Synthesis of 5-fluorouridine nucleolipid derivatives and their cytostatic/cytotoxic activities on human HT-29 colon carcinoma cells

Article abstract of DOI:10.1002/cbdv.201300219

One of the major drawbacks of chemotherapeutics is their insufficient penetration through cell membranes due to a high hydrophobicity. Thus, we have synthesized a series of selected nucleolipid derivatives of 5-fluorouridine (5-FUrd; 2a), carrying lipophilic moieties at N(3) and/or in the 2′,3′-O-position (i.e., 3a-7a and 3c), and tested their cytostatic/cytotoxic activities using HT-29 human colon carcinoma cells, in comparison with, e.g., 5-FU (1) and 5-FUrd (2a). Incorporation and intracellular localization of the substances under test were performed after conjugation with the fluorochrome Atto 425. We showed that all 5′-O-labelled Atto 425 derivatives were incorporated by the human HT-29 cells and accumulated in their cytoplasm. Moreover, after 24-h treatment of HT-29 human colon carcinoma cells, 1 or 2a (10, 20, 40, or 80 μM) revealed a significant (14-23 or 33-45%, resp.) decrease of the viability in comparison with the (negative) control. Interestingly, derivatives 3a and 3c (40 and 80 μM) led to a significant (77-95 or 89-96%, resp.) inhibition of survival of human HT29 cells, i.e., these two substances were ca. 63-72% or ca. 75%, respectively more effective than 5-FU (1; positive control). Furthermore, derivative 5a showed a significant, i.e., 30 and 86%, inhibition of the survival at 40 and 80 μM, respectively in comparison with the (negative) control. Some synthesized 5-FUrd derivatives turned out to be more effective than 5-FU (1) or 5-FUrd (2a). Copyright

Full text of DOI:10.1002/cbdv.201300219

CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
2235
Synthesis of 5-Fluorouridine Nucleolipid Derivatives and Their Cytostatic/
Cytotoxic Activities on Human HT-29 Colon Carcinoma Cells
by Edith Malecki1)a), Anisa Farhat1)b), Gabriel A. Bonaterra*b), Doris Rçthleinc), Martin Wolfc),
Jꢀrgen Schmittc), Ralf Kinscherfb), and Helmut Rosemeyer*a)
a) Organic Materials Chemistry and Bioorganic Chemistry, Institute for Chemistry of New Materials,
Department of Biology and Chemistry, University of Osnabrꢀck, Barbarastr. 7, D-49069 Osnabrꢀck
b) Anatomy and Cell Biology, Department of Medical Cell Biology, University of Marburg, Robert-
Koch-Str. 8, D-35032 Marburg
c) B. Braun Melsungen AG, Hospital Care Division, Am Schwerzelshof, D-34212 Melsungen
Dedicated to Prof. Dr. Piet Herdewijn, Katholieke Universiteit Leuven, in admiration of his outstanding
contributions to medicinal chemistry
One of the major drawbacks of chemotherapeutics is their insufficient penetration through cell
membranes due to a high hydrophobicity. Thus, we have synthesized a series of selected nucleolipid
derivatives of 5-fluorouridine (5-FUrd; 2a), carrying lipophilic moieties at N(3) and/or in the 2’,3’-O-
position (i.e., 3a–7a and 3c), and tested their cytostatic/cytotoxic activities using HT-29 human colon
carcinoma cells, in comparison with, e.g., 5-FU (1) and 5-FUrd (2a). Incorporation and intracellular
localization of the substances under test were performed after conjugation with the fluorochrome Atto
425. We showed that all 5’-O-labelled Atto 425 derivatives were incorporated by the human HT-29 cells
and accumulated in their cytoplasm. Moreover, after 24-h treatment of HT-29 human colon carcinoma
cells, 1 or 2a (10, 20, 40, or 80 mm) revealed a significant (14–23 or 33–45%, resp.) decrease of the
viability in comparison with the (negative) control. Interestingly, derivatives 3a and 3c (40 and 80 mm) led
to a significant (77–95 or 89–96%, resp.) inhibition of survival of human HT29 cells, i.e., these two
substances were ca. 63–72% or ca. 75%, respectively more effective than 5-FU (1; positive control).
Furthermore, derivative 5a showed a significant, i.e., 30 and 86%, inhibition of the survival at 40 and
80 mm, respectively in comparison with the (negative) control. Some synthesized 5-FUrd derivatives
turned out to be more effective than 5-FU (1) or 5-FUrd (2a).
1. Introduction. – Colorectal cancer (CRC) is one of the six most frequently
diagnosed malignancies in humans in the western world [1], with a five-year relative
survival rate of 44% and 60% in the United Kingdom and North America [1][2]. The
clinical problem arises from the fact that tumor cells circulate and metastasize to the
liver and to other distant sites, these metastases being the main cause of death in colon
cancer [3]. CRC is frequently adjuvantly treated with the most commonly used drug in
colon cancer chemotherapy, the cytostatic 5-fluorouracil (5-FU, 1; Fig. 1), which is
widely known as ꢁgold treatmentꢂ in the post-operative of stage III [4].
However, mechanism of action of 5-FU (1) is not fully understood. The antitumor
activity of 1 depends on the ability of the drug to bind and inactivate the enzyme
1
)
These authors contributed equally.
ꢃ 2013 Verlag Helvetica Chimica Acta AG, Zꢀrich

2236
CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
Fig. 1. Structures of 5-Fluorouracil (5-FU; 1) and its derivatives 5-fluorouridine (5-FUrd; 2a), and 2’-
deoxy-5-fluorouridine (floxuridine, 5-FdUrd; 2b)
thymidylate synthase (TS). After incorporation, the cells use 5-FU (1) for the synthesis
of deoxyuridylic and uridylic acids (5FdUMP and 5FUMP, resp.), which are
incorporated in DNA and RNA, respectively [5]. However, there are patients with
tumors intrinsically resistant to 1 [6]. More recently, the pattern and extent of cell
damage induced by chemotherapeutics (e.g., fluoropyrimidines) in human cancer cells
have been suggested to depend also on a pathway downstream from drugꢀtarget
interactions that, once triggered, initiates programmed cell death (apoptosis) [7]. 5-FU
(1), as well as its b-d-ribo- and 2’-deoxy-b-d-ribonucleosides, 2a and 2b, respectively
(Fig. 1) possess antitumor activities against various types of carcinomas, particularly of
the bladder, breast, and the gastrointestinal tract [8][9]. Promising results have been
also obtained in the topical treatment of premalignant keratosis of the skin and basal
cell carcinomas [10][11]. The intrathecal use of 2’-deoxy-5-fluorouridine (5-FdUrd; 2b)
has been studied for meningeal dissemination of malignant brain tumors, and it has
been found that this nucleoside has an excellent antitumor activity and minimal
neurotoxicity [12]. To improve the therapeutic effect of 5-FU (1), several modalities of
drug administration, e.g., iv bolus injections or continuous infusion for extended time
[13], or modification of the molecular structure of the drug might be performed to
optimize the diffusion rate. In this context, one of the major drawbacks of many
chemotherapeutics is their insufficient penetration through cell membranes, as well as
the crossing of the bloodꢀbrain barrier due to high hydrophobicity.
One method to overcome these problems is the introduction of lipophilic residues
to the drug in order to render them hydrophobic, thereby, and to improve their
pharmacokinetics [14]. In the case of low-molecular-weight drugs, this kind of chemical
modification is intended to fulfill ꢁLipinskiꢂs Rule of Fiveꢂ [15], which describes
molecular properties important for a drugꢂs pharmacokinetics in the human body
(upon oral administration), including their absorption, distribution, metabolism, and
excretion (ADME). It is highly important for drug development where a pharmaco-
logically active lead structure is optimized stepwise for increased activity and
selectivity. One part of the rule involves the drugꢂs partition coefficient (log P between
octan-1-ol and H2O) within the range of ꢀ0.4 and þ5.5. We synthesized a series of
selected nucleolipid derivatives of 5-fluorouridine (5-FUrd), carrying lipophilic

CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
2237
moieties at the N(3)-atom and/or in the 2’,3’-O-position (i.e., 3a–7a and 3c) [16][17],
and we labelled most of them with the coumarine fluorophore Atto 425 which was
coupled as N(9)-butanoate to the 5’-OH group of the corresponding nucleolipid. In this
context, the nucleolipid 7a (cf. Fig 2) was prepared from an unfarnesylated precursor as
described in [18]. For comparison, also the 2’,3’-O-nonadecylidene derivative of
uridine, 3c, was synthesized [19]. The HT-29 human colon carcinoma cell line has been
extensively used for scientific investigations and, therefore, is well-characterized [20–
22] Using human HT-29 cells, growth parameters can easily be determined by
evaluating cell death (e.g., viability, apoptosis), cell number, and protein content as a
function of the time during which cells are cultured and treated [20][21]. Herein, the
syntheses of different nucleolipid derivatives from 5-FUrd (2a) are described, including
their cancerostatic/cytotoxic activities in HT-29 human colon carcinoma cells in vitro.
2. Results. – 2.1. Synthesis. We have synthesized a series of selected nucleolipid
derivatives of 5-FUrd (2a), carrying lipophilic moieties at the N(3)-atom and/or in the
2’,3’-O-position (! 3a–7a and 3c) [16][17] and labelled most of them with the
coumarine fluorophore Atto 425ꢄ which was coupled as N(9)-butanoate to the 5’-OH of
the corresponding nucleolipid (! 3b–7b) (Fig. 2).
Fig. 2. Structures of 5-FUrd derivatives 3a–7a, as well as of Atto 425-conjugated 5-FUrd derivatives 3b–
7b, and of the uridine derivative 3c

2238
CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
This dye exhibits the longest irradiation wavelength of all Atto fluorochromes
without carrying additional ionic groups except the butanoate moiety. In this context,
the nucleolipid 7a was first obtained from a non-farnesylated precursor [16] as
described in [17]. For comparison, also the 2’,3’-O-nonadecylidene derivative of
uridine, 3c, was prepared [19a]. Coupling of compounds 3a–7a with the fluorophore
was performed applying the Steglich reaction (N,N’-dicyclohexylcarbodiimide (DCC),
4-(dimethylamino)pyridine (DMAP)) [23]. The products 3b–7b were purified by
silica-gel column chromatography and characterized by fluorescence spectroscopy, as
well as by HR-ESI-MS (for details, see Exper. Part).
Fig. 3 shows electrospray ionization (ESI) mass spectra of compounds 3b (a), 4b
(b), 5b (c), 6b (d), and 7b (e), indicating the correct, single-charged molecular-ion
peaks as [MþH]þ and/or [MþNa]þ peaks. Moreover, characteristic single-charged
fragment-ion peaks such as [Mꢀ(2’,3’-O-ketal group)]þ or [Mꢀ(Atto 425)]þ could be
detected.
2.2. Anticancer Activity. Then, derivates of 5-FUrd (2a) derivatives were
investigated regarding their anticancer activities using HT-29 human colon carcinoma
cells in comparison with 5-FU (1) or with 5-FUrd (2a). Effects of all tested substances
on viability/survival of human HT-29 cells after 24 h treatment are depicted in Fig. 4.
After 24 h of incubation, 1 revealed, at concentrations of 10, 20, 40, and 80 mm, a
significant (14–23%) decrease of the viability of HT-29 human colon carcinoma cells in
comparison with the (negative) control (Fig. 4). Moreover, after 24-h treatment with
10, 20, 40, and 80 mm 5-FUrd (2a), the survival of human HT-29 cells was significantly
decreased (33–45%), in comparison with the (negative) control (Fig. 4).
Substance 3a showed a 77 and 95% significant decrease (p<0.001) of the viability
of human HT-29 cells at 40 or 80 mm, respectively, whereas the derivative 3c
significantly (p<0.001) diminished the survival of HT-29 human colon carcinoma
cells by 89–96% at 40 and 80 mm (Fig. 4). Additionally, after 24-h treatment at
concentrations of 40 and 80 mm, derivatives 3a and 3c were more effective than 5-FU
(1), because both substances significantly reduced the viability of human HT-29 cells by
ca. 63–72% or 75%, respectively, in comparison with 1 (positive) control (Fig. 4).
Derivatives 4a and 6a did not significantly affect the survival of HT-29 human colon
carcinoma cells compared to the (negative) control; however, at concentrations of 40
and 80 mm, both derivatives indicated significant supportive effects on cell growth of
human HT-29 cells in comparison with 1 (positive) control (Fig. 4). Furthermore,
substance 5a showed a significant 30 and 86% inhibition of the survival at 40 and 80 mm,
respectively, in comparison with the (negative) control (Fig. 4). The derivative 7a did
not exhibit any significant effect on survival of HT-29 human colon carcinoma cells
after 24-h treatment at concentrations of 10, 20, 40 and 80 mm (Fig. 4). The compounds
3b–7b were the derivatives of 3a–7a after 5’-O-labeling with the fluorochrome Atto
425 (Fig. 2).
Fig. 5 shows the data concerning survival of human HT-29 cells after 24 h
incubation with different concentrations (10, 20, 40, and 80 mm) with the Atto 425-
conjugated derivatives. In this context, we showed that former derivatives 3a and 5a,
after conjugation with Atto 425, have lost their survival inhibitory activity of human
HT-29 cells (Fig. 5). However, 5’-O-labelled Atto 425 5-FUrd (2c) significantly
decreased the survival of HT-29 human colon carcinoma cells by 31–52% after 24-h

CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
2239
Fig. 3. ESI-MS Spectra of compounds 3b (a), 4b (b), 5b (c), 6b (d), and 7b (e). For details, see Exper.
Part.

2240
CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
Fig. 4. Viability/survival of human HT-29 colon carcinoma cells after 24-h incubation with 5-FU (1) (as
positive control), 5-FUrd (2a), and its derivatives 3a, 4a, 5a, 6a, 7a, and 3c. Values are given in % survival
of control (incubation with medium alone, 100% survival), as meanꢁSEM; p, significance vs. [negative]
control (medium alone; *: p<0.05, **: p<0.01, ***: p<0.001 and significance vs. 5-FU (positive
control) þ: p<0.05, þ þ: p<0.01, þ þ þ: p<0.001.
Fig. 5. Viability/survival of human colon carcinoma cell line HT-29 after 24 h incubation with Atto 425-
conjugated 5-FdUrd (2c), its derivatives 3b, 4b, 5b, 6b, 7b, and Atto 425. 1ꢂAtto¼Atto 425 alone (control
for derivatives conjugated with one Atto 425 molecule); 3ꢂAtto¼threefold concentration of Atto 425
(control for 5-FU conjugated with three Atto 425 molecules). Values are given in % survival of control
(without treatment/medium alone, 100% survival), as meanꢁSEM; p, significance vs. control without
treatment, *: p<0.05, **: p<0.01, ***: p<0.001. N¼4 independent experiments, using 4–6 wells per
treatment and experiment.
treatment at concentrations of 10, 20, 40, and 80 mm (Fig. 5). All 5’-O-labelled Atto 425
derivatives were incorporated by the human HT-29 cells and accumulated homoge-
neously in their cytoplasm (exemplarily shown in Fig. 6). However, the derivatives

CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
2241
Fig. 6. Representative images of the intracellular localization of Atto 425-conjugated 5-FUrd (2c) and
some of its derivatives in human HT-29 adenocarcinoma cells. a) 2c, b) 4b, c) 5b after 24-h treatment.
Magnification, 200ꢂ.
avoided the nuclei of the cells (Fig. 6). These distributions were observed by all 5-FUrd
derivatives with the exception of the one of 5-FUrd, 2c, itself, which showed a
granulated distribution in the cytoplasm (Fig. 6).
3. Discussion. – Fluoropyrimidines are antimetabolite drugs widely used in the
treatment of cancer. The effects of the fluoropyrimidines, such as 5-FU (1) and 5-FUrd
(2a) arise from a complex combination of different molecular interactions between the
fluoropyrimidines, their metabolites, and the plasma and/or the nuclear membrane that
facilitate or complicate their availability to be effective. Anticancer therapy, well-
known to clinical oncologists, depends on the efficiency of a therapeutic agent, which is
limited by the specific distribution from the peripheral blood circulation to the tumor
[24]. As a consequence, only a small percentage of the administered dose reaches the
tumor, whereas the remaining drug causes unwanted side effects [24].
5-FU (1) is most commonly intravenously injected; however, more than 80% of this
drug is metabolized in the liver [25]. Moreover, an important consideration in the use
of 1 and related drugs is the development of drug resistance by the tumor. Drug
resistance can also involve changes in drug transport. In this context, to improve the
therapeutic efficacy of 5-FU (1), several modalities of drug administration or
modification of the molecular structure may be considered and investigated. In
accordance with these therapeutic strategies we synthesized several 5-FUrd derivatives
and investigated their cytotoxic effects against HT-29 human colon carcinoma cells.
Our results clearly revealed that the derivatives 3a and 5a, as well as the uridine
nucleolipide 3c, do enter the cancer cells and exhibit the highest in vitro activity, i.e.,
significantly inhibited survival of the human HT-29 cells. The internalization of cationic
cell-penetrating drugs depends on lipophilic interactions on the cell surface, leading to
a direct entry by diffusion or endocytosis. According to Lipinskiꢂs rule [15], the most
important molecular properties for a drugꢂs pharmacokinetics in the human body are
ADME. Our data confirm the absorption by detection of the cytoplasmic localization
of the 5’-O-Atto 425-labelled 5-FUrd derivatives 3b–7b, but they cannot penetrate into
the nucleus, their intracellular target, which may explain the loss of their function/
cytotoxicity due to 5’-O-Atto 425-labelling of 3a–7a (! 3b–7b). A conceivable
interpretation of this result might be the blocking of the 5’-positions of the nucleolipids

2242
CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
by covalent attachment of the dye, preventing a phosphorylation of the corresponding
compound to its 5’-triphosphate. On the other hand, the unlabelled (i.e., without Atto
425 conjugation) nucleolipids 3a–7a can in principle be phosphorylated and enzymati-
cally appended to the growing nucleic acid chain. All nucleolipids, however, possess
terminator properties with respect of nucleic acid elongation; after a single terminal
incorporation of an 2’,3’-O-blocked nucleotide, an enzymatic elongation of the growing
nucleic acid chain is abolished, and cell apoptosis is initiated. Many biochemical
mechanisms seem to be responsible for the cytotoxicity of 5-FU (1), e.g., i) inhibition of
thymidylate synthase, following its conversion to 2’-deoxy-5-fluorouridine 5’-mono-
phosphate; ii) conversion to 5-fluorouridine 5’-triphosphate and its incorporation into
RNA; iii) its incorporation into DNA, leading to an impairment of the integrity of
DNA [26][27]. In case of the active nucleolipids, however, it is reasonable that not this
classical biochemical mechanism causes apoptosis of the tumor cells but the above
described abolishment of the nucleic acid elongation due to an 2’,3’-O-blocking. This
termination mechanism is analogous to that described for the action of 3’-azido-2’,3’-
dideoxythymidine (AZT) as antiretroviral drug. Nuclear delivery of a given drug into
living cells needs to overcome two major obstacles: 1) the cellular internalization by
direct translocation in an energy-independent manner or through endocytosis where
escape from endosomes is essential for translocation into various cellular compart-
ments, and 2) the passage through the nuclear envelope into the nuclei [28]. The
cellular internalization of cationic cell-penetrating peptides (CPPs) is dependent on
lipophilic interactions on the cell surface which lead to direct entry by diffusion or
endocytosis [28]. The active 5-FUrd derivatives, 3a and 5a, after 24-h treatment of
human HT-29 cells, have more lipid moieties than compounds 4a and 6a. An increase of
the lipophilic properties may be associated with an increase of the cytotoxic effect. The
substances 5-FU (1) and 5-FUrd (2a) have displayed an inhibition curve characteristic
of a cytostatic effect described in [29–31]. Regarding the derivatives 3a and 3c, the
cytostatic effect switches to a cytotoxic effect after loss of the F-atom in the molecule
3c. Because 5-FU (1) cytotoxicity has been primarily focused on increasing activation,
decreasing degradation, and enhancing the enzyme thymidylate synthase inhibition,
the derivative 3c might have another or a complementary target, or distinct
mechanisms of action in the cell, which cannot be disclosed in detail at the moment.
In conclusion, we have described novel synthesized 5-FUrd derivatives which are
incorporated into HT-29 human colon cancer cells, and are more effective than 5-FU
(1) or 5-FUrd (2a) concerning inhibition of survival of these cells. However, further
investigations are required to decipher the molecular mechanism of 5-FUrd (2a) and its
derivatives, as well as their cytotoxic effects after long-time treatment of human HT-29
cells, and also of other cancer cell lines.
Experimental Part
General. All chemicals were purchased from SigmaꢀAldrich (D-Deisenhofen) and 5-FU (1) from
TCI Europe (B-Zwijndrecht). Atto 425 was purchased from Atto-Tec GmbH (D-Siegen). Solvents were
of laboratory grade and were distilled before use. TLC: Al sheets, silica gel 60 F254, 0.2-mm layer (Merck).
M.p. Bꢀchi SMP-20; uncorrected. UV Spectra: Cary 1E spectrophotometer (Varian, D-Darmstadt).
NMR Spectra: AMX-500 spectrometer (Bruker, D-Rheinstetten); at 500.14 (1H) and 125.76 MHz (13C);

CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
2243
chemical shifts, d, are in ppm rel. to TMS as internal standard; J values in Hz. ESI-MS: Bruker Daltronics
Esquire HCT instrument (Bruker Daltronics, D-Leipzig); ionization performed with a 2% aq. HCOOH
soln. 5-FU (1) and the synthesized 5-FUrd derivatives were dissolved in DMSO (SigmaꢀAldrich Chemie
GmbH).
Synthesis of 5-FUrd Derivatives. Compounds 4a and 6a were prepared as described in [16a];
compound 3a was synthesized as described in [17]; compound 5a was prepared according to [16b].
Compound 3c was prepared as described in [19].
2’,3’-O-Cyclododecylidene-5-fluoro-3-[(2E,6E)-3,7,11-trimethyl-2,6,10-dodecatrien-1-yl]uridine
(7a). Compound 6a (463.5 mg, 1.086 mmol) was dissolved in anh. DMF. Under N2, K2CO3 (390 mg,
2.82 mmol) was added, and the suspension was stirred for 10 min at r.t. Then, farnesyl bromide (0.33 ml,
1.19 mmol) was added dropwise within 2 h. After stirring overnight, the mixture was filtered, and the
solid residues was washed with a small amount of CH2Cl2. The combined filtrates were evaporated to
dryness in high vacuum overnight. The oily residue was subjected to CC (SiO2 (6ꢂ10 cm); CH2Cl2/
MeOH 97:3 (v/v)) to give a main zone, from which 7a (579 mg, 85%) was isolated. Colorless oil. Rf
(SiO2; CH2Cl2/MeOH 97:3 (v/v)) 0.57. UV (MeOH): 268 (12.400). 1H-NMR ((D6)DMSO): 8.24 (d,
3J(HꢀC(6),F)¼7.0, HꢀC(6)); 5.90 (d, 3J(1’,2’)¼2.4, HꢀC(1’)); 5.20 (br. s, HOꢀC(5’)); 5.13 (t, 3J(2’’,1’’)¼
6.6, HꢀC(2’’)); 5.04 (d, 3J(1’’,2’’)¼6.0, CH2(1’’)); 4.85 (dd, 3J(2’,1’)¼3.5, 3J(2’,3’)¼6.5, HꢀC(2’)); 4.75 (dd,
3J(3’,2’)¼6.5, 3J(3’,4’)¼3.0, HꢀC(3’)); 4.41–4.39 (m, HꢀC(6’’), HꢀC(10’’)); 4.14 (Yq, 3J(4’,3’)¼3.5,
2
3J(4’,5’ and 5’’)¼3.5, HꢀC(4’)); 3.65–3.56 (Y-octet, J(5’,5’’)¼ ꢀ15, CH2(5’)); 2.04–1.89 (4m, CH2(8’’),
CH2(9’’), CH2(5’’), CH2(4’’)); 1.74 (m, Me(13’’), 2 HendoꢀC(a’)); 1.63 (s, Me(14’’)); 1.56–1.53 (3m,
Me(15’’), Me(12’’), 2 HexoꢀC(a)); 1.45 (m, 2 HendoꢀC(b’)); 1.33–1.31 (m, 8 CH2(ketal)). 13C-NMR
((D6)DMSO): 156.1 (d, 2J(C(4),F)¼26.0, C(4)); 148.7 (C(2)); 139.4 (d, 1J(C(5),F)¼228.9, C(5)); 139.3
(C(3’’)); 134.5 (C(7’’)); 130.5 (C(11’’)); 124.3 (d, 2J(C(6),F)¼34.7, C(6)); 124.0 (C(6’’)); 123.6 (C(10’’));
118.1 (C(2’’)); 117.0 (OꢀCꢀO); 91.8 (C(1’)); 86.7 (C(4’)); 83.6 (C(2’)); 80.1 (C(3’)); 61.1 (C(5’)); 39.0,
38.9, 38.3 (C(1’’), C(4’’), C(8’’)); 33.3 (C(a’)); 30.3 (C(a)); 25.9 (C(5’’)); 25.4 (C(9’’)); 25.2 (C(12’’)); 17.2
(C(15’’)); 15.9 (C(14’’)); 15.5 (C(13’’)); 25.5, 25.4, 25.2, 25.0, 21.8, 21.5, 21.4, 19.6, 19.5 (9 CH2). ESI-MS:
631.50 ([MþH]þ ), 335.2 ([N(3)-farnesyl-5-fluorouracil]þ ).
Synthesis of 5-FUrdꢀAtto 425 Conjugates. Coupling of Selected Nucleolipids of 5-FUrd, 3a–7a, with
4-[3-(Ethoxycarbonyl)-7,8-dihydro-6,8,8-trimethyl-2-oxo-2H-pyrano[3,2-g]quinolin-9(6H)-yl]butanoic
acid (¼Atto 425 N(9)-butanoate). All Steglich esterifications were performed as described in [23] but
with slight modifications for each compound. The corresponding products were isolated as diastereoiso-
meric mixtures due to the stereogenic center C(6) of the fluorophore.
5’-{4-[3-(Ethoxycarbonyl)-7,8-dihydro-6,8,8-trimethyl-2-oxo-2H-pyrano[3,2-g]quinolin-9(6H)-yl]-
butanoyl}-5-fluoro-2’,3’-O-(1-nonyldecylidene)uridine (3b). Atto 425 N(9)-butanoate (5 mg,
12.244 mmol) was dissolved in anh. CH2Cl2 (1.5 ml), and 4-(dimethylamino)pyridine (DMAP; 0.6 mg,
0.01244 mmol), dissolved in CH2Cl2 (0.5 ml), and 3a (6.55 mg, 12.44 mmol), dissolved in CH2Cl2 (1.3 ml),
were added under N2 and cooling in an ice bath. Thereupon, N,N’-dicyclohexylcarbodiimide (DCC;
2.57 mg, 12.44 mmol), dissolved in CH2Cl2 (0.11 ml), was added dropwise during 45 min. After 5 min, the
soln. was allowed to warm to r.t., and stirring was continued overnight under exclusion of light. The
reaction was monitored by TLC (CH2Cl2/MeOH 93 :7 (v/v)). After addition of further 30 mol-% of
DMAP, DCC, and 3a, stirring was continued for 48 h in total. The mixture was evaporated in vacuo, and
the residue was purified by repeated CC (SiO2 60 (2ꢂ21.5 cm); CH2Cl2/MeOH 98 :2 (v/v)) to afford 3b
quant. Green fluorescent solid (ca. 13.3 mg). Rf (SiO2; CH2Cl2/MeOH 93 :7 (v/v)) 0.88. Fluorescence
spectroscopy: lmax (irrad.), 426; lmax (emission), 465 nm. ESI-MS: 911.1 ([MþH]þ ), 932.9 ([MþNa]þ ),
630.4 ([MþNaꢀ1-nonyldecylidene]þ ), 608.4 ([MþHꢀ1-nonyldecylidene]þ ).
5’-{4-[3-(Ethoxycarbonyl)-7,8-dihydro-6,8,8-trimethyl-2-oxo-2H-pyrano[3,2-g]quinolin-9(6H)-yl]-
butanoyl}-2’,3’-O-[(1R)-4-ethoxy-1-methyl-4-oxobutylidene]-5-fluorouridine (4b). Atto 425 N(9)-buta-
noate (5 mg, 12.244 mmol) was dissolved in anh. CH2Cl2 (1.5 ml), and DMAP (0.6 mg, 0.01244 mmol),
dissolved in CH2Cl2 (0.5 ml), and 4a (4.83 mg, 12.44 mmol) were added under N2 and cooling in an ice
bath. Thereupon, DCC (2.57 mg, 12.44 mmol), dissolved in CH2Cl2 (0.11 ml), was added dropwise during
45 min. After 5 min, the soln. was allowed to warm to r.t., and stirring was continued overnight under
exclusion of light. The reaction was monitored by TLC (CH2Cl2/MeOH, 9 :1 (v/v)). The product
appeared as two fluorescent spots. After addition of further 30 mol-% of DMAP, DCC, and 4a, stirring

2244
CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
was continued for 48 h in total. The mixture was evaporated in vacuo, and the residue was subjected to
CC (SiO2 60 (2ꢂ15 cm); CH2Cl2/MeOH 93 :7 (v/v)) to give 4b (quant.). Green fluorescent solid (ca.
9.6 mg). Rf (SiO2; CH2Cl2/MeOH, 9 :1 (v/v)) 0.85 and 0.75. Fluorescence spectroscopy: lmax (irrad.), 426;
lmax (emission) 462 nm. ESI-MS: 772.82 ([MþH]þ ), 794.4 ([MþNa]þ ), 630.4 ([MþNa]ꢀethyl
levulinate]þ ), 449.4 ([MþNa]ꢀ(Atto 425 N(9)-butanoate)]þ ).
5’-{4-[3-(Ethoxycarbonyl)-7,8-dihydro-6,8,8-trimethyl-2-oxo-2H-pyrano[3,2-g]quinolin-9(6H)-yl]-
butanoyl}-2’,3’-O-[(1R)-4-ethoxy-1-methyl-4-oxobutylidene]-5-fluoro-3-[(2E,6E)-3,7,11-trimethyldode-
ca-2,6,10-trien-1-yl]uridine (5b). Atto 425 N(9)-butanoate (5 mg, 12.244 mmol) was dissolved in anh.
CH2Cl2 (1.5 ml), and DMAP (0.6 mg, 0.01244 mmol), dissolved in CH2Cl2 (0.5 ml), and 5a (7.37 mg,
12.44 mmol), dissolved in CH2Cl2 (1.3 ml), were added under N2 and cooling in an ice bath. Thereupon,
DCC (2.57 mg, 12.44 mmol), dissolved in CH2Cl2 (0.11 ml), was added dropwise during 45 min. After
5 min, the soln. was allowed to warm to r.t., and stirring was continued overnight under exclusion of light.
The reaction was monitored by TLC (CH2Cl2/MeOH 95 :5 (v/v)). After addition of further 30 mol-% of
DMAP, DCC, and 5a, stirring was continued for 48 h in total. The mixture was evaporated in vacuo, and
the residue was purified by repeated CC (SiO2 60 (2ꢂ23 cm); CH2Cl2/MeOH 96 :4 (v/v)) furnish 5b
(7.7 mg, 63.4%). Green fluorescent solid. Rf (SiO2; CH2Cl2/MeOH 95 :5 (v/v)): 0.83 and 0.64.
Fluorescence spectroscopy: lmax (irrad.), 426; lmax (emission), 460 nm. ESI-MS: 976.6 ([MþH]þ ),
999.1 ([MþNa]þ ), 608.5 ([MþHꢀ(Atto 425 N(9)-butanoate)]þ ).
2’,3’-O-Cyclododecylidene-5’-{4-[3-(ethoxycarbonyl)-7,8-dihydro-6,8,8-trimethyl-2-oxo-2H-pyra-
no[3,2-g]quinolin-9(6H)-yl]butanoyl}-5-fluorouridine (6b). Atto 425 N(9)-butanoate (5 mg,
12.244 mmol) was dissolved in anh. CH2Cl2 (1.5 ml), and DMAP (0.72 mg, 0.01244 mmol), dissolved in
CH2Cl2 (0.6 ml), and 6a (6.37 mg, 14.93 mmol), dissolved in CH2Cl2 (1.8 ml), were added under N2 and
cooling in an ice bath. Because of the insufficient solubility of the reagents, MeCN (1 ml) was added.
Thereupon, DCC (3.08 mg, 14.93 mmol), dissolved in CH2Cl2 (0.13 ml), was added dropwise during
45 min. After stirring for 5 min, the soln. was allowed to warm to r.t., and stirring was continued overnight
under exclusion of light. The reaction was monitored by TLC (CH2Cl2/MeOH 93 :7 (v/v)). After addition
of further 10 mol-% of DMAP, DCC, and 6a, stirring was continued for 48 h in total. The mixture was
evaporated in vacuo, and the residue was purified by CC (SiO2 60 (2ꢂ15.5 cm); CH2Cl2/MeOH 94 :6 (v/
v)) to afford 6b (9.5 mg, 94.6%). Green fluorescent solid. Rf (SiO2; CH2Cl2/MeOH 93 :7 (v/v)) 0.84 and
0.75. Fluorescence spectroscopy: lmax (irrad.) 426; lmax (emission) 460 nm. ESI-MS: 810.5 ([MþH]þ ),
832.5 ([MþNa]þ ), 630.5 ([MþNaꢀcyclododecyl]þ ), 608.5 ([MþHꢀcyclododecyl]þ ).
2’,3’-O-Cyclododecylidene-5’-{4-[3-(ethoxycarbonyl)-7,8-dihydro-6,8,8-trimethyl-2-oxo-2H-pyra-
no[3,2-g]quinolin-9(6H)-yl]butanoyl}-5-fluoro-3-[(2E,6E)-3,7,11-trimethyl-2,6,10-dodecatrien-1-yl]uri-
dine (7b). Atto 425 N(9)-butanoate (5 mg, 12.244 mmol) was dissolved in anh. CH2Cl2 (1.5 ml), and
DMAP (0.6 mg, 0.01244 mmol), dissolved in CH2Cl2 (0.5 ml), and 7a (7.85 mg, 12.44 mmol), dissolved in
CH2Cl2 (1.3 ml), were added under N2 and cooling in an ice bath. Thereupon, DCC (2.57 mg,
12.44 mmol), dissolved in CH2Cl2 (0.11 ml), was added dropwise during 45 min. After 5 min, the soln.
was allowed to warm to r.t., and stirring was continued overnight under exclusion of light. The reaction
was monitored by TLC (CH2Cl2/MeOH 96 :4 (v/v)). After addition of further 30 mol-% of DMAP, DCC,
and 7a, stirring was continued for 48 h in total. The mixture was evaporated in vacuo, and the residue was
purified by CC (SiO2 60 (2ꢂ25.5 cm); CH2Cl2/MeOH 98 :2 (v/v)) to give 7b (10.9 mg, 86.6%). Green
fluorescent solid. Rf (SiO2; CH2Cl2/MeOH 95 :5 (v/v)) 0.88 and 0.68. Fluorescence spectroscopy: lmax
(irrad.) 426; lmax (emission) 460 nm. ESI-MS: 1015.3 ([MþH]þ ), 1036.7 ([MþNa]þ ), 630.4 ([MþNaꢀ
cyclododecyl]þ ), 608.4 ([MþHꢀcyclododecyl]þ ).
Cell Lines and Culture Conditions. In vitro experiments were performed using the human colon
carcinoma cell line HT-29 (DSMZ, GmbH, D-Braunschweig), which originated from the primary tumor
of a 44-year-old Caucasian woman with colon adenocarcinoma. The cells were cultured in 90% RPMI
1640 supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, and 0.1 mg/ml
streptomycin. Cells were maintained at 378 in a humidified atmosphere of 5% CO2 and 95% air.
Determination of Viability/Survival of 5-Fluorouridine and Its Derivatives. HT-29 Human colon
carcinoma cells (5ꢂ103; 100 ml medium/well) were seeded in 96-well plates (BD FalconTM, Becton
Dickinson GmbH, D-Heidelberg). After 24 h, the medium was changed and different concentrations of
5-fluorouracil (5-FU; 1), 5-fluorouridine (5-FUrd; 2a), 5-FUrd derivatives 3a–7a and 3c, or Atto 425

CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
2245
conjugated derivatives, 2c and 3b–7b, were tested at concentrations of 10, 20, 40, or 80 mm. After
incubation for 24 h, viability/survival was measured using PrestoBlueTM reagent (Invitrogen-Life
Technologies GmbH, DE-Darmstadt). PrestoBlueTM reagent is more sensible than AlamarBlueꢄ, which
is a redox indicator of enzyme activity widely used in whole organism screening for viability/cytotoxicity
[32][33]. PrestoBlueTM was directly added into the culture medium at a final concentration of 10%. Then,
the plates were returned to the incubator, and 30 min, 1 h, 2 h, 3 h, 4 h, and 6 h after addition of
PrestoBlueTM the optical density (OD) was measured at 570 and 600 nm (as reference) with a SUNRISE
ELISA-reader (Tecan, AT-Salzburg). Results are expressed in % of survival (OD 570/600 nm of
samplesꢂ100/OD 570/600 nm of control without substances)). As control (100% viability), cells were
cultured with medium alone (i.e., without addition of test substances).
Intracellular Localization of Atto 425-Conjugated Derivatives. HT-29 Human carcinoma cells (4ꢂ
104; 400 ml medium/well) were cultured on sterile microscope glass slides with 8-well silicone inserts
(flexiPERMꢄ cell culture chamber; Sarstedt AG, CH-Sevelen), and after 24 h the substances under test
were added at 20 mm during 24 h. Subsequently, the cells were washed with PBS and covered with Pro
Longꢄ Gold antifade reagent (Life Technologies GmbH, D-Darmstadt). Afterwards, the cells were
analyzed using a specific Atto filter (ex. 436 nm/em. 484 nm) and photographed with an Axio Image.M2
microscope (Carl Zeiss GmbH, D-Jena), including the digital high-resolution imaging system AxioCam
HRc/AxioVision Rel. 4.8 (Carl Zeiss GmbH). Digitalized figures were processed, arranged, and lettered
by standard imaging software.
Statistical Analysis: The SigmaPlot 12.2 software (Systat Software Inc., Chicago) was used to carry
out statistical analyses by the unpaired Studentꢂs t test or WilcoxonꢀMannꢀWhitney test. Data are shown
as meanꢁSEM. A p value<0.05 was considered statistically significant.
This work was supported by B. Braun Melsungen AG, DE-Melsungen. The authors wish to thank
Mrs. Andrea Cordes for excellent technical assistance, as well as Mrs. Ellen Essen and Mrs. Gabriella
Stauch for manuscript preparation.
REFERENCES
[1] S. H. Rim, L. Seeff, F. Ahmed, J. B. King, S. S. Coughlin, Cancer 2009, 115, 1967.
[2] M. P. Coleman, M. Quaresma, F. Berrino, J.-M. Lutz, R. De Angelis, R. Capocaccia, P. Baili, B.
Rachet, G. Gatta, T. Hakulinen, A. Micheli, M. Sant, H. K. Weir, J. M. Elwood, H. Tsukuma, S.
Koifman, G. Azevedo e Silva, S. Francisci, M. Santaquilani, A. Verdecchia, H. H. Storm, J. L.
Young, Lancet Oncol. 2008, 9, 730.
[3] S. B. Ho, P.-S. Yan, R. Dahiya, B. A. Neuschwander-Tetri, C. Basbaum, Y. S. Kim, J. Cell Physiol.
1994, 160, 213.
[4] A. Carrato, Gastrointest. Cancer Res. 2008, 2 (4 Suppl. 2), S42.
[5] C. Marin-Vicente, Y. Lyutvinskiy, P. Romans Fuertes, R. A. Zubarev, N. Visa, J. Proteom. Res. 2013,
12, 1969.
[6] M. E. Nita, H. Nagawa, O. Tominaga, N. Tsuno, S. Fujii, S. Sasaki, C.-G. Fu, T. Takenoue, T. Tsuruo,
T. Muto, Br. J. Cancer 1998, 78, 986.
[7] S. W. Lowe, H. E. Ruley, T. Jacks, D. E. Housman, Cell 1993, 74, 957.
[8] H. Furuse, Y. Hirano, M. Harada, L. H. Ming, T. Aoki, Y. Kurita, S. Mugiya, T. Ushiyama, S. Ozono,
Anticancer Res. 2009, 29, 1001.
[9] M. A. Garcꢅa, E. Carrasco, M. Aguilera, P. Alvarez, C. Rivas, J. M. Campos, J. C. Prados, M. A.
´
Calleja, M. Esteban, J. A. Marchal, A. Aranega, PLoS One 2011, 6, e23887.
[10] A. Albert, ꢁSelective toxicity: the physico-chemical basis of therapyꢂ, Chapman and Hall, London,
New York, 7th edn., 1985, pp. 60, 125, 126.
[11] C. Heidelberger, L. Griesbach, O. Cruz, R. Schnitzer, F. Grunberg, Exper. Biol. Med. 1958, 97, 470.
[12] M. Yamada, H. Nakagawa, M. Fukushima, K. Shimizu, T. Hayakawa, K. Ikenaka, J. Neuro-Oncol.
1998, 37, 115.
[13] G. J. Peters, ꢁAntimetabolitesꢂ, in ꢁOxford Textbook of Oncologyꢂ, Vol. 1, Eds. M. Peckham, H. M.
Pinedo, U. Veronesi, Oxford University Press, Oxford, 1995, pp. 524–552.

2246
CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
[14] H. Rosemeyer, Chem. Biodiversity 2005, 2, 977; D. R. Lu, S. Øie, ꢁCellular Drug Delivery: Principles
and Practiceꢂ, Humana Press, Totowa, New Jersey, 2004; P. Dasgupta, R. Mukherjee, Br. J.
Pharmacol. 2000, 129, 101; C. F. Torres, D. Martin, G. Torrelo, V. Casado, O. Fernandez, D. Tenllado,
L. Vazquez, M. I. Moran-Valero, G. Reglero, Curr. Nutr. Food Sci. 2011, 7, 160.
[15] C. A. Lipinski, F. Lombardo, B. W. Dominy, P. J. Feeney, Adv. Drug Delivery Rev. 1997, 23, 3; A. K.
Ghose, V. N. Viswanadhan, J. J. Wendoloski, J. Comb. Chem. 1999, 1, 55.
[16] a) E. Malecki, H. Rosemeyer, Helv. Chim. Acta 2010, 93, 1500; b) E. Malecki, R. Viere, H.
Rosemeyer, Chem. Biodiversity 2013, 10, 1828.
[17] E. Malecki, V. Ottenhaus, E. Werz, H. Rosemeyer, Chem. Biodiversity 2014, in print, doi: 10.1002/
cbdv.201300127; H. Rosemeyer, E. Malecki, Eur. Pat. Appl. EP 12 186 564.6, 2012, priority date:
September 28, 2012.
[18] K. Kçstler, E. Werz, E. Malecki, M. Montilla-Martinez, H. Rosemeyer, Chem. Biodiversity 2013, 10,
39.
[19] a) E. Werz, R. Viere, G. Gassmann, S. Korneev, E. Malecki, H. Rosemeyer, Helv. Chim. Acta 2013,
96, 872; b) H. Rosemeyer, E. Malecki, E. Werz, S. Korneev, K. Gall, Eur. Pat. Appl. EP 12 186 576.0,
2012, priority date: September 28, 2012.
[20] P. M. Calabro-Jones, J. E. Byfield, J. F. Ward, T. R. Sharp, Cancer Res. 1982, 42, 4413.
[21] X.-Y. Tang, Y.-Q. Zhu, W.-H. Tao, B. Wei, X.-L. Lin, Postgrad. Med. J. 2007, 83, 338.
[22] G. A. Bonaterra, O. Kelber, D. Weiser, R. Kinscherf, Phytomedicine 2013, 20, 691.
[23] B. Neises, W. Steglich, Angew. Chem., Int. Ed. 1978, 17, 522.
[24] R. M. Mader, M. Kalipciyan, P. Ohana, A. Hochberg, G. G. Steger, Int. J. Clin. Pharmacol. Ther.
2011, 49, 69.
[25] R. B. Diasio, B. E. Harris, Clin. Pharmacokinet. 1989, 16, 215.
[26] S. D. Soni, T. Srikrishnan, Nucleosides, Nucleotides Nucleic Acids 2004, 23, 1779.
[27] J. Pasztor, PhD Thesis, University of Vienna, 2010.
[28] F. Noor, R. Kinscherf, G. A. Bonaterra, S. Walczak, S. Wçlfl, N. Metzler-Nolte, ChemBioChem 2009,
10, 493.
[29] H. Eda, K. Fujimoto, S. Watanabe, M. Ura, A. Hino, Y. Tanaka, K. Wada, H Ishitsuka, Cancer
Chemother. Pharmacol. 1993, 32, 333.
[30] J. Weinreich, S. Schott, I. Kçnigsrainer, D. Zieker, A. Kçnigsrainer, H. Schott, Invest. New Drugs
2011, 29, 1294.
[31] J. Weinreich, T. C. Schott, I. Kçnigsrainer, M. Kꢀper, A. Kçnigsrainer, H. Schott, Anticancer Res.
2012, 32, 4299.
[32] M. R. van der Harst, S. Bull, C. M. Laffont, W. R. Klein, Vet. Res. Commun. 2005, 29, 247.
[33] S. Al-Nasiry, N. Geusens, M. Hanssens, C. Luyten, R. Pijnenborg, Hum. Reprod. 2007, 22, 1304.
Received July 2, 2013