate derivatives of stavudine can be metabolized by
lipase-mediated hydrolysis. Modeling studies and com-
parison of the hydrolysis rate constants revealed a chiral
preference of the lipase active site for the putative S-
stereoisomer of these compounds. The in vitro anti-HIV
activity of the compounds correlated with their suscep-
tibility to lipase- (but not esterase-) mediated hydrolysis.

T. K. Venkatachalam et al. / Bioorg. Med. Chem. 12 (2004) 3371–3381
3379
methanol, chloroform, hexane, methylene chloride,
THF, and ether. Analytical thin-layer chromatography
(TLC) was performed on Merck pre-coated glass plates
(silica gel 60, F254, 250-lm thick), and visualized under
254-nm UV light. Column chromatography was per-
formed using EM silica gel 60, 230–400 mesh. Synthesis
of phosphoramidate analogs of stavudine was achieved
starting from substituted phenols and phosphorous
oxychloride. The resulting phosphodichloridate was re-
500 lL of lipase solution was added and the contents
were shaken to form a homogeneous solution. From this
reaction mixture 50 lL was used for HPLC analysis. The
column used was a Lichrospher (RP) analytical column
of 4 · 250 mm. The eluent used for HPLC was water/
TFA/TEA (0.1%) and CH3CN in the ratio of 65:35. The
column was maintained at room temperature. The flow
rate was maintained at 1 mL minꢀ1, the detection
wavelength was adjusted to 265 nm and the reference
wavelength was kept at 400 nm. Aliquots of the sample
were collected from the reaction vial at various time
intervals and analyzed. The amounts of products
observed during the reaction of these phosphoramidates
were estimated from the area obtained from the HPLC
profiles. In addition, authentic samples of the products
in most of the cases were run to identify the peaks ob-
served during the reaction. Further confirmation of the
product structures was obtained using an LC/mass
instrument. The rate of reaction was computed by using
first order rate constants and an average of eight to nine
time points were used for this estimate. The rate con-
stants reported refers to rate/hour since some of the
reactions were too slow to obtain meaningful results.
acted with L-alanine methylester hydrochloride followed
by stavudine to furnish the required compounds. The
detailed synthetic scheme and methods are already re-
ported in our earlier publications.1 Synthesis of chloro-
ethyl
accomplished as described below:
substituted
stavudine
derivatives
were
4.1. General procedure for synthesis of chloroethyl
substituted phosphoramidate derivatives of stavudine
The general procedure followed for the synthesis of the
above derivatives of stavudine is shown in Scheme 1. A
RB flask was charged with 20 mL of anhydrous aceto-
nitrile and triethylamine (1.2 equiv) followed by triazole
(2.50 equiv). To this mixture was added a solution of
appropriately substituted phenol (1.0 equiv) and bis(2-
chloroethyl)amine phosphodichloridate (1.00 equiv) in
anhydrous acetonitrile. The contents were stirred at 0 ꢁC
by external cooling using ice bath. After addition the
contents were brought to room temperature and stirred
for 5 h. Upon completion of the reaction, d4T
(1.0 equiv) was added to the mixture and the contents of
the flask were brought to 50 ꢁC and allowed to stir for
one week. After this period, the solvent was evaporated
under vacuum and the residue was subjected to column
chromatography using chloroform and methanol (10:1)
and the appropriate fractions containing the product
were pooled together. Evaporation of the solvent and
further purification by preparative thin layer chroma-
tography furnished analytically pure product in 10%
yield as solids.
31P NMR studies with enzyme was conducted using a
5 mm NMR tube. Initially a capillary tube was charged
with 1% phosphoric acid in water and fused carefully.
The sample was dissolved in either 100 lL of deuterated
methanol, followed by D2O (0.4 mL) and was trans-
ferred into the NMR tube using a Pasteur pipette. The
capillary tube containing the standard was introduced
into the NMR tube. The NMR spectrum of the sub-
strate was taken under these conditions and the tube was
taken out of the machine, using a pipette man, 100 lL of
the esterase solution was introduced into the NMR tube.
The contents were carefully shaken for a minute the
spectrum was run and periodically monitored to ascer-
tain the disappearance of the starting material.
4.3. Statistical analysis
Hydrolysis rates were determined by fitting single
exponential decay equations to the disappearance of
each isomer substrate in the presence of enzyme. Ham-
mett Sigma values were correlated to the log trans-
4.2. Experimental conditions for lipase-mediated hydro-
lysis
For the kinetic study, a known amount of the phos-
phoramidate derivative was carefully weighed (5–7 mg)
using a Metller analytical balance and transferred into a
scintillation glass vial. Using a pipette man, 3 mL of
methanol were added and the contents were vortexed for
2 min until a homogeneous solution resulted. Using
another pipette man, 100 lL of the above solution was
transferred into another scintillation vial and to this was
added 950 lL of water and the contents vortexed. In
parallel, 5 mg of solid lipase powder was weighed and
transferred to a volumetric flask. To this was added
8 mL of water and the contents were shaken to dissolve
the enzyme. The reaction mixture was prepared as fol-
lows for the kinetic study: From the stock solution of
the compound as mentioned above, 500 lL of the
methanolic solution of the phosphoramidate derivative
was pipetted out into another glass vial and to this
formed hydrolysis rate constants using a linear
regression model (JMP Software, SAS Institute Inc.).
All p-values less than 5% were deemed significant. The
effect of sigma was tested for the fast rate constant, the
slow rate constant and the selectivity index (fast/slow).
4.4. Physical constants for the chloroethyl substituted
phosphoramdiate derivatives
4.4.1. 50-(20,30-Didehydro-30-deoxythymidine)phenyl N-
bis(2-chloroethyl)phosphoramidate. IR (KBr) 3184,
3061, 2365, 2329, 1690, 1490, 1455, 1249, 1087, 934, 774,
1
699 cmꢀ1. H NMR (CDCl3): d 1.86 (d, 3H, 5-CH3),
3.39–3.61 (m, 8H, NCH2 CH2Cl), 4.23–4.40 (m, 2H,
50H), 5.03 (s, 1H, 40H), 5.87–5.96 (m, 1H, 10H), 6.28–
6.37 (m, 1H, 20H), 7.00 (m, 1H, 30H), 7.17–7.20 (m, 2H,

3380
T. K. Venkatachalam et al. / Bioorg. Med. Chem. 12 (2004) 3371–3381
Ar-2, 6), 7.21 (s, 1H, 6H), 7.31–7.34 (m, 2H, Ar-3, 5),
8.55 (s, 1H, 3-NH); 13C NMR (CD3OD): d 12.90, 42.19,
49.67, 67.95, 84.79, 90.13, 111.61, 120.24, 125.66,
127.70, 130.10, 133.25, 135.71, 150.79, 163.60; 31P NMR
d 4.66, 5.27; HPLC: Rt: 12.4, 14.1 min.
4.5.2. 50-[4-Bromophenyl methoxy alaninylphosphate]-
200,30-didehydro-30-deoxylthymidine, isomer 2 (R). TLC:
Rf : 0.51 (CHCl3/MeOH, 9:1), mp 160 ꢁC; 1H NMR
(CDCl3): d 8.21(b s), 7.45–7.43 (m), 7.42–7.27 (m), 7.26–
7.11 (m), 7.11–7.09 (m), 7.01–6.99 (m), 6.31–6.28 (d),
5.94–5.91(d), 5.01(s), 4.30–4.26 (m), 3.96–3.62 (m), 1.88
(s), 1.59 (b s), 1.40–1.35 (m); 31P NMR (CDCl3) 2.56; IR
m 3423, 3242, 2924, 2852, 1741, 1693, 1589, 1485, 1385,
4.4.2. 50-(20,30-Didehydro-30-deoxythymidine)4-methylphenyl
N-bis(2-chloroethyl)phosphoramidate. IR (KBr) 3037,
2959, 2360, 1692, 1507, 1465, 1248, 1089, 941, 818,
756 cmꢀ1. 1H NMR (CDCl3): d 1.85 (s, 3H, 5-CH3), 2.32
(s, 3H, CH3), 3.37–3.60 (m, 8H, NCH2 CH2Cl), 4.22–
4.41 (m, 2H, 50H), 5.03 (s, 1H, 40H), 5.92 (d, 1H,
J ¼ 18:0 Hz, 1H), 6.33 (m, 1H, J ¼ 18:0 Hz, 20H), 6.97–
7.14 (m, 5H, Ar-2, 3, 5, 6; 6H; 30H), 8.23 (br, 1H, 3-NH);
13C NMR (CD3OD): d 12.89, 21.11, 42.21, 49.71, 67.32,
84.79, 90.11, 111.59, 119.93, 127.67, 130.53, 133.32,
135.41, 135.77, 150.70, 163.45; 31P NMR (CDCl3): d
4.79, 5.42; HPLC: Rt: 21.9, 25.3 min.
1248, 1223, 1153, 1112, 1090, 1038, 1018, 930 cmꢀ1
;
HPLC Rt: 12.1 min, ½aꢂ ꢀ27:5.
D
References and notes
1. Uckun, F. M.; Qazi, S.; Pendergrass, S.; Venkatachalam,
T. K.; Mao, C.; Richman, D. Antimicrob. Agents Chemo-
ther. 2002, 46, 3613.
2. Uckun, F. M.; Vig, R.; Aryl phosphate derivatives of D4T
having anti-HIV activity. U.S. Patent Number 6,030,957.
Issue date: 2-29-2000. U.S. Patent Number 6,350,736.
Issue date: 2-26-2002.
4.4.3. 50-(20,30-Didehydro-30-deoxythymidine)4-methoxy-
phenyl N-bis(2-chloroethyl)phosphoramidate. IR (KBr)
3066, 2958, 2360, 2339, 1695, 1506, 1462, 1249, 1091,
1034, 936, 838 cmꢀ1. 1H NMR (CDCl3): d 1.86 (s, 3H, 5-
CH3), 3.37–3.60 (m, 8H, NCH2 CH2Cl), 3.79 (s, 3H,
OCH3), 4.22–4.39 (m, 2H, 50H), 5.03 (s, 1H, 40H), 5.92
(m, 1H, 10H), 6.33 (m, 1H, 20H), 6.83–6.87 (m, 2H, Ar-2,
6), 7.09–7.14 (m, 2H, Ar-3, 5), 7.34 (s, 1H, 6H), 7.27 (m,
1H, 30H), 8.14 (s, 1H, 3-NH); 13C NMR (CD3OD): d
12.90, 42.22, 49.69, 55.95, 67.83, 84.79, 90.11, 114.99,
121.14, 127.68, 133.31, 135.36, 135.76, 150.70, 163.80;
31P NMR (CDCl3): d 5.09, 5.69; HPLC: Rt: 14.4,
16.6 min.
3. Uckun, F. M.; Chen, C. L.; Lisowski, E.; Mitcheltree, G.
C.; Venkatachalam, T. K.; Erbeck, D.; Chen, H.; Wau-
rzyniak, B. Arzneimittelforschung Drug Res. 2003, 53(5), 357.
4. Uckun, F. M.; Samuel, P.; Qazi, S.; Chen, C.; Pendergrass,
S.; Venkatachalam, T. K. Antiviral Chem. Chemother.
2002, 13, 197.
5. Uckun, F. M.; Qazi, S.; Pendergrass, S.; Lisowski, E.;
Waurzyniak, B.; Chen, C.; Venkatachalam, T. K. Anti-
microb. Agents Chemother. 2002, 46, 3428.
6. Uckun, F. M.; Chen, C. L.; Samuel, P.; Pendergrass, S.;
Venkatachalam, T. K.; Waurzyniak, B.; Qazi, S. Antimic-
rob. Agents Chemother. 2003, 47, 1233.
7. Chen, C. L.; Yu, G.; Venkatachalam, T. K.; Uckun, F. M.
Drug Metab. Dispos. 2002, 30, 1523.
8. Chen, C. L.; Venkatachalam, T. K.; Zhu, Z. H.; Uckun,
F. M. Drug Metab. Dispos. 2001, 29, 1035.
9. McIntee, E. J.; Remmel, R. P.; Schinazi, R. F.; Abraham,
T. W.; Wagner, C. R. J. Med. Chem. 1997, 40, 3323.
10. Venkatachalam, T. K.; Tai, H. L.; Vig, R.; Chen, C. L.;
Jan, S.; Uckun, F. M. Bioorg. Med. Chem. Lett. 1998, 8,
3121.
11. McGuigan, C.; Cahard, D.; Sheeka, H. M.; De Clercq, E.;
Balzarini, J. J. Med. Chem. 1996, 39, 1748.
12. McGuigan, C.; Pathirana, R. N.; Balzarini, J.; De Clercq,
E. J. Med. Chem. 1993, 36, 1048.
13. McGuigan, C.; Pathirana, R. N.; Mahmood, N.; Devine,
K. G.; Hay, A. J. Antiviral Res. 1992, 17, 311.
14. Wagner, C. R.; Iyer, V. V.; McIntee, E. J. Med. Res. Rev.
2000, 26, 417.
15. Song, H.; Johns, R.; Griesgraber, G. W.; Wagner, C. R.;
Zimmerman, C. L. Pharm. Res. 2003, 20, 448.
16. Meier, C. Mini. Rev. Med. Chem. 2002, 2, 219.
17. Mazzon, C.; Rampazzo, C.; Scaini, M. C.; Gallinaro, L.;
Karlsson, A.; Meier, C.; Balazarini, J.; Reichard, P.;
Bianchi, V. Biochem. Pharmacol. 2003, 66, 471.
18. Balzarini, J.; Haller-Meier, F.; De Clercq, E.; Meier, C.
Antiviral Chem. Chemother. 2001, 5, 301.
19. Song, H.; Griesgraber, G. W.; Wagner, C. R.; Zimmer-
man, C. L. Antimicrob. Agents Chemother. 2002, 46, 1357.
20. McGuigan, C.; Cahard, D.; Sheeka, H. M.; De Clercq, E.;
Balzarini, J. Bioorg. Med. Chem. Lett. 1996, 6, 1183.
21. Siddiqui, A. Q.; McGuigan, C.; Ballatore, C.; Zuccotto,
F.; Gilbert, I. H.; De Clercq, E.; Balzarini, J. J. Med.
Chem. 1999, 42, 4122.
4.4.4. 50-(20,30-Didehydro-30-deoxythymidine)4-bromophen-
yl N-bis(2-chloroethyl)phosphoramidate. IR (KBr) 3171,
3046, 2361, 2339, 1693, 1485, 1467, 1251, 1090, 930, 837,
778 cmꢀ1 1H NMR (CDCl3): d 1.90 (s, 3H, 5-CH3),
.
3.42–3.84 (m, 8H, NCH2 CH2Cl), 4.18–4.40 (m, 2H,
50H), 5.04 (s, 1H, 40H), 5.93–5.96 (m, 1H, 10H), 6.29–
6.37 (m, 1H, 20H), 7.00 (m, 1H, 30H), 7.07–7.12 (m, 2H,
Ar-2, 6), 7.34 (s, 1H, 6H), 7.44–7.47 (m, 2H, Ar-3, 5),
8.36 (s, 1H, 3-NH); 13C NMR (CD3OD): d 12.90, 42.19,
49.62, 67.83, 84.87, 90.18, 122.02, 127.71, 133.12,
134.47, 135.27, 135.58, 136.13, 150.67, 163.80; 31P NMR
(CDCl3): d 4.71, 5.22; HPLC: Rt: 31.8, 35.8 min.
4.5. Physical constants of individual separated individual
stereoisomers of DDE113
4.5.1. 50-[4-Bromophenyl methoxy alaninylphosphate]-
20,30-didehydro-30-deoxylthymidine, isomer 1 (S). TLC:
Rf : 0.51 (CHCl3/MeOH, 9:1), mp 59 ꢁC. 1H NMR
(CDCl3): d 8.15 (b s), 7.45–7.42 (m), 7.42–7.27 (m),
7.26–7.19 (m), 7.08–7.03 (m), 6.36–6.35 (d), 5.91–5.89
(d), 5.04 (s), 4.36–4.32 (m), 4.0–3.92 (m), 3.71–3.65 (m),
3.61–3.54 (m), 1.82 (s), 1.58 (b s), 1.34–1.32 (m); 31P
NMR (CDCl3) 3.11; IR m 3228, 3064, 2918, 2848, 1743,
1689, 1485, 1385. 1246, 1223, 1153, 1113, 1088, 1036,
22. Balzarini, J.; Heredewijn, P.; De Clercq, E. J. Biol. Chem.
1989, 264, 6127.
928, 837 cmꢀ1; HPLC Rt: 11.23 min, ½aꢂ ꢀ30.
D

T. K. Venkatachalam et al. / Bioorg. Med. Chem. 12 (2004) 3371–3381
3381
23. Wagner, C. R.; McIntee, E. J.; Schinazi, R. F.; Abraham,
T. W. Bioorg. Med. Chem. Lett. 1995, 5, 1819.
24. Vig, R.; Venkatachalam, T. K.; Uckun, F. M. Antiviral
Chem. Chemother. 1998, 9, 445.
25. Cygler, M.; Grochulski, P.; Kazlauskas, R. J.; Schrag,
J. D.; Bouthillier, F.; Rubn, B.; Serreqi, A. N.; Gupta,
A. K. J. Am. Chem. Soc. 1989, 116, 3180.
26. Tsi, M. D.; Chang, T. T. J. Am. Chem. Soc. 1980, 102,
5416.
27. Bruzik, K.; Tsi, M. D. J. Am. Chem. Soc. 1982, 104, 863.
28. Bruzik, K.; Jiang, R. T.; Tsi, M. D. J. Am. Chem. Soc.
1983, 105, 5455.
38. Lang, D.; Hofmann, B.; Haalck, L.; Hecht, H. J.; Spener,
F.; Schmid, R. D.; Schomburg, D. J. Mol. Biol. 1996, 259,
704.
39. Roussel, A.; Miled, N.; Berti Dupuis, L.; Riviere, M.;
Spinelli, S.; Berna, P.; Gruber, V.; Verger, R.; Cambillau,
C. J. Biol. Chem. 2002, 277, 2266.
40. van Pouderoyen, G.; Eggert, T.; Jaeger, K. E.; Dijkstra, B.
W. J. Mol. Biol. 2001, 309, 215.
41. Terzyan, S.; Wang, C. S.; Downs, D.; Hunter, B.; Zhang,
X. C. Protein Sci. 2000, 9, 1783.
42. Nardini, M.; Lang, D. A.; Liebeton, K.; Jaeger, K. E.;
Dijkstra, B. W. J. Biol. Chem. 2000, 275, 31219.
43. Canaan, S.; Roussel, A.; Verger, R.; Cambillau, C.
Biochim. Biophys. Acta 1999, 1441, 197.
29. Jiang, R. T.; Shyy, Y. J.; Tsi, M. D. Biochemistry 1984, 23,
1661.
30. Cavalier, J. F.; Verger, R.; Buono, G. Acc. Chem. Res.
2000, 33, 579.
31. Marguet, F.; Douchet, I.; Cavalier, J. F.; Buono, G.;
Verger, R. Colloids Surf. B: Biointerfaces. 1999, 13, 37.
32. Marguet, F.; Cavalier, J. F.; Verger, R.; Buono, G. Eur. J.
Org. Chem. 1999, 7, 1671.
33. Tyndall, J. D.; Sinchaikul, S.; Fothergill; Gilmore, L. A.;
Taylor, P.; Walkinshaw, M. D. J. Mol. Biol. 2002, 323, 859.
34. Wei, Y.; Contreras, J. A.; Sheffield, P.; Osterlund, T.;
Derewenda, U.; Kneusel, R. E.; Matern, U.; Holm, C.;
Derewenda, Z. S. Nat. Struct. Biol. 1999, 6, 340.
35. Wang, X.; Wang, C. S.; Tang, J.; Dyda, F.; Zhang, X. C.
Structure (London, England) 1997, 5, 1209.
36. Kim, K. K.; Song, H. K.; Shin, D. H.; Hwang, K. Y.; Suh,
S. W. Structure (London, England) 1997, 5, 173.
37. Kohno, M.; Funatsu, J.; Mikami, B.; Kugimiya, W.;
Matsuo, T.; Morita, Y. J. Biochem. (Tokyo) 1996, 120, 505.
44. Bourne, Y.; Martinez, C.; Kerfelec, B.; Lombardo,
D.; Chapus, C.; Cambillau, C. J. Mol. Biol. 1994, 238,
709.
45. Noble, M. E.; Cleasby, A.; Johnson, L. N.; Egmond,
M. R.; Frenken, L. G. FEBS Lett. 1993, 331, 123.
46. Derewenda, Z. S.; Derewenda, U.; Dodson, G. G. J. Mol.
Biol. 1992, 227, 818.
47. Hata, Y.; Matsuura, Y.; Tanaka, N.; Kakudo, M.;
Sugihara, A.; Iwai, M.; Tsujisaka, Y. J. Biochem. (Tokyo)
1979, 86, 1821.
48. Mao, C.; Vig, R.; Venkatachalam, T. K.; Sudbeck, E. A.;
Uckun, F. M. Bioorg. Med. Chem. Lett. 1998, 8, 2213.
49. Bohm, H. J. J. Comput. Aided Mol. Des. 1992, 6, 593.
50. Bohm, H. J. J. Comput. Aided Mol. Des. 1994, 8, 243.
51. Mao, C.; Sudbeck, E. A.; Venkatachalam, T. K.; Uckun,
F. M. Biochem. Pharmacol. 2000, 60, 1251.
52. Insight, 1996, San Diego, CA.