Photochemical DNA cleavage by a Berenil analog

Article abstract of DOI:10.1016/j.tetlet.2003.08.002

Berenil [bis(4-amidinophenyl)1,3-triazene] is a photostable DNA-binding ligand. We describe here the synthesis of N-(3-hydroxypropyl)-Berenil, which in contrast to Berenil is photosensitive to 360 nm irradiation, behaving as a caged diazonium salt. The 4-amidinobenzenediazonium fragment produced by photolysis induces DNA modification and cleavage.

Full text of DOI:10.1016/j.tetlet.2003.08.002

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 7307–7309
Photochemical DNA cleavage by a Berenil analog
Sally J. Burr, Abdulghani Mselati and Emrys W. Thomas*
Division of Biological Sciences, Salford University, Salford M5 4WT, UK
Received 6 May 2003; revised 22 July 2003; accepted 1 August 2003
Abstract—Berenil [bis(4-amidinophenyl)1,3-triazene] is a photostable DNA-binding ligand. We describe here the synthesis of
N-(3-hydroxypropyl)-Berenil, which in contrast to Berenil is photosensitive to 360 nm irradiation, behaving as a caged diazonium
salt. The 4-amidinobenzenediazonium fragment produced by photolysis induces DNA modification and cleavage.
© 2003 Elsevier Ltd. All rights reserved.
A wide variety of compounds1 are known to act as
DNA and RNA photocleavage agents: interest in such
compounds lies in their potential as specific nucleases
and as phototherapeutic agents. Berenil [bis(4-
amidinophenyl)-1,3-triazene] is a paradigmatic minor-
groove binder interacting with DNA at AT-rich sites.2
During an investigation of a series of Berenil analogues
incorporating alkyl groups on the triazene nitrogen, it
was noticed that several compounds were unstable in
daylight; these compounds also showed light-dependent
activity in plasmid relaxation assays for DNA cleavage.
This report describes the photochemical and DNA-
cleaving properties of N-(3-hydroxpropyl)-Berenil as a
representative active compound in the series.
N-(3-Hydroxypropyl)-Berenil 1 was synthesised3 as out-
lined in Scheme 1 (umax 366 nm, m=20,700 M−1 cm−1
(H2O)).
Scheme 1.
* Corresponding author. E-mail: e.w.thomas@salford.ac.uk
0040-4039/$ - see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2003.08.002

7308
S. J. Burr et al. / Tetrahedron Letters 44 (2003) 7307–7309
The light-dependent plasmid relaxing ability of 1
appeared to be due in part to the initial photoproduct
2: plasmid relaxation was observed in a dark reaction
using 2 as prepared by diazotisation of 4-amidinobenz-
eneamine (Fig. 1, lanes 5 and 6). Similar results were
obtained in dark reactions using pre-irradiated 1 (result
not shown). Plasmid relaxation did not require an
added electron donor (mediating homolytic dediazonia-
tion of the diazonium group) as shown for other diazo-
nium reagents.6 With broadband 350 nm irradiation,
however, an enhanced DNA cleavage by 2 (relative to
the dark reaction) was observed (Fig. 1, cf. lanes 5 and
6 with lanes 7 and 8). Photolytic dediazoniation of
diazonium salts is well documented7 and could generate
carbocationic species with DNA cleavage potential.
Footprinting and binding studies (results not shown)
revealed that 1 showed a strong preference for AT-rich
binding sites in DNA, with an affinity similar to that of
Berenil itself. In contrast to Berenil, however, 1 proved
to be photosensitive† to 350 nm irradiation, with rapid
bleaching of the 366 nm absorption band in aqueous
solution. Actinometry showed a modest (ꢀ0.04) quan-
tum yield for photodecomposition: a number of other
analogues (including N-methyl- and N-(2-hydroxy-
ethyl-) had quantum yields in the same range.
Product analyses showed that the initial photoproducts
from 1 were the 4-amidinobenzenediazonium dication 2
and 4-amidino-N-(3-hydroxypropyl)aniline:
1
can
therefore be regarded as a caged diazonium salt. Pro-
longed 350 nm irradiation led to photolysis of the
initial photoproduct 2, presumably resulting from red
edge absorption (2, umax 257 nm, m=15,200 M−1 cm−1
(H2O), with m=20 M−1 cm−1 at 350 nm): 4-hydroxy-
benzamidine was identifiable as a major product. Pho-
tolysis of N-alkylated diaryl triazenes in non-polar
organic solvents, as reported previously,5 is dominated
by homolytic decomposition pathways. A heterolytic
photolysis pathway for diaryl triazenes (presumably
facilitated in polar media), as observed here, has not
been reported to date.
The dark reaction of 2 with double-stranded and single-
stranded 5%[32P]-labelled oligonucleotides is shown in
Figure 2(A/B).
The appearance of slower-moving bands (see lanes 1, 2,
5, and 6) most likely indicates the formation of covalent
adducts between 2 (a relatively electrophilic diazonium
salt) and DNA. Some spontaneous chain cleavage takes
place (lanes 1 and 5) which is enhanced by heat treat-
ment (lanes 2 and 6). Significantly, DNA modification
at pH 7.5 differs from that at pH 5.5 in that specific
chain cleavage at purines is apparent (cf. lanes 2 with
6). Detailed studies on the chemistry of DNA modifica-
tion by 1 and 2 are in progress. In this context, the
covalent modification of DNA by electrophilic diazo-
nium reagents, involving both C- and N-coupling reac-
tions at purine residues, are well documented,8 as are
DNA photocleavage processes involving diazonium
salts.9
References
1. (a) Armitage, B. Chem. Rev. 1998, 98, 1171–1200; (b)
Fernandez, M.-J.; Grant, K. B.; Herraiz, F.; Yang, X.;
Lorente, A. Tetrahedron Lett. 2001, 42, 5701–5704.
2. (a) Brown, D. G.; Sanderson, M. R.; Skelly, J. V.; Jenkins,
T. C.; Brown, T.; Garman, E.; Stuart, D. I.; Neidle, S.
EMBO J. 1990, 9, 1329–1334; (b) Brown, D. G.; Sander-
son, M. R.; Garman, E.; Neidle, S. J. Mol. Biol. 1992, 226,
481–490.
3. Compound 1 was isolated as the dihydrochloride salt, mp
210°C. MS (ES): m/z 340.35 (M++1) (calcd for C17H21N7O
339.40).The synthetic route follows that described for N-
methyl-Berenil (Ashley, J. N.; Berg, S. S. J. Chem. Soc.
1957, 3089–3093).
Figure 1. Plasmid relaxation: reaction mixtures contained
pBR322 (0.5 mg) in 0.02 M sodium acetate–0.005 M EDTA
buffer, and compounds 1 and 2 at the concentrations indi-
cated below. All reaction times were 5 min at 4°C, followed
by quenching of excess diazonium by vortexing samples with
an equal volume of butanol containing dimethylaniline (0.1%
w/v), centrifugation, and rejection of the butanol phase. The
quenched aqueous phases were extracted twice with diethyl
ether prior to electrophoresis in 1% agarose gels. Lane (1) 0.5
mg DNA+hw control; lane (2) 0.5 mg DNA+50 mM 1 dark;
lane (3) 0.5 mg DNA+50 mM 1+hw; lane (4) 0.5 mg DNA+5
mM 1+hw; lane (5) 0.5 mg DNA+50 mM 2, dark; lane (6) 0.5
mg DNA+5 mM 2,dark; lane (7) 0.5 mg DNA+50 mM 2+hw;
lane (8) 0.5 mg DNA+5 mM 2+hw.
4. Heller, H. G.; Langan, J. R. J. Chem. Soc., Perkin Trans.
2 1981, 341–343.
5. Julliard, M.; Vernin, G.; Metzger, J. Helv. Chim. Acta
1980, 63, 467–472.
6. (a) Griffiths, J.; Murphy, J. A. J. Chem. Soc., Chem.
Commun. 1992, 24–26; (b) Arya, D. P.; Warner, P. M.;
Jebaratnum, D. Tetrahedron Lett. 1993, 34, 7823–7826; (c)
Arya, D. P.; Jebaratnum, D. Tetrahedron Lett. 1995, 36,
4369–4372.
7. Gasper, S. M.; Devadoss, C.; Schuster, G. B. J. Am.
Chem. Soc. 1995, 117, 5206–5211.
† Photolyses were conducted with a 350 nm broadband UV source
calibrated by actinometry using Aberchrome.4 Product analyses
were by HPLC (reverse phase C18 silica column; 0.1 M sodium
dodecyl sulphate, 13% acetonitrile, 0.06 M phosphoric acid for
elution; detection at 206 nm).

S. J. Burr et al. / Tetrahedron Letters 44 (2003) 7307–7309
7309
Figure 2. Dark reactions of 2 with: (dCdGdCdGdAdAdTdTdCdGdCdG)2 (dsDNA; panel A) and dAdGdAdCdAdCdAdG-
dAdAdTdC (ssDNA; panel B). Oligos were 5%end-labelled with 32P, and treated with 2 (0.0005 M) in either 0.02 M sodium
acetate–0.005 M EDTA (pH 5.5), or in 0.05 M sodium cacodylate–0.005 M EDTA (pH 7.5). Reactions were terminated by
vortexing the samples with an equal volume of butanol containing 0.1% v/v dimethylaniline, centrifugation, and rejection of the
butanol layer. The quenched aqueous phases were extracted twice with diethyl ether prior to electrophoresis in 20% denaturing
polyacrylamide gels. Panel A: lane (1) ds DNA+0.5 mM compound 2, pH 5.5, 3 h 20°C; lane (2) as lane 1 but heated at 90°C
for 15 min after quenching; lane (3) ds DNA control; lane (4) ds DNA control heated at 90°C 15 min; lane (5) ds DNA+0.5 mM
2, pH 7.5, 3 h 20°C; lane (6) as lane 5 but heated at 90°C for 15 min after quenching; lane (7) ds DNA control; lane (8) ssDNA
control, heated at 90°C for 15 min at pH 7.5. Panel B: lane (1) ss DNA+0.5 mM compound 2, pH 5.5, 3 h 20°C; lane (2) as lane
1 but heated at 90°C for 15 min after quenching; lane (3) ss DNA control; lane (4) ss DNA control heated at 90°C for 15 min;
lane (5) ssDNA+0.5 mM 2, pH 7.5, 3 h 20°C; lane (6) as lane 5 but heated at 90°C for 15 min after quenching; lane (7) ss DNA
control; lane (8) ssDNA control, heated at 90°C for 15 min at pH 7.5. Arrows in both panels indicate cleavage products.
8. (a) Chin, A.; Hung, M.-H.; Stock, L. M. J. Org. Chem.
1981, 46, 2203–2207; (b) Hung, M.-H.; Stock, L. M. J.
Org. Chem. 1982, 47, 448–453; (c) Gannett, P. M.; Powell,
J. H.; Rao, R.; Shi, X.; Lawson, T.; Kolar, C.; Toth, B.
Chem. Res. Toxicol. 1999, 12, 297–304.
9. (a) Behr, J.-P. J. Chem. Soc., Chem. Commun. 1989,
101–103; (b) Garcia, A.; Giege, R.; Behr, J.-P. Nucleic
Acids Res. 1990, 18, 89–95; (c) Schmid, N.; Behr, J.-P.
Biochemistry 1991, 30, 4357–4361.