Alternatives to the 4,4′-dimethoxytrityl (DMTr) protecting group

Article abstract of DOI:10.1016/j.tetlet.2004.01.152

The 9-phenyl- and the 9-(p-tolyl)-xanthen-9-yl groups 2a and 2b are recommended as alternatives to the 4,4′-dimethoxytrityl group 1 for the protection of the 5′-hydroxy functions in oligonucleotide synthesis.

Full text of DOI:10.1016/j.tetlet.2004.01.152

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 2567–2570
Alternatives to the 4,40-dimethoxytrityl (DMTr) protecting group
Colin B. Reese* and Hongbin Yan
Department of Chemistry, KingÕs College London, Strand, London WC2R 2LS, UK
Received 15 December 2003; revised 20 January 2004; accepted 30 January 2004
Abstract—The 9-phenyl- and the 9-(p-tolyl)-xanthen-9-yl groups 2a and 2b are recommended as alternatives to the 4,40-dimeth-
oxytrityl group 1 for the protection of the 50-hydroxy functions in oligonucleotide synthesis.
Ó 2004 Elsevier Ltd. All rights reserved.
R
Thy
Thy
RO
RO
O
O
AcO
HO
MeO
OMe
O
4 a ; R = 1
b ; R = 2a
c ; R = 2b
3 a ; R = 1
b ; R = 2a
c ; R = 2b
1 (DMTr)
2 a ; R = H (Px)
b ; R = Me (Tx)
Thy = thymin-1-yl
Forty years ago, Khorana and his co-workers introduced1
the acid-sensitive 4,40-dimethoxytrityl (DMTr) 1 group for
the protection of 50-hydroxy functions in oligonucleotide
synthesis. This protecting group has since been used very
widely indeed, especially in the solid phase synthesis of
oligonucleotides and their analogues.2 Twenty-five years
ago, we introduced3 the 9-phenylxanthen-9-yl (Px) group
2a as an alternative to the DMTr protecting group 1.
Unlike the corresponding 50-O-DMTr derivatives (as
in 3a), 50-O-Px derivatives (as in 3b) of the commonly
used N-acyl-20-deoxyribonucleosides crystallise readily.3
Another advantage of the Px group that was apparent
from our original study3 was that it is marginally (ca. 33%)
more labile than the DMTr group in acetic acid–water (4:1
v/v) solution. This is advantageous because the glycosidic
linkages of purine (especially 6-N-acyladenine) deoxyribo-
nucleosides are readily cleaved2 under acidic conditions.
Nevertheless, the Px group 2a has so far been used much
less widely in the solid phase synthesis both of DNA and
RNA sequences than the DMTr protecting group 1. One
possible reason for this is that the required monomeric
building blocks are not commercially available.
decreased by the introduction of electron-donating or
electron-withdrawing substituents. Of particular signifi-
cance in the context of oligonucleotide synthesis was the
observation that the introduction of a para-methyl
substituent in the 9-phenyl residue (as in 2b) of the Px
group increased its acid-lability without compromising
its robustness. Thus 50-O-[9-(p-tolyl)xanthen-9-yl]thy-
midine (50-O-Tx-thymidine) 3c was converted4 into
thymidine (3; R ¼ H) by treatment with trifluoroacetic
acid (TFA) in dichloromethane–ethanol (95:5 v/v)
solution at 23 °C ca. 2.5 times more rapidly than the
corresponding 50-O-Px derivative 3b. We now report a
more detailed study of the comparative unblocking rates
of the DMTr, Px and Tx protecting groups (1, 2a and
2b, respectively). In the deoxy-series, we chose the
50-protected 30-O-acetylthymidine derivatives5 4a–c as
model substrates and dichloroacetic acid as the
unblocking agent. In order to measure the relative
unblocking rates in solution, it is necessary to add a
reagent that irreversibly scavenges ÔtritylÕ cations (i.e.,
DMTrþ, Pxþ and Txþ). We have found pyrrole6 to be
particularly useful for this purpose and have demon-
strated its efficacy as a rapid scavenger for Pxþ and
DMTrþ cations.6;7 The pKa of pyrrole ()0.27)8 is such
that it is only partially protonated by dichloroacetic acid
(pKa 1.25)8 and is incompletely protonated both by tri-
chloro- and trifluoro-acetic acids (pKas 0.66 and 0.23,
respectively).8 Although ÔtritylÕ scavengers may also be
In a later study, we demonstrated4 that the acid-lability
of the Px protecting group could either be increased or
* Corresponding author. Tel.: +44-20-7848-2260; fax: +44-20-7848-
1771; e-mail: colin.reese@kcl.ac.uk
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.01.152

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
2568
C. B. Reese, H. Yan / Tetrahedron Letters 45 (2004) 2567–2570
Table 1. Unblocking of 30-O-acetyl-50-protected thymidine derivatives 3
The 50-O-DMTr protecting group has also been used
widely in the solid phase synthesis of RNA sequences.2
As the glycosidic linkages of ribonucleoside derivatives
are much more resistant to acidic hydrolysis10 than those
of the corresponding 20-deoxyribonucleoside derivatives,
concomitant depurination should not be a problem
during the removal of the 50-O-DMTr group. A crucial
aspect of oligonucleotide synthesis is the choice11 of the
protecting group for the 20-hydroxy functions. The tert-
butyldimethyl-silyl (TBDMS)12 5, 1-(2-fluorophenyl)-4-
methoxy-piperidin-4-yl (Fpmp)13 6 and (tri-isopropyl-
silyloxy)methyl (TOM)14 7 groups are among the
20-protecting groups that have commonly been used in
combination with the 50-O-DMTr group in solid phase
oligonucleotide synthesis. We have favoured the use of
the Fpmp and related 1-aryl-4-alkoxypiperidin-4-yl
protecting groups. Recently, the 1-(4-chlorophenyl)-4-
ethoxypiperidin-4-yl (Cpep) group 8, which has
improved hydrolysis properties (i.e., greater stability at
low and greater lability at high pH), has been
developed15 as an alternative to the Fpmp group 6. A
unique and most important property of the Fpmp, Cpep
and related protecting groups is that their use15–17 allows
chemically- and ribonuclease-stable 20-protected oligo-
nucleotides to be isolated in a pure state and then con-
verted into fully-unblocked RNA sequences under mild
conditions of acidic hydrolysis, such that cleavage and
migration of the internucleotide linkages can occur only
to a negligible extent. Although the Fpmp and Cpep
groups have also been designed to resist hydrolytic
cleavage under relatively strong acidic conditions, it is
nevertheless desirable that the 50-protecting group
should be removable as rapidly as possible. We have
therefore carried out a comparative unblocking study on
a group of these 50-protected (i.e., 50-O-DMTr, 50-O-
Px and 50-O-Tx)-20-O-Cpep-ribonucleoside derivatives
9a–c.
Entry no
Substratea
Acid
molarityb
Pyrrole
molarity
t1=2 (s)
1
2
3
4
5
6
7
8
4a
4b
4c
4a
4b
4c
4a
4b
0.125
0.125
0.125
0.375
0.375
0.375
0.625
0.625
0.375
0.375
0.375
0.50
450
198
60
48
14
0.50
0.50
10
12
0.75
0.75
ca. 4
a All reactions were carried out in dichloromethane solution at 0 °C
with a substrate molarity of 0.025.
b Molarity of dichloroacetic acid.
useful in solid phase synthesis, they are rarely used as
released ÔtritylÕ cations are generally washed away.
All the unblocking experiments involving the 50-pro-
tected-30-O-acetylthymidine derivatives 4a–c were car-
ried out9 by treating a 0.025 M solution of the substrate
in dichloromethane with an excess each of dichloro-
acetic acid and pyrrole at 0 °C. In the first series of
experiments, 5 mol equiv of acid (i.e., ca. 1% by volume)
and 15 mol equiv of pyrrole were used. It can be seen
from Table 1 (entries 1–3) that the half-times (t1=2) for
the removal of the DMTr, Px and Tx protecting groups
were 450, 198 and 60 s, respectively. Despite the fact that
the Px and especially the Tx groups offer a rate advan-
tage over the DMTr protecting group, the time required
for the complete unblocking even of the Tx group is
likely to be ca. 8–10 min under these conditions.
Although unblocking is likely perhaps to be 4–5 times
faster at room temperature, removal, especially of the
DMTr protecting group, would still be unacceptably
slow. If the quantity of acid is increased to 15 mol equiv
(i.e., ca. 3% by volume) and 20 mol equiv of pyrrole is
added (entries 4–6), the Px and Tx groups are again
removed at significantly faster rates than the DMTr
protecting group. Indeed it can be estimated that, under
the latter conditions, the times required for virtually
complete removal of the Px and Tx groups at room
temperature would be ca. 30 and 20 s, respectively,
compared with ca. 1.5 min for the DMTr protecting
group. Finally, it can be seen that the unblocking rates
of the DMTr and Px groups at 0 °C can be further
increased by a factor of 3–4 (entries 7 and 8) by
increasing the acid concentration to 0.625 M (ca. 5% by
volume) and the pyrrole concentration to 0.75 M. Under
these conditions, the unblocking rate of the Tx-pro-
tected substrate 4c was too fast to measure with any
accuracy.
The ribonucleoside substrates18 used in this part of the
study were the 50-O-DMTr-, 50-O-Px- and 50-O-Tx-
derivatives (9a–c, respectively) of 4-N-benzoyl-20-O-[1-
(4-chlorophenyl-4-ethoxypiperidin-4-yl]-30-O-levulinyl-
cytidine 10. As the rate of cleavage of the Cpep pro-
tecting group15 8 is virtually constant in the pH range
0.5–2.5, it seemed sensible to use a stronger acid than
dichloroacetic acid in the unblocking reactions. These
reactions (Scheme 1) were carried out by treating
0.025 M solutions of the substrates 9a–c with
6.0 mol equiv of trifluoroacetic acid and 15 mol equiv of
Bz
Bz
C
C
RO
HO
O
O
i
O(Cpep)
(Lev)O
O(Cpep)
(Lev)O
EtO
10
9 a ; R = DMTr 1
b ; R = Px 2a
c ; R = Tx 2b
MeO
N
Me
Me C Si
i
O
Pr SiO
3
N
Bz
3
C
= 4-N-benzoylcytosin-1-yl; Lev =
7 (TOM)
F
Me
O
5 (TBDMS)
Cl
Scheme 1. Reagents and conditions: (i) CF3CO2H, pyrrole, CH2Cl2,
0 °C.
6 (Fpmp)
8 (Cpep)

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
C. B. Reese, H. Yan / Tetrahedron Letters 45 (2004) 2567–2570
2569
4a [1.43 (3H, s), 2.04 (3H, s), 2.32 (1H, m), 2.44 (1H, m),
3.22 (1H, dd, J 3.9 and 10.4), 3.74 (6H, s), 4.07 (1H, m),
5.30 (1H, m), 6.22 (1H, dd, J 6.0 and 8.5), 6.09 (4H, d, J
7.9), 7.12–7.40 (9H, m), 7.53 (1H, s), 11.40 (1H, br s)]; 4b
[1.46 (3H, s), 2.03 (3H, s), 3.36 (1H, m), 2.50 (1H, m), 3.15
(1H, dd, J 3.6 and 10.3), 3.20 (1H, dd, J 3.0 and 10.3), 4.05
(1H, m), 5.29 (1H, m), 6.20 (1H, dd, J 6.1 and 8.3), 7.08–
7.45 (13H, m), 7.58 (1H, s), 11.40 (1H, br s)]; 4c [1.47 (3H,
s), 2.02 (3H, s), 2.22 (3H, s), 2.34 (1H, m), 2.45 (1H, m),
3.13 (1H, dd, J 3.6 and 10.3), 3.18 (1H, dd, J 3.8 and 10.3),
4.04 (1H, m), 5.26 (1H, m), 6.19 (1H, dd, J 6.1 and 8.3),
7.12 (7H, m), 7.25–7.44 (5H, m), 7.58 (1H, s), 11.40 (1H,
br s)].
pyrrole in dichloromethane solution at 0 °C. No
detectable loss of the 20-O-Cpep protecting group was
observed in any of these experiments and the half-times
for the removal of the 50-O-DMTr, 50-O-Px and 50-O-Tx
groups from 9a, 9b and 9c, respectively, were found20 to
be 42, ca. 6 and ca. 3 s. It would therefore seem to be
advantageous to use the Px (or Tx) rather than the
DMTr group to protect the 50-hydroxy functions in solid
phase oligoribonucleotide synthesis. Indeed, in our ori-
ginal study16 involving the use of the 20-O-Fpmp pro-
tecting group in solid phase synthesis, the 50-hydroxy
functions were protected with the Px group. All sub-
sequent work was carried out with 50-O-DMTr-20-O-
Fpmp-protected monomers as they were commercially
available. As in the case of DMTr-protected building
blocks, coupling yields obtained with Px (and presum-
ably also with Tx)-protected building blocks16 can be
assayed spectrophotometrically.
6. Reese, C. B.; Serafinowska, H. T.; Zappia, G. Tetrahedron
Lett. 1986, 27, 2291–2294.
7. Reese, C. B.; Yan, H. J. Chem. Soc., Perkin Trans. 1 2002,
2619–2633.
8. Albert, A.; Serjeant, E. P. Ionisation Constants of Acids &
Bases. A Laboratory Manual; Methuen: London, 1962; p
124, p 145.
9. The rates of the 50-unblocking reactions were determined
as follows. Pyrrole (0.10–0.21 mL, 1.5–3.0 mmol) was
added to a solution of substrate 4a, 4b or 4c (0.10 mmol)
and 20,30,50-tri-O-acetyluridine (0.037 g, 0.10 mmol) in
dichloromethane (2.0 mL). The stirred solution was cooled
to 0 °C and a precooled (to 0 °C) 0.30–1.50 M solution of
dichloroacetic acid in dichloromethane (2.0 mL) was
added. After appropriate intervals of time, aliquots
(0.1 mL) of the reaction solution were removed and
basified with 0.7 M methanolic triethylamine. The samples
were analysed by HPLC on a Jones C18 reversed phase
column. Straight lines were obtained by plotting log10 [%
substrate remaining] against time. The times required for
50% unblocking (t1=2) are indicated in Table 1.
10. Kochetkov, N. K.; Budovskii, E. I. Organic Chemistry of
Nucleic Acids. Part B; Plenum: London and New York,
1970; pp 425–448.
In conclusion, we recommend that the 50-O-DMTr
group 1 should be replaced either by the 50-O-Px or by
the 50-O-Tx protecting group in the solid phase synthesis
both of DNA and RNA sequences. In reaching this
conclusion, it should be borne in mind that, if solid
phase synthesis is to be carried out with 50-O-Px- or
50-O-Tx-protected phosphoramidites, and perhaps this is
also true for 50-O-DMTr-protected phosphoramidites, it
may be advisable to use a less acidic activating agent
than 1H-tetrazole (pKa 4.8),21 such as 1-phenylimidazo-
lium triflate (pKa 6.2)21 or imidazolium perchlorate
(pKa 7.0).21 This should ensure that absolutely no
50-unblocking occurs during the coupling process, even
in the synthesis of RNA sequences when coupling times
tend to be relatively long.21
11. Reese, C. B. In Nucleic Acids and Molecular Biology;
Eckstein, F., Lilley, D. M. J., Eds.; Springer: Berlin, 1989;
Vol. 3, pp 164–181.
12. Ogilvie, K. K.; Sadana, K. L.; Thompson, A. E.; Quilliam,
M. A.; Westmore, J. B. Tetrahedron Lett. 1974, 2861–
2863.
13. Reese, C. B.; Thompson, E. A. J. Chem. Soc., Perkin
Trans. 1 1988, 2881–2885.
Acknowledgements
We thank Avecia Ltd for generous financial support.
14. Pitsch, S.; Weiss, P. A.; Jenny, L.; Steitz, A.; Wu, X. Helv.
Chim. Acta 2001, 84, 3773–3795.
References and notes
15. Lloyd, W.; Reese, C. B.; Song, Q.; Vandersteen, A. M.;
Visintin, C.; Zhang, P.-Z. J. Chem. Soc., Perkin Trans. 1
2000, 165–176.
16. Rao, M. V.; Reese, C. B.; Schehlmann, V.; Yu, P. S.
J. Chem. Soc., Perkin Trans. 1 1993, 43–55.
1. Schaller, H.; Weimann, G.; Lerch, B.; Khorana, H. G.
J. Am. Chem. Soc. 1963, 85, 3821–3827.
2. Current Protocols in Nucleic Acid Chemistry; Beaucage,
S. L., Bergstrom, D. E., Glick, G. D., Jones, R. A., Eds.;
Wiley: New York, 2000–2002; Vol. 1.
17. Capaldi, D. C.; Reese, C. B. Nucleic Acids Res. 1994, 22,
2209–2216.
3. Chattopadhyaya, J. B.; Reese, C. B. J. Chem. Soc., Chem.
Commun. 1978, 639–640.
4. Gaffney, P. R. J.; Liu, C.; Rao, M. V.; Reese, C. B.; Ward,
J. G. J. Chem. Soc., Perkin Trans. 1 1991, 1355–1360.
5. 50-Protected 30-O-acetylthymidine derivatives 4a–c were
chosen as model substrates rather than the corresponding
50-protected thymidine derivatives 3a–c for two reasons.
First, in solid phase oligonucleotide synthesis, the terminal
nucleoside residues are always acylated on their 30-hydroxy
functions and this would be expected to affect the
50-unblocking rates. Secondly, HPLC analysis of the
partially-unblocked mixtures was facilitated by using
acetylated substrates. The three substrates 4a–c were all
prepared in greater than 90% yields by treating the
corresponding 50-protected thymidine derivatives 3a–c
with acetic anhydride in pyridine solution. Their
1H NMR spectra (360 MHz, in (CD3)2SO) are as follows:
18. Ribonucleoside substrates 9a–c were prepared in 86–90%
yield from 4-N-benzoyl-20-O-[1-(4-chlorophenyl)-4-ethoxy-
piperidin-4-yl]-30-O-levulinylcytidine19 10 and DMTr–Cl,
Px–Cl and Tx–Cl, respectively. Their 1H NMR spectra
(360 MHz, in (CD3)2SO) are as follows: 9a [0.96 (3H, t,
J 7.0), 1.60 (1H, m), 1.83 (3H, m), 2.10 (3H, s), 2.56 (2H,
m), 2.72 (2H, m), 2.89 (1H, m), 3.04 (1H, m), 3.13 (1H, m),
3.25 (3H, m), 3.36 (1H, m), 3.47 (1H, dd, J 4.1 and 10.7),
3.75 (6H, s), 4.20 (1H, m), 4.86 (1H, m), 5.27 (1H, m), 6.20
(1H, d, J 7.4), 6.92 (6H, m), 7.16–7.42 (12H, m), 7.52 (2H,
m), 7.64 (1H, m), 8.01 (2H, m), 8.17 (1H, m), 11.38 (1H,
br)]; 9b [0.97 (3H, t, J 6.9), 1.64 (1H, m), 1.83 (3H, m),
2.10 (3H, s), 2.53 (2H, m), 2.70 (2H, m), 2.91 (1H, m), 3.03
(1H, m), 3.10–3.42 (6H, m), 4.18 (1H, m), 4.80 (1H, m),
5.18 (1H, m), 6.17 (1H, d, J 7.2), 6.93 (2H, d, J 9.1),

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
2570
C. B. Reese, H. Yan / Tetrahedron Letters 45 (2004) 2567–2570
7.14–7.47 (16H, m), 7.53 (2H, m), 7.64 (1H, m), 8.01 (2H,
m), 8.16 (1H, m), 11.38 (1H, br)]; 9c [0.93 (3H, t, J 7.0),
1.65 (1H, m), 1.83 (3H, m), 2.10 (3H, s), 2.26 (3H, s), 2.52
(2H, m), 2.70 (2H, m), 2.92 (1H, m), 3.04–3.42 (7H, m),
4.18 (1H, m), 4.79 (1H, m), 5.17 (1H, m), 6.17 (1H, d, J
7.2), 6.93 (2H, d, J 9.1), 7.11–7.47 (15H, m), 7.53 (2H, m),
7.64 (1H, m), 8.02 (2H, d, J 7.3), 8.17 (1H, m), 11.38 (1H,
br)].
19. Reese, C. B.; Yan, H. Unpublished observations.
20. The experimental procedure used for the determination of
the rates of 50-deprotection of the ribonucleoside sub-
strates 9a–c was essentially the same as that used above9 in
the 50-deprotection of the thymidine derivatives 4a–c.
21. Hayakawa, Y.; Kawai, R.; Hirata, A.; Sugumoto, J.;
Kataoka, M.; Sakakura, A.; Hirose, M.; Noyori, R.
J. Am. Chem. Soc. 2001, 123, 8165–8176.