Lewis acid deprotection of silyl-protected oligonucleotides and base-sensitive oligonucleotide analogues

Article abstract of DOI:10.1016/j.tetlet.2004.06.081

a- a- a- Oligonucleotides protected with N-(trimethylsilylethoxycarbonyl) (Teoc) and P-(trimethylsilylethanol) (Tse) groups were synthesized and deprotected by a single ZnBr₂ treatment. Teoc group stabilized dA against depurination. This strategy was applied to the synthesis of base-sensitive oligonucleotide prodrugs bearing S-acetyl-2-thioethyl (Sate) phosphotriesters.

Full text of DOI:10.1016/j.tetlet.2004.06.081

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 6287–6290
Lewis acid deprotection of silyl-protected oligonucleotides and
base-sensitive oligonucleotide analogues
Fernando Ferreira, Jean-Jacques Vasseur and Franc¸ois Morvan*
´ ` ´
Laboratoire de Chimie Organique Biomoleculaire de Synthese, UMR 5625 CNRS-UM II, Universite de Montpellier II,
CC008, Place E. Bataillon, 34095 Montpellier Cedex 5, France
Received 4 June 2004; accepted 21 June 2004
Abstract—Oligonucleotides protected with N-(trimethylsilylethoxycarbonyl) (Teoc) and P-(trimethylsilylethanol) (Tse) groups were
synthesized and deprotected by a single ZnBr2 treatment. Teoc group stabilized dA against depurination. This strategy was applied
to the synthesis of base-sensitive oligonucleotide prodrugs bearing S-acetyl-2-thioethyl (Sate) phosphotriesters.
Ó 2004 Elsevier Ltd. All rights reserved.
Prodrugs of oligonucleotides (prooligos) are designed to
circumvent the main limitations of the oligonucleotides
as therapeutics: that is degradation by nucleases and
poor uptake. Thus we have shown that prooligos are
nuclease resistant, and rapidly and highly taken up by
cells.1 Since prooligos are constituted of phosphate pro-
tected with base-sensitive S-acetyl-2-thioethyl (Sate)
groups, it is compulsory to use protecting groups that
will be removed without ammonia treatment. For that
purpose, our group develop several strategies.2–4
Thus after deprotection they will yield phosphodiester
linkages. The Tse groups will be removed in the same
manner than the Teoc ones.
Firstly we synthesized the three nucleosides correspond-
ing to dC, dA and dG with the Teoc protecting group.
The dC was efficiently protected using a transient 30
and 50-OH protection with trimethylsilyl group and then
with 4-nitrophenyl-2-(trimethylsilyl) ethylcarbonate in
presence of DMAP as catalyst. After 16h, a 20min
treatment with ammonia yielded the expected N4-
Teoc-dC 1a (80%), (Scheme 1). Finally it was 50-dimeth-
oxytritylated to yield 2a (85%).
Herein we present the synthesis of base-sensitive prooli-
gos using silyl protecting groups either on the nucleo-
bases or on the phosphate.5
Since the amino function of dA and dG is not enough
nucleophilic the 2-(trimethylsilyl)ethyl-carbonochlori-
date (TeocCl) was used instead of the 4-nitrophenyl
derivative. This reagent was fleshly prepared starting
from 2-(trimethylsilyl)ethanol and phosgene in toluene
in the presence of powdered anhydrous potassium car-
bonate.7 Introduction of Teoc group was performed
We chose to use the trimethylsilylethoxycarbonyl (Teoc)
group that could be rapidly introduced on the nucleo-
bases starting either from its p-nitrophenol derivative6
for C or from its chloride derivative7 for A and G. Teoc
group was described for amino protection8 in peptide
synthesis and for hydroxyl protection9 in oligonucleo-
tide synthesis, but never as protection of nucleobases.
This protecting group could be removed under several
conditions10 and specially under a mild treatment with
Lewis acid like ZnCl2 or ZnBr2.9,11 Furthermore, since
prooligos should exhibit some charges to be soluble in
aqueous media, they were introduced through trimethyl-
silylethyl (Tse) phosphotriester as already reported.4,12
O
N
O
N
Si
Si
NH2
N
HN
N
O
O
HN
N
O
O
N
O
HO
HO
DmtrO
i, ii, iii
iv
O
O
O
OH
OH
OH
1a
2a
Scheme 1. Reagents: (i) TMSCl, dry pyridine; (ii) 4-nitrophenyl-2-
(trimethylsilyl) ethylcarbonate, DMAP; (iii) NH4OH; (iv) DmtrCl dry
pyridine.
Keywords: Silyl protecting group; DNA; Teoc; Tse; Sate.
*
Corresponding author. Tel.: +33-467-144-961; fax: +33-467-042-
029; e-mail: morvan@univ-montp2.fr
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.06.081

6288
F. Ferreira et al. / Tetrahedron Letters 45 (2004) 6287–6290
O
N
O
dC>dA>dG. It is noteworthy that Tse group on O-6
NH2
N
Si
ii, iii
Si
HN
N
O
HN
N
O
N
N
of dG was removed by the same way. With BiCl3 we ob-
served some depurination. As a DNA synthesis cycle in-
volves an acidic treatment, we studied the stability of
30,50-di-O-Ac-N6-Teoc-dA 1b and N6-Bz-dA in acidic
conditions. While N6-Bz-dA was depurinated in 80%
acetic acid at 20°C within 1h, 1b was fully stable up
to 2h. On the contrary 1b was degraded in a 5% TFA
CH2Cl2 solution, but fortunately stable in 3% TCA or
2.5% DCA in CH2Cl2. Hence we could use the standard
detritylation solution on the synthesizer.
i
N
N
N
N
N
N
AcO
AcO
DmtrO
O
O
O
OAc
OAc
OH
2b
1b
Scheme 2. Reagents: (i) TeocCl, N-methyl imidazolo CH2Cl2; (ii) 0.2N
NaOH, THF–MeOH–H2O; (iii) DmtrCl dry pyridine.
starting from 30,50-di-O-acetyl dA and dG.13 On the one
hand (Scheme 2), the N-6 of adenine was protected by
reaction with TeocCl in presence of N-methyl imidazole
in dichloromethane for 16h (95%). Then a 10-min treat-
ment with 0.2N NaOH in THF–MeOH–H2O (25:15:10,
v/v/v) led to the expected N6-Teoc-dA. After work up,
the crude was directly dimethoxytritylated to give 2b
(80%).
As Teoc and Tse groups were rapidly removed by ZnBr2
treatment and were stable under DCA or TCA treat-
ment, they are fully compatible to be used for the syn-
thesis of base-sensitive oligonucleotides like prooligos.
To be soluble in aqueous media the prooligos should
have some charges. For that purpose, we have shown
that Tse group on the phosphate could be removed by
a Et3N–3HF treatment without the hydrolysis of the
Sate groups.4 Thus the 50-O-Dmtr-N-protected nucleo-
sides (2a–d) were converted into Sate (3a–d) and Tse
(4a–d) phosphoramidite derivatives (Scheme 4) in pres-
ence of diisopropyl ammonium tetrazolide as catalyst
in CH2Cl2 using Sate bis-N,N0-diisopropyl phosphine15
and Tse bis-N,N0-diisopropyl phosphine,12 respectively.
3a, (80%); 3b (70%); 3c (83%); 3d (75%), 4a (90%); 4b
(80%); 4c (85%); 4d (90%).
On the other hand, the protection of the O-6 position of
dG was necessary to obtain a good yield.14 Without, the
di-O-30,50-acetyl N2-Teoc-dG was obtained with only
15% yield. Thus, the O-6 of guanine was protected with
Tse group in two steps. First (Scheme 3) 2,4,6-tri-
isopropylbenzenesulfonyl chloride in presence of
DMAP and TEA reacted on the O-6, then the trimeth-
ylsilylethanol in presence of DABCO displaced this leav-
ing group to yield the O6-Tse-dG (60%). The Teoc group
was finally introduced on N-2 by treatment with TeocCl
in presence of tert-butyl magnesium chloride in THF for
16h (70%). Acetyl groups were removed by means of a
solution of 0.2N NaOH in THF–MeOH–H2O
(25:15:10, v/v/v) for 10min. After work up, the crude
was directly dimethoxytritylated to give 2c (75%).
An oligo phosphodiester GCATTAGCATpOCH3 was
synthesized from the Tse phosphoramidites to evaluate
their efficacy. They were used at a standard 0.1M con-
centration in dry acetonitrile with a 120s coupling step.
Since acetic anhydride usually used for capping step
could also react on the exocyclic amino function spe-
cially of adenine we capped with di-tert-butyl diethyl
phosphoramidite (0.05M in CH3CN) for 10s. Oxidation
was performed with a 0.067% 2-butanoneperoxyde solu-
tion in CH2Cl2 for 60s16 and detritylation with standard
3% TCA solution for 60s.
In order, to find the best treatment for the removal of
Teoc group we tried several conditions at the nucleoside
level (N4-Teoc-dC 1a, 30,50-di-O-Ac-N6-Teoc-dA 1b and
30,50-di-O-Ac-O6-Tse-N2-Teoc-dG 1c). Treatment with
fluorine reagents (Et3N–3HF, TBAF, TBAF–AcOH
and HF–pyridine) led to no or incomplete deprotection,
while treatment with acid Lewis like ZnBr2 and ZnCl2
gave rapid deprotection of the Teoc group. The full de-
protection was faster with ZnBr2 (30–60min) than with
ZnCl2 (about 2h). The order of silyl removal was
In order, to be released from the solid support without
ammonia treatment we used a solid support with a disul-
DMTrO
B
O
O
N
P
O
P
S
O
S
O
N
Si
S
O
N
O
O
O
N
O
N
N
N
N
N
N
N
NH
NH2
O
N
N
N
NH2
ii
NH2
i
AcO
AcO
AcO
3a-d
O
O
i
i
O
B
DMTrO
OAc
OAc
OAc
O
iii
OH
2a-d
i
B
DMTrO
Si
O
Si
O
O
N
O
N
O
O
N
N
N
N
N
2a B : N4-teoc C
2b N6-teoc A
2c N2-teoc,O6-tse G
N
iv, v
Si
Si
N
H
O
N
N
H
O
AcO
DmtrO
P
Si
O
P
Si
O
O
O
N
N
OAc
OH
2c
1c
2d
T
4a-d
Scheme 3. Reagents: (i) 2,4,6-triisopropylbenzenesulfonyl chloride,
DMAP, Et3N, CH2Cl2; (ii) TseOH, DABCO; (iii) TeocCl, t-Bu–MgCl,
THF; (iv) 0.2N NaOH, THF–MeOH–H2O; (v) DmtrCl dry pyridine.
Scheme 4. Synthesis of Sate and Tse phosphoramidites derivatives.
Reagents: (i) diisopropyl ammonium tetrazolide, CH2Cl2.

F. Ferreira et al. / Tetrahedron Letters 45 (2004) 6287–6290
6289
Gteoc,tse
Gteoc,tse
Cteoc
Ateoc
T
T
T
T
Ateoc
Cteoc
Ateoc
T
T
Ateoc
Gteoc,tse
Cteoc
T
Ateoc
Gteoc,tse
Cteoc
Ateoc
T
O
O
P
O
P
O
P
O
P
O
P
O
O
P
O
P
O
P
O
O
P
O
P
O
P
O
P
O
P
O
O
P
O
P
O
P
S
S
O
HO
O
P
O
O
O
O
O
O
O
O
O
O
O
O
P
O
O
O
O
O
O
HO
HO
HO
O
O
O
P
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
P
O
O
O
O
O
O
O
O
O
O
O
O
hν
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
H3C
O
Si
Si
Si
Si
Si
Si
Si
Si
Si
S
S
S
S
Si
Si
Si
Si
S
S
O
O
O
O
O
sat. ZnBr2 in CH3NO2/iPrOH
sat. ZnBr2 in CH3NO2/iPrOH
G
C
A
T
A
G
C
T
A
G
C
A
T
T
A
G
C
A
T
O
P
O
P
O
P
O
O
O
O
O
P
O
P
O
P
O
P
O
P
O
P
O
O
O
P
O
O
P
O
O
O
O
O
O
P
P
O
P
O
O
P
O
O
O
O
hν
S
HO
O
O
O
O
O
O
O
P
O
O
P
O
O
O
O
P
O
O
O
O
P
O
O
P
O
O
O
O
O-
O-
O-
O-
O
O
O
O-
O-
O-
O-
O-
O-
O-
O-
O-
O
S
H3C
O
S
S
S
S
S
S
TCEP, CH3CN, H2
O
O
O
O
O
O
Photolysis
CH3CN H2O 1:1: v/v
G
C
A
T
T
A
G
C
A
T
O
O
O
O
O
O
P
O
O
O
O
G
C
A
T
A
G
C
T
A
HO
O
P
O
O
P
O
O
P
O
O
P
O
O
P
O
O
O
O
P
O
O
P
O
O
P
O
O
P
O-
O
P
O
P
O
P
O
O
O
O
O
P
O
P
O
P
O-
O-
O-
O-
O-
O-
O-
O-
O-
O
O-
O
O
O
P
P
O
P
O
O
P
O
O
O
O
H3
C
O
O
O
O-
O-
O-
O-
O
O
O
S
S
S
S
S
S
O
O
O
O
O
Scheme 5. Deprotection of oligo GCATTAGCATpOCH3 by ZnBr2
treatment and release from solid support by TCEP treatment.
Scheme 6. Deprotection of oligo GCATTAGCTAT by ZnBr2 treat-
ment and release from solid support by photolysis.
fide linkage17 that will be cleaved by tris-2-carboxyethyl-
phosphine (TCEP).
strategy opens the way to the synthesis of base-sensitive
prooligos.
After elongation, the CPG-supported oligo was treated
overnight with a saturated solution of ZnBr2 in nitro-
methane–isopropanol (1:1, v/v) (Scheme 5). Then the
beads were thoroughly washed with water and with a
0.1M EDTA solution to scavenge the Zn2+ cations. Fi-
nally, the oligo was cleaved from the solid support by
treatment with TCEP in triethylammonium acetate buf-
fer pH7 with 80% acetonitrile for 2h. This treatment led
to a 30-phosphotriester with a thio-ethyl group that elim-
inated spontaneously to a 30-phosphodiester after elimi-
nation of episulfide. As that elimination is very slow on
diester due to the negative charge, we started the synthe-
sis with methoxyphosphoramidite18 to obtain after
ZnBr2 treatment a 30-methyl triester linkage and then
the expected 30-methyl phosphodiesters as confirmed
by MALDI-TOF MS (negative mode m/z for
C99H126N37O61P10 calcd 3120.07, found 3121.90). The
crude HPLC showed a broad peak (Fig. 1 left) with
peaks at higher retention times corresponding to the
short mers 50-di-tert-butylphosphotriester as determined
by MALDI-TOF MS (Fig. 1 middle). After purification
the pure decamer was obtained (Fig. 1 right).
Using the eight phosphoramidites derivative 3a–d and
4a–d a prooligo exhibiting the four nucleobases and
bearing on each side three phosphotriester SATE link-
ages (GpSateCpSateApSateTTAGCpSateTpSateApSate) was
synthesized on a solid support with a photolabile lin-
ker.19
After elongation according to the same cycle than for
the previous oligo, the prooligo was deprotected
(Scheme 6) by treatment with ZnBr2 in nitromethane
isopropanol (7h). The deprotection was monitored by
MALDI-TOF MS20 thanks to the photolabile linker.
We observed that this treatment also led to few hydrol-
ysis of one Sate group.
Then a 20min photolysis released the prooligo in solu-
tion. Nevertheless, one Sate group was partially
removed as showed by HPLC (Fig. 2) peaks 13–
14.5min and MALDI-TOF MS (Mꢀ102Da). The
prooligo with the six Sate groups was eluted as a massif
between 14.5 and 17min owing to the 26 diastereoiso-
mers. Each main peak was characterized by MALDI-
TOF MS and exhibited the same mass corresponding
to the expected prooligo with six Sate (Negative mode
m/z for C122H160N37O67P10S6 calcd 3718.97, found
3722.10).
A first oligo was synthesized using silyl protecting
groups on the nucleobase (Teoc+Tse for G) and on
the phosphorous (Tse). By a simple treatment with
Lewis acid like ZnBr2 all the protecting groups were effi-
ciently removed, without any ammonia treatment. This
Figure 1. C18 reverse phase HPLC profile of GCATTAGCATpoCH3 crude (left) and purified (right). MALDI-TOF MS of crude (middle).

6290
F. Ferreira et al. / Tetrahedron Letters 45 (2004) 6287–6290
`
the ÔMinistere de la recherche et de la TechnologieÕ for
the award of a researchship.
3722.10
References and notes
1. Bologna, J. C.; Vives, E.; Imbach, J. L.; Morvan, F.
Antisense Nucl. Acid Drug Dev. 2002, 12, 33–41.
2. Alvarez, K.; Vasseur, J. J.; Beltran, T.; Imbach, J. L. J.
Org. Chem. 1999, 64, 6319–6328.
3. Spinelli, N.; Meyer, A.; Hayakawa, Y.; Imbach, J. L.;
Vasseur, J. J. Eur. J. Org. Chem., 2002, 49–56.
4. Guerlavais Dagland, T.; Meyer, A.; Imbach, J. L.;
Morvan, F. Eur. J. Org. Chem., 2003, 2327–2335.
5. Part of this work was presented at the Nucleic Acids
Chemistry & Biology: 5th Cambridge Symposium, Cam-
bridge, UK, August 31, September 3, 2003.
6. Dhanak, D.; Reese, C. B. J. Chem. Soc., Perkin Trans. 1,
1986, 2181–2186.
7. Shute, R. E.; Rich, D. H. Synthesis, 1987, 346–349.
8. Rosowsky, A.; Wright, J. E. J. Org. Chem. 1983, 48,
1539–1541.
1500 2100 2700 3300 3900 4500
Mass (m/z)
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0.0
9. Gioeli, C.; Balgobin, N.; Josephson, S.; Chattopadhyaya,
J. B. Tetrahedron Lett. 1981, 22, 969–972.
10. Greene, T. W.; Wuts, P. G. M. Protective Groups in
Organic Synthesis; Wiley & Sons Inc, 1999.
11. Wunsch, E.; Moroder, L. Hoppe-SeylerÕs Z. Physiol.
¨
Chem. 1981, 362, 1289–1292.
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
12. Wada, T.; Sekine, M. Tetrahedron Lett. 1994, 35, 757–760.
13. Waldmann, H.; Reidel, A.; Heuser, A.; Muhlegger, K.;
Von der Eltz, H.; Birkner, C. EP 0 649 855 A1, 1994.
14. Hayakawa, Y.; Wakabayashi, S.; Kato, H.; Noyori, R. J.
Am. Chem. Soc. 1990, 112, 1691–1696.
Minutes
Figure 2. MALDI-TOF MS and C18 reverse phase HPLC of the crude
prooligo GpSateCpSateApSateTTAGCpSateTpSateApSate
.
15. Tosquellas, G.; Alvarez, K.; DellÕAquila, C.; Morvan, F.;
Vasseur, J. J.; Imbach, J. L.; Rayner, B. Nucl. Acids Res.
1998, 26, 2069–2074.
16. Kataoka, M.; hattori, A.; okino, S.; Hyodo, M.; Asano,
M. R. K.; Hayakawa, Y. Org. Lett. 2001, 3, 815–818.
17. Thuong, N. T.; Asseline, U. In Oligonucleotides and
Analogues. A Practical Approach; Eckstein, F., Ed.;
Oxford University Press: Oxford, 1991; pp 283–308.
18. Beaucage, S. L.; Caruthers, M. H. Tetrahedron Lett. 1981,
22, 1859–1862.
In conclusion a strategy using silyl protecting groups on
the nucleobases and on the phophate was successfully
applied to the synthesis of regular and base-sensitive
oligonucleotide. The Teoc and Tse silyl groups were effi-
ciently removed by treatment with ZnBr2. It is notewor-
thy that the Teoc is poorly removed by fluoride reagents.
Furthermore Teoc group stabilizes dA against depurina-
tion.
19. DellÕAquila, C.; Imbach, J. L.; Rayner, B. Tetrahedron
Lett. 1997, 38, 5289–5292.
20. Meyer, A.; Spinelli, N.; Bres, J. C.; DellÕ Aquila, C.;
Morvan, F.; Lefebvre, I.; Rayner, B.; Imbach, J. L.;
Vasseur, J. J. Nucleos. Nucleot. Nucl. Acids 2001, 20,
963–966.
Acknowledgements
This work was supported by grant from ÔAssociation
pour la Recherche sur le CancerÕ (ARC), F.F. thanks