Radical reactions in vivo - an overview

Radiat Environ Biophys (1990) 29:249-262 Radiation and Environmental Biophysics © Springer-Verlag 1990

Radical reactions in vivo - an overview*

M. Saran and W. Bors GSF-Institut ffir Strahlenbiologie, Ingolstfidter Landstrasse 1, D-8042 Neuherberg, Federal Republic of Germany

Received May 4, 1990/Accepted in revised form June 21, 1990

Summary. Generation of radicals in vivo depends on metabolic activities. The reactions are usually influenced by (i) the presence and concentration of oxygen; (ii) the availability of transition metals (effects of binding and compartimentali- zation); (iii) the level of reductants and antioxidants (e.g. nutritional effects). The effects of radicals are thought to be due to (i) membrane damage (affecting passive or active transport through altered fluidity/function interrelationships, intercellular messenging through modifications in the synthesis of prostaglandins and leukotrienes); (ii) protein damage (e.g. affecting membrane transporters, channel proteins, receptor or regulatory proteins, immunomodulators); (iii) damage to DNA. Defense mechanisms consist of (i) prevention of the 'spreading' of primary damage by low molecular weight antioxidants (e.g. vitamin E, GSH, vitamin C, fi-carotene, uric acid); (ii) prevention or limitation of 'secondary' damage by enzymes (e.g. GSH-peroxidase, catalase, superoxide dismutase, DT-diaphor- ase) and/or chelators; (iii) repair processes, e.g. lipid degradation/membrane repair enzymes (phospholipases, peroxidases, some transferases and reductases), protein disposal or repair enzymes (proteases, GSSG-reductase), DNA degradation repair enzymes (exonuclease III, endonucleases III and IV, glycosylases, polymerases). Recent hypotheses on a messenging function of the superoxide anion O2 are discussed and possible implications of cross-reactions between O2 and nitric oxide (endothelium-derived relaxing factor EDRF) are shortly mentioned.

Introduction

Precise predictions of mechanisms of radical reactions are only possible for well defined chemical systems. Any general discussion of in vivo radical reactions is hampered by the existence of boundaries between cellular compartments having diverse structural and metabolic features. This does not allow to treat the

* Paper given at the workshop "Molecular Radiation Biology". German Section of the DNA Repair Network, Mfinchen-Neuherberg, 21.-23.3.90

250

cell as a homogeneous entity and any statement being true for one compartment might not hold for another. The only way to arrive at a crude description of radical processes within the living cell is to clarify at any instance whether the effects observed in vivo can be interpreted by results obtained in model systems under controlled conditions.

R a d i c a l r e a c t i o n s i n v i t r o

In order to facilitate the understanding of radical reactions in vivo some general features of radical reactions should be shortly recalled. 1) By virtue of the fact of having an unpaired electron, i.e. a 'half-open' chemical bond, radicals are usually very reactive. 2) The reaction of a radical with any other molecule (usually having an even number of electrons) will lead to a new radical, which means, that radical reactions by their very nature are self perpetuating chain processes. 3) Radicals may be produced by a) homolytic bond cleavage: R 1 - - R 2 - ~ ' R I + ' R 2 (because of being highly

endergonic, this process is very unlikely in vivo); b) electron 'withdrawal' , e.g.

- oxidation by other radicals: RH 2 + ' O H ~ "RH + H20, - radiation: RHz -~ "RH + H ÷ + e - ;

c) electron 'donation ' , e.g. through electron 'leaks' (vide infra): O 2 + e - ~O~-;

d) catalyzed electron transfer to peroxidic compounds: M(, - 1)+ + H202 __. Mn÷ + "OH + O H - (Fenton reaction), M ("-1)÷ + R O O H - - * M ~÷ + R O ' + O H - (°organic' Fenton reaction), (M stands for complexed transition metals in various oxidation states).

4) Oxygen is involved in most radical reactions because of the following features: a) it is a bi-radical (having two unpaired electrons), b) it is in a triplet ground state (the spins of these two lone electrons are

parallel). Two important aspects can be deduced from these facts: (i) oxygen is especially reactive towards other radicals (it has °half open' chemical bonds itself), (ii) oxygen has to be activated before it can enter into a reaction with a normal molecule (since reactions between singlet states - which is the general configura- tion of almost all possible reaction partners of oxygen - and triplets are 'spin- forbidden'). In principle three different ways exist to circumvent the restrictions set by the spin conservation rule: - excitation of oxygen to a singlet state (~Oz), - 'partial abolition' of the triplet character of oxygen by having electrons shared with complexes of transition metals (Martell and Sawyer 1988 and references therein), - introduction of oxygen into radical processes. Whereas the significance of ~Oz in vivo is still a matter of debate (Krinsky 1984; Kanofsky 1989), the latter two processes certainly do occur. 5) The propagation of radical reactions is affected by several parameters: If a radical is generated somewhere in an aerobic solution of biomolecules, peroxyl (ROO') and alkoxyl radicals (RO') are very likely intermediates in the

251

~ ~ _ _ ~ CHEMOATTRACTANTS PLATELET AGGREGATION, CELL ADHESION PROSTAGLANDINS,LEUKOTRIENES/ ~ FA-OOH /~ LDL - MODIFICATION, VASOCONSTRICTION

-- PORTER PROTEINS

/ ~ ~ :~ ~ (- MITOCHONDRIAL 0 2 O~ O~ RESPIRATION

I I .-Q_ . NAD~ E~DO~

/ IW~ I \ ~ ~ ~ . COMPARTMENT i I ~°2<, I ? RooM...o.. o:. FA-O0.. H,O, O: OF"

/ I . . . . . V ~ / 'PRIMARY' / ~ REACTIONS

ENZYMES OF ANTIOXIDA~IVE DEFENSE: CATALASE. GSH - PEROXIDASE, SUPEROXIDE DISMUTASE ETC. l

NUCLEUS

'SITE SPECIFIC' CHEMISTRY " " ° / ~ 'FDNTON'

CHEMISTRY ( "OH, "~' %..~ ) ~d

COMPARTMENT OF 'SECONDAR~ RADICAL REACTIONS

Fig. 1. The scheme roughly symbolizes a cell having: an extracellular space (containing serum, hormones, blood particles); - a membrane (containing phospholipids and functional proteins); - a cytosolic space (containing metabolically active compartments, proteins, antioxidants); - a hypothetical compartment of'secondary' (metal dependent) reactions

chain reactions that proceed and among the most abundant end products will be peroxides (ROOH and HzO2) and hydroxylated (ROH) or carbonyl (HR = O) products. The spatial dimension of the ' cluster ' in which these reactions proceed depends on the mean diffusion path of the radicals involved and on the chain length of the pertinent reactions. For highly reactive radicals it will be small in terms of cellular distances (Saran et al. 1988), only rather unreactive species like O~- or stable products like peroxides and aldehydes will be able to move for example the distance from one organelle to the other. Yet if a radical is generated in vivo through a Fenton-mechanism, it owes its life to a stable product of a former radical reaction. This means that in vivo a primary radical reaction might manifest itself, after some time and at a rather distant location, by initiating a secondary radical chain (Benedetti et al. 1979).

Sources of radicals in vivo

The scheme (Fig. 1) might serve as a guideline for the discussion. For the special case of sparsely ionizing radiation - dashed arrow in left part of Fig. 1 - it is evident that radical sites can be produced anywhere in the cell. Yet a gross estimate based on the assumption of 1 hit per cell occurring per year by the average natural background radiation of 1 m G y (100 torero) (Feinendegen et al. 1988) as opposed to an estimated number of at least 104 'oxidat ive ' hits per day occurring on any cell's D N A (Ames 1989), makes radiation under normal

252

conditions a rather unimportant side process for in vivo radical generation. The dominant part of cellular radical reactions is supposed to be due to processes for which different descriptive terms have been coined (right part of the scheme): a) 'radical leakage': in the course of the four electron reduction of oxygen to water in mitochondrial respiration the two formal states 02 and 032- directly possess radical character and the intermediate state 0 2. (i.e. H2Oa) bears the potential to generate radicals through Fenton reactions. Radical leakage refers to the idea of some of these radical states accidentally crossreacting with the environment and initiating radical chains there. b) 'electron leakage': one-electron redox processes, e.g. NAD(P) H-dependent redox cycling between quinoid and semi-quinoid states occurring in metabolism of xenobiotics, are prone to accidental electron transfer to suitable acceptors and thus will be able to initiate radical chains. A convincing example of such a process is the plant chloroplast where ferredoxin, which channels electrons from photosystem II to the plant's reductive pathways, will inevitably transfer electrons directly to 02 as soon as the pool of available NAD ÷ is fully reduced. If the induced radical chains are not intercepted by low molecular weight scavengers (see below) or come to an end by chain terminating reactions such as radical recombination and dismutation, hydroperoxides are produced which might reach a metal containing site and initiate secondary radical reactions through a Fenton process unless they are disposed by enzymes such as catalase or GSH-peroxidase. Thus mainly two regions have to be considered as origins for intracellular radicals: metabolically active centers and the vicinity of metal-storing structures or metal binding sites.

Reactions of radicals in vivo

If a radical process spreads within a cell, low molecular weight antioxidants may interfere with the chain reaction by donating H-atoms to radicals. This results in 'restitution' of the original radical site. The antioxidant, however, has now become a radical itself and only if it is stable enough to slow the chain process down to such an extent that it can await being 'healed' by H-atoms derived from metabolism, the process will be truly chain-breaking. For the antioxidants depicted in the left part of the scheme such a sequence - involving vitamin E, vitamin C, GSH and H-atoms from NAD(P)H - has been shown to operate in vitro (Niki 1988).

At present it is still not possible to give an exact description of radical reactions within cells by putting together the huge body of data that has accumu- lated on model systems. It seems feasible, however, to group some of the findings together into some 'plausible scenarios': A) Metabolism/membranes/antioxidants B) Metal-dependent reactions C) Proteins and cytosolic reactions D) Processes involving the extraceUular space E) DNA, genetic changes, carcinogenesis. (References may be found e.g. in Slater 1984; Wolff 1986; or in recent congress- reports: Armstrong et al. 1984; Rotilio 1986; Elstner et al. 1987; Simic et al. 1988; Singal 1988; L'Abbate and Ursini 1989).

253

Fig. 2. Model for the removal of peroxidized fatty acids from the membrane by the Ca z +-dependent enzyme phospholipase A2 (PLaseA2) and subsequent disposal of the hydroperoxide by the antioxidative enzyme glutathione peroxidase (GSH-Px). Adapted from van Kuijk et al.; TIBS (1987)

P L a s e A z + C a 2+ + H z O ~- ~ J

C 0 I C 0

C-O-C C-O-C

o

H I o I

0 O~ GSH- Px+OSH

C-OH C - 0 - C

I

° i 0=C 0=

l Oxidized Bilayer with Oxidized Fatty Acid Disrupted St ruc ture Removed for Reaction

with GSH-Px

Scenario A. Metabolism/antioxidants/membranes

These topics are intimately interconnected and in fact cannot be treated separate- ly. Any metabolic activity seems to be reflected by changes in the antioxidant pool. As has been shown, even the moderate effort of riding a bicycle leads to a marked decrease in the content of reduced GSH and a concomitant rise in oxidized GSSG (Gohil et al. 1988). Pentane exhalation during high-altitude mountain-climbing, which is an indication for radicals having reacted with membranes, can be greatly decreased by a diet which is enriched in the membrane- protecting antioxidant a-tocopherol (Simon-Schnass et al. 1988). Furthermore, after induction of artificial oxidant stress conditions in blood plasma, it could be established that the pool of cellular antioxidants was used up in the following sequence: vitamin C - SH groups - bilirubin - urate - ~-tocopherol and that increasing amounts of hydroperoxides (triglyceride-OOH, phospholipid-OOH, cholesterol-OOH) appeared as soon as the pool of reduced ascorbate had van- ished (Frei et al. 1988). The interrelationship between antioxidants and membrane damage seems to be twofold (symbolized in the membrane part of Fig. 1): almost any radical site being induced somewhere in the phospholipid matrix will after intra- and intermolecular transfer first be stabilized as a pentadienyl radical around the bisallylic position of a polyunsaturated fatty acid (Patterson and Hasegawa 1978). This radical site will then either be transferred in lateral direction to a neighbouring molecule in something like a ' relay ' mechanism (Ivanov 1985) or eventually add oxygen to become a peroxyl radical. Interception of these chains by the membrane bound chain breaker a-tocopherol and the 'handing over ' of the radical site to soluble antioxidant compounds like ascorbate or GSH is an example of the first category of membrane/antioxidant interrelationships. The fate of the radicals that escape this line of defense and end up as lipid hydroperoxides is exemplified in Fig. 2 (van Kuijk et al. 1987). They are cut out by phospholipase Az and thereby become accessible

254

MECHANISMS OF METAL-DEPENDENT RADICAL PROCESSES

1) SITE OF ACTION 'NON-SPECIFIC' e,g, FENTON REACTIONS IN SOLUTION BY FREELY DIFFUSIBLE METAL-COMPLEXES as USED FOR DNA-FOOTPRINTING (see Lext)

2) SITE OF ACTION DETERMINED BY PROTEIN MOIETY OF RADICAL GENERATING AGENT

3a) SITE OF ACILON DETERMINED BY METAL BINDING CAPACITY OF TARGET SITE

METAL IS PART OF 'GENERATOR'

3b) SITE OF ACTION DETERMINED BY METAL BINDING CAPACITY OF TARGET SITE - METAL IB 'NOT PART OF GENERATOR',

i.e. COMES FROM OTHER SOURCES

~'e(ED TA17 +2~42 O2

~ "OH

O 2 " ~ ~

4) SITE OF ACTION DETERMINED BY 'RESULT OF PRECEEDING OXIDATION'

HYDROPEROXIDE EITHER PREFORMED BY 'UNSPECIFIC' OR BY METAL- DEPENDENT ke. 'SPECIFIC' PROCESS

Fig. 3. Models for metal-ion dependent enzymatic and non- enzymatic reaction mechanisms (see text)

to detoxification by the GSH peroxidase system. Clearly in this case the interplay between the membrane and the antioxidant defense is more complex than in the first example. Aside from the availability of the substrate GSH, several enzymatic steps are involved and it can easily be visualized that after excessive membrane oxidation both lines of defense may be overridden and the cell may be flooded with hydroperoxides which are able to induce secondary °Fenton- damage'. It is evident that the interdependencies in this scenario are so intricate that they are virtually non-separable under in vivo conditions.

Scenario B. Metal-dependencies

It is hypothesized that almost all in vivo radical reactions are metal catalyzed, i.e. (i) depend on the availability or binding of metal ions, (ii) can be influenced by chelation, and (iii) are inherently 'site-specific'. Figure 3 depicts several mechanistic possibilities for metal dependent processes. Mechanism 1) clearly is an artificial example. It shows that masking DNA by protein (the shaded bar above the symbolized DNA) or binding DNA to calcium phosphate (grey bar below DNA) greatly modifies the pattern of DNA fragmen- tation by the Fenton reagent FeEDTA + H202 (Tullius et al. 1987). Mechanisms 2) to 4) have been shown to occur under physiological conditions. All of them

255

fulfill the requirement to be sensitive to chelation. The main difference between them lies in the availability of metals and their respective binding sites. Cases 2) and 3 a), merely differing by the mechanism by which site specificity is determined but which do not depend on the availability of 'free metals' seem to be realized in some biological hydroxylation systems (multifunctional oxidases; Hamilton 1974; Stadtman 1986 and references therein). The most convincing examples for processes analogous to case 3 b) come from radiation biological studies where it could be shown that damage by radiation induced or experimen- tally added peroxides is 'site directed', e.g. predominantly occurs at the sites in DNA where metals are bound (Samuni et al. 1983; Pr/itz 1984; Aboul-Enein and Schulte-Frohlinde 1988). Mechanism 4) can be visualized as the 'reverse' of case 3b. Whereas in the latter the site of action is determined by the site where the metal is bound, in the first case the 'action' will be in close vicinity to the position of the hydroperoxide, which is induced to decay by metal catalysts coming close to it. Some protein degradation processes (see below) seem to operate via such a mechanism.

One of the central questions in this context concerns the pattern of metal binding within the cell. None of the relevant transition metals can actually be regarded as 'free', the concentration and bioavailability are carefully controlled. For iron the transport protein transferrin (binding two metal atoms) as well as the storage granules ferritin (containing up to 4500 molecules of Fe(III)) are known. According to investigations by Aust and others (Samokyszyn et al. 1989; Gutteridge 1985) the purpose of the 'ferro-oxidase' enzyme ceruloplasmin is to store oxidized iron in ferritin from where it can be released by yet not fully elucidated reductive processes. In vitro experiments have shown that O~- is able to release iron from this store (Koster and Slee 1986; Buettner et al. 1986). This means that a radical would not only be responsible for the release of a bound metal but that it would also be involved in subsequent site-specific radical reactions considering the fact that the metal will immediately be chelated at a suitable site in its environment. The whole complexity of in vivo metal dependencies is evidenced by experiments with antineoplastic drugs. Adriamycin, for example, has no effect without Fe or Cu ions. It is such a good chelator, however, that it is able to extract iron even from ferritin or transferrin. The adriamycin/iron-complex binds to membranes or DNA and catalyzes lipid peroxidation and HzOa-dependent DNA damage. If the iron is extracted from the complex by the still more potent chelator ICRF-187 the binding of adriamycin vanishes and the cardiotoxicity disappears (Rajagopalan et al. 1988 and references therein). Another important question concerns the details of the mechanism of the Fenton reaction. Even though it has been under investigation for more than 90 years since its discovery by Fenton (1894) and has been 'revisited' several times (Wall- ing 1975; Cohen and Sinet 1980; Stadtman and Berlett 1988) there are still some unexplained features. The mechanism depends on the one hand on the accessibility of the bound metal to reduction (Buettner et al. 1983) and is gov- erned by the availability of a free, water accessible coordination site in the complex (Graf et al. 1984). On the other hand the Fenton reaction per se, the transfer of an electron from the reduced metal to peroxide is far from being entirely understood. Several observations indicate that no free OH radicals are produced (Koppenol 1985) but rather act in bound form (as crypto OH species

256

(Youngman and Elstner 1981; Youngman 1984)) in some sort of a cage reaction. The observation that the oxygen which is incorporated in the product phenol if benzene is exposed to a Fenton reaction originates from both hydrogen peroxide and molecular oxygen suggests a complex ternary interplay between metal- complex, oxygen and peroxide (Cohen and Ofodile 1983). Even though it is at present very difficult to even roughly calculate the extent to which metal dependent reactions are responsible for in vivo radical damage, the general statement that they are inextricably involved is in accord with all experimental evidence.

Scenario C. Proteins and cytosolic reactans

The amount of oxidized protein that can be found within a cell depends on the following parameters:

- partial pressure of 02, - concentration of Fe or Cu ions, - scavengers of activated oxygen species, - concentration of metal chelators, - dietary intake of vitamins, - age. All these parameters (at least the first five) suggest that metal dependent oxygen radical processes are the cause of protein oxidation. Two additional findings merit special attention: first, the suggestion that proteins are 'marked' for pro- teolysis by oxidative steps (Stadtman 1986; Davies 1987), and second, the finding that the cellular content of oxidized protein is age-dependent (Oliver et al. 1987; Ames 1989). According to observations by Stadtman, the oxidation of amino acid residues in proteins occurs predominantly at metal binding sites (e.g. the histidine residues of glutamate dehydrogenase) and recently Davies and co- workers (Pacifici et al. 1989) described an enzyme which is triggered to degrade the respective protein if it detects changes in hydrophobicity due to oxidation- dependent unfolding. The function of this enzyme is independent of energy supply by ATP and independent of Ca 2 ÷ but strongly influenced by the presence of chelators. It is not yet entirely understood which of the mechanisms quoted in the previous chapter pertains to these proteinase functions. The observed age dependency of the overall status of protein oxidation adds to the complexity of in vivo radical reactions. In principle, the age dependency could be due to: i) a decrease in metabolic rate, ii) decreased or less effective defense by low-molecular weight antioxidants, iii) decreased or less effective defense by antioxidant enzymes, iv) generally increased 'leakage' processes (e.g. from metabolic centers) (perhaps due to general membrane deterioration?) v) decreased activity of disposing proteases (or other clearance systems), vi) decreased or less effective control of metal storage, vii) age-dependent changes at the genetic level, e.g. accumulation of mutations which dis-equilibrate the cellular balance to become responsible for one of the above observations. For each of these dependencies there is experimental support. Although it is not possible at present to give an estimate of the relative contributions of any

257

Fig. 4. Model for the possible involve- ment of fatty acid hydroperoxides (FA- OOH), low density lipoprotein particles (LDL), and macrophages/monocytes in the genesis of atherosclerosis

( F A - O O H )

EXCRETION?)

p,,~PLATELETS

MICROLESION TO VASCULAR ENDOTHELIUM

PLATELET AGGREGATION - SECRETION OF PDGF - PROLIFERATION OF SMC

~ LDL

o - ACCUMULATION O F LDL PARTICLES

- INVASION OF MONOCYTES

MONOCYTES

, ¶ ~ ) 2 ) ) LDL

- OXIDATIVE DAMAGE TO LDL ~ "J~PHOSPHO (PL AND/OR APOPROTEIN) o-- ~UPID$

e" CHOLESTERYL - INCREASED INGESTION BY , ESTER

MACROPHAGES

o B-lOG

AND ATHEROSCLEROTIC PLAQUES ~-. MEMBRANE COOH

of them to the overall process of protein oxidation, changes on the level of DNA are most probably involved. It would be difficult to visualize changes in metabolism persisting for long periods of time that are not controlled by the genome.

Scenario D. Processes involving the extracellular space

Analogous dependencies of 'damage' on the concentrations of oxygen, metals, scavengers and chelating substances as discussed above for the cell's interior also exist in the extracellular space. Tissue damage due to reperfusion after ischemia is one of the most prominent phenomena and has been reviewed several times (Darley-Usmar et al. 1989 and refs therein). Figure 4 shortly summarizes another example. The processes depicted are thought to be responsible for atherosclerotic degenerations of the vascular system (Yagi 1986). They give an idea to what extent oxidative processes on a molecular basis (alterations of phospholipids e.g. in endothelial cells or low density lipoprotein particles) might lead to consequences at the level of intercellular relationships (deteriorated recog- nition of LDL by receptors, increased phagocytosis through generation of ehe- moattractants). Involving a series of physiological reactions (emission of hor- monal messengers, growth factors etc.), they can finally manifest themselves in permanent modifications of tissues or whole organs and thus be causative for disorders such as apoplexias, myocardial infarctions and aging in general terms (see Singal 1988; L'Abbate and Ursini 1988).

258

Scenario E. DNA, genetic changes, carcinogenesis

Oxidative alterations dependent on age are also found in DNA. They are partial- ly repaired by excision repair as evidenced by the urinary excretion of DNA fragments such as thymine glycol, thymidine glycol, hydroxymethyluracil, 8- hydroxyguanosine (Ames 1989). The product distribution closely resembles the pattern found after in vitro irradiation of DNA under aerobic conditions and suggests the DNA lesions being due to direct action of oxidants. This poses two questions: first, do the oxidants come from either the mitochondria or the cytoplasm and cross the nuclear membrane, or are they generated in the nuclear membrane or possibly within the nucleus itself? The second point resembles the philosophical 'egg or hen' question who is first: is DNA first oxidatively altered by some stochastic process which later on leads to changes in the protein/ cytosol realm or are (stochastic?) changes at the periphery of the cell, i.e. outside the nucleus, causative for DNA damage? The experimental findings that the amount of DNA lesions increases with metabolic rate (Cutler 1986), is diminished after increasing the supply of dietary antioxidants (Summerfield and Tappel 1984) or depends on the cellular level of antioxidative enzymes (Cerutti 1985; Egner and Kensler 1985) and can be influenced by chelation, at first sight seem to support the idea that oxidants originate from the extranuclear space and interact with DNA after having crossed the nuclear membrane. Yet there are also many observations suggesting interrelationships between prooxidant conditions and nuclear damage which can not be due to direct interaction. These rather support the view that oxidative reactions in the extranuclear space influence gene expression via common regulatory pathways, e .g . involving G-protein-mediated signal transduction, Ca 2 +-dependent kinase reactions and corresponding changes in oncogene activation (Cantoni et al. 1989; Shibanuma et al. 1988). Since this complex matter is beyond the scope of this article it is only symbolically depicted in the lower right part of Fig. 5 but will not be discussed in more detail.

Future aspects: the messenger hypothesis

The preceeding chapters have shown that oxygen radicals may be toxic or dele- terious. This is certainly correct for highly reactive species like hydroxyl radicals yet perhaps not for 0 2 . The latter proved to be rather unreactive chemically, it is not even able to initiate lipid peroxidation (Bors et al. 1980; Bielski et al. 1983). There is, however, growing evidence that besides phagocytes (for which 0 2 -production has been known for many years) many other cells produce superoxide ions at very low concentration levels (Heinecke et al. 1987; Steinbrecher 1988). It has been hypothesized that this O~ production might have the teleologi- cal purpose of functioning as a 'message'. (Vanhoutte and Katusic 1988; Saran and Bors 1989; Halliwell 1989). Under this aspect 02 has been proposed to be correlated to reactions of nitric oxide, the endothelium-derived relaxing factor EDRF, which is another stable radical serving as an in vivo messenger (Palmer et al. 1987). The physiological target of NO is the heme of soluble guanylate cyclase (Ignarro et al. 1986), which regulates, via the second messenger cGMP, a great variety of cellular processes. The possibility outlined in Fig. 5 that 02

259

VASCULAR TONE ONO0- PLATELET AGGREGATION

~--<-------~ CELL ADHESION cGbIP~ LDL- MODIFICATION

GUANYLATE )<-- < CYCLASE~ ,as

PHORBOL NO (EDRF) O~ (EDCF2)? ESTER

S,TR~c O~Dr i +~ '~'~ SYNTHETASE ~ (~ )+n +

PHOSPHORYLATED PROTEINS

i" T v

PROTEIN ~ RNA ~- DNA

Fig. 5. Middle and right part: mechanism of 0 2 production in phagocytes (rounded shapes symbolizing inactive, rectangular shapes active forms of the respective protein). Attachment of a chemotactic protein CP to its receptor protein activates the latter to trigger, via a GTP- binding protein (G), the enzyme phospholipase C (PLC) to convert phosphatidylinositol (PI) to inositoltriphosphate (IP3) and diacylglycerol (DG). The 1P3-mediated Ca2+-release from internal stores together with DG activates the enzyme protein kinase C (PKC) to phosphorylate the NADPH-oxidase complex to initiate Of-production (adapted from Badwey and Karnovsky 1986). The left part outlines the generation of nitric oxide (NO) from arginine by the enzyme nitric oxide synthetase (Bredt and Snyder 1990). Symbolized in the extracellular space is the hypothesis that either the excreted NO or an excreted Of activate or deactivate soluble guanylate cyclase to produce the second messenger cyclic guanosine monophosphate (cGMP)

ei ther reacts wi th N O to p r o d u c e pe roxon i t r i t e (Saran et al. 1990) o r in terac ts d i rec t ly wi th the soluble guany la t e cyclase (sGC) (Ullr ich et al. 1989) is cu r ren t ly unde r inves t iga t ion . The conjec ture tha t rad ica l s (at least the supe rox ide anion) migh t pa r t i c ipa t e in b io logica l messenging processes opens in teres t ing new areas for fur ther research.

References

Aboul-Enein A, Schulte-Frohlinde D (1988) Biological deactivation and single strand breakage of plasmid DNA by photoensitization using tris (2,2'-bipyridyl)ruthenium(II) and peroxo- disulfate. Photochem Photobiol 48:27-34

Ames BN (1989) Endogeneous oxidative DNA damage, aging and cancer. Free Rad Res Comm 7:121 128

Armstrong D, Sohal RS, Cutler RG, Slater TF (eds) (1984) Free radicals in molecular biology, aging and disease. Raven Press, New York

Badwey JA, Karnovsky ML (1986) Production of superoxide by phagocytic leukocytes: a paradigm for stimulus-response phenomena. Curr Top Cell Regul 28:183-208

Benedetti A, Casini AF, Ferrali M, Comporti M (1979) Effects of diffusable products of peroxidation of rat liver microsomal lipids. Biochem J 180:303-312

Bielski BHJ, Arudi RL, Sutherland MW (1983) A study of the reactivity of HO2/O 2 with unsaturated fatty acids. J Biol Chem 258:4759-61

Bors W, Saran M, Czapski G (1980) The nature of intermediates during biological oxygen activation. In: Bannister WH, Bannister JW (eds) Biological and clinical aspects of superoxide and superoxide dismutase. Elsevier/North-Holland, New York, pp 1-31

260

Bredt DS, Snyder SH (1990) Isolation of nitric oxide synthetase, a calmodulin-requiring enzyme. Proc Natl Acad Sci USA 87:68~685

Buettner GR, Doherty TP, Patterson LK (1983) The kinetics of the reaction of superoxide radical with Fe(III) complexes of EDTA, DETAPAC and HEDTA. FEBS Lett 158:143-146

Buettner GR, Saran M, Bors W (1986) The kinetics of the reaction of ferritin with superoxide. Free Rad Res Comm 2:369-372

Cantoni O, Sestili P, Cattabeni F, Bellomo G, Pou S, Cohen M, Cerutti P (1989) Calcium chelator Quin 2 prevents hydrogen-peroxide-induced DNA breakage and cytotoxicity. Eur J Biochem 182:209-212

Cerutti PA (1985) Prooxidant states and tumor promotion. Science 227:375 381 Cohen G, Ofodile SE (1983) Activation of molecular oxygen during a Fenton reaction: A

study with 180 2. In: Cohen G, Greenwald RA (eds) Oxy radicals and their scavenger systems, vol h Molecular aspects. Elsevier, New York, pp 74-83

Cohen G, Sinet PM (1980) Fenton's reagent once more revisited. In: Bannister JV, Hill HAO (eds) Chemical and biological aspects of 02 and SOD. Elsevier/North-Holland, New York, pp 27-37

Cutler RG (1986) Aging and oxygen radicals. In: Taylor AE, Matalon S, Ward PA (eds) Physiology of oxygen radicals. American Physiological Society, Bethesda, Md, pp 251-285

Darley-Usmar VM, Stone D, Smith DR (1989) Oxygen and reperfusion damage: on overview. Free Rad Res Comm 7:247-254

Davies KJA (1987) Protein damage and degradation by oxygen radicals. I. General aspects. J Biol Chem 262:990~9907

Egner PA, Kensler TW (1985) Effects of biomimetic SOD on complete and multistage carcinogenesis in mouse skin. Carcinogenesis 6:1167-72

Elstner EF, Bors W, Wilmanns W (eds) (1987) Reaktive Sauerstoffspezies in der Medizin. Springer, Berlin Heidelberg New York

Feinendegen LE, Bond VP, Booz J, Muehlensiepen H (1988) Biochemical and cellular mechanisms of low-dose effects. Int J Radiat Biol 53:23-37

Frei B, Stocker R, Ames BN (1988) Antioxidant defenses and lipid peroxidation in human blood plasma. Proc Natl Acad Sci USA 85:9748 52

Gohil K, Viguie C, Stanley WC, Brooks GA, Packer L (1988) Blood GSH oxidation during human exercise. J Appl Physiol 64:115 119

Graf E, Mahoney JR, Bryant RG, Eaton JW (1984) Iron-catalyzed hydroxyl radical formation. Stringent requirements for free iron coordination site. J Biol Chem 259:3620 3624

Gutteridge JMC (1985) Antioxidant activity of ceruloplasmin. In: Greenwald RA (ed) CRC Handbook of methods for oxygen radical research. CRC Press, Boca Raton, FL, pp 303 307

Halliwell B (1989) Superoxide, iron, vascular endothelium and reperfusion injury. Free Rad Res Comm 5:315-318

Hamilton GA (1974) Chemical models and mechanisms for oxygenases. In: Hayalshi O (ed) Molecular mechanisms of oxygen activation. Academic Press, New York, pp 405-451

Heinecke JW, Rosen H, Suzuki LA, Chait A (1987) The role of sulfur containing amino acids in superoxide production and modification of low density lipoprotein by arterial smooth muscle cells. J Biol Chem 262:10098-10103

Ignarro LJ, Adams JB, Horwitz PM, Wood KS (1986) Activation of soluble guanylate cyclase by NO-hemoproteins involves NO-heine exchange. Comparison of heine-containing and heme-deficient enzyme forms. J Biol Chem 261:4997-5002

Ivanov II (1985) A relay model of lipid peroxidation in biological membranes. J Free Rad Biol Med 1:247-253

Kanofsky JA (1989) Singlet oxygen production by biological systems. Chem-Biol Interact 70:1 28

Koppenol WH (1985) The reaction of ferrous EDTA with H202: evidence against OH radical formation. J Free Rad Biol Med 1:281 285

Koster JF, Slee RG (1986) Ferritin, a physiological iron donor for microsomal lipid peroxidation. FEBS Lett 199:85-88

Krinsky NI (1984) Biology and photobiology of singlet oxygen. In: Bors W, Saran M, Tait D (eds) Oxygen radicals in chemistry and biology, de Gruyter, Berlin, pp 453-464

261

L'Abbate A, Ursini F (eds) (1989) The role of oxygen radicals in cardiovascular diseases. Kluwer, Dordrecht

Martell AE, Sawyer DT (eds) (1988) Oxygen complexes and oxygen activation by transition metals. Plenum Press, New York

Niki E, Yamamoto K, Takahashi M (1988) Role of iron, ascorbic acid and tocopherol in the oxidation of lipids. In: Ando W, Morooka Y (eds) The role of oxygen in chemistry and biology. Studies in organic chemistry. Elsevier, Amsterdam, pp 509-514

Oliver CN, Ahn B, Moerman EJ, Goldstein S, Stadtman ER (1987) Age related changes in oxidized proteins. J Biol Chem 262:5488 5491

Pacifici RE, Salo DC, Davies KJA (1989) Macroxyproteinase (M.O.P.): a 670 kDa proteinase complex that degrades oxidatively denatured proteins in red blood cells. Free Rad Biol Med 7:521-536

Palmer RMJ, Ferrige AG, Moncada S (1987) Nitric oxide release accounts for the biological activity of endothelium-derived relaxing factor. Nature 327:524-526

Patterson LK, Hasegawa K (1978) Pulse radiolysis studies in model lipid systems. The influence of aggregation on kinetic behavior of OH induced radicals in aqueous sodium linoleate. Ber Bunsenges Phys Chem 82:951-956

Prfitz WA (1984) Inhibition of DNA ethidium bromide intercalation due to free radical attack upon DNA. II. Copper(II)-catalyzed DNA damage by O~. Radiat Environ Biophys 23:7-18

Rajagopalan S, Politi PM, Sinha BK, Myers CE (1988) Adriamycin-induced free radical formation in the perfused rat heart: implications for cardiotoxicity. Cancer Res 48:4766-4769

Rotilio G (ed) (1986) Superoxide and superoxide dismutase in chemistry, biology and medicine. Elsevier, Amsterdam

Samokyszyn VM, Miller DM, Reif DW, Aust SD (1989) Inhibition of superoxide and ferritin dependent lipid peroxidation by ceruloplasmin. J Biol Chem 264:21 26

Samuni A, Aronovitch J, Chevion M, Czapski G (1983) Metal-mediated hydroxyl radical damage. A site specific mechanism. Life Chem Reports [Suppl] 2:39-47

Saran M, Bors W (1989) Oxygen radicals acting as chemical messengers: a hypothesis. Free Rad Res Comm 7:213-220

Saran M, Michel C, Bors W (1988) Reactivities of free radicals. In: Schulte-Hostede S, Darrall NM, Blank LW, Wellburn AR (eds) Air pollution and plant metabolism. Elsevier, London, pp 76-93

Saran M, Michel C, Bors W (1990) Reactions of NO with 02 . Implications for the action of endothelium-derived relaxing factor (EDRF). Free Rad Res Comm (in press)

Shibanuma M, Kuroki T, Nose K (1988) Superoxide as a signal for increase in intracellular pH. J Cell Physiol 136:379-383

Simic MG, Taylor KA, Ward JF, von Sonntag C (eds) (1988) Basic life sciences, vol 49: Oxygen radicals in biology and medicine. Plenum Press, New York

Simon-Schnass I, Pabst H (1988) Influence of vitamin E on physical performance. Int J Vitamin Nutr Res 58:49-54

Singal PK (ed) (1988) Oxygen radicals in the pathophysiology of the heart disease. Kluwer, Boston

Slater TF (1984) Free-radical mechanisms in tissue injury. Biochem J 222:1 15 Stadtman ER (1986) Oxidation of proteins by mixed-function oxidation systems: implication

in protein turnover, ageing and neutrophil function. Trends Biochem Sci 11:11-12 Stadtman ER, Berlett BS (1988) Fenton chemistry revisited: amino acid oxidation. In: Simic

MG, Taylor KA, Ward JF, von Sonntag C (eds) Basic life sciences, vo149: Oxygen radicals in biology and medicine. Plenum Press, New York, pp 131-136

Steinbrecher UP (1988) Role of superoxide in the endothelial cell modification of low density lipoproteins. Biochim Biophys Acta 959:20-30

Summerfield FW, Tappel AL (1984) Effects of dietary polyunsaturated fats and vitamin E on aging and peroxidative damage to DNA. Arch Biochem Biophys 233:408-416

Tullius TD (1987) Chemical snapshots of DNA: using the hydroxyl radical to study the structure of DNA and DNA-protein complexes. Trends Biochem Sci 12:297 300

Ullrich V, Brfine B, Hecker G, Schmidt K-U, Mfilsch A, Busse R (1989) Physiological targets of superoxide anion and hydrogen peroxide in reperfusion injury. Free Rad Res Comm 7:265~74

262

Vanhoutte PM, Katusic ZS (1988) Endothelium-derived contracting factor: endothelioum and/ or superoxide anion? Trends Pharmacol Sci 9:229-230

Van Kuijk FJGM, Sevanian A, Handelman GJ, Dratz EG (1987) A new role for phospholipase Az: protection of membranes from lipid peroxidation damage. Trends Biochem Sci 12:31-34

Walling C (1975) Fenton's reagent revisited. Acc Chem Res 8:125-131 Wolff SP, Garner A, Dean RT (1986) Free radicals, lipid and protein degradation. Trends

Biochem Sci 11:27-31 Yagi K (1986) A biochemical approach to atherogenesis. Trends Biochem Sci 11:18-19 Youngman RJ (1984) Oxygen activation: is the OH radical always biologically relevant ? Trends

Biochem Sci 9:280-283 Youngman RJ, Elstner EF (1981) Oxygen species in paraquat toxicity: the crypto-OH radical.

FEBS Lett 129:265-268

Radical reactions in vivo - an overview

Documents

Transcript of Radical reactions in vivo - an overview