New 3′-O-aromatic acyl-5-fluoro-2′-deoxyuridine derivatives as potential anticancer agents

DOI: 10.1016/j.ejmech.2016.03.010

Source and publish data:

European Journal of Medicinal Chemistry p. 41 - 52 (2016)

Update date:2022-08-28

Topics:

Authors:

Szymańska-Michalak, Agnieszka

Wawrzyniak, Dariusz

Framski, Grzegorz

Kujda, Marta

Zgo?a, Paulina

Stawinski, Jacek

Barciszewski, Jan

Boryski, Jerzy

Kraszewski, Adam

Read Full Text PDF DownLoad Join now for total 90,000,000 free articles

Article abstract of DOI:10.1016/j.ejmech.2016.03.010

New aromatic and aliphatic 3′-O-acyl-5-fluoro-2′-deoxyuridine derivatives were synthesized and evaluated as candidates for prodrugs against various cancer cell lines. As the most promising candidate for antimalignant therapeutics was found a dual-acting a

Full text of DOI:10.1016/j.ejmech.2016.03.010

European Journal of Medicinal Chemistry 1 15 (2016) 41e52

Contents lists available at ScienceDirect

European Journal of Medicinal Chemistry

journal homepage: h ttp://www.elsevier.com/locate/ejmech

Research paper

New 3⁰-O-aromatic acyl-5-ﬂuoro-2⁰-deoxyuridine derivatives as

potential anticancer agents

Agnieszka Szymanska-Michalak ¹, Dariusz Wawrzyniak ^{a 1}, Grzegorz Framski ^a,

ꢀ

Marta Kujda ^b, Paulina Zgoła ^a, Jacek Stawinski ^a, Jan Barciszewski ^a, Jerzy Boryski ^a,

Adam Kraszewski ^a

ꢀ

Institute of Bioorganic Chemistry, Polish Academy of Sciences, Noskowskiego 12/14, 61-704, Poznan, Poland

^bJerzy Haber Institute of Catalysis and Surface Chemistry, Polish Academy of Sciences, Niezapominajek 8, 30-239, Cracow, Poland

a r t i c l e i n f o

a b s t r a c t

New aromatic and aliphatic 3⁰-O-acyl-5-ﬂuoro-2⁰-deoxyuridine derivatives were synthesized and eval-

uated as candidates for prodrugs against various cancer cell lines. As the most promising candidate for

antimalignant therapeutics was found a dual-acting acyl derivative 7h, which apparently released not

only the known anticancer nucleoside, 5-ﬂuoro-2⁰-deoxyuridine (FdU), but also an additional active

metabolite, acetylsalicylic acid, reinforcing thus therapeutic effect of FdU. Promising therapeutic indices

showed also some aromatic dicarboxylic acids derivatives decorated with FdU esters (11 and 12).

Article history:

Received 3 December 2015

Received in revised form

2 March 2016

Accepted 3 March 2016

Available online 4 March 2016

Keywords:

5-ﬂuoro-2⁰-deoxyuridine

Nucleoside analogues

Double barrelled

Multi decorated

Anticancer

Glioblastoma multiforme

1. Introduction

target side-effects in the gastrointestinal tract and bone marrow

that may pose problems during therapy [5e7]. In addition, due to

5-Fluoro-2⁰-deoxyuridine (FdU) as an antimetabolite has been

discovered nearly 60 years ago [1] and after approval by FDA in

1970, it has been widely used in anticancer therapy [2]. Cytostatic

activity of FdU is due to depletion of thymidine 5⁰-triphosphate

(TTP) concentration as a result of thymidylate synthase inhibition.

To acquire this activity, FdU has to be ﬁrst phosphorylated to 5-

ﬂuoro-2⁰-deoxyuridine-5⁰-phosphate (FdUMP) that is a true and

the speciﬁc enzyme inhibitor [3]. Unfortunately, efﬁciency of the

FdU transformation into FdUMP in cell is hampered by a simulta-

neous cleavage of the glycosyl bond of the nucleoside by thymidine

phosphorylase that releases nucleobase 5-ﬂuorouracil (FUra). To

regain its anticancer activity FUra has to go again through the

whole biosynthetic cycle from nucleobase via FdU to FdUMP, and

this diminished anticancer potency of the original FdU drug [4].

Although clinically valuable and effective, FdU discloses off

low oral absorption, FdU requires expensive and often harmful

intravenous infusion [5]. To improve bioavailability of FdU, the

prodrug approach was recently explored for its delivery [8e10],

notably by using 3⁰, 5⁰-O-di-acyl and 3-N, 3⁰-, 5⁰-O-acyl esters, that

in several instances improved anticancer potency and therapeutic

value of FdU. The most important feature of these prodrugs was an

ability to modulate their conversion into a biologically active drug

in physiological milieu (via chemical and enzymatic hydrolysis

[11,12]), solubility and physicochemical properties in aqueous

environment, and cellular uptake (cellular membrane interaction

and transmembrane transport). Recent papers have shown [10,13]

that simple acyl esters of FdU possess the presupposed features

in terms of lipophilicity, resistance to chemical hydrolysis, and

variation in susceptibility to enzymatic hydrolysis. Although none

of the newly synthesized compounds delivered outstanding anti-

cancer properties, several valuable conclusions could be drawn

from those studies. The most interesting observation was that lip-

ophilicity, commonly accepted to be responsible for the trans-

membrane transport of a compound, although important, can't be

correlated directly with the observed cytostatic activity. Despite of

* Corresponding author.

E-mail address: akad@ibch.poznan.pl (A. Kraszewski).

Contribution of both authors was equal.

http://dx.doi.org/10.1016/j.ejmech.2016.03.010

ꢀ

A. Szymanska-Michalak et al. / European Journal of Medicinal Chemistry 115 (2016) 41e52

intensive studies on FdU acyl esters [8,11,12,14e16], none of the 3⁰-

O- or 5⁰-O-acyl-FdU derivatives found application in anticancer

therapy yet. However, several trends in modulation of cytostatic

activity of acyl-FdU as function of the acyl group used could be

distinguished. In general, presence of an acyl group usually in-

creases total lipophilicity of the compound and often improves its

IC₅₀, and thus can be used for tuning stability in physiological-like

media (RPMI/FBS). The same seems to be true for amino acid de-

rivatives of FdU [10,13,17] but in these cases, active intracellular

transport emerged as an additional factor strongly affecting cyto-

static activity. It should be also mentioned that 3⁰-carboxylic ester

of 5-ﬂuoro-2⁰-deoxyuridine bearing tumor-homing cyclic peptides

CNGRC [18] or plasminogen activator inhibitor of type II (PAI-2) [19]

disclose very selective cytostatic activity when targeting the tumor

marker APN/CD13 of HT-1080 tumor cell lines and urokinase over-

expressing malignant cells.

the desired products 5. To this end, the crude reaction mixture

containing products 5a and 6a was treated for 30 min with 3 M

excess of morpholine in methylene chloride (DCM). It was

rewarding to observe that under these conditions 3-N-benzoyl

group was removed selectively from bisacylated product 6a

without affecting 3⁰-O-benzoyl ester group in 5a. This procedure

was found to be also effective for other N-acylated products 6a-g

having different aromatic acyl groups, and allowed to obtain the

desired 3⁰-O-acyl-5-ﬂuoro-2⁰-deoxyuridines 7a-g in satisfactory

overall yields (>70%, vide Experimental) [37,38].

During synthesis of 7h, however, somewhat unexpectedly

acylation of nucleoside 1 with 2-acetylsalicylyl chloride 2h in pyr-

idine afforded several products (TLC analysis). Treatment of such

reaction mixture with morpholine (to remove possible N-acyl

groups), followed by 80% acetic acid aq. (to remove the 5⁰-O-DMT

group) produced two compounds (ratio ca 1:1), which after puri-

ﬁcation were identiﬁed (¹H, ¹³C NMR, and HRMS analyses) as target

product 7h and 3⁰-O-(2-acetylsalicylyl)salicylyl derivative 8 (Fig. 1).

To explain formation of compound 8 we considered two scenarios,

with a key role played by pyridine. In the ﬁrst one, the initially

formed product 5h was assumed to undergo a pyridine-catalyzed

transacylation of the acetyl group in the 2-acetylsalicyl moiety by

excess of acyl chloride 2h to produce 8 [39], and in the second one,

a similar phenomenon (a pyridin-catalyzed transacylation of the

acetyl group) could occur in 2-acetylsalicylyl chloride 2h itself,

prior to acylation of nucleoside 1 [40]. Since the second scenario

appeared to be more likely, in order to avoid formation of salicylyl-

salicylated products of type 8 the reactions of 1 with 2h was carried

out in methylene chloride (DCM), in the absence of pyridine. The

acylation was slow (completion overnight) and at the end of the

reaction, the TLC analysis revealed, apart from 3⁰-O-acylated-, and

3⁰-O,3-N-bisacylated products, 5h and 6h, formation of some side

products due to instability of the DMT group under the reaction

conditions. To remedy this additional problem we added to the

reaction mixture 2,6-lutidine as moderately strong (pK_a6.60) [41]

non-nucleophilic base (6 M equiv.). This secured fast (3 h), clean,

and quantitative conversion of nucleoside 1 in the reaction with 2h

into 2-acetylsalicylated products 5h and 6h. These, after a succes-

sive treatment with morpholine and 80% acetic acid, afforded after

puriﬁcation the desired 3⁰-O-(2-acetylsalicylyl)-5-ﬂuoro-2⁰-deoxy-

uridine 7h in 68% yield. The only disadvantage of this approach was

a tedious procedure for the removal of lutidine before the depro-

tection of the DMT group (at least ﬁve extractions with phosphoric

buffer pH 4.0). This inconvenience could be alleviated by using

stronger, non-nucleophilic base, namely diisopropylethylamine

(DIPEA, pK_a11.44) [42] (3 M equiv.) and this resulted also in higher

isolated yield of 7h (79%).

Inspired by the studies on 3⁰-O-retinoyl-5-ﬂuoro-2⁰-deoxyur-

idine [20,21] as a dual-acting anticancer nucleosidic compound, we

designed 3⁰-O-acyl-FdU bearing esters of carboxylic acids, that by

simple chemical or carboxylesterase-assisted hydrolysis can

generate additional active metabolites, e.g. lipoic acid [22e24],

acetylsalicylic acid [25e29], and indazole-3-carboxylic acid deriv-

ative (lonidamine [30,31]) reinforcing thus therapeutic effect of

FdU. Modulation of FdU anticancer potency of the compounds

designed herein was empowered by a set of aromatic multi car-

boxylic acid esteriﬁed with FdU that supposedly may serve as an

effective vehicle delivering in a form of one conjugated molecule

more than one of cytostatic metabolite.

2. Results and discussion

2.1. Chemistry

2.1.1. 3⁰-O-aromatic acyls-5-ﬂuoro-2⁰-deoxyuridines

The starting material for the synthesis of all 3⁰-O-acyl-5-ﬂuoro-

2⁰-deoxyuridines of type 7 (Scheme 1) was 5⁰-O-(4,4⁰-dimethoxy-

trityl)-5-ﬂuoro-2⁰-deoxyuridine, obtained by a classical Khorana's

procedure [32,33]. The choice of methods for the introduction of

acyl groups into the 3⁰-O-position of 5-ﬂuoro-2⁰-deoxyuridine

based exclusively on simplicity of the procedure and commercial

availability of the acylating reagents. Thus, 3⁰-O-acylated nucleo-

sides 7aeh bearing aromatic acyls were obtained by acylation of

suitably protected 5-ﬂuoro-2⁰-deoxyuridine derivative 1 with acyl

chlorides 2. To this end 5⁰-O-dimethoxytrityl-5-ﬂuoro-2⁰-deoxyur-

idine 1 was dissolved in pyridine and treated with the respective

acylating agent of type 2 (2 M equiv.) until the nucleosidic substrate

reacted completely (ca 2 h at room temp.). Inspection of the reac-

tion mixtures by TLC analysis usually revealed the presence of two

acylated compounds, 3⁰-O-monoacyl nucleoside of type 5 and, most

likely, 3⁰-O, 3-N-bisacyl nucleoside of type 6 (¹H and ¹³C NMR

spectroscopy analysis).

2.1.2. Double-barrelled 3⁰-O-acyl-5-ﬂuoro-2⁰-deoxyuridines

Compound 7h was ﬁrst in the series of double barrelled com-

pounds (7h-k) bearing additional to FdU pharmacophore such as

lonidamine, lipoic acid, and dichloroacetic acid. With the exception

of dichloroacetyl derivative 7k, for the synthesis of the other

compounds we chose an approach in which dimethoxytritylated

ﬂuorodeoxyuridine 1 was acylated with the corresponding car-

boxylic acid (3i and 3j respectively) activated with N-ethyl-N⁰-(3-

dimethylaminopropyl)carbodiimide hydrochloride (EDC) in the

presence of 4-(dimethylamino)pyridine (DMAP) (Scheme 1). The

choice of EDC over other carbodiimides [e.g. N,N⁰-dicyclohex-

ylcarbodiimide [43] (DCC), N,N⁰-diisopropylcarbodiimide [44] (DIC)

or unsymmetrical N-cyclohexyl-N⁰-isopropylcarbodiimide [45]] as

carboxylic acids activator was based on its proven efﬁcacy in similar

reactions of lonidamine with various carbohydrates, trouble-free

work-up, and simple isolation procedure of the ﬁnal products [46].

In a typical procedure 5⁰-protected nucleoside 1 and the desired

Formation of the bisacylated by-products 6 was studied in detail

using benzoyl chloride 2a as an acylating agent in pyridine [34]. The

structure of the isolated 6a was unambiguously assigned as a 3⁰-O,

3-N-bisacyl derivative on the basis of comparative ¹³C NMR spec-

troscopic analysis of compounds 5a, 6a and 7a [35,36]. The use of a

stoichiometric amount of acylating reagent 2a reduced signiﬁcantly

formation of the undesired N-acylation products, but at the same

time a considerable portion of the starting nucleoside 1 remained

unreacted.

Considering economy of the procedure and rather high costs of

5-ﬂuoro-2⁰-deoxyuridine, we chose an approach in which the use

of higher excess of benzoyl chloride (2 M equiv.) resulted in com-

plete consumption of the starting nucleoside 1, and then the

formed N-acylated by-products 6 were attempted to convert into

ꢀ

A. Szymanska-Michalak et al. / European Journal of Medicinal Chemistry 115 (2016) 41e52

Reagents and conditions: (i) morpholine, 3 molar equiv.; (ii) 80% acetic acid aq.

Scheme 1. Synthesis of 3⁰-O-acyl 5-ﬂuoro-2⁰-deoxyuridines.

yield (72%). No N-acylation of the nucleobase was observed under

the reaction conditions.

2.1.3. Multi carboxylic aromatic acids esteriﬁed with 3⁰-O-5-ﬂuoro-

2⁰-deoxyuridine

As a conceptual extension of gathering anticancer potency in

one molecule, we have synthesized and studied 5-ﬂuro-2⁰-deoxy-

uridine esters of di- and tricarboxylic aromatic acids i.e. iso-

phthalic acid, pyridine 3,5-dicarboxylic acid and benzene-1,3,5-

tricarboxylic acid (compounds 11, 12 and 14 respectively, Scheme

2).

Fig. 1. A side product formed during acylation of 1 with 2-acetylsalicyl chloride.

Di-(5-ﬂuoro-2⁰-deoxyuridin-3⁰-yl) benzene-1,3-dicarboxylate

and di-(5-ﬂuoro-2⁰-deoxyuridin-3⁰-yl) pyridine-3,5-dicarboxylate

11 and 12 respectively, and tri-(5-ﬂuoro-2⁰-deoxyuridin-3⁰-yl)-

benzene-1,3,5-tricarboxylate 14, were obtained in the reaction of

dicarboxylic acid chlorides 9 or 10 or benzene-1,3,5-tricarboxylic

acid chloride 13 with a slight excess of 5⁰-O-dimethoxytrityl-5-

ﬂuoro-2⁰-deoxyuridine 1 (1.05 M equiv. per one acid chloride

function) in the presence of excess of DMAP (2.5 M equiv. per one

acid chloride function) (Scheme 2). The reactions were carried out

in MeCN and the molar ratio of reagents used ensured completion

of the reaction in 2 h. After excess of DMAP was shed off the re-

action mixture was treated with acetic acid (80% aq.) to obtain the

unprotected compound. Such a two-step one-pot approach was

possible to apply since both reactions i.e. esteriﬁcation of 9, 10 and

13 as well as dimethoxytrityl group removal, proceeded practically

quantitatively. After acetic acid was evaporated, ﬁnal products 11,

12 and 14 were isolated by the silica gel column chromatography

using a stepwise gradient (0e20% v/v) of propan-2-ol in dichloro-

methane. Pure products were solidiﬁed by freeze-drying from

benzene/methanol and were obtained as white powders in high

carboxylic acid 3 (1.2 M equiv.) were dissolved in acetonitrile and

treated with EDC (1.7 M equiv.) in the presence of DMAP

(2.2 M equiv.) at room temperature. After 20 h (usually overnight)

the reactions were complete and TLC analysis of the reaction

mixtures usually revealed presence of a major product of higher R_f

value (acylated nucleosides of type 5). After evaporation of aceto-

nitrile and removal of the DMT group with 80% acetic acid, the ﬁnal

products 7i,j were isolated by the silica gel column chromatography

(yields 64e88%). In none of the instances the formation of possible

N-acylated products (vide supra) was observed, apparently, due to

lower acylating potency of the in situ activated carboxylic acid vs

acyl chlorides.

For the synthesis of dichloroacetyl derivative 7k, the last in the

series of double barrelled compounds, dichloroacetic anhydride 4

(2 M excess) was used as an acylating agent. The reaction was

carried out in pyridine in the presence of DMAP (2.5 M equiv.) to

afford 3⁰-O-acylated nucleoside 5k (TLC analysis), which upon

treatment with 80% acetic acid aq., furnished 3⁰-O-dichloroacetyl-

5-ﬂuoro-2⁰-deoxyuridine 7k. After standard work-up and the silica

gel column puriﬁcation, the ﬁnal product 7k was obtained in high

ꢀ

A. Szymanska-Michalak et al. / European Journal of Medicinal Chemistry 115 (2016) 41e52

Reagents and conditions: (i) DMAP 7.5 equiv., in MeCN. (ii) 80 % AcOH aq., 30 min.

Scheme 2. Synthesis of FdU esteriﬁed aromatic di- and tricarboxylic acids.

yields (11, 76%; 12, 72%, 14, 68%). Their structure were unambigu-

ously conﬁrmed with ¹H, ¹³C NMR and HRMS analysis.

Considering drug-likeness parameters alone i.e., M.W., logP^a,

PSA, HBD, HBA, tPapp, similar conclusions can be drawn for double

barrelled FdU derivatives 7i-k (Scheme 1, Table 1), although in

these instances the SAR might not be as straightforward as for

compounds 7a-h. For more complex and conceptually different

molecules 11, 12 and 14 (Scheme 2, Table 2) which do not obey

standard the Lipinski “rule of 5” [47e50], (MW > 500, higher

number of HBD and particularly HBA), drug-likeness prediction

needs probably other type of evaluation regime (for instance, ki-

netics of liberation of biologically active FdU, vide infra).

2.2. Physicochemical and chemical data

To evaluate drug-likeness of chemical compounds at the early

stages of a drug discovery process, several descriptors have been

devised. For example, Lipinski et al., proposed “rule of 5” [47e50],

M. Congreve et al., “rule of 3” [51], and A. Rayna et al., a set of rules

for prediction of oral drug-likeness [52]. At the moment these

“rules” are unquestionably useful in introductory considerations of

drug candidate's structure but are constantly discussed and many

modiﬁcations and extensions were postulated [53e64]. In our

studies we chose a common set of physicochemical and chemical

parameters, which would provide the most useful information on

structure relation activity (SAR) and biological potency of the

studied compounds. Additionally, because our compounds contain

aromatic rings we also included aromatic ring count which

appeared to be an important parameter deﬁning aqueous solubility,

lipophilicity, or protein binding [65]. The collective physicochem-

ical data for all the compounds investigated in this paper are listed

in Tables 1 and 2.

2.2.2. Stability in cell culture media

By checking stability of compounds 7a-h in neat RPMI, RPMI/

FBS 9:1 (v/v), and human blood plasma (HBP) one can get rough

information what can happen with the investigated compounds

under physiological conditions. Using linear regression method, the

concentrations of substrates 7a-h and their metabolites were

calculated on the basis of integration of peaks area obtained by

HPLC analysis. As the neat RPMI is free from enzymatic activities,

half-life times in this media reﬂect susceptibility of the investigated

compounds to chemical degradation (hydrolysis), while in RPMI/

FBS, and particularly in HBP, in which enzymatic activities are ex-

pected (e.g. carboxyesterases), both chemical and enzymatic factors

can be responsible for stability. As it is apparent from data in

Table 1, all the analysed compounds were rather stable under the

investigated reaction conditions and this could be correlated with

pK_avalues of the respective carboxylic acids [67,68] (the stronger

the acid, the less stable the ester). A potential carboxyesterases

activity expected for RPMI/FBS and HBP had no pronounced effect

along the series 7a-h. Similar t_1/2values in RPMI, RPMI/FBS, and

HBP supports the assumption that in these media contribution of

enzymatic transformation of 7a-h is not essential. In addition,

considering high half-times it might be presumed that during 48 h

the experimental cells are exposed mainly to unchanged com-

pounds 7a-h what should facilitate an effective cellular up-take of

the intact drug molecules.

2.2.1. Drug likeness

Except of multicarboxylic acids derivatives 11, 12 and 14 (vide

infra) the compounds investigated in this paper obey, within

acceptable tolerance, all the criteria of the Lipinski drug-likeness

“rule of 5” [47] (Table 1). In majority of cases physicochemical pa-

rameters of compounds 7a-k argue against the “rule of 3” [51] but

this disagreement seems to be rather apparent because this rule is

dedicated to leads and not to initial screenings presented herein.

The other essential for cellular up-take and bloodebrain barrier

penetration parameter, namely polar surface area (PSA), in most of

the cases do not exceed value of 90 Å²and should favour effective

cellular internalisation of a given molecule [66]. Considering anal-

ysis by T. J. Ritchie and J. F. Macdonald [65] two aromatic rings

present in compounds 7a-h should not affect their drug develop-

ability. Thus, all physicochemical parameters of compounds 7a-k

are within limits of values for high drug-likeness of potential drug

candidates.

Stabilities in cell culture media of potentially dual-acting com-

pounds 7i-k derived from FdU, bearing lonidaminyl, lipoyl and

dichloroacetyl moieties (Scheme 1) differed signiﬁcantly. The FdU-

lonidaminyl conjugate 7i was stable in all the investigated media

ꢀ

A. Szymanska-Michalak et al. / European Journal of Medicinal Chemistry 115 (2016) 41e52

Table 1

Selected physicochemical data of 3⁰-O-aryl acyl-5-ﬂuoro-2⁰-deoxyuridines 7a-k.

Cpd

FdU

LogP^a

ꢀ1.28

153.7

119.3

0.72

88.7

123.7

2.27

82.7

134.7

0.61

78.1

136.2

1.21

83.4

129.0

1.25

84.7

125.3

ꢀ0.45

92.8

113.8

0.64

143.6

87.3

0.54

86.0

128.9

3.24

78.1

167.9

1.45

143.9

78.4

0.06

153.7

160.8

PSA [Å²]

aPSA [Å²]

HBD

HBA

246.19

20.62

350.30

19.74

>240

228

144

406.41

21.42

>240

380.33

18.49

>240

384.74

20.47

81.6

105

384.74

18.37

82.5

144.4

58.5

351.29

19.65

21.5

105

395.30

11.38

12.7

28.8

408.33

18.73

21.4

41.3

21.8

549.34

20.68

>240

434.50

14.72

72.2

105

14.5

357.12

20.11

<0.1

tP_app(nm/s)

t_1/2[h] RPMI

t_1/2[h] RPMI/FBS

t_1/2[h] HBP

No of Ar

LogP^ae ALOGPS 2.1 software (vide Experimental). PSA e polar surface area, HBD e hydrogen bonds donors, HBA e hydrogen bonds acceptors, tPapp e apparent theoretical

permeability; RPMI e RPMI 1640 media; FBS e foetal bovine serum, HBP e human blood plasma, MRC-5 e normal human foetal lung ﬁbroblast, SI e IC₅₀non-tumor cell line/

IC₅₀tumor cell line.

Table 2

^aSelected physicochemical and pharmacokinetic data of multicarboxylic acid esters decorated with 3⁰-O-5-ﬂuoro-2⁰-deoxyuridine, 11,12 and 14.

Cpds

FdU

Structure

RPMI

RPMI/FBS

ꢀ0.46

107.9

159.5

6.3 h

2.8 h

5.2 h

ꢀ1.64

111.3

155.8

ꢀ1.61

105.8

203.9

1/2

LogP

ꢀ1.28

153.7

119.3

PSA (Å²)

aPSA (Å²)

HBD

HBA

246.19

622.49

623.47

894.68

Absorption

tP_app(nm/s)

20.62

17.25

17.33

15.25

IC₅₀

M]/SI

HeLa

Caco-2

T-47D

5.31/4

12.85/2

5.61/4

5.57/4

23.40/1

10.37/2

22.46

2.57/17

>100/<1

3.83/12

17.97/2

31.16/1

26.03/1

44.15

8.18/10

>100/<1

14.73/6

6.45/13

18.86/4

10.63/8

84.50

7.68/4

31.16/1

8.17/4

7.03/4

29.71/1

10.49/3

30.74

T98G

U-118 MG

U-87 MG

MRC-5

Abbreviations as in Table 1.

(t_1/2> 240 h), while 3⁰-O-dichloroacetyl-FdU 7k under the same

conditions hydrolysed readily (t_1/2< 0.1 h). Lipoyl-FdU 7j conjugate

showed moderate stability in neat RPMI (chemical stability), but in

the RPMI/FBS and HBP media, which contain enzymatic activity, t_1/2

value of 7j changed somewhat unpredictable, possibly as a result of

different substrate afﬁnity towards enzyme present in these media

(Table 1).

It seems that stability of FdU multicarboxylates in cell culture

media as well as in the cell might be crucial for their cytostatic

activity because the active part after hydrolysis of multiester is free

nucleoside e FdU. It might be presumed then, that their activity

should correlate straightforward with unleashing of FdU. Unfortu-

nately, we were able to measure half-time only for diFdU ester of

3,5-dicarboxy pyridinic acid 12 because of solubility of compound

11 and 14 in aqueous environment (RPMI or HBP) was well below

2 mM and was insufﬁcient for monitoring of their metabolism by

HPLC under our standard analytical conditions (initial concentra-

tion 2 mM). Stability of compound 12 in the discussed media

followed the order RPMI > HBP > RPMI/FBS (Table 2). This stability

order points to involvement of enzymatic hydrolysis of the inves-

tigated compounds. In addition, what potentially might be bene-

ﬁcial for pharmacodynamics of compounds of type 12 is that

hydrolysis of the second carboxylic ester was much slower even in

RPMI/FBS (see Supplementary data). This is in agreement with

known features of carboxyestrases which disclose much lower

reactivity towards polar or charged substrates [69].

Although stability of compounds 11 and 14 could not be

measured, it is likely that all rules governing their kinetics of hy-

drolysis remain similar as those for compound 12 described above.

2.2.3. Hydrodynamic diameters

Since most of the compounds studied herein are amphiphilic

and consist of a polar part i.e. 5-ﬂuorouracil-1-yl and 2⁰-deoxyri-

bose residue at one side, and a lipophilic aromatic substituent at the

3⁰-position of nucleoside at the other one, one can suspect forma-

tion of microstructures (e.g. micromicelles) in aqueous

ꢀ

A. Szymanska-Michalak et al. / European Journal of Medicinal Chemistry 115 (2016) 41e52

environment. This might reinforced their drug-likeness and facili-

tate the cellular and/or intestinal up-take [70,71]. For this reason

we attempted to measure hydrodynamic diameters using a dy-

namic light scattering (DLS) method, expecting that if any micro-

structures will be formed they should be detected by this method.

The obtained results for 7a indicated that this type of compounds

may form stable, DLS-detectable aggregates of size close to

200e400 nm. After dilution by pure water to the ﬁnal concentra-

tion in the range of 1e0.25 mg/mL, the values of hydrodynamic

diameter remains unchanged, what may suggests that the inves-

tigated compounds under the experimental conditions did not

form micelle-like structures.

O-acyl-5-ﬂuoro-2⁰-deoxyuridines 7a-h are more cytotoxic than the

parent FdU in the investigated cancer cell lines and in the reference

normal human foetal lung ﬁbroblasts (MRC-5) (Table 3). Consid-

ering character of the 3⁰-O-acyl groups, it seems reasonable to

postulate that the observed higher antimalignant potency of 7a-h

vs FdU might be attributed to their higher lipophilicity (as is

apparent from the corresponding logP, PSA and aPSA values,

Table 1), that should facilitate cell membrane penetration and

cellular up-take.

Regardless of a higher cytotoxicity (IC₅₀) of 7a-g in all examined

carcinoma cells, their potential therapeutic value as measured by a

selectivity index (SI), is lower than that of the parent FdU (in most

of the cases SI < 1) and thus do not justify their further develop-

ment as potential anticancer drugs. The only exception in this

subseries of the aromatic acyl derivatives, was a double barrelled

3⁰-O-(2-acetylsalicyl)-5-ﬂuoro-2⁰-deoxyuridine (7h) whose cyto-

toxicity in each cancer cell lines investigated was distinctly higher

than FdU (in some cases well above ten times) with superior

selectivity indices (SI), e.g. for HeLa and Caco-2 cell lines (SI 18 and

33 respectively), and for the series of glioma cell lines T98G, U-

118 MG and U-87MG (SI 5, 6 and 14 respectively). Considering that

apart from FdU, the other metabolite generated form 7h is aspirin

or salicylic acid, the observed phenomenon is in line with the

ﬁnding that both compounds can induce growth inhibition of

various malignancies while leaving normal cells unaffected

[25e27,29]. The pronounced differences in SI values of 7a-g vs 7h

were rather unexpected but it might be speculated that kinetics of

internalization and intracellular metabolism of 7h towards bio-

logically active FdU and aspirin or salicylic acid is so far distinct

from the analogous decomposition of 7a-g that may result in much

higher selectively of growth inhibition of cancer cells vs normal

cells (MRC-5).

2.3. In vitro cytostatic activity

Glioblastoma multiforme (GBM) is the most common, most

aggressive, and lethal type of primary brain tumor. Current thera-

pies for GBM (surgical resection, radiotherapy and chemotherapy)

are weakly successful [72] with ﬁve-year survival rates less than 5%

(according to The Central Brain Tumor Registry of the United States,

2012). Therefore, novel effective therapeutic agents for treating

GBM are urgently needed. It has been shown recently [73] that the

anticancer drugs from distinct pharmacological classes (other than

temozolomide e widely used drug in GBM treatment) exert

inhibitory effects on GBM cell growth and invasion.

Few years ago Barciszewski et al., from our institute, reported on

a successful application of a sequence speciﬁc iRNA in brain tumor

therapy [74,75]. Since FdU was hardly explored as drug against

GBM, in this studies we focused on law molecular weight, novel

FdU derivatives 7a-7k, 11, 12 and 14 which were evaluated for their

antiproliferative activity against six human cancer cell lines,

including three glioma cell lines i.e. T98G, U-118 MG, U-87 MG.

Similar investigations were performed also on the cervical (HeLa),

breast (T-47D) and colon (Caco-2) cancer cell lines because of

slightly different reasons. HeLa cells were used as commonly

accepted standard of human malignant cells, T47D cells were used

because 5-ﬂuorouracil (FU) is often one of the components in

breast cancer multi-drug therapy [76,77], and Caco-2 cells were

used for preliminary evaluation of intestinal up-take potency of the

examined compounds when administrated orally [78]. Finally, non-

cancerous lung ﬁbroblasts (MRC-5) were used as the reference for

estimation of therapeutic value of the examined compounds for

malignant vs normal cells. The results of antiproliferative activity of

FdU derivatives 7a-k, 11, 12 and 14 are listed in Table 3 and Table 2,

respectively.

2.3.2. Double barrelled FdU derivatives

In comparison to the parent FdU, with a few exceptions (i.e. 7a,

7f and Caco-2 cells, 7a, 7d, 7f, 7h, 7k and T-47D cells, 7h and T98G

cell line, 7d, 7e and U-118MG cells), all investigated compounds

disclosed higher (up to 90 times, vide U-118G cells and compound

7a) cytostatic activity in cancer cell lines but with exception of 7h,

also clearly higher cytotoxicity for referential normal MRC-5 cells

(Table 3). This makes application of these compounds rather

doubtful in cancer therapy. Unfortunately, this tentative conclusion

concerns also to compounds 7i-k armed with two type of anti-

cancer activities (FdU and second metabolite lonidamine, lipoic

acid, dichloroacetic acid, respectively). Although some of them

showed antiproliferating activity but none of them disclosed anti-

cancer potency advantageous over the parent FdU.

2.3.1. 3⁰-O-acyl- FdU derivatives of aromatic carboxylic acids

The obtained results indicate that, with a few exceptions, all 3⁰-

However, selectivity indices SI of 3⁰-O-benzoyl- and 3⁰-O-

Table 3

^aCytostatic and SI values [IC₅₀M)/SI] of 3⁰-O-aryl acyl-5-ﬂuoro-2⁰-deoxyuridines 7a-k.

Cpd

FdU

HeLa

Caco-2

T-47D

5.31/4

12.85/2

5.61/4

5.57/4

23.40/1

10.37/2

22.46

0.57/3

0.39/<1

0.18/<1

2.98/<1

1.11/<1

3.88/<1

0.34/<1

0.07

0.55/1

0.38/2

2.06/<1

0.39/2

19.26/<1

0.52/1

0.66

6.46/<1

1.26/2

13.77/<1

1.50/2

41.35/<1

0.64/4

2.46

1.75/<1

0.42/1

0.68/1

0.67/1

46.27/<1

0.42/1

0.60

0.94/4

0.87/1

2.12/<1

1.06/1

0.40/3

0.65/2

0.88/1

1.14

2.75/18

1.53/33

15.04/3

9.69/5

8.79/6

3.58/14

50.24

1.12/4

1.03/4

4.41/1

3.97/1

9.21/1

5.65/1

4.15

0.86/1

6.92/<1

7.23/<1

0.33/2

6.19/<1

0.56/1

0.59

6.95/<1

12.23/<1

7.02/<1

5.09/<1

2.71/1

n.d*.

36.96/<1

7.52/<1

0.44/<3

0.26/<6

0.53/<3

1.48

70.55/<1

16.35/<1

0.66/<1

5.48/<1

13.35/<1

3.30

T98G

U-118 MG

U-87 MG

MRC-5

3.75

*n.d. e not determined.

Abbreviations as in Table 1.

ꢀ

A. Szymanska-Michalak et al. / European Journal of Medicinal Chemistry 115 (2016) 41e52

acetylsalicylyl derivatives of FdU (7a and 7h respectively), stand out

from the others due to their SI parameter for GMB cell lines that are

in each case exceeding those of FdU (Table 3). It is worthy to notice

that in the case of 7a, favorably values of SI parameters were due to

its high cytostatic activity (low IC₅₀) while in the case of compound

7h the high SI values were related to its low toxicity towards non-

malignant MRC-5 cells. It is also worth to notice that 7h disclosed

highest therapeutic potency in Caco-2 and HeLa cells (SI ¼ 33 and

18, respectively, Table 3) what makes this compound therapeuti-

cally promising in ﬁghting malignancy of other tissues. To lesser

extend similar conclusions concern also compound 7g.

deoxyuridin-3⁰-yl)-1,3,5-tricarboxybenzoate 14 did not showed any

advantages over 5-ﬂuoro-2⁰-deoxyuridine in all of the investigated

cell lines.

3. Conclusions

We have synthesised new aromatic and aliphatic 3⁰-O-acyl-5-

ﬂuoro-2⁰-deoxyuridine derivatives 7a-k, 11, 12 and 14 as potential

anticancer agents. The chemistry involved acylation of 5⁰-O-dime-

toxytrityl-5-ﬂuoro-2⁰-deoxyuridine with various acylating reagents

(acyl chlorides, carboxylic acid anhydrides) or reagent systems

(carboxylic acid in combination with condensing agents) that

afforded the corresponding 3⁰-O-acylated FdU derivatives in good

yield. In some cases the formation of O,N-bisacylated nucleosides

was observed, but this problem was remedied by a selective

Some comments deserve results of experiments performed on

Caco-2 cells in which with exception of compounds 7a and 7f all

other investigated compounds showed much higher anticancer

activity (IC₅₀values in the range 0.18e6.92

mM) than FdU

(IC₅₀¼ 12.85

M). Considering that monolayer Caco-2 cells are

removal of the N-acyl groups by a short treatment with

morpholine.

widely used as a in vitro model for prediction of drug absorption in

humans [79e81], it might be concluded that compounds 7b-e and

7g-l should be well absorbed from intestinal ﬂuid. Derivatives 7a

and 7f proved to be poorly active in Caco-2 cells (IC₅₀36.96 and

For evaluation of antimalignant potency of the synthesised

compounds three lines of glioblastoma cells (T98G, U-118 MG and

U-87 MG), HeLa cells, breast cancer cells (T-47D), and Caco-2 cells

were chose. The ﬁrst series of the investigated compounds, 7a-g,

consisted of FdU acylated with electronically and structurally

different aromatic groups, and did not provide any signiﬁcant im-

provements in terms of SI over the parent FdU. In contrast to this,

for the double-barrelled 3⁰-O-(2-acetylsalicylyl)-5-ﬂuoro-2⁰-deox-

yuridine 7h, the therapeutic indices were notably higher than those

of FdU (except for breast cancer cells, T-47D). Compound 7i-k,

bearing potentially active additional metabolites (lipoic acid, 7i;

lonidamine, 7j; dichloroacetic acid, 7k) turned out to be disap-

pointing as anticancer agents in the investigated cell lines.

Finally, in the third series of compounds investigated, multi-FdU

esters of di- and tri-carboxylic aromatic acids i.e. 11, 12 and 14, di-

FdU 1,3-dicarboxybenzoate 11 appeared to be active in two cell

lines (HeLa and breast cancer cells T-47D), with the SI indices ca

three times higher than that of FdU. Compound 12 had therapeutic

parameters evidently superior over the parent nucleoside (FdU) in

glioblastoma cell lines T98G, U-118 MG and U-87 MG, mainly due to

its much lower toxicity. The IC₅₀and SI parameters of benzene-

tricarboxylic acid derivative 14 in all the investigated cell lines were

similar to those of FdU.

70.55 mM respectively) and their low antiproliferative activity may

suggests that they are substrates for P-glycoprotein, a protein

belonging to the ATP-binding cassette (ABC) transporters super-

family, that has clinical relevance due to its role in drug metabolism

and multidrug resistance (MDR). The last partially argues against

calculated tP_appvalue because all of the compounds including 7a

and 7f (19.74 nm/s and 19.65 nm/s respectively) belong to class of

medium permeability [82] with tP_appin the range of 11.38 nm/s (for

7g) and 21.42 nm/s (for 7b). Apparently, low antiprolifeartive po-

tency of 7a and 7f in Caco-2 cells was an effect of unrecognized yet

intracellular and not outer membrane “molecular” events.

Considering therapeutic potency of the examined derivatives of

FdU 7a-h and the double barreled 7i-k (Table 3) in ﬁghting breast

cancer cell line T-47D, their parameters IC₅₀and SI clearly point out

that these compounds (with exception of 7h) did not exceed

therapeutic potency of FdU. Although antiproliferative activities

(IC₅₀) of 7b, 7c, 7e, 7g, 7i were higher than that of FdU but at the

same time their cytoxicity towards non-malignant cells was also

higher (Table 3), the SI values for these compounds did not exceed 1

(except for 7h, SI ¼ 3).

Although not impressive in terms of biological activity, some of

the investigated compounds (for instance 7h and 12) can constitute

a launch pad for further studies on double-barrelled and multi

loaded antimalignant potential drugs.

2.3.3. FdU esters of multi carboxylic acids

Biological activity found for FdU esters of multi carboxylic acids

11, 12 and 14 are somehow slightly difﬁcult to rationalise. In com-

parison to FdU, all these compounds disclosed comparable anti-

proliferative activity but at the same time were less toxic (up to four

times, compound 12) for normal MRC-5 cells (Table 2). As a result

their SI value in several cases were essentially improved as compare

to those of FdU. For example, di-(5-ﬂuoro-2⁰-deoxyuridin-3⁰-yl)-

3,5-dicarboxypyridinylate 12 showed higher therapeutic potency in

nearly all the examined cell lines (with the exception of Caco-

2 cells) and the SI exceeded even six times (for T98G cells) that of

FdU. This apparent increase in value of SI parameter of compound

12 vs FdU, might points to this compound as a good candidate for a

new antimalignant pro-drug. As such conclusion might be prema-

ture, the results obtained for dicarboxylate 12 should certainly

warrant further studies on development of anticancer compounds

based on FdU skeleton. The second noteworthy in this series

4. Experimental

4.1. Material and methods

¹H, ¹³C NMR spectra were recorded on Bruker Avance II 400 or

500 MHz machines. Mass spectra were recorded with the ESI

technique with negative or positive ionization with accuracy below

5 ppm. Amount of water in anhydrous solvents was controlled

using Karl Fischer coulometric titration (Metrohm 684 KF coulo-

meter). Thin-layer chromatography was performed on Merck silica

gel 60F₂₅₄plates and visualized with UV. For column chromatog-

raphy Kieselgel 60 Merck was used. HPLC analyses were performed

on a Nucleosil 100-5C18 column (5.0

mm, 4.6 mm ꢁ 150 mm) using

compound

di-(5-ﬂuoro-2⁰-deoxyuridin-3⁰-yl)-1,3-

and Waters Breeze HPLC systems with A þ B solvent systems (A,

0.01 M aqueous triethylammonium acetate pH 7.4; B, A/acetoni-

trile, 1:4, v/v) at 35 ^ꢂC, ﬂow rate 1.5 mL/min; events: 5 min A 100%,

linear gradient of B 0e100% in 20 min, 5 min B 100% and A 100%

10 min wash. Lonidamine was obtained as described previously by

I.G. George et al. [83]. Aromatic di-carboxylic acid dichlorides were

obtained according to R. Martin et al. [84]. Hydrodynamic

dicarboxybenzoate 11 that showed much higher antiproliferative

potency than FdU in two cell lines, i.e. HeLa (four times) and breast

cancer T-47D cells (three times). The last ﬁnding can be of interest

because none of the investigated herein compounds disclosed ad-

vantageous over the parent FdU therapeutic potency in this cell

line. Finally, it was slightly surprising that tri-(5-ﬂuoro-2⁰-

ꢀ

A. Szymanska-Michalak et al. / European Journal of Medicinal Chemistry 115 (2016) 41e52

diameters were measured with dynamic light scattering (DLS)

method using Zetasizer Nano ZS Malvern. Samples were prepared

in DMSO at the following concentrations: 7 mg/mL, 2 mg/mL and

1 mg/mL. Next, the samples were diluted by pure water and the

hydrodynamic size of compounds was measured using the glass

cuvette. To check stability of prepared solutions the measurements

were performed in intervals of time during 12 h.

(d, J ¼ 8.4 Hz, 2H, o-H of Ar), 7.57 (d, J ¼ 8.4 Hz, 2H, m-H of Ar), 6.26

(t, J ¼ 6.8 Hz, 1H, H-1⁰), 5.48e5.49 (m, 1H, H-3⁰), 4.20e4.21 (m, 1H,

H-4⁰), 3.71e3.72 (m, 2H, H-5⁰, H-5⁰⁰), 2.39e2.43 (m, 2H, H-2⁰, H-2⁰⁰),

1.31 (s, 9H, CH₃of t-Bu) ppm, ¹³C NMR (100 MHz, DMSO-d₆) d_C

165.56 (C]O), 157.40 (d, J ¼ 26 Hz, C-4), 157.12 (C of Ar), 149.47 (C-

2), 140.53 (d, J ¼ 230 Hz, C-5), 129.68, 127.045 & 126.013 (C of Ar),

124.91 (d, J ¼ 34 Hz, C-6), 85.35 (C-4⁰), 84.93 (C-1⁰), 75.84 (C-3⁰),

61.74 (C-5⁰), 37.47 (C-2⁰), 35.28 [C(CH₃)₃], 31.19 [C(CH₃)₃] ppm;

HRMS (ESI): calcd for C₂₀H₂₂N₂O₆F [M-H]^ꢀ: 405.14619, found:

405.14736.

4.1.1. General procedure for the synthesis of 3⁰-O-acyl-5-ﬂuoro-2⁰-

deoxyuridines of type 7

Method A (for compounds 7a-h). 5⁰-O-dimethoxytrityl-5-ﬂuoro-2⁰-

deoxyuridine 1 (1 mmol) was rendered anhydrous by evaporation

of the added pyridine and then dissolved in pyridine (10 mL) (or in

DCM with 3 M equiv. excess of DIPEA for compound 7 h). To this,

the respective acyl chloride of type 2 (2 mmol) was added, and the

reaction mixture was left for 2e3 h at room temp yielding two

acylated compounds of type 5 and 6. Then the solvent was evap-

orated, the crude mixture dissolved in DCM (10 mL) was treated

for 30 min with 3 M excess of morpholine. After this time the

resulting mixture was washed twice with water (2 ꢁ 5 mL), the

organic layer separated, evaporated and treated with 80% acetic

acid aq. (5 mL) for 30 min. After deprotection with AcOH, the

mixtures were concentrated to an oil under reduced pressure.

3⁰-O-acyl-5-ﬂuoro-2⁰-deoxyuridines 7 were isolated by a silica gel

60 column chromatography using a stepwise gradient (0e10%) of

propan-2-ol in methylene chloride. The fractions containing pure

products were collected and evaporated yielding non-hygroscopic

foams. After freeze-drying from benzene, compounds 7 were

obtained as amorphous solids.

3⁰-O-(4-methoxybenzoyl)-5-ﬂuoro-2⁰-deoxyuridine

(7c). White

solid; yield 0.28 g, 75%; RP HPLC Rt 14.14 min; ¹H NMR (400 MHz,

DMSO-d₆): d_H11.88 (s, 1H, NH), 8.26 (d, J _6-F5¼ 7.2 Hz, 1H, H-6), 7.96

(d, J ¼ 8.8 Hz, 2H, o-H of Ph), 7.06 (d, J ¼ 8.8 Hz, 2H, m-H of Ar), 6.26

(t, J ¼ 6.8 Hz,1H, H-1⁰), 5.43e5.45 (m,1H, H-3⁰), 4.17e4.18 (m,1H, H-

4⁰), 3.84 (s, 3H, OCH₃), 3.70e3.73 (m, 2H, H-5⁰, H-5⁰⁰), 2.38e2.43 (m,

2H, H-2⁰, H-2⁰⁰) ppm; ¹³C NMR (100 MHz, DMSO-d₆) d_C165.31 (C]

O),163.81 (C of Ar),157.40 (d, J ¼ 26 Hz, C-4),149.47 (C-2),140.52 (d,

J ¼ 230 Hz, C-5), 131.91 (C of Ar), 124.92 (d, J ¼ 34 Hz, C-6), 121.90,

114.48 (C of Ar), 85.40 (C-4⁰), 84.94 (C-1⁰), 75.67 (C-3⁰), 61.73 (C-5⁰),

55.73 (OCH₃), 37.27 (C-2⁰) ppm; HRMS (ESI): calcd for C₁₇H₁₆N₂O₇F

[M-H]^ꢀ: 379.09415, found: 379.09524.

3⁰-O-(2-chlorobenzoyl)-5-ﬂuoro-2⁰-deoxyuridine (7d). White solid;

yield 0.33 g, 85%; RP HPLC Rt 14.52 min; ¹H NMR (400 MHz, DMSO-

d₆): d_H11.88 (s, 1H, NH), 8.25 (d, 1H, J _6-F5¼ 7.2 Hz, H-6), 7.88e7.91

(m, 1H, o-H of Ar), 7.59e7.62 (m, 2H, m-H of Ar), 7.48e7.52 (m, 1H,

p-H of Ar), 6.23e6.26 (m, 1H, H-1⁰), 5.49e5.51 (m, 1H, H-3⁰),

4.19e4.21 (m, 1H, H-4⁰), 3.71e3.73 (m, 2H, H-5⁰, H-5⁰⁰), 2.41e2.45

(m, 2H, H-2⁰, H-2⁰⁰) ppm; ¹³C NMR (100 MHz, DMSO-d₆): d_C164.90

(C]O), 157.42 (d, J ¼ 27 Hz, C-4), 149.47 (C-2), 141.54 (d, 230 Hz, C-

5), 133.94, 132.36, 131.80, 131.27, 129.98, 127.89 (C of Ar), 124.91 (d,

J ¼ 35 Hz, C-6), 85.16 (C-4⁰), 84.94 (C-1⁰), 76.73 (C-3⁰), 61.73 (C-5⁰),

37.27 (C-2⁰) ppm; HRMS (ESI): calcd for C₁₆H₁₃N₂O₆ClF [M-H]^ꢀ:

383.04462, found: 383.04573.

Method B (for compounds 7i-k). 5⁰-O-dimethoxytrityl-5-ﬂuoro-2⁰-

deoxyuridine 1 (1 mmol) and the respective carboxylic acids 3

(1.2 mmol) were rendered anhydrous by the evaporation of the

added pyridine and then were dissolved in acetonitrile (10 mL) and

treated by EDC (1.7 mmol) in the presence of DMAP (2.2 mmol) at

room temperature. After 20 h the reaction was complete (TLC

analysis), acetonitrile was evaporated, and the residue was treated

with 80% acetic acid aq., (5 mL) for 30 min. Further work-up as for

Method A.

3⁰-O-(4-chlorobenzoyl)-5-ﬂuoro-2⁰-deoxyuridine (7e). White solid;

yield 0.29 g, 75%; RP HPLC Rt 15.32 min; ¹H NMR (400 MHz, DMSO-

d₆): d_H11.88 (s, 1H, NH), 8.26 (d, 1H, J

¼ 7.2 Hz, H-6), 8.02 (d,

6-F5

Method C (for compound 7l). 5⁰-O-dimethoxytrityl-5-ﬂuoro-2⁰-

deoxyuridine 1 (1 mmol) was rendered anhydrous by the evapo-

ration of the added pyridine and then dissolved in pyridine

(10 mL). To this, 2 M excess of dichloroacetic anhydride 4 in the

presence of 2.5 M equiv. of DMAP was added. After 4 h the reaction

was complete (TLC analysis), solvent was evaporated, and the

residue was treated with 80% acetic acid (5 mL). Further work-up

as for Method A.

J ¼ 8.4 Hz, 2H, o-H of Ar), 7.62 (d, J ¼ 8.4 Hz, 2H, m-H of Ar), 6.27 (t,

J ¼ 7.2 Hz, 1H, H-1⁰), 5.47e5.48 (m, 1H, H-3⁰), 4.19e4.21 (m, 1H, H-

4⁰), 3.66e3.76 (m, 2H, H-5⁰, H-5⁰⁰), 2.32e2.50 (m, 2H, H-2⁰, H-2⁰⁰);

¹³C NMR (100 MHz, DMSO-d₆): d_C164.43 (C]O), 157.13 (d,

J ¼ 26 Hz, C-4), 149.47 (C-2), 140.11 (d, J ¼ 229 Hz, C-5), 138.53,

131.24, 128.96, 128.21 (C of Ar), 124.17 (d, J ¼ 34 Hz, C-6), 84.86 (C-

4⁰), 84.52 (C-1⁰), 75.97 (C-3⁰), 61.32 (C-5⁰), 36.96 (C-2⁰) ppm; HRMS

(ESI): calcd for

383.04562.

₁₆H₁₃N₂O₆ClF [M-H]^ꢀ: 383.04407, found:

3⁰-O-benzoyl-5-ﬂuoro-2⁰-deoxyuridine (7a). White solid; yield

0.29 g, 83%; RP HPLC Rt 13.96 min; ¹H NMR (400 MHz, DMSO-d₆):

d_H11.88 (s, 1H, NH), 8.27 (d, 1H, J _6-F5¼ 7.2 Hz, H-6), 8.03e8.01 (m,

2H, o-H of Ar), 7.71e7.67 (m, 1H, p-H of Ar), 7.58e7.54 (m, 2H, m-H

of Ar), 6.24 (t, J ¼ 6.6 Hz, 1H, H-1⁰), 5.48 (d, J ¼ 5.6 Hz, 1H, H-3⁰),

4.20e4.21 (m, 1H, H-4⁰), 3.70e3.74 (m, 2H, H-5⁰, H-5⁰⁰), 2.40e2.46

(m, 2H, H-2⁰, H-2⁰⁰) ppm; ¹³C NMR (100 MHz, DMSO-d₆): d_C165.65

(C]O), 157.40 (d, J ¼ 26 Hz, C-4), 149.47 (C-2), 141.53 (d, J ¼ 230 Hz,

C-5), 134.03, 129.75, 129.20 (C of Ar), 124.93 (d, J ¼ 34 Hz, C-6),

85.33 (C-4⁰), 84.95 (C-1⁰), 76.07 (C-3⁰), 61.74 (C-5⁰), 37.43 (C-2⁰)

ppm; HRMS (ESI): calcd for C₁₆H₁₄N₂O₆F [M-H]^ꢀ: 349.08359,

found: 349.08459.

3⁰-O-(nicotinoyl)-5-ﬂuoro-2⁰-deoxyuridine (7f). White solid; yield

0.27 g, 78%; RP HPLC Rt 11.75 min; ¹H NMR (400 MHz, DMSO-d₆):

d_H11.88 (s, 1H, NH), 9.15e9.16 (m, 1H, o-H of Ar), 8.83e8.85 (m, 1H,

p-H of Ar), 8.34e8.37 (m, 1H, o-H of Ar), 8.26 (d, 1H, J _6-F5¼ 7.2 Hz,

H-6), 7.58e7.61 (m, 1H, m-H of Ar), 6.29 (t, J ¼ 7.2 Hz, 1H, H-1⁰),

5.50e5.51 (m, 1H, H-3⁰), 4.24 (br s, 1H, H-4⁰), 3.68e3.77 (m, 2H, H-

5⁰, H-5⁰⁰), 2.39e2.44 (m, 2H, H-2⁰, H-2⁰⁰) ppm; ¹³C NMR (125 MHz,

DMSO-d₆): d_C164.77 (C]O), 157.47 (d, J ¼ 27 Hz, C-4), 154.34,

150.70 (C of Ar), 149.53 (C-2), 141.59 (d, J ¼ 228 Hz, C-5), 137.56,

125.91 (C of Ar), 125.03 (d, J ¼ 34 Hz, C-6), 124.39 (C of Ar), 85.24 (C-

4⁰), 85.03 (C-1⁰), 76.55 (C-3⁰), 61.78 (C-5⁰), 37.4 (C-2⁰) ppm; HRMS

(ESI): calcd for C₁₅H₁₃N₃O₆F [M-H]^ꢀ: 350.07884, found: 350.07959.

3⁰-O-(4-tert-butylbenzoyl)-5-ﬂuoro-2⁰-deoxyuridine

(7b). White

solid; yield 0.30 g, 73%; RP HPLC Rt 17.56 min; ¹H NMR (400 MHz,

3⁰-O-(4-nitrobenzoyl)-5-ﬂuoro-2⁰-deoxyuridine (7g). Yellow solid;

DMSO-d₆): d_H11.89 (s, 1H, NH), 8.26 (d, 1H, J _6-F5¼ 7.2 Hz, H-6), 7.94

yield 0.28 g, 71%; RP HPLC Rt 15.54 min; ¹H NMR (400 MHz, DMSO-

ꢀ

A. Szymanska-Michalak et al. / European Journal of Medicinal Chemistry 115 (2016) 41e52

d₆): d_H11.89 (s,1H, NH), 8.35e8.38 (m, 2H, o-H of Ar), 8.24e8.28 (m,

4.1.2. General procedure for the synthesis of 3⁰-O-esters of 5-ﬂuoro-

3H, m-H of Ar and H-6), 6.28e6.32 (m, 1H, H-1⁰), 5.52e5.53 (m, 1H,

H-3⁰), 4.25 (br s, 1H, H-4⁰), 3.68e3.78 (m, 2H, H-5⁰, H-5⁰⁰), 2.40e2.55

(m, 2H, H-2⁰, H-2⁰⁰) ppm; ¹³C NMR (125 MHz, DMSO-d₆): d_C164.30

(C]O), 157.46 (d, J ¼ 27 Hz, C-4), 150.84 (C of Ar), 149.55 (C-2),

141.61 (d, J ¼ 230 Hz, C-5), 135.35, 131.38 (C of Ar), 125.49 (d,

J ¼ 34 Hz, C-6), 124.34 (C of Ar), 85.24 (C-4⁰), 85.01 (C-1⁰), 77.08 (C-

3⁰), 61.82 (C-5⁰), 37.38 (C-2⁰) ppm; HRMS (ESI): calcd for

2⁰-deoxyuridines with aromatic multi carboxylic acids

Nucleoside 1 (1.05 equiv. per one acid chloride function) was

rendered anhydrous by evaporation with dry pyridine and dis-

solved in acetonitrile (10 mL/1 mmol of nucleoside 1). To this so-

lution DMAP (2.5 equiv per one acid chloride function) and the

respective carboxylic acid chloride (1 mmol) were added and the

reaction mixture was left for 2 h at room temperature (TLC anal-

ysis). The reaction was quenched by addition of water (1 mL) and

the solution concentrated to a viscous oil. The residue was dis-

solved in methylene chloride (10 mL/1 mmol) and washed three

times with water (3 ꢁ 5 mL), the organic layer separated, and after

drying (Na₂SO₄anhyd.) it was evaporated under vacuum. After

dimethoxytrityl group was removed (80% acetic acid, 5 mL, 30 min)

and the ﬁnal product was isolated by a silica gel 60 column chro-

matography using a stepwise gradient 0 / 20% (v/v) propane-2-ol

in dichloromethane. The fractions containing pure products were

combined, evaporated and solidiﬁed by freeze-drying from ben-

zene/methanol.

₁₆H₁₃N₃O₈F [M-H]^ꢀ: 394.06867, found: 394.06964.

3⁰-O-(2-acetylsalicylyl)-5-ﬂuoro-2⁰-deoxyuridine (7h). White solid;

yield 0.32 g, 79%; RP HPLC Rt 14.59 min; ¹H NMR (400 MHz, DMSO-

d₆): d_H11.88 (s, 1H, NH), 8.26 (d, 1H, J _6-F5¼ 7.2 Hz, H-6), 8.01 (d, 1H,

J ¼ 8.0 Hz o-H of Ar), 7.71 (t, 1H, J ¼ 8.0 Hz p-H of Ar), 7.44 (m,

J ¼ 7.8 Hz, 1H, m-H of Ar), 7.28 (d, J ¼ 8.0 Hz, 1H, m-H of Ar), 6.24 (t,

J ¼ 7.6 Hz, 1H, H-1⁰), 5.45 (br s, 1H, H-3⁰), 4.13 (br s, 1H, H-4⁰), 3.70

(m, 2H, H-5⁰, H-5⁰⁰), 2.38e2.41 (m, 2H, H-2⁰, H-2⁰⁰), 2.31 (s, 3H, CH₃)

ppm; ¹³C NMR (100 MHz, DMSO-d₆): d_C169.60 (C]O), 164.07 (C]

O),157.48 (d, J ¼ 26 Hz, C-4),150.39 (C of Ar),149.54 (C-2),140.59 (d,

J ¼ 228 Hz, C-5), 135.07, 131.95, 128.68, 126.85, 129.98, 127.89 (C of

Ar), 124.66 (d, J ¼ 35 Hz, C-6), 124.52, 123.28 (C of Ar), 85.26 (C-4⁰),

84.94 (C-1⁰), 76.33 (C-3⁰), 61.76 (C-5⁰), 37.35 (C-2⁰) ppm; HRMS

(ESI): calcd for C₁₈H₁₆N₂O₈F [M-H]^ꢀ: 407.08907, found: 407.09018.

Di-(5-ﬂuoro-2⁰-deoxyuridin-3⁰-yl)-benzene-1,3-dicarboxylate

(11).

White solid 0.47 g, yield: 76%; RP HPLC Rt 13.97 min; ¹H NMR

(400 MHz, DMSO-d₆): d_H11.87 (s, 2H, NH), 8.53 (t, J ¼ 2 Hz, 2H, H of

Ar), 8.30 (dd, J ¼ 1.8 Hz, J ¼ 7.8 Hz, 4H, H of Ar), 8.27 (d, J

3⁰-O- [1-(2,4-dichlorobenzyl)-1H-indazole-3-carbonyl]-5-ﬂuoro-2⁰-

deoxyuridine (7i). White solid; yield 0.35 g, 64%; RP HPLC Rt

20.97 min; ¹H NMR (400 MHz, DMSO-d₆): d_H11.90 (s, 1H, NH), 8.29

(d, 1H, J _6-F5¼ 7.2 Hz, H-6), 8.12 (d, 1H, J ¼ 8.0 Hz, H of Ar), 7.81 (d,

1H, J ¼ 8.8 Hz, H of Ar), 7.64 (d, 1H, J ¼ 2.0 Hz, H of Ar), 7.52 (t, 1H,

J ¼ 7.8 Hz, H of Ar), 7.39 (t, J ¼ 7.4 Hz, 1H, H of Ar), 7.35 (dd,

J ¼ 8.4 Hz, J ¼ 2.0 Hz, 1H, H of Ar), 6.92 (d, J ¼ 8.4 Hz, 1H, H of Ar),

6.30 (t, J ¼ 6.4 Hz, 1H, H-1⁰), 5.88 (s, 2H, CH₂), 5.59e5.61 (m, 1H, H-

3⁰), 4.29 (br s, 1H, H-4⁰), 3.72e3.80 (m, 2H, H-5⁰, H-5⁰⁰), 2.50e2.57

¼ 7.2 Hz, 2H, H-6), 7.78 (t, J ¼ 2 Hz, 1H, H of Ar), 6.26e6.30 (m,

2H, H-1⁰), 5.40 (t, J ¼ 5.2 Hz, 2H, H-3⁰), 4.21e4.24 (m, 2H, H-4⁰),

3.68e3.73 (m, 4H, H-5⁰, H-5⁰⁰), 2.38e2.53 (m, 4H, H-2⁰, H-2⁰⁰); ¹³C

NMR (100 MHz, DMSO-d₆): d_C164.46 (C]O),157.01 (d, J ¼ 26 Hz, C-

4), 149.07 (C-2), 140.12 (d, J ¼ 229 Hz, C-5), 134.13, 134.01, 130.02,

129.62 (C of Ar), 124.49, 124.44 (2xd, J ¼ 34 Hz, C-6), 84.78 (C-4⁰),

84.51 (C-1⁰), 76.17 (C-3⁰), 61.32 (C-5⁰), 36.93 (C-2⁰) ppm; HRMS

(ESI): calcd for

621.12904.

₂₆H₂₃N₄O₁₂F₂[MꢀH]^ꢀ: 621.12805 found:

(m, 2H, H-2⁰, H-2⁰⁰); ¹³C NMR (100 MHz, DMSO-d₆):

d 161.13 (C]O),

157.02(d, J ¼ 27 Hz, C-4), 149.10 (C-2), 140.89 (C of LND), 140.15 (d,

229.5 Hz, C-5), 134.50, 133.47, 133.19, 133.00, 130.76, 129.08, 127.76,

127.35 (C of LND), 124.52 (d, J ¼ 35 Hz, C-6), 123.68, 122.89, 121.37,

110.86 (C of LND), 84.94 (C-4⁰), 84.58 (C-1⁰), 75.59 (C-3⁰), 61.34 (C-

5⁰), 49.98 (C of LND), 37.10 (C-2⁰); HRMS (ESI): calcd for

Di-(5-ﬂuoro-2⁰-deoxyuridin-3⁰-yl)-pyridine-3,5-dicarboxylate (12).

White solid; yield 0.45 g, 72%; RP HPLC Rt 12.68 min; ¹H NMR

(400 MHz, DMSO-d₆): d_H11.87 (s, 2H, NH), 9.38 (d, J ¼ 2 Hz, 2H, o-H

of Ar), 8.72 (t, 1H, J ¼ 2 Hz, 1H, p-H of Ar), 8.28 (d, 2H, J _6-F5¼ 6.8 Hz,

H-6), 6.29e6.33 (m, 2H, H-1⁰), 5.39 (t, J ¼ 5.0 Hz, 2H, H-3⁰),

4.28e4.29 (m, 2H, H-4⁰), 3.69e3.78 (m, 4H, H-5⁰, H-5⁰⁰), 2.40e2.58

(m, 4H, H-2⁰, H-2⁰⁰); ¹³C NMR (125 MHz, DMSO-d₆): d_C163.59 (C]

O), 157.07 (d, J ¼ 26 Hz, C-4), 153.99(C of Ar), 149.12 (C-2), 140.17 (d,

J ¼ 229 Hz, C-5), 137.42, 125.70 (C of Ar), 124.57 (d, J ¼ 34 Hz, C-6),

84.69 (C-4⁰), 84.60 (C-1⁰), 76.65 (C-3⁰), 61.35 (C-5⁰), 36.92 (C-2⁰)

ppm; HRMS (ESI): calcd for C₂₅H₂₃N₅O₁₂F₂[M-H]^ꢀ: 622.12330,

found: 622.12475.

₂₄H₁₈N₄O₆Cl₂F [M-H]^þ: 547.05874, found: 547.06033.

3⁰-O-

[(R)-5-(1,2-Dithiolan-3-yl)-pentanoyl]-5-ﬂuoro-2⁰-deoxyur-

idine (7j). Yellow solid; yield 0.38 g, 88%; RP HPLC Rt 17.83 min; ¹H

NMR (400 MHz, DMSO-d₆): d_H11.87 (s, 1H, NH), 8.21 (d, 1H, J

¼ 7.2 Hz, H-6), 6.15 (t, J ¼ 7.6 Hz, 1H, H-1⁰), 5.31 (t, J ¼ 5.2 Hz,

1H, H-3⁰), 4.00 (br s, 1H, H-4⁰), 3.62e3.65 [m, 3H, H-5⁰, H-5⁰⁰, H of

lipoic acid (LA)], 3.08e3.22 (m, 2H, H of LA), 2.31e2.45 (m, 3H, H-2⁰,

H-2⁰⁰& H of LA), 2.22e2.30 (m, 2H, H of LA), 1.34e1.91 (m, 7H, H of

LA) ppm; ¹³C NMR (100 MHz, DMSO-d₆): d_C172.44 (C]O), 156.97

(d, J ¼ 26 Hz, C-4), 149.02 (C-2), 140.08 (d, J ¼ 229 Hz, C-5), 124.44

(d, J ¼ 34 Hz, C-6), 84.94 (C-4⁰), 84.40 (C-1⁰), 74.60 (C-3⁰), 61.24 (C-

5⁰), 56.00, 40.17, 38.08 (C of LA), 36.88 (C-2⁰), 34.76, 33.23, 28.00,

24.04 (C of LA) ppm; HRMS (ESI): calcd for C₁₇H₂₂N₂S₂O₆F [M-H]^ꢀ:

433.08978, found: 433.09117.

Tri-(5-ﬂuoro-2⁰-deoxyuridin-3⁰-yl)-benzene-1,3,5-tricarboxylate (14).

White solid; yield 0.61 g, 68%; RP HPLC Rt 13.52 min; ¹H NMR

(400 MHz, DMSO-d₆): d_H11.86 (s, 3H, NH), 8.72 (s, 3H, H of Ar), 8.26

(d, J _6-F5¼ 7.2 Hz, 2H, H-6), 6.29 (t, J ¼ 6.6 Hz, 3H, H-1⁰), 5.54e5.56

(m, 3H, H-3⁰), 4.24e4.29 (m, 3H, H-4⁰), 3.58e3.60 (m, 6H, H-5⁰, H-

5⁰⁰), 2.40e2.58 (m, 6H, H-2⁰, H-2⁰⁰); ¹³C NMR (100 MHz, DMSO-d₆):

d_C163.73 (C]O), 157.18 (d, J ¼ 25.75 Hz, C-4), 149.21 (C-2), 140.16

(d, J ¼ 229 Hz, C-5), 133.93, 130.94, 128.30 (C of Ar), 124.44 (d,

J ¼ 35 Hz, C-6), 84.61 (C-4⁰), 84.49 (C-1⁰), 76.70 (C-3⁰), 61.31 (C-5⁰),

36.84 (C-2⁰) ppm; HRMS (ESI): calcd for C₃₆H₃₂N₆O₁₈F₃[M-H]^ꢀ:

893.17197, found: 893.17464.

3⁰-O-dichloroacetyl-5-ﬂuoro-2⁰-deoxyuridine (7k). Yellow solid;

yield 0.26 g, 72%; RP HPLC Rt 13.88 min; ¹H NMR (400 MHz, DMSO-

d₆): d_H11.89 (s, 1H, NH), 8.20 (d, 1H, J _6-F5¼ 7.2 Hz, H-6), 6.92 (s, 1H,

CH), 6.18 (t, J ¼ 7.6 Hz, 1H, H-1⁰), 5.38e5.40 (m, 1H, H-3⁰), 4.08 (br s,

1H, H-4⁰), 3.66 (t, J ¼ 4 Hz, 2H, H-5⁰, H-5⁰⁰), 2.33e2.42 (m, 2H, H-2⁰,

H-2⁰⁰); ¹³C NMR (100 MHz, DMSO-d₆): d_C163.94 (C]O), 156.96 (d,

J ¼ 26 Hz, C-4), 149.04 (C-2), 140.12 (d, J ¼ 229 Hz, C-5), 124.39 (d,

J ¼ 34 Hz, C-6), 84.28 (C-4⁰& C-1⁰), 78.38 (C-3⁰), 66.34 (CH), 61.21

(C-5⁰), 36.39 (C-2⁰) ppm; HRMS (ESI): calcd for C₁₁H₁₀N₂Cl₂O₆F [M-

H]^ꢀ: 354.98999, found: 354.99103.

3⁰-O- [2-(2-acetylysalicylyl)-salicylyl)-5-ﬂuoro-2⁰-deoxyuridine (8).

White solid; ¹H NMR (400 MHz, DMSO-d₆): d_H11.87 (s, 1H, NH),

8.23 (dd, J ¼ 1.6 Hz, J ¼ 8 Hz, 1H, H of Ar), 8.19 (d, J _6-F5¼ 6.8 Hz, 1H,

H-6), 8.06 (dd, J ¼ 1.6 Hz, J ¼ 7.6 Hz,1H, H of Ar), 7.75e7.81 (m, 2H, H

of Ar), 7.51 (t, J ¼ 7.6 Hz, 2H, H of Ar), 7.28e7.37 (m, 2H, H of Ar), 6.11

(t, J ¼ 6.4 Hz,1H, H-1⁰), 5.37e5.38 (m, 1H, H-3⁰), 3.99 (br s, 1H, H-4⁰),

ꢀ

A. Szymanska-Michalak et al. / European Journal of Medicinal Chemistry 115 (2016) 41e52

3.60 (m, 2H, H-5⁰, H-5⁰⁰), 2.22e2.31 (m, 2H, H-2⁰, H-2⁰⁰), 2.21 (s, 3H,

CH₃) ppm; ¹³C NMR (125 MHz, DMSO-d₆): d_C169.07 (C]O), 163,44,

(C]O), 162.22 (C]O), 156.95 (d, J ¼ 26 Hz, C-4), 150.77, 149.52 (C of

Ar), 148.93 (C-2), 140.04 (d, J ¼ 230 Hz, C-5), 135.27, 134.75, 131.93,

131.54, 126.72, 126.50 (C of Ar), 124.46 (d, J ¼ 34 Hz, C-6), 122.94,

121.67 (C of Ar), 84.53 (C-4⁰), 84.35 (C-1⁰), 75.66 (C-3⁰), 61.07 (C-5⁰),

36.87 (C-2⁰), 20.64 (CH₃) ppm; HRMS (ESI): calcd for

and counted. To each well of the 96-well plate, 100 mL of the diluted

cell suspension (1 ꢁ 10⁴cells) was added. After 24 h, when a partial

monolayer was formed, 100

mL of fresh medium with different

compound concentrations (7.81, 15.625, 31.25, 62.5, 125, 250, 500

and 1000 g/mL) were added to the cells. After 48 h, the super-

natant was washed out and 100 L of MTT solution in medium (ﬁnal

concentration 0.5 mg/mL) were added to each well for 2 h. After the

incubation time was complete, unreacted dye was removed by

₂₅H₂₁N₂O₁₀FNa [MþNa]^þ: 551.1072, found: 551.1091.

aspiration. The formazan crystals were dissolved in 100

mL/well

5⁰-O-(4,4⁰-dimethoxytrityl)-3⁰-O-benzoyl-5-ﬂuoro-2⁰-deoxyuridine

(5a). ¹H NMR (400 MHz, DMSO-d₆): d_H11.93 (s, 1H, NH), 7.98e8.01

(m, 3H, H-6, o-H of Bz), 7.67e7.71 (m, 1H, p-H of Bz), 7.53e7.57 (m,

2H, m-H of Bz), 7.39e7.41 (m, 2H, H of DMTr), 7.20e7.30 (m, 7H, H of

DMTr), 6.86 (d, J ¼ 8.8 Hz, 4H, H of DMTr), 6.26 (t, J ¼ 6.8 Hz, 1H, H-

1⁰), 5.51 (t, J ¼ 3.2 Hz, 1H, H-3⁰), 4.26e4.27 (m, 1H, H-4⁰), 3.72 (s, 6H,

OCH₃), 3.28e3.48 (m, H-5⁰, H-5⁰⁰), 2.53e2.64 (m, 2H, H-2⁰, H-2⁰⁰)

ppm; ¹³C NMR (100 MHz, DMSO-d₆): d_C165.58 (C]O), 158.56 (C of

DMTr), 157.40 (d, J ¼ 26 Hz, C-4), 149.40 (C-2), 145.05 (C of DMTr),

140.55 (d, J ¼ 230 Hz, C-5), 135.72, 135.61 (C of DMTr), 134.04 (C of

Bz), 130.11 (C of DMTr), 130.09, 129.63 (C of Bz), 128.39, 128.01,

127.19 (C of DMTr), 125.01 (d, J ¼ 34 Hz, C-6), 113.64 (C of DMTr),

86.46 (C-4⁰), 85.03 (C-1⁰), 83.22 (C of DMTr), 75.13 (C-3⁰), 66.76

(OCH₃), 64.03 (C-5⁰), 36.86 (C-2⁰) ppm.

DMSO and measured spectrophotometrically in a multi-well Syn-

ergy2 plate reader (BioTek Instruments, USA) at a test wavelength

of 492 nm and a reference wavelength of 690 nm. The results were

calculated as an IC₅₀(inhibitory concentration 50) e the IC₅₀cor-

responds to the concentration of tested compound that inhibits cell

viability/proliferation by 50%. Results are presented as mean of at

least three independent experiments.

4.2.3. In silico pharmacokinetic prediction

Calculations of pharmacokinetic proﬁle descriptors of the syn-

thesized compounds were performed by various software solutions

accessible on-line. The transformation of the stoichiometric for-

mulas of the compounds into a SMILES code (Simpliﬁed Molecular

Input Line Entry System) was carried out by ChemBioDraw Ultra

version 12.0 program (Cambridge Software). The SMILES code was

applied to calculate logP values with ALOGPS 2.1 software [86]

(mean value obtained from ALOGPs, AC_logP, miLogP, ALOGP,

MLOGP, LogKOWWIN, XLOGP2, XLOGP3 methods), PSA (topological

polar surface area) and aPSA (apolar surface area) values. PSA and

aPSA descriptors were calculated using the VEGA ZZ program [87].

The pharmacokinetic proﬁle was evaluated according to the Lip-

inski “rule of ﬁve” [48] by using Molinspiration application (http://

www.molinspiration.com), which includes also analyses of mo-

lecular weight (MW), number of hydrogen-bond acceptors (HBA)

and number of hydrogen-bond donors (HBD). The Caco-2 predic-

tion model based on descriptors generated by preADMET (http://

preadmet.bmdrc.org) was used to compute Caco-2 apparent

permeability (tP_app). In this model a number of hydrogen bond

donors and three molecular surface area properties determine

membrane permeability of compounds.

5⁰-O-(4,4⁰-dimethoxytrityl)-3⁰-O-benzoyl-5-ﬂuoro-3-N-benzoyl-2⁰-

deoxyuridine (6a). ¹H NMR (400 MHz, DMSO-d₆): d_H8.24 (d, J

¼ 6.8 Hz, 1H, H-6), 8.12 (d, J ¼ 7.6 Hz, 2H, o-H of N-Bz), 7.99 (d,

J ¼ 7.2 Hz, 2H, o-H of O-Bz), 7.81 (t, J ¼ 7.3 Hz, 1H, p-H of N-Bz), 7.68

(t, J ¼ 7.3 Hz, 1H, p-H of O-Bz), 7.61 (t, J ¼ 7.1 Hz, 2H, m-H of N-Bz),

7.54 (t, J ¼ 7.1 Hz, 2H, m-H of O-Bz), 7.41e7.43 (m, 2H, H of DMTr),

7.21e7.30 (m, 7H, H of DMTr), 6.88 (d, J ¼ 7.92 Hz, 4H, H of DMTr),

6.26e6.27 (m, 1H, H-1⁰), 5.54 (br s, 1H, H-3⁰), 4.3 (br s, 1H, H-4⁰),

3.72 (s, 6H, OCH₃), 3.47e3.61 (m, H-5⁰, H-5⁰⁰), 2.60e2.76 (m, 2H, H-

2⁰, H-2⁰⁰) ppm; ¹³C NMR (100 MHz, DMSO-d₆): d_C168.49 (C of Bz),

165.57 (C]O), 158.57 (C of DMTr), 156.28 (d, J ¼ 26 Hz, C-4), 147.99

(C-2), 145.03 (C of DMTr), 140.22 (d, J ¼ 230 Hz, C-5), 136.33 (C of

Bz), 135.71, 135.59 (C of DMTr), 134.06 (C of Bz), 131.16, 131.08 (C of

Bz), 130.12 (C of DMTr), 129.93, 129.93, 129.60, 129.17 (C of Bz),

128.32, 128.03, 127.21 (C of DMTr), 126.14 (d, J ¼ 34 Hz, C-6), 113.67

(C of DMTr), 86.55 (C-4⁰), 85.67 (C-1⁰), 83.43 (C of DMTr), 74.71 (C-

3⁰), 66.76 (OCH₃), 63.84 (C-5⁰), 37.06 (C-2⁰) ppm.

Acknowledgements

4.2. Biological assays

Financial support from the National Science Centre of Poland,

Project No. 2011/01/B/NZ4/04936 is greatly acknowledged.

4.2.1. Cell line and culture conditions

GBM cell lines (T98G, U-118 MG, U-87 MG), HeLa (cervical

cancer cell line), T-47D (breast cancer cell line), Caco-2 (colon

cancer cell line), and non-cancerous lung ﬁbroblast cell line (MRC-

5) were purchased from ATCC (Manassas, USA). All cell lines were

from human origin. HeLa and T-47D were cultured in RPMI 1640

medium. Caco-2 and U-118 MG were cultured in DMEM medium.

T98G, U-87 MG and MRC-5 were cultured in EMEM medium. Each

medium was supplemented with 10% fetal bovine serum (FBS) and

10 mg/mL antibiotics (penicillin and streptomycin). Cells were

cultured at 37 ^ꢂC with 5% CO₂in humidiﬁed air. Cell media (RPMI

1640, DMEM, EMEM), Human Blood Plasma (HBP) were obtained

from SigmaeAldrich Chemie GmbH (Steinheim, Germany) and

ATCC. Cell concentrations in culture were adjusted to allow for

exponential growth.

Appendix A. Supplementary data

Supplementary data associated with this article can be found in

the online version, at http://dx.doi.org/10.1016/j.ejmech.2016.03.

010. These data include MOL ﬁles and InChiKeys of the most

important compounds described in this article.

References

[1] C. Heidelberger, L. Griesbach, O. Cruz, R.J. Schnitzer, E. Grunberg, Fluorinated

pyrimidines VI. Effects of 5-ﬂ uorouridine and 5-ﬂ uoro-2⁰-deoxyuridine on

transplanted tumors, Proc. Soc. Exp. Biol. 97 (1958) 470e475.

[2] W.B. Parker, Enzymology of purine and pyrmidine antimetabolites used in the

treatment of cancer, Chem. Rev. 109 (2009) 2880e2893.

[3] L. Bosch, E. Harbers, C. Heidelberger, Studies on ﬂ uorinated pyrimidines .5.

Effects on nucleic acid metabolism invitro, Cancer Res. 18 (1958) 335e343.

[4] N.K. Chaudhuri, K.L. Mukherjee, C. Heidelberger, Studies on ﬂ uorinated py-

4.2.2. Cell viability/proliferation assay

rimidines .7. The degradative pathway, Biochem. Pharmacol.

1 (1958)

328e341.

Cell viability/proliferation was evaluated by a dye staining

[5] W.D. Ensminger, A. Rosowsky, V. Raso, D.C. Levin, M. Glode, S. Come, G. Steele,

E. Frei, Clinical-pharmacological evaluation of hepatic arterial infusions of 5-

ﬂ uoro-2⁰-deoxyuridine and 5-ﬂ uorouracil, Cancer Res. 38 (1978) 3784e3792.

[6] D.G. Power, N.E. Kemeny, The role of ﬂ oxuridine in metastatic liver disease,

method

using

3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-

tetrazolium bromide (MTT). The protocol was adapted from the

literature methods [85]. The monolayer cell culture was trypsinized

ꢀ

A. Szymanska-Michalak et al. / European Journal of Medicinal Chemistry 115 (2016) 41e52

Mol. Cancer Ther. 8 (2009) 1015e1025.

[33] H. Schaller, G. Weimann, B. Lerch, H.G. Khorana, Studies on polynucleotides.

XXIV. The stepwise synthesis of speci ﬁ c deoxyribopolynucleotides (4). Pro-

tected derivatives of deoxyribonucleotides and new synthesis of deoxy-

nucleoside-3⁰phosphates, J. Am. Chem. Soc. 85 (1963) 3821e3828.

[34] R. Lohrmann, H.G. Khorana, Studies on polynucleotides .34. Speci ﬁ c synthesis

of C3]-C5]-linked ribooligonucleotides. New protected derivatives of

ribonucleosides þ ribonucleoside 3]-phosphates. Further syntheses of diri-

bonucleoside phosphates, J. Am. Chem. Soc. 86 (1964) 4188e4194.

[35] T. Kamimura, T. Masegi, M. Sekine, T. Hata, Structural assignment of N-3-

acylated uridine derivatives by means of C-13 Nmr-spectroscopy, Tetrahe-

dron Lett. 25 (1984) 4241e4244.

[7] J.K. Martin, M.J. Oconnell, H.S. Wieand, R.J. Fitzgibbons, J.A. Mailliard, J. Rubin,

D.M. Nagorney, L.K. Tschetter, J.E. Krook, Intraarterial ﬂ oxuridine vs systemic

ﬂ uorouracil for hepatic metastases from colorectal-cancer e a randomized

trial, AMA Arch. Surg. 125 (1990) 1022e1027.

[8] S. Fujii, United States Patent. 5-Fluorouracil derivatives, United States Patent,

no. 4,983,609 (1991).

[9] C.P. Landowski, B.S. Vig, X.Q. Song, G.L. Amidon, Targeted delivery to PEPT1-

overexpressing cells: acidic, basic, and secondary ﬂ oxuridine amino acid

ester prodrugs, Mol. Cancer Ther. 4 (2005) 659e667.

[10] Y. Tsume, B.S. Vig, J. Sun, C.P. Landowski, J.M. Hil ﬁ nger, C. Ramachandran,

G.L. Amidon, Enhanced absorption and growth inhibition with amino acid

monoester prodrugs of ﬂ oxuridine by targeting hPEPT1 transporters, Mole-

cules 13 (2008) 1441e1454.

[36] M. Sekine, General-method for the preparation of N-3-substituted and O-4-

substituted uridine derivatives by phase-transfer reactions, J. Org. Chem. 54

(1989) 2321e2326.

[11] T. Kawaguchi, S. Fukushima, Y. Hayashi, M. Nakano, Nonenzymatic and

enzymatic-hydrolysis of 5-Fluoro-2'-Deoxyuridine (Fudr) esters, Pharm. Res. 5

(1988) 741e744.

[12] T. Kawaguchi, M. Saito, Y. Suzuki, N. Nambu, T. Nagai, Speci ﬁ city of esterases

and structure of prodrug esters .2. Hydrolytic regeneration behavior of 5-

ﬂ uoro-2⁰-deoxyuridine (Fudr) from 3',5'-diesters of Fudr with rat-tissue ho-

mogenates and plasma in relation to their antitumor-activity, Chem. Pharm.

Bull. 33 (1985) 1652e1659.

[37] Morpholine was used previously for selective removal of one of the benzoyl

protecting groups from 6-N,Nedibenzoyl-2⁰-deoxyadenosine and 4-N,N-

dibenzoyl-2⁰-deoxycytidine [38].

[38] R. Kierzek, The synthesis of 5⁰-O-dimethoxytrityl-N-acyl-2'-deoxynucleosides

improved transient protection approach, Nucleos. Nucleot.

4 (1985)

641e649.

[39] This hypothesis was veriﬁed negatively in experiment in which 3'-O-salicy-

lated 5h dissolved in neat pyridine was allowed to react with 1 molar equiv. of

2h. After 12 h the reaction mixture was treated with morpholine and then the

DMT protecting group was removed with 80% acetic acid. Inspection of the

reaction mixture with TLC analysis did not show the presence of compound 8

[40] This scenario was checked by keepieng O-acetylsalicyl chloride 2h in neat

pyridine-d5 and recording periodically the ¹H NMR spectra of the reaction

mixture. After 10 min in the region of chemical shifts d_H2.3 e 2.5 ppm typical

for resonance frequency of the methyl group of the acetyl moiety in 2h,

several additional signals appeared, and their number and intensities were

increasing in time. After 3 h intensity of the signal of the methyl group of 2h

became comparable to those of the new signals (data not shown). This

seemed to indicate that O-acetylsalicyl chloride 2h when left in pyridine

underwent deacetylation and produced dimeric (and most likely also poly-

meric) acylating species, which in the reaction with nucleoside 1 could pro-

[13] C.P. Landowski, X.Q. Song, P.L. Lorenzi, J.M. Hil ﬁ nger, G.L. Amidon, Floxuridine

amino acid ester prodrugs: enhancing Caco-2 permeability and resistance to

glycosidic bond metabolism, Pharm. Res. 22 (2005) 1510e1518.

[14] Y. Nishizaw, J.E. Casida, S.W. Anderson, C. Heidelbe, 3⁰,5⁰-diesters of 5-ﬂ uoro-

2⁰-deoxyuridine e synthesis and biological activity, Biochem. Pharmacol. 14

(1965) 1605e1619.

[15] J.E. Casida, J.L. Engel, Y. Nishizaw, 3⁰,5⁰-diesters of 5-ﬂ uoro-2⁰-deoxyuridine

and thymidine e hydrolysis by esterases in human mouse and insect tissue,

Biochem. Pharmacol. 15 (1966) 627e644.

[16] F. Kanzawa, A. Hoshi, K. Kuretani, M. Saneyoshi, T. Kawaguchi, Anti-tumor

activity of 3⁰,5⁰-diesters of 5-ﬂ uoro-2⁰-deoxyuridine against Murine

Leukemia-L1210 cells, Cancer Chemoth. Pharm. 6 (1981) 19e23.

[17] D. Vivian, J.E. Polli, Synthesis and in vitro evaluation of bile acid prodrugs of

ﬂ oxuridine to target the liver, Int. J. Pharm. 475 (2014) 597e604.

[18] Z. Zhang, H. Hatta, K. Tanabe, S. Nishimoto, A new class of 5-ﬂ uoro-2⁰-deox-

yuridine prodrugs conjugated with a tumor-homing cyclic peptide CNGRC by

ester linkers: synthesis, reactivity, and tumor-cell-selective cytotoxicity,

Pharm. Res. 22 (2005) 381e389.

duce O-salicyl-salicylated 5-ﬂuoro-2⁰-deoxyuridine

8 as a predominant

product.

[41] D.D. Perrin, Dissociation Constants of Organic Bases in Aqueous Solution,

Butterworth, London, 1965.

[42] T. Fujii, H. Nishida, Y. Abiru, M. Yamamoto, M. Kise, Studies on synthesis of the

antibacterial agent Nm441 .2. Selection of A Suitable base for alkylation of 1-

substituted piperazine with 4-(Bromomethyl)-5-methyl-1,3-dioxol-2-one,

Chem. Pharm. Bull. 43 (1995) 1872e1877.

[19] K.L. Vine, J.M. Locke, J.B. Bremner, S.G. Pyne, M. Ranson, Selective targeting of

2⁰-deoxy-5-ﬂ uorouridine to urokinase positive malignant cells in vitro, Bio-

org. Med. Chem. Lett. 20 (2010) 2908e2911.

[20] Z. Xia, E.E. Knaus, L.I. Wiebe, Dual-acting 5-ﬂ uorodeoxyuridine (FUdR) pro-

[43] M. Smith, J.G. Moffatt, H.G. Khorana, Carbodiimides .8. Observations on the

reactions of carbodiimides with acids and some new applications in the

synthesis of phosphoric acid esters, J. Am. Chem. Soc. 80 (1958) 6204e6212.

[44] J. Izdebski, A. Orlowska, R. Anulewicz, E. Witkowska, D. Fiertek, Reinvesti-

gation of the reactions of carbodiimides with alkoxycarbonylamino acid

symmetrical anhydrides e isolation of 2 N-Acylureas, Int. J. Pept. Prot. Res. 43

(1994) 184e189.

drugs: cell death induced by 3⁰-O-retinoyl-5-ﬂ uoro-2⁰-deoxyuridine and 5⁰-O-

[bis(2,2,2,-trichloroethyl)

phosphoryl-3⁰-O-butanoyl]-5-ﬂ uoro-2⁰-deoxyur-

idine, Clin. Exp. Pharmacol. 3 (2013) 143.

[21] Z. Xia, E.E. Knaus, L.I. Wiebe, Pharmacokinetics of 3⁰-O-retinoyl-5-ﬂ uoro-2⁰-

deoxyuridine (RTUdR), a dual acting mutually masking prodrug, and its me-

tabolites in tumor bearing mice, Curr. Drug Deliv. 10 (2013) 557e563.

[22] D.Y. Shi, H.L. Liu, J.S. Stern, P.Z. Yu, S.L. Liu, Alpha-lipoic acid induces apoptosis

in hepatoma cells via the PTEN/Akt pathway, FEBS Lett. 582 (2008)

1667e1671.

[45] J.

propylcarbodiimide e a hybrid that combines the structural features of Dcc

and Dic, Int. J. Pept. Prot. Res. 44 (1994) 414e419.

[23] E. Dozio, M. Ruscica, L. Passafaro, G. Dogliotti, L. Steffani, A. Pagani,

G. Demartini, D. Esposti, F. Fraschini, P. Magni, The natural antioxidant alpha-

lipoic acid induces p27(Kip1)-dependent cell cycle arrest and apoptosis in

MCF-7 human breast cancer cells, Eur. J. Pharmacol. 641 (2010) 29e34.

[24] A. Goraca, H. Huk-Kolega, A. Piechota, P. Kleniewska, E. Ciejka, B. Skibska,

Lipoic acid e biological activity and therapeutic potential, Pharmacol. Rep. 63

(2011) 849e858.

[25] E. Cuesta, J. Boada, J.C. Perales, T. Roig, J. Bermudez, Aspirin inhibits NF-kappa

B activation in a glycolysis-depleted lung epithelial cell line, Eur. J. Pharmacol.

517 (2005) 158e164.

[46] G. Giorgioni, S. Ruggieri, A. Di Stefano, P. Sozio, B. Cinque, L. Di Marzio,

G. Santoni, F. Claudi, Glycosyl and polyalcoholic prodrugs of lonidamine,

Bioorg. Med. Chem. Lett. 18 (2008) 2445e2450.

[47] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Experimental and

computational approaches to estimate solubility and permeability in drug

discovery and development settings, Adv. Drug Deliv. Rev. 23 (1997) 3e25.

[48] C.A. Lipinski, Drug-like properties and the causes of poor solubility and poor

permeability, J. Pharmacol. Toxicol. Method 44 (2000) 235e249.

[49] C.A. Lipinski, Lead- and drug-like compounds: the rule-of-ﬁ ve revolution,

Drug Discov. Today 1 (2004) 337e341.

[26] G.A. Spitz, C.M. Furtado, M. Sola-Penna, P. Zancan, Acetylsalicylic acid and

salicylic acid decrease tumor cell viability and glucose metabolism modu-

lating 6-phosphofructo-1-kinase structure and activity, Biochem. Pharmacol.

77 (2009) 46e53.

[27] K. Schror, Pharmacology and cellular/molecular mechanisms of action of

aspirin and Non-aspirin NSAIDs in colorectal cancer, Best. Pract. Res. Cl. Ga. 25

(2011) 473e484.

[28] X. Garcia-Albeniz, A.T. Chan, Aspirin for the prevention of colorectal cancer,

Best. Pract. Res. Cl. Ga. 25 (2011) 461e472.

[29] X.F. Li, Y.S. Zhu, H.B. He, L.Q. Lou, W.W. Ye, Y.X. Chen, J.H. Wang, Synergisti-

cally killing activity of aspirin and histone deacetylase inhibitor valproic acid

(VPA) on hepatocellular cancer cells, Biochem. Biophys. Res. Co. 436 (2013)

259e264.

[50] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Experimental and

computational approaches to estimate solubility and permeability in drug

discovery and development settings, Adv. Drug Deliv. Rev. 46 (2001) 3e26.

[51] M. Congreve, R. Carr, C. Murray, H. Jhoti, A rule of three for fragment-based

lead discovery? Drug Discov. Today 8 (2003) 876e877.

[52] A. Rayan, D. Marcus, A. Goldblum, Predicting oral druglikeness by iterative

stochastic elimination, J. Chem. Inf. Model 50 (2010) 437e445.

[53] D.F. Veber, S.R. Johnson, H.Y. Cheng, B.R. Smith, K.W. Ward, K.D. Kopple,

Molecular properties that in ﬂ uence the oral bioavailability of drug candidates,

J. Med. Chem. 45 (2002) 2615e2623.

[54] T. Luker, L. Alcaraz, K.K. Chohan, N. Blomberg, D.S. Brown, R.J. Butlin,

T. Elebring, A.M. Grif ﬁ n, S. Guile, S. St-Gallay, B.M. Swahn, S. Swallow,

M.J. Waring, M.C. Wenlock, P.D. Leeson, Strategies to improve in vivo toxi-

cology outcomes for basic candidate drug molecules, Bioorg. Med. Chem. Lett.

21 (2011) 5673e5679.

[30] H. Pelicano, D.S. Martin, R.H. Xu, P. Huang, Glycolysis inhibition for anticancer

treatment, Oncogene 25 (2006) 4633e4646.

[31] M.T. Gatto, B. Tita, M. Artico, L. Saso, Recent studies on lonidamine, the lead

compound of the antispermatogenic indazol-carboxylic acids, Contraception

65 (2002) 277e278.

[32] M. Smith, H.G. Khorana, D.H. Rammler, I.H. Goldberg, Studies on poly-

nucleotides .14. Speci ﬁ c synthesis of C3⁰-C5⁰interrribonucleotide linkage e

syntheses of uridylyl-(3⁰-]5⁰)-uridine and uridylyl-(3⁰-]5⁰)-adenosine, J. Am.

Chem. Soc. 84 (1962) 430e439.

[55] Y. Takaoka, Y. Endo, S. Yamanobe, H. Kakinuma, T. Okubo, Y. Shimazaki, T. Ota,

S. Sumiya, K. Yoshikawa, Development of a method for evaluating drug-

likeness and ease of synthesis using a data set in which compounds are

assigned scores based on chemists' intuition, J. Chem. Inf. Comp. Sci. 43 (2003)

1269e1275.

[56] J.R. Proudfoot, The evolution of synthetic oral drug properties, Bioorg. Med.

Chem. Lett. 15 (2005) 1087e1090.

ꢀ

A. Szymanska-Michalak et al. / European Journal of Medicinal Chemistry 115 (2016) 41e52

[57] T.H. Keller, A. Pichota, Z. Yin, A practical view of 'druggability', Curr. Opin.

Chem. Biol. 10 (2006) 357e361.

[58] C. Abad-Zapatero, A Sorcerer's apprentice and the rule of ﬁ ve: from rule-of-

thumb to commandment and beyond, Drug Discov. Today 12 (2007)

995e997.

Trends 6 (2012) 192e200.

[74] K. Rolle, S. Nowak, E. Wyszko, M. Nowak, R. Zukiel, R. Piestrzeniewicz,

I. Gawronska, M.Z. Barciszewska, J. Barciszewski, Promising human brain tu-

mors therapy with interference RNA intervention (iRNAi), Cancer Biol. Ther. 9

(2010) 396e406.

[59] M.Q. Zhang, B. Wilkinson, Drug discovery beyond the 'rule-of-ﬁ ve', Curr. Opin.

Biotech. 18 (2007) 478e488.

[75] M. Piwecka, K. Rolle, E. Wyszko, R. Zukiel, S. Nowak, M.Z. Barciszewska,

J. Barciszewski, Nucleic acid-based technologies in therapy of malignant gli-

omas, Curr. Pharm. Biotechnol. 12 (2011) 1805e1822.

[76] D.A. Cameron, H. Gabra, R.C.F. Leonard, Continuous 5-ﬂ uorouracil in the

treatment of breast-cancer, Brit. J. Cancer. 70 (1994) 120e124.

[77] W. Zoli, P. Ulivi, A. Tesei, F. Fabbri, M. Rosetti, R. Maltoni, D.C. Giunchi,

L. Ricotti, G. Brigliadori, I. Vannini, D. Amadori, Addition of 5-ﬂ uorouracil to

doxorubicin-paclitaxel sequence increases caspase-dependent apoptosis in

breast cancer cell lines, Breast Cancer Res. 7 (2005) R681eR689.

[78] I.J. Hidalgo, T.J. Raub, R.T. Borchardt, Characterization of the human-colon

[60] P.D. Leeson, B. Springthorpe, The in ﬂ uence of drug-like concepts on decision-

making in medicinal chemistry, Nat. Rev. Drug Discov. 6 (2007) 881e890.

[61] J.D. Hughes, J. Blagg, D.A. Price, S. Bailey, G.A. DeCrescenzo, R.V. Devraj,

E. Ellsworth, Y.M. Fobian, M.E. Gibbs, R.W. Gilles, N. Greene, E. Huang,

T. Krieger-Burke, J. Loesel, T. Wager, L. Whiteley, Y. Zhang, Physiochemical

drug properties associated with in vivo toxicological outcomes, Bioorg. Med.

Chem. Lett. 18 (2008) 4872e4875.

[62] M.J. Waring, De ﬁ ning optimum lipophilicity and molecular weight ranges for

drug candidates-Molecular weight dependent lower logD limits based on

permeability, Bioorg. Med. Chem. Lett. 19 (2009) 2844e2851.

[63] K. Ohno, Y. Nagahara, K. Tsunoyama, M. Orita, Are there differences between

launched drugs, clinical candidates, and commercially available compounds?

J. Chem. Inf. Model 50 (2010) 815e821.

carcinoma cell-line (Caco-2) as

a model system for intestinal epithelial

permeability, Gastroenterology 96 (1989) 736e749.

[79] P. Artursson, J. Karlsson, Correlation between oral-drug absorption in humans

and apparent drug permeability coef ﬁ cients in human intestinal epithelial

(Caco-2) cells, Biochem. Biophys. Res. Co. 175 (1991) 880e885.

[64] G.R. Bickerton, G.V. Paolini, J. Besnard, S. Muresan, A.L. Hopkins, Quantifying

the chemical beauty of drugs, Nat. Chem. 4 (2012) 90e98.

[65] T.J. Ritchie, S.J.F. Macdonald, The impact of aromatic ring count on compound

developability e are too many aromatic rings a liability in drug design? Drug

Discov. Today 14 (2009) 1011e1020.

[80] S. Yamashita, T. Furubayashi, M. Kataoka, T. Sakane, H. Sezaki, H. Tokuda,

Optimized conditions for prediction of intestinal drug permeability using

Caco-2 cells, Eur. J. Pharm. Sci. 10 (2000) 195e204.

[81] G. Camenisch, J. Alsenz, H. van de Waterbeemd, G. Folkers, Estimation of

permeability by passive diffusion through Caco-2 cell monolayers using the

[66] D.E. Clark, S.D. Pickett, Computational methods for the prediction of 'drug-

likeness', Drug Discov. Today 5 (2000) 49e58.

drugs' lipophilicity and molecular weight, Eur. J. Pharm. Sci. 6 (1998)

313e319.

[67] G. Kortum, W. Vogel, K. Andrussow, Dissociation Constants of Organic Acids in

Aqueous Solution, Butterworths Scienti ﬁ c Publications, Ltd., London, 1961.

[68] Y. Ishihama, M. Nakamura, T. Miwa, T. Kajima, N. Asakawa, A rapid method for

pK(a) determination of drugs using pressure-assisted capillary electrophoresis

with photodiode array detection in drug discovery, J. Pharm. Sci. 91 (2002)

933e942.

[82] M. Yazdanian, S.L. Glynn, J.L. Wright, A. Hawi, Correlating partitioning and

Caco-2 cell permeability of structurally diverse small molecular weight

compounds, Pharm. Res. 15 (1998) 1490e1494.

[83] I.G. Georg, J.S. Tash, R. Chakrasali, S.R. Jakkaraj, and K. Roby, Lonidamine an-

alogues for fertility management, US. Patent Application, US 2011/0060003

A1 (2011) 1e44.

[69] S. Peyrottes, D. Ergon, I. Lefebvre, G. Gosselin, J.-L. Imbach, C. Perigaud, SATE

pronucleotide approaches: an overview, Mini Rev. Med. Chem. 4 (2004)

395e408.

[84] R. Martin, T.-L. Wang, B.T. Flatt, X.-H. Gu, and R. Grifﬁth, Azepinoindole and

pyridoindole derivatrives as pharmaceutical agents., International Patent, WO

03/099821 A1 (2003) 1e268.

[70] Y. Lei, Y.S. Lai, Y.L. Li, S. Li, G. Cheng, D. Li, H.P. Li, B. He, Z.W. Gu, Anticancer

drug delivery of PEG based micelles with small lipophilic moieties, Int. J.

Pharm. 453 (2013) 579e586.

[85] C.P.R. Xavier, C.F. Lima, M. Rohde, C. Pereira-Wilson, Quercetin enhances 5-

ﬂ uorouracil-induced apoptosis in MSI colorectal cancer cells through p53

modulation, Cancer Chemoth. Pharm. 68 (2011) 1449e1457.

[71] M. Mendez-Perez, B. Vaz, L. Garcia-Rio, M. Perez-Lorenzo, Polymeric pre-

micelles as ef ﬁ cient lipophilic nanocarriers: extending drug uptake to the

submicellar regime, Langmuir 29 (2013) 11251e11259.

[72] S.A. Azizi, C. Miyamoto, Principles of treatment of malignant gliomas in adults:

an overview, J. Neurovirol. 4 (1998) 204e216.

[73] L.L. Wang, H. Zhao, K. Cui, L.L. Yao, M. Ren, A.J. Hao, P. Smollen, F. Nie, G.X. Jin,

Q. Liu, S.T.C. Wong, Identi ﬁ cation of novel small-molecule inhibitors of glio-

blastoma cell growth and invasion by high-throughput screening, Biosci.

[86] I.V. Tetko, J. Gasteiger, R. Todeschini, A. Mauri, D. Livingstone, P. Ertl,

V. Palyulin, E. Radchenko, N.S. Ze ﬁ rov, A.S. Makarenko, V.Y. Tanchuk,

V.V. Prokopenko, Virtual computational chemistry laboratory e design and

description, J. Comput. Aided Mol. Des. 19 (2005) 453e463.

[87] A. Pedretti, L. Villa, G. Vistoli, VEGA: a versatile program to convert, handle

and visualize molecular structure on Windows-based PCs, J. Mol. Graph.

Model. 21 (2002) 47e49.

Products guided by the article

Product name:(2R,3S,5R)-5-(5-fluoro-2,4-dioxo-3,4-dihydropyrimidin-1(2H)-yl)-2-(hydroxymethyl)tetrahydrofuran-3-yl benzoate

Cas No:1995-83-1

R&D Labs maybe for 1995-83-1

Chengdu Cogon Bio-tech Co., Ltd.

Contact:86-28-85171192

Address:NO.52.YongFeng Rd. Chengdu,610041,P.R.China.
Hangzhou Maytime Bio-Tech Co.,Ltd.

website:http://www.maytime.com.cn

Contact:+86-571-88925295 88920965

Address:NO.2-1701 Ganghui Central Ningwei Street, Xiaoshan Hangzhou Zhejiang China
Guanmat Optoelectronic Materials (Jiangxi) Inc.

Contact:0799-3609933

Address:Chishan Industrail Park, Pingxiang, Jiangxi PR China
JIEJIE GROUP CO.,LTD.

Contact:+86 021-51698675

Address:1701, Jielong Plaza, No.618 Pingliang Rd, ShangHai,China
Zouping Fuhai Technology Development Co., Ltd.

Contact:+0532-86934525 18660293207

Address:jiuhu town industrial park Zouping County,bingzhou Provincejiuhu town industrial park

Relevant to this article

Spray coatable electrochromic dioxythiophene polymers with high coloration efficiencies

Doi:10.1021/ma049222y
(2004)
Synthesis and structure activity relationships of selected isomeric oxime O ethers as anticholinergic agents

Doi:10.1002/jps.2600661125
(1977)
Bredereck's reagent revisited: Latent anionic ring-opening polymerization and transesterification reactions

Doi:10.1002/adsc.200404097
(2004)
STUCTURES OF FLAZIN AND YS, HIGHLY FLUORESCENT COMPOUNDS ISOLATED FROM JAPANESE SOY SAUCE

Doi:10.1016/S0040-4039(00)84806-6
(1986)
Hindered organoboron groups in organic chemistry. 28 The solvolyses of bis(2,6-dimethyl-4-methoxyphenyl)organylboranes, (DMP)₂BR

Doi:10.1016/S0040-4020(01)85690-8
(1994)
?-CYCLOHEXADIENONE DERIVATIVES OF RHODIUM(I) ACETYLACETONATE. CRYSTAL AND MOLECULAR STRUCTURE OF RHODIUM (η⁴-4-METHYL-4-exo-TRICHLOROMETHYL-2,5-CYCLOHEXADIENE-1-ONE)(PENTANE-O,O'-2,4-DIONATE)

Doi:10.1016/S0022-328X(00)98960-5
(1986)

Article Doi

New 3′-O-aromatic acyl-5-fluoro-2′-deoxyuridine derivatives as potential anticancer agents

DOI: 10.1016/j.ejmech.2016.03.010

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.ejmech.2016.03.010

Full text of DOI:10.1016/j.ejmech.2016.03.010

Products guided by the article

R&D Labs maybe for 1995-83-1

Relevant to this article

Hot Product