Skip to main content

Release of Ca2+ and Mg2+ from yeast mitochondria is stimulated by increased ionic strength



Divalent cations are required for many essential functions of mitochondrial metabolism. Yet the transporters that mediate the flux of these molecules into and out of the mitochondrion remain largely unknown. Previous studies in yeast have led to the molecular identification of a component of the major mitochondrial electrophoretic Mg2+ uptake system in this organism as well as a functional mammalian homolog. Other yeast mitochondrial studies have led to the characterization of an equilibrative fatty acid-stimulated Ca2+ transport activity. To gain a deeper understanding of the regulation of mitochondrial divalent cation levels we further characterized the efflux of Ca2+ and Mg2+ from yeast mitochondria.


When isolated mitochondria from the yeast Saccharomyces cerevisiae were suspended in a salt-based suspension medium, Ca2+ and Mg2+ were released from the matrix space. Release did not spontaneously occur in a non-ionic mannitol media. When energized mitochondria were suspended in a mannitol medium in the presence of Ca2+ they were able to accumulate Ca2+ by the addition of the electrogenic Ca2+ ionophore ETH-129. However, in a KCl or choline Cl medium under the same conditions, they were unable to retain the Ca2+ that was taken up due to the activation of the Ca2+ efflux pathway, although a substantial membrane potential driving Ca2+ uptake was maintained. This Ca2+ efflux was independent of fatty acids, which have previously been shown to activate Ca2+ transport. Endogenous mitochondrial Mg2+ was also released when mitochondria were suspended in an ionic medium, but was retained in mitochondria upon fatty acid addition. When suspended in a mannitol medium, metal chelators released mitochondrial Mg2+, supporting the existence of an external divalent cation-binding site regulating release. Matrix space Mg2+ was also slowly released from mitochondria by the addition of Ca2+, respiratory substrates, increasing pH, or the nucleotides ATP, ADP, GTP, and ATP-gamma-S.


In isolated yeast mitochondria Ca2+ and Mg2+ release was activated by increased ionic strength. Free nucleotides, metal ion chelators, and increased pH also stimulated release. In yeast cells this release is likely an important mechanism in the regulation of mitochondrial matrix space divalent cation concentrations.


Mg2+ is the most abundant divalent cation in cells and is important for many metabolic processes. Free Mg2+ in cells is quite low (0.3 – 3%) as compared to the total amount of Mg2+ due to buffering by polyphosphates, nucleic acids, and free nucleotides [1, 2]. In mammalian cells the free concentration is usually maintained at a concentration between 0.25 and 1 mM [2]. The Mg2+ level in the cell varies during changing physiological conditions. The free Mg2+ level increases slightly as ATP levels decrease. The total cellular Mg2+ level changes upon the addition of several hormones to cells [3]. Recently the TRPM family of ion channels has been found to play an important role in cellular Mg2+ homeostasis in several tissues [4]. This suggests that regulated pathways for Mg2+ transport exist to maintain proper intracellular levels.

Mammalian mitochondrial Mg2+ transport

The pathways involved in mammalian mitochondrial Mg2+ flux are largely unknown. The human homolog of a component of the main electrophoretic mitochondrial Mg2+ transporter in yeast has been cloned [5]. The protein can functionally substitute for its yeast homolog. These double membrane-spanning proteins likely homo-oligomerize to form ion-transporting complexes. However definitive evidence that this protein functions in mammalian mitochondrial ion transport has yet to be obtained.

Heart mitochondria take up and release Mg2+ in a respiration-dependent manner [68]. Uncouplers of respiration that dissipate the membrane potential release mitochondrial Mg2+ into the cytoplasm [9]. There is evidence that hormones alter mitochondrial Mg2+ levels [10]. Hormones may increase cyclic AMP (cAMP) levels to stimulate Mg2+ transport [11]. Fatty acids have been shown to stimulate the efflux of mitochondrial Mg2+ at alkaline pH [12]. Even though mitochondrial Mg2+ has been well characterized at the isolated organelle level, the cellular signals regulating mitochondrial Mg2+ transport have not been clearly elucidated.

Mg2+ transport in yeast

In yeast cells when medium Mg2+ is plentiful (> 1 mM), yeast cells contain around 400 nmole Mg2+/mg protein [1]. As medium Mg2+ decreased (0.02 mM), cellular Mg2+ decreased to levels near 80 nmole/mg protein. In times of ample Mg2+, the Mg2+ is taken up and stored in the vacuole. When these cells were then placed in a Mg2+ free growth medium, they survived by utilizing their vacuole stores. But once the cellular Mg2+ level became too low, the yeast cells died, in contrast to limitations in many other nutrients in which the cells survive in the G0 phase of the cell cycle [1].

Two genes comprising the yeast Mg2+ uptake system in the plasma membrane have been identified [13]. Yeast plasma membrane Mg2+ transport is inhibited by aluminum [13] and polyamines [14]. Since overexpression of either of the Mg2+ transport proteins confers resistance to aluminum toxicity, the inhibition of Mg2+ uptake may be the primary cause for aluminum toxicity in yeast [13]. Mg2+ may also provide benefits to cells confronted with heavy metals. Increased cellular Mg2+ levels protected cells from toxic concentrations of Mn2+ by down-regulating Mn2+ transport [15].

Yeast mitochondrial Mg2+

Mg2+ is essential for many mitochondrial reactions including those of DNA and RNA metabolism [16, 17] as well as ATP synthesis [18]. Mg2+ blocks the yeast mitochondrial ATP-induced unspecific channel (YMUC) [19]. Mg2+ also inhibits mitochondrial anion uniport [20]. Therefore, the regulation of matrix space Mg2+ could be important for the mitochondrial distribution of many other ions. In this report we discovered that Mg2+ and Ca2+ are released from yeast mitochondria under conditions of increased ionic strength. In yeast cells this release pathway is likely an important regulator of mitochondrial divalent cation levels.


Ionic media prevent ETH-129-mediated accumulation of mitochondrial Ca2+

In contrast to mammalian mitochondria, energized yeast mitochondria do not elecrophoretically take up Ca2+ or undergo an increase in non-specific inner membrane permeability in the presence of Ca2+ [21]. However energized yeast mitochondria open an unspecific channel (YMUC), which dissipates the membrane potential in the absence of the YMUC inhibitors phosphate or decavanadate. Yeast mitochondria can accumulate Ca2+ by the addition of the electrogenic Ca2+ ionophore ETH-129 to intact cells [22] or to energized mitochondria in a mannitol medium [21]. When suspended in a 0.3 M KCl medium, energized mitochondria were unable to accumulate Ca2+ after ETH-129 addition as measured by mitochondrial-targeted aequorin [22] or the presence of the Ca2+-indicating dye antipyrylazo III (Figure 1A). Less Ca2+ was accumulated as the KCl concentration increased. Increasing the KCl concentration to 0.3 M only allowed less than 30 percent of the medium Ca2+ to be accumulated. It is important to note that ETH-129 has no transport affinity for K+ and the presence of KCl does not affect the binding of ETH-129 to Ca2+ [23]. After all of the medium Ca2+ was taken up by mitochondria suspended in a mannitol medium, the addition of 0.3 M KCl caused the slow efflux of Ca2+ from the matrix space. This efflux of Ca2+ did not occur when an equal amount of sorbitol was added. Similarly to mitochondria suspended in KCl media, mitochondria suspended in increasing concentrations of choline Cl media osmotically balanced with mannitol were increasingly unable to accumulate high levels of Ca2+ by the addition of ETH-129 (Figure 1B).

Figure 1
figure 1

ETH-129 mediated mitochondrial Ca2+ uptake is greatly reduced in an ionic medium. panel A, the medium contained a 0.6 osmolar (osM) mixture of 0.3 M KCl and 0.6 M mannitol, 10 mM HEPES (TEA+), pH 7.20, 10 mM KPi, 0.5 mg/ml BSA, 10 μM EGTA (K+), and 0.1 mM antipyrylazo III. Where indicated 1 mM ethanol, 150 μM CaCl2, 3.6 μM ETH-129, 0.3 M KCl, and 0.6 M sorbitol were added. panel B, the conditions were the same as panel A, except 0.3 M choline Cl was present instead of KCl.

After suspending mitochondria in a 0.3 M KCl medium in the presence of Ca2+, ETH-129, and 1 mM ethanol (respiratory substrate), no mitochondrial swelling occurred (Figure 2A). Therefore no non-specific pore opened to allow Ca2+ efflux or KCl influx. If a non-specific pore were to open, an influx of KCl into the matrix space would result in an increase in matrix space volume and a decrease in light scattering of the mitochondrial suspension. Further evidence for the lack of mitochondrial pore opening is that a substantial membrane potential (ΔΨ) was maintained in a KCl medium after ethanol, Ca2+ and ETH-129 addition (Figure 2B). The ΔΨ was monitored by the absorbance of the dye safranine O [24].

Figure 2
figure 2

Ca2+ and ETH-129 don't cause swelling or a large ΔΨ drop when mitochondria are suspended in KCl. panel A, the medium contained 0.3 M KCl, 10 mM HEPES, pH 7.20, 10 mM KPi, and 0.5 mg/ml BSA. 1 mM ethanol, 80 μM CaCl2, and 3.6 μM ETH-129 were present as shown. panel B, the medium contained a 0.6 osM mixture of 0.3 M KCl and 0.6 M mannitol, 10 mM HEPES (TEA+), pH 7.20, 10 mM KPi, 0.5 mg/ml BSA, and 12 μM safranine O. 1 mM ethanol, 80 μM CaCl2, 3.6 μM ETH-129, and 4 μM FCCP were added where indicated. 1 mM ethanol was chosen because this concentration prevents mitochondrial swelling in a 0.3 M KCl medium.

In the KCl medium in the absence of Ca2+ and ETH-129, the ΔΨ was roughly twenty percent smaller than that observed when the mitochondria were suspended in a mannitol medium. Immediately following Ca2+ and ETH-129 addition the ΔΨ in the mannitol medium dropped to a lower value than that observed after the same treatment in the KCl medium. At the same time Ca2+ was taken up more rapidly in the mannitol medium. Therefore a sufficient driving force for mitochondrial Ca2+ uptake by ETH-129 was maintained in the KCl medium although the Ca2+ was not retained in the organelle. The steady state ΔΨ attained in the 0.3 M KCl media after Ca2+ and ETH-129 addition was approximately two-thirds of the value of that in the mannitol medium. These results are consistent with the activation of a Ca2+/H+ exchange when mitochondria are suspended in the ionic media. This activity together with the electrogenic ETH-129-mediated Ca2+ uptake causes proton re-entry into the matrix space to slightly decrease the ΔΨ. After the ΔΨ reached a steady state, it was completely dissipated by addition of the uncoupler FCCP (carbonyl cyanidep-trifluoromethoxyphenylhydrazone) to give the baseline level.

In a previous report a large decrease in the ΔΨ occured after ETH-129 addition to mitochondria suspended in a mannitol medium in the absence of BSA (bovine serum albumin) [25]. Together with ionophore-catalyzed Ca2+ influx, activation of a fatty acid stimulated Ca2+/H+ exchange led to the decreased ΔΨ. Because the ΔΨ was not dissipated to as great of an extent when mitochondria were suspended in a KCl medium, the Ca2+ efflux pathway described here appears to be less active than the one activated by fatty acids [25].

Increasing ionic strength induces release of Mg2+ from mitochondria

Using atomic absorption spectrophotometry, we determined that isolated yeast mitochondria contain between 25 and 35 nmole Mg2+/mg protein. This is fairly consistent with a previously measured value of between 20 and 30 nmole Mg2+/mg protein [26]. When non-energized mitochondria were suspended in a 0.3 M KCl medium, endogenous Mg2+ was lost from the matrix space in a time-dependent manner (Figure 3A). A slightly higher rate of Mg2+ efflux also occurred in the presence of ethanol, ETH-129 and Ca2+ under identical conditions to Figure 1A (data not shown). To study the effects of varied ionic strength on the Mg2+ efflux rate, non-energized mitochondria were suspended in 0.6 osmolar (osM) media containing a varied KCl concentration osmotically balanced with mannitol (Figure 3Ainset.) Increasing the KCl concentration increased the rate of Mg2+ release from the matrix space. When no KCl was present virtually all of the Mg2+ remained in the mitochondria.

Figure 3
figure 3

An ionic medium induces a loss of mitochondrial Mg2+. The medium contained 0.3 M KCl, 10 mM HEPES, pH 7.20, and 0.1 mM decavanadate (Na+). The mitochondria were spun down at the indicated time after suspension and analyzed for Mg2+. inset, the conditions were the same as panel A except the osmolarity of the medium was kept constant by replacing a portion of the KCl with mannitol. The mitochondria were spun down 4 minutes after suspension for analysis. panel B, the medium contained 0.3 M choline Cl, 10 mM HEPES (TEA+), pH 7.20, and 0.1 mM decavanadate (Na+). 4 μM FCCP, 2 mM ADP (Na+), 1 mM ethanol, and 2 mM EDTA (K+) were present where indicated.

Mg2+ was also released from the matrix space when mitochondria were suspended in all other ionic media tested including K+ TES, NaCl, and tetramethylammonium (TMA) Cl (data not shown). Endogenous Mg2+ was released from the matrix space at slightly different rates in the different ionic media in increasing order TMA Cl < choline Cl < KCl < Tris Cl (compare K+ in Figure 3A to choline in Figure 3B). These rates of Mg2+ release corresponded to the rates of spontaneous swelling of yeast mitochondria in these salt media. However, Mg2+ was released even faster than any of these spontaneous rates in a KCl medium in the presence of 1 mM ethanol (data not shown) where no swelling occurred (see Figure 2A). Mg2+ was also released at a substantial rate in K+ TES media where no swelling occurred. Therefore organelle lysis due to matrix space swelling was not responsible for the Mg2+ release. The entry of ions into the matrix space may help displace bound Mg2+ to stimulate the rate of efflux.

When non-energized yeast mitochondria were suspended in a choline Cl medium, the addition of either EDTA (ethylenediaminetetraacetic acid) or ADP slightly increased the Mg2+ efflux rate (Figure 3B). The combined presence of EDTA and ADP increased the rate of Mg2+ efflux more than either compound alone. The addition of ethanol as a respiratory substrate also increased the efflux of Mg2+. Since ethanol opens the yeast mitochondrial unselective channel (YMUC), decavanadate, a YMUC inhibitor [27, 28], was present in the medium during all experiments. Since the efflux of Mg2+ is fairly rapid even in the absence of ADP, EDTA, or ethanol, the addition of these compounds only slightly stimulated the rate of efflux.

EDTA, respiration, ADP, and ATP stimulate loss of Mg2+ from mitochondria

To verify the effects of the addition of EDTA, ethanol, and ADP observed in a choline Cl medium, similar experiments were performed suspending mitochondria in a mannitol medium where only a very small rate of spontaneous Mg2+ release occured. As shown in Figure 4A, adding 2 mM ethanol only led to the release of 20–25 percent of the endogenous matrix space Mg2+. Adding 2 mM EDTA resulted in a quick initial and then subsequently slower release of Mg2+. The addition of EDTA and ethanol together led to a very quick release of almost all of the endogenous mitochondrial Mg2+. The addition of ATP caused a slightly greater rate of Mg2+ release than ADP. The presence of carboxyatractyloside, an inhibitor of the adenine nucleotide translocator, did not alter the rate of Mg2+ efflux when ATP or ADP was added. Therefore ADP and ATP likely bind to a site on the outer surface of the inner mitochondrial membrane to release Mg2+. ADP and ATP were half maximally effective at concentrations of 0.43 mM and 0.22 mM respectively (Figure 4B).

Figure 4
figure 4

ADP, ATP, ethanol, and EDTA stimulate mitochondrial Mg2+ efflux in a mannitol medium. panel A, the medium contained 0.6 M mannitol, 10 mM HEPES (TEA+), pH 7.20 and 0.1 mM decavanadate (Na+). Where indicated 1 mM ethanol, 2 mM ATP (Na+), 2 mM ADP (Na+), and 2 mM EDTA (K+) were present. panel B, the medium was the same as for panel A. The amount of mitochondrial Mg2+ was measured 6 min. after suspension varying the ADP and ATP concentrations as indicated.

As described in Table 1, the nucleotides AMP and cAMP were not very effective at causing efflux of endogenous mitochondrial Mg2+ in a mannitol medium. 0.1 mM EDTA was effective in releasing the majority of Mg2+, while 0.1 mM EGTA (ethylene glycol-bis-(beta-aminoethyl ether)-N, N, N', N'-tetracetic acid) was not effective. EDTA and EGTA are metal ion chelators. These results suggest that an inhibitory metal ion blocks the Mg2+ release pathway at a site on the inner membrane facing the cytoplasm. EDTA likely binds the inhibitory metal ion to stimulate efflux. GTP and ATP-γS were nearly as effective as ATP in activating the Mg2+ efflux pathway. The nucleotide analog 5'-fluorosufonylbenzoyladenosine (FSBA) partially blocked the ADP-induced Mg2+ release suggesting that nucleotides do not activate release solely by chelating an inhibitory metal ion. The addition of 0.1 mM Ca2+ also led to a slight increase in the Mg2+ release rate. Unlike yeast plasma membrane Mg2+ transporters [13], Al3+ did not block mitochondrial Mg2+ efflux (data not shown). La3+ and ruthenium red, which block the Ca2+ uniporter in mammalian mitochondria [29] were also unable to block Mg2+ release in yeast mitochondria.

Table 1 Effectors of yeast mitochondrial Mg2+ release. Yeast mitochondria were suspended in 0.6 M mannitol, 10 mM HEPES (TEA+), pH 7.20, and 0.1 mM decavanadate (Na+). Mitochondrial Mg2+ was measured at 20 minutes. The percentage of Mg2+ released was compared to that released in the presence 2 mM ATP (Na+).

The concentrations of EDTA and EGTA added to the medium were varied to identify the amount of chelator needed to cause Mg2+ release (Figure 5). 100 times more EGTA than EDTA was needed to cause an equal amount of Mg2+ release. Both chelators stimulated efflux to a greater extent than an equal amount of chloride medium indicating that these compounds do not just release Mg2+ by increasing the ionic strength of the medium.

Figure 5
figure 5

EDTA stimulates mitochondrial Mg2+ efflux much more potently than EGTA. The medium contained 0.6 M mannitol, 10 mM HEPES (TEA+), pH 7.20, and 0.1 mM decavanadate (Na+). TEA EDTA, TEA EGTA, and TEA Cl were present as shown. Timepoints were taken at 20 minutes.

Fatty acids do not stimulate mitochondrial Mg2+ efflux

Since fatty acids stimulate Ca2+ transport from yeast mitochondria in a mannitol medium [25], we sought to determine if they also stimulate Mg2+ release. We therefore performed experiments under the exact conditions which have previously been shown to stimulate Ca2+ efflux from yeast mitochondria. As shown in Figure 6, free fatty acids or BSA, which binds free fatty acids, did not alter the rate of Mg2+ efflux from the matrix space even in the presence of Ca2+ and ETH-129 that were present in the past study. The presence of Ca2+ slightly stimulated Mg2+ efflux in a similar manner as is shown in Table 1. BSA or fatty acids also did not significantly alter the rate of mitochondrial Mg2+ release when the mitochondria were suspended in a KCl medium (data not shown). This data suggest that the mitochondrial Mg2+ efflux pathway studied in this report and the fatty acid-induced Ca2+/H+ exchange are distinct activities.

Figure 6
figure 6

Fatty acids do not stimulate Mg2+ efflux from mitochondria. The medium contained 0.6 M mannitol 10 mM HEPES (TEA+), pH 7.20, and 10 mM KPi. Where indicated 1 mM ethanol, 25 μM oleate (Na+), 80 μM CaCl2, 3.6 μM, ETH-129, and 1 mg/ml BSA were present. Mitochondria were spun down and analyzed for Mg2+ by atomic absorption spectrophotometry.

Increasing pH stimulates mitochondrial Mg2+ efflux

Increasing the pH increased the rate of Mg2+ release from the mitochondrial matrix space in a KCl medium (Figure 7). The rate of release at pH 7.8 was nearly twice the rate at pH 6.5. This could be explained by the presence of a site for proton inhibition of Mg2+ release. Protons may inhibit from the matrix space side of the inner membrane because pumping protons during respiration also stimulated Mg2+ efflux.

Figure 7
figure 7

Increasing pH increases mitochondrial Mg2+ efflux. The medium contained 0.3 M KCl, 10 mM HEPES for pH > 6.7 or 10 mM MES for pH < 6.7, the pH indicated, and 0.1 mM decavanadate (Na+). Mitochondria were spun down 4 min. after suspension and analyzed for Mg2+.

Mitochondrial divalent cation influx

To determine if divalent cation influx into non-energized yeast mitochondria could occur, the mitochondria were suspended in a KCl medium in the presence of increasing concentrations of the free divalent cations Ca2+, Co2+, Mg2+, or Mn2+. The total mitochondrial levels of the cations increased substantially (Figure 8). This increase is due to the large divalent cation buffering capacity present in the mitochondrial matrix space. This finding is consistent with previous results using mitochondrial-targeted aequorin indicating that free external and free mitochondrial Ca2+ levels equilibrate [22]. Assuming that this nonspecific divalent cation influx activity also equilibrates free matrix space and external Mg2+ concentrations, the free matrix space Mg2+ concentration would be approximately 1.2 mM for our isolated mitochondria that contained 31.5 nmol Mg2+/mg protein. A free matrix space Mg2+ level of 0.7 mM was measured in isolated yeast mitochondria using a fluorescent indicator [30].

Figure 8
figure 8

Mg2+, Ca2+, Co2+, and Mn2+ influx into yeast mitochondria. The medium contained 0.3 M KCl, 10 mM HEPES, pH 7.2, and 0.1 mM decavanadate (Na+). Either 4 mM MgCl2, MnCl2, CoCl2, or CaCl2 were present as indicated. EGTA (K+) was present at different concentrations to yield the indicated free divalent cation concentrations as described in Methods. The mitochondria were incubated for 5 min. in the indicated free ion concentration, then spun down for 2 minutes in a microcentrifuge, washed once with a cold solution containing 0.6 M mannitol, 10 mM MES (TEA+), pH 6.0, and then spun down again. Ion concentrations were then determined as described in Methods.


The present study describes conditions in which isolated yeast mitochondria are able to release Mg2+ and Ca2+ from the matrix space. The efflux is most active when mitochondria are suspended in a high ionic strength medium. Respiration, ATP, ADP, EDTA, and Ca2+ are also able to stimulate Mg2+ efflux.

The essential role of Mg2+ in mitochondrial metabolism

Genetic studies in yeast have established that Mg2+ is essential for mitochondrial RNA metabolism [17]. Mg2+ also regulates the yeast mitochondrial ATP synthase [18]. The free concentration of Mg2+ in beef heart mitochondria is estimated to be between 0.7 and 0.9 mM [10]. Within the physiological range, Mg2+ can modulate the efficiency of the mammalian 2-oxoglutarate dehydrogenase and mitochondrial ATP synthase to affect the rate of oxidative phosphorylation [31]. Therefore the matrix space Mg2+ level needs to be maintained within a specific range to maintain proper mitochondrial function. However in other species and tissues mitochondrial Mg2+ has been shown to vary over a larger range from 0.2 to 1.5 mM [32, 33].

The role of EDTA and EGTA in stimulating Mg2+ efflux

EDTA and EGTA have very different affinities for activating mitochondrial Mg2+ release. The most likely candidate for a metal ion bound by EDTA and EGTA to activate Mg2+ release is Mg2+ itself because it has a greatly different affinity for each of these chelators. Many other divalent cations such as Ca2+ do not satisfy this criterion. The ability of EDTA to activate mitochondrial Mg2+ efflux from the cytoplasmic side of the inner membrane suggests that when the cytoplasmic concentration of the inhibiting ion is low, mitochondrial Mg2+ equilibrates with the cytoplasmic concentration. However, the significance of mitochondrial Mg2+ efflux functioning to supply Mg2+ to the cytoplasm is questionable, since the vacuole is the major storage organelle for Mg2+ in yeast cells [1].

The presence of a cytoplasmically facing Mg2+ inhibiting the mitochondrial efflux of divalent cations does not appear to be entirely consistent with a mitochondrial influx of Mg2+ (see Figure 8). Binding of Mg2+ to the external inhibitory site may be expected to inhibit Mg2+ influx into the matrix space. However, a salt-based medium may displace inhibitory Mg2+. Yeast mitochondria do not swell when suspended in a 0.3 M Mg2+ acetate medium [25] indicating that an influx of Mg2+ into the matrix space does not occur at this very high concentration. Therefore mitochondrial influx may depend on the specific KCl and Mg2+ concentrations employed. In a previous report Mg2+ was transported into yeast mitochondria in the absence of a membrane potential, however at a highly reduced rate compared to energized mitochondria [30].

Comparison to yeast mitochondrial cation transporters

Fatty acids and BSA do not regulate the mitochondrial Mg2+ and Ca2+ release studied in this report as they did the yeast mitochondrial fatty acid-induced Ca2+/H+ exchanger described previously [25]. Since yeast mitochondria in situ are exposed to an environment containing KCl and do not appear to have a high activity Ca2+ efflux pathway open [22], perhaps some regulatory element inhibiting mitochondrial Ca2+ release is lost during mitochondrial isolation or upon suspension of isolated mitochondria in ionic media.

The yeast mitochondrial Mg2+ release activity may be of a similar nature as the yeast mitochondrial K+/H+ exchange. The similar inhibitors and conditions stimulating mammalian mitochondrial K+ and Mg2+ transport have been noted previously [12, 34]. The activities of both yeast proteins are non-specific in nature [26] (and see Figure 8). Cations also stimulate both activities [26] (and see Figure 6). Further experiments are needed to determine if divalent cations can exchange for many other cations across the inner mitochondrial membrane or just for protons. Both yeast mitochondrial Mg2+ release and K+/H+ exchange appear to be inhibited by divalent cations. The yeast K+/H+ exchange, although only moderately inhibited by Mg2+, is strongly inhibited by Zn2+ [26, 35, 36], whereas the ability of chelators to activate Mg2+ release suggests a role for divalent cations in regulating this activity as well. However there is at least one striking difference in the regulation of yeast mitochondrial Mg2+ release and K+/H+ exchange. The K+/H+ exchange appears to be spontaneously active in isolated yeast mitochondria [26] when they are suspended in a mannitol medium as monitored by the efflux of endogenous K+ (data not shown) while Mg2+ release does not occur spontaneously in a mannitol medium.

In a previous study no loss of free mitochondrial Mg2+ was detected in yeast mitochondria when they were suspended in a 0.135 M KCl medium for thirty minutes [30], while we detect a loss of over 70 percent of the total mitochondrial Mg2+ levels in just ten minutes in 0.3 M KCl media using mitochondria from strain W303-1A. However, in the previous study the DBY747 yeast strain was used and the mitochondria were preloaded in a 10 mM Mg2+ 135 mM KCl medium before Mg2+ release was monitored. It will be important to determine which of these factors is responsible for the different Mg2+ efflux results.

Yeast mitochondrial Mg2+ efflux candidates

Mitochondrial matrix space Mg2+ is important for many aspects of nucleotide metabolism [37, 38]. Two inner mitochondrial membrane transporters, Mrs2p and Lpe10p, are needed for group II intron splicing [16, 39]. MRS2 and LPE10 have slight sequence similarity with the bacterial Mg2+transporter CorA. Assays with a fluorescent Mg2+ indicator dye indicate that Mrs2p is part of an electrophoretic mitochondrial Mg2+ influx pathway inhibited by cobalt(III)hexaammine [30]. Mitochondrial Mg2+ levels changed with the levels of Mrs2p and Lpe10p. Mitochondrial electrophoretic Mg2+ uptake was absent in an MRS2 deletion strain. Mrs2p and Lpe10p are essential for yeast growth on nonfermentable carbon sources [38]. However they cannot substitute for each other suggesting non-redundant functions. It is possible that Mrs2p or Lpe10p is responsible for the mitochondrial Mg2+ release described in this report. However, in the previous experiments Mg2+ was taken up by energized mitochondria in an Mrs2p-dependent manner where we found that Mg2+ was released from energized mitochondria [30]. In the previous study there was also no difference in the uptake of Mg2+ in ionic or non-ionic media, while greatly differing Mg2+ efflux results occurred in this report under these conditions. In addition cobalt(III)hexaammine, an inhibitor of Mrs2p, blocked Mg2+ transport while having no effect on Ca2+ transport [30], while the efflux described here appears not to discriminate between Mg2+ and Ca2+.

The YOL027 and YPR125 genes are multicopy suppressors of the respiratory-deficient phenotype caused by deletion of MRS2 [30] and are therefore candidate mitochondrial Mg2+ transporters. YOL027p has been demonstrated to mediate K+/H+ exchange activity in yeast mitochondria [30]. Because of the similarities in mitochondrial Mg2+ and K+ transport properties mentioned above either YOL027p or YPR125p may also function as a Mg2+/H+ exchanger. Mg2+ inhibited K+/H+ exchange activity in YOL027p-contaning submitochondrial particles. Deletion of YOL027 resulted in slightly altered mitochondrial transport rates of both Mg2+ and Ca2+ [40]. It is possible that the homologous YPR125 gene (40 % sequence identity) could be responsible for the remaining Mg2+ and Ca2+ transport activity because overexpression of YPR125 rescued growth of the YOL027 deletion strain and YPR125p also localized to mitochondria.

When added to yeast cells certain ionophores such as nigericin selectively target mitochondria [41]. Nigericin is an ionophore that possesses K+/H+ exchange activity. A screen for nigericin-resistant mutants revealed that the Mdm31 and Mdm32 genes may be involved in mitochondrial cation homeostasis [42]. Mdm31 and Mdm32 deletion mutants contained a two to three-fold higher level of mitochondrial Mg2+ than the wild type. Since yeast mitochondrial K+/H+ exchange is inhibited by Mg2+ [36, 43], it was hypothesized that Mdm31p and Mdm32p may be proteins that mediate the efflux of mitochondrial Mg2+ [42]. Knocking out their function could have elevated mitochondrial Mg2+ levels to inhibit the endogenous mitochondrial K+/H+ exchanger to rescue the cells from nigericin-mediated death.

Mammalian mitochondrial Mg2+ transport

We have shown that yeast mitochondria contain a Mg2+ release activity, while a similar activity has been demonstrated in mammalian mitochondria [33]. Mammalian mitochondria are hypothesized to take up Mg2+ through a uniport mechanism or by respiration-dependent 'unspecific leak' and release Mg2+ by a Mg2+/H+ exchange [33, 44]. Our current data combined with data from the Schweyen laboratory [30] indicate that a similar system is likely present in yeast. They have characterized Mrs2 as the respiration-dependent mitochondrial Mg2+ uptake system, while the efflux described in this report is likely a Mg2+/H+ exchange. In this regard the respiratory substrate-induced release that occurs in yeast mitochondria in the absence of external Mg2+ resembles the respiration-induced release in rat heart attributed to Mg2+/H+ exchange [33]. The ionic strength-induced release of Mg2+ in yeast mitochondria resembles a Mg2+ release activity in rat liver mitochondria that is stimulated by alkaline pH in the presence of fatty acids [12]. This rat liver mitochondrial Mg2+ release also only occurred in an ionic medium. This suggests that yeast and mammalian mitochondrial Mg2+ transport may occur through similar mechanisms.


By using a Ca2+-indicating dye and measuring mitochondrial Mg2+ levels we have identified Ca2+ and Mg2+ efflux activities stimulated by increased ionic strength in yeast mitochondria. Mitochondrial efflux did not appear to be spontaneously active when isolated mitochondria were suspended in a sugar-containing medium. Nucleotide di- and tri- phosphates, increasing pH, and divalent cation chelators also activated mitochondrial Mg2+ release. The function of this mitochondrial efflux pathway is likely important for regulating the matrix space concentrations of Mg2+, Ca2+, and perhaps other divalent cations.


Yeast growth and mitochondrial isolation

The yeast strain W303-1A was grown aerobically at 30°C in a media containing 2% lactate, 1% yeast extract, 2% peptone, 0.05% dextrose, and 0.01% adenine at pH 5.0. Yeast cells were harvested during logarithmic growth phase (A600 = 1.8–2.2). Mitochondria were isolated from spheroplasts as previously described [21, 45], except 0.6 M sucrose was used in the homogenization buffer instead of 0.6 M mannitol. The isolated yeast mitochondria were suspended in 0.6 M mannitol, 20 mM HEPES (K+), and 0.1 mM EGTA, pH 6.8. Protein concentration was determined using a mini-Biuret method using BSA as a standard. Samples were solubilized in deoxycholate (Na+), which was present at a final concentration of 1% by weight.


For ion concentration determination atomic absorption spectrophotometry was performed using an AA-575 spectrophotometer (Varian). 2 mg of mitochondrial protein was incubated in the indicated media, spun down for 2 minutes in a microcentrifuge and the supernatant poured off. The mitochondrial pellet was solubilized with 0.6 mL 2 N perchlorate overnight. 0.5 mL was removed and diluted to 2 mL with deionized water prior to sample reading. For swelling experiments, solute permeability was monitored by light scattering using an SLM- Aminco DW-2C spectrophotometer in split beam mode at A540 using mitochondria suspended at 1 mg protein/ml. The uptake of Ca2+ by mitochondria was followed using antipyrylazo III as an indicator of the extramitochondrial Ca2+ concentration [46]. Changes in ΔA720-A790 were recorded in the DW-2C operated in the dual wavelength mode and converted to Ca2+ concentration values using a calibration curve that was generated with a standard solution of Ca2+. Estimates of membrane potential were made using safranine O (12 μM) as an indicator. Safranine accumulation was followed spectrophotometrically, as a change in ΔA511-A533. Initial absorbances without mitochondria in the media were always set to identical values. After addition of mitochondria to the medium absorbances in different ionic strength media were shifted up or down compared to runs in the same media. However, the magnitude of the dye response was similar in the different ionic strength media. Therefore, the different runs were comparable after aligning their baseline membrane potentials after FCCP addition.

Free ion concentration determination

Medium free ion concentrations in the presence of chelators were determined using a computer program [47].


  1. Beeler T, Bruce K, Dunn T: Regulation of cellular Mg2+ by Saccharomyces cerevisiae. Biochim Biophys Acta. 1997, 1323: 310-318.

    CAS  PubMed  Article  Google Scholar 

  2. Grubbs RD: Intracellular magnesium and magnesium buffering. Biometals. 2002, 15: 251-259. 10.1023/A:1016026831789.

    CAS  PubMed  Article  Google Scholar 

  3. Romani AM, Maguire ME: Hormonal regulation of Mg2+ transport and homeostasis in eukaryotic cells. Biometals. 2002, 15: 271-283. 10.1023/A:1016082900838.

    CAS  PubMed  Article  Google Scholar 

  4. Schlingmann KP, Gudermann T: A critical role of TRPM channel-kinase for human magnesium transport. J Physiol. 2005, 566: 301-308. 10.1113/jphysiol.2004.080200.

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  5. Li FY, Nikali K, Gregan J, Leibiger I, Leibiger B, Schweyen R, Larsson C, Suomalainen A: Characterization of a novel human putative mitochondrial transporter homologous to the yeast mitochondrial RNA splicing proteins 3 and 4. FEBS Lett. 2001, 494: 79-84. 10.1016/S0014-5793(01)02319-5.

    CAS  PubMed  Article  Google Scholar 

  6. Brierley GP, Bachmann E, Green DE: Active transport of inorganic phosphate and magnesium ions by beef heart mitochondria. Proc Natl Acad Sci U S A. 1962, 48: 1928-1935.

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  7. Brierley G, Murer E, Bachmann E, Green DE: Studies on Ion Transport. Ii. The Accumulation of Inorganic Phosphate and Magnesium Ions by Heart Mitochondria. J Biol Chem. 1963, 238: 3482-3489.

    CAS  PubMed  Google Scholar 

  8. Crompton M, Capano M, Carafoli E: Respiration-dependent efflux of magnesium ions from heart mitochondria. Biochem J. 1976, 154: 735-742.

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  9. Kubota T, Shindo Y, Tokuno K, Komatsu H, Ogawa H, Kudo S, Kitamura Y, Suzuki K, Oka K: Mitochondria are intracellular magnesium stores: investigation by simultaneous fluorescent imagings in PC12 cells. Biochim Biophys Acta. 2005, 1744: 19-28. 10.1016/j.bbamcr.2004.10.013.

    CAS  PubMed  Article  Google Scholar 

  10. Jung DW, Brierley GP: Magnesium transport by mitochondria. J Bioenerg Biomembr. 1994, 26: 527-535. 10.1007/BF00762737.

    CAS  PubMed  Article  Google Scholar 

  11. Romani A, Dowell E, Scarpa A: Cyclic AMP-induced Mg2+ release from rat liver hepatocytes, permeabilized hepatocytes, and isolated mitochondria. J Biol Chem. 1991, 266: 24376-24384.

    CAS  PubMed  Google Scholar 

  12. Schonfeld P, Schuttig R, Wojtczak L: Rapid release of Mg(2+) from liver mitochondria by nonesterified long-chain fatty acids in alkaline media. Arch Biochem Biophys. 2002, 403: 16-24. 10.1016/S0003-9861(02)00206-0.

    PubMed  Article  Google Scholar 

  13. MacDiarmid CW, Gardner RC: Overexpression of the Saccharomyces cerevisiae magnesium transport system confers resistance to aluminum ion. J Biol Chem. 1998, 273: 1727-1732. 10.1074/jbc.273.3.1727.

    CAS  PubMed  Article  Google Scholar 

  14. Maruyama T, Masuda N, Kakinuma Y, Igarashi K: Polyamine-sensitive magnesium transport in Saccharomyces cerevisiae. Biochim Biophys Acta. 1994, 1194: 289-295.

    CAS  PubMed  Article  Google Scholar 

  15. Blackwell KJ, Tobin JM, Avery SV: Manganese uptake and toxicity in magnesium-supplemented and unsupplemented Saccharomyces cerevisiae. Appl Microbiol Biotechnol. 1997, 47: 180-184. 10.1007/s002530050909.

    CAS  PubMed  Article  Google Scholar 

  16. Gregan J, Bui DM, Pillich R, Fink M, Zsurka G, Schweyen RJ: The mitochondrial inner membrane protein Lpe10p, a homologue of Mrs2p, is essential for magnesium homeostasis and group II intron splicing in yeast. Mol Gen Genet. 2001, 264: 773-781. 10.1007/s004380000366.

    CAS  PubMed  Article  Google Scholar 

  17. Gregan J, Kolisek M, Schweyen RJ: Mitochondrial Mg(2+) homeostasis is critical for group II intron splicing in vivo. Genes Dev. 2001, 15: 2229-2237. 10.1101/gad.201301.

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  18. Jenkins WT: Yeast mitochondrial F1-ATPase--effects of metal ions. Arch Biochem Biophys. 1994, 313: 89-95. 10.1006/abbi.1994.1363.

    CAS  PubMed  Article  Google Scholar 

  19. Guerin B, Bunoust O, Rouqueys V, Rigoulet M: ATP-induced unspecific channel in yeast mitochondria. J Biol Chem. 1994, 269: 25406-25410.

    CAS  PubMed  Google Scholar 

  20. Manon S, Guerin M: Evidence for three different electrophoretic pathways in yeast mitochondria: ion specificity and inhibitor sensitivity. J Bioenerg Biomembr. 1993, 25: 671-678.

    CAS  PubMed  Google Scholar 

  21. Jung DW, Bradshaw PC, Pfeiffer DR: Properties of a cyclosporin-insensitive permeability transition pore in yeast mitochondria. J Biol Chem. 1997, 272: 21104-21112. 10.1074/jbc.272.34.21104.

    CAS  PubMed  Article  Google Scholar 

  22. Jung DW, Bradshaw PC, Litsky M, Pfeiffer DR: Ca2+ transport in mitochondria from yeast expressing recombinant aequorin. Anal Biochem. 2004, 324: 258-268. 10.1016/j.ab.2003.10.029.

    CAS  PubMed  Article  Google Scholar 

  23. Wang E, Erdahl WL, Hamidinia SA, Chapman CJ, Taylor RW, Pfeiffer DR: Transport properties of the calcium ionophore ETH-129. Biophys J. 2001, 81: 3275-3284.

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  24. Akerman KE, Wikstrom MK: Safranine as a probe of the mitochondrial membrane potential. FEBS Lett. 1976, 68: 191-197. 10.1016/0014-5793(76)80434-6.

    CAS  PubMed  Article  Google Scholar 

  25. Bradshaw PC, Jung DW, Pfeiffer DR: Free fatty acids activate a vigorous Ca(2+):2H(+) antiport activity in yeast mitochondria. J Biol Chem. 2001, 276: 40502-40509. 10.1074/jbc.M105062200.

    CAS  PubMed  Article  Google Scholar 

  26. Manon S, Guerin M: K+/H+ exchange in yeast mitochondria: sensitivity to inhibitors, solubilization and reconstitution of the activity in proteoliposomes. Biochim Biophys Acta. 1992, 1108: 169-176.

    CAS  PubMed  Article  Google Scholar 

  27. Roucou X, Manon S, Guerin M: Conditions allowing different states of ATP- and GDP-induced permeability in mitochondria from different strains of Saccharomyces cerevisiae. Biochim Biophys Acta. 1997, 1324: 120-132.

    CAS  PubMed  Article  Google Scholar 

  28. Manon S, Roucou X, Guerin M, Rigoulet M, Guerin B: Characterization of the yeast mitochondria unselective channel: a counterpart to the mammalian permeability transition pore?. J Bioenerg Biomembr. 1998, 30: 419-429. 10.1023/A:1020533928491.

    CAS  PubMed  Article  Google Scholar 

  29. Gunter TE, Pfeiffer DR: Mechanisms by which mitochondria transport calcium. Am J Physiol. 1990, 258: C755-86.

    CAS  PubMed  Google Scholar 

  30. Kolisek M, Zsurka G, Samaj J, Weghuber J, Schweyen RJ, Schweigel M: Mrs2p is an essential component of the major electrophoretic Mg2+ influx system in mitochondria. Embo J. 2003, 22: 1235-1244. 10.1093/emboj/cdg122.

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  31. Rodriguez-Zavala JS, Moreno-Sanchez R: Modulation of oxidative phosphorylation by Mg2+ in rat heart mitochondria. J Biol Chem. 1998, 273: 7850-7855. 10.1074/jbc.273.14.7850.

    CAS  PubMed  Article  Google Scholar 

  32. Jung DW, Apel L, Brierley GP: Matrix free Mg2+ changes with metabolic state in isolated heart mitochondria. Biochemistry. 1990, 29: 4121-4128. 10.1021/bi00469a015.

    CAS  PubMed  Article  Google Scholar 

  33. Rutter GA, Osbaldeston NJ, McCormack JG, Denton RM: Measurement of matrix free Mg2+ concentration in rat heart mitochondria by using entrapped fluorescent probes. Biochem J. 1990, 271: 627-634.

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  34. Diwan JJ: Mitochondrial transport of K+ and Mg2+. Biochim Biophys Acta. 1987, 895: 155-165.

    CAS  PubMed  Article  Google Scholar 

  35. Manon S, Guerin M: Investigation of the effects of Zn2+ and Cu2+ on the K+ transport in yeast mitochondria. Evidences for the involvement of a Zn(2+)-binding protein in the K+/H+ exchange. Biochem Mol Biol Int. 1995, 35: 585-593.

    CAS  PubMed  Google Scholar 

  36. Froschauer E, Nowikovsky K, Schweyen RJ: Electroneutral K+/H+ exchange in mitochondrial membrane vesicles involves Yol027/Letm1 proteins. Biochim Biophys Acta. 2005, 1711: 41-48.

    CAS  PubMed  Article  Google Scholar 

  37. Wiesenberger G, Link TA, von Ahsen U, Waldherr M, Schweyen RJ: MRS3 and MRS4, two suppressors of mtRNA splicing defects in yeast, are new members of the mitochondrial carrier family. J Mol Biol. 1991, 217: 23-37. 10.1016/0022-2836(91)90608-9.

    CAS  PubMed  Article  Google Scholar 

  38. Van Dyck E, Jank B, Ragnini A, Schweyen RJ, Duyckaerts C, Sluse F, Foury F: Overexpression of a novel member of the mitochondrial carrier family rescues defects in both DNA and RNA metabolism in yeast mitochondria. Mol Gen Genet. 1995, 246: 426-436. 10.1007/BF00290446.

    CAS  PubMed  Article  Google Scholar 

  39. Bui DM, Gregan J, Jarosch E, Ragnini A, Schweyen RJ: The bacterial magnesium transporter CorA can functionally substitute for its putative homologue Mrs2p in the yeast inner mitochondrial membrane. J Biol Chem. 1999, 274: 20438-20443. 10.1074/jbc.274.29.20438.

    CAS  PubMed  Article  Google Scholar 

  40. Nowikovsky K, Froschauer EM, Zsurka G, Samaj J, Reipert S, Kolisek M, Wiesenberger G, Schweyen RJ: The LETM1/YOL027 gene family encodes a factor of the mitochondrial K+ homeostasis with a potential role in the Wolf-Hirschhorn syndrome. J Biol Chem. 2004, 279: 30307-30315. 10.1074/jbc.M403607200.

    CAS  PubMed  Article  Google Scholar 

  41. Kovac L, Bohmerova E, Butko P: Ionophores and intact cells. I. Valinomycin and nigericin act preferentially on mitochondria and not on the plasma membrane of Saccharomyces cerevisiae. Biochim Biophys Acta. 1982, 721: 341-348. 10.1016/0167-4889(82)90088-X.

    CAS  PubMed  Article  Google Scholar 

  42. Kucejova B, Kucej M, Petrezselyova S, Abelovska L, Tomaska L: A screen for nigericin-resistant yeast mutants revealed genes controlling mitochondrial volume and mitochondrial cation homeostasis. Genetics. 2005, 171: 517-526. 10.1534/genetics.105.046540.

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  43. Welihinda AA, Trumbly RJ, Garlid KD, Beavis AD: On the regulation of Na+/H+ and K+/H+ antiport in yeast mitochondria: evidence for the absence of an Na(+)-selective Na+/H+ antiporter. Biochim Biophys Acta. 1993, 1144: 367-373.

    CAS  PubMed  Article  Google Scholar 

  44. Jung DW, Panzeter E, Baysal K, Brierley GP: On the relationship between matrix free Mg2+ concentration and total Mg2+ in heart mitochondria. Biochim Biophys Acta. 1997, 1320: 310-320.

    CAS  PubMed  Article  Google Scholar 

  45. Daum G, Bohni PC, Schatz G: Import of proteins into mitochondria. Cytochrome b2 and cytochrome c peroxidase are located in the intermembrane space of yeast mitochondria. J Biol Chem. 1982, 257: 13028-13033.

    CAS  PubMed  Google Scholar 

  46. Scarpa A, Brinley FJJ, Dubyak G: Antipyrylazo III, a "middle range" Ca2+ metallochromic indicator. Biochemistry. 1978, 17: 1378-1386. 10.1021/bi00601a004.

    CAS  PubMed  Article  Google Scholar 

  47. Brooks SP, Storey KB: Bound and determined: a computer program for making buffers of defined ion concentrations. Anal Biochem. 1992, 201: 119-126. 10.1016/0003-2697(92)90183-8.

    CAS  PubMed  Article  Google Scholar 

Download references


We would like to thank Dr. Dennis Jung for helpful discussions and Warren Erdahl for technical assistance. This research was supported by a grant from the NIH to DP (GM071396).

Author information

Authors and Affiliations


Corresponding author

Correspondence to Patrick C Bradshaw.

Additional information

Authors' contributions

PB designed and performed the experiments and wrote the manuscript. DP was involved in the conception of the experiments and supervised the work.

Authors’ original submitted files for images

Rights and permissions

Open Access This article is published under license to BioMed Central Ltd. This is an Open Access article is distributed under the terms of the Creative Commons Attribution License ( ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and Permissions

About this article

Cite this article

Bradshaw, P.C., Pfeiffer, D.R. Release of Ca2+ and Mg2+ from yeast mitochondria is stimulated by increased ionic strength. BMC Biochem 7, 4 (2006).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI:


  • Divalent Cation
  • Matrix Space
  • Ionic Medium
  • Yeast Mitochondrion
  • Mammalian Mitochondrion