Mechanism of reactions involving singlet
Volume
83, number
FEBS LETTERS
1
November
Discussion
MECHANISM
OF REACTIONS
INVOLVING
SUPEROXIDE
SINGLET
OXYGEN
1977 zyxwvutsr
Letter
AND THE
ANION
W. H. KOPPENOL and J. BUTLER* zyxwvutsrqponmlkjihgfedcbaZYXWVUT
Department
of M olecular Biophy sics, Phy sics Laboratory,
University of Utrecht, Sorbonnelaan
4, Utrecht, The Netherlands
and Christie Hospital and Holt Radium Institute, Paterson Laboratories, M anchester M 20 9BX, England
Received
1. Introduction
__,
‘AgO2 + HzOz
0, + HzOz + H’ -
‘Ago, tHzO+OH
O;+OH’tH+
d
%;02
+ Hz0
Hz02
__t
%;02
+ 2H20
+
Hz02
1977
2. Discussion
Recently it was shown [l] on thermodynamic
grounds that singlet oxygen could be a product of the
following redox reactions:
O;tO;t2H+
25 August
Since superoxide anions and hydrogen peroxide are
formed in living systems and since singlet oxygen as
well as the hydroxyl radical are considered to be
harmful species, it is important to know in which
reactions the formation or involvement of these
species is kinetically feasible. For instance, does the
dismutation of 0, yield ‘Z+, ‘A or %Z-02? In two
recent papers by Khan [2,31 it zas calAlated that
‘Ci02 would be formed if 0, were surrounded by six
or more water molecules. It is obvious that the calculations of Khan require the strict control of many
parameters; therefore we maintain that the formation
of ‘XL02 as a product of the dismutation reaction is
most improbable [l] . To decide whether oxygen will
be formed in its ‘Ag or “Cg state we must know the
relative rates of the respective reactions.
*Present address: Hahn-Meitner-Institut
Berlin GmbH, Sektor Strahlenchemie,
North- Holland Publishing Company
fiir Kernforschung
1 Berlin 39, Germany
- Amsterdam
According to the well-known
Arrhenius equation
k =A&~/RT
the rate constant, k, of a reaction is determined by
(i) zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLK
A, a frequency factor, which is influenced by effects
such as diffusion and electrostatic interaction, as well
as the entropy of the activated complex with respect
to the reactants, and (ii) E,, the activation energy,
which can be related to the height of the energy barrier.
For an electron transfer reaction it is necessary that
during the transition state overlap should occur
between the electron donating and the electron accepting orbital [4] . This is possible if they have the correct
topology and symmetry. These effects are not considered in the Marcus theory [5] . If such an overlap
does occur, making mixing in of higher states unnecessary, we expect the reaction to have a small activation
energy and a high rate constant.
0; has the electron configuration:
In 0; one of the two ng* orbitals is filled and the Dther
one contains one electron. It is the latter orbital
which can easily overlap with a half-filled 7~: orbital
of HOz (figs. la-c). After electron transfer, an oxygen
molecule will be formed which has an empty and a
filled T$ orbital: ‘Ago2 (figs. 2a-c). A mechanism
leading to “2; or - if it were thermodynamically
I
Volume
83, number
1
FEBS LETTERS
November
la
lb
1977
zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQ
IC
Fig.la-c.
Schematic representation
of orbitals involved in overlap. Empty orbitals are not shaded, orbitals containing one electron
are half-shaded
and filled orbitals are shaded. The orbitals in the zx- and zy-plane are not shown, nor are the empty 0: orbitals
in the xy-plane.
of other reactions, see table 1. Some of the reactions
possible - ’ Zi02 would involve overlap between the
need comment.
filled rrg*orbital of 0; and the half-filled np* orbital of
Reactions (5) and (6) form the Haber-Weiss cycle
HOa and is therefore unlikely. Experimental evidence
[20] of which reaction [S] has been proposed as a
for the formation of ‘A,02 has been presented by
source of OH’ radicals in biochemical systems [21].
Mayeda and Bard [6]. HOz, which has a pK of 4.9 [7],
Recently McClune and Fee [9] , Halliwell [ 141 and
reacts relatively fast with 0; : kl = 8.5 . lo7 Me1 s-l.
Rigo et al. [22] have investigated this reaction. Of
The reaction of 0; with itself is much slower: zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA
these authors McClune and Fee [9] and Rigo et al.
k < 0.3 M-’ s-’ [8]. This low-rate constant cannot
be caused by electrostatic interaction alone. One can
[22] did not appreciate the fact that reaction (5) forms
part of a cycle, but this has had no effect on their
calculate, using Debye’s correction factor [lo] , that
it should not be less than 0.25 kl. We believe that the
results. These results as well as those of Halliwell [ 141
slow rate is caused by another effect. In the case of a
show that reaction (5) has a very low rate constant.
reaction of 0; with O;, an O,‘- will be formed after
If reaction (5) is to occur, the rate determining step
electron transfer, and in the case of a reaction of 0;
will probably be the transfer of an electron from the
with HOz, a HO; ion. Clearly the HO, ion is a more
rrg*orbital of 0; to the u,* orbital of HzOz. It can be
feasible intermediate than the double-charged Oz- ion.
seen in fig.4 that the filled 71; orbitals of Hz02 hinder
In fact H202 has a pK of 11.8, such that around and
the overlap. If reaction (5) occurs, it will yield ‘Ag or
above pH 11 HO, exists as a relatively stable species.
3E O- depending on whether the electron donating
g 2,
The reaction of HOz with itself zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA
(k = 7.6 . 10’ M-’ s-l)
orbital is the filled or half-filled one.
is slower than the reaction of HO2 with 0; [7]. We
The disproportionation
reaction of hydrogen
assume that this is caused by the stabilizing effect of
peroxide, reaction (4), is very similar to reaction (5).
the proton on the electron donating orbital and the
Smith and Kulig [ 131 have found evidence for ‘A,O,
effect of this proton on hindering the overlap.
production through the base catalysed disproportionaThe concept of overlap between electron donating
tion of Hz02 under conditions which show that this
and accepting orbital helps us to understand the course
reaction has an appreciable amount of activation energy.
000
e”e3
HO2
Fig.2a-c.
2
Encounter,
eB”G$z3-e
0;
overlap
and products
2b
2a
of reaction
(1).
HO,
Effect
of orbital
symmetry
and topology
Donor
orbital
Reaction
Table 1
on the rate of reaction between
Acceptor
orbital
Overlap
possible
Figure
k*)’
g
yes
l(a-c)
(n*P
g
yes,
2x
oxygen
containing
radicals
State of
oxygen if
a product
and/or
molecules
Experimental
evidence for
this state
Rate constant
(M-'s-‘)
zyxwvutsrq
zyxwvutsrqponmlkj
(I) 0 ; + HO, + H+ -+ H,O,
(2) ‘Ago,
+ ‘Ago,
-+ 20,
+ 0,
+ hv
(Q
(QO
yes
(Q’
($)O
difficult
(n*Y
g
(o:Y
difficult
(Q’
(o$O
difficult
(n*Y
g
(PUS
no
(7)0;+OH’+H++H,O+O,
(Q’
(P,)’
yes
(8) OH’ + OH’ + H,O,
(Pu)’
(Pu)’
yes
(3) 0;
+ ‘Ago,
(4) H,O,
(5) 0;
+ H,O,
+,H,O,
(6) H,O,
+ 0, + 0;
-+ 0, + 2H,O
+ H+ -+ 0, + OH’ + H,O
+ OH’ -+ H,O + 0;
a In aprotic media
b The IAgO, in these particular
+ H+
experiments
8.5
3
from the reaction
yes
%zg
yes
yes,
1131
no
lAg
?zg
4
may have been produced
%g
unknown,
high (21
(3.6 * 0.1) . lo’,
1.6 lo9 [11,18Ja
unknown, very
low [12]
yesb
,
lAg
[I51
with HO,
.,,,,,,,.
-^
cc
_
_
_
-,,
__
M-r s“
2.3
10’
[ 161
1.0
10’9
[ 171
5.5
of 0;
CA
zyx
< lo+
no
'A
ii
w
y,,,,/,,.
. 10’ [7]
1’31
. lo9
[19]
[22]
KY
2
Volume
83, number
November
FEBS LETTERS
1
1977
O0
0
;c=c,
‘4302
s
Fig.3.
Double
overlap
Fig.4. Electrons
between
in ni orbitals
Fig.5. Overlap
of a d orbital
Fig.6.
between
Overlap
nl orbitals
of H,O,
of two ‘Ago,
(left) hinder
overlap
of Fe’+ with the 0; orbital
the empty
ng* orbital
molecules,
of ‘Ago,
between
of H,O,,
+ R-S-H
(3pu --f 3d,) d
(2).
n; orbital
reaction
of 0;
(right)
and 0: orbital
of an alkene,
reaction
4
anion produced
[25] may react with
(5).
(10).
*Overlap between the filled $ orbital of ‘Ago, and the
empty rr; Tf the alken& does not seem likely in view of the
difference m electronegativity
between oxygen and carbon;
however if it occurred, the compound
formed would be a
dioxetane:
O_O
I
The superoxide
reaction
another thiol group in exactly the same manner. The
products of this reaction will be another sulphur
radical and hydrogen peroxide. The reaction of singlet
oxygen with a thiol is of particular interest in that the
sulphur atom need not be activated. A direct overlap
between an empty I$ orbital of ‘AgO, and the filled
3pu orbital of the sulphur will directly lead to a reaction giving the same products as reaction (9).
3E;02 and O;, having no empty rri orbitals, cannot
easily react with a double bond. As shown in fig.6
‘A 0 has one empty 71: orbital which can overlap
g *
with the 71” orbital of the double bond*. Addition
0; +RS’+H+
(9)
of H,O,,
(12).
and the filled xu orbital
The oxidation of HzOz by OH’ is very slow for an
OH’-reaction [23] . No direct overlap is possible and
therefore it is likely that the reaction proceeds by a
mechanism that involves a high-activation energy. The
effect of direct overlap is nicely illustrated by the
reaction of 0; with OH’ [7] : this reaction is 450 times
faster than reaction (6). Note that we expect ‘AgOa,
not ‘CiOa to be formed, even though ‘Xi02 is thermodynamically possible.
We can now discuss two other reactions which are
of biol-ogical importance: the oxidation of a thiol and
the ‘ene’ reaction of ‘Aa02 with an alkene.
The oxidation of a thiol by “Xi oxygen is brought
about by the availability of the 3d, state of sulphur.
A half-filled TINorbital of 3Ci02 can overlap with a
3p, -+ 3d, excited sulphur atom:
3Z;02
reaction
I
-c-c
_
I I
FEBS LETTERS
Volume 83, number 1
will preferably occur at that carbon atom which has the
higher electron density. The reaction product is a
hydroperoxide:
/
/=\
+‘Ago2
7-
November 1977
The production of hydroxyl radicals via the modified Haber-Weiss reactions (11) and (12), and of
singlet oxygen from the dismutation reaction (1) and
the disproportionation
reaction (4) can occur simultaneously. Therefore it is not surprising that both
superoxide dismutase and catalase are needed to
protect living systems [34,35] .
(10)
Experimental evidence shows that when ‘Ago, reacts
with alkenes in organic solvents the expected product
is found but not exclusively. The rate constants range
from 4.4 . lo3 to 2.4 . 10’ M-’ s-r and the activation
energies are low: 5.4-0.5 kcal/mol [26]. Since the
reaction of 0, with HOa will yield ‘AgOa, reaction
(10) may be of particular importance with respect to
lipid peroxide formation, provided ‘Ago2 is produced
near a membrane; its lifetime in water is very short:
r = 2 gs [27]. In the lipid bilayer this may be increased,
permitting selective reactions. It will be noted that we
do not regard 0; in itself as a harmful radical. From
reaction (9) and (10) it is clear that we expect 0; to
behave like triplet oxygen.
As shown above, it is unlikely that OH’ radicals can
be formed from the direct reaction of 0; with HaOz.
However, iron compounds may be able to catalyze this
reaction:
Fe(III)complex
+ 0; P
Fe(I1) complex + O2
(11)
Evidence for reduction of Fe(III)EDTA by 0; has
been presented by Halliwell [28]. Overlap between a
d-orbital of iron and the $-orbital
of Hz02 is just
possible as shown in fig.5, making reaction (12)
possible:
Fe(II)complex
+ Hz02 +
Fe(III)complex
+ OH’ + OH- (12)
This leads to a modified Haber-Weiss cycle (see also
[29] , [30] and [3 l] ) which would explain the results
of Haber and Weiss and the fact that iron compounds
appear to be necessary for lipid peroxide formation
[31-331.
References
[l] Koppenol, W. H. (1976) Nature 262,420-421.
[2] Khan, A. U. (1976) J. Whys.Chem. 80,2219-2228.
[3] Khan, A. U. (1977) J. Am. Chem. Sot. 99,370-371.
[4a]
Pearson, R. G. (1976) in: Symmetry
Rules for Chemical
Reactions,
pp. 440-441,
Wiley, New York.
[4b] Pearson, R. G. (1971) Pure Appl. Chem. 27, 145-160.
151 Marcus, R. A. (1964) Ann. Red. Phys. Chem. 15,
155-196.
161 Mayeda, E. A. and Bard, A. J. (1973) J. Am. Chem. Sot.
96,4023-4024.
171 Behar, D., Czapski, G., Rabani, J., Dorfman, L. M. and
Schwarz, H. A. (1970) J. Phys. Chem. 74,3209-3213.
181 Bielski, B. H. J. and Allen, A. 0. (1977) J. Phys. Chem.
81, 1048-1050.
I91 McClune, G. J. and Fee, J. A. (1976) FEBS Lett. 67,
294-298.
1101 Debye, P. (1942) Trans. Electrochem. Sot. 82,265-272.
Rosenthal,
I. (1975) Isr. J. Chem. 13,86-90.
IllI
1121 Schumb, W. C., Sattertield, Ch. N. and Wenthworth,
R. L. (1955) Hydrogen Peroxide, p. 515, Reinhold,
New York.
iI31 Smith, L. L. and Kulig, M. J. (1976) J. Am. Chem.
Sot. 98,1027-1029.
[I41 Halliwell, B. (1976) FEBS Lett. 72,8-10.
1151 Arneson, R. M. (1970) Arch. Biochem. Biophys. 136,
352-360.
1161 Thomas, J. K. (1965) Trans. Faraday Sot. 61,
702-707.
I171 Sehested, K., Rasmussen, 0. L. and Fricke, H. (1968)
J. Phys. Chem. 72,626-631.
[I81 Guiraud, H. J. and Foote, C. S. (1976) J. Am. Chem.
Sot. 98,1984-1986.
[I91 Rabani, J. and Matheson, M. S. (1966) J. Phys. Chem.
70,761-769.
Haber, F. and Weiss, J. (1934) Proc. Royal SOC. A 147,
WI
332-351.
[211 Beauchamp, C. and Fridovich, I. (1970) J. Biol..Chem.
245,4641-4646.
Rigo, A., Stevanato,
R., Finazzi-Agro,
A. and Rotilio, G.
PI
(1977) FEBS Lett. 80,130-132.
1231 Dorfman, L. M. and Adams, G. E. (1973) Reactivity
of the Hydroxyl Radical in Aqueous Solutions.
NSRDS-NBS 46, National Bureau of Standards,
Washington.
5
Volume
[24]
[25]
[26]
[27]
[28]
[29]
[30]
8 3, number
1
I:EBS LETTERS
Koppenol,
W. H., Butler, J. and Van Leeuwen, J. W.
(1977) in: International
Conference
on Singlet Oxygen
and Related Species in Chemistry
and Biology,
Abstract K-9, Pinawa, Canada, August 22-26.
Misra, H. P. (1974) J. Biol. Chem. 249,2151-2155.
Ashford, R. D. and Ogryzlo, E. A. (1975) J. Am. Chem.
Sot. 97, 3604-3607,
and references mentioned
therein.
Merkel, P. B. and Kearns, D. R. (1972) J. Am. Chem.
Sot. 94,7244-7253.
Halliwell, B. (1975) FEBS Lett. 56, 34-38.
Barb, W. G., Baxendale, J. H., George, P. and Hargrave,
K. R. (1951) Trans. Faraday Sot. 47,591-616.
Misra, H. P. and Fridovich, I. (1976) Arch. Biochem.
Biophys. 176,577-581.
6
View publication stats
[31a]
November
1977
Fang, K.-L., McCay, P. B., Poyer, J. L., Misra, H. P.
and Keele, B. B. (1976) Chem.-Biol. Interactions,
15,
77-89.
[31b]McCay,P.B.,Floyd,R.A.,Lai,E.K.,Poyer,J.L.and
Fond, K.-L.,(1976)
Biophys. J. 16,66a, Abstract No.
WPOS-D14.
[32]
Kaschnitz, R. M. and Hatefi, Y. (1975) Arch. Biochem.
Biophys. 171,292-304.
[33]
Pederson, T. C. and Aust, S. D. (1973) Biochem.
Biophys. Res. Commun. 52, 1071-1078.
[34]
Fee, J. A. and Teitelbaum,
H. D. (1972) Biochem.
Biophys. Res. Commun. 49, 150-158.
[35]
Yost, F. J. and Fridovich, I. (1976) Arch. Biochem.
Biophys. 175,514-519.
83, number
FEBS LETTERS
1
November
Discussion
MECHANISM
OF REACTIONS
INVOLVING
SUPEROXIDE
SINGLET
OXYGEN
1977 zyxwvutsr
Letter
AND THE
ANION
W. H. KOPPENOL and J. BUTLER* zyxwvutsrqponmlkjihgfedcbaZYXWVUT
Department
of M olecular Biophy sics, Phy sics Laboratory,
University of Utrecht, Sorbonnelaan
4, Utrecht, The Netherlands
and Christie Hospital and Holt Radium Institute, Paterson Laboratories, M anchester M 20 9BX, England
Received
1. Introduction
__,
‘AgO2 + HzOz
0, + HzOz + H’ -
‘Ago, tHzO+OH
O;+OH’tH+
d
%;02
+ Hz0
Hz02
__t
%;02
+ 2H20
+
Hz02
1977
2. Discussion
Recently it was shown [l] on thermodynamic
grounds that singlet oxygen could be a product of the
following redox reactions:
O;tO;t2H+
25 August
Since superoxide anions and hydrogen peroxide are
formed in living systems and since singlet oxygen as
well as the hydroxyl radical are considered to be
harmful species, it is important to know in which
reactions the formation or involvement of these
species is kinetically feasible. For instance, does the
dismutation of 0, yield ‘Z+, ‘A or %Z-02? In two
recent papers by Khan [2,31 it zas calAlated that
‘Ci02 would be formed if 0, were surrounded by six
or more water molecules. It is obvious that the calculations of Khan require the strict control of many
parameters; therefore we maintain that the formation
of ‘XL02 as a product of the dismutation reaction is
most improbable [l] . To decide whether oxygen will
be formed in its ‘Ag or “Cg state we must know the
relative rates of the respective reactions.
*Present address: Hahn-Meitner-Institut
Berlin GmbH, Sektor Strahlenchemie,
North- Holland Publishing Company
fiir Kernforschung
1 Berlin 39, Germany
- Amsterdam
According to the well-known
Arrhenius equation
k =A&~/RT
the rate constant, k, of a reaction is determined by
(i) zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLK
A, a frequency factor, which is influenced by effects
such as diffusion and electrostatic interaction, as well
as the entropy of the activated complex with respect
to the reactants, and (ii) E,, the activation energy,
which can be related to the height of the energy barrier.
For an electron transfer reaction it is necessary that
during the transition state overlap should occur
between the electron donating and the electron accepting orbital [4] . This is possible if they have the correct
topology and symmetry. These effects are not considered in the Marcus theory [5] . If such an overlap
does occur, making mixing in of higher states unnecessary, we expect the reaction to have a small activation
energy and a high rate constant.
0; has the electron configuration:
In 0; one of the two ng* orbitals is filled and the Dther
one contains one electron. It is the latter orbital
which can easily overlap with a half-filled 7~: orbital
of HOz (figs. la-c). After electron transfer, an oxygen
molecule will be formed which has an empty and a
filled T$ orbital: ‘Ago2 (figs. 2a-c). A mechanism
leading to “2; or - if it were thermodynamically
I
Volume
83, number
1
FEBS LETTERS
November
la
lb
1977
zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQ
IC
Fig.la-c.
Schematic representation
of orbitals involved in overlap. Empty orbitals are not shaded, orbitals containing one electron
are half-shaded
and filled orbitals are shaded. The orbitals in the zx- and zy-plane are not shown, nor are the empty 0: orbitals
in the xy-plane.
of other reactions, see table 1. Some of the reactions
possible - ’ Zi02 would involve overlap between the
need comment.
filled rrg*orbital of 0; and the half-filled np* orbital of
Reactions (5) and (6) form the Haber-Weiss cycle
HOa and is therefore unlikely. Experimental evidence
[20] of which reaction [S] has been proposed as a
for the formation of ‘A,02 has been presented by
source of OH’ radicals in biochemical systems [21].
Mayeda and Bard [6]. HOz, which has a pK of 4.9 [7],
Recently McClune and Fee [9] , Halliwell [ 141 and
reacts relatively fast with 0; : kl = 8.5 . lo7 Me1 s-l.
Rigo et al. [22] have investigated this reaction. Of
The reaction of 0; with itself is much slower: zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA
these authors McClune and Fee [9] and Rigo et al.
k < 0.3 M-’ s-’ [8]. This low-rate constant cannot
be caused by electrostatic interaction alone. One can
[22] did not appreciate the fact that reaction (5) forms
part of a cycle, but this has had no effect on their
calculate, using Debye’s correction factor [lo] , that
it should not be less than 0.25 kl. We believe that the
results. These results as well as those of Halliwell [ 141
slow rate is caused by another effect. In the case of a
show that reaction (5) has a very low rate constant.
reaction of 0; with O;, an O,‘- will be formed after
If reaction (5) is to occur, the rate determining step
electron transfer, and in the case of a reaction of 0;
will probably be the transfer of an electron from the
with HOz, a HO; ion. Clearly the HO, ion is a more
rrg*orbital of 0; to the u,* orbital of HzOz. It can be
feasible intermediate than the double-charged Oz- ion.
seen in fig.4 that the filled 71; orbitals of Hz02 hinder
In fact H202 has a pK of 11.8, such that around and
the overlap. If reaction (5) occurs, it will yield ‘Ag or
above pH 11 HO, exists as a relatively stable species.
3E O- depending on whether the electron donating
g 2,
The reaction of HOz with itself zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA
(k = 7.6 . 10’ M-’ s-l)
orbital is the filled or half-filled one.
is slower than the reaction of HO2 with 0; [7]. We
The disproportionation
reaction of hydrogen
assume that this is caused by the stabilizing effect of
peroxide, reaction (4), is very similar to reaction (5).
the proton on the electron donating orbital and the
Smith and Kulig [ 131 have found evidence for ‘A,O,
effect of this proton on hindering the overlap.
production through the base catalysed disproportionaThe concept of overlap between electron donating
tion of Hz02 under conditions which show that this
and accepting orbital helps us to understand the course
reaction has an appreciable amount of activation energy.
000
e”e3
HO2
Fig.2a-c.
2
Encounter,
eB”G$z3-e
0;
overlap
and products
2b
2a
of reaction
(1).
HO,
Effect
of orbital
symmetry
and topology
Donor
orbital
Reaction
Table 1
on the rate of reaction between
Acceptor
orbital
Overlap
possible
Figure
k*)’
g
yes
l(a-c)
(n*P
g
yes,
2x
oxygen
containing
radicals
State of
oxygen if
a product
and/or
molecules
Experimental
evidence for
this state
Rate constant
(M-'s-‘)
zyxwvutsrq
zyxwvutsrqponmlkj
(I) 0 ; + HO, + H+ -+ H,O,
(2) ‘Ago,
+ ‘Ago,
-+ 20,
+ 0,
+ hv
(Q
(QO
yes
(Q’
($)O
difficult
(n*Y
g
(o:Y
difficult
(Q’
(o$O
difficult
(n*Y
g
(PUS
no
(7)0;+OH’+H++H,O+O,
(Q’
(P,)’
yes
(8) OH’ + OH’ + H,O,
(Pu)’
(Pu)’
yes
(3) 0;
+ ‘Ago,
(4) H,O,
(5) 0;
+ H,O,
+,H,O,
(6) H,O,
+ 0, + 0;
-+ 0, + 2H,O
+ H+ -+ 0, + OH’ + H,O
+ OH’ -+ H,O + 0;
a In aprotic media
b The IAgO, in these particular
+ H+
experiments
8.5
3
from the reaction
yes
%zg
yes
yes,
1131
no
lAg
?zg
4
may have been produced
%g
unknown,
high (21
(3.6 * 0.1) . lo’,
1.6 lo9 [11,18Ja
unknown, very
low [12]
yesb
,
lAg
[I51
with HO,
.,,,,,,,.
-^
cc
_
_
_
-,,
__
M-r s“
2.3
10’
[ 161
1.0
10’9
[ 171
5.5
of 0;
CA
zyx
< lo+
no
'A
ii
w
y,,,,/,,.
. 10’ [7]
1’31
. lo9
[19]
[22]
KY
2
Volume
83, number
November
FEBS LETTERS
1
1977
O0
0
;c=c,
‘4302
s
Fig.3.
Double
overlap
Fig.4. Electrons
between
in ni orbitals
Fig.5. Overlap
of a d orbital
Fig.6.
between
Overlap
nl orbitals
of H,O,
of two ‘Ago,
(left) hinder
overlap
of Fe’+ with the 0; orbital
the empty
ng* orbital
molecules,
of ‘Ago,
between
of H,O,,
+ R-S-H
(3pu --f 3d,) d
(2).
n; orbital
reaction
of 0;
(right)
and 0: orbital
of an alkene,
reaction
4
anion produced
[25] may react with
(5).
(10).
*Overlap between the filled $ orbital of ‘Ago, and the
empty rr; Tf the alken& does not seem likely in view of the
difference m electronegativity
between oxygen and carbon;
however if it occurred, the compound
formed would be a
dioxetane:
O_O
I
The superoxide
reaction
another thiol group in exactly the same manner. The
products of this reaction will be another sulphur
radical and hydrogen peroxide. The reaction of singlet
oxygen with a thiol is of particular interest in that the
sulphur atom need not be activated. A direct overlap
between an empty I$ orbital of ‘AgO, and the filled
3pu orbital of the sulphur will directly lead to a reaction giving the same products as reaction (9).
3E;02 and O;, having no empty rri orbitals, cannot
easily react with a double bond. As shown in fig.6
‘A 0 has one empty 71: orbital which can overlap
g *
with the 71” orbital of the double bond*. Addition
0; +RS’+H+
(9)
of H,O,,
(12).
and the filled xu orbital
The oxidation of HzOz by OH’ is very slow for an
OH’-reaction [23] . No direct overlap is possible and
therefore it is likely that the reaction proceeds by a
mechanism that involves a high-activation energy. The
effect of direct overlap is nicely illustrated by the
reaction of 0; with OH’ [7] : this reaction is 450 times
faster than reaction (6). Note that we expect ‘AgOa,
not ‘CiOa to be formed, even though ‘Xi02 is thermodynamically possible.
We can now discuss two other reactions which are
of biol-ogical importance: the oxidation of a thiol and
the ‘ene’ reaction of ‘Aa02 with an alkene.
The oxidation of a thiol by “Xi oxygen is brought
about by the availability of the 3d, state of sulphur.
A half-filled TINorbital of 3Ci02 can overlap with a
3p, -+ 3d, excited sulphur atom:
3Z;02
reaction
I
-c-c
_
I I
FEBS LETTERS
Volume 83, number 1
will preferably occur at that carbon atom which has the
higher electron density. The reaction product is a
hydroperoxide:
/
/=\
+‘Ago2
7-
November 1977
The production of hydroxyl radicals via the modified Haber-Weiss reactions (11) and (12), and of
singlet oxygen from the dismutation reaction (1) and
the disproportionation
reaction (4) can occur simultaneously. Therefore it is not surprising that both
superoxide dismutase and catalase are needed to
protect living systems [34,35] .
(10)
Experimental evidence shows that when ‘Ago, reacts
with alkenes in organic solvents the expected product
is found but not exclusively. The rate constants range
from 4.4 . lo3 to 2.4 . 10’ M-’ s-r and the activation
energies are low: 5.4-0.5 kcal/mol [26]. Since the
reaction of 0, with HOa will yield ‘AgOa, reaction
(10) may be of particular importance with respect to
lipid peroxide formation, provided ‘Ago2 is produced
near a membrane; its lifetime in water is very short:
r = 2 gs [27]. In the lipid bilayer this may be increased,
permitting selective reactions. It will be noted that we
do not regard 0; in itself as a harmful radical. From
reaction (9) and (10) it is clear that we expect 0; to
behave like triplet oxygen.
As shown above, it is unlikely that OH’ radicals can
be formed from the direct reaction of 0; with HaOz.
However, iron compounds may be able to catalyze this
reaction:
Fe(III)complex
+ 0; P
Fe(I1) complex + O2
(11)
Evidence for reduction of Fe(III)EDTA by 0; has
been presented by Halliwell [28]. Overlap between a
d-orbital of iron and the $-orbital
of Hz02 is just
possible as shown in fig.5, making reaction (12)
possible:
Fe(II)complex
+ Hz02 +
Fe(III)complex
+ OH’ + OH- (12)
This leads to a modified Haber-Weiss cycle (see also
[29] , [30] and [3 l] ) which would explain the results
of Haber and Weiss and the fact that iron compounds
appear to be necessary for lipid peroxide formation
[31-331.
References
[l] Koppenol, W. H. (1976) Nature 262,420-421.
[2] Khan, A. U. (1976) J. Whys.Chem. 80,2219-2228.
[3] Khan, A. U. (1977) J. Am. Chem. Sot. 99,370-371.
[4a]
Pearson, R. G. (1976) in: Symmetry
Rules for Chemical
Reactions,
pp. 440-441,
Wiley, New York.
[4b] Pearson, R. G. (1971) Pure Appl. Chem. 27, 145-160.
151 Marcus, R. A. (1964) Ann. Red. Phys. Chem. 15,
155-196.
161 Mayeda, E. A. and Bard, A. J. (1973) J. Am. Chem. Sot.
96,4023-4024.
171 Behar, D., Czapski, G., Rabani, J., Dorfman, L. M. and
Schwarz, H. A. (1970) J. Phys. Chem. 74,3209-3213.
181 Bielski, B. H. J. and Allen, A. 0. (1977) J. Phys. Chem.
81, 1048-1050.
I91 McClune, G. J. and Fee, J. A. (1976) FEBS Lett. 67,
294-298.
1101 Debye, P. (1942) Trans. Electrochem. Sot. 82,265-272.
Rosenthal,
I. (1975) Isr. J. Chem. 13,86-90.
IllI
1121 Schumb, W. C., Sattertield, Ch. N. and Wenthworth,
R. L. (1955) Hydrogen Peroxide, p. 515, Reinhold,
New York.
iI31 Smith, L. L. and Kulig, M. J. (1976) J. Am. Chem.
Sot. 98,1027-1029.
[I41 Halliwell, B. (1976) FEBS Lett. 72,8-10.
1151 Arneson, R. M. (1970) Arch. Biochem. Biophys. 136,
352-360.
1161 Thomas, J. K. (1965) Trans. Faraday Sot. 61,
702-707.
I171 Sehested, K., Rasmussen, 0. L. and Fricke, H. (1968)
J. Phys. Chem. 72,626-631.
[I81 Guiraud, H. J. and Foote, C. S. (1976) J. Am. Chem.
Sot. 98,1984-1986.
[I91 Rabani, J. and Matheson, M. S. (1966) J. Phys. Chem.
70,761-769.
Haber, F. and Weiss, J. (1934) Proc. Royal SOC. A 147,
WI
332-351.
[211 Beauchamp, C. and Fridovich, I. (1970) J. Biol..Chem.
245,4641-4646.
Rigo, A., Stevanato,
R., Finazzi-Agro,
A. and Rotilio, G.
PI
(1977) FEBS Lett. 80,130-132.
1231 Dorfman, L. M. and Adams, G. E. (1973) Reactivity
of the Hydroxyl Radical in Aqueous Solutions.
NSRDS-NBS 46, National Bureau of Standards,
Washington.
5
Volume
[24]
[25]
[26]
[27]
[28]
[29]
[30]
8 3, number
1
I:EBS LETTERS
Koppenol,
W. H., Butler, J. and Van Leeuwen, J. W.
(1977) in: International
Conference
on Singlet Oxygen
and Related Species in Chemistry
and Biology,
Abstract K-9, Pinawa, Canada, August 22-26.
Misra, H. P. (1974) J. Biol. Chem. 249,2151-2155.
Ashford, R. D. and Ogryzlo, E. A. (1975) J. Am. Chem.
Sot. 97, 3604-3607,
and references mentioned
therein.
Merkel, P. B. and Kearns, D. R. (1972) J. Am. Chem.
Sot. 94,7244-7253.
Halliwell, B. (1975) FEBS Lett. 56, 34-38.
Barb, W. G., Baxendale, J. H., George, P. and Hargrave,
K. R. (1951) Trans. Faraday Sot. 47,591-616.
Misra, H. P. and Fridovich, I. (1976) Arch. Biochem.
Biophys. 176,577-581.
6
View publication stats
[31a]
November
1977
Fang, K.-L., McCay, P. B., Poyer, J. L., Misra, H. P.
and Keele, B. B. (1976) Chem.-Biol. Interactions,
15,
77-89.
[31b]McCay,P.B.,Floyd,R.A.,Lai,E.K.,Poyer,J.L.and
Fond, K.-L.,(1976)
Biophys. J. 16,66a, Abstract No.
WPOS-D14.
[32]
Kaschnitz, R. M. and Hatefi, Y. (1975) Arch. Biochem.
Biophys. 171,292-304.
[33]
Pederson, T. C. and Aust, S. D. (1973) Biochem.
Biophys. Res. Commun. 52, 1071-1078.
[34]
Fee, J. A. and Teitelbaum,
H. D. (1972) Biochem.
Biophys. Res. Commun. 49, 150-158.
[35]
Yost, F. J. and Fridovich, I. (1976) Arch. Biochem.
Biophys. 175,514-519.