Modelling, synthesis and biological evaluation of an ethidium-arginine conjugate linked to a ribonuclease mimic directed against TAR RNA of HIV-1

Article abstract of DOI:10.1016/S0223-5234(02)01380-6

Using molecular modelling studies, an active anti-HIV ethidium-arginine conjugate targeted against the viral TAR RNA sequence has been linked to an artificial ribonuclease, with the aim to obtain an irreversible inhibitor. The ribonuclease moiety consists of an N-[N-(3-aminopropyl)-3-aminopropyl] glycine and has been constructed via two successive N-alkylations following the Fukuyama procedure.

Full text of DOI:10.1016/S0223-5234(02)01380-6

Eur. J. Med. Chem. 37 (2002) 573–584
www.elsevier.com/locate/ejmech
Original article
Modelling, synthesis and biological evaluation of an
ethidium–arginine conjugate linked to a ribonuclease mimic
directed against TAR RNA of HIV-1
Nadia Patino a, Christophe Di Giorgio a, Cristina Dan-Covalciuc a, Vale´rie Peytou a,
Raphae¨l Terreux b, Daniel Cabrol-Bass b, Christian Bailly c, Roger Condom a,*
a Laboratoire de Chimie Bio-organique, UNSA-CNRS UMR 6001, Uni6ersite´ de Nice Sophia-Antipolis, 06108 Nice, France
b LARTIC, Uni6ersite´ de Nice Sophia-Antipolis, 06108 Nice, France
c INSERM Unite´ 524, IRCL, 1 Place de Verdun, 59045 Lille, France
Received 2 December 2001; accepted 4 April 2002
Abstract
Using molecular modelling studies, an active anti-HIV ethidium–arginine conjugate targeted against the viral TAR RNA
sequence has been linked to an artificial ribonuclease, with the aim to obtain an irreversible inhibitor. The ribonuclease moiety
consists of an N-[N-(3-aminopropyl)-3-aminopropyl] glycine and has been constructed via two successive N-alkylations following
´
the Fukuyama procedure. © 2002 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved.
Keywords: HIV-1; TAR; RNA; Ribonuclease; Fukuyama; N-alkylation
or of polyamide nucleic acids [9]. In our case, the most
1. Introduction
promising conjugate of the ethidium–arginine series
(compound 1, Fig. 1) was shown (i) to inhibit HIV-1
replication (IC50=1 mM, without any toxicity up to
100 mM), and (ii) to interact strongly with the TAR
RNA. Furthermore, melting temperature and Rnase
protection experiments suggested that the ethidium
moiety of 1 was inserted next to the A17 residue while
the arginine side chain occupied the pyrimidine bulge of
TAR (Fig. 2), in agreement with molecular modelling
studies.
To make compound 1 a more potent inhibitor of the
HIV replication, we envisaged to link it to an artificial
ribonuclease, which could induce an irreversible hydro-
lysis of the TAR RNA. The glycine aminoacid was
selected as ribonuclease mimic rather than well known
polyamines [10–12] or imidazole moiety [13–16] be-
cause glycine–anthraquinone conjugates were shown to
hydrolyse efficiently and selectively 5%-cytidine(C)ꢀ
adenosine(A)-3% sites of tRNA, via an intramolecular
cooperation of carboxylate and ammonium ions [17].
As shown in Fig. 2, such a 5%-CA-3% sequence is found
in close proximity of the postulated intercalation site on
TAR RNA of the ethidium moiety of compound 1
The remarkable efficacy of protease and reverse trans-
criptase inhibitor associations for the treatment of
HIV-1 infection has been clearly established [1]. How-
ever, due to the rapid development of viral resistance
against these drugs [2], it is necessary to elaborate new
chemotherapeutic agents directed against other viral
targets. The interaction of the viral Tat protein with the
TAR RNA region of HIV-1 is essential for the regula-
tion of gene expression [3–5] and represents an attrac-
tive target for inhibiting HIV replication. Recently, we
rationally designed bifunctional molecules capable of
binding to two close sites of TAR which are involved in
the interaction with Tat [6]. These compounds were
constituted by an arginine residue covalently linked, via
a polyamide spacer, to an ethidium-intercalating moie-
ty. During the last decade, several arginine-containing
molecules directed against TAR RNA have been deve-
loped, including Tat-derived peptides and peptoids [7],
as well as arginine conjugates of aminoglycosides [7,8]
* Correspondence and reprint
E-mail address: condom@unice.fr (R. Condom).
´
0223-5234/02/$ - see front matter © 2002 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved.
PII: S0223-5234(02)01380-6

574
N. Patino et al. / European Journal of Medicinal Chemistry 37 (2002) 573–584
(C19ꢀA20). Thus, using molecular modelling studies,
compound 2 was designed in which the active ethi-
dium–arginine conjugate 1 is covalently attached, via a
polyamine linker, to a glycine residue (Fig. 1). The
nature and the optimal length of the linker was esta-
blished to allow the location of the glycine residue in
front of the potentially hydrolysable 5%-C19ꢀA20-3% base
pair of TAR (Fig. 2). This spacer consists of two
Fig. 4. Molecular model of the interaction between compound 2 and
TAR RNA of HIV-1. The ethidium moiety is in green, the arginine
residue in blue, the linker in yellow and the nucleasic moiety in red.
consecutive aminopropyl units in which the central
ammonium ion is expected to enhance the binding of 2
to RNA through electrostatic interactions.
In this paper, we report on the synthesis of com-
pound 2 following the Fukuyama procedure [18,19] as
well as its biological evaluation. In this context, an
unusual issue of the Fukuyama N-alkylation is also
reported.
Fig. 1. Structures of ethidium–arginine conjugates 1 and 2.
2. Chemistry
2.1. Dynamic molecular modelling studies
A model for the interaction of TAR RNA and
compound 2 was built starting from the previous one
established for compound 1 [4], and according to the
proposed mechanism for the 5%-CA-3% hydrolysis by a
glycine residue [17] (Fig. 3). First, the nucleasic moiety
was positioned in front of the phosphate group of the
potentially hydrolysable 5%-C19ꢀA20-3% site of TAR. Two
distance constraints were defined: the first one between
an oxygen atom of the carboxylate moiety of the
glycine residue and an hydrogen atom of the 2%-OH
group of C19, the second one between an hydrogen of
the ammonium ion of the glycine and an oxygen of the
targeted phosphate group. A linker was then designed
in order to connect the nucleasic moiety to the car-
boxylic group of the arginine residue of compound 1
and its length was geometrically optimised. Molecular
dynamic simulations were performed in order to evalua-
te the stability of the resulting hypothetical compound
2–TAR complex (Fig. 4). The means of the energy
interaction, mainly electrostatic, was found of −1023
kCal mol−1. The interaction mechanism was investi-
gated and its analysis showed that compound 2 first
interacts by its arginine domain. In a second step, the
Fig. 2. Schematic potential interaction of ethidium–arginine conju-
gate 2 with the TAR RNA of HIV-1.
Fig. 3. Mechanism proposed by Endo for the hydrolysis of tRNAPhe
by a glycine moiety.

N. Patino et al. / European Journal of Medicinal Chemistry 37 (2002) 573–584
575
main linker between arginine and ethidium penetrates
in the minor groove of TAR. Then, the ethidium moie-
ty intercalates between the base pairs C18ꢀG44 and
C19ꢀG43. Lastly, the nucleasic domain interacts with the
accessible phosphate group of the C19ꢀA20 base pair.
and the arginine–ribonuclease containing moiety 4, as
outlined in Fig. 5. The synthesis of the ethidium deriva-
tive 3 was previously described [6]. Compound 4 was
constructed starting from the linear N-Fmoc aminoacid
chain
(5),
the
commercially
available
N-
FmocꢀArg(Pmc)ꢀOH (6), and the suitably protected
polyamine key synthon 7 (see below).
2.2. Synthesis
The synthesis of compound 5 was carried out in two
steps starting from the commercially available N-Fmoc-
6-aminocaproic acid (8) and b-alanine methyl ester (9)
(Fig. 6). Coupling of 8 with 9 using Bop reagent led to
10 (91%) which was then hydrolysed with HCl to yield
5 (85%).
As the amide bond formation from suitably pro-
tected compound 1 and the ribonuclease moiety would
most probably lead to substantial epimerisation of the
arginine residue in 1, the synthesis of compound 2 was
alternatively performed from the ethidium derivative 3
The major difficulty in synthesising compound 4 lay
in the construction of the ribonuclease-containing
polyamine moiety 7 (Fig. 5). Various synthetic ap-
proaches have been described for the synthesis of selec-
tively N-functionalised polyamines [20]. These strategies
require either orthogonal protections of the amino
groups incorporated into an available polyamine skele-
ton, or stepwise construction of the desired polyamine
chain using reactions, such as nucleophilic substitution
[21], reductive alkylation [22] or, more recently, N-alky-
lation of sulfonamides [18,19,23,24]. More particularly,
the Fukuyama’s 2-nitrobenzenesulfonyl (o-Ns) group
has been extensively used, both in solution and solid
phase, for the selective preparation of N-protected pri-
mary amines [25], N-substituted amino acid derivatives
[26–30], N-alkylated peptides [31], oligoureas [32],
polyamides [33] and PNA [34]. Moreover, its utility in
organic chemistry as an amino protecting group is most
promising, as its removal occurs under mild conditions
(PhS−) which are highly selective as they do not affect
the most commonly used amino protecting groups (i.e.
Boc, Z, Alloc).
To explore the potential of the o-Ns group in multi-
step syntheses, and in particular as a permanent amino
protecting group, we first elaborated strategy A to
prepare compound 2, as outlined in Fig. 7. Although
the key di-o-Ns-protected compound 21 could be syn-
thesised, our attempts to remove the two o-Ns groups
by a conventional treatment (PhSH–DIEA) were un-
successful, as we obtained a complex mixture from
which our target derivative 2 could not be isolated.
Moreover the N-alkylation of the di-o-Ns-protected
compounds 15a,b applying the Fukuyama’s procedure,
was found to be hampered by the presence of the o-Ns
group on the amino function of the glycine residue.
Indeed, and even under mild conditions (i.e. 1 equiv. of
Cs2CO3 and room temperature (r.t.)), compounds 16a,b
were obtained from the di-o-Ns-protected 15a,b and
from the N-protected aminopropyl bromide 11a,b with
only low to medium yields (25 and 50%, respectively)
together with the unexpected by-products 17a,b (about
20%) and 18a (10%), thus complicating substantially
their purification. The structures of 17a,b and 18a,
Fig. 5. Retrosynthesic scheme of compound 2. (atom numbering
corresponds to NMR assignments).
Fig. 6. Synthesis of compound 5. Reagents and conditions: (i) Bop,
DIEA, CH2Cl2. (ii) 1/1 4 N HCl–dioxane.

576
N. Patino et al. / European Journal of Medicinal Chemistry 37 (2002) 573–584
ester function (for the glycine derivatives 11a to 16a,
this signal was measured within a 3.65–3.80 ppm
1
range). Moreover, a H signal corresponding either to
the two Ha of the glycine residue, as in the starting
material, or to a single Ha, as in intermediate Ia (see
1
Fig. 8) was not detected in the H-NMR spectrum of
neither 17a nor 18a, confirming that rearrangement,
cyclisation and water elimination involved the glycine
methyl ester residue, following the mechanism proposed
in Fig. 8. This mechanism consists in an unusual based-
catalysed rearrangement of 15a,b to give the nitroani-
line intermediates Ia,b, which then undergo an
intramolecular cyclisation, producing the six-member
ring derivatives 17a,b. The formation of 18a results
from N-alkylation of 17a with bromide 11a. The com-
plex rearrangement of 15a,b into intermediates Ia,b is in
line with the proposed one in the literature [28]. It is
initiated by the deprotonation of the glycine methylene
followed by a 1,2-alkyl shift, probably by a Stevens
rearrangement whose mechanism is not fully under-
stood [35]. A SNAr rearrangement, then a loss of SO2
leads to the 2-nitroaniline derivatives Ia–b. Concerning
the conversion of I into 17, it is noteworthy that
replacement of the glycine methyl ester group of 15a by
Fig. 7. Attempt in the synthesis of compound 2 following the strategy
A Reagents and conditions: (i) for 11a: (Boc)2O, TEA, CH2Cl2 or for
11b: MmtꢀCl, TEA, CH2Cl2; (ii) o-NsꢀCl, NMM, CH2Cl2, 0 °C; (iii)
Cs2CO3, DMF; (iv) for 11a: 50% TFA–CH2Cl2 or for 11b: 1%
TFA–CH2Cl2; (v) Bop, DIEA, CH2Cl2; (vi) DEA, CH2Cl2; (vii)
Fmoc-deprotected 19, HOSu, DCC, DMF; (viii) Fmoc deprotected
20, Bop, HOBt, DIEA, DMF.
derivating from 2-nitro aniline, were unambiguously
1
attested by mass spectrometry and H-NMR. Further-
more, the formation of such by-products is consistent
with the reported formation of 2-nitroaniline deriva-
tives, which occurs during the Fukuyama N-alkylation
of o-Ns-protected valine methyl ester [28]. The mass
spectrum of 17a confirmed, with respect to structure
16a, the loss of SO2 and of H2O, indicating that, as
displayed in Fig. 8 (see discussion below), rearrange-
ment, cyclisation and subsequent water elimination pro-
ducing 17a, have occurred. Concerning the mass
spectrum of 18a, it confirmed, with respect to 17a, the
presence of a N-Bocꢀaminopropyl group (also attested
1
by H-NMR) and consequently, supported alkylation
of 17a with bromide 11a. That this alkylation involved
the amine group of aniline in 17a was suggested by the
1
absence, in the H-NMR spectrum of 18a, of the NH
signal, measured at 6.9 ppm for 17a. One should under-
score that, excepted the specific NH and
BocꢀNH(CH2)3 signals for 17a and 18a, respectively,
1
these two derivatives display a very similar H pattern,
indicating a very close chemical structure. Furthermore,
1
their H-NMR spectra exhibited a methoxy singlet at
Fig. 8. Putative mechanism for the formation of by-products 17a,b
and 18a.
4.05 ppm that is no more characteristic of a methyl

N. Patino et al. / European Journal of Medicinal Chemistry 37 (2002) 573–584
577
The difficulties to obtain the di-(o-Ns)-protected
compounds 16a,b and our unsuccessful attempts to
deprotect the two o-Ns groups at the last stages of the
synthesis of 2 led us to elaborate the strategy B shown
in Fig. 9. It consisted in replacing the troublesome o-Ns
groups by the Boc amino protecting group after each
N-alkylation step. This replacement allowed us to ob-
tain our target compound 2 after classical deprotection
of 30 (i.e. the Boc analog of the o-Ns derivative 21).
Moreover, this substitution was found to be more
efficient for the preparation of 25 (i.e. the Boc analog of
the o-Ns derivative 16), as the key Fukuyama N-alkyla-
tion of 24 proceeded smoothly and cleanly.
Thus, starting from 13b, compound 22 was isolated
in 70% yield after the cleavage of the o-Ns group with
PhSH and DIEA, and its replacement by the Boc
group, by means of Boc2O. These two steps were
performed without isolation of the deprotected interme-
diate, thus shortening substantially this synthetic path-
way. The selective cleavage of the Mmt protecting
group of the Mmt-, Boc and O-t-Bu-protected com-
pound 22 was achieved almost quantitatively by means
of 1% TFA–CH2Cl2, affording the TFA salt 23. Reac-
tion between 2-nitro-benzenesulfonyl chloride and 23
gave the corresponding sulfonamide 24 in 86% yield,
which was then alkylated with N-Mmt aminopropyl
bromide 11b, affording compound 25 in high yield
(95%). The replacement, ‘in one pot’, of the o-Ns group
of 25 by the Boc one led to 26 in 85% yield. Finally, the
cleavage of the Mmt protecting group in 26 was
achieved selectively with 1% TFA–CH2Cl2, giving
quasi quantitatively the key synthon 7c.
The next step consisted in the preparation of 27, by
coupling 7c with 6 using Bop reagent (90%). Removal
of the Fmoc protecting group in 27 by means of
diethylamine yielded 28 (95%), which was then con-
densed with the N-Fmoc aminoacid 5, via a DCC–
HOSu preactivation, to lead the fully protected
compound 29 (60%). At last, Fmoc cleavage with die-
thylamine afforded compound 4 in 95% yield.
Finally, condensation of the ethidium derivative 3
onto the free terminal amino group of 4 (Bop–HOBt
activation), afforded the polyprotected conjugate 30 in
65% yield. Simultaneous cleavage of the Pmc, Boc and
tert-butyl protecting groups of the various amine and
acid functions was performed quantitatively using
TFA–H2O. At last, compound 2 was purified by semi-
preparative HPLC. Its purity was demonstrated by
HPLC analyses and its structure confirmed by HRMS
experiments.
Fig. 9. Synthesis of compound 2 following the strategy B. Reagents
and conditions: (i) PhSH, DIEA, DMF, 2–6 h then (Boc)2O; (ii) 1%
TFA–CH2Cl2; (iii) o-NsꢀCl, NMM, CH2Cl2; (iv) Cs2CO3, DMF; (v)
Bop, DIEA, CH2Cl2; (vi) DEA, CH2Cl2; (vii) DCC, HOSu, DMF, 12
h then 29, NMM; (viii) Bop, HOBt, DIEA, DMF; (ix) 9/1 TFA–
H2O.
the bulky tertiobutyl one, as in 15b, did not prevent the
intramolecular cyclisation of Ib.
This side reaction highlights the enhanced acidic
character of the a protons in the N-alkylated N-sul-
fonylated glycine residue of 15a,b which, in this con-
text, consistently limits the potential of the o-Ns group
as a permanent amino protecting group of the glycine.
3. Pharmacology
The binding of compound 2 to TAR RNA was
evaluated by melting temperature experiments, by

578
N. Patino et al. / European Journal of Medicinal Chemistry 37 (2002) 573–584
measuring its ability to alter the thermal denaturation
profile of the TAR RNA sequence. The capacity of 2 to
hydrolysis the TAR RNA sequences was evaluated by
polyacrylamide gel electrophoresis. Finally, the anti-
HIV-1 activity and cytotoxicity of compound 2 were
evaluated in infected CEM-SS, MT4 and PBMC cells.
6. Experimental
Unless otherwise stated, all reagents were obtained
from commercial suppliers and were used without fur-
ther purification. All solvents were freshly distilled. The
following abbreviations are employed: Ethidium (Eth),
N,N%-dicyclohexylcarbodiimide (DCC), N-hydroxysuc-
cinimide (HOSu), 1-hydroxybenzotriazole (HOBt), ben-
zotriazole-1-yl-oxy-tris-(dimethylamino)-phosphonium-
hexafluorophosphate (Bop), 2-nitrobenzenesulfonyl
chloride (o-NsꢀCl), monomethoxytrityl chloride
(MmtꢀCl), tert-butyloxycarbonyl (Boc), allyloxycar-
bonyl (Alloc), benzyloxycarbonyl (Z), 2,2,5,7,8-pen-
tamethylchroman-6-sulfonyl (Pmc), N-a-9-fluorenylme-
thoxycarbonyl - NG - (2,2,5,7,8 - pentamethylchroman-6-
sulfonyl) arginine [FmocꢀArg(Pmc)OH], N-methylmor-
pholine (NMM), diethylamine (DEA), diisopropyl-
ethylamine (DIEA), triethylamine (TEA), trifluo-
roacetic acid (TFA), acetic acid (AcOH), dimethylfor-
mamide (DMF), hexane (Hex). Melting points were
determined using an electrothermal digital melting
point apparatus IA900 Series and were uncorrected.
1H-NMR spectra were recorded on a Brucker AC 200
(200 MHz) Fourier Transform spectrometer. Thin-layer
chromatographies (TLC) were run on precoated silica
gel plates (Merck, 60F 254). Column chromatographies
were carried out on silica gel 60 (Merck 230–400 mesh
ASTM) or on LH20 Sephadex (Sigma, 25–100 mm).
HPLC analyses and purifications were performed on a
HP1100 equipped with UV detector (at 254 nm), using
a C18 Merck Interchrom column as support. Elution
gradients of H2O (0.1% TFA)–CH3CN (0.1% TFA)
were used with a flow of 1 mL min−1. Optical rotations
were evaluated on a Bellingham and Stanley polarime-
ter (ADP220) and were at 91°. Mass spectrometry
analyses were carried out on a TSQ 7000 FINNIGAN
MAT (ESIMS) instrument. FAB–HRMS was per-
formed by the Service Central de Microanalyses, CNRS
(Vernaison-France).
4. Results and discussion
Melting temperature studies clearly indicated that the
ligand 2 maintained a high affinity for the TAR RNA
sequence, which was further comparable with that of 1.
Indeed, in the presence of conjugate 2, at a conjugate/
RNA ratio of 0.4, the DTm of the TAR RNA (DTm=
Tm complex−Tm RNA) was increased by 10 °C
whereas at the same ratio, the DTm increase was only
of 7.5 °C in the presence of ethidium bromide, and of
11.5 °C in the presence of conjugate 1 [6]. Disappoin-
tingly, however, no hydrolysis of the TAR RNA by 2
was observed at 37 °C and pH 7, as attested by poly-
acrylamide gel electrophoresis, suggesting an inappro-
priate positioning of the cleaver towards the potential
scission site.
Moreover, unlike conjugate 1, compound 2 was de-
void of any antiviral activity nor of toxicity (IC50 and
CC50\10−5 M). As melting temperature experiments
showed that compound 2 was able to bind to TAR in
vitro as efficiently as compound 1, the lack of any
anti-HIV activity for 2 could be due to a poor intracel-
lular uptake.
5. Conclusion
An efficient procedure has been developed to synthe-
sise a designed ethidium–arginine–ribonuclease conju-
gate targeting the HIV-1 TAR RNA sequence. In the
context of the Fukuyama N-alkylation of the N-(3-
aminopropyl) N-(2-nitrobenzenesulfonyl) glycine moie-
ty, the 2-nitrobenzenesulfonyl group was found to be
unsuitable as permanent amino protecting group of the
glycine residue. Nevertheless, it could be used provi-
sionally for the N-alkylation and then efficiently re-
6.1. Dynamic molecular modelling studies
All molecular modelling studies were performed
using the Insight II/Discover molecular modelling pac-
kage. Simulations were performed for a duration of 200
ps at constant energy using a dielectric continuum to
simulate the solvent, and the co-ordinates of the diffe-
rent atoms of TAR RNA were frozen. Energy was
computed with CFF93 as force field.
placed in
a
‘one pot’ procedure by the Boc
urethane-protecting group.
Biological studies have shown that although the
ethidium–arginine–ribonuclease conjugate 2 strongly
interacted with TAR RNA, it did not induce its clea-
vage nor displayed an antiviral activity on HIV-infected
cells. Changes in both the length and the nature of the
spacer linking the glycine and arginine residue in view
of inducing TAR–RNA hydrolysis are currently under
consideration. Moreover, a prodrug approach involving
conversion of the carboxylic acid into an ester must be
considered, in view of increasing the cell membrane
permeability.
6.2. Chemistry
6.2.1. FmocꢀNH(CH2)5CONH(CH2)2CO2Me (10)
Bop (442 mg, 1.0 mmol) was added at 0 °C to a
stirred solution of N-Fmoc-6-aminocaproic acid (8)
(353 mg, 1.0 mmol), HCl·b-alanine methyl ester (9) (168

N. Patino et al. / European Journal of Medicinal Chemistry 37 (2002) 573–584
579
mg, 1.2 mmol), DIEA (0.7 mL, 4.0 mmol) in CH2Cl2 (4
mL). After 2 h stirring, the mixture was diluted with
CH2Cl2 (20 mL). The organic layer was washed succes-
sively with a 1 M aqueous KHSO4 solution, a 10%
aqueous NaHCO3 solution, brine and dried over
MgSO4. The solvent was evaporated in vacuo. Com-
pound 10 was obtained as an amorphous powder (91%)
and used without further purification. TLC (7/3 Hex–
washed with an aqueous 10% NaHCO3 solution, with
an aqueous 10% citric acid solution and finally with
H2O. The solution was dried over MgSO4 and evapo-
rated to dryness. Et2O was added to precipitate com-
pound 12a, which was then filtered off. Compound 12a
was obtained in 84% yield, as a white amorphous
powder. TLC (3/7 Hex–AcOEt, UV) Rf=0.54; 1H-
NMR (CDCl3) l: 8.15 (1H, o-Ns, dd), 8.0 (1H, o-Ns,
dd), 7.8 (2H, o-Ns, td), 6.15 (1H, NH, s), 4.1 (2H, CH2
a (Gly), s), 3.65 (3H, OCH3, s); MS m/z=275.1 [M+
H]+.
1
EtOAc, UV) Rf=0.45; H-NMR (CDCl3) l: 7.9–7.2
(8H, Fmoc, m), 6.1 (1H, NH amide, m), 5.25 (1H, NH
(Fmoc), m), 4.25 (2H, CH2 (Fmoc), d), 4.15 (1H, CH
(Fmoc), t), 3.7 (3H, OCH3, s), 3.5 (2H, CH2-8, q), 3.2
(2H, CH2-13, q), 2.55 (2H, CH2-7, t), 2.15 (2H, CH2-9,
t), 1.8–1.2 (6H, CH2-10,11,12, m); MS m/z=439.2
[M+H]+.
6.3.3. BocꢀNH(CH2)3Nꢀ(o-Ns)CH2CO2CH3 (13a)
Compound 12a (274 mg, 1.0 mmol) was dissolved in
DMF (2 mL) then Cs2CO3 was added (358 mg, 1.1
mmol). The solution was stirred 30 min at r.t., then
compound 11a (357 mg, 1.5 mmol) was added. Stirring
was continued for 18 h. DMF was removed under
reduced pressure and the residue was dissolved in
EtOAc. The organic layer was washed with H2O, dried
over MgSO4 and evaporated to dryness. The product
was purified on a silica gel column (7/3–1/1 Hex–
EtOAc) and compound 13a was obtained as a colour-
less oil in 80% yield. TLC (1/1 Hex–AcOEt, UV)
Rf=0.67; 1H-NMR (CDCl3) l: 8.15 (1H, o-Ns, m),
7.85 (3H, o-Ns, m), 5.0 (1H, BocꢀNH, m), 4.3 (2H,
CH2 a (Gly), s), 3.8 (3H, OCH3, s), 3.5 (2H, CH2-1, t),
3.25 (2H, CH2-3, m), 1.8 (2H, CH2-2, m), 1.5 (9H, Boc,
s); MS m/z=432.1 [M+H]+.
6.2.2. FmocꢀNH(CH2)5CONH(CH2)2CO2H (5)
To a solution of compound 10 (438 mg, 1.0 mmol) in
dioxane (10 mL) was added 4N aqueous solution HCl
(10 mL). The mixture was heated at 50 °C for 1 h
EtOAc (50 mL) was then added. The organic layer was
washed with water until pH 7, with brine then dried
over MgSO4. The solvent was evaporated in vacuo and
the residue was triturated with Et2O. The white amor-
phous powder was filtered off giving 5 in 85% yield.
TLC (99/1 EtOAc–AcOH, UV) Rf=0.30; 1H-NMR
(CD3OD) l: 7.75 (2H, Fmoc, d), 7.65 (2H, 2H, Fmoc,
d), 7.4 (4H, Fmoc, m), 4.35 (2H, CH2 Fmoc, d), 4.25
(1H, CH Fmoc, t), 3.5 (2H, CH2-8, t), 3.1 (2H, CH2-13,
t), 2.5 (2H, CH2-7, t), 2.25 (2H, CH2-9, t), 1.6–1.4 (6H,
CH2-10,11,12, m); MS m/z=425.1 [M+H]+.
6.3.4. TFA·NH2(CH2)3Nꢀ(o-Ns)CH2CO2CH3 14a
Compound 13a (1.1 g, 2.55 mmol) was dissolved in 6
mL of a solution of 1/1 TFA–CH2Cl2. After stirring
for 1 h at 0 °C, the solvent was evaporated in vacuo
and the residue was taken up in Et2O. The white solid
that precipitated was filtered off, giving the TFA salt
14a in 95% yield. TLC (1/1 EtOAc–MeOH, ninhy-
6.3. Synthesis of compounds 11a–16a, by-products 17a
and 18a
6.3.1. N-Boc-3-bromopropylamine (11a)
TEA (1.40 mL, 10 mmol) was slowly added to a cold
solution (0 °C) of HBr·3-bromopropylamine (1.1 g, 5.0
mmol) and Boc2O (1.09 g, 5.0 mmol) in 30 mL of
CH2Cl2. The solution was stirred 2 h at r.t., then
washed with water. The organic layer was evaporated
under reduced pressure. The residue was purified by
silica gel column chromatography (Hex 100% to 1/1
Hex–AcOEt). Compound 11a was obtained as an oil in
82% yield. TLC Rf (CH2Cl2, ninhydrine) Rf=0.39;
1H-NMR (CDCl3) l: 4.9 (1H, NH, m), 3.4 (2H,
CH2Br, t), 3.15 (2H, BocꢀNHCH2, m), 2.0 (2H,
CꢀCH2ꢀC, m), 1.5 (9H, Boc, s); MS m/z=238.3, 240.4
[M+H]+.
1
drine) Rf=0.30; H-NMR (CDCl3) l: 8.1 (1H, o-Ns,
m), 7.85 (3H, o-Ns, m), 4.3 (2H, CH2a (Gly), s), 3.8
(3H, OCH3, s), 3.6 (2H, CH2-1, m), 3.3 (2H, CH2-3, m),
2.1 (2H, CH2-2, m); MS m/z=332.2 [M+H]+.
6.3.5. o-NsꢀNH(CH2)3Nꢀ(o-Ns)CH2CO2CH3 (15a)
Compound 15a was obtained from 14a (445 mg, 1.0
mmol) and o-NsꢀCl (221 mg, 1.0 mmol), following the
above procedure for the preparation of 12a. Purifica-
tion by silica gel column chromatography (7/3–1/1
Hex–EtOAc) afforded 15a in 75% yield, as an amor-
phous solid. TLC (1/1 Hex–EtOAc, UV) Rf=0.42;
1H-NMR (CDCl3) l: 8.2 (1H, o-Ns, m), 8.1 (1H, o-Ns,
m), 8.0–7.6 (6H, o-Ns, m), 5.85 (1H, NH, t), 4.1 (2H,
CH2a (Gly), s), 3.8 (3H, OCH3, s), 3.55 (2H, CH2-1, t),
3.3 (2H, CH2-3, q), 1.9 (2H, CH2-2, m); MS m/z=
517.4 [M+H]+.
6.3.2. N-(o-Ns) methyl glycinate (12a)
This compound was previously described in the liter-
ature [19]. NMM (1.1 mL, 10 mmol) was added to a
cold solution of CH2Cl2 (30 mL) containing
HCl·GlyꢀOMe (502 mg, 4.0 mmol) and o-NsꢀCl (886.5
mg, 4.0 mmol). The solution was stirred 2 h at r.t., then

580
N. Patino et al. / European Journal of Medicinal Chemistry 37 (2002) 573–584
6.3.6. BocꢀNH[(CH2)3Nꢀ(o-Ns)]2CH2CO2CH3 (16a),
by-products 17a and 18a
6.4.2. N-(o-Ns) tertiobutyl glycinate (12b)
This compound was prepared from AcOH·GlyꢀO
t-Bu (1.05 g, 5.35 mmol) and o-NsꢀCl (1.3 g, 5.50
mmol), following the above procedure for the prepara-
tion of 12a. Compound 12b was obtained in 90% yield,
as a white amorphous powder. TLC (7/3 Hexane–
Cs2CO3 (391 mg, 1.2 mmol) was added to a solution
of 15a (598.5 mg, 1.16 mmol) and 11a (552 mg, 2.32
mmol) in DMF (2 mL). The mixture was stirred for 4
h at r.t. then the solvent was evaporated in vacuo. The
residue was taken up with EtOAc and the organic layer
was successively washed with 10% aqueous NaHCO3
solution, 10% aqueous citric acid solution and water.
After drying over MgSO4, the solution was evaporated
to dryness. Purification by silica gel column chromatog-
raphy (1/1–0/1 Hex–EtOAc) gave 16a, 17a and 18a in
25, 11 and 18% yield, respectively.
1
AcOEt, UV) Rf=0.42; H-NMR (CDCl3) l: 8.20 (1H,
o-Ns, dd), 8.0 (1H, o-Ns, dd), 7.8 (2H, o-Ns, td), 6.1
(1H, NH, s), 4.0 (2H, Ha (Gly), s), 1.4 (9H, Ot-Bu, s);
MS m/z=317.0 [M+H]+.
6.4.3. MmtꢀNH(CH2)3Nꢀ(o-Ns)CH2CO2t-Bu (13b)
This compound was prepared from 12b (3.0 g, 9.5
mmol), 11b (7.75 g, 18.9 mmol) and Cs2CO3 (3.4 g, 10.4
mmol), following the above procedure for the prepara-
tion of 13a. The crude product was purified on silica gel
column (7/3–1/1 Hex–EtOAc) and compound 13b was
obtained as a colourless oil in 95% yield. TLC (7/3
6.3.7. Compound 16a
TLC (3/7 Hex–EtOAc, UV) Rf=0.65; 1H-NMR
(CDCl3) l: 8.0 (2H, o-Ns, m), 7.75–7.5 (6H, o-Ns, m),
4.8 (1H, NH, m), 4.15 (2H, CH2a (Gly), s), 3.65 (3H,
OCH3, s), 3.3 (6H, CH2-1,3,4, m), 3.1 (2H, CH2-6, m),
1.85 (2H, CH2-2, m), 1.7 (2H, CH2-5, m), 1.4 (9H, Boc,
s); MS m/z=696.3 [M+Na]+.
1
Hex–EtOAc, UV) Rf=0.52; H-NMR (CDCl3) l: 8.0
(1H, o-Ns, m), 7.7–7.1 (15H, o-Ns, Mmt, m), 6.85 (2H,
Mmt, d), 4.1 (2H, Ha (Gly), s), 3.8 (3H, OCH3, s), 3.5
(2H, CH2-1, m), 2.2 (2H, CH2-3, m), 1.9–1.6 (3H, NH,
CH2-2, m), 1.4 (9H, Ot-Bu, bs); MS m/z=668.45
[M+Na]+.
6.3.8. Compound 17a
TLC (3/7 Hex–EtOAc, UV) Rf=0.53; 1H-NMR
(CDCl3) l: 8.1 (1H, o-Ns, m), 8.0 (1H, Ar, m), 7.9–7.5
(4H, Ar and o-Ns, m), 7.3 (2H, Ar, m), 6.9 (1H, NH,
bs), 5.0 (2H, CH2Nꢀ(o-Ns), t), 4.05 (3H, OCH3, s), 3.1
(2H, CH2ꢀ(CꢁC), m), 2.1 (2H, CꢀCH2ꢀC, m);); MS
m/z=457.0 [M+Na]+.
6.4.4. TFA·NH2(CH2)3Nꢀ(o-Ns)CH2CO2t-Bu (14b)
Compound 13b (600 mg, 0.93 mmol) was dissolved in
15 mL of 1/99 TFA–CH2Cl2. The mixture was stirred
for 2 h then the solvent was evaporated in vacuo. The
residue was purified by silica gel column chromatogra-
phy (1/0–1/1 EtOAc–MeOH), giving 14b (92%) as a
colourless resin. TLC (8/2 EtOAc–MeOH, ninhydrine)
6.3.9. Compound 18a
TLC (3/7 Hex–EtOAc, UV) Rf=0.49; 1H-NMR
(CDCl3) l: 8.0 (1H, o-Ns, m), 7.9 (1H, Ar, m), 7.89–
7.5 (4H, Ar and o-Ns, m), 7.3 (2H, Ar, m), 5.0 (1H,
BocꢀNH, m), 4.80 (2H, CH2Nꢀ(o-Ns), t), 4.0 (3H,
OCH3, s), 3.6 (2H, BocꢀNHCH2, m), 3.4 (2H, CH2Nꢀ,
m), 3.15 (2H, CH2ꢀCꢁC), m), 2.1 (2H, CꢀCH2ꢀC (cycl),
m), 1.75 (2H, CꢀCH2ꢀC, m), 1.4 (9H, Boc, s); MS
m/z=614.2 [M+Na]+.
1
Rf=0.33; H-NMR (CDCl3) l: 8.1 (2H, o-Ns, m), 7.85
(2H, o-Ns, m), 4.1 (2H, Ha (Gly), s), 3.6 (2H, CH2-1,
m), 3.3 (2H, CH2-3, m), 2.1 (2H, CH2-2, m), 1.4 (9H,
Ot-Bu, s); MS m/z=374.2 [M+H]+.
6.4.5. o-NsꢀNH(CH2)3Nꢀ(o-Ns)CH2CO2t-Bu (15b)
Compound 15b was obtained from 14b (487 mg, 1.0
mmol) and o-NsꢀCl (222 mg, 1.0 mmol) following the
above procedure for the preparation of 15a. Purifica-
tion by silica gel column chromatography (7/3–1/1
Hex–EtOAc) afforded 15b in 77% yield, as a yellow
amorphous solid. TLC (1/1 Hex–EtOAc, UV) Rf=
0.53; 1H-NMR (CDCl3) l: 8.2–8.1 (2H, o-Ns, m),
8.0–7.6 (6H, o-Ns, m), 5.85 (1H, NH, t), 4.1 (2H, Ha
(Gly), s), 3.55 (2H, CH2-1, t), 3.3 (2H, CH2-3, q), 1.9
(2H, CH2-2, m), 1.4 (9H, Ot-Bu, s); MS m/z=559.1
[M+H]+.
6.4. Synthesis of compounds 11b–17b
6.4.1. N-Mmt-3-bromopropylamine (11b)
TEA (2.675 mL, 19 mmol) was slowly added to a
cold solution (0 °C) of HBr·3-bromopropylamine (2.0
g, 9.13 mmol) and MmtꢀCl (2.81 g, 9.13 mmol) in 10
mL of CH2Cl2. The solution was stirred for 2 h at r.t.,
then washed with water. The organic layer was evapo-
rated under reduced pressure. The residue was purified
by chromatography on silica gel column (1/0–9/1 Hex–
AcOEt). Compound 11b was obtained as a yellow oil in
1
95% yield. TLC (9/1 Hex–AcOEt, UV) Rf=0.72; H-
6.4.6. MmtꢀNH[(CH2)3Nꢀ(o-Ns)]2CH2CO2t-Bu (16b),
by-product 17b
Cs2CO3 (358 mg, 1.1 mmol) was added to a solution
of 15b (603 mg, 1.08 mmol) and 11b (886 mg, 2.16
mmol) in DMF (1.5 mL). The mixture was stirred for
18 h at r.t. then the solvent was evaporated in vacuo.
NMR (CDCl3) l: 7.6–7.1 (12H, Mmt, m), 6.9 (2H,
Mmt, dd), 3.85 (3H, OCH3, s), 3.65 (2H, CH2Br, t, J
6.7 Hz), 2.4 (2H, CH2NH, t, J 6.5 Hz), 2.1 (2H,
CH2CH2CH2, m), 1.7 (1H, NH, s); MS m/z=410.2,
412.1 [M+H]+.

N. Patino et al. / European Journal of Medicinal Chemistry 37 (2002) 573–584
581
The residue was taken up in EtOAc and the organic
6.4.11. Compound 20
layer was washed with 10% aqueous NaHCO3 solution
then with water. The solution was dried over MgSO4,
then evaporated to dryness. Purification by silica gel
column chromatography (7/3–1/1 Hex–EtOAc) gave
16b and 17b in 50 and 25% yield, respectively.
DEA (2 mL) was added to a solution of 19 (340 mg,
0.27 mmol) in CH2Cl2 (2 mL). After stirring for 2 h, the
solvent was evacuated in vacuo. The residue was
purified by silica gel column chromatography (1/0–1/1
EtOAc–MeOH), giving the free amino compound in
93% yield. This one (260 mg, 0.25 mmol), compound 5
(106 mg, 0.25 mmol), HOSu (43 mg, 0.375 mmol) and
NMM (28 mL, 0.25 mmol) were then placed in DMF (2
mL) at 0 °C. DCC (57.5 mg, 0.28 mmol) was added
and the mixture was stirred for 20 h at r.t. The solvent
was evaporated under reduced pressure. The residue
was taken up in EtOAc, DCU was filtered off and
washed with EtOAc. The filtrate was washed with a
10% aqueous NaHCO3 solution, brine and finally dried
over MgSO4. The solvent was evaporated in vacuo.
Purification by silica gel column chromatography (1/0–
8/2 EtOAc–MeOH) gave 20 as a colourless resin, in
70% yield. TLC (8/2 EtOAc–MeOH, UV) Rf=0.46;
1H-NMR (CDCl3) l: 7.9 (2H, o-Ns, m), 7.8–7.4 (10H,
o-Ns and Fmoc, m), 7.3 (4H, Fmoc, m), 7.1 (1H, NH
amide, m), 6.7 (1H, NH, m), 6.4–6.1 (3H, NH guani-
dinium, m), 5.3 (1H, NH (Fmoc), m), 4.45 (1H, Ha
(Arg), m), 4.25 (2H, CH2 (Fmoc), d), 4.15 (1H, CH
(Fmoc), t), 3.95 (2H, Ha (Gly), s), 3.45 (CH2-8, m),
3.4–3.0 (12H, CH2l (Arg), CH2-1,3,4,6,13, m), 2.50
(10H, 2CH3 (Pmc), CH2 (Pmc), CH2-7, m), 2.1 (2H,
CH2-9, m), 2.05 (3H, CH3m (Pmc), s), 1.9–1.65 (10H,
CH2-2,5,10,11,12, m), 1.6–1.4 (6H, CH2-b,g (Arg), CH2
(Pmc), m), 1.3 (9H, Ot-Bu, m), 1.15 (6H, 2CH3 (Pmc),
s); MS m/z=1467.1 [M+Na]+.
6.4.7. Compound 16b
TLC (1/1 Hex–EtOAc, UV) Rf=0.64; 1H-NMR
(CDCl3) l: 8.0 (2H, o-Ns, m), 7.6–7.0 (18H, Ar and
o-Ns, m), 6.7 (2H, Mmt, d), 4.0 (2H, Ha (Gly), s), 3.7
(3H, OCH3, s), 3.3 (6H, CH2-1,3,4, m), 2.0 (2H, CH2-6,
m), 1.7 (5H, CH2-2,5, NH, m), 1.4 (9H, Ot-Bu, s); MS
m/z=887.9 [M+H]+.
6.4.8. Compound 17b
TLC (1/1 Hex–EtOAc, UV) Rf=0.52; 1H-NMR
(CDCl3) l: 8.0 (2H, Ar and o-Ns, m), 7.7 (2H, Ar and
o-Ns, m), 7.3 (2H, Ar, m), 6.9 (1H, NH, m), 4.95 (2H,
CH2Nꢀ(o-Ns), t), 3.1 (2H, CH2ꢀ(CꢁC), m), 2.1 (2H,
CꢀCH2ꢀC, m), 1.65 (9H, Ot-Bu, s); MS m/z=477.2
[M+H]+.
6.4.9. TFA·NH2[(CH2)3Nꢀ(o-Ns)]2CH2CO2t-Bu (7b)
Compound 16b (220 mg, 0.25 mmol) dissolved in 4
mL of 1/99 TFA–CH2Cl2 was stirred for 2 h at 0 °C,
then the solvent was evaporated in vacuo. The residue
was purified by silica gel column chromatography (1/0–
1/1 EtOAc–MeOH), giving 7b (93%) as a colourless
resin. 1H-NMR (CDCl3) l: 8.15 (4H, o-Ns, m), 7.8–7.6
(4H, o-Ns, m), 4.0 (2H, Ha (Gly), s), 3.4 (6H, CH2-1, 3,
4, m), 3.3 (2H, CH2-6, m), 2.1 (4H, CH2-2, 5, m), 1.4
(9H, Ot-Bu, s); MS m/z=616.2 [M+H]+.
6.4.12. Compound 21
DEA (1 mL) was added to a solution of 20 (274 mg,
0.19 mmol) in CH2Cl2 (1 mL). After stirring for 2 h, the
solvent was evacuated in vacuo. The residue was
purified by silica gel column chromatography (1/0–1/1
EtOAc–MeOH) to give the free amino compound in
90% yield. This one (208 mg, 0.17 mmol), compound 3
(65 mg, 0.17 mmol), HOBt (24 mg, 0.17 mmol) and
DIEA (0.12 mL, 0.68 mmol) were then placed in DMF
(0.5 mL). Bop (75.1 mg, 0.17 mmol) was added at 0 °C
and the mixture was stirred for 2 h. Then, CH2Cl2 (10
mL) was added and the organic layer was washed with
a 10% aqueous NaHCO3 solution, water, dried over
MgSO4 and evaporated in vacuo. A LH 20 column
chromatography (1/0–1/1 H2O–MeOH) afforded com-
pound 21 as a purple resin in 50% yield. HPLC [8/2–0/
1 H2O (1/1000 TFA)–CH3CN (1/1000 TFA) in 30 min,
6.4.10. FmocꢀArg(Pmc)CONH[(CH2)3Nꢀ(o-Ns)]2-
CH2CO2t-Bu (19)
FmocArg(Pmc)OH (6) (378 mg, 0.57 mmol), 7b
(415.5 mg, 0.57 mmol) and DIEA (0.4 mL, 2.28 mmol)
were placed at 0 °C in CH2Cl2. Bop (252 mg, 0.57
mmol) was then added. The mixture was stirred for 2 h
at r.t. then CH2Cl2 (20 mL) was added. The organic
layer was washed with a 10% aqueous NaHCO3 solu-
tion, brine and dried over MgSO4. The solvent was
evaporated in vacuo. Purification by silica gel column
chromatography (1/0–1/1 EtOAc–MeOH) gave 19 as a
colourless resin, in 88% yield. TLC (EtOAc, UV) Rf=
1
0.42; H-NMR (CDCl3) l: 7.9 (2H, o-Ns, m), 7.8–7.4
(10H, o-Ns and Fmoc, m), 6.7 (1H, NH, m), 6.25 (3H,
guanidino, bs), 5.3 (1H, NH, m), 4.3 (1H, Ha (Arg),
m), 4.25 (2H, CH2 (Fmoc), d), 4.15 (1H, CH (Fmoc),
m), 4.0 (2H, CH2a (Gly), s), 3.5–3.0 (10H, CH2l (Arg),
CH2-1,3,4,6, m), 2.6–2.4 (8H, 2CH3 (Pmc), CH2 (Pmc),
m), 2.05 (3H, CH3 (Pmc), s), 1.8 (4H, CH2-2,5, m),
1.65–1.5 (6H, CH2b-g (Arg), CH2 (Pmc)), 1.4 (9H,
Ot-Bu, s), 1.4 (18H, Boc, m), 1.15 (6H, 2CH3 (Pmc), s);
MS m/z=1261.2 [M+H]+.
1
288 nm]: Rt=23.1 min; H-NMR (DMSO) l: 8.6 (2H,
H-e3, H-e4, m), 8.2 (2H, H-e9, H-e10, m), 7.9 (2H,
o-Ns, m), 7.8 (3H, H-e5, H-e7, H-e8, m), 7.6–7.1 (11H,
H-e1, H-e2, m), 7.0 (2H, NH2, bs), 6.6 (2H, NH2, bs),
6.35 (1H, H-e6, s), 4.2 (1H, Ha (Arg), m), 4.0 (3H,
N+(Me), s), 3.9 (2H, Ha (Gly), s), 3.4–3.0 (14H,
CH2-1,3,4,6,8,13, CH2-d (Arg), m), 2.5–2.35 (10H,

582
N. Patino et al. / European Journal of Medicinal Chemistry 37 (2002) 573–584
2CH3 (Pmc), CH2 (Pmc), CH2-7, m), 2.15 (2H, CH2-9,
m), 2.05 (3H, CH3m (Pmc), s), 1.8–1.6 (10H, CH2-
2,5,10,11,12, m), 1.6–1.4 (6H, CH2-b,g (Arg), CH2
(Pmc), m), 1.4 (9H, Ot-Bu, s), 1.15 (6H, 2CH3 (Pmc),
s); MS m/z=1548.3 [M]+.
6.4.16. MmtꢀNHꢀ(CH2)3ꢀN(o-Ns)ꢀ(CH2)3ꢀ
N(Boc)ꢀCH2CO2t-Bu (25)
Compound 24 (331 mg, 0.7 mmol) was dissolved in
DMF (2 mL) then Cs2CO3 was added (342 mg, 1.05
mmol). After stirring for 30 min at r.t., compound 11b
was added (570 mg, 1.4 mmol). Stirring was continued
at r.t. for 18 h. DMF was removed under reduced
pressure and the residue was dissolved in EtOAc. The
organic layer was washed with water, brine then dried
over MgSO4 and evaporated to dryness. The product
was purified by silica gel column chromatography (7/3–
1/1 Hex–EtOAc) and compound 25 was obtained as an
yellow oil in 95% yield. TLC (7/3 Hex–EtOAc, UV)
Rf=0.31; 1H-NMR (CDCl3) l: 8.0 (1H, o-Ns, m),
7.7–7.1 (15H, o-Ns and Mmt, m), 6.75 (2H, Mmt, d),
3.75 (3H, OCH3, s), 3.7 (2H, Ha (Gly), d) 3.1–3.4 (6H,
CH2-1,3,4, m), 2.05 (2H, CH2-6, m), 1.85–1.55 (5H,
NH, CH2-2,5, NH, m), 1.4 (9H, Ot-Bu, s), 1.35 (9H,
Boc, d); MS m/z=825.7 [M+Na]+.
6.4.13. MmtꢀNH(CH2)3Nꢀ(Boc)CH2CO2t-Bu (22)
PhSH (421 mL, 4.11 mmol) and DIEA (573 mL, 3.29
mmol) were added to a solution of compound 13b (529
mg, 0.82 mmol) in 3 mL of DMF. After 2 h of stirring,
Boc2O (218 mg, 1 mmol) was added. The mixture was
stirred two additional hours then DMF was evaporated
in vacuo. The residue was taken up with EtOAc. The
organic layer was washed with 10% aqueous NaHCO3
solution then with water and dried over MgSO4. The
solvent was evaporated in vacuo. The crude residue was
purified by silica gel column chromatography (CH2Cl2
then 7/3 Hex–EtOAc) to yield 22 in 75% yield. TLC
1
(9/1 Hex–EtOAc, UV) Rf=0.27; H-NMR (CDCl3) l:
7.6–7.1 (12H, Mmt, m), 6.85 (2H, Mmt, d), 3.8 (3H,
OCH3, s), 3.75 (2H, Ha (Gly), d), 3.35 (2H, CH2-1, m),
2.15 (2H, CH2-3, m), 1.9–1.6 (3H, NH, CH2-2, m), 1.5
(9H, Ot-Bu, s), 1.4 (9H, Boc, d); MS m/z=561.3
[M+H]+.
6.4.17. MmtꢀNHꢀ[(CH2)3ꢀN(Boc)]2ꢀCH2CO2t-Bu (26)
PhSH (338 mL, 3.3 mmol) and DIEA (460 mL, 2.64
mmol) were added to a solution of compound 25 (529
mg, 0.66 mmol) in 3 mL of DMF. After disappearance
of the starting material (about 6 h, TLC monitoring),
Boc2O (0.218 g, 1 mmol) was added. The mixture was
stirred for 4 h additional then DMF was evaporated in
vacuo. The residue was taken up with EtOAc. The
organic layer was washed with water and dried over
MgSO4. The solvent was evaporated in vacuo and the
crude residue was purified by silica gel column chro-
matography (CH2Cl2 then 7/3 Hex–EtOAc) to yield 26
in 85% yield. TLC (7/3 Hex–EtOAc, UV) Rf=0.67;
1H-NMR (CDCl3) 7.5–7.0 (12H, Mmt, m), 6.7 (2H,
Mmt, d), 3.75 (2H, Ha Gly), d), 3.7 (3H, OCH3, s),
3.35–3.0 (6H, CH2-1,3,4, m), 2.05 (2H, CH2-6, m),
1.75–1.55 (5H, NH, CH2-2,5, m), 1.4 (9H, Ot-Bu, s),
1.45 (18H, Boc, m); MS m/z=740.7 [M+Na]+.
6.4.14. TFA·NH2(CH2)3Nꢀ(Boc)CH2CO2t-Bu (23)
Compound 22 (605 mg, 1.08 mmol), dissolved in 15
mL of 1/99 TFA–CH2Cl2, was stirred at r.t. for 3 h.
The solvent was evaporated under reduced pressure and
the crude residue purified by silica gel column chro-
matography (1/0–0/1 EtOAc–MeOH). Compound 23
was obtained as a colourless resin (84%). TLC (8/2
EtOAc–MeOH, ninhydrine) Rf=0.47; 1H-NMR
(CDCl3) l: 3.8 (2H, Ha (Gly), d), 3.45 (2H, CH2-1, m),
3.25 (2H, CH2-3, m), 1.9 (3H, NH, CH2-2, m), 1.5 (9H,
Ot-Bu, s), 1.4 (9H, Boc, d); MS m/z=289.3 [M+H]+.
6.4.15. o-NsꢀNH(CH2)3Nꢀ(Boc)CH2CO2t-Bu (24)
Compound 23 (462 mg, 1.15 mmol) and o-NsꢀCl
(255 mg, 1.15 mmol) were placed at 0 °C in 6 mL of
CH2Cl2. NMM (380 mL, 3.45 mmol) was slowly added.
The mixture was stirred at r.t. for 2 h. The solvent was
then evaporated in vacuo. The residue was taken up in
EtOAc and the organic layer was washed with 10%
aqueous NaHCO3 solution, with brine and finally dried
over MgSO4. Evaporation of the solvent under reduced
pressure led to an oil that was purified by silica gel
column chromatography (1/1 Hex–EtOAc), giving 24
(86%) as a colourless resin. TLC (1/1 Hex–EtOAc, UV)
Rf=0.57; 1H-NMR (CDCl3) l: 8.1 (1H, o-Ns, m),
7.85–7.6 (3H, o-Ns, m), 6.35 (1H, NH, m), 3.65 (2H,
Ha (Gly), d), 3.35 (2H, CH2-3, m), 3.15 (2H, CH2-1,
m), 1.65 (2H, CH2-2, m), 1.5–1.3 (18H, Ot-Bu and
Boc, m); MS m/z=474.3 [M+H]+.
6.4.18. TFA·NH2ꢀ[(CH2)3ꢀN(Boc)]2ꢀCH2CO2t-Bu (7c)
Compound 26 (351 mg, 0.49 mmol) was dissolved in
8 mL of 1/99 TFA–CH2Cl2.The mixture was stirred at
r.t. for 3 h. The solvent was then evaporated under
reduced pressure and the crude residue purified by silica
gel column chromatography (EtOAc 100% to 8/2
EtOAc–MeOH), allowing the isolation of 7c as an oil
(85%). TLC (8/2 EtOAc–MeOH, ninhydrine) Rf=
0.60; 1H-NMR (CDCl3) l: 3.75 (2H, Ha (Gly), d),
3.35–3.0 (6H, CH2-1,3,4, m), 3.3 (2H, CH2-6, m), 1.9–
1.7 (4H, CH2-2,5, m), 1.4 (9H, Ot-Bu, s), 1.45 (18H,
Boc, m); MS m/z=445.5 [M+H]+.
6.4.19. FmocꢀNHArg(Pmc)CONHꢀ[(CH2)3ꢀN(Boc)]2ꢀ
CH2CO2t-Bu (27)
This compound was synthesised following the same
protocol as for compound 19, starting from

N. Patino et al. / European Journal of Medicinal Chemistry 37 (2002) 573–584
583
FmocꢀArg(Pmc)OH (6) (332 mg, 0.5 mmol) and com-
6.4.22. Compound 4
pound 7c (280 mg, 0.5 mmol). Purification by silica gel
column chromatography (EtOAc) gave 27 as a colour-
less resin (92%). TLC (EtOAc, UV) Rf=0.38; 1H-
NMR (CDCl3) l: 7.7–7.1 (8H, Fmoc, m), 6.4–6.0 (3H,
guanidino, bs), 5.3 (1H, NH, m), 4.3 (1H, Ha (Arg),
m), 4.25 (2H, CH2 (Fmoc), d), 4.15 (1H, CH (Fmoc),
m); 3.7 (2H, Ha (Gly), d), 3.5–2.9 (10H, CH2d (Arg),
CH2-1,3,4,6, m), 2.65–2.40 (8H, 2CH3 (Pmc), CH2
(Pmc), m), 2.0 (3H, CH3 (Pmc), s), 1.85–1.45 (10H,
CH2-b,g (Arg), CH2 (Pmc), CH2-2,5, m), 1.4 (9H, Ot-
Bu, s), 1.35 (18H, Boc, m), 1.15 (6H, 2CH3 (Pmc), s);
MS m/z=1113.2 [M+Na]+.
To a solution of 29 (128 mg, 0.1 mmol) in CH2Cl2 (1
mL) was added DEA (1 mL). After stirring for 3 h, the
solvent and DEA were removed under reduced pres-
sure. The residue was triturated three times with a 1/1
Et2O–Hex mixture, allowing the isolation of 4 in 85%
yield. Compound 4 was used in the next step without
further purification. HPLC [H2O (1/1000 TFA)–
CH3CN (1/1000 TFA): from 80/20 to 0/100% in 30
1
min, 254 nm] Rt=20.2 min; H-NMR (CDCl3) l: 6.3
(3H, NH guanidinium, m), 4.45 (1H, Ha (Arg), m),
3.75 (2H, Ha (Gly), d), 3.45 (2H, CH2-8, m), 3.15 (10H,
CH2-d (Arg), CH2-1,3,4,6, m), 2.95 (2H, CH2-13, m),
2.7–2.4 (10H, 2CH3 (Pmc), CH2 (Pmc), CH2-7, m),
2.15 (2H, CH2-9, m), 2.05 (3H, CH3m (Pmc), s), 1.9–
1.2 (16H, CH2-2,5,10,11,12, CH2-b,g (Arg), CH2 (Pmc),
m), 1.4 (27H, Ot-Bu, 2 Boc, 2 s), 1.2 (6H, 2CH3 (Pmc),
s); MS m/z=1074.6 [M+Na]+
6.4.20. H2NꢀArg(Pmc)CONHꢀ[(CH2)3ꢀN(Boc)]2ꢀ
CH2CO2t-Bu (28)
DEA (2 mL) was added to a solution of 27 (185 mg,
0.17 mmol) in CH2Cl2 (2 mL). After stirring for 2 h, the
solvent was evaporated in vacuo. The residue was
purified by silica gel column chromatography (from
EtOAc to 1/1 EtOAc–MeOH) to give 28 as an oil in
93% yield. TLC (EtOAc–MeOH 1:1, ninhydrine) Rf=
6.4.23. Compound 30
Bop (42 mg, 95 mmol) was added to a solution of
ethidium derivative 3 (36 mg, 95 mmol), compound 4
(100 mg, 95 mmol), HOBt (13 mg, 95 mmol) and DIEA
(70 mL, 0.4 mmol) in DMF (0.5 mL). The mixture was
stirred for 2 h, then CH2Cl2 (10 mL) was added. The
1
0.35; H-NMR (CDCl3) l: 6.25 (3H, guanidinium, bs),
4.5 (1H, Ha (Arg), m), 3.7 (2H, Ha (Gly), d), 3.4–3.0
(10H, CH2d (Arg), CH2-1,3,4,6, m), 2.65–2.40 (8H,
2CH3 (Pmc), CH2 (Pmc), m), 2.0 (3H, CH3 (Pmc), s),
1.85–1.45 (10H, CH2-b,g (Arg), CH2 (Pmc), CH2-2,5,
m), 1.45–1.3 (27H, Ot-Bu, 2 Boc, 2 s), 1.15 (6H, 2CH3
(Pmc), s); MS m/z=891.2 [M+Na]+.
organic layer was washed with
a 10% aqueous
NaHCO3 solution, water then dried over MgSO4 and
evaporated in vacuo. A LH 20 column chromatography
(from H2O to 1/1 H2O–MeOH) afforded compound 30
as a purple resin in 70% yield. HPLC [H2O (1/1000
TFA)–CH3CN (1/1000 TFA): from 80/20 to 0/100% in
30 min, 288 nm]: Rt=22.2 min; [h]D= −87° (MeOH,
6.4.21. Compound 29
Compound 5 (153 mg, 0.36 mmol) and HOSu (62
mg, 0.54 mmol) were placed in DMF (2 mL) at 0 °C.
DCC (82 mg, 0.4 mmol) was then added. The reaction
mixture was stirred for 12 h at r.t. then compound 28
(312.5 mg, 0.36 mmol) and NMM (60 mL, 0.54 mmol)
were added. The solution was stirred for 4 h, then the
solvent was evaporated under reduced pressure. The
residue was taken up in EtOAc, DCU was filtered off
and washed with EtOAc. The filtrate was washed with
a 10% aqueous NaHCO3 solution, brine and finally
dried over MgSO4. The solvent was evaporated in
vacuo. The residue was purified by silica gel column
chromatography (EtOAc to 8/2 EtOAc–MeOH), giving
29 in 60% yield. TLC (EtOAc, UV) Rf=0.25; 1H-NMR
(CDCl3) l: 7.8–7.15 (8H, Fmoc, m), 7.1 (1H, NH
amide, m), 6.45–6.15 (3H, NH guanidinium, m), 5.5
(1H, NH (Fmoc), m), 4.45 (1H, Ha (Arg), m), 4.30
(2H, CH2 (Fmoc), d), 4.15 (1H, CH (Fmoc), t), 3.75
(2H, Ha (Gly), d), 3.45 (CH2-8, m), 3.15 (12H, CH2-d
(Arg), CH2-1,3,4,6,13, m), 2.50 (10H, 2CH3 (Pmc), CH2
(Pmc), CH2-7, m), 2.1 (2H, CH2-9, m), 2.05 (3H, CH3m
(Pmc), s), 1.9–1.2 (16H, CH2-2,5,10,11,12, CH2-b,g
(Arg), CH2 (Pmc), m), 1.45–1.3 (27H, Ot-Bu, 2 Boc, 2
s), 1.15 (6H, 2CH3 (Pmc), s); MS m/z=1297.25 [M+
Na]+.
c=1,15 g l−1, 24 °C); H-NMR (CD3OD) l: 8.6 (2H,
1
H-e3, H-e4, m), 8.3 (1H, H-e9, m), 8.2 (1H, H-e10, s),
7.9 (2H, H-e8, H-e7, m), 7.6 (1H, H-e5, m), 7.4 (2H,
H-e1, H-e2, m), 6.6 (1H, H-e6, s), 4.3 (1H, Ha (Arg),
m), 4.15 (3H, N+(Me), s), 3.9 (2H, Ha (Gly), s),
3.55–3.1 (14H, CH2-8,13, 1,3,4,6, CH2-d (Arg), m),
2.8–2.4 (10H, 2CH3 (Pmc), CH2 (Pmc), CH2-7, m),
2.15 (2H, CH2-9, m), 2.05 (3H, CH3m (Pmc), s), 1.9–
1.3 (16H, CH2-2,5,10,11,12, CH2-b,g (Arg), CH2 (Pmc),
m), 1.45 (27H, Ot-Bu, 2 Boc, 2 s), 1.35 (6H, 2CH3
(Pmc), s); MS m/z=1377.6 [M]+
6.4.24. Compound 2
Compound 30 (69 mg, 50 mmol) was dissolved in 1
mL of a 90/10 TFA–H2O solution. After 30 min of
stirring, the solvent was evaporated in vacuo. The crude
product was purified by reverse-phase semi-preparative
HPLC, giving 40 mg of pure compound 2. HPLC [H2O
(1/1000 TFA)–CH3CN (1/1000 TFA): from 80/20 to
0/100% in 30 min, 288 nm]: Rt=8.2 min. [h]D= +10°
(H2O, c=1.03 g l−1, 24 °C); FAB–HRMS calc. for
C44H63N12O6 m/z: 855.4993, Found: m/z: 855.4982%
[M+].

584
N. Patino et al. / European Journal of Medicinal Chemistry 37 (2002) 573–584
[13] R. Breslow, S.D. Dong, Y. Webb, R. Xu, J. Am. Chem. Soc. 118
6.5. Pharmacology
(1996) 6588–6600.
[14] C.H. Tung, Z. Wei, M.J. Leibowitz, S. Stein, Proc. Natl. Acad.
Sci. USA 89 (1992) 7114–7118.
[15] V.V. Vlassov, G. zuber, B. Felden, J.P. Behr, R. Giege, Nucleic
Acids Res. 23 (1995) 3161–3167.
[16] A. Lorente, J.F. Espinosa, M. Fernandez-Saiz, J.M. Lehn, W.D.
Wilson, Y.Y. Zhong, Tetrahedron Lett. 37 (1996) 4417–4420.
[17] M. Endo, K. Hirata, T. Ihara, S. Sueda, M. Takagi, M.
Komiyama, J. Am. Chem. Soc. 118 (1996) 5478–5479.
[18] T. Fukuyama, C.-K. Jow, M. Cheung, Tetrahedron Lett. 36
(1995) 6373–6374.
Hydrolysis assays were realised as described by Endo
et al., using polyacrylamide gel electrophoresis [17].
Melting temperature studies were performed as de-
scribed for compound 1 [6]. In vitro anti-HIV-1 activity
and cytotoxicity of 2 were evaluated on infected CEM-
SS, MT4 and PBMC cells, using well-described proce-
dures [6,36,37].
[19] T. Fukuyama, M. Cheung, C.-K. Jow, Y. Hidai, T. Kan,
Tetrahedron Lett. 38 (1997) 5831–5834.
[20] For a review see: V. Kuksa, R. Buchan, P.K.T. Lin, Synthesis, 9
(2000) 1189–1207.
Acknowledgements
[21] R.N. Salvatore, S.E. Schmidt, E.I. Shin, A.S. Nagle, J.H. Wor-
rell, K.W. Jung, Tetrahedron Lett. 41 (2000) 9705–9708 and
references therein.
This work was supported by the ‘Agence Nationale
de Recherches sur le SIDA’ (ANRS).
[22] A.F. Abdel-Magid, K.G. Carson, B.D. Harris, C.A. Maryanoff,
R.D. Shah, J. Organomet. Chem. 61 (1996) 3849–3862 and
references therein.
[23] A.S. Kolodziejczyk, E. Sugajska, B. Falkiewicz, K. Wisniewski,
Synlett 10 (1999) 1606–1608.
References
[24] K. Wisniewski, A.S. Kolodziejczyk, Tetrahedron Lett. 38 (1997)
483–486.
[25] T. Fukuyama, M. Cheung, T. Kan, Synlett 8 (1999) 1301–1303.
[26] X. Lin, H. Dorr, J.M. Nuss, Tetrahedron Lett. 41 (2000) 3309–
3313.
[27] L. Yang, K. Chiu, Tetrahedron Lett. 38 (1997) 7307–7310.
[28] W.R. Bowman, D.R. Coghlan, Tetrahedron 53 (1997) 15787–
15798.
[29] D. Albanese, D. Landini, V. Lupi and M. Penso, Eur. J. Org.
Chem. (2000) 1443–1449.
[30] J.J. Turner, N. Wilschut, H.S. Overkleeft, W. Klaffke, G.A. Van
der Marel, J.H. Van Boom, Tetrahedron Lett. 40 (1999) 7039–
7042.
[31] J.F. Reichwein, R.M.J. Liskamp, Tetrahedron Lett. 39 (1998)
1243–1246.
[1] C.K. Mc Donald, D.R. Kuritzkes, Arch. Int. Med. 157 (1997)
951–959.
[2] R.F. Schinazi, Int. Antivir. 4 (1996) 95–107.
[3] S.M. Kingsman, A.J. Kingsman, Eur. J. Biochem. 240 (1996)
491–507.
[4] K.A. Jones, B.M. Peterlin, Annu. Rev. Biochem. 63 (1994)
717–743.
[5] S. Gunnery, S.R. Green, M.B. Mathews, Proc. Natl. Acad. Sci.
USA 89 (1992) 11557–11561.
[6] V. Peytou, R. Condom, N. Patino, R. Guedj, A.M. Aubertin, N.
Gelus, C. Bailly, R. Terreux, D. Cabrol-Bass, J. Med. Chem. 42
(1999) 4042–4053.
[7] For an Overview, see: A. Lapidot, A. Litovchick, Drug Dev.
Res. 50 (2000) 502–515 and references herein.
[8] K. Hamasaki, A. Ueno, Bioorg. Med. Chem. Lett. 11 (2001)
591–594.
[32] M.E. Wilson, J.S. Nowick, Tetrahedron Lett. 39 (1998) 6613–
6616.
[9] A. Farese-Di Giorgio, S. Pairot, N. Patino, R. Condom, C. Di
Giorgio, A. Aumelas, A.M. Aubertin, R. Guedj, Nucleosides
Nucleotides Nucleic Acids 18 (1999) 263–275.
[10] K. Shinozuka, Y. Nakashima, K. Shimizu, H. Sawai, Nu-
cleosides Nucleotides Nucleic Acids 20 (2001) 117–130.
[11] A. Kuzuya, Y. Azuma, T. Inokawa, K. Yoshinari, M.
Komiyama, Nucleic Acids Symp. Ser. 37 (1997) 209–210.
[12] K. Shinozuka, K. Shimizu, Y. Nakashima, H. Sawai, Bioorg.
Med. Chem. Lett. 4 (1994) 1979–1982.
[33] N.D.and Hone, L.J. Payne, Tetrahedron Lett. 41 (2000) 6149–
6152.
[34] B. Falkiewicz, A.S. Kolodziejczyk, B. Liberek, K. Wisniewski,
Tetrahedron 57 (2001) 7909–7917.
[35] S.H. Pine, Org. React. 18 (1970) 403–464.
[36] C. Moog, A. Wick, P. Le Ber, A. Kirn, A.M. Aubertin, Antivir.
Res. 24 (1994) 275–288.
[37] R. Pauwels, J. Balzarini, M. Baba, D. Snoeck, P. Herdewijn, J.
Desmyter, E. De Clercq, J. Virol. Methods 20 (1988) 309–321.