- Top of page
We examined the controversial notion of whether lactate is directly oxidized by subsarcolemmal (SS) and intermyofibrillar (IMF) mitochondria obtained from red and white rat skeletal muscle. Respiratory control ratios were normal in SS and IMF mitochondria. At all concentrations (0.18–10 mm), and in all mitochondria, pyruvate oxidation greatly exceeded lactate oxidation, by 31- to 186-fold. Pyruvate and lactate oxidation were inhibited by α-cyano-4-hydroxycinnamate, while lactate oxidation was inhibited by oxamate. Excess pyruvate (10 mm) inhibited the oxidation of palmitate (1.8 mm) as well as lactate (1.8 mm). In contrast, excess lactate (10 mm) failed to inhibit the oxidation of either palmitate (1.8 mm) or pyruvate (1.8 mm). The cell-permeant adenosine analogue, AICAR, increased pyruvate oxidation; in contrast, lactate oxidation was not altered. The monocarboxylate transporters MCT1 and 4 were present on SS mitochondria, but not on IMF mitochondria, whereas, MCT2, a high-affinity pyruvate transporter, was present in both SS and IMF mitochondria. The lactate dehydrogenase (LDH) activity associated with SS and IMF mitochondria was 200- to 240-fold lower than in whole muscle. Addition of LDH increased the rate of lactate oxidation, but not pyruvate oxidation, in a dose-dependent manner, such that lactate oxidation approached the rates of pyruvate oxidation. Collectively, these studies indicate that direct mitochondrial oxidation of lactate (i.e. an intracellular lactate shuttle) does not occur within the matrix in either IMF or SS mitochondria obtained from red or white rat skeletal muscle, because of the very limited quantity of LDH within mitochondria.
At rest l-lactate is present in blood, skeletal muscle and other tissues. During prolonged exercise, when lactate is produced by skeletal muscle, a considerable portion of this substrate is transported to the liver and converted to glucose via hepatic gluconeogenesis. In addition, lactate is also oxidized by heart and oxidative types of skeletal muscle (Baldwin et al. 1978). Uptake of lactate into skeletal muscle and other tissues is facilitated by a family of monocarboxylate transporters (MCTs), of which 14 have been identified (Halestrap & Meredith, 2004). Distinct kinetic properties for lactate and pyruvate have been shown for MCT1, 2 and 4 (Broer et al. 1998, 1999; Dimmer et al. 2000), all of which are expressed in skeletal muscle (Bonen et al. 2000b, 2006; Benton et al. 2004). MCT4 is highly expressed in glycolytic muscles (Bonen et al. 1979, 2000a,c), while MCT1 content is greater in oxidative muscles (Bonen et al. 1979, 2000a,c) and is correlated with rates of lactate uptake into this tissue (McCullagh et al. 1996, 1997). MCT2 has a great affinity for pyruvate (Lin et al. 1998; Broer et al. 1999) and is associated with both subsarcolemmal (SS) and intramyofibrillar (IMF) mitochondria (Benton et al. 2004). In contrast, MCT1 and 4 are associated only with SS mitochondria (Benton et al. 2004). Since it appears that SS and IMF mitochondria have different capacities for substrate oxidation (Krieger et al. 1980; Cogswell et al. 1993; Chilibeck et al. 2002; Koves et al. 2005), the MCT differences in SS and IMF mitochondria (Benton et al. 2004) suggest that these mitochondria may have different capacities for the oxidation of pyruvate.
In recent years there has been a suggestion that direct oxidation of lactate to pyruvate can also occur within mitochondria (Brooks et al. 1999b), rather than in the cytosol, as has been accepted for many years. This process has been termed the intracellular lactate shuttle, in which it is postulated that lactate is transported into mitochondria via MCT1, after which LDH within mitochondria converts lactate to pyruvate to allow its oxidation (Brooks et al. 1999b). The notion that lactate might be converted to pyruvate in mitochondria is not new, as more than 30 years ago it was reported that LDH is located near or perhaps within mitochondria (Baba & Sharma, 1971). Recently, with more modern techniques, such proximity of LDH near mitochondria has been reconfirmed (Brooks et al. 1999b; Hashimoto et al. 2006). However, while LDH may be detectable near and/or in mitochondria, there is considerable consensus that in skeletal muscle and heart, LDH content and/or activity are only 0.5–2% of that found in the cytosol (Kline et al. 1986; Brandt et al. 1987; Rasmussen et al. 2002; Sahlin et al. 2002). This would suggest that lactate conversion to pyruvate within mitochondria is probably negligible. Nevertheless, evidence has been presented that rates of pyruvate and lactate oxidation are identical in isolated skeletal muscle mitochondria (Brooks et al. 1999b). However, all other experimental studies to date have been unable to replicate these findings (Popinigis et al. 1991; Rasmussen et al. 2002; Sahlin et al. 2002; Willis et al. 2003; Ponsot et al. 2005). In addition, the concept of direct oxidation of lactate within mitochondria has been challenged on theoretical grounds (Sahlin et al. 2002).
Despite theoretical concerns and the considerable lack of experimental support for direct oxidation of lactate within mitochondria, recent studies have shown an apparent co-localization of proteins (MCT1, CD147, LDH) that may be involved in an intracellular lactate shuttle in L6 muscle cells (Hashimoto et al. 2006), while in rodent skeletal muscle MCT1 and its chaperone CD147 may (Hashimoto et al. 2005) co-localize at or near mitochondria. Yet previously, this group had observed co-localization of MCT1 at or near mitochondria, but without its chaperone CD147 (Butz et al. 2004). Nevertheless, despite some discordant observations, these data have been interpreted as providing evidence for an intracellular lactate shuttle (Butz et al. 2004; Hashimoto et al. 2005, 2006). However, these descriptive studies have not been supported by evidence of direct oxidation of lactate by mitochondria. Indeed, the lactate shuttle hypothesis remains controversial.
We examined (a) whether the controversy concerning direct oxidation of lactate within mitochondria was attributable to a difference in the metabolic capacities of SS and IMF mitochondria obtained from either highly oxidative (red) or highly glycolytic (white) rat skeletal muscles. In addition (b) we also compared the rates of pyruvate and lactate oxidation in SS and IMF mitochondria obtained from red and white muscle. If lactate and pyruvate are equivalently metabolized by skeletal muscle mitochondria, as has been proposed elsewhere (Brooks et al. 1999b), then the following observations would be expected: (i) at given concentrations of lactate and pyruvate, CO2 production from each substrate would be similar, (ii) excess lactate could displace pyruvate oxidation, (iii) the AMPK-induced activation of pyruvate dehydrogenase (PDH) (Smith et al. 2005) would enhance lactate as well as pyruvate oxidation, and (iv) exogenous provision of LDH would not alter lactate or pyruvate oxidation rates. However, in the present study none of these suppositions (i–iv), derived from the putative intracellular lactate shuttle hypothesis (Brooks et al. 1999b), were supported by the experimental data. Indeed, our studies, which are the first to compare pyruvate and lactate oxidation in isolated SS and IMF mitochondria from red and white muscles, demonstrate that pyruvate oxidation in SS and IMF mitochondria is greatly in excess of lactate oxidation, due to the absence of LDH in mitochondria.
- Top of page
Mature, male Sprague–Dawley rats (250–350 g) were used in these studies. They were bred on site and maintained at 22.5 ± 0.5°C with a 12 h light (19.00–07.00) and 12 h dark (07.00–19.00) cycle in animal holding facilities. Food and water were provided ad libitum. After animals had been anaesthetized (Somnotol, 60 mg kg−1, i.p.) red (RTA) and white tibialis anterior (WTA), and red (RG) and white gastrocnemius (WG) were removed. Since the muscle fibre composition of the RG and RTA are similar (McCullagh et al. 1996), and those of the WG and WTA are also similar (McCullagh et al. 1996), these respective red and white muscles were pooled to ensure that sufficient quantities of mitochondria were isolated for the experiments. Immediately after harvesting the muscles, the animals were killed with an overdose of Somnotol. The procedures for harvesting the muscle tissue and killing of the animals were approved by the Animal Care Committee at the University of Guelph.
Antibodies against MCT1 and 4 were custom made (Qiagen, Tokyo, Japan). Antibodies against MCT2 were obtained from Chemicon International (Temecula, CA, USA). All of these MCT antibodies have been used in our previous work (Bonen et al. 2000b, 2006; Yoshida et al. 2004). The monoclonal antibody MO25 (Matsuno et al. 1996) was used to detect fatty acid translocase (FAT)/CD36. Commercially available antibodies were used to detect cytochrome c oxidase IV (COX IV) (Santa Cruz Biotechnology, Santa Cruz, CA, USA), LDH (Abcam Inc., Cambridge, MA, USA), and Na+–K+-ATPase (Upstate Biotechnology, Lake Placid, NY, USA). A kit for determining LDH isozymes was also purchased (Paragon LD gel, Beckman, Fullerton, CA, USA). Radiolabelled [U-14C]-lactate, [l-14C]-pyruvate and [l-14C]-palmitate, were obtained from Amersham Biosciences (Buckinghamshire, UK). 5-Aminoimidazole-4-carboxamide-1-β-d-ribofuranoside (AICAR) was obtained from Toronto Research Chemicals Inc. (North York, ON, Canada). All other reagents were obtained from Sigma-Aldrich (St Louis, MO, USA).
Isolation of SS and IMF mitochondria
All studies were performed in freshly obtained subsarcolemmal (SS) and intermyofibrillar (IMF) mitochondria isolated separately from pooled RG and RTA muscles, and from pooled WG and WTA muscles. SS and IMF mitochondria were prepared on ice as we have recently described (Benton et al. 2004; Campbell et al. 2004), using modifications of the procedures published by Cogswell et al. (1993). Briefly, extracted muscles were placed in buffer 1 (100 mm KCl, 50 mm Tris HCl, 5 mm MgSO4 and 5 mm EDTA, pH 7.4). Then, the muscles were minced and diluted 10-fold in buffer 2 (buffer 1, supplemented with 0.84 mm ATP). Thereafter, muscle samples were homogenized only briefly using a polytron at 7500 r.p.m. (Polytron PT3100, Kinematica, Litttau-Lucerne, Switzerland) for 2 × 15 s. Subsequently the mitochondria were isolated by centrifugation at 800 g for 10 min. The resultant supernanant and pellet fractions were used to obtain SS and IMF mitochondria, respectively.
SS mitochondria The supernatant fraction derived from the initial centrifugation step was recentrifuged at 10 000 g for 10 min. Thereafter, the pellet was washed twice in buffer 2 (buffer 1 supplemented with 0.84 mm ATP) and centrifuged at 10 000 g for 10 min. The final pellet was resuspended in a final volume of 700 μl buffer (220 mm sucrose, 70 mm d-mannitol, 10 mm Tris-HCl, and 1 mm EDTA, pH 7.4)
IMF mitochondria The pellet fraction derived from the initial centrifugation step was rehomogenized using the polytron and centrifuged again at 800 g for 10 min. The supernatant fraction was discarded and the pellet was diluted 10-fold in buffer 2 and treated with subtilisin A (Sigma-Aldrich) for exactly 5 min. Addition of 5 ml of ice-cold buffer 2 arrested the protease activity. The samples were then centrifuged at 5000 g for 5 min. The pellet was resuspended in a 10-fold dilution of buffer 2 and centrifuged at 800 g for 10 min. The resultant supernatant fraction was centrifuged at 10 000 g for 10 min. The resultant pellet was washed twice in buffer 2, and was centrifuged at 10 000 g for 10 min. The final pellet was resuspended in a final volume of 700 μl buffer (220 mm sucrose, 70 mm d-mannitol, 10 mm Tris-HCl, and 1 mm EDTA, pH 7.4)
Mitochondrial purification for Western blotting To detect proteins associated specifically with mitochondria, it was necessary to remove contaminating subcellular materials, as we have done in our previous work (Benton et al. 2004; Campbell et al. 2004). For these purposes, the SS and IMF mitochondria were purified using a 60% Percoll gradient. The SS and IMF mitochondria were centrifuged at 20 000 g for 1 h after which the mitochondrial layer was removed from the interface between the Percoll and the aqueous phase. The Percoll was removed from the samples by further centrifugation at 20 000 g for 5 h. Thereafter, the purified mitochondria were frozen at −80°C until the samples were analysed.
Subcellular fractionation of skeletal muscle We used pooled rat hindlimb muscles (1.5 g) to examine the subcellular distribution of MCTs. For these purposes we used a procedure designed to retain all muscle constituents, except debris and mitochondria. We have described this fractionation procedure in detail previously (Bonen et al. 2000a). After the last centrifugation step, plasma membrane and low-density membrane fractions were removed and frozen (−80°C) until analysed by Western blotting.
Giant sarcolemmal vesicles Giant sarcolemmal vesicles, which consist of plasma membrane proteins (10%) and cytosolic protein (90%), but contain no mitochondria, were prepared from red and white skeletal muscle, as we have previously described in detail (Bonen et al. 1998, 2000a; Tonouchi et al. 2002; Enoki et al. 2006). Once prepared, the vesicles were frozen (−80°C) until analysed for LDH and COX IV with Western blotting.
Mitochondrial respiratory capacity
To determine the respiratory capacities of the red and white muscle SS and IMF mitochondria we have used the procedures outlined elsewhere (Ljubicic et al. 2004). Briefly, samples of isolated SS and IMF mitochondrial subfractions were incubated with VO2 buffer (250 mm sucrose, 50 mm KCl, 25 mm Tris-HCl, 10 mm K2HPO4 and 0.2% BSA, pH 7.4) at 30°C in a water-jacketed respiratory chamber with continuous stirring. Oxygen consumption (n atoms O2 (mg protein)−1 min−1) was assessed in the presence of 11 mm glutamate (state 4 respiration), or glutamate plus 0.4 mm ADP (state 3 respiration) using a Clark oxygen electrode (YSI, Yellow Springs, OH, USA), as we have reported previously (Ljubicic et al. 2004). The degree of coupling was described by the respiratory control ratio (RCR), which expresses how strongly ADP is capable of stimulating oxygen consumption compared with the basal (or state 4) rate. The RCR was calculated by dividing the state 3 rate by the state 4 rate of each isolated mitochondrial preparation.
Lactate and pyruvate oxidation in SS and IMF mitochondria
Oxidation of lactate and pyruvate were determined using radiolabelled [U-14C]-lactate and [l-14C]-pyruvate in a sealed system as we have previously described (Campbell et al. 2004; Bezaire et al. 2006; Holloway et al. 2006). Briefly, oxidation experiments were performed in sealed 20 ml scintillation vials. Mitochondria (100 μl, protein concentration > 2 mg ml−1) were added to a 900 μl aliquot of pre-gassed (5% CO2 and 95% O2) modified Krebs–Ringer buffer (MKR: 115 mm NaCl, 2.6 mm KCl, 1.2 mm KH2PO4, 10 mm NaHCO3, 10 mm Hepes, pH 7.4), which was supplemented with ATP (5 mm), NAD+ (1 mm), l-carnitine (0.5 mm), l-malate (0.5 mm), coenzyme A (0.1 mm) and cytochrome C (25 μm) for the oxidation studies, as we have reported previously (Campbell et al. 2004; Bezaire et al. 2006; Holloway et al. 2006). Oxidation (30 min, 37°C) was initiated by the addition of 0.18, 1.8 or 10 mm l-lactate and l-[U-14C]-lactate (0.4 μCi vial−1) or 0.18, 1.8 and 10 mm l-pyruvate and [1-14C]-pyruvate (0.4 μCi vial−1) (Amersham Biosciences, Buckinghamshire, UK). Oxidation was terminated with the addition of 70% perchloric acid (Sigma-Aldrich). During the incubation the 14CO2 produced was trapped in 1 m benzethonium hydroxide (500 μl) contained in a microcentrifuge tube suspended within the scintillation vial. Acidifying the reaction mixture with 1.0 ml of 1 m H2SO4, liberated 14CO2 from the incubating buffer and captured it in the benzethonium hydroxide trap (60 min at room temperature). Thereafter, the microcentrifuge tube containing the 14CO2 was placed in a scintillation vial, and radioactivity was counted.
Inhibition of lactate (1.8 mm) and pyruvate (0.18 mm) oxidation was examined with 5 mmα-cyano-4-hydroxycinnamate (CINN) and with 50 mm sodium oxamate. For these purposes mitochondria were preincubated with each inhibitor for 30 min. The same volumes of ethanol (for CINN) or H2O (for oxamate) used to dissolve the inhibitors were added to control vials. After the 30 min preincubation, mitochondria were centrifuged at 10 000 g for 10 min and washed twice. Thereafter, lactate and pyruvate oxidation were determined as above.
Effects of LDH on mitochondrial lactate and pyruvate oxidation
To examine the effects of exogenously provided LDH on lactate (1.8 mm) and pyruvate (0.18 mm) oxidation, red and white muscle SS and IMF mitochondria were incubated with progressively increased quantities of LDH-1 (0–10 units). Lactate and pyruvate oxidation were determined as described above.
Acute AICAR stimulation
To examine the stimulatory effects of AMPK on mitochondrial lactate and pyruvate oxidation, the cell-permeant adenosine analogue AICAR (1 mg (g body weight)−1 in 0.45% saline, 3 ml (300 g body weight)−1) was injected subcutaneously (s.c.) into the back neck of rats. Control rats were injected with an equivalent volume of 0.45% saline. After 1 h, RTA, RG, WTA and WG muscles were isolated and rates of lactate (1.8 mm) and pyruvate (0.18 mm) oxidation were also determined in SS and IMF mitochondria from red and white muscles as described above.
Substrate competition studies
To determine whether mitochondria preferred oxidation of pyruvate or lactate, competition studies were performed. For these purposes we determined whether excess quantities of lactate (10 mm) inhibited the oxidation of pyruvate (1.8 mm), and whether excess pyruvate (10 mm) inhibited oxidation of lactate (1.8 mm). In addition, we also examined whether lactate (10 mm) and pyruvate (10 mm) inhibited palmitate (1.8 mm) oxidation in red and white muscle SS and IMF mitochondria. Palmitate oxidation (1.8 mm, [1-14C]-palmitate, 0.5 μCi vial−1) was determined in a similar manner as was done for lactate and pyruvate, and as we have previously described (Campbell et al. 2004).
MCT1, 2, 4 as well as other proteins (LDH, Na+–K+-ATPase, FAT/CD36), were detected on mitochondria, and on plasma membranes in some instances, using Western blotting as we have previously reported (Benton et al. 2004; Campbell et al. 2004). Proteins were visualized by chemiluminescence detection, according to the manufacturer's instructions (Hyperfilm-ECL; Amersham). Blots were quantified using the ChemiGenius 2 Bioimaging system (SynGene, Cambridge, UK).
LDH activity and LDH isoenzymes detection
LDH activity and isoenzymes were determined as previously described (Chi et al. 1983; McCullagh et al. 1996). Briefly, mitochondrial fractions and whole muscle homogenates were diluted with 10 mm Tris-buffer (pH 7.5). The LDH activity was determined in the pyruvate-to-lactate direction using a fluorometric assay (Chi et al. 1983; McCullagh et al. 1996). LDH isoenzymes in mitochondrial fractions and whole muscle homogenates were separated by adding 20 μg or 1 μg protein, respectively, to agarose gels followed by electrophoresis for 40 min at 100 V according to the manufacture's instructions (Paragon LD isoenzyme electrophoresis kit, Beckman, Fullerton, CA, USA). Thereafter, gels were covered with a blotter saturated in substrate (208 mm lithium l-lactate, 5.6 mm NAD, 2.4 mmp-nitro blue tetrazolium chloride and 0.33 mm phenazine methosulphate (PMS)) and incubated for 20 min at 45°C. Gels were fixed in 5% acetic acid and dried. LDH isoenzyme bands were scanned and quantified using the ChemiGenius 2 Bioimaging system (SynGene).
The data were analysed with analyses of variance. Fisher's least-squares difference test was used for post hoc analyses, when appropriate. The significance level was set at P≤ 0.05 and all data are reported as means ±s.e.m.