SEARCH

SEARCH BY CITATION

Keywords:

  • oscillations;
  • glycolysis;
  • mitochondria;
  • membrane potential;
  • cyanine dye

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We employed the fluorescent cyanine dye DiOC2(3) to measure membrane potential in semi-anaerobic yeast cells under conditions where glycolysis was oscillating. Oscillations in glycolysis were studied by means of the naturally abundant nicotinamide adenine dinucleotide (NADH). We found that the mitochondrial membrane potential was oscillating, and that these oscillations displayed the same frequency and duration as the NADH oscillations. It was confirmed that DiOC2(3) localizes itself in the mitochondrial membrane and thus reports qualitative changes solely in mitochondrial membrane potential. Our studies showed that glycolytic oscillations perturb the mitochondrial membrane potential and that the mitochondria do not have any controlling effect on the dynamics of glycolysis under these conditions. Depolarization of the mitochondrial membrane by addition of FCCP quenched mitochondrial membrane potential oscillations and delocalized DiOC2(3), while glycolysis continued to oscillate unaffected. Copyright © 2007 John Wiley & Sons, Ltd.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Glycolytic oscillations in a suspension of intact cells of Saccharomyces cerevisiae have been known for several decades; for a review on oscillatory behaviour of yeast cells, see Richard (2003). Yeast cells in a high-density suspension are able to synchronize glycolysis, most likely by pulsatile secretion of acetaldehyde (Richard et al., 1994, 1996b; Dano et al., 1999; Poulsen et al., 2004, 2007), and each individual cell oscillates in phase with the cell population. The oscillations are monitored routinely by fluorescence measurements of the naturally abundant fluorophores NADH and NADPH (Richard et al., 1993), where the NAD+/NADH couple participates in redox reactions in glycolysis. NADH and NADPH cannot be distinguished by their fluorescence but only NADH is found to oscillate in yeast, while NADPH remains constant (Richard et al., 1993) during glycolytic oscillations. From here on we therefore refer only to NADH.

Glycolytic oscillations are evoked under anaerobic (Poulsen et al., 2004) or semi-anaerobic conditions by the addition of cyanide (Betz and Chance, 1965a). Hence, oxidative phosphorylation in mitochondria is shut down before oscillations appear. However, whether the mitochondria still have a controlling role with respect to glycolytic oscillations has been subject to speculation.

According to Aon et al. (1991), depolarization of the mitochondrial membrane annihilated cyanide-induced oscillations. The authors found that it was not possible to induce a train of glycolytic oscillations when yeast cells were preincubated with carbonyl cyanide p-(trifluoromethoxy)phenylhydra- zone (FCCP), which discharges the electrochemical proton gradient across the mitochondrial membrane. They therefore concluded that mitochondrial function played a role in the triggering and dynamics of glycolytic oscillations through either NADH, ATP or membrane potential.

In addition to oscillations in NADH, oscillations in several glycolytic intermediates, including ATP, have been investigated in yeast (Richard et al., 1993, 1996a; Betz and Chance, 1965a). However, to our knowledge no measurements of mitochondrial membrane potential during glycolytic oscillations have yet been presented. On-line non-invasive methods for monitoring of membrane potential exist in the form of fluorescent probe molecules, such as the carbocyanines, oxonols and rhodamines (Plasek and Sigler, 1996). Each dye possesses specific attributes, in the form of hydrophobicity, charge, size, etc., which optimize them for specific organisms, organelles and response times.

DiOC2(3) is a cationic carbocyanine dye which is distributed across cellular membranes according to charge. Due to the positive charge, the dye is expected to accumulate in the mitochondria. At low dye concentrations (low nM scale) this allows quenching of the green fluorescence of the monomeric dye to mirror the membrane potential. In addition, DiOC2(3) displays a red fluorescence upon aggregation at higher concentrations of dye (low µ M range). This aggregation is also dependent on charge and thus the dye displays an increase in red fluorescence upon hyperpolarization of the membrane (Novo et al., 1999; Shapiro, 2000). There is some discrepancy in the literature as to whether the green fluorescence of DiOC2(3) increases or decreases upon hyperpolarization, and the changes of dye behaviour unrelated to membrane potential still need further investigation (Wolk and Hofer, 1987; Plasek and Sigler, 1996; Shapiro, 2000; Breeuwer and Abee, 2004). However, it is generally agreed that DiOC2(3) localizes itself in the mitochondrial membrane due to the mitochondrial transmembrane potential (see Shapiro, 2000; Johnson et al., 1981; for review, see Chen, 1989).

In this study we show that DiOC2(3) in the µ M concentration range can report qualitative changes in mitochondrial membrane potential in S. cerevisiae. In addition, we found that although DiOC2(3) is toxic to yeast cells, the fluorescent probe does not alter the dynamics of glycolysis in these cells. Monitoring DiOC2(3) in yeast cells while glycolysis was oscillating revealed oscillations in mitochondrial membrane potential with a frequency and shape similar to those observed in NADH fluorescence. To our knowledge, this not been shown previously. Studies of the effect of the proton gradient uncoupler FCCP showed that DiOC2(3) delocalized in the cells, as a consequence of mitochondrial membrane potential depolarization and oscillations in Δψ, stopped immediately upon the addition of FCCP, while the NADH continued to oscillate unaffected.

We conclude that the mitochondrial membrane potential is perturbed by the glycolytic oscillations and that the mitochondria have little or no regulating effect on glycolytic oscillations.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Reagents and chemicals

Phthalic acid buffer, potassium phosphate buffer, KCN and glucose were obtained from Merck (Germany). Carbonyl cyanide p-(trifluoromethoxy) phenylhydrazone (FCCP) and oligomycin were from Sigma-Aldrich (Germany). The fluorescent probes 3,3′-diethyloxacarbocyanine iodide (DiOC2 (3)) and bis-(1.3-dibutylbarbituric acid)-trimethine oxonol (DiBAC4(3)) were purchased from Molecular Probes (USA), while yeast nitrogen base (YNB) was obtained from DIFCO Laboratories (MI, USA).

Microorganism and growth conditions

Yeast cells, Saccharomyces cerevisiae diploid strain X2180 was grown at 30 °C on a rotary shaker at 150 r.p.m. The growth medium contains 10 g/l glucose, 6.7 g/l YNB in 100 mM phthalic acid buffer, pH 5.0. When glucose in the growth medium was depleted, as measured with a glucose test strip (Macherey-Nagel, Germany), the cells were washed twice with 100 mM phosphate buffer, pH 6.8 (centrifugation for 5 min at 5000 r.p.m.; GSA Sorvall, USA). The yeast cells were resuspended in the same buffer to a cell density of 10% wet weight and starved for 3 h on a rotary shaker (150 r.p.m. at 30 °C). Following the period of starvation, the cells were kept on ice until use. Unless stated otherwise, cells were supplied with 6 µ M of the membrane potential-sensitive dye [DiOC2(3)] at time 0 in all experiments.

Spectrofluorometric measurements

NADH auto-fluorescence (excitation 366/3 nm, emission 450/3 nm) and membrane potential-dependent DiOC2(3) fluorescence (excitation 488/3 nm, emission 600/3 nm) in yeast cell suspension were measured by an Edinburgh Spectrofluorometer Model FS920 (Edinburgh, Scotland). The suspension was contained in a quartz cuvette and stirred. Fluorescence microscopy was carried out using an inverted Leica microscope (DMIRE2) equipped with a DC350FX cooled colour camera. The light source was a 100 W mercury lamp. The fluorescent dye DiOC2(3) in cells was excited using 480/40 nm and 560/40 nm bandpass filters and 527/30 nm and 645/75 nm bandpass emission filters, respectively.

Determination of DiOC2(3) toxicity

Glucose-fermenting yeast cells at a cell density of 10% wet weight were exposed to 6 µ M DiOC2(3) for approx 25 min; the cells were then washed in 100 mM potassium phosphate buffer, pH 6.8, and resuspended in dye-free buffer. The cell suspension was diluted five, six, seven and eight times and spread onto agar plates in triplicate. The plates were incubated at 30 °C until colonies were visible (ca. 48 h) and the colonies were counted. A control without exposure to 6 µ M DiOC2(3) was performed for comparison.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Validation of dye specificity and toxicity

From experiments on simple organisms such as bacteria, it is known that changes in both green fluorescence (530 nm) and red fluorescence (600 nm) of the fluorescent dye DiOC2(3) can reflect changes in plasma membrane potential (for review, see Plásek and Sigler, 1996). In contrast to bacteria, yeast cells maintain a membrane potential across both the plasma and the mitochondrial membranes. Due to the positive charge of the fluorescent probe, the dye is expected to equilibrate according to the Nernst equation across both the plasma and the mitochondria membranes; thus, the dye is expected to accumulate in the mitochondria. This is a common feature of cationic dyes and has also been observed with a similar dye in a previous study on yeast (Pena et al., 1984). To examine whether the dye fluorescence mainly stems from mitochondria as predicted, the dye distribution was studied by fluorescence microscopy. Figure 1 shows yeast cells incubated with 6 µ M DiOC2(3), 30 mM glucose and 5 mM KCN (conditions which induce oscillations in NADH), with (A) a phase contrast image of a representative yeast cell sample and (B and C) images of green and red fluorescence, respectively, of DiOC2(3) in the same cells.

thumbnail image

Figure 1. Phase contrast image (A) and DiOC2(3) fluoresc- ence images of green (B) and red (C) fluorescence. Note how the intensity of the fluorescence, both red and green, is unevenly distributed in the cells and seems to be accumulated in intracellular organelles, more specifically in the mitochondria. In addition, the dye intensity varies from cell to cell but all cells display dye confined in several intracellular compartments

Download figure to PowerPoint

The microscopy images reveal that the dye is accumulated in intracellular compartments, as predicted, and the plasma membrane is not visible. In addition, the nuclei, which are visible in the cells of Figure 1A, are not stained, and can be seen in Figure 1B as dark patches. Thus, the vast majority of fluorescence stems from the mitochondria. We can therefore conclude that any changes in red or green fluorescence result from changes in the mitochondrial membrane potential, and potential changes in fluorescence intensity due to changes in plasma membrane potential are insignificant, due to overriding fluorescent signal from the mitochondria.

To investigate whether the plasma membrane displayed a significant contribution to changes in membrane potential, a plasma membrane-specific membrane potential dye was tested. DiBAC4(3) is an anionic distributional potential dye which, due to its negative charge and hydrophobicity, should be restricted to the plasma membrane (de Poorter and Keltjens, 2001). It was found that this dye was indeed limited to the extracellular space and plasma membrane, but at the low concentration needed to allow membrane potential-dependent quenching by the cells, the background light from NADH auto-fluorescence exceeded the light emitted by the dye (data not shown). In conclusion, the dye DiOC2(3) was found to be specific for the mitochondrial membrane potential at the concentrations used, whereas the use of the plasma membrane-specific probe DiBAC4(3) was found to be unsuited for this type of measurement.

As DiOC2(3) has previously been shown to be toxic (Novo et al., 1999), the dye may have additional and perhaps undesired effects on yeast metabolism, specifically on glycolysis. In order to investigate the immediate toxicity of the dye, the effects of adding dye to the yeast cells during glycolytic oscillations was studied. As can be seen in Figure 2 (black trace), adding 6 µ M DiOC2(3) to oscillating yeast cells caused an immediate drop in NADH fluorescence but the oscillations continued un-altered in frequency and duration. The absorption spectra of NADH and DiOC2(3) were therefore examined and it was found that the drop was exclusively due to light absorption by DiOC2(3) (data not shown). The effect can be seen in Figure 2 (grey trace), where the measurements subsequent to DiOC2(3) addition have been multiplied by a correction factor. The factor corresponds to the value of the last peak fluorescence before probe addition, divided by the third (allows signal stabilization) peak fluorescence after DiOC2(3) addition. As can be seen in the time series, this completely eliminated the drop in NADH fluorescence and it is evident that the glycolytic oscillations are unperturbed by addition of the fluorescent probe. A similar absorption relationship was found between NADH and the proton gradient uncoupler FCCP.

thumbnail image

Figure 2. Effect on NADH fluorescence of adding 6 µ M DiOC2(3) to oscillating yeast cells. Oscillations were induced by the addition of 30 mM glucose and 5 mM KCN at times 1.3 and 2.5 min, respectively. The NADH emission at 450 nm (black) displays an immediate drop subsequent to probe addition. Applying a correction factor to the data (grey), based on DiOC2(3) absorption data, reveals that DiOC2(3) has no physiological effect on the glycolytic oscillations but partially absorbs the light at the NADH excitation wavelength (366 nm)

Download figure to PowerPoint

Although there is no immediate effect of adding the cyanine dye to oscillating yeast cells, the cells may be damaged by exposure to the dye for a longer period of time. The extent of this damage was tested by performing growth experiments on yeast cells exposed to DiOC2(3). The colony count revealed that 14% of the cells died subsequent to 25 min exposure to 6 µM dye; thus, although the dye is toxic to yeast, the majority of the exposed cells are viable.

Membrane potential oscillations

The measurement of mitochondrial membrane potential in yeast cells displaying glycolytic oscilla- tions revealed that mitochondrial membrane potential (Δψ) oscillated as well. To our knowledge, this has never been shown previously. Two representative time series showing NADH and Δψ oscillations (black and grey traces, respectively) are plotted in Figure 3. The oscillations were induced by addition of 5 mM potassium cyanide and 30 mM glucose. The addition of 5 mM cyanide causes respiration to cease; this was tested in separate experiments in which dissolved oxygen was measured with membrane inlet mass spectrometry. It was verified that the addition of cyanide effectively blocked oxygen uptake by glucose-fermenting yeast cells, implying that respiration was 100% inhibited. Due to instrumental limitations, only one emission wavelength could be measured at a time, and therefore the time series in Figure 3 stem from two separate but otherwise identical experiments. As cytosolic and mitochondrial NADH cannot be distinquished by fluorometry, it cannot be ruled out that mitochondrial NADH contributes to the signal observed in Figure 3. However, it is well established that the oscillations observed stem from glycolysis; thus, mitochondrial NADH must either oscillate in perfect synchrony or provide a stable background signal. The observed initial slow increase in DiOC2(3) fluorescence is caused by the slow staining of the mitochondria upon dye addition at time 0. When cells are pre-incubated with DiOC2(3) for 15 min, membrane potential oscillations start simultaneously with oscillations in glycolysis (data not shown). The membrane potential and NADH oscillations shown in Figure 3 are initiated using identical experimental conditions and times of addition of glucose and cyanide. Hence, the oscillations have the same duration and, in addition, the frequencies seem to be identical. As the measurements of both NADH and Δψ are strictly qualitative, it is not possible to conclude anything from the amplitude of the oscillations.

thumbnail image

Figure 3. Oscillations in NADH auto-fluorescence (black) and mitochondrial membrane potential (grey). Glycolytic oscillations in 10% wet weight starved yeast cell suspensions were initiated by the addition of 30 mM glucose at time 1 min and 5 mM KCN at 1.5 min. The two time series were obtained in separate but identically designed experiments, and show that the period of the oscillations is on average 33 s; however, this varied slightly during the experiments

Download figure to PowerPoint

Note in Figure 3 how the oscillations are 180° out of phase at time 13 min but in phase at time 17 min. This observation illustrates that the period of the glycolytic oscillations varies slightly, which hampers the examination of the phase relationship between NADH and Δψ oscillations. In order to determine whether or not the oscillations were in phase, a fast (15–20 s) switch between measurement of NADH and Δψ was performed and timed during oscillations. The oscillating transients were extended for three periods in order to obtain an overlap. This can be seen in Figure 4, in which the oscillations of NADH (black and dark grey trace) and Δψ (grey trace) seem to be almost in phase, with Δψ lagging slightly behind NADH. The experiment was performed several times, but as the accuracy relies mainly on the manual timing of the shift between NADH and Δψ measurements, it is not possible to determine whether the delay is in fact a feature of the system or just an error in the timing. A determination of the exact phase relationship would require precise and simultaneous measurements and data acquisition. However, the experiments conducted revealed that the period of the oscillations in NADH and Δψ were identical when compared within the same experiment (as opposed to comparing periods of separate experiments, as in Figure 3).

thumbnail image

Figure 4. Examination of phase relationship between oscillations in NADH and mitochondrial membrane potential. In order to circumvent the fact that only one emission wavelength can be monitored at a time with our set-up, the time series of NADH oscillations (black) were copied and horizontally displaced in order to extend the NADH time series (dark grey). This replacement results in an overlap between the NADH and Δψ (grey), which allows the phase relationship to be visualized

Download figure to PowerPoint

Preincubation and perturbation experiments with FCCP

In order to investigate the relation between glycolytic oscillations and mitochondrial membrane potential, we applied the well-known uncoupler of mitochondrial proton gradient, FCCP, to oscillating yeast cells. The addition of 20 µ M FCCP to an oscillating yeast cell suspension caused an immediate depolarization of the mitochondrial membrane potential; in addition, all Δψ oscillations were abolished (see Figure 5, grey trace). This was expected, as concentrations as low as 2.5 µ M have been shown to abolish yeast mitochondrial membrane potential for hours (Lloyd, 2003). As can be seen in Figure 5 (black trace), no change in NADH fluorescence was observed. Thus, Δψ was depolarized, whereas the NADH oscillations continued with no detectable change in period, amplitude or train duration.

thumbnail image

Figure 5. The effect of addition of 20 µ M FCCP at time 15 min to yeast cells displaying glycolytic oscillations. Upon addition of FCCP, NAD(P)H oscillations continued unaltered (black), whereas the membrane potential decreased rapidly and oscillations disappeared (grey)

Download figure to PowerPoint

It has previously been shown that yeast cells preincubated for 20 min with as little as 10 nM FCCP were unable to initiate glycolytic oscillations (Aon et al., 1991). Therefore, we preincubated 10% wet weight yeast cell suspensions (five times the cell density used by Aon et al.) with 50 and 500 nM FCCP for 20–25 min. Upon addition of 30 mM glucose and 5 mM KCN, the preincubated cells displayed oscillations in both glycolysis and mitochondrial membrane potential (data not shown). At preincubation with 500 nM FCCP, the oscillations were visibly affected. This could be observed as a prolonged lag phase before oscillations could be seen and a decrease in NADH oscillation amplitude. Nevertheless, oscillations were clearly present in both NADH and mitochondrial Δψ.

As preincubation with FCCP may cause different effects from perturbations with FCCP during glycolytic oscillations, perturbation experiments identical to that presented in Figure 5 were conducted with 50 nM, 500 nM and 2 µ M FCCP. This had no effect visible on NADH oscillations (data not shown), whereas the response of the membrane potential varied with the added concentration of FCCP (see Figure 6). At 50 nM FCCP (Figure 6, black trace), no significant change in mitochondrial Δψ was observed. Upon addition of 500 nM FCCP (Figure 6, dark grey trace), the membrane potential decreased while displaying damped oscillations. Finally, at 2 µ M FCCP (Figure 6, grey trace), the membrane potential decreased immediately, but with a slower rate than observed in Figure 5. Low-amplitude oscillations were visible during this decrease, but these can be ascribed to background NADH fluorescence. Thus, neither preincubation nor perturbation with 50 nM FCCP caused any change in the oscillating yeast in our experiments. 500 nM FCCP did not prevent initiation of glycolytic oscillations, but did affect the cells. This was observed as a prolonged lag phase before initiation of oscillations and a damping of mitochondrial membrane potential oscillations. Finally, perturbing the cells with 2 µ M FCCP caused a response similar to that observed in Figure 5. Note that no background oscillations are observed at 20 µ M FCCP in Figure 5. This is due to the fact that FCCP at this concentration absorbs a significant amount of the background light emitted by NADH. As FCCP uncouples the mitochondrial proton gradient, we tested the possible effect of inhibiting the F0F1-ATPase. This was done by perturbing oscillating yeast cells with 30 µ M oligomycin (data not shown). No effect of adding this inhibitor was seen, either on NADH or on membrane potential oscillations.

thumbnail image

Figure 6. Qualitative membrane potential measurements during FCCP perturbations of yeast cells displaying glycolytic oscillations. The cells were perturbed at time 15 min with 50 nM (black), 500 nM (dark grey) and 2 µ M (grey) FCCP, during oscillations induced in starved yeast cell suspensions (10% wet weight) by the addition of 30 mM glucose at time 1.3 min and 5 mM KCN at time 2.5 min

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

In the present study, observations of oscillations in the mitochondrial membrane potential of yeast are shown. To our knowledge, this has not been presented previously. The oscillations are visualized by on-line fluorescence measurements of the membrane potential-sensitive probe DiOC2(3). By the use of fluorescence microscopy, we established that DiOC2(3) is localized in the mitochondrial membrane and that the observed oscillations are caused solely by oscillations in mitochondrial membrane potential. In addition, the autofluorescence of the nicotinamide nucleotides was recorded. NADH participates in several of the glycolytic reactions and therefore allows the observation of glycolytic oscillations in yeast.

Comparing glycolytic oscillations with the re- corded membrane potential oscillations revealed that both the frequency and the duration of the oscillations were identical; however, it was not possible to determine whether the oscillations were in phase or whether the oscillations in mitochondrial membrane potential lagged slightly behind NADH oscillations (<5 s). Upon uncoupling of the mitochondrial proton gradient by FCCP, the mitochondrial membrane potential collapsed and oscillations in membrane potential were abolished. The glycolytic NADH oscillations continued unaffected. The observation that mitochondrial membrane potential oscillations can be disrupted without affecting the glycolytic oscillations can point to two different explanations: either the mechanisms controlling mitochondrial membrane potential include an oscillator unrelated to the glycolytic oscillations, or the glycolytic oscillations impose oscillations on the mitochondrial membrane potential. Taking into account that the oscillations display the same frequency and duration, we strongly believe in the latter explanation. Thus, we propose that during glycolytic oscillations induced by glucose and potassium cyanide, the mitochondrial membrane potential is perturbed by these oscillations without exerting any measurable effect back into glycolysis.

A study of the electron transport chain (ETC) in the inner mitochondrial membrane reveals a number of possible explanations for this behaviour. S. cerevisiae is known to be petite-positive, which refers to its ability to survive and proliferate despite the loss of functional mitochondrial DNA, encoding proteins of complex III (bc1 complex), complex IV (cytochrome c oxidase) and the proton-translocating subunit of complex V. Despite the loss of almost the entire ETC, S. cerevisiae maintains a mitochondrial membrane potential, which is vital for the targeting of proteins for the mitochondria (Chen and Clark-Walker, 2000). Prior to the observed oscillations, 5 mM of potassium cyanide was added to the cell suspension. Cyanide is a known inhibitor of complex IV of ETC and at this concentration respiration was effectively blocked, which could be observed by the cessation of oxygen consumption (Richard et al., 1994). This means that the citric acid cycle was inactive and did not load electrons into the ETC. In addition, S. cerevisiae under these conditions ferment glucose almost exclusively to ethanol, which is a redox neutral reaction (Vandijken et al., 1993). This means that electrons from cytosolic NADH accessible to the NADH oxidoreductase or the glycerol 3-phosphate shuttle are scarce. The NADH oxidoreductase and the glycerol 3-phosphate shuttle are mechanisms positioned on the exterior of the inner mitochondrial membrane, which use ubiquinone as electron acceptor for shuttling electrons into the ETC (Boumans et al., 1998; Velazquez and Pardo, 2001; Bakker et al., 2001; Hunte et al., 2003). They cause no direct transfer of protons across the membrane, but cytosolic NADH could indirectly perturb Δψ through these mechanisms.

The most likely way for the cells to maintain a mitochondrial membrane potential under the applied conditions is therefore through cytosolic ATP. Cytosolic ATP concentration has been shown to oscillate during glycolytic oscillations (Richard et al., 1996a; Betz and Chance, 1965a). Upon inhibition of the ETC, ATP4− is translocated into the mitochondria in exchange for ADP3− through the adenine nucleotide translocator. This is an electrogenic transport which has been shown to be sufficient to create and maintain a mitochondrial membrane potential (Giraud and Velours, 1997). In addition, hydrolysis of the imported ATP may fuel the translocation of protons across the inner mitochondrial membrane through the F0F1-ATPase, thus creating a proton gradient, which adds to the membrane potential (Lefebvre-Legendre et al., 2003). As ATP in the cytosol oscillates with an amplitude of almost 50% relative to maximum ATP concentration (Richard et al., 1996a), these oscillations may very well be conveyed in the membrane potential. However, the failure of oligomycin to quench the membrane potential oscillations indicates that the F0F1-ATPase proton-pumping activity plays a minor role at most in maintaining oscillations in mitochondrial Δψ.

The above represent reasonable suggested explanations for mechanisms involved in the coupling of glycolytic oscillations and mitochondrial membrane potential oscillations. A thorough study is necessary in order to deduce the true connection. However, based upon the observations presented here, we conclude that mitochondrial membrane potential does indeed oscillate and suggest that these oscillations represent perturbations from glycolysis.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The authors wish to thank Ole Michelsen (Technical University of Denmark) for help with establishing the membrane potential measurement method, and laboratory technician Anita Lunding (University of Southern Denmark) for her commitment and always flawless work in the laboratory. This research was supported by the Danish Natural Science Research Council.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References