Plant Physiol. (1998) 118: 759-771

Mitochondrial Contribution to the Anoxic Ca²⁺ Signal in Maize Suspension-Cultured Cells¹

Chalivendra C. Subbaiah^*, Douglas S. Bush², and Martin M. Sachs

Department of Crop Sciences, University of Illinois, Urbana, Illinois 61801 (C.C.S., M.M.S.); Department of Biological Sciences, Rutgers University, Newark, New Jersey 07102 (D.S.B.); and United States Department of Agriculture/Agricultural Research Service, Plant Physiology and Genetics Research Unit, Urbana, Illinois 61801 (M.M.S.)

	ABSTRACT

Top Abstract Introduction Methods Results Discussion References

Anoxia induces a rapid elevation of the cytosolic Ca²⁺ concentration ([Ca²⁺]_cyt) in maize (Zea mays L.) cells, which is caused by the release of the ion from intracellular stores. This anoxic Ca²⁺ release is important for gene activation and survival in O₂-deprived maize seedlings and cells. In this study we examined the contribution of mitochondrial Ca²⁺ to the anoxic [Ca²⁺]_cyt elevation in maize cells. Imaging of intramitochondrial Ca²⁺ levels showed that a majority of mitochondria released their Ca²⁺ in response to anoxia and took up Ca²⁺ upon reoxygenation. We also investigated whether the mitochondrial Ca²⁺ release contributed to the increase in [Ca²⁺]_cyt under anoxia. Analysis of the spatial association between anoxic [Ca²⁺]_cyt changes and the distribution of mitochondrial and other intracellular Ca²⁺ stores revealed that the largest [Ca²⁺]_cyt increases occurred close to mitochondria and away from the tonoplast. In addition, carbonylcyanide p-trifluoromethoxyphenyl hydrazone treatment depolarized mitochondria and caused a mild elevation of [Ca²⁺]_cyt under aerobic conditions but prevented a [Ca²⁺]_cyt increase in response to a subsequent anoxic pulse. These results suggest that mitochondria play an important role in the anoxic elevation of [Ca²⁺]_cyt and participate in the signaling of O₂ deprivation.

	INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

O₂ deprivation is the primary stress experienced by plants during flooding. Our previous work showed that [Ca²⁺]_cyt elevation may mediate the rapid molecular and long-lasting physiological responses to O₂ limitation (Subbaiah et al., 1994a, 1994b). Furthermore, the kinetics and magnitude of the anoxic [Ca²⁺]_cyt increase were different from the patterns of [Ca²⁺]_cyt changes induced by other stimuli in wheat aleurone cells (Bush, 1996) and Arabidopsis seedlings (Sedbrook et al., 1996). The [Ca²⁺]_cyt elevation occurring in maize (Zea mays L.) cells under anoxia did not depend on extracellular Ca²⁺ but was prevented by ruthenium red, suggesting that the Ca²⁺ signal originated from one or more of the ruthenium-red-sensitive intracellular Ca²⁺ stores (Subbaiah et al., 1994a).

The origin and spatiotemporal patterns of the [Ca²⁺]_cyt elevation are currently recognized as important elements of Ca²⁺ signaling, and the characteristic variations in these features appear to encode the qualitative and quantitative divergence of stimuli (Bush, 1995). Therefore, there has been a growing interest in the identification of the Ca²⁺ stores or channels responsible for the initiation and propagation of the [Ca²⁺]_cyt changes in specific signaling pathways (for a recent example, see Franklin-Tong et al., 1996). In the present study we traced the origin of the Ca²⁺ signal as a part of our attempt to elucidate the nature and intracellular location of the O₂ sensor. Being the primary site of O₂ consumption and also an important target of ruthenium red action, the mitochondrion could serve as a Ca²⁺ store in response to anoxia in maize cells.

Mitochondria isolated from mung bean seedlings (Moore et al., 1986), rat liver (Nishida et al., 1989), and intact rat hepatocytes (Aw et al., 1987) were shown to release Ca²⁺ from their matrix immediately after O₂ deprivation. However, these earlier analyses were carried out using organelles isolated out of the cell either before or after stimulation and thus may not represent real-time changes in an intact, living cell. In addition, the role of mitochondria in intracellular Ca²⁺ homeostasis had not been firmly established until recently (Rizzuto et al., 1994, and refs. therein). Only during the last few years has the interest in mitochondrial Ca²⁺ in the context of stimulus-response coupling been rekindled after a spurt of experimental observations (Martínez-Serrano and Satrústegui, 1992; Rizzuto et al., 1992, 1994; for review, see Gunter et al., 1994; Hajnoczky et al., 1995; Jouaville et al., 1995; Rutter et al., 1996; Babcock et al., 1997, and refs. therein). These reports indicate that mitochondria accumulate and release large quantities of Ca²⁺ and actively participate in cellular Ca²⁺ signaling.

Our knowledge of the role of mitochondria in intracellular Ca²⁺ homeostasis or cellular signaling in plant systems has been limited to only a few studies (Moore et al., 1986; Rugolo et al., 1990; Silva et al., 1992; Zottini and Zannoni, 1993; Aubert et al., 1996; Naton et al., 1996). Furthermore, there are other cellular compartments (and the plasma membrane) in plant cells in addition to mitochondria that have ruthenium-red-sensitive Ca²⁺ transporters (Brosnan and Sanders, 1993; Chason, 1994; Marshall et al., 1994; Allen et al., 1995); therefore, the scenario is more complicated than in animal cells. Recently, confocal microscopy or compartment-specific Ca²⁺ probes have been successfully used to address many long-standing questions about Ca²⁺ signaling, particularly regarding the role of subcellular compartments (Franklin-Tong et al., 1993, 1996; Rutter et al., 1996; Simpson and Russell, 1996; Babcock et al., 1997; for review, see Pozzan et al., 1994; Gilroy, 1997). In this study we combined the power of these two tools to investigate the relationship between mitochondrial and cytosolic Ca²⁺ changes in anoxic maize cells. The results indicate that mitochondria are involved in the Ca²⁺-mediated signaling of O₂ deprivation in plants.

	MATERIALS AND METHODS

Top Abstract Introduction Methods Results Discussion References

Cell Culture

Chemicals

Dye Loading

To confirm the selective uptake of rhod-2 AM into mitochondria, cells were coloaded with rhod-2 AM and a mitochondria-specific fluorophore, MitoTracker Green FM (100 nM final concentration), and their distribution was monitored. MitoTracker Green FM loads specifically into mitochondria in a membrane-potential-independent manner. The two dyes showed no spectral overlap at the settings used (at the highest sensitivity, rhod-2 AM showed a negligible blur in the MitoTracker Green FM settings). JC-1 was used at 3 µM and the cells were incubated for 15 min in the dye. All dye incubations were at room temperature (26°C ± 2°C). At the end of the incubation, cells were washed with the perfusion buffer and were embedded in agarose in a perfusion plate, as described previously (Subbaiah et al., 1994a). The perfusion plate was connected to a buffer reservoir and perfused aerobically for 15 to 20 min before the imaging was started (Subbaiah et al., 1994a).

Confocal Microscopy

The microscope was adjusted at a particular z-position of the selected cells at the beginning of an experiment and was not manipulated again until the end of the experiment. In some experiments fluo-3-loaded cells were incubated with 1 µM rhodamine B or 170 nM DiOC₆(3) for organelle localization while still on the stage. In these experiments fluo-3 contributed <1% to the total fluorescence from either rhodamine B or DiOC₆(3). Anoxia was imposed or withdrawn by withholding or restoring perfusion, as described previously (Subbaiah et al., 1994a). We attempted to obtain the spatial information concerning [Ca²⁺]_cyt from maize cells loaded with indo-1 or fura-1, the fluorescence indicators that respond to Ca²⁺ changes by spectral shifts. However, none of the ratiometric dyes or their ester forms was taken up sufficiently by the P3377 maize cells. The fluorescence from the dye-loaded cells was rarely quantifiable against the strong background of cells at the short wavelengths necessary to excite these dyes. Since confocal microscopy eliminates most of the intensity differences due to cell thickness, ratio imaging was not critical, particularly in the absence of ratiometric dyes that work in the visible range. Our choice to use nonratiometric Ca²⁺ dyes was based on careful preliminary analyses, as detailed in ``Results''.

Calibration

No attempt was made to calibrate JC-1 fluorescence with actual membrane potential values. Because JC-1 is a ratiometric dye and our interest was only in comparing the mitochondrial potential in single cells, we simply present the ratio of dye emission at 588 and 522 nm. Therefore, the data provide only a qualitative comparison of the membrane potential.

Measurement of Mitochondrial Volume

Identification of Stationary Mitochondria Using Three-Dimensional Reconstructions

Analysis of Data

View larger version (92K): [in this window] [in a new window]   Figure 2. Anoxia-induced changes in the mitochondrial Ca2+ of maize cells. The cells were loaded with rhod-2 AM and an optical section was collected where mitochondria showed no displacement. The bright spots in the images represent mitochondria. A, Cell under aerobic perfusion. B, The same cell after 10 min of anoxia. C, Anoxia-induced [Ca2+]m decrease. The optical sections were corrected for the background and the anoxic image shown in B was subtracted from the aerobic image in A. The derivative image is represented as a percentage of the aerobic image shown in A. D, Comparison of [Ca2+]m in aerobic and anoxic images. rhod-2 AM fluorescence was quantified from 16 discrete areas in each optical section (after correcting for the background) and is presented as the mean ± SE. Calibrated Ca2+ values are given on the parallel axis. The coordinates and areas of the selected locations were the same for both of the images. The changes were statistically significant (P < 0.05). The color bar is applicable only to A and B.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

[Ca²⁺]_m Changes in Response to Anoxia

View larger version (64K): [in this window] [in a new window]   Figure 1. Colocalization of rhod-2 AM and MitoTracker Green FM in maize cells incubated for 15 min before imaging by confocal microscopy. A, Intracellular distribution of rhod-2 AM in a single optical section of a maize cell. The cell was excited with a 543-nm He-Ne laser, and emission was collected at 575 nm. B, Localization of MitoTracker Green FM in the same cell shown in A. The image was collected at the same z-position as in A but with an excitation at 488 nm using an Ar ion laser and emission at 515 nm. C, Nomarski optics revealing mitochondria in the same cell. Scale bar = 5 µm.

View this table: [in this window] [in a new window]   Table I. [Ca2+]m changes during anoxia-reoxygenation cycles and FCCP treatment [Ca2+]m changes were measured as the percentage change in rhod-2 AM fluorescence. "" indicates a decrease and "+" indicates an increase in [Ca2+]m from the preexisting value. The resting [Ca2+]m ranged from 180 to 240 nM, based on our in vivo calibrations. Values are ±SE.

A majority of mitochondria, 80% of those sampled, showed a decrease in Ca²⁺ following anoxia and regained Ca²⁺ upon reoxygenation (Figs. 2 and 3B; Table I). The changes were statistically significant (P > 0.05, Fig. 2D) and reversible in subsequent anoxia-reoxygenation cycles (Table I). The data are from 16 cells in four different experiments. The reversibility of the fluorescence changes also indicated that the changes were not due to the loss of the dye but to changes in [Ca²⁺]_m. Furthermore, there was a small but significant proportion of mitochondria (10%-25%) within single cells that did not show a fluorescence decrease in response to O₂ deprivation. In a fraction of this subpopulation, fluorescence levels did not change significantly in response to the anoxia-reoxygenation treatments (Fig. 3A, green arrow), and in the rest the [Ca²⁺]_m response was inverse to the typical response (i.e. fluorescence increased under anoxia and decreased upon reperfusion; Fig. 3A, red arrow). In spite of this heterogeneity, the average response was in the direction of a net [Ca²⁺]_m release during anoxia and a net gain following reoxygenation (Table I). The pattern and magnitude of Ca²⁺ changes were similar in the motile mitochondria whose positions could be tracked within single optical slices. There was no distinct cell-to-cell variation in the responses of mitochondria to anoxia.

View larger version (41K): [in this window] [in a new window]   Figure 3. Heterogeneity in the response of mitochondria to anoxia-reoxygenation treatments in maize cells. A, Imaging of [Ca2+]m carried out as described for Figure 2. Mitochondria indicated by white arrows (an enlargement of this region is shown in B) followed the typical pattern of response, i.e. anoxic release of Ca2+ and reperfusion-induced uptake of the ion, with an average fluorescence decrease of 28% upon deaeration and a mean increase of 19.5% upon reoxygenation. The mitochondrion, indicated by the green arrow, showed only a small (<10%) decrease in fluorescence in response to anoxia and no change after reaeration. In the group marked by the red arrow, mitochondrial fluorescence increased upon anoxia (by 20%) and was restored upon reoxygenation (a 25% decrease). B, Part of A enclosing mitochondria marked by white arrows shown at a greater enlargement, highlighting the mitochondria that showed the typical anoxic response.

Do [Ca²⁺]_m Changes Contribute to Changes in the [Ca²⁺]_cyt in Anoxic Maize Cells?: Colocalization of Anoxia-Induced [Ca²⁺]_cyt Changes with Mitochondria

In the present study the [Ca²⁺]_cyt changes were imaged in 1- to 2-µm-thick optical sections of fluo-3-loaded cells using confocal microscopy. Under our loading conditions fluo-3 generally remained in the cytosol and nucleus of P3377 maize cells for at least 2 h, with very little sequestration into mitochondria or other cellular compartments (Subbaiah et al., 1994a). Furthermore, a number of related dyes that stain the cytoplasm but do not report Ca²⁺ (e.g. fluorescein diacetate and CellTracker Orange) showed no spatial redistribution within the cytosol following anoxia. In addition, there was no discernible change in the structure of the cytoplasm or organelles at least for 1 h of continued anoxia, and our O₂-deprivation treatments were generally terminated within 30 min. These observations validated our use of the nonratiometric Ca²⁺ indicator fluo-3 to report spatial changes in [Ca²⁺]_cyt and correlate them with the distribution of mitochondria and other potential Ca²⁺ stores.

From the confocal analysis it was evident that the [Ca²⁺]_cyt changes were localized only to distinct areas in the cytosol even after 10 to 20 min of anoxia (Fig. 4A). Figure 4B shows the costaining of cells with the mitochondrial dye rhodamine B. An overlay of these two images (Fig. 4C) indicates that a considerable spatial association of the domains of [Ca²⁺]_cyt increases with the mitochondrial populations. We quantified the association of the [Ca²⁺]_cyt changes with mitochondria by measuring and correlating the cytosolic distribution of Ca²⁺ increases and mitochondrial densities as shown in Figure 4D. The Ca²⁺ increases were found to be correlated significantly with the mitochondria in a majority of the measured areas (P > 0.05, Fig. 4D). In the experiment shown in Figure 4, Ca²⁺ increases greater than 15 nM (or 10 fluorescence units) showed a stronger linear correlation with mitochondrial distribution than smaller Ca²⁺ increases. Such a relationship indicates that the largest [Ca²⁺]_cyt increases occurred around mitochondria. We hypothesize that it was because of the release of Ca²⁺ from these organelles. The smaller Ca²⁺ increases may have been due to a slow diffusion of the ion away from mitochondrial locations, which is probably why they were seen at locations other than around mitochondria.

View larger version (58K): [in this window] [in a new window]   Figure 4. Colocalization of anoxia-induced [Ca2+]cyt changes with the mitochondrial distribution in maize cells. A, Confocal imaging of anoxia-induced [Ca2+]cyt changes in fluo-3-loaded maize cells. The fluorescence image is the result of subtracting an optical section of aerobically perfused cells from a corresponding section of the same set of cells after 10 min of anoxia. There are three cells in this image, which are numbered counterclockwise. In cell 1, [Ca2+]cyt increased very little; in cell 2, the largest [Ca2+]cyt increases occurred at the periphery of the cell; and in cell 3, the changes are more uniform throughout the cytoplasm. N, Nucleus. B, Mitochondrial distribution in the cells shown in A. At the end of the experiment, the cells were perfused with 1 µM rhodamine B to stain mitochondria and the optical section was obtained from the same region of the cells shown in A. C, Overlay of A with B. For the sake of clarity, the image in B was binarized and inverted to convert mitochondria into dark particles. D, Linear-regression analysis of anoxic [Ca2+]cyt changes and mitochondria. The images in A and B were aligned using the MetaMorph program. Lines (n = 11) were randomly drawn across all of the representative areas of the cytoplasm (excluding vacuoles) in the three cells in the overlay shown in C. The Ca2+ and the mitochondrial pixel intensities that intercept these lines were measured. The values were plotted, and linear curve fitting was carried out. The fluorescence values are indicated in blue and the calibrated Ca2+ values are indicated in red on the x axis. The correlation was statistically significant (P < 0.05). The results shown here are representative of nine cells. This plane was chosen for illustration because it gave a wide range of [Ca2+]cyt changes and mitochondrial distribution.

A spatial comparison of [Ca²⁺]_cyt changes was also carried out with the other potential Ca²⁺ stores using DiOC₆(3) staining (Fig. 5A). The dye stained the ER as a diffuse, network-like fluorescence and, being potentiometric, also marked mitochondria as bright, particulate fluorescence (Fig. 5A; Liu et al., 1987). DiOC₆(3) also identified vacuoles by negative staining (Fig. 5A). Because the ER was evenly distributed throughout the cytoplasm, the anoxic Ca²⁺ changes aligned with some parts of the ER network but were more spatially discrete than the ER (data not shown). However, the possibility of a heterogeneous distribution of Ca²⁺ channels on the ER (as reported for some animal cells; Rooney and Meldolesi, 1996) does not allow us to rule out the ER as a potential Ca²⁺ store under anoxia. Measurement of [Ca²⁺]_cyt changes around the vacuole revealed a significant ionic gradient, which increased away (at least up to 4 µm) from the tonoplast under anoxia (Fig. 5B), indicating that this organelle might act predominantly as a sink rather than a source for the anoxic Ca²⁺ signal.

View larger version (39K): [in this window] [in a new window]   Figure 5. Spatial comparison of anoxia-induced [Ca2+]cyt changes with the distribution of ER and vacuoles in maize cells. A, Localization of ER and vacuoles in maize cells. Cells were perfused in 170 nM DiOC6(3) and the confocal image was collected. The fluorescent regions include the ER and mitochondria, and spaces that were not stained represent the vacuoles. The brightest spots represent mitochondria, and the diffuse network-like fluorescence is the ER. B, [Ca2+]cyt gradients around the tonoplast. In subtraction images from fluo-3 experiments that depict anoxic [Ca2+]cyt changes (such as the one shown in Fig. 4A), radial lines were randomly drawn from the tonoplast to the cell periphery. The fluorescence increases that intercept the lines were measured and averaged from 22 replicates (comprising six cells). The mean fluorescence intensities and the calibrated Ca2+ values were plotted against the distance from the vacuole, using an empirical fit derived from an exponential equation, shown on the top.

We previously showed that caffeine, an agonist of the ryanodine channel located on the tonoplast and ER (Sanders et al., 1995; Muir and Sanders, 1996), induced an aerobic elevation of Ca²⁺ but failed to block a subsequent anoxia-induced [Ca²⁺]_cyt increase (Subbaiah et al., 1994a). We also found that TMB-8, an antagonist of the inositol trisphosphate-receptor (Schumaker and Sze, 1987); neomycin sulfate, an inhibitor of phospholipase C (Franklin-Tong et al., 1996); and ryanodine, an inhibitor of ryanodine channel when used at micromolar concentrations (Allen et al., 1995), had no effect on the anoxic [Ca²⁺]_cyt increase (data not shown). Although the spatial correlation studies do not completely rule out the ER as a possible Ca²⁺ source, these results indicated that the vacuole did not significantly contribute to the [Ca²⁺]_cyt elevation under O₂ deprivation. At the same time, the inverse relationship between mitochondrial and cytosolic Ca²⁺ changes in response to anoxia, the spatially discrete nature of the anoxic Ca²⁺ signal, and a significant coincidence between the [Ca²⁺]_cyt increases and the mitochondrial distribution strongly implied that Ca²⁺ entered the cytosol from mitochondria.

Although no detailed studies were made, we consistently noticed rapid dye accumulation around the nuclei of fluo-3-loaded maize cells in our previous studies (Subbaiah et al., 1994a). Our current observations using confocal microscopy showed unambiguously that fluo-3 entered the nuclei of maize cells and that the dye can be used to report nuclear Ca²⁺ changes (Fig. 4). Large fluorescence increases in the nuclear-localized fluo-3 accompanied the [Ca²⁺]_cyt increases under anoxia (Fig. 4, cells 1 and 3). However, the actual magnitude of these changes needs to be carefully determined, since a separate calibration may be necessary owing to altered properties of the dye localized in the nucleus (Burnier et al., 1994). Nuclear Ca²⁺ changes under anoxia will be an important area of focus of our future studies.

Is Anoxic [Ca²⁺]_cyt Elevation Dependent on Mitochondrial Ca²⁺ Status?

In rhod-2 AM-loaded cells, treatment with FCCP induced a rapid (within the first 5 min) loss of the mitochondrial signal (Table I). Such a loss in rhod-2 AM fluorescence was previously shown to be mainly due to a release ofmitochondrial Ca²⁺ and not to dye leakage (Hajnoczky et al., 1995; Babcock et al., 1997). These inhibitors also induce mitochondrial Ca²⁺ release in cells expressing mitochondrial aequorin, the mitochondrial localization of which is insensitive to membrane potential changes (Rizzuto et al., 1992). This suggests that the rhod-2 AM fluorescence loss in inhibitor-treated maize cells was indeed due to a loss of mitochondrial Ca²⁺. To complement the observations on [Ca²⁺]_m, the effects of these inhibitors on [Ca²⁺]_cyt were also studied in fluo-3-loaded cells. Maize cells showed a normal anoxic elevation and reoxygenation-induced recovery of [Ca²⁺]_cyt prior to treatment with FCCP (Fig. 6). Perfusion with FCCP induced a small, transient increase in the [Ca²⁺]_cyt of aerobic cells, but subsequently the inhibitor-treated cells failed to respond to anoxia by an increase in [Ca²⁺]_cyt (Fig. 6). Together, these results suggest that the [Ca²⁺]_cyt elevation in response to anoxia is dependent on the availability of Ca²⁺ in mitochondria.

View larger version (41K): [in this window] [in a new window]   Figure 6. Effect of FCCP pretreatment on [Ca2+]cyt changes in maize cells. Cells loaded in fluo-3 were given an anoxia-reoxygenation cycle. Later they were perfused with 100 nM FCCP for 10 min and anoxia was imposed on the cells. Anoxia was extended for 20 min to detect any delayed Ca2+ elevation. The [Ca2+]cyt changes were monitored by confocal imaging and quantified using the MetaMorph program. The differences in fluorescence intensity and the calibrated Ca2+ changes (means ± SE) at the end of each of the indicated treatments were plotted. Oligomycin A/antimycin A treatment (both at 2 µM) had similar effects on [Ca2+]cyt (data not shown).

Prolonged FCCP treatment under anoxia (to detect any delayed [Ca²⁺]_cyt increase) also led to a severe loss of cytoplasmic Ca²⁺ signal (Fig. 6). This could have been due to the additional effects of FCCP on other metabolic processes, particularly on cellular ATP levels (Budd and Nicholls, 1996; Jou et al., 1996, and refs. therein); therefore, the results need to be carefully interpreted. However, in view of the [Ca²⁺]_m-releasing effect of FCCP from intact cells (Table I) and from isolated maize mitochondria (Silva et al., 1992), the failure of FCCP-treated cells to respond to anoxia may be a consequence of nonavailability of Ca²⁺ in mitochondria. This is further supported by the effects of the mitochondria-specific inhibitors oligomycin A and antimycin A, which also induced a loss in mitochondrial Ca²⁺ (data not shown). Recent studies of mammalian cells also indicate mitochondria to be the sole or predominant FCCP-sensitive cellular Ca²⁺ pool (Fulceri et al., 1991; Drummond and Fay, 1996; Babcock et al., 1997, and refs. therein).

Attempts were also made to test the effects of ruthenium red on the mitochondrial Ca²⁺ changes, because ruthenium red is a potent blocker of the anoxic [Ca²⁺]_cyt increase in maize cells (Subbaiah et al., 1994a). Within minutes of adding ruthenium red to rhod-2 AM-loaded cells, the mitochondrial fluorescence decreased (data not shown). In vitro studies showed that ruthenium red quenches the fluorescence of rhod-2 AM and that this dye-drug combination could not be used for Ca²⁺ imaging in mitochondria (data not shown). However, the rapid quenching of rhod-2 AM fluorescence observed in intact cells suggests that soon after its entry into the cell ruthenium red localizes into the mitochondria and may therefore act on the mitochondrial Ca²⁺ fluxes in maize cells.

Mitochondrial Potential Changes in Relation to the Kinetics of [Ca²⁺]_m Release under Anoxia

JC-1 entered the maize cells immediately after incubation and localized into mitochondria, as observed by the punctate fluorescence (data not shown), which was very similar to rhod-2 AM or rhodamine B staining (Figs. 1A and 4B). The difference in the emission ratio at 585/522 nm of different mitochondria in the same cell and among cells ranged from 1.1 to 4.8 (mean = 3.5), signifying a great intracellular and intercellular heterogeneity in the membrane potential of maize mitochondria. Furthermore, the rapid decrease in the 585/522 nm ratio after FCCP treatment and the increase in the ratio following nigericin (a K⁺/H⁺ ionophore that abolishes a pH gradient but induces a compensatory increase in membrane potential) treatment (Table II) indicated that JC-1 responds to the changes in mitochondrial potential in plants cells as well. The effect of oligomycin A/antimycin A treatment on the emission ratio of JC-1 was ambiguous. There were no measurable changes in the mitochondrial potential during the first 20 min of anoxia (Table II). Any significant decrease in the membrane potential required at least 30 min or more of O₂ deprivation. However, [Ca²⁺]_m decreased substantially in these cells by even shorter periods of anoxia (Fig. 2C; Table I), indicating that the anoxic efflux of Ca²⁺ from mitochondria occurred predominantly by membrane-potential-independent pathways.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The long-term goal of our studies is to elucidate the pathway of O₂ sensing in plants. Our previous work in maize seedlings and cultured cells indicated that mobilization of Ca²⁺ from the intracellular pool(s) leads to an increase in the [Ca²⁺]_cyt and triggers the responses to anoxia (Subbaiah et al., 1994a, 1994b). In this report we sought the identity of the Ca²⁺ source that generates the anoxic Ca²⁺ signal. The identification of the store that releases Ca²⁺ in response to anoxia may indicate the nature of upstream events that lead to Ca²⁺ mobilization, and possibly the character of the O₂ sensor, aside from the downstream events that translate the ionic signal into molecular responses. Our analysis indicates that mitochondria act as a significant Ca²⁺ store contributing to anoxic [Ca²⁺]_cyt elevation.

Evidence for the Mitochondrial Origin of the Anoxic Ca²⁺ Signal

Until recently, the direct measurement of [Ca²⁺]_m within living cells has been problematic. The highly specific Ca²⁺ indicators of the fura-2 family enabled the measurement of changes in mitochondrial Ca²⁺ with some success but only after the fluorescence of the cytosolic dye was quenched with cytotoxic heavy metal ions (Miyata et al., 1991). A direct and unequivocal measurement of changes in [Ca²⁺]_m became possible with the development of chimeric constructs that express the Ca²⁺ indicator protein aequorin specifically in mitochondria (Rizzuto et al., 1992, 1994; Brini, 1997). These studies yielded an averaged response of [Ca²⁺]_m from several thousands of cells, although not in single cells or organelles (Rutter et al., 1996). In a more recent study, Rutter et al. (1996) measured [Ca²⁺]_m changes successfully at the single-cell level, and this analysis revealed Ca²⁺ dynamics of two major groups of mitochondria. However, mitochondrial-targeted aequorin has not yet been tested in plants.

In the present study a combination of confocal microscopy and a compartment-specific probe (rhod-2 AM) enabled us to measure real-time Ca²⁺ fluxes in a single mitochondrion or in small groups of mitochondria within single, intact, living cells. Our measurements showed that Ca²⁺ in individual mitochondria rapidly and reversibly changed in response to changes in O₂ availability (Figs. 2 and 3). Furthermore, the [Ca²⁺]_m changes were inversely related to the changes in [Ca²⁺]_cyt, suggesting that mitochondria indeed contributed to the elevation of [Ca²⁺]_cyt under anoxia and also to its restoration upon reoxygenation.

The inference that the mitochondria acted as a potential store of releasable Ca²⁺ for the anoxic [Ca²⁺]_cyt elevation was further validated by a predominant (although not complete) colocalization of [Ca²⁺]_cyt changes with mitochondria (Fig. 4). Confocal imaging technology offered sufficient spatial resolution of the Ca²⁺ increase such that we could compare the distribution of intracellular Ca²⁺ stores in the vicinity of the ionic changes. These spatial correlation studies indicated that the anoxia-induced changes in [Ca²⁺]_cyt occurred close to mitochondria (Fig. 4C) and away from the vacuole (Fig. 5B). A greater spatial coincidence of the largest [Ca²⁺]_cyt increases with mitochondria (Fig. 4D) also suggests that mitochondria could act as a primary source of the signal. The lack of a perfect correlation between the Ca²⁺ changes and mitochondria suggests that Ca²⁺ might enter the cytosol from other sources, such as by an influx of extracellular Ca²⁺ or from intracellular stores other than mitochondria. Our spatial correlation analysis with the ER did not completely rule out this organelle as a source of the anoxic [Ca²⁺]_cyt increase. However, the vacuole, which is the major intracellular Ca²⁺ pool in plant cells, appears to play no role in the anoxic [Ca²⁺]_cyt elevation (Fig. 5B), as was also indicated by the failure of pharmacological agents that are targeted to this store (and the ER as well) to inhibit the anoxic [Ca²⁺]_cyt increase (Subbaiah et al., 1994a; data not shown).

A partial correlation between Ca²⁺ changes and mitochondria also could be due to one or more of the following: a slow diffusion of the ion away from mitochondrial locations, a lack of response from some mitochondria (Fig. 3A), and the movement of some mitochondria out of the field during the later part of the imaging period. Since the intense fluorescence of rhodamine B showed a considerable overlap with the fluo-3-measurement channel, we could not stain the cells with rhodamine B either before or during a fluo-3 experiment. However, for our spatial localization analysis, only those optical planes in which the mitochondrial movement was minimum were selected. This was based on bright-field observations before the start of the fluo-3 measurements and on the serial optical sections of rhodamine B staining at the end of experiments. The spatial association of anoxic [Ca²⁺]_cyt increases with mitochondria (Fig. 4) reinforces the view that the [Ca²⁺]_cyt increase is related to the anoxia-induced decrease in [Ca²⁺]_m (Fig. 2). In addition, the emptying of [Ca²⁺]_m either by the protonophores or by the electron-transport inhibitors prevented the [Ca²⁺]_cyt elevation to a subsequent anoxic treatment (Fig. 6). Taken together, evidence supports the proposal that mitochondria play a significant role in the initiation of the anoxic Ca²⁺ signal.

Can Mitochondria Act as a Store of Mobilizable Ca²⁺?

Although resting Ca²⁺ values in mitochondria do not exceed 1 µM, the organelles appear to accumulate large pools (15-20 µM) of Ca²⁺ in actively metabolizing cells (Rutter et al., 1996; Brini et al., 1997). Even if maize mitochondria do not attain such large quantities of the ionic Ca²⁺, the capacity of mitochondria for reversible Ca²⁺ binding would sufficiently account for the [Ca²⁺]_cyt elevation under anoxia. Freshly isolated mitochondria contain a total matrix Ca²⁺ level of 5 to 30 nmol/mg and can rapidly take up Ca²⁺ to levels of 60 to 80 nmol/mg in maize and other species (Rugolo et al., 1990; Silva et al., 1991). With a reversible Ca²⁺-binding capacity estimated to be 40 times that of the cytoplasm (equivalent to >200 mM matrix Ca²⁺; Budd and Nicholls, 1996; Babcock et al., 1997), mitochondria can potentially contribute to all of the anoxic Ca²⁺ increase. For example, in hepatocytes 50% of the mitochondrial matrix Ca²⁺ was released during the first 30 min of anoxia (Aw et al., 1987).

Heterogeneity in the Mitochondrial Anoxic Response

Structural and functional variations among mitochondria within a single cell have been extensively reported. In particular, heart and skeletal muscle cells possess two distinct populations of mitochondria, intermyofibrillar and subsarcolemmal fractions, which are localized in specialized cellular compartments. Aside from the composition of membrane lipids and ions, of enzyme and respiratory activities, and of protein-import properties, etc. (Takahashi and Wood, 1996, and refs. therein), the intermyofibrillar and subsarcolemmal mitochondria also differ in their Ca²⁺-uptake/release properties (Pinsky et al., 1981, and refs. therein) and in their tolerance to anoxia-reperfusion injury (Duan and Karmazyn, 1989). In addition, plant mitochondria within single cells differ in their genome size, structure, and contents (for review, see Wolstenholme and Fauron, 1995).

Mitochondrial Ca²⁺ Release under Anoxia Is Independent of Transmembrane Depolarization

The overlapping optical properties of rhod-2 AM and JC-1 did not allow us to investigate the relationship between the membrane potential of individual mitochondria and their ability to release Ca²⁺. Nevertheless, we were able to examine whether a loss in [Ca²⁺]_m was always preceded by mitochondrial depolarization. A complete collapse of the potential brought about by FCCP treatment was correlated with a large loss in mitochondrial Ca²⁺ (Table I; Fig. 6). However, under O₂ deprivation Ca²⁺ release occurred prior to the loss of the membrane potential. In rat hepatocytes the mitochondrial membrane potential decreased by less than 20% from the aerobic values, whereas 50% of the matrix Ca²⁺ was released during the first 30 min of anoxia (Aw et al., 1987). Nishida et al. (1989), who studied the sequence of anoxia-induced events in isolated mitochondria, also observed that Ca²⁺ release preceded the loss in membrane potential. In fact, isolated maize mitochondria were shown to have a Ca²⁺-release pathway that was not dependent on a decrease in the membrane potential (Silva et al., 1992). In addition, the ability of ruthenium red to block both the electroneutral Ca²⁺ efflux from mitochondria (Gunter et al., 1994) and the anoxia-induced [Ca²⁺]_cyt elevation (Subbaiah et al., 1994a) also suggest that the Ca²⁺ release in anoxic maize cells might occur by membrane-potential-independent mechanisms.

Significance of Mitochondrial Ca²⁺ Release in Anoxia Signaling

Our results argue that the [Ca²⁺]_cyt-mediated signaling of anoxia involves the release of Ca²⁺ from mitochondria rather than mere inhibition of energy-coupled removal of the ion from the cytosol. The identification of [Ca²⁺]_m as the source of the anoxic Ca²⁺ signal would allow us to focus on the mitochondria-related events or components in our search for the O₂ sensor. The elucidation of how O₂ deprivation initiates Ca²⁺ release from mitochondria may indicate exactly where the changes in O₂ levels are sensed in the cell. A potential candidate for the O₂ sensor is the mitochondrial electron-transport chain. However, in view of the sensitivity of gene-expression changes to even mild alterations in O₂ availability (Paul and Ferl, 1991), such that the genes are induced at much higher concentrations than the K_m (O₂) of Cyt a₃, a low-affinity system could be a more appropriate sensor (such as a component of the plasma membrane redox system). The Ca²⁺ released from mitochondria may communicate the metabolic changes occurring in mitochondria under O₂ deprivation to the cytoplasm and the nucleus. Consistent with this, our preliminary observations indicate that anoxia induces large changes in the nuclear-localized fluo-3 fluorescence (e.g. cells 1 and 3, Fig. 4; also see fig. 4 in Subbaiah et al., 1994a). Furthermore, our earlier results revealed that an elevation of this ion in maize cells is indeed a requisite step in initiating rapid, compensatory metabolic and gene-expression responses to anoxia (Subbaiah et al., 1994a, 1994b).

	FOOTNOTES

   Received June 3, 1998; accepted August 9, 1998.
³   Names are necessary to report factually on available data; however, the U.S. Department of Agriculture neither guarantees nor warrants the standard of the product, and the use of the name by the U.S. Department of Agriculture implies no approval of the product to the exclusion of others that may also be suitable.

	ABBREVIATIONS

Abbreviations: [Ca²⁺]_cyt and [Ca²⁺]_m, cytosolic and mitochondrial free Ca²⁺ concentrations, respectively. DiOC₆(3), 3,3-dihexyloxacarbocyanine iodide. FCCP, carbonylcyanide p-trifluoromethoxyphenyl hydrazone. JC-1, 5,5,6,6-tetrachloro-1,1,3,3-tetraethylbenzimidazylcarbocyanine iodide.

	ACKNOWLEDGMENTS

We thank Prof. Peter Hepler (University of Massachusetts) and Dr. Imad N. Saab (University of Illinois) for critically reading the manuscript and the Beckman Institute Optical Visualization Facility (University of Illinois) for the use of their confocal microscope for the analyses shown in Figure 1. We are grateful to Dr. Ed Bonder and Mr. M. Rodriguez (Department of Cell Biology, Rutgers University, Newark, NJ), for facilities and help with the data analysis.

	LITERATURE  CITED

Top Abstract Introduction Methods Results Discussion References

Allen GJ, Muir SR, Sanders D (1995) Release of Ca²⁺ from individual plant vacuoles by both InsP₃ and cyclic ADP-ribose. Science 268: 735-737 [Abstract/Free Full Text]

Ankarcrona M, Dypbukt JM, Bonfoco E, Zhivotovsky B, Orrenius S, Lipton SA, Nicotera P (1995) Glutamate-induced neuronal death: a succession of necrosis or apoptosis depending on mitochondrial function. Neuron 15: 961-973 [CrossRef][ISI][Medline]

Aubert S, Gout E, Bligny R, Marty-Mazars D, Barrieu F, Alabouvette J, Marty M, Douce R (1996) Ultrastructural and biochemical characterization of autophagy in higher plant cells subjected to carbon deprivation: control by the supply of mitochondria with respiratory substrates. J Cell Biol 133: 1251-1263 [Abstract/Free Full Text]

Aw TY, Anderson BS, Jones DP (1987) Suppression of mitochondrial respiratory function after short-term anoxia. Am J Physiol 252: C362-C368 [Abstract/Free Full Text]

Babcock DF, Herrington J, Goodwin PC, Park YB, Hille B (1997) Mitochondrial participation in the intracellular Ca²⁺ network. J Cell Biol 136: 833-844 [Abstract/Free Full Text]

Brini M, De Giorgi F, Marsault R, Massimino ML, Cantini M, Rizzuto R, Pozzan T (1997) Subcellular analysis of Ca²⁺ homeostasis in primary cultures of skeletal muscle myotubes. Mol Biol Cell 8: 129-143 [Abstract]

Brosnan JM, Sanders D (1993) Identification and characterization of high-affinity binding sites for inositol triphosphate in red beet. Plant Cell 5: 931-940 [Abstract/Free Full Text]

Budd SA, Nicholls DG (1996) Mitochondria, calcium regulation, and acute glutamate excitotoxicity in cultured cerebellar granule cells. J Neurochem 67: 2282-2291 [ISI][Medline]

Burnier M, Centeno G, Burki E, Brunner HR (1994) Confocal microscopy to analyze cytosolic and nuclear calcium in cultured vascular cells. Am J Physiol 35: C1118-C1127

Bush DS (1995) Calcium regulation in plant cells and its role in signaling. Annu Rev Plant Physiol Plant Mol Biol 46: 95-122 [CrossRef][ISI]

Bush DS (1996) Effects of gibberellic acid and environmental factors on cytosolic calcium in wheat aleurone cells. Planta 199: 89-99

Chason A (1994) Characterization of the tonoplast Ca²⁺/H⁺ antiport system from maize roots. Plant Physiol Biochem 32: 341-346

Di Lisa F, Blank PS, Colonna R, Gambassi G, Silverman HS, Stern MD, Hansford RG (1995) Mitochondrial membrane potential in single living adult rat cardiac myocytes exposed to anoxia or metabolic inhibition. J Physiol 486: 1-13 [ISI][Medline]

Drummond RM, Fay FS (1996) Mitochondria contribute to Ca²⁺ removal in smooth muscle cells. Pflugers Arch Eur J Physiol 431: 473-482 [ISI][Medline]

Duan J, Karmazyn M (1989) Acute effects of hypoxia and phosphate on two populations of heart mitochondria. Mol Cell Biochem 90: 47-56 [CrossRef][ISI][Medline]

Fiskum G, Lehninger AL (1980) The mechanisms and regulation of mitochondrial Ca²⁺ transport. Fed Proc 39: 2432-2436 [Medline]

Franklin-Tong VE, Drfbak BK, Allan AC, Watkins PAC, Trewavas AJ (1996) Growth of pollen tubes of Papaver rhoeas is regulated by a slow-moving calcium wave propagated by inositol 1,4,5-trisphosphate. Plant Cell 8: 1305-1321 [Abstract]

Franklin-Tong VE, Ride JP, Read ND, Trewavas AJ, Franklin FCH (1993) The self-incompatibility response in Papaver rhoeas is mediated by cytosolic free calcium. Plant J 4: 163-177

Fulceri R, Bellomo G, Mirabelli F, Gamberucci A, Beneditti A (1991) Measurement of mitochondrial and non-mitochondrial Ca²⁺ in isolated intact hepatocytes: a critical re-evaluation of the use of mitochondrial inhibitors. Cell Calcium 12: 431-439 [CrossRef][ISI][Medline]

Gasbarrini A, Borle AB, Farghali H, Bender C, Francavilla A, Van Thiel D (1992) Effect of anoxia on intracellular ATP, Na⁺_i, Ca²⁺_i, Mg²⁺_i, and cytotoxicity in rat hepatocytes. J Biol Chem 267: 6654-6663 [Abstract/Free Full Text]

Gilroy S (1997) Fluorescent probes of plant cell function. Annu Rev Plant Physiol Plant Mol Biol 48: 165-190 [CrossRef][ISI]

Gunter TE, Gunter KK, Sheu S-S, Gavin CE (1994) Mitochondrial calcium transport: physiological and pathological relevance. Am J Physiol 267: C313-C339 [Abstract/Free Full Text]

Hajnoczky G, Robb-Gaspers LD, Seitz MB, Thomas AP (1995) Decoding of cytosolic calcium oscillations in the mitochondria. Cell 82: 415-424 [CrossRef][ISI][Medline]

Jou M-J, Peng T-I, Sheu S-S (1996) Histamine reduces oscillations of free mitochondrial free Ca²⁺ concentration in single cultured rat brain astrocytes. J Physiol 497: 299-308 [ISI][Medline]

Jouaville LS, Ichas F, Holmuhamedov EL, Camacho P, Lechleiter JD (1995) Synchronization of calcium waves by mitochondrial substrates in Xenopus laevis oocytes. Nature 377: 438-441 [CrossRef][Medline]

Lawrie AM, Rizzuto R, Pozzan T, Simpson AWM (1996) A role for calcium influx in the regulation of mitochondrial calcium in endothelial cells. J Biol Chem 271: 10753-10759 [Abstract/Free Full Text]

Liu, Z, Bushnell WR, Brambl R (1987) Potentiometric cyanine dyes are sensitive probes for mitochondria in intact plant cells. Plant Physiol 84: 1385-1390 [Abstract/Free Full Text]

Marshall J, Corzo A, Leigh RA, Sanders