Human mitochondrial pyruvate carrier 2 as an autonomous membrane transporter

Abstract

The active transport of glycolytic pyruvate across the inner mitochondrial membrane is thought to involve two mitochondrial pyruvate carrier subunits, MPC1 and MPC2, assembled as a 150 kDa heterotypic oligomer. Here, the recombinant production of human MPC through a co-expression strategy is first described; however, substantial complex formation was not observed, and predominantly individual subunits were purified. In contrast to MPC1, which co-purifies with a host chaperone, we demonstrated that MPC2 homo-oligomers promote efficient pyruvate transport into proteoliposomes. The derived functional requirements and kinetic features of MPC2 resemble those previously demonstrated for MPC in the literature. Distinctly, chemical inhibition of transport is observed only for a thiazolidinedione derivative. The autonomous transport role for MPC2 is validated in cells when the ectopic expression of human MPC2 in yeast lacking endogenous MPC stimulated growth and increased oxygen consumption. Multiple oligomeric species of MPC2 across mitochondrial isolates, purified protein and artificial lipid bilayers suggest functional high-order complexes. Significant changes in the secondary structure content of MPC2, as probed by synchrotron radiation circular dichroism, further supports the interaction between the protein and ligands. Our results provide the initial framework for the independent role of MPC2 in homeostasis and diseases related to dysregulated pyruvate metabolism.

Figure 1

Recombinant expression and purification of human MPC. (A) A schematic of the initial human MPC protein constructs expressed in yeast (co-expression set 1). Left panel: Proper localization to the mitochondria was confirmed via confocal microscopy. DIC: Differential interference contrast. Middle panel: Electrophoretic analysis of MPC1-8xHis samples purified using Co-IMAC and GF. *N-terminally truncated portion of 60S RPL28. The inset displays the silver staining and in-gel fluorescence detection of trace amounts of MPC2-GFP, which co-purified with MPC1. Right panel: The corresponding GF trace showed a predominant protein-detergent monodisperse peak at approximately 64 kDa. (B) Diagram of the alternative MPC constructs (co-expression set 2) and corresponding confocal microscopy (left panel). Middle panel: Electrophoretic (Tricine-SDS-PAGE) analysis of representative chromatography steps. PP: PreScission Protease. Right panel: GF peak and silver staining analysis revealed that pure monodisperse MPC2 (associated with DDM) was obtained at an equivalent molecular weight of 70 kDa. In all GF profiles above, v_o indicates void volume, and v_t indicates the total liquid volume of the GF column. The corresponding elution volumes for calibration standards are shown in red. Full-length gels from which silver-stained or fluorescent lanes were cropped are included in Supp. Figure 1.

Figure 2

In vitro MPC activity and its dependence on time, electrochemical gradient and chemical inhibition. (A) Quantification of intravesicular and extravesicular pyruvate, as detected based on ¹⁴C radiolabeled assay (left panel) and enzymatic assay (right panel), respectively, in liposomes (L) and proteoliposomes (PL) reconstituted with MPC1 and MPC2 and a control condition free of lipid vesicles (Ctrl (-L/PL)). The inset indicates the pH gradient across the outer and inner vesicle environments. An asterisk (*) indicates that MPC1 co-purified with yeast RPL28. (B) Quantification of intravesicular and extravesicular pyruvate, as detected based on ¹⁴C radiolabeled assay (left panel) and enzymatic assay (right panel), respectively in the MPC2-proteoliposome as a function of different incubation times with a ΔpH of 1.5 units. The inset indicates the pH gradient across the outer and inner vesicle environments. Half-maximum times were obtained by the fitting of hyperbolic saturation curves (solid blue lines, R² = 0.99 in both graphs). (C) Quantification of intravesicular pyruvate as detected based on ¹⁴C radiolabeled assay in the MPC2-proteoliposome as a function of different pyruvate concentrations (0.125 mM - 3 mM) with a ΔpH of 1.5 units. The inset indicates the pH gradient across the outer and inner vesicle environments. The Km was obtained by fitting a hyperbolic saturation curve (solid green line, R² = 0.97). (D) Quantification of intravesicular and extravesicular pyruvate, as detected based on ¹⁴C radiolabeled assay (left panel) and enzymatic assay (right panel), respectively in liposomes (L) and MPC2-proteoliposome (PL) after 30 min of incubation for ΔpH = 1.5 and ΔpH = 0. The inset indicates the pH gradient across the outer and inner vesicle environments. (E) Quantification of intravesicular and extravesicular pyruvate, as detected based on ¹⁴C radiolabeled assay (left panel) and enzymatic assay (right panel), respectively in liposomes (L) and MPC2-proteoliposome (PL) in the presence of DMSO (vehicle control) and the well-established MPC inhibitors UK-5099 and rosiglitazone (Rgz). The inset indicates the pH gradient across the outer and inner vesicle environments. For all experiments, results are reported as mean ± standard deviations from triplicates to sextuplicates. Statistical significances were assessed by Welch’s unpaired t test, where ns = non-significant, *p < 0.05 and **p < 0.001.

Figure 3

Autonomous transport function of MPC2 in vivo. (A) Differential growth rates of 3Δ cells or JRY472 bearing an empty plasmid (green circles) or expressing either wild-type MPC2 (blues squares) or MPC2.C54S (red triangles) in the absence and presence of external leucine and valine. Lines indicate the mean ± standard error for three independent experiments. (B) Differential respiration rates of 3Δ cells or JRY472 (green circles) expressing either wild-type MPC2 (blues squares) or MPC2.C54S when exposed to glucose (filled geometric shapes) and ethanol (empty geometric shapes). Lines indicate the mean ± standard error for two independent experiments. (C) Quantification of imported pyruvate, as detected based on ¹⁴C radiolabeled assay in isolated mitochondria from 3Δ yeast cells and 3Δ + MPC2 yeast cells in the presence of DMSO (control) and the inhibitors rosiglitazone (Rgz) and UK-5099. Results are reported as mean ± standard deviations from triplicates. In all cases, statistical significances were assessed by Welch’s unpaired t test, where ns = non-significant, *p < 0.05, **p < 0.001 and ***p < 0.001.

Figure 4

Oligomeric state, secondary structure composition and orientation of MPC2. (A) Chemical cross-linking suggested that MPC2 assembled into higher oligomers in a lipid environment. Left panel: Cross-linking of human MPC2 from isolated yeast mitochondrial extracts by 0.2 mM DSS. Middle panel: Cross-linking, using increasing DSS concentrations (0.2–4 mM), of purified MPC2-detergent (DDM) complex. Right panel: Cross-linking, using increasing DSS concentrations (0.2–4 mM), of purified MPC2 reconstituted in asolectin-derived lipid vesicles. (B) Secondary structure analysis of MPC2 in presence and absence of ligands, as probed by Circular Dichroism. Far-UV CD spectra, as well as calculated percentage of secondary structure content in the form of pie-chart, for MPC2 alone (Ctrl; top left), in the presence of 25 µM pyruvate (+Pyr; top right), in the absence of TCEP (−TCEP; bottom left), and in the presence of 50 µM rosiglitazone (+Rgz; bottom right). (C) Right panel: A schematic representation indicating the possible MPC2-GFP-10xHis orientations in the artificial lipid bilayer of the proteoliposomes, as well as the two possible outcomes upon treatment with PreScission protease (PP): ➊ particles with the C-terminal fusion facing the interior of vesicles are protected from digestion by PP and ➋ cleavage of GFP-10xHis tag when the C-terminal fusion is oriented towards the outside of the lipid membrane. Middle panel: The corresponding samples were quantified in a plate reader for GFP fluorescence, and the relative populations are indicated as percentages. Left panel: Qualitative electrophoretic analysis of the proteoliposomes (PL) and supernatant (SN) samples, before and after treatment with PP.