Chemical mechanism of DNA scission by (1,10-phenanthroline)copper. Carbonyl oxygen of 5-methylenefuranone is derived from water

Full text of DOI:10.1021/ja962409n

J. Am. Chem. Soc. 1997, 119, 1135-1136
1135
Chemical Mechanism of DNA Scission by
1,10-Phenanthroline)copper. Carbonyl Oxygen of
-Methylenefuranone Is Derived from Water
(
5
Michael M. Meijler, Ottilie Zelenko, and David S. Sigman*
Department of Biological Chemistry, School of Medicine
Department of Chemistry and Biochemistry
Molecular Biology Institute
UniVersity of California, Los Angeles
Los Angeles, California 90095-1570
ReceiVed July 15, 1996
The tetrahedral 2:1 (1,10-phenanthroline)copper(I) complex
(OP)2Cu ), cleaves the phosphodiester backbone of B-DNA
+
(
Figure 1. Mass spectra of 5-methylenefuranone (5-MF) isolated from
reactions containing 18O-H
O and H
Middle panel: 5-MF isolated from an anaerobic reaction mixture
2
2 2
O . Left panel: authentic 5-MF.
under physiological conditions by oxidation of the deoxyribose
moiety using hydrogen peroxide as an essential coreactant (eq
1
8
16
1,2
containing 22% H
2 2 2
O and 1 mM H O . Right panel: 5-MF isolated
1
). The principal products are free base, 5-methylenefuranone
16
from a reaction mixture containing H
2
18O
and H O.
2 2
2
+
generated in situ by the (OP)2Cu -catalyzed oxidation of MPA
in the presence of O2 or included at a concentration of 1 mM
when required. After a 15 min reaction at 25 °C, the reactants
were heated at 90 °C for another 15 min to insure complete
(1)
3
4
(
5-MF), and 3′- and 5′-phosphorylated ends (eq 2). In this
paper, we investigate the atom source of the carbonyl oxygen
of 5-MF to elucidate the mechanism of this efficient nucleolytic
5
conversion of the metastable intermediate at the 3′ end to 5-MF,
which was isolated as indicated below. Anaerobic reactions
6
1
8
activity. Using O-enriched hydrogen peroxide and water, we
find that the carbonyl oxygen is derived from H2O.
were performed in glass tubes, sealed with a rubber septum,
and purged with argon. Following the addition of reactants via
a syringe, the reaction was purged continuously for 30 min prior
6
to heating the tube at 90 °C and isolation of 5-MF. As
3
anticipated from our previous studies, we isolated 5-MF from
2+
reaction mixtures containing OP, Cu , and MPA in the presence
of oxygen. Under these conditions, H2O2 is generated in situ,
although our present studies indicate that addition of 1 mM H2O2
generates slightly higher yields. Under anaerobic conditions,
the coreactant H2O2 must be added to achieve scission. The
simultaneous presence of OP, copper ion, and reducing agent
(2)
(
e.g., MPA) is essential for the isolation of 5-MF under both
anaerobic and aerobic conditions.
The oxygen source of the carbonyl in 5-MF was examined
Poly dA-T was used as the substrate to determine the source
of the carbonyl oxygen in 5-MF because of its ready availability
and the ease of preparing internally labeled phosphodiester
by performing the reaction under (a) anaerobic conditions in
the presence of 1 mM H 18O , (b) anaerobic conditions using
2
2
22% H 18
16
2
O and 1 mM H2 O2, (c) aerobic conditions in the
3
2
18
bonds using Escherichia coli DNA polymerase I and (R- P)-
presence of 20 and 30% H2 O, and (d) aerobic conditions with
dATP. In our initial demonstration of 5-MF as the product of
18O as the source of molecular oxygen. Since the carbonyl
2
3
60
DNA scission, superoxide generated by Co irradiation was
used as the source of reducing equivalents to activate the
nuclease activity in order to avoid possible addition of nucleo-
philic reducing agents, such as 3-mercaptopropionic acid (MPA),
to 5-MF. Since this precaution was unnecessary in the present
studies, activation by MPA was used because it is more efficient
and permits the use of small (1 mL) reaction volumes. Control
experiments have demonstrated that 5-MF is stable (less than
group of 5-MF is not exchangeable with solvent, the possibility
that oxygen is incorporated during the isolation of the product
can be excluded.7
In Figure 1, the mass spectra of synthetic
-MF (left panel) is compared to that of 5-MF isolated from an
5
18
anaerobic reaction mixture containing 22% H2 O and 1 mM
16
H2 O2 (middle panel) and to that of 5-MF isolated from an
18
anaerobic reaction mixture containing 1 mM H2 O2(right panel).
18
The results clearly indicate that H2 O is the exclusive source
1
0% loss observed) in the presence of MPA at 25 and 90 °C
(
5) Kuwabara, M.; Yoon, C.; Goyne, T. E.; Thederahn, T.; Sigman, D.
S. Biochemistry 1986, 25, 7401-7408.
6) After the sample was cooled to room temperature, the 5-MF was
for up to 1 h.
Aerobic reactions were initiated by the addition of a freshly
(
2
+
prepared solution of OP (1,10-phenanthroline) and Cu to a
solution of 20 µg of poly dA-T in pH 7.5 Tris-HCl (50 mM),
NaCl (25 mM), and MPA (1 mM) to a final concentration of
extracted into dichloromethane, concentrated, and dissolved either in 10%
acetonitrile for HPLC analysis or acetone for GC/MS analysis. The
dichloromethane soluble reaction products were separated on a Phenomex
Nucleosil C-18 column (5 µM, 4.6 × 0.250 mm) and detected at 260 nm
using a Waters LC spectrophotometer. The mobile phase consisted of 14%
acetonitrile and 86% water; results were plotted with a Shimadzu Chro-
matopac integrator. Calibration of the HPLC analyses from 200 nM to 2
µM was accomplished using authentic 5-MF prepared according to
Grundmann and Kober.17 The GC-MS system was composed of a Finnigan
9610 capillary GC fitted with a J & W DB-5MS 30 m × 0.28 mm i.d.
fused silica capillary column interfaced directly to a Finnigan 4000 with
an Incos 2300 data system. Fragments generated by electron impact
ionization at 70 eV were scanned from 38 to 500 m/z in 1 s intervals. The
source temperature was 250 °C.
1
35 µM OP and 30 µM CuSO4. Hydrogen peroxide was
*
Address correspondence to this author at the Molecular Biology
Institute.
(
1) Sigman, D. S.; Chen, C.-h. B. Ann. ReV. Biochem. 1990, 59, 207-
2
2
2
36.
(
2) Sigman, D. S.; Mazumder, A.; Perrin, D. M. Chem. ReV. 1993, 93,
295-2316.
(3) Goyne, T. E.; Sigman, D. S. J. Am. Chem. Soc. 1987, 109, 2846-
848.
4) Pope, L. M.; Reich, K. A.; Graham, D. R.; Sigman, D. S. J. Biol.
Chem. 1982, 257, 12121-12128.
(
(7) Pitie, M.; Bernadou, J.; Meunier, B. J. Am .Chem. Soc. 1995, 117,
2935-2936.
S0002-7863(96)02409-2 CCC: $14.00 © 1997 American Chemical Society

1136 J. Am. Chem. Soc., Vol. 119, No. 5, 1997
Communications to the Editor
Scheme 1
(OP)2Cu+ by hydrogen peroxide and denoted as a resonance
hybrid of a copper(II)-hydroxyl radical and a putative
copper(III)-oxo species generates a deoxyribose-centered radi-
cal by hydrogen abstraction (drawn here C-1 localized) (a, step
i) . The DNA bound cupric ion then oxidizes a to form the
carbocation (b, step ii) which is stabilized by partial double-
bond formation from the furanose oxygen. Species c, formed
by loss of the C-2 proton (stereochemistry unknown, step iii),
eliminates the 3′-phosphate resulting in strand scission, a new
5
′-phosphorylated terminus and resonance-stabilized d (step iv).
Attack by water on d leads to the observed incorporation of
18
O into the C-1 position of the deoxyribose (step v). Elimina-
tion of the purine/pyrimidine from d would lead to e (step vi),
which has properties consistent with a metastable intermediate
trapped in previous studies that decomposes into the new 3′-
1
8
phosphorylated end and the O-labeled 5-methylenefuranone
3
,5,11
(step vii).
An important feature of the mechanism of Scheme 1 is that
the phosphodiester backbone is cleaved prior to attack of water.
This avoids the formation of a 2-deoxyribonolactone prior to
strand scission. This intermediate has been proposed both in
7,9
the scission of DNA by a manganese porphyrin or ene-
1
2,13
diynes
and in the photoreaction of a deoxyoligonucleotide
1
4
containing 2′-iododeoxyuridine. It is stable in the absence of
heat or base treatment, two steps which are not necessary for
of the carbonyl oxygen of 5-MF. This conclusion is further
reinforced by the finding that 5-MF isolated from aerobic
18
reaction mixtures containing 20 and 30% H2 O is enriched in
1
8
O to the corresponding extent (condition c above) and the
finding that 5-MF isolated from a reaction mixture containing
1
8
O2 as the source of molecular oxygen (condition d above) is
1
8
not O enriched.
-MF is also the product of the oxidative degradation of DNA
5
by the manganese complex of the cationic porphyrin derivative
meso-tetrakis(4-N-methylpyridiniumyl)porphyrin using the water-
soluble potassium monopersulfate as oxidant.7,8 In this case,
the carbonyl oxygen in 5-MF is derived from the monopersulfate
and not H2 O, consistent with the reaction proceeding by an
oxygen rebound mechanism.9 In addition to the different source
of oxygen of the carbonyl group of 5-MF, the reactions of (1,-
the cleavage of the phosphodiester backbone by (OP)2Cu+.
Although the present study has focused on the products formed
by the tetrahedral (OP)2Cu , the results are probably applicable
+
18
to the targeted scission of DNA by (1,10-phenanthroline)copper
15,16
directed with proteins and oligonucleotides.
Acknowledgment. We thank Dr. Kym Faull of the Mass Spec-
trometry Facility of UCLA and Mr. Ed Ruth of the Environmental
Engineering Analytical Chemistry Laboratory, Department of Civil and
1
0-phenanthroline)copper and the manganese porphyrin with
DNA also differ because the latter chelate reacts with guanosine
of DNA when tethered to an oligonucleotide, generating alkaline
labile sites which require piperidine treatment for strand
scission.9 In contrast, the reaction of (1,10-phenanthroline)-
copper with the deoxyribose moiety of DNA leads to strand
scission without the requirement of either alkaline or heat
18
Environmental Engineering, UCLA, for assistance in the O analyses.
We acknowledge useful comments provided during the reviewing
process. This research was supported by USPHS GM 21199.
JA962409N
4
,10
treatment.
(
11) Sigman, D. S. Acc. Chem. Res. 1986, 19, 180-186.
Several possible reaction pathways are consistent with the
oxygen of water as the source of the carbonyl oxygen of 5-MF.
We favor the pathway presented in Scheme 1 in which the
copper-oxo species formed by the oxidation of the DNA bound
(12) Meschwitz, S. M.; Schultz, R. G.; Ashley, G. W.; Goldberg, I. H.
Biochemistry 1992, 31, 9117-9121.
(13) Goldberg, I. H.; Kappen, Y. J.; Xu, Y. J.; Stassinopolous, A.; Zeng,
X.; Xi, Z.; Yang, C. F. in DNA and RNA CleaVers and Chemotherapy of
Cancer and Viral Diseases; Meunier, B., Ed.; Kluwer Academic Publish-
ers: Dordrecht, Boston, London, 1996; Vol. Series C-Vol. 479, pp 1-22.
(14) Sugiyama, H.; Fujimoto, K.; Saito, I. J. Am. Chem. Soc. 1995, 117,
2945-2946.
(
8) Bernadou, J.; Pratviel, G.; Bennis, F.; Girardet, M.; Meunier, B.
Biochemistry 1989, 28, 7268-7275.
(
9) Bernadou, J.; Pratviel, G.; Meunier, B. in DNA and RNA CleaVers
(15) Sigman, D. S.; Bruice, T. W.; Mazumder, A.; Sutton, C. L. Acc.
Chem. Res. 1993, 26, 98-104.
and Chemotherapy of Cancer and Viral Diseases; Meunier, B., Ed.; Kluwer
Academic Publishers: Dordrecht, Boston, London, 1996; Vol. Series C-Vol.
(16) Pan, C. Q.; Landgraf, R.; Sigman, D. S. Mol. Microbiol. 1994, 12,
335-342.
4
79, pp 1-22.
(
Chem. 1979, 254, 12269-12272.
10) Sigman, D. S.; Graham, D. R.; D’Aurora, V.; Stern, A. M. J. Biol.
(17) Grundmann, C.; Kober, E. J. Am. Chem. Soc. 1955, 77, 2332-
2333.