Liquid-phase combinatorial synthesis of diPNA-arginine conjugates

Article abstract of DOI:10.1016/S0040-4039(02)00067-9

Considering the promising anti HIV activity displayed by some diPNA-arginine conjugates, a library has been generated to determine the target(s) and mode(s) of action of these presumed multi targets drugs and to optimize the antiviral properties of lead compounds. This library has been prepared using a combinatorial liquid-phase strategy, involving easily available N-protected PNA dimeric backbones as building blocks.

Full text of DOI:10.1016/S0040-4039(02)00067-9

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 1641–1644
Liquid-phase combinatorial synthesis of diPNA-arginine
conjugates
Mary-Lore`ne Leroux, Christophe Di Giorgio, Nadia Patino and Roger Condom*
Laboratoire de Chimie Bio-Organique, UMR 6001, Universite´ de Nice Sophia-Antipolis, F-06108 Nice Cedex 2, France
Received 4 December 2001; revised 7 January 2002; accepted 8 January 2002
Abstract—Considering the promising anti HIV activity displayed by some diPNA-arginine conjugates, a library has been
generated to determine the target(s) and mode(s) of action of these presumed multi targets drugs and to optimize the antiviral
properties of lead compounds. This library has been prepared using a combinatorial liquid-phase strategy, involving easily
available N-protected PNA dimeric backbones as building blocks. © 2002 Elsevier Science Ltd. All rights reserved.
The generation of combinatorial chemical libraries has
emerged as a highly efficient method for the identifica-
tion of novel chemical leads or for the optimization of
a promising lead compound.1
lead compounds, we planned to elaborate an extended
library of these compounds.
Combinatorial syntheses of PNA libraries have been
described using solid-phase procedures.4–7 However, in
our case, a liquid-phase combinatorial approach
seemed better adapted, considering the small size of the
conjugates and the large amounts (multi-milligrams)
required for extensive and broad screening purposes.
We report here the combinatorial synthesis of a diPNA-
arginine conjugates library, following an easy and rapid
liquid-phase FPB (Fully Protected Backbone) strategy.
In this context, we recently reported2 a new anti-HIV
family constituted by a Polyamide Nucleic Acid3 (PNA)
dimer moiety bearing the adenine–guanine or the gua-
nine–adenine base-pair, tethered to an arginine residue
through a linker of variable length (Fig. 1).
Several of these PNA-arginine conjugates were indeed
found to display a micromolar anti-HIV activity
(IC50=1 mM) without any cellular toxicity (CC50 >200
mM). However, these compounds seemed to act by at
least two different modes of action. To elucidate the
target(s) and mode(s) of action of these presumed
multi-targeted drugs, to establish structure–activity
relationships and to optimize the antiviral activity of
Molecular diversity for the family of the diPNA-
arginine conjugates shown in Fig. 1 can be generated by
varying their structural elements such as the nature of
the nucleobases, the length of the spacer linking the
arginine residue to the PNA dimer backbone, and the
nature of C-extremity of arginine (R). Therefore, we
envisaged to prepare such a library using (i) the five
natural nucleobases, adenine (A), cytosine (C), guanine
(G), thymine (T), uracile (U), and the universal base
analogue 5-nitroindole8 (In), (ii) a spacer that would be
composed of 2, 3, 4 and 5 methylenes, and (iii) the
arginine residue under its acid and methyl ester forms.
The complete size of this library will be of 288 compo-
nents (Fig. 2). Indeed, there are six2 nucleobase combi-
nations for PNA dimer conjugates, which further
display two different arginine C-extremities (R=H and
Me.). If we consider the four linkers, we get 288
structures.
Figure 1. diPNA-Arginine conjugate of HIV-1.
To anticipate deconvolution problems, the main pool is
partitioned into four sublibraries with respect to each
of the four-spacer lengths. Each of these sublibraries is
* Corresponding author. Tel.: +3-349-207-6152; fax: +3-349-207-6151;
e-mail: condom@unice.fr
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00067-9

1642
M.-L. Leroux et al. / Tetrahedron Letters 43 (2002) 1641–1644
split in two groups, in respect to the C-extremity of the
arginine residue (acid or ester). Each of them is consti-
tuted by six batches of six compounds. Moreover, each
batch contains a well-identified PNA at the second
position (Fig. 2).
To generate this library, we applied a strategy, named
Fully Protected Backbone (FPB), that was recently
developed for the synthesis of linear9 and cyclic10
polyPNA.
Starting from fully orthogonally protected derivatives
of type 1, this strategy consists of the selective deprotec-
tion of the P1-protected amino function, which is then
condensed stoichiometrically with an equimolar mix-
ture of the six B1–6 nucleobase acetic acid units. After
subsequent release of the second P2 amino protecting
group, the Bi unit chosen among the six B1–6 nucleobase
acetic acid units is then introduced. A first sublibrary of
diPNA-arginine conjugates is obtained after deprotec-
tion of the guanidinium group of arginine and of
nucleobases (ester groups) followed by a subsequent
hydrolysis (acid groups) (Fig. 3).
Figure 3. General strategy for the generation of diPNA-
arginine conjugates library. P1 and P2 are two orthogonal
protecting groups.
The synthesis of the N-protected dimer 2 was achieved
in eight steps in 37% overall yield (Scheme 2). Z- and
Mmt-protection of the primary amino function of
backbone 4 gave 5 and 6, respectively. Boc- and Alloc-
protection of the secondary amine of 5 and 6, respec-
tively, led to the corresponding fully protected
monomers 7 and 8. Saponification of 7 and detrityla-
tion of 8 afforded acid 9 and the TFA ammonium salt
As an illustration of this original method, we describe
herein the preparation of the sublibrary deriving from
the aminobutanoic acid (n=3) spacer. The key ortho-
gonally fully protected di(aminoethylglycinamide) back-
bone 1 (Scheme 1) was prepared by condensing the
dimeric moiety 2 and aminobutanoyl arginine methyl
ester 3 by means of DCC/HOSu, in 69% yield. The
arginine-based linker 3 was prepared from commercial
NG-(Pmc)-arginine and N-Z aminobutanoic acid, fol-
lowing a procedure previously described.2
Scheme 1. Synthesis of key compound 1. (a) DCC/HOSu,
DMF (69%).
Scheme 2. Preparation of Boc- and Alloc-protected backbone
2. (a) ZOSu, NMM, CH3CN (66%); (b) Mmt-Cl, TEA,
CH2Cl2 (90%); (c) (Boc)2O, TEA, CH2Cl2 (100%); (d) Alloc-
Cl, TEA, CH2Cl2 (88%); (e) 1 M LiOH, THF (82%); (f) TFA/
CH2Cl2 (92%); (g) ClCO2iBu, DIEA, CH2Cl2 (96%); (h) 1 M
LiOH, 0.8 M CaCl2, iPrOH/H2O 7/3 (85%).
Figure 2. Composition and subdivision of the libraries of
diPNA-arginine conjugates.

M.-L. Leroux et al. / Tetrahedron Letters 43 (2002) 1641–1644
1643
10, respectively. Subsequent condensation of 9 and 10
using isobutyl chloroformate led to the fully protected
dimer 11 which, after saponification, gave 2.
The last steps of the preparation of the targeted sub-
library are described in Scheme 3. Starting from 1, the
Boc group was selectively and smoothly cleaved using a
5.5 M methanolic solution of HCl, generated in situ by
addition of acetyl chloride to methanol.11 The resulting
hydrochloride salt 12 was then reacted stoichiometri-
cally with an equimolar mixture of the six base-acetic
acids [protected (G(OBn), A(diBoc), C(diBoc)) or free (T, U,
In)] by means of HATU in DMF. The depletion of 12
and base units B1–6 was monitored by HPLC.12 As
shown in Fig. 4B, HPLC analysis of the mixture iso-
lated after a rough purification (DMF removal and
triturating with water) clearly indicated the presence of
the six neo-formed 13 (B1–6-P2) conjugates.
The selective Alloc-deprotection of 13 (B1–6-P2) was
cleanly performed using Pd[PPh3]4 and DEA. The cor-
responding mixture of the six compounds 14 (B1–6-H)
(as attested by HPLC, data not shown) was purified by
triturating with diethylether. Then, condensation with
each of the six B1–6 base units onto 14 (B1–6-H), via a
HATU activation, gave six 15 (B1–6-Bi) batches which
were purified as described for mixture 13 (B1–6-P2).
Treatment of each 15 (B1–6-Bi) batch with 1/1 TFA/
CH2Cl2 induced the simultaneous deprotection of the
Boc and OBn-nucleobases (A(diBoc), G(OBn), C(diBoc)) and
of the NG-Pmc arginine residue, affording the corre-
sponding 16 (B1–6-Bi) batch. Hydrolysis with 2 M HCl
led to the corresponding acids 17 (B1–6-Bi). The six
batches 16 (B1–6-Bi) and the six 17 (B1–6-Bi) ones were
purified by semi-preparative HPLC. The HPLC analy-
sis of each of these clearly indicated the presence of the
six conjugates. This is illustrated in Fig. 4C for the
batch 16 (B1–6-In).
Figure 4. HPLC traces of: (A) the six B1–6 base acetic acids
and compound 12. (B) Z-[B1–6-Alloc]-(3)-Arg(Pmc)OMe 13
(B1–6-P2). (C) Batch 16 (B1–6-In) Z-[B1–6-In]-(3)-ArgOH.12
All batches of type 16 (B1–6-Bi) (esters) and 17 (B1–6-Bi)
(acids) constituting the four sublibraries are currently
under biological evaluation, for their antiviral activity
on cells as well as for their ability to inhibit viral targets
(RT, protease, integrase….). Moreover, to allow the
easy identification of active compounds in batches
showing some biological activity, without the need to
resynthezise every component separately, the corre-
sponding orthogonal library containing a well-identified
PNA at the first position (i.e. Z-[Bi-B1–6]-(n)-ArgOR) is
under preparation. Indeed, any batch of a given subli-
brary from library Z-[B1–6-Bi]-(n)-ArgOR and any one
of a given sublibrary from library Z-[Bi-B1–6]-(n)-
ArgOR share only one synthon. Therefore, for a given
screen, an active compound will confer activity to one
single batch of each libraries.
Sublibraries containing b-alanine (n=2), aminopen-
tanoic acid (n=4) and aminocaproic acid (n=5) spacers
have been prepared following the same strategy as
described for n=3.
In conclusion, to establish structure–activity relation-
ships for some promising antiHIV diPNA-arginine con-
jugates, we generated a library using a FPB strategy,
which requires easily obtainable building blocks. This
method allowed the preparation of 288 components,
split into four sublibraries. Each group constituting
these sublibraries has been synthesized in five steps
(esters) or six steps (acids) from a fully protected back-
Scheme 3. Generation of the sublibrary Z-[B1–6,Bi]-(3)-
ArgOR: (a) AcCl/MeOH (100%); (b) B1–6CH2CO2H, HATU,
DMF (87%); (c) Pd[P(Ph)3]4, DEA, DMF (98%); (d)
BiCH2CO2H, HATU, DMF (68–96%); (e) TFA/CH2Cl2 1/1
(100%); (f) 2 M HCl, CH3CN (98%).

1644
M.-L. Leroux et al. / Tetrahedron Letters 43 (2002) 1641–1644
bone, in an overall yield of 70% and with a high final
mixture purity. Biological evaluation of this library
should permit us to determine the target(s) and the
mode(s) of action of this family of compounds, as well
as to identify new active molecules and/or to optimize
the structure of active conjugates.
3. (a) Nielsen, P. E.; Egholm, M. Curr. Issues Mol. Biol.
1999, 1, 89–104; (b) Uhlmann, E.; Peyman, A.; Breipohl,
G.; Will, D. W. Angew. Chem., Int. Ed. Engl. 1998, 37,
2796–2823; (c) Dueholm, K. L.; Nielsen, P. E. New J.
Chem. 1997, 21, 19–31.
4. (a) Cook, P. D.; Kiely, J.; Sprankle, K. WO Patent
9,523,163, 1995; (b) Cook, P. D.; Kiely, J.; Sprankle, K.
US Patent 5,831,014, 1998; (c) Cook, P. D.; Kiely, J.;
Sprankle, K. US Patent 5,539,083, 2001.
It should be noted that one could extend this strategy
to prepare libraries of short linear and/or cyclic polyP-
NAs (less than decamer). Moreover, the monomeric
and polymeric protected backbones could also find an
application as building blocks in combinatorial solid-
phase syntheses of PNA libraries.
5. Soonkap, H.; Patron, A.; Fagnani, R. WO Patent 2000
002 899, 2000.
6. (a) Matysiak, S.; Reuthner, F.; Hoheisel, J. D. BioTech-
niques 2001, 31, 896–904; (b) Matysiak, S.; Hauser, N. C.;
Wurtz, S.; Hoheisel, J. D. Nucleosides Nucleotides 1999,
18, 1289–1291 and references cited therein.
7. Nielsen, P. E. Method Enzymol. 1996, 267, 426–433.
8. Challa, H.; Styers, M. L.; Woshi, S. A. Org. Lett. 1999,
10, 1639–1641.
Acknowledgements
9. Di Giorgio, C.; Pairot, S.; Schwergold, C.; Patino, N.;
Condom, R.; Farese-Di Giorgio, A.; Guedj, R. Tetra-
hedron 1999, 55, 1937–1958.
We thank the ‘Agence Nationale de Recherche sur le
Sida’ (ANRS), SIDACTION, CNRS, the ‘Re´gion
Provence-Alpes-Coˆte d’Azur’ and the Association pour
la Recherche sur le Cancer (ARC) for their support.
10. (a) Verheijen, J. C.; Grotenbreg, G. M.; Hart de Ruyter,
L.; van der Klein, P. A. M.; van der Marel, G. A.; van
Boom, J. H. Tetrahedron Lett. 2000, 41, 3991–3995; (b)
Depecker, G.; Schwergold, C.; Di Giorgio, C.; Patino,
N.; Condom, R. Tetrahedron Lett. 2001, 42, 8306–8308.
11. Nudelman, A.; Bechor, Y.; Falb, E.; Fisher, B.; Wexler,
B. A.; Nudelman, A. Synth. Commun. 1998, 28, 471–474.
12. HPLC analyses were performed with a 996 photodiode
array detector (195–290 nm), using a RP-18 (5 mm)
Licrospher 100 (250×4 mm) column as support. Elution
gradient A/B: 80/20 to 0/100 in 30 mn of solvents A
(0.1% TFA H2O) and B (0.1% TFA CH3CN) was used
with a flow of 1 mL/min. (B) Peaks attribution was
determined by means of UV spectra (umax are caracteris-
tic of each nucleobases). (C) Peaks attribution was per-
formed by LC/MS.
References
1. (a) Gallop, M. A.; Barrett, R. W.; Dower, W. J.; Fodor,
S. P. A.; Gordon, E. M. J. Med. Chem. 1994, 37,
1233–1280; (b) Terrett, N. K.; Gardner, M.; Gordon, D.
W.; Kobylecki, R. J.; Steele, J. Tetrahedron 1995, 51,
8135–8173; (c) Ellman, J. A.; Gallop, M. A. Curr. Opin.
Chem. Biol. 1998, 2, 317–319; (d) Guillier, F.; Orain, D.;
Bradley, M. Chem. Rev. 2000, 100, 2091–2158.
2. Farese-Di Giorgio, A.; Pairot, S.; Patino, N.; Condom,
R.; Di Giorgio, C.; Aumelas, A.; Aubertin, A. M.; Guedj,
R. Nucleosides Nucleotides 1999, 18, 263–275.