Matrix volume measurements challenge the existence of diazoxide/glibencamide-sensitive KATP channels in rat mitochondria

Abstract

A mitochondrial sulphonylurea-sensitive, ATP-sensitive K+ channel (mitoKATP) that is selectively inhibited by 5-hydroxydecanoate (5-HD) and activated by diazoxide has been implicated in ischaemic preconditioning. Here we re-evaluate the evidence for the existence of this mitoKATP by measuring changes in light scattering (A520) in parallel with direct determination of mitochondrial matrix volumes using 3H2O and [14C]sucrose. Incubation of rat liver and heart mitochondria in KCl medium containing Mg2+ and inorganic phosphate caused a decrease in light scattering over 5 min, which was accompanied by a small (15-30 %) increase in matrix volume. The presence of ATP or ADP in the buffer from the start greatly inhibited the decline in A520, whilst addition after a period of incubation (1-5 min) induced a rapid increase in A520, especially in heart mitochondria. Neither response was accompanied by a change in matrix volume, as measured isotopically. However, the effects of ATP and ADP on A520 were abolished by carboxyatractyloside and bongkrekic acid, inhibitors of the adenine nucleotide translocase (ANT) that lock the transporter in two discrete conformations and cause distinct changes in A520 in their own right. These data suggest that rather than matrix volume changes, the effects of ATP and ADP on A520 reflect changes in mitochondrial shape induced by conformational changes in the ANT. Furthermore, we were unable to demonstrate either a decrease in A520 or increase in matrix volume with a range of ATP-sensitive K+ channel openers such as diazoxide. Nor did glibencamide or 5-HD cause any reduction of matrix volume, whereas the K+ ionophore valinomycin (0.2 nM), produced a 10-20 % increase in matrix volume that was readily detectable by both techniques. Our data argue against the existence of a sulphonylurea-inhibitable mitoKATP channel.

Figure 1. The effects of ATP and ADP on the light scattering response and measured matrix volume of heart (A) and liver (B) mitochondria

Heart mitochondria were added through an injection port into 3.5 ml standard KCl buffer contained within the sample cuvette of the spectrophotometer and incubated with continuous stirring at 25 °C under a gas phase of 100 % O₂, as outlined in Methods. This enabled changes in light scattering to be recorded immediately following complete mixing of the mitochondrial suspension with the buffer (within 5 s of addition of the mitochondria). Where indicated by the labelling to the right of the trace, 0.2 mm ATP or ADP were present in the medium from the start. After 5 min, ³H₂O and [¹⁴C]sucrose were added to the cuvette and matrix volumes determined as described under Methods. For determination of matrix volumes at time zero, mitochondria were added to ice-cold buffer containing isotopes, and volumes were determined immediately. Data from three separate experiments on different batches of mitochondria are summarised in the accompanying bar graphs. Error bars represent s.e.m. and the numbers above each bar the mean (±s.e.m.) volume after incubation expressed as a percentage of the time zero value. Cont = control.

Figure 2. The effects of adenine nucleotide translocase ligands on the changes in light scattering of heart mitochondria induced by ATP and ADP

Heart mitochondria were added through an injection port into incubation buffer contained within the sample cuvette of the spectrophotometer as described in Fig. 1, and additions of ATP (0.2 mm), ADP (0.2 mm), bongkrekic acid (BKA, 10 μm) or carboxyatractyloside (CAT, 10 μm) made as indicated by the arrows. In some incubations (indicated by the label to the left of the trace) these effectors were present in the buffer before addition of mitochondria. A and B, representative traces from two separate experiments with different mitochondrial preparations. All traces shown are typical of at least three separate experiments.

Figure 3. Changes in light scattering of heart mitochondria induced by ATP and CAT are not accompanied by changes in isotopically measured matrix volume

Heart mitochondria were incubated under the same conditions as described for Figs 1 and 2, but using both sample and reference cuvettes of the spectrophotometer as described under Methods. After 1 min incubation in the cuvette the indicated reagent was added to the mitochondrial suspension in the sample cuvette only, and after a further 2 min ³H₂O and [¹⁴C]sucrose were added to both cuvettes, and matrix volumes were determined as described under Methods. In the bar graph, the changes in matrix volume induced by the reagent added are expressed as the percentage of the volumes in the absence of reagent (reference cuvette). Data are presented as the means ±s.e.m. (error bars) of the number of experiments shown, each performed with a separate mitochondrial preparation. When CAT (10 μm) was added, 0.2 mm ATP was present in both the sample and reference cuvettes from the start of the incubation. The mean control matrix volume (reference cuvette) for the total of six mitochondrial preparations represented was 1.26 ± 0.10 μl (mg protein)⁻¹. Glib, glibenclamide; Val, valinomycin.

Figure 4. Effects of phosphate and Mg²⁺ ions on the increase in light scattering of heart mitochondria induced by ATP

A, experiments were performed as in Fig. 3, with mitochondria present in both sample and reference cuvettes, but the medium was varied as indicated. Where shown, ATP (0.2 mm) was added to the sample cuvette only. The control medium lacked the 2.5 mm MgCl₂ (Mg²⁺) and 2.5 mm potassium phosphate (P_i) usually present, and these reagents were added back as indicated. In addition, a parallel experiment was performed using the buffer described by Garlid and colleagues (Kowaltowski et al. 2001). B, mitochondria were present only in the sample cuvette as for Figs 1 and 2. Incubation was performed in the absence and presence of 0.2 mm ATP, 2.5 mm MgCl₂ and 2.5 mm phosphate, as shown by the label to the left of the trace, and further additions of CAT (10 μm) or 0.2 mm ATP were made as indicated by the arrows.

Figure 5. ATP-sensitive K⁺ (K_ATP) channel openers and blockers are without effect on heart or liver mitochondrial matrix volume

The experimental protocol was the same as described in Fig. 3, but the buffer used was supplemented with 10 mm KHCO₃ and 1 mm ATP, and the gas phase 95 % O₂:5 % CO₂ to mimic physiological conditions, as explained under Methods. Where noted, to the left of the trace of light scattering, the medium was supplemented with 50 μm glibencamide (Glib), 5-hydroxydecanoate (5-HD) or diazoxide (Diaz). Where indicated, additions were made to the sample cuvette of 0.2 nm valinomycin (Val) or 50 μm glibencamide (Glib), 5-HD, diazoxide (Diaz), pinacidil (Pinac), cromokalin (Cromo) and the nicorandil analogue N-[2-(acetoxy)ethyl]-3-pyridinecarboxamide (Nico). At the end of each run ³H₂O and [¹⁴C]sucrose were added to both cuvettes and matrix volumes were determined as described under Methods. In the bar graph beneath, the change in matrix volume induced by the reagent added is expressed as the percentage of the volume in the absence of reagent (reference cuvette), presented as the means ±s.e.m. (error bars) of the number of experiments shown, each performed with a separate mitochondrial preparation. Data are given for both heart (black bars) and liver (cross-hatched bars) mitochondria. Traces of light scattering are only shown for heart mitochondria, but essentially identical results were found for liver mitochondria. The effects of valinomycin were statistically significant by Student's t test (*P < 0.02; **P < 0.01).