73
597th MEETING, LONDON
such act as inhibitors of iron-catalysed free-radical damage to
biomolecules (Stocks et al., 1974; Gutteridge et al., 1981a). For
iron to be able to catalytically participate in reaction ( 3 ) it
requires access to its two valence states. Iron as part of the
cytochromes, haemoglobin, catalase and ferritin is unable to
catalyse reaction ( 3 ) (Halliwell, 19786).
A method has recently been developed to detect micromolar
concentrations of iron capable of participating in the HaberWeiss reaction. As previously mentioned, bleomycin does not
degrade DNA unless ferrous ions and dioxygen are present. Its
ability to chelate loosely bound iron and to bind to DNA has
therefore been exploited as a quantitative measurement of iron
(Gutteridge et al., 1981b). This assay has provided us with
direct evidence that ‘oxygen activating’ iron is present in
extracellular fluids such as the synovial and pleural fluids and
the cerebrospinal fluid. Loosely bound iron could not be
detected in normal human plasma, a finding consistent with its
unsaturated-transferrin levels, which are partly responsible for
its observed potent antioxidant properties (Stocks et al., 1974).
dation and of the degradation of DNA, carbohydrates and
amino acids by iron salts (Gutteridge et al., 1980: J. M. C .
Gutteridge & B. Halliwell, unpublished work). This extracellular
copper protein is an acute-phase reactant appearing in response
to inflammatory stimuli and tissue damage. The remarkable
antioxidant properties of caeruloplasmin and its inducible
presence in extracellular fluids, which are inadequately protected
against oxygen intermediates, can be related to mechanisms
which do not scavenge active oxygen intermediates but limit
their reactivity with iron.
4. Protection against oxygen radicals in extracellular fluids
Cells utilizing dioxygen are, under normal conditions,
adequately protected against its toxic effects. Extracellular fluids
are generally poor in superoxide dismutase, catalase and
peroxidases, but are nevertheless, subjected to superoxide and
hydrogen peroxide generated by active leucocytes, released
enzymes or autoxidizing substrates. In the presence of inorganic
iron, hydroxyl-radical formation can take place. Under such
circumstances, primary protection may be dependent on two
different proteins both involved in iron metabolism. These are
the iron-binding protein lactoferrin and the copper protein
caeruloplasmin. catalysing ferrous-ion oxidation with the coupled reduction of dioxygen to water. The iron-binding protein
lactoferrin shares many common physical and chemical properties with transferrin. Both proteins bind two molecules of ferric
ion per molecule of protein with high affinity. Lactoferrin is,
however, immunologically distinct from transferrin and it is
found at both intracellular and extracellular sites. Activated
neutrophils can secrete lactoferrin into the surrounding medium.
When this protein has available binding sites for iron it can. like
transferrin. act as an effective antioxidant by protecting lipids
from iron-catalysed peroxidation (Gutteridge et 01.. I98 la). If
ferrous ions in extracellular fluids are rapidly oxidized to the
ferric state. reaction ( 3 ) (resulting in the formation of the OH’
radical) cannot take place. The ferric ions can then be removed
by transferrin of lactoferrin. Caeruloplasmin is an extremely
effective inhibitor of iron- and copper-catalysed lipid peroxi-
Gutteridge, J. M. C., Paterson, S. K.. Segal, A. W. & Halliwell. B.
Ambe, K. S. & Tappel, A. L. ( I 96 I) J. Food Sci. 2 6 , 4 4 8 4 5 I
Gutteridge,J. M. C. (1981) FEBS Lerf. 182, 343-346
Gutteridge, J. M. C. & Fu, X.-C. (1981a) FEBS Lett. 123, 71-74
Gutteridge, J. M. C. & Fu, X.-C. (1981b) Biochem. Biophys. Res.
Commun. 99, 1354-1360
Gutteridge, J. M. C. & Shute, D. J. (1982) J . Inorg. Biochem. in the
press
Gutteridge, J. M. C., Richmond, R. & Halliwell. B. (1980) FEBS Lett.
112.269-272
(1981a) Biochem. J . 199,259-261
Gutteridge, J. M. C., Rowley, D. A., & Halliwell. B. (1981b) Biochem.
J . 199,263-265
Halliwell, B. (1978a) FEBS Lett. 96, 238-242
Halliwell, B. (19786) FEBS Lett. 92, 321-326
Halliwell, B. & Gutteridge.J. M. C. (1981) FEBS Lett. 128. 347-352
Halliwell, B., Richmond, R., Wong, S. F. & Gutteridge,J. M. C. (1980)
in Biological and Clinical Aspecls of Superoxide and Superoxide
Dismutase (Bannister, W. H. & Bannister, J. V., eds.). pp. 3 2 4 0 .
Elsevier/North-Holland,Amsterdam
Kapp, D. S. & Smith, K. C. (1970)Radial. Res. 42. 34-49
Morre, J. & Morazzani-Pelletier,S. (1966) C.R. Hebd. Sdances Acad.
Sci. Ser. C, 262, 1729- I73 1
Oberley. L. W. & Buettner. G. R. (1979) FEBS Lett. 9 7 . 4 7 4 9
Packer, J. E., Mahood. J. S., More-Arellamo. V. 0.. Slater. T. F..
Willson, R. C. & Wolfenden. B. S. (1981) Biochem. Biophjx Res.
Commun. 98.901-906
Saran, M.. Michel. C. & Bors, W. (1980) in Chemical and Biochemical
Aspects of Superoxide and Superoxide Dismutase (Bannister. J. V.
& Hill. H. A. O., eds.), pp. 38-44. Elsevier/North Holland.
Amsterdam
Sausville, E. A., Peisach, J. & Horwitz. S. B. (1976) Biochem. Biophvs.
Res. Commun. 73.814-822
Sausville, E. A., Peisach. J. & Horwitz. S. B. (1978) Biochemistrj, 17.
2740-2746
Stocks, J., Gutteridge. J. M. C.. Sharp. R. J. & Dormandy, T. L. (1974)
Clin. Sci. Mol. Med. 47. 223-233
Wong, S . F.. Halliwell. B.. Richmond. R. & Skowroneck. W. R. (1981)
J. Inorg. Biochem. 14, 127-1 34
Oxygen radicals and herbicide action
ALAN D. DODGE
School of Biological Sciences, Unit1ersit.v of Bath, Bath B A 2
7A Y, U.K.
The success of plants, and of weeds in particular, is due to their
ability to colonize and survive in a wide range of environments.
To achieve this they are endowed with extensive protective
devices ranging from structural to biochemical. Herbicides
initiate plant death in a variety of ways, but in many instances
they do so by overtaxing or destroying the biochemical
protective mechanisms which control toxic oxygen species, free
radicals or excess excitation energy.
The bipyridylium herbicides paraquat and diquat have been
widely used as rapid, ‘total kill’ herbicides since the 1950s.
Investigations have demonstrated that oxygen is important in
promoting the action of these herbicides (Mees. 1960: Van
Rensen, 1975) and is rapidly taken up after herbicide application
(Boger & Kunert, 1978). The restraining action of the
VOl. 10
photosynthetic-electron-transport inhibitor, monuron, on the
development of toxic symptoms (Mees. 1960; Dodge. 197 1)
indicates that an active electron-transport system is also
important. It is generally agreed that chloroplast electron flow is
diverted from the normal reduction of N A D P + to reduce the
paraquat ion, clearly demonstrated to accumulate in anaerobic
experiments with isolated chloroplasts (Zweig et al., 1965). In
the presence of oxygen. however, the univalently reduced
paraquat radical is reoxidized to generate superoxide (Farrington et a[., 1973) also been demonstrated in, as has
isolated chloroplasts (Elstner & Frommeyer, 1978, Youngmhn
et al., 1980). Superoxide and hydrogen peroxide may be
generated as a normal part of pseudocyclic phosphorylation in
chloroplasts, but scavanged by superoxide dismutase enzymes
(Asada et al., 1973: Hall & Foyer, 1980) and hydrogen
peroxide-removal systems (Foyer & Halliwell, 1976; Nakano &
Asada, 1980). It is assumed that these systems are totally
overtaxed in the presence of paraquat, and the superoxide
BIOCHEMICAL SOCIETY TRANSACTIONS
74
radical could reach a steady-state concentration of around
10-6~
up to 5 . m from the chloroplast grana (Farrington et al.,
1973). The sequence of cellular damage initiated by paraquat
involves the destruction of cellular membranes, such as the
tonoplast and plasmalemma (Harris & Dodge, 1972), thus
destroying cellular compartmentalization. This is followed by
the uncontrolled decay of cellular organelles, leading to the
typical chlorotic appearance of the leaf tissue. In parallel with
pigment breakdown there is an increase in ethane generation, a
secondary product of lipid peroxidation (Youngman et al.,
1979). Experiments have shown that the appearance of
phytotoxic symptoms is decreased when paraquat-treated leaves
are incubated with a superoxide scavenger such as copper
penicillamine (Youngman & Dodge, 1979). Furthermore,
paraquat-tolerant lines of Lolium perenne (rye-grass) (Harper &
Harvey, 1978) and Convza (Youngman & Dodge, 1982)
have been shown to possess higher activities of superoxide
dismutase.
The initiation of lipid peroxidation in the unsaturated fatty
acids of the cellular membranes may be accomplished by free
radicals such as hydroxyl or by singlet oxygen. It appears
possible that, in vivo, hydroxyl radicals might be produced from
superoxide in a Haber-Weiss type reaction (McCord, 1979).
Once the peroxidation reactions are initiated, a rapid deteriorative chain reaction will follow, possibly involving the reaction
of lipid peroxides with superoxide to produce hydroxyl radicals
(Peters & Foote, 1976)and also the generation of singlet oxygen
from lipid peroxides (Svingen et al., 1978).
Approx. 50% of all herbicides operate by inhibiting photosynthetic electron transport. These range from the ureas discovered in the 1950s (Wessels & Van der Veen, 1956) through
the S-triazines, uracils, triazinones, hydroxybenzonitriles,
pyridazinones to the acylanilides. In general there appears to be
one major site of action, a protein component located on the
outside of the thylakoid in the vicinity of Q, the quencher
of Photosystem 11, although the specific points of interaction
may vary from one herbicide group to another (Pfister &
Arntzen, 1979). The result of an inhibition of electron flow
is that CO, fixation will cease and plant starvation will follow.
However, many investigators have shown that the appearance
of phytotoxic symptoms is promoted by increased light intensity
(Ashton, 1965; Van Oorschot & van Leeuwen, 1974; Pallett
& Dodge, 1980). In addition, chlorophyll bleaching is retarded
when monuron- and ioxynil-treated leaves are incubated under
argon (Pallett & Dodge, 1980).
In a normal photosynthesizing plant, excitation energy
absorbed by the chloroplast pigments is utilized to drive electron
transport. When there is an excess of light energy, singlet
chlorophyll undergoes intersystem crossing to the longer-lived
triplet state, but the energy may be harmlessly dissipated by the
carotenoids. When electron transport is prevented by an
inhibitor herbicide, this protective device is overloaded and
gradually destroyed (Pallett & Dodge, 1980). Triplet chlorophyll
may induce cellular damage by type-I reactions in which electron
or hydrogen abstraction is induced from susceptible molecules
such as unsaturated fatty acids to yield lipid free radicals. In the
presence of oxygen, lipid peroxides are formed and general
membrane destruction will continue with the production of
secondary products such as ethane (Pallett & Dodge, 1980). In
addition to the type-I reactions, type-I1 reactions may occur as
the triplet chlorophyll interacts with triplet oxygen to generate
singlet oxygen. This, among other reactions, is capable of
initiating lipid peroxidation in unsaturated fatty acids of
membranes (Rawls and Van Santen, 1970). Experiments have
shown that the non-specific singlet-oxygen quencher DABCO
(diazobicyclo[2.2.2.loctane) restrained the action of the
herbicide monuron (Youngman et al., 1980).
Van Oorschot (1974) has likened the damage caused by
photosynthetic-inhibitor herbicides to that initiated when leaves
are incubated in the absence of CO,. Isolated chloroplast
incubated in the absence of an electron acceptor provide a useful
analogous system to this. Recent experiments have shown that
the breakdown of chlorophyll and linolenic acid, and the
production of both ethane and malonaldehyde in pea chloroplasts is retarded under nitrogen or when maintained in darkness
(Percival & Dodge, 1982). Furthermore, the non-specific singlet
oxygen quenchers DABCO and crocin retard the breakdown,
but on the other hand, the singlet-oxygen generator Rose
Bengal, immobilized on Sepharose, leads to enhanced
breakdown.
The role of carotenoids in protecting the photosynthetic
apparatus from excess excitation energy has already been
mentioned, and carotenoidless mutants are particularly susceptible to photo-oxidation (Anderson & Robertson, 1960). In these
experiments, chlorophyll destruction was decreased when the
mutants were illuminated under nitrogen. When normal photosynthesizing organisms are incubated under very high light
conditions, photo-oxidation is limited by the removal of oxygen
(Sironval & Kandler, 1958;Van Hasselt, 1972). In recent years
a number of herbicides and experimental herbicides have been
developed which have the inhibition of carotenoid biosynthesis as
a major site of action. These include the substituted pyridazinone
herbicides norflurazon and metflurazon, difunone, dichlormate
and aminotriazole. Although all of these herbicides appear to
have additional sites of action (Ridley, 1982), the major site is
on the desaturation reactions between phytoene and lycopene.
The chlorotic symptoms produced by these herbicides are due
therefore not to the failure of chlorophyll formation (A. D.
Dodge & M. C. Thompson, unpublished work), but to the greater
susceptibility of these pigments to destruction. Although no
workers have attempted to investigate the role of oxygen in
promoting the action of these herbicides, it is clear that the
generation of singlet oxygen by type-I1 reactions could be a
major feature of their action.
The primary site of action of the diphenyl ether herbicides
such as nitrofen and oxyflurofen is not clear, although there is a
requirement for light (Vanstone & Stobbe, 1979). Recent
experiments by Kunnert & Boger (1981) with oxyflurofen have
suggested that, as with the bipyridyls, light activation is due to
electron transport, which causes the reduction of the parent
compound to the anion radical. Although no light-induced
oxygen uptake was evident, these workers suggest that the
radical may react with oxygen to yield superoxide, thus
providing the initiating precursor of sufficient oxygen radicals to
promote membrane lipid peroxidation.
The toxicity of oxygen radicals is clearly of importance in the
action of many currently used herbicides. Future developments
in this area might involve the use of photosensitizing compounds
which promote the generation of toxic oxygen species. Rose
Bengal and other xanthine dyes are effective experimental
insecticides (Fondren et al., 1978), and Zweig & Nachtigall
(1975)have shown that the herbicidal properties of fluranthene
are possibly due to the u.v.-induced generation of singlet oxygen.
Anderson, I. C. & Robertson, D. S. (1960) Plant Physiol. 35,531-534
Asada, K., Urano, M. & Takahashi, M. (1973) Eur. J. Biochem. 36,
257-266
Ashton, F. M. (1965) Weeds 13,164-168
Boger, P. & Kunert, K. J. (1978) Z . Naturforsch C. 33,688-694
Dodge, A. D. (1971) Endeavour 30, 130-135
Elstner, E. F. & Frommeyer, D. (1978) FEBS Lett. 86, 143-146
Farrington, J. A., Ebert, M., Land, E. J. & Fletcher, K. (1973) Biochim.
Biophys. Acta 314,372-381
Fondren, J. E., Norment, B. R. & Heitz, J. R. (1978) Enuiron. Entomol.
7,205-208
Foyer, C . H. & Halliwell, B. (1976) Planta 133, 21-25
Hall, D. 0. & Foyer, C. H. (1980) in Chemical and Biochemical
Aspects of Superoxide and Superoxide Dismutase (Bannister, J. V.
& Hill, H. A. O., eds.), pp. 380-389, Elsevier/North-Holland, New
York
Harper, D. B. & Harvey, B. M. R. (1978) Plant Cell. Environ. 1,
211-215
1982
75
597th MEETING, LONDON
Harris, N. & Dodge, A. D . (1972) Planta 104,201-209
Kunert, K. J. & Boger, P. (1981) Weed Sci. 29, 169-173
McCord, J. M. (1979) in Reuiews in Biochemical Toxicology
(Hodgson, E., Bend, J. R. & Philpot, R. M., eds.), vol. 1, pp.
109- 124, Elsevier/North-Holland, New York
Mees, G. C. (1960) Ann. Appl. Biol. 48,601-612
Nakano. Y. & Asada, K. (1980) Plant Cell Physiol. 21, 1295-1307
Pallett. K. E. & Dodge, A. D. (1980)J. Exp. Bot. 31, 1051-1066
Percival, M. P. & Dodge, A. D. (1982) Proc. Int. Congr. Photosynthesis 5th in the press
Peters, J. W. & Foote, C. S. (1976)J. Am. Chem. Soc. 98,873-875
Ptister, K. & Arntzen, C. J. (1979) Z. Naturforsch C. 34,996-1009
Rawls, H. R. & Van Santen, P. J. (1970) J. Am. Oil Chem. Soc. 47,
12 1-1 25
Ridley, S. R. (1982) Proc. In?. Symp. Carotenoids 6rh. In the press
Sironval, C. & Kandler. 0. (1958) Biochim. Biophys. Acta 29, 359-368
Svingen, B. A., @Neal, F. 0. & Aust, S. D. (1978) Photochem.
PhotobioL 28,803-809
Van Hasselt, Ph. R. (1972) Acta Eot. Neerl. 21,539-548
Van Oorschot, J. L. P. (1974) Weed Res. 14,75-79
Van Oorschot, J. L. P. & Van Leeuwen, P. H. (1974) Weed Res. 14,
8 1-86
Van Rensen, J. J. S. (1975) Physiol. Plant. 33,42-46
Vanstone, D. E. & Stobbe, E. H. (1979) Weed Sci. 27,88-90
Wessels, J. S. C. & Van der Veen (1956) Biochim. Biophys. Acta 19,
548-549
Youngman, R. J. & Dodge, A. D. (1979) Z. Naturforsch C. 34,
I03 2- 1035
Youngman, R. J. & Dodge, A. D. (1982) Proc. In?. Congr.
Photosynthesis. 5th in the press
Youngman, R. J., Dodge, A. D., Lengfelder, E. & Elstner, E. F. (1979)
Experientia 35, 1295-1 296
Youngman, R. J., Pallett, K. E. & Dodge, A. D. (1980) in Chemical
and Biochemical Aspects of Superoxide and Superoxide Dismutase
(Bannister, J. V. & Hill, H. A. O., eds.), pp. 4 0 2 4 1 1. Elsevier/
North-Holland, New York
Zweig, A. & Nachtigall, G. W. (1975) Phorochem. Phorobiol. 2 2
257-259
Zweig, G., Shavit, N . & Avron, M. (1965) Biochim. Biophys. Acra 109,
3 3 2-346
Regulation of manganese superoxide dismutase by oxygen in Succhuromyces cerevisiae
ANNE P. AUTOR
Department of Pharmacology, University of Iowa, Iowa City, I A
52242, U S A .
Saccharomyces cereuisiae, in common with all other eukaryotic
cells, contains two forms of superoxide dismutase, which,
although identical in enzymic activity, are widely dissimilar in
structure. Manganese (Mn) superoxide dismutase located in the
mitochondrial matrix and copper/zinc (Cu/Zn) superoxide
dismutase in the cytosol are different in molecular weight,
sub-unit number and amino acid composition (Weisiger &
Fridovich, 1973; Johansen et al., 1979; Harris et al., 1980). The
synthesis of both enzymes is controlled by nuclear DNA. Little
else is known about the regulation of these enzymes, although
recently the synthesis of Cu/Zn superoxide dismutase was
reported in a cell-free rabbit reticulocyte lysate system programmed with human placental mRNA (Bannister et al., 1980).
Both forms of the enzyme are induced by hyperoxia. Much
experimental evidence demonstrating the regulatory effect of
elevated 0, has been obtained from the study of prokaryotes
that also contain Mn superoxide dismutase (Gregory &
Fridovich, 1973) and simple eukaryotes (Asada et al., 1976). By
manipulating growth conditions, it has been concluded that one
or more metabolic products of 0,, rather than molecular 0,
itself, induce the synthesis of the enzyme in single-cell
organisms (Hassen & Fridovich, 1977, 1979). 0, or its
metabolites also induce both Mn and Cu/Zn superoxide
dismutase in S. cereuisiae (Gregory et al., 1974; Autor, 1981a)
and mammalian cells, particularly those of lung tissue (Crapo &
McCord, 1976; Stevens & Autor, 1977).
Growth and maintenance of cells under normoxia results in a
small but continuous production of partially reduced oxygen
radicals including superoxide anion (O,-'), H,O, and an oxygen
moiety at the redox level of hydroxyl radical (HO') (Fridovich,
1978). Available evidence to date indicates that HO' is formed
by a non-enzymic but iron-catalysed reaction between 0,-'
and
H,O, (Halliwell, 1981). Furthermore, the rate of generation of
these partially reduced oxygen species appears to increase when
organisms are exposed to hyperoxic conditions (Fridovich,
1978). Because it is one of the most powerful oxidizing agents
known, HO' is accepted as the most likely candidate for the
cytotoxic agent associated with 0,. Removal of the reactants
that produce this radical would be a necessary step in
detoxification. 'Superoxide dismutase and catalase catalytically
eliminate O,-* and H,O, respectively. Glutathione peroxidase
also uses H,O, as a substrate but its activity is not discussed
here.
Variability in the intracellular content of superoxide
VOl. 10
dismutase (and catalase) has consequences for the ability of the
affected cell to withstand oxidative stress. Work from this
laboratory has shown that a 4-fold increase in intracellular Mn
superoxide dismutase activity (and concomitantly in catalase) in
rat pulmonary cells results in a much higher LD,, (median lethal
dose) for ionizing radiation and autoxidizing dihydroxyfumarate, both of which manifest their toxicity through oxygen
radicals (Autor et al., 1979). Furthermore, pulmonary cells,
which induce both enzymes under hyperoxic conditions, also
under hyperoxia are capable of a higher than normal rate of
generation of oxygen radicals from their mitochondria. Thus in
this and other eukaryotic-cell systems elevated oxygen-radical
flux and elevated superoxide dismutase activity (Autor &
Stevens, 1980; Dryer et al., 1980) may be related in a manner
that resembles substrate induction of enzymes, and further, this
elevated enzyme activity appears to be important in providing
resistance to oxygen-radical toxicity.
0, control of the biosynthesis of other mitochondrial enzymes
has been studied in S. cerevisiae by measuring both the catalytic
activity and the content of immunoreactive polypeptides in
anaerobically versus aerobically grown yeast. Anaerobically
grown yeast do contain mitochondria but they are devoid of
cytochromes and are termed promitochondria (Schatz &
Kovac, 1974). Of the mitochondrial enzymes studied thus far
that are regulated by 0,, i.e. cytochrome oxidase (Woodrow &
Schatz, 1979), cytochrome c, cytochrome c peroxidase
(Djavadi-Ohaniance et al., 1978) and catalase (Zimniak et al.,
1976), all contain haem as a prosthetic group. Anaerobic yeast
cells do not synthesize haem; therefore the absence of
haem-containing proteins in anaerobiosis has been attributed to
the absence of the prosthetic group (Woodrow & Schatz, 1979;
Barlas et al., 1979). Mn superoxide dismutase, also regulated by
0, but containing only manganese as a prosthetic group,
presents an interesting exception.
Since so little is known about the biosynthesis and regulation
of Mn superoxide dismutase in eukaryotes and the influence of
0, on the process we undertook a series of studies to investigate
its site of synthesis as well as the nature of the initial gene
product. The effect of 0, and haem on the synthesis of Mn
superoxide dismutase was studied by comparing wild-type yeast
(D 10B-273) grown aerobically with anaerobically grown wildtype yeast and with a haemless mutant (Autor, 1981b). Mn
superoxide dismutase from S. cereuisiae is composed of four
identical 24 000-mol.wt. subunits (Ravindranath & Fridovich,
1375). All analytical procedures in these studies were based on
analysis and comparison of the polypeptide subunits separated
by sodium dodecyl sulphate/polyacrylamide-slab-gelelectrophoresis and localized by reaction with antisera prepared from