Modified N,O-nucleosides: Design, synthesis, and anti-tumour activity

Article abstract of DOI:10.1071/CH13511

A preliminary library of modified N,O-nucleosides was prepared and tested on a selected number of human cancer lines that include SKOV3, SW480, and K562. Thymine, N-benzyl substituents, and aromatic rings contribute to an increase of the biological activity, up to 10-25μM, that appeared also reliant on the calculated lipophilicity of the nucleosides, expressed as cLogP, where P represents the partition coefficient of a solute between n-octanol and water. CSIRO 2014.

Full text of DOI:10.1071/CH13511

CSIRO PUBLISHING
Aust. J. Chem. 2014, 67, 670–674
Full Paper
http://dx.doi.org/10.1071/CH13511
Modified N,O-Nucleosides: Design, Synthesis,
and Anti-tumour Activity
A D
,
B D
,
A
Loredana Maiuolo,
Olga Bortolini,
Antonio De Nino,
A
C
C
Beatrice Russo, Riccardo Gavioli, and Fabio Sforza
A
Dipartimento di Chimica e Tecnologie Chimiche, Universita` della Calabria,
Via P. Bucci, 87036 Rende (CS), Italy.
B
Dipartimento di Scienze Chimiche e Farmaceutiche, Universita` di Ferrara,
Via Borsari 46, 44121 Ferrara, Italy.
C
Dipartimento di Biochimica e Biologia Molecolare, Universita` di Ferrara,
Via Fossato di Mortara 74, 44121 Ferrara, Italy.
D
Corresponding authors. Email: maiuolo@unical.it; brl@unife.it
A preliminary library of modified N,O-nucleosides was prepared and tested on a selected number of human cancer lines
that include SKOV3, SW480, and K562. Thymine, N-benzyl substituents, and aromatic rings contribute to an increase
of the biological activity, up to 10–25 mM, that appeared also reliant on the calculated lipophilicity of the nucleosides,
expressed as cLogP, where P represents the partition coefficient of a solute between n-octanol and water.
Manuscript received: 25 September 2013.
Manuscript accepted: 4 December 2013.
Published online: 23 January 2014.
R
O
B
O
B
Introduction
N
N
In recent years, a large number of nucleoside analogues with
antiviral and/or antitumour properties have been designed and
synthesized.[1,2] Nucleosides, in fact, comprise the largest class
of clinically useful antiviral agents and they continue to be
excellent candidates as anticancer drugs.[3]
Rꢀ
Rꢀ
1
2
Fig. 1. General structure of N,O-nucleosides.
In the search for effective, selective, and non-toxic agents, a
variety of modifications of the naturally occurring structure
have been devised on both the sugar and the nucleobase. Of
particular interest are nucleoside analogues in which the fura-
nose ring has been replaced by N,O-heterocyclic systems, as
in isoxazolidine 1 and isoxazoline 2 derivatives, in which B
represents pyrimidine or purine nucleobases (Fig. 1).[4–6]
Most of the N,O-containing nucleoside analogues described in
the literature possess remarkable antiviral activity but a modest
potential activity towards human cancer cell lines,[7,8] contrary
to what is observed with substituted nucleosides and thio-
nucleosides, in particular 40-thio-b-D-arabinofuranosylcytosine
(40-Thio-Ara-C).[3,9–11]
O
N
O
X
X
ꢂ
NH
R
O
NH
ꢁ
MW
N
ꢁ
O
R
O
N
O
N
Rꢀ
H
Rꢀ
H
H
3
4
5
Scheme 1. Synthesis of N,O-nucleosides via 1,3-dipolar cycloaddition of
nitrones 3 with vinyl nucleobases 4 by microwave (MW) irradiation.
In a previous study, we reported the efficient synthesis
of several N,O-nucleosides by direct 1,3-dipolar cyclization
methodology.[12,13] The cyclization was carried out between
selected nitrones 3 and a set of unprotected vinyl nucleobases 4
under microwave (MW) irradiation, in the absence of solvent or
catalyst (Scheme 1). The cycloadducts are formed in good yield,
with complete regioselectivity and notable cis diastereoselec-
tivity, up to 98 %, resulting from a dominant (Z)-exo nitrone–
alkene approach.[13]
The N,O-nucleosides of the previous study were evaluated by
in vitro assays for their antiproliferative activity on LCLs,
JiJoye, and Jurkat cell lines, and were found particularly
promising.[13]
Results and Discussion
In analogy with specific absorption rate (SAR) tests designed to
prove the minimal structural requirements for antiproliferative
activity in the NCI 60 panel of human cancers,[14] we investi-
gated variations in the four canonical quadrants of our com-
pounds 5 (Fig. 2), and some of these were submitted to growth
inhibition assays on SKOV3 (cisplatin-resistant) and SW480
cell lines established for ovarian and colon cancers respectively
(Table 1). It was found that compound 5d exhibited greater
potency (25–50 mM) than either compounds 5a or 5e on these
Journal compilation Ó CSIRO 2014
www.publish.csiro.au/journals/ajc

Synthesis of modified N,O-nucleosides
671
N
tumoural lines, comparable with that of cisplatin cis-
[PtCl2(NH3)2] used as model reference.
NW
Motivated by the above finding, a novel series of N,O-
isoxazolidines was designed, modifying the nucleobases and/
or the substituents on the isoxazolidine ring. Taking into
account these findings and as pyrimidine nucleoside analogues
are essential components of haematological malignancy therapy
and are also used in the treatment of solid tumours,[15] we
prepared a new set of N,O-nucleosides varying the SE quadrant
with thymine or fluorouracil as nucleobases, and the R and R0
substituents in the NW and SW quadrants respectively.
Initially, in analogy to compound 5d, we chose to maintain
unchanged the R and B groups (benzyl and thymine respectively),
varying only the SW quadrant.
R
O
E
W
N
B
SE
Rꢀ
SW
S
Fig. 2. Representation of isoxazolidines in quadrants for changes of B, R
and R0.
The procedure for the preparation of the new nucleosides is
based on the direct 1,3-dipolar cycloaddition of the nitrone 3
with the vinyl nucleobase 4 in the absence of solvent, under
microwave irradiation.
In particular, nitrones 3f–j were obtained in excellent yields
and high purity by condensation of the precursor aldehydes 6
with suitable hydroxylamines 7 in acetate-buffered water/
ethanol solution (Scheme 2).[16]
Table 1. Estimated IC50 values in cisplatin-resistant SKOV3 and
SW480 cell lines, as determined by the 3-(4,5-dimethylthiazol-2-yl)2,5-
diphenyltetrazolium bromide solution (MTT) assay
Data are the mean ꢀ s.d. of three independent experiments performed
in triplicate
R
R0
Nucleobase (B)
Compound
IC50 (mM)
SKOV3
SW480
With the set of nitrones 3f–j in hand, we initially synthesized
compounds 5f–h by 1,3-dipolar cycloaddition with vinylthymine
as nucleobase 4, according to the general procedure described in
the Experimental section. Then, the N,O-nucleoside derivatives
5f–h were evaluated by in vitro assays for their antiproliferative
activity against SKOV3 and SW480 cell lines. Among the new
derivatives, the ortho-chloro 5f and ortho-fluoro 5g compounds
displayed a biological activity similar to or even enhanced
relative to that shown by 5d, thus confirming that thymine,
N-benzyl substituents, and aromatic rings are the optimal
combination for biological activity.
But
Me
But
Bn
Ph
Thy
Thy
F-Ura
Thy
Thy
5a
.100
.100
.100
43 ꢀ 5
.100
.50
.100
.100
.100
50 ꢀ 6
.100
Ph
5b
Ph
5c
Ph
5d
Me
2-Cl-Ph
5e
Cisplatin
Bn
Bn
Bn
2-Cl-Ph
2-F-Ph
3-Py
Thy
Thy
Thy
5f
24 ꢀ 6
39 ꢀ 1
.100
36 ꢀ 1
39 ꢀ 1
.100
5g
5h
The small library of N,O-nucleosides 5a–h was analysed
taking into consideration physicochemical data related to
solubility and permeability.[17] These properties are easily
obtained from research databases such as SciFinderÒ or may
be calculated.[18] Table 2 lists the calculated lipophilicity
expressed as cLogP, where P represents the partition coefficient
of a solute between n-octanol and water, the molecular weight
(MW), and the number of hydrogen donor (Hdon) and acceptor
(Hacc) sites of our N,O-nucleosides, which in all cases is in
ꢂ
R
O
3f: R ꢃ Bn; Rꢀ ꢃ 2-Cl-Ph
3g:R ꢃ Bn; Rꢀ ꢃ 2-F-Ph
3h:R ꢃ Bn; Rꢀ ꢃ 3-Py
3i: R ꢃ Bn; Rꢀ ꢃ 4-Cl-Ph
ꢁ
O
N
ꢁ RNHOH
H
Rꢀ
H
Rꢀ 3j: R ꢃ Bn; Rꢀ ꢃ 1-Naphthyl
6
7
3
Scheme 2. Synthesis of aryl-substituted nitrones 3f–j.
Table 2. Estimated IC50 values in cisplatin-resistant SKOV3, SKOV480, K562, MDA, HCT116, and A2780 cancer cell lines, as determined by the
3-(4,5-dimethylthiazol-2-yl)2,5-diphenyltetrazolium bromide solution (MTT) assay
Data are the mean ꢀ s.d. of three independent experiments performed in triplicate. cLogP, calculated lipophilicity of the nucleosides, where P represents the
partition coefficient of a solute between n-octanol and water; MW, molecular weight; Hacc, number of hydrogen acceptor sites of the N,O-nucleosides; Hdon
,
number of hydrogen donor sites of the N,O-nucleosides
R
Ar
B
Compound
IC50 (mM)
cLogP
MW
Hacc
Hdon
SKOV3
SW480
K562
MDA
HCT116
A2780
But
Me
But
Bn
Me
Bn
Bn
Bn
Bn
Bn
Bn
Ph
Ph
Ph
Ph
Thy
Thy
F-Ura
Thy
Thy
Thy
Thy
Thy
F-Ura
Thy
Thy
5a
5b
5c
5d
5e
5f
.100
.100
.100
43 ꢀ 5
.100
24 ꢀ 6
39 ꢀ 1
.100
.100
25 ꢀ 3
19 ꢀ 2
.100
.100
.100
56 ꢀ 6
.100
36 ꢀ 1
39 ꢀ 1
.100
.100
15 ꢀ 1
28 ꢀ 1
.100
.100
.100
30 ꢀ 3
.100
30 ꢀ 3
45 ꢀ 4
.100
50 ꢀ 5
31 ꢀ 3
11 ꢀ 3
.100
.100
.100
60 ꢀ 3
.100
31 ꢀ 3
33 ꢀ 5
.100
.100
34 ꢀ 1
31 ꢀ 1
.100
.100
.100
45 ꢀ 1
.100
41 ꢀ 4
55 ꢀ 8
.100
.100
24 ꢀ 3
17 ꢀ 3
.100
.100
.100
60 ꢀ 3
.100
47 ꢀ 3
49 ꢀ 9
.100
.100
21 ꢀ 1
29 ꢀ 1
2.906
1.656
2.586
3.579
2.381
4.066
3.555
2.165
3.850
4.070
4.397
329
287
333
363
321
397
381
364
401
397
413
6
6
6
6
6
6
6
7
6
6
6
1
1
1
1
1
1
1
1
1
1
1
2-Cl-Ph
2-Cl-Ph
2-F-Ph
5g
5h
5i
3-Py
2-Cl-Ph
4-Cl-Ph
1-Naphthyl
5j
5k

672
L. Mauiolo et al.
accordance with the ‘rule of five’, several requirements used for
the development of orally bioavailable drug candidates.[17,19]
These rules have been enhanced by other authors with additional
parameters such as the number of rotatable bonds; however, the
original proposal still maintains its validity.[20] In the present
case, all the compounds display similar parameters with respect
to MW and H donor and acceptor numbers. The only exception
is seen in calculated lipophilicity, which is higher for the most
biologically active compounds 5d, 5f, and 5g. Following this
line of reasoning, N,O-nucleosides substituted with N-benzyl
groups and aromatic rings, preferably with thymine as nucleo-
base and displaying calculated lipophilicity values in the range
3.5–4.5 should represent the optimal combination for increased
biological activity. To support this hypothesis, we planned the
synthesis of three new N,O-nucleosides (5i–k) with cLogP
values higher than 3.5 and we extended the number of human
cancer cell lines for biological evaluation with A2780 (cisplatin-
sensitive human ovarian cancer), HCT116 (colon cancer), MDA
(breast cancer), and K562 (human Caucasian chronic myeloge-
nous leukaemia). The related results are collected in the lower
part of Table 2, which includes the biological activities of 5d, 5f,
and 5g with the new cancer lines as well.
logP of substrates 5a–5k was calculated with the ACDlogP
program (SciFinderÒ).
Nitrones 3f–j and vinyl nucleobases 4 were synthesized
according to published procedures.[12,16,21–27] The full charac-
terization of compounds 3f–j can be found in references [16] and
[24–26], and for 5a–e in reference [13].
Synthesis
General Procedure of Synthesis of Compounds 5f–k
The selected nitrone 3 (0.2 mmol) and vinyl nucleobase 4
(0.1 mmol) were ground together in a mortar and further mixed
in a vortexer. The mixture of the two solids was transferred into a
50-mL Pyrex container, which was placed within an unmodified
household microwave oven, at 600-W irradiation power. After
the appropriate time, from 10 to 15 min, the reaction mixture
was dissolved in a minimum quantity of CHCl3 and submitted to
flash chromatographic separation, using variable mixtures of
chloroform and methanol. The excess nitrone was recovered and
can be re-used. The cycloadducts were analysed by HPLC and
1H NMR to establish the diastereoisomeric cis : trans ratio,
which ranges from 80 : 20 to 85 : 15. Yields were calculated
based on isolated compounds.
The presence of nucleobases other than thymine is detri-
mental for cytotoxicity, as demonstrated by the results found
with compound 5f (B ¼ Thy), the half-maximal inhibitory con-
centration (IC50) in the range 25–50 mM, and 5i (B ¼ F-Ura)
IC50 . 100 mM. However, within the set of nucleosides bearing
N-benzyl, thymine, and aromatic substituents, increasing bio-
logical activity was observed for compounds with increased
values of cLogP. The best results in fact, with cytotoxicity
,10–25 mM, were displayed by compounds 5f, 5j, and 5k, all
possessing a cLogP value higher than 4. The position of chlorine
substituents on the aromatic ring, ortho for 5f and para for 5j,
cannot be underestimated because a quite significant decrease of
IC50 was observed with the latter compound in four of the six
cancer lines investigated.
Further purification and separation of the cis cycloadducts
were carried out by semi-preparative HPLC chromatography.
cis-40-Aza-40-(N-benzyl)-30-(2-chlorophenyl)-20,30-
dideoxythymidine 5f
White solid, yield 77 %, mp 148–1498C. Diastereoisomer
cis : trans ratio 80 : 20. dH (500 MHz, CDCl3) 1.80 (d, J 1.1, 3H,
Thy CH3), 2.2 (ddd, J 3.8, 9.4, 13.9, 1H, H2C), 3.51 (ddd, J 7.6,
7.9, 13.9, 1H, H20 C), 3.82 (d, J 14.3, 1H, HBn), 4.08 (d, J 14.3, 1H,
HBn), 4.48 (dd, J 7.9, 9.4, 1H, H3C), 6.06 (dd, J 3.8, 7.6, 1H, H1C),
7.24–7.45 (m, 9H, Ar), 7.55–7.63 (m, 1H, Thy H6), 8.75 (bs, 1H,
NH). dC (500 MHz, CDCl3) 12.54, 46.37, 60.03, 66.08, 83.56,
109.86, 127.41, 127.68, 127.87, 128.54, 128.84, 129.25, 130.10,
133.93, 135.09, 135.76, 136.81, 150.35, 163.96. m/z (electro-
spray ionisation (ESI)-HRMS) calc. for [C21H20ClN3O3 þ H]þ
398.1271; found 398.1264.
Conclusion
In conclusion, we prepared and tested a preliminary library of
N,O-nucleosides that demonstrate the impact of substitution in
the three assessable canonical quadrants of lead antiproliferative
agent compounds 5j and 5k, with a cytotoxic activity in the
range 10–25 mM. To our knowledge, this is the first example of
modified N,O-nucleosides showing such promising inhibitory
activity against different lines of ovarian (SKOV3, A2780) and
colon (SW480, HCT116) carcinoma.
cis-40-Aza-40-(N-benzyl)-30-(2-fluorophenyl)-20,30-
dideoxythymidine 5g
White solid, yield 75 %, mp 176–1778C. Diastereoisomer
cis : trans ratio 85 : 15. dH (500 MHz, CDCl3) 1.81 (d, J 0.76, 3H,
Thy CH3), 2.40 (ddd, J 3.8, 9.8, 13.9, 1H, H2C), 3.34 (dt, J 7.5,
13.9, 1H, H20 C), 3.78 (d, J 14.4, 1H, HBn), 4.05 (d, J 14.4, 1H,
HBn), 4.26 (dd, J 7.6, 9.8,1H, H3C), 6.09 (dd, J 3.8, 7.5, 1H, H1C),
7.04–7.52 (m,10H, Ar þ Thy H6), 8.94 (bs, 1H, NH). dC
(500 MHz, CDCl3) 12.51, 46.25, 59.96, 63.46, 83.45, 109.93,
116.16, 123.97, 124.86, 127.81, 128.29, 128.49, 129.93, 135.99,
136.81, 150.47, 160.12, 162.09, 164.11. m/z (ESI-HRMS) calc.
for [C21H20FN3O3 þ Na]þ 404.1386; found 404.1382.
Experimental
General
Commercial starting materials were used without further puri-
1
fication. Solvents were distilled before use. H and 13C NMR
spectra were recorded at 500 and 125 MHz respectively in
CDCl3 using tetramethylsilane (TMS) as internal standard
(Bruker ACP 500 MHz). Chemical shifts are given in parts per
million and coupling constants in Hertz. High-resolution mass
spectra (HRMS) were acquired on a Q-star pulsar-i (MDS Sciex
Applied Biosystems, Toronto, Canada) equipped with an ion-
spray source at 10000 atomic mass unit (amu) resolution.
GC-MS spectra were carried out on a Shimadzu QP 5000. HPLC
analyses were performed on a Hewlett-Packard 1100 Series,
monitored by a UV detector at 254 nm, using a Jupiter 10 m C-18
(25 cm) column, H2O/MeOH 9 : 1, 4.0 mL minꢁ1. The value of
cis-40-Aza-40-(N-benzyl)-30-(3-pyridyl)-20,30-
dideoxythymidine 5h
White solid, yield 74 %, mp 160–1618C. Diastereoisomer
cis : trans ratio 87 : 13. dH (500 MHz, CDCl3) 1.80 (d, J 0.76, 3H,
Thy CH3), 2.37 (ddd, J 3.1, 9.8, 14.0, 1H, H2C), 3.40 (ddd, J 7.6,
14.0, 1H, H20 C), 3.77 (d, J 13.9, 1H, HBn), 3.97 (dd, J 7.9, 9.8, 1H,
H3C), 4.20 (d, J 13.9, 1H, HBn), 6.03 (dd, J 3.1, 7.6, 1H, H1C),
7.20–7.42 (m, 7H, Ar), 7.68–7.74 (m,1H, Thy H6), 8.58–8.64
(m, 2H, Ar), 9.25 (bs, 1H, NH). dC (500 MHz, CDCl3) 12.56,
48.36, 59.66, 67.95, 83.44, 110.09, 124.04, 127.98, 128.54,

Synthesis of modified N,O-nucleosides
673
128.93, 132.68, 135.09, 135.41, 136.31, 149.43, 150.15, 150.51,
164.08. m/z (ESI-HRMS) calc. for [C20H20N4O3 þ H]þ
365.16140; found 365.16091; m/z calc. for [C20H20N4O3 þ
Na]þ 387.1433; found 387.1427.
of 25 ꢂ 103 cells in 50 mL fully defined serum-free medium
(AIM V medium; Life Technologies) and treated with the test
compounds at concentrations ranging from 10 to 200 mM. Stock
solutions (10 mM) of each compound were made up in dime-
thylsulfoxide (DMSO) and diluted in AIM V medium to give
final concentrations. The final concentrations of DMSO in the
solutions ranged from 0.1 % to 1 %. At 1 %, DMSO alone had no
effect on cell viability.
Untreated cells were placed in every plate as a negative
control. Cisplatin was employed as a control for cisplatin-
resistant SKOV3. After 72 h, 25 mL of a 3-(4,5-dimethylthiozol-
2-yl)-2,5-diphenyltetrazolium bromide solution (MTT) (12mM)
was added to each well and plates were incubated at 378C and
in 5 % CO2. After 2 h, the MTT crystals were solubilized with
100 mL of lysing buffer (50 % DMF þ 20 % sodium dodecyl
sulphate (SDS), pH 4.7); after 24 h, the spectrophotometric
absorbance of each sample was measured at 570 nm.
cis-40-Aza-40-(N-benzyl)-30-(2-chlorophenyl)-20,30-
dideoxy-5-fluorouridine 5i
White solid, yield 65 %, mp 155–1568C. Diastereoisomer
cis : trans ratio 80 : 20. dH (500 MHz, CDCl3) 2.32 (ddd, J 3.4,
8.7, 14.0, 1H, H2C), 3.37 (ddd, J 7.5, 8.8, 14.0, 1H, H20 C), 3.90 (d,
J 14.1, 1H, HBn), 4.13 (d, J 14.1, 1H, HBn), 4.48 (dd, J 7.5, 8.7,
1H, H3C), 6.03 (dd, J 3.4, 8.8, 1H, H1C), 7.15–7.65 (m, 9H, Ar),
F-Ura H6 not detected owing to large J(HF), 9.05 (bs, 1H, NH).
dC (500 MHz, CDCl3) 47.57, 61.10, 68.08, 85.43, 110.23,
127.54, 127.70, 127.95, 128.65, 128.90, 129.53, 131.43,
134.85, 135.09, 135.76, 142.81, 152.45, 164.56. m/z
(ESI-HRMS) calc. for [C20H17ClFN3O3 þ H]þ 402.1021; found
402.1032.
MDA, SKOV3, SW480, HCT116, and K562 cell lines were
purchased from American Tissue Culture Collection (ATCC;
Manassas, VA). The A2780 cell line was kindly donated by
Dr Paola Bergamini, Department of Chemical and Pharmaceu-
tical Sciences, University of Ferrara (Italy).
cis-40-Aza-40-(N-benzyl)-30-(4-chlorophenyl)-20,30-
dideoxythymidine 5j
White solid, yield 75 %, mp 163–1648C. Diastereoisomer
cis : trans ratio 91 : 9. dH (500 MHz, CDCl3) 1.79 (d, J 1.3, 3H,
Thy CH3), 2.31 (ddd, J 3.8, 9.8, 14.2, 1H, H2C), 3.34 (dt, J 7.7,
14.2,1H, H20 C), 3.70 (d, J 14.2, 1H, HBn), 3.91 (dd, J 7.7, 9.8, 1H,
H3C), 4.01 (d, J 14.2, 1H, HBn), 6.02 (dd, J 3.8, 7.7, 1H, H1C),
7.29–7.45 (m, 10H, Ar þ Thy H6), 9.23 (bs,1H, NH).
dC (500 MHz, CDCl3) 12.46, 48.27, 59.51, 69.72, 83.38,
109.91, 127.80, 128.44, 128.84, 128.92, 129.32, 134.41,
135.25, 135.55, 136.61, 150.40, 164.02. m/z (ESI-HRMS) calc.
for [C21H20ClN3O3 þ H]þ 398.1272; found 398.1269.
Supplementary Material
HPLC chromatograms, and H and 13C NMR spectra of com-
pounds 5f–k are available from the Journal’s website. Graphs of
biological activity for compounds 5d and 5f–k are provided as
supplementary material.
1
Acknowledgements
The authors thank Dr E. Bianchini, Department of Chemical and Pharma-
ceutical Sciences, University of Ferrara, for technical assistance.
cis-40-Aza-40-(N-benzyl)-30-(1-naphthyl)-20,30-
dideoxythymidine 5k
References
White solid, yield 76 %, mp 179–1808C. Diastereoisomer
cis : trans ratio 77 : 23. dH (500 MHz, CDCl3) 1.83 (d, J 1.1, 3H,
Thy CH3), 2.41 (ddd, J 3.9, 9.6, 14.2, 1H, H2C), 3.56 (dt, J 7.9,
14.2, 1H, H20 C), 3.80 (d, J 14.2, 1H, HBn), 4.16 (d, J 14.2, 1H,
HBn), 4.70 (dd, J 7.9, 9.6,1H, H3C), 6.15 (dd, J 3.9, 7.9, 1H, H1C),
7.18–8.28 (m,13H, Ar þ Thy H6), 8.85 (bs, 1H, NH). dC
(500 MHz, CDCl3) 12.39, 43.84, 47.16, 60.11, 66.51, 83.54,
109.83, 112.61, 123.73, 125.66, 125.95, 126.45, 127.71, 128.43,
128.45, 128.67, 128.84, 129.15, 133.10, 135.80, 136.92, 150.32,
163.82. m/z (ESI-HRMS) calc. for [C25H23N3O3 þ H]þ
414.1818; found 414.1821.
[1] B. D. Cheson, M. J. Keating, W. Plunkett, Nucleoside Analogs in
Cancer Therapy 1997 (Marcel Dekker: New York).
[2] E. De Clercq, Clin. Microbiol. Rev. 1997, 10, 674.
[3] J. A. Secrist, III, Nucleic Acids Symp. Ser. 2005, 49, 15. doi:10.1093/
NASS/49.1.15
[4] G. Romeo, U. Chiacchio, A. Corsaro, P. Merino, Chem. Rev. 2010,
110, 3337. doi:10.1021/CR800464R
[5] A. Procopio, S. Alcaro, A. De Nino, L. Maiuolo, F. Ortuso, G. Sindona,
Bioorg. Med. Chem. Lett. 2005, 15, 545. doi:10.1016/J.BMCL.2004.
11.048
[6] O. Bortolini, A. De Nino, L. Maiuolo, B. Russo, G. Sindona, A. Tocci,
Tetrahedron Lett. 2007, 48, 7125. doi:10.1016/J.TETLET.2007.07.
219
[7] U. Chiacchio, A. Padwa, G. Romeo, Curr. Org. Chem. 2009, 13, 422.
doi:10.2174/138527209787582268
[8] R. Romeo, S. V. Giofre`, D. Iaria, M. T. Sciortino, S. Ronsisvalle,
A. Chiacchio, A. Scala, Eur. J. Org. Chem. 2011, 5690. doi:10.1002/
EJOC.201100767
[9] J. R. Shelton, S. R. Burt, M. A. Peterson, Bioorg. Med. Chem. Lett.
2011, 21, 1484. doi:10.1016/J.BMCL.2011.01.003
[10] M. A. Peterson, M. Oliveira, M. A. Christiansen, C. E. Cutler, Bioorg.
Med. Chem. Lett. 2009, 19, 6775. doi:10.1016/J.BMCL.2009.09.083
[11] S. Miura, Y. Yoshimura, M. Endo, H. Machida, A. Matsuda,
M. Tanaka, T. Sasaki, Cancer Lett. 1998, 129, 103. doi:10.1016/
S0304-3835(98)00089-5
[12] O. Bortolini, M. D’Agostino, A. De Nino, L. Maiuolo, M. Nardi,
G. Sindona, Tetrahedron 2008, 64, 8078. doi:10.1016/J.TET.2008.
06.074
Methodology for In Vitro Biological Evaluation
MDA, A2780, SKOV3, SW480, HCT116, and K562 cell lines
were maintained in RPMI-1640 supplemented with 10 % new-
born bovine serum (HyClone; Thermo Fisher Scientific Inc.,
Waltham, MA), penicillin (100 units mLꢁ1), and streptomycin
(100 U mLꢁ1), and glutamine (2 mM); the pH of the medium
was 7.2 and incubation was at 378C in a 5 % CO2 atmosphere.
Adherent cells were routinely used at 70 % confluence and
passaged every 3 days by treatment with 0.05 % trypsin-EDTA.
K562 cells were routinely fed every 3 days. The antiproliferative
activity of compounds was tested with the 3-(4,5-dimethyl-
thiazol-2-yl)2,5-diphenyltetrazolium bromide solution (MTT)
assay, which is based on the conversion of the yellow tetrazo-
lium salt MTT into purple formazan crystals by metabolically
active cells and provides a quantitative determination of viable
cells. Cells were seeded in triplicate in 96-well plates at a density
[13] O. Bortolini, A. De Nino, T. Eliseo, R. Gavioli, L. Maiuolo, B. Russo,
F. Sforza, Bioorg. Med. Chem. 2010, 18, 6970. doi:10.1016/J.BMC.
2010.08.024

674
L. Mauiolo et al.
[14] R. H. Shoemaker, Nat. Rev. Cancer 2006, 6, 813. doi:10.1038/
NRC1951
[22] M. M. Andrade, M. T. Barros, R. C. Pintos, Tetrahedron 2008, 64,
10521. doi:10.1016/J.TET.2008.08.101
[15] S. Hatse, E. De Clercq, J. Balzarini, Biochem. Pharmacol. 1999, 58,
539. doi:10.1016/S0006-2952(99)00035-0
[23] T. Kano, T. Hashimoto, K. Maruoka, J. Am. Chem. Soc. 2005, 127,
11926. doi:10.1021/JA0523284
[16] O. Bortolini, I. Mulani, A. De Nino, L. Maiuolo, M. Nardi, B. Russo,
S. Avnet, Tetrahedron 2011, 67, 5635. doi:10.1016/J.TET.2011.
05.098
[17] C. A. Lipinski, F. Lombardo, B. W. Dominy, P. J. Feeney, Adv. Drug
Deliv. Rev. 1997, 23, 3. doi:10.1016/S0169-409X(96)00423-1
[18] F. Csizmadia, A. Tsantili-Kakoulidu, I. Panderi, F. Darvas, J. Pharm.
Sci. 1997, 86, 865. doi:10.1021/JS960177K
[24] D. A. Evans, H.-J. Song, K. R. Fardrick, Org. Lett. 2006, 8, 3351.
doi:10.1021/OL061223I
[25] P. Merino, S. Anoro, F. Merchan, T. Tejero, Heterocycles 2000, 53,
861. doi:10.3987/COM-99-8826
[26] P. Merino, S. Anoro, E. Cerrada, M. Laguna, A. Moreno, T. Tejero,
Molecules 2001, 6, 208. doi:10.3390/60300208
[27] R. Dalpozzo, A. De Nino, L. Maiuolo, A. Procopio, R. Romeo,
G. Sindona, Synthesis 2002, 172. doi:10.1055/S-2002-19795
[19] J. Owens, Drug Discov. Today 2003, 8, 12. doi:10.1016/S1359-6446
(02)02556-4
[20] M. Congreve, R. Carr, C. Murray, H. Jhoti, Drug Discov. Today 2003,
8, 876. doi:10.1016/S1359-6446(03)02831-9
[21] C. R. Strauss, R. S. Varma, Top. Curr. Chem. 2006, 266, 199.
doi:10.1007/128_060