Mitochondria provide the main source of cytosolic ATP for activation of outward-rectifying K + channels in mesophyll protoplast of chlorophyll-deficient mutant rice ( OsCHLH ) seedlings*

The role of mitochondria in providing intracellular ATP that controls the activity of plasma membrane outward-rectifying K + channels was evaluated. The OsCHLH rice mutant, which lacks chlorophyll in the thylakoids, was isolated by T-DNA gene-trapping (Jung et al., 2003, Plant Cell Physiol ., 44 , 463-472). The OsCHLH mutant is unable to fix CO 2 and exhibits reduced growth. Wild type and mutant plants exhibit similar rates of respiratory O 2 uptake in the dark while the rate of photosynthetic O 2 evolution by the mutant was negligible during illumination. During dark respiration the wild type and mutant exhibited similar levels of cytoplasmic ATP. In the mutant oligomycin treatment (an inhibitor of mitochondrial F 1 F 0 -ATPase) drastically reduced ATP production. The fact that this was reversed by the addition of glucose suggested that the mutant produced ATP exclusively from mitochondria but not from chloroplasts. In whole-cell patch-clamp experiments, the activity of outward-rectifying K + channels of rice mesophyll cells showed ATP-dependent currents, which were 1.5-fold greater in wild type than in mutant cells. Channels in both wild type and mutant cells were deactivated by the removal of cytosolic ATP, whereas in the presence of ATP the channels remained active. We conclude that mesophyll cells in the OsCHLH rice mutant derive ATP from mitochondrial respiration, and that this is critical for the normal function of plasma membrane outward-rectifying K + channels. cytosolic ATP that controls K out channels. Here we show that the activity of K out channel in chlorophyll-deficient mutant cells is controlled by a cytosolic pool of ATP that is predominantly derived from oxidative phosphorylation. Cytosolic ATP production was further enhanced by the addition of glucose. These results strongly suggest that there is metabolic cross-talk between mitochondria and K out channel activity in rice chlorophyll-deficient mesophyll cells.


INTRODUCTION
Mitochondrial metabolism not only impacts on other cellular biosynthetic and catabolic processes, but also plays an important role in providing the overall cellular energy supply.
Dark respiration and photosynthesis are metabolic pathways that produce redox equivalents and ATP to meet the energy requirements to support cell growth and maintenance (2). The central role of mitochondria in plants is reflected by the array of mitochondrial transporters and shuttles that transfer metabolites and reducing equivalents back and forth between mitochondria and the cytosol. Oxidative phosphorylation provides the cytosol with ATP, which is required for sucrose synthesis and other cytosolic processes. Studies with specific respiratory inhibitors have shown that oxidative phosphorylation occurs both in darkness and in light (2)(3)(4). However, very little is known about the regulation of mitochondrial respiration in photosynthetic mesophyll cells, and the role of mitochondrial ATP production as a source of cytosolic ATP in this process is not fully understood.
Many biochemical reactions in plant and mammalian cells depend on a tightly controlled ratio of ATP to ADP. In mammalian mitochondria, this ratio is preserved by regulatory mechanisms that couple the rate of cellular ATP consumption to the rate of ATP production by oxidative phosphorylation (5)(6)(7)(8). Hence, efficient communication between cellular energy stores and membrane metabolic sensors is necessary for regulation of membrane excitability and associated functions. In most mammalian cells, there is a class of K + channels whose activity is closely coupled to metabolism by ATP (9,10). Recent studies have shown that oxidative phosphorylation, through its effect on ATP synthesis, plays an essential role in the regulation of ATP-sensitive K + channels. (11)(12)(13). However, it has also been reported that glycolytic ATP preferentially controls these channels (11,14,15). by guest on July 9, 2020 http://www.jbc.org/ Downloaded from In plant mesophyll cells, outward-rectifying K + (K out ) channels activated by membrane depolarization serve as the primary pathway for K + efflux. As K out channel activity is completely abolished (within 15 min) in the absence of cytosolic ATP (16)(17)(18)(19), it can be concluded that K + efflux in plant cells is ATP-dependent. However, the roles of photophosphorylation and oxidative phosphorylation in the regulation of K out channels remain controversial. Light activates the H + pump current (20)(21)(22)(23), K + channels (16,(24)(25)(26), and regulates membrane potential (27)(28)(29) in photosynthetic cells, suggesting that these cells are sensitive to some aspects of metabolic energy within the cytosol. Light-induced activation of the H + pump in mesophyll cells may require both photosynthetic products (i.e., sugars) and a non-photosynthetic light effect (20,28). On the other hand, the light-modulated transport system in mesophyll cells induced a rapid depolarization, which was immediately followed by repolarization and a subsequent slower hyperpolarization (16,(27)(28)(29)(30)(31). The hyperpolarization seemed to be regulated by changes in activity of the plasma membrane H + -ATPase and voltage-dependent ion channels (31)(32)(33). Analysis of metabolites in the presence of oligomycin, which blocked the F 1 F 0 -ATPase, revealed a dramatic decrease in the mitochondrial and cytosolic ATP concentration (34), indicating that mitochondria-driven ATP production was decreased. While some reports suggest that the activation of K out channels and membrane potential in mesophyll cells is regulated by ATP derived from photosynthesis (16,29), a more recent study shows that leaf mitochondria modulate whole cell redox homeostasis (35). These conflicting findings prompted us to evaluate the effects of ATP on the K out channels in mutant, chlorophyll-deficient rice cells.
To explore this in mesophyll cells, we exploited a chlorophyll-deficient rice mutant (OsCHLH), which lacks the largest subunit of Mg 2+ -chelatase (1). This rice mutant is defective in photoperception, and was therefore useful for investigating the source of by guest on July 9, 2020 http://www.jbc.org/ Downloaded from 5 cytosolic ATP that controls K out channels. Here we show that the activity of K out channel in chlorophyll-deficient mutant cells is controlled by a cytosolic pool of ATP that is predominantly derived from oxidative phosphorylation. Cytosolic ATP production was further enhanced by the addition of glucose. These results strongly suggest that there is metabolic cross-talk between mitochondria and K out channel activity in rice chlorophylldeficient mesophyll cells. 7 introduced into a leaf disc O 2 electrode chamber (Hansatech Instruments, King's Lynn, Norfolk, UK) in a closed gas exchange system of air with 5% CO 2 at 26 ± 2°C as described previously (38). After calibration, the steady-state rate of O 2 consumption was monitored until a stable rate was reached. Leaf discs were then illuminated at an irradiance of 60 and 900 µmol m -2 s -1 . Rates of net O 2 evolution and net O 2 uptake were determined as described by Walker (39). The actinic light was provided by a 150-W quartz-halogen slide-projectors fitted with heat filters through neutral density filters.
Analysis of total cellular ATP levels-The amount of ATP was measured by the luciferin-luciferase method (40) in intact seedlings (Fig. 4A) and isolated mesophyll protoplasts (Fig. 4B). The OsCHLH rice mutants were harvested about 2 h into the dark and light (90 µmol m -2 s -1 ) periods, respectively and stored in liquid nitrogen until analysis. The leaves were extracted and analyzed on the day they were sampled. The protoplasts were incubated in a temperature-controlled room (28 ± 1°C) for 40 min in the dark. The protoplasts (30 µg/mL) were incubated in incubation medium (0.6 M sorbitol, 1 mM CaCl 2 , 10 mM KCl, 10 mM Mes-NaOH (pH 6.2)) with various combinations of substrate and inhibitors. ATP was extracted from 100 µl of cell suspension by adding 2.5% trichloroacetic acid using an ENLITEN ATP assay bioluminescence detection kit (Promega, Madison, WI, USA), according to the manufacturer's recommended protocol. The extracted ATP solution was neutralized by 0.75 M Tris-acetate buffer (pH 7.75) and centrifuged for 10 min at 1,600 x g. The resulting supernatant was used in ATP determination assays. The reaction was initiated by adding 100 µl of the extract to 100 µl of the luciferin-luciferase luminous reagent (Promega). The luminescence was integrated for 5 s using a luminometer (Luminoskan Ascent 2.1 Int. Germany). The actual ATP levels were calculated from an ATP standard curve constructed using commercially supplied ATP. for the bath solution. During whole-cell recordings, the membrane potential was held at -40 mV, except during voltage steps. Currents across the membrane were measured upon imposition of a series of voltage pulses from -160 to +100 mV with 20 mV increments, as described previously (42). Liquid junction potential was measured and corrected (43). Seal resistances were in the range of 1.5 to 4 GΩ. For whole-cell experiments, application of voltage programs and handling of the data were performed using a Digidata 1200 interface and patch-clamp software pClamp8.0 with Calmpex and Clampfit (Axon instruments).
Whole-cell data were low-pass filtered with an eight-pole Bessel filter at a cut-off frequency of 2 kHz. Capacitance and pipette offset potentials were compensated during the patchclamp experiments.
Data Analysis-Group data are expressed as mean ± SE. Comparisons among groups 9

RESULTS
The OsCHLH mutation affects chlorophyll pigmentation of the rice seedlings-To study the mitochondrial contribution to the cytosolic ATP pool in photosynthetic cells, we used the chlorophyll-deficient OsCHLH rice mutant (Fig. 1A). We previously identified this mutant as a knockout of OsCHLH, the gene encoding Mg 2+ -chelatase which is involved in The growth of OsCHLH mutant seedlings was comparable to the wild type (Fig. 1A, panel a). The fresh weight of 8-day-old seedlings was 0.061 ± 0.011 and 0.031 ± 0.008 g for wild type and OsCHLH mutant plants, respectively. The addition of sucrose, which stimulates ATP generation by respiration (36), enhanced the growth in both wild type and mutant plants, showing 33.3% and 52.9% increases for the wild type and OsCHLH mutant plants, respectively (data not shown). Although the mutants did not live longer than 4-5 weeks, these data indicate that mitochondria can contribute to the OsCHLH mutants' cellular energy requirements.
Photosynthetic CO 2 uptake-The photosynthetic capacity of the OsCHLH mutant was 1 0 studied in intact plants by CO 2 gas exchange ( Fig. 2A). Whereas in the wild type there was significant CO 2 uptake in the light, there was no CO 2 uptake in the OsCHLH mutant plants.
The OsCHLH mutant plants also released more CO 2 than the wild type during dark respiration (Fig. 2B). Although it showed a substantial decrease, CO 2 (47), we next measured total cellular ATP levels. In dark treated mutant seedlings ATP level was 1.88 whereas in the light it was 0.97 µmol ATP g -1 fresh weight ( showing that oligomycin is effective in both types of cell. When glycolysis was blocked by the addition of 1 mM iodoacetic acid (I-Ac) to a substrate-free bath solution, cytosolic ATP content was decreased by approximately 20% in both mutant and wild type cells, indicating that glycolysis produced a relatively small amount of ATP. In the presence of glucose in the bath solution, ATP levels were increased by approximately 50% in both wild type and mutant cells. This effect of glucose was largely decreased when oligomycin was included with glucose in the bath solution (oligomycin decreased ATP production by approximately 66% in wild type and 71% in the mutant cells). I-Ac inhibited the production of ATP by approximately 25% in both cell types, relative to the effects of glucose alone. This implies that glycolysis in the mutant cells proceeds at similar rates to that found in the wild type.
Taken together, these findings suggest that the cytosolic ATP pool was predominantly produced by mitochondrial oxidative phosphorylation in the mutant.
Voltage-and time-dependent ATP-sensitive outward-rectifying K + channels (K out channels)-To test the voltage-and time-dependence of membrane currents, a series of voltage pulses was applied to the plasma membrane in the whole-cell configuration. Typical whole-cell current time courses in the presence or absence of ATP are shown in Fig. 5. The whole-cell patch-clamp recordings showed that rice mesophyll cells lacked the inwardrectifying K + channel activity described in tobacco (49), oat (50), and Vicia mesophyll cells (51). These mesophyll cells display a similar pattern of K + currents as a yeast mutant 1 3 down effects of outward current were observed in both cell types (Figs. 5B and 5C, right panels). The effect of ATP was statistically significant at positive membrane potentials in both wild type and mutant cells (Fig. 6). In the absence of ATP, whole cell currents were run down in similar rates in both cells; they showed about 78-80% inhibition at +100 mV as compared in those of +ATP conditions in both cells, respectively. The run-down effect of K out channel activity could be prevented in both cell types by the addition of 2 mM ATP, although the magnitude of channel current was smaller in mutant cells. It is possible that the number of K + channels in the plasma membrane is greater in wild type cells than in the mutant. The modulation of channel density is regarded as an important mechanism for controlling the transport rate across the plasma membrane in an increasing number of systems (53). This might bring about reduced growth in the mutant seedlings compared to the wild type (Fig. 1A) and therefore result in the reduced size of protoplasts (compare wild type, 16.6 ± 0.43 µm with mutant 15.1 ± 0.35 µm) (Fig. 1C). This ATP dependency of the K out channel is similar to that observed in a variety of cell types and plant species (17)(18)(19)(54)(55)(56)(57)(58).

Analysis of K out channels activated by membrane depolarization in OsCHLH rice
mutant cells-The activity of ATP-dependent K + efflux in plasma membranes was maintained in the presence of ATP in the cytosolic solution for 10 min in both wild type (Fig.   7A) and mutant cells (Fig. 7B). When cytosolic ATP was depleted using an ATP-free pipette solution, K out channel activities declined in a time-dependent manner. K out channel activity at 1 5

DISCUSSION
In animal cells there is good evidence that mitochondria are the predominant source of ATP (5)(6)(7)(8)(9)(10)(11)(12)(13). In plants, biomass production is ultimately determined by the ratio between photosynthesis and mitochondrial respiration (59). These metabolic pathways produce redox equivalents and ATP to meet the cellular energy requirements for growth and maintenance (2). However, in plants it remains unclear to what extent mitochondria contribute to cytosolic ATP levels.
In order to address this question, we investigated the activities of ATP-dependent outward-rectifying K + channels (K out channels) in mesophyll cells from seedlings of the photosynthetic capacity (Fig. 1). This was further supported by measurements of photosynthetic activity (CO 2 gas exchange (Fig. 2) and O 2 evolution (Fig. 3)). Light -induced growth of the mutant was approximately 50% of that observed in wild type plants (data not shown). Despite impaired photosynthesis and slower growth, the light-grown OsCHLH rice mutant seedlings were well developed with fully expanded leaves, indicating that growth can occur in the absence of photosynthesis (20,30). It is likely that the cytochrome and its branched alternative respiratory pathways in mitochondria coordinately balance ubiquinone pool oxidation/reduction and carbon skeleton turnover in response to cytosolic ATP levels (60). Sucose enhanced the growth of the wild type and mutant seedlings, although the growth of the mutant in the presence of sucrose was still only 58.2% of that of wild type 1 6 (data not shown). Sucrose-induced increases in growth were most likely due to enhanced metabolic pathways that produced redox equivalents and ATP via the activation of mitochondria functions.
Mitochondrial functions of the OsCHLH mutant were similar to those of the wild type, as determined by measuring O 2 concentrations in intact leaves (Fig. 3B). No statistically significant differences in O 2 consumption were obsered in dark respiration. On the other hand, the OsCHLH mutant plants released higher levels of CO 2 during dark respiration, indicating that mitochondrial activity of the OsCHLH mutant were 62.5% higher than that of wild type (Fig. 2B). This quantitative difference between O 2 uptake and CO 2 release of the OsCHLH mutant plants in the dark seems to depend on the plant age and/or the experimental conditions. Photosynthetic O 2 evolution was modest (10-15% when it was normalized at 900 µmol m -2 s -1 ) in wild type seedlings at a low irradiance of 60 µmol m -2 s -1 and robust (~4  These results are consistent with light-induced inhibition of mitochondrial respiration, which has been reviewed previously (2). Light regulation of mitochondrial ATP production was also observed in the mutant seedlings as evidenced by a dramatic reduction of ATP levels ( Fig. 4A) as O 2 consumption is closely coupled to mitochondrial ATP synthesis (2,48).
However, the decrease of ATP levels in the light is not due to CO 2 assimilation in the OsCHLH mutant as in the mutant light did not induce CO 2 uptake as it did in the wild type (compare traces in Fig. 2A). It is important to note that the decrease in O 2 consumption that  (70). A large body of evidence suggests that mitochondria are involved in photosynthetic metabolism, particularly with respect to ATP production (2). The level of ATP increased by 11-26% after exposure to light (71). In this case, the cytochrome During oxidation of the redox equivalents in the mitochondrial respiratory electron transport chain, O 2 is consumed and a proton gradient is formed across the inner mitochondrial membrane. This proton gradient provides energy for ATP synthesis in the mitochondria, a process termed oxidative phosphorylation. ATP production in darkness was similar in wild type and mutant cells (Fig. 4B), which is consistent with the similar rates of O 2 uptake (Fig. 3B). Oligomycin (2 µg ml -1 ), which is known to inhibit mitochondrial H + -ATP synthase (72), significantly inhibited the level of cytosolic ATP in both wild type and mutant cells, effectively inhibiting mitochondrial respiration. DCMU also inhibited ATP production in both wild type and mutant cells, although OsCHLH plants lack photosynthetic capacity. This appears to be due to the inhibition of cytochrome b oxidation-reduction by DCMU at one ubiquinone site, presumably ubiquinone (Q1) redox near the inner side of the mitochondrial membrane (48). The combined inhibitory effects of oligomycin and DCMU on ATP production were synergistic although the binding sites for the two compounds were specific.
In the presence of glucose as a substrate, oxidative phosphorylation was 154% and 152% of control levels, for wild type and mutant cells, respectively. Oligomycin reduced ATP levels by 66% and 70% in wild type and mutant cells, whereas it showed a much 1 9 greater effect in the absence of glucose, decreasing ATP levels by ~75% in both cell types. I-Ac inhibited ATP production by 36% in both wild type and mutant cells. Taken together, these findings suggest that glycolysis is functional in both wild type and mutant cells.
However, glycolysis, which is the initial stage of glucose metabolism, produces a relatively small amount of ATP in plant cells (Fig. 4B). Therefore, oxidative phosphorylation represents the primary source of ATP in the chlorophyll-deficient OsCHLH rice mutant.
These cells are capable of maintaining a steady-state flux of energy from mitochondrial oxidative phosphorylation, which provides ATP for the cytosolic ATPases to perform essential cellular functions.
K + is the most abundant cation in the cytoplasm of living cells, and it regulates ionic strength, osmotic potential, and membrane polarization. Permeability to K + is mediated by voltage-dependent K + channels, which are by far the best-characterized plasma membrane ion channels in plant cells (73). It has been established in a variety of cell types that activation of K out channels under steady-state conditions is dependent on cytoplasmic ATP, and a lack of ATP causes channel "run-down" (16,17,19,24,(54)(55)(56)(57)(58). The activation of various protein kinases, PP1, PP2A, and PP2C are required for the up-regulation of K out channel currents in Samanea, Tobacco and Vicia mesophyll cells (56,74,75). Our data show that rice mesophyll cells also exhibit voltage-dependent K + current and that ATP is required for its activation (Figs. 5-7). The magnitude of current in the wild type cells was greater in the presence of ATP than that of mutant (Fig. 6). The smaller whole cell current in the mutant cells might be due to a combination of environmental conditions and epistatic interactions, which are the two main factors controlling the effects of deleterious mutations (76). One could therefore expect that the number of K + channels in the plasma membrane is different in the wild type and mutant cells because the mutant seedlings were poorly by guest on July 9, 2020 http://www.jbc.org/ Downloaded from 2 0 developed (Fig. 1A), although the mitochondrial respiratory activity is quite similar to that of wild type (Figs. 3B and 4B). Interestingly, rice mesophyll cells lacked the inward K + current that is found in guard cells (16, 19, 26, 49-52, 77, 78). The ATP-activated K out channel current was maintained over periods of 10 min from the beginning of patch recording in both wild type and mutant cells.
We also found that the ATP-dependent K out channel current was largely inhibited by oligomycin, which blocks mitochondrial F 1 F 0 -ATPase, in both wild type and mutant cells (data not shown). These data further supported the idea that a cytosolic ATP pool derived from mitochondria maintained the activation of K out channel current in the mutant cells.
However, a previous study suggested that activation of K + channels, including K out and nonselective cation channels, was mediated by ATP produced photosynthetically from albino mutant (alb-1) plant cells (16). Thus, the role of mitochondria in the stimulation of the K out channel current remains controversial. More recent studies showed that light-induced membrane hyperpolarization is modulated by photosynthesis-dependent plasma membrane H + -ATPases in mesophyll cells (27,29). However, this hyperpolarization clearly occurred in response to light absorption by pigments other than chlorophyll, demonstrating that chlorophyll-deficient cells also exhibit light-induced membrane hyperpolarization (30), which differs from the findings of the earlier study (16). Taken together, the data suggest that light-driven photosynthesis is not directly involved in the activation of K out channel current in the plasma membrane, although the possibility of direct K + transport by other ATPases (i.e. K + -ATPases) cannot be ruled out (17). Mitochondrial respiration therefore provides the source of cytosolic ATP. From these results, we strongly suggest that mitochondria in the  Other experimental conditions were described in Materials and Methods.    by guest on July 9, 2020