Pyrrolidino DNA with bases corresponding to the 2-oxo deletion mutants of thymine and cytosine: Synthesis and triplex-forming properties

Article abstract of DOI:10.1002/ejoc.200700217

The dual recognition properties of pyrrolidino DNA species as parallel triplex-forming oligonucleotides were previously found to be strongly dependent upon the nature of the pyrimidine bases. In the structure-activity study presented here we were able to

Full text of DOI:10.1002/ejoc.200700217

FULL PAPER
DOI: 10.1002/ejoc.200700217
Pyrrolidino DNA with Bases Corresponding to the 2-Oxo Deletion Mutants of
Thymine and Cytosine: Synthesis and Triplex-Forming Properties
Alain Mayer[a] and Christian J. Leumann*[a]
Keywords: Oligonucleotides / Pyrrolidino DNA / Antigene agents / Dual recognition / Nucleoside analogues
The dual recognition properties of pyrrolidino DNA species
as parallel triplex-forming oligonucleotides were previously
found to be strongly dependent upon the nature of the pyrim-
idine bases. In the structure–activity study presented here we
were able to exclude this differential binding being due to
their 2-oxo function. We had previously reported on the in-
corporation of pyrrolidino C-nucleosides into triplex-forming
2Ј-deoxyoligonucleotides (TFOs). The basic nitrogen atom
that replaces the 4Ј-oxygen atom of the 2Ј-deoxysugar in
such modified units introduces a positive charge in the third
strand, and this is able to produce favourable electrostatic
interaction with the negatively charged DNA target duplex.
A first series of pyrrolidino pseudonucleosides with the bases
isocytosine and uracil proved successful for GC base-pair re-
cognition, but was unsuccessful for AT base-pair recognition
within the parallel triplex binding motif. Here we report on
the synthesis of the two novel 2Ј-deoxypyrrolidino nucleo-
sides carrying the bases pyridin-2-one and 2-aminopyridine,
their phosphoramidite building blocks and their
incorporation into TFOs. Pyrrolidinylpyridin-2-one (dp2P)
and -2-aminopyridine (dp2AP), prepared as part of a struc-
ture–activity profiling of pyrrolidino DNA in triplex binding,
are deletion mutants of T and C, respectively. We found by
Tm measurements that neither modification increased triplex
binding efficiency relative to the iso-C- and -U-containing
pyrrolidino TFOs. These experiments clearly show that the
C4 carbonyl function, although important for triplex binding
through indirect contributions in general, is not responsible
for the differential binding of the latter two aminonucleosides
and suggest that TFO conformation is more important.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
(and at physiological pH completely protonated) ana-
logue[10–13] or with a charge-neutral analogue of cyto-
sine.[14–18] Several approaches address the targeting of pyr-
imidine-purine inversion sites with high affinity. While rea-
sonable solutions have been found for CG recognition,[19,20]
no fully satisfactory solution for TA base pair recognition
is so far available.[21–24]
Cuenoud and co-workers recently introduced the dual re-
cognition concept to increase triplex stability.[25,26] Modi-
fied TFOs containing an aminoethoxy side-chain at C2Ј of
the ribose units were prepared in order to generate positive
charges within the third strand. The target dsDNA can then
be recognised not only through selective base–base interac-
tions but also through additional, nonspecific salt bridges
between these appropriately placed positive charges in the
TFO and negatively charged backbone phosphate oxygen
atoms.
We recently developed a similar dual recognition ap-
proach based on pyrrolidino nucleosides. Replacement of
the 4Ј-oxygen atom in a deoxynucleoside of a TFO by a
basic nitrogen atom, protonated at physiological pH, intro-
duces an ideally located positive charge that can interact
with a negatively charged pro-(R)-phosphate oxygen atom
of the purine strand of a target dsDNA. Pyrrolidino pseu-
donucleosides containing the bases uracil (dpψU) or N-1-
methyl uracil (dpψT) for AT base-pair recognition, together
with the base isocytosine (dpψiC) for GC base-pair recogni-
Introduction
Stable and selective binding to genomic DNA by an oli-
gonucleotide through triple-helix formation allows the flow
of the genetic information in the cell to be modulated di-
rectly at the level of DNA. A triplex-forming oligonucleo-
tide (TFO) binds to the exposed purine bases in the major
groove of a DNA duplex either by Hoogsteen or by reverse-
Hoogsteen base-pairing in the parallel or antiparallel tri-
plex-binding motifs, respectively.[1] The so-called antigene
strategy in theory confers complete control of gene expres-
sion and therefore constitutes an ideal method for gene
therapy.[2–6] In reality, however, this strategy suffers from a
number of significant limitations. Some are those encoun-
tered with most classical drugs, such as cellular uptake or
in vivo stability, while others – such as pH dependence, re-
striction to homopurine target sites, and low thermal triplex
stability, for instance – are specific to antigene agents. A
variety of strategies to overcome these weaknesses and to
improve the efficiency and applicability of antigene agents
have been developed in the past.[7–9] The pH dependence,
originating from the necessity of cytosine N-3 protonation,
can be eliminated by replacement either with a more basic
[a] Department of Chemistry & Biochemistry, University of Bern,
Freiestrasse 3, 3012 Bern
Fax: +41-31-631-3422
E-mail: leumann@ioc.unibe.ch
4038
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 4038–4049

Pyrrolidino DNA with 2-Oxo Deletion Mutants of Thymine and Cytosine
FULL PAPER
Figure 1. Top left: previously studied pyrrolidino pseudonucleoside analogues of deoxyuridine (dpψU), thymidine (dpψT) and deoxycytid-
ine (dpψiC); top right: O4 deletion mutants dp2P and dp2AP of the corresponding nucleoside analogues; bottom: expected triplex pairing
patterns.
tion within the parallel triplex binding motif, were explored
(Figure 1). The dpψiC unit was demonstrated to be a suc-
Results and Discussion
cessful substitute for cytidine in TFOs, resulting in increases
Synthesis of Phosphoramidite Building Blocks
in thermal triplex stability of about 2–2.5 °C per modifica-
tion.[27] The TFOs containing the corresponding pyrrolid-
Of the different methods available for the preparation of
ino nucleosides developed to replace T (dpψU and dpψT) C-nucleosides[29] and imino-sugar-based C-nucleosides,[30]
instead showed target affinities inferior to those of the natu- our efforts first concentrated on palladium-catalysed Heck
ral reference for an as yet unknown reason.[28]
couplings of a pyrroline sugar analogue with the corre-
To determine the effect of the C4-carbonyl function that sponding aglycons. This reaction had been previously been
is not directly involved in hydrogen bonding to the corre- successfully applied for the synthesis of the first generation
sponding target purine base, pyrrolidino C-nucleosides of pyrrolidino pseudonucleosides.[31] However, reactions be-
bearing pyridin-2-one (dp2P) and 2-aminopyridine tween enamine 1 (Figure 2) and different halogenated deriv-
(dp2AP) units as nucleobases were designed. Pyridin-2-one atives of the required aglycons with use of various P- and
represents the most simplified T analogue, keeping only the As-ligands and different reaction conditions never pro-
minimum of functional groups required for T–AT base trip- duced the desired nucleosides. Alternative methods involv-
let formation. 2-Aminopyridine had already previously ing the addition of the organometallic derivative of the nuc-
been shown to be an excellent cytosine analogue in parallel leobases variously to the corresponding lactam 2,[32,33] to
triplexes in the deoxyribo backbone series.[10–13] In this arti- the corresponding hemiaminal 3 or to the N-acyliminium
cle we report on the synthesis of the phosphoramidite build- ion[34–38] obtained in situ from 3 proved unsuccessful as
ing blocks of dp2P and dp2AP, their incorporation into well.
TFOs as substitutes for T and C, respectively, and the
A number of pyrrolidino C-nucleosides with aromatic
evaluation of their triplex pairing properties. Finally we (hetero)cycles, such as substituted phenyl groups, imid-
compare the results with those observed with the first gen- azoles or 9-deazapurines, as aglycons were synthesised by
eration of pyrrolidino pseudonucleosides.
Schramm and co-workers as transition state inhibitors for
Figure 2. Chemical structures of the different N-glycosyl donors used in C-glycoside formation.
Eur. J. Org. Chem. 2007, 4038–4049
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4039

A. Mayer, C. J. Leumann
FULL PAPER
Scheme 1. Synthesis of imine 4. (a) Fmoc-Cl, dioxane, 5% NaHCO3 soln., 0 °C to room temp., 9 h, 98%; (b) BH3·(CH3)2S, THF, reflux,
2 h, 97%; (c) TBDMS-Cl, imidazole, THF, room temp., 2 h; (d) piperidine, THF, room temp., 12 h, 90% over 2 steps; (e) NCS, hexane,
room temp., 1 h; (f) LTMP, THF, –78 °C, 2 h, 74% over two steps.
nucleoside hydrolases or nucleoside phosphatases. The agly- the next step. The benzyl protecting group was removed by
con was introduced through addition of the corresponding catalytic hydrogenolysis and the pure β anomer of 14 was
aryllithium or aryl-Grignard reagents to the imine function isolated in 35% yield over two steps. The configuration at
1
of substituted 3,4-dihydro-2H-pyrroles.[39–42] A similar the pseudo-anomeric centre was determined by H NMR-
strategy with imine 4 as sugar component (Figure 2) finally NOE experiments: a strong mutual NOE was observed be-
proved also to be successful in the synthesis of dp2P and tween 2-H and 5-H of the pyrrolidine ring, establishing the
dp2AP.
β configuration.
The imine 4 was synthesised in six steps and 62% overall
The Fmoc group had previously been shown to be com-
patible with oligonucleotide synthesis and was therefore
used to protect the pyrrolidine nitrogen atom. Bevers et al.
reported the necessity of protecting the pyridin-2-one oxy-
gen atom in oligonucleotide synthesis without giving fur-
ther details.[43] In preliminary experiments we also experi-
enced side reactions between the unprotected pyridone moi-
ety and the incoming phosphoramidite building block dur-
ing oligonucleotide synthesis, dramatically decreasing the
efficiency of the synthesis. Therefore, protection of the pyri-
done seemed to be required, and the 2-(p-nitrophenyl)ethyl
group (NPE) was chosen for this purpose. The NPE group
was incorporated through the Mitsunobu reaction. Al-
though all the starting material was consumed and only one
new product seemed to be formed, the fully protected com-
pound 16 was obtained only in a rather low yield of 35%.
The silyl ethers were cleaved off with acidic methanol to
afford the diol 17 in 67% yield. Standard tritylation of the
primary alcohol (Ǟ 18), followed by phosphitylation, fi-
nally afforded phosphoramidite building block 19.
yield starting from trans-3-hydroxy--proline (Scheme 1).
The amino function of the starting material was Fmoc-pro-
tected and the carboxylic acid was then selectively reduced
to the primary alcohol to provide diol 7. Both hydroxy
groups were converted into TBDMS ethers and the pyrroli-
dino ring nitrogen atom was deprotected. Treatment of in-
termediate 8 with NCS in hexane resulted in the N-chloro
analogue. Subsequent LTMP-mediated elimination of HCl
at –78 °C by analogy to published procedures[39,40] afforded
imine 4 in good yield. Although similar imines had been
reported to decompose easily, 4 was isolated by column
chromatography and proved to be stable at room temp.
The appropriately protected aglycons were prepared as
shown in Scheme 2. 5-Bromo-2-hydroxypyridine (9) was se-
lectively O-benzyl-protected to yield 10 by treatment with
benzyl bromide and silver carbonate, whereas the 2-amino
derivative 11 was converted into its silyl derivative 12.
A similar route was applied for the synthesis of dp2AP
and its phosphoramidite derivative (Scheme 4). The stabase
adduct of 2-amino-5-bromopyridine (12) was subjected to
bromo–lithium exchange followed by treatment with imine
4 to furnish the β-pyrrolidino pseudonucleoside 20 as deter-
mined after purification by 1H NMR-NOE experiments
(see Experimental Section). The stabase protective group
was hydrolysed during column chromatography. Fmoc pro-
tection was performed selectively at the pyrrolidine nitrogen
atom. For the exocyclic amino group of 2-aminopyridine,
classical protective groups suitable for oligonucleotide syn-
Scheme 2. Synthesis of the aglycons 10 and 12. (a) BnBr, Ag2CO3,
THF, reflux, 5 h, 93 %; (b) butyllithium, Cl(Me)2SiCH2CH2Si-
(Me)2Cl, THF, –78 °C, 2 h30, 79%.
Treatment of benzyl-protected pyridone 10 with butyl- thesis (acetyl, benzoyl, etc) had been reported to exhibit
lithium and subsequent addition of imine 4 to the obtained increased stability against ammonolysis and not to be re-
lithio derivative at –78 °C resulted in the formation of a movable under classical oligonucleotide deprotection condi-
1:10 α/β mixture of the desired product 13 in 50% yield tions.[11] In order to avoid resorting to harsher deprotection
(Scheme 3). The separation of the two anomers presented conditions, the more labile phenoxyacetyl (Pac) group was
some difficulties at this stage and was conducted only after incorporated on the primary amine and intermediate 22
4040
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 4038–4049

Pyrrolidino DNA with 2-Oxo Deletion Mutants of Thymine and Cytosine
FULL PAPER
Scheme 3. Synthesis of phosphoramidite 19. (a) Butyllithium, THF, –78 °C, 1 h; (b) 4, THF, –78 °C; (c) 10% Pd/C, H2, MeOH, room
temp., 30 min., 35% over two steps; (d) Fmoc-OSu, THF, dioxane, 1  aq. NaHCO3, room temp., 1 h 30, 80%; (e) PPh3, DIAD, p-
nitrophenylethanol, dioxane, room temp., 8 h, 35%; (f) 1  aq. HCl/MeOH 1:10, room temp., 24 h, 67%; (g) DMTCl, pyridine, room
temp., 5 h, 74%; (h) CEPCl, iPr2NEt, THF, room temp., 2 h, 82%.
Scheme 4. Synthesis of phosphoramidite 25. (a) Butyllithium, THF, –78 °C, 1 h; (b) 4, THF, –78 °C, 2 h, 19%; (c) Fmoc-OSu, THF,
dioxane, 1  aq. NaHCO3, room temp., 30 min, 75%; (d) Pac20, pyridine, room temp., 2 h, 99%; (e) 1  aq. HCl/MeOH 1:10, room
temp., 24 h, 99%; (f) DMTCl, pyridine, room temp., 2 h, 88%; (g) CEPCl, iPr2NEt, THF, room temp., 1 h30, 90%.
was formed in excellent yield. The diol 23 was obtained ment with piperidine (Scheme 5). Pyrrolidino C-nucleoside
quantitatively from 22 by acidic methanolysis of the silyl 27 was fully characterised and the β configuration at the
1
ethers. DMT protection afforded compound 24, and intro- anomeric centre was confirmed by H NMR-NOE experi-
duction of the phosphoramidite moiety yielded the building ments, indicating that no anomerisation had occurred dur-
block 25.
ing any step of the synthesis, in particular during the
Pac-deprotected product 26, formed on extended expo- TBDMS deprotection step performed under strongly acidic
sure of compound 22 to acidic conditions, was isolated and conditions (see Experimental Section). The phosphoramid-
converted into the fully deprotected dp2AP (27) by treat- ite building blocks of dp2P and dp2AP (19 and 25, respec-
Scheme 5. Synthesis of dp2AP. (a) 1  aq. HCl/MeOH 1:10, room temp., 30 h, 76% of 23 and 23% of 26; (b) piperidine, DMF, room
temp., 12 h, 91%.
Eur. J. Org. Chem. 2007, 4038–4049
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4041

A. Mayer, C. J. Leumann
FULL PAPER
tively) were then incorporated into oligonucleotides by ing properties of the modified oligodeoxynucleotides were
automated solid-phase oligonucleotide synthesis and their subsequently examined by UV melting curve analysis.
triplex-forming affinities were evaluated.
Pairing Properties of dp2P-Containing TFOs with DNA
Synthesis of Oligodeoxynucleotides
Duplexes in the Parallel Motif
Oligodeoxynucleotides were synthesised by standard so-
The melting temperatures (Tms) for TFO dissociation
lid-phase phosphoramidite chemistry with a DNA syn- from a 21-mer dsDNA target are summarised in Table 2.
thesiser. Minor modifications to the synthesis cycle were in- Relative to the natural third strand Ref1, TFO dp1 dis-
troduced for the incorporation of the unnatural building played a significant loss in triplex stability of about 11 °C
blocks. More precisely, the coupling time was extended per modification (Entry 4). This corresponds to a further
from 1.5 to 6 min and the standard activator tetrazole was destabilisation relative to dpψU- and even dpψT-containing
replaced by the more powerful (S-benzylthio)-1H-tetrazole. triplexes (Entries 2 and 3, respectively).
For the sequences containing dp2AP, the classical acetyl
This result was confirmed by circular dichroism (CD)
capping step was replaced by a phenoxyacetyl (Pac) capping spectroscopy. The CD spectra of the triplexes were mea-
step. Coupling efficiencies for the modified units were typi- sured at different temperatures below and above the third
cally Ͼ97%, according to trityl assay. Standard ammonia strand melting temperature. The resulting curves exhibited
deprotection resulted in cleavage of the crude oligonucleo- almost identical shapes except in the region below 235 nm,
tides from the solid support and removal of all protecting with a maximal change in the CD spectral properties at
groups (including Fmoc). For all sequences containing 220 nm (Figure 3 left). Consequently, the variation of the
NPE protective groups, a two-step deprotection procedure CD intensity at a fixed wavelength of 220 nm was moni-
first involving treatment with 1,5,7-triazabicyclo[4.4.0]dec- tored between 10 and 50 °C and a sigmoidal curve was ob-
5-ene (TBD) in acetonitrile (1 ) at 60 °C for 8 h, followed tained (Figure 3, right). The first derivative of this curve
by ammonolysis, was applied. Use of a weaker base (DBU, gave a Tm value of 32 °C, which is in perfect agreement with
0.5  in pyridine or 1  in acetonitrile), lower temperatures the one determined by UV-melting experiments.
or shorter reaction times resulted in incomplete NPE re-
To determine target binding selectivity, another target
moval. Isomerisation at the anomeric centre by elimination/ duplex (Target 2, Table 3) exposing a guanine instead of an
addition of the pyrrolidine ring nitrogen atom under these adenine opposite to the pyrrolidine analogue was prepared.
conditions could be excluded on the basis of an NMR ex- The data obtained from melting curve experiments are sum-
periment with the monomer 17.
marised in Table 3, together with the Tm values measured
Crude oligomers were purified by ion-exchange HPLC with the matched duplex (Target 1) for comparison. The
and characterised by ESI– mass spectrometry. A summary unmodified TFO Ref1 used as reference gave a strongly de-
of the modified oligomers synthesised and their mass spec- stabilised triplex with a Tm of about 30 °C and a ∆Tm of
trometric analysis is provided in Table 1. The triplex-form- –12.6 relative to Target 1, as expected for a mismatch situa-
Table 1. TFO sequences and ESI-MS data for the oligonucleotides investigated in this study.
TFO
Sequence
Mod.
[M – H]– (found)
[M – H]– (calcd)
dp1
dp2
dp3
dp4
dp5
5Ј-d(TTTTXCTCTCTCTCT)
5Ј-d(TTTTTCTXTCTCTCT)
5Ј-d(TTTTTXTCTXTCTCT)
5Ј-d(TTTTTXTXTXTXTXT)
5Ј-d(TTTXXXXTTTTXTTT)
dp2P
4393.0
4407.1
4389.5
4334.8
4335.0
4392.9
4406.9
4388.9
4335.0
4335.0
dp2AP
dp2AP
dp2AP
dp2AP
Table 2. Sequence of TFOs and Tm data (°C) for third strand dissociation from UV melting curves (260 nm).[a]
[b]
Entry
TFO
Sequence
Tm
∆Tm per modification.
1
2
3
4
Ref1
Ref2
Ref3
dp1
5Ј-d(TTTTTCTCTCTCTCT)
5Ј-d(TTTTXCTCTCTCTCT)
5Ј-d(TTTTYCTCTCTCTCT)
5Ј-d(TTTTZCTCTCTCTCT)
43.1
38.4
37.0
32.0
0
–4.7
–6.1
–11.1
[a] Target duplex = d(GCTAAAAAGAGAGAGAGATCG)·d(CGATCTCTCTCTCTTTTTAGC), Tm = 57.0Ϯ1.0 °C. [b] Single strand
concentration = 1.2 µ in 140 m KCl, 7 m NaH2PO4, 0.5 m MgCl2, pH 6.0.
4042
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 4038–4049

Pyrrolidino DNA with 2-Oxo Deletion Mutants of Thymine and Cytosine
FULL PAPER
Figure 3. CD spectra of the triplex-containing TFO dp1 (Table 4); left: spectra measured at different temperatures; right: temperature
scan from 10 to 50 °C at 220 nm; single strand concentration = 1.2 µ in 140 m KCl, 7 m NaH2PO4, 0.5 m MgCl2, pH 6.0.
Table 3. Sequence of TFOs and Tm data (°C) for third strand dissociation from UV melting curves (260 nm).
Target 1
3Ј-d(CGATTTTTCTCTCTCTCTAGC)
5Ј-d(GCTAAAAAGAGAGAGAGATCG)
5Ј-d(TTTTXCTCTCTCTCT)
Target 2
3Ј-d(CGATTTTCCTCTCTCTCTAGC)
5Ј-d(GCTAAAAGGAGAGAGAGATCG)
5Ј-d(TTTTXCTCTCTCTCT)
TFO
Entry
TFO
TFO
X
Tm[a](Target 1)
Tm[a](Target 2)
∆Tm
1
2
3
Ref1
Ref2
dp1
T
dpψU
dp2P
43.1
38.4
32.0
30.5
30.5
29.5
–12.6
–7.9
–2.5
[a] Single strand concentration = 1.2 µ in 140 m KCl, 7 m NaH2PO4, 0.5 m MgCl2, pH 6.0. Tm of target duplex 1 = 57.0Ϯ1.0 °C.
Tm of target duplex 2 = 61.0Ϯ1.0 °C.
tion (Entry 1). The dpψU-containing TFO Ref2 and the du- well known for cytosines: the relative stabilities of triple hel-
plex Target 2 formed a triplex that was also substantially ices containing C+-GC triplets drop as the number of adja-
less stable than that with the duplex Target 1 (Entry 2). The cent C+-GC triplets increases, due to a further decrease in
Tm value was in the same range as for the natural triplex, the pKa value of cytosine N3 and local repulsion of contigu-
indicating that this system presented a mismatch as well. ous positive charges in the pile of protonated cytosines.[44]
Not much difference in thermal stability between match and However, these composition effects have been reported to
mismatch was encountered in the case of the triplex with be less pronounced for 2-aminopyridine C-nucleosides.[11] It
dp1 (Entry 3). The pyrrolidinylpyridin-2-one unit thus appears to be verified in our case, since the additional loss
seems to recognise the AT base pair within the parallel in stability is low.
binding motif with very low affinity.
More interesting is the pH variation study of the dp2AP
unit. Triplexes containing TFOs dp2 and dp3 were strongly
affected by the pH value: an increase in pH from 6.0 to 7.0
resulted in a significant drop in Tm (Entries 2–3 and 4–5),
which was consistent with the presence of pH-dependent
functionalities in the TFO (cytosine and 2-aminopyridine
bases, as well as pyrrolidine units). However, the pH depen-
dence seemed to arise mainly from the cytosine, since the
replacement of a natural C with a dp2AP unit (dp2 Ǟ dp3)
resulted in a lower Tm decrease when the pH was increased
from 6.0 to 7.0. This suggestion was confirmed with the
TFOs dp4 and dp5, with all natural cytosines replaced by
The missing C4-carbonyl oxygen atom thus seems to play
an important role in triplex stability. It may positively influ-
ence the stacking interactions and/or the solvation proper-
ties. From the experiments presented here, however, it can
be ruled out that this oxygen atom may be responsible for
the loss in affinity of a dpψU (or dpψT) residue relative to
an unmodified T as reported previously.
Pairing Properties of dp2AP-Containing TFOs with DNA
Duplexes in the Parallel Motif
At pH 6.0, all the modified TFOs (Table 4) gave triple dp2AP moieties. These produced triple helical structures
helical structures exhibiting diminished stability relative to that showed no drop in Tm over the same pH range, despite
the natural system. However, the destabilisation decreases the five pyrrolidinyl-2-aminopyridine units (Entries 6–7 and
with an increased number of modifications, from –5 °C per 13–14). Only at pH Ͼ 8 were no triplexes formed any more
modification in the case of monomodified TFO dp2 to (Entries 8–9 and 15–16). Although the dp2AP gave less
–3 °C per modification for TFO dp4 containing five modifi- stable triplexes than the references at pH 6.0, the weaker
cations (Entries 2 and 6, respectively). In a different se- pH dependence provided an advantage at pH 7.0. TFO dp5,
quence context, runs of four contiguous dp2AP units led to for instance, exhibited a loss in stability of 4.3 °C per modi-
further destabilisation (dp5, Entry 9). This phenomenon is fication at pH 6.0, but at pH 7.0 it formed a triplex with a
Eur. J. Org. Chem. 2007, 4038–4049
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4043

A. Mayer, C. J. Leumann
FULL PAPER
Table 4. Sequence of TFOs and Tm data (°C) for third strand dissociation from UV melting curves (260 nm).[a]
[b]
Entry
TFO
Sequence
Target duplex
pH
Tm
∆Tm per modification
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Ref1
dp2
5Ј-d(TTTTTCTCTCTCTCT)
5Ј-d(TTTTTCTXTCTCTCT)
1
1
6.0
6.0
7.0
6.0
7.0
6.0
7.0
8.0
9.0
6.0
7.0
8.0
6.0
7.0
8.0
9.0
43.1
38.0
19.1
36.1
22.1
26.9
27.0
22.1
n.d.[c]
27.2
n.d.[c]
n.d.[c]
5.7
0
–5.1
–
–3.5
–
–3.2
–
–
–
0
0
0
dp3
dp4
5Ј-d(TTTTTXTCTXTCTCT)
5Ј-d(TTTTTXTXTXTXTXT)
1
1
Ref5
dp5
5Ј-d(TTTCCCCTTTTCTTT)
5Ј-d(TTTXXXXTTTTXTTT)
2
2
–4.3
–
–
7.0
n.d.[c]
n.d.[c]
–
[a] C = 5-methylcytosine, X = dp2AP; Target duplex 1 = d(GCTAAAAAGAGAGAGAGATCG)·d(CGATCTCTCTCTCTTTTTAGC),
Tm = 57.0 Ϯ 1.0 °C; Target duplex 2 = d(GCTAAAGGGGAAAAGAAATCG)·d(CGATTTCTTTTCCCCTTTAGC), Tm
=
60.0 °CϮ1.0 °C. [b] Single strand concentration = 1.2 µ in 140 m KCl, 7 m NaH2PO4, 0.5 m MgCl2, pH 6.0. [c] No Tm detectable.
similar Tm, whereas the reference Ref5 showed no detect- plex recognition, possibly through favourable stacking in-
able affinity for the target duplex (Entries 10–11 and 13– teractions and/or solvation effects.
14).
More importantly, these results clearly show that the C4-
Incorporation of dp2AP into TFOs gives rise to triplexes carbonyl function is not responsible for the differential sta-
displaying astonishingly low pH dependence in the pH 6–8 bilities of dpψiC-GC and dpψU/T-AT base triplets. Since
range, relative to corresponding systems containing cytid- the dual recognition motif strongly relies on electrostatic
ines, methylcytidines or any pyrrolidino nucleosides studied interactions between an amino and a phosphate function,
so far. Both the pyrrolidine and the pyridine nitrogen atoms and since such interactions are strongly distance-dependent,
seem to benefit from a high protonation state up to pH 8, it seems more likely that local structural differences between
generating triplexes that would not suffer from pH effects the dpψiC-GC and the dpψU-AT (or dpψT-AT) triplets, or
under physiological conditions. This remarkable property is of the triple helical conformation in general, could result in
compromised, however, by the triplex destabilisation prop- variations in the N-O(P) distances that are important
erties induced by dp2AP. The missing carbonyl group might enough to result in the loss of the favourable electrostatic
also play a role in these negative effects here for the same contact in the latter case. Pyrrolidine-modified TFOs with
reasons as given for dp2P.
an RNA backbone, which would be expected to give triple
helices of the A conformation based on the known 3Ј-endo
conformation of the nucleosides, might therefore shed light
on the influence of conformational parameters on triplex
stability. Corresponding experiments are currently un-
derway in our laboratory.
Conclusions
Pyrrolidino C-nucleosides have the potential to be excel-
lent candidates for the dual recognition approach in DNA
duplex recognition as constituents of parallel TFOs. How-
ever, positive evidence is so far limited to pyrrolidino pseu-
doiso-C as a nucleobase, forming a stable dpψiC-GC trip-
let. In order to promote pyrrolidino DNA as an antigene
agent the critical parameters for effective and general dual
recognition still need to be identified and tuned appropri-
ately.
Of the multitude of factors that can be modified to in-
crease the binding efficiency of pyrrolidino DNA, that of
base variation was examined in this study. More specifically,
pyrrolidinylpyridin-2-one (dp2P) and -2-aminopyridine
(dp2AP), both deletion mutants – of dpψU and dpψiC,
respectively – were developed as tools to test the impor-
tance of the C4-carbonyl oxygen atom in triplex affinity. We
found a general tendency towards decreased triplex stability
in both cases. The obtained results show that this functional
Experimental Section
General: Reactions were carried out under Ar in anhydrous sol-
vents. Solvents for extractions were technical grade and were dis-
tilled prior to use. Reagents were purchased from Fluka AG, Ald-
rich and Merck (highest quality available). NMR spectra were mea-
sured with a Bruker AC 300 or a Bruker DRX 400 spectrometer.
1H NMR spectra were recorded at 300 or 400 MHz. Chemical
shifts are reported in ppm relative to the residual undeuterated
NMR solvent (CHCl3: δ = 7.27 ppm, [D5]DMSO: δ = 2.50 ppm,
CD2HOH: δ = 3.35 ppm, HDO: δ = 4.8 ppm). Coupling constants
J are in Hz. 13C NMR spectra were recorded at 75 or 100 MHz.
Chemical shifts are reported in ppm relative to the residual undeu-
terated NMR solvent (CHCl3: δ = 77.00 ppm, [D5]DMSO: δ =
39.70 ppm, CD2HOD: δ = 49.30 ppm). 31P NMR spectra were re-
corded at 162 MHz. Chemical shifts are reported in ppm relative
1
to 85% H3PO4 as an external standard. H NMR difference-NOE
group plays an important, general, but indirect, role in tri- experiments were recorded at 400 MHz. Electron impact (EI) mass
4044 Eur. J. Org. Chem. 2007, 4038–4049
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Pyrrolidino DNA with 2-Oxo Deletion Mutants of Thymine and Cytosine
FULL PAPER
spectra were recorded with an AutoSpeq Q VG instrument with an
ionisation energy of 70 eV. Fast atom bombardment (FAB+) mass
spectra were also recorded with an AutoSpeq Q VG. Electrospray
ionisation (ESI) mass spectra were recorded either with a Fisons
Instrument VG Platform (oligonucleotides) or with an Applied Bi-
osystems, Sciex QSTAR Pulsar (monomers). TLC was performed
with precoated silica gel plates (SIL G-25 UV254) from Macherey–
Nagel. Visualisation was achieved by use of UV light (254 nm) or
staining solutions (cerium reagent, anisaldehyde reagent, ninhydrin
reagent). Flash chromatography (FC) was performed with silica
gel 60 (particle size 40–63 µm) from Fluka. The compounds based
on C-nucleosides dp2P and dp2AP were named as substituted pyr-
1 H, HO-C-3), 7.33 (t, JH,H = 7.2 Hz, 2 H, Fmoc-H), 7.42 (t, JH,H
= 7.0 Hz, 2 H, Fmoc-H), 7.66 (d, JH,H = 7.4 Hz, 2 H, Fmoc-H),
7.90 (d, JH,H = 7.4 Hz, 2 H, Fmoc-H) ppm. MS (FAB+): m/z (%)
= 340 (100, [M + H]+), 309 (22), 179 (60), 155 (52), 119 (83).
HR-MS (FAB+, [M + H]+): 340.15485 (C20H22NO4 requires:
340.15488).
(2R,3S)-3-{[(tert-Butyl)dimethylsilyl]oxy}-2-{[(tert-butyl)dimethyl-
silyl]oxymethyl}pyrrolidine (8): tert-Butyldimethylsilyl chloride
(11.7 g, 77.6 mmol) was added to a solution of 7 (12.0 g,
35.4 mmol) and imidazole (8.4 g, 123 mmol) in dry THF (150 mL).
After 2 h, piperidine (23 mL, 233 mmol) was added and stirring
was continued for 12 h. The solution was concentrated and the resi-
due was distributed between AcOEt and H2O. The org. layer was
dried (MgSO4) and the solvents were evaporated. FC (CH2Cl2/
MeOH/25% aq. NH3 50:1:1 to 25:1:1) afforded 11.0 g (31.8 mmol,
1
rolidines and numbered as described below (Figure 4). H and 13C
NMR spectroscopic data for compounds 15–19 and 21–26, bearing
Fmoc groups, were not reported, due to broad signals from coalesc-
ence (carbamate bond).
1
90%) of 8 as a yellow oil. H NMR (300 MHz, CDCl3, 25 °C): δ
= 0.06 (s, 12 H, 4ϫCH3-Si), 0.88, 0.89 (2ϫs, 18 H, 2ϫ(CH3)3C-
Si), 1.63–1.72 (m, 1 H, 4-H), 1.83–1.95 (m, 1 H, 4-H), 2.22 (s, 1
H, NH), 2.90–3.11 (m, 3 H, 2-H, 5-H), 3.54–3.64 (m, 2 H, CH2-
OTBDMS), 4.10–4.15 (m, 1 H, 3-H) ppm. 13C NMR (75 MHz,
CDCl3, 25 °C): δ = –5.50, –5.45, –4.76, –4.59 (4ϫq, 4ϫCH3-Si),
18.02, 18.27 (2ϫs, 2 ϫ(CH3)3C-Si), 25.82, 25.90 (2ϫq, 2ϫ(CH3)3-
C-Si), 35.28 (t, C-4), 44.82 (t, C-5), 63.12 (t, CH2-OTBDMS), 68.14
(d, C-2), 73.77 (d, C-3) ppm. MS (FAB+): m/z (%) = 346 (100, [M
+ H]+), 288 (8), 200 (17). HR-MS (FAB+, [M + H]+): 346.25998
(C17H40NO2Si2 requires: 346.25976).
Figure 4. Numbering scheme for pyrrolidino pseudonucleosides
13–19 derived from dp2P and 20–27 derived from dp2AP used for
NMR assignments.
(4S,5R)-4-{[(tert-Butyl)dimethylsilyl]oxy}-5-{[(tert-butyl)dimethyl-
silyl]oxymethyl}-1-azacyclopent-1-ene (4): N-Chlorosuccinimide
(500 mg, 3.74 mmol) was added to a solution of 8 (1.0 g,
2.89 mmol) in hexane (40 mL). After 1 h, the suspension was fil-
tered through Celite and the solvents were evaporated. The residue
was dissolved in dry THF (50 mL) and cooled to –78 °C. A solu-
tion of tetramethylpiperidine (1.18 mL, 6.97 mmol) in THF
(30 mL), previously treated with BuLi (1.6  in hexane, 4.06 mL,
6.50 mmol) at 0 °C and stirred at 0 °C for 30 min, was added to
the solution at –78 °C over 1 h until no more N-chloro compound
was detected by TLC. After 1 h, the mixture was warmed to room
temp., most of the THF was evaporated, and the residue was dis-
tributed between tert-butyl methyl ether and H2O. The org. layer
was dried (MgSO4) and the solvents were evaporated. FC (hexane/
AcOEt 5:1 to 4:1) gave 730 mg (74%, 2.12 mmol) of 4 as a yellow
oil. TLC (hexane/AcOEt 4:1): Rf 0.34. 1H NMR (300 MHz,
CDCl3, 25 °C): δ = 0.02, 0.04, 0.08 (3ϫs, 12 H, 4ϫCH3-Si), 0.87,
0.88 (2ϫs, 18 H, 2ϫ(CH3)3C-Si), 2.44 (d, JH,H = 18.4 Hz, 1 H, 3-
H), 2.75 (dd, JH,H = 6.2, 18.4 Hz, 1 H, 3-H), 3.56 (dd, JH,H = 4.8,
10.3 Hz, 1 H, CH2-OTBDMS), 3.86 (dd, JH,H = 3.3, 10.3 Hz, 1 H,
CH2-OTBDMS), 4.06 (m, 1 H, 4-H), 4.37 (d, 1 H, JH,H = 6.6 Hz,
5-H), 7.61 (s, 1 H, 2-H) ppm. 13C NMR (75 MHz, CDCl3, 25 °C): δ
= –5.55, –5.44, –4.72, –4.64 (4ϫq, 4ϫCH3-Si), 18.01, 18.25 (2ϫs,
2 ϫ (CH3)3C-Si), 25.79, 25.85 (2 ϫ q, 2 ϫ (CH3)3C-Si), 47.39 (t,
C(3)), 63.08 (t, CH2-OTBDMS), 71.12 (d, C-4), 83.47 (d, C-5),
165.95 (d, C-2) ppm. MS (FAB+): m/z (%) = 344 (100, [M + H]+),
154 (10), 115 (17).
trans-N-[(9-Fluorenylmethoxy)carbonyl]-3-hydroxy-L-proline (6): A
solution of 9-fluorenylmethyl chloroformate (11.0 g, 42.5 mmol) in
dioxane (150 mL) was added at 0 °C to a suspension of trans-3-
hydroxy--proline (5, 5.5 g, 41.9 mmol) in dioxane (300 mL) and
NaHCO3 (5%, 150 mL). After 2 h, the cooling bath was removed
and stirring was continued for 7 h. Most of the dioxane was evapo-
rated, and the aq. layer was washed twice with Et2O, acidified to
pH 3.0 with aq. HCl (2 ) and extracted with AcOEt (3ϫ). The
combined org. layers were dried (MgSO4) and concentrated to give
14.7 g (98%) of 6 as a white powder. 1H NMR (300 MHz, [D6]-
DMSO, 25 °C): δ = 1.80–1.98 (m, 2 H, 4-H), 3.48–3.54 (m, 2 H, 5-
H), 4.05–4.37 (m, 5 H, 2-H, 3-H, Fmoc-H, Fmoc-CH2), 5.52, 5.59
(2ϫbrs, 1 H, OH), 7.29–7.44 (m, 4 H, Fmoc-H), 7.62–7.69 (m, 2
H, Fmoc-H), 7.88–7.91 (m, 2 H, Fmoc-H), 12.91 (brs, 1 H,
COOH) ppm. 13C NMR (75 MHz, [D6]DMSO, 25 °C): δ = 34.38,
35.37 (2ϫt, C-4), 47.33, 47.97 (2ϫt, C-5), 49.61 (d, Fmoc-C),
69.70, 69.98 (2ϫt, Fmoc-C), 70.91, 71.18 (2ϫd, C-3), 75.81, 76.91
(2ϫd, C-2), 123.15 (d, Fmoc-C), 128.14, 128.24 (2ϫd, Fmoc-C),
130.16 (d, Fmoc-C), 130.71 (d, Fmoc-C), 143.66, 143.75 (2ϫs,
Fmoc-C), 146.69, 146.82 (2ϫs, Fmoc-C), 157.03, 157.21 (2ϫs, -
NCO2-), 174.79, 175.09 (2ϫs, COOH) ppm. MS (FAB+): m/z (%)
= 354 (26) [M + H]+, 338 (24), 309 (100). HR-MS (FAB+, [M +
H]+): 354.13574 (C20H20NO5 requires: 354.13415).
(2R,3S)-N-[(9-Fluorenylmethoxy)carbonyl]-3-hydroxy-2-(hydroxy-
methyl)pyrrolidine (7): BH3·(CH3)2S (2  in THF, 96.2 mL,
192.4 mmol) was added to a solution of 6 (13.0 g, 36.8 mmol) in
dry THF (130 mL). The solution was heated at reflux for 2 h and
cooled to room temp., and the reaction was quenched by addition
of MeOH. The solution was concentrated, the residue was dis-
solved in AcOEt, washed with H2O and dried (MgSO4), and the
solvents were evaporated. FC (toluene/THF 1:1 to 5:6) gave 12.1 g
2-Benzyloxy-5-bromopyridine (10): BnBr (4.1 mL, 34.5 mmol) and
Ag2CO3 (4.75 g, 17.2 mmol) were added to a solution of 9 (5.00 g,
28.7 mmol) in dry THF (250 mL). The mixture was heated at reflux
in the dark for 5 h and was then filtered through Celite and washed
with diethyl ether, and the solvents were evaporated. The residue
was purified by FC (AcOEt 2–5% in hexane) to yield 7.04 g (93%)
1
(97%) of 7 as a white solid. TLC (toluene/THF 1:1): Rf 0.24. H
NMR (300 MHz, [D6]DMSO, 25 °C): δ = 1.66–1.75 (m, 1 H, 4-H),
1.94–2.07 (m, 1 H, 4-H), 3.08–3.20 (m, 1 H, CH2-OH), 3.28–3.57 of 10 as a white solid. TLC (AcOEt/hexane 2:98): Rf 0.26. 1H NMR
(m, 4 H, 2-H, 5-H, CH2-OH), 4.17–4.21 (m, 1 H, 3-H), 4.27 (s, 3
(300 MHz, CDCl3, 25 °C): δ = 5.35 (s, 2 H, CH2-Bn), 6.73 (d, JH,H
H, Fmoc-H, Fmoc-H), 4.76–4.82 (m, 1 H, HO-CH2), 4.92–4.94 (m, = 8.7 Hz, 1 H, 3-H), 7.29–7.46 (m, 5 H, Bn-H), 7.66 (dd, JH,H
=
Eur. J. Org. Chem. 2007, 4038–4049
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4045

A. Mayer, C. J. Leumann
FULL PAPER
2.5, 8.7 Hz, 1 H, 4-H), 8.21 (d, JH,H = 2.5 Hz, 1 H, 6-H) ppm. 13C
4.42 (5.3%, 2-H), 5.36 (1.0%, CH2-Bn). MS (ESI+): m/z (%) = 1058
NMR (75 MHz, CDCl3, 25 °C): δ = 67.98 (Bn-CH2), 111.89 (C5), (34), 530 (100, [M + H]+), 344 (8). HR-MS (ESI+, [M + H]+):
112.93 (C3), 127.97, 128.00, 128.49, 136.89 (Bn-C), 141.18 (C4),
147.44 (C6), 162.40 (C2) ppm. MS (EI+): m/z (%) = 263/265 (56/
56, [M]+), 184 (8), 157/159 (15/15), 91 (100), 65 (55). HR-MS (EI+,
[M]+): 262.9941 (C12H10BrNO requires: 262.9946).
529.3275 (C29H49N2O3Si2 requires: 529.3281).
4-(tert-Butyldimethylsilanyloxy)-5-(tert-butyldimethylsilanyloxy-
methyl)-2-(2-oxopyridin-5-yl)pyrrolidine (14): Pd/C (10 % w/w,
81 mg) was added under argon to a solution of 13 (α/β mixture,
809 mg, 1.5 mmol) in MeOH (50 mL) and the black suspension
was stirred under hydrogen. The reaction was monitored by TLC in
AcOEt/hexane 1:4, stained with ninhydrin. As soon as no starting
material was detectable any more, typically after 30–45 min, the
palladium catalyst was filtered off over Celite and washed with
MeOH, and the filtrate was concentrated and coevaporated with
AcOEt to yield a greenish foam. FC (AcOEt/THF 1:0 to 4:1) af-
forded 352 mg of 14 (35% over two steps) as a yellowish foam.
TLC (CH2Cl2/MeOH 9:1): Rf 0.43. 1H NMR (300 MHz, CDCl3,
25 °C): δ = 0.05, 0.06, 0.07, 0.07 (4ϫs, 12 H, 4ϫCH3-Si), 0.89
(2ϫs, 18 H, 2ϫ(CH3)3C), 1.62–1.69 (m, 1 H, 3-Hβ), 1.93 (ddd,
JH,H = 1.7, 4.7, 9.6 Hz, 1 H, 3-Hα), 3.13–3.16 (m, 1 H, H5), 3.47
(dd, JH,H = 4.9, 7.6 Hz, 1 H, CH2O), 3.59 (dd, JH,H = 3.9, 7.6 Hz,
1 H, CH2O), 4.18–4.21 (m, 1 H, 4-H), 4.26 (dd, JH,H = 4.8, 7.4 Hz,
5-Bromo-2-(2,2,5,5-tetramethyl-1,2,5-azadisilolidin-1-yl)pyridine
(12): A solution of 2-amino-5-bromopyridine (11, 5.00 g,
28.9 mmol) in dry THF (80 mL) was treated at –78 °C with nBuLi
(1.6  in hexane, 18.1 mL, 28.9 mmol). After 1 h at –78 °C, a solu-
tion of 1,2-bis(chlorodimethylsilyl)ethane (6.22 g, 28.9 mmol) in
THF (15 mL) was added dropwise. After another 90 min at –78 °C,
nBuLi (1.6  in hexane, 18.1 mL, 28.9 mmol) was added, and the
mixture was allowed to reach room temp. and stirred for an ad-
ditional 2 h. Ice-cold brine (50 mL) was added and the mixture
was quickly extracted with Et2O (2ϫ200 mL). The combined org.
phases were dried (MgSO4) and concentrated. Kugelrohr distil-
lation (110–120 °C/0.1 mbar) gave 7.20 g (79%) of the stabase ad-
duct 12 as a white solid. TLC (AcOEt/hexane 1:99): Rf 0.8. 1H
NMR (300 MHz, CDCl3, 25 °C): δ = 0.30 (s, 12 H, 4ϫCH3-Si),
0.82 (s, 4 H, 2ϫCH2-Si), 6.46 (dd, JH,H = 0.8, 8.9 Hz, 1 H, 3-H),
7.46 (dd, JH,H = 2.6, 8.9 Hz, 1 H, 4-H), 8.14 (dd, JH,H = 0.8, 2.6 Hz,
1 H, 6-H) ppm. 13C NMR (75 MHz, CDCl3, 25 °C): δ = –0.58
(4ϫCH3-Si), 8.54 (2ϫCH2-Si), 108.52 (C-5), 113.26 (C-3), 139.29
(C-4), 148.67 (C-6), 159.43 (C-2) ppm.
1 H, 2-H), 6.55 (d, JH,H = 7.1 Hz, 1 H, 3Ј-H), 7.37 (d, JH,H
=
1.9 Hz, 1 H, 6Ј-H), 7.51 (dd, JH,H = 1.9, 7.1 Hz, 1 H, 4Ј-H) ppm.
13C NMR (75 MHz, CDCl3, 25 °C): δ = –5.39, –5.37, –4.69, –4.61
(4ϫCH3-Si),18.02,18.34(2ϫ(CH3)3C-Si),25.83,25.95(2ϫ(CH3)3-
C-Si), 42.88 (C-3), 56.91 (C-2), 65.38 (CH2O), 68.53 (C-5), 74.19
(C4), 120.10 (C-3Ј), 123.22 (C-5Ј), 131.49 (C-6Ј), 141.36 (C-4Ј),
165.05 (C-2Ј) ppm. Difference-NOE (400 MHz, CDCl3, 25 °C): δ
= 1.62–1.69 (3-Hβ) Ǟ 1.93 (7.3 %, 3-Hα), 3.47 (1.5 %, CH2O),
4.18–4.21 (5.6%; 4-H), 4.26 (1.8%, 2-H), 7.37 (3.0%, 6Ј-H), 7.51
(3.6%, 4Ј-H); 1.93 (3-Hα) Ǟ 1.62–1.69 (7.6%, 3-Hβ), 4.18–4.21
(1.7%, 4-H), 4.26 (3.8%, 2-H), 7.37 (2.9%, 6Ј-H), 7.51 (1.0%, 4Ј-
H); 3.13–3.16 (5-H) Ǟ 3.47 and 3.59 (5.7 %, CH2O), 4.18–4.21
(2.9 %, 4-H), 4.26 (2.5 %, 2-H); 3.47 (CH2O) Ǟ 3.59 (19.9 %,
CH2O), 4.18–4.21 (2.6%, 4-H); 3.59 (CH2O) Ǟ 3.13–3.16 (4.1%,
5-H), 3.47 (10.7%, CH2O), 4.18–4.21 (2.3%, 4-H); 4.18–4.21 (4-H)
Ǟ 1.62–1.69 (3.1%, 3-H3), 1.93 (1.6%, 3-Hα), 3.13–3.16 (2.3%, 5-
H), 3.47 and 3.59 (2.4%, CH2O); 4.26 (2-H) Ǟ 1.93 (2.8%, 3-Hα),
3.13–3.16 (2.0%, 5-H), 7.37 (3.1%, 6Ј-HЈ), 7.51 (2.8%, 4Ј-H); 6.55
(3Ј-H) Ǟ 7.51 (4.5%, 4Ј-H); 7.37 (6Ј-H) Ǟ 1.62–1.69 (1.2%, 3-
Hβ), 4.26 (3.3%, 2-H); 7.51 (4Ј-H) Ǟ 1.62–1.69 (2.5%, 3-Hβ), 4.26
(2.6%, 2-H), 6.55 (9.1%, 3Ј-H). MS (ESI+): m/z (%) = 878 (6),
439 (100, [M + H]+), 346 (5). HR-MS (ESI+, [M + H]+): 439.2819
(C22H43N2O3Si2 requires: 439.2812).
2-(2-Benzyloxypyridin-5-yl)-4-(tert-butyldimethylsilanyloxy)-5-(tert-
butyldimethylsilanyloxymethyl)pyrrolidine (13): nBuLi (1.6  in hex-
ane, 4.49 mL, 7.2 mmol) was added dropwise at –78 °C to a stirred
solution of 10 (1.90 g, 7.2 mmol) in dry THF (33 mL). After 1 h, a
solution of 4 (823 mg, 2.4 mmol) in dry THF (12 mL) was slowly
added and the mixture was stirred for 2 h. The reaction was
quenched at –78 °C with water (40 mL) and the layers were sepa-
rated. The aq. phase was extracted with ether (40 mL). The com-
bined org. phases were dried (MgSO4) and concentrated to yield a
yellow oil. FC (AcOEt/hexane 1:8 to 1:5) afforded 809 mg of an α/
β mixture of 13 as a yellow oil, which was used for the next step
without further purification. A pure fraction of the β anomer was
isolated for analytical characterisation. TLC (AcOEt/hexane 1:7):
1
Rf 0.27. H NMR (300 MHz, CDCl3, 25 °C): δ = 0.06, 0.07, 0.08
(3ϫs, 12 H, 4 ϫCH3-Si), 0.90, 0.91 (2 ϫs, 18 H, 2 ϫ(CH3)3C),
1.73–1.90 (m, 2 H, 3-Hβ, NH), 1.97–2.04 (m, 1 H, 3-Hα), 3.14–
3.18 (m, 1 H, 5-H), 3.55–3.69 (m, 2 H, CH2O), 4.23–4.27 (m, 1 H,
4-H), 4.42 (dd, JH,H = 6.4, 9.7 Hz, 1 H, 2-H), 5.36 (s, 2 H, CH2-
Bn), 6.77 (d, JH,H = 8.8 Hz, 1 H, 3Ј-H), 7.30–7.47 (m, 5 H, Bn-H),
7.62 (dd, JH,H = 2.6, 8.8 Hz, 1 H, 4Ј-H), 8.12 (d, JH,H = 2.6 Hz, 1
H, 6Ј-H) ppm. 13C NMR (75 MHz, CDCl3, 25 °C): δ = –5.39,
–4.69, –4.57 (4ϫ CH3-Si), 18.05, 18.33 [2ϫ (CH3)3C-Si], 25.86,
25.94 [2ϫ(CH3)3C-Si], 43.78 (C3), 57.64 (C2), 64.81 (CH2O), 67.57
(CH2-Bn), 68.86 (C-5), 74.26 (C4), 111.01 (C-3Ј), 127.74, 127.91,
128.42 (5 C-Bn), 132.66 (C-5Ј), 137.40 (C-4Ј), 137.45 (C-Bn),
144.97 (C-6Ј), 162.93 (C-2Ј) ppm. Difference-NOE (400 MHz,
CDCl3, 25 °C): δ = 1.73–1.90 (H3β) Ǟ 1.97–2.04 (6.5%, H3α),
3.55–3.69 (2.6%, CH2O), 4.23–4.27 (6.8%, H4), 7.62 (3.5%, 4Ј-H),
8.12 (2.6%, 6Ј-H); 1.97–2.04 (3-Hα) Ǟ 1.73–1.90 (19.1%, 3-Hβ),
3.55–3.69 (2.1%, CH2O), 4.23–4.27 (3.1%, 4-H), 4.42 (5.1%, 2-H),
7.62 (2.0%, 4Ј-H), 8.12 (2.1%, 6Ј-H); 3.14–3.18 (5-H) Ǟ 3.55–3.69
(5.1%, CH2O), 4.23–4.27 (2.8%, 4-H), 4.42 (3.4%, 2-H); 3.55–3.69
(CH2O) Ǟ 3.14–3.18 (5.1%, 5-H), 4.23–4.27 (5.7%, 4-H); 4.23–
4.27 (4-H) Ǟ 1.73–1.90 (6.0 %, 3-Hβ), 1.97–2.04 (1.4 %, 3-Hα),
3.14–3.18 (2.4 %, 5-H), 3.55–3.69 (2.8 %, CH2O); 4.42 (2-H) Ǟ
1.97–2.04 (5.7%, 3-Hα), 3.14–3.18 (3.1%, 5-H), 7.62 (2.2%, 4Ј-H),
8.12 (7.9 %, 6Ј-H); 6.77 (3Ј-H) Ǟ 5.36 (1.1 %, CH2-Bn), 7.62
Fluoren-9-ylmethyl 4-(tert-Butyldimethylsilanyloxy)-5-(tert-butyldi-
methylsilanyloxymethyl)-2-(2-oxopyridin-5-yl)pyrrolidine-1-carb-
oxylate (15): A solution of Fmoc-OSu (541 mg, 1.6 mmol) in THF
(8.5 mL) was added to a suspension of 14 (352 mg, 0.8 mmol) in
dioxane (8.5 mL) and aq. NaHCO3 (1 , 8.5 mL). When no start-
ing material was detected any more by TLC (1.5 to 2 h), 60 mL of
aq. sat. NaCl was added and the product was extracted with
3ϫ30 mL AcOEt. The org. layers were dried with MgSO4 and con-
centrated to yield a yellow foam. The product was used for the next
step without further purification. A pure sample of 15 was obtained
as white foam after FC (AcOEt) for characterisation. TLC (THF):
Rf 0.63. MS (ESI+): m/z (%) = 1322 (20), 661 (100, [M + H]+), 509
(4). HR-MS (ESI+, [M + H]+): 661.3488 (C37H53N2O5Si2 requires:
661.3493).
Fluoren-9-ylmethyl 4-(tert-Butyldimethylsilanyloxy)-5-(tert-butyldi-
methylsilanyloxymethyl)-2-{2-[2-(4-nitrophenyl)ethoxy]pyridin-5-
yl}pyrrolidine-1-carboxylate (16): Ph3P (408 mg, 1.6 mmol) and (p-
nitrophenyl)ethanol (260 mg, 1.6 mmol) were added under argon to
(10.1%, 4Ј-H); 7.62 (4Ј-H) Ǟ 6.77 (14.2%, 3Ј-H); 8.12 (6Ј-H) Ǟ a solution of 15 (278 mg, 0.4 mmol) in dry dioxane (4 mL). DIAD
4046
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 4038–4049

Pyrrolidino DNA with 2-Oxo Deletion Mutants of Thymine and Cytosine
FULL PAPER
(302 µL, 1.6 mmol) was then added dropwise to the stirred solu-
tion. After 8 h, the reaction mixture was concentrated and the yel-
low oily residue was purified by FC (AcOEt/hexane 2:3 to 1:1) to
yield 118 mg (35%) of 16 as a yellow foam. TLC (AcOEt): Rf 0.41.
3.15 (m, 1 H, 5-H), 3.57–3.69 (m, 2 H, CH2O), 4.21–4.25 (m, 1 H,
4-H), 4.30–4.37 (m, 3 H, 2-H, NH2), 6.47 (d, JH,H = 8.5 Hz, 1 H,
3Ј-H), 7.46 (dd, JH,H = 2.3, 8.5 Hz, 1 H, 4Ј-H), 8.02 (d, JH,H
=
2.3 Hz, 1 H, 6Ј-H) ppm. 13C NMR (75 MHz, CDCl3, 25 °C): δ =
MS (ESI+): m/z (%) = 1621 (18), 1089 (59), 810 (87, [M + H]+), –5.43, –4.73, –4.60 (4ϫCH3-Si), 18.01, 18.28 (2 ϫ (CH3)3C-Si),
658 (100), 279 (81). HR-MS (ESI+, [M + H]+): 810.3963
(C45H60N3O7Si2 requires: 810.3969).
25.82, 25.90 (2 ϫ (CH3)3C-Si), 43.68 (C-3), 57.82 (C-2), 64.47
(CH2O), 68.85 (C-5), 74.26 (C-4), 108.55 (C-3Ј), 129.39 (C-5Ј),
136.52 (C-4Ј), 146.26 (C-6Ј), 157.58 (C-2Ј) ppm. Difference-NOE
(400 MHz, CDCl3, 25 °C): δ = 3.10–3.15 (5-H) Ǟ 3.57–3.69 (4.8%,
CH2O), 4.21–4.25 (2.8%, 4-H), 4.30–4.37 (2.6%, 2-H); 3.57–3.69
(CH2O) Ǟ3.10–3.15 (4.9%, 5-H), 4.21–4.25 (3.4%, 4-H); 6.47 (3Ј-
H) Ǟ 7.46 (9.6%, 4Ј-H); 7.46 (4Ј-H) Ǟ 1.71–1.81 (1.6%, 3-Hβ),
4.30–4.37 (1.5%, 2-H), 6.47 (10.4%, 3Ј-H); 8.02 (6Ј-H) Ǟ 1.71–
1.81 (0.9%, 3-Hβ), 4.30–4.37 (4.3%, 2-H). MS (ESI+): m/z (%) =
438 (100) [M + H]+, 324 (26), 255 (13), 192 (20). HR-MS (ESI+,
[M + H]+): 438.2986 (C22H44N3O2Si2 requires: 438.2972).
Fluoren-9-ylmethyl 4-Hydroxy-5-(hydroxymethyl)-2-{2-[2-(4-nitro-
phenyl)ethoxy]pyridin-5-yl}pyrrolidine-1-carboxylate (17): A solu-
tion of aq. HCl (1 , 256 µL) was added to a solution of 16
(151 mg, 0.2 mmol) in MeOH (2.56 mL). The solution was stirred
at room temp. for 24–27 h until no starting material was detectable
by TLC and was then concentrated to give a violet solid. FC
(CH2Cl2/MeOH 19:1) afforded 73 mg (67%) of 17 as a slightly pink
solid. TLC (AcOEt): Rf 0.10. MS (ESI+): m/z (%) = 1163 (10), 582
(100, [M + H]+), 360 (12). HR-MS (ESI+, [M + H]+): 582.2223
(C33H32N3O7 requires: 582.2240).
Fluoren-9-ylmethyl 2-(2-Aminopyridin-5-yl)-4-(tert-butyldimethyl-
silanyloxy)-5-(tert-butyldimethylsilanyloxymethyl)pyrrolidine-1-car-
boxylate (21): A solution of Fmoc-OSu (583 mg, 1.7 mmol) in THF
(9 mL) was added to a suspension of 20 (378 mg, 0.9 mmol) in
dioxane (9 mL) and aq. NaHCO3 (1  , 9 mL). After 30 min, sat.
aq. NaCl (75 mL) was added and the product was extracted with
3ϫ40 mL AcOEt. The org. phase was dried (MgSO4) and the sol-
vents were evaporated. The obtained yellow foam was purified by
FC (AcOEt/hexane 6:4) to yield 425 mg (75%) of 21 as a white
foam. TLC (AcOEt): Rf 0.55. MS (ESI+): m/z (%) = 1321 (15), 660
(100) [M + H]+, 433 (30). HR-MS (ESI+, [M + H]+): 660.3662
(C37H54N3O4Si2 requires: 660.3652).
Fluoren-9-ylmethyl 5-[Bis(4-methoxyphenyl)(phenyl)methoxy-
methyl]-4-hydroxy-2-{2-[2-(4-nitrophenyl)ethoxy]pyridin-5-yl}pyrrol-
idine-1-carboxylate (18): DMT-Cl (35 mg, 0.1 mmol) was added in
three portions (16.3 mg) every 30 min to a solution of 17 (50 mg,
0.1 mmol) in dry pyridine (0.5 mL). Additional portions were
added every hour until no starting material was detectable by TLC.
The solution was diluted with AcOEt (5 mL) and washed with
water (10 mL), and the aqueous phase was extracted with AcOEt
(5 mL). The combined organic phases were dried with MgSO4 and
concentrated to give a yellow oil. The product was subjected to FC
(AcOEt/THF 1:0 to 4:1) to yield 56 mg (74%) of 18 as a white
foam. The FC column was conditioned with elution solvent + 1%
Et3N. TLC (AcOEt/THF 4:1): Rf 0.38. MS (ESI+): m/z (%) = 906
(40, [M + Na]+), 273 (78), 215 (100). HR-MS (ESI+, [M + Na]+):
906.3337 (C54H49N3O9Na requires: 906.3366).
Fluoren-9-ylmethyl 4-(tert-Butyldimethylsilanyloxy)-5-(tert-butyldi-
methylsilanyloxymethyl)-2-[2-(2-phenoxyacetylamino)pyridin-5-yl]-
pyrrolidine-1-carboxylate (22): Pac2O (664 mg, 2.3 mmol) was
added to a solution of 21 (425 mg, 0.6 mmol) in dry pyridine
(6.4 mL). After 2 h, the reaction was quenched with water (10 mL)
and the mixture was concentrated. The residue was dissolved in
AcOEt (120 mL), washed with aq. NaOH (0.1  , 2ϫ16 mL), dried
(MgSO4) and concentrated to yield a yellow oil. FC (AcOEt/hex-
ane 1:4) afforded 506 mg (99 %) of 22 as a white foam. TLC
(AcOEt/hexane 3:7): Rf 0.53. MS (ESI+): m/z (%) = 794 (55) [M +
H]+, 739 (91), 381 (100), 319 (16). HR-MS (ESI+, [M + H]+):
794.4027 (C45H60N3O6Si2 requires: 794.4020).
Fluoren-9-ylmethyl 5-[Bis(4-methoxyphenyl)(phenyl)methoxy-
methyl]-4-[(2-cyanoethoxy)(diisopropylamino)phosphanyloxy]-2-{2-
[2-(4-nitrophenyl)ethoxy]pyridin-5-yl}pyrrolidine-1-carboxylate (19):
iPr2NEt (55 µL, 0.3 mmol) was added to a solution of 18 (56 mg,
0.06 mmol) in dry THF (1.4 mL), followed by CEP-Cl (35 µL,
0.2 mmol). After 2 h, a solution of aq. NaHCO3 (1 , 5 mL) was
added and the product was extracted with AcOEt (2ϫ5 mL). The
org. phases were dried with MgSO4 and the solvents were evapo-
rated. The resulting yellow foam was purified by FC (AcOEt/hex-
ane 9:1) to afford 56 mg (82%) of 19 as a white foam. The FC
column was conditioned with elution solvent + 1% Et3N. TLC
(AcOEt/hexane 9:1): Rf 0.46, 0.52. 31P NMR (161.9 MHz, CDCl3):
δ = 147.84, 147.90, 148 ppm.
Fluoren-9-ylmethyl 4-Hydroxy-5-(hydroxymethyl)-2-[2-(2-phenoxy-
acetylamino)pyridin-5-yl]pyrrolidine-1-carboxylate (23): Aq. HCl
(1 , 220 µL) was added to a solution of 22 (126 mg, 0.2 mmol) in
MeOH (2.2 mL) and the solution was stirred at room temp. As
soon as all the starting material had reacted, according to TLC (20
to 24 h), the pink solution was concentrated and the resulting pink
foam was purified by FC (CH2Cl2/MeOH 19:1 to 9:1) to yield
89 mg (99%) of 23 as a slightly yellow foam. TLC (AcOEt/THF
9:1): Rf 0.28. MS (ESI+): m/z (%) = 566 (100) [M + H]+, 149 (34),
121 (34). HR-MS (ESI+, [M + H]+): 566.2294 (C33H32N3O6 re-
quires: 566.2291).
2-(2-Aminopyridin-5-yl)-4-(tert-butyldimethylsilanyloxy)-5-(tert-bu-
tyldimethylsilanyloxymethyl)pyrrolidine (20): nBuLi (1.6  in hex-
ane, 8.51 mL, 13.6 mmol) was added dropwise at –78 °C to a solu-
tion of 12 (4.30 g, 13.6 mmol) in dry THF (60 mL). After 1 h, a
solution of 4 (1.56 g, 4.5 mmol) in dry THF (20 mL) was added
slowly to the first solution and the mixture was stirred at –78 °C
for 2 h. The reaction was quenched with 80 mL of water, the tem-
perature was allowed to reach room temperature, and the mixture
was stirred for an additional hour. The layers were then separated
and the aqueous layer was extracted with ether (80 mL). The com-
bined org. phases were dried with MgSO4 and concentrated to give
a yellow oil. FC, first with AcOEt/hexane 1:4, allowing 1.19 g of
remaining 4 (76%) to be recovered, and then with AcOEt/THF 1:0
to 4:1, gave 378 mg (19%) of the desired product 20. TLC (AcOEt):
Fluoren-9-ylmethyl 5-[Bis(4-methoxyphenyl)(phenyl)methoxy-
methyl]-4-hydroxy-2-[2-(2-phenoxyacetylamino)pyridin-5-yl]pyrrol-
idine-1-carboxylate (24): DMT-Cl (199 mg, 0.6 mmol) was added
portionwise over 1 h to a solution of 23 (277 mg, 0.5 mmol) in dry
pyridine (2.7 mL). After an additional hour, the solution was di-
luted with AcOEt (20 mL) and washed with water (45 mL), and the
aqueous phase was extracted with AcOEt (20 mL). The combined
organic layers were treated with brine (30 mL) and dried with
MgSO4, and the solvents were evaporated. FC (AcOEt/hexane 7:3
to 8:2) of the yellow foam afforded 372 mg (88%) of 24 as a white
foam. The FC column was conditioned with elution solvent + 1%
1
Rf 0.19. H NMR (300 MHz, CDCl3, 25 °C): δ = 0.05, 0.06, 0.07,
0.07 (4ϫs, 12 H, 4ϫCH3-Si), 0.89, 0.90 (2ϫs, 18 H, 2ϫ(CH3)3-
C), 1.71–1.81 (m, 1 H, 3-Hβ), 1.93–2.01 (m, 2 H, 3-Hα, NH), 3.10–
Eur. J. Org. Chem. 2007, 4038–4049
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4047

A. Mayer, C. J. Leumann
FULL PAPER
Et3N. TLC (AcOEt): Rf 0.62. MS (ESI+): m/z (%) = 868 (7) [M + solid-phase phosphoramidite chemistry. Modifications to the syn-
H]+, 609 (100). HR-MS (ESI+, [M + H]+): 868.3557 (C54H50N3O8
requires: 868.3597).
thesis cycle were introduced for the incorporation of the nonnatural
building blocks as follows: (i) the coupling time was extended from
1.5 to 6 min, (ii) (S-benzylthio)-1H-tetrazole (0.25  in CH3CN)
was used as activator, and (iii) for the sequences containing dp2AP,
the classical acetyl capping step was replaced by a phenoxyacetyl
(Pac) capping step (Capping A solution: 6.15 g DMAP in 100 mL
CH3CN; Capping B solution: 7.5 g phenoxyacetic anhydride and
10 mL sym-collidine in 40 mL CH3CN). Standard ammonia depro-
tection (33% aq. ammonia at 55 °C for 16 h) led to cleavage of the
crude oligonucleotides from the solid support and removal of all
protecting groups (including Fmoc). For all sequences containing
NPE protective groups, a two-step deprotection procedure first in-
volving treatment with TDB (1 ) in acetonitrile at 60 °C for 8 h,
followed by ammonolysis, was applied. The crude oligonucleotides
were purified by ion exchange HPLC (DEAE-HPLC) (ET 125/4
NUCLEOGEN DEAE 60–7 column, 125 ϫ4.0 mm, Macherey–
Nagel) or FPLC (Mono Q HR 5/5 column, Pharmacia biotech)
and were then desalted over Sep-Pack C18 cartridges (Waters) ac-
cording to the manufacturer’s protocol. Purified oligonucleotides
were monitored for purity by reversed-phase HPLC (Aquapore RP-
300, 7 µm, 220ϫ4.6 mm, Brownlee or NUCLEOSIL 100–5, C18,
220ϫ5 mm, Macherey–Nagel) or FPLC (Pep RPC HR 5/5 col-
umn, Pharmacia biotech). The integrity of all oligonucleotides was
confirmed by ESI-MS.
Fluoren-9-ylmethyl 5-[Bis(4-methoxyphenyl)(phenyl)methoxy-
methyl]-4-[(2-cyanoethoxy)(diisopropylamino)phosphanyloxy]-2-[2-
(2-phenoxyacetylamino)pyridin-5-yl]pyrrolidine-1-carboxylate (25):
iPr2NEt (373 µL, 2.1 mmol) was added to a solution of 24 (372 mg,
0.4 mmol) in dry THF (9.5 mL), followed by CEP-Cl (239 µL,
1.1 mmol). After 2 h, a solution of aq. NaHCO3 (1 , 30 mL) was
added and the product was extracted with AcOEt (2ϫ30 mL). The
organic phases were dried with MgSO4 and concentrated to give a
colourless oil. FC (AcOEt/hexane 1:1) afforded 412 mg (90%) of
25 as a white foam. The FC column was conditioned with elution
solvent + 1% Et3N. TLC (AcOEt/hexane 1:1): Rf 0.33, 0.42. 31P
NMR (161.9 MHz, CDCl3): δ = 147.95, 148.08, 148.30 ppm. MS
(ESI+): m/z (%) = 1068 (100) [M + H]+, 359 (41), 341 (56), 313
(33), 253 (32), 200 (25). HR-MS (ESI+, [M + H]+): 1068.4706
(C63H67N5O9P requires: 1068.4676).
Fluoren-9-ylmethyl 2-(2-Aminopyridin-5-yl)-4-hydroxy-5-(hydroxy-
methyl)pyrrolidine-1-carboxylate (26): Aq. HCl (1 , 900 µL) was
added to a solution of 22 (511 mg, 0.6 mmol) in MeOH (9 mL) and
the solution was stirred at room temp. After 30 h, the pink solution
was concentrated and the resulting pink foam was purified by FC
(CH2Cl2/MeOH 19:1 to 9:1) to yield 277 mg (76%) of 23 (described
earlier), together with 64 mg (23%) of 26 as a white solid. TLC
(CH2Cl2/MeOH 19:1): Rf 0.04. MS (ESI+): m/z (%) = 432 (100) [M
+ H]+, 254 (21), 210 (10).
Thermal Denaturation Experiments: UV-melting experiments were
carried out with a Varian Cary 100 BIO UV/Vis spectrophotometer
equipped with a temperature controller. Data were collected with
the Cary WinUV thermal software. All measurements were con-
ducted in buffered solution (140 m KCl, 7 m NaH2PO4, 0.5 m
MgCl2) at oligonucleotide concentrations of 1.2 µ; the pH values
were measured directly in the sample. Melting curves were recorded
at 260 nm in a consecutive heating-cooling-heating cycle (0–90 °C
or 20–90 °C) with a temperature gradient of 0.5 °Cmin–1. For tem-
peratures Ͻ20 °C, nitrogen was passed through the spectrophotom-
eter to avoid condensation. To avoid evaporation of the solutions,
6–8 drops of dimethylpolysiloxane were added at the top of the
samples in the cell. The absorbance melting curves were smoothed
and the first derivative curves were obtained by use of Varian
WinUV or Microcal Origin software.
2-(2-Aminopyridin-5-yl)-4-hydroxy-5-(hydroxymethyl)pyrrolidine
(27): Piperidine (94 µL, 1.0 mmol) was added to a suspension of 26
(61 mg, 0.1 mmol) in DMF (650 µL) and the mixture was stirred
at room temp. overnight. The solution was then concentrated to
yield a brown solid. FC (AcOEt/MeOH 1:0 to 1:1) afforded 21 mg
(71%) of 27 as a slightly yellow solid. TLC (CH2Cl2/MeOH 9:1):
1
Rf 0.04. H NMR (300 MHz, MeOD, 25 °C): δ = 2.17–2.25 (m, 1
H, 3-Hα), 2.39–2.50 (m, 1 H, 3-Hβ), 3.55–3.60 (m, 1 H, 5-H), 3.85
(d, JH,H = 4.9 Hz, 2 H, CH2O), 4.42–4.45 (m, 1 H, 4-H), 4.76 (dd,
JH,H = 6.0, 12.1 Hz, 1 H, 2-H), 6.69 (d, JH,H = 8.8 Hz, 1 H, 3Ј-H),
7.68 (dd, JH,H = 2.4, 8.8 Hz, 1 H, 4Ј-H), 8.03 (d, JH,H = 2.3 Hz, 1
H, 6Ј-H) ppm. 13C NMR (75 MHz, MeOD, 25 °C): δ = 41.24 (C-
3), 61.01 (C-2), 61.59 (CH2O), 70.82 (C-5), 73.80 (C-4), 110.71 (C-
3Ј), 121.23 (C-5Ј), 139.09 (C-4Ј), 148.15 (C-6Ј), 161.66 (C-2Ј) ppm.
Difference-NOE (400 MHz, MeOD, 25 °C): δ = 2.17–2.25 (3-Hα)
Ǟ 2.39–2.50 (15.6%, 3-Hβ), 4.42–4.45 (2.2%, 4-H), 4.76 (6.5%, 2-
H); 2.39–2.50 (3-Hβ) Ǟ 2.17–2.25 (19.5%, 3-Hα), 4.42–4.45 (9.3%,
4-H), 4.76 (2.1%, 2-H), 7.68 (6.0%, 4Ј-H), 8.03 (3.2%, 6Ј-H); 3.55–
3.60 (5-H) Ǟ 3.85 (4.0 %, CH2O), 4.42–4.45 (2.6%, 4-H), 4.76
(4.0%, 2-H); 3.85 (CH2O) Ǟ 3.55–3.60 (4.8 %, 5-H), 4.42–4.45
(3.0%, 4-H); 4.42–4.45 (4-H) Ǟ 2.17–2.25 (1.4%, 3-Hα), 2.39–2.50
(4.8%, 3-Hβ), 3.55–3.60 (2.5%, 5-H), 3.85 (2.8%, CH2O); 4.76 (2-
H) Ǟ 2.17–2.25 (4.7%, 3-Hα), 3.55–3.60 (3.5%, 5-H), 7.68 (2.6%,
4Ј-H), 8.03 (5.8%, 6Ј-H); 6.69 (3Ј-H) Ǟ 7.68 (7.6%, 4Ј-H); 7.68
(4Ј-H) Ǟ 2.39–2.50 (2.9%, 3-Hβ), 4.76 (3.4%, 2-H), 6.69 (9.8%,
3Ј-H); 8.03 (6Ј-H) Ǟ 2.39–2.50 (2.1%, 3-Hβ), 4.76 (7.0%, 2-H).
MS (ESI+): m/z (%) = 210 (100) [[M + H]+], 192 (88), 122 (21). HR-
MS (ESI+, [M + H]+): 210.1246 (C10H16N3O2 requires: 210.1242).
Circular Dichroism Spectroscopy: CD spectra were recorded with a
Jasco J-715 spectropolarimeter equipped with a Jasco PFO-350S
temperature controller. The temperature was measured directly in
the sample. The graphs were subsequently smoothed with a noise
filter. The buffer was used as blank.
Acknowledgments
We thank the Swiss National Science Foundation (grant no.
200020–107692) for financial support of this project.
[1] D. Praseuth, A. L. Guieysse, C. Helene, Biochim. Biophys. Acta
1999, 1489, 181–206.
[2] C. Helene, N. T. Thuong, A. Harel-Bellan, Ann. N. Y. Acad.
Sci. 1992, 660, 27–36.
[3] J. M. Kalish, M. M. Seidman, D. L. Weeks, P. M. Glazer, Nu-
cleic Acid Res. 2005, 33, 3492–3502.
[4] M. M. Seidman, N. Puri, A. Majumdar, B. Cuenoud, P. S.
Miller, R. Alam, Ann. N. Y. Acad. Sci. 2005, 1058, 119–127.
[5] F. A. Rogers, J. A. Lloyd, P. M. Glazer, Curr. Med. Chem. Anti-
Cancer Agents 2005, 5, 319–326.
Oligonucleotides: All oligodeoxyribonucleotides were prepared
from the modified phosphoramidites 19 or 25 and commercially
available natural building blocks on (deoxy)nucleoside-CPG solid
supports (Glen Research). Synthesis was performed in the trityl-off
mode on the 1.3 µmol scale of a Pharmacia LKB Gene Assembler
Special DNA-synthesiser or on the 1.0 µmol scale of a PerSeptive
Biosystems Expedite Nucleic Acid Synthesis System by standard
[6] J. R. Goñi, X. de la Cruz, M. Orozco, Nucleic Acid Res. 2004,
32, 354–360.
4048
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 4038–4049

Pyrrolidino DNA with 2-Oxo Deletion Mutants of Thymine and Cytosine
FULL PAPER
[7] S. Buchini, C. J. Leumann, Curr. Opin. Chem. Biol. 2003, 7,
717–726.
[8] K. R. Fox, Curr. Med. Chem. 2000, 7, 17–37.
[9] J. Robles, A. Grandas, E. Pedroso, F. J. Luque, R. Eritja, M.
Orozco, Curr. Org. Chem. 2002, 6, 1333–1368.
[10] S. Hildbrand, C. J. Leumann, Angew. Chem. Int. Ed. Engl.
1996, 35, 1968–1970.
[11] S. Hildbrand, A. Blaser, S. P. Parel, C. J. Leumann, J. Am.
Chem. Soc. 1997, 119, 5499–5511.
[12] P. J. Bates, C. A. Laughton, T. C. Jenkins, D. C. Capaldi, P. D.
Roselt, C. B. Reese, S. Neidle, Nucleic Acids Res. 1996, 24,
4176–4184.
[26]
[27]
M. J. J. Blommers, F. Natt, W. Jahnke, B. Cuenoud, Biochemis-
try 1998, 37, 17714–17725.
A. Mayer, A. Häberli, C. J. Leumann, Org. Biomol. Chem.
2005, 3, 1653–1658.
A. Häberli, C. J. Leumann, Org. Lett. 2002, 4, 3275–3278.
Q. Wu, C. Simons, Synthesis 2004, 10, 1533–1553.
V. L. Schramm, P. C. Tyler, Curr. Top. Med. Chem. 2003, 3,
525–540.
A. Häberli, C. J. Leumann, Org. Lett. 2001, 3, 489–492.
A. Momotake, H. Togo, M. Yokoyama, J. Chem. Soc. Perkin
Trans. 1 1999, 1193–1200.
M. Yokoyama, T. Ikenogami, H. Togo, J. Chem. Soc. Perkin
Trans. 1 2000, 2067–2071.
S. Hanessian, R. Margarita, A. Hall, S. Johnstone, M. Trem-
blay, L. Parlanti, J. Am. Chem. Soc. 2002, 124, 13342–13343.
S. Hanessian, M. Tremblay, J. F. W. Petersen, J. Am. Chem.
Soc. 2004, 126, 6064–6071.
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[13] S. A. Cassidy, P. Slickers, J. O. Trent, D. C. Capaldi, P. D. Ro-
selt, C. B. Reese, S. Neidle, Nucleic Acids Res. 1997, 25, 4891–
4898.
[14] U. V. Krosigk, S. A. Benner, J. Am. Chem. Soc. 1995, 117,
5361–5362.
[15] A. Ono, P. O. P. Ts’O, L.-S. Kan, J. Am. Chem. Soc. 1991, 113,
4032–4033.
[36] S. Hanessian, S. Claridge, S. Johnstone, J. Org. Chem. 2002,
67, 4261–4274.
[37] I. Collado, J. Ezquerra, C. Pedegral, J. Org. Chem. 1995, 60,
[16] A. Ono, P. O. P. Ts’O, L.-S. Kan, J. Org. Chem. 1992, 57, 3225–
3230.
[17] G. Xiang, W. Soussou, L. W. McLaughlin, J. Am. Chem. Soc.
1994, 116, 11155–11156.
[18] G. Xiang, L. W. McLaughlin, Tetrahedron 1998, 54, 375–392.
[19] S. Buchini, C. J. Leumann, Angew. Chem. Int. Ed. 2004, 43,
3925–3928.
5011–5015.
[38] W. N. Speckamp, M. J. Moolenaar, Tetrahedron 2000, 56,
3817–3856.
[39] B. A. Horenstein, R. F. Zabinski, V. L. Schramm, Tetrahedron
Lett. 1993, 34, 7213–7216.
[20] R. T. Ranasinghe, D. A. Rusling, V. E. C. Powers, K. R. Fox,
T. Brown, Chem. Commun. 2005, 2555–2557.
[21] D. GuianvarcЈh, J.-L. Fourrey, R. Maurisse, J.-S. Sun, R.
Benhida, Bioorg. Med. Chem. 2003, 11, 2751–2759.
[22] Y. Wang, D. A. Rusling, V. E. C. Powers, O. Lack, S. D. Os-
borne, K. R. Fox, T. Brown, Biochemistry 2005, 44, 5884–5892.
[23] A. B. Eldrup, O. Dahl, P. E. Nielsen, J. Am. Chem. Soc. 1997,
119, 11116–11117.
[24] A. G. Olsen, O. Dahl, P. E. Nielsen, Bioorg. Med. Chem. Lett.
2004, 14, 1551–1554.
[25] B. Cuenoud, F. Casset, D. Hüsken, F. Natt, R. M. Wolf, K.-H.
Altmann, P. Martin, H. E. Moser, Angew. Chem. Int. Ed. Engl.
1998, 37, 1288–1291.
[40] R. H. Furneaux, G. Limberg, P. C. Tyler, V. L. Schramm, Tet-
rahedron 1997, 53, 2915–2930.
[41] G. B. Evans, R. H. Furneaux, G. J. Gainsford, V. L. Schramm,
P. C. Tyler, Tetrahedron 2000, 56, 3053–3062.
[42] G. B. Evans, R. H. Furneaux, T. L. Hutchison, H. S. Kezar,
P. E. J. Morris, V. L. Schramm, P. C. Tyler, J. Org. Chem. 2001,
66, 5723–5730.
[43] S. Bevers, S. B. Ha, L. W. McLaughlin, Biochemistry 1999, 38,
7710–7718.
[44] L. L. Kiessling, L. C. Griffin, P. B. Dervan, Biochemistry 1992,
31, 2829–2834.
Received: March 12, 2007
Published Online: July 3, 2007
Eur. J. Org. Chem. 2007, 4038–4049
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4049