Potential multifunctional inhibitors of HIV-1 reverse transcriptase. Novel [AZT]-[TSAO-T] and [d4T]-[TSAO-T] heterodimers modified in the linker and in the dideoxynucleoside region

Article abstract of DOI:10.1021/jm991092+

In an attempt to combine the anti-HIV-inhibitory capacity of nucleoside reverse transcriptase (RT) inhibitors (NRTI) and non-nucleoside RT inhibitors (NNRTI), several heterodimer analogues of the previously reported [AZT]- (CH₂)₃-[TS

Full text of DOI:10.1021/jm991092+

5188
J . Med. Chem. 1999, 42, 5188-5196
P oten tia l Mu ltifu n ction a l In h ibitor s of HIV-1 Rever se Tr a n scr ip ta se. Novel
[AZT]-[TSAO-T] a n d [d 4T]-[TSAO-T] Heter od im er s Mod ified in th e Lin k er a n d
in th e Did eoxyn u cleosid e Region
Sonsoles Vela´zquez,† Victoria Tun˜o´n,† Mar´ıa Luisa J imeno,† Cristina Chamorro,† Erik De Clercq,‡
J an Balzarini,‡ and Mar´ıa J ose´ Camarasa*,†
Instituto de Quı´mica Me´dica (C.S.I.C.), J uan de la Cierva 3, 28006 Madrid, Spain, and Rega Institute for Medical Research,
Katholieke Universiteit Leuven, B-3000 Leuven, Belgium
Received J une 3, 1999
In an attempt to combine the anti-HIV-inhibitory capacity of nucleoside reverse transcriptase
(RT) inhibitors (NRTI) and non-nucleoside RT inhibitors (NNRTI), several heterodimer
analogues of the previously reported [AZT]-(CH2)3-[TSAO-T] prototype have been prepared. In
these novel series, other NRTIs, an expanded range of linkers with different conformational
freedom and other attachment sites for these linkers on the base part of the NRTI analogue
have been explored. Moreover, in order to circumvent the dependence of the NRTI moiety of
the heterodimer on activation by cellular nucleoside kinases, novel heterodimers in which the
NRTI is bearing a masked monophosphate group at the 5′-position are described. Among the
novel heterodimers, several derivatives show a potent anti-HIV-1 activity, which proved
comparable, or even superior, to that of the AZT heterodimer prototype. The nature of the
NRTI was important for the eventual anti-HIV-1 activity. In particular, the d4T heterodimer
derivative containing a propyl linker between the N-3 positions of the base of TSAO-T and
d4T was ∼5- to 10-fold more inhibitory to HIV-1 than the corresponding AZT heterodimer
prototype.
In tr od u ction
TSAO with the HIV-1 RT.12-15 Our experimental data
strongly suggest a specific interaction of the 3′-spiro
moiety of TSAO molecules with the glutamic acid
residue at position 138 (Glu-138) of the p51 subunit of
HIV-1 RT.13b,14
Reverse transcriptase (RT) represents an attractive
target for the chemotherapy of human immunodefi-
ciency virus (HIV) infection because of its key role in
virus replication.1,2 Several members of different classes
of HIV inhibitors that are targeted at the virus-encoded
RT are currently approved for the treatment of HIV-
infected individuals. Among them, the nucleoside RT
inhibitors (NRTIs) such as 3′-azido-2′,3′-dideoxythymi-
dine (AZT), 2′,3′-dideoxyinosine (ddI), 2′,3′-dideoxycy-
tidine (ddC), 2′,3′-didehydro-2′,3′-dideoxythymidine (d4T),
and (-)2′,3′-dideoxy-3′-thiacytidine (3TC) have been
widely used to treat AIDS patients.3,4 Another impor-
tant group of RT inhibitors is the so-called non-nucleo-
side RT inhibitors (NNRTIs), i.e., nevirapine, delavir-
dine, efavirenz, emivirine, PETT, Quinoxaline, GW420-
867X, TSAO, etc., which are structurally diverse but
highly specific to HIV-1 RT.5-7 Of the NNRTIs, nevi-
rapine, delavirdine, and efavirenz have been formally
approved for the treatment of HIV-1 infection in com-
bination with NRTIs and/or protease inhibitors.5-7 They
interact noncompetitively with the enzyme at an allo-
steric and highly hydrophobic nonsubstrate binding site
that is distinct from, but functionally and also spatially
associated with, the substrate binding site.8-10
To obtain a sustained benefit from antiviral therapy,
combination of different anti-HIV agents has become the
standard clinical practice to prevent emergence of virus-
drug resistance.16-19 An interesting alternative ap-
proach to combination therapy, supported by struct-
ural8-10 and biochemical studies,20 would be the use of
heterodimers resulting from the linking of a NNRTI and
an NRTI through an appropriate spacer in an attempt
to combine the inhibitory capacity of these two different
classes of molecules. With this aim, in 1995, we reported
for the first time21,22 the synthesis and anti-HIV activity
assessment of a series of heterodimers which combine
in their structure an NRTI analogue such as AZT, or
the natural substrate dThd, and an NNRTI such as
TSAO-T or HEPT linked through flexible polymethylene
spacers between the N-3 positions of the thymine bases
of both compounds. The most active compound of this
series was the [TSAO-T]-(CH2)3-[AZT] heterodimer (1).
Polymethylene linkers [-(CH2)n-] with n ) 4-6 showed
good antiviral activity, while longer spacers n > 7
dramatically disminished activity. However, they were
less potent inhibitors than the parent compounds from
which they were derived.21
Among NNRTIs, the TSAO nucleosides11,12 occupy a
unique position in that they interfere at the interface
between the p51 and p66 subunits of HIV-1 RT.13a Well-
defined amino acids at both the p51 and p66 RT
subunits are needed for an optimum interaction of
To obtain better insights in the feasibility of this
heterodimer approach to increase the inhibitory efficacy
of the test compounds against HIV-RT, this paper
describes the synthesis and anti-HIV evaluation of a
novel series of [NRTI]-spacer-[TSAO] heterodimers.23 In
† Instituto de Qu´ımica Me´dica.
‡ Katholieke Universiteit Leuven.
10.1021/jm991092+ CCC: $18.00 © 1999 American Chemical Society
Published on Web 11/18/1999

Multifunctional Inhibitors of HIV-1-RT
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 25 5189
the corresponding 5′-phenylphosphoramidate derivative
of AZT (2b),25 d4T (2c),26 and dThd (2d ) (Scheme 1).
Intermediates 2b and 2c were prepared as described.25,26
Compound 2d was prepared by reaction of thymidine
with (phenylmethoxy)alaninyl chlorophosphate28 in THF
in the presence of N-methylimidazole. These starting
materials were a mixture of two diastereoisomers, in
an approximately 1:1 ratio, resulting from the mixed
stereochemistry at the phosphate center and were used
as isomeric mixtures in the synthetic sequence.
Next, we focused on modifications on the spacer in
the model heterodimer. The synthesis of [AZT]N3-
spacer-N3[TSAO-T] heterodimers 8a -d was carried out
as shown in Scheme 2. Spacers no longer than six atoms
were introduced based on our previous biological re-
sults.21 Reaction of AZT with commercially available
2-bromoethyl ether, 1,4-dibromo-2-butene (E), R,R′-
dibromo-p-xylene, and 1,4-dichloro-2-butyne reagents,
in the presence of K2CO3, yielded the N-3-bromo inter-
mediates of AZT 7a -d (92%, 80%, 75%, and 35%,
respectively). Treatment of these intermediates with
TSAO-T gave the corresponding heterodimers 8a -d
(75%, 73%, 92%, and 23%, respectively). The lower
yields of 7d and 8d could be explained by the lower
reactivity of the corresponding reagent (1,4-dichloro-2-
butyne) that led to reaction mixtures of the final product
together with unreacted starting material. Longer reac-
tion times did not improve the yields due to partial
deprotection of the tert-butyl-dimethylsilyl groups.
F igu r e 1. Modifications carried out on the TSAO-spacer-
NRTI heterodimer molecule.
these novel series, several modifications in the spacer
and in the NRTI moiety of the model heterodimer (1)
were addressed (Figure 1).
First, we focused on the extension of the heterodimer
approach to other approved NRTIs, and therefore, we
replaced the AZT in the model heterodimer (1) by d4T.
Since NRTIs, and also dThd, interact with RT only
after they have been converted to their 5′-triphosphate
form, and since it is also likely that the [TSAO-T]-
(CH2)n-[NRTI] heterodimers would not be recognized by
the cellular kinases involved in the three phosphoryl-
ation steps, we prepared novel heterodimers in which
the NRTI (AZT, d4T) or the natural substrate (dThd)
bear at the 5′-position a masked monophosphate group.
Among the most established and successful pro-nucle-
otide approaches the aryloxyaminoacyl-phosphorami-
dates introduced by McGuigan et al.24 have been chosen
based on the pronounced anti-HIV activity of the
phenylphosphoramidate derivatives of AZT25 and d4T.26
We also prepared a series of [AZT]-spacer-[TSAO-T]
heterodimers in which the flexible polymethylene spacer
of the heterodimer prototype (1) was replaced by an
expanded range of spacers of different nature and
conformational freedom.
Finally, for the synthesis of heterodimers 11a -c,
where the position of the linker on the NRTI (AZT, d4T,
or dThd) was changed from the N-3 to the C-5 position
of the base moiety, we developed a facile and straight-
forward two-step procedure outlined in Scheme 3. The
key step for the synthesis involves the palladium cross-
coupling reaction29-34 of terminal alkyne 9 with 5-iodo
nucleosides 10a -c. The requisite 5-iodonucleosides
10a -c were prepared as follows. Compound 10a 35 was
synthesized by direct iodination of 3′-azido-2′-deoxyuri-
dine (AZU)36 which in turn was prepared in two stepss
by direct conversion of 2′-deoxyuridine into 2,3′-anhy-
drouridine, under Mitsunobu conditions, followed by
azidation at the 3′-position.37 The synthesis of 5-iodo-
nucleoside 10b38 was carried out following the procedure
developed by Larsen et al.39 but improved by using the
easily synthesized40 or commercially available 5-iodo-
2′-deoxyuridine as starting material. The other coupling
partner, the bulky highly functionalized N-3 propargyl
TSAO analogue 9, was easily prepared in excellent yield
(90%) by reaction of TSAO-T with propargylbromide in
the presence of K2CO3. The coupling reaction of terminal
alkyne intermediate 9 with 5-iodoprecursors 10a -c,
performed in DMF with tetrakis(triphenylphosphine)-
palladium(0), copper(I) iodide, and triethylamine under
argon at room temperature, gave heterodimers 11a -c
(55%, 46%, and 61% yields, respectively) together with
compound 12, as a minor side product, due to partial
self-dimerization of the alkyne 9 (Scheme 3). The
catalyst (Ph3P)4Pd(0) was found to be specific for the
reaction. Other catalysts, i.e., bis(triphenylphosphine)-
palladium(II) chloride, failed to give the coupled prod-
ucts. Under such mild conditions, the target het-
erodimers were obtained in moderate to good yields
without hydroxyl group protection of the NRTI (or dThd)
With the TSAO derivatives, we have ascertained that
linkage of the methylene spacer to the N-3 position of
the thymine base maintains the antiviral activity.27 In
contrast, attachment points of the spacer to the NRTI
other than the N-3 position were explored to preserve
potential base-pairing of the template with the het-
erodimer.
Ch em istr y
In an attempt to investigate the role of the NRTI in
the activity of the heterodimers we first addressed our
attention to the synthesis of [d4T]-(CH2)n-[TSAO-T] (5a ,
6a ) heterodimers. The synthesis of these compounds
was accomplished according to the method described in
our previous paper.21 Thus, as shown in Scheme 1,
treatment of d4T (2a ) with 2 equiv of 1,3-dibromopro-
pane or 1,6-dibromohexane in dry acetone:DMF (1:1)
and in the presence of potassium carbonate (K2CO3)
gave the N-3-substituted derivatives 3a (75%) and 4a
(75%). Subsequent reaction of these intermediates with
TSAO-T, under basic conditions (K2CO3), gave het-
erodimers 5a (50%) and 6a (60%). The synthesis of a
series of 5′-(phenylphosphoramidates) heterodimers 5b-d
and 6b-d was carried out in a similar way starting from

5190 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 25
Vela´zquez et al.
Sch em e 1
Sch em e 2
Sch em e 3
moiety or deprotection of the labile tert-butyldimethyl-
silyl groups. Finally, compound 11c was subjected to
hydrogenolysis on Pd/C to provide heterodimer 13c with
a flexible polymethylene linker in 83% yield (Scheme
3).
Structures of the novel heterodimers 5a -d , 6a -d ,
8a -d , 11a -c, and 13c were assigned on the basis of
their analytical and spectroscopic data. AZT het-
erodimers with different spacers 8a -d and AZT and
dThd heterodimers 11a ,c and 13 linked through the C-5
position of their base moiety were assigned by compari-
1
son of their H NMR with the 1H NMR of the previously

Multifunctional Inhibitors of HIV-1-RT
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 25 5191
F igu r e 2. 1H NMR (in bold) and 13C NMR (in parentheses)
chemical shifts (ppm) assignments for [d4T]-(CH2)3-[TSAO-T]
heterodimer 5a .
F igu r e 4. 13C NMR chemical shifts (ppm) assignments for
the [5′-MMP-d4T]-(CH2)3-[TSAO-T] heterodimer 5c. This com-
pound is a mixture of diastereoisomers that could not be
separated. The chemical shifts for both diastereoisomers are
shown.
Biologica l Resu lts a n d Discu ssion
The inhibitory activity of the test compounds was
explored against HIV-1 (IIIB) and HIV-2 (ROD) replica-
tion in MT-4 and CEM cell cultures (Table 1). TSAO-T
was included as reference compound. The prototype
heterodimer [AZT]-(CH2)3-[TSAO-T] 1 was inhibitory to
HIV-1-induced cytopathicity at a 50% effective concen-
tation (EC50) ranging between 0.09 and 0.38 µM.
Replacement of AZT in the prototype heterodimer 1
by d4T (heterodimer 5a ) resulted in a 5- to 10- fold
higher inhibitory effect against HIV-1 than against 1
and in a 2-fold more potent inhibitory effect than
unsubstituted TSAO-T.11a,b Longer methylene linkers
(i.e., n ) 6, derivative 6a ) decreased the antiviral
activity of 5a by 6-fold.
In the prototype heterodimer 1 we also varied the
nature of the spacer. No marked differences in anti-
HIV-1 activity were observed when the spacer consisted
of a butynyl (8b), butenyl (8d ), or ethoxyethyl (8a )
moiety (EC50: 0.18-0.34 µM in CEM cells and 0.31-
1.00 µM in MT-4 cells). Only when a benzylic spacer
was introduced (8c) was a markedly lower antiviral
activity noted.
Within the phenyloxy phosphoramidate series, the
d4T (5c, 6c) or thymidine (5d , 6d ) analogues showed
marked anti-HIV-1 activity ranging between 0.17 and
0.33 µM for 5c and 5d and between 0.25 and 0.93 µM
for 6c and 6d . In these series, the most potent derivative
was the d4T analogue 5c, but no improved antiviral
potency could be noted when compared to the corre-
sponding nonphosphorylated analogue 5a . The AZT
phenyloxy phosphoramidate heterodimers 5b and 6b
were markedly less active than their corresponding d4T
(5c, 6c) and dThd (5d , 6d ) analogues. In contrast with
the AZT phosphoramidate heterodimers 5b and 6b, the
AZT phosphoramidate monomers 3b and 4b lacking the
F igu r e 3. 1H NMR chemical shifts (ppm) assignments for the
[5′-MMP-d4T]-(CH2)3-[TSAO-T] heterodimer 5c. This com-
pound is a mixture of diastereoisomers that could not be
separated. The chemical shifts for both diastereoisomers are
shown.
reported [AZT or dThd]-(CH2)n-[TSAO-T] heterodimers
whose structures were unequivocally determined.21 For
the assigment of d4T heterodimers 5a , 6a , and 11b and
5′-(phenylphosphoramidates) heterodimers 5b-d and
6b-d , one- and two-dimensional (1D and 2D) NMR
tecniques were used. The methodology followed is il-
lustrated with compounds 5a and 5c, which were chosen
as model compounds. Full asignment of 5′-(phenylphos-
phoramidate) heterodimer 5c was not obvious because
this compound was a mixture of the two diastereoiso-
mers resulting from the mixed stereochemistry at the
phosphate center, as was evident from the 31P NMR,
1H NMR, and 13C NMR spectra. The full assignment of
1H spectra were achieved by COSY and TOCSY 2D
experiments and selective 1D-TOCSY experiments. The
full assignment of 13C spectra was achieved by HSQC
and HMBC 2D experiments. 1H NMR and 13C NMR
spectral data for compound 5a are shown in Figure 2.
1H NMR and 13C NMR spectral data for nucleotide 5c
are shown in Figures 3 and 4, respectively. Structures
of heterodimers 6a , 11b, 5b,d , and 6b-d were deter-
1
mined through comparison of their H NMR spectra to
those of the model compounds.

5192 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 25
Ta ble 1. Anti-HIV and Cytostatic Activity of Test Compounds
Vela´zquez et al.
a
EC50 (µM)
b
MT-4
CEM
CC50 (µM)
compd
HIV-1
HIV-2
HIV-1
HIV-2
MT-4
>100
CEM
1
0.38 ( 0.07
0.09 ( 0.01
>100
3b
4b
5a
5b
5c
5d
6a
6b
6c
6d
8a
8b
8c
8d
11a
11b
11c
12
>50
>10
>50
>50
>10
0.02 ( 0.03
>50
44 ( 3.3
22 ( 1.9
16 ( 12
41 ( 4.9
22 ( 1.5
189 ( 64
>10
>50
>10
>50
0.04 ( 0.01
3.2 ( 2.1
0.18 ( 0.02
0.33 ( 0.14
0.23 ( 0.06
2-250c
0.57 ( 0.37
0.93 ( 0.08
0.31 ( 0.14
0.70 ( 0.15
1.7 ( 1.2
1.0 ( 0.01
0.33 ( 0.20
0.18 ( 0.01
0.44 ( 0.31
>250
0.65 ( 0.1
0.17 ( 0.03
0.25 ( 0.15
0.12 ( 0.03
1.7 ( 1.1
>250
>250
>250
123 ( 0.7
>250
26 ( 8.1
15 ( 7.1
28 ( 19
31 ( 15
>50
>250
>250
>50
>50
>50
>250
>50
116 ( 55
>250
>250
>250
>250
0.4 ( 0.0
37 ( 13
25 ( 4.6
>50
0.25 ( 0.04
0.18 ( 0.04
0.34 ( 0.09
5.3 ( 4.2
>50
>50
57 ( 40
>250
>50
104 ( 6.7
>250
>250
>50
>250
>250
>250
>250
>250
>250
>250
>250
>250
>10
0.23 ( 0.14
0.20 ( 0.1
0.11 ( 0.1
0.55 ( 0.21
106 ( 9.7
31 ( 19
8.0 ( 9.4
15 ( 2.9
65 ( 20
24
>10
>10
>10
>10
>250
>10
>6
21 ( 1.1
21 ( 0.1
>250
21 ( 1.3
>50
>250
>10
>50
13
0.51 ( 0.43
0.06
0.38 ( 0.28
0.05
TSAO-Td
>6
12 ( 2.4
a
b
Effective concentration of 50% or the compound concentration required to inhibit HIV-induced cytopathicity. Cytostatic concentration
of 50%. c Within the concentration range of 2 and 250 µM, the inhibitory effect of the compound was close to 50%. Therefore, an exact
EC50 value could not be determined. Data taken from ref 11a,b.
d
TSAO moiety but containing the linker at the N-3
position of the base were devoid of anti-HIV activity.
This observation suggests that the AZT moiety did not
contribute to the antiviral activity of the heterodimers
5b and 6b and that the NH-3 fulfill a crucial role for
enzyme recognition of AZT.
Con clu sion s
In summary, in this paper we report on novel deriva-
tives of the prototype heterodimer (1) by modifying the
NRTI moiety, the conformational freedom of the linker,
and the anchoring point of the linker to the NRTI
moiety. Moreover, a series of arylphosphoramidate
heterodimers (5b-d and 6b-d), designed as membrane-
soluble prodrugs of the corresponding 5′-monophosphate
derivatives, have been prepared. Several members of
this class of compounds show potent anti-HIV-1 activi-
ties comparable and even superior to those of the
prototype heterodimer (1). The nature of the spacer and
the position of the linker on the NRTI can be changed
while keeping the antiviral activity. The phosphorami-
date series of compounds had no improved anti-HIV-1
activity over the corresponding nonphosphorylated ana-
logues. In contrast, the nature of the NRTI was shown
to be important for increasing the inhibitory efficacy of
the model heterodimer 1. In particular, the [d4T]-
(CH2)3-[TSAO-T] heterodimer analogue (5a ) was 5- to
10-fold more inhibitory to HIV-1 in MT-4 and CEM cells
cultures than the previously reported AZT counterpart
1, and even 2-fold more potent than the unsubstituted
TSAO-T parent compound.
Given the above-mentioned considerations, other at-
tachment points of the spacer to the NRTI have also
been explored by anchoring the linker at the C-5
position of the pyrimidine base of the NRTI. The
introduction of a propynyl spacer linked to the C-5
position (11a ) showed comparable antiviral activity to
that of the AZT heterodimer prototype (1), whereas in
the corresponding d4T analogue (11b) a 5-fold decrease
in anti-HIV potency was observed (compare 11b with
5a ). However, the spacer rigidity in the C-5 series of
compounds did not markedly influence the antiviral
potency, and thus, the dThd heterodimer 13c (bearing
a more flexible propyl group as spacer) was endowed
with anti-HIV-1 activity comparable to that of the
corresponding propynyl analogue 11c (EC50s ranging
between 0.38 and 0.55 µM).
With the exception of 5c, none of the compounds
investigated proved active against HIV-2 in MT-4 and
CEM cell cultures, pointing to the predominant role of
the NNRTI (TSAO-T) moiety in the eventual activity
of the test compounds. It is intriguing why the phos-
phoramidate heterodimer 5c (containing d4T as the
nucleoside) had marginal, but significant, anti-HIV-2
activity, whereas the corresponding AZT derivative had
not. This is the first TSAO derivative that was found
to show some, albeit marginal, activity, against HIV-2.
Although most of the compounds were poorly cytotoxic
to CEM cell cultures, the toxicity was in several cases
more pronounced in MT-4 cell cultures. This difference
could amount up to l0-fold (i.e., compounds 5a , 5d , 6c,
and 11a ). It is unclear which structural or functional
determinants on the TSAO-T heterodimers play a role
in the eventual toxicity.
Exp er im en ta l Section
Ch em ica l P r oced u r es. Microanalyses were obtained with
a Heraeus CHN-O-RAPID instrument. HRMS spectra were
registered in a VG autospec; the mode of ionization was fast
1
atom bombardment (FAB) using MNOBA as matrix. H and
13C NMR spectra were recorded on a Varian UNITY 500
spectrometer operating at 499.84 MHz (1H) and 125.71 MHz
(13C), respectively, using acetone-d6 or CDCl3 as solvent at 30
°C with TMS as internal standard. 31P spectra were recorded
on a Varian INOVA 400 spectrometer operating at 161.89
MHz, using acetone-d6 or CDCl3 as solvent at 30 °C with
phosphoric acid as external standard. Monodimensional 1H,
13C, and 31P spectra were obtained using standard conditions.
Homonuclear 2D spectra (COSY and TOCSY) were acquired
in the phase-sensitive mode. Data were collected in a 2048 ×

Multifunctional Inhibitors of HIV-1-RT
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 25 5193
512 matrix with a spectral width of 3461 Hz and 1.2 s of
relaxation delay and then processed in a 2048 × 1024 matrix.
In the TOCSY spectra a mixing time of 0.128 s was used. 2D
inverse proton detected heteronuclear one-bond shift correla-
tion spectra were obtained using the pulsed field gradient
HSQC pulse sequence. Data were collected in a 2048 × 512
matrix with a spectral width of 3460 Hz in the proton domain
and 22500 Hz in the carbon domain and were processed in a
2048 × 1024 matrix. The experiment was optimized for a one-
bond heteronuclear coupling constant of 150 Hz. 2D inverse
proton detected heteronuclear long-range shift correlation
spectra were obtained using the pulsed field gradient HMBC
pulse sequence. The HMBC experiment was acquired in the
same conditions of the HSQC experiment and optimized for
long-range coupling constants of 7 Hz. 1D-selective TOCSY
experiments were acquired using a selective pulse EBURP2-
256 with a B1 corresponding to 10 Hz and a spin-lock time of
128 ms.
Gen er a l P r oced u r e for th e Syn th esis of 3-N-(br om o-
sp a cer ) Nu cleosid e In ter m ed ia tes 3a , 4a , a n d 9a -d a n d
3-N -(B r o m o a lk y l)-5′-(p h e n y lm e t h o x y a la n in y l)p h o s -
p h a t e Nu cleot id e In t er m ed ia t es 3b-d a n d 4b-d . To a
solution of the nucleoside [AZT or d4T] (1 equiv) in acetone/
DMF (1:1) or the nucleotide (2d ) (1 equiv) in acetone were
added K2CO3 (1.1 equiv) and the corresponding dibromoalkyl,
alkenyl, aryl, or dichloroalkynyl reagent (2.0-6.0 equiv). The
reaction mixture was refluxed for 5-16 h. After evaporation
of the solvent, the residue was dissolved in ethyl acetate (14
mL), washed with water (2 × 14 mL), dried (Na2SO4), filtered,
and evaporated to dryness. The residue was purified by
CCTLC on the chromatotron. The reaction time, number of
equivalents of the dibromo reagent, chromatography eluent,
1
yield of the isolated products, and H NMR data are indicated
below for each compound.
N-(3-Br om op r op yl)-2′,3′-d id eh yd r o-2′,3′-d id eoxyth ym i-
d in e (3a ): reaction time 16 h; 6 equiv of 1,3-dibromopropane;
eluent dichloromethane/methanol, 20:1; yield of 3a (75%) as
Analytical TLC was performed on silica gel 60 F254 (Merck).
Separations on silica gel were performed by preparative
centrifugal circular thin-layer chromatography (CCTLC) on a
1
a syrup; H NMR (CDCl3) δ 1.80 (d, 3H, CH3-5, J ) 1.2 Hz),
2.17 (m, 2H, CH2), 3.51 (t, 2H, CH2Br, J ) 7.0 Hz), 3.79 (m,
2H, 2H-5′), 4.03 (t, 2H, CH2N, J ) 7.0 Hz), 4.12 (t, 1H, OH-
5′), 4.46 (m, 1H, H-3′), 4.86 (m, 1H, H-4′), 5.92 (ddd, 1H, H-2′,
J 2′,3′ ) 6.0, J 1′,2′ ) 1.3, J 2′,4′ ) 2.3 Hz), 6.42 (dt, 1H, H-3′, J 1′,3′
) J 3′,4′ ) 1.7 Hz), 6.99 (m, 1H, H-1′), 7.78 (q, 1H, H-6). Anal.
(C13H17BrN2O4) C, H, N.
Chromatotron (Kiesegel 60 PF
gipshaltig (Merck)), layer
254
thickness (1 mm), flow rate (5 mL/min). Flash column chro-
matography was performed with silica gel 60 (230-400 mesh)
(Merck). Analytical HPLC was carried out on a Waters 484
system using a mBondapak C18 (3.9 × 300 mm: 10 mm). The
systems used were as follows: (A) isocratic conditions, mobile
phase CH3CN/H2O (0.05% TFA); (B) gradient conditions,
mobile phase CH3CN/H2O (0.05% TFA); (C) linear gradient
conditions, 2-22 min 95%-10% water, 22-32 min 10% water,
32-50 min 10%-95% water. Flow rate: 1 mL/min. Detec-
tion: UV (214 nm). All retention times are quoted in minutes.
The phosphoramidate intermediates and heterodimers (2d ,
3b-d , 4b-d , 5b-d , and 6b-d ) were isolated as mixtures of
diastereoisomers, with the isomerism arising from mixed
stereochemistry at the phosphate center. Many NMR peaks
of the phosphoramidate compounds are split due to the
presence of diastereoisomers in the sample. These compounds
were noted to be hygroscopic and did not give useful microana-
lytical data but were found to be pure by high-field multi-
nuclear NMR spectroscopy, mass spectrometry, and rigorous
HPLC analysis.
3′-Azid o-3-N-(3-b r om op r op yl)-3′-d eoxyt h ym id in e-5′-
(p h en ylm eth oxya la n in yl)p h osp h a te (3b): reaction time 9
h; 4 equiv of 1,3-dibromopropane; eluent dichloromethane/
methanol, 20:1; yield of 3b (74%) as a syrup; HPLC Rt ) 5.70
1
min (system A, 45:55); 31P NMR (CDCl3) 3.14, 3.49 ppm; H
NMR (CDCl3) δ 1.34, 1.37 (d, 3H, Ala-CH3, J ) 7.0 Hz), 1.88,
1.90 (s, 3H, CH3-5), 2.05-2.48 (m, 4H, 2H-2′, CH2), 3.39 (t,
2H, CH2Br, J ) 6.8 Hz), 3.60 (m, 1H, Ala-NH), 3.68, 3.69 (s,
3H, Ala-OCH3), 3.92-4.40 (m, 7H, CH2N, Ala-CH, H-3′, H-4′,
2H-5′), 6.21 (m, 1H, H-1′), 7.16-7.38 (m, 6H, H-6, Ph); MS
m/e FAB 628.1040 (MH+, C23H30BrN6O8P requires 628.1045).
3-N-(3-Br om op r op yl)-2′,3′-d id eh yd r o-2′,3′-d id eoxyth y-
m id in e-5′-(p h en ylm eth oxya la n in yl)p h osp h a te (3c): reac-
tion time 7 h; 2 equiv of 1,3-dibromopropane; eluent dichlo-
romethane/methanol, 10:1; yield of 3c (48%) as a syrup; HPLC
Rt ) 5.51 min (system A, 40:60); 31P NMR (CDCl3) 3.13, 3.69
1
ppm; H NMR (CDCl3) δ 1.24, 1.28 (d, 3H, Ala-CH3, J ) 7.0,
All experiments involving water-sensitive compounds were
conducted under scrupulously dry conditions. Triethylamine
and acetonitrile were dried by refluxing over calcium hydride.
Tetrahydrofuran was dried by refluxing over sodium/ben-
zophenone. Anhydrous N,N′-dimethylformamide was pur-
chased from Aldrich.
7.0 Hz), 1.77, 1.81 (d, 3H, CH3-5, J ) 1.2, 1.2 Hz), 2.16 (m,
2H, CH2), 3.36, 3.36 (t, 2H, CH2Br, J ) 7.1, 7.1 Hz), 3.42, 3.50
(m, 1H, Ala-NH), 3.63, 3.64 (s, 3H, Ala-OCH3), 3.91 (m, 1H,
Ala-CH), 4.06 (m, 2H, CH2N), 4.17-4.34 (m, 2H, 2H-5′), 4.94,
4.97 (m, 1H, H-4′), 5.83 (m, 1H, H-2′), 6.21, 6.28 (dt, 1H, H-3′,
J 2′,3′ ) 6.0, 6.0 Hz, J 1′,3′ ) J 3′,4′ ) 1.8, 1.6 Hz), 7.00, 7.01 (m,
1H, H-1′), 7.08-7.28 (m, 6H, H-6, Ph); MS m/e FAB 585.0877
(MH+, C23H29BrN3O8P requires 585.0875).
5′-(P h en ylm eth oxyalan in yl)ph osph ate Th ym idin e (2d).
This synthesis was conducted in accordance with the procedure
reported by McGuigan et al.24,28 from thymidine and (phenyl-
methoxy)alaninyl phosphorochloridate. To a suspension of
thymidine (0.6 g, 2.47 mmol) in dry THF (9 mL) was added
N-methylimidazole (1.18 mL, 14.87 mmol). The resulting
solution was cooled to -20 °C, and phenyl methoxyalani-
nylphosphorochloridate (4.95 mmol) dissolved in THF (5 mL)
was slowly added. The clear solution became cloudy and slowly
demixed. The solution was stirred for 2 h at -20 °C and
overnight at room temperature. Water was added, and the
solvent was removed under reduced pressure. The residue was
dissolved in CHCl3 (25 mL) and washed with saturated NH4-
Cl (2 × 15 mL), water (15 mL), and brine (15 mL). The organic
phase was dried over Na2SO4, filtered, and concentrated to
dryness. The residue was purified by flash chromatography
on silica gel with dichloromethane/methanol, 10:1, to give
compound 2d as a white foam (0.49 g, 41%): HPLC Rt ) 5.51
min (40:60); 31P NMR (acetone-d6) 4.18, 4.39 ppm; 1H NMR
(acetone-d6) δ 1.32, 1.36 (d, 3H, Ala-CH3, J ) 7.1, 7.1 Hz), 1.80,
1.82 (d, 3H, CH3-5, J ) 1.1, 1.1 Hz), 2.15 (m, 2H, 2H-2′), 3.65
(s, 3H, Ala-OCH3), 4.11-4.56 (m, 5H, H-3′, H-4′, 2H-5′, Ala-
CH), 4.59, 5.00 (2m, 2H, OH, Ala-NH), 6.32 (m, 1H, H-1′),
7.24-7.37 (m, 5H, Ph), 7.57, 7.60 (d, 1H, H-6, J ) 1.1, 1.1 Hz),
9.98 (bs, 1H, NH-3); MS m/e FAB 483.1414 (MH+, C20H26N3O9P
requires 483.1406).
3-N-(3-Br om opr opyl)th ym idin e-5′-(ph en ylm eth oxyalan -
in yl)p h osp h a te (3d ): reaction time 30 h; 5 equiv of 1,3-
dibromopropane; eluent dichloromethane/methanol, 10:1; yield
of 3d (55%) as a syrup; HPLC Rt ) 3.45 min (system A, 45:
55); 31P NMR (CDCl3) 3.66, 4.10 ppm; 1H NMR (CDCl3) δ 1.29,
1.29 (d, 3H, Ala-CH3, J ) 7.0, 7.1 Hz), 1.84 (d, 3H, CH3-5, J )
0.9 Hz), 1.97 (m, 1H, H-2′a), 2.14 (m, 2H, CH2), 2.27 (m, 1H,
H-2′b), 3.35 (t, 2H, CH2Br, J ) 6.9 Hz), 3.63, 3.65 (s, 3H, Ala-
OCH3), 3.59-3.77 (m, 1H, Ala-NH), 3.89-4.44 (m, 8H, CH2N,
H-3′, H-4′, 2H-5′, Ala-CH, OH), 6.20, 6.24 (t, 1H, H-1′, J 1′,2′a
)
J 1′,2′b ) 6.4, 6.3 Hz), 7.09-7.29 (m, 6H, Ph, H-6); MS m/e FAB
603.0987 (MH+, C23H31BrN3O9P requires 603.0980).
N-(6-Br om oh exyl)-2′,3′-d id eh yd r o-2′,3′-d id eoxyth ym i-
d in e (4a ): reaction time 10 h; 4 equiv of 1,6-dibromohexane;
eluent dichloromethane/methanol, 20:1; yield of 4a (75%) as
a syrup; 1H NMR (CDCl3) δ 1.36-1.63 (m, 6H, 3CH2), 1.79 (d,
3H, CH3-5, J ) 1.1 Hz), 1.85 (m, 2H, CH2), 3.50 (t, 2H, CH2-
Br, J ) 7.0 Hz), 3.83 (m, 4H, CH2N, 2H-5′), 4.17 (t, 1H, 5′-
OH, J ) 5.8 Hz), 4.85 (m, 1H, H-4′), 5.92 (m, 1H, H-2′), 6.42
(m, 1H, H-3′), 6.99 (m, 1H, H-1′), 7.77 (d, 1H, H-6). Anal.
(C16H23BrN2O4) C, H, N.
3′-Azido-3-N-(6-br om oh exyl)-3′-deoxyth ym idin e-5′-(ph e-
n ylm eth oxya la n in yl)p h osp h a te (4b): reaction time 6 h; 3

5194 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 25
Vela´zquez et al.
equiv of 1,6-dibromohexane; eluent dichloromethane/methanol,
refluxed for 8-16 h. After evaporation of the solvent, the
residue was dissolved in ethyl acetate (14 mL), washed with
water (2 × 14 mL), dried (Na2SO4), filtered, and evaporated
to dryness. Repeated chromatography of the residue by
preparative CCTLC on the chromatotron was required to give
pure heterodimers.
The reaction time, chromatography eluent, and yield of the
isolated compounds are indicated below for each compound.
Heter odim er [TSAO-T]N3-(CH2)3-N3[d4T] (5a).41 TSAO-T
and 3a reacted for 5 h. The residue was chromatographed
(dichloromethane/methanol, 20:1) to give 5a (50%) as a white
foam: HPLC Rt ) 16.18 min (system A, 60:40); 1H NMR and
13C NMR are shown in Figure 2. Anal. (C37H59N5O12SSi2) C,
H, N, S.
20:1; yield of 4b (72%) as a syrup; 31P NMR (CDCl3) 3.09, 3.42
1
ppm; H NMR (CDCl3) δ 1.36, 1.37 (d, 3H, Ala-CH3, J ) 6.9,
6.8 Hz), 1.41-1.66 (m, 6H, 3CH2), 1.85 (m, 2H, CH2), 2.14 (m,
1H, H-2′a), 2.38 (m, 1H, H-2′b), 3.39 (t, 2H, CH2Br, J ) 6.8
Hz), 3.60 (m, 1H, Ala-NH), 3.70, 3.71 (s, 3H, Ala-OCH3), 3.91
(t, 2H, CH2N, J ) 7.6 Hz), 3.99-4.37 (m, 5H, Ala-CH, H-3′,
H-4′, 2H-5′), 6.18, 6.23 (t, 1H, H-1′, J 1′,2′a ) J 1′,2′b ) 6.6 Hz),
7.15-7.36 (m, 6H, H-6, Ph); MS m/e FAB 670.1521 (MH+,
C
26H36BrN6O8P requires 670.1515).
3-N-(6-Br om oh exyl)-2′,3′-d id eh yd r o-2′,3′-d id eoxyt h y-
m id in e-5′-(p h en ylm eth oxya la n in yl)p h osp h a te (4c): reac-
tion time 24 h; 4 equiv of 1,6-dibromohexane; eluent dichlo-
romethane/methanol, 20:1; yield of 4c (50%) as a syrup; HPLC
Rt ) 20.15 min (system A, 55:45); 31P NMR (CDCl3) 3.19, 3.73
Heter od im er [TSAO-T]N3-(CH2)3-N3[5′-MMP -AZT] (5b).
TSAO-T and 3b reacted for 7 h. The residue was chromato-
graphed with hexane/ethyl acetate (1:1) and subsequently with
dichloromethane/methanol (20:1) to give 5b (65%) as a white
foam: HPLC Rt ) 5.70 min (system A, 45:55), Rt ) 23.73 min
(system B); MS m/e FAB 1137.4087 (MH+, C47H72N9O16P SSi2
requires 1137.4093).
1
ppm; H NMR (CDCl3) δ 1.27-1.86 (m, 14H, 4CH2, Ala-CH3,
CH3-5), 3.39 (t, 2H, CH2Br, J ) 6.8 Hz), 3.52 (m, 1H, Ala-
NH), 3.69, 3.69 (s, 3H, Ala-OCH3), 3.90 (m, 3H, CH2N, Ala-
CH), 4.12-4.38 (m, 2H, 2H-5′), 5.00 (m, 1H, H-4′), 5.89 (m,
1H, H-2′), 6.27, 6.34 (m, 1H, H-3′), 7.06-7.35 (m, 7H, H-1′,
H-6, Ph); MS m/e FAB 627.1339 (MH+, C26H35BrN3O8P re-
quires 627.1344).
Heter od im er [TSAO-T]N3-(CH2)3-N3[5′-MMP -d 4T] (5c).
TSAO-T and 3c reacted for 5 h. The residue was chromato-
graphed (dichloromethane/acetone, 10:1) to give 5c (75%) as
a syrup: HPLC Rt ) 12.48 min (system A, 55:45); 31P NMR
3-N-(6-Br om oh exyl)th ym idin e-5′-(p h en ylm eth oxya la n -
in yl)p h osp h a te (4d ): reaction time 22 h; 4 equiv of 1,3-
dibromohexane; eluent dichloromethane/methanol, 20:1; yield
1
(CDCl3) 3.20, 3.70 ppm; H NMR and 13C NMR are shown in
1
of 4d (50%) as a syrup; 31P NMR (CDCl3) 3.56, 3.90 ppm; H
Figures 3 and 4; MS m/e FAB 1094.3931 (MH+, C47H71N6O16
P
NMR (CDCl3) δ 1.29-1.40 (m, 16H, Ala-CH3, CH3-5, 2H-2′,
4CH2), 3.34 (t, 2H, CH2Br, J ) 6.7 Hz), 3.63, 3.65 (s, 3H, Ala-
OCH3), 3.81-4.43 (m, 9H, CH2N, Ala-CH, Ala-NH, H-3′, H-4′,
2H-5′, OH), 6.24 (m, 1H, H-1′), 7.11-7.38 (m, 6H, H-6, Ph);
MS m/e FAB 645.1450 (MH+, C26H37BrN3O9P requires
645.1450).
3′-Azid o-3-N-(5-br om oeth yl eth er )-3′-d eoxyth ym id in e
(7a ): reaction time 24 h; 3 equiv of 2-bromoethyl ether; eluent
dichloromethane/methanol, 10:1; yield of 7a (92%) as a syrup;
1H NMR (acetone-d6) δ 1.90 (d, 3H, CH3-5, J ) 1.2 Hz), 2.39
(m, 1H, H-2′a), 2.50 (m, 1H, H-2′b), 2.65 (bs, 1H, 5′-OH), 3.40
(t, 2H, CH2Br, J ) 6.0 Hz), 3.80 (m, 5H, 2CH2O, H-4′), 3.92
(m, 2H, 2H-5′), 4.15 (t, 2H, NCH2, J ) 5.9 Hz), 4.37 (m, 1H,
H-3′), 6.01 (t, 1H, H-1′, J 1′,2′a ) J 1′,2′b ) 6.4 Hz), 7.34 (d, 1H,
H-6). Anal. (C14H20BrN5O5) C, H, N.
3′-Azid o-3-N-(4-b r om o-2-(E )b u t en yl)-3′-d eoxyt h ym i-
d in e (7b): reaction time 7 h; 2 equiv of 1,4-dibromo-2-butene
(E); eluent hexane/ethyl acetate, 1:1; yield of 7b (80%) as a
syrup; 1H NMR (acetone-d6) δ 1.91 (d, 3H, CH3-5, J ) 1.2 Hz),
2.31-2.62 (m, 3H, 2H-2′, 5′-OH), 3.75-4.03 (m, 5H, H-4′, 2H-
5′, CH2Br), 4.39 (m, 1H, H-3′), 4.51, 4.54 (2s, 2H, CH2N), 5.84
(m, 2H, CHdCH), 6.03 (t, 1H, H-1′, J 1′,2′ ) 6.5 Hz), 7.33 (d,
1H, H-6). Anal. (C14H18BrN5O4) C, H, N.
SSi2 requires 1094.3922).
Heter odim er [TSAO-T]N3-(CH2)3-N3[5′-MMP -dTh d] (5d).
TSAO-T and 3d reacted for 8 h. The residue was chromato-
graphed (dichloromethane/ethyl acetate, 1:2) to give 5d (81%)
as a syrup: HPLC Rt ) 12.68 min (system A, 55:45), Rt )
21.93 min (system B); MS m/e FAB 1112.4031 (MH+, C47H73
-
N6O17P SSi2 requires 1112.4028).
Heter od im er [TSAO-T]N3-(CH2)6-N3[d 4T] (6a ). TSAO-T
and 4a reacted for 3 h. The residue was chromatographed
(dichloromethane/methanol, 20:1) to give 6a (60%) as a white
foam: HPLC Rt ) 18.25 min (system A, 55:45), Rt ) 22.81
min (system B). Anal. (C40H65N5O12SSi2) C, H, N, S.
Heter od im er [TSAO-T]N3-(CH2)6-N3[5′-MMP -AZT] (6b).
TSAO-T and 4b reacted for 12 h. The residue was chromato-
graphed (dichloromethane/methanol, 20:1) to give 6b (75%) as
a white foam: HPLC Rt ) 18.20 min (system A, 60:40), Rt )
24.82 min (system B); MS m/e FAB 1179.4561 (MH+, C50H78
-
N9O16PSSi2 requires 1179.4562).
Heter od im er [TSAO-T]N3-(CH2)6-N3[5′-MMP -d 4T] (6c).
TSAO-T and 4c reacted for 9 h. The residue was chromato-
graphed (dichloromethane/acetone, 10:1) to give 6c (70%) as
a white foam: HPLC Rt ) 19.80 min (system A, 60:40), Rt )
24.17 min (system B); MS m/e FAB 1136.4399 (MH+, C50H77
-
3′-Azid o-3-N-(r-b r om o-p -xylen yl)-3′-d eoxyt h ym id in e
(7c): reaction time 4 h; 2 equiv of R,R′-bromo-p-xylene; eluent
dichloromethane/methanol, 20:1; yield of 7c (75%) as a syrup;
1H NMR (CDCl3) δ 1.83 (d, 3H, CH3-5, J ) 1.1 Hz), 2.35 (m,
3H, 2H-2′, 5′-OH), 3.66-3.92 (m, 3H, H-4′, 2H-5′), 4.30 (m,
1H, H-3′), 4.36 (s, 2H, CH2Br), 4.99 (s, 2H, CH2N), 5.97 (t, 1H,
H-1′, J 1′,2′a ) J 1′,2′b ) 6.5 Hz), 7.27 (m, 5H, H-6, Ph). Anal.
(C18H20BrN5O4) C, H, N.
3′-Azid o-3-N-(4-ch lor o-2-b u t yn yl)-3′-d eoxyt h ym id in e
(7d ): reaction time 30 h; 4 equiv of 1,4-dichloro-2-butyne;
eluent dichloromethane/methanol, 10:1; yield of 7d (35%) as
a syrup, recovered starting material (20%); 1H NMR δ 1.85 (s,
3H, CH3-5), 2.47 (m, 2H, 2H-2′), 3.82 (m, 2H, 2H-5′), 3.95 (m,
1H, H-4′), 4.28 (t, 2H, CH2Cl, J ) 1.9 Hz), 4.47 (m, 2H, H-3′,
5′-OH), 4.69 (d, 2H, CH2N, J ) 1.8 Hz), 6.24 (t, 1H, H-1′, J 1′,2′a
) J 1′,2′b ) 6.3 Hz), 7.85 (s, 1H, H-6). Anal. (C14H16ClN5O4) C,
H, N.
Gen er a l P r oced u r e for th e Syn th esis of th e Het-
er od im er s [TSAO-T]N3-sp a cer -N3[d 4T or AZT] 5a , 6a , a n d
8a -d a n d [TSAO-T]N3-sp a cer -N3[5′-MMP -(AZT, d 4T, or
d Th d )] 5b-d a n d 6b-d . To a solution of the key 3-N-(n-
bromo-spacer) nucleoside (3a , 4a , 7a -d ) or the 3-N-(n-bro-
moalkyl) nucleotide-5′-(phenylmethoxyalaninyl)phosphate (3b-
d , 4b-d ) (1 equiv) in dry acetonitrile were added K2CO3 (1.1
equiv) and TSAO-T (1.1 equiv). The reaction mixture was
N6O16PSSi2 requires 1136.4392).
Heter odim er [TSAO-T]N3-(CH2)6-N3[5′-MMP -dTh d] (6d).
TSAO-T and 4d reacted for 16 h. The residue was chromato-
graphed (dichloromethane/ethyl acetate, 1:2) to give 6d (50%)
as a syrup: HPLC Rt ) 23.07 min (system A, 55:45); MS m/e
FAB 1154.4492 (MH+, C50H79N6O17P SSi2 requires 1154.4498).
Heter odim er [TSAO-T]N3-(CH2)2-O-(CH2)2-N3[AZT] (8a).
TSAO-T and 7a reacted for 15 h. The residue was chromato-
graphed (dichloromethane/methanol, 20:1) to give 8a (75%) as
a white foam: HPLC Rt ) 13.96 min (system A, 55:45). Anal.
(C38H62N8O13SSi2) C, H, N, S.
Heter odim er [TSAO-T]N3-CH2-CHdCH(E)-CH2-N3[AZT]
(8b). TSAO-T and 7b reacted for 4 h. The residue was
chromatographed (hexane/ethyl acetate, 1:1) to give 8b (73%)
as a white foam: HPLC Rt ) 16.28 min (system A, 55:45).
Anal. (C38H60N8O12SSi2) C, H, N, S.
Heter od im er [TSAO-T]N3-CH2-P h -CH2-N3[AZT] (8c).
TSAO-T and 7c reacted for 3 h. The residue was chromato-
graphed (dichloromethane/ethyl acetate, 20:1) to give 8c (92%)
as a white foam: HPLC Rt ) 29.81 min (system A, 55:45).
Anal. (C42H62N8O12SSi2) C, H, N, S.
Heter od im er [TSAO-T]N3-CH2-CtC-CH2-N3[AZT] (8d ).
TSAO-T and 7d reacted for 30 h. The residue was chromato-
graphed (dichloromethane/methanol, 10:1) to give 8d (23%) as
a
white foam together with unreacted starting material

Multifunctional Inhibitors of HIV-1-RT
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 25 5195
(36%): HPLC Rt ) 15.72 min (system A, 55:45). Anal.
(C38H58N8O12SSi2) C, H, N, S.
virus-induced cytopathicity (CEM) or cell viability (MT-4) by
50% or to reduce by 50% the number of viable cells in mock-
infected cell cultures, respectively.
[1-[2′,5′-Bis-O-(ter t-bu tyld im eth ylsilyl)-â-D-r ibofu r a n o-
syl]-3-N -(3-b r om op r op a r gyl)t h ym in e ]-3′-sp ir o-5′′-(4′′-
a m in o-1′′,2′′-oxa th iole 2′′,2′′-d ioxid e) (9). TSAO-T was re-
acted with 2 equiv of propargyl bromide for 7 h, according to
the general procedure for the synthesis of 3-N-(bromo-spacer)
nucleoside and nucleotide intermediates (3a -d , 4a -d , and
7a -d ). The residue was purified by CCTLC on the chroma-
totron (hexane/ethyl acetate, 1:1) to give intermediate 9 (90%)
as a white foam: HPLC Rt ) 21.33 min (system A, 55:45); 1H
NMR (CDCl3) δ -0.10, 0.05, 0.19, 0.21 (4s, 12H, 4CH3), 0.82,
0.97 (2s, 18H, 2t-Bu), 2.00 (d, 3H, CH3-5, J ) 1.2 Hz), 2.14 (t,
1H, CtCH), 3.91 (m, 1H, H-5′a, J 5′a, 5′b ) 12.4, J 4′,5′a ) 2.6 Hz),
4.01 (m, 1H, H-5b′, J 4′,5′b ) 2.8 Hz), 4.34 (t, 1H, H-4′), 4.54 (d,
1H, H-2′, J 1′,2′ ) 7.9 Hz), 4.71 (m, 2H, CH2-CtCH), 5.61 (bs,
2H, NH2-4′′), 5.63 (s, 1H, H-3′′), 5.98 (d, 1H, H-1′), 7.23 (d,
1H, H-6). Anal. (C27H45BrN3O8SSi2) C, H, N, S.
Gen er a l P r oced u r e for th e Syn th esis of Heter od im er s
[TSAO-T]N3-CH2-CtC-C5[AZT, d 4T, or d Th d ] 11a -c. A
solution of the corresponding 5-iodonucleoside 10a -c (1 mmol)
in 3-5 mL of dry DMF was deoxygenated under argon. The
terminal alkyne 9 (2 mmol), (Ph3P)4Pd (0.1 mmol), CuI (0.2
mmol), and NEt3 (2 mmol) were added, and the mixture was
stirred at room temperature under argon atmosphere. After
24 h the solvent was removed in vacuo, and the residue was
dissolved in CH2Cl2 (50 mL), washed with 5% EDTA (2 × 25
mL) and saturated aqueous NaCl (25 mL), dried (Na2SO4), and
evaporated to dryness. The residue was purified by CCTLC
on the chromatotron. The apolar yellow impurities were first
eluted with hexane. Further elution with hexane/ethyl acetate
1:1 and then with ethyl acetate gave, from the faster moving
fractions, pure [TSAO-T]N3-CH2-CtC-CtC-CH2-N3[TSAO-T]
dimer 12 as a white foam. The slower moving fractions were
subjected to a second purification by CCTLC on the chroma-
totron (CH2Cl2/MeOH, 20:1, and then 10:1) to afford the
corresponding pure heterodimer 11a -c. The yield of the
isolated compounds and analytical data are indicated below
for each compound.
Heter odim er [TSAO-T]N3-CH2-CtC-C5[AZT] (11a): yield
of dimer 12 (12%); yield of 11a (55%) as a white foam.
Analytical data of 12: Anal. (C54H88N6O16S2Si4) C, H, N, S.
Analytical data of 11a : HPLC Rt ) 7.67 min (system A, 55:
45). Anal. (C36H54N8O12SSi2) C, H, N, S.
Heter od im er [TSAO-T]N3-CH2-CtC-C5[d 4T] (11b): yield
of dimer 12 (16%); yield of 11b (46%) as a white foam; HPLC
Rt ) 12.84 min (system A, 55:45). Anal. (C36H53N5O12SSi2) C,
H, N, S.
Heter odim er [TSAO-T]N3-CH2-CtC-C5[dTh d] (11c): yield
of dimer 12 (6%); yield of 11c (61%) as a white foam; HPLC
Rt ) 9.35 min (system A, 45:55). Anal. (C36H55N5O13SSi2) C,
H, N, S.
Heter odim er [TSAO-T]N3-CH2-CH2-CH2-C5[dTh d] (13c).
Compound 11c (0.125 g, 0.15 mmol) was dissolved in ethanol
(35 mL) and hydrogenated in the presence of 10% Pd/C (0.035
g) at 40 psi at 30 °C for 3 h. The mixture was filtered off and
evaporated to give 13c (0.106 g, 83%) as an amorphous solid.
Ack n ow led gm en t. We thank Ann Absillis and
Lizette van Berckelaer for excellent technical assistance.
This research was supported in part by grants from the
Spanish CICYT (Project SAF97-0048-C02-01), from
the Fonds voor Wetenschappelijk Onderzoek (Project
G.0104.98), the Belgian Geconcerteerde Onderzoeksac-
ties (Project 95/5), and the Biomedical Research Pro-
gramme of the European Commission (Project ERBBM-
H4CT972161).
Su p p or tin g In for m a tion Ava ila ble: 1H NMR chemical
shift assignments of heterodimers 5b,d , 6a -d , 8a -d , 11a -
c, and 13c and dimer 12. This material is available free of
charge via the Internet at http://pubs.acs.org.
Refer en ces
(1) Katz, R. A.; Skalka, A. M. The retroviral enzymes. Annu. Rev.
Biochem. 1994, 63, 133-173.
(2) Vaishnav, Y. N.; Wong-Staal, F. The biochemistry of AIDS.
Annu. Rev. Biochem. 1991, 60, 577-630.
(3) Schinazi, R. F. Competitive inhibitors of human immunodefi-
ciency virus reverse transcriptase. Perspect. Drug Discovery Des.
1993, 1, 151-180.
(4) For an overview, see: De Clercq, E. Toward improved anti-HIV
chemotherapy: therapeutic strategies for intervention with HIV
infections. J . Med. Chem. 1995, 38, 2491-2517.
(5) Young, S. D. Non-nucleoside inhibitors of HIV-1 reverse tran-
scriptase. Perspect. Drug Discovery Des. 1993, 1, 181-192.
(6) De Clercq, E. HIV-1 specific RT inhibitors: Highly selective
inhibitors of human immunodeficiency virus type 1 that are
specifically targeted at the viral reverse transcriptase. Med. Res.
Rev. 1993, 13, 229-258.
(7) (a) De Clercq, E. Non-nucleoside reverse transcriptase inhibitors
(NNRTIs) for the treatment of human immunodeficiency virus
type 1 (HIV-1) infections: strategies to overcome drug resistance
development. Med. Res. Rev. 1996, 16, 125-157. (b) De Clercq,
E. The role of non-nucleoside reverse transcriptase inhibitors
(NNRTIs) in the therapy of HIV-1 infection. Antiviral Res. 1998,
38, 153-179.
(8) (a) J acobo-Molina, A.; Ding, J .; Nanni, R. G.; Clark, A. D., J r.;
Lu, X.; Tantillo, C.; Williams, R. L.; Kamer, G.; Ferris, A. L.;
Clark, P.; Hizi, A.; Hughes, S. H.; Arnold, E. Crystal structure
of human immunodeficiency virus type 1 reverse transcriptase
complexed with double-stranded DNA at 3.0 Å resolution shows
bent DNA. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 6320-4. (b)
Tantillo, C.; Ding, J .; J acobo-Molina, A.; Nanni, R. G.; Boyer, P.
L.; Hughes, S. H.; Pauwels, R.; Andries, K.; J anssen, P. A. J .;
Arnold, E. Locations of anti-AIDS drug binding sites and
resistance mutations in the three-dimensional structure of HIV-1
reverse transcriptase. Implications for mechanisms of drug
inhibition and resistance. J . Mol. Biol. 1994, 243, 369-387.
(9) Nanni, R. G.; Ding, J .; J acobo-Molina, A.; Hughes, S. H.; Arnold,
E. Review of HIV-1 reverse transcriptase three-dimensional
structure: Implications for drug design. Perspect. Drug Discovery
Des. 1993, 1, 129-150.
(10) Smerdon, S. J .; J a¨ger, J .; Wang, J .; Kohlstaedt, L. A.; Chirino,
A. J .; Friedman, J . M.; Rice, P. A.; Steitz, T. A. Structure of the
binding site for non-nucleoside inhibitors of the reverse tran-
scriptase of human immunodeficiency virus type 1. Proc. Natl.
Acad. Sci. U.S.A. 1994, 91, 3911-3915.
HPLC Rt ) 2.95 min (system A, 55:45). Anal. (C36H59N5O13
SSi2) C, H, N, S.
-
An ti-HIV Eva lu a tion . Human immunodeficiency virus
type 1 [HIV-1 (IIIB)] was obtained from Dr. R. C. Gallo (when
at the National Cancer Institute, Bethesda, MD). HIV-2 (ROD)
was provided by Dr. L. Montagnier (when at the Pasteur
Institute, Paris, France). First, 4 × 105 CEM or 3 × 105 MT-4
cells per milliliter were infected with HIV-1 or HIV-2 at ∼100
CCID50 (50% cell culture infective dose) per milliliter of cell
suspension. Then 100 µL of the infected cell suspension was
transferred to microtiter plate wells and mixed with 100 µL
of the appropriate dilutions of the test compounds. After 4 days
giant cell formation (CEM) or HIV-induced cytopathicity (MT-
4) was recorded microscopically in the HIV-infected cell
cultures. The 50% effective concentration (EC50) and 50%
cytotoxic concentration (CC50) of the test compounds were
defined as the compound concentrations required to inhibit
(11) (a) Balzarini, J .; Pe´rez-Pe´rez, M. J .; San-Fe´lix, A.; Schols, D.;
Perno, C. F.; Vandamme, A. M.; Camarasa, M. J .; De Clercq, E.
2′,5′-Bis-O-(tert-butyldimethylsilyl)-3′-spiro-5′′-(4′′-amino-1′′,2′′-
oxathiole-2′′,2′′-dioxide)pyrimidine (TSAO) nucleoside ana-
logues: Highly selective inhibitors of human immunodeficiency
virus type 1 that are targeted at the viral reverse transcriptase.
Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 4392-4396. (b) Camarasa,
M. J .; Pe´rez-Pe´rez, M. J .; San-Fe´lix, A.; Balzarini, J .; De Clercq,
E. 3′-Spironucleosides (TSAO derivatives), a new class of specific
human immunodeficiency virus type 1 inhibitors: Synthesis and
antiviral activity of 3′-spiro-5′′-[4′′-amino-1′′,2′′-oxathiole-2′′,2′′-
dioxide]pyrimidine nucleosides. J . Med. Chem. 1992, 35, 2721-
2727.
(12) (a) Camarasa, M. J .; Pe´rez-Pe´rez, M. J .; Vela´zquez, S.; San-Fe´lix,
A.; Alvarez, A.; Ingate, S.; J imeno, M. L.; Karlsson, A.; De Clercq,
E.; Balzarini, J . TSAO derivatives: Highly specific inhibitors of
human immunodeficiency virus type-1 (HIV-1) replication.

5196 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 25
Vela´zquez et al.
Nucleosides Nucleotides 1995, 14, 585-594. (b) Balzarini, J .;
Camarasa, M. J .; Karlsson, A. TSAO derivatives: Highly specific
human immunodeficiency virus type 1 (HIV-1) reverse tran-
scriptase inhibitors. Drugs Future 1993, 18, 1043-1055.
(13) (a) J onckheere, H.; Taymans, J . M.; Balzarini, J .; Vela´zquez, S.;
Camarasa, M. J .; Desmyter, J .; De Clercq, E.; Anne´, J . Resis-
tance of HIV-1 reverse transcriptase against [2′,5′-bis-O-(tert-
butyldimethylsilyl)-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-
dioxide)] (TSAO) derivatives is determined by the mutation
Glu138 f Lys on the p51 subunit. J . Biol. Chem. 1994, 269,
25255-25258. (b) Harris, D.; Lee, R.; Misra, H. S.; Pandey, P.
K.; Pandey, V. N. The p51 subunit of human immunodeficiency
virus type 1 reverse transcriptase is essential in loading the p66
subunit on the template primer. Biochemistry 1998, 37, 5903-
5908.
(14) Balzarini, J .; Kleim, J . P.; Riess, G.; Camarasa, M. J .; De Clercq,
E.; Karlsson, A. Sensitivity of (138 GLU f LYS) mutated human
immunodeficiency virus type 1 (HIV-1) reverse transcriptase
(RT) to HIV-1-specific RT inhibitors. Biochem. Biophys. Res.
Commun. 1994, 201, 1305-1312.
(15) Boyer, P. L.; Ding, J .; Arnold, E.; Hughes, S. H. Subunit
specificity of mutations that confer resistance to non-nucleoside
inhibitors in human immunodeficiency virus type 1 reverse
transcriptase. Antimicrob. Agents Chemother. 1994, 38, 1909-
1914.
(29) (a) Heck, R. F. Arylation, methylation and carboxyalkylation of
olefins by group VIII metal derivatives. J . Am. Chem. Soc. 1968,
90, 5518-5526. (b) Heck, R. F. Palladium reagents in Organic
Synthesis; Academic Press: New York, 1985. (c) Heck, R. F.
Palladium-catalyzed reactions of organic halides. Acc. Chem. Res.
1979, 12, 146-151. (d) Heck, R. F. Palladium-catalyzed vinyla-
tion of organic halides. Org. React. 1982, 27, 345-390.
(30) Sonogashira, K.; Tohda, Y.; Hagihara; N. Convenient synthesis
of acetylenes catalytic substitutions of acetylenic hydrogen with
bromoalkenes, iodoarenes and bromopyridines. Tetrahedron Lett.
1975, 4467-4470.
(31) (a) Bergstrom, D. E.; Ruth, J . L. Synthesis of C-5 substituted
pyrimidine nucleosides via organopalladium intermediates. J .
Am. Chem. Soc. 1976, 98, 1587-1589. (b) Bergstrom, D. E.;
Ogawa, M. K. C-5 substituted pyrimidine nucleosides. 2. Syn-
thesis via olefin coupling to Organopalladium intermediates
derived from uridines and 2′-deoxyuridines. J . Am. Chem. Soc.
1978, 100, 8106-8112. (c) Bergstrom, D. E.; Beal, P.; J enson,
J .; Lin, X. Palladium mediated synthesis of C-5 pyrimidines
nucleoside thioethers from disulfides and mercurinucleosides.
J . Org. Chem. 1991, 56, 5598-5602. (d) Bergstrom, D. E.; Lin,
X.; Wang, G.; Rotsein, D.; Beal, P.; Norrix, K.; Ruth, J . L. Synlett.
1992, 179-188.
(16) De Clercq, E. HIV-1 resistance to reverse transcriptase inhibi-
tors. Biochem. Pharmacol. 1994, 47, 155-169.
(17) Schinazi, R. F.; Larder, B. A.; Mellors, J . W. Mutations in HIV-1
reverse transcriptase and protease associated with drug-
resistance. Int. Antiviral News 1996, 4, 95-107.
(32) Robins, M. J .; Barr, P. J . Nucleic Acid Related Compounds. 39.
Efficient conversion of 5-iodo to 5-alkynyl and derived 5-substi-
tuted uracil bases and nucleosides. J . Org. Chem. 1983, 48,
1854-1862.
(33) (a) Hobbs, F. W., J r. Palladium-catalyzed synthesis of alkynyl-
amino nucleosides. A universal linker for nucleic acids. J . Org.
Chem. 1989, 54, 3420-3422. (b) Robins, M. J .; Vinayak, R. S.;
Wood, S. G. Solvent not palladium oxidation state is the primary
determinant for successful coupling of terminal alkynes with
iodo-nucleosides. Tetrahedron Lett. 1990, 31, 3731-3734.
(34) (a) Kundu, N. G.; Dasgupta, S. K. Synthesis of 5-(acylethynyl)-
uracils and their corresponding 2′-deoxyribonucleosides through
Palladium-catalyzed reactions. J . Chem. Soc., Perkin Trans. 1
1993, 2657-2663. (b) Kundu, N. G.; Dasgupta, S. K.; Chaudhuri,
L. N.; Mahanty, J . S.; Spears, C. P.; Shahinian, A. H. Synthesis
and biological activities of (E)-5-(2-acylvinyl)uracils. Eur. J . Med.
Chem. 1993, 28, 473-479.
(35) Lin, T. S.; Gao, Y. S.; Mancini, W. R. Synthesis and biological
activity of various 3′-azido and 3′-amino analogues of 5-substi-
tuted pyrimidine deoxyribonucleosides. J . Med. Chem. 1983, 26,
1691-1696.
(36) Lin, T. S.; Mancini, W. R. Synthesis and antineoplastic activity
of 3′-azido and 3′-amino analogues of pyrimidine deoxyribo-
nucleosides. J . Med. Chem. 1983, 26, 544-548.
(37) Balagopala, M. I.; Ollapally, A. P.; Lee, H. H. An improved
synthesis of azidothymidine. Nucleosides Nucleotides 1996, 15,
899-906.
(38) J oshi, B. V.; Rao, T. S.; Reese, C. B. Conversion of some
pyrimidine 2′-deoxyribonucleosides into the corresponding 2′,3′-
didehydro-2′,3′-dideoxynucleosides. J . Chem. Soc., Perkin Trans.
1 1992, 2537-2544.
(39) Larsen, E.; Kofoed, T.; Pedersen, E. B. Synthesis of 2′,3′-anhydro-
2′-deoxyuridines and 2′,3′-didehydro-2′,3′-dideoxyuridines using
polymer supported fluoride. Synthesis 1995, 1121-1125.
(40) Robins, M. J .; Barr, P. J .; Giziewicz, J . Nucleic acid related
compounds. 38. Smooth and high-yield iodination and chlorina-
tion at C-5 of uracil bases and p-toluyl-protected nucleosides.
Can. J . Chem. 1982, 60, 554-557.
(18) Schmit, J . C.; Cogniaux, J .; Hermans, P.; Van Vaeck, C.;
Sprecher, S.; Van Remoortel, B.; Witvrouw, M.; Balzarini, J .;
Desmyter, J .; De Clercq, E.; Vandamme, A. M. Multiple drug
resistance to nucleoside analogues and non-nucleoside reverse
transcriptase inhibitors in an efficiently replicating human
inmunodeficiency virus type 1 patient strain. J . Infect. Dis. 1996,
174, 962-968.
(19) Vandamme, A. M.; Van Vaerenbergh, K.; De Clercq, E. Anti-
human inmunodeficiency virus drug combination strategies.
Antiviral Chem. Chemother. 1998, 9, 187-203.
(20) Spence, R. A.; Kati, W. M.; Anderson, K. S.; J ohnson, K. A.
Mechanism of inhibition of HIV-1 reverse transcriptase by non-
nucleoside inhibitors. Science 1995, 267, 988-993.
(21) Vela´zquez, S.; Alvarez, R.; San-Fe´lix, A.; J imeno, M. L.; De
Clercq, E.; Balzarini, J .; Camarasa, M. J . Synthesis and Anti-
HIV Activity of [AZT]-[TSAO-T] and [AZT-[HEPT] Dimers as
potential multifunctional inhibitors of HIV-1 reverse tran-
scriptase. J . Med. Chem. 1995, 38, 1641-1649.
(22) For
a recent paper, see: Renoud-Grappin, M.; Fossey, C.;
Fontaine, G.; Ladure´e, D.; Aubertin, A. M.; Kirn, A. Imidazo-
[1,5-b]pyridazine-d4T conjugates: Synthesis and anti-human
immunodeficiency virus evaluation. Antiviral Chem. Chemother.
1998, 9, 205-223.
(23) A preliminary account of this work was presented at the XIII
International Round Table Conference on Nucleosides, Nucle-
otides and their biological Applications. Montpellier, France.
September 6-10, 1998. Poster 155. Nucleosides Nucleotides
1999, 18, 1029-1030.
(24) McGuigan, C.; Devine, K. G.; O’Connor, T. J .; Galpin, S. A.,
J effries, D. J .; Kinchington, D. Synthesis and evaluation of some
novel phosphoramidate derivatives of 3′-azido-3′-deoxythimidine
(AZT) as anti-HIV compounds. Antiviral Chem. Chemother.
1990, 1, 107-113.
(25) McGuigan, C.; Pathirana, R. N.; Balzarini, J .; De Clercq, E.
Intracellular delivery of bioactive AZT nucleotides by aryl
phosphate derivatives of AZT. J . Med. Chem. 1993, 36, 1048-
1052.
(26) McGuigan, C.; Cahard, D.; Sheeka, H. M.; De Clercq, E.;
Balzarini, J . Arylphosphoramidate derivatives of d4T have
improved anti-HIV efficacy in tissue culture and may act by the
generation of a novel intracellular metabolite. J . Med. Chem.
1996, 39, 1748-1753.
(27) Pe´rez-Pe´rez, M. J .; San-Fe´lix, A.; Balzarini, J .; De Clercq, E.;
Camarasa, M. J . TSAO analogues. Stereospecific synthesis and
anti-HIV-1 activity of 1-[2′,5′-bis-O-(tert-butyldimethylsilyl-â-D-
ribofuranosyl]-3′-spiro-5′′-[4′′-amino-1′′,2′′-oxathiole-2′′,2′′-diox-
ide]pyrimidine and pyrimidine modified nucleosides. J . Med.
Chem. 1992, 35, 2988-2995.
(41) For reasons of simplicity we have adopted this nomenclature
for the whole set of heterodimers synthesized. A more correct
name, for example, of the heterodimer [TSAO-T]N3-(CH2)3-N3-
[d4T] (5a ), chosen as a model, would be [1-[2′,5′-Bis-O-(tert-
butyldimethylsilyl)-â-D-ribofuranosyl]-3-N-[(2′′′′,3′′′′-didehydro-
2′′′′,3′′′′′-dideoxythymidin-3′′′-yl)propyl]thymine]-3′-spiro-5′′-(4′′-
amino-1′′,2′′-oxathiole 2′′,2′′-dioxide). Although the oxathiole ring
has priority over the TSAO nucleoside system, double primes
have been used in the numbering of the oxathiole ring in order
to keep the same numbering system as in the previous TSAO
papers. Triple and quadruple primes have been used for the
second nucleoside system chosen as radical.
(28) McGuigan, C.; Pathirana, R. N.; Mahmood, N.; Devine, K. G.;
Hay, A. J . Aryl phosphate Derivatives of AZT retain activity
against HIV1 in cell lines which are resistant to the action of
AZT. Antiviral Res. 1992, 17, 311-321.
J M991092+