N-Nitration, 15N-Labeling, and N-to-N Linking of Hydroxyl-Silylated Pyrimidine Nucleosides

Full text of DOI:10.1021/jo9615514

J . Org. Chem. 1997, 62, 1547-1549
1547
N-Nitr a tion , 15N-La belin g, a n d N-to-N
Sch em e 1
Lin k in g of Hyd r oxyl-Silyla ted P yr im id in e
Nu cleosid es
Xavier Ariza, J aume Farra`s,* Carme Serra, and
J aume Vilarrasa*
Department of Organic Chemistry, Faculty of Chemistry,
University of Barcelona, 08028 Barcelona, Catalonia, Spain
Received August 9, 1996
Very recently, we reported on a novel method for the
15N-labeling, N-alkylation, and N-amination of nucleoside
derivatives based on the N-nitration of appropriate
precursors with nitronium trifluoroacetate followed by
treatment with 15NH3, alkylamines, or hydrazine (either
unlabeled or 15N-labeled), respectively, which is shown
in Scheme 1.1 Since the hydroxy groups of the sugar
rings underwent concomitant O-nitration, we protected
them or most of them as acetates, benzoates, or isopro-
pylidene acetals, but not as tert-butyldimethylsilyl ethers,
because in preliminary experiments we noted that TB-
DMS groups were partially removed in the nitration
medium.1a,c However, the widespread use of silyl ethers
in the nucleoside field,2 mainly of the TBDMS and 1,1,3,3-
tetraisopropyldisiloxane-1,3-diyl (TIPDS) derivatives,
prompted us to restudy the performance of these protect-
ing groups in connection with the process shown in
Scheme 1. We report here that the above-mentioned
limitation can be overcome by improving the N-nitration
step and, furthermore, that TBDMS and TIPDS groups
are stable under the conditions inherent in the second
step (the ring-opening/ring-closing step1).
Sch em e 2
Resu lts a n d Discu ssion
Uridine and thymidine derivatives 1a -5a were sub-
jected at 0 °C to the action of fresh nitrating solutions
arising from mixing NH4NO3 (n mol per mol) and (CF3-
CO)2O (TFAA, 2n mol per mol)3 in CH2Cl2. Rather than
using moderate excesses of reagents (n ) 1.5-2.0) and
lengthening the reaction times, the highest conversion
and lowest deprotection percentages were obtained by
using large excesses of the nitrating mixture (n ) 4, in
the cases of 1a , 3a , and 5a , Scheme 2). Under these
conditions, the isolated yields of N-nitro compounds 1b,
3b, and 5b were always around 90% (see the Experi-
mental Section). With the partially protected substrates
2a and 4a , in which the alcohol function reacted more
quickly than the imide-like NH, the highest yields of N,O-
dinitro derivatives 2b and 4b (88% and 85%, respectively)
were obtained by employing a larger excess of nitrating
mixture (n ) 6-8).4 The loss of O-silyl groups was hardly
detected. Apparently, with freshly prepared nitrating
mixtures, when there is an excess of mixed anhydride
NO2OCOCF3 in the reaction medium, the silyl ether
cleavage practically does not take place; on the other
hand, when the nitrating mixture is almost exhausted
(i.e., when most anhydrides have disappeared and the
major component of the mixture is TFA)3 the chance of
O-Si bond cleavage and nitro-exchange reactions is
much higher.5
(1) Ariza, X.; Bou, V.; Vilarrasa, J . J . Am. Chem. Soc. 1995, 117,
3665. (b) For an application to the synthesis of [3-15N]-labeled AZT,
see: Ariza, X.; Bou, V.; Vilarrasa, J .; Tereshko, V.; Campos, J . L.
Angew. Chem., Int. Ed. Engl. 1994, 33, 2454. (c) Bou, V. Nuevas
reacciones en el campo de los nucleo´sidos. Ph.D. Thesis, Universitat
de Barcelona, 1992.
(2) For reviews, see: (a) Ueda, T. Chemistry of Nucleosides and
Nucleotides; Townsend, L. B., Ed.; Plenum Press: New York, 1988.
(b) Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis; Wiley: New York, 1991. (c) Kocienski, P. J . Protecting
Groups; Thieme: Stuttgart, 1994. (d) van Look, G.; Simchen, G.;
Heberle, J . Silylating Agents; Fluka Chemie AG: Buchs, 1995.
(3) Romea, P.; Aragone`s, M.; Garcia, J .; Vilarrasa, J . J . Org. Chem.
1991, 56, 7038.
When 1b was treated with 15NH4Cl and K2CO3 in CH3-
CN-H2O at rt for a long time, [3-15N]-labeled 1a was
(4) With smaller excesses (n ) 2-4), uridine 2a afforded mainly the
2′-O-nitro compound, while thymidine 4a gave rise to mixtures of
O-nitro and N,O-dinitro compounds (with n ) 2, 80% of 3′-O-nitro
derivative plus 10% of N,O-dinitro 4b; with n ) 4, 30% of O-nitro plus
62% of 4b).
S0022-3263(96)01551-4 CCC: $14.00 © 1997 American Chemical Society

1548 J . Org. Chem., Vol. 62, No. 5, 1997
Notes
Sch em e 3
obtained in 78% yield. As this yield is similar to those
achieved in related experiments with other standard
protecting groups,1 it appears that the corresponding
O-Si bond is reasonably stable under these conditions,
i.e., the TBDMS group can be reliably used in 15N labeling
experiments.
a good choice for the 15N labeling of nucleosides and
related reactions.
Exp er im en ta l Section
For the general section, see ref 1a. Compounds 1a -5a were
prepared according to standard procedures.2 Coupling constants
(J ) are reported in Hz.
Nevertheless, these conditions are milder than those
often required for reactions with other N-nucleophiles,
where substrates and reagents are heated in the presence
of K2CO3 to accelerate the rate-limiting and nonquanti-
tative ring-closing step.1 Thus, we first checked the
stability of the TBDMS and TIPDS groups of 1a and 5a ,
respectively, in refluxing CH3CN or CH3CN-H2O for 7
h, in the presence of an excess of K2CO3 and/or an
alkylamine (ethylenediamine).6 Since in all experiments
>97% of the starting material was recovered, it was
concluded that the basic media in which the second step
of Scheme 1 is usually carried out are not strong enough
for the cleavage of the above-mentioned O-Si bonds.
Moreover, the reaction of 1b with ethylenediamine (in
the 2:1 molar ratio) and likewise with 1,3-propanedi-
amine, envisaged as a preliminary approach to a new
cross-linking procedure7 for N-nitro-containing oligo-
nucleotide chains, afforded the desired N-to-N “dimers”
6 and 7 (see Scheme 3) in 55% and 57% yields, respec-
tively, after purification by flash chromatography. These
figures mean a yield of ca. 70-80% per each ring closing,
which agree with the best yields found for the reaction
of N-nitropyrimidine nucleosides with simple alkyl-
amines.1a
5′-O-(ter t-Bu t yld im et h ylsilyl)-2′,3′-O-isop r op ylid en e-3-
n itr ou r id in e (1b). TFAA (1.12 mL, 8.0 mmol) was added to a
suspension of finely powdered NH4NO3 (320 mg, 4.00 mmol) in
anhydrous CH2Cl2 (10 mL), at 0 °C. The mixture was vigorously
stirred at rt until the solid was dissolved (45 min) and then
cooled again. Compound 1a 8 (399 mg, 1.00 mmol) in CH2Cl2
(0.5 mL) was added, the syringe was rinsed with an additional
volume of CH2Cl2 (1.5 mL), and the resulting solution was stirred
for 15 min. Addition of more solvent, washing with cold
phosphate buffer, drying, and filtration through a pad of silica
gel afforded chromatographically pure 1b (400 mg, 90% yield)
as a viscous oil: [R]D ) -22.28 (c 0.61, CHCl3); 1H NMR (CDCl3)
δ 0.10 (s, 3 H), 0.11 (s, 3 H), 0.91 (s, 9 H), 1.37 (s, 3 H), 1.59 (s,
3 H), 3.80 (dd, J ) 11.8, 2.6, 1 H), 3.94 (dd, J ) 11.8, 1.8, 1 H),
4.44 (m, 1 H), 4.65-4.80 (m, 2 H), 5.81 (d, J ) 8.4, 1 H), 5.95 (d,
J ) 2.2, 1 H), 7.82 (d, J ) 8.4, 1 H); 13C NMR (CDCl3) δ -5.6,
-5.5, 18.3, 25.2, 25.8, 27.2, 63.4, 80.5, 85.6, 87.2, 93.4, 100.8,
114.2, 139.9, 145.3, 155.2; FABMS m/ z 444 (M + 1).
N3,O2′-Din itr o-3′,5′-O-(tetr a isop r op yld isiloxa n e-1,3-d iyl)-
u r id in e (2b). From 2a plus NH4NO3 (6 mol/mol) and TFAA
(12 mol/mol); reaction time ) 40 min; 88% yield; white gum;
1
[R]D ) +28.74 (c 0.43, CHCl3); H NMR (CDCl3) δ 1.0-1.1 (m,
28 H), 4.00 (dd, J ) 13.3, 2.7, 1 H), 4.09 (dd, J ) 9.5, 2.7, 1 H),
4.26 (d, J ) 13.3, 1 H), 4.49 (dd, J ) 9.5, 4.6, 1 H), 5.59 (d, J )
4.4, 1 H), 5.80 (s, 1 H), 5.87 (d, J ) 8.4, 1 H), 7.67 (d, J ) 8.4, 1
H); 13C NMR (CDCl3) δ 12.5, 12.8, 12.9, 13.4, 16.7, 16.8, 17.2,
In summary, both TBDMS and TIPDS ethers do
survive to the action of the excess of nitrating mixture
utilized in the first step (from N-H to N-NO2) and to
the basic media required in the second step (from N-NO2
to N-R) of our N-exchange procedure.1 Contrary to our
former suggestion,1a these common protecting groups are
(7) For illustrative reports regarding formation of bridges among
nucleosides through their pyrimidine or purine rings and/or of inter-
strand covalent cross-links in oligonucleotide and DNA duplexes, see
the following papers (and references cited therein): (a) Wang, H.;
Zuiderweg, E. R. P.; Glick, G. D. J . Am. Chem. Soc. 1995, 117, 2981
(T-to-T disulfide bridges, from N3-CH2CH2SH). (b) Milton, J .; Con-
nolly, B. A.; Nikiforov, T. T.; Cosstick, R. J . Chem. Soc., Chem.
Commun. 1993, 779 (T-to-In disulfide bridges through 4-thiothymine
and 6-mercaptopurine bases). (c) Agathocleous, D. C.; Page, P. C. B.;
Cosstick, R.; Galpin, I. J .; McLennan, A. G.; Prescott, M. Tetrahedron
1990, 46, 2047 (dinucleosides bridged by a carbon chain through the
amino groups). (d) Asseline, U.; Thuong, N. T. Tetrahedron Lett. 1994,
35, 5221 (dC-dC derivative linked through the amino groups). (e)
Kirchner, J . J .; Sigurdsson, S. T.; Hopkins, P. B. J . Am. Chem. Soc.
1992, 114, 4021 (HNO2-promoted G-to-G links, C2-NH-C2). (f) Rink,
S. M.; Solomon, M. S.; Taylor, M. J .; Rajur, S. B.; McLaughlin, L. W.;
Hopkins, P. B. J . Am. Chem. Soc. 1993, 115, 2551 (nitrogen mustard-
promoted G-to-G links through N7). (g) Armstrong, R. W.; Salvati, M.
E.; Nguyen, M. J . Am. Chem. Soc. 1992, 114, 3144 (G-to-G and G-to-A
links through N7 caused by carzinophilin). (h) Bose, D. S.; Thompson,
A. S.; Ching, J .; Hartley, J . A.; Berardini, M. D.; J enkins, T. C.; Neidle,
S.; Hurley, L. H.; Thurston, D. E. J . Am. Chem. Soc. 1992, 114, 4939
(pyrrolobenzodiazepine derivative, G-to-G links through the amino
groups).
(5) When a sample of 1a was added to an excess of nitrating mixture
that had been left for 18 h under N2, N-nitro compound 1b was not
obtained; only starting material (1a ) and deprotected product (desi-
lylated 1a ) were present in similar amounts. While the nitration of
1a with an excess of freshly prepared nitrating mixture for 15-30 min
always gives yields above 90% of 1b (deprotected product being not
observed), when the reaction was left overnight the yield of 1b
decreased to 31% and the crude mixture contained, in addition to 16%
of 1a , 18% of 2′,3′-O-isopropylideneuridine (desilylated 1a ) and 33%
of N3,O5′-dinitro derivative.
(6) (a) The TIPDS group is known to be sensitive to aqueous basic
media: Markiewicz, W. T. J . Chem. Res., Synop, 1979, 24. (b) There
are also examples of the moderate stability of TBDMS protecting
groups under certain basic conditions: Reddy, M. P.; Farooqui, F.;
Hanna, N. B. Tetrahedron Lett. 1995, 36, 8929 and references 1-3
therein. (c) Also see: Fallois, L. L. H.; De´cout, J .-L.; Fontecave, M.
Tetrahedron Lett. 1995, 36, 9479.
(8) Hayakawa, H.; Tanaka, H.; Maruyama, Y.; Miyasaka, T. Chem.
Lett. 1985, 1401. Hayakawa, H.; Tanaka, H.; Maruyama, Y.; Obi, K.;
Miyasaka, T. Nucleic Acids Symp. Ser. 1985, 16, 109.

Notes
J . Org. Chem., Vol. 62, No. 5, 1997 1549
17.4, 59.0, 67.8, 82.3, 82.5, 87.6, 101.4, 138.6, 144.9, 155.0; CIMS
(NH4+) m/ z 594 (M + 18).
Rea ction of 1b w ith Eth ylen ed ia m in e. To a solution of
1b (89 mg, 0.2 mmol) in anhyd CH3CN (4 mL) was added finely
powdered anhyd K2CO3 (21 mg, 0.15 mmol). To the resulting
2′-O-(Met h a n esu lfon yl)-3-n it r o-3′,5′-O-(t et r a isop r op yl-
d isiloxa n e-1,3-d iyl)u r id in e (3b). From 3a 9 plus NH4NO3 (4
mol/mol) and TFAA (8 mol/mol): reaction time ) 20 min; 90%
yield; white gum; [R]D ) +34.25 (c 0.36, CH2Cl2); 1H NMR
(CDCl3) δ 1.0-1.3 (m, 28 H), 3.22 (s, 3 H); 4.00 (dd, J ) 13.5,
2.4, 1 H), 4.13 (dd, J ) 9.6, 2.4, 1 H), 4.27 (d, J ) 13.5, 1 H),
4.34 (dd, J ) 9.6, 4.5, 1 H), 5.08 (d, J ) 4.5, 1 H), 5.82 (s, 1 H),
5.84 (d, J ) 8.4, 1 H), 7.85 (d, J ) 8.4, 1 H); 13C NMR (CDCl3)
δ 12.5, 12.7, 12.8, 13.4, 16.7, 16.8, 17.1, 17.3, 39.1, 58.8, 66.6,
82.1, 82.2, 89.4, 101.1, 138.5, 145.1, 154.9; CIMS (NH4+) m/ z
627 (M + 18).
suspension was added
a solution of ethylenediamine (1,2-
ethanediamine, 7 µL, 0.1 mmol) in anhyd CH3CN (1 mL)
dropwise (over 1 h). After vigorous stirring for 2 days at rt, slow
warming to 80 °C, and then refluxing for 7 h, the mixture was
poured into an excess of phosphate buffer solution and extracted
with CH2Cl2. Drying of the organic phase, removal of the solvent
under vacuum, and separation of the desired compound from
polar byproducts by column chromatography (SiO2, CH2Cl2-
MeOH 98:2) gave 1,2-bis(5′-O-(tert-butyldimethylsilyl)-2′,3′-O-
isopropylideneuridin-3-yl)ethane, 6 (45 mg, 55%), as a white
gum: 1H NMR (CDCl3) δ 0.08 (s, 2 × 6 H), 0.88 (s, 2 × 9 H),
1.36 (s, 2 × 3 H), 1.57 (s, 2 × 3 H), 3.75 (dd, J ) 11.6, 3.5, 2 ×
1 H), 3.91 (dd, J ) 11.6, 2.4, 2 × 1 H), 4.2-4.4 (m, 6 H), 4.6-4.8
(m, 4 H), 5.58 (d, J ) 8.2, 2 × 1 H), 5.89 (d, J ) 1.0, 2 × 1 H),
7.64 (d, J ) 8.2, 2 × 1 H); 13C NMR (CDCl3) δ -5.6, -5.5, 18.3,
25.4, 25.8, 27.2, 38.8, 63.1, 79.9, 85.9, 87.2, 92.9, 100.9, 113.7,
138.3, 150.9, 163.1; HRMS (FAB) calcd for C38H63N4O12Si2 (M
+ H+) m/ z 823.3981, found 823.3906.
Rea ction of 1b w ith 1,3-P r op a n ed ia m in e. To a solution
of 1b (222 mg, 0.50 mmol) in anhyd CH3CN (5 mL) was added
finely powdered anhyd K2CO3 (55 mg, 0.4 mmol). To the
resulting suspension was added a solution of 1,3-propanediamine
(21 µL, 0.25 mmol) in anhyd CH3CN (3 mL) dropwise, over 1 h.
After vigorous stirring for 2 days at rt, slow warming to 80 °C,
and then refluxing for ca. 2 h, the solvent was removed and the
residue was purified by column chromatography (SiO2, CH2Cl2-
MeOH 98:2) to afford 1,3-bis[5′-O-(tert-butyldimethylsilyl)-2′,3′-
O-isopropylideneuridin-3-yl]propane, 7 (120 mg, 57%), as a white
gum: 300-MHz 1H NMR (CDCl3) δ 0.08 (s, 2 × 6 H), 0.89 (s, 2
× 9 H), 1.37 (s, 2 × 3 H), 1.59 (s, 2 × 3 H), 2.00 (m, 2 H), 3.80
(dd, J ) 11.5, 3.1, 2 × 1 H), 3.92 (dd, J ) 11.5, 2.5, 2 × 1 H),
4.02 (m, 4 H), 4.36 (m, 2 × 1 H), 4.73 (m, 2 × 2 H), 5.68 (d, J )
8.1, 2 × 1 H), 5.92 (d, J ) 2.1, 2 × 1 H), 7.64 (d, J ) 8.1, 2 × 1
H); 13C NMR (CDCl3) δ -5.6, -5.5, 18.2, 25.3, 25.8, 26.1, 27.2,
38.9, 63.3, 80.3, 85.9, 87.0, 93.2, 101.3, 113.8, 138.1, 150.7, 162.6.
HRMS (FAB) calcd for C39H65N4O12Si2 (M + H+) m/ z 837.4138,
found 837.4098.
5′-O-(ter t-Bu t yld im et h ylsilyl)-N3,O3′-d in it r ot h ym id in e
(4b). From 4a plus NH4NO3 (6 mol/mol) and TFAA (12 mol/
mol): reaction time ) 20 min; 85% yield; white solid; mp 89.0-
1
90.5 °C; H NMR (CDCl3) δ 0.16 (s, 6 H), 0.94 (s, 9 H), 2.01 (d,
J ) 1.2, 3 H), 2.28 (ddd, J ) 14.7, 9.2, 6.4, 1 H), 2.64 (dd, J )
14.7, 5.7, 1 H), 3.93 (dd, J ) 11.5, 1.9, 1 H), 4.01 (dd, J ) 11.5,
2.0, 1 H), 4.30 (m, 1 H), 5.47 (d, J ) 6.4, 1 H), 6.27 (dd, J ) 9.2,
5.7, 1 H), 7.54 (d, J ) 1.2, 1 H); 13C NMR (CDCl3) δ -5.5, -5.4,
12.8, 18.3, 25.8, 36.7, 63.5, 83.4, 84.1, 85.5, 110.9, 134.1, 145.2,
156.4; CIMS (NH4+) m/ z 464 (M + 18). Anal. Calcd for
C
16H26N4O9Si: C, 43.04; H, 5.87; N, 12.55. Found: C, 42.90;
H, 6.18; N, 12.30.
3-Nitr o-3′,5′-O-(tetr a isop r op yld isiloxa n e-1,3-d iyl)th ym i-
d in e (5b). From 5a plus NH4NO3 (4 mol/mol) and TFAA (8
mol/mol): reaction time ) 20 min; 92% yield; white solid; mp
1
106.5-107.5 ˚C; H NMR (CDCl3) δ 0.99-1.10 (m, 24 H), 1.99
(d, J ) 1.2, 3 H), 2.33 (ddd, J ) 13.8, 7.6, 1.8, 1H), 2.55 (ddd, J
) 13.8, 10.3, 7.2, 1 H), 3.79 (ddd, J ) 8.4, 2.8, 2.2, 1 H), 4.02
(dd, J ) 13.4, 2.8, 1 H), 4.16 (dd, J ) 13.4, 2.2, 1 H), 4.48 (ddd,
J ) 10.3, 8.4, 7.6, 1 H), 6.02 (dd, J ) 7.2, 1.8, 1 H), 7.52 (q, J )
1.2, 1 H); 13C NMR (CDCl3) δ 12.4, 12.7, 12.8, 12.9, 13.5, 16.8,
16.9, 17.3, 17.4, 39.8, 59.7, 66.8, 85.1, 85.4, 109.8, 134.6, 145.0,
156.3; CIMS (NH4+) m/ z 547 (M + 18); FABMS m/ z 530 (M +
1). Anal. Calcd for C22H39N3O8Si2: C, 49.88; H, 7.42; N, 7.93.
Found: C, 49.69; H, 7.70; N, 7.81.
Rea ction of 1b w ith 15NH3. To a solution of 15NH4Cl (22
mg, 0.4 mmol) and K2CO3 (111 mg, 0.80 mmol) in water (1 mL),
in a septum-closed vial, was added a solution of 1b (178 mg,
0.40 mmol) in CH3CN (1 mL; plus 2 mL to rinse the syringe).
The mixture was either shaken or vigorously stirred for 6 days
at rt. Afterwards, it was poured into a phosphate buffer solution
and extracted with CH2Cl2. The organic layer was dried with
Na2SO4, and the solvent was removed under vacuum. Purifica-
tion of the residue by column chromatography over silica gel,
with CH2Cl2-MeOH 99:1 to 98:2 as the eluents, gave a pure
product (chromatographically identical to 1a ), the NMR spectra
of which showed the splittings expected for 15N3-labeled 1a : 300-
MHz 1H NMR (CDCl3) δ 0.09 (s, 3 H), 0.10 (s, 3 H), 0.91 (s, 9
H), 1.36 (s, 3 H), 1.59 (s, 3 H), 3.80 (dd, J ) 11.6, 3.0, 1 H), 3.94
(dd, J ) 11.6, 2.4, 1 H), 4.33 (m, 1 H), 4.68 (dd, J ) 6.0, 3.0, 1
H), 4.78 (dd, J ) 6.4, 2.7, 1 H), 5.70 (ddd, J ) 8.0, J H5-N ) 2.8,
J H5-NH ) 2.2, 1 H), 5.98 (d, J ) 2.7, 1 H), 7.71 (d, J ) 8.0, 1 H),
8.1 (concentration-dependent value, J HN ) 90.8, J ) 2.2, 1 H);
13C NMR (CDCl3) δ -5.6, -5.5, 18.3, 25.3, 25.8, 27.2, 63.3, 80.2,
85.3, 86.6, 91.8, 102.1 (d, J CN ) 7.3, C5), 114.0, 140.6, 150.2 (d,
J CN ) 18.2, C2), 163.4 (d, J CN ) 9.1, C4); CIMS (NH4+) m/ z 417
(M + 18, 100), 400 (M + 1, 53) [M(C18H3014N15NO9Si) ) 399].
Ack n ow led gm en t. Thanks are due to the CIRIT
and CICYT (Programa de Qu´ımica Fina) for financial
support (QFN93-4422) and to the CIRIT for a doctorate
studentship to one of us (C.S., 1995-1998). We are also
indebted to Dr. Irene Ferna´ndez (MS Service, in our
Department) for CI and FAB mass spectra and to Dr.
Xavier Huguet (Servei d’Espectrometria de Masses,
CSIC, Barcelona) for the high-resolution FAB mass
spectra.
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra of
compounds 1b, 2b, 3b, 1a *, 6, and 7 (7 pages). This material
is contained in libraries on microfiche, immediately follows this
article in the microfilm version of the journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
(9) Markiewicz, W. T. Chem. Scr. 1986, 26, 123.
J O9615514