HIV-1 non-nucleoside reverse transcriptase inhibitors: Incorporation of benzylphosphonate moiety for solubility improvement

Article abstract of DOI:10.1016/j.mencom.2016.03.009

Benzylphosphonates of 5'-norcarbocyclic analogue of 2',3'-dideoxy-2',3'-didehydrouridine and its N³-benzyl derivatives with different substituents at the phosphorus atom were designed and synthesized in attempt to improve solubility of potentia

Full text of DOI:10.1016/j.mencom.2016.03.009

Mendeleev
Communications
Mendeleev Commun., 2016, 26, 114–116
HIV-1 non-nucleoside reverse transcriptase inhibitors: incorporation
of benzylphosphonate moiety for solubility improvement
Elena S. Matyugina,a Vladimir T. Valuev-Elliston,a Alexander O. Chizhov,b
Sergei N. Kochetkova and Anastasia L. Khandazhinskaya*a
a V. A. Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, 119991 Moscow,
Russian Federation. Fax: +7 499 135 1405; e-mail: khandazhinskaya@bk.ru
b N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 119991 Moscow,
Russian Federation
DOI: 10.1016/j.mencom.2016.03.009
Benzylphosphonates of 5'-norcarbocyclic analogue of 2',3'-di-
deoxy-2',3'-didehydrouridine and its N3-benzyl derivatives
O
O
with different substituents at the phosphorus atom were
designed and synthesized in attempt to improve solubility
of potential non-nucleoside reverse transcriptase inhibitor.
Solubility of the most hydrophilic representative was 400 mg
in 100 ml of 1% DMSO solution in water. The target phos-
phonates were examined against HIV-1 reverse transcriptase
with KI for all compounds being higher than 100 mm.
Ph
N
N
P
O
R'
OR
O
R = H, Et, Bn
R' = H, Bn
According to WHO to date, AIDS has killed more than 39 million
lives, in late 2013 the number of people infected with HIV was
about 36 million.1 Currently, the main technique in the treatment
of HIV infection is the so-called highly active antiretroviral
therapy (HAART), wherein a pharmacological agent is applied
as a cocktail of several antivirals [commonly reverse transcriptase
(RT) and protease inhibitors]. There are two main classes of HIV
RT inhibitors, the most studied and widely used in anti-HIV
therapy target enzyme.2 Nucleoside RT inhibitors (NRTIs) are
analogues of natural nucleosides lacking free 3'-OH group and
mechanism of their action consists in the intracellular triphos-
phorylation, followed by incorporation into a growing strand of
proviral DNA and, consequently, the termination of its synthesis.
Non-nucleoside RT inhibitors (NNRTIs), in contrast to the NRTIs,
do not require intracellular triphosphorylation. Once in the cell,
they bind to hydrophobic ‘pocket’ of HIV RT, alter the structure
of the active centre and thus effectively inhibit viral replication via
a noncompetitive mode. Currently there are more than 50 classes
of compounds with different chemical structures, emerged as
NNRTIs, but only five approved for therapy by FDA (efavirenz,
delaverdin, nevirapine, etravirine and rilpivirine).3 The main
disadvantages of NNRTI are drug toxicity, side effects and the
rapid emergence of resistant strains of the virus, resulting in
the need to replace drug. Thus, new effective low-toxic NNRTIs
with resistance profile, different from the existing ones are in
demand.
The ‘phosphonate strategy’ has been developed for HIV NRTIs.
One of the major problems of the latter is the low efficiency
of the three-stage intracellular phosphorylation required for the
manifestation of the inhibitory activity of the drug. Introduction
of phosphonate moiety which is isosteric mimic of natural phos-
phate residue allows one to overcome the first stage (of three
ones), and to obtain a nucleoside monophosphate analogue which
is resistant to dephosphorylating enzymes. This analogue is suitabl
e
for subsequent phosphorylation. The most successful example
of such a strategy is the Tenofovir.4 Strategy of introduction of
the phosphonate moiety into AZT molecule led to a design of its
depot forms. It was shown that the 5'-phosphonates of AZT are
able to generate AZT in the body. In this case, pharmacokinetic
parameters of released nucleoside are significantly better than
when AZT is administered itself (intact). Hence depot forms
allow one to significantly reduce the toxicity of the drug while
maintaining its anti-HIV activity.5–9 Other examples of successful
biological applications of the phosphonate moiety are phosphon-
oxins and polyoxin phosphonate analogues with antimicrobial
activity.10–13 Methylene phosphonate analogues of adenine di-
nucleoside can inhibit inosine monophosphate dehydrogenase.14
Thus, the presence of the phosphonate moiety in different bio-
logically active molecules leads to improved efficiency and/or
reduced toxicity.
Recently, we reported that 3,4-substituted 5'-norcarbocyclic
analogues of 2',3'-dideoxy-2',3'-didehydrouridine A act as
NNRTIs (IC50 5–50 mM) on enzyme of the wild and mutant virus
strains.15,16 However, the antiviral potential of these compounds
is substantially reduced because of poor solubility, which was
the prerequisite for the design of their phosphonate analogues B.
In this work, commercially available diethyl benzylphosphonate
was used as a donor of the phosphonate moiety. Two methods of
the phosphonate fragment administration into the 5'-norcarbo-
cyclic uracil analogues were evaluated (Scheme 1). Treatment of
diethyl benzylphosphonate with phosphorus pentachloride gave
the corresponding dichloride whose subsequent reaction with
A major problem of many NNRTI candidates is low solubility
which results in almost zero oral bioavailability. The reason is
that the molecules required for binding to the hydrophobic pocket
of HIV RT should possess high lipophilicity. A lot of different
modifications of potential HIV NNRTI molecules were tried with
varying effectiveness to overcome this problem; various hydro-
philic groups (hydroxyl, amino, nitro, sulfo, sulfamino, etc.)
were used.3 This work is devoted to assessing the prospects of
phosphonate fragment introduction into HIV NNRTI molecule
which was not applied for this purpose previously.
© 2016 Mendeleev Communications. Published by ELSEVIER B.V.
on behalf of the N. D. Zelinsky Institute of Organic Chemistry of the
Russian Academy of Sciences.
– 114 –

Mendeleev Commun., 2016, 26, 114–116
O
O
N
HO
O
N
NH
N
O
O
O
C
O
n
Ar
Ph
P
OR
Cl
O
Ph
P
O
O
N
NH
R
3
OR
Py
O
1 R = H
2 R = Et
5 R = H
6 R = Et
A
Ar = Ph, 3,5-Me2C6H3, 3,5-Cl2C6H3
R = H, Me
BnBr,
K2CO3,
DMF
n = 0, 1
O
O
O
O
N
Ph
1 or 2
O
N
N
Ph
P
O
N
N
HO
Ph
N
O
P
O
Bn
R'
OR
O
O
OR
4
7 R = H, R' = Bn
8 R = Bn, R' = H
9 R = Bn, R' = Bn
10 R = Et, R' = Bn
B
R = H, Et, Bn
1-(4-hydroxycyclopent-2-en-1-yl)uracil afforded target product in
10–15% yield.
Scheme 2
The more effective was the use of monochlorides 1 and 2,
which were prepared by treatment of diethyl benzylphosphonate
with bromotrimethylsilane or sodium hydroxide followed by
refluxing the resulting phosphonic acid or its monoethyl ester with
thionyl chloride in 1,2-dichloroethane.†,16 The target reagents were
used without further purification in the reaction with 1-(4-hydroxy-
cyclopent-2-en-1-yl)uracil 317 in pyridine (Scheme 2).‡ After
purification by anion exchange chromatography on DEAE-cel-
lulose and then by reverse phase chromatography (C18 column),
the yields of target phosphonates 5 and 6 were 30 and 37%,
respectively. Attempts to synthesize benzylphosphonates of
3-benzyl-1-(4-hydroxycyclopent-2-en-1-yl)uracil 4 by reaction
with phosphonyl chlorides 1 and 2 were unsuccessful. However,
products 7–10 were accessed by treatment of compounds 5 and
6 with benzyl bromide and potassium carbonate in dry DMF.15
O
O
P
O
P
Me3SiBr
DMF
NaOH
EtOH
Ph
Ph
Ph
P
OH
OEt
OH
OH
OEt
OEt
SOCl2
SOCl2
PCl5
(CH2Cl)2
reflux
(CH2Cl)2
reflux
31P NMR (DMSO-d6) d: 18.37. HRMS, m/z: 347.0799 [M–H]– (calc. for
C16H17N2O5P, m/z: 347.0802).
4-[2,4-Dioxo-3,4-dihydropyrimidin-1(2H)-yl]cyclopent-2-en-1-yl ethyl
benzylphosphonate 6 (mixture of diastereomers). O-Ethyl benzylphos-
phonyl chloride (0.75 mmol) was added to a solution of 1-(4-hydroxy-
2-cyclopent-2-en-1-yl)uracil 3 (145 mg, 0.75 mmol) in dry pyridine
(10 ml). The mixture was stirred for 18 h at room temperature, then the
solvent was evaporated and the residue was subjected to column chro-
matography (CHCl3–MeOH, 98:2) to give 6 as a colourless syrup (104 mg,
0.27 mmol, 37%). Rf = 0.4. 1H NMR (CDCl3) d: 1.28–1.23 (m, 3H, Me),
1.50–1.46, 1.65–1.62 (2m, 1H, Hb-5'), 2.75–2.72, 2.88–2.84 (2m, 1H,
Ha-5'), 3.20–3.13 (m, 2H, CH2), 4.06–4.01 (m, 2H, CH2P), 5.29–5.26
(m, 1H, H-4'), 5.58–5.57 (m, 1H, H-1'), 5.56–5.54 (m, 1H, H-5), 5.87–5.82
(m, 1H, H-3'), 6.02–6.01, 6.18–6.16 (2m, 1H, H-2'), 7.07–7.05 (m, 1H,
H-6), 7.26–7.22 (m, 5H, Ph), 9.49 (s, 1H, NH). 31P NMR (DMSO-d6) d:
24.31, 24.20. HRMS, m/z: 399.1073 [M+Na]+ (calc. for C18H21N2O5P,
m/z: 399.1080).
4-[2,4-Dioxo-3-benzyl-4-hydropyrimidin-1(2H)-yl]cyclopent-2-en-1-yl
benzylphosphonate 7. Potassium carbonate (18 mg, 0.13 mmol) and BnBr
(17 ml, 0.13 mmol) were added to a solution of 5 (35 mg, 0.1 mmol) in dry
DMF (10 ml). The reaction mixture was stirred for 2 h at room tempe-
rature, the solvent was evaporated, and the residue was separated by PLC
(dioxane–aq. NH3, 4:1) to give 7 as a pale yellow oil (11 mg, 0.025 mmol,
25%). Rf = 0.31. 1H NMR (DMSO-d6) d: 1.46–1.37 (m, 1H, Hb-5' ),
2.68–2.64 (m, 1H, Ha-5'), 2.91–2.87 (m, 2H, CH2P), 4.98 (s, 2H, CH2N),
5.03–5.00 (m, 1H, H-4'), 5.39–5.36 (m, 1H, H-1'), 5.75–5.73 (d, 1H,
H-5, J 7.98 Hz), 5.86–5.85 (m, 1H, H-3'), 6.11–6.09 (m, 1H, H-2'), 6.63
(s, 1H, OH), 6.87 (m, 1H, H-6), 7.30–7.24 (m, 8H, 2Bn), 7.59–7.52 (m,
2H, 4'',4'''-Bn). 31P NMR (DMSO-d6) d: 19.99. HRMS, m/z: 461.1233
[M+Na]+ (calc. for C23H23N2O5P, m/z: 461.1237).
O
O
P
O
P
Ph
Ph
Ph
P
Cl
Cl
Cl
OH
Cl
OEt
1
2
Scheme 1
†
Commercial reagents (Acros, Aldrich, and Fluka) were used; anhydrous
solvents were purified according to the standard procedures. Column chro-
matography was performed on Silica Gel 60 0.040–0.063 mm (Merck)
columns. Preparative liquid chromatography (PLC) was performed on
Silica Gel 60 F254. TLC was performed on Silica Gel 60 F254. NMR spectra
were recorded on a Bruker Avance 400 spectrometer (400 MHz for 1H and
162 MHz for 31P). HRMS spectra were measured on Bruker micrOTOF II
or maXis instruments using electrospray ionization.
Benzylphosphonic acid and ethyl benzylphosphonate chlorides 1 and 2.
Benzylphosphonic acid (800 mg, 4.6 mmol) or ethyl hydrogen benzyl-
phosphonate (500 mg, 2.5 mmol) were refluxed with thionyl chloride
(20–40 ml) in 1,2-dichloroethane (30–50 ml) for 4 h. The solvent was
then removed, the residue was dissolved in tetrachloromethane and twice
evaporated. Phosphonyl chlorides were used without further purification.
1-(4-Hydroxycyclopent-2-en-1-yl)uracil 3 was obtained according to
published procedures.18
‡
4-[2,4-Dioxo-3,4-dihydropyrimidin-1(2H)-yl]cyclopent-2-en-1-yl benzyl-
phosphonate 5. Benzylphosphonyl chloride (1.3 mmol) was added to
a solution of 1-(4-hydroxy-2-cyclopent-2-en-1-yl)uracil 3 (200 mg,
1.03 mmol) in dry pyridine (10 ml). The mixture was stirred for 18 h at
room temperature. After the solvent was evaporated and the residue was
separated by column chromatography on DEAE-cellulose (aq. NH4HCO3,
0–0.2 m) followed by column reversed phase chromatography (silica
gel C18, eluted with water), product 5 was obtained as a colourless syrup
(107 mg, 0.31 mmol, 30%). Rf = 0.35 (dioxane–aq. NH3, 4:1). 1H NMR
(DMSO-d6) d: 1.42–1.38 (m, 1H, Hb-5'), 2.63–2.59 (m, 1H, Ha-5'),
2.79–2.74 (m, 2H, CH2P), 4.96–4.93 (m, 1H, H-4'), 5.33–5.30 (m, 1H,
H-1'), 5.56–5.54 (d, 1H, H-5, J 7.98 Hz), 5.79–5.77 (m, 1H, H-3'),
6.08–6.06 (m, 1H, H-2'), 7.11–7.09 (m, 1H, H-6), 7.20–7.18 (m, 2H,
2'',6''-HPh), 7.23–7.26 (m, 2H, 3'',5''-HPh), 7.23–7.26 (m, 1H, 4''-HPh).
4-[2,4-Dioxo-3,4-dihydropyrimidin-1(2H)-yl)]cyclopent-2-en-1-yl benzyl
benzylphosphonate 8 (mixture of diastereomers) was obtained as a side
product in the synthesis of 7. The purification by PLC (CHCl3–MeOH,
9:1) gave product 8 as a pale yellow oil (9 mg, 20%). Rf = 0.36. 1H NMR
(CDCl3) d: 1.43–1.35 (m, 1H, Hb-5'), 2.64–2.60 (m, 1H, Ha-5'), 3.15–3.09
(m, 2H, CH2P), 5.01–4.93 (m, 2H, CH2O), 5.15–5.12 (m, 1H, H-4'),
5.46–5.44 (m, 1H, H-1'), 5.54–5.52 (d, 1H, H-5, J 8.05 Hz), 5.75–5.73
(m, 1H, H-3'), 5.95–5.91 (m, 1H, H-2'), 6.93–6.89 (m, 1H, H-6), 7.29–7.20
(m, 10H, 2Ph), 8.52 (s, 1H, NH). 31P NMR (DMSO-d6) d: 24.74, 24.91.
HRMS, m/z: 461.1236 [M+Na]+ (calc. for C23H23N2O5P, m/z: 461.1237).
– 115 –

Mendeleev Commun., 2016, 26, 114–116
Reaction of phosphonate 5 with excess of benzyl bromide
performed according to previously published method,15,16 where
activated DNA was used as the primer-template and the efficiency
of the enzyme in the elongation was estimated by incorporation
of radioactively labeled substrate, followed by calculation of KI
by Dixon method.19 KI for all of the tested compounds was more
than 100 mm, which means that no significant activity was found
in this experiment.
(1.3 equiv.) gave a mixture of products 7–9. Phosphonates 7–9
were isolated by preparative liquid chromatography, elution with
chloroform–methanol (9:1); yields of products 8 and 9 were
20 and 15%, respectively. To isolate target product 7 (yield
15%), we used additional elution with aqueous ammonia–dioxane
(4:1). Treatment of compound 6 with two-fold excess of benzyl
bromide furnished product 10 in 31% yield. Due to chirality of
the phosphorus atom, compounds 6, 8, 9 and 10 were mixtures
of diastereomers.
Thus, we have synthesized benzylphosphonates of 5'-nor-
carbocyclic analogue of 2',3'-dideoxy-2',3'-didehydrouridine and
its N3-benzyl derivatives with different substituents at the phos-
phorus atom. Compounds 8 and 9 carrying benzyl residue were
more hydrophobic (solubility was 400 mg in 100 ml of 30 and
33% DMSO in water, respectively) in comparison with 6 and 10
bearing ethoxy substituent at the phosphorus atom (solubility
was 400 mg in 100 ml of 18 and 20% DMSO in water, respec-
tively). The most hydrophilic were phosphonates 5 and 7 carrying
free hydroxyl group at the phosphorus atom (solubility was 400 mg
in 100 ml of 1 and 2% DMSO in water, respectively). Note that
solubility in water was poor even for compounds 5 and 7.
The inhibitory properties of the synthesized phosphonates
5–10 were evaluated in vitro against wild type HIV-1 reverse
transcriptase. Experiments to determine the inhibition (KI) were
References
1 http://www.who.int/mediacentre/factsheets/fs360/ru
2 J. A. Levy, HIV and the Pathogenesis of AIDS, 3rd edn., ASM Press,
Washington, DC, 2010.
3 Y. Song, Z. Fang, P. Zhan and X. Liu, Curr. Med. Chem., 2014, 21, 329.
4 E. De Clercq, Biochem. Pharmacol., 2011, 82, 99.
5 E. A. Shirokova, M. V. Jasko, A. L. Khandazhinskaya, A. V. Ivanov,
D. V. Yanvarev, Yu. S. Skoblov, V. A. Mitkevich, E. V. Bocharov, T. R.
Pronyaeva, N. V. Fedyuk, M. K. Kukhanova and A. G. Pokrovsky, J. Med.
Chem., 2004, 47, 3606.
6 A. L. Khandazhinskaya, D. V. Yanvarev, M. V. Jasko, A. V. Shipitsin,
V. A. Khalizev, S. I. Shram, Y. S. Skoblov, E. A. Shirokova and M. K.
Kukhanova, Drug Metab. Dispos., 2009, 37, 494.
7 E. A. Shirokova, M. V. Jasko, A. L. Khandazhinskaya, D. V. Yanvarev,
Y. S. Skoblov, T. R. Pronayeva, N. V. Fedyuk, A. G. Pokrovsky and
M. K. Kukhanova, Nucleosides Nucleotides Nucleic Acids, 2003, 22, 981.
8 A. Khandazhinskaya, E. Matyugina and E. Shirokova, Expert Opin.
Drug Metab. Toxicol., 2010, 6, 701.
9 A. L. Khandazhinskaya and E. A. Shirokova, Acta Naturae, 2013, 5 (3), 54.
10 M. Staake, J. Chauhan, D. Zhou, A. Shanker, A. De Chatterjee, S. Das
and S. E. Patterson, Org. Lett., 2010, 12, 4596.
4-[2,4-Dioxo-3-benzyl-3,4-dihydropyrimidin-1(2H)-yl]cyclopent-2-en-
1-yl benzyl benzylphosphonate 9 (mixture of diastereomers) was obtained as
a side product in the synthesis of 7. The purification by PLC (CHCl3–MeOH,
9:1) gave product 9 as a pale yellow oil (8 mg, 15%). Rf = 0.48. 1H NMR
(CDCl3) d: 1.39–1.36 (m, 1H, Hb-5'), 2.62–2.58 (m, 1H, Ha-5'), 3.13–3.03
(m, 2H, CH2P), 5.05–5.03 (m, 2H, CH2O), 5.00–4.94 (m, 2H, CH2O),
5.14–5.11 (m, 1H, H-4'), 5.59–5.44 (m, 1H, H-1'), 5.46–5.42 (d, 1H, H-5,
J 8.00 Hz), 5.72–5.71 (m, 1H, H-3'), 5.93–5.87 (m, 1H, H-2'), 6.88–6.84
(m, 1H, H-6), 7.24–7.21 (m, 15H, 3Ph). 31P NMR (DMSO-d6) d: 24.68,
24.86. HRMS, m/z: 551.1702 [M+Na]+ (calc. for C30H29N2O5P, m/z:
551.1706).
11 D. Zhou, M. Staake and S. E. Patterson, Org. Lett., 2008, 10, 2179.
12 D. H. Suk, D. Rejman, C. C. Dykstra, R. Pohl, K. W. Pankiewicz and
S. E. Patterson, Bioorg. Med. Chem. Lett., 2007, 17, 2811.
13 D. H. Suk, L. Bonnac, C. C. Dykstra, K. W. Pankiewicz and S. E. Patterson,
Bioorg. Med. Chem. Lett., 2007, 17, 2064.
14 D. Rejman, M. Olesiak, L. Q. Chen, S. E. Patterson, D. Wilson, H. N.
Jayaram, L. Hedstrom and K.W. Pankiewicz, J. Med. Chem., 2006, 49,
5018.
15 E. S. Matyugina, V. T. Valuev-Elliston, D. A. Babkov, M. S. Novikov,
A. V. Ivanov, S. N. Kochetkov, J. Balzarini, K. L. Seley-Radtke and
A. L. Khandazhinskaya, MedChemComm, 2013, 4, 741.
16 E. S. Matyugina, V. T. Valuev-Elliston, A. N. Geisman, M. S. Novikov, A. O
.
4-[2,4-Dioxo-3-benzyl-3,4-dihydropyrimidin-1(2H)-yl)]cyclopent-2-en-
1-yl ethyl benzylphosphonate 10 (mixture of diastereomers) was prepared
from reactant 6 using the procedure described for 7. Purification by
column chromatography on silica gel (CHCl3–MeOH, 98:2) afforded
product 10 as a colourless oil (25 mg, 31%). Rf = 0.42. 1H NMR (CDCl3)
d: 1.28–1.26 (m, 6H, 2Me), 1.47–1.45 (m, 1H, Hb-5'), 1.64–1.63 (m, 1H,
Hb-5'), 2.74–2.70 (m, 1H, Ha-5'), 2.84–2.82 (m, 1H, Ha-5'), 3.19–3.13
(m, 4H, 2CH2), 4.07–4.02 (m, 4H, 2CH2P), 5.14–5.09 (m, 4H, 2CH2Bn),
5.26–5.23 (m, 2H, 2H-4'), 5.60–5.55 (m, 2H, 2H-1'), 5.73–5.70 (m, 2H,
2H-5), 5.86–5.82 (m, 2H, 2H-3'), 6.00–5.98 (m, 1H, H-2'), 6.17–6.15
(m, 1H, H-2'), 7.03–7.01 (m, 2H, 2H-6), 7.26–7.22 (m, 16H, 2Ph),
7.48–7.46 (m, 4H, 2Ph). 31P NMR (DMSO-d6) d: 24.26, 24.14. HRMS,
m/z: 489.1548 [M+Na]+ (calc. for C25H27N2O5P, m/z: 489.1550).
Chizhov, S. N. Kochetkov, K. L. Seley-Radtke andA. L. Khandazhinskaya,
MedChemComm, 2013, 4, 1443.
17 V. Hegde, K. Seley and S. Schneller, Nucleosides Nucleotides Nucleic
Acids, 2000, 19, 269.
18 E. Matyugina, A. Khandazhinskaya, L. Chernousova, S. Andreevskaya,
T. Smirnova, A. Chizhov, I. Karpenko, S. Kochetkov and L. Alexandrova,
Bioorg. Med. Chem., 2012, 20, 6680.
19 M. Dixon, Biochem. J., 1953, 55, 170.
Received: 14th October 2015; Com. 15/4750
– 116 –