4-Thio-5-bromo-2′-deoxyuridine: Chemical synthesis and therapeutic potential of UVA-induced DNA damage

Article abstract of DOI:10.1016/j.bmcl.2003.11.069

4-Thio-5-bromo-2′-deoxyuridine (3a) is prepared from 5-bromo-2′-deoxyuridine (BrdU) and its key properties are explored. The thionucleoside (3a) can react readily with monobromobimane and produces high fluorescence. 3a has UV maximum absorption at 340 nm and can be incorporated into cellular DNA. The cells containing 3a become sensitive to UVA light, offering therapeutic potential for UVA-induced cell killing.

Full text of DOI:10.1016/j.bmcl.2003.11.069

Bioorganic & Medicinal Chemistry Letters 14 (2004) 995–997
4-Thio-5-bromo-20-deoxyuridine: chemical synthesis and
therapeutic potential of UVA-induced DNA damage
Yao-Zhong Xu,a,* Xiaohui Zhang,a Hai-Chen Wu,a Andrew Masseyb
and Peter Karranb
aDepartment of Chemistry, The Open University, Walton Hall, Milton Keynes MK7 6AA, UK
bCancer Research UK London Research Institute, Clare Hall Laboratories, South Mimms EN6 3LD, UK
Received 7October 2003; revised 24 November 2003; accepted 25 November 2003
Abstract—4-Thio-5-bromo-20-deoxyuridine (3a) is prepared from 5-bromo-20-deoxyuridine (BrdU) and its key properties are
explored. The thionucleoside (3a) can react readily with monobromobimane and produces high fluorescence. 3a has UV maximum
absorption at 340 nm and can be incorporated into cellular DNA. The cells containing 3a become sensitive to UVA light, offering
therapeutic potential for UVA-induced cell killing.
# 2003 Elsevier Ltd. All rights reserved.
1. Introduction
proliferating cells selectively. While investigating its
killing mechanism, we also embarked on a search for
novel or improved pro-drugs for UVA-induced cell killing.
DNA is frequently damaged by internal and external
agents. Damaged DNA, if not repaired, can lead to
mutation and cancer.1 On the other hand, purposely
designed DNA damage can be used as a means to kill
cancer cells. This approach underlies chemotherapy and
radiotherapy in which a toxic chemical or high energy
radiation is employed to inflict lethal DNA damage on
cancer cells. However, these treatments are undiscrimi-
nating and also harm normal cells. An ideal treatment
would be to target DNA in cancer cells selectively while
minimising damage to DNA in normal cells. Recently
we reported2,3 that 4-thiothymidine (S4-TdR), an ana-
logue of the naturally occurring nucleoside thymidine,
can be incorporated into cellular DNA. More impor-
tantly, cells containing thiothymidine DNA are highly
sensitive to low doses of UVA light.3 These findings
suggested a new therapeutic approach in which a mini-
mally mutagenic, non-toxic pro-drug (e.g., S4-TdR) is
introduced into the DNA of target cells and then acti-
vated by a localised treatment with UVA light to exe-
cute cell killing. This mild and synergistic approach has
the advantage over conventional therapy that it targets
To be suitable, a pro-drug of this kind should fulfil two
basic requirements: (1) ready incorporation into cellular
DNA, ideally with some selectivity towards cancer cells;
(2) having UV maximum absorption substantially
remote from 260 nm. As a part of our ongoing work, we
have focused initially on thymidine analogues. Incor-
poration of thymidine analogues into cellular DNA
requires their phosphorylation by the enzyme thymidine
kinase (TK). TK activity is up-regulated in growing cells-
including cancer cells. This can offer a level of selection
against cancer cells. An additional level of targeting
cancer cells would be achieved by selective UVA light
activation and damage of the cells incorporating these
thymidine analogues. Here we describe the first synth-
esis of 4-thio-5-bromo-20-deoxyuridine (S4-BrdU) 3a
and its key properties.
The synthesis of S4-BrdU was carried out by a modifi-
cation of the protocol we developed4 for 4-thiothymi-
dine. 5-Bromo-20-deoxyuridine (BrdU) 1a (Scheme 1)
was reacted with trimethylchlorosilane to form 30,50-O-
bis-(trimethylsilyl)-5-bromo-20-deoxyuridine 1b5 which
was then reacted with tri-triazolo oxyphosphorus pro-
duced in situ from triazole and POCl3 to form 4-triazolo
derivative 2b.7 2b was then transformed by a treatment
with thioacetic acid into a 4-thio derivative 3b which,
after acidic deprotection, produced the title compound
Abbreviations: S4-T, 4-thiothymine; S4-TdR, 4-thiothymidine; dU, 20-
deoxyuridine; BrdU, 5-bromo-20-deoxyuridine; S4-BrdU, 4-thio-5-
bromo-20-deoxyuridine; TdR, thymidine; ELISA, enzyme linked
immuno sorbent assay; TK, thymidine kinase; UVA, UV light at 320–
400 nm; XP, Xeroderma Pigmentosum; mBB, monobromobimane.
* Corresponding author. Tel.: +44-1908-654-064; fax: +44-1908-858-
327; e-mail: y.z.xu@open.ac.uk
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.11.069

996
Y.-Z. Xu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 995–997
Scheme 1. Synthetic route to 4-thio-5-bromodeoxyuridine (S4-BrdU).
Reagents: (i) Cl–Si(Me)3/TEA; (ii) Triazole/POCl3/TEA; (iii) CH3COSH;
(iv) 1N HCl.
3a.8 The overall yield from 1 to 3a (4 steps) was 30%.
To the best of our knowledge, this is the first report of
the synthesis of 4-thio-5-bromo-20-deoxyuridine. Inter-
estingly, a great number of 5-substituted-20-deoxyur-
idine analogues have been synthesised and tested for
anti-viral or anti-cancer activity.9 However, to our sur-
prise, only very few 4-thio-5-substituted-20-deoxyuridine
analogues have been prepared10 and little biological
exploitation carried out.11
Figure 1. Top panel is the UV spectra of 3a (50 mM in water) and 4
(50 mM in CH3CN); Low panel is the fluorescence spectra of 3a and
4 (excited at 380 nm). 3a has little fluorescent emission.
The current report of 4-thio-5-bromo-20-deoxyuridine
should provide a good model for chemical synthesis and
biological evaluation of other 4-thio-5-substituted-20-
deoxyuridine analogues, particularly 4-thio-5-iodo-20-
deoxyuridine.
Table 1. Chemical structures and maximum UV absorption of dU,
TdR, BrdU and their thio-analogues
2. Properties of S4-BrdU
2.1. Cellular incorporation
X
Y
lmax (nm)
Initial experiments indicate that the synthetic S4-BrdU
can be incorporated into DNA by cultured human
tumour cells. This incorporation is via TK activation to
form its corresponding nucleotide and triphosphate,
offering a potential for targeting TK-rich cells (such as
tumour cells). Interestingly unlike S4-TdR, S4-BrdU in
DNA can be recognised by the monoclonal antibodies
used against DNA containing BrdU.12 This recognition
can be ascribed to the presence of a bromo group in S4-
BrdU and would endow S4-BrdU with a unique prop-
erty for its rapid detection in DNA by a simple ELISA
procedure. Furthermore, like S4-dT, S4-BrdU was not
toxic to cells over a range of concentrations from 0–300
mM. Several cell lines were tested and all of the cell
lines13 proliferated normally in the presence of up to 300
mM S4-BrdU.
O
O
O
S
S
S
H
CH3
Br
H
CH3
Br
262
267
280
332
335
340
2.3. UVA irradiation and its mechanisms
S4-BrdU is not cytotoxic on its own but sensitises cells
to killing by a subsequent exposure to low doses of
UVA light. UVA irradiation (wavelength 320–400 nm,
lmax 365 nm) has been used to cause lethal DNA
damage selectively in cells that have incorporated S4-
BrdU into their DNA. Previously we suggested3 that the
extreme sensitivity of XP cells (Xeroderma Pigmentosum
cells defective in nucleotide excision repair enzymes) to
S4T-dR/UVA reflects the production of the 6-4 pyr-
imidine–pyrimidone-like DNA lesions that have been
identified among the products of UVA irradiation of
oligonucleotides containing S4-TdR.14 However, our
observation that XP cells did not show hyper-sensitivity
to S4-BrdU/UVA treatment indicates S4-BrdU may act
by a mechanism different from that of S4-TdR. This is
of potential use regarding effective DNA damage. It is
well documented that BrdU (the oxy analogue of
S4-BrdU) can be readily incorporated into cell DNA
and its sensitivity to UVA light is dependent on the
2.2. Photo absorption
The synthesised 3a (S4-BrdU) absorbs maximally at 340
nm (Fig. 1). This is substantially away from the absorp-
tion range (260–270 nm) of normal DNA nucleosides.
As anticipated, the enhancement of conjugation leads to
absorption by the nucleoside at longer wavelength
(Table 1). Since the maximum wavelength for S4-BrdU
is further away, UVA activation of S4-BrdU should
inflict less damage to normal DNA and enhance the
selectivity of cell-targeting.
presence of a photo-sensitiser (i.e., Hoechst dye).15
A

Y.-Z. Xu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 995–997
997
combination of BrdU and Hoechst dye dramatically
sensitises cells to the lethal effect of UVA16 as the bound
dye can absorb UVA photons and provide the energy
for dissociation of the bromine atom,16 which in turn
promotes DNA chain breakage. Our synthetic S4-BrdU
itself can act as the primary chromophore and obviate
the requirement for the dye. In addition, the bromo
group in S4-BrdU could dissociate itself from its excited
state as in the case of BrdU/dye to cause DNA chain
breakage. Therefore S4-BrdU/UVA treatment would
have an alternative avenue to execute cell death.
product was directly used for the next step. Pure product
was obtained by a silica gel column and characterised by
1
NMR. H NMR (300 MHz, CDCl3) d: 1.93–2.17(m, 2H,
20-H and 200-H), 3.56ꢀ3.74 (m, 2H, 50-H), 3.84 (m, 1H, 40-
H), 4.22 (m, 1H, 30-H), 6.15 (t, J=6.4, 1H, 10-H), 8.12 (q,
J=13.1, 1H, 6-H), 8.42 (s, 1H, NH).
6. Alauddin, M. M.; Conti, P. S. Tetrahedron 1994, 50, 1699.
7. Xu, Y.-Z.; Zheng, Q.; Swann, P. F. Nucleic Acids Res.
1990, 18, 4061.
8. 3b in THF was acidified with 1 N HCl (pH 3) and the
reaction monitored by TLC. The desilylated product 3a
was purified by a silica gel column to give pure 4-thio-5-
bromo-20-deoxyuridine. 1H NMR (300 MHz, DMSO-d6)
d: 2.17–2.25 (m, 2H, 20-H and 200-H), 3.56ꢀ3.65 (m, 2H,
50-H), 4.09 (m, 1H, 40-H), 4.22 (m, 1H, 30-H), 5.21 (t,
J=4.8, 1H, 50-OH), 5.26 (d, J=4.4, 1H, 30-OH), 6.02 (t,
J=6.0, 1H, 10-H), 8.52 (s, 1H, 6-H), 13.08 (s, 1H, NH).
UV: lmax=340 nm (e=17ꢁ103). HRMS: m/z 322.9699
([M++1]) (calcd for C9H12O4N2Br1S1: 322.9696).
9. De Clercq, E. Med. Res. Rev. 2002, 22, 531. Perigaud, C.;
Gosselin, G.; Imbach, J. L. Nucleosides Nucleotides 1992,
11, 903 (more references therein).
10. 4-ThiodU; Coleman, R. S.; Kesicki, E. A. J. Am. Chem. Soc.
1994, 116, 11636. 4-Thio-5-methyl-dU (i.e., S4-dT);4 4-
thio-5-ethyl-dU; Kulikowski, T.; Shugar, D. J. Med.
Chem. 1974, 17, 269. 4-Thio-5-fluoro-dU;11 4-thio-5,6-
dihydro-dU; Peyrane, F.; Fourrey, J. L.; Clivio, P. Chem.
Comm. 2003, 736.
2.4. Chemical properties
S4-BrdU retains the strong nucleophilicity of the sulfur
atom and can be readily functionalised with SN2-type
electrophiles. For instance, monobromobimane (mBB),17
a thiol-reactive fluorophore for tagging cysteine in pep-
tides, reacts with S4-BrdU to produce a stable, highly
fluorescent compound 4 (Fig. 1). The reaction is con-
firmed by spectral methods18 to take place at the S4-
position of the nucleoside. The fluorescent functional-
isation of S4-BrdU is thiol-specific and almost quant-
itative even in the presence of a large excess of thymidine
or BrdU. The marked differences in fluorescence spectra
of S4-BrdU (3a) and fluorophore-conjugated-S4-BrdU
(4) are demonstrated in Figure 1. The extremely high
fluorescence of 4 can offer a sensitive monitoring of
minute amounts of S4-BrdU in treated cells or patients.
11. Bretner, M.; Felczak, K.; Dzik, J. M.; Golos, B.; Rode,
W.; Drabikowska, A.; Poznanski, J.; Krawiec, K.; Piasek,
A.; Shugar, D.; Kulikowski, T. Nucleosides Nucleotides
1997, 16, 1295.
12. MRC5VA cells were grown in a medium containing dia-
lysed fetal calf serum and no added nucleoside for 72 h,
then 100 mM S4-BrdU for 72 h. 104 cells were plated per
well of a 96 well plate and the incorporated nucleoside
detected using a Cell Proliferation ELISA, BrdU (colori-
metric) kit (from Roche).
13. The tested cell lines include SV-40 transformed normal
MRC5VA fibroblasts, the nuclear excision repair-defec-
tive XPA fibroblast XP12RO and its hMSH2-defective
variant XP12ROB4.
14. Warren, M. A.; Murray, J. B.; Connolly, B. A. J. Mol.
Biol. 1998, 279, 89.
15. Rosenstein, B. S.; Setlow, R. B.; Ahmed, F. E. Photo-
chem. Photobiol. 1980, 31, 215.
16. Limoli, C. L.; Ward, J. F. Radiat. Res. 1994, 138, 312.
17. Kosower, E. M.; Pazhenchevsky, B.; Dodiuk, H.; Kanety,
H.; Faust, D. J. Org. Chem. 1981, 46, 891.
3. Summary
We have made the first synthesis of S4-BrdU and
defined some of its important characteristics. Further
exploration of its other biological and clinic use is
clearly warranted. The synthesis of S4-BrdU reported
here should offer a straightforward route to other 4-
thio-5-substituted pyrimidine nucleosides and provide
the key precursor for preparation of DNA oligomers
containing the modified base. The availability of such
modified DNA will facilitate the examination of its
coding properties and possible mutagenicity. These
studies are now under way.
18. 4-Thio-5-bromo-20-deoxyuridine (61.54 mg, 0.191 mmol) was
dissolved in 0.5 mL of 0.5 M phosphate buffer (pH=8) and
0.5 mL of CH3CN. The solution was treated with mono-
bromobimane (50 mg, 0.185 mmol) and stirred at rt for 5 h,
by which period, TLC showed that the starting material was
completely converted into a new compound with higher
Rf [in CH2Cl2:CH3OH (9:1)]. The solution was then mixed
with 10 mL of saturated aqueous NaCl and extracted with
EtOAc (3ꢁ25 mL). The organic extracts were combined,
dried over Na2SO4, and concentrated under vacuum. Pur-
ification of the residue by column chromatography in silica
gel (eluting with 5% MeOH/CH2Cl2) afforded compound
4 (30 mg, 31%). Data: 1H NMR (300 MHz, DMSO-d6) d:
1.72 (s, 3H, CH3), 1.89 (s, 3H, CH3), 2.45 (s, 3H, CH3), 3.53–
3.69 (m, 2H, 50-H), 3.85 (m, 1H, 40-H), 4.21 (m, 1H, 30-H),
4.51 (s, 2H, ꢂSCH2), 5.24 (m, 2H, 30 and 50, ꢂOH), 5.98 (t,
J=5.85 and 5.88, 1H, 10-H), 8.62 (s, 1H, 6-H); UV: lmax=
263.8 (e=12ꢁ103), 328 (e=9ꢁ103). HRMS: m/z 530.0707
([M+NH4]+) (calcd for C19H25O6N5SBr: 530.0703);
Fluorescence: Emission=440 nm while excited at 380 nm).
Acknowledgements
The work is partly supported by Cancer Research UK
and the Open University. Dr. Xu is grateful to Prof.
David Shuker for valuable discussion and to Graham
Jeffs for excellent technical support. We are grateful to
EPSRC national mass spectrometry service centre for
accurate mass measurements.
References and notes
1. Friedberg, E. C.; Walker, G. C.; Siede, W. DNA Repair
andMutation ; ASM Press: Washington DC, 1995.
2. Massey, A.; Xu, Y.-Z.; Karran, P. DNA Repair 2002, 1, 27 5.
3. Massey, A.; Xu, Y.-Z.; Karran, P. Curr. Biol. 2001, 11, 1142.
4. Xu, Y.-Z.; Zheng, Q.; Swann, P. F. Tetrahedron Lett.
1991, 32, 2817.
5. The protection of 30,50-hydroxyl groups by silylation was
carried out according to a published protocol.6 The