ELSEVIER

Fundamental and Molecular Mechanisms of Mutagenesis

Mutation Research 357 (1996) 83-88

Oxygen is not required for degradation of DNA by glutathione and Cut II)

Aparna Jain a, Nasir K. Alvi a, J.H. Parish b, S.M. Hadi a3* a Department of Biochemistry, Faculty of Life Sciences, Aligarh Muslim Unirersity. Aligarh-202 002, India

b Department of Biochemistry and Molecular Biology, Unioersity of Leeds, Leeds LS2 WT. UK

Received 7 February 1996; revised 10 April 1996: accepted 12 April 1996

Abstract

Previous studies by others have shown that thiols, such as glutathione, cause cleavage of DNA in the presence of Cu(II) ions and that the hydroxyl radical derived from molecular oxygen is the major cleaving species. In this paper, we present several lines of evidence that strongly suggest that molecular oxygen is not essential for DNA cleavage and that thiyl radicals may also be involved. Indirect evidence is presented to indicate that glutathione may substitute oxygen as an electron acceptor. In addition, DNA degradation occurs to a significant extent under anaerobic conditions and no inhibition of single-strand cleavage of supercoiled plasmid DNA is seen in the presence of superoxide dismutase and catalase. In view of the ubiquitous presence of glutathione, these results could be of interest under certain diseased conditions where copper concentrations are elevated.

Keywords; Glutathione; DNA cleavage; Thiyl radical

1. Introduction

Glutathione (GSH; y-glutamylcysteinyl glycine) is a universal reducing agent found in all cells. It is present in cells at a concentration of up to 5 mM and provides a strong reducing environment [l]. It is generally considered that reducing compounds (H- donors) behave as antioxidants against free radical damage. Prutz [2] has shown that such compounds can also act as pro-oxidants in combination with Cu(I1). In such reactions, C&I) is reduced to Cu(1) by GSH through the formation of potentially damag- ing free glutathione radicals (GS ‘). We have been

* Corresponding author.

interested in the mechanism of action of known

antioxidants of both plant and animal origin. We have shown that antioxidants, such as flavonoids [3-61, that occur in dietary plant material and the antioxidant uric acid [7], which is present at a rela- tively high concentration in human serum, are capa- ble of causing DNA breakage in the presence of Cu(I1) through the generation of active oxygen species. In view of these examples together with those in the literature, we have suggested that several of the biological antioxidants are themselves capable of generating oxygen radicals under appropriate con- ditions.

Reed and Douglas [8,9] have studied chemical cleavage of DNA by glutathione in the presence of C&I) ions. According to these authors, the hydroxyl

0027-5107/96/$15.00 Copyright 0 1996 Elsevier Science B.V. All rights reserved.

PII SOO27-5 107(96)00085- 1

radical may be involved in the cleavage mechanism. In addition, there is extensive base damage before

cleavage. In this report, we show that oxygen is not essential for DNA cleavage. suggesting the involve- ment of glutathione free radicals.

2. Materials and methods

Glutathione and calf thymus DNA (sodium salt. average mol. wt. l,OOO,OOO Dal was obtained from Sigma (St. Louis. MO). Supercoiled plasmid pBR322 DNA was prepared according to standard methods [lo]. All other chemicals were of analytical grade.

2. I. Reactiorl of’ GSH with c&j’ thymus DNA cmd digestiorz with S, r~ucle~~se

Reaction mixture (0.5 ml) contained 10 mM Tris HCI (pH 7.5). 500 pg of DNA, varying amounts of GSH and cupric chloride. Incubation was performed at 37°C for specified time periods. For anaerobic experiments, nitrogen was bubbled through the solu- tions for 5 min. This method has been successfully

used in this laboratory to create an anaerobic atmo- sphere where quercetin-Cu(II)-catalysed degradation of calf thymus DNA is completely inhibited [5.6]. S, nuclease digestion was performed and acid soluble deoxyribonucleotides were determined as described previously [6].

Superoxide was determined by the reduction of nitroblue tetrazolium to a formazan by the method of Nakayama et al. [I 11 in a total volume of 3.0 ml. Superoxide dismutase was introduced in the reaction mixture to confirm the production of 0, ~.

2.3. Reaction of GSH with plasmid pBR322 DNA

Reaction mixture (30 pll contained 10 mM Tris- HCl (pH 7.51 and 1 pg of plasmid DNA. The concentrations of GSH and CuCl, are specified in legends. Incubation at 37°C was for I h. After incubation, 10 pl of a solution containing 40 mM EDTA, 0.05% bromophenol blue and 50% (v/v) glycerol was added. and the solution was subjected

to electrophoresis on 1% agarose gels. The gels were stained with ethidium bromide (0.5 pg/p,l), viewed

and photographed.

2.4. Detrrminutiorl of’ GSH

Glutathione was determined by reaction with 5.5’-dithiobis-2 nitrobenzoic acid (DTNB) and read- ing the absorbance at 412 nm as described by Sedlak and Lindsay [12].

3. Results

3. I. Glutathione-C’u(IIi interuction

In our earlier studies on the flavonoid quercetin- Cu(IIl-mediated DNA degradation, we have studied the complexes involved in quercetin, DNA and Cu(II1 interaction [4]. We have shown that quercetin and Cu(II) fonn a charge transfer complex which absorbs at around 450 nm and decays in an oxygen-depen- dent reaction. A ternary complex of quercetin. Cu(II1 and DNA is formed which generates DNA damaging hydroxyl radicals from molecular oxygen through the formation of superoxide anion. DNA degradation does not occur in the absence of oxygen [5,6]. The

decay of charge transfer complex is a slow process. but is complete in about I5 min. However, in the

presence of DNA, it is considerably faster and is completed in about 1 min. This was interpreted to indicate that Cu(II)-induced oxidation of quercetin is promoted by DNA. Promotion by DNA of Cu(II) induced oxidation of catechols such as DOPA, dopamine and epinephrine has also been reported [2]. We observed a similar acceleration of the decay of quercetin-Cu(II) charge transfer complex by glu- tathione (Fig. la). The acceleration is also seen in the absence of oxygen (Fig. lb). Depending on the extent of oxidation, a range of quercetin-derived oxidation products may be formed [3]. Under aerobic conditions these products absorb below 325 nm [3,4]. This is also seen in the experiment in Fig. la where. under aerobic conditions, the products absorb below 325 nm (peak not seen in Fig. la). However, under anaerobic conditions, a peak is seen at around 350 nm (Fig. lb). Apparently the oxidation of quercetin does not proceed to completion in the absence of air.

A. Jain et al./Mutation Research 357 (1996) 83-88 85

\ 0.0000 I I - ._

325.0 370.0 415 0 460 0 505 0 5’

Wovelength

14400 I

0

.O

WAVELENGTH (nm)

Fig. 1. Effect of time on absorption spectra of quercetin (50 p,M)

in the presence of Cu(II) (150 p,M) and GSH (500 FM). a:

spectra in the presence of air. Trace 1 is the spectrum of quercetin

alone: traces 2, 3 and 4 are recordings after 1, 5 and 15 min,

respectively. b: spectra under anaerobic conditions. Trace 1 is

quercetin alone: traces 2, 3 and 4 are recordings after 1. 5 and 15 min, respectively. In both experiments, the reaction was started by

adding GSH. For the anaerobic experiment, each solution was

bubbled with nitrogen for 5 min before mixing in the cuvette after

which nitrogen was again bubbled for 2 min and the cuvette

sealed with paraffin.

20 40 60

Time (min)

Fig. 2. Effect of GSH on the generation of superoxide anion by

quercetin. The concentration of quercetin used in the reaction

mixture was: 0, 40 p_M quercetin alone; 0. quercetin plus 40

FM GSH; X , quercetin plus 100 pM GSH: A, quercetin plus 50

kg/ml SOD.

Since the quercetin-Cu(I1) charge transfer complex does not decay in the absence of oxygen, it is further suggested that GSH interacts with the complex by

accepting an electron and initiating the oxidation process of quercetin. We have earlier shown that quercetin alone is capable of reducing molecular

oxygen to superoxide anion [5]. Fig. 2 shows the effect of two concentrations of GSH on the rate of formation of superoxide by quercetin. Both concen-

trations of GSH inhibit the rate of superoxide pro- duction in a progressive manner. One explanation of this result would be that GSH acts as an electron acceptor and therefore competes with molecular oxy- gen. Taken together, the results of Fig. 1 and Fig. 2 would be consistent with GSH acting as an electron acceptor either from quercetin alone or from

quercetin-Cu(I1) charge transfer complex.

3.2. Stoichiometry of GSH-CdII) interaction

In the experiment shown in Fig. 3, increasing concentrations of GSH were added to a fixed con- centration of Cu(I1) (300 p,M) and the concentration of free GSH was determined in the solution. It is seen that free GSH is not detected in the solution to a significant extent up to a concentration of 300 ~.LM. The results indicate that 1 p,mol GSH is able to reduce 1 pmol of Cu(I1). That Cu(1) is formed was

X6

2.ooyj 1.60 -

E c

" * c 1.20- rJ

::

s 0 0.80 -

i

: 0.40 -

I 1 I I 0 200 400 600 800

[GS++l JJ,M

Fig. 3. Determination of the stoichiometry of GSH-Cu(II) interac-

tion. Increasing concentrations of GSH were added to 300 FM

Cu(II) solutions before estimating free GSH concentrations ah

described in the text.

confirmed by using the Cu(l)-specific sequestering reagent, bathocuproine [6].

3.3. DNA degradation by glutathione alow and irl the presence of MII)

Glutathione alone and in the presence of Cu(II) generated S, nuclease-sensitive sites in calf thymus DNA [6]. The reaction was assessed by recording the proportion of double-stranded DNA converted to acid-soluble nucleotides by S, nuclease. As seen.

Table I S, nuclease hydrolysis of calf thymu:, DNA treated with GSH and

cuw

Reaction conditions ” % DNA hydrolysed

I. Denatured DNA IO0 IO0 2. ds DNA alone 6.0 7.0

GSH alone GSH+Cu(II)

XdsDNA+GSH( 25kM) 7.1 19.5

1. ds DNA+GSH ( 50 FM) I I.0 77.0

5. ds DNA+GSH (100 FM) 14.0 21.2

6. ds DNA + GSH ( I50 FM) 15.6 76.6

7. ds DNA + GSH (200 FM) 17.5 19.6 8. ds DNA + GSH (300 FM) 20.0 33.5

9. ds DNA+GSH (I00 FM) 11.3 30.0

The concentration of CuCI, used in all experiments was 300 FM.

” Anaerobic incubation.

increasing concentrations of GSH lead to a progres- sive increase in the production of acid-soluble mate-

rial and this is enhanced in the presence of Cu(II) (Table I). Both with GSH alone and in the presence of Cu(I1). the production of acid soluble material was still observed when the reaction mixture was incu-

bated under anaerobic conditions. Supercoiled plas- mid DNA was also examined as a substrate with glutathione alone and as shown in Fig. 4. the super- coiled molecules were converted to relaxed open circles in a dose-dependent reaction. Thiols are con- sidered to cause DNA cleavage in the absence of Cu(I1) at millimolar concentrations [S]. However. we

were able to detect strand cleavage at considerably lower GSH concentration.

3.4. .!$tect of radical scal~erzgers on GSH-CufIII- mediated DNA clearsage

Fig. 5 shows the effect of various scavengers of

radicals on GSH-C&I) catalysed cleavage of plas- mid DNA. Sodium azide and potassium iodide are singlet and triplet state quenchers, respectively [13]. SOD and catalase remove superoxide anion and hy- drogen peroxide, respectively, Benzoate and manni- to1 are scavengers of hydroxyl radicals. Only sodium

azide and potassium iodide afforded partial protec- tion to the conversion of supercoiled molecules to relaxed open circles. Also. no inhibition of DNA cleavage is seen by scavengers of hydroxyl radicals viz. benzoate and mannitol. Hydroxyl radicals can

t’ig. 1. Cleavage of supercoiled pBR322 plasmid DNA by GSH

alone. Figure shows the effect of increasing concentrations of 50.

100 and 200 &M GSH, respectively. The reaction mixtures were

incubated for 2 h at 37°C.

A. Jain et al. /Mutation Research 357 (IYY6J 83-88 87

Fig. 5. Effect of active oxygen scavengers on GSH-CuW medi-

ated DNA breakage. a: DNA alone. b: GSH (100 )LM) and CuW

(200 PM). c-h: same as lane b with KI (50 mM), NaN, (50 mM).

Na benzoate (50 mM), mannitol(50 mkl). SOD (100 pg/ml) and

catalase (100 wg/ml), respectively.

arise by either the Haber-Weiss reaction (0, + H,Oz --+ OH ’ + OH- + 0,) or Fenton reaction (H?O, + Cu(1) + OH ’ + Ok + Cu(I1). Formation of superoxide anion can also, in turn, lead to produc- tion of hydrogen peroxide [14]. Thus, the formation of superoxide anion from molecular oxygen is essen- tial for the generation of hydroxyl radicals. Since sodium azide (which scavengers singlet oxygen) af- fords only partial protection to supercoiled DNA, these results further confirm that molecular oxygen is not essential for GSH-Cu(II)-mediated DNA

cleavage.

4. Discussion

Several lines of indirect evidence presented above

strongly suggest that molecular oxygen is not essen- tial for DNA cleavage by glutathione and Cu(I1). In the absence of species derived from molecular oxy- gen, such as hydroxyl radical and singlet oxygen, the other reactive species involved could be thiyl radi- cals. Although we have not directly demonstrated the formation of thiyl radicals, we suggest that DNA is degraded by GSH-Cu(I1) system through a variety of species including thiyl radicals. The results of Fig. 1 and Fig. 2 suggest that glutathione can act as an electron acceptor which would result in the forma- tion of a thiyl radical. Fig. 3 shows that Cu(I1) is directly reduced by GSH giving rise to Cu(1) and

possibly thiyl radicals (GS ‘1 (see below). DNA

degradation occurs, to a significant extent, under anaerobic conditions, by both GSH alone and in the presence of Cu(I1) (Table 1). The fact that glu- tathione alone is capable of DNA cleavage would also be supportive of the generation of thiyl radicals. Finally, no inhibition of cleavage of plasmid DNA is seen by SOD and catalase (scavengers 0; - + H20,,

respectively). Reed and Douglas [8] have earlier examined the chemical cleavage of plasmid DNA by glutathione in the presence of Cu(I1) ions. According

to these authors, the hydroxyl radical scavengers mannitol and glycerol gave conflicting results; glyc- erol protected supercoiled DNA up to 8596, whereas mannitol afforded no protection. SOD, at concentra- tions as high as 1 mg/ml, provided only limited protection which was not abolished on using boiled

enzyme. Clearly, such protection could not have been due to quenching of superoxide anion. How- ever, sodium azide inhibited DNA cleavage, which is

in agreement with our own results, where potassium iodide also provides partial inhibition. As already mentioned, sodium azide and potassium iodide are singlet and triplet oxygen quenchers. Although Reed and Douglas [8] have implicated hydroxyl radicals as the major DNA cleaving species, their results clearly do not rule out the involvement of additional proxi- mal DNA cleaving agents. It is, therefore, suggested that thiyl radicals in addition to hydroxyl radicals

and singlet oxygen are responsible for DNA degrada- tion by GSH-Cu(I1). It is well established that one- electron oxidation of the labile mercapto group of thiols leads to an equilibrium between free thiyl radicals and disulphide radical anions [ 151.

(1) GS’+GS- $ GSS’-G

(-1)

This reaction has been shown to be conveniently initiated by radiolysis of aqueous solutions. All 3 radical species may arise in GSH-C&I) system.

GSH + Cu( II) + GS + Cu( I)

GS‘+ Cu(1) + GS-+ Cu(I1)

(1) GS’+GS- P GSS’-G

(-I)

In view of ubiquitous presence of GSH in living cells, the formation of thiyl radicals may have impli-

cations for a number of pathological conditions. For example, increased levels of oxidative DNA damage have been suggested to be important in the pathology of Parkinson’s and Alzheimer’s diseases. Further- more, Halliwell and Gutteridge [ 141 have shown that

copper ion concentrations in human brain tissue damaged in Parkinson’s disease are at a level that

could promote oxidative DNA damage in the pres- ence of endogenous substrates, such as L-DOPA [ 161. In this context, the endogenous presence of GSH in cells at a relatively high concentration could be of interest.

Acknowledgements

We are grateful to the Commission of the Euro- pean Communities for financial support under a joint

research project of S.M.H. and J.H.P. (Contract CII-

CT92-0002).

References

[II

@I

[31

[Al

[51

Schulz. G.E. and Schirmer. R.H. ( 1088) Principle\ of pl-otein

structure. Springer-Verlag. New York, p. 53.

Prutz. W.A. (1993) Cu(II)-induced oxidation of catechols.

ascorbate and o-phenylenediamine ih promoted by DNA. Z.

Natureforsch., 48C. 872-878.

Ahmad. M.S.. Fazal. F., Rahman. A.. Hadi, S.M. and Parish.

J.H. (1992) Activities of flavonoids for the cleavage of DNA

in the presence of Cu(II): correlation with generation of

active oxygen specie>, Carcinogenesi\. 13. 605-608.

Rahman, A., Shahabuddin. Hadi. S.M. and Parish. J.H. ( 1990)

Complexes involving quercetin, DNA and C&I). Carcino-

genesia. I 1. X0-2003.

Fazal, F.. Rahman. A.. Green\ill, J.. Ainley. IL. Hadi. S.M.

and Parish. J.H. ( 1990) Strand scihsion in DNA by quercetin

and Cu(II): identification of free radical intermediates and

biological consequences of scission, Carcinogenesis, I I. 200-2008.

161 Rahman. A., Shahabuddin, Hadi, S.M., Parish J.H. and Ain-

ley, K. (1989) Strand scission in DNA induced by quercetin

and CuUI): role of Cu(I) and oxygen free radicals, Carcino-

genesis, 10(10), 1833-1839.

[7] Shamsi, F.A. and Hadi, S.M. (1995) Photo-induction of

strand scission in DNA by uric acid and Cu(II), Free Rad.

Biol. Med.. 19(2). 189-196.

[x] Reed, C.J. and Douglas, K.T. (1991) Chemical cleavage 01

plasmid DNA by glutathione in the presence of Cu(II) ions.

Biochem. J., 275. 601-608.

[Y] Reed. C.J. and Douglas. K.T. (1989) Single strand cleavage

of DNA by Cu(II) and thiols: a powerful chemical DNA-

cleaving system, Biochem. Biophys. Res. Commun.. 132(3).

111-117.

[ I()1

II II

1121

[I31

[I41

[I51

[IhI

Maniatis. T.. Fritsch, E.F. and Sambrook. J. (1982) Molecu-

lar toning: A Laboratory Mannual. Cold Spring Harbor, NY.

Nakayama. T., Kimura. T., Kodama. M. and Nagata. C..

(1083) Generation of hydrogen peroxide and superoxide

anion from active metabolites of naphthylamines and

aminoazodyes: its possible role in carcinogenesis, Carcim-

genesis. 4, 765-769.

Sedlah. J.. and Lindsay. R.H. (1968) Estimation of total

protein bind and protein sulphydryl group with Ellman’h

reagent, Anal. Biochem., 25. 192-205.

Korycka-Dahl. M. and Richardson, T. (I 980) Photodegrada-

tion of DNA with fluorescent light in the presence of ri-

boflavin and photo protection by flavin triplet state quenchers.

Biochem. Biophys.. Acta. 610. 729.

Halliwell, B.. and Gutteridge. J.M. (1984) Review article:

oxygen toxicity, oxygen radicals. transition metals and dis-

ease. Biochem. J.. 219. 1-14.

Prutz, W.A.. Butter. J. and Land. Prutz. E.J. (1994) The

glutathione free radical equilibrium. GS + GS + GSS ‘G.

mediating clcctron transfer to Fe(III)-cytochrome C. Bio-

phys. Chem.. 49, 10-1 I I. Husain. S. and Hadi. S.M. (lY95) Strand scihhion in DNA

induced by L-DOPA in the presence of Cu(II). FEBS Lett.,

36-1. 75-78.