Novel approach to the synthesis of AZT 5′-O-hydrogen phospholipids

Article abstract of DOI:10.1016/S0040-4039(02)01045-6

A series of AZT 5′-O-hydrogen phospholipids were synthesized by a tandem transesterification of diphenyl phosphite with AZT and a long-chain alcohol. The method possesses the merits of ease of operation and high yields. It can also be extended to synthesize other biological phospholipids.

Full text of DOI:10.1016/S0040-4039(02)01045-6

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 5281–5283
Novel approach to the synthesis of AZT 5%-O-hydrogen
phospholipids
Qiang Xiao,a Jing Sun,b Yong Ju,a,* Yu-fen Zhaoa,* and Yu-xin Cuib
aThe Key Laboratory of Bioorganic Phosphorus Chemistry, Ministry of Education, Department of Chemistry, School of Life
Sciences and Engineering, Tsinghua University, Beijing 100084, PR China
bNational Research Laboratory of Natural and Biomimetic Drugs, School of Pharmaceutical Sciences, Peking University,
Beijing 100083, PR China
Received 28 March 2002; revised 15 May 2002; accepted 31 May 2002
Abstract—A series of AZT 5%-O-hydrogen phospholipids were synthesized by a tandem transesterification of diphenyl phosphite
with AZT and a long-chain alcohol. The method possesses the merits of ease of operation and high yields. It can also be extended
to synthesize other biological phospholipids. © 2002 Elsevier Science Ltd. All rights reserved.
The development of effective antiviral therapy for the
treatment of individuals infected with human
immunodeficiency virus (HIV) presents a unique chal-
lenge. Nucleosides and nucleotides have demonstrated
widespread utility as antiviral and anticancer therapeu-
tics.1 One of them is 3%-azido-2%, 3%-dideoxythymidine
(AZT), which was the first clinically approved drug
against HIV infection, despite its undesirable side
effects, such as bone marrow suppression, myopathy
and hepatic abnormalities.2 In order to reduce its toxic-
ity and increase the anti-HIV activity, a lot of work has
been reported aimed at developing 5%-O-ester prodrugs
of AZT.3 The 5%-hydrogen phosphate of AZT (PZT,
Scheme 1), which is much less toxic than AZT (CC50
values of 2.5 mM and 210 mM for PZT and AZT,
respectively), is one of the most significant compounds.
The overall selectivity (SI) for PZT (CC50/IC50) is supe-
rior to that of AZT,4 and now PZT is currently in
Phase I clinical trials.5 PZT, unlike highly acidic
nucleoside 5%-phosphonates, which cannot enter the cell,
may penetrate the cell membrane due to its weakly
acidic undissociated nature. Encouraged by the results
with PZT, 5%-O-hydrogen phospholipids were synthe-
sized in a convenient manner in order to: (1) increase
transmembrane transport characteristics; (2) determine
their antiviral effects and (3) further investigate the
intracellular metabolism of 5%-O-hydrogen phospho-
nates, which may be different from that of AZT.6
Two methods can be used to synthesize AZT 5%-O-
hydrogen phospholipids. The first uses the phospho-
ramidate strategy, which has been extensively
developed for the synthesis of oglionucleotides.7 Using
this methodology, phosphoramidate diesters of phos-
phites were synthesized and then hydrolyzed under the
activation of 1H-tetrazole in the presence of water. The
second is the H-phosphonate method.8 In this method,
it is necessary to prepare H-monophosphonate and
then couple to an alcohol in the presence of a condens-
ing reagent. Both of these methods suffer from the
disadvantage of laborious synthetic procedures and
chromatographic purification. In order to obtain multi-
gram quantities of AZT 5%-O-hydrogen phospholipids,
we describe, herein, a mild and one-pot synthesis.
O
O
HO P O
Me
T
NH
O
H
T=
O
N
N3
PZT
Scheme 1.
Keywords: phosphorylation; transesterification; H-phosphonate;
AZT; phospholipid.
* Corresponding authors. Tel.: +86-10-62772259; fax: +86-10-
Diphenyl phosphite (DPP) is a commercially available,
inexpensive phosphorylation reagent, which undergoes
fast transesterification with alcohols in pyridine to
afford mixtures of the corresponding dialkyl and alkyl
62781695;
e-mail:
juyong@tsinghua.edu.cn;
yfzhao@
tsinghua.edu.cn
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01045-6

5282
Q. Xiao et al. / Tetrahedron Letters 43 (2002) 5281–5283
phenyl H-phosphonates.9 In order to obtain mono-
phenyl H-phosphonate only, a large excess of DPP was
needed to obviate the formation of dialkyl H-phospho-
nates.10 Later research showed that the amount of DPP
can be greatly reduced if DPP was added to the desired
monohydroxylic nucleoside very slowly.11 Inspired by
these results, we intended to synthesize the target phos-
pholipids with a tandem transesterification reaction.
method, the other phospholipids were also synthesized
in good yields (Table 1). It is noteworthy to mention
that these 5%-O-hydrogen phospholipids are also impor-
tant reaction intermediates: (1) they can be used to
synthesize other phosphoramidate prodrugs of AZT by
the Todd reaction13, and (2) they can be oxidized by
iodine, sulfur or selenium to obtain phosphonate, phos-
phorothioate and phosphoroselenoate derivatives of
AZT respectively.14
Phospholipid 5d was selected to check the viability of
this method. Under an argon atmosphere, a solution of
AZT (1 mmol) in 10 ml dry pyridine was added to a
stirred solution of DPP (2 mmol) in 10 ml dry pyridine
over 1 hour. After 15 min, the 31P NMR spectrum
showed four new signals other than that for DPP (lP
ca. 1.7 ppm). Two main peaks at 6.75 and 5.98 ppm are
the signals of a pair of diastereoisomers of phenyl
H-phosphonate 3d due to the chirality of AZT. The
small peak at 14.48 ppm is the symmetric H-phospho-
nate 4. Then, hexadecanol (3 mmol) was added in one
portion. After 1 hour, the 31P NMR showed that the
signals at 6.75 and 5.98 ppm had disappeared and a
further three new signals had appeared. The signals at
9.83 and 9.11 ppm correspond to the target phospho-
lipid 5d and the signal at 8.65 ppm is the signal due to
symmetric H-phosphonate 6d, which was identified by
comparison with an authentic sample (Scheme 2). After
purification, the phospholipid 5d was obtained in 87%
yield12 based on the quantity of AZT. Using the same
In summary, AZT 5%-O-hydrogen phospholipids were
synthesized in an easy one-pot reaction and high yields.
The extension of this methodology to other hydrogen
phospholipids should prove to be very useful for the
development of phospholipid prodrugs of nucleosides
such as d4T and ddI. This methodology can also be
used to synthesize phospholipids of carbohydrate and
other biological molecules. Further efforts in this direc-
tion and the assessment of biological activities against
HIV are in progress.
Acknowledgements
The authors would like to acknowledge financial sup-
port from the National Natural Science Foundation of
China (No. 20132020, No. 20175026, No. 29672022 and
No. 20172033), the Ministry of Science and Technol-
N3
O
HO
O
P
H
O
P
H
T
O
P
H
pyridine
PhO
P
H
O
O
T
O
T
PhO
OPh
O
O
PhO
OPh
O
T
O
N3
N3
N3
2
3 a-e
2
1
4
Trace amount
N3
O
OH
O
O
n
O
P
H
O
O
T
O
P
H
O
T
n
P
H
O
O
O
n
T
n
O
pyridine
N3
N3
Trace amount
6 a-e
5 a-e
4
Scheme 2.
Table 1. Phospholipids 5a–e prepared using the tandem transesterification reaction
Compds.
n
31P NMRa
31P NMR (l ppm)
ESI MS
(M+Na)+
Yield (%)c
b
1JP–H
(M+H)+
(M+K)+
5a
5b
5c
5d
5e
8
9.19, 8.62
9.20, 8.63
9.28, 8.76
9.28, 8.69
9.39, 8.84
696
697
695
698
697
472
500
528
556
584
494
522
550
578
606
510
538
566
594
622
90
85
85
87
84
10
12
14
16
a The values were determined in CDCl3 using a Bruker AMP 200 at 81 Hz (85% H3PO4 as internal standard).
b In Hz.
c Based on the reactant AZT.

Q. Xiao et al. / Tetrahedron Letters 43 (2002) 5281–5283
5283
9. Kers, A.; Kers, I.; Stawinski, J.; Sobkowski, M.;
Kraszewski, A. Synthesis 1995, 427.
10. Jankowska, J.; Sobkowski, M.; Stawinski, J.; Kraszewski,
ogy, the Chinese Ministry of Education and Tsinghua
University.
A. Tetrahedron Lett. 1994, 35, 3355.
11. Iyer, V. V.; Griesgraber, G. W.; Radmer, M. R.; McIn-
tee, E. J.; Wagner, C. R. J. Med. Chem. 1999, 42, 5188.
12. Characteristics are given for a representative compound
5d: colorless oil; 1H NMR (CDCl3, 300 MHz) l 9.47
References
1. Jones, R. J.; Bischofberger, N. Antivir. Res. 1995, 27, 1.
2. Moore, K. H. P.; Raasch, Rh.; Brouwer, K. L. R.;
Opheim, K.; Cheeseman, S. H.; Eyster, E.; Lemon, S. M.;
Vanderhorst, C. M. Antimicrob. Agents Chemother. 1995,
39, 2732.
1
(br, 1H, H-3), 6.86, 6.85* (d, 1H, JP–H=705 Hz, P-H),
7.32, 7.29* (s, 1H, H-6), 6.14, 6.13* (t, 1H, J1%,2%=6.3 Hz,
H-1%), 3.97–4.33 (m, 6H, CH2-5%, H-4%, H-3%,
HOCH2(CH2)14CH3), 2.34 (m, 2H, H-2%), 1.87 (s, 3H,
CH3-5), 1.64 (m, 2H, HOCH2CH2(CH2)13CH3), 1.14–1.18
(br, 26H, H-OCH2CH2(CH2)13CH3), 0.81 (t, 3H, J=6.6
Hz, HOCH2CH2(CH2)13CH3); 13C NMR (75 MHz):
169.66 (C-4), 150.47, 150.30* (C-2), 135.47, 135.37* (C-
6), 111.49, 111.43* (C-5), 84.88 (C-4%), 82.17, 82.08*
(C-1%), 66.71, 66.64* (C-5%, 2J=3.1 Hz), 64.22 (C-3%),
59.96 (CH3(CH2)14CH2O), 37.32, 37.27* (C-2%), 31.83
(CH3(CH2)13CH2CH2O), 30.34, 30.26* (1C), 29.13–29.47
(br, 10C), 25.41, 25.37* (1C), 22.60 (1C), 14.03
(CH3(CH2)14CH2O), 12.37 (CH3-5); ESI MS (+): m/z 556
(M+H)+, 578 (M+Na)+, 594 (M+K)+; HRMS (ESI):
found 556.3253, C26H47N5O6P (M+H)+ requires
556.3258.
3. Parang, K.; Wiebe, L. I.; Knaus, E. E. Curr. Med. Chem.
2000, 7, 995.
4. (a) Machado, J.; Salomon, H.; Oliveira, M. Nucleos.
Nucleot. 1999, 18, 901; (b) Tarussova, N. B.; Kukhanova,
M. K.; Krayevsky, A. A.; Karamov, E. K.; Lukashov, V.
V.; Kornilayeva, G. B.; Rodina, M. A.; Galegov, G. A.
Nucleos. Nucleot. 1991, 10, 351.
5. Yurin, O.; Kravtchenko, A.; Serebrovskaya, L.; Goloch-
vastova, E.; Burova, N.; Voronin, E.; Pokrovsky, V.
AIDS 1998, 12, 240.
6. Boal, J. H.; Iyer, R. P.; Egan, W. Nucleos. Nucleot. 1993,
12, 1075.
7. Beaucage, S. L.; Iyer, R. P. Tetrahedron 1993, 49, 6123.
8. Stawinski, J. In Handbook of Organophosphorus Chem-
istry; Engel, R., Ed.; Marcel Dekker: New York, 1992;
pp. 377–434.
13. Ji, C. J.; Xue, C. B.; Zeng, J. N.; Zhao, Y. F. Synthesis
1988, 444.
14. Lindh, I.; Stawinski, J. J. Org. Chem. 1989, 54, 1338.