New AZT conjugates as potent anti-HIV agents

Article abstract of DOI:10.1080/15257770500377789

In an attempt to discover anti-HIV agents with much reduced cytotoxicity from the currently available HIV-reverse transcriptase inhibitors, AZT conjugates of cholanic acids, 2-imidazolidone-4-carboxylic acid and its derivatives, and N,N′-disubstituted 5-hydroxy-tetrahydropyrimidin-2-ones have been synthesized and their anti-HIV profiles determined with CEM-SS cell line. The AZT conjugates with 2-imidazolidone-4-carboxylic acid and 2-pyrrolidone-5-carboxylic acid through an ester linkage, and with N,N′-diphenyl-5-hydroxy-tetrahydropyrimidin-2-one through a succinate tether showed significantly higher therapeutic indexes than AZT while they also retained or enhanced AZT's anti-HIV activity. Thus, structural features that favor the desired therapeutic profile of the conjugates appear to include a five-membered ring cyclic urea or lactam, and six-membered ring cyclic urea with N,N′-diphenyl substitution. Copyright Taylor & Francis Group, LLC.

Full text of DOI:10.1080/15257770500377789

This article was downloaded by: [UQ Library]
On: 15 October 2014, At: 10:37
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Nucleosides, Nucleotides and Nucleic
Acids
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/lncn20
New Azt Conjugates as Potent Anti-HIV
Agents
Zhengqing You a & Henry Joung Lee a
a College of Pharmacy and Pharmaceutical Sciences, Florida A&M
University , Tallahassee, Florida, USA
Published online: 20 Aug 2006.
To cite this article: Zhengqing You & Henry Joung Lee (2006) New Azt Conjugates as Potent Anti-HIV
Agents, Nucleosides, Nucleotides and Nucleic Acids, 25:1, 37-54
To link to this article: http://dx.doi.org/10.1080/15257770500377789
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions

Nucleosides, Nucleotides, and Nucleic Acids, 25:37–54, 2006
Copyright ꢀ Taylor & Francis Group, LLC
C
ISSN: 1525-7770 print / 1532-2335 online
DOI: 10.1080/15257770500377789
NEW AZT CONJUGATES AS POTENT ANTI-HIV AGENTS
Zhengqing You and Henry Joung Lee
College of Pharmacy and Pharmaceutical
2
Sciences, Florida A&M University, Tallahassee, Florida, USA
In an attempt to discover anti-HIV agents with much reduced cytotoxicity from the currently
2
available HIV-reverse transcriptase inhibitors, AZT conjugates of cholanic acids, 2-imidazolidone-
4-carboxylic acid and its derivatives, and N,N ꢁ-disubstituted 5-hydroxy-tetrahydropyrimidin-2-ones
have been synthesized and their anti-HIV profiles determined with CEM-SS cell line. The AZT
conjugates with 2-imidazolidone-4-carboxylic acid and 2-pyrrolidone-5-carboxylic acid through an
ester linkage, and with N,N ꢁ-diphenyl-5-hydroxy-tetrahydropyrimidin-2-one through a succinate
tether showed significantly higher therapeutic indexes than AZT while they also retained or enhanced
AZT’s anti-HIV activity. Thus, structural features that favor the desired therapeutic profile of the
conjugates appear to include a five-membered ring cyclic urea or lactam, and six-membered ring cyclic
urea with N,N ꢁ-diphenyl substitution.
Keywords AZT; Prodrug; Steroid; 2-Imidazolidone-4-carboxylic acid; 2-Pyrrolidone-5-
carboxylic acid; Tetrahydropyrimidinone; Anti-HIV
INTRODUCTION
Azidothymidine (AZT), the first clinically approved agent for treatment
of acquired immunodeficiency syndrome (AIDS), acts by inhibiting the re-
verse transcriptase of human immunodeficiency virus (HIV),[1−3] Although
AZT has attained a mainstay status in the treatment of AIDS, its serious
dose-related side effects have necessitated ever growing research to find
safer and more effective analogs,[4] One promising approach to improve
the therapeutic index has been the prodrug strategy. This approach relies
mostly on derivatization at the 5ꢁ-O of AZT. The derivative is cleaved by
metabolism to gradually release AZT. Since AZT has a short plasma half-
life due in part to glucuronidation in the liver at the 5ꢁ-OH[5] and has a
poor cellular penetrating capability,[6] the derivatization could increase drug
Received 19 January 2005; accepted 1 July 2005.
This research has been supported by NIH grants AM 21627, RR 08111, RR 02660 and AI 20360.
The National Cancer Institute (NCI) has performed all of the bio-assay experiments, and we would like
to especially thank Dr. Shizuko Sei at the NCI for helpful discussions and suggestions.
Address correspondence to Henry Joung Lee, College of Pharmacy and Pharmaceutical Sciences,
Florida A&M University, Tallahassee, FL, USA. E-mail: henry.lee@famu.edu
37

38
Z. You and H. J. Lee
stability against metabolic destruction and improve its cellular bioavailabil-
ity, thus reducing dose-related toxicity. One important class of AZT prodrugs
has been carboxylic esters derived from acids such as bicyclams attached to
an acid group,[7] retinoic acid,[8] 1,4-dihydronicotinic acid,[8] and steroid
acids.[9] Another class is phosphoesters, some of which are derived from sal-
icyl alcohols,[10] ether lipids,[11] or lipophilic glycosides.[12] More recently, a
new class of AZT prodrugs, 5ꢁ-O-carbonates of AZT, has been presented.[13]
In our design of new steroid-AZT conjugates, four rationales were employed:
1) Biologically inactive steroid acids obtained on the basis of the antedrug
concept[14] are devoid of systemic toxicity. 2) The acids are designed to im-
prove lipophilicity and bioavailability of the resulting prodrugs. 3) Protection
of the 5ꢁ-OH could increase drug half-life in vivo. 4) Binding of steroid conju-
gates to transcortin could further improve metabolic stability.[15] In our pre-
vious study,[16] three AZT conjugates 4–6 from steroid acids 1–3 (Figure 1)
were shown to have either no anti-HIV activity (conjugate 4), or reduced
activity (conjugates 5 and 6). The inactive conjugate, however, showed very
strong anti-growth activity against a number of cancer cell lines. The un-
usual profile of the conjugate 4 prompted us to study AZT conjugates from a
few more cholanic acid derivatives. Steroid acids 7–9 (Figure 1) were there-
fore used to conjugate with AZT to form prodrugs 10–12 (Figure 1) for
comparison.
In another front of our prodrug study, we proposed to conjugate AZT
with certain HIV-protease inhibitors. This is of the same idea as the “double
drugs,” a new class of prodrug form of an HIV protease inhibitor conjugated
with a reverse transcriptase inhibitor by a spontaneously cleavable linker.[17]
Not only could such conjugates achieve some of the envisaged advantages of
FIGURE 1

New AZT Conjugates
39
the AZT-steroid acid conjugates, they could also deliver a synergistic anti-HIV
effect by having simultaneous attacks at two different stages of the viral life
cycle. Of special interest to us was the cyclic urea HIV-protease inhibitor 13
(Figure 2), which gave very potent anti-HIV activity.[18] As an alcohol,
compound 13 can not directly form an ester with AZT. But with a tether
of succinate, it could form a diester conjugate with AZT. Succinic acid,
which is released upon hydrolytic cleavage of the diester prodrug, is not
expected to introduce serious toxic side effects since it is an endogenous
compound in the body. Thus a conjugate between AZT and the cyclic urea
appeared to be an interesting target. But before the acquisition of the
cyclic urea 13 which requires a long synthetic procedure, a model study
to conjugate AZT with cyclic urea acid 14, and cyclic urea alcohols 15 and
16 (Figure 2) was undertaken to establish the conjugation methodology.
Acid 14 is commercially available, while alcohols 15 and 16 were thought
to be accessible through simple synthetic preparations. Two of the model
conjugates were made, subjected to anti-HIV bioassay, and gave some sur-
prising results. While AZT-16 conjugate through a succinate tether showed
lower anti-HIV activity, and a slightly better therapeutic index than AZT,
AZT-14 ester conjugate gave higher anti-HIV activity, and much higher
therapeutic index than AZT. The better than expected results in the lat-
ter conjugate prompted us to investigate AZT conjugates with derivatives
of cyclic urea acid 14, and new targets were designed from acids 17–22
(Figure 3). This study describes both the synthesis and anti-HIV bioassay
FIGURE 2

40
Z. You and H. J. Lee
FIGURE 3
results of these new AZT conjugates. Another AZT conjugate, which was
prepared from N,Nꢁ-diphenyl-5-hydroxytetrahydropyrimidin-2-one during
our attempts to synthesize the tetrahydropyrimidin-2-ones, is also discussed
here.
RESULTS AND DISCUSSIONS
Chemistry
A standard procedure for ester formation between AZT and an acid was
used in the preparation of the conjugates from the three cholanic acids 7–9
and the conjugate from cyclopentane-carboxylic acid 17. Good yields were
obtained (73–89%) for these four targets.
Preparation of AZT/N,Nꢁ-dibenzyltetrahydropyrimidin-2-one conjugate
30 is shown in Scheme 1. The first intermediate 15 was obtained by a
procedure modified from the literature.[19] Instead of a sealed tube reaction
at higher temperatures as reported, the two reagents (ethyl carbonate
and 1,3-diamino-2-propanol) were simply heated to reflux under a reflux
condenser. The yield of our reaction (39%), however, was lower than
the literature result (75%). The second intermediate N,Nꢁ-dibenzyl-5-
O-benzyltetrahydropyrimidin-2-one 29 (Scheme 1) was designed so that
a selective O-benzyl cleavage would yield the desired N,Nꢁ-dibenzyl-5-
hydroxytetrahydropyrimidinone 16. While tribenzylation of 15 proceeded
smoothly with benzyl chloride and sodium hydride to give 29, a selective
debenzylation at the O turned out to be more difficult. Initial attempts to
selectively remove the benzyl group with catalytic hydrogenation (Pd/C)

New AZT Conjugates
41
SCHEME 1
failed. What was obtained as a result appeared to be N-benzyl-5-hydroxy-
[20]
tetrahydropyrimidin-2-one. Thus a literature procedure with SnCl4
was
tried, and it did selectively cleave O-benzyl to give the desired alcohol
16, though very slowly. Conjugation of AZT with 16 went without much
complication. As shown in the last two steps of Scheme 1, succinic anhydride
was condensed with 16 to form an acid tether, which was then condensed
with AZT in the presence of DCC to yield the diester conjugate 30 (45%).
Although successful for the N,Nꢁ-dibenzyl derivative, the procedure of
Scheme 1 could not produce the corresponding N,Nꢁ-dialkyl derivatives,
mainly because of difficulties in cleavage of the O-alkyl bond. Therefore,
an alternative procedure was designed to obtain other N,Nꢁ-disubstituted
derivatives.
Scheme 2 gives another route to access AZT/N,Nꢁ-disubstituted tetrahyd-
ropyrimidinone conjugates. It was envisaged that 3-chloro-2-chloromethyl-
propene should condense, in the presence of a base, with N,Nꢁ-disubstituted
urea to form a six-membered ring with a methylene substitution at the C-5.
When N,Nꢁ-dimethyl or N,Nꢁ-dibutyl urea was used, unfortunately, very low
yield of the desired product was detected, and many side products made
purification difficult. Carbanilide, however, turned out to be quite different.
The diphenyl urea condensed with 3-chloro-2-chloromethylpropene in the
presence of NaH to give N,Nꢁ-diphenyl-5-methylenetetrahydropyrimidin-2-
one 31 in good yield (83%) without difficulty in isolation. The next interme-
diate would be alcohol 32, which could have been most efficiently obtained
by ozonolysis (O3) and reduction (NaBH4). However, due to a lack of the
ozonolysis apparatus in our laboratory, a longer procedure was used to con-
vert 31 to 32 (Scheme 2). Potassium permanganate treatment at low temper-
atures turned 31 to a diol, which was oxidized by periodic acid to the ketone,

42
Z. You and H. J. Lee
SCHEME 2
and then reduced to alcohol 32. Conjugation of AZT with 32 proceeded to
form 33 just as smoothly as in the synthesis of conjugate 30.
Scheme 3–5 each provide a three step procedure to convert either 2-
imidazolidone-4-carboxylic acid (racemic) or 2-pyrrolidone-5-carboxylic acid
(racemic) to their corresponding AZT conjugates with N,Nꢁ-dimethylation,
N-methylation, or N-butylation. N,Nꢁ-Dimethylation and N-methylation
(Schemes 3 and 4) were achieved by treating the imidazolidone and pyrroli-
done with methyl iodide and potassium hydroxide in refluxing acetone.
N-Butylation (Scheme 5), on the other hand, was achieved by treatment
of the pyrrolidone with butyl bromide and potassium hydroxide in DMSO.
Recovery of the acid group was through a treatment with aqueous sodium
SCHEME 3

New AZT Conjugates
43
SCHEME 4
hydroxide, while ester formation between the resulting acid and AZT pro-
ceeded under the standard conditions to yield conjugates 28, 26, and 27.
Synthesis of AZT conjugates with 2-imidazolidone-4-carboxylic acid
(racemic), 2-pyrrolidone-5-carboxylic acid (racemic), and 4-imidazolecar-
boxylic acid required a slight modification of the standard procedure, mainly
in the purification. This was necessary since the resulting products were
significantly more polar than AZT, while all of the other conjugates men-
tioned above were less polar than AZT. The polarity issue necessitated the
use of methanol in the column separation of conjugates 34, 25, and 24,
while all the other conjugates were separated with either benzene/acetone
or EtOAc/hexanes system.
Biology
The ester conjugates were subjected to in vitro anti-HIV bioassays with
CD4-expressing lymphocytes (CEM-SS cell line). Drug efficacy (EC50) was
obtained by adding HIV and the conjugate to the cell to measure protection
SCHEME 5

44
Z. You and H. J. Lee
TABLE 1 Anti-HIV Activity and Cytotoxicity in CEM-SS Cell
Agent
EC50 (M)
IC50 (M)
TI50 (IC50/EC50)
10
11
12
30
33
34
28
25
26
27
23
24
32
14
19
AZT
4.2 × 10−8
8.9 × 10−8
3.1 × 10−7
7.4 × 10−8
1.6 × 10−8
7.6 × 10−9
1.0 × 10−7
1.8 × 10−8
8.1 × 10−8
8.5 × 10−8
4.0 × 10−8
3.1 × 10−7
Inactive
3.7 ×10−5
>1.0 × 10−4
>1.0 × 10−4
9.9 × 10−5
8.8 × 102
>1.1 × 103
>3.2 × 102
1.3 × 103
8.9 × 10−4
5.6 × 104
>2.0 × 10−4
>1.0 × 10−4
>2.0 × 10−4
>1.0 × 10−4
>1.0 × 10−4
>1.0 × 10−4
>1.0 × 10−4
>2.6 × 104
>1.0 × 103
>1.1 × 104
>1.2 × 103
>1.2 × 103
>2.5 × 103
>3.2 × 102
Inactive
Inactive
1.8 × 10−8a
1.8 × 10−5c
1.0 × 103b
aEC50 data of AZT obtained by the NCI at the same time as the other
agents.
bTI50 data of AZT cited from an earlier NCI paper. See reference 21.
cIC50 value derived from the EC50 and TI50 values given. For a ratio-
nale, see reference 22.
of the drug against HIV-induced cell death. Cytotoxicity (IC50) data was ob-
tained from adding only the conjugate to the cell to measure cell death
caused by the agent. Therapeutic index (TI) is the ratio IC50/EC50. As sum-
marized in Table 1, the three steroid acid conjugates 10–12 all showed anti-
HIV activity, which is different from conjugate 4, which gave no activity in
our previous study.[16] Also worthy of notice is the decreased cytotoxicity of
these three conjugates compared with conjugate 4. They gave IC50 values of
3.7 × 10−5, >1.0 × 10−4, and >1.0 × 10−4, while conjugate 4 showed a value
of 4.7 × 10−6. When a comparison is made with AZT, it is clear that these
conjugates (10–12) were weaker anti-HIV agents than AZT (EC50 values of
4.2 × 10−8, 8.9 × 10−8, and 3.1 × 10−7 vs. 1.8 × 10−8). However, 11 had a
somewhat higher TI value than AZT (IC50 value of >1.1 × 103 vs. 1.0 × 103).
The two AZT-tetrahydropyrimidinone conjugates 30 and 33 gave quite
different results. While the dibenzyl derivative 30 showed significantly lower
anti-HIV activity than AZT (EC50 of 7.4 × 10−8 vs. 1.8 × 10−8) and similar
TI value to that of AZT (1.3 × 103 vs. 1.0 × 103), the diphenyl derivative 33
gave similar anti-HIV activity to AZT, but significantly higher TI value (5.6 ×
104 vs. 1.0 × 103), and thus significantly lower cytotoxicity than AZT.
The AZT conjugate of 2-imidazolidone-4-carboxylic acid 34 gave very
strong anti-HIV activity and significantly higher TI than AZT. This acid has a
five-membered ring with a urea group in the ring. The derivatives (Scheme 3)
designed to compare with this acid all have a five-membered ring, with either
no functional group in the ring (17), two nitrogens just as the urea but no

New AZT Conjugates
45
carbonyl (18), a lactam group (19), or the lactam/urea with their nitrogens
alkylated (20–22). Table 1 shows that except for conjugate 25, which was
derived from acid 19, all the AZT conjugates from these derivatives (17–22)
gave significantly lower anti-HIV activity. Conjugate 25 exhibited the same
EC50 value as AZT, but much higher TI and thus IC50. It appears that a non-
substituted amide or urea group in the five-membered ring is necessary for
potent anti-HIV activity.
The question of whether the acid component or the other alcohol teth-
ered to AZT by a succinate group contributed to the anti-HIV activity of
the three potent conjugates 34, 25, and 33 has been answered by the data
for these components 14, 19, and 32 as shown in the table. These three
compounds all failed to give any anti-HIV activity, thus they could not have
contributed to the activity after the breakage of the conjugates. The active
compound after the breakage must be AZT, and only AZT.
EXPERIMENTAL/CHEMISTRY
The steroid acids 7–9 were purchased from Steraloids, Newport, RI. 2-
Pyrrolidone-5-carboxylic acid and solvents were bought from Fisher Scien-
tific, Fair Lawn, NJ. All the other reagents were ordered from Aldrich Chem-
ical, Milwaukee, WI. NMR spectra were obtained with a Bruker HX-300 spec-
trometer and the chemical shifts reported in parts per million (ppm) down
field from tetramethylsilane as an internal standard. Keys: s (singlet), d (dou-
blet), t (triplet), q (quartet). Elemental analysis was carried out by Galbraith
Laboratories, Knoxville, TN. Melting points were determined on a Thomas
Hoover Capillary Melting Point Apparatus and are uncorrected.
A standard procedure for the preparation of 10–12 and 23 is described
as follows. A mixture of the acid (1.6 mmol), AZT (0.82 mmol), DCC
(1.6 mmol), and DMAP (0.06 mmol) in dry THF (5 mL) was stirred un-
der a rubber septum for 3 days. The mixture was then filtered, and the solid
thoroughly washed with EtOAc. The combined organic solution was con-
densed, and the resulting residue chromatographed (3:1 benzene/acetone)
to yield the major product.
5β-Cholanic Acid, 3 ꢀ-Azido-3 ꢀ-deoxythymidin-5 ꢀ-yl Ester (10).
White
solid (73%). 1H NMR (CDCl3, 270 MHz) δ: 8.92 (1H, br. s, NH), 7.23 (1H, s,
H-6ꢁꢁ), 6.12 (1H, t, J = 6.3 Hz, H-1ꢁ), 4.34 (2H, d of AB-quartet, ꢀν = 24.0 Hz,
JAB = 12.2 Hz, JA = 4.4 Hz, JB = 3.9 Hz, H-5ꢁ), 4.18 (1H, m), 4.09 (1H, m), 1.94
(3H, s, Me of thymidine), 0.92 (3H, d, J = 6 Hz, Me of steroid), 0.91 (3H, s, Me
of steroid), 0.64 (3H, s, Me of steroid). 13C NMR (CDCl3, 68 MHz): 173.51,
163.34, 149.96, 135.11, 111.25, 85.62, 81.95, 63.25, 60.84, 56.71, 56.04, 43.82,
42.87, 40.64, 40.37, 37.68, 36.00, 35.42, 35.36, 31.14, 31.02, 28.24, 27.57,
27.29, 27.11, 26.62, 24.27, 21.40, 20.91, 18.34, 12.60, 12.11. Analysis Calcd
for C34H51N5O5: C66.97, H8.43, N11.49. Found: C66.83, H8.51, N11.38.

46
Z. You and H. J. Lee
5β-Cholanic Acid-3,7,12-trione, 3 ꢁ-Azido-3 ꢁ-deoxythymidin-5 ꢁ-yl Ester
1
(12).
White solid (78%). H NMR (CDCl3, 270 MHz) δ: 9.09 (1H, br.
s, NH), 7.23 (1H, d, J = 1.0 Hz, H-6ꢁꢁ), 6.09 (1H, t, J = 6.3 Hz, H-1ꢁ), 4.34 (2H,
d of AB-quartet, ꢀν = 25.4 Hz, JAB = 12.2 Hz, JA = 4.4 Hz, JB = 3.9 Hz, H-5ꢁ),
4.20 (1H, m), 4.08 (1H, m), 1.93 (3H, s, Me of thymidine), 1.39 (3H, s, Me of
steroid), 1.06 (3H, s, Me of steroid), 0.85 (3H, d, J = 6.8 Hz, Me of steroid).
13C NMR (CDCl3, 68 MHz): 211.73, 208.80, 208.44, 173.33, 163.43, 149.96,
135.26, 111.22, 85.71, 81.86, 63.22, 60.72, 56.90, 51.79, 48.98, 46.81, 45.56,
44.95, 42.78, 38.60, 37.59, 36.43, 36.00, 35.45, 35.30, 31.17, 30.29, 27.60,
25.09, 21.88, 18.61, 12.57, 11.80. Analysis Calcd for C34H45N5O8: C62.66,
H6.96, N10.75. Found: C62.72, H6.87, N10.63.
3α-Acetyloxy-5β-cholanic Acid, 3 ꢁ-Azido-3 ꢁ-deoxythymidin-5 ꢁ-yl Ester
1
(11).
White solid (89%). H NMR (CDCl3, 270 MHz) δ: 9.34 (1H, br. s,
NH), 7.23 (1H, s, H-6ꢁꢁ), 6.11 (1H, t, J = 6.3 Hz, H-1ꢁ), 4.70 (1H, m, H-3), 4.33
(2H, d of AB-quartet, ꢀν = 24.4 Hz, JAB = 12.2 Hz, JA = 4.4 Hz, JB = 3.4 Hz,
H-5ꢁ), 4.17 (1H, m), 4.08 (1H, m), 2.01 (3H, s, Ac), 1.93 (3H, s, Me of thymi-
dine), 0.91 (3H, s, Me of steroid), 0.90 (3H, d, J = 6.5 Hz, Me of steroid),
0.63 (3H, s, Me of steroid). Analysis Calcld for C36H53N5O7: C64.74, H8.00,
N10.49. Found: C64.58, H8.21, N10.31.
Cyclopentanecarboxylic Acid, 3 ꢁ-Azido-3 ꢁ-deoxythymidin-5 ꢁ-yl Ester (23).
Wax-like solid (83%). 1H NMR (CDCl3, 300 MHz) δ: 9.820 (1H, br. s, NH),
7.206 (1H, s, H-6ꢁꢁ), 6.103 (1H, t, J = 6.3 Hz, H-1ꢁ), 4.299 (2H, d of AB-quartet,
ꢀν = 26.8 Hz, JAB = 12.1 Hz, JA = 4.3 Hz, JB = 3.6 Hz, H-5ꢁ), 4.160 (1H, m),
4.050 (1H, m), 2.727 (1H, m), 2.450 (1H, m), 2.277 (1H, m), 1.879 (3H, s,
Me), 1.46–1.95 (8H, m, cyclopentane). 13C NMR (CDCl3, 75 MHz): 176.07,
163.92, 150.24, 135.07, 111.21, 85.26, 81.77, 63.19, 60.64, 43.56, 37.53, 30.00,
25.66, 12.47. Analysis Calcd for C16H21N5O5: C52.89, H5.83, N19.27. Found:
C52.95, H5.93, N19.04.
5-Hydroxytetrahydropyrimidin-2-one (15).
This compound has been
reported in the literature,[19] based on which we have developed a mod-
ified synthetic procedure. Thus, a mixture of 1,3-diamino-2-propanol (24
g, 266 mmol) and ethyl carbonate (17 g, 144 mmol) was heated to reflux
under a reflux condenser for 8 h. More ethyl carbonate (14 g, 115 mmol)
was added, and reflux continued for 2 days. After being cooled to room
temperature, the reaction mixture was treated with methanol/acetone (2:1,
25 mL) to induce crystal formation. Filtration separated the solid product,
which was washed with methanol/acetone (1:1) to yield 8.5 g of the de-
sired compound. The mother liquor was condensed, and then treated with
methanol/acetone (1:1) to give more white solid (3.5 g). A total of 12 g of
5-hydroxy-tetrahydropyrimidin-2-one was isolated (39%).

New AZT Conjugates
47
N,N ꢁ-Dibenzyl-5-O-benzyltetrahydropyrimidin-2-one (29).
A mixture
of 5-hydroxy-tetrahydropyrimidin-2-one (3 g, 25 mmol), benzyl chloride (15
g, 120 mmol) and sodium hydride (4.8 g, 60%, 120 mmol) in anhydrous
DMF (35 mL) was heated to 60◦C for 2 days. The mixture was then poured
into 3% aqueous HCl (550 mL) and extracted with ethyl acetate/hexanes
(1:1, 2 × 250 mL). The combined organic solution was washed with water (80
mL) and then dried over sodium sulfate. Filtration and condensation gave
a brown oil, which slowly yielded crystals upon standing. Removal of the liq-
uid and washing with ethyl acetate/petroleum ether yielded large crystals
(3.44 g). The mother liquor was chromatographed (1:1 EtOAc/hexanes)
to give more desired product (3.25 g). A total of 6.69 g of N,Nꢁ-dibenzyl-5-
1
O-benzyltetrahydropyrimidin-2-one (67%) was obtained. M.P. 76–77◦C. H
NMR (CDCl3, 270 MHz) δ: 7.15–7.35 (15H, m, aromatic), 4.60 (4 H, AB-
quartet, CH2Ph), 4.314 (2H, s, OCH2), 3.74 (1H, m, H-5), 3.19–3.37 (4H, m,
H-4 & H-6). Analysis Calcd for C25H26N2O2: C77.69, H6.78, N7.25. Found:
C78.05, H6.98, N7.15.
N,N ꢁ-Dibenzyl-5-hydroxytetrahydropyrimidin-2-one (16).
To a solu-
tion of N,Nꢁ-dibenzyl-5-O-benzyltetrahydropyrimidin-2-one (2.2 g, 5.7 mmol)
in dichloromethane (70 mL) was added, via a syringe, SnCl4 (12 g, 45 mmol).
The mixture was stirred under a nitrogen atmosphere for 2 weeks, and
then quenched with water (70 mL). The resulting mixture was extracted
with dichloromethane (2 × 100 mL), and the combined extracts dried over
sodium sulfate. Rotary evaporation of the dried solution and flash chro-
matography (1:1 and 2:1 ethyl acetate/hexanes) isolated N,Nꢁ-dibenzyl-5-
hydroxytetrahydropyrimidin-2-one as a white solid (1.15 g, 68%) after vac-
1
uum drying. H NMR (CDCl3, 270 MHz) δ: 7.25–7.37 (10H, m, aromatic),
4.61 (4H, AB-quartet, NCH2Ph), 4.02 (1H, m, H-5), 3.35 (2H, dd, J = 11.7, 3.4
Hz, H-4 & H-6), 3.15 (2H, dd, J = 11.7, 3.9 Hz, H-4 & H-6). 13C NMR (CDCl3,
75 MHz): 155.76, 137.97, 128.61, 127.99, 127.34, 61.86, 51.64, 51.53. Analysis
Calcd for C18H20N2O2: C72.95, H6.80, N9.45. Found: C73.25, H6.97, N9.19.
N,N ꢁ -Dibenzyltetrahydropyrimidin-2-on-5-yl 3 ꢁ-Diazo-3 ꢁ-deoxythymidin-
5 ꢁ4-yl Succinate (30).
The 5-hydroxypyrimidinone (16) (200 mg, 0.67
mmol), succinic anhydride (130 mg, 1.3 mmol), and DMAP (100 mg, 0.8
mmol) were stirred in dry DMF (2.5 mL) under nitrogen for 7.5 h. DCC
(270 mg, 1.3 mmol) was then added, followed by addition of AZT (290 mg,
1.1 mmol) 10 min later. Stirring continued for another 11 h and the mix-
ture was poured in aqueous HCl (130 mL, 0.3%). The aqueous suspension
was extracted with 1:1 ethyl acetate/hexanes (400 mL), and the organic
layer washed with water (50 mL). Rotary evaporation and flash chromatog-
raphy (4:1 EtOAc/hexanes and EtOAc) isolated the succinate (195 mg,
45%) as a white solid after vacuum evaporation. 1H NMR (CDCl3, 270 MHz)

48
Z. You and H. J. Lee
δ: 8.42 (1H, br. s, NH), 7.18–7.37 (11H, m, aromatic & H-6ꢁꢁ), 6.05 (1H, t,
J = 6.3 Hz, H-1ꢁ), 5.05 (1H, m, H-5), 4.63 (2H, AB-quartet, ꢀν = 28.8 Hz,
JAB = 14.6 Hz, CH2Ph), 4.62 (2H, AB-quartet, ꢀν = 57.1 Hz, JAB = 15.1 Hz,
CH2Ph), 4.45 (1H, dd, J = 12.2, 5.4 Hz, H-5ꢁ), 4.30 (1H, dd, J = 12.2, 3.9 Hz,
H-5ꢁ), 4.21 (1H, m, H-4ꢁ), 4.01 (1H, m, H-3ꢁ), 3.41–3.50 (2H, m, H-4 & H-
6), 3.18–3.27 (2H, m, H-4 & H-6), 2.37–2.59 (6H, m, CH2CH2, H-2ꢁ), 1.88
(3H, d, J = 1.0 Hz, Me). 13C NMR (CDCl3, 68 MHz): 171.59, 171.25, 163.34,
155.40, 149.96, 137.89, 135.51, 128.48, 128.02, 127.29, 111.22, 85.77, 81.77,
64.66, 63.31, 60.47, 51.37, 48.56, 37.34, 28.88, 28.70, 12.44. Analysis Calcd
for C32H35N7O8: C59.53, H5.46, N15.18. Found: C59.90, H5.70, N14.85.
N,N ꢁ-Diphenyl-5-methylenetetrahydropyrimidin-2-one (31). To a sus-
pension of carbanilide (10 g, 47 mmol) and NaH oil suspension (60%,
4.9 g, 122 mmol) in freshly distilled THF (150 mL) was added 3-chloro-2-
chloromethylpropene (6.5 g, 52 mmol). The resulting mixture was heated to
a slow reflux under a reflux condenser equipped with a drying tube (CaSO4).
Heating was stopped after 12 h. The cooled solution was quenched with aque-
ous NH4Cl (saturated, 130 mL), and then condensed via rotary evaporation
to remove THF. The remaining aqueous solution and slightly yellow solid
was extracted with EtOAc/hexanes (3:1, 2 × 300 mL). The combined or-
ganic solution was washed with water (2 × 70 mL) and dried over Na2SO4.
Condensation of the dried solution to a volume of about 30 mL resulted in
precipitation, which, after a wash with petroleum ether, gave 10.3 g of white
1
solid (83%). H NMR (CDCl3, 300 MHz) δ: 7.14–7.37 (10H, m, aromatic),
5.127 (2H, m, methylene), 4.377 (4H, t, J = 1.2 Hz, H-4 & H-6). 13C NMR
(CDCl3, 75 MHz): 155.43, 142.97, 136.14, 128.57, 125.37, 125.18, 111.11,
53.60.
N,N ꢁ-Diphenyl-5-hydroxytetrahydropyrimidin-2-one (32).
To an ace-
tone solution (20 mL) of N,Nꢁ-diphenyl-5-methylenetetrahydropyrimidin-
2-one (630 mg, 2.3 mmol) was added through a dropping funnel 25 mL
of an aqueous KMnO4 (500 mg, 3 mmol) during a period of 10 min at
0◦C. The resulting mixture was stirred at the same temperature for 10 min
and then acidified with aqueous HCl (50 mL, 3%). Condensation via rotary
evaporation removed acetone, which was followed by extraction with EtOAc
(3 × 200 mL). The combined organic solution was condensed to give an oil.
The oil was dissolved in methanol (60 mL), and 12 mL of aqueous HIO4
solution (700 mg, 3.0 mmol) added. The mixture was stirred for 4 h, and
then condensed via rotary evaporation to remove methanol. The remaining
aqueous mixture was extracted with EtOAc/hexanes (2:1, 250 mL), and the
organic layer washed with saturated aqueous NaHCO3 (30 mL) and water
(30 mL). The washed solution was dried over sodium sulfate, condensed, and
further dried under vacuum. The resulting residue was dissolved in dry THF

New AZT Conjugates
49
(10 mL) and treated with NaBH4 (200 mg, 5.4 mmol). The resulting mixture
was stirred for 20 min and then poured into 2% aqueous NH4Cl (50 mL).
Rotary evaporation removed THF, and the remaining aqueous mixture was
extracted with EtOAc (3 × 100 mL). The combined organic solution was con-
densed, and the remaining residue chromatographed (3:1 EtOAc/hexanes)
to yield N,Nꢁ-diphenyl-5-hydroxytetrahydropyrimidin-2-one as a white solid
(268 mg, 42%).
N,N ꢁ-Diphenyltetrahydropyrimidin-2-on-5-yl 3ꢁ-Diazo-3ꢁ-deoxythymidin-
5ꢁ-yl Succinate (33).
A mixture of N,Nꢁ-diphenyl-5-hydroxytetrahydro–
pyrimidin-2-one (215 mg, 0.8 mmol), succinic anhydride (156 mg, 1.56
mmol), and DMAP (114 mg, 0.93 mmol) in dry THF (8 mL) was stirred
under nitrogen for 10 h. DCC (275 mg, 1.3 mmol) and AZT (320 mg, 1.2
mmol) were then added. The resultant mixture was stirred under nitrogen
for 30 hours. The solution was condensed and its residue chromatographed
(3:1 and 2:1 benzene/acetone) to yield the desired succinate as a white solid
(192 mg, 39%). 1H NMR (CDCl3, 270 MHz) δ: 9.13 (1H, br s, NH), 7.37–7.11
(11H, m, phenyls and vinyl of thymine), 6.04 (1H, t, J = 6.1 Hz, H-1ꢁ), 5.39
(1H, m, H-5), 4.28 (2H, d of AB-quartet, ꢀν = 37.1 Hz, JAB = 11.7 Hz, JA = 4.4
Hz, JB = 4.2 Hz, H-5ꢁ), 4.10–3.82 (6H, m), 2.79–2.67 (4H, m, CH2CH2), 2.26
(2H, m, H-2ꢁ), 1.80 (3H, s, Me). Analysis Calcd for C30H31N7O8: C58.34,
H5.06, N15.88. Found: C58.02, H5.24, N15.53.
N,N ꢁ-Dimethyl-2-imidazolidone-4-carboxylic Acid (22).
A mixture of 2-
imidazolidone-4-carboxylic acid (1.5 g, 11.5 mmol), KOH pellets (3.0 g, 87%,
46 mmol) and methyl iodide (8.5 g, 57 mmol) in acetone (20 mL) was stirred
at room temperature for 40 min, and then heated to reflux under a reflux
condenser for 2 days. The mixture was filtered, and the solid washed with
acetone. The combined acetone solution was condensed to remove acetone,
and then mixed with aqueous HCl (3%, 40 mL). The aqueous mixture was
extracted with EtOAc (4 × 100 mL). The combined organic solution was
condensed and then subjected to flash chromatography (EtOAc) to isolate
the major product as a brownish oil. The oil was added to an aqueous NaOH
(3%, 30 mL), and stirred for 5 min. Aqueous HCl (3%, 40 mL) was added,
and the acidified mixture extracted with EtOAc (3 × 100 mL). The combined
organic solution was washed with water (30 mL) and then dried over sodium
sulfate. The solution was filtered, condensed, and dried under vacuum to
yield a brownish solid (370 mg, 20%). 1H NMR (CDCl3, 300 MHz) δ: 6.096
(1H, br. s, COOH), 4.091 (1H, dd, J = 10.01, 5.96 Hz, H-4), 3.6228 (1H, dd,
J = 10.01, 9.18 Hz, H-5), 3.4478 (1H, dd, J = 9.18, 5.96 Hz, H-5), 2.898 (3H,
s, Me), 2.807 (3H, s, Me).
N,N ꢁ-Dimethyl-2-imidazolidone-4-carboxylic Acid, 3 ꢁ-Diazo-3 ꢁ-deoxythy-
midin-5 ꢁ-yl Ester (28).
A mixture of N,Nꢁ-dimethyl-2-imidazolidone-4-
carboxylic acid (300 mg, 1.9 mmol), DCC (400 mg, 1.94 mmol), DMAP

50
Z. You and H. J. Lee
(14 mg, 0.1 mmol) and AZT (297 mg, 1.1 mmol) was stirred in freshly dis-
tilled THF (4 mL) under a rubber septum for 2 days. The resultant mixture
was filtered, and the solid washed with acetone. The combined organic so-
lution was condensed, and the residue chromatographed (EtOAc and 2:1
EtOAc/acetone) to yield the main product as a yellow solid (440 mg, 57%).
1H NMR (CDCl3, 300 MHz) δ: 10.195 (1H, s, NH), 7.160 (1H, s, H-6ꢁꢁ), 6.010
(1H, t, J = 6.6 Hz, H-1ꢁ), 4.448 (2H, m, H-5ꢁ), 4.328 (1H, m, H-4ꢁ), 4.136
(1H, dd, J = 9.5, 5.7 Hz, H-4), 4.055 (1H, m, H-3ꢁ), 3.599 (1H, t, J = 9.3 Hz,
H-5), 3.389 (1H, dd, J = 9.2, 5.7 Hz, H-5), 2.866 (3H, s, NMe), 2.773 (3H,
s, NMe), 2.43–2.65 (2H, m, H-2ꢁ), 1.925 (3H, s, CMe). 13C NMR (CDCl3,
75 MHz): 170.19, 163.99, 160.46, 150.12, 136.31, 111.15, 86.51, 81.13, 64.28,
60.60, 56.97, 47.94, 36.65, 30.96, 30.27, 12.27. Analysis Calcd for C16H21N7O6:
C47.17, H5.20, N24.07. Found: C46.91, H5.36, N23.82.
N-Methyl-2-pyrrolidone-5-carboxylic Acid (20).
This acid was synthe-
sized from 2-pyrrolidone-5-carboxylic acid using the same procedure as that
for N,Nꢁdimethyl-2-imidazolidone-4-carboxylic acid, and a white solid (25%)
1
obtained. H NMR (CDCl3, 300 MHz) δ: 13.04 (1H, br. s, COOH), 4.096
(1H, m, H-5), 2.677 (3H, s, Me), 2.12–2.34 (3H, m), 1.903 (1H, m). 13C
NMR (DMSO-d6, 75 MHz): 174.35, 173.52, 60.76, 28.93, 28.37, 22.25.
N-Methyl-2-pyrrolidone-5-carboxylic Acid, 3 ꢁ-Diazo-3 ꢁ-deoxythymidin-5 ꢁ-
yl Ester (26).
This ester was prepared from the acid and AZT using the
same procedure as that for N,Nꢁ-dimethyl-imidazolidone-4-carboxylic acid,
3ꢁ-azido-3ꢁ-deoxythymidin-5ꢁ-yl ester. A yield of 65% was achieved (slightly
yellow solid). 1H NMR (CDCl3, 300 MHz) δ: 9.679 (1H, s, NH), 7.102 (1H,
s, H-6ꢁꢁ), 5.965 (1H, m, H-1ꢁ), 4.410 (2H, m, H-5ꢁ), 4.255 (1H, m), 4.170
(1H, m), 4.011 (1H, m), 2.842 (3H, s, NMe), 2.30–2.58 (5H, m), 2.084 (1H,
m), 1.924 (3H, s, CMe). 13C NMR (CDCl3, 75 MHz): 175.16, 171.38, 163.91,
150.15, 136.31, 111.40, 86.62, 81.21, 64.23, 61.35, 60.59, 36.83, 29.07, 28.88,
22.64, 12.37. Analysis Calcd for C16H20N6O6: C48.98, H5.14, N21.42. Found:
C48.82, H5.35, N21.23.
N-Butyl-2-pyrrolidone-5-carboxylic Acid (21).
A
mixture of 2-
pyrrolidone-5-carboxylic acid (2.0 g, 15.5 mmol), KOH pellets (3.9 g, 85%, 59
mmol), and butyl bromide (10.0 g, 72.5 mmol) in DMSO (10 mL) was stirred
for 6 h and then heated to a slow reflux under a reflux condenser for 2 days.
After being cooled down to room temperature, the mixture was acidified with
aqueous HCl (3%, 190 mL), and then extracted with EtOAc/hexanes (3:2,
4 × 150 mL). The combined organic solution was condensed and the residue
chromatographed (1:1 and 2:1 EtOAc/hexanes) to give the major non-
polar product as a yellow oil. The oil was added to an aqueous NaOH (3%,
15 mL), and stirred for 5 min. More water (50 mL) was then added,

New AZT Conjugates
51
followed by extraction with EtOAc (3 × 80 mL).The combined organic solu-
tion was washed with water (20 mL), and dried over Na2SO4. Condensation
of the EtOAc solution followed by vacuum evaporation yielded a brownish
oil (520 mg, 18%). 1H NMR (CDCl3, 300 MHz) δ: 5.707 (1H, br. s, COOH),
4.242 (1H, dd, J = 9.0, 3.0 Hz, H-5), 3.733 (1H, m, NCH2), 2.959 (1H, m,
NCH2), 2.12–2.65 (4H, m), 1.497 (2H, m, CH2 in butyl), 1.303 (2H, m, CH2
in butyl), 0.919 (3H, t, J = 7.3 Hz, Me).
N-Butyl-2-pyrrolidone-5-carboxylic Acid, 3 ꢁ-Azido-3 ꢁ-deoxythymidin-5 ꢁ-yl
Ester (27).
This ester was prepared from the acid and AZT using the same
procedure as that for N,Nꢁ-dimethyl-2-imidazolidone-4-carboxylic acid, 3ꢁ-
azido-3ꢁ-deoxythymidin-5ꢁ-yl ester. The product was obtained as a slightly yel-
low thick oil (82%). H NMR (CDCl3, 300 MHz) δ: 9.896 (1H, br. s, NH),
1
7.143 (1H, s, H-6ꢁꢁ), 6.022 (1H, t, J = 6.4 Hz, H-1ꢁ), 4.417 (2H, d of AB-
quartet, ꢀν = 29.0 Hz, J = 11.9, 6.0 Hz, H-5ꢁ), 4.22–4.32 (2H, m), 4.029 (1H,
m), 3.710 (1H, m, NCH2), 2.888 (1H, m, NCH2), 2.26–2.61 (5H, m), 2.105
(1H, m), 1.949 (3H, s, Me), 1.469 (2H, m, CH2 in butyl), 1.306 (2H, m, CH2
in butyl), 0.909 (3H, t, J = 7.2 Hz, Me in butyl). 13C NMR (CDCl3, 75 MHz):
175.14, 171.71, 163.88, 150.16, 136.05, 111.48, 86.26, 81.11, 64.20, 60.57,
59.43, 41.51, 36.75, 29.36, 29.09, 23.03, 19.95, 13.59, 12.35. Analysis Calcd
for C19H26N6O6: C52.53, H6.03, N19.34. Found: C52.29, H6.31, N19.08.
2-Imidazolidone-4-carboxylic Acid, 3 ꢁ-Azido-3 ꢁ-deoxythymidin-5 ꢁ-yl Ester
(34).
A mixture of 2-imidazolidone-4-carboxylic acid (500 mg, 3.8 mmol),
DCC (790 mg, 3.8 mmol), DMAP (25 mg, 0.2 mmol) and AZT (600 mg, 2.2
mmol) in dry THF (10 mL) was stirred under a rubber septum for 3 days. The
resulting suspension was filtered, and the solid washed with EtOAc (100 mL).
The combined THF-EtOAc solution was condensed, and the residue chro-
matographed (EtOAc/MeOH 10:1) to yield the polar product as a white
1
solid (1.2 g, 82%). H NMR (DMSO-d6, 300 MHz) δ: 11.353 (1H, s, H-3ꢁꢁ),
7.394 (1H, d, J = 1.1 Hz, H-6ꢁꢁ), 6.765 (1H, s, NH of imidazolidone), 6.376
(1H, s, NH of imidazolidone), 6.105 (1H, t, J = 6.8 Hz, H-1ꢁ), 4.495 (1H,
m), 4.26–4.35 (3H, m), 3.996 (1H, m), 3.593 (1H, t, J = 9.1 Hz, H-5), 3.345
(1H, m), 2.429 (1H, m, H-2ꢁ), 2.304 (1H, m, H-2ꢁ), 1.784 (3H, d, J = 1.1 Hz,
Me). 13C NMR (DMSO-d6, 75 MHz): 172.07, 163.73, 162.81, 150.44, 135.98,
110.18, 83.63, 80.46, 64.19, 60.09, 53.36, 43.06, 35.51, 12.13. Analysis Calcd
for C14H17N7O6: C44.33, H4.52, N25.85. Found: C44.44, H4.68, N25.82.
2-Pyrrolidone-5-carboxylic Acid, 3 ꢁ-Azido-3 ꢁ-deoxythymidin-5 ꢁ-yl Ester
(25).
It was synthesized by the same procedure as that for 2-imidazolidone-
4-carboxylic acid, 3ꢁ-azido-3ꢁ-deoxythymidin-5ꢁ4-yl ester. A white solid (85%)
was obtained. 1H NMR (CDCl3, 270 MHz) δ: 10.30 (1H, br. s, NH of thymi-
dine), 7.71 (1H, s, NH of pyrrolidone), 7.10 (1H, s, H-6ꢁꢁ), 5.85 (1H, t,

52
Z. You and H. J. Lee
J = 6.2 Hz, H-1ꢁ), 4.23–4.61 (4H, m), 4.01 (1H, m), 2.13–2.75 (6H, m),
1.89 (3H, s, Me). 13C NMR (CDCl3, 68 MHz): 178.77, 171.71, 164.04, 150.33,
137.00, 111.16, 87.54, 81.52, 63.65, 60.32, 55.71, 36.61, 29.25, 24.60, 12.23.
Analysis Calcd for C15H18N6O6: C47.62, H4.80, N22.21. Found: C47.35,
H5.12, N21.89.
4-Imidazolecarboxylic Acid, 3 ꢁ-Azido-3 ꢁ-deoxythymidin-5 ꢁ-yl Ester (24).
The procedure for the synthesis of 2-imidazolidone-5-caboxylic acid, 3ꢁ-azido-
3ꢁ-deoxythymidin-5ꢁ-yl ester was used to yield a white solid (40%). 1H NMR
(CDCl3, 300 MHz) δ: 10.075 (1H, br. s, NH of thymidine), 8.281 (1H, t, J =
0.8 Hz, imidazole), 7.643 (1H, t, J = 1.4 Hz, imidazole), 7.109 (1H, dd, J =
1.6, 0.8 Hz, imidazole), 7.042 (1H, d, J = 1.1 Hz, H-6ꢁꢁ), 5.878 (1H, dd, J = 7.7,
4.8 Hz, H-1ꢁ), 4.674 (2H, d of AB-quartet, ꢀν = 21.3 Hz, JAB = 12.0 Hz, JA =
3.5 Hz, JB = 4.9 Hz, H-5ꢁ), 4.509 (1H, m, H-4ꢁ), 4.086 (1H, m, H-3ꢁ), 2.700 (1H,
m, H-2ꢁ), 2.524 (1H, m, H-2ꢁ), 1.886 (3H, d, J = 1.1 Hz, Me). Analysis Calcd
for C14H15N7O5: C46.54, H4.18, N27.14. Found: C46.23, H4.39, N26.88.
BIOLOGICAL ASSAYS
Anti-HIV assay was carried out with CD4 expressing lymphocytes (CEM-
SS cell line) through the following standard procedure. Agent is dissolved
dimethyl sulfoxide and diluted 1:100 in cell culture medium before prepar-
ing serial half-log10 dilutions. CD4 expressing lymphocytes are added and
after a brief interval HIV-1 is added, resulting in a 1:200 final dilution of the
compound. Uninfected cells with the compound serve as a toxicity control,
and infected and uninfected cells without the compound serve as basic con-
trols. Cultures are incubated at 37◦C in a 5% carbon dioxide atmosphere
for 6 days. The tetrazolium salt, XXT, is added to all wells, and cultures are
incubated to allow formazan color development be visible viable cells. In-
dividual wells are analyzed spectraphotometrically to quantitate formazan
production, and in addition are viewed microscopically for detection of vi-
able cells and confirmation of protective activity. Drug-treated virus-infected
cells are compared with drug-treated non-infected cells and with other appro-
priate controls (untreated infected and untreated non-infected cells, drug-
containing wells without cells, etc.) on the same plate. Data are reviewed
in comparison with other tests done at the same time and a determination
about activity is made.
CONCLUSION
We have synthesized three AZT conjugates with cholanic acids, two
conjugates with N,Nꢁ-disubstituted tetrahydropyrimidinones, seven AZT
prodrugs from 2-imidazolidone-4-carboxylic acid and six of its derivatives.

New AZT Conjugates
53
Anti-HIV and cytotoxicity bioassay with CEM-SS cell line shows that three
of these conjugates, AZT conjugates with 2-imidazolidone-4-carboxylic
acid, 2-pyrrolidone-5-carboxylic acid, and N,Nꢁ-diphenyl-5-hydroxy-tetrahyd-
ropyrimidin-2-one gave anti-HIV activities similar to or greater than AZT,
and they exhibited much lower cytotoxic effects and thus much greater TI
values than AZT. Since the compounds other than AZT used to synthesize
the conjugates are inactive against HIV, the anti-HIV activity after the
breakage of the conjugates could only have come from AZT. Structural
features that favor the desired therapeutic profile of the conjugates appear
to include a five-membered ring cyclic urea or lactam, and six-membered
ring cyclic urea with N,Nꢁ-diphenyl substitution. Thus a good stride has been
made towards our project goals with the three conjugates 34, 25, and 33.
REFERENCES
1. Mitsuya, H.; Weinhold, K.J.; Furman, P.A.; St. Clair, M.H.; Nusinoff-Lehrman, S.; Gallo, R.C.;
Bolognesi, D.; Barry, D.W.; Broder, S. 3ꢁ-Azido-3ꢁ-deoxythymidine (BW A509U): An antiviral
agent that inhibits the infectivity and cytopathic effect of human T-lymphotropic virus type
III/lymphadenopathy-associated virus in vitro. Proc. Natl. Acad. Sci. U.S.A. 1985, 82(20), 7096–7100.
2. Richman, D.D.; Fischl, M.A.; Grieco, M.H.; Gottlieb, M.S.; Volberding, P.A.; Laskin, O.L.; Leedom,
J.M.; Groogman, J.E.; Mildvan, D.; Schooley, R.T.; Jackson, G.G.; Durack, D.T.; Phil, D.; King, D. The
efficacy of azidothymidine (AZT) in the treatment of patients with AIDS and AIDS-related complex.
N. Engl. J. Med. 1987, 317(4), 185–191.
3. Mitsuya, H.; Broder, S. Strategies for antiviral therapy in AIDS. Nature (London) 1987, 325(6107),
773–778.
4. Fischl, M.A.; Richman, D.D.; Grieco, M.H.; Gottlieb, M.S.; Volberding, P.A.; Laskon, O.L.; Leedom,
J.M.; Grooman, J.E.; Mildvan, D.; Hirsch, M.S.; Jackson, G.G.; Durack, D.T.; Musinoff-Lehrman, S.
The toxicity of azidothymidine (AZT) in the treatment of patients with AIDS and AIDS-related
complex. N. Engl. J. Med. 1987, 317(4), 192–197.
5. De Clercq, E. Perspectives for the chemotherapy of AIDS. Anticancer Res. 1987, 7(5B), 1023–1038.
6. Zimmerman, T.P.; Mahony, W.B.; Prus, K.L. 3ꢁ-Azido-3ꢁ-deoxythymidine: An unusual nucleoside ana-
logue that permeates the membrane of human erythrocytes and lmphocytes by non-facilitated dif-
fusion. J. Biol. Chem. 1987, 262(12), 5748–5754.
7. Dessolin, J.; Galea, P.; Vlieghe, P.; Chermann, J.-C.; Kraus, J.-L. New bicyclam-AZT conjugates: Design,
synthesis, anti-HIV evaluation, and their interaction with CXCR-4 coreceptor. J. Med. Chem. 1999,
42(2), 229–241.
8. Aggarwal, S.K.; Gogu, S.R.; Rangan, S.R.S.; Agrawal, K.C. Synthesis and biological evaluation of
zidovudine. J. Med. Chem. 1990, 33(5), 1505–1510.
9. Sharma, A.P.; Ollapally, A.P.; Lee, H.J. Synthesis and anti-HIV activity of prodrugs of azidothymidine.
Antiviral Chem. Chemother. 1993, 4(2), 93–96.
10. Meier, C.; Knispel, T.; De Clercq, E.; Balzarini, J. cycloSal-pronucleotides of 2ꢁ,3ꢁ-dieoxyadenosine
and 2ꢁ,3ꢁ-dideoxy-2ꢁ,3ꢁ-didehydroadenosine: Synthesis and antiviral evaluation of a highly efficient
nucleotide delivery system. J. Med. Chem. 1999, 42(9), 1604–1614.
11. Hong, C.I.; Nechaev, A.; Kirisits, A.J.; Vig, R.; West, C.R.; Manouilov, K.K.; Chu, C.K. Nucleoside
conjugates. 15. Synthesis and biological activity of anti-HIV nucleoside conjugates of ether and
thioether phospholipids. J. Med. Chem. 1996, 39, 1771.
12. Namane, A.; Gouyette, C.; Fillion, M.-P.; Fillion, G.; Huynh-Dinh, T. Improved brain delivery of AZT
using a glycosyl phosphotriester prodrug. J. Med. Chem. 1992, 35(16), 3039–3044.
13. Vlieghe, P.; Clerc, T.; Pannecouque, C.; Witvrouw, M.; De Clercq, E.; Salles, J.-P.; Kraus, J.-L. New
3ꢁ-Azido-3ꢁ-deoxythymidin-5ꢁ-yl O-(4-hydroxyalkyl or -alkenyl or -alkylepoxide) carbonate prodrugs:
Synthesis and anti-HIV evaluation. J. Med. Chem. 2001, 44(18), 3014–3021.

54
Z. You and H. J. Lee
14. Lee, H.J.; Soliman, M.R.I. Anti-inflammatory steroids without pituitary-adrenal suppression. Science
1982, 215(4535), 989–991.
15. Rosner, W. Plasma steroid-binding proteins. Endocrinol. Metab. Clin. North Am. 1991, 20(4), 697.
16. You, Z.; Mian, A.M.; Agarwal, R.P.; Lee, H.J. Synthesis, anti-HIV, and cytotoxic activity of new AZT
conjugates of steroid acids. Nucleos. Nucleot. Nucl. Acids 2003, 22(11), 2049–2060.
17. Matsumoto, H.; Hamawaki, T.; Ota, H.; Kimura, T.; Goto, T.; Sano, K.; Hayashi, Y.; Kiso, Y. “Double-
drugs”—A new class of prodrug form of an HIV protease inhibitor conjugated with a reverse tran-
scriptase inhibitor by a spontaneously cleavable linker. Bioorganic & Med. Chem. Letters 2000,
10(11), 1227.
18. De Lucca, G.V.; Liang, J.; Aldrich, P.E.; Calabrese, J.; Cordova, B.; Klabe, R.M.; Rayner, M.M.; Chang,
C.-H. Design, synthesis, and evaluation of tetrahydropyrimidinones as an example of a general
approach to nonpeptide HIV protease inhibitors. J. Med. Chem. 1997, 40(11), 1707–1709.
19. Fyles, T.M.; James, T.D.; Pryhitka, A.; Zojaji, M. Assembly of ion channel mimics from a modular
construction set. J. Org. Chem. 1993, 58, 7456–7468.
20. Hori, H.; Nishida, Y.; Ohrui, H.; Meguro, H. Regioselective de-O-benzylation with Lewis acids. J. Org.
Chem. 1989, 54, 1346–1353.
21. Weislow, O.S.; Kiser, R.; Fine, D.L.; Bader, J.; Shoemaker, R.H.; Boyd, M.R. New soluble-formazan assay
for HIV-1 cytopathic effects: Application to high-flux screening of synthetic and natural products for
AIDS-antiviral activity. J. Natl. Cancer Inst. 1989, 81(8), 577–586.
22. Since AZT has been extensively studied in the past 20 years or so, it is the NCI’s policy that when
it is used as a standard compound in anti-HIV screening assays, only its EC50 value is determined.
The IC50 and TI50 values of AZT, as is suggested by the NCI, should be obtained from previously
published articles such as reference 21.