Vol. 182, No. 2, 1992 January 31, 1992
RESEARCH COMMUNICATIONS Pages 588-592
CLEAVAGE OF PLASMID DNA BY [email protected]
), [email protected]
) AND Co(Hi) DESFERAL COMPLEXES Rajendra R. Joshi and K. Nagappa Ganesh’
Biorganic Chemistry Unit, Division of Organic Chemistry, National Chemical Laboratory, pbona-411cK)8, India Received
Summary: It is demonstrated that the Cu(II), Co(II1) and Ni(I1) complexes of a siderophore chelating drug desferal cleave DNA, in contrast to the corresponding [email protected]
) complex which does not bring about DNA scission. Hydroxy radical scavengers inhibit the cleavage reaction. 0 19'32 Academic Press, Inc.
Interaction of metal complexes with nucleic acids is currently attracting wide attention due to their potential utility as drugs, regulators of gene expression and tools for molecular biology’“. Many metal complexes exhibit nucleolytic activity, the most important examples being [email protected]
)-OP’, Fe(E)-BLM3, Fe(E)-EIYTA4,metalloporphyrins5, Ru and Co complexes of 4,7-diphenyl-1, lo-phenanthroline6 and more recently by Ni(II) complexes7p8. These cleave DNA either by oxidative degradation of deoxyribose moiety or by base modification and such chemical methods of nicking DNA have found useful applications in probing sequence dependant conformational variability of DNA9, identifying ligand/protein binding sites on DNA” and in the design of artificial sequence-specific nucleases4*“. Desferal 1, a well known siderophore12 and a highly effective drug in chelation therapy of iron overload diseases, forms a stable octahedral co-ordination Fe(II1) complex Ferrioxamine B 2. In contrast to Fe-EDTA, 2, cannot undergo a red-ox cycling, thus preventing iron-catalysed hydroxyl radical formation13 which is a useful property for its clinical application. Indeed, desferal is employed to arrest hydroxyl radical production in DNA scission reactions caused by Fe(I1) complexes”. In this communication, we demonstrate that while the Fe(m) complex of desferal is passive in DNA scission, the corresponding Cu(I1) 5, Co(II1) 4 and Ni(I1) 5 complexes of desferal, actively cleave DNA, similar to other metallonucleases. should be addressed. . . : EDTA = Ethylenediamine tetraacetic acid; OP = 1,lO Abb evr&w pheknthroline BLM = Bleom tin; Des = Desferal; ME = 2-mercaptoethanol. NCL Communkation number 5I 03. *To whom correspondence
0006-291x/92 Copyright All rights
6 1992 by Academic Press, of reproduction in any form
182, No. 2, 1992
M = Cu";
M = CO' II
Scheme 1 MATERIALS
Desferal was obtained as a generous gift from Hindustan Ciba Geigy. The Cu(II) 3, Co(III) 4, and Ni(II) 5, complexes were synthesised, by stirring desferal (1 eq) separately with 10 eq of CuCl , CoCl and NiC in $0. The complexes were purified by chromatography over Sephadex bl0 and tz eir homogeneity checked by HPLC. Column: Cl8 Reverse phase; Solvent system: 30% CH CN in Trieth lammonium acetate buffer (O.O5M), pH 7.0; 2, R = 5.80, Xmax = 439 nm (E = 2 ii 00); 5, R = 5.97, Xmax = 427 nm (E = 1200); 4,k = 6.10, Xmax = 511 nm (E = 268); 5, R’ = 6.55., Xmax = 394 nm. Among the comblexes Z-5, only 2 showed a reversible signal in cyclic voltammogram (range: +0.6V to -0.5V) corresponding to an E” of +0.24V. The DNA cleavage reactions were carried out on double strand3 plasmid DNA pBR322 (> 90 % form-I, Bangalore Genei) by incubation for 30 min at 37°C with various reagents (see legend Figure 1). It was followed by addition of bromophenol blue and loaded directly into different wells on a 1% agarose gel for analysis by electrophoresis at lOOV, 25mA, till the dye reached about 75 % of the gel length. The gel was stained with ethidium bromide for visualisation on a transilluminator followed by photography. RESULTS AND DISCUSSION As can be seen from the results of Figure 1, Fe(III)-desferal 2 does not cleave the plasmid DNA under any conditions (lane 3), while the respective Cu(I1) 3 (lane 4), [email protected]
) 4 (lane 5) and Ni(II) 5 (lane 6) complexes cleave the plasmid to produce form-II DNA, similar to that observed with Cu(II)-OP (lane 7). The cleavage reaction with Cu(I1) complex requires a reducing agent such as 2-mercaptoethanol (ME), dithiothreitol or ascorbate and the reaction is made more efficient by addition of I-$Os. The cleavage reactions with Co(III) and Ni(II) complexes proceed even in the absence of a reducing agent. The cleavage efficiency is dependent on metal ion concentration, the optimal concentrations for 100% cleavage being 235pm, 42.5pM and 1OpM for [email protected]
), [email protected]
) and Ni(II) complexes respectively. Among the three, the Ni(II) complex showed maximum efficiency at low concentrations and the excess complex did not further degrade the DNA to the linear form. The main product of the scission reactions with Cu(I1) and Ni(I1) is the randomly nicked form-II DNA. In case of Co(III), a slight increase in concentration over the optimal value led to extensive degradations, 589
Figure 1. Cleavage of pBR322 with various metal complexes. The reaction mixtures contained (in a total volume of 20~1) 25mM Tris-borate, pH 7.8, 2.5 mM sodium acetate and plasmid pBR322 (final concentration 140pM) with the following additions: & 1, none; & 2, H 0 (4mM) + ME (O.SmM); & 3, H 0 (4mM) + ME (0.5mM) + Fe(III)-Des t8jgM); lane 4, H,O, (4mM) + ME (0.3mM) +Cu(II)-Des (255,~M); lane 5, H,O, (4mM) + Co(III)-Des (42.5pM); lane 6, H 0, (4mM) + Ni(II)-Des (10pM); & 2, $02 (4mM) + ME (0.5mM) + Cu(II)-OP [email protected]
resulting in disappearence of bands on agarose gel. The cleavage seems to be specific for double stranded DNA, since the single strand plasmid M13mp18 did not show any scission with various metal complexes. Free desferal in the presence of either an oxidising or a reducing agent or both did not exhibit any DNA cleavage (not shown). Incubation of plasmid with Cu(II)-desferal plus ME in the presence of hydroxyl radical scavengers such as mannitol, glycerol, NaN, and catalase significantly inhibited (>75%) the cleavage, while boiled catalase was ineffective in the inhibition reaction. Recently Cu(II)-thiol induced chemical cleavage of plasmid DNA has been reported14 and the suggested mechanistic pathway involves the generation of either a reactive intermediate (probably HO) in free solution or a cleavage species which is DNAbound. DNA scission by Cu(II)-desferal complex may perhaps follow a similar course, since the reaction specificity (red-ox reagent requirements and inhibitory effects) are identical. The non-production of hydroxy radicals and the consequent absence of DNA cleavage with Ferrioxamine B (2) has been previously attributed to the unavailability of a vacant coordination site on the metal for binding a $0 molecule13. The requirement of a reducing agent in case of Cu(I1) complex indicates that the metal reduction [Cu(II)+Cu(I)] is a primary requisite for hydroxyl radical generation. The observed selective cleavage of DNA using Ni complexes of either a tripeptide’ or tetraazamacrocycles in the presence of extraneous oxygen atom donors, has been attributed to oxidative modifications of base residues on DNA. Mn(II1) and Mn(IV) complexes of desferal have recently been shown’” to possess superoxide dismutase activity+. In +The reversible signal in cyclic voltammogram (Earn, +0.24V) observed presently for the Cu complex 3 indicates red-ox process which is metal centered rather than originating from the ligand oxidation. In contrast the Mn(IV) complex” exhibited an irreversible signal (Eoln, -0.35V) corresponding to ligand oxidation.
conjunction with present work, it is therefore interesting that depending on the nature of the bound metal, the desferal metallo-complexes exhibit a variety of chemical activities ranging from passivity in Fenton reaction (Fe complex) to production of hydroxy radicals (Cu complex) and as a mimic for superoxide dismutase activity (Mn complexes). It has been reported16 that desferal in presence of ascorbate may act as a prooxidant in interfering with the activity of some oxidative enzymes and alkaline phosphatase. Also relevant are the clinical findings of increased toxicity when desferal is supplemented with ascorbate during therapy l7 . It may perhaps be suggested that the oxidative damage of enzyme systems in presence of a reducing agent such as ascorbate and the accompanying toxicity may originate from complexes of desferal with trace metals [e.g., Cu(II)] present in the system. This communication is the first report of oxidative damage of DNA by metal complexes (Cu, Co and Ni) of desferal. Desferal 1,, possesses three hydroxamic and two peptide groups which are potential interaction sites with DNA and a terminal NH, group useful for linking chemical moieties such as an intercalator or a groove binder of nucleic acids to modulate the cleavage properties. Further studies in this direction and investigation of the accompanying sequence specificity and mechanistic aspects of cleavage reactions are in progress.
RRJ thanks University Grants Commission (UGC) India for award of a research fellowship. Acknowledgment:
1. For recent reviews, see‘IUlius T.D. ,(1989) Ed. “Metal-DNAChemistry”ACS Symposium Series, No.402, American Chemical Society, Washington, DC. 2. Sigman D.S.(1986), Act. Chem.Res.,19, 180-186. 3. (a) Hecht S.M. (1986) Act. Chem.Res.& 383-387.(b) Stubbe J., Kozarich J.W. (1987) Chem.Rev.,82, 1107-1136. 4. Dervan P.B. (1986) Science,Z , 464-471. 5. (a) Le Doan T., Perrouault L., Helene C., Chassignol M., Thoung N. J. (1986) Biochemistry,3, 6736-6739. (b) Wood B., Skorobogaty A., Dabrowiak J.C. (1987)
Biochemistry,25, 68756883. 6. (a) Barton J.K. ,(1986) Science,Z,
727-734.(b) Basile L.A., Barton J.K. (1987) J. Am. C&m. Sot., me, 7548-7550. (c) Basile L.A., Raphael A.L. and Barton J.K. (1987) J. Am. C%em.Sot., U& 7550-7551. 7. Mack D.P., Dervan P.B. ,(1990) .7. Am. @em. Sot., 112,4604-4606. 8.5;r6n X., Rokita S.E. and Burrows C.J. (1991), J. Am. Chem.Sot., 113,58849. (a) Burkhoff A.M., Tullius T.D. (1987) Cell, 48, 935-943 (h) Yoon C., Kuwabara M.D., Law R., Wall R., Sigman D.S. (1988) J. Biol. Chem.,Z, 8458-8463. 10. (a) Tullius T.D. (1989) Ann. Rev.Biophys.BiophysChem.,18, 213-237. (b) ‘Bdlius T.D., Dombroski B.A., Churchill M.E.A., Kam K. (1988) Methodsin Enzymology,Vol.155, Ed. Wu R., 537-558. 11. (a) Hertzberg R.P. and Dervan PB. (1984) Biochemistry,23, 3934-3945. (b) Strobe1 S.A., Moser H.E., Dervan P.B. (1988) .7. Am. Chem.Sot., 110, 7927-
182, No. 2, 1992
12. (a) Keberle H. (1964) Ann. N. Z Acad. Sci, m, 758-768. (b) Raymond K.N., Muller Cf., Matzamke B.F. (1984) in “Topics in current chemistry “, Vol. 123, p.50102, Spmger-Verlag. (c) Martell A. E., Anderson W.F., Badman D.G. (1981) (Eds): Development of Zron chelators for cZinicaZwe, Elsevier Holland, New York. (d) Dobbin P.S., Hider R.C. (1990) Chem. Brit., & 565-568. 13. Graf E., Mahoney J.R., Bryant R.G. and Eaton J.W. (1984) J. Biol. Chem., U, 3620-3624. 14. Reed C.J. and Douglas K.T. (1991) Biochem. .I., 275, 601-608. 15. (a) Rush J.D.,Maskos Z and Koppenol W.H. (1991) Arch. Biochem. Biophys., 2&$215-219. (b) Hahn S.M., Krishna C.M., Samuni A., Mitchell J.B., Russo A. (1991) Arch. Biochem. Biophys., 288, 215-219. 16. Mordente A., Mencis E., Miggiano G.A.D., Martorana G.E. (1990) Arch. Biochem. Biophysics, 277 , 234-240. 17. (a) Davies S. C., Hungarford J.L., Arden G.B., Marcus R.E., Miller M.H., Huchins E.R. (1983) Lancet, 181-184. (b) Nieuhaus A.W. (1981) N. EngZ. J. Med., jQ& 170-174.