Role of the transmembrane potential in the membrane proton leak

Abstract

The molecular mechanism responsible for the regulation of the mitochondrial membrane proton conductance (G) is not clearly understood. This study investigates the role of the transmembrane potential (DeltaPsim) using planar membranes, reconstituted with purified uncoupling proteins (UCP1 and UCP2) and/or unsaturated FA. We show that high DeltaPsim (similar to DeltaPsim in mitochondrial State IV) significantly activates the protonophoric function of UCPs in the presence of FA. The proton conductance increases nonlinearly with DeltaPsim. The application of DeltaPsim up to 220 mV leads to the overriding of the protein inhibition at a constant ATP concentration. Both, the exposure of FA-containing bilayers to high DeltaPsim and the increase of FA membrane concentration bring about the significant exponential Gm increase, implying the contribution of FA in proton leak. Quantitative analysis of the energy barrier for the transport of FA anions in the presence and absence of protein suggests that FA- remain exposed to membrane lipids while crossing the UCP-containing membrane. We believe this study shows that UCPs and FA decrease DeltaPsim more effectively if it is sufficiently high. Thus, the tight regulation of proton conductance and/or FA concentration by DeltaPsim may be key in mitochondrial respiration and metabolism.

Figure 1

Representative volt-ampere characteristics of bilayer membranes from E. coli total lipid extract with and without hUCP2. hUCP2 concentration was ∼7 μg protein/mg of lipid respectively. Buffer solution contained 50 mM K₂SO₄, 25 mM TES, 0.6 mM EGTA at pH 7.35, and T = 37°C. Data were recorded using 10 kHz and 2.9 kHz filters. To obtain a spline line, the acquired 16,368 points were averaged (163 points per line point).

Figure 2

(A) Membrane conductance of bilayer lipid membranes reconstituted with various amounts of arachidonic acid at membrane potentials up to 230 mV. Bilayer lipid membranes were made from E. coli total lipid extract (1.5 mg/mL). Buffer contained 50 mM K₂SO₄, 25 mM TES, 0.6 mM EGTA at pH 7.35. (B) Conductance of bilayer lipid membranes reconstituted with linoleic (gray squares) and arachidonic (black circles) acids at membrane potentials ≤210 mV. The concentration of AA and OA was 15 mol % of lipid. Dashed lines represent the fit of the Eq. 1 to the experimental data.

Figure 3

(A) Dependence of the total membrane conductance on membrane composition and applied voltage. hUCP2 concentration was ∼7 μg/mg of lipid. Buffer solution contained 50 mM K₂SO₄, 25 mM TES, 0.6 mM EGTA at pH 7.35, and T = 37°C. Each point with a vertical bar represents a mean value and a standard deviation from 9–18 experiments, carried out on three different days. (B) UCP1-mediated membrane conductance in the presence of arachidonic (AA, 20:4) and oleic (OA, 18:1) acids at different ΔΨ_m. Dashed lines represent the fit of the Eq. 1 to the experimental data. Inset: Comparison of G calculated at zero potential. Bilayer lipid membranes were made from E. coli total lipid extract. The concentration of AA and OA was 15 mol %, mUCP1 (charge 2) ∼2.6 μg/mg of lipid. Buffer solution contained 50 mM Na₂SO₄, 25 mM TES, 0.6 mM EGTA at pH 7.0.

Figure 4

Dependence of the total membrane conductance on pH in membranes containing UCP and AA (black symbols) or only AA (gray symbols). The concentration of AA was 15 mol %, mUCP1(charge 12) ∼14.8 μg/mg of lipid. Buffer solution contained 50 mM Na₂SO₄, 10 mM MES, 10 mM TRIS, 0.6 mM EGTA, T = 32°C. Each point with a vertical bar represents a mean value and a standard deviation from 8–12 experiments carried out on three different days.

Figure 5

Comparison of the ATP inhibitory effect at low and at high voltages. Buffer solution contained 50 mM Na₂SO₄, 25 mM TES, 0.6 mM EGTA at pH = 7.4, and T = 32°C. Lipid membranes were made from E. coli polar lipid extract (1 mg/mL) and reconstituted with both UCP1 (∼10.4 μg/mg of lipid, charge 9) and 15 mol % arachidonic acid (AA). Each point with a vertical bar represents a mean value and a standard deviation from 8–15 experiments, carried out on five different days.

Figure 6

Comparison of relative conductances (G/G₀) of the membranes containing UCP and AA (black symbols), AA alone (gray symbols) and pure membranes (white symbols). G is the conductance at the respective voltage and G₀ is the conductance calculated at 0 mV. Lipid membranes were made from E. coli polar lipid extract (1.5 mg/mL) and reconstituted with ∼7 μg hUCP2 per mg of lipid. Arachidonic acid was added at a concentration of 15 mol %. Buffer solution contained 50 mM K₂SO₄, 25 mM TES, 0.6 mM EGTA at pH 7.35 and T = 37°C.