Synthesis and anticancer activity of 5'-chloromethylphosphonates of 3'-azido-3'-deoxythymidine (AZT)

Article abstract of DOI:10.1016/j.bmc.2011.08.069

A series of novel N-alkyl 5'-chloromethylphosphonates of 3'-azido-3'-deoxythymidine (6-15) was synthesized by means of phosphonylation of 3'-azido-3'-deoxythymidine (4) with P-chloromethylphosphonic ditriazolide (3) followed by a reaction with the appropriate amine. The synthesized phosphonamidates 6-15 were evaluated for their cytotoxic activity in two human cancer cell lines: oral (KB) and breast (MCF-7) using the sulforhodamine B (SRB) assay. The highest activity in KB human cancer cells was displayed by phosphonamidate 8 (IC₅₀ = 5.8 μg/mL), however, this compound was less potent than the parent AZT (IC₅₀ = 3.1 μg/mL). Phosphonamidate 10 showed only moderate activity (IC₅₀ = 12.1 μg/mL) whereas the other phosphonamidates proved inactive. Similarly, the highest activity in MCF-7 human cancer cells was displayed by phosphonamidate 8 (IC₅₀ = 3.7 μg/mL) but it proved somewhat less active than AZT (IC₅₀ = 2.6 μg/mL). Some activity was also displayed by phosphonamidate 10 (IC₅₀ = 12.8 μg/mL) but the other phosphonamidates were found inactive. Hydrolysis studies indicate that the synthesized phosphonamidates are likely to act as prodrugs of the parent nucleoside (AZT). Transport measurements showed that the most active phosphonamidates (8 and 10) were able to permeate across the intestinal epithelium in vitro. The apparent permeability coefficients determined in Caco-2 cell monolayers indicated that these compounds could be moderately absorbed in humans.

Full text of DOI:10.1016/j.bmc.2011.08.069

Bioorganic & Medicinal Chemistry 19 (2011) 6375–6382
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis and anticancer activity of 50-chloromethylphosphonates
of 30-azido-30-deoxythymidine (AZT)
Lech Celewicz a, , Agnieszka Józwiak , Piotr Ruszkowski , Halina Laskowska , Anna Olejnik ,
⇑
a
b
b
c
´
a
a
b
´
Anna Czarnecka , Marcin Hoffmann , Bogusław Hładon
a Faculty of Chemistry, Adam Mickiewicz University, Grunwaldzka St. 6, 60-780 Poznan´, Poland
b Department of Pharmacology, Poznan´ University of Medical Sciences, Rokietnicka St. 5a, 60-806 Poznan´, Poland
c
´
´
Department of Biotechnology and Food Microbiology, Poznan University of Life Sciences, Wojska Polskiego St. 48, 60-627 Poznan, Poland
a r t i c l e i n f o
a b s t r a c t
A series of novel N-alkyl 50-chloromethylphosphonates of 30-azido-30-deoxythymidine (6–15) was syn-
thesized by means of phosphonylation of 30-azido-30-deoxythymidine (4) with P-chloromethylphosphon-
ic ditriazolide (3) followed by a reaction with the appropriate amine. The synthesized phosphonamidates
6–15 were evaluated for their cytotoxic activity in two human cancer cell lines: oral (KB) and breast
(MCF-7) using the sulforhodamine B (SRB) assay. The highest activity in KB human cancer cells was dis-
Article history:
Received 24 January 2011
Revised 29 August 2011
Accepted 30 August 2011
Available online 5 September 2011
played by phosphonamidate 8 (IC50 = 5.8
ent AZT (IC50 = 3.1 g/mL). Phosphonamidate 10 showed only moderate activity (IC50 = 12.1
whereas the other phosphonamidates proved inactive. Similarly, the highest activity in MCF-7 human
cancer cells was displayed by phosphonamidate 8 (IC50 = 3.7 g/mL) but it proved somewhat less active
than AZT (IC50 = 2.6 g/mL). Some activity was also displayed by phosphonamidate 10 (IC50 = 12.8
g/mL) but the other phosphonamidates were found inactive. Hydrolysis studies indicate that the
l
g/mL), however, this compound was less potent than the par-
Keywords:
l
l
g/mL)
N-Alkyl 50-chloromethylphosphonates of 30-
azido-30-deoxythymidine
Phosphonylation
l
l
Cytotoxic activity
Cancer cell lines
l
KB and MCF-7
Caco-2 permeability assay
synthesized phosphonamidates are likely to act as prodrugs of the parent nucleoside (AZT). Transport
measurements showed that the most active phosphonamidates (8 and 10) were able to permeate across
the intestinal epithelium in vitro. The apparent permeability coefficients determined in Caco-2 cell
monolayers indicated that these compounds could be moderately absorbed in humans.
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
area of research.8–12 The 50-monophosphate of AZT itself can not
be employed as a prodrug since it is negatively charged at physio-
Several nucleoside analogues have found important use as anti-
viral1 and anticancer2–4 therapeutics. Notably, 30-azido-30-deoxy-
thymidine (AZT, zidovudine) has been developed as an anti-HIV
drug for the treatment of acquired immunodeficiency syndrome
(AIDS).5 On the other hand, there are some examples of application
of AZT as an antitumor agent in combination with either cisplatin,
methotrexate or 5-fluorouracil in the therapy of advanced colon
cancer.6 In addition, Wagner reported the potent inhibitory activity
of AZT in cultured human breast cancer cells.7 It is assumed that
the mechanism of anticancer action of AZT involves its intracellular
conversion to the 50-triphosphate (via the corresponding mono-
and diphosphate). The 50-triphosphate of AZT acts as a competitive
inhibitor of DNA polymerases and a chain terminator of the nascent
DNA strand due to the lack of a 30-hydroxyl group.6 The develop-
ment of AZT prodrugs which can easily penetrate the lipid-rich cell
membranes and then undergo hydrolysis to the 50-monophosphate
or its analogues has therefore become an increasingly important
logical pH and consequently too polar to cross the cell
membranes.9 Moreover, the blood and cell surface phosphohydro-
lases effectively convert nucleoside 50-phosphates to the parent
nucleosides.9
A number of AZT monophosphate prodrugs was synthesized
and evaluated for their antiviral and anticancer activity.13 Meier
developed the so-called cycloSal-pronucleotide approach using
the bifunctional salicylic alcohol to protect the phosphate group.14
McGuigan found that some alkyloxy phosphoramidates,15,16 diaryl
phosphates17 and aryloxy phosphoramidate diesters18,19 of AZT
were effective pronucleotides12 whereas amino acid phosphoram-
idate monoesters20,21 of AZT were investigated by Wagner.9,11 Péri-
gaud elaborated S-acyl-2-thioethyl (SATE) phosphotriesters22,23
and phosphoramidate diesters24 of AZT as useful pronucleotides.25
It should be noted that there were also several attempts to
synthesize phosphonate prodrugs of AZT. Thus, Rosowsky pub-
lished the synthesis and antiretroviral activity of 50-phosphonofor-
mates of AZT having the carboxyl group esterified with long chain
alcohols.26 Other 50-phosphonoformate esters of AZT incorporating
S-(pivaloyl)thioethyl (t-butyl-SATE) group were described by
⇑
Corresponding author.
E-mail address: celewicz@amu.edu.pl (L. Celewicz).
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.08.069

6376
L. Celewicz et al. / Bioorg. Med. Chem. 19 (2011) 6375–6382
O
P
O
P
N
N
a
N
ClCH2
Cl
+
N
ClCH2
N
N
H
N
Cl
N
1
2
N
3
O
N
O
N
O
N
CH3
CH3
CH3
HN
HN
HN
O
O
P
N
O
O
O
b
c
ClCH2
P
N
O
ClCH2
R1
O
HO
O
O
O
R2
N
N
N3
N3
N3
4
5
6 - 15
d
6: R1 = H, R2 = CH3
7: R1 = H, R2 = CH2CH3
8: R1 = H, R2 = CH2CH2CH3
9: R1 = H, R2 = CH2CH2CH2CH3
O
CH3
HN
10: R1 = H, R2 = CH(CH3)2
11: R1 = H, R2 = CH2CH=CH2
12: R1 = H, R2 = CH2CH2Ph
13: R1 = H, R2 = CH2CH2-(3-indolyl)
14: R1 = CH3, R2 = CH3
O
P
O
N
ClCH2
O
O
O
HNEt3
15: R1 = CH2CH3, R2 = CH2CH3
N3
16
Scheme 1. Synthesis of the 50-chloromethylphosphonates of 30-azido-30-deoxythymidine. Reagents and conditions: (a) NEt3, CH3CN, rt; (b) 3, pyridine, rt; (c) R1–NH–R2, rt;
(d) H2O, pyridine, NEt3, rt.
Meier.27 Shirokova reported the synthesis and anti-HIV activity of
two series of 50-phosphonoformate and 50-phosphonoacetate
derivatives of AZT.28 Recently, Kraszewski examined dinucleoside
31P NMR spectra of products 6–15 revealed the presence of two
diastereoisomers due to a chiral center being formed at the phos-
phorus atom. There were two close signals, in the ratio of approx-
imately 1:1, in each 31P NMR spectrum. Also thin layer
chromatography of compounds 6–15 was consistent with the pres-
ence of two diastereoisomers showing two overlapping spots but
we were unable to resolve them by silica gel column chromatogra-
phy. However, it was possible to resolve the two diastereoisomers
by HPLC on reversed-phase column (see Experimental section,
compound 8 and 10).
It is worth mentioning that the use of P-chloromethylphosphonic
dichloride (1), rather than its triazolide counterpart 3, resulted in the
formation of a considerable amounts of the symmetrical (50–50)
dinucleoside phosphonate. Application of 2- and 4-chlorophenyl
phosphoroditriazolides31–33 for the phosphorylation and methyl-
phosphonic ditriazolide34–36 for the phosphonylation of 50-pro-
tected nucleosides has been reported in the phosphotriester
synthesis of oligonucleotides and provided an inspiration for the
development of our method.
It should be noted that P-chloromethyl group remains intact
throughout the synthesis of phosphonamidates 6–15. Apparently,
nucleophilic substitution of the chlorine atom at the P-chloro-
methyl group by 1,2,4-triazole anion or alkyl amines does not take
place during the course of the reaction. However, our attempts to
perform nucleophilic substitution reaction on phosphonamidate
6 with azide or iodide anion (using sodium azide or iodide in
DMF at 100 °C) were unsuccessful. Below 90 °C the reaction did
not proceed but at higher temperatures it resulted in the mixture
of products due to attack of the nucleophiles on the 50-carbon atom
of 6, which leads to the cleavage of the ester bond. Reactivity of the
P-chloromethyl group towards nucleophilic substitution is
a
phosphonate–phosphates29 and 50- -hydroxyphosphonates30 of
AZT as potential anti-HIV agents.
The aim of our study was to synthesize novel phosphonate pro-
drugs of AZT with potential anticancer or antiviral properties. In
this paper, we report the synthesis of N-alkyl 50-chloro-
methylphosphonates of 30-azido-30-deoxythymidine (AZT) (6–15)
and evaluation of their cytotoxic activity in two human cancer cell
lines: oral (KB) and breast (MCF-7). Moreover, chemical stability
and permeability of the most active compounds across intestinal
barrier in vitro was investigated.
2. Results and discussion
2.1. Chemistry
A series of novel phosphonamidates 6–15 was synthesized by
phosphonylation of 30-azido-30-deoxythymidine (4) with P-chlo-
romethylphosphonic ditriazolide (3) according to the procedure
outlined in Scheme 1. P-Chloromethylphosphonic ditriazolide (3)
was prepared by reaction of P-chloromethylphosphonic dichloride
(1) with 1,2,4-triazole (2) in the presence of triethylamine in ace-
tonitrile. Reaction of compound 3 with 30-azido-30-deoxythymi-
dine (4) in the presence of pyridine afforded the key
intermediate 5, which, in situ, was treated with the appropriate
amine (or amine hydrochloride in the presence of triethylamine)
to give desired products 6–15 in 70–85% yield. Treatment of inter-
mediate 5 with water, in the presence of triethylamine and pyri-
dine, afforded P-chloromethylphosphonate 16.

L. Celewicz et al. / Bioorg. Med. Chem. 19 (2011) 6375–6382
6377
enhanced in phosphonate esters bearing electron-withdrawing
groups (e.g., trifluoroethyl).37,38
diphosphate.43 Since N-alkyl P-methylphosphonates of AZT did not
exhibit any significant cytotoxic activity (unpublished results) we
turned our attention to the P-chloromethylphosphonates. The
P-chloromethyl group is more polar then the P-methyl and to some
extent can mimic the hydroxyl on a phosphorus atom.44 Moreover,
we anticipated that the P-chloromethylphosphonate could be con-
verted to the P-hydroxymetylphosphonate and then oxidized to the
P-carboxylphosphonate which after decarboxylation is transformed
to the phosphate via the transient H-phosphonate.26,28 From phos-
phonamidemoietyweexpectedmorestabilitythanphosphonodiest-
er could give to the structure of the target compound. Furthermore,
lipophilicity ofthe phosphonamidates can be modulated bychanging
the nature of N-alkyl substituents.
Although the biochemical mode of action of the AZT phospho-
namidates requires more detailed studies, we think that these
compounds, due to their nonionic character, can penetrate the cell
membranes as indicated by encouraging values of their partition
coefficients (Table 1). Once inside the cells, the phosphonamidates
could be hydrolyzed or metabolized to AZT either directly or via
50-chloromethylphosphonate 16. Finally, AZT after enzymatic
phosphorylation to its 50-triphosphate, can act as an inhibitor of
DNA polymerases.
2.2. In vitro cytotoxicity
The synthesized phosphonamidates 6–15 were tested for their
cytotoxic activity in two human cancer cell lines: oral (KB) and
breast (MCF-7) employing sulforhodamine B (SRB) assay.39 The
highest activity in KB human cancer cells was displayed by phos-
phonamidate 8 (IC50 = 5.8
parent AZT (IC50 = 3.12 g/mL). Phosphonamidate 10 showed only
moderate activity (IC50 = 12.1 g/mL) whereas the other phospho-
namidates proved inactive. Likewise, the highest activity in MCF-7
human cancer cells was displayed by phosphonamidate
lg/mL), however, it was less potent than
l
l
8
(IC50 = 3.7
(IC50 = 2.6
idate 10 (IC50 = 12.8
l
l
g/mL) which was somewhat less active than AZT
g/mL). Some activity was also shown by phosphonam-
g/mL) whereas the other phosphonamidates
l
were inactive. The findings show that the phosphonamidate 8 with
N-n-propyl substituent was the most potent compound in both the
cancer cell lines (KB and MCF-7) and its activity is comparable with
that of the parent AZT. Also phosphonamidate 10 with the N-i-pro-
pyl substituent exhibited moderate activity in both the cell lines.
However phosphonamidate 9 with a longer N-n-butyl chain substi-
tuent was 13-fold less potent than 8 in KB cancer cells and 22-fold
less potent than 8 in MCF-7 cancer cells. It is possible several rea-
sons for this. The phosphonamidate 9 could become so hydropho-
bic that it is poorly soluble in aqueous phase. Alternatively, it could
be caught by fat depots and never reach the intended site. Finally,
hydrophobic compounds are often more susceptible to metabolism
and subsequent elimination.40
Partition coefficient (logP) values of the compounds 6–15 were
calculated41 to determine a possible correlation between the cyto-
toxicity data and lipophilicity (Table 1). All of the AZT phospho-
namidates were more lipophilic than AZT (logP = 0.06)30, with
logP values ranging from 0.11 to 2.37. The most active compounds
8 and 10 showed moderate values of logP, 1.08 and 1.02 respec-
tively. However, linear regression analysis did not reveal any corre-
lation between logP values and the cytotoxicity data.
2.3. Chemical stability
Chemical stability of the most active compounds 8 and 10 was
studied in the phosphate buffer (pH 7.2, 7.4 and 7.8) at 37 °C and
the concentration of the compound under study of 0.5 mM
(Table 2). The composition of the reaction mixtures was analyzed
by HPLC on reversed-phase column. It was established that the
hydrolysis of both compounds 8 and 10 at pH 7.2 was very slow
and after 7 h only a small fraction of each compound (6% and 4%
respectively) underwent hydrolysis with formation of AZT as the
only product. At pH 7.4 the hydrolysis was somewhat faster, 17%
and 13% of compound 8 and 10, respectively was hydrolyzed after
7 h. This time, AZT was detected as a major product and also a
small amount of phosphonate 16 was formed. A further increase
of the pH to 7.8 resulted in about 50% hydrolysis of compound 8
after 7 h; AZT was detected as a major product and phosphonate
16 as a minor one. The amount of phosphonate 16 did not exceed
11% of the total amount of products. The hydrolysis of compound
10 at pH 7.8 was slower than that of 8 and only 36% of 10 reacted
after 7 h. Authentic samples of AZT and phosphonate 16 were used
to identify the products.
Beforeweproposeamechanismofactionofthe targetcompounds
some explanation for their design rationale is needed. We chose
phosphonate compounds with relatively stable P–C bond and non-
ionic character, to ensure sufficient lipophilicity and hydrolytic sta-
bility. Some of them can be further phosphorylated to triphosphate
analogues. For example, anti-HIV drug tenofovir (a phosphonate) is
phosphorylated bycellular kinases toitsactive metabolite—tenofovir
More detailed stability studies of compound 8 and 10 at pH 7.8
are presented in Fig. 1.
The hydrolysis studies suggest that, the synthesized phospho-
namidates act as depot forms of AZT rather than 50-phosphonate
prodrugs of AZT.28
Table 1
In vitro cytotoxic activity of the synthesized compounds 6–15 in two human cancer
cell lines: oral (KB) and breast (MCF-7)
Compound
Cytotoxicity (IC50
,
l
g/mL)a SDb
LogPc
2.4. Caco-2 permeability assay
KB
MCF-7
6
7
8
9
10
11
12
13
52.0 1.7
>100
63.2 1.4
>100
0.11
0.60
1.09
1.58
1.02
0.96
2.31
2.37
0.66
1.64
In view of their cytotoxic activity and chemical stability, poten-
tial oral absorption of phosphonamidates 8 and 10 was also evalu-
ated. In order to determine their permeability across the intestinal
barrier we used the Caco-2 cell culture system. The expected good
5.8 0.8
77.0 3.5
12.1 1.9
47.0 4.1
>100
80.0 4.9
65.0 1.8
85.0 1.7
3.12 0.26
0.99 0.02
3.7 0.7
82.0 3.1
12.8 1.6
34.0 3.1
>100
Table 2
>100
Hydrolysis of compounds 8 and 10 in phosphate buffer (pH 7.2, 7.4 and 7.8) at 37 °C
after 7 h
14
15
22.8 1.2
67.0 2.5
2.60 0.25
0.93 0.06
Zidovudine (AZT)
Cytarabine (standard)
0.0630
pH
Amount of the compound consumed (%)
ꢀ2.3241
8
10
a
b
c
IC50 is the compound concentration required to inhibit cell growth by 50%.
SD (standard deviation) of three independent experiments.
LogP (logarithm of partition coefficient) was calculated using ‘log PKnowwin
7.2
7.4
7.8
6
17
50
1
3
5
4
13
36
1
2
4
’
method.42

6378
L. Celewicz et al. / Bioorg. Med. Chem. 19 (2011) 6375–6382
0.16
AB transport
BA transport
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
0
20
40
60
80
100
120
Time [min]
Fig. 2. Cumulative fraction of phosphonamidate 8 in bidirectional AB (absorptive)
and BA (exsorptive) transport across Caco-2 cell monolayer.
transported (CTF) in the receiver compartment increased linearly
with time, independently on transport direction (Fig. 2). The
time-course experiments showed that the CFT of compounds 8
and 10 was significantly higher in the secretory direction (BA
transport) than in the absorptive one (AB transport) (Fig. 2).
It was reflected in the apparent permeability coefficient (Papp
)
values determined for compounds transported in AB (Papp,AB) and
BA direction (Papp,BA). As shown in Table 3, the Papp,BA values were
approximately fivefold higher than the Papp,AB values suggesting
that secretory transport rather than absorptive transport of phos-
phonamidates 8 and 10 is predominant in the Caco-2 system.
These results indicate that phosphonamidates tested, similarly to
the parent AZT, may be a substrate for efflux carrier modulating
its absorptive permeability. The current data show that AZT efflux
in the intestine is mainly mediated by the ATP-binding cassette
(ABC) family of proteins, suggesting the intervention of P-glycopro-
tein (P-gp), multi-drug resistance proteins (MRP-2, MRP-4) and
breast cancer-related protein (BCRPs) in these phenomena.47–51
In spite of the impact of efflux mechanism in phosphonami-
dates 8 and 10 absorption, the apical-to-basolateral permeability
coefficients (3.22 and 3.36 ꢁ 10ꢀ6 cm/s, respectively) determined
in the Caco-2 cell assay, suggest relatively high oral availability.
The drug substances with Papp values ranging from 1 to 10 ꢁ
10ꢀ6 cm/s can be classified as moderately (20–70%) absorbed com-
pounds.46 In vitro permeability of phosphonamidates 8 and 10 is
comparable to permeability of the parent AZT reported in the liter-
ature. The Papp values of AZT determined in the Caco-2 cell assay
are usually in the range of 2.85 to 8.5 ꢁ 10ꢀ6 cm/s52–55 and confirm
moderate oral bioavailability of this drug (63 10%).56 A better
prediction of phosphonamidates (8 and 10) absorption in humans
could be obtained by using a P-gp inhibitor in bidirectional perme-
ability studies. The inclusion of a P-gp inhibitor would probably re-
sult in increased absorptive and decreased secretory transport of
the compounds tested.
Fig. 1. Hydrolysis of compound 8 (a) and 10 (b) at 0.5 mM initial concentration in
phosphate buffer (pH 7.8) at 37 °C.
correlation between permeability in vitro and oral drug absorption
in the human was based on previous studies in this cell line.45,46
Prior to carrying out a transport study, the cytotoxicity and the
effect of compounds tested on Caco-2 monolayer integrity and
intestinal barrier properties was examined. No cytotoxicity of sub-
stances at 100 lM concentration in transporting medium after a
2 h exposure was observed. The transepithelial electrical resistance
(TEER) measurement indicated high integrity of Caco-2 cell mono-
layer in both cultures, before and after transport experiments (data
not shown).
The transport of phosphonamidates 8 and 10 was evaluated at a
target concentration of 100 lM and was initiated by adding the
test solution to the apical (A) compartment for absorptive
transport or basolateral (B) compartment for exsorptive transport
determination. The solution from acceptor compartment was sam-
pled at 15 min intervals and the concentration of a transported
compound was analyzed. The results indicated that phosphonam-
idates 8 and 10 were able to cross the Caco-2 monolayer in both
absorptive and secretory direction and no significant difference
in their permeability was observed. The cumulative fraction
Table 3
The apparent permeability coefficients of phosphonamidates 8 and 10 and the parent nucleoside (zidovudine, AZT)
a
c
Compound
Papp (ꢁ10ꢀ6 cm/s)a SDb
Log Papp
LogtPapp
A–B
B–A
A–B
B–A
8
10
3.22 0.76
3.36 0.64
2.85 0.0754
4.0 0.252
16.10 0.33
18.00 0.47
ꢀ5.49
ꢀ4.79
ꢀ4,75
ꢀ5.53
ꢀ5.52
ꢀ5.16
ꢀ5.47
Zidovudine (AZT)
ꢀ5.5554
ꢀ5.4052
ꢀ5.1653
ꢀ5.0755
6.93 0.1753
8.5 0.255
a
b
c
Papp (apparent permeability coefficient) was determined using method described by Tavelin et al.58, A – apical side, B – basolateral side.
SD (standard deviation) of three independent experiments.
tPapp (theoretical value of Papp) calculated by method described by Guangli and Yiyu.57

L. Celewicz et al. / Bioorg. Med. Chem. 19 (2011) 6375–6382
6379
It is noteworthy that the Papp,AB values determined experimen-
tally for both the compounds in the Caco-2 system were compara-
ble to theoretically calculated ones (tPapp) using the method
described by Guangli and Yiyu.57 The log Papp values, experimental
as well as calculated, are presented in Table 3.
The reaction mixture was stirred at room temperature for a further
1 h and the appropriate amine (4.57 mmol) or amine hydrochlo-
ride (4.57 mmol) and triethylamine (694 mg, 6.86 mmol) was
added. After 1 h, the reaction mixture was evaporated under
reduced pressure. To the residue was added saturated aqueous
sodium bicarbonate (10 mL) and the mixture was extracted with
chloroform (3 ꢁ 10 mL). The combined chloroform extracts were
washed with water (10 mL), dried over anhydrous magnesium sul-
fate, filtered and evaporated to dryness. The residue was purified
by silica gel column chromatography using chloroform – methanol
(50:1, v/v) as eluent to give pure products 6–15 (yield 70–85%).
3. Conclusion
In summary, we have developed an efficient method for the syn-
thesis of N-alkyl 50-chloromethylphosphonates of 30-azido-30-deox-
ythymidine employing P-chloromethylphosphonic ditriazolide as
phosphonylating agent. P-Chloromethylphosphonic ditriazolide
was more selective than its dichloro counterpart and its use did
not result in the formation of symmetrical (5–50)dinucleoside phos-
phonates. It was established that the P-chloromethyl group in phos-
phonamidates 6–15 was chemically stable and did not undergo
reactions with nucleophiles, such as amines and azide anion, at
room temperature. The obtained compounds 6–15 were examined
for their cytotoxic activity in two human cancer cell lines: oral
(KB) and breast (MCF-7). The highest activity in KB human cancer
cells was displayed by phosphonamidate 8 with the N-n-propyl sub-
4.1.1.1.
methylphosphonate) (6).
30-Azido-30-deoxythymidine 50-(N-methyl-P-chloro-
1H NMR (CDCl3): d 1.84 (d, 3H,
J = 1.0 Hz, 5-CH3); 2.13–2.32 (m, 2H, H-200, H-20); 2.64–2.72 (m,
3H, N–CH3); 3.27–3.41 (m, 3H, P–CH2–Cl, NH–C); 3.47–3.58 (m,
1H, H-40); 4.20–4.35 (m, 3H, H-50, H-500, H-30); 6.24 (t, 1H,
J = 6.1 Hz, H-10); 7,43 (d, 1H, J = 1.1 Hz, H-6); 9.51 (s, 1H, 3-NH).
13C NMR (CDCl3): d 12.55 (5-CH3); 27.06 (N–CH3); 35.28 (d,
JP-C = 155 Hz, P–CH2–Cl); 37.11 (C-20); 60.15 (C-30); 63.07 (C-50);
82.41 (C-10); 85.47 (C-40); 111.22 (C-5); 135.80 (C-6); 150.50 (C-
2); 166.40 (C-4). 31P NMR (CDCl3): d 25.28; 25.34. MS-ESI m/z:
393; 395 [M+H]+; 415; 417 [M+Na]+; 391; 393 [MꢀH]ꢀ; 427;
429; 431 [M+Cl]ꢀ. Anal. Calcd for C12H18ClN6O5P: C, 36.70; H,
4.62; N, 21.40. Found: C, 36.72; H, 4.61; N, 21.42.
stituent (IC50 = 5.8
(IC50 = 3.1 g/mL). Similarly, the highest activity in MCF-7 human
cancer cells was displayed by phosphonamidate 8 (IC50 = 3.7
lg/mL) but it was less potent than the parent AZT
l
l
l
g/mL) which was somewhat less potent than AZT (IC50 = 2.6
g/mL). In view of our hydrolysis studies, the synthesized phospho-
30-Azido-30-deoxythymidine 50-(N-ethyl-P-chloro-
1H NMR (CDCl3): d 1.20 (t, 3H,
namidates are most probably the prodrug forms of the parent AZT.
The results of transport studies suggest that phosphonamidates 8
and 10 with Papp values of 3.22 and 3.36 ꢁ 10ꢀ6 cm/s, respectively,
can be considered as moderately absorbed in humans.
4.1.1.2.
methylphosphonate) (7).
J = 7.0 Hz, N-C-CH3); 1.93 (d, 3H, J = 3.0 Hz, 5-CH3); 2.33–2.42 (m,
2H, H-20, H-200); 3.06–3.25 (m, 3H, N-CH2-C, NH-Et); 3.60–3.71
(m, 2H, P-CH2-Cl); 4.04–4.08 (m, 1H, H-40); 4.21–4.40 (m, 2H, H-
50, H-500); 4.42–4.51 (m, 1H, H-30); 6.14 and 6.21 (t, 1H, J = 6.5 Hz,
H-10); 7.38 (s, 1H, H-6); 9.54 (s, 1H, 3-NH). 13C NMR (CDCl3): d
4. Experimental section
4.1. Chemistry
12.48 (5-CH3); 17.74 (N-C-CH3); 35.18 (d, JP-C = 155 Hz, P–CH2–
2
Cl); 36.27 (N–CH2–C); 37.24 (C-20); 60.53 (C-30); 63.10 (d, JP-C
=
6.0 Hz, C-50); 82.30 (C-10); 85.44 (C-40); 111.54 (C-5); 135.64 (C-
6); 150.30 (C-2); 163.86 (C-4). 31P NMR (CDCl3): d 24.75; 24.87.
MS-ESI m/z: 407; 409 [M+H]+; 429; 431 [M+Na]+; 445; 447
[M+K]+; 405; 407 [M-H]ꢀ; 441; 443; 445 [M+Cl]ꢀ. Anal. Calcd for
NMR spectra were recorded on a Varian-Gemini 300 MHz spec-
trometer. Chemical shifts (d) are reported in ppm relative to the
tetramethylsilane (TMS) peak. For 31P NMR spectra 85% phospho-
ric(V) acid in D2O was used as an external standard (coaxial inner
tube). Mass spectra were measured on a Waters Micromass ZQ
electrospray (ES) mass spectrometer. Elemental analyses were per-
formed on EL III elemental analyzer (Elementar Analysensysteme
GmbH, Germany). Thin layer chromatography (TLC) was per-
formed on silica gel 60 F254 precoated (0.2 mm) plates and vacuum
C13H20ClN6O5P: C, 38.39; H, 4.96; N, 20.66. Found: C, 38.43; H,
4.97; N, 20.64.
4.1.1.3. 30-Azido-30-deoxythymidine 50-(N-n-propyl-P-chloro-
methylphosphonate) (8).
1H NMR (CDCl3): d 0.94 (t, 3H,
flash column chromatography on silica gel 60 H (5–40 lm) pur-
J = 7.3 Hz, N–C–C–CH3); 1.55 (sextet, 2H, J = 7.3 Hz, N–C–CH2–C);
1.93 (d, 3H, J = 1.1 Hz, 5-CH3); 2.33–2.47 (m, 2H, H-20, H-200);
2.95–3.05 (m, 2H, N–CH2–C–C); 3.28–3.36 (m, 1H, NH–Pr);
3.57–3.68 (m, 2H, P–CH2–Cl); 4.02–4.10 (m, 1H, H-40); 4.17–4.51
(m, 3H, H-50, H-500, H-30); 6.14 and 6.21 (t, 1H, J = 6.6 Hz, H-10);
7.32 and 7.38 (d, 1H, J = 1.1 Hz, H-6); 9.75 (s, 1H, 3-NH). 13C NMR
(CDCl3): d 11.01 (N–C–C–CH3); 12.45 (5-CH3); 25.29 (N–C–CH2–
C); 35.06 (d, JP-C = 142 Hz, P–CH2–Cl); 37.21 (C-20); 43.08 (N–CH2–
C–C); 60.53 (C-30); 63.61 (C-50); 82.29 (C-10); 85.43 (C-40); 111.52
(C-5); 135.59 (C-6); 150.34 (C-2); 163.88 (C-4). 31P NMR (CDCl3): d
25.06; 25.23. MS-ESI m/z: 421; 423 [M+H]+; 443; 445 [M+Na]+;
459; 461 [M+K]+; 419; 421 [M-H]ꢀ; 455; 457; 459 [M+Cl]ꢀ. Anal.
Calcd for C14H22ClN6O5P: C, 39.96; H, 5.27; N, 19.97. Found: C,
39.98; H, 5.28; N, 19.96. HPLC: eluent (a), retention time (tR) of
15.6 and 17.2 min in the ratio 1:1; eluent (c), retention time (tR) of
13.9 and 15.4 min in the ratio 1:1.
chased from Merck. High performance liquid chromatography
(HPLC) was performed on a Waters chromatograph equipped with
a Waters 996 UV-Vis photodiode array detector. Analytical HPLC
was carried out on Waters XTerra RP18 reversed-phase column
(4.6 ꢁ 150 mm, 5
lm) using three eluting systems: (a) water–
methanol (70:30), (b) 25 mM ammonium formate (pH 7.0)–
methanol (80:20) and (c) phosphate buffer (20 mM Na2HPO4, pH
was adjusted to 7.1 with H3PO4)–acetonitrile (80:20). Water con-
tained 5% methanol to prevent bacterial growth. The flow rate
was 1 mL/min and detection at 266 nm. Chemical reagents were
purchased from Sigma-Aldrich. 30-Azido-30-deoxythymidine was
prepared based on the literature procedure.59
4.1.1. General procedure for the preparation of compounds
6–15
To
a solution of P-chloromethylphosphonic dichloride (1)
4.1.1.4.
methylphosphonate) (9).
J = 7.2 Hz, N–C–C–C–CH3); 1.35 (sextet, 2H, J = 7.4 Hz, N–C–C–CH2–
C); 1.43–1.56 (m, 2H, N–C–CH2–C–C); 1.93 (d, 3H, J = 1.4 Hz, 5-CH3);
2.21–2.49 (m, 2H, H-20, H-200); 2.97–3.08 (m, 2H, N–CH2–C–C–C);
30-Azido-30-deoxythymidine 50-(N-n-butyl-P-chloro-
1H NMR (CDCl3): d 0.92 (t, 3H,
(153 mg, 0.914 mmol) in acetonitrile (2 mL) was added 1,2,4-tria-
zole (2) (164 mg, 2.374 mmol) followed by triethylamine
(189 mg, 1.865 mmol) and the reactants were stirred for 30 min
at room temperature. Then to the mixture 30-azido-30-deoxythymi-
dine (4) (100 mg, 0.374 mmol) and pyridine (2.30 mL) were added.

6380
L. Celewicz et al. / Bioorg. Med. Chem. 19 (2011) 6375–6382
3.30–3.40 (m, 1H, NH–C–C–C–C); 3.57–3.72 (m, 2H,
P–CH2–Cl); 4.02–4.17 (m, 1H, H-40); 4.18–4.27 (m, 1H, H-30); 4.32–
4.51 (m, 2H, H-50, H-500); 6.14 and 6.21 (t, 1H, J = 6.8 Hz, H-10); 7.32
and 7.38 (d, 1H, J = 1.4 Hz, H-6); 9.87 (s, 1H, 3-NH). 13C NMR (CDCl3):
d 7.19 (5-CH3); 8.36 (N–C–C–C–CH3); 14.43 (N–C–C–CH2–C); 28.91
(N–C–CH2–C–C); 29.80 (d, JP-C = 144 Hz, P-CH2-Cl); 31.96 (C-20);
2H, N–C–CH2–ind); 3.09–3.20 (m, 1H, NH–C–C–ind); 3.26–3.53
(m, 5H, P–CH2–Cl, N–CH2–C–ind, H-40); 3.65–3.78 (q, 1H,
J = 12.5 Hz, H-30); 3.98–4.17 (m, 2H, H-50, H-500); 6.13 (t, 1H,
J = 6.7 Hz, H-10); 6.94–7.56 (m, 6H, H-6, ind); 10.83 (s, 1H, NH–
ind); 11.38 (s, 1H, 3-NH). 13C NMR (DMSO-d6): d 12.10 (5-CH3);
27.13 (d, JP-C = 142 Hz, P–CH2–Cl); 34.32 (N–C–CH2–ind); 35.73
2
2
35.81 (N–CH2–C–C–C); 55.29 (C-30); 58.22 (d, JP-C = 6.0 Hz, C-50);
(C-20); 41.29 (N–CH2–C–ind); 60.13 (C-30); 63.11 (d, JP-C = 5.5 Hz,
3
77.03 (C-10); 80.13 (C-40); 106.18 (C-5); 130.22 (C-6); 144.98 (C-2);
158.49 (C-4). 31P NMR (CDCl3): d 24.85; 25.03. MS-ESI m/z: 435; 437
[M+H]+; 457; 459 [M+Na]+; 433; 435 [MꢀH]ꢀ; 469; 471; 473
[M+Cl]ꢀ. Anal. Calcd for C15H24ClN6O5P: C, 41.43; H, 5.56; N, 19.33.
Found: C, 41.40; H, 5.57; N, 19.35.
C-50); 81.73 (d, JP-C = 6.7 Hz, C-40); 83.51 (C-10); 109.93 (ind);
111.37 (C-5); 112.03 (ind); 118.20 (ind); 120.86 (ind); 122.60
(ind); 122.86 (ind); 127.12 (ind); 135.78 (ind); 136.19 (C-6);
150.37 (C-2); 163.66 (C-4). 31P NMR (DMSO-d6): d 25.38; 25.58.
MS-ESI m/z: 522; 524 [M+H]+; 544; 546 [M+Na]+; 560; 562
[M+K]+; 520; 522 [MꢀH]ꢀ; 556; 558; 560 [M+Cl]ꢀ. Anal. Calcd
for C21H25ClN7O5P: C, 48.33; H, 4.83; N, 18.79. Found: C, 48.35;
H, 4.82; N, 18.77.
4.1.1.5. 30-Azido-30-deoxythymidine 50-(N-i-propyl-P-chloro-
methylphosphonate) (10).
1H NMR (CDCl3): d 1.23 (d, 6H,
J = 6.0 Hz, N-C(CH3)2); 1.94 (d, 3H, J = 1.4 Hz, 5-CH3); 2.29–2.45
(m, 2H, H-200, H-20); 2.95–3.06 (m, 1H, NH-iPr); 3.47–3.71 (m, 3H,
P-CH2-Cl, N-CH); 4.02–4.17 (m, 1H, H-40); 4.19–4.40 (m, 2H, H-50,
H-500); 4.45–4.50 (m, 1H, H-30); 6.14 and 6.21 (t, 1H, J = 6.7 Hz,
H-10); 7.31 and 7.38 (d, 1H, J = 1.1 Hz, H-6); 9.41 (s, 1H, 3-NH).
13C NMR (CDCl3): d 12.53 (5-CH3); 25.71, 25.91 (N–C(CH3)2);
35.56 (d, JP-C = 142 Hz, P–CH2–Cl); 37.27 (C-20); 44.00 (N–CH);
60.54 (C-30); 63.56 (C-50); 82.33 (C-10); 85.43 (C-40); 111.53
(C-5); 135.50 (C-6); 150.16 (C-2); 163.59 (C-4). 31P NMR (CDCl3):
d 23.34; 23.52. MS-ESI m/z: 421; 423 [M+H]+; 443; 445 [M+Na]+;
459; 461 [M+K]+; 455; 457; 459 [MꢀCl]ꢀ. Anal. Calcd for
4.1.1.9. 30-Azido-30-deoxythymidine 50-(N,N-dimethyl-P-chloro-
methylphosphonate) (14).
1H NMR (CDCl3): d 1.93 (d, 3H,
J = 1.1 Hz, 5-CH3); 2.92–2.49 (m, 2H, H-20, H-200); 2.82 (d, 6H,
J = 3.7 Hz, N(CH3)2); 3,63 and 3,64 (dd, 2H, J = 4.8 Hz, J = 9.5 Hz,
P-CH2-Cl); 4.03–4.17 (m, 2H, H-50, H-500); 4.30–4.41 (m, 2H, H-40,
H-30); 6.15 and 6.23 (t, 1H, J = 6.6 Hz, H-10); 7.32 and 7.40 (d, 1H,
J = 1.1 Hz, H-6); 9.40 (s, 1H, 3-NH). 13C NMR (CDCl3): d 12.50
(5-CH3); 34.10 (d, JP-C = 144 Hz, P–CH2–Cl); 36.56 (N–CH3); 37.33
2
(C-20); 60.57 (C-30); 63.34 (d, JP-C = 6.5 Hz, C-50); 82.27 (d,
3JP-C = 6.4 Hz, C-40); 85.29 (C-10); 111.49 (C-5); 135.49 (C-6);
150.22 (C-2); 163.75 (C-4). 31P NMR (CDCl3): d 25.82; 25.89. MS-
ESI m/z: 429; 431 [M+Na]+; 445; 447 [M+K]+; 405; 407 [MꢀH]ꢀ;
441; 443; 445 [M+Cl]ꢀ. Anal. Calcd for C13H20ClN6O5P: C, 38.39;
H, 4.96; N, 20.66. Found: C, 38.43; H, 4.95; N, 20.65.
C14H22ClN6O5P: C, 39.96; H, 5.27; N, 19.97. Found: C, 39.99; H,
5.26; N, 19.98. HPLC: eluent (a), retention time (tR) of 14.5 and
16.1 min in the ratio 1:1.
4.1.1.6.
methylphosphonate) (11).
30-Azido-30-deoxythymidine 50-(N-allyl-P-chloro-
1H NMR (CDCl3): d 1.93 (d, 3H,
4.1.1.10. 30-Azido-30-deoxythymidine 50-(N,N-diethyl-P-chloro-
J = 1.1 Hz, 5-CH3); 2.26–2.49 (m, 2H, H-20, H-200); 3.53–3.72 (m,
5H, N–-CH2–C@C, P–CH2–Cl, H-40); 4.02–4.18 (m, 1H, NH–C–
C@C); 4.19–4.41 (m, 2H, H-50, H-500); 4.44–4.50 (m, 1H, H-30);
5.14–5.30 (m, 2H, N–C–C@CH2); 5.82–5.94 (m, 1H, N–C–CH@C);
6.13 and 6.19 (t, 1H, J = 6.7 Hz, H-10); 7.30 and 7.36 (d, 1H,
J = 1.1 Hz, H-6); 9.81 (s, 1H, 3-NH). 13C NMR (CDCl3): d 12.53 (5-
CH3); 35.25 (d, JP-C = 143 Hz, P–CH2–Cl); 43.58 (C-20); 50.59 (N–
CH2–C@C); 60.55 (C-30); 63.75 (d, JP-C = 6,3 Hz, C-50); 82.21 (C-10);
85.43 (C-40); 111.45 (C-5); 115.94 (N–C–C@CH2); 135.36 (N–C–
CH@C); 135.53 (C-6); 150.23 (C-2); 163.80 (C-4). 31P NMR (CDCl3):
d 24.70; 24.86. MS-ESI m/z: 419; 421 [M+H]+; 441; 443 [M+Na]+;
417; 419 [MꢀH]ꢀ; 453, 455; 457 [M+Cl]ꢀ. Anal. Calcd for
methylphosphonate) (15).
1H NMR (CDCl3): d 1.04 (t, 6H,
J = 7.0 Hz, N(C-CH3)2); 1.86 (d, 3H, J = 1.1 Hz, 5-CH3); 2.12–2.40
(m, 2H, H-20, H-200); 3.24–3.41 (m, 6H, N(CH2–C)2, P–CH2–Cl);
3.81–4.11 (m, 3H, H-40, H-50, H-500); 4.16–4.26 (m, 1H, H-30); 6.14
and 6.22 (t, 1H, J = 6.6 Hz, H-10); 7.37 (d, 1H, J = 0.9 Hz, H-6);
10.25 (s, 1H, 3-NH). 13C NMR (CDCl3): d 13.10 (5-CH3); 13.55
(N(C–CH3)2); 33.46 (d, JP-C = 144 Hz, P–CH2–Cl); 37.21 (C-20);
37.97 (N(CH2–C)2); 60.24 (C-30); 63.08 (C-50); 82.55 (d,
3JP-C = 6.6 Hz, C-40); 85.11 (C-10); 111.20 (C-5); 135.48 (C-6);
150.30 (C-2); 163.71 (C-4). 31P NMR (CDCl3): d 25.64; 25.72. MS-
ESI m/z: 457; 459 [M+Na]+; 433; 435 [MꢀH]ꢀ; 469; 471; 473
[M+Cl]ꢀ. Anal. Calcd for C15H24ClN6O5P: C, 41.43; H, 5.56; N,
19.33. Found: C, 41.41; H, 5.57; N, 19.34.
C14H20ClN6O5P: C, 40.15; H, 4.81; N, 20.07. Found: C, 40.11; H,
4.80; N, 20.05.
4.1.2. Procedure for the preparation of compound 16
4.1.1.7. 30-Azido-30-deoxythymidine 50-(N-phenethyl-P-chloro-
methylphosphonate) (12).
To a solution of P-chloromethylphosphonic dichloride (1)
1H NMR (CDCl3): d 1.90 (d, 3H,
(153 mg, 0.914 mmol) in acetonitrile (2 mL) was added 1,2,4-tria-
zole (2) (164 mg, 2.374 mmol) followed by triethylamine
(189 mg, 1.865 mmol) and the reactants were stirred for 30 min
at room temperature. Then to the mixture 30-azido-30-deoxythymi-
dine (4) (100 mg, 0.374 mmol) and pyridine (2.30 mL) were added.
The reaction mixture was stirred at room temperature for a further
1 h and a mixture of water (0.112 mL), triethylamine (240 mg,
2.372 mmol) and pyridine (0.748 mL) was added. After 20 min,
the reaction mixture was evaporated under reduced pressure. To
the residue was added saturated aqueous sodium bicarbonate
(20 mL) and the mixture was extracted with chloroform
(3 ꢁ 10 mL). The combined chloroform extracts were washed with
water (10 mL), dried over anhydrous magnesium sulfate, filtered
and evaporated to dryness. The residue was dissolved in chloro-
form (1 mL) and added dropwise to stirred n-hexane (80 mL).
The resulting colorless precipitate was collected by filtration and
dried in vacuo over phosphorus pentoxide to afford product 16
(128 mg, yield 71%).
J = 1.0 Hz, 5-CH3); 2.24–2.45 (m, 1H, H-20, H-200); 2.82 (t, 2H,
J = 6.2 Hz, N–C–CH2–Ph); 3.21–3.56 (m, 6H, H-40, NH–C–C–Ph, N–
CH2–C–Ph, P–CH2–Cl); 3.94–4.07 (m, 2H, H-50, H-500); 4.19–4.31
(m, 1H, H-30); 6.11 and 6.18 (t, 1H, J = 6.6 Hz, H-10); 7.19–7.34
(m, 6H, H-6, Ph); 9.68 (s, 1H, 3-NH). 13C NMR (CDCl3): d 12.52
(5-CH3); 35.17 (d, JP-C = 143 Hz, P–CH2–Cl); 37.27 (C-20); 38.25
(N–C–CH2–Ph); 42.63 (N–CH2–C–Ph); 60.56 (C-30); 63.57 (d,
2JP-C = 5.8 Hz, C-50); 82.25 (C-10); 85.39 (C-40); 111.43 (C-5);
126.71 (Ph); 128.61 (Ph); 128.81 (Ph); 135.42 (C-6); 138.06 (Ph);
150.13 (C-2); 163.61 (C-4). 31P NMR (CDCl3): d 24.40; 24.62.
MS-ESI m/z: 483; 485 [M+H]+; 505; 507 [M+Na]+; 481; 483
[MꢀH]ꢀ; 517; 519; 521 [M+Cl]ꢀ. Anal. Calcd for C19H24ClN6O5P:
C, 47.26; H, 5.01; N, 17.40. Found: C, 47.22; H, 5.00; N, 17.39.
4.1.1.8. 30-Azido-30-deoxythymidine 50-[N-2-(1H-indol-3-yl) ethyl-
P-chloromethylphosphonate] (13).
1H NMR (DMSO-d6): d
1.77 (s, 3H, 5-CH3); 2.26–2.45 (m, 2H, H-20, H-200); 2.76–2.88 (m,

L. Celewicz et al. / Bioorg. Med. Chem. 19 (2011) 6375–6382
6381
4.1.2.1. 30-Azido-30-deoxythymidine 50-(P-chloromethylphosph-
onate) triethylammonium salt (16).
1H NMR (DMSO-d6): d
cultured in DMEM medium (Sigma), supplemented with 10%
heat-inactivated fetal bovine serum (Gibco BRL), 1% non-essential
amino acids 100X (NEAA, Sigma) and 50 mg/L gentamycin (Gibco
BRL). For transport experiments, Caco-2 cells were plated at a den-
sity of 4.4 ꢁ 105 cells/cm2 on polycarbonate membranes (Millicell,
1.16 (t, 9H, J = 7.1 Hz, N(C–CH3)3); 1.91 (d, 3H, J = 1.0 Hz, 5-CH3);
2.03–2.22 (m, 2H, H-200, H-20); 2.94 (q, 6H, J = 7.1 Hz, N(CH2–C)3);
3.28–3.41 (m, 2H, P–CH2–Cl); 3.47–3.58 (m, 1H, H-40); 4.20–4.35
(m, 2H, H-50, H-500); 4.48 (m,1H, H-30); 6.14 (t, 1H, J = 6.6 Hz, H-10);
7.56 (d, 1H, J = 1.0 Hz, H-6); 11.21 (bs, 1H, 3-NH). 13C NMR (DMSO-
d6): d 8.55 (N(C–CH3)3); 12.08 (5-CH3); 36.19 (N(CH2–C)3); 39.28
(d, JP-C = 155 Hz, P–CH2–Cl); 44.96 (C-20); 60.85 (C-30); 63.37 (C-50);
83.01 (C-10); 84.47 (C-40); 109.22 (C-5); 136.16 (C-6); 150.54 (C-2);
163.40 (C-4). 31P NMR (DMSO-d6): d8.95. MS-ESI (negative mode)
m/z: 378; 380. HPLC: eluent (a), retention time (tR) of 7.6 min.
12 mm diameter) with pore size of 0.4 lm (Millipore). Cells were
maintained at 37 °C, in a humidified, 5% CO2, 95% air atmosphere.
Growth medium was replaced twice weekly. Cell confluence was
monitored by transepithelial electrical resistance (TEER) measure-
ments. TEER was determined employing the Millicell Electrical
Resistance System (Millipore). Transport experiments were per-
formed on monolayers, 21 days post-seeding.
4.1.3. Chemical stability studies
4.3.2. Transport experiments
Hydrolysis of the compounds 8 and 10 was carried out in the
0.066 M phosphate buffer at pH 7.2, 7.4 and 7.8 at 37 °C and the
concentration of the compound under study of 0.5 mM. The ali-
quots were taken out after certain intervals and frozen. The com-
position of the reaction mixtures was analyzed by HPLC under
the conditions described above.
Solutions of compounds 8 and 10 at concentration of 100 lM
were prepared in transport medium (Hank0s balanced salt solution
containing 10 mM HEPES buffer, pH 7.4). The solutions were fil-
tered through a 0.22 lm membrane (Millex GP, Millipore) and
pre-heated at 37 °C before permeability experiments.
Prior to the experiments, the monolayers were washed twice
with transport medium and pre-incubated at 37 °C for 30 min.
After the pre-incubation, TEER of the monolayers was measured
to check cell monolayer integrity. Only monolayers with TEER val-
4.2. In vitro cytotoxicity
4.2.1. Cell cultures
ues above 800 X
cm2 were used in experiments. Transport was ini-
Human cancer cells KB (carcinoma nasopharynx) were cultured
in RPMI 1640 medium and human cancer cells MCF-7 (breast can-
cer cell line) were cultured in DMEM medium. Each medium was
tiated by adding transport medium to the acceptor side and the
solution of the test compound to the donor side. The Caco-2 cell
monolayers were placed in the incubator (37 °C) on the plate sha-
ker. At 15 min intervals, the sample from acceptor compartment
was taken and the concentration of compound transported was
analyzed by HPLC. Each sample volume was replaced with fresh
pre-heated (37 °C) transport medium. To quantify transport of
compounds tested across the Caco-2 cell monolayer, the apparent
permeability coefficient (Papp) was used. The Papp values were
determined as previously described.58 Experimentally derived data
are shown as mean SD (n = 3).
supplemented with 10% fetal bovine serum, 1% L-glutamine and
1% penicillin/streptomycin solution. The cell lines were kept in
the incubator at 37 °C. The optimal plating density of cell lines
was determined to be 5 ꢁ 104. Both the cell lines were obtained
from The European Collection of Cell Cultures (ECACC) supplied
by Sigma–Aldrich.
4.2.2. In vitro cytotoxicity assay
The protein-staining sulforhodamine B (SRB, Sigma–Aldrich)
microculture colorimetric assay, developed by the National Cancer
Institute (USA) for in vitro antitumor screening was used in this
study, to estimate the cell number by providing a sensitive index
of total cellular protein content, being linear to cell density.39 The
monolayer cell culture was trypsinized and the cell count was ad-
justed to 5 ꢁ 104 cells. To each well of the 96 well microtiter plate,
0.1 mL of the diluted cell suspension (approximately 10,000 cells)
was added. After 24 h, when a partial monolayer was formed, the
4.4. Theoretical calculation of Caco-2 cell permeability
The Caco-2 prediction model, developed by Guangli and Yiyu,57
based on descriptors generated by open source software Chemistry
Development Kit (CDK) was used to compute Caco-2 apparent per-
meability for the studied compounds. In this model membrane
permeability of compounds is determined by a number of hydro-
gen bond donors and three molecular surface area properties.
The correlation coefficients of the experimental and predicted
Caco-2 apparent permeability for their test set was as high as 0.85.
supernatant was washed out and 100
concentrations (0.1, 0.2, 1, 2, 10 and 20
in microtitre plates. The tested compounds were dissolved in DMSO
(20 M) and the content of DMSO did not exceed 0.1%; this concen-
tration was found to be nontoxic to the cell lines. The cells were ex-
posed to compounds for 72 h. After that, 25 L of 50% trichloroacetic
lL of six different compound
lM) were added to the cells
l
Acknowledgments
l
Research support from Ministry of Science and Higher Educa-
tion (Grant No. N N204 343037) is gratefully acknowledged.
acid was added to the wells and the plates were incubated for 1 hour
at 4 °C. The plates were then washed out with the distilled water to
remove traces of medium and next dried by the air. The air-dried
References and notes
plates were stained with 100 lL SRB and kept for 30 min at room
temperature. The unbound dye was removed by rapidly washing
with 1% acetic acid and then air dried overnight. The optical density
was read at 490 nm. All cytotoxicity experiments were performed
three times. Cell survival was measured as the percentage absor-
bance compared to the control (non-treated cells). Zidovudine and
cytarabine (Sigma–Aldrich) were used as the internal standards.
1. De Clercq, E. J. Clin. Virol. 2004, 30, 115.
2. Galmarini, C. M.; Popowycz, F.; Joseph, B. Curr. Med. Chem. 2008, 15, 1072.
3. Longley, D. B.; Harkin, D. P.; Johnston, P. G. Nat. Rev. Cancer 2003, 3, 330.
4. Périgaud, C.; Gosselin, G.; Imbach, J.-L. Nucleosides Nucleotides 1992, 11, 903.
5. (a) Border, S. Antiviral Res. 2010, 85, 1; (b) Border, S. Med. Res. Rev. 1990, 10,
419; (c) 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. Proc. Natl. Acad.
Sci. U.S.A. 1985, 82, 7096.
6. (a) Brunetti, I.; Falcone, A.; Calabresi, P.; Goulette, F. A.; Darnowski, J. W. Cancer
Res. 1990, 50, 4026; (b) Darnowski, J. W.; Goulette, F. A. Biochem. Pharmacol.
1994, 48, 1797.
4.3. In vitro permeability assay
7. Wagner, C. R.; Ballato, G.; Akanni, A. O.; McIntee, E. J.; Larson, R. S.; Chang, S. L.;
Abulhajj, Y. J. Cancer Res. 1997, 57, 2341.
8. Iyer, V. V.; Griesgraber, G. W.; Radmer, M. R.; McIntee, E. J.; Wagner, C. R. J. Med.
Chem. 2000, 43, 2266.
4.3.1. Cell culture
Caco-2 cell line was obtained from the European Collection of
Cell Cultures (ECACC) and supplied by Sigma-Aldrich. Cells were

6382
L. Celewicz et al. / Bioorg. Med. Chem. 19 (2011) 6375–6382
9. Wagner, C. R.; Iyer, V. V.; McIntee, E. J. Med. Res. Rev. 2000, 20, 417.
10. Zemlicka, J. Biochim. Biophys. Acta 2002, 1587, 276.
11. Drontle, D. P.; Wagner, C. R. Mini-Rev. Med. Chem. 2004, 4, 409.
12. Cahard, D.; McGuigan, C.; Balzarini, J. Mini-Rev. Med. Chem. 2004, 4, 371.
13. Hecker, S. J.; Erion, M. D. J. Med. Chem. 2008, 51, 2328.
14. Meier, C.; Ruppel, M. F. H.; Vukadinovic, D.; Balzarini, J. Mini-Rev. Med. Chem.
2004, 4, 383.
31. Stawin´ ski, J.; Hozumi, T.; Narang, S. A.; Bahl, C. P.; Wu, R. Nucleic Acids Res.
1977, 4, 353.
32. Agarwal, K. L.; Riftina, F. Nucleic Acids Res. 1978, 5, 2809.
33. Chattopadhyaya, J. B.; Reese, C. B. Tetrahedron Lett. 1979, 5059.
34. Agarwal, K. L.; Riftina, F. Nucleic Acids Res. 1979, 6, 3009.
35. Dorman, M. A.; Noble, S. A.; McBride, L. J.; Caruthers, M. H. Tetrahedron 1984,
40, 95.
36. Marugg, J. E.; de Vroom, E.; Dreef, C. E.; Tromp, M.; van der Marel, G. A.; van
Boom, J. H. Nucleic Acids Res. 1986, 14, 2171.
37. Berté-Verrando, S.; Dizière, R.; Samadi, M.; Savignac, P. J. Chem. Soc. Perkin
Trans 1 1995, 3125.
15. Devine, K. G.; McGuigan, C.; O0Connor, T. J.; Nicholls, S. R.; Kinchington, D. AIDS
1990, 4, 371.
16. McGuigan, C.; Devine, K. G.; O0Connor, T. J.; Galpin, S. A.; Jeffries, D. J.;
Kinchington, D. Antiviral Chem. Chemother. 1990, 1, 107.
17. McGuigan, C.; Pathirana, R. N.; Mahmood, N.; Devine, K. G.; Hay, A. J. Antivir.
Res. 1992, 17, 311.
18. McGuigan, C.; Pathirana, R. N.; Balzarini, J.; De Clercq, E. J. Med. Chem. 1993, 36,
1048.
38. Waschbüsch, R.; Carran, J.; Marinetti, A.; Savignac, P. Chem. Rev. 1997, 97, 3401.
39. Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.; McMahon, J.; Vistica, D.;
Warren, J. T.; Bokesch, H.; Kenney, S.; Boyd, M. R. J. Natl. Cancer Inst. 1990, 82,
1107.
19. Saboulard, D.; Naesens, L.; Cahard, D.; Salgado, A.; Pathirana, R.; Velazquez, S.;
McGuigan, C.; De Clercq, E.; Balzarini, J. Mol. Pharmacol. 1999, 56, 693.
20. Song, H.; Griesgraber, G. W.; Wagner, C. R.; Zimmerman, C. L. Antimicrob. Agents
Chemother. 2002, 46, 1357.
40. Patrick, G. L. An Introduction to Medicinal Chemistry; Oxford University Press,
2001. p. 308.
41. Yamaguchi, H.; Noshita, T.; Yu, T.; Kidachi, Y.; Kamiie, K.; Umetsu, H.; Ryoyama,
K. Eur. J. Med. Chem. 2010, 45, 2943.
21. Song, H.; Johns, R.; Griesgraber, G. W.; Wagner, C. R.; Zimmerman, C. L. Pharm.
Res. 2003, 20, 448.
22. Tan, X.; Boudinot, F. D.; Chu, C. K.; Egron, D.; Périgaud, C.; Gosselin, G.; Imbach,
J.-L. Antiviral Chem. Chemother. 2000, 11, 203.
23. Shafiee, M.; Deferme, S.; Villard, A. L.; Egron, D.; Gosselin, G.; Imbach, J.-L.;
Lioux, T.; Pompon, A.; Varray, S.; Aubertin, A. M.; Van Den Mooter, G.; Kinget,
R.; Périgaud, C.; Augustijns, P. J. Pharm. Sci. 2001, 90, 448.
24. Beltran, T.; Egron, D.; Lefebvre, I.; Périgaud, C.; Pompon, A.; Gosselin, G.;
Aubertin, A.-M.; Imbach, J.-L. Nucleosides Nucleotides, Nucleic Acids 1999, 18,
973.
42. (a) Pyka, A.; Babuska, M.; Zachariasz, M. Acta Pol. Pharm. Drug Res. 2006, 63,
159; (b) Meylan, W. M.; Howard, P. H. J. Pharm. Sci. 1995, 84, 83.
43. Chapman, T. M.; McGavin, J. K.; Noble, S. Drugs 2003, 63, 1597.
44. Bergstrom, D. E.; Shum, P. W. J. Org. Chem. 1988, 53, 3953.
45. Artursson, P.; Karlsson, J. Biochem. Biophys. Res. Commun. 1991, 175, 880.
46. Yee, S. Pharm. Res 1997, 14, 763.
47. Chan, L. M. S.; Lowes, S.; Hirst, B. H. Eur. J. Pharm. Sci. 2004, 21, 25.
48. Szakács, G.; Váradi, A.; Ozvegy-Laczka, C.; Sarkadi, B. Drug Discov. Today 2008,
13, 379.
49. Villard, A.-L.; Coussot, G.; Lefebvre, I.; Augustijns, P.; Aubertin, A.-M.; Gosselin,
G.; Peyrottes, S.; Périgaud, C. Bioorg. Med. Chem. 2008, 16, 7321.
50. Varatharajana, L.; Thomas, S. A. Antiviral. Res. 2009, 82, A99.
51. Quevedo, M. A.; Nieto, L. E.; Briñón, M. C. Eur. J. Pharm. Sci. 2011, 43, 151.
52. Park, G.-B.; Mitra, A. K. Pharm. Res. 1992, 9, 326.
25. Peyrottes, S.; Egron, D.; Lefebvre, I.; Gosselin, G.; Imbach, J.-L.; Périgaud, C.
Mini-Rev. Med. Chem. 2004, 4, 395.
26. Rosowsky, A.; Fu, H.; Pai, N.; Mellors, J.; Richman, D. D.; Hostetler, K. Y. J. Med.
Chem. 1997, 40, 2482.
27. Meier, C.; Aubertin, A. M.; de Monte, M.; Faraj, A.; Sommadossi, J. P.; Périgaud,
C.; Imbach, J.-L.; Gosselin, G. Antivir. Chem. Chemother. 1998, 9, 41.
28. Shirokova, E. A.; Jasko, M. V.; Khandazhinskaya, A. L.; Ivanov, A. V.; Yanvarev, D.
V.; Skoblov, Y. S.; Mitkevich, V. A.; Bocharov, E. V.; Pronyaeva, T. R.; Fedyuk, N.
V.; Kukhanova, M. K.; Pokrovsky, A. G. J. Med. Chem. 2004, 47, 3606.
29. Szymczak, M.; Szyman´ ska, A.; Stawin´ ski, J.; Boryski, J.; Kraszewski, A. Org. Lett.
2003, 5, 3571.
53. Yazdanin, M.; Glynn, S. L.; Wright, J. L.; Hawi, A. Pharm. Res. 1998, 15, 1490.
54. Batrakova, E. V.; Li, S.; Miller, D. W.; Kabanov, A. V. Pharm. Res. 1999, 16, 1366.
55. Salama, N. N.; Scott, K. R.; Eddington, N. D. Biopharm. Drug Dispos. 2004, 25,
227.
56. Morse, G. D.; Shelton, M. J.; O0Donnell, A. M. Clin. Pharmacokinet. 1993, 24, 101.
57. Guangli, M.; Cheng, Y. J. Pharm. Pharmaceut. Sci. 2006, 9, 210.
58. Tavelin, S.; Gråsjö, J.; Taipalensuu, J.; Ocklind, G.; Artursson, P. Methods Mol.
Biol. 2002, 188, 233.
´
´
30. Szymanska, A.; Szymczak, M.; Boryski, J.; Stawinski, J.; Kraszewski, A.; Collu, G.;
Sanna, G.; Giliberti, G.; Loddo, R.; La Colla, P. Bioorg. Med. Chem. 2006, 14, 1924.
59. Czernecki, S.; Valery, J.-M. Synthesis 1991, 239.