exo-N-[2-(4-Azido-2,3,5,6-tetrafluorobenzamido)ethyl]-dC: a novel intermediate in the synthesis of dCTP derivatives for photoaffinity labelling

Article abstract of DOI:10.1016/j.tetlet.2007.12.083

An alternative route for the synthesis of a photoaffinity labelling (PAL) dCTP derivative is reported. This method involves the intermediacy of exo-N-[2-(4-azido-2,3,5,6-tetrafluorobenzamido)ethyl]-dC. The latter is prepared from the coupling of known N-(2-aminoethyl)-4-azido-2,3,5,6-tetrafluorobenzamide, prepared in an improved three-step sequence, with an activated 4-triazolyl derivative of dU, followed by deprotection. ¹⁹F NMR spectroscopy proved extremely useful in following the synthetic transformations, and enabled control of any adventitious reduction of the azides.

Full text of DOI:10.1016/j.tetlet.2007.12.083

Available online at www.sciencedirect.com
Tetrahedron Letters 49 (2008) 1336–1339
exo-N-[2-(4-Azido-2,3,5,6-tetrafluorobenzamido)ethyl]-dC:
a novel intermediate in the synthesis of dCTP derivatives
for photoaffinity labelling
Crina Cismas , Thanasis Gimisis *
ß
Organic Chemistry Laboratory, Department of Chemistry, University of Athens, 157 84 Athens, Greece
Received 17 November 2007; revised 14 December 2007; accepted 18 December 2007
Available online 23 December 2007
Abstract
An alternative route for the synthesis of a photoaffinity labelling (PAL) dCTP derivative is reported. This method involves the inter-
mediacy of exo-N-[2-(4-azido-2,3,5,6-tetrafluorobenzamido)ethyl]-dC. The latter is prepared from the coupling of known N-(2-amino-
ethyl)-4-azido-2,3,5,6-tetrafluorobenzamide, prepared in an improved three-step sequence, with an activated 4-triazolyl derivative of
dU, followed by deprotection. 19F NMR spectroscopy proved extremely useful in following the synthetic transformations, and enabled
control of any adventitious reduction of the azides.
Ó 2007 Elsevier Ltd. All rights reserved.
Keywords: Photoaffinity labelling; dCTP Derivatives; 4-Azido-2,3,5,6-tetrafluorobenzamide derivatives; Synthesis; 19F NMR
Modelling clustered DNA damage and repair requires
oligonucleotides with DNA lesions in close proximity to
each other, either in tandem, or on opposing DNA strands.
These oligonucleotides act as model systems for clustered
DNA damaged sites.1 It has now been established that
the overall repair of clustered sites is retarded when com-
pared to the repair of the constituent isolated lesions, and
the mechanism of repair is dependent on the lesions within
the cluster, the interlesion distance and the orientation of
the lesions to each other. However, little is known about
which base excision repair proteins play the greatest role
in the processing of clustered damage sites.
Photoaffinity labelling is a technique for marking the
binding sites of proteins.2 Perfluorophenyl azides have pro-
ven to be versatile labelling reagents and have already been
used for the selective labelling of DNA binding proteins,3–5
photoaffinity modification of human ribosome,6,7 function-
alization of bifunctional chelating agents for c-imaging and
radiotherapy8 and, more recently, biotin-tagged photo-
affinity probes.9,10 A biological ligand modified with a
photoactive moiety, such as the azido group, serves as a
precursor of highly reactive intermediate nitrenes generated
upon photolysis and capable of forming covalent crosslink
bonds. By incorporating a photoaffinity labelling derivative
of dCTP into the repair gap arising from the processing of
a clustered damaged site, it is possible to capture base exci-
sion repair proteins involved in the repair of the cluster and
compare them with the proteins involved in the repair of
single lesions. In addition, capturing proteins at different
times during the repair may provide insight into the
dynamics of the repair processes.
Therefore, we were interested in the synthesis of the
photoaffinity labelling (PAL) derivative of dCTP (1). The
need to develop a new synthetic strategy for this compound
emerged from the problems encountered when following
the literature procedure (Scheme 1, route a).11 The
sequence of Scheme 1 could be reproduced by extending
the reaction times (3 days vs 24 h), but, in our hands, the
yields were unsatisfactory. Convergent to this, our attempt
*
Corresponding author. Tel.: +30 210 727 4928; fax: +30 210 727 4761.
E-mail address: gimisis@chem.uoa.gr (T. Gimisis).
Present address: Organic Chemistry Department, Babes-Bolyai Uni-
versity, 11 Arany Janos str., 400028 Cluj-Napoca, Romania.
0040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.12.083

C. Cismaßs, T. Gimisis / Tetrahedron Letters 49 (2008) 1336–1339
1337
route a
F
F
N3
F
H
N
HN
N
NH2
N
O
F
4 M+
O
4 M+
O
N
O
N
O
O
O
O
O
-O P O P O P O
O- O- O-
-O P O P O P O
O- O- O-
O
O
OH
OH
dCTP
1
route b
route c
F
F
F
F
N3
F
F
N3
H
H
N
N
H2N
HN
O
N
F
O
O
3
F
N
NH
O
N
O
HO
HO
O
O
3
+
2
dU
OH
OH
Scheme 1. Previous synthetic routes and current retrosynthetic analysis of PAL-dCTP (1).
for a one-step coupling to obtain the target derivative
(Scheme 1, route b) met with little success because of the
lability of the azide group in the reaction media. We found
that, under the reaction conditions, the perfluoroazide
derivative 3 reduced to the corresponding perfluoroaniline.
A retrosynthetic strategy for the synthesis of the target
compound 1, described in this Letter (route c), is also pre-
sented in Scheme 1. It relies on the intermediacy of a stable
product such as 2, whose protected precursor may be puri-
fied through normal phase chromatographic methods. A
potential problem to overcome in this scheme would be
the relative ease of reduction of the azide group, which
has been mainly accomplished as presented below.
For the synthesis of compounds 6 and 3 (Scheme 2), we
started from procedures described in the literature with
essential changes that improved yield and selectivity.
Diazotation of the commercially available 4-amino-
2,3,5,6-tetrafluorobenzoic acid with sodium nitrite, followed
by displacement with NaN3 was performed in trifluoro-
acetic acid12 and afforded the aromatic azide in excellent
yield after increasing the reaction time (Scheme 2).
N-Hydroxy-succinimidyl ester 613 was isolated in 90% yield
with no traces of dicyclohexylurea by performing the reac-
tion in acetonitrile instead of dichloromethane, followed by
a standard ethyl acetate work-up. Monoacylation of ethy-
lenediamine with derivative 6 gave a hard-to-separate 2.5:1
mixture of azide 3 and the corresponding reduced amine
when following the literature procedure.8 We overcame this
problem by working at 0–5 °C, and selectively obtained the
desired product, in an improved yield (Scheme 2).
A clear differentiation between those derivatives bearing
the azide function and those with amine could be observed
in 19F NMR spectra. The signals corresponding to the two
fluorine atoms next to azide (in the range À151 to
À153 ppm) are at least 10 ppm shifted compared with the
signals of the fluorine atoms next to a reduced amine group
(in the range À164 to À166 ppm, Table 1). It is interesting
to note that, although they are particularly useful, no 19F
NMR data had been previously reported for intermediates
5, 6 or 3.
The TBDMS-protected deoxyuridine 714 was chosen as
starting material for the synthesis of the nucleoside build-
ing-block. The C-4 position on uracil was activated for
nucleophilic substitution by conversion into either the
triazolylpyrimidinone 815 or the mesitylenesulfonyl deriva-
F
F
F
F
F
NH2
F
F
N3
F
1
tive 916 (Scheme 3, H and 13C NMR spectra are reported
a
b
HO
HO
in the Supplementary data).
We next examined the method to obtain the new
TBDMS-protected derivative 10 (see Supplementary data)
O
O
4
5
F
F
F
Table 1
F
N3
F
N3
F
19F NMR shifts for compounds 4, 5, 6 and 3 (in CDCl3)
O
H
N
c
O
N
F
H2N
Compd.
Fa
Fb
O
F
O
O
6
3
4
5
6
3
À162.60
À151.08
À150.26
À150.99
À139.14
À137.31
À133.97
À141.45
Scheme 2. Reagents and conditions: (a) (i) 2 equiv NaNO2, TFA, 1 h; (ii)
10 equiv NaN3, Et2O, 2 h (95%); (b) 1 equiv NHS, 1.02 equiv DCC,
CH3CN, overnight (90%); (c) 15 equiv ethylenediamine, CH3CN, 0–5°C,
1 h (75%).
Fa represents the fluorine atoms ortho to the azide (or amino) groups.

1338
C. Cismaßs, T. Gimisis / Tetrahedron Letters 49 (2008) 1336–1339
Scheme 4. Reagents and conditions: (a) (HF)n/Py, THF, 2.5 h (50%), (b)
(i) 1.3 equiv POCl3, 1.5 equiv proton-sponge, 43 equiv triethyl phosphate,
3 h, 0 °C; (ii) 6.1 equiv Bu3N, 2.4 equiv TBAP, 30 min. (15%).
For the cleavage of the TBS-protecting groups we found
that the perfluorinated aromatic azide was not compatible
with the more commonly used tetrabutylammonium fluo-
ride (TBAF) reagent.18,19 We isolated the new derivative
2 (Scheme 4) by using the milder (HF)n/Py reagent20 and
the 19F NMR spectrum showed that the azide function
was stable under these conditions.
The final stage was the two-step, one-pot, triphospho-
rylation procedure in the presence of a proton-sponge21
(Scheme 4), followed by semi-preparative HPLC purifica-
tion of the reaction product. The reaction was followed
by analytical HPLC to ensure the formation of the mono-
phosphate intermediate and its conversion to the corres-
ponding triphosphate.
The spectral data of the final compound 1 are compara-
ble with those reported (1H, 31P NMR),11 while the 19F
NMR spectrum confirms the presence of the azide group.
The small differences in the proton NMR spectrum are
probably due to the different counter-ion, that is, Na+ ver-
sus Et3NH+ in our case.
The signals corresponding to the a and b phosphorus
atoms in the 31P NMR (pH 4) have similar shifts to the
ones reported.11 The c phosphorus exhibited a 5 ppm high
field shift. This is a well-documented shift,22 attributable to
the pH difference between the published (pH not reported)
and our (pH 4) 31P NMR spectra.
The expected two sets of signals are clearly shown in the
19F NMR spectrum, with one at À144 ppm corresponding
to the fluorine atoms (Fb) ortho to the amidic group and
the other at À151 ppm corresponding to the fluorine atoms
(Fa) next to the azide group.
Scheme 3. Reagents and conditions: (a) 4.4 equiv TBDMS–Cl, 8.8 equiv
imidazole, dry DMF, overnight (96%), (b) 2.2 equiv MsCl, 0.25 equiv
DMAP, Et3N, anhyd DMF, 24 h (35%); (c) 1.5 equiv 3, THF, 20 h (65%),
(d) 2 equiv 3, cat. K2CO3, CH2Cl2, 6 h (40%).
by the nucleophilic displacement of 8 or 9 with perfluoro-
azide 3. The reaction of the mesitylene derivative 9 in the
presence of a catalytic amount of K2CO3 yielded product
10 in moderate yield (40 %), even in the presence of a large
excess of amine (10 equiv). A comparable yield was
obtained when the triazolyl derivative 8 was reacted with
amine 3, in dry pyridine, but changing the solvent to
tetrahydrofuran significantly increased the yield (65 %).
1
The H NMR spectrum of compound 10 exhibited two
sets of signals in CDCl3, while just one set of sharp signals
was observed in DMSO-d6 solution. This indicated that 10
exists in a tautomeric equilibrium between 4-amino and the
E and Z isomers of 4-imino forms in CDCl3. A similar
behaviour as well as isomeric ratio (3:1) was previously
reported for a 4-amino-substituted pyrimidinone.17 Inter-
estingly, this pattern was better observed in the 19F
NMR spectra and the two sets of signals were registered
in the CDCl3 spectrum in a 3:1 ratio (Fig. 1).
Considering the above analyses, it was surprising to find
that the ESI-MS exhibited molecular peak only for the
reduced azide. Nevertheless, it has been reported that in
the presence of a phosphoryl group, ESI induces the trans-
formation of an azide to the corresponding amine.23 This
effect was not observed in the ESI-MS spectra of any of
the other azides (compounds 3, 5, 6, 10 and 2), in this
study.
In conclusion, we have developed an alternative route
for the synthesis of PAL-dCTP derivative 1. This involves
the intermediacy of protected and free PAL-dC (com-
pounds 10 and 2, respectively), two new and easy-to-handle
intermediates. 19F NMR spectroscopy has proven extre-
mely helpful in following the synthetic transformations
Fig. 1. Compound 10 19F NMR spectra in DMSO-d6 (A) and CDCl3 (B);
Fa represents the fluorine atoms ortho to the azide group.

C. Cismaßs, T. Gimisis / Tetrahedron Letters 49 (2008) 1336–1339
1339
5. Lavrik, O. I.; Prasad, R.; Beard, W. A.; Safronov, I. V.; Dobrikov, M.
I.; Srivastava, D. K.; Shishkin, G. V.; Wood, T. G.; Wilson, S. H. J.
Biol. Chem. 1996, 271, 21891–21897.
6. Demeshkina, N. A.; Laletina, E. S.; Meschaninova, M. I.; Repkova,
M. N.; Ven’yaminova, A. G.; Graifer, D. M.; Karpova, G. G. Mol.
Biol. 2003, 37, 132–139.
and in avoiding adventitious reduction of the azide to the
corresponding amine. Compound 2 is useful for the prepa-
ration of the corresponding phosphoramidites and after
insertion to PAL-oligonucleotides, together with 1, in
photoaffinity labelling studies.
7. Demeshkina, N.; Laletina, E.; Meschaninova, M.; Ven’yaminova, A.;
Graifer, D.; Karpova, G. Biochim. Biophys. Acta 2003, 1627, 39–
46.
Acknowledgements
8. Rajagopalan, R.; Kuntz, R. R.; Sharma, U.; Volkert, W. A.;
Pandurangi, R. J. Org. Chem. 2002, 67, 6748–6757.
9. Han, S.-Y.; Park, S.-S.; Lee, W. G.; Min, Y. K.; Kim, B. T. Bioorg.
Med. Chem. Lett. 2006, 16, 129–133.
10. Han, S.-Y.; Choi, S. H.; Kim, M. H.; Lee, W. G.; Kim, S. H.; Min, Y.
K.; Kim, B. T. Tetrahedron Lett. 2006, 47, 2915–2919.
11. Wlassoff, W. A.; Dobrikov, M. I.; Safronov, I. V.; Dudko, R. Y.;
Bogachev, V. S.; Kandaurova, V. V.; Shishkin, G. V.; Dymshits, G.
M.; Lavrik, O. I. Bioconjugate Chem. 1995, 6, 352–360.
12. Pinney, K. G.; Katzenellenbogen, J. A. J. Org. Chem. 1991, 56(9),
3125–3133.
13. Keana, J. F. W.; Cai, S. X. J. Org. Chem. 1990, 55, 3640–3647.
14. Zhang, W.; Rieger, R.; Iden, C.; Johnson, F. Chem. Res. Toxicol.
1995, 8, 148–156.
15. Quinn, J. R.; Zimmerman, S. C. J. Org. Chem. 2005, 70, 7459–
7467.
The project was co-funded by the European Social Fund
and National Resources (EPEAEK II, PYTHAGORAS).
C.C. acknowledges financial aid by the European Commu-
nities’ Marie Curie Research Training Network (Contract
MRTN-CT-2003-505086 [CLUSTOXDNA]).
Supplementary data
Experimental procedures, 1H NMR and 13C NMR spec-
tra of 9, 19F NMR spectra of 1, 2 and 10, HPLC trace and
31P NMR spectrum of 1 and LR ESI-MS spectra of 1, 2
and 10. Supplementary data associated with this article
can be found, in the online version, at doi:10.1016/
j.tetlet.2007.12.083.
16. Bischofberger, N. Tetrahedron Lett. 1987, 28, 2821–2824.
17. Katoh, A.; Hida, Y.; Kamitani, J.; Ohkanda, J. J. Chem. Soc., Dalton
Trans. 1998, 3859–3864.
References and notes
18. Golinski, M.; Heine, M.; Watt, D. S. Tetrahedron Lett. 1991, 32,
1553–1556.
19. Chehade, K. A. H.; Spielmann, H. P. J. Org. Chem. 2000, 65, 4949–
4953.
20. Huang, Y.; Torres, M. C.; Iden, C. R.; Johnson, F. Bioorg. Chem.
2003, 31, 136–148.
1. Shikazono, N.; Pearson, C.; O’Neill, P.; Thacker, J. Nucleic Acids
Res. 2006, 34, 3722–3730.
2. For an excellent introduction to the topic see: Soundararajan, N.; Liu,
S. H.; Soundararajan, S.; Platz, M. S. Bioconjugate Chem. 1993, 4,
256–261.
21. Kovacs, T.; Otvos, L. Tetrahedron Lett. 1988, 29, 4525–4528.
22. Yoza, N.; Ueda, N.; Nakashima, S. Fresenius J. Anal. Chem. 1994,
348, 633–638.
23. Xiao, Q.; Ju, Y.; Yang, X.; Zhao, Y.-F. Rapid Commun. Mass
Spectrom. 2003, 17, 1405–1410.
3. Dezhurov, S. V.; Khodyreva, S. N.; Plekhanova, E. S.; Lavrik, O. I.
Bioconjugate Chem. 2005, 16, 215–222.
4. Lavrik Olga, I.; Kolpashchikov Dmitry, M.; Prasad, R.; Sobol
Robert, W.; Wilson Samuel, H. Nucleic Acids Res. 2002, 30, 7371–
7373.