General method for the synthesis of 2′-O-carboranyl-nucleosides

Article abstract of DOI:10.1016/j.tetlet.2005.04.046

The carboranyl cage is a new modifying entity for nucleosides, DNA oligonucleotides, and other biomolecules. Herein, the first reliable method for the synthesis of nucleosides modified with a carborane cluster at the 2′-position is described.

Full text of DOI:10.1016/j.tetlet.2005.04.046

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 3969–3972
General method for the synthesis of 20-O-carboranyl-nucleosides
Błaej Wojtczak,a Andrey Semenyuk,b Agnieszka B. Olejniczak,a Marek Kwiatkowskib
and Zbigniew J. Lesnikowskia,*
aCenter of Medical Biology PAS, Laboratory of Molecular Virology and Biological Chemistry, Lodz 93-232, Poland
bDepartment of Genetics and Pathology, Rudbecklaboratory, Uppsala University, Uppsala 751 85, Sweden
Received 14 February 2005; revised 5 April 2005; accepted 12 April 2005
Abstract—The carboranyl cage is a new modifying entity for nucleosides, DNA oligonucleotides, and other biomolecules. Herein,
the first reliable method for the synthesis of nucleosides modified with a carborane cluster at the 20-position is described.
ꢀ 2005 Elsevier Ltd. All rights reserved.
The initial stimulus for the design and synthesis of bor-
on-containing nucleosides was their potential applica-
tion as boron-rich carriers for boron neutron capture
therapy (BNCT).1–4 More recently, it has been demon-
strated that carboranes can be used to enhance hydro-
phobic interactions between pharmaceuticals and their
receptors and to increase the in vivo stability of com-
pounds that are normally rapidly metabolized. Carbor-
ane modification may also facilitate cellular uptake,
increasing lipophilicity and penetration through cellular
membranes of compounds bearing carboranes. Several
antiviral nucleosides modified with carborane cages
have been synthesized and their biological properties
studied.2 Carborane clusters have also been explored
as new and versatile modifying entities for DNA oligo-
nucleotides.5,6 Thus, carborane-containing nucleosides
are desirable precursors for the preparation of modified
DNA oligomers. Nearly all the carborane-modified
nucleosides described so far belong to the pyrimidine
series and are thymidine or uridine derivatives contain-
ing the modification confined within the nucleic
base.3,7–9
by Anisuzzaman for the synthesis of 30-O- and 20-O-
(o-carboran-1-yl)methyluridine 4.10 The method is based
on the reaction of 20,30-O-(dibutylstannylene)uridine
with 3-bromopropyne, acetylation of the resulting mix-
ture of 20- and 30-O-(3-propynyl)uridines, and then reac-
tion with a bis(acetonitrile)decaborane adduct yielding a
difficult to separate mixture of 30,50-di-O-acetyl-20-O-(o-
carboran-1-yl)methyluridine and 20,50-di-O-acetyl-30-O-
(o-carboran-1-yl)methyluridine. The applicability of this
method of the synthesis of carborane-modified purine
nucleosides was not reported.
Since RNA and DNA oligomers modified at the 20-posi-
tion have proven to be more valuable as biopharma-
ceuticals and molecular probes than those bearing
modification at the 30-position, finding a method that
produces only one isomer is desirable.11,12 Our initial
approach to the site-specific synthesis of 20-O-(o-carbo-
ran-1-yl)methyluridine 4 (Scheme 1) was based on
the method for selective preparation of 20-O-pro-
pargylnucleosides.13,14a
The propargyl group was introduced selectively at
the 20-position of compound 114b using 1 equiv of
propargyl bromide and 2-tert-butylimino-2-diethylamino-
1,3-dimethyl-perhydro-1,3,2-diazaphosphorine (BEMP)
providing 2. However, we found that during the
subsequent reaction, the tetraisopropyldisiloxane-1,3-
diyl protection was unstable in the presence of
the bis(acetonitrile)decaborane adduct and therefore
the disilyl protection was removed and replaced by
acetyl groups. Further reaction of the resultant
The only method for attachment of a carborane group
to the 20-position of a nucleoside unit is that described
Keywords: Carborane; Nucleoside; Pharmacophore; Antiviral; Boron
neutron capture therapy.
*
Corresponding author. Tel.: +48 42 6771249; fax: +48 42 6771230;
e-mail: zlesnik@cbm.pan.pl
0040-4039/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.04.046

3970
B. Wojtczak et al. / Tetrahedron Letters 46 (2005) 3969–3972
Scheme 1. Reagents and conditions: (i) Propargyl bromide, BEMP; (ii) TBAF/THF; (iii) Ac2O/Py; (iv) B10H14ÆCH3CN/toluene; (v) NH3aq/CH3CN.
Scheme 2. Reagents and conditions: (i) DMSO/AcOH/Ac2O; (ii) TBABr/1-(3-hydroxypropyl)-para-carborane [5, HO(CH2)3C2B10H11]/CuBr2 in
CH2Cl2; (iii) TBAF/THF; (iv) NH3 in CH3CN/H2O.
3-N-benzoyl-30,50-di-O-acetyl-20-O-propargyluridine
3
Target compounds 8a–d were obtained from 7a–d in a
three-step procedure without isolation or purification
of the intermediate products. First, compounds 7a–d
were reacted with 1-(3-hydroxypropyl)-para-carborane
(5) yielding the fully protected 30,50-O,O-TIPDSi-20-
O-[(para-carboran-1-yl)propyleneoxymethyl]nucleosides.
Next, the disiloxane protection was removed using a
solution of TBAF in THF yielding N-protected 20-
O-[(para-carboran-1-yl)propyleneoxymethyl]nucleosides.
The N-acyl protection was removed using concentrated
aqueous ammonia solution providing 20-O-[(para-carbo-
ran-1-yl)propyleneoxymethyl]nucleosides 8a–d (Scheme
2). Compounds 8a–dꢀ were obtained in yields of 40–
with the bis(acetonitrile)decaborane adduct provided
fully protected 20-O-(o-carboran-1-yl)methyluridine.
Though this method, after deprotection, successfully
provided 4, an approach applying a similar procedure
to other nucleosides failed. Instead of the desired 20-
O-isomer of carborane-modified cytosine, adenosine or
guanosine, darkening of the reaction mixture was ob-
served and a complex mixture of products was detected.
To avoid potential damage to the nucleoside component
during carborane cage formation in the presence of the
dodecaborane complex, we turned our attention to
methods allowing attachment of the modifying unit
bearing a preformed carborane cage. As the modifying
entity, a para-carborane cage was chosen. Though
ortho-carborane derivatives are easily available and less
expensive than their para-counterparts, the closed ortho-
carborane cage decomposes in basic conditions into its
open cage, nido-form which is often associated with
significantly increased toxicity.
ꢀ General procedure for the synthesis of 8a–d: A mixture of solid 30,50-
TIPDSi-20-O-MTM (7) (1 mmol), tetrabutylammonium bromide
(1.1 mmol) and 1-(3-hydroxypropyl)-para-carborane 5 (2.8 mmol)
˚
was dried under vacuum with 4 A molecular sieves, and then
dissolved in dry 1,2-dichlorethane (10 mL). The resultant solution
was stirred at room temperature for 24 h and after this solid CuBr2
(1.1 mmol) was added. After reaction completion (ca. 24 h), the
reaction mixture was partitioned between dichloromethane and satd
sodium bicarbonate solution. The organic phase was separated, dried
over anhydrous magnesium sulfate, and the solvent was evaporated
yielding the crude product. This material, without purification, was
dissolved in THF (10 mL) and treated with TBAF/THF (0.5 M,
6 mL). Removal of the TIPDSi group (40 min) was followed by
addition of pyridine/methyl alcohol/water (3:1:1, 10 mL) and ion
exchange resin Dowex 50W · 8 (pyridinium form) (ca. 1 g wet
weight). After 30 min, the ion exchange resin was filtered off and
washed with pyridine/methyl alcohol/water (3 · 5 mL). The filtrate
and washings were combined and evaporated to dryness. When
appropriate, the remaining base protecting groups were removed by
dissolving the evaporated material in acetonitrile (2 mL) and addition
of (25%) aqueous ammonia solution (5 mL). After 2 h incubation at
55 ꢁC, the solvent was evaporated and the product was dried by
coevaporation with toluene. Crude compound 8 was purified by flash
column chromatography.
The successful approach to the synthesis of 20-O-[(para-
carboran-1-yl)propyleneoxymethyl] derivatives of all
four nucleosides 8a–d is based on the formacetal linkage
formation described by Matteucci et al.15 and modified
by Sawada and Ito.16 The method involves nucleophilic
substitution of the activated methylthiomethyl group in
the fully protected 30,50-O,O-(tetraisopropyldisiloxane-
1,3-diyl)-20-O-methylthiomethyl-nucleosides (7a–d) with
a suitable alcohol bearing the carborane cage (Scheme
2). The key intermediates 7a–d were obtained from the
reaction of N-protected 30,50-O,O-(tetraisopropyl-
disiloxane-1,3-diyl)nucleosides 6a–d with DMSO in a
mixture of acetic acid/acetic anhydride.17 The N-pro-
tected nucleosides 6a–d are easily available and can be
prepared according to the literature procedure from
suitable nucleosides in high yields.13

B. Wojtczak et al. / Tetrahedron Letters 46 (2005) 3969–3972
3971
10
8
70% and were fully characterized by chromatographic
and spectroscopic methods.ꢁ
8a, B = U
Uridine
Introduction of a large, bulky, and hydrophobic carbo-
rane unit to the sugar part of a nucleoside may be
associated with substantial changes of the natural con-
formation of the nucleoside18 that, in the worst scenario,
could prevent the use of these modified compounds in
many biological experiments.
6
4
θ
] [
2
0
The conformation of a nucleoside is determined by the
rotation of the base about the single bond that joins it
to the 10-carbon of the furanose, the pucker of the atoms
in the five-membered furanose ring, the rotation of the
exocyclic CH2OH group, and the rotation of the phos-
phate attached to this group. There are two general ori-
entations of the base: syn and anti, either form is allowed
although the anti conformation is more common physi-
ologically due to both lower steric strain and to the fact
that this conformation is required for nucleic acid struc-
ture. In addition, the anti form is favored for the pyrim-
idine nucleosides; this is due to a steric clash between the
-2
-4
-6
200
220
240
260
280
300
320
340
[nm]
λ
Figure 1. Circular dichroism spectra of 20-O-[(para-carboran-1-
yl)propyleneoxymethyl]uridine 8a and unmodified uridine.
keto oxygen on the C-2 base with the ribose when in the
syn conformation.19 It was therefore important to com-
pare the conformation of the derivatives obtained with
their non-modified counterparts. We decided to use circ-
ular dichroism (CD), since these spectra are most aff-
ected by restricted rotation around the glycosidic bond.
ꢁ Selected data 8a: Yield: 52%; TLC (CH2Cl2/MeOH, 9:1): Rf = 0.4;
UV (96% C2H5OH): kmin = 231.2 nm, 292.75 nm, kmax = 262.9 nm;
1H NMR, 250.131 MHz (CD3OD): d 1.00–3.34 (bm, 10H, BH-p-
carborane), 1.35–1.75 (m, 4H, –OCH2CH2CH2-p-carborane), 3.30–
3.45 (m, 2H, –OCH2CH2CH2-p-carborane), 3.81–4.02 (m, 2H, 2H-
50,500), 4.10–4.29 (m, 1H, H-40), 4.30–4.39 (m, 2H, H-30, H-20), 4.72–
4.79 (m, 2H, –OCH2O–), 5.74 (d, 1H, H-6, JH6–H5 = 8.09 Hz), 5.75 (d,
The CD spectra§ of the 20-O-(o-carboran-1-yl)-modified
nucleosides under analogous conditions were almost
identical in terms of their shape and molecular ellipticity
values with unmodified nucleosides (Fig. 1), with the
exception that a much higher molecular ellipticity of
the maximum at 275 nm was observed for modified
nucleoside 8b than for unmodified cytidine. The CD
measurement suggests, therefore, that the lipophilic
and bulky o-carborane has little effect on the overall
nucleoside conformation if linked through a propylene-
oxymethyl linker to the 20-position of the nucleoside.
1H, H-10, JH1 –H2 = 4.34 Hz), 7.66 (d, 1H, H-5, JH5–H6 = 8.14 Hz),
8.02 (s, NH); FAB-MS (+ve, Gly) 458.1 [M]+ (molecular formula:
C15H30B10N2O7, calculated exact mass = 457.333). IR (KBr): mmax
(BꢀH) = 2607 cmꢀ1. 8b Yield: 71%; TLC (CH2Cl2/MeOH, 9:1):
Rf = 0.12; UV (96% C2H5OH): kmin = 227.9 nm, 254.0 nm, 304.2 nm,
0
0
k
max = 242.0 nm, 273.7 nm; 1H NMR, 250.131 MHz (CD3OD): d 1.0–
3.5 (bm, 10H, BH-p-carborane), 1.35–1.75 (m, 4H, OCH2CH2CH2-p-
carborane), 3.25–3.45 (m, 2H, OCH2CH2CH2-p-carborane), 3.83–
4.04 (m, 3H, H-50,500, H-40), 4.22–4.30 (m, 2H, H-20, H-30), 4.77–4.86
0
0
(m, 2H, OCH2O), 5.75 (s, 1H, H-6), 5.95 (d, 1H, JH1 -H2 = 7.70 Hz),
7.98 (d, 1H, H-5, JH5–H6 = 7.72); FAB-MS (+ve, Gly): 459.2 [M+H]+
(molecular formula: C15H31B10N3O6: calculated exact mass 458.317);
IR (KBr): mmax (BꢀH) = 2607 cmꢀ1. 8c Yield: 39%; TLC (CH2Cl2/
The present approach provides a route to nucleoside
conjugates modified with different types of carborane
cages or other functional groups as long as a suitable
alcohol terminated with the intended functional group
is available. Our method provides an opening for the
synthesis and study of nucleic acids modified with
carborane clusters at desired locations5,12 and of other
biologically important derivatives of nucleosides.
MeOH, 9:1): Rf = 0.26; UV (96% C2H5OH):
kmin = 239.6 nm,
298.6 nm, kmax = 256.9 nm, 313.6 nm; 1H NMR, 250.131 MHz
(CD3OD): d 0.96–3.30 (bm, 10H, BH-p-carborane), 0.97–1.10 (m,
2H, OCH2CH2CH2-p-carborane), 1.34–1.41 (m, 2H, OCH2CH2CH2-
p-carborane), 2.41 (br s, 1H, 1H-C-p-carborane), 2.94–3.08 (m, 2H,
OCH2CH2CH2-p-carborane), 3.72 and 3.86 (dd, 1H, H-50, JH5 –
0
= 2.41, JH5 –H5 = 12.54; dd, 1H, H-500, JH500–H40 = 2.34, JH5 -
0
00
00
H40
= 12.62), 4.16 (d, 1H, 1H-40, JH4 –H5 = 2.24), 4.35–4.45 (m, 1H,
0
00
H50
1H-30), 4.50–4.70 (m, 2H, OCH2O; 1H, H-20), 6.04 (d, 1H, H-10,
JH10–H20 = 6.75), 8.20 (s, 1H, H-2), 8.29 (s, 1H, H-8); 11B {1H}
NMR (CD3OD): d ꢀ10.23 (s, 5B), ꢀ12.67 (s, 5B). FAB-MS (+ve,
Gly) 483.4 [M+] (molecular formula: C17H37B10N5O4, calculated
exact mass: 483.360); IR (KBr): mmax (BꢀH) = 2607 cmꢀ1. 8d Yield:
15–30%; UV (96% C2H5OH): kmin = 224.4 nm, 299.0,
Acknowledgments
This work was supported in part by the Polish Commit-
tee for Scientific Research (KBN), grant PBZ-KBN-059/
T09/08.
k
max = 251.5 nm, 274.6 nm; TLC (CH2Cl2/MeOH, 9:1): Rf = 0.21;
1H NMR (CD3OD): d 0.88–3.5 (bm, 10H, BH-p-carborane), 1.01–
1.50 (m, 4H, CH2CH2CH2-p-carborane), 3.05–3.14 (m, 2H,
OCH2CH2CH2-p-carborane), 3.68–3.86 (m, 2H, 2H-50,500), 4.06–
§ The absorption maxima and minima in circular dichroism (CD)
spectra of modified nucleosides 4a–d are as follows: 4a: kmax = 271
(H = 6.9) and 232 nm (H = ꢀ2.8), kmin = 242 (H = ꢀ3.3) and 218 nm
(H = ꢀ4.1); 4b: kmax = 273 nm (H = 7.4), kmin = 223 (H = ꢀ5.2);
4c: kmax = 292 (H = 1.4), 235 (H = 1.8), 219 nm (H = 2.1),
4.09 (m, 1H, H-40), 4.21 (d, 1H, H-20, JH2 –H1 = 6.19 Hz), 4.35 (dd,
1H, H-30, JH3 –H2 = 4.93, JH3 –H4 = 3.14 Hz), 4.59–4.65 (m, 2H,
0
0
0
0
0
0
OCH2O), 5.89 (d, 1H, H-10, JH1 –H2 = 6.15 Hz), 7.93 (s, 1H, H-8);
FAB-MS (+ve, Gly): 497.1 [M]+, (C16H31B10N5O6: 497.340); IR
(Film): mmax = 2608 cmꢀ1 (BꢀH).
0
0
k
max = 270.0 nm (H = 0.0), 227.0 nm (H = 1.4); 4d:
kmax = 291
(H = 1.1), 218 (H = 2.2), kmin = 245 (H = 0.6).

3972
B. Wojtczak et al. / Tetrahedron Letters 46 (2005) 3969–3972
9. Byun, Y.; Yan, J.; Al-Madhoun, A. S.; Johnsamuel, J.;
Supplementary data
Yang, W.; Barth, R. F.; Eriksson, S.; Tjarks, W. Appl.
Radiat. Isot. 2004, 61, 1125–1130.
10. Anisuzzaman, A. K. M.; Alam, F.; Soloway, A. H.
Polyhedron 1990, 9, 891–892.
IR, CD and MS spectra of 8a–d are available. Supple-
mentary data associated with this article can be found
in the online version, at doi:10.1016/j.tetlet.2005.04.
046.
11. Manoharan, M.; Kawasaki, A. M.; Prakash, T. P.; Fraser,
A. S.; Prhavc, M.; Inamati, G. B.; Casper, M. D.; Cook, P.
D. Nucleos. Nucleot. Nucl. Acids 1999, 18, 1737–1746.
12. Olejniczak, A. B.; Koziolkiewicz, M.; Lesnikowski, Z. J.
Antisense Nucl. Acid Drug Develop. 2002, 12, 79–94.
13. Sproat, B. S. In Methods in Molecular Biology. Protocols
for Oligonucleotides and Analogs. Synthesis and Properties;
Agraval, S., Ed.; Humana: Totowa, NJ, 1993; Vol. 20, pp
115–141.
14. (a) Grotli, M.; Douglas, M.; Eritja, R.; Sproat, B. S.
Tetrahedron 1998, 54, 5899–5914; (b) Zhu, L.; dos Santos,
O.; Seeman, N. C.; Canary, J. W. Nucleos. Nucleot. Nucl.
Acids 2002, 21, 723–735.
References and notes
1. Sjoberg, S.; Carlsson, J.; Ghaneolhosseini, H.; Gedda, L.;
Hartman, T.; Malmquist, J.; Naeslund, C.; Olsson, P.;
Tjarks, W.. J. Neurooncol. 1997, 33, 41–52.
2. Lesnikowski, Z. J.; Shi, J. X.; Schinazi, R. F. J. Organo-
met. Chem. 1999, 581, 156–169.
3. Tjarks, W. J. Organomet. Chem. 2000, 614–615, 37–
47.
4. Barth, R. F.; Yang, W.; Al-Madhoun, A. S.; Johnsamuel,
J.; Byun, Y.; Chandra, S.; Smith, D. R.; Tjarks, W.;
Eriksson, S. Cancer Res. 2004, 64, 6287–6295.
5. Lesnikowski, Z. J. Eur. J. Org. Chem. 2003, 4489–
4500.
6. Olejniczak, A. B.; Plesek, J.; Kriz, O.; Lesnikowski, Z. J.
Angew. Chem., Int. Ed. 2003, 42, 5740–5743.
7. Goudgaon, N. M.; Elkattan, G. F.; Schinazi, R. F.
Nucleos. Nucleot. 1994, 13, 849–880.
15. Matteucci, M.; Lin, K. Y.; Butcher, S.; Moulds, C. J. Am.
Chem. Soc. 1991, 113, 7767–7768.
16. Sawada, D.; Ito, Y. Tetrahedron Lett. 2001, 42, 2501–
2504.
17. Hovinen, J.; Azhayeva, E.; Azhayev, A.; Guzaev, A.;
Lonnberg, H. J. Chem. Soc., Perkin Trans. 1 1994, 211–
217.
18. Miles, D. W.; Townsend, L. B.; Robins, M. J.; Robins, R.
K.; Inskeep, W. H.; Eyring, H. J. Am. Chem. Soc. 1971,
93, 1600–1608.
8. Mourier, N. S.; Eleuteri, A.; Hurwitz, S. J.; Tharnish, P.
M.; Schinazi, R. F. Bioorg. Med. Chem. 1999, 7, 2759–
2766.
19. Seanger, W. Nucleic Acid Structure; Springer:
New York, 1984.