Liquid-phase synthesis of a cyclic hexameric peptide nucleic acid

Article abstract of DOI:10.1016/S0040-4039(01)01776-2

A cyclic fully N-protected hexameric (aminoethylglycinamide) can be readily obtained by using a divergent approach in liquid phase and consists of coupling orthogonal fragments of suitable oligomers. This cyclic N-protected backbone is then converted into a peptide nucleic acid by a series of selective deprotection-coupling steps affording the desired structure in good overall yields. Such a procedure could provide a general approach for targeting any short cyclic peptide nucleic acids (containing every kind of nucleobase).

Full text of DOI:10.1016/S0040-4039(01)01776-2

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 8303–8306
Liquid-phase synthesis of a cyclic hexameric peptide nucleic acid
Geoffrey Depecker, Caroline Schwergold, Christophe Di Giorgio, Nadia Patino and
Roger Condom*
Laboratoire de Chimie Bio-Organique, UMR 6001 CNRS, Universit e´ de Nice Sophia-Antipolis, F-06108 Nice Cedex 2, France
Received 23 July 2001; accepted 19 September 2001
Abstract—A cyclic fully N-protected hexameric (aminoethylglycinamide) can be readily obtained by using a divergent approach
in liquid phase and consists of coupling orthogonal fragments of suitable oligomers. This cyclic N-protected backbone is then
converted into a peptide nucleic acid by a series of selective deprotection–coupling steps affording the desired structure in good
overall yields. Such a procedure could provide a general approach for targeting any short cyclic peptide nucleic acids (containing
every kind of nucleobase). © 2001 Elsevier Science Ltd. All rights reserved.
1
. Introduction
required length of the antisense (15–20 mer) to ensure
selectivity, their high cost limits considerably their ther-
apeutic potential.
Some functional RNAs biologically act as folding stem-
loop (or hairpin) structures which can be highly con-
served and play an active role by interacting with
proteins and/or nucleotidic sequences. For example,
HIV trans-activation is dependent, among others, on
the formation of a complex between TAR loop RNA
With the view of addressing these drawbacks, we are
exploring the feasibility of targeting loops that are
involved in crucial steps of viral expression by means of
short and cyclic complementary oligonucleotide ana-
logues. Specificity and selectivity could be brought by
their cyclic nature, thus mimicking kissing-loop interac-
tions. Furthermore, the use of RNA data bank 3D
structures and molecular design could allow a perfect
fitting between both the RNA target and the synthetic
loop.
1
and cellular factors. Two complementary stem-loops
may form ‘kissing-loop’ complexes which were found to
regulate the replication of ColE1 Escherichia coli plas-
mid and to promote dimerization of viral RNAs (SL1
loop of HIV-1). Several studies have further shown
that kissing-loop interactions involving only a few
nucleotide residues in each loop can be specific and
selective. Recently, formation of such complexes was
shown to occur between the natural TAR loop of HIV
RNA (which only contains six bases) and its comple-
2
3
4
In this paper, we describe the synthesis of a cyclic
hexameric PNA (peptide or polyamide nucleic acids)
derivative 1 bearing two different pyrimidine nucle-
obases (see Fig. 1). This hexameric PNA derivative 1
represents the first generation of this type of synthetic
loop. It can serve as a model to validate a more general
5
mentary synthetic one. Moreover, kissing-loop com-
plexes exhibit a particularly enhanced stability as
compared to complexes resulting from the interaction
between a loop and its linear complementary sequence.
RNA loops which are involved in pathogenic cellular,
viral or bacterial expression processes therefore repre-
sent potential targets in biomedicinal chemistry. A large
number of linear RNA inhibitors have been synthesized
in the context of antisense strategies. However, their
effectiveness is hampered by their poor cellular uptake
6
and low in vivo stability. Furthermore, owing to the
Keywords: kissing-loop complex; peptide nucleic acids; cyclization;
orthogonally synthons; coupling steps.
*
Corresponding author. Tel.: +3-349-207-6152; fax: +3-349-207-6151;
e-mail: condom@unice.fr
Figure 1. Cyclic hexameric PNA structure.
0
040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01776-2

8
304
G. Depecker et al. / Tetrahedron Letters 42 (2001) 8303–8306
approach including natural and synthetic nucleobases to
target natural biologically functional loops. Compound
Syntheses of linear PNA oligomers using solid-phase
chemistry have been largely described and consist of an
iterative process (coupling monomers or submonomers)
1
is constituted by the U-C-U-C-C-U PNA sequence
10,11
containing a linker tethered to N- and C-extremities of
the PNA backbone through amide bonds. This hexam-
eric PNA is complementary of a HIV RNA loop which
to achieve elongation.
To synthesize the cyclic hex-
americ PNA derivative 1, we used the fully protected
backbone (FPB) approach that was been recently devel-
7
12
is involved in the packaging of the HIV virus. PNAs,
oped in our laboratory. This approach consists of
which consist of N-(2-aminoethyl)glycine repeating units
attached to a nucleic base through a methylenecarbonyl
linker, have been chosen for their remarkable ability to
recognize their complementary oligonucleotides (espe-
elaborating, using liquid-phase synthetic procedures, a
fully protected linear poly(aminoethylglycinamide) con-
taining as many different and orthogonal protecting
groups as they are different potential elongation sites (P1,
P2) and different base units to introduce (PA–D) (Fig. 2).
The hexameric backbone is then connected, at one or at
both of its N- and/or C-extremities, to a linker. The
following head-to-tail cyclization affords the cyclic N-
protected hexameric backbone. Selective and sequential
deprotection then allows the simultaneous condensation
of the required number of identical nucleobase acetic acid
units on to the backbone (Fig. 2).
8
cially RNAs) with high specificity and affinity. More-
over, their remarkable resistance to nucleases, their
achiral, uncharged pseudopeptidic backbone and their
increased lipophilicity make them more attractive than
conventional charged antisense inhibitors for therapeutic
9
use. The length of the linker (17 atoms) was optimized
by molecular modelization with the aim of allowing the
head-to-tail cyclization to occur favourably.
Figure 2. Synthetic strategy to cyclic PNA analogues.
Scheme 1. Reagents and conditions: (a) Troc-Cl or Alloc-Cl, TEA, CH Cl , O°C, 10 min (90%); (b) LiOH/CaCl , iPrOH/H O,
2
2
2
2
2
h (90%); (c) TFA, CH Cl , 1 h (95%); (d) Cl-CO iBu, NMM, CH Cl , −15°C, 15 min (80%).
2 2 2 2 2

G. Depecker et al. / Tetrahedron Letters 42 (2001) 8303–8306
8305
The synthesis of the cyclic N-protected hexameric back-
bone, shown in Scheme 1, was performed via a divergent
approach using classical deprotection–activation cou-
pling reactions in peptide chemistry. The orthogonally
N-protected aminoethylglycine 5, 6, and 7, 8 monomer
synthons were first prepared and condensed two by two
to afford the fully protected 9 and 10 dimers, respectively,
and, after suitable deprotections, their corresponding
dimers 11, 12 and 13. The two dimer units 11 and 12 that
were obtained from 9 were condensed together to produce
tetramer 14 which, after C-terminal deprotection, was
conjugated to dimer 13 to yield linear hexamer 16. This
hexamer, as its acid 17, was then grafted onto the
a-amino-v-ester linker 22 giving derivative 23. All Boc
deprotections were performed with TFA/CH Cl . All
cytosine and uracyl acetic acid base units 28 and 29,
respectively, followed by cytosine deprotection. First, the
three Alloc protecting groups were cleanly removed from
25 with Pd[PPh ] as allyl acceptor and DEA, and three
N-Z-cytosine acetic acid units 28 were then condensed
onto the triamine thus produced by means of HATU/
HOAt activation giving the mixed polyamide compound
26. Next, the selective cleavage of the three Troc groups
by means of cadmium in acetic acid followed by the
HATU/HOAt-mediated condensation of three uracyl
acetic acid units 29 on to the triamine thus generated,
afforded the cyclic Z-cytosine and uracyl hexameric PNA
27. Benzyloxycarbonyl (Z) removal was performed by
treatment with HBr in glacial acetic acid.
3
4
2
2
2
+
saponifications of methyl esters were Ca catalyzed using
LiOH (1N), CaCl , iPrOH/H O medium. The coupling
In conclusion, a cyclic protected hexameric amino-
ethylglycinamide framework carrying as many different
and orthogonal protecting groups as there are different
types of nucleobases in the target PNA sequence has been
prepared. This backbone allowed a cyclic hexameric
PNA complementary to the PACK HIV RNA loop
to be successfully synthesized following a liquid-
phase procedure (17 steps starting from monomers, acetic
2
2
reactions were achieved via isobutylchloroformate activa-
tion. Linker 22 was synthesized, as shown in Scheme 2,
from v-aminooctanoic and v-aminoheptanoic acids
using routine deprotection–activation processes (six-step
procedure from v-aminooctanoic acid).
†
The one-step and simultaneous N- and C-deprotection
of 23 with TFA/CH Cl afforded the bifunctional TFA
acid unit bases and linker afforded 5 mg of pure 1 in
6% overall yield). Although this particular hexameric
PNA loop 1 contains only two different nucleobases, our
FPB approach could be applied to any short cyclic
PNA (less than decamer) containing the four natural
or synthetic bases as a large palette of orthogonal
protecting groups is available. This liquid-phase FPB
approach allows selective, sequential and simultaneous
attachment of several nucleobase units on the protected
hexameric (aminoethylglycinamide) backbone. It also
offers several advantages such as a greater choice of
solvents and reagents, high solubility of the different
protected backbones, high coupling efficiencies, intro-
2
2
v-amino-a-acid 24 salt in quasi quantitative yield. The
key cyclization step of 24 leading to the cyclic fully
protected N-poly(aminoethylglycinamide) backbone 25
was achieved using HOAt/HATU activation. The head-
to-tail cyclization of 24 was performed under semi-high
dilution conditions (10 mM) in order to minimize inter-
molecular condensation and afford 25. Its structure and
purity were clearly attested by NMR, mass spectrometry
and HPLC analyses. Conversion of the orthogonally
hexa-protected cyclic precursor 25 into the target PNA
loop 1 required the introduction of the protected Z-
Scheme 2. Reagents and conditions: (a) Alloc-Cl, 1N NaOH, dioxane, 1 h (90%); (b) N,N-dimethylformamide di-tert-butyl acetal,
toluene, 80°C, 2 h (90%); (c) Pd[P(Ph) ] , DEA, CH Cl , 10 min (90%); (d) Bop, NMM, CH Cl , 15 min (80%); (e) TFA, CH Cl ,
3
4
2
2
2
2
2
2
(
Z)
2
h (95%); (f) HOAt, HATU, DIPEA, DMF, 12 h (90%) and 28 (C -CH CO H) for compound 26, 15 min (84%) or 29
2 2
(
U-CH CO H) for compound 27, 15 min (70%); (g) Cd/AcOH, DMF, 5 h (80%); (h) HBr/AcOH, 2 days (70%).
2 2
†
HPLC (A/B 97:3 to 40:60 over 30 min): t =16.60 min, umax=201.5 and 270.9 nm. MS (ESI+) m/z 1778.7 (M+H)+.
R

8
306
G. Depecker et al. / Tetrahedron Letters 42 (2001) 8303–8306
duction of the same natural nucleobase units in one
step and at advanced stages of the synthesis.
References
1
2
3
. Gatignol, A. Gene Expr. 1996, 5, 217.
. Tomizawa, J. I. J. Mol. Biol. 1990, 212, 695.
. Mujeed, A.; Clever, J. L.; Billeci, T. M.; James, T. L.;
Parslow, T. G. Nat. Struct. Biol. 1998, 5, 432.
. Eisinger, J. Biochem. Biophys. Res. Commun. 1971, 43,
2. Experimental
4
5
Analytical HPLC chromatograms were obtained using
a WATERS 600 equipped with a 996 photodiode array
detector (PDA, UV detector from 195 to 290 nm) and
a column (250×4 mm) packed with Lichrospher 100-
RP-18 (5 mm). A gradient with water (0.1% TFA) as
solvent A and acetonitrile (0.1% TFA) as solvent B was
used with a 1 ml/min flow. Mass spectrometry analyses
were carried out on a TSQ 7000 Finnigan MAT (ESI)
performed by J. M. Guigonis of GUMPAC, Nice
8
54.
. Comolli, L. R.; Pelton, J. G.; Tinoco, I. Nucleic Acids
Res. 1998, 26, 4688.
. Crooke, S. Adv. Pharmacol. 1997, 40, 1.
. Hayashi, T.; Shioda, T.; Iwakura, Y.; Shibuta, H. Virol-
ogy 1992, 188, 590.
. Nielsen, P. E. Pure Appl. Chem. 1998, 70, 105.
. Uhlmann, E.; Peyman, A.; Breipohl, G.; Will, D. W.
Angew. Chem. Int. Ed. 1998, 37, 2796.
0. Christensen, L.; Fitzpatrick, R.; Gildea, B.; Petersen, K.
H.; Hansen, H. F.; Koch, T.; Egholm, M.; Buchardt, O.;
Nielsen, P. E.; Coull, J.; Berg, R. H. J. Pept. Sci. 1995, 3,
6
7
8
9
(
France).
1
Acknowledgements
1
75.
11. Richter, L. S.; Zuckermann, R. N. Bioorg. Med. Chem.
Lett. 1995, 5, 1159.
We thank the ‘Agence Nationale de Recherche sur le
SIDA’ (ANRS) and SIDACTION for their support
and the LARTIC (Professor Cabrol-Bass, D.) labora-
tory for molecular design.
12. Di Giorgio, C.; Pairot, S.; Schwergold, C.; Patino, N.;
Condom, R.; Farese-Di Giorgio, A.; Guedj, R. Tetra-
hedron 1999, 55, 1937.