Direct mitochondrial import of lactate supports resilient carbohydrate oxidation

Abstract

Lactate is the highest turnover circulating metabolite in mammals. While traditionally viewed as a waste product, lactate is an important energy source for many organs, but first must be oxidized to pyruvate for entry into the tricarboxylic acid cycle (TCA cycle). This reaction is thought to occur in the cytosol, with pyruvate subsequently transported into mitochondria via the mitochondrial pyruvate carrier (MPC). Using ¹³C stable isotope tracing, we demonstrated that lactate is oxidized in the myocardial tissue of mice even when the MPC is genetically deleted. This MPC-independent lactate import and mitochondrial oxidation is dependent upon the monocarboxylate transporter 1 (MCT1/Slc16a1). Mitochondria isolated from the myocardium without MCT1 exhibit a specific defect in mitochondrial lactate, but not pyruvate, metabolism. The import and subsequent mitochondrial oxidation of lactate by mitochondrial lactate dehydrogenase (LDH) acts as an electron shuttle, generating sufficient NADH to support respiration even when the TCA cycle is disrupted. In response to diverse cardiac insults, animals with hearts lacking MCT1 undergo rapid progression to heart failure with reduced ejection fraction. Thus, the mitochondrial import and oxidation of lactate enables carbohydrate entry into the TCA cycle to sustain cardiac energetics and maintain myocardial structure and function under stress conditions.

Fig.1 |. Lactate contributes to the TCA cycle independent of the MPC.

a, Schematic representation of the competition for cytosolic NAD+ between glycolysis and lactate consumption, illustrating that rRedox balanced metabolism is lactate producing. Glyc. denotes glycolysis. b, Schematic of carbon labeling of heart TCA cycle metabolites resulting from infusion of [U-¹³C]glucose or [U-¹³C]lactate. c, Heart ¹³C labeling of lactate, pyruvate and TCA cycle metabolites following [U-¹³C]glucose infusion. d, Heart ¹³C metabolite labeling following [U-¹³C]lactate infusion in Mpc1^iCKO (n=10) and Mpc1^fl/fl (n=8). e, ¹³C enrichment of citrate from primary cultured adult cardiomyocytes (ACMs) cultured with [U-¹³C]glucose with or without 1.6 mM lactate, n=3. f, Survival fraction of ACMs grown in DMEM and HPLM-FA medias with and without lactate, n=3–4 ACM preparations from distinct hearts. *p < 0.05, **p < 0.01, ***p < 0.001 for HPLM-FA with and without lactate by unpaired t-test corrected for multiple comparisons at FDR 0.01. g, Immunoblots of MCT1 and MPC1 from purified mitochondria isolated from Mpc1^fl/fl and Mpc1^iCKO murine hearts. h, Quantification of MPC1 and MCT1 protein levels (Mpc1^fl/fl, n=5, Mpc1^iCKO, n=7). l, Heart ¹³C labeling of glycolytic and TCA cycle metabolites from Mpc1^fl/fl and Mpc1^iCKO mice treated with either AZD3965 (AZD) or vehicle (Veh) and infused with [U-¹³C]glucose (Mpc1^fl/fl Veh n=4, Mpc1^iCKO Veh n=3, Mpc1^fl/fl AZD n=8, Mpc1^iCKO n=5). Glc: glucose; Lac: lactate; Pyr: pyruvate; Succ: succinate; Mal: Malate. All data represent mean ± SEM. Significance determined by multiple comparisons corrected (Holm-Sídák method) unpaired t tests (c and d), one-way ANOVA with Dunnett’s multiple comparison test (f, h, and k), and two-way ANOVA corrected for multiple comparisons (l). ns, not significant (p>0.05).

Fig. 2 |. Lactate oxidation requires MCT1.

a, b, Representative images of ACMs from Mct1^fl/fl and Mct1^iCKO mice labeled with MitoTracker Red CMXRos and co-stained for endogenous MCT1 (green). MCT1 and MitoTracker Red pixel intensity plots for the yellow line are graphed on the right of each panel. Scale bars represent 10 μM (full image) and 6.7 μM (magnified). c, Representative images of Mct1^fl/fl ACMs stained for endogenous SLC25A6 (red) and MCT1 (green) and d, Endogenous TOMM20 (red) and MCT1 (green). Corresponding pixel intensity plots for the yellow line shown in the magnification are graphed on the right of each panel. Scale bars represent 10 μM (full image) and 6.7 μM (magnified). e, Immunoblot of Proteinase K protection assay conducted on mitochondria isolated from human cardiac tissue. f, Coomassie stained SDS-Page gel (left) of mitochondrial preparations isolated from human cardiac tissue immunoprecipitated for MCT1 or IgG and corresponding anti-MCT1 immunoblot (right). g, ¹³C enrichment in lactate, alanine, pyruvate or citrate from Mct1^fl/fl or Mct1^iCKO ACMs cultured with [U-¹³C]lactate. n=3 independent ACM preparations from unique hearts. h, Survival fraction of Mct1^fl/fl or Mct1^iCKO ACMs grown in DMEM or HPLM-FA medias supplemented with lactate. i, Heart ¹³C labeling of glucose, pyruvate and TCA cycle metabolites from [U-¹³C]glucose infusions in Mct1^fl/fl (n=4), or Mct1^iCKO (n=6) mice. j, Heart ¹³C labeling of lactate and TCA cycle metabolites from [U-¹³C]lactate infusions in Mct1^fl/fl (n=7) or Mct1^iCKO (n=8) mice. Glc: glucose; Lac: lactate; Pyr: pyruvate; Succ: succinate; Mal: Malate. All data represent mean ± SEM. Significance determined by two-way ANOVA (g), and by multiple comparisons corrected (Holm-Sídák method) unpaired t tests (I,j). ns, not significant (p>0.05).

Fig. 3 |. Mitochondrial MCT1 is necessary for respiration on lactate.

a, Oxygen consumption rates (JO₂) of isolated cardiac mitochondria incubated with pyruvate from the indicated genotypes (Mct1^fl/fl n=9, Mct1^iCKO n=10, Mpc1^iCKO n=9). b, Citrate M+2 labeling fraction from isolated cardiac mitochondria incubated with [U-¹³C]pyruvate (n=3 individual hearts). c, Oxygen consumption rates of isolated cardiac mitochondria incubated with lactate (WT n=10, Mct1^iCKO n=10, Mpc1^iCKO n=10). d, Citrate M+2 labeling fraction from isolated cardiac mitochondria incubated with [U-¹³C]lactate (n=3 individual hearts). e, Oxygen consumption rates of isolated cardiac mitochondria incubated with lactate treated with Veh (DMSO) n=6, MCT1i (7ACC2) n=6, LDHi (GSK 2837808A) n=6. f, ¹³C labeling fractions of lactate (M+3), pyruvate (M+3) and citrate (M+2) from human donor cardiac mitochondria incubated with [U-¹³C]lactate (n=3 human hearts). g, Schematic illustrating the transfer of electrons from lactate (traced by ²H hydride transfer) into the mitochondrial NADH pool for respiration. h, Ratio of NADH (M+1) over NAD⁺ from ACMs incubated with [2-²H]lactate (n=3 for Mct1^fl/fl and Mct1^iCKO). i, Normalized NADH (M+1) ion intensities from isolated mitochondria incubated with [2-²H]lactate (n=3 for Mct1^fl/fl and Mct1^iCKO). j, k, Oxygen consumption rates of isolated cardiac mitochondria incubated with pyruvate or lactate and treated with DMSO (Veh), or CPI-613 (n=6 hearts for pyruvate, n=8 hearts for lactate). l, Normalized NADH ion intensities from isolated mitochondria treated with DMSO (Veh), or CPI-613, n=3. m, Citrate M+2 labeling fraction from isolated cardiac mitochondria incubated with [U-¹³C]lactate and treated with DMSO (Veh) or CPI-613, n=3. All data represent mean ± SEM. Significance determined by one-way ANOVA with Dunnett’s multiple comparison test (a-e), two-way ANOVA (f) and by unpaired two-tailed t-tests (h-m). ns, not significant (p>0.05).

Fig. 4 |. Loss of MCT1 impairs cardiac function upon injury.

a, Heart weight to body weight ratio of mice treated with angiotensin II and phenylephrine (Ang/PE) by osmotic minipump for 42 days (Mct1^fl/fl saline n=7, Mct1^fl/fl Ang/PE n=3, Mct1^iCKO saline n=4, Mct1^iCKO Ang/PE n=6). b, Quantification of Mpc1 transcript levels from Ang/PE treated hearts (Mct1^fl/fl saline n=5, Mct1^fl/fl Ang/PE n= 3, Mct1^iCKO Ang/PE n=3). c, Left ventricular ejection fraction (LVEF) of same Ang/PE mice (Mct1^fl/fl saline n=7, Mct1^fl/fl Ang/PE n=8, Mct1^iCKO saline n=4, Mct1^iCKO Ang/PE n=10). Statistical significance was determined by a mixed-effects model (repeated measures ANOVA) with the Geisser-Greenhouse correction was used. If significant, then a Tukey’s multiple comparison test was applied with individual variances computed for each comparison. *=p<0.05, **=p<0.01 comparing Mct1^fl/fl Ang/PE vs. Mct1^iCKO Ang/PE. #=p<0.05, ##=p<0.01, comparing Mct1^iCKO Ang/PE vs. Mct1^iCKO saline. d, LVEF of mice subjected to trans-aortic constriction (TAC) and monitored for 6 weeks (Mct1^fl/fl sham n=5, Mct1^fl/fl TAC n=9, Mct1^iCKO sham n=4, Mct1^iCKO TAC n=10). Statistical significance was determined by a mixed-effects model as in c. *=p<0.05, **=p<0.01 comparing Mct1^fl/fl TAC vs. Mct1^iCKO TAC. #=p<0.05, ##=p<0.01, ###=p<0.001, comparing Mct1^iCKO TAC vs. Mct1^iCKO sham. +=p<0.05, comparing Mct1^fl/fl sham vs. Mct1^iCKO sham. e, Venn diagram showing intersection of differentially expressed genes between hearts from Mct1^iCKO and Mct1^fl/fl Ang/PE and vehicle treated animals (Mct1^fl/fl saline n=6, Mct1^fl/fl Ang/PE n=3, Mct1^iCKO saline n=4, Mct1^iCKO Ang/PE n=5). f, Volcano plot showing differential abundance of polar metabolites in saline treated Mct1^iCKO and Mct1^fl/fl hearts (Mct1^fl/fl saline n=7, Mct1^iCKO saline n=4). g, Volcano plot showing differentially abundant polar metabolites in Ang/PE treated Mct1^iCKO and Mct1^fl/fl hearts (Mct1^fl/fl Ang/PE n=3, Mct1^iCKO Ang/PE n=5). h, M+2 citrate labeling fraction from ACMs cultured with [U-¹³C]lactate and treated with Ang/PE (n=3). i, Normalized NADH (M+1) ion intensities from ACMs cultured with [2-²H]lactate, n=3. j, Quantified MCT1 immunoblot bands from mitochondria isolated from human cardiac tissue (donor n=8, HF n=6). k, Evans Blue staining of heart sections showing myocardial salvage and necrosis following in vivo I/R injury and quantification of necrotic tissue within the area at risk (n=5). l, LVEF of mice subjected I/R injury and their recovery over 9 weeks (Mct1^fl/fl n=11, Mct1^iCKO n=9). Data are plotted as mean ± SEM. Significance determined by unpaired two-tailed t-test (j,k) with multiple comparisons correction (l) or one-way ANOVA with Dunnett’s multiple comparison test (a,b, and h-i). ns, not significant (p>0.05).