Bis- and tris-DNA intercalating porphyrins designed to target the major groove: Synthesis of acridylbis-arginyl-porphyrins, molecular modelling of their DNA complexes, and experimental tests

Article abstract of DOI:10.1002/ejoc.200300311

In order to increase the DNA binding affinity of a bis-arginylporphyrin which has been previously shown to bind preferentially in the major groove of the d(GGCGCC)₂ sequence (Mohammadi et al., Biochemistry 1998, 37, 9165), we have synthesized bis- and tris-intercalating derivatives in which one or both arginyl arms are connected through a flexible chain to an acridine ring. We report here the synthesis of these two molecules along with the molecular modelling of their complexes with a GC-rich oligonucleotide encompassing the central d(GGCGCC)₂ hexamer. The modelling computations showed that when the porphyrin was intercalated into the central d(CpG)₂ site with both arginyl side-chains bonded to the guanines flanking the intercalation site, the acridine ring(s) could intercalate immediately upstream from the central hexamer, but at the cost of substantial DNA conformational energy. A significant preference for major-groove binding over minor-groove binding was found. The results of circular dichroism studies and topoisomerase I-unwinding experiments supported the bis- and tris-intercalation of these derivatives. The bis-acridyl derivative provided, as expected, greater stabilization against thermal denaturation than the mono-acridyl and the parent bis-arginyl-porphyrin compounds. Based on the modelling results, the structures of derivatives can be tailored to facilitate tris-intercalation in rigid GC-rich sequences, and thereby enhance the selective targeting of GC base pairs by the arginyl side-chains, by lengthening the porphyrin-acridine connectors. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejoc.200300311

FULL PAPER
Bis- and Tris-DNA Intercalating Porphyrins Designed to Target the Major
Groove: Synthesis of Acridylbis-arginyl-porphyrins, Molecular Modelling of
Their DNA Complexes, and Experimental Tests
[
a]
[a]
[a]
[b]
Samia Far, Alain Kossanyi, Catherine Verch e` re-B e´ aur, Nohad Gresh,
[
c]
[a]
Eliane Taillandier, and Martine Perr e´ e-Fauvet*
Keywords: Cationic porphyrins / Molecular modelling / DNA intercalation / DNA recognition / Topoisomerase I
In order to increase the DNA binding affinity of a bis-arginyl-
conformational energy. A significant preference for major-
porphyrin which has been previously shown to bind prefer- groove binding over minor-groove binding was found. The
entially in the major groove of the d(GGCGCC)
2
sequence
results of circular dichroism studies and topoisomerase I-un-
winding experiments supported the bis- and tris-intercala-
tion of these derivatives. The bis-acridyl derivative provided,
as expected, greater stabilization against thermal denatura-
tion than the mono-acridyl and the parent bis-arginyl-por-
phyrin compounds. Based on the modelling results, the struc-
(Mohammadi et al., Biochemistry 1998, 37, 9165), we have
synthesized bis- and tris-intercalating derivatives in which
one or both arginyl arms are connected through a flexible
chain to an acridine ring. We report here the synthesis of
these two molecules along with the molecular modelling of
their complexes with a GC-rich oligonucleotide encom- tures of derivatives can be tailored to facilitate tris-intercala-
passing the central d(GGCGCC) hexamer. The modelling tion in rigid GC-rich sequences, and thereby enhance the
computations showed that when the porphyrin was intercal- selective targeting of GC base pairs by the arginyl side-
2
ated into the central d(CpG)
2
site with both arginyl side-
chains, by lengthening the porphyrin-acridine connectors.
chains bonded to the guanines flanking the intercalation site,
the acridine ring(s) could intercalate immediately upstream
from the central hexamer, but at the cost of substantial DNA
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
Introduction
iron[ as well as of rhodium or ruthenium metallointercala-
8]
[
9]
[10Ϫ14]
tors, but by only very few smaller-sized ligands.
Most DNA sequence-specific drugs target the minor
groove, as is the case for nonintercalating groove binders
[
1]
On the basis of theoretical computations, we previously
[2]
and the majority of intercalators and oligopeptide-inter- synthesized a bis-arginyl derivative of a tricationic porphy-
calator conjugates.[ Major-groove recognition can be rin (Figure 1, BAP) which was designed to bind specifically
3]
[
15]
achieved by extended oligomeric motifs such as synthetic to the palindromic sequence d(GGCGCC)2.
The DNA
[4]
antigene oligonucleotides and their intercalator conju- binding of the parent compound, meso-tetrakis(N-methyl-
[
5]
[6]
gates, peptide nucleic acids, oligopeptides stitched to- 4-pyridiniumyl)porphyrin or H TMPyP-4, has been widely
2
[7]
[16]
gether by disulfide bonds, and peptide complexes of studied,
and its ability to intercalate into CpG or GpC
steps or to bind in a groove at A-T rich sequences has been
[
17]
shown. This porphyrin also has DNA-photosensitizing
and antitumour properties.
[
18]
BAP modelling studies have
[
a]
Laboratoire de Chimie Bioorganique et Bioinorganique
Institut de Chimie Mol e´ culaire et des Mat e´ riaux d’Orsay,
CNRS UMR 8124, B aˆ t. 420 Universit e´ Paris-Sud,
shown that the two Arg arms of the bis-arginyl derivative
bind preferentially to the major groove of DNA, the best-
91405 Orsay Cedex, France
bound sequence being the palindromic sequence
Fax: (internat.) ϩ33-169157281
[19]
d(GGCGCC)2.
In the energy-minimized structure, the
E-mail: mperreef@icmo.u-psud.fr
[
[
b]
c]
porphyrin ring was intercalated into the central d(CpG)2
Laboratoire de Pharmacochimie Mol e´ culaire et Cellulaire,
CNRS FRE 2718, UFR Biom e´ dicale, Universit e´ Ren e´ - sequence so that each Arg arm was bound to a distinct
Descartes (Paris V)
DNA strand and the guanidinium groups were hydrogen-
4
5, rue des Saints-P e` res, 75006 Paris, France
6
7
Laboratoire de Spectroscopie Biomol e´ culaire, UFR de Sant e´ , bonded to two O /N atoms of the two successive guanines
M e´ decine et Biologie Humaine, Universit e´ Paris-Nord
upstream from the intercalation site. Simultaneous interac-
tions of an Arg side-chain with two successive G bases on
the same strand have been observed in an X-ray diffraction
7
4 rue Marcel Cachin, 93017 Bobigny, France
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2004, 1781Ϫ1797
DOI: 10.1002/ejoc.200300311
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1781

M. Perr e´ e-Fauvet et al.
FULL PAPER
Figure 1. Molecular structure of BAP, MABAP and BABAP compounds
study of the complex of the DNA-binding domain of yeast
Results and Discussion
RAP1 with telomeric DNA.[
20]
Spectroscopic IR studies carried out on BAP-oligonucle-
otide complexes have confirmed that the Arg arms are
located in the major groove of G-C rich complexes and
thermal denaturation studies showed that the best relative
Molecular Modelling of DNA Complexes of Acridylbis-
arginyl-porphyrins
Oligonucleotide Sequence
stabilization occurs on d(GGCGCC) -encompassing se-
2
[
21]
quences.
The GGCGCC sequence occurs frequently in onco-
genes and is also found in the Primer Binding Site (PBS)
of the Long Terminal Repeat (LTR) of the HIV retro-
We have modelled the interaction of acridyl derivatives of
bis-arginyl-porphyrin (BAP) with the palindromic sequence
[
22]
d(CTGTTC GGGC GCCC GAACAG) (Figure 2). The in-
2
6
7
14Ј
15Ј
14
tercalation sites are d(C -G )·d(C -G ) and d(C -
G )·d(C -G ) for the acridine rings and d(C -
G )·d(C -G ) for the porphyrin ring. We present below
the most favourable binding modes of two selected BAP
derivatives, denoted as mono-acridyl-bis-arginyl-porphyrin
[
23]
virus. Thus, ligands that recognize this sequence with en-
hanced affinity and selectivity could have promising antitu-
mour or antiviral properties. Towards this aim, we have ex-
tended the structure of BAP by linking to one, or both,
Arg arms a second, or third, intercalator, of acridine type,
through an additional Gly residue and a polymethylenic
chain. Considerable enhancements of DNA-binding affin-
ities have previously been demonstrated upon passing from
15
6Ј
7Ј
10
11
10Ј
11Ј
(
MABAP) and bis-acridyl-bis-arginyl-porphyrin (BA-
[26]
BAP).
Their structures are given in Figure 1. In these
derivatives, the acridine ring is bound to the arginine
through a glycine and a hexamethylene diamine linker.
[
24]
mono- to bis- and tris-intercalating derivatives.
The
Mono-acridyl-bis-arginyl-porphyrin (MABAP)
syntheses of porphyrin-mono- and bis-acridine conjugates
have recently been reported[ and these molecules are en-
25]
Three binding modes, each with the phenyl group (bear-
[
25a,25c,25d]
dowed with significant nuclease
ic
and cytotox- ing the arginyl arms) located in the major groove, were com-
[25b]
activities. Although acridines have low sequence selec- pared: they have three, two, and one N-methylpyridinium
tivities, except possibly in favour of intercalation at pyr-pur rings in the minor groove and were denoted as A, C, and
steps over pur-pyr steps, the purpose of their inclusion is B, respectively, in our preceding papers on the mono-inter-
[
19,21]
essentially that of reinforcing the DNA-binding affinities of calating parent compound BAP.
porphyrin-peptide conjugates. modes, the acridyl ring can intercalate at the
For each of these
6
7
14Ј
15Ј
Here we report the molecular modelling of the complexes d(C pG )·d(C pG ) site from either the major-groove
of two acridylbis-arginyl-porphyrins with a double-stranded side (denoted as ЈIЈ, internal mode), or from the minor-
sequence d(CTGTTCGGGCGCCCGAACAG)2 encom- groove side (denoted as ЈEЈ, external mode). In the latter
[
23]
passing an undecanucleotide fragment of the HIV-1 PBS.
case, the aliphatic acridine connector winds around the su-
This constitutes a prerequisite to their chemical synthesis, gar-phosphate backbone.
which is subsequently reported together with DNA-interac-
tion experiments.
The two most favourable modes of binding to the PBS-
encompassing sequence are modes A-I and C-I (Table 1).
1782
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 1781Ϫ1797

DNA Intercalating Porphyrins Designed to Target the Major Groove
FULL PAPER
Figure 2. Base numbering of the double-stranded d(CTGTTC GGGC GCCC GAACAG)
2
Table 1. Binding energies (kcal·molϪ1) of the intercalation of MABAP in the sequence d(CTGTTC GGGC GCCC GAACAG)
arginine side-chains interact in the major groove
; the
2
Intercalation mode of the porphyrin
Intercalation mode of the acridine
A
I
B
C
I
E
I
E
E
∆
∆
inter
Ϫ325.0
142.8
Ϫ330.8
145.8
Ϫ320.2
134.2
Ϫ350.8
144.4
Ϫ314.0
130.2
E
DNA
E
lig
35.4
60.1
71.3
57.5
80.0
E ϭ Einter ϩ ∆EDNA ϩ∆Elig
Ϫ146.8
Ϫ124.9
Ϫ114.7
Ϫ148.9
Ϫ103.8
They are represented in Figure 3 (a) and (b) respectively. In acridine intercalates from the minor-groove side and the
mode A-I, the Arg side-chain, which precedes the acridine, other from the major-groove side, were also considered. The
6
7
7
8
9
interacts with O /N of guanines G , G and G upstream most favourable complexes of BABAP with the PBS-en-
from the intercalation site. The side-chain of the other argi- compassing sequence in modes AϪC are shown in Figure 4
8
Ј
6
9Ј
nine interacts with N of G and O of G upstream from (aϪc). The stabilities of the three complexes are ranked in
7
6
7
11
the intercalation site, and with O /N of G of the interca- the order A-II Ͼ B-EE Ͼ C-II (Table 2).
lation site. In binding mode C-I, the acridine-connected ar-
The most favourable A binding mode is II, in which both
ginine binds to O of G and G and to N of G . The other acridines intercalate from the major groove side. The side-
6
7
8
7
9
7
Ј
8Ј
9Ј
6
arginine binds to G , G and to G as well as to O of chain of one Arg interacts with the three successive guan-
1
1
7
8
9
G
of the intercalation site.
ines G , G , G upstream from the intercalation site. The
7Ј
8Ј
side-chain of the other Arg similarly spans G , G , and
on the other strand.
The most favourable B binding mode is EE, favoured
9Ј
G
Bis-acridyl-bis-arginyl-porphyrin (BABAP)
As in the case of MABAP, the three modes A, B, and C over II because the porphyrin ring, which protrudes more
were considered. Mixed modes of the type IE, in which one in the major groove, causes the peptide backbone to be
Figure 3. Molecular modelling of two intercalated complexes of MABAP in the sequence d(CTGTTC GGGC GCCC GAACAG)
arginine side-chains interact in the major groove: (a) mode A-I; (b) mode C-I (hydrogens are omitted for clarity)
2
; the
Eur. J. Org. Chem. 2004, 1781Ϫ1797
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1783

M. Perr e´ e-Fauvet et al.
FULL PAPER
Figure 4. Molecular modelling of three intercalated complexes of BABAP in the sequence d(CTGTTC GGGC GCCC GAACAG)
2
; the
arginine side-chains interact in the major groove: (a) mode A-II; (b) mode B-EE; (c) mode C-II (hydrogens are omitted for clarity)
8Ј
shifted more towards the periphery thus facilitating winding primed strand, and similarly the other Arg binds to G and
9
Ј
of the aliphatic acridine connector around the DNA back-
G
on the primed strand. They also interact respectively
8
9
11Ј
11
bone. One Arg side-chain binds to G and G on the un- with G and G of the intercalation site.
1784
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org Eur. J. Org. Chem. 2004, 1781Ϫ1797

DNA Intercalating Porphyrins Designed to Target the Major Groove
FULL PAPER
Table 2. Binding Energies (kcal·molϪ1) of the intercalation of BABAP in the sequence d(CTGTTC GGGC GCCC GAACAG)
arginine side-chains interact in the major groove
. The
2
Intercalation mode of the porphyrin
Intercalation mode of the acridine
A
II
B
EE
C
II
IE
II
IE
E
∆
∆
inter
Ϫ432.5
148.0
136.0
Ϫ332.5
138.9
103.0
Ϫ399.7
128.6
137.1
Ϫ410.5
154.0
129.3
Ϫ375.1
157.9
Ϫ405.2
158.9
132.0
E
DNA
E
lig
89.9
E ϭ Einter ϩ ∆EDNA ϩ∆Elig
Ϫ148.5
Ϫ90.6
Ϫ134.0
Ϫ127.2
Ϫ127.3
Ϫ114.3
Similar to mode A, the most favourable C binding mode imental data suggest that BABAP has a stronger affinity
7
is II. On the unprimed strand, one Arg interacts with G , than MABAP for DNA. Further refinements of the energy
8
9
G and G and on the primed strand, the other Arg inter- balances could include an explicit, rather than implicit, in-
7
Ј
8Ј
9Ј
acts with G , G and G .
clusion of solvation effects using a Continuum reaction
For both MABAP and BABAP, the amount of DNA field procedure in the JUMNA procedure (Zakrzeswka et
conformational distortion energy, ∆EDNA, in the range al., work in progress).
Ϫ1
1
29Ϫ159 kcal·mol , is considerable. For comparison, it is
Finally, it was important to evaluate to what extent the
recalled that in the complex of a DNA double-stranded do- two predominant structural features shown by energy-mini-
decamer with the mono-intercalating porphyrin-netropsin mization of the BABAP-DNA complex would be main-
[
27]
derivative,
∆
also studied by the JUMNA procedure, tained after molecular dynamics in a water bath. These two
EDNA had values in the range 45Ϫ70 kcal·molϪ1, yet the features are tris-intercalation and the onset of H-bonds be-
DNA was shorter than in the present study and had under- tween the Arg arms and the guanine bases. We wished to
gone less severe distortions. Such large values may reflect avoid the uncertainties due to the number and distribution
the sensitivity of the JUMNA potential to DNA distor- of the counterions and prevent excessive distortions of the
tions. Note that in the present case it is the Einter term that DNA backbone which could occur if environmental effects
determines the most stable binding complexes for MABAP are not fully accounted for. Thus, for the present purpose,
(C-I) and BABAP (A-II).
and because of the qualitative nature of the present evalu-
Taking into account the fact that a more attractive elec- ation, we have used screened phosphate charges of Ϫ0.5 (as
trostatic potential is exerted on positive charges in the in JUMNA), and the hydrogen bonds of the base pairs (bp)
minor groove by A-T-rich sequences than by G-C-rich were enforced with harmonic restraints. This evaluation was
[
28]
ones, we have also computed the energy balances for the limited to the best binding complex A-II. Two separate MD
intercalated complexes of BABAP in the sequence simulations were performed in a bath of 3744 water mol-
d(CTGTAT ATAC GTAT ATACAG) . The central hexa- ecules with periodic boundary conditions. The first was
2
meric sequence in this case is d(TACGTA)2 instead of with rigid DNA, and the second with one DNA strand re-
[
19]
d(GGCGCC) , and the phenyl group (bearing the arginyl laxed (as in our former MD study on BAP ). The lowest
2
arms) is therefore located in the minor groove of the oligo- frames from these simulations of 50 ps duration are given in
nucleotide. The energy balance in the best minor-groove the Supporting Information (see S1-S2). Tris-intercalation
binding mode (A-II) was found to be less favourable than is preserved in both cases. While for the first dynamics
that in the best major groove binding mode of BABAP. simulation, the Arg side-chain on the primed strand main-
Such a difference is somewhat larger than that previously tains its H-bonds with all three G bases, G7Ј-G9Ј, the Arg
reported for the parent mono-intercalating compound on the unprimed strand retains its H-bonds with only G8
[
19]
BAP.
Notwithstanding the quantitative limitations in- and G9. In the second simulation, the Arg side-chain on
herent to the simplified approach followed in this study, the primed strand is held in place by ionic H-bonds with
with only implicit modelling of environmental effects, this the phosphate groups on the 3Ј- and 5Ј-sites of guanine G8Ј
could imply that enhancement of the major- versus minor- and by hydrogen bonds with guanines G7Ј and G9Ј. It
groove preference would result on passing from mono- to binds to G8Ј through water molecules. However the Arg on
tris-intercalation, provided the targeting peptide side-chains the unprimed strand loses its H-bonds with the G bases,
are properly oriented.
and interacts with the water. This indicates that the receptor
The most favourable major groove BABAP complex, A- competes with the solvent in its interactions with a ligand’s
[
29]
II, has a significantly lower Einter value than those found functional group.
However, longer simulation times and
for the most stable MABAP complexes A-I and C-I. This is averaging are needed to better quantify such competition.
compensated, however, by much larger ligand conformation It could also be very important in the future to evaluate how
energy rearrangements, ∆E , for BABAP than for MA- such competition is reflected by other molecular mechanics
lig
BAP. As a result, the energy balances, which consist of potentials that explicitly include polarization and coopera-
small differences between large and opposing contributions, tive effects and the energy balance for direct versus through-
are virtually identical for both compounds, while the exper- water binding of a cationic entity to an electron-rich ligand.[
30]
Eur. J. Org. Chem. 2004, 1781Ϫ1797
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1785

M. Perr e´ e-Fauvet et al.
FULL PAPER
Scheme 1. Retrosynthetic scheme for MABAP synthesis (PG ϭ Protective Group)
Synthesis of the Two Acridylbis-arginyl-porphyrins MABAP 9-acridinecarboxylic acid, hexamethylene diamine, glycine
and BABAP
and arginine. This requires the preliminary protection of the
functional groups carried by reagents that are likely to react
during the coupling. The acridine derivative 5, which was
obtained in an overall yield of 60%, was coupled to an argi-
nine protected both on its amine function, to avoid second-
ary coupling reactions, and on its guanidinium group. The
protection of the latter yields arginine derivatives that are
easier to purify than their salt form and this protection was
retained until the ultimate stage of the synthesis. We chose
to protect the guanidinium with the 2,2,4,6,7-pentamethyldi-
hydrobenzofuran-5-sulfonyl (Pbf) protective group, as it is
The strategy used for synthesizing MABAP is given in
the retrosynthetic Scheme 1. It implies the preliminary syn-
thesis of the porphyrinic synthon A and the acridyl synthon
C. BABAP synthesis results from the coupling of synthons
A and C, and BAP was earlier obtained by the coupling of
synthons A and B.[
15]
The synthesis of the porphyrin diacid A has been de-
scribed previously.
[
15]
The amino function of the arginine
(synthons B and C) will be used as the anchoring point on
the porphyrin A. Synthon B is a commercial -arginine
methyl ester and synthon C consists of a guanidinium-pro-
tected arginine connected to an acridine moiety through a
glycine and a hexamethylenic chain. The last steps of the
synthesis involve the removal of the protective group of
guanidinium and methylation of the pyridyl rings of the
porphyrin.
[31]
easily cleaved by trifluoroacetic acid,
while the arginine
amine function was protected by the 9-fluorenylmethoxycar-
bonyl (Fmoc) group, one of the very rare amine protective
[32]
groups to be cleaved by bases. The eventual deprotection
of the Fmoc group by the amine function of 5 (and to a
lesser extent by the base used for the coupling) was reduced
if the DMF solution of 5 was diluted and added slowly to
an ice-cooled solution of the activated ester of (Fmoc)-
Synthesis of Synthon C
[33]
Arg(Pbf).
Cleavage of the Fmoc group by 1,8-diazabicy-
The synthon C was prepared according to Scheme 2 clo[5.4.0]undec-7-ene (DBU) gave the synthon C which was
using a series of peptidic couplings involving four reagents: obtained with an overall yield of 30%.
1786
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 1781Ϫ1797

DNA Intercalating Porphyrins Designed to Target the Major Groove
FULL PAPER
Scheme 2. Synthesis of synthon C: (a) dioxane, room temp., 72 h; (b) BOP (1.1 equiv.), DIEA (1.2 equiv.), DMF, room temp., Ar, 24 h;
(
c) CH
2 2
Cl /TFA (v/v), room temp. 1 h; (d) PyBOP (1.2 equiv.), DIEA (5 equiv./3 ϩ1.2 equiv./Gly), DMF, room temp., Ar, 48 h; (e) BOP
(
1.2 equiv.), DIEA (5 equiv./5 ϩ1.2 equiv./Arg), DMF, room temp., Ar, 24 h; (f) DBU, THF, room temp., 15 min
Synthesis of MABAP
Synthesis of BABAP
MABAP was synthesised by the coupling of the porphy-
BABAP was obtained by the coupling of the porphyrin
rin diacid (synthon A) with synthons B and C, using diacid with an excess of synthon C, in the presence of the
benzotriazol-1-yloxytris(dimethylamino)phosphonium BOP reagent, according to the procedure described in
hexafluorophosphate (BOP reagent), according to the pro- Scheme 4. The porphyrin precursor 10 was obtained in 48%
cedure shown in Scheme 3. This procedure yielded mainly yield. Cleavage of the Pbf protective groups followed by
the symmetrical porphyrins resulting from the coupling of methylation yielded the bis-acridyl-bis-arginyl-porphyrin
A with two B (17%) or A with two C (40%). The unsym- BABAP.
metrical porphyrin 8 (12%) was separated by chromatogra-
phy on a reverse phase silica column using CH COOH/ NMR spectroscopy and mass spectrometry (see Expt.
MABAP and BABAP were fully characterized by 1
H
3
[
34]
CH OH/H O, 1:1:2, as eluent.
Cleavage of the Pbf pro- Sect.). These techniques indicated that the acridine rings
3
2
tective group followed by methylation of the pyridyl were neither methylated by methyl iodide, nor protonated
[
15]
rings
yielded the water-soluble porphyrin MABAP.
www.eurjoc.org
at physiological pH.
Eur. J. Org. Chem. 2004, 1781Ϫ1797
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1787

M. Perr e´ e-Fauvet et al.
FULL PAPER
Scheme 3. Synthesis of MABAP
Binding of Bis-arginyl-porphyrin Derivatives to DNA
that we have modelled in the presence of the BAP deriva-
tives. The normalized melting curves are given in Figure 5
Tm Measurements
(a) and the data are reported in Table 3. At a NaCl concen-
Thermal denaturation studies were performed on the do- tration of 100 m, the thermal denaturation of this G-C
decanucleotide d(TC GGGC GCCC GA) alone and in the rich dodecanucleotide alone occurs at a very high tempera-
2
presence of BAP, MABAP and BABAP. d(TC GGGC ture (68 °C). As a result, and similar to our previous find-
[
21]
GCCC GA) is the central sequence of the oligonucleotide ings,
the amount of BAP-induced stabilization is very
2
1788
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 1781Ϫ1797

DNA Intercalating Porphyrins Designed to Target the Major Groove
FULL PAPER
Scheme 4. Synthesis of BABAP
small (∆Tm ϭ 1 °C) [with the shorter decanucleotide sorbance of single-strand DNA, even at 90 °C, in contrast
d(GTGGC GCCAC) that has a Tm of 56 °C, ∆Tm to the absorbance of the BAP- and MABAP-bound oligon-
°C ]. MABAP produces no more stabilization than ucleotides. To lower the Tm value, we repeated the thermal
ϭ
2
[
21]
4
BAP. However, in the case of BABAP, we did not observe denaturation studies in 10 m NaCl, which resulted in a Tm
the beginning of a plateau, which corresponds to the ab- value of 50 °C for the oligonucleotide alone. Under these
Eur. J. Org. Chem. 2004, 1781Ϫ1797
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1789

M. Perr e´ e-Fauvet et al.
FULL PAPER
Table 3. Melting temperature of d(TC GGGC GCCC GA)
2
and d(TC GTAC GTAC GA)
2
in the presence of BAP, MABAP and BABAP
[Oligonucleotide(bp)]/[Ligand] ϭ 12
d(TCGGGCGCCCGA)
2
d(TCGGGCGCCCGA)
10 m NaCl
2
d(TCGTACGTACGA)
2
100 m NaCl
100 m NaCl
T
m
(°C)
T
m
(°C)
m
T (°C)
No ligand
BAP
MABAP
BABAP
68
69
69
50
66
65
48
50
52
60
n.d. probably Ͼ 78
56, 80
conditions the BAP and MABAP-bound oligonucleotides
had Tm values of 66 and 65 °C, respectively, an increase of
16 and 15 °C. At this point we have no explanation for
the virtually equal ∆Tm values of the BAP and MABAP
complexes of this dodecanucleotide. The curve of the
BABAP-bound oligonucleotide had a more complex behav-
iour, with two inflexion points at 56 and 80 °C. The latter
corresponds to a ∆T of 30 °C (Figure not shown). Because
m
of such complex behaviour, we repeated these experiments
with another dodecanucleotide, d(TCGTACGTACGA)2,
which has fewer hydrogen bonds, and thus a lower melting
temperature, namely 48 °C [Figure 5 (b)]. In this case, its
BAP-, MABAP-, and BABAP-bound complexes had Tm
values of 50, 52, and 60 °C, respectively. We consider this
finding to be a clear indication of the increased DNA stabil-
ization upon passing from mono- to bis- and tris-interca-
lation. Nevertheless, a stabilization of 12 °C for this oligo-
nucleotide is low relative to the stabilization observed in the
BABAP-d(TC GGGC GCCC GA) complex. This suggests
2
that BABAP stabilizes GGC GCC more than TAC GTA
sequences. Finally we note that the 2 °C BAP-induced ther-
mal stabilization is much smaller than the 4 °C one found
for the sequence d(GTGGC GCCAC) yet the latter has a
2
higher Tm of 56 °C.[ This is an additional indication for
21]
the intrinsic preference of BAP for sequences having a
d(GGC GCC) central core.
2
In order to complement these T measurements, we have
m
also undertaken a combination of circular dichroism (CD)
and topoisomerisation experiments to demonstrate the in-
tercalation of the acridine rings of MABAP and BABAP
in DNA.
Figure 5. UV melting curves for (a) d(TC GGGC GCCC GA)
2
and (b) d(TC GTAC GTAC GA) alone (-) and in the presence of
2
BAP (Ϫ Ϫ), MABAP (---) and BABAP (···); [Oligonucleotide](bp)/
[Ligand] ϭ 12
Circular Dichroism
Intercalation of porphyrins is characterized by the ap-
pearance of a negative induced circular dichroism band in nau et al., the appearance of a weak and broad positive
the Soret region (420Ϫ480 nm), the appearance of a posi- CD signal in the 350Ϫ400 nm region is indicative of the
[
35]
tive induced band being indicative of a non-intercalated intercalation of the acridine ring.
[
16a]
binding mode.
The acridine moiety in MABAP and BA-
Our experiments were carried out on CT-DNA, synthetic
BAP has a strong absorption band at 250 nm and a weak polynucleotides and oligonucleotides with variable GC con-
absorption band at 360 nm, and its binding to DNA will tents. We present here the interactions of BAP, MABAP
induce CD signals in both regions. However, as spectral and BABAP with poly(dG-dC) and the central fragment
2
changes in short-wavelength region could result either from d(TC GGGC GCCC GA) of the oligonucleotide that we
2
a conformational change in the DNA upon binding of the have modelled.
ligand or from induced circular dichroism of the bound
As seen in Figure 6 (a and b), the three porphyrins dis-
acridine, we have investigated only the longer wavelength play a broad negative signal in the Soret region, the signa-
region. As demonstrated by Wirth et al. and Gimenez-Ar- ture of the intercalation of the porphyrin ring in the polynu-
1790
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 1781Ϫ1797

DNA Intercalating Porphyrins Designed to Target the Major Groove
FULL PAPER
linking number.[36] After extraction, DNA gel electropho-
resis shows topoisomers having mobilities intermediate be-
tween the relaxed (I ) DNA (or the nicked form II also
R
present in the native DNA) and the native supercoiled (IS)
DNA. This method allows to demonstrate the intercalation
of ligands in DNA and the efficiency of the intercalation.
Figure 7 shows the results of the unwinding of a pTZ
plasmid DNA in the presence of BAP, MABAP and BA-
BAP. In these experiments, DNA was preincubated with to-
poisomerase I and increasing amounts of each of the three
porphyrin compounds were added. At low porphyrin con-
centrations (lane 11), DNA had the same migration as re-
laxed DNA (I ) (lane 3). Increasing amounts of added li-
R
gand provided topoisomers with increasing numbers of
negative supercoils, which have increasing mobilities inter-
mediate between those of relaxed and supercoiled DNA. A
ligand-dependent concentration could be characterized for
which most topoisomers migrate like supercoiled DNA,
which indicates ligand-induced modification of the topo-
logy of topoisomerase I relaxed DNA. This occurred in lane
7
for BAP, lane 8 for MABAP and lane 9 for BABAP. The
corresponding ligand concentrations provide information
on the extent of ligand intercalation. These corresponded
to [DNA](bp)/[Ligand] ratios (1/r ) of 8 for BAP, 12 for
0
MABAP and 24 for BABAP. The concentrations of
MABAP and BABAP were thus 1.5- and three-times lower
than the corresponding BAP concentration. These results
are consistent with the involvement of the acridine rings in
DNA binding, which, according to the modelling compu-
tations, should occur by intercalation.
Figure 6. Induced circular dichroism (ICD) of BAP (-), MABAP
(
---) and BABAP (···) in the presence of (a) poly(dG-dC)
2
,
,
[
[
DNA](bp)/[Ligand] 30; (b) d(TC GGGC GCCC GA)
ϭ
2
Conclusion and Perspectives
DNA](bp)/[Ligand] ϭ 12; (c) ICD of BABAP in the presence of
d(TC GTAC GTAC GA)
2
, [DNA](bp)/[BABAP] ϭ 12
We have resorted to molecular modelling to design
mono- and bis-acridyl derivatives of the bis-arginyl porphy-
rin mono-intercalator BAP, denoted as MABAP and BA-
BAP respectively. It was necessary to find an appropriate
connector between the Arg backbone and the added acri-
dine ring(s) that would enable the ring(s) to intercalate be-
tween DNA base pairs. This was a prerequisite to the
chemical synthesis reported in this paper. In the absence of
available experimental (X-ray or NMR) structural data, the
computations reported here suggest the existence of several
possibilities for complex formation between MABAP and
cleotide and the dodecamer. The weak and broad positive
induced band around 380 nm in the MABAP and BABAP
spectra contrasts with the corresponding monotonous de-
crease of the band in the 360Ϫ380 nm region in the case of
the BAP complex. This could indicate that the acridines are
intercalated. This band is more intense in the induced CD
spectra of BABAP than in those of MABAP. This is in
agreement with the bis-intercalation of the acridines in the
BABAP complexes. These experiments were also performed
BABAP and d(CTGTTC GGGC GCCC GAACAG) . The
2
on the dodecanucleotide d(TC GTAC GTAC GA) investi-
2
most stable complexes have both Arg side-chains in the
gated in the thermal denaturation studies. In the BABAP-
6
7
major groove, bound to O /N of the two successive G
bases on a given strand upstream from the intercalation site.
Those complexes having three, one, or two N-methylpyridi-
nium rings in the minor groove are denoted by ЈAЈ, ЈBЈ,
d(TC GTAC GTAC GA) complex [curve (c) represented
2
in Figure 6 (b)], the induced acridine signal is more intense.
This could reflect a facilitated intercalation due to the
greater flexibility of this oligonucleotide.
and ЈCЈ, respectively. For a given GlyϪC ϪAcr extension,
6
those complexes in which the GlyϪC chain remains in the
major groove or winds across the sugar-phosphate chain to
6
Topoisomerase I-Uunwinding Experiments
Topoisomerase I catalyzes the relaxation of supercoiled reach the minor groove are denoted by ЈIЈ (internal) or ЈEЈ
plasmid DNA (form I ) into the I form. In the presence (external). The acridine ring intercalates from the major or
S
R
of intercalators, the topology of DNA is modified and the the minor groove in these complexes respectively. As de-
topoisomerisation provides relaxed DNA with a smaller duced fom the present calculations, the most stable complex
Eur. J. Org. Chem. 2004, 1781Ϫ1797
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1791

M. Perr e´ e-Fauvet et al.
FULL PAPER
Figure 7. Topoisomerase I-unwinding of supercoiled DNA, incubated in the presence of increasing amounts of BAP, MABAP and
BABAP. N: native DNA, E: extracted DNA. 1/r ϭ [DNA](bp)/[Ligand]
0
is C-I, followed by complexes A-I and B-I for MABAP, and Ishikawa et al.[25d] that in cationic porphyrin-9-aminoacrid-
A-II, followed by complexes B-EE, C-II and B-IE for BA- ine hybrids, intercalation of the acridine actually occurred
BAP. The preference of the B complex for the EE mode can at the expense of that of the porphyrin ring, although this
be understood by the fact that sliding the porphyrin ring could have been due to the rather short length of the con-
along the diad axis allows the two arms to move more freely nector arm. Synthesis of derivatives in which the acridines
and reach the sugar-phosphate backbone than in mode A are connected to the linker by an NH, rather than a CϭO
which restricts their mobility in the major groove.
group is underway in our laboratory (Ramiandrasoa et al.,
Circular dichroism and topoisomerase I-unwinding to be published). This should afford acridine protonation
experiments support the bis- and tris-intercalation of MA- at neutral pH, thereby enhancing DNA binding, provided
BAP and BABAP in DNA. Furthermore, thermal denatur- that porphyrin intercalation is not prevented. Derivatives
ation studies have shown the melting temperatures of with longer porphyrin-acridine connectors will be designed
double-stranded oligonucleotides to be significantly higher to facilitate tris-intercalation without excessive DNA distor-
in the presence of BABAP than in the presence of the tion that could penalize the more rigid GC-rich sequences.
mono-intercalator BAP.
The design and synthesis of BABAP thus constitutes a
The present calculations imply that substantial of A ver- starting point in the design of tris-intercalating compounds
sus B as well as I versus E preferences could be achieved by that have three anchoring points, each of which has intrinsic
modifying the nature as well as the length of the connector. antitumour properties, and whose connecting arms should
This could be important from the perspective of replacing be responsible for both sequence-specific recognition and
the acridine ring (which is mutagenic) by derivatives such modulatable II versus EE binding modes.
as m-amsacrine (which is antitumoral). Thus compounds in
which the -NHSO CH substituent protrudes into the
2
3
Experimental Section
minor groove or the major groove could be tailored by
modifying the nature of the Arg arms and their connectors
to the added intercalators. In the present work, we have
used 9-carboxyacridine, rather than a cationic 9-aminoacri-
dine to provide additional intercalating groups because we
General Methods: Coupling reactions were performed under an ar-
gon atmosphere and in a commercial acid- and base-free DMF
(
99.8% Aldrich). Most of the reactions were carried out in the dark.
THF and dioxane were dried over sodium and distilled prior to
wanted the porphyrin to intercalate in the d(CpG) site use. All commercially available chemical reagents were used with-
2
prior to intercalation of the acridine. It has been shown by out purification. Reactions were monitored by thin layer chroma-
1792
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 1781Ϫ1797

DNA Intercalating Porphyrins Designed to Target the Major Groove
tography (TLC) performed on silica gel sheets containing UV flu-
FULL PAPER
a molar extinction coefficient of 1.31 ϫ
260 nm using
1
13
10 Ϫ1·cm
.
4
Ϫ1 [40]
orescent indicator (60 F254 Merck). H and C NMR spectra were
recorded on Bruker AC 200, Bruker AC 250 and Bruker AC 400
Oligonucleotide Thermal Denaturation Measurements: Optical ther-
mal denaturation measurements were performed in quartz cuvettes
1 mL, 10 mm path length) on a Kontron Uvikon 933 spectropho-
tometer. The temperature of the cell holder was varied from 10 °C
1
spectrometers (200, 250 and 400 MHz for H, respectively, and 50,
1
3
6
3 and 100 MHz for C, respectively). Chemical shifts, δ, are re-
(
ported in ppm taking residual CHCl or CHD OD as the reference.
3
2
Infrared spectra were recorded on a Bruker IFS 66 Fourier Trans-
form spectrophotometer and UV/Visible spectra on a Safas 190
DES spectrophotometer. Mass spectra were recorded either on a
Riberg Mag R10Ϫ10 or a Finningan-MAT-95-S. Elemental analy-
ses were performed by the Service Central de Microanalyse du
CNRS.
Ϫ1
to 90 °C (0.15 °C min ) by circulating water using a Huber water
bath, controlled by a Huber PD415 temperature programmer. Oli-
gonucleotides were dissolved in sodium cacodylate buffer (10 m,
pH 7), containing NaCl (10 m or 100 m). Thermal denaturation
curves of the complexes were measured at an oligonucleotide(base
pair):ligand ratio of 12. The oligonucleotide concentration, in base
pairs, was 2 ϫ 10 . The thermal denaturation profiles of the free
duplexes and of the complexes were monitored at 260 nm. Melting
m
temperatures (T ) were determined from the first derivative of the
Ϫ5
Computational Procedure: We have used the JUMNA 10 molecular
[
37]
mechanics procedure. Similar to our previous studies devoted to
porphyrin-oligopeptide conjugates, solvation effects were implicitly
accounted for by the use of a sigmoidal dielectric function, and, to
account for screening effects, the phosphodiester group had a net
melting curves.
Circular Dichroism: Circular dichroism spectra were recorded on a
Jobin Yvon CD6 autodichrograph in a room thermostatted at 20
C. All the measurements were made in sodium cacodylate buffer
10 m, pH 7) containing NaCl (100 m). A 3 mL, 10 mm path-
length quartz cuvette was used to avoid dichroic effects from the
cuvette. The spectra shown in this study were recorded at a DNA-
2
base pair):ligand ratio of 30 for poly(dG-dC) and 12 for the do-
decanucleotides. They were obtained by an average of four accumu-
lations recorded with steps of 0.2 nm and a response time of 0.2 s.
For each measurement, the spectrum of the porphyrin derivative at
the same concentration in the same buffer and same cuvette was
subtracted. The observed ∆(O.D.) was divided by the initial por-
[
19,21]
charge of Ϫ0.5.
Using the JUMNA procedure we generated
two and three intercalation sites for bis- and tris-intercalating de-
rivatives respectively. We then followed a similar strategy to that
used in these previous studies, which involved a succession of con-
°
(
strained and unconstrained energy-minimization steps in which the
DNA was first held rigid and then relaxed.[
19,21]
The distances be-
(
6
7
tween the amino groups of the Arg side-chains and the O /N sites
of the targeted bases were constrained as well as the distances be-
1
4
5
tween the N and C and C atoms of the acridine and the nitrogens
of the G and C bases of the corresponding acridine intercalation
sites. The latter constraint is obviously necessary to ensure that the
acridine rings overlap properly with the bases of the intercalation
sites in the initial energy-minimization steps. Several simulations
involving rigid or relaxed DNA with or without distance con-
straints were performed in parallel, and all ended up with relaxed
DNA and unconstrained ligand minimization.
phyrin concentration to give the apparent ∆ε values, ∆εapp
.
Topoisomerisation Experiments: Reaction mixtures containing pTZ
supercoiled plasmid (ca. 250 ng, final concentration in bp 0.265
µ), an aqueous solution of porphyrin derivative of known concen-
tration (5 µL), 10X topoisomerase buffer (2 µL) and deionized
water (up to 19 µL), were incubated for 5 min at 37 °C. Topoiso-
Molecular dynamics studies were carried out using the Cff91 force-
field and the Accelrys package. The DNA-BABAP complex was
immersed in a water box of 3744 water molecules, and periodic
boundary conditions were applied. No explicit counterions were
included, but the phosphate charges were set to Ϫ0.5. To avoid
distortions of DNA in the course of dynamics simulations, the hy-
drogen bonds between the base pairs were enforced by a harmonic
restraining potential. Three different dynamics simulations were
run at 300 K. In the first, DNA was held rigid, in the second, one
DNA strand was relaxed and in the third, the two strands were
relaxed. Starting from the JUMNA structure for complex A-II, en-
ergy minimization was first carried out with the conjugate gradient
algorithm. Dynamics simulations were then initialized for 5 ps, and
resumed during 100 ps, with frames recorded every 5 ps and sub-
jected to energy minimization.
Ϫ1
merase I (1 µL, 1 u·µL ) was then added and the reaction mixture
incubated for 30 min at 37 °C. The topoisomerase I reaction was
Ϫ1
stopped by adding Proteinase K (2.2 µL, 500 µg·mL ), EDTA (2
µL, 0.5 , pH 8) and SDS (1 µL, 10%), and the mixture was then
incubated for 30 min at 50 °C. The mixture was extracted with
phenol/chloroform, and DNA was precipitated,[ redissolved in
TE (10 µL), and BBSE 6X (2 µL) was added. This plasmid solution
40]
(
10 µL) was submitted to electrophoresis on a 1.3% agarose gel in
TBE buffer in a refrigerated tank maintained at 6 °C. The running
Ϫ1
buffer was recirculated. A constant voltage of 8 V·cm was applied
for 200 min. The DNA was then visualized using the gel that had
been stained for 20 min in an ethidium bromide solution (1
µg·mL ). The gel was photographed under transillumination
Ϫ1
(
312 nm).
DNA Solutions: Oligonucleotides were purchased from Cyberg e` ne
Abbreviations: EDTA: Ethylenediaminetetracetic acid, sodium salt,
(France). They were dissolved in ultra-pure water and their concen-
0.5 , pH 8; SDS: sodium dodecyl sulfate, 10%; BBSE 6X: Bromo-
trations, expressed in bases, were determined by measuring the ab-
phenol Blue, SDS, EDTA (100 µL Bromophenol Blue 2%, 250 µL
SDS, 100 µL EDTA, 250 µL glycerol, 300 µL TE); TBE: Tris, Bo-
rate, EDTA; TE: Tris, EDTA.[
sorbance at 260 nm at 90 °C, applying the following molar extinc-
Ϫ1
Ϫ1
tion coefficients (in  ·cm ): d(TCGGGCGCCCGA): 7758;
40]
d(TCGTACGTACGA): 9975.[ Poly(dG-dC)
Pharmacia Biotech) was dissolved in phosphate buffer (1 mL, pH
) containing phosphate (25 m) and NaCl (150 m). Its concen-
tration in base pairs (bp) was determined spectrophotometrically
at 255 nm using molar extinction coefficient of 1.68
38]
(from Amersham
2
tert-Butyl N-(6-Aminohexyl)carbamate (1): Di-tert-butyl dicarbon-
ate [(Boc) O] (7.09 g, 32.5 mmol) was dissolved in freshly distilled
2
dioxane (80 mL) and added dropwise over 8 h to a stirred solution
of 1,6-diaminohexane (28.3 g, 244 mmol) in dioxane (250 mL) un-
7
a
ϫ
Ϫ1
Ϫ1 [39]
1
04  ·cm
isolated from JM109 Escherichia coli by the alkaline lysis method
and purified using a Hybaid plasmid maxi kit. The concentration bonyl)diaminohexane] was washed with diethyl ether (3 ϫ 10 mL).
of DNA in base pairs was determined spectrophotometrically at The filtrate was concentrated in vacuo, and then water (250 mL)
.
Supercoiled pTZ plasmid (2861 base pairs) was der argon. The mixture was stirred at room temp. for 3 days. The
solution was filtered and the solid residue [N,NЈ-bis(tert-butoxycar-
Eur. J. Org. Chem. 2004, 1781Ϫ1797
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1793

M. Perr e´ e-Fauvet et al.
FULL PAPER
was added to precipitate the residual N,NЈ-bis(tert-butoxycarbonyl)- formation of the activated ester. The solution of 3 was then added
diaminohexane. The precipitate was filtered and washed with water and the reaction mixture was stirred for 48 h at room temp. The
(
3 ϫ 10 mL). The filtrate, containing excess diamine and the desired
product, was extracted with CH Cl (4 ϫ 50 mL). The CH Cl
solution was washed with water, dried over Na SO and the solvent
evaporated to give pure tert-butyl N-(6-aminohexyl)carbamate (1)
crude product was concentrated under reduced pressure and the
residue was dissolved in acetone (25 mL) and added dropwise to a
2
2
2
2
2
4
stirred solution of 5% NaHCO
product by a solution of 5% NaHCO
boxamide (194 mg, 0.41 mmol) in 80% yield.
): δ ϭ 1.23 (m, 2 H), 1.32 (m, 8 H), 1.43 (s, 9 (250 MHz, CDCl ): δ ϭ 1.34 (s, 9 H), 1.40 (m, 6 H), 1.67 (m, 2
3
(250 mL). Treatment of the crude
3
as for 2 afforded acridinecar-
(5.02 g, 23.2 mmol) in 71% yield [relative to (Boc)
250 MHz, CDCl
2
O]. 1H NMR
4
1
H NMR
(
3
3
H), 2.67 (t, J ϭ 6.7 Hz, 2 H), 3.10 (q, J ϭ 6.5 Hz, 2 H), 4.55 (t
unresolved, 1 H). 13C NMR (63 MHz, CDCl
): δ ϭ 26.1 (CH ),
6.2 (CH ), 28.0 (CH ), 29.6 (CH ), 33.0 (CH ), 40.0 (CH
), 78.2 (C),155.7 (C). MS (CI): m/z ϭ 217 (M ϩ H). 7.97 (d, J ϭ 8.8 Hz, 2 H). C NMR (50 MHz, CDCl
(216.33): calcd. C 61.08, H 11.18, N 12.95, O 14.79; (CH ), 27.8 (CH ), 28.7 (CH ), 30.3 (CH ), 30.4 (CH ), 40.2 (CH
40.9 (CH ), 44.6 (CH ), 80.6 (C), 123.6 (C), 126.5 (CH), 128.1
CH), 129.5 (CH), 132.3 (CH), 143.9 (C), 149.5 (C), 158.2 (C),
H), 3.12 (q, J ϭ 6.3 Hz, 2 H), 3.55 (m, 4 H), 5.28 (t unresolved, 1
3
2
H), 6.39 (t unresolved, 1 H), 7.28 (t unresolved, 1 H), 7.38 (t, J ϭ
2
2
3
2
2
2
), 41.5 8.8 Hz, 2 H), 7.61 (t, J ϭ 8.8 Hz, 2 H), 7.81 (d, J ϭ 8.8 Hz, 2 H),
ϩ
13
(CH
2
3
): δ ϭ 27.5
),
C
11
H
24
N
2
O
2
2
2
3
2
2
2
found C 61.01, H 11.02, N 12.73, O 15.02.
2
2
(
1
1
N-[N-(tert-Butoxycarbonyl)-6-aminohexyl]-9-acridinecarboxamide
68.4 (C), 172.4 (C). IR (KBr): ν˜ ϭ 3261, 2932, 1707, 1677, 1644,
(
2): Diisopropylethylamine (DIEA) (1.42 mL, 8.16 mmol) was ad-
ded to a suspension of 9-acridinecarboxylic acid hydrate (1.64 g,
.8 mmol) and BOP (3.31 g, 7.48 mmol) in DMF (30 mL) and the
Ϫ1
ϩ
ϩ
518, 1279, 1168 cm . MS (ES ): m/z found 501.5 (M ϩ Na).
(478.60): calcd. C 67.76, H 7.16, N 11.71; found C
7.81, H 7.38, N 11.37.
27 34 4 4
C H N O
6
6
resulting solution was stirred for 20 min. tert-Butyl N-(6-aminohex-
yl)carbamate (1) (1.47 g, 6.8 mmol) was added and the reaction
mixture was stirred for 24 h at room temp. The crude product was
concentrated under reduced pressure and the residue was dissolved
in acetone (20 mL) and added dropwise to a stirred solution of 5%
N-[N-(Glycyl)-6-aminohexyl]-9-acridinecarboxamide, TFA Salt (5):
Dissolution of acridinecarboxamide 4 (141 mg, 0.29 mmol) in TFA/
CH Cl (v/v, 3 mL) followed by the same treatment as applied to 2
2 2
afforded the crude TFA salt 5 (143 mg, 0.29 mmol) which was used
in the next step without further purification. H NMR (250 MHz,
NaHCO
oped by Kossanyi et al.[ The mixture was allowed to stand for
4 h at room temp. This afforded a precipitate that was filtered,
washed with water and dried to give 2 (2.19 g, 5 mmol, 74%) as a
3
(200 mL), according to the procedure previously devel-
1
41]
CD
4
3
OD): δ ϭ 1.61 (m, 6 H), 1.78 (m, 2 H), 3.35 (s, 2 H), 3.65 (m,
2
H), 7.69 (t, J ϭ 8.8 Hz, 2 H), 7.91 (t, J ϭ 8.8 Hz, 2 H), 8.07 (d,
13
J ϭ 8.8 Hz, 2 H), 8.21 (d, J ϭ 8.8 Hz, 2 H). C NMR (63 MHz,
CD OD): δ ϭ 27.5 (CH ), 27.8 (CH ), 30.2 (CH ), 30.3 (CH ), 40.5
), 40.8 (CH ), 41.4 (CH ), 123.7 (C), 126.5 (CH), 128.1 (CH),
29.6 (CH), 132.3 (CH), 145.2 (C), 149.6 (C), 167.1 (C), 171.2 (C).
1
pale yellow powder. H NMR (200 MHz, CDCl
H), 1.36 (m, 8 H), 3.16 (q, J ϭ 6.4 Hz, 2 H), 3.67 (q, J ϭ 6.4 Hz,
H), 4.51 (t unresolved, 1 H), 6.48 (t unresolved, 1 H), 7.50 (t,
J ϭ 7.8 Hz, 2 H), 7.72 (t, J ϭ 7.8 Hz, 2 H), 7.96 (d, J ϭ 7.8 Hz, 2
H), 8.10 (d, J ϭ 7.8 Hz, 2 H). 13C NMR (63 MHz, CDCl
): δ ϭ
6.9 (CH ), 27.3 (CH ), 28.7 (CH ), 29.8 (CH ), 30.4 (CH ), 40.6
CH ), 40.8 (CH ), 79.6 (C), 123.1 (C), 126.0 (CH), 127.5 (CH),
29.0 (CH), 131.6 (CH), 143.1 (C), 148.9 (C), 157.8 (C), 168.4 (C).
Na): calcd. 444.22631;
O (439.56): calcd. C 68.31, H
.57, N 9.56, O 14.56; found C 67.71, H 7.64, N 9.51, O 14.28.
3
): δ ϭ 1.33 (s, 9
3
2
2
2
2
(CH
2
2
2
2
1
ϩ
ϩ
MS (ES ): m/z found 379.2 (M ϩ H).
3
2
(
1
2
2
3
2
2
α
η
N-(N-{N-[N -Fluoren-9-ylmethoxycarbonyl-N -(2,2,4,6,7-penta-
methyldihydrobenzofuran-5-sulfonyl)arginyl]glycyl}-6-aminohexyl)-
2
2
9
(
0
-acridinecarboxamide (6):
100 mg, 0.15 mmol), BOP (79 mg, 0.18 mmol) and DIEA (31 µL,
.18 mmol) in DMF (4 mL) was stirred at room temp. for 15 min,
A mixture of Fmoc-Arg(Pbf)-OH
MS (HRMS, m/z calcd. for C25
31 3 3
H N O
found 444.22610. C25 ·H
H
31
N
3
O
3
2
7
then cooled to 0 °C. Acridinecarboxamide 5 (74 mg, 0.15 mmol)
and DIEA (130 µL, 0.75 mmol) were dissolved in DMF (4 mL)
and added dropwise to the activated ester of Fmoc-Arg(Pbf)-OH.
This reaction mixture was stirred for 20 min at 0 °C, then at room
temp. for 24 h. DIEA and DMF were evaporated under reduced
pressure and the residue was dissolved in acetone (20 mL) and ad-
ded dropwise to a stirred solution of 5% NaHCO (200 mL). The
3
mixture was allowed to stand for 24 h and the resulting precipitate
was filtered, dried and purified by silica gel 60 (Merck, 40Ϫ63 µm)
N-(6-Aminohexyl)-9-acridinecarboxamide, TFA Salt (3): N-[N-(tert-
Butoxycarbonyl)-6-aminohexyl]-9-acridinecarboxamide (2)
270 mg, 0.64 mmol) was dissolved in TFA/CH Cl (v/v, 4 mL) and
(
2
2
the resulting solution was stirred at room temp. for 1 h. The solvent
and most of the trifluoroacetic acid were removed under reduced
pressure to give a brown gum, which upon trituration with diethyl
ether (2 ϫ 10 mL) yielded a yellow solid. The solid was dried under
reduced pressure to afford the crude TFA salt 3 (277 mg, 99%)
1
column chromatography (CH
pound 6 (109 mg, 0.11 mmol) in 70% yield. H NMR (400 MHz,
CD OD): δ ϭ 1.40 [m, 10 H, (CH alkyl chain ϩ 2 CH Pbf],
.50 (m, CH γ), 1.67 (m, H,
β), 2.05 (s, 3 H, CH Pbf), 2.48 (s, 3 H,
Pbf), 2.95 (s, 2 H, CH Pbf), 3.14 (m,
CH δ), 3.55 (t, 6.8 Hz, H,
NHCOAcr), 3.70 (d, J ϭ 16.8 Hz, 1 H, CH Gly), 3.88 (d, J ϭ
2 2
Cl /MeOH, 95:5) to afford com-
which was used in the next step without further purification. H
1
NMR (250 MHz, CD
3
OD): δ ϭ 1.56 (m, 4 H), 1.72 (m, 2 H), 1.82
m, 2 H), 2.97 (t, J ϭ 7.3 Hz, 2 H), 3.68 (t, J ϭ 6.8 Hz, 2 H), 7.76
t, J ϭ 8.8 Hz, 2 H), 8.02 (t, J ϭ 8.8 Hz, 2 H), 8.15 (d, J ϭ 8.8 Hz,
3
2
)
2
3
(
(
2
2
1
CH
CH
4
CH
2
1
4
H, CH
2
CH
2
NHGly
ϩ
2
4
H), 8.26 (d, J ϭ 8.8 Hz, 2 H). 13C NMR (50 MHz, CD
2
CH
2
NHCOAcr ϩ CH
2
3
3
OD): δ ϭ
), 40.6 (CH ), 40.8
), 123.7 (C), 126.7 (CH), 128.4 (CH), 128.7 (CH), 133.0 (CH),
3
Pbf), 2.55 (s, 3 H, CH
NHGly
3
2
7.1 (CH
2
), 27.7 (CH
2
), 28.5 (CH
2
), 30.1 (CH
2
2
H, CH
2
ϩ
2
J
ϭ
2
(CH
2
ϩ
ϩ
145.0 (C), 148.7 (C), 168.8 (C). MS (ES ): m/z found: 322.1 (M
6.8 Hz, 1 H, CH Gly), 3.95 (t, J ϭ 7.8 Hz, 1 H, CHα), 4.19 (t,
ϩ H).
J ϭ 6.8 Hz, 1 H, CH Fmoc), 4.37 (m, 2 H, CH Fmoc), 7.28 (t,
2
N-{N-[N-(tert-Butoxycarbonyl)glycyl]-6-aminohexyl}-9-acridine-
carboxamide (4): DIEA (436 µL, 2.5 mmol) was added to a stirred
solution of N-(6-aminohexyl)-9-acridinecarboxamide, trifluo-
J ϭ 7.5 Hz, 2 H, 2 CH Fmoc), 7.36 (t, J ϭ 7.5 Hz, 2 H, 2 CH
Fmoc), 7.63 (m, 4 H, 2 CH Fmoc ϩ 2 CH Acr), 7.78 (d, J ϭ
7.5 Hz, 2 H, 2 CH Fmoc), 7.85 (t, J ϭ 8.1 Hz, 2 H, 2 CH Acr),
roacetate salt (3) (222 mg, 0.51 mmol) in DMF (2 mL). Simul- 8.04 (d, J ϭ 8.1 Hz, 2 H, 2 CH Acr), 8.17 (d, J ϭ 8.1 Hz, 2 H, 2
1
3
taneously, a mixture of N-(tert-butoxycarbonyl)glycine (107 mg, CH Acr). C NMR (100 MHz, CD
3
OD): δ ϭ 12.5 (CH
), 27.4 (CH ), 27.7 (CH ), 28.7 (CH
), 30.9 (CH ), 40.4 (CH ), 40.8 (CH ), 41.3
3
), 18.4
0
0
.61 mmol), PyBOP (382 mg, 0.73 mmol) and DIEA (127 µL,
.73 mmol) in DMF (2 mL) was stirred for 15 min to allow the
(CH
30.2 (CH
3
), 19.6 (CH
3
), 26.9 (CH
2
2
2
3
),
2
), 30.3 (CH
2
2
2
2
1794
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 1781Ϫ1797

DNA Intercalating Porphyrins Designed to Target the Major Groove
FULL PAPER
(
8
1
CH
2
), 43.6 (CH
7.6 (C), 118.4 (C), 121.0 (CH), 123.7 (C), 126.0 (CH), 126.1 (CH), (br., 12 H, 8 CH pyrrole ϩ 4 CH pyridyl), 8.86 (d, 2 H, 2 CH
26.6 (C), 128.2 (CH), 128.9 (CH), 129.6 (CH), 132.3 (CH), 133.5 pyridyl). UV/Vis (CH OH): λmax (% Abs) ϭ 250 (53), 358 (8.3),
2 2 2
), 43.9 (CH ), 48.0 (CH), 56.6 (CH), 68.0 (CH ), 2 H, 2 CH Acr), 7.87 ϩ 7.98 (m ϩ d, 6 H, 6 CH pyridyl), 8.55Ϫ8.74
3
ϩ
(C), 134.3 (C), 139.4 (C), 142.6 (C), 143.9 (C), 145.1 (C), 149.6 (C), 421 (100), 512 (1.6), 555 (4.2), 594 (1.2), 643 nm (0.3). MS (ES ):
ϩ
2ϩ
1
3
58.1 (C), 158.8 (C), 169.0 (C), 171.4 (C). IR (KBr): ν¯ ϭ 3340, m/z (%) ϭ 1704.9 (100) (M ϩ H), 853.3 (26) (M ϩ2 H)/2, 569.2
064, 2930, 1703, 1648, 1549, 1255, 1105 cm . MS [HRMS, m/z
Ϫ1
3ϩ
(8) (M ϩ3 H)/3.
ϩ
64 8 8
calcd. for C56H N O SNa (M ϩ Na)]: calcd. 1031.44743;
found 1031.44655.
Porphyrin 10: A mixture of porphyrin diacid (synthon A) (17 mg,
0.022 mmol), BOP (44 mg, 0.10 mmol) and DIEA (40 µL,
η
0.23 mmol) in DMF (2 mL) was stirred for 1 h under argon. Acri-
dinecarboxamide 7 (39 mg, 0.05 mmol) was then added and the
reaction mixture was stirred for 48 h at room temp. After removal
of DIEA and DMF under reduced pressure, the crude product was
purified by silica gel (Polygoprep 6Ϫ12 µm) column chromatogra-
N-(N-{N-[N -(2,2,4,6,7-Pentamethyldihydrobenzofuran-5-sulfonyl)-
arginyl]glycyl}-6-aminohexyl)-9-acridinecarboxamide (7): DBU (52
µL, 0.35 mmol) was added to a solution of acridinecarboxamide 6
(
91 mg, 0.09 mmol) in THF (2.6 mL) and the solution was stirred
at room temp. for 15 min. The reaction mixture was poured into
diethyl ether (100 mL) whilst stirring. The precipitate was filtered,
washed with diethyl ether and dried. It was then dissolved in
phy (CH
.011 mmol) in 48% yield. H NMR (400 MHz, CDCl
m, 20 H, 2 (CH alkyl chain ϩ 4 CH Pbf], 1.38 (m, 8 H, 2
CH CH NHGly ϩ 2 CH γ), 1.52 (m, 4 H, 2 CH CH NHCOAcr),
.61 (m, 2 H, 2 CHβ ), 1.73 (m, 2 H, 2 CHβ ), 1.88 (s, 6 H, 2 CH
Pbf), 2.30 (s, 6 H, 2 CH Pbf), 2.38 (s, 6 H, 2 CH
H, 2 CH Pbf), 2.91 (m, 4 H, 2 CH δ), 3.12 (m, 4 H, 2 CH
.41 (m, 4 H, 2 CH
Gly), 3.78 (d, J ϭ 16.7 Hz, 2 H, 2 CH Gly), 4.32 (m, 2 H, 2 CHα),
2 2
Cl /MeOH, 90:10) to afford the porphyrin 10 (26 mg,
1
0
3
): δ ϭ 1.28
[
2
)
2
3
CH
ous NaCl solution (50 mL), then with water (3 ϫ 50 mL), and dried
over Na SO . After evaporation in vacuo, the acridinecarboxamide
(54 mg, 0.07 mmol) was obtained in 75% yield. H NMR
400 MHz, CD OD): δ ϭ 1.44 (s, 6 H, 2 CH Pbf), 1.56 [m, 10 H,
CH alkyl chain CH CH γ), 1.78 (m, H,
NHCOAcr), 2.07 (s, 3 H, CH Pbf), 2.49 (s, 3 H, CH
Pbf), 2.55 (s, 3 H, CH Pbf), 2.98 (s, 2 H, CH Pbf), 3.14 (m, 2 H,
CH δ), 3.24 (t, J ϭ 6.8 Hz, 2 H, CH NHGly), 3.64 (t, J ϭ 6.8 Hz,
H, CH NHCOAcr), 3.81 (d, J ϭ 16.5 Hz, 1 H, CH Gly), 3.86 (d,
J ϭ 16.5 Hz, 1 H, CH Gly), 3.95 (m, 1 H, CHα), 7.68 (t, J ϭ
.8 Hz, 2 H, 2 CH Acr), 7.89 (t, J ϭ 8.8 Hz, 2 H, 2 CH Acr), 8.07
d, J ϭ 8.8 Hz, 2 H, 2 CH Acr), 8.19 (d, J ϭ 8.8 Hz, 2 H, 2 CH
2 2
Cl (50 mL) and the solution was washed with saturated aque-
2
2
2
2
2
1
1
2
3
2
4
1
3
3
Pbf), 2.75 (s, 4
7
2
2
2
NHGly),
(
(
3
3
3
2
NHCOAcr), 3.67 (d, J ϭ 16.7 Hz, 2 H, 2 CH
2
)
3
ϩ
2
β
ϩ
2
2
CH
2
CH
2
3
3
4
2
.47 (d, J ϭ 15.0 Hz, 2 H, 2 CHOPh), 4.56 (d, J ϭ 15.0 Hz, 2 H,
CHOPh), 6.83 (s, 1 H, CH Ph), 7.27 (d, J ϭ 2.4 Hz, 2 H, 2 CH
3
2
2
2
Ph), 7.31 (m, 4 H, 4 CH Acr), 7.51 (m, 4 H, 4 CH Acr), 7.78 (d,
J ϭ 8.5 Hz, 4 H, 4 CH Acr), 7.93 (d, J ϭ 8.5 Hz, 4 H, 4 CH Acr),
2
2
8
2
.07 (d, J ϭ 5.8 Hz, 4 H, 4 CH pyridyl), 8.10 (d, J ϭ 5.8 Hz, 2 H,
CH pyridyl), 8.70 (br., 8 H, 8 CH pyrrole), 8.83 (d, J ϭ 5.8 Hz,
8
(
13
4 H, 4 CH pyridyl), 8.87 (d, J ϭ 5.8 Hz, 2 H, 2 CH pyridyl). UV/
Vis (CH OH): λmax (% Abs) ϭ 250 (90), 360 (16), 415 (100), 511
Acr). C NMR (100 MHz, CD
3
OD): δ ϭ 12.5 (CH
), 27.5 (CH ), 27.8 (CH ), 28.7 (CH
), 40.9 (CH ), 41.5 (CH ), 43.3 (CH
2
3
), 18.4 (CH
), 30.2
),
3
),
3
1
9.6 (CH
3
), 26.6 (CH
), 30.3 (CH
3.9 (CH ), 55.6 (CH), 87.6 (C), 118.4 (C), 123.6 (C), 126.0 (CH),
2
2
2
2
3
ϩ
(
1
6.7), 544 (2.0), 587 (2.1), 644 nm (1.2). MS (ES ): m/z (%) ϭ
152.4 (100) (M ϩ 2 H)/2, 768.6 (48) (M ϩ 3 H)/3.
(CH
2
2
), 40.3 (CH
2
2
2
2
ϩ
3ϩ
4
1
26.5 (C), 128.1 (CH), 129.5 (CH), 132.3 (CH), 133.4 (C), 134.3
Synthesis of Porphyrins 9 and 11 by Removal of the Pbf Protective
Group: Porphyrin 8 (10 mg, 5.75 ϫ 10 mmol) was dissolved in
(C), 139.3 (C), 143.9 (C), 145.1 (C), 149.5 (C), 158.0 (C), 169.0 (C),
Ϫ3
ϩ
171.3 (C). MS [HRMS, m/z calcd. for C41
H
54
N
8
O
6
SNa (M
ϩ
TFA/H
room temp. for 0.5 h. Deprotection was monitored by reverse phase
TLC (AcOH/MeOH/H O, 2:1:2). The solvents were removed under
reduced pressure and the crude residue was dissolved in MeOH
2 mL). This solution was slowly added to Et O (50 mL) whilst
stirring to afford porphyrin 9 as a red precipitate which was filtered
2
O (95:5, 1 mL) and the resulting solution was stirred at
Na)]: calcd. 809.37847; found 809.37792.
2
Porphyrin 8: Porphyrin diacid (synthon A) (36 mg, 0.047 mmol)
was dissolved in DMF (2 mL) and the solution was stirred under
argon at 0 °C. After complete dissolution, BOP (22 mg, 0.05 mmol)
and DIEA (80 µL, 0.47 mmol) were added and the mixture was
stirred for 0.5 h. Acridinecarboxamide 7 (39 mg, 0.05 mmol) was
then added and the mixture was stirred in the dark for 3 h. -Argi-
nine methyl ester hydrochloride (11.5 mg, 0.044 mmol) and BOP
(
2
Ϫ3
2
and washed with Et O. Porphyrin 9 (9.6 mg, 5.7 ϫ 10 mmol) was
obtained in quantitative yield and was immediately used in the next
step without further purification. Porphyrin 11 was obtained in
quantitative yield from porphyrin 10 using the same procedure, and
was immediately used in the next step without further purification.
(19.5 mg, 0.044 mmol) were then added and the mixture was stirred
at room temp. for 24 h. After evaporation of the solvent, the residue
was dissolved in acetone (5 mL) and the resulting solution was
slowly added to 50 mL of water whilst stirring. The mixture was
allowed to stand overnight at 6 °C. The resulting red precipitate
was filtered off through a filter funnel (porosity 4) coated with Ce-
Mono-acridyl-bis-arginyl-porphyrin (MABAP): A large excess of
CH
(7 mg, 4.16 ϫ 10 mmol) in DMF (6 mL). The mixture was stirred
at 40 °C for 3 h. After removal of DMF and CH I under reduced
3
I (52 µL, 0.85 mmol) was added to a solution of porphyrin 9
Ϫ3
3
2 2
lite, then dissolved in CH Cl /MeOH, 95:5, and the solvent evapo- pressure, the crude residue was dissolved in water (1 mL) and this
Ϫ
rated in vacuo. The crude product, consisting of a mixture of por-
phyrins, was passed through a reverse phase column Lichroprep
solution was passed through a column of Dowex Cl ion exchange
resin. The aqueous layer was evaporated and the product was pre-
Ϫ3
RP-18 (40Ϫ63 µm) (AcOH/MeOH/H
2
O, 1:1:2 then 2:1:2). Porphy-
cipitated in H
rin 8 (10 mg, 5.75.10 mmol) was obtained in 12% yield. H NMR mmol) in 72% yield. H NMR (400 MHz, CD
400 MHz, CD OD): δ ϭ 1.52 [m, 16 H, (CH alkyl chain ϩ 2 [m, 8 H, (CH alkyl chain], 1.64 (m, 4 H, 2 CH
CH β ϩ 2 CH γ], 1.76 (s, 6 H, 2 CH Pbf), 2.16 (s, 3 H, CH Pbf), 2 CHβ ), 1.94 (m, 2 H, 2 CHβ ), 3.03 (m, 2 H, CH
.27 (s, 3 H, CH Pbf), 2.58 (s, 2 H, CH Pbf), 2.90 (m, 4 H, 2 (m, 4 H, 2 CH δ), 3.60 (m, 2 H, CH NHCOAcr), 3.54 (s, 3 H,
CH δ), 3.04 (m, 2 H, CH NHGly), 3.34 (m, 2 H, CH NHCOAcr), COOCH ), 3.71 (d, J ϭ 16.6 Hz, 1 H, CH Gly), 3.88 (d, J ϭ
.44 (s, 3 H, COOCH ), 3.68 (m, 2 H, CH Gly), 4.24 (m, 1 H, 16.6 Hz, 1 H, CH Gly), 4.41 (t, 1 H, CHα), 4.59 (m, 5 H, CHα ϩ
CHα), 4.37 (s, 2 H, CH OPh), 4.49 (m, 3 H, CHα ϩ CH OPh), 2 CH
.81 (s, 1 H, CH Ph), 7.31 (m, 4 H, 4 CH Acr), 7.44 (m, 2 H, 2
CH Ph), 7.67 (d, J ϭ 8.0 Hz, 2 H, 2 CH Acr), 7.81 (d, J ϭ 8.0 Hz,
2
O/acetone to afford MABAP (5 mg, 2.98 ϫ 10
Ϫ3
1
1
3
OD): δ ϭ 1.19Ϫ1.45
γ), 1.82 (m, 2 H,
NHGly), 3.17
(
3
2
)
4
2
)
4
2
2
2
3
3
1
2
2
2
3
2
2
2
2
2
2
3
3
3
2
ϩ
2
2
2 3
OPh), 4.81 (s, 9 H, 3 N CH ), 7.20 (s, 1 H, CH Ph), 7.25 (t,
6
1 H, CH Acr), 7.41 (t, 1 H, CH Acr), 7.46 (t, 1 H, CH Acr), 7.52
(t, 1 H, CH Acr), 7.58 (s, 1 H, CH Ph), 7.65 (m, 3 H, CH Ph ϩ 2
Eur. J. Org. Chem. 2004, 1781Ϫ1797
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1795

M. Perr e´ e-Fauvet et al.
FULL PAPER
Gresh, Tetrahedron Lett. 1993, 34, 7263Ϫ7266. [3d] H. Goula-
ouic, S. Carteau, F. Subra, J.-F. Mouscadet, C. Auclair, J.-S.
Sun, Biochemistry 1994, 33, 1412Ϫ1418. [ C. Bourdouxhe-
Housiaux, P. Colson, C. Houssier, M. J. Waring, C. Bailly, Bio-
chemistry 1996, 35, 4251Ϫ4264. [ P. H e´ lissey, C. Bailly, J. N.
Vishwakarma, C. Auclair, M. J. Waring, S. Giorgi-Renault,
Anti-Cancer Drug Des. 1996, 11, 527Ϫ551.
CH Acr), 7.71 (d, 1 H, CH Acr), 7.85 (d, 1 H, CH Acr), 8.88 (d,
2
8
H, 2 CH pyridyl), 8.98 (m, 4 H, 4 CH pyridyl), 9.06Ϫ9.30 (br.,
H, 8 CH pyrrole), 9.33 (d, 2 H, 2 CH pyridyl), 9.39 (d, 4 H, 4
3e]
CH pyridyl). UV/Vis (H
2
O): λmax (% Abs) ϭ 252 (74), 361 (22.7),
3f]
ϩ
4
34 (100), 556 (7.6), 644 nm (1.1). MS (ES ): m/z (%) ϭ 374.6
4ϩ
3ϩ
(
100) (M Ϫ 5 Cl Ϫ H)/4, 498.8 (53) (M Ϫ 5 Cl Ϫ 2 H)/3, 299.7
ϩ
23) (M5 Ϫ 5 Cl)/5.
(
[4] [4a]
[4b]
H. E. Moser, P. B. Dervan, Science 1987, 238, 645Ϫ650.
T. Le Doan, L. Perrouault, D. Praseuth, N. Habhoub, J.-L.
3
Bis-acridyl-bis-arginyl-porphyrin (BABAP): A large excess of CH I
Decout, N. T. Thuong, J. Lhomme, C. H e´ l e` ne, Nucleic Acids
(
3
43 µL, 0.7 mmol) was added to a solution of porphyrin 11 (7 mg,
.45 ϫ 10 mmol) in DMF (6 mL). The mixture was stirred at 40
[4c]
Res. 1987, 15, 7749Ϫ7760.
C. H e´ l e` ne, J.-J. Toulm e´ , Biochim.
Ϫ3
[4d]
Biophys. Acta 1990, 1049, 99Ϫ125.
A. De Mesmaeker, R.
°C for 3 h. After removal of DMF and CH
3
I under reduced pres-
Häner, P. Martin, H. E. Moser, Acc. Chem. Res. 1995, 28,
366Ϫ374. [ C. Escud e´ , C. H. Nguyen, J.-L. Mergny, J.-S. Sun,
E. Bisagni, T. Garestier, C. H e´ l e` ne, J. Am. Chem. Soc. 1995,
4e]
sure, the crude residue was dissolved in water (1 mL) and this solu-
Ϫ
tion was passed through a column of Dowex Cl ion exchange
117, 10212Ϫ10219.
resin. The aqueous layer was evaporated and the product was pre-
[
[
5]
Ϫ3
J.-S. Sun, J.-C. Fran c¸ ois, T. Montenay-Garestier, T. Saison-
Behmoaras, V. Roig, N. T. Thuong, C. H e´ l e` ne, Proc. Natl.
Acad. Sci. U. S. A. 1989, 86, 9198Ϫ9202.
cipitated in H
mmol) in 86% yield. H NMR (400 MHz, CD
H, 2 (CH alkyl chain], 1.37 (m, 8 H, 2 CH
NHCOAcr), 1.68 (m, 4 H, 2 CH γ), 1.90 (m, 4 H, 2
β), 2.99 (m, 4 H, 2 CH δ), 3.12 (m, 4 H, 2 CH NHGly), 3.19
NHCOAcr), 3.75 (d, J ϭ 17.0 Hz, 2 H, 2 CH Gly),
.83 (d, J ϭ 17.0 Hz, 2 H, 2 CH Gly), 4.33 (m, 2 H, 2 CHα), 4.78
2
O/acetone to afford BABAP (6 mg, 2.96 ϫ 10
1
3
OD): δ ϭ 1.15 [m,
CH NHGly ϩ 2
8
2
)
2
2
2
6] [6a]
[6b]
B. Hyrup, P. E. Nielsen, Bioorg. Med. Chem. 1996, 4, 5Ϫ23.
P. E. Nielsen, Acc. Chem. Res. 1999, 32, 624Ϫ630.
CH
2
CH
2
2
CH
2
2
2
[7]
C. Park, J. L. Campbell, W. A. Goddard III, J. Am. Chem. Soc.
(m, 4 H, 2 CH
2
1
995, 117, 6287Ϫ6291.
3
[8]
B. Cuenoud, A. Scheparz, Science 1993, 259, 510Ϫ513.
ϩ
[9] [9a]
(s, 4 H, 2 CH
2
OPh), 4.82 (s, 9 H, 3 N CH
3
), 7.19 (m, 9 H, CH Ph
N. Y. Sardesai, K. Zimmerman, J. K. Barton, J. Am. Chem.
Soc. 1994, 116, 7502. [ N. Y. Sardesai, J. K. Barton, J. Biol.
9b]
ϩ 8 CH Acr), 7.53 (m, 10 H, 2 CH Ph ϩ 8 CH Acr), 8.94 (d, J ϭ
5
9
.6 Hz, 4 H, 4 CH pyridyl), 9.00 (d, J ϭ 5.6 Hz, 2 H, 2 CH pyridyl),
.15 (br., 8 H, 8 CH pyrrole), 9.35 (d, J ϭ 5.6 Hz, 4 H, 4 CH
Inorg. Chem. 1997, 2, 762Ϫ771.
[
[
10] [10a]
K. C. Murdock, R. C. Child, P. F. Fabio, R. B. Angier, R.
E. Wallace, F. E. Durr, R. V. Citarella, J. Med. Chem. 1979,
pyridyl), 9.40 (d, J ϭ 5.6 Hz, 2 H, 2 CH pyridyl). UV/Vis (H
2
O):
2, 1024Ϫ1030. [
10b]
R. E. Wallace, K. C. Murdock, R. B. Ang-
2
λ
max (% Abs) ϭ 251 (100), 360 (21.3), 438 (73), 552 (9.0), 641 nm
ϩ
4ϩ
ier, F. E. Durr, Cancer Res. 1979, 39, 1570Ϫ1574.
(
0.5). MS (ES ): m/z (%) ϭ 461.1 (100) (M Ϫ 5 Cl Ϫ H)/4, 614.1
11] [11a]
C. Garbay-Jaureguiberry, C. Esnault, M. Delepierre, P.
ϩ
5ϩ
(
44) (M3 Ϫ 5 Cl Ϫ 2 H)/3, 368.8 (19) (M Ϫ 5 Cl)/5.
Laugaa, S. Laalami, J.-B. Le Pecq, B. P. Roques, Drugs Exptl.
Clin. Res. 1987, 13, 353Ϫ357. [
11b]
C. Garbay-Jaureguiberry,
M.-C. Barsi, A. Jacquemin-Sablon, J.-B. Le Pecq, B. P. Roques,
J. Med. Chem. 1992, 35, 72Ϫ81.
G. J. Atwell, B. F. Cain, B. C. Baguley, G. J. Finlay, W. A.
Denny, J. Med. Chem. 1984, 27, 1481Ϫ1485.
Acknowledgments
[
[
[
12]
13]
This work was supported by the Agence Nationale de Recherche
sur le Sida (ANRS), which is personally acknowledged by one of
us (A. K.) for financial support. We also thank the Ligue Nationale
Contre le Cancer (Comit e´ de Paris) for support. The molecular
dynamics studies were done on the computers of CINES (Montpel-
lier, France). Finally, we thank the Commissariat a` lЈEnergie Ato-
mique (CEA) and the D e´ partement dЈIng e´ nierie des Prot e´ ines
C. Bailly, M. Brana, M. J. Waring, Eur. J. Biochem. 1996,
2
40, 195Ϫ208.
14] [14a]
R. H. Terbrueggen, J. K. Barton, Biochemistry 1995, 34,
14b]
8
227Ϫ8234. [
D. T. Odom, C. S. Parker, J. K. Barton, Bio-
chemistry 1999, 38, 5155Ϫ5163.
M. Perr e´ e-Fauvet, N. Gresh, Tetrahedron Lett. 1995, 36,
[15]
(
Prof. A. M e´ nez) for access to their dichrograph.
4
227Ϫ4230.
[
16] [16a]
R. F. Pasternack, E. J. Gibbs, J. J. Villafranca, Biochemistry
16b]
[
1] [1a]
1983, 22, 2406Ϫ2414. [
R. F. Pasternack, E. J. Gibbs, in
N. Zein, M. Poncin, R. Nilakantna, G. A. Ellestad, Science
989, 244, 697Ϫ699. [
1b]
Metal DNA Chemistry (Ed.: T. Tullius), American Chemical
1
C. Zimmer, U. Wahnert, Progr. Bi-
16c]
[1c]
Society, Washington, DC, 1989, pp. 59Ϫ73. [
R. F. Paster-
ophys. Mol. Biol. 1986, 41, 31Ϫ112.
Wade, T. J. Dwyer, B. H. Geierstanger, D. E. Wemmer, P. B.
M. Mrksich, W. S.
nack, E. J. Gibbs, in Metal Ions in Biological Systems, Vol. 33
(
Eds.: A. Sigel, H. Sigel), Marcel Dekker Inc., New York, 1996,
[16d]
Dervan, Proc. Natl. Acad. Sci. U. S. A. 1992, 89, 7586Ϫ7590.
[
1d]
pp. 367Ϫ397.
R. J. Fiel, J. C. Howard, E. H. Mark, N.
T. J. Dwyer, B. H. Geierstanger, Y. Bathini, J. W. Lown, D.
E. Wemmer, J. Am. Chem. Soc. 1992, 114, 5911Ϫ5919. [ S.
N. Ho, S. H. Boyer, S. L. Schreiber, S. J. Danishefshky, G. R.
1e]
Datta-Gupta, Nucleic Acids Res. 1979, 6, 3093Ϫ3118.
[
[
[
17] [17a]
D. Praseuth, A. Gaudemer, J. B. Verlhac, I. Kraljic, I. Sis-
[
17b]
so e¨ ff, E. Guill e´ , Photochem. Photobiol. 1986, 44, 717Ϫ724.
R. W. Byrnes, R. J. Fiel, N. Datta-Gupta, Chem.-Biol. Interact.
988, 67, 225Ϫ241.
Crabtree, Proc. Natl. Acad. Sci. U. S. A. 1994, 91, 9203Ϫ9207.
[
1f]
V. A. Nikolaev, S. L. Grokhovsky, A. N. Surovaya, T. A.
1
Leinsoo, N. Y. Sidorova, A. S. Zasedatelev, A. L. Zhuze, G. A.
18] [18a]
A. Villanueva, L. Caggiari, G. Jori, C. Milanesi, J. Photo-
Strahan, R. H. Shafer, G. V. Gursky, J. Biomol. Struct. Dyn.
chem. Photobiol., B: Biol. 1994, 23, 49Ϫ56. [
18b]
A. Villanueva,
[
1g]
1996, 14, 31Ϫ47.
H. Xuereb, M. Maletic, J. Gildersleeve, I.
A. Juarranz, V. Diaz, J. Gomez, M. Canete, Anti-Cancer Drug
Des. 1992, 7, 297Ϫ303.
N. Gresh, M. Perr e´ e-Fauvet, J. Comput.-Aided Mol. Des. 1999,
13, 123Ϫ137.
P. Konig, R. Giraldo, L. Chapman, D. Rhodes, Cell 1996, 85,
125Ϫ136.
S. Mohammadi, M. Perr e´ e-Fauvet, N. Gresh, K. Hillairet, E.
Pelczer, D. Kahne, J. Am. Chem. Soc. 2000, 122, 1883Ϫ1890.
[
1h]
S. Neidle, Nat. Prod. Rep. 2001, 18, 291Ϫ309. [1i] P. B.
19]
Dervan, Bioorg. Med. Chem. 2001, 9, 2215Ϫ2235.
[
[
2] [2a]
F. Arcamone, Doxorubicin, Anticancer Antibiotics, Aca-
demic Press, New York, 1981. [ S. H. Lee, I. H. Goldberg,
Biochemistry 1989, 28, 1019Ϫ1026.
2b]
[20]
[21]
3] [3a]
C. Bailly, N. Helbecque, J.-P. H e´ nichart, P. Colson, C.
Houssier, K. E. Rao, R. G. Shea, J. W. Lown, J. Mol. Recognit.
Taillandier, Biochemistry 1998, 37, 6165Ϫ6178.
[
3b]
[22] [22a]
1
990, 3, 26Ϫ35.
Chem. 1991, 2, 379Ϫ393.
r e´ e-Fauvet, A. Gaudemer, P. H e´ lissey, S. Giorgi-Renault, N.
C. Bailly, J.-P. H e´ nichart, Bioconjugate
A. Tsimanis, V. Bichko, D. Dreilina, J. Meldrais, V. Lozha,
R. Kukaine, E. Gren, Nucleic Acids Res. 1983, 11, 6079Ϫ6087.
[3c]
G. Anneheim-Herbelin, M. Per-
[22b]
F. D. Hong, H.-J. S. Huang, H. To, L.-J. S. Young, A. Oro,
1796
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 1781Ϫ1797

DNA Intercalating Porphyrins Designed to Target the Major Groove
FULL PAPER
[
26]
R. Bookstein, E. Y.-H. P. Lee, W.-H. Lee, Proc. Natl. Acad.
Sci. U. S. A. 1989, 86, 5502Ϫ5506. [ M. Dvorak, P. Urbanek,
P. Bartunek, V. Paces, J. Vlach, V. Pecenka, L. Arnold, M. Trav-
nicek, J. Riman, Nucleic Acids Res. 1989, 17, 5651Ϫ5664. [
The binding of MABAP and BABAP to other sequences has
been investigated and the results will be presented in a forth-
coming paper.
22c]
22d]
[27]
M. Perr e´ e-Fauvet, N. Gresh, J. Biomol. Struct. Dyn. 1994, 11,
S. Smith, D. Baker, L. Jardines, Biochem. Biophys. Res. Com-
1203Ϫ1224.
mun. 1989, 160, 1397Ϫ1402. [
Biol. 1995, 251, 629Ϫ647.
23] [23a]
22e]
Y. Timsit, D. Moras, J. Mol.
[28] [28a]
A. Pullman, B. Pullman, Q. Rev. Biophys. 1981, 14,
28b]
289Ϫ380. [
1985, 5, 1021Ϫ1032.
R. Lavery, B. Pullman, J. Biomol. Struct. Dyn.
[
S. Wain-Hobson, P. Sonigo, O. Danos, S. Cole, M. Alizon,
Cell 1985, 40, 9Ϫ17. [23b] L. Ratner, W. Haseltine, R. Patarca,
K. J. Livak, B. Starcich, S. F. Josephs, E. R. Doran, J. A. Rafal-
ski, E. A. Whitehorn, K. Baumeister, L. Ivanoff, S. R. Petteway
Jr, M. L. Pearson, J. A. Lautenberger, T. S. Papas, J. Ghrayeb,
N. T. Chang, R. C. Gallo, F. Wong-Staal, Nature 1985, 313,
[29]
B. Honig, A. Nicholls, Science 1995, 268, 1144Ϫ1149.
30] [30a]
[
[
H. Guo, N. Gresh, B. P. Roques, D. R. Salahub, J. Phys.
Chem. B 2000, 104, 9746Ϫ9754. [
30b]
N. Gresh, J. Sponer, J.
Phys. Chem. B 1999, 103, 11415Ϫ11427. [
30c]
G. Tiraboschi, B.
P. Roques, N. Gresh, J. Comput. Chem. 1999, 20, 1379Ϫ1390.
L. A. Carpino, H. Shroff, S. A. Triolo, M. E. Mansour, H.
Wenschuh, F. Albericio, Tetrahedron Lett. 1993, 34,
7829Ϫ7832.
31]
2
77Ϫ284.
[
24] [24a]
J.-B. Le Pecq, M. Le Bret, J. Barbet, B. P. Roques, Proc.
Natl. Acad. Sci. U. S. A. 1975, 72, 2915Ϫ2919. [
24b]
L. P. G.
[
[
[
32]
33]
34]
Wakelin, M. Romanos, T. K. Chen, D. Glaubiger, E. S. Canel-
L. A. Carpino, G. Y. Han, J. Am. Chem. Soc. 1970, 92,
lakis, M. J. Waring, Biochemistry 1978, 17, 5057Ϫ5063. [
24c]
R.
5748Ϫ5749.
G. Wright, L. P. G. Wakelin, A. Fieldes, R. M. Acheson, M. J.
C.-D. Chang, M. Waki, M. Ahmad, J. Meienhofer, E. O. Lun-
dell, J. D. Haug, Int. J. Pept. Protein Res. 1980, 15, 59Ϫ66.
Best yields were obtained recently for the synthesis of new bis
and tris-intercalating acridylbis-arginyl-porphyrins by using a
Pbf-protected arginine as synthon B.
Waring, Biochemistry 1980, 19, 5825Ϫ5836. [
24d]
H. D. King,
W. D. Wilson, E. J. Gabbay, Biochemistry 1982, 21, 4982Ϫ4989.
[24e]
G. J. Atwell, W. Leupin, S. J. Twigden, W. A. Denny, J.
Am. Chem. Soc. 1983, 105, 2913Ϫ2914. [
24f]
J. B. Hansen, T.
[35] [35a]
Koch, O. Buchardt, P. E. Nielsen, B. Nord e´ n, M. Wirth, J.
M. Wirth, O. Buchardt, T. Koch, P. E. Nielsen, B. Nord e´ n,
Chem. Soc., Chem. Commun. 1984, 509Ϫ511. [
24g]
B. Gaugain,
J. Am. Chem. Soc. 1988, 110, 932Ϫ939.
[35b]
E. Gimenez-Ar-
J. Markovits, J.-B. Le Pecq, B. P. Roques, FEBS Lett. 1984,
nau, S. Missailidis, M. F. G. Stevens, Anti-Cancer Drug Des.
1998, 13, 125Ϫ143 and 431Ϫ451.
[36] [36a]
[
24h]
169, 123Ϫ126.
P. Laug aˆ a, J. Markovits, A. Delbarre, J.-B.
[24i]
Le Pecq, B. P. Roques, Biochemistry 1985, 24, 5567Ϫ5575.
W. Keller, Proc. Natl. Acad. Sci. U. S. A. 1975, 72,
4876Ϫ4880. [
Biochemistry 1981, 20, 3598Ϫ3603.
36b]
M. Duguet, C. Bonne, A.-M. de Recondo,
M. Wirth, O. Buchardt, T. Koch, P. E. Nielsen, B. Nord e´ n, J.
Am. Chem. Soc. 1988, 110, 932Ϫ939. [ J. Markovits, C. Gar-
24j]
[
[
37]
38]
bay-Jaureguiberry, B. P. Roques, J.-B. Le Pecq, Eur. J. Biochem.
R. Lavery, K. Zakrzewska, H. Sklenar, Comput. Phys. Com-
mun. 1995, 91, 135Ϫ158.
C. R. Cantor, M. M. Warshaw, H. Shapiro, Biopolymers 1970,
24k]
1
989, 180, 359Ϫ366. [
S. C. Zimmerman, C. R. Lamberson,
M. Cory, T. A. Fairley, J. Am. Chem. Soc. 1989, 111,
6
805Ϫ6809.
9, 1059Ϫ1077.
[
25] [25a]
[39]
[40]
J. W. Lown, S. M. Sondhi, C.-W. Ong, Biochemistry 1986,
W. Muller, D. M. Crothers, J. Mol. Biol. 1968, 35, 251Ϫ290.
J. Sambrook, E. F. Fritsch, T. Maniatis, Molecular Cloning, a
Laboratory Manual, 2nd ed., Cold Spring Harbor Laboratory
Press, Plainview, NY, 1989.
[
25b]
2
5, 5111Ϫ5117.
G. Karagianis, J. A. Reiss, R. Marchesini,
M. De Cesare, F. Zunino, D. R. Phillips, Anti-Cancer Drug
Des. 1996, 11, 205Ϫ220. [
25c]
C. Drexler, M. W. Hosseini, G.
[25d]
[41]
Pratviel, B. Meunier, Chem. Commun. 1998, 1343Ϫ1344.
Y. Ishikawa, A. Yamashita, T. Uno, Chem. Pharm. Bull. 2001,
A. Kossanyi, B. Mestre, M. Perr e´ e-Fauvet, Synth. Commun.
1999, 29, 4341Ϫ4346.
49, 287Ϫ293.
Received May 20, 2003
Eur. J. Org. Chem. 2004, 1781Ϫ1797
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1797