The importance of oxidative stress in apoptosis

Sharon Glutton
MRC Radiation and Genome Stability Unit, Harwell, UK

In vitro studies have shown that apoptosis can be triggered by many distinct and different stimuli including exposure to physical and chemical agents or by the removal of growth factors. Free radicals and oxidative stress have been implicated in apoptosis, although there is much uncertainty regarding their importance.This review aims to address much of the existing ambiguity.
Downloaded from http://bmb.oxfordjournals.org/ by guest on March 3, 2012

Correspondence to.

Sharon Clutton,
MRC Radiation and Genome Stability Unit, Harwell, Didcot, Oxfordshire OX11 ORD, UK

Apoptosis is a morphologically distinct form of cell death which plays a key role in embryogenesis and tissue homeostasis. The co-ordinated ultrastructural changes that typify apoptosis were first described by Kerr etalx. These are the compaction and marginalisation of chromatin in the nucleus, accompanied by plasma membrane blebbing and cell shrinkage, ultimately forming apoptotic bodies which in vivo are rapidly removed by phagocytosis by neighbouring parenchymal cells or by professional phagocytes. Since the subcellular organelles remain intact and there is little leakage of the contents of the dying cell, apoptosis provides the organism with a safe method of maintaining genome integrity by allowing the organism to remove damaged or abnormal cells without compromising neighbouring cells. This contrasts with necrotic cell death caused by severe cell injury where the cells lyse releasing destructive enzymes and potentially toxic chemicals, which may often cause inflammation. There is increasing evidence that the molecular mechanism of apoptosis has been evolutionary conserved since it occurs in both nematodes and vertebrates. Of particular importance are the genetic studies on the nematode Caenorhabditis elegans which have led to the identification of two genes, ced-3 and ced-4, essential in developmental cell death2. Mutation in either gene permits cells which would normally die to survive. The ced-3 gene product encodes a protein which has considerable homology with a family of mammalian cysteine proteases related to interleukin ip converting enzyme (ICE)3-4. It is becoming increasingly apparent that cysteine proteases play a key role in the downstream events governing apoptosis in both nematodes and mammals. Resembling the observations in C. elegans, overexpression of a variety of cysteine proteases in mammalian cells has been found to trigger apoptosis5'6, while inhibition of their activity can protect the cells7'8.
Brituh Medical Bulletin 1997;33 (No 3) 662-668

The Brihth Councill997

Oxidative stresi in apoptosis

Oxidative stress
A diverse number of stimuli have been shown to induce apoptosis, many of them are also known to compromise the fine balance between intracellular oxidants and their defence systems. Under aerobic situations, the participation of oxygen in redox reactions is unavoidable and a variety of highly reactive chemical entities are produced. These are commonly referred to as reactive oxygen species (ROS) and the list comprises the hydroxyl radical, hydrogen peroxide, lipid peroxides, nitric oxide and superoxide, to name but a few. Many of these agents have a beneficial role in the cell but, when present in excess, the cell becomes oxidatively stressed9. A number of gross biochemical changes occur as a consequence of oxidative stress, the extent depending on the severity of the insult. Extreme, nonphysiological concentrations of oxidants or oxidant generating agents cause necrosis rather than apoptosis, consistent with the hypothesis and observations of Duvall and Wyllie10. Oxidative overload causes gross cellular damage resulting in alteration of redox state (e.g. depletion of nucleotide coenzymes, disturbance of sulphur containing enzymes), saturation and destruction of the defence and repair systems. If the cellular balance is not restored, a number of pathological processes are elicited. Predominant processes resulting from oxidative stress include oxidative lipid degradation (lipid peroxidation), the loss of intracellular calcium homeostasis and alteration of metabolic pathways11. All of these processes have been recorded in exhaustive lists of apoptotic models. However, an interesting challenge remains to dissect out the biochemical effector from the biochemical changes occurring as a consequence of cell death. The breakdown of membrane lipids increases the concentration of biologically active lipids, such as arachidonic acid metabolites, and the release of diacylglycerols and ceramide from sphingomyelin. These in turn can stimulate the influx of calcium and its release from intracellular stores resulting in loss of calcium homeostasis on which many integral enzyme systems depend. For example, activation of calcium dependent poly-(ADP-ribose)-synthetase consumes NAD, which consequently limits the availability of NAD(P)H to take part in redox reactions, such as ATP and glutathione generation12'13. Additionally, calcium-dependent endonuclease activation is believed to initiate the chromatin fragmentation frequently observed in apoptotic cells14-15. In association with diacylglycerol release, calcium can activate protein kinase-C dependent signal transduction pathways, thus modifying gene transcription16.
8nhjhMedreo/Bu//.t.n 1997^3 (No 3) 663

Downloaded from http://bmb.oxfordjournals.org/ by guest on March 3, 2012

Apoptosis

Involvement of ROS in apoptosis

Apoptosis first occurs during embryogenesis in the inner cell mass during blastulation and, in a series of elegant experiments from Parchment and Pierce17, hydrogen peroxide generated from amine oxidase activity was identified in the blastocoel fluid as being responsible for this cell death. Their results were surprising since hydrogen peroxide was traditionally associated with pathogenic events including necrotic cell death. Similarly, toxicological studies revealed that exposure of cells to low doses of hydrogen peroxide triggered apoptosis instead of necrosis seen to occur at higher doses18. Subsequently, a wide range of oxidants have been found to be apoptotic, these include organic and lipid peroxides, thiol oxidising compounds and redox cycling quinones19-20*23'24. Moreover, oxidative changes have been found in a variety of cell types even when non-oxidative stimuli have been employed, for example thymocytes treated with glucocorticoids or etoposide25. There is great confusion in the literature regarding the use of antioxidants as modifiers of an apoptotic response. In some systems there is little doubt that free radicals and ROS are produced, such as exposure to ionising radiation, and so it is of no surprise that, in such examples, antioxidants have been shown to reduce or delay apoptosis. A good example of this is radiation and hydrogen peroxide induced apoptosis in thymocytes which can be inhibited by trolox, an inhibitor of lipid peroxidation25, or reduction of apoptosis in blastocysts when treated with glutathione, N-acetyl cysteine or catalase17. In other systems the oxidative insult is implied, such as after treatment with tumour necrosis factor (TNF) or inflammatory cytokines (e.g. IL-1) which are believed to initiate signal transduction pathways via ROS26. Again, agents such as superoxide dismutase have been found to attenuate the effects27. However, the situation is ambiguous since the Fas antigen which is structurally related to the TNF receptor can induce apoptosis when incubated with the anti-Fas/APO-1 but does not readily respond to antioxidant treatment, although it may be that there is a divergent signalling pathway between Fas/APO-1 and TNF receptor28. Another confounding problem with the use of exogenous antioxidants is that they may also act as pro-oxidants depending on both the buffering capacity of the cell and the concentrations employed.

Downloaded from http://bmb.oxfordjournals.org/ by guest on March 3, 2012

Is BCL-2 an antioxidant?
Perhaps the best studied negative regulator of apoptosis is the protooncogene product BCL-2. BCL-2 is a 26kDa integral membrane protein originally cloned from the t(14:18) translocation breakpoint found in B-cell
664 Bnhih M,d,cal Bulletin 199723 (No 3)

Oxidative stress in apoptosis

lymphomas. Expression of this protein has been found to rescue lethally insulted cells from a wide range of stimuli29'30. BCL-2 is only one member of a larger family of related proteins, some of which also prevent cell death, whilst others, like Bax, accelerate it. These related proteins can homo- and heterodimerise, and the final commitment to death depends upon the stoichiometry of the effector and blocking proteins. The importance of the bcl-2 gene in apoptosis can be highlighted by the presence of a structurally and functionally, related homologue ced-9 that plays a pivotal role in preventing wide spread cell death in C. elegans2. Since free radicals have frequently been implicated in apoptosis, Hockenbery et at23 tested the hypothesis that BCL-2 acted as an antioxidant. Indeed they reported that sustained hpid peroxidation was a feature of apoptosis induced by the redox cycling quinone, menedione and by hydrogen peroxide, and both apoptosis and lipid peroxidation could be modified by conventional antioxidants and also by overexpression of bcl-2. Similarly, Kane et al3i found that bcl-2 expression inhibited apoptotic and necrotic cell death in neural GTl-7 cells treated with thiol depleting agents, and suggested that the bcl-2 gene product modulated intracellular levels of free radicals, which had a knock-on effect in suppressing apoptosis. However, BCL-2 facilitated cell survival even after application of the thiol blocking agent diethyl maleate (DEM) and, therefore, it is unlikely that it could function as a direct free radical scavenger. Perhaps the strongest argument against BCL-2 having direct antioxidant properties came from experiments of Jacobson and Raff32. They demonstrated that prograrnmed cell death could occur when cells were cultured at very low concentrations of oxygen, although not by agents known to exert their effects via ROS such as hydrogen peroxide and menedione. Similar experiments by Shimizu et at33, failed to detect ROS during hypoxia induced apoptosis. In both of these experiments, BCL-2 still offered protection against apoptosis. The mechanism of BCL-2 remains unknown, although from these experiments it is clear that BCL-2 is unlikely to act in the capacity as a free radical scavenger. Moreover, such evidence suggests that ROS are not the exclusive effectors in the death programme. The quest remains to find a common pathway downstream of the possible initiating agents which unifies all observations, and the relevance of oxygen independent redox reactions is in question.

Downloaded from http://bmb.oxfordjournals.org/ by guest on March 3, 2012

Intracellular redox state

The tripeptide glutathione plays an integral role in cellular redox biochemistry and, for this reason, the concentrations between reduced
Bnhth Medrnjl Buhn 1997;33 (No 3) 665

Apoptosis

and oxidised glutathione are stringently controlled. The machinery of the cell, the proteins, depend upon both the pH and the redox state of their micro-environment for their reactivity. For example, it is known that serine protease zymogens have a local requirement for high local concentrations of oxidised glutathione34. In addition, protein-thiol modification is believed to be involved in the initiating steps for protein proteolytic cleavage35. Activation of the ICE-like enzymes which are believed to orchestrate apoptosis, requires proteolytic processing, which has lead to speculation that oxidative thiol modification of ICE-like enzymes may be integral to controlling apoptosis36. As yet, there is only indirect evidence to support his hypothesis. Agents which disrupt thiol homeostasis, such as DEM and (d-1)buthionine sulphoximine have been shown to induce apoptosis, which can then be rescued by agents that restore this balance21"23-33. Intracellular glutathione levels also drop during apoptosis when both oxidative and non-oxidative stimuli are applied, and only in the case of apoptosis perpetrated by ROS has there been a concomitant increase in oxidised glutathione37. Intriguingly, Slater et al36 found that protease inhibitors can prevent both apoptosis and glutathione efflux. In search of a common pathway for apoptosis, investigators have suggested that BCL-2 may play a central role in intracellular signalling by regulating calcium38 or glutathione efflux37.

Downloaded from http://bmb.oxfordjournals.org/ by guest on March 3, 2012

Summary
There is no doubt that oxidative stress can elicit cell death, and that mild oxidative stress can initiate apoptosis rather than necrosis; although reactive oxygen species can cause oxidative stress, they are not essential for the apoptotic processes to occur. ICE-family proteins play the part of cellular executioner, but the biochemical messengers remain to be determined. A tempting, unifying hypothesis is that perturbation of cellular redox homeostasis may control these key events.

References
1 Kerr JFK, Wylhe AH, Curne AR Apoptosis a basic biological phenomenon with wide ranging implications in tissue kinetics. Br ] Cancer 1972; B2 , 239-57 2 Steller H. Mechanisms and genes of cellular suicide Science 1995; 267: 1445-62 3 Cerretti DP, Kozlosky CJ, Mosley B et al. Molecular cloning of the interleukin ip converting enzyme Science 1992; 256: 97-9 4 Thornberry NA, Bull HG, Calaycay JR et al. A novel heterodimeric cysteine protease is required for interleukin 1(3 processing in monocytes Nature 1992, 356: 768-74 5 Kumar S. ICE-like proteases in apoptosis. Trends Btochem Set 1995; 20. 198-202

666

Bnhih Medico/ BuHehn 1997;53 (No 3)

Oxidative itreis in apoptojij

6 Muira M, Zhu H, Rotello R et al Induction of apoptosis in fibroblasts by IL-lp"-converting enzyme, a mammalian homologue of the C. elegans cell death gene ced 3. Cell 1993; 75. 65360 7 Tewan M, Quan LT, O'Rouke K et al. Yama/CPP32, a mammalian homologue of CED 3 is a CRrmA-inhibitable protease that cleaves the death substrate poly(ADP-nbose) polymerase Cell 1995; 81. 801-9 8 Nicholson DW, Ah A, Thornberry D et al. Identification and inhibition of ICE/CED-3 protease necessary for mammalian apoptosis Nature 1995; 376: 37-43 9 Sies H. Introduction In Oxtdatwe Stress. London: Academic Press, 1985; 1-7 10 Duvall E, Wylhe AH. Life and death of the cell. Immunol Today 1986; 7: 115-9 11 Slater TFS. Free radicals fact or fiction' Br J Cancer 1987; 55: 5-10 12 Schaufstatter IU, Hyslop PA, Hinshaw DB, Spragg RG, Sldar LA, Cochrane CG. Hydrogen peroxide induced injury of cells and its prevention by inhibitors of poly(ADP-nbose) polymerase. Proc Natl Acad Set USA 1986; 83: 4908-12 13 Hyslop PA, Hinshaw DB, Halsey WA et al. Mechanisms of antioxidant mediated cell injury; the glycolytic and mitochondrial pathways of ADP-phosphorylation are major targets inactivated by hydrogen peroxide. / Biol Chem 1987; 263: 1665-75 14 Lemasters JL, DiFuiseppi J, Nieminen AL, Herman B. Blebbmg, free calcium and mitochondrial membrane potential preceding cell death in hepatocytes Nature 1987; 325: 78-81 15 Orrenius S, McConkey DJ, Bellomo G, Nicotera P. Role of Ca** in toxic cell killing. Trends Pharmacol Set 1989; 10: 281-5 16 Wolf M, Cautrecasas P, Sahyoun N. Interaction of protein kinase C with membranes is regulated by Caz+, phorbol esters and ATP. / Btol Chem 1985; 260: 15718-22 17 Paxchment RE. The implications of a unified theory of programmed cell death, polyamines, oxyradicals and histogenesis m the embryo. Int ] Dev Biol 1993; 37: 75-83 18 Lennon SV, Martin SJ, Cotter TG. Dose dependent induction of apoptosis in human tumour cell lines by widely diverging stimuli. Cell Prohf 1991; 25: 203-4 19 Sandstrom PA, Tebbey PW, Van Cleave S, Buttke TM. Lipid hydrogen peroxides mduce apoptosis in T cells displaying an HTV-associated glutathione peroxidase deficiency. / Btol Chem 1994, 269: 798-801 20 Fernandes RS, Cotter TG. Apoptosis or necrosis. Intracellular levels of glutathione influence mode of cell death. Btochem Pharmacol 1994; 48: 675-81 21 Sato N, Iwata S, Nakamura K, Hon T, Mon K, Yodoi J Thiol mediated regulation of apoptosis. / Immunol 1995, 154: 3194-203 22 Dypbuckt JM, Ankarcona M, Sjoholm A, Strom K, Orrenius S, Nicotera P. Different prooxidant levels stimulate growth, trigger apoptosis, or produce necrosis of insulin secreting RINm5F cells. / Biol Chem 1994; 269- 30553-60 23 Hockenbury DM, Oltavai ZN, Yin XM, Milhman CL, Korsmeyer SJ. Bcl-2 functions in an antioxidant pathway to prevent apoptosis. Cell 1993; 75. 241-51 24 Torres-Roca JF, Lecoeur H, Armatore C, Gougeon ML. The early intracellular production of reactive oxygen intermediates mediates apoptosis in dexamethasone treated thymocytes. Cell Death Differ 1995; 2: 309-19 25 Forrest VJ, Kang YH, McClam DE, Robinson DH, Ramaknshnan N Oxidative stress induced apoptosis prevented by Trolox. Free Radic Biol Med 1994; 16: 675-84 26 Burdon RH. Superoxide and hydrogen peroxide in relation to mammalian cell proliferation. Free Radtc Btol Med 1995; 18: 775-94 27 Hirose K, Longo DL, Oppenheim JJ, Matsushima K Overexpression of mitochondrial manganese SOD promotes the survival of tumour cells exposed to interleulan 1, tumour necrosis factor, selected anticancer drugs and ionising radiation. FASEB } 1993; 7: 361-8 28 Schulze-Osthoff K, Krammer PH, Droge W. Divergent signalling via APO-1/FAS and the TNF receptor, two homologous molecules involved in physiological cell death. EMBO ] 1994; 13: 4587-96 29 Vaux DL, Cory S, Adams JM. Bcl-2 gene promotes the haemopoietic cell survival and cooperates with c-myc to immortalise pre-B cells. Nature 1988; 335: 440-2 30 Reed JC. Mirureview: Bcl-2 and regulation of cell death. / Cell Biol 1994; 124: 1-6

Downloaded from http://bmb.oxfordjournals.org/ by guest on March 3, 2012

Bnfiifp Mtdtcal Bullthn 199743 (No 3)

667

Apoptosis

31 Kane DJ, Sarafian TA, Anton R et al. Bcl-2 inhibition of neural death: decreased generation of reactive oxygen species. Science 1993; 262 1274-7 32 Jacobson MD, Raff MC. Programmed cell death and Bcl-2 protection in very low oxygen. Nature 1995; 374: 814-6 33 Shjnuzu S, Eguchi Y, Kosaka H, Kamuke W, Matsuda H, Tsujimoto Y. Prevention of hypoxia induced cell death by Bcl-2 and Bcl-xL. Nature 1995; 374- 811-3 34 Bngehus R Mixed disulphidcs: biological function and increase in oxidative stress. In: Sies H (ed) Oxidative Stress. London: Academic Press, 1985; 256 35 Grune T, Remheckel T, Joshi M, Davies KJA Proteolysis in cultured liver epithelial cells during oxidative stress. / Btol Chem 1995; 270: 2344-51 36 Slater AGF, Stefan C, Nobel I, Van den Dobblestein DJ, Orrenius S. Intracellular redox changes during apoptosis. Cell Death Differ 1996; 3 57-62 37 Ghibelli L, Coppola G, Rotilio E, Lafavia E, Maresca V, Cinolo MK. Non-oxidative loss of glutathione in apoptosis via GSH extrusion. Btocbem Bwphys Res Commun 1995; 216: 313 20 38 Lam M, Dubyak G, Chen L, Nunez G, Miesfeld RL, Distelhorst CW. Evidence that BCL-2 represses apoptosis by regulating endoplasmic reticulum-associated Ca2* fluxes. Proc Natl Acad Set USA 1994, 91: 6569-73

Downloaded from http://bmb.oxfordjournals.org/ by guest on March 3, 2012

668

Bnhsh Medical Bulbtin 1997;53 (No. 3)