Project Leader: Vladimir P. Skulachev, D.Sci., Director of A.N. Belozersky Institute of Physico-Chemical Biology. Address: A.N.Belozersky Institute of Physico-Chemical Biology, Moscow State University, 119899, Moscow, Russia. Fax.: +7 (095) 939-0338 E-mail: skulach@belozersky.msu.ruList of participants:
Modulation of mitochondrial functions may be an important element in a complex treatment of TNF-dependent diseases (septic shock, cachexia). The same approaches could provide a rational basement for overcoming the NF-kB-dependent antiapototic defense, which sometimes strongly decreases the efficiency of anti-tumor chemotherapy.
BackgroundDuring last several years, intensive studies on apoptosis revealed a crucial role of mitochondria and promoted the second wave of mitochondria research. It was discovered that these organelles are not only “intracellular power stations” but determine the very fate of the cell being a source of pro-apoptotic agents [cytochrome c, specific apoptosis-inducing factor (AIF), some procaspases] and a target for anti-apoptotic defense ( oncoprotein Bcl-2 and its relatives). For cell biologists, mitochondria came back to the focus of interest as a probable “stress-sensor”, as a “point of no-return” in apoptotic program and as a potential target for therapy based on pro- and anti-apoptotic interventions. Unfortunately, this interest was usually not supported by the adequate experience in classical mitochondriology and bioenergetics. This gap did not filled even after three years of the boom in the field, which resulted in publication of many obvious artifacts confusing the readers. The following project is planed in a major part as a contribution of mitochondriologists equipped with their specific methods to studies of TNF-dependent effects.
More than a decade ago it was found that (at least in some cell lines) TNF caused rapid inhibition of mitochondrial respiration [1, 2]. On the other hand, inhibition of respiration by some specific inhibitors or by depletion of mitochondrial DNA prevented TNF-induced death of some cell lines [3, 4]. The TNF-induced inhibition of respiration might be related to formation of ceramide which is released from sphingomyelin and is reported to inhibit initial and middle spans of the respiratory chain.
Pleiotropic effects of TNF include activation of transcription factor NF-kB and induction of general anti-apoptotic defense [5] which is probably also mediated in some way by mitochondria [6].
Another poorly understood effect of TNF is a change of location of mitochondria in the cell [7], which point to perturbation in cytoskeletal systems (microtubular system and actin-myosin cortex) and its interaction with microtubule-associated motor molecules such as kinesin and dynein.
In the both pro- and anti-apoptotic branches of TNF-induced responses, mitochondrion-independent signaling pathways were also identified. TNF-induced apoptosis initiated by activation of “primary protease” (caspase 8) in specific TNF receptor-adapter complex could be executed by direct proteolytic activation of the caspase cascade with no mitochondria involved. On the other hand, the same caspase 8 was shown to cleave inactive precursor of Bid protein to active Bid which activates the mitochondrion-bound protein Bax forming complex with porin in the outer mitochondrial membrane. The Bax-porin complex forms large protein-permeable holes in this membrane, an event resulting in release of the pro-apoptotic proteins to cytosol.
Activation of NF-kB (anti-apoptotic signaling) could be catalyzed by another TNF receptor-adapter complex and cascade of kinases which phosphorylate and inactivate an inhibitory subunit of the transcription factor I-kB [8]. ROS formed by mitochondria were hypothesized to activate in some way the both chains of events.
The mechanisms that underlay a key role of mitochondria in TNF-induced effects remains unresolved. The working hypotheses formulated suggest the interplay of (i) mitochondrial overproduction of ROS and (ii) endogenous mitochondrial antioxidant defense mechanisms. The hypotheses may be regarded as a specific application of a more general concept introduced recently in our group [9]. The concept includes an idea that mitochondria, being a major intracellular source of ROS, possess several lines of antioxidant defense. The first one includes mechanisms preventing ROS formation in mitochondria by specific changes in their energy state, namely “mild” uncoupling. The second one is oxidation of already formed O2-. back to O2 by cytochrome c dissolved in the intermembrane space of mitochondria. The third line of defense is elimination of dangerous ROS-overproducing mitochondria due to permeability transition in the inner membrane (“mitoptosis”). The fourth one includes elimination of ROS-producing cells via mitochondria-dependent apoptosis. In agreement with this concept, several physiological mechanisms of “mild” uncoupling were described in our group [10-12]. These findings were supported by recent discovery of specific uncoupling proteins (UCP) in various tissues [13,14]. In experiments with isolated mitochondria, we have shown that partial uncoupling almost completely inhibited ROS generation in respiratory chain [15]. The next line of defense was also examined in our group. It was found that submicromolar concentrations of cytochrome c, added to mitochondria and submitochondrial particles, strongly inhibit the H2O2 formation by respiratory chain (Korshunov and Skulachev, in preparation). As to the third line of defense, the evidence was presented that the permeability transition in intact cell can be induced by combine action of oxidative stress and "mild" uncoupling [16]. The last part of the concept, mitochondria-mediated apoptosis, received strong experimental support and is widely accepted now. The main efforts of our group in this direction are concentrated on elucidation of the mechanisms of release of pro-apoptotic proteins (cytochrome c, first of all ) into cytosol [17 ].
Within the framework of the above concept, TNF looks like a specific signal which actuates the antioxidant defense system inducing a burst of mitochondrial ROS production to activate NF-kB and transcription of NF-kB-dependent genes (SOD, UCP2, etc.). If this defense appears to be insufficient, TNF induces elimination of mitochondria (“mitoptosis”) and, finally, execution of the cell suicide program (apoptosis) using the other TNF-dependent signaling pathway that may be mitochondria-dependent or independent.
Specific aims.
1. TNF-induced inhibition of electron transfer and generation of reactive oxygen species (ROS) in different segments of respiratory chain.
In the early studies, an inhibition of cellular respiration by TNF was observed [1,2] . The inhibition took place 10-30 min after the TNF addition to the cells. It remains obscure what sites of the respiratory system was inhibited and what was the inhibition mechanism. These questions hardly can be answered in experiments with intact cells. Here the entire arsenal of bionergetics approaches should be applied to permeabilized cells and to mitochondria isolated from the TNF-pretreated cells. The preliminary data indicate that Complex I and Complex III of respiratory chain are the most sensitive targets for the TNF-induced inhibition. At the first step of the study, these data will be verified using specific electron donor/acceptor pairs and specific inhibitors. In the experiments, the time courses of inhibition in different segments of the chain will be compared.
It should be stressed that Complex I is one of the most complicated molecular devices (consists of 43 polypeptides and at least 6 various redox-active prosthetic groups). The mechanism of its functioning remains unresolved even in general terms (sequence of redox events, stoichiometry of proton/electron transfer, etc.). The experiments designed to clarify the details of TNF-induced inhibition will include measurements of “partial reactions” using artificial electron acceptors, comparison of the forward and reverse electron transfers and generation of ROS. Similar approaches will be applied to Complex III where the redox components and the mechanism of electron transfer (“Q-cycle”) are defined much better than in Complex I.
We have suggested that the most important consequence of TNF-induced inhibition of electron transfer is generation of ROS by respiratory chain. It is hypothesized that inhibition of the both Complex I and Complex III results in an increased life-time of semiquinone form of CoQ, which could be a major source for one-electron reduction of oxygen and formation of superoxide anion radical, the primary reactive oxygen species generated in mitochondria. This mechanism is established for ROS generation in Complex III induced by antimycin A and can be blocked by the other Q-cycle inhibitor, mixothiazol. Generation of ROS in Complex III is stimulated by uncouplers in the presence of antimycin A but is inhibited in its absence. This behavior will be used to test the possible antimycin A-like effect of the TNF treatment.
Our recent data indicate that ROS generation in Complex I is much more efficient during the reverse electron transfer than during the forward electron transfer [15, 20]. As a result, even partial uncoupling causes strong inhibition of ROS production. We have proposed that the mechanisms of physiological uncoupling (see below) play a key role in regulation of ROS production in the living cell. This hypothesis will be tested in experiments with intact cell where uncoupling will be caused by exogenous uncoupling agents (FCCP, SF6843, etc.). Such studies would be supplemented by measuring reversal of the uncoupling by “recoupler” 6-ketocholestanol, described recently in our group [12]. ROS generation induced by TNF will be determined under these conditions using intracellular fluorescent dyes.
2. Possible induction of permeability transition pores in the inner mitochondrial membrane by the TNF challenge.
It is suggested that TNF treatment can activate the mechanisms of physiological uncoupling based on increase of proton conductivity of the inner mitochondrial membrane. Our recent studies revealed an important role of mitochondrial anion carriers (including adenine nucleotide traslocase and aspartate/glutamate carrier) in uncoupling mediated by free fatty acids (FFA) [21]. These mechanisms coexist with the specialised "uncoupling proteins" described recently in different tissues. It was reported [22] that expression of the gene coding for one of these proteins, UCP2, is initiated in liver after the TNF challenge. Significant TNF-dependent induction of UCP3 mRNA in muscle was also observed [23]. The functional consequences of possible UCP activation induced by TNF in various cells will be examined taking an advantage of selective inhibition of UCP-dependent uncoupling by GDP. Uncoupling caused by anion carriers will also be revealed using selective inhibitors - carboxyatractylate and glutamate [21].
Overproduction of ROS per se or in combination with partial uncoupling can induce opening of non-selective pores (“permeability transition pores”, PTP) in the inner mitochondrial membrane. This phenomenon discovered earlier in isolated mitochondria was recently demonstrated in intact lymphocytes by our group [16]. Opening of the PTP induced by ROS (and probably also by uncoupling) during the TNF challenge is hypothesized. Measurements of membrane potential, Ca2+ transport and ultrastructural changes (swelling) in mitochondria will help to verify this suggestion. Specific inhibitors of the PTP , cyclosporin A and its analogues, will be used in these studies.
The PTP opening was implicated by several research groups as a central event in apoptosis induced by various agents including TNF. It was suggested that the PTP opening caused swelling of mitochondrial matrix and, as a consequence, disruption of the outer mitochondrial membrane and release of several pro-apoptotic proteins (cytochrome c, AIF and procaspases) from intermembrane space into cytosol. An alternative hypothesis suggested formation of the protein-permeable holes in the outer mitochondrial membrane (Bax-porin complexes). It seems possible that both mechanisms coexist even in one and the same type of cells. Such a possibility will be examined in studies of cytochrome c release from mitochondria into cytosol during the TNF-induced apoptosis. The inhibitors of TNF-dependent signaling upstream of mitochondria, such as inhibitor of caspase 8, zVADfmk, probably will allow to dissociate the induction of the PTP- and Bax-mediated mechanisms of the cytochrome c release. If the TNF-induced PTP opening is caused by ROS, the inhibition of the cytochrome c release by antioxidants is expected.
3. Role of mitochondria in the TNF-induced, NF-kB-dependent anti-apoptotic effects.
To prevent induction of anti-apoptotic defense, we shall use the inhibitors of protein synthesis and various NF-kB-negative models (cell lines and knock-out mice). It is suggested that induction of uncoupling protein(s) (in particular, UCP2) is involved in antioxidant defense of mitochondria and in anti-apoptotic defense. The well documented TNF-dependent induction of antioxidant enzyme, mitochondrial superoxide dismutase (MnSOD), will be taken into account.
Different mitochondrial functions modified by TNF (electron transfer in different segments of the chain, ROS generation and the PTP opening) may be of different sensitivity to NF-kB-dependent defense. If it is the case, there is a chance to overcome the NF-kB-dependent anti-apoptotic effects by modulation of mitochondrial functions.
4. Involvement of different mitochondrial functions in the TNF-induced apoptosis and the TNF-dependent induction of anti-apoptotic mechanisms.
The aim of these studies is to the great extent complementary to the previous ones. The earlier studies clearly indicated that some functions of intact mitochondria are necessary for realization of TNF-induced pro- and anti-apoptotic programs. We plan to undertake a systematic study of this phenomenon using selective inhibitors of mitochondrial functions. The major role of ROS production and following PTP opening is suggested.
In experiments with Bcl-2 and Bcl-XL overexpressing cells, we shall study the protective action of these oncoproteins on the TNF-induced changes in mitochondria, the TNF-induced apoptosis and the TNF-dependent induction of anti-apoptotic mechanisms. The new perspectives in these studies were opened after development of the methods for preparation of recombinant Bcl, Bax and Bid proteins. In vitro addition of these protein to mitochondria from the TNF-treated cells and measurements of permeability of the other mitochondrial membrane for cytochrome c may allow to discriminate the irreversible disruption of the membrane from formation of non-selective protein-permeable holes. Bid is one of the candidates for a role of signaling molecule upstream mitochondria in TNF-induced apoptosis. This protein is activated by caspase 8 and can induce the cytochrome c release from mitochondria into cytosol [24, 25]. It would be important to study the effect of activated Bid on mitochondrial respiration and ROS production. On the other hand, the effect of respiratory inhibitors on the interaction of Bid with mitochondria could be expected.
Special attention will be paid to induction by TNF of uncoupling proteins (UCPs) [22, 23] and to role of such a process in the TNF-induced thermogenesis which may be related to thermoregulatory uncoupling under cold stress, discovered in this group [26].
5. Investigation of possible role of interactions between mitochondria and cytoskeleton in the TNF-induced apoptosis.
We plan to analyze the cytoskeletal structures with the aid of immunomorphological methods, and dynamic distribution of mitochondria by means of videomicroscopy with computer-enhanced contrast. The course of TNF-induced apoptosis will be modified by specific drugs and transfected genetic constructions affecting some cytoskeletal proteins.
References