Temperature dependence of the mitochondrial inner membrane anion channel: the relationship between temperature and inhibition by magnesium.

The mitochondrial inner membrane anion channel (IMAC) carries a wide variety of anions and is postulated to be involved in mitochondrial volume homeostasis in conjunction with the K+/H+ antiporter, thus allowing the respiratory chain proton pumps to drive salt efflux. How it is regulated is uncertain; however, it is inhibited by matrix Mg2+ and matrix protons. Previously determined values for the IC50 suggested that the channel would be closed under physiological conditions. In a previous study (Liu, G., Hinch, B., Davatol-Hag, H., Lu, Y., Powers, M., and Beavis, A. D. (1996) J. Biol. Chem. 271, 19717-19723), it was demonstrated that the channel is highly temperature-dependent, and that a large component of this sensitivity resulted from an effect on the pIC50 for protons. We have now investigated the effect of temperature on the inhibition by Mg2+ and have found that it too is temperature-dependent. When the temperature is raised from 20 degrees C to 45 degrees C, the IC50 increases from 22 to 350 microm at pH 7.4 and from 80 to 1.5 mm at pH 8.4, respectively. The Arrhenius plot for the IC50 is linear with a slope = -80 kJ/mol. The IC50 is also strongly pH-dependent, and at 37 degrees C increases from 90 microm at pH 7.4 to 1230 microm at pH 8.4. In view of the extremely rapid fluxes that IMAC is capable of conducting at 37 degrees C, we conclude that inhibition by matrix Mg2+ and protons is necessary to limit its activity under physiological conditions. We conclude that the primary role of Mg2+ is to ensure IMAC is poised to allow regulation by small changes in pH in the physiological range. This control is mediated by a direct effect of H+ on the activity, in addition to an indirect effect mediated by a change in the Mg2+ IC50. The question that remains is not whether IMAC can be active at physiological concentrations of Mg2+ and H+, but what other factors might increase its sensitivity to changes in mitochondrial volume.


INTRODUCTION
The mitochondrial inner membrane anion channel (IMAC) is a non-selective anion channel that carries a wide variety of anions ranging from small singly charged ions, such as Cland HCO 3 -, to multicharged anions, such as citrate, ferrocyanide and even ATP (reviewed in ref. 1). In view of the variety of anions transported and the fact that under physiological conditions mitochondria generate a membrane potential that is about 180mV negative on the inside, we have hypothesized that IMAC in conjunction with the K + /H + antiporter is involved in mitochondrial volume homeostasis (1). The combined action of these transporters coupled to the proton pumps of the respiratory chain provides a mechanism for respiratory energy to drive salt efflux. In recent years, other roles have been proposed for IMAC. Vanden Hoek, et al (2) have proposed that IMAC may also be involved in the efflux of the superoxide anion from mitochondria during ischemic preconditioning, and O'Rourke's group (3,4) has provided evidence suggesting IMAC is involved in synchronized oscillations of mitochondrial membrane potential in isolated cardiac myocytes.
Although, in energized mitochondria, the anion flux through IMAC is expected to be in the outward direction, IMAC is most easily assayed in de-energized mitochondria by monitoring

Assay of anion transport
Anion transport was assayed by following the swelling that accompanies net salt transport, using the light scattering technique as described in detail elsewhere (9)(10)(11). In brief, reciprocal absorbance, which as a function of mitochondrial volume, is monitored and normalized for mitochondrial protein concentration in the assay to yield a parameter referred to as β, the rate of change of which can be converted to a rate of ion transport in nmol/min per mg mitochondrial protein. In a previous study of the effect of temperature on IMAC (7), we used malonate as the substrate anion; however, in order to investigate inhibition by Mg 2+ we chose to use Cl -, since the association between Mg 2+ and Clis much weaker than between Mg 2+ and malonate and much less Mg 2+ must be added to obtain a given concentration of free Mg 2+ .

Pretreatment of Mitochondria with A23187
In order to deplete the mitochondria of endogenous Mg 2+ , the normal stock suspension was pretreated as described in ref. 7, except the amount of A23187 was increased to 10nmol/mg mitochondrial protein to allow rapid equilibration of Mg 2+ across the inner membrane following transfer to the assay medium. In brief, the mitochondrial stock suspension (50mg/ml) was diluted 1:5 into a medium containing potassium salts of MOPS (25 mM) and EDTA (5 mM) adjusted to pH 7.4 at 25°C and maintained at 0°C. A23187 (10 nmol/mg), nigericin (0.5 nmol/mg) and rotenone (0.5 µg /mg) were added and at least 10 minutes allowed to elapse before transfer of aliquots to the assay medium.

Assay Media for Anion Transport
The potassium chloride medium for assay of IMAC contained the potassium salts of Cl-(55 mM), EGTA (0.1 mM) and MOPS (5 mM for assays at pH 7.4) or TAPS (5 mM for assays at pH 8.4). After addition of the pretreated mitochondria, the final assay medium also contained EDTA (50 µM) sucrose (0.5 mM). All media were 110 mOsm. For experiments in which the temperature was to be varied, the pH of the medium was adjusted at 25°C to a value, calculated on the basis of -0.0095 for ∆pK/°C for MOPS and -0.021 for ∆pK/°C for TAPS, that would yield the desired pH at the assay temperature. Thus, separate assay media were prepared for each temperature to be studied. The value of ∆pK/°C was determined experimentally in the assay media described above. The temperature of the medium in the assay tube was measured during the experiment to ensure that a steady value had been achieved. The free concentration of Mg 2+ was calculated for each temperature and pH value using the program WinMaxC, available from C. Patton, Ph.D., Stanford University, Hopkins Marine Station, CA. The same program was used to calculate the total Mg 2+ to be added to obtain a specific free concentration.

Drugs, Reagents and Mitochondria
Most drugs were obtained from Sigma. Valinomycin, nigericin and rotenone were dissolved in ethanol and A23187 (20 mM stock) was dissolved in dimethylsulfoxide. Rat liver mitochondria were prepared from 30-35 day old Sprague-Dawley rats as described previously (9), except that the first slow-spin pellet was not resuspended, but discarded and the rats were not starved.

Analysis of data
All figures were prepared and nonlinear regression accomplished using the program GraphPad Prism version 3.03 for Windows, GraphPad Software, San Diego, California USA, www.graphpad.com.

Effect of Temperature on Inhibition of IMAC by Mg 2+
In a previous paper (6), we showed that when the mitochondrial inner membrane anion channel (IMAC) is assayed at 25°C, it is inhibited by matrix Mg 2+ and that the IC 50 rises from 38µM at pH 7.4 to 250µM at pH 8.4. The values obtained were not affected by the method used to remove endogenous Mg 2+ nor the time of addition of Mg 2+ to the assay medium. In the present study, because at high temperature the rate of swelling is very rapid, the mitochondria were pretreated with A23187 and EDTA to ensure complete depletion of endogenous Mg 2+ , and Mg 2+ and valinomycin were added to the assay at zero time, i.e. before the mitochondria. The data contained in Fig. 1 show typical traces obtained at 3 temperatures, 25°C for comparison with previous data, 37°C to determine the flux at physiological temperatures and 15°C to illustrate behavior at low temperatures. In each case, there is a short acceleration phase before maximum swelling rates are observed. Mg 2+ inhibited the fluxes at all three temperatures and the IC 50 values obtained from the dose response curves shown in Fig. 2 are 114 µM, 37µM and 17µM at 37°C, 25°C and 15°C respectively. Note, however, that the Hill coefficients tend to increase with the temperature. Consistent with our previous studies (7), the control rates are strongly temperature dependent with values equal to 4.0, 1.2 and 0.23 µmol Cl -/min.mg at 37°C, 25°C and 15°C respectively.

Relationship Between Mg 2+ IC 50 and Temperature
To examine more closely the relationship between the Mg 2+ IC 50 and temperature and how this might be affected by pH, we carried out dose response studies at a series of temperatures ranging from 5°C to 45°C and at two pH values pH7.4 and 8.1. For each temperature, the pH of the assay medium was adjusted separately to ensure that the assay pH did not vary with temperature.
The data contained in Fig. 3 show Arrhenius plots of the control rates of Cltransport in the absence of Mg 2+ at the two pH values. Non-linear relationships are observed, consistent with the data obtained previously using malonate as the substrate for IMAC (7). Note that there is a 275-fold increase in rate when the temperature is raised from 10°C to 45°C at pH 7.4, and at 10°C, there is a 15-fold increase in rate when the pH is raised from 7.4 to 8.1.
The curves are drawn using equation 1 (Eq. 9 of ref. 7), which was derived on the basis of a model in which the open probability of the channel is temperature dependent. Like the transport rate, the Mg 2+ IC 50 is strongly dependent on the temperature increasing exponentially as the temperature is raised (Fig. 4A). In contrast to the Arrhenius plots for the transport rates, the Arrhenius plots of the IC 50 values are essentially linear. Using values interpolated from the curves fitted to the data, the ratio of IC 50 values range from about 4.6 at 45°C to 5.3 at 20°C. The narrow temperature range and scatter in the data do not allow one to determine to what extent the effect of pH results from a change in the slope or intercept.
However, whatever the cause of the shift, there is about a five-fold increase in Mg 2+ IC 50 as the pH is raised from 7.4 to 8.1 over the temperature range examined.

Influence of Temperature on the Relationship between Mg 2+ IC 50 and pH
To examine further the effect of temperature on the relationship between the IC 50 for Mg 2+ and pH, the IC 50 was determined at a series of pH values ranging from 7.35 to pH 8.35 at 25°C and 37°C. The results of two independent experiments contained in Fig. 5 reveal that the IC 50 increases 10-fold and 16-fold at 25°C and 37°C respectively. The curves drawn were fitted to the data using a model, previously shown to describe the relationship between the Mg 2+ IC 50 and pH at 25°C (6), in which Because the intercept on the ordinate is so close to the origin, the data do not allow one to determine whether the effect of temperature is mediated via an effect on K Mg or the pK for the protonation site. The slopes, however, which are determined by the product K Mg .K H and n differ by a factor of 7. Using the constants provided in the legend, values of IC 50 can be calculated.
This latter value is considerably closer to published estimates of the physiological free matrix Mg 2+ concentration (8).

Estimate of IMAC Activity at Physiological Mg 2+ Concentration
It is now well accepted that the physiological concentration of free Mg 2+ in the mitochondrial matrix is close to 0.5 mM (8 To better illustrate the effect of Mg 2+ and temperature on the activity of IMAC, we have plotted the data as percent inhibition of the rate at 45°C in the absence of Mg 2+ (Fig. 6B) Note that at 37°C there is 34% inhibition. These data suggest that Mg 2+ poises the channel to be very sensitive to changes in mitochondrial matrix pH at physiological temperatures.

DISCUSSION
In this paper, we have examined the temperature dependence of the IC 50 for matrix Mg 2+ for inhibition of the mitochondrial inner membrane anion channel. In view of the combined effects of pH and temperature on the Mg 2+ IC 50 and their direct effects on the activity of IMAC, the data presented provide strong evidence that IMAC may have considerable activity under physiological conditions.
The data presented demonstrate that Mg 2+ is an efficacious inhibitor over a temperature range that is associated with a 350-fold change in anion flux. We have observed, however, that the Hill coefficient for the inhibition tends to increase with temperature. It is quite likely that this reflects the involvement of more than one inhibitory Mg 2+ binding site at high temperatures.
A similar but more pronounced phenomenon was observed for inhibition by protons (7), which was explained by the presence of a second inhibitory site with a lower temperature dependence. respectively.
The dependence of the Mg 2+ IC 50 on pH (Fig. 5) is consistent with the model previously proposed (6), although the Hill coefficient appears to increase slightly with temperature so that the increase in IC 50 with pH is greater at 37°C than it is at 25°C. Because both the curves extrapolate to values of IC 50 very close to the origin, it is not possible to determine whether K Mg is affected by temperature. Both curves shown were fitted by setting K Mg = 8 µM and allowing pK H and the Hill coefficients (n) to vary. Using this approach, the effect of raising the temperature from 25°C to 37°C is explained by a decrease in the value of pK H from 6.9 to 6.5, and an increase in n from 0.89 to 1.17. Curves can be fit equally well, however, if it is assumed that pK H = 6.7 at both temperatures and K Mg and n are allowed to vary. In this case, the increase in IC 50 can be explained by an increase in K Mg from 6.3 to 13 µM together with the above- (results not shown). Thus, under physiological conditions, the K + /H + antiporter could also have significant activity. Note that the relatively high activity of IMAC, would ensure that respiration (H + pump) driven anion efflux would elevate matrix pH and further activate both the K + /H + antiporter and IMAC, which are both regulated by matrix protons (1,14). Note also that the very high J max for IMAC relative to the rate of mitochondrial respiration is also consistent with the suggestion by Aon et al (4) that IMAC is responsible for depolarization of mitochondria in myocytes observed in their studies.