Degradation of DNA by metalloanthracyclines: Requirement for metal ions

Vol.

136,

No.

1, 1986

April

14,

1986

BIOCHEMICAL

AND

BIOPHYSICAL

COMMUNICATIONS Pages

DEGRADATION OF DNA BY MBTALLOANTHRACYCLINES: Yitbarek Department Received

RESEARCH

H. Mariam

of Chemistry,

January

21,

REQUIREMENT FOR METAL

and Gwendolyn

Atlanta

l-7

IONS

P. Glover

University,

Atlanta,

Georgia

30314

1986

Metallodaunomycin has been shown to cleave DNA only in the presence of agent and a metal ion under reaction conditions similar to oxygen, a reducing those used for the cuprous-phenanthroline complex. The intermediacy of 02T and H202 has been substantiated by experiments with superoxide dismutase and catalase, respectively. Only partial inhibition by OH* scavengers was observed. An important feature of the reaction is that no specificity for Cu(I1) was This observation has led us to propose a reaction mechanism different observed. from that proposed for the cuprous-phenanthroline complex. The mechanism proposed includes a catalytic role for metal ions other than Cu(I1) as well as the direct participation of products of metal-catalyzed redox reactions such as semiquinone and/or hydroquinone of daunomycin. 0 1986 Academic Press, Inc.

The anthracyclines and C and

certain

have

found

been

to

ions(l),

and this

mediation

presence

ing

the

in

part,

may actually the

of

scission Unlike

In this

and the the

role

is

reaction

mechanism,

OP2-Cu(II),

of

l,lO-phenanthroline

reaction

not

the

streptonigrin, latter

may also

mediate

DNA strand

responsible

specific different

of metal report,

complex

of

case,

endows

to cuprous

ions the

ion

from that

oxygen scission

therapeutic cleaves

model This

them with

the

suggestion,

proposed

the

ability

to metal-

daunomycin-induced is

shows

the

This for

to other

degradation

evidence

only(2).

study-

us to investigate

on the in

DNA in

for

of the anthracyclines

has prompted

metal

which

antibiotics(2).

structure ions,

their

to be a simple

new evidence

certain

for

,that

has led

B

and bleomycin

on molecular

similarity

quinone

agents

depends

a variety

structural

mitomycins

that

be partially

of

whose

DNA in the presence

loanthracyclines.

Since

has been suggested

the

agents

to

l,lO-phenanthroline-cuprous

action

primarily

akin

a reaction

metalloanthracyclines

mechanism but

DNA in

of oxygen(2,3),

antineoplastic cleave

cleave

Recently,

the

structurally

5,8-quinolinediones.

and metal

effects.

are

DNA

described. cleavage

us to propose

a

the phenanthroline-cuprous,

complex(2).

1

Afl

Copyright 0 1986 rights of reproduction

0006-291X/86 $1.50 by Acadtmtc Press, Inc. in arly ,form reserved.

vol.

136,

No.

1, 1986

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

EXPERIMENTAL [Methy-3H] thymidine 5'-triphosphate, tetrasodium salt (0.0070 mg/ml, specific activity 81.5 Ci/mmole) was purchased from SchwarxfMann and E. Coli DNA polymerase I was from Bethesda Research Laboratories, Inc., while Chelex 100, minus 400 mesh was from Bio-Rad Laboratories. All metal salts, reagent grade, were supplied by Fisher Chemical Co. Other materials, unless otherwise specified, were from Sigma Chemical Co. DNA and daunomycin concentrations were determined spectrophotometrically (using the respective extinction coefficients of 6600 and 11,500 ~-1 CM-~ at 260 and 480 nm respectively) on a Varian 210 Spectrophometer. Poly(dA-[3H]T) was prepared by usin a modified procedure of Modrich and PHIT) was determined Lehman(4). The degradation of poly(dA-[ by acid solubilization following the method of Downey et al(5). The reaction mixture contained, in a final vol me of 0.1 ml, 50 1M daunomycin, 50 mM Hepes buffer, pH 7.4, 20 pM poly(dA-[ Y HIT), 1 mM 2-Mercaptoethanol and 1 l&l CuS04. After 30 min at 25"C, the reaction was stopped by the addition of 0.01 ml of (M EDTA, pH 8.0, 0.1 ml of Salmon sperm DNA, 2 mgjml, and 0.5 ml of 1M perchloric acid. After 10 min at O'C, the solution was centrifuged at 7,500 rpm for 10 min and an aliquot of the supernatant was counted in 10 ml Biofluor in a liquid scintillation counter to determine the percent acid soluble product. Superoxide dismutase(6) and Catalase(7) were assayed for purity using standard procedures before use. Distilled water and all solutions other than those of metal salts were Chelex-treated to scavenge metal ion contaminants. RESULTS The degradation agent

of poly(dA-[3H]T)

and molecular

a 50-fold strongly

excess

oxygen of Dm over

suggests

that

Table components completea

only

1.

(Table

by Dm requires 1).

Cu(II).

The This

catalytic

Requirements

reaction

can proceed

and the obligatory

levels

of

Cu(II)

for the Degradation

average d

a metal

are

ion,

a reducing

effectively

presence

% acid solublee

minus Cu(II)b

209

4

minus 2-mercaptoethanol'-'

102

2

minus daunomycinh

117

2

oxygenbac

76

The

of DNA

51

minus

of Cu(I1)

required.

2631


aThe complete reaction mix ure contained in a final volume of 0.1 ml 5Om~ ‘3 Hepes, pH 7.4, 1mM Z-mercaptoethanol, 1 IM CuSO4, and 20 N-f poly(dA-[ HIT), 50 pM daunomycin. After 30 min at 24'C, acid soluble radioactivity was determined. bThe reaction mixture contained all components as in a minus the component indicated. 'This experiment was carried out by bubbling N2 gas through the reaction mixture. dCounts indicate averages of at least three runs and are for a 10 ul-aliquot of the supernatant from the reaction mixture. eThe control cpm used for the calculation (see Experimental) was 5107 corn. 2

in

re-

Vol.

136,

No. 1, 1986

Table

2.

BIOCHEMICAL

Effect

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

of Superoxide Diamutase and Catalase on the Degradation DNA by Daunomycin-Cu(II) Complexa

conditions

averaged

completea

2190

52

plus catalaseb (10 ug/ml)

130

3

plus superoxide dismutaseb (40 ugg/ml)

2452

58

plus boiled cata1aseb.c (10 idml)

1230

29

of

4; acid solublee

*Poly(dA-[3H]T) (20 uM) was cleaved in 0.1 ml solution containing 5'$" Hepes, pH 7.4, 1mM 2-mercaptoethanol, 50 u M daunomycin, 1.0 p M CuS04. Indicates the complete system plus the component indicated. CThe boiled catalase was prepared by heating the catalase for 10 min before addition. dAs in Table 1. econtrol cpm: 4206.

quirements form

of

for

a reducing

oxygen

may be

The involvement superoxide catalase

on the

slightly

increased

degradation

that

would

expect

inhibition.

02r

of

Ii202

Production

catalyzed

Haber-Weiss

that

be made here

leads

established

by

experiments

then

The experiment

compared

if

provided

(52%). were

with

one

complete

Apparently,

may then

a Fenton

SOD gave

it

peroxide-dependent.

to B202 which

SOD and

control

since

hand,

with

of with

to the

species,

other

degradation.

The effects

2.

of OH. from H202 via

react

further

reaction

is

well

documented.

The

that

the

OH* radical

is

the

by

several

reaction(8) is

the

reactive

is

a reduced

for

(58%)

on the

that

responsible

in Table

reaction

to 0,:

suggest

respectively.

a primary

Catalase,

oxygen

was

product

the

oxygen

species

shown

not

that

OH. radicals.

can

is

molecular

and catalase,

soluble

indicating

the reduction

and

are

acid

indicates

and

reactive

02;

(SOD)

This

form

the

of

dismutase

protection,

agent

or

implicit

to

a metal-

assumption

primary

reactive

species(9). The role Table vengers. present

3 shows

of the

Bensoate

the

OH* radical

results

of

and

acetate

at a concentration

of

was

the

then

effects

were 50 mM.

only

examined of

partially

Sodium 3

some hydroxyl

chloride

effective

experiments.

free

radical

inhibitors

had virtually

scawhen

no effect

Vol.

136,

No. 1, 1986

Table

3.

BIOCHEMICAL

Effect

of

scavenger

Various

concn

AND

Hydroxyl Radical Daunomycin-Cu(II)

(mM)b

RESEARCH

Scavengers Complex

on the

averagebsd

COMMUNICATIONS

Thiol-Dependent

%inhibitionc,b

acetatea

50(100)

821(1505)

18(33)

benzoatea

50(100)

912(2189)

20(48)

NaCla

50(100)

91(91)

aThe complete reaction mixture 50 UM daunom tin, 50mM Hepes, pH 7.4, 1 Values indicated. in parentheses are inhibition is calculated by substracting plete system represented as 100%. dAs

even at

100 mM concentration.

was doubled, that

the

hydroxyl

reaction

for

the metal

ions

increase

ion

for

metal

Table

metal

only

concentration

part

benzoate

The results

modest(lO).

in

of

via

a mechanism

and acetate

seem to suggest which

involves

a

radical.

specificity

loo-fold

was rather

may proceed

Z(2)

contained, in a final volume of 0.1 ml, 20 pM poly(dA-[3H]T, plus the component for 1OOmM concentrations. cThe percent the percent acid soluble from the comin Table 1.

When the

the inhibition

The requirement

for

BIOPHYSICAL

a metal

ions

(Table

tested in

4.

ion 4).

was about

the

was

The percent a third

concentration

Effect

of

conc(

pt+l)b

Metal

examined

Ions

acid

of that

of

the

on the averagebsd

further

metal

to

soluble

product

of Cu(II). ions

Degradation

establish

the

observed

Furthermore, resulted

of

Xacid

in

only

DNAa

solublebsc

Zn(I1)

l.O(lOO)

927(1264)

18(33)

b?(W

l.O(lOO)

931(1337)

18(35)

Fe(II)

l.O(lOO)

910(1075)

17(28)

Fe(II1)

l.O(lOO)

906(1081)

17(2&X)

Ca(I1)

l.O(lOO)

912(1416)

17(27)

Yb(III)e

l.O(lOO)

1967(2810)

38(54)

Cu(I1)

1.0

2633

51

aPoly(dA-13H]T) (20 @l) was cleaved in a reaction mixture containing 5OmM Hepes, pH 7.4, 1mM 2-mercaptoethanol, 50 d daunomycin, and a given concentration of the metal ion. The acid soluble radioactivity was then determined after 30 min at 24°C. bValues in parentheses are for 100 l.M concentrations. CThe control cpm for the 1 @ concentrations (metal) was 5107 and that for the 100 pM, 3728. dAs in Table 1. econtrol cpm: 5205.

4

a a

Vol.

136,

No.

1, 1986

BIOCHEMICAL

two-fold

increase

increase

in

the

cleavage

increase

in

the

concentration

Since

the

ions,

except

extent

stability ions

in

the

acid

reaction of

this

constants

of

to the polymer

does

stability

not

the

appears

metal

not

constants

with

the

two-fold

almost

loo-fold

that

is

expected

to

form(l1).

to be about

the

same for

all

the metal

not

reflect

Moreover,

explain

COMMUNICATIONS

The observed

consistent

does

ions.

RESEARCH

product.

complex

observation

the

BIOPHYSICAL

soluble

is

of degradation

Cu(II),

to different

percent

AND

the data

the

simple

because

and different

differences

binding

such

the

of the

metal

should

lead

binding

degradation

in

effects.

DISCUSSION The several participate with

in

those

anthroline system

lines

are

the

degradation

reported

for

system

showed

(1)

the

were

no marked

metal

ions

examined,

These

ion

is

only

a requirement

determining

role,

step,

is

the

more

with

the

several

(2)

and

in

that

(3)

that

two

except

of metal

extent effective

(1)

probably

do not may be

consistent

features (Table

compared that

and II202

the orthophenof the Dm 4),

degradation

implications:

pathways

that

ions

of

02:

are

The novel

observation

they

that

observations

Cu(II).

of a number

Cu(I1)

suggest

These

for

differences

indicate

systems(2,3),

specificity

coupled

strongly

reaction.

the OP2-Cu(II)

(3)

findings

a catalytic

presented

effectiveness

there

ions.

of evidence

(2)

by

to the

the

other

other

metal

the presence the metal

of a metal ions

participate

may play

in

operative

that

in

a rate-

the

Cu(II)

system. The partial (1)

OH* may be only

produced is

in

the

generated;

latter

is

other

hand,

present,

reaction

(2)

in

divalent

proximity

probably

the

partially

hence,

proceeds

systems metal

in ions

for several from

the

may

for

will

Cu(II)

only case.

H202,

no Cu(II) actually

is

probably via

in

the degradation, site

considerations:

predominantly which

can be interpreted

degradation

OH* scavengers

for

derived

responsible of the

true

calls

OH',

by OH* scavengers

inhibition

(2)

and may react be partially

The

in systems

the

primary

redox

occur

to

a significant

after

reactive

The

Cu(II) species

by Que et.

reactions extent

it

on the

in which

outlined

present,

soon

possibility,

(1)

is

OH* may be

effective.

former

the mechanism

two ways:

is and

a1.(2);

involving leading

the to

Vol.

136,

BIOCHEMICAL

No. 1, 1986

degradation;

(3)

ing

Dm intermediates

reactive Since

able

the data

to

rule

Cu(II)(8).

protection

which

out

credence

provided

absence

an attractive

this

but

DmH2 + ZHORS'

Dm + 2HORS' @DmH'

+ HORS-S-H

DIM'-lhl' DmR' + 2HORSH-DmH2 Step

1 represents

an excess mation

reduction

of a reducing

a free

turn,

can be converted

as a reactive 5-7 are

intermediate.

self-explanatory.

the requirements that

processes

require

metal

ion.

quinone are

consistent Both

the

with

reduction that

these

and

proposing

a metal

anaerobic

the aerobic proposed

we are

Dm in

202;

(3)

H202 + 0,:

+

methide

of

a reducing is

that

direct

participation

reductive

the

case

3

following: (5)

(6)

OH' + OH- + O2

(7)

Step

the

which

both

6

is

A novel the

and the

are

in-

DmH*,

to

in

steps

consistent feature

with of

the

reductive

and oxidative

the degradation

requires

semiquinone redox

oxygen responsible

in

to DmH2. Steps

processes

as proposed

of Dm have

processes

probably

has been alluded

reductive

of the

to reactive

for-

be converted

agent.

of

2 shows the

by a solvent.

of Lown and Sim on similar reduction

the presence

reduction,

, since

reaction

of Dm leads

be applica-

H202 + O2

work).

(Dm*)

The mechanism

as a catalyst

and aerobic

complete

ions,

metal

(DmH2) in

be protonated

ion

the work

the

that

+ HORS-S-ROH

+ 2H+ -

DmH' can also

here

degradation

of

pathway.

Dm by a one-electron

products

to Dm.

oxygen

The proposed

of

(2)

can then

The

for

reactions

ZHORS' + 02-02:

reaction

also

we propose

(1)

reason-

the observation

other

Consequently,

is

redox

should

of

it

as from

reaction

in the presence

Finally,

back

operative,

as a minor

to a semiquinone

can be oxidized

as well

(HOCH2CH2SH, in this

which

generat-

(4)

(DmH*) from

radical,

known

comes from

of Dm to the hydroquinone

agent

of a semiquinone

volving

l-4

+ HORS-S-ROH

the

reaction

mechanism.

reactions

effects,

involves

mechanism

only

COMMUNICATIONS

redox

no differential

Case 3, if

of Cu(I1)

RESEARCH

to degradation.

on the

probably

reaction

Dm + ZHORSH -

for effect

system

lead

Case 1

by catalase.

in the Cu(II)

BIOPHYSICAL

metal-catalyzed,

4 indicated

case 2(12).

significant

In the is

may undergo

in Table

Further

SOD had no

ble

Dm itself

AND

and/or

processes

a

hydro-

themselves

systems(13). been

reported.

Moreover,

species(l4).

It

has

been

at

in

part

for

least

Vol.

136,

No.

1, 1986

the biological lines(l5).

activity The

anthracyclines is

pertinent

action

of

the

in our

laboratory

Grant

reported this to

the here

proposed the

anthracyclines.

ACKNOWLEDGMENTS:

AND

including

results mediate

therefore

Program,

BIOCHEMICAL

will

be reported

This

research

BIOPHYSICAL

acute

strongly

biological

of suggest

activity(l6).

understanding of

COMMUNICATIONS

cardiotoxicity

seem to

Results

RESEARCH

of

the

further

anthracy-

that

metallo-

Work in this

mechanism work

the

of

currently

the in

area

mode of progress

elsewhere. was

supported

by

grants

from

the

NIH/MBRS

No. RR-08006.

REFERENCES P. B. (1982) J. Am. Chem. Sot. 104, 3131. Hertzberg,R. P. and Deravan, 315, and references therein. 2. Que, B. G., Downey, K. M., and So, A. G. (1980) Biochemistry 19, 59875991. 3. Marshall, L. E., Graham, D. R., Reich, K. A., and Sigman, D. S. (1981) Biochemistry 20, 244-250. 4. Modrich, P., and Lehman, J. R. (1970) J. Biol. Chem. 245, 3626-3631. Downey, K. M., Que, B. G., and So, A. G. (1980) Biochem. Biophys. Res. 5. Commun. 93, 264-270. I. J. (1969) J. Biol. Chem. 244, 6049-6055. 6. McCord, J. M., and Fridovich, Beers, R. F., Jr., and Sizer, I. W. (1952) J. Biol. Chem. 195, 133-140. 7. Wilshire, J., and Sawyer, D. (1979) Act. Chem. Res. 12, 105-140; 8. 86, 139-142. McCord, J. M., and Day, E. D. (1978) FEBS Lett. The first order rate constant for the conversion of 02: to H202 by SOD is 9. 2 x 109 ~-1 S-l, (McCord, et. al. (1977) in Superoxide and Superoxide Dismutases, Acad. Press P. ll), i.e., a diffusion-controlled reaction. The only slight increase in degradation (Table 2) in the presence of SOD strongly suggests that production of H202 in the absence of SOD is not the rate-determining step. 10. The second order rate constants for the reaction of a hydroxyl radical and sodium chloride were determined to be 4.2 x with acetate, benzoate, 7 3.3 x 109, and 103 ~-1 s-1, respectively (J. Biol. Chem. 249, 2151 tY9;4)* Int. J. Appl. Radiat. Isot. 18, 493 (1970)). Therefore, at least acetat: and benzoate should be effective scanvengers of the hydroxyl radical. (using a log stability constant of 1.1 (Glover, P. 11. A sample calculation M. S. Thesis, Atlanta University, 1985) at pH 7.4) of the concentrations of the complexes that are expected to form at the two concentrations employed, 1 and 100 M, gives 2.5 x 10D4 M and 2.5 x 10e2 M, respectively. 12. For case 2 to occur, metal ions such as Mg(II), Zn(II), and Ca(I1) would Such reactions are not have to be reduced and oxidized quite readily. if the degradation reaction were dependent however common. Furthermore, on the redox capabilities of the different metal ions, one would have then observed differential effects in the extent of the degradation. Lown, w., and Sim, S. (1976) Can. J. Chem., 54, 2563-2572. 13. 14. Kleyer, D. L., Gaudiano, G., and Koch, T. H. (1984) J. Am. Chem. Sot. 106, 1105-1109, and references therein. Meyers, C. E., McGuire, W., and Young, R. (1976) Cancer Treat. Rep. 60, 15. 961-962. 16. We have considered the possibility that the nonspecificity observed in this investigation might be due to Cu(I1) impurity in the different metal since the concentrations of the metal ions ion concentrations. However, used were exceedingly low, any Cu(II) impurity is unlikely to give the results observed.