New Insights into the Regulation of Plant Succinate Dehydrogenase ON THE ROLE OF THE PROTONMOTIVE FORCE *

Regulation of succinate dehydrogenase was investigated using tightly coupled potato tuber mitochondria in a novel fashion by simultaneously measuring the oxygen uptake rate and the ubiquinone (Q) reduction level. We found that the activation level of the enzyme is unambiguously reflected by the kinetic dependence of the succinate oxidation rate upon the Q-redox poise. Kinetic results indicated that succinate dehydrogenase is activated by both ATP (K1⁄2 3 M) and ADP. The carboxyatractyloside insensitivity of these stimulatory effects indicated that they occur at the cytoplasmic side of the mitochondrial inner membrane. Importantly, our novel approach revealed that the enzyme is also activated by oligomycin (K1⁄2 16 nM). Time-resolved kinetic measurements of succinate dehydrogenase activation by succinate furthermore revealed that the activity of the enzyme is negatively affected by potassium. The succinate-induced activation ( K ) is prevented by the presence of an uncoupler. Together these results demonstrate that in vitro activity of succinate dehydrogenase is modulated by the protonmotive force. We speculate that the widely recognized activation of the enzyme by adenine nucleotides in plants is mediated in this manner. A mechanism that could account for such regulation is suggested and ramifications for its in vivo relevance are discussed.

dehydrogenase involve artificial electron acceptors (4,5) or are based on oxygen uptake measurements (14)(15)(16). These methods both have limitations: the former cannot be readily used in well-coupled mitochondria whilst the latter does not yield data that reflect specific succinate dehydrogenase activity. Evidently, data that are obtained in uncoupled systems using nonendogenous substrates are not easily interpreted in terms of their potential physiological relevance. By measuring the oxygen uptake rate and, simultaneously, the reduction level of the Q-pool [cf. (17)], information can be obtained as to the kinetic behaviour of mitochondrial respiratory enzymes (including succinate dehydrogenase) towards their natural substrates within a tightly coupled environment. Such modular kinetic measurements have previously yielded valuable data concerning the kinetic interplay between Q-reducing and QH 2 -oxidising enzymes both in isolated plant and yeast mitochondria (18)(19)(20). Mathematical modelling of these kinetic data is proving an increasingly powerful tool in understanding the in vivo role of mitochondrial electron transfer (18)(19)(20)(21)(22)(23).
In this paper we have followed a modular kinetic approach to determine unequivocally the various activation states of succinate dehydrogenase in tightly coupled potato tuber mitochondria. It is confirmed that the enzyme is activated by both ATP and ADP which, interestingly, appears to occur at the cytoplasmic side of the inner-membrane. Importantly, our novel approach reveals that the activation state of succinate dehydrogenase is affected by oligomycin as well as by potassium. These observations suggest that in vitro activity of the plant succinate dehydrogenase can be modulated by the protonmotive force.

Experimental Procedures
Fresh potato tubers were purchased in a local supermarket and stored overnight at 4˚C. Mitochondria were isolated and purified according to a previously described protocol (24) either in the presence or the absence of added K + . The isolation and purification media used in the respective procedures were buffered at pH 7.4 with either MOPS/KOH or MOPS/NaOH. Rat liver mitochondria were prepared from male Wistar rats in the presence of K + and essentially as described elsewhere (25). Protein content was estimated using the bicinchoninic acid method with bovine serum albumin as a standard (26). Respiratory activity and the reduction level of the Q-pool were simultaneously measured voltametrically in a specially constructed chamber (University of Sussex) housing a Rank oxygen electrode and glassy carbon and platinum electrodes connected to a Ag/AgCl reference electrode similar to that described by Moore et al. (17). In experiments with mitochondria isolated in the absence of K + , the reference electrode was connected via a NaCl-rather than a KCl-agar bridge. Mitochondria were incubated in 2.2 ml reaction medium (medium A) that contained mannitol (0.3 M), MgCl 2 (1 mM), K 2 HPO 4 (5 mM), KCl (10 mM), MOPS (20 mM, pH adjusted to 7.2 with KOH) and ubiquinone-1 (1 µM); K + -salts were replaced by Na + -salts (medium B) in experiments with mitochondria isolated in the absence of K + . Other chemicals were added as indicated in the legends to the figures. Data were recorded digitally using a PowerLab/4SP system (ADInstruments Pty Ltd., UK) connected to iMac running Chart v3.6s software (ADInstruments Pty Ltd., UK).

Results
Similar to their mammalian counterparts, fresh potato tuber mitochondria contain only the cytochrome pathway through which reducing equivalents are transferred from QH 2 to oxygen. Upon oxidation of succinate, the respiratory chain in isolated potato mitochondria can therefore be considered to consist of two kinetic units that 'communicate' through a single intermediate, the Q-pool. This pool is reduced by the combined action of succinate dehydrogenase and the dicarboxylate carrier (to allow succinate entry into the mitochondrial matrix) and subsequently oxidised by the cytochrome pathway [cf. (23)]. Activation of the Qreducing side of the respiratory system (from this point simply referred to as SDH) will result in an increased rate of electron transfer and, concomitantly, in an increased reduction level of the Q-pool. On the other hand, stimulation of the QH 2 -oxidising side will also cause an increased respiratory activity, but this will be accompanied by a decreased rather than an increased Qreduction level. Simultaneous steady state measurement of the O 2 -uptake rate and the Q-redox poise will therefore readily reveal which side of the respiratory system is activated. This is clearly illustrated by the data shown in Fig.1.
Addition of succinate to a reaction mixture containing potato tuber mitochondria results in a 20-35% reduction of the Q-pool and in an oxygen uptake rate of ~30 nmol O 2 /min/mg ( Fig.1). The subsequent addition of ATP causes a slight increase in the respiratory rate and, significantly, in a striking further reduction of the Q-pool (to ~80%). Addition of ADP results in a substantially increased rate (typical respiratory control ratio ~ 4) and in an oxidation of the Q-pool to ~60% (Fig.1A), demonstrating that electron transfer is tightly coupled to phosphorylation. The considerable ATP-dependent rise in the Q-redox poise, together with the small but detectable increase in respiratory rate confirms that SDH is activated by ATP. This activation is insensitive to CAT (Fig.1B), an inhibitor of the mitochondrial adenine nucleotide translocator (27), which suggests that ATP exerts its stimulatory effect at the cytoplasmic side of the inner-membrane. Addition of ADP in the presence of CAT does not significantly affect the respiratory rate or the Q-redox poise (Fig.1B), confirming that CAT prevents ADP entry into the matrix. As anticipated, CCCP-induced uncoupling also causes an increase in the respiratory rate (~5 fold) as well as a substantial oxidation of the Q-pool (to ~35%; Fig.1B).
ATP-induced activation of SDH has, to our knowledge, not been quantified to date. The relatively large and fast change observed in the Q-redox poise (Figs 1A and 1B), however, readily enables quantification (Fig.1C). The level of activation can be defined as the difference between the Q-redox poise that is attained in the presence of succinate+ATP and the Qreduction level that is reached in the presence of succinate alone (Fig.1C, inset). A measure of relative activation is obtained by expressing the activation level observed at a limiting ATP concentration (A) as a fraction of that seen at a saturating concentration (T). The titration data presented in Fig.1C show that SDH is activated half-maximally by ~3 µM ATP.
It has been reported previously (5) that SDH in mung bean and cauliflower mitochondria is not only activated by ATP, but also by ADP. Addition of ADP to fresh potato mitochondria that were pre-incubated in the presence of CAT and Ap5A [inhibitor of adenylate kinase (28)] causes an immediate and significant reduction of the Q-pool (from ~20 to ~80%) as well as an increase in the respiratory rate ( Fig.2A). This indicates that SDH is indeed activated by ADP. That this activation is neither affected by CAT nor by Ap5A shows that the effect is not due to indirect ATP formation through phosphorylation of ADP within the mitochondrial matrix or through adenylate kinase activity within the inter-membrane space, respectively. These observations demonstrate that SDH activity is stimulated by ADP per se which, similar to the activation by ATP, occurs at the cytoplasmic side of the inner-membrane.
Such an activation is in contrast with the observation that SDH is deactivated by ADP in Arum mitochondria (13). It should be noted, however, that the deactivation in Arum was observed in the absence of CAT. Addition of ADP under such conditions causes an oxidation of the Q-pool (cf. Fig.1A) and most likely a decrease of the membrane potential ( ) due to proton-flux into the matrix through the ATP synthase [cf. (29)]. It can therefore not be excluded that the observed deactivation of SDH by ADP in Arum mitochondria is accounted for by these effects Effects of adenine nucleotides may implicate the mitochondrial ATP synthase, even though activation of SDH by these compounds occurs in the presence of CAT. To assess this possibility, ATP-induced activation was studied in the presence of oligomycin. It was found that oligomycin does not affect the activation by ATP (data not shown). Importantly, oligomycin alone appears to activate SDH. This is illustrated by the data presented in Fig.3A which show that addition of oligomycin to mitochondria oxidising succinate results in a considerable and immediate reduction of the Q-pool (to ~70%) and in a small but significant increase in the respiratory activity (from ~30 to ~60 nmol O 2 /min/mg). Similar effects are observed when oligomycin is substituted by N,N'-dicyclohexylcarbodiimide, another inhibitor of the ATP synthase (data not shown). Subsequent addition of CCCP results in a strong oxidation of the Q-redox poise (to ~25%) and does not affect the oxygen uptake rate significantly. This indicates that the oligomycin-induced activation of SDH is reversed by CCCP. When ATP is added under these conditions, re-activation of SDH occurs, which is reflected in a gradually increasing respiratory rate and Q-reduction level (Fig.3A). This re-activation, however, is significantly slower than the ATP-induced activation observed in the absence of oligomycin and CCCP (cf. Figs 3A and 1A). In Fig.3B the oligomycin-induced activation is presented quantitatively and expressed as a fraction of the activation that is achieved in the presence of a saturating amount of ATP. It can be inferred from Fig.3B that activation of SDH by oligomycin is not to the same extent as that by ATP (maximal activation by oligomycin is ~85% of the stimulation observed at a saturating ATP concentration) and that half-maximal stimulation occurs at ~16 nM. It has previously been reported that oligomycin causes an increase in when added to potato mitochondria oxidising succinate under state 2 conditions (30). It is therefore likely that SDH is not directly activated by oligomycin, but indirectly by an increased . This notion agrees well with the observed deactivating effect of CCCP (Fig.3A).
It has recently become apparent that plants contain a mitochondrial ion channel that mediates unidirectional K + -entry into the matrix (31). Activity of this K + -channel effectively deenergises the inner-membrane and, importantly, is inhibited by both ATP and ADP (31). Thus, when mitochondria are incubated in the presence of KCl, succinate alone is incapable of generating a . However, is readily established in the additional presence of ATP, or indeed in the absence of KCl (31). It is therefore conceivable that activation of SDH by adenine nucleotides, similar to that by oligomycin, is mediated by . To explore this possibility further, activation experiments were performed with mitochondria isolated either in the presence or the total absence of K + .  In rat liver mitochondria, the Q-pool is rapidly reduced to almost 100% upon addition of succinate (Fig.4C, trace 1; note that K + was present during isolation but absent in the assay). In this system, SDH is therefore fully activated by succinate alone, which is also apparent from the observation that ATP has no significant additional effect (Fig.4C, trace 2). The activation by succinate (± ATP) appears to be even somewhat faster than the ATP-induced activation of SDH observed in potato tuber mitochondria, isolated ± KCl (Fig.4C, compare traces 1 and 2 with traces 3 and 4).
It is evident from the above results that Q-reduction measurements provide important information on SDH activation, particularly in terms of time-resolved kinetics. However, the Qkinetic method is most powerful when Q-reduction levels are related to the corresponding oxygen uptake rates. Titration of the respiratory activity with an appropriate inhibitor thus reveals the kinetic dependency of both Q-reducing and QH 2 -oxidising enzymes upon the reduction level of the Q-pool (18,19,21,23).
In Fig.5A  conditions increases only slightly with the Q-redox poise. The presence of ADP alters these kinetics such that respiratory activity is substantially increased at any Q-reduction level. A further slight stimulation of the oxygen uptake rate, at Q-reduction levels greater than ~20% only, is observed when respiration is uncoupled by CCCP. Also shown in Fig.5A are steady states that were attained upon oxidation of NADH, both in the presence and the absence of ADP. It is clear that the NADH data are readily described by the appropriate kinetic curves fitted through the succinate data. This confirms that activity of the cytochrome pathway is kinetically independent of Q-reducing enzymes.
The kinetic behaviour of SDH with respect to the Q-reduction level was determined by titrating the succinate-dependent respiratory activity with antimycin A. From Fig.5B it is evident

Discussion
In this paper we have demonstrated how a modular Q-kinetic approach can improve our understanding of the regulation of mitochondrial respiration. It is evident that determination of the kinetic dependency of respiratory activity on the Q-reduction level unequivocally reveals the different activation levels attained by SDH (Figs 5B and 6). It is also clear that continuous Qreduction measurements provide valuable information on the time-resolved kinetics of SDH activation (Fig.4). Importantly, activation of SDH was studied within a tightly coupled environment (Fig.5A) in which the enzyme reacts with its natural substrate (Q) and product (QH 2 ). Under such conditions, the Q-redox poise is significantly increased as a result of SDH stimulation, whereas the respiratory rate is hardly affected (e.g. Fig.1). The sensitivity of the Qreduction signal has enabled the quantification of ATP-induced SDH activation (Fig.1C) and, more importantly, has revealed that the enzyme is activated by oligomycin (Fig.3), a phenomenon that has remained unreported until now.
An interesting feature of SDH activation by ATP is its insensitivity to CAT, suggesting that activation occurs at the cytoplasmic side of the inner-membrane (Fig.1). This interpretation is not fully conclusive, however, since ATP may enter the matrix via a CAT-insensitive nucleotide carrier (32). SDH is also activated by ADP, again in a CAT-insensitive manner ( Fig.2). In the presence of CAT, ADP does not induce a state 3, which unambiguously confirms that it does not enter the matrix. Since ATP-formation through adenylate kinase action has been excluded ( Fig.2), these data indicate that SDH is activated by ADP per se at the cytoplasmic side of the inner-membrane. This strengthens the notion that activation by ATP also occurs in the inter-membrane space, since it is intuitively unlikely that ADP and ATP activate SDH by different mechanisms.
The results presented in this paper disclose an important general aspect of (in vitro) mitochondrial SDH regulation, namely that the enzyme's initial activation level is dependent on the source of mitochondria as well as on the procedure employed to isolate them. SDH is only slowly activated by succinate in potato tuber mitochondria isolated in the presence of K + , whereas activation occurs faster and to a greater extent in organelles isolated in the absence of this cation (Fig.4A). In both potato samples, activation by succinate is positively affected by ATP (Fig.4A), whilst such activation is ATP-independent in rat liver mitochondria (Fig.4C).
These observations indicate that the initial activation level of SDH, upon mitochondrial isolation, is higher in rat liver than in potato and that the presence of K + during the preparation significantly lowers the activation level in potato tuber mitochondria.
Activation of SDH by succinate is prevented by CCCP, completely in potato (Fig.4A) and to a large extent in rat liver (Fig.6B) mitochondria. This strongly suggests that this activation is mediated by the protonmotive force. That SDH activity can be modulated in this fashion is corroborated by the observed stimulatory effect of oligomycin, which is reversed as well as prevented by CCCP (Figs 3A and 6A). Earlier publications further support this novel regulatory mechanism. For example, extraction of succinate dehydrogenase from its native membrane results in a considerable decrease in the turnover number of the enzyme (33-35).
Upon membrane-reconstitution this number is restored to its original value, a result which has been interpreted as a positive modulation of succinate dehydrogenase activity by the membrane or one of its components (33-35). Related to this, it is also apparent that activation by succinate occurs much more rapidly in tightly coupled mitochondria (from rat liver) than in membranous or soluble enzyme preparations (3). Moreover, the activation energy of the process by which NAD + -linked substrates (via generation of QH 2 ) activate succinate dehydrogenase is relatively low in intact mitochondria (3). It has been suggested that protein conformational changes, likely required for activation, are facilitated within a coupled environment (3,8).
The data presented in this paper remain inconclusive as to whether activation of SDH by adenine nucleotides is also wholly mediated by the protonmotive force. However, several lines of circumstantial evidence suggest that this could indeed be the case.
From our results it is clear that activation by ATP is slower in the presence than in the absence of CCCP (Figs 3A and 1A), which suggests that activation occurs more readily within a coupled environment. This observation agrees well with the earlier findings that activation of succinate dehydrogenase by adenine nucleotides in cauliflower mitochondria is negatively affected by sonication or freeze-thawing treatments (5) and that the mammalian succinate dehydrogenase is not at all activated by ATP when studied in a solubilised form or in submitochondrial particles (3). The fact, however, that SDH is activated in the presence of CCCP, albeit relatively slowly (Fig.3A), could indicate an additional stimulatory effect of ATP that is not mediated via the protonmotive force.
As mentioned before in the Results section, a discrepancy appears to exist between the SDH-activating effect of ADP in potato tuber mitochondria (Fig.2) and the SDH-deactivating effect of this nucleotide in A. maculatum (13). The potato experiments were performed in the presence of CAT. Under such conditions is anticipated to remain relatively high upon addition of ADP, thus allowing a ready activation of SDH. In the Arum study, on the other hand, CAT was absent, allowing ADP to enter the matrix where it is phosphorylated to ATP.

Under those conditions the magnitude of is expected to drop (29). That SDH activity in
Arum mitochondria is not stimulated by ADP or ATP at a low , but instead is deactivated, is in accordance with the relatively slow ATP-induced activation of SDH observed in potato mitochondria under uncoupled conditions (cf. Fig.3A).
It has been reported that potato tuber mitochondria become relatively de-energised upon isolation, such that a membrane potential is not established prior to the addition of a respiratory an absolute prerequisite to achieve maximum activation of SDH (Fig.4A). In the more energised rat liver organelles, on the other hand, succinate alone is sufficient to accomplish this (Fig.4C).
This would suggest that ATP is not required for maximum SDH activation under coupled conditions and, in turn, that ATP exerts its stimulatory effect by increasing the protonmotive force.
As previously discussed, SDH activation by both ATP and ADP is likely to occur in the inter-membrane space. It is difficult to envisage how these nucleotides would act directly on SDH at this side of the inner-membrane. Several mechanisms, however, have been reported by which ATP increases (and hence the protonmotive force) when added to plant mitochondria oxidising succinate under state 2 conditions. An increase in is for example induced by ATP through its inhibitory action on the mitochondrial plant uncoupling protein (36,37). The inhibition of this uncoupling protein is unaffected by CAT and in that respect is similar to the stimulatory effect of ATP on SDH (Fig.1B). However, it is unclear whether SDH activation by ADP could be explained by this mechanism, since ADP has no reported effect(s) on the plant uncoupling protein.
Both ATP and ADP could potentially increase through their respective inhibitory action on the mitochondrial K + -channel (38) that, recently, has also been identified in plants (31). In the presence of KCl, plant mitochondria are effectively de-energised by the activity of this channel, such that succinate alone is not able to generate a (31). In the additional presence of ATP, however, a is readily established in a manner that is insensitive to atractyloside (31). The plant K + -channel is half-maximally inhibited by ATP at ~ 290 µM (31), a concentration that is two orders of magnitude higher than that required to stimulate SDH (Fig.1C). However, the ATP concentration needed to inhibit the mammalian K + -channel halfmaximally is only ~ 2.3 µM (38), a value that is in striking accordance with that calculated from  (Figs 4A and 4B).
As mentioned before, the Q-pool in potato mitochondria oxidising succinate is reduced by the combined action of the dicarboxylate carrier and succinate dehydrogenase (i.e. SDH activity). From Q-kinetic in vitro measurements it is therefore not a priori clear whether a stimulatory effect on SDH is exerted at the level of succinate transport or at the level of succinate oxidation. In vivo, succinate is generated intra-mitochondrially and hence an activating effect on transport would have little physiological relevance. It is conceivable that the observed regulation of SDH by occurs at the level of succinate-entry, since this process is affected by the energy status of the inner-membrane (40). It has been observed previously, however, that SDH activation by adenine nucleotides in cauliflower mitochondria is negatively affected by sonication or freeze-thawing treatments (5). In thawed or sonicated mitochondria, succinate should readily reach the active site of succinate dehydrogenase without the need of the dicarboxylate carrier. Such observations would therefore suggest that SDH is regulated at the level of succinate oxidation. It is arguable in general, however, whether regulation by has any in vivo importance at all, even if it would be exerted at the level of succinate oxidation. The reason for this is that it is highly unlikely that the severe degree of de-energisation, brought about by addition of CCCP or indeed by the relative force required to isolate potato tuber mitochondria, would ever occur within the plant cell under physiological conditions.
In conclusion, we have shown that in vitro SDH activity can be regulated by the energy status of the mitochondrial inner-membrane. The results suggest that the widely recognised activation of succinate dehydrogenase by ATP may also be the indirect result of such regulation.
This implies that serious caution should be taken when physiological meaning is attributed to the regulation of succinate dehydrogenase by adenine nucleotides, both in plant and mammalian systems.    with a saturating amount of ATP (T) to obtain a measure of relative activation. All assays were performed in medium A that contained ~0.9 mg mitochondrial protein (from potato tuber).  independent mitochondrial preparations (from potato tuber). Q-reduction is expressed as the fraction QH 2 in the pool and the results were modelled according to (18).