Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Jan 1.
Published in final edited form as: Exp Physiol. 2012 May 21;98(1):207–219. doi: 10.1113/expphysiol.2012.066688

Effect of exercise on mouse liver and brain bioenergetic infrastructures

E Lezi 1, Jianghua Lu 2, Jeffrey M Burns 2,3, Russell H Swerdlow 2,3,4
PMCID: PMC3540163  NIHMSID: NIHMS429511  PMID: 22613742

Abstract

To assess the effects of exercise on liver and brain bioenergetic infrastructures, we exposed C57BL/6 mice to 6 weeks of moderate-intensity treadmill exercise. During the training period, fasting blood glucose was lower in exercised mice than in sedentary mice, but serum insulin levels were not reduced. At week 6, trained mice showed a paradoxical decrease in plasma lactate during exercise, which was accompanied by an increase in the liver monocarboxylate transporter 2 protein level (~30%, P < 0.05). Exercise increased liver peroxisomal proliferator-activated receptor-γ coactivator 1α expression (approximately twofold, P < 0.001), NAD-dependent deacetylase sirtuin-1 protein (~30%, P < 0.05), p38 protein (~15%, P < 0.05), cytochrome c oxidase subunit 4 isoform 1 protein (~50%, P < 0.05) and AMP-activated protein kinase phosphorylation (~40%, P < 0.05). Despite this, liver mitochondrial DNA copy number (~30%, P = 0.05), mitochondrial transcription factor A expression (~15%, P < 0.05), cytochrome c oxidase subunit 2 expression (~10%, P < 0.05), cAMP-response element binding protein phosphorylation (~60%, P < 0.05) and brain-derived neurotrophic factor expression (~40%, P < 0.05) were all reduced, while cytochrome oxidase and citrate synthase activities were unchanged. The only altered brain parameter observed was a reduction in tumour necrosis factor α expression (~35%, P < 0.05); tumour necrosis factor α expression was unchanged in liver. Our data suggest that lactate produced by exercising muscle modifies the liver bioenergetic infrastructure, and enhanced liver uptake may in turn limit the ability of exercise-generated lactate to modify brain bioenergetics. Also, it appears that, at least in the liver, a dissociated mitochondrial biogenesis, in which some components are strategically enhanced while others are minimized, can occur.

Physical exercise has systemic benefits. It has a favourable effect on blood glucose levels, cholesterol and blood pressure (Richter et al. 1992; Stewart, 2004; Jia et al. 2012). Some of the effects of exercise result from an increase in muscle mitochondrial mass and bioenergetic efficiency (Rockl et al. 2008; Little et al. 2010). Mitochondrial mass is increased through the process of mitochondrial biogenesis, which in turn is regulated by energy and redox-sensing pathways that converge on the peroxisomal proliferator-activated receptor-γ coactivator 1α (PGC-1α), nuclear respiratory factor 1 (NRF-1) and mitochondrial transcription factor A (TFAM) cotranscription and transcription factors (Vinã et al. 2009). Among these proteins, engagement of PGC-1α is an upstream event. Peroxisomal proliferator-activated receptor-γ coactivator 1α activates NRF-1, and this initiates TFAM expression (Vinã et al. 2009). Mitochondrial transcription factor A plays roles in both mitochondrial DNA (mtDNA) replication and gene expression.

Although muscle is the primary tissue used during exercise, both the liver and the brain are also engaged and modified. The expression and activities of antioxidant enzymes increase in both tissues (Somani & Husain, 1996; Wilson & Johnson, 2000; Devi & Kiran, 2004; Navarro et al. 2004). In brain, brain-derived neurotrophic factor (BDNF) and vascular endothelial growth factor levels appear to rise, consequently driving angiogenesis and neurogenesis (Fabel et al. 2003; Vaynman et al. 2004; Berchtold et al. 2010). One recent study even reported that high-intensity exercise increases brain PGC-1α expression and several other markers of mitochondrial mass (Steiner et al. 2011).

Why exercise-induced muscle activation modifies non-muscle tissues is not fully understood. This phenomenon presumably must be mediated by either a removal from or release into the blood of specific molecules. Regarding the latter possibility, lactate constitutes a potential candidate, because lactate produced by contracting muscles enters the bloodstream, from where it is taken up by liver, heart and skeletal muscle monocarboxylate transporters (MCTs; Bonen et al. 2006).

Lactate figures prominently in both liver and brain bioenergetic metabolism. In the Cori cycle, blood lactate is transferred to hepatocytes by monocarboxylate transporter 2 (MCT2) and enters gluconeogenesis. The glucose produced is released into the blood and helps to prevent hypoglycaemia during sustained exercise (Hoene & Weigert, 2010). When blood lactate levels are sufficiently elevated, lactate from the blood can also enter the brain (Ide et al. 2000; van Hall et al. 2009). This could affect brain physiology, because lactate is used by neurons to support oxidative phosphorylation (Pellerin et al. 2007), and lactate also appears to play a role in memory formation (Veneman et al. 1994; Berthet et al. 2009; Suzuki et al. 2011).

The primary purpose of this study was to evaluate the effects of moderate exercise on mouse liver and brain bioenergetic infrastructures. To do this, we analysed pathways and proteins that sense cell energy and redox states, mediate adaptive responses to cell energy and redox state changes, and execute these responses. As part of this analysis, we especially considered the potential role that muscle-generated lactate might play.

Methods

Animals

The animal work described in this study was approved by the Institutional Animal Care and Use Committee of the University of Kansas Medical Center, and efforts were made to minimize animal discomfort. Twelve male, 4-month-old, C57BL/6 mice were obtained from the Jackson Laboratory. All mice were maintained on an ad libitum diet. The mice were housed four per cage on a 12 h–12 h light–dark schedule. After a 1 week adaptation period, the 4.25-month-old mice were randomly placed into two groups, a control/sedentary group (CT, n = 6) or an exercise group (EX, n = 6).

Exercise training

Mice in the EX group were exercised for two sessions per day on a six-lane treadmill designed for mice (Columbus Instruments, Columbus, OH, USA). Each session consisted of a 3 min warm-up at 15 m min−1 plus 42 min at 18 m min−1. This speed approximates the lactate threshold for untrained C57BL/6 mice (Billat et al. 2005). The training was performed 5 days per week for 6 weeks. Mice in the CT group were not subjected to any exercise training. In order to avoid confounding factors such as sound and light, during the training sessions the CT mice were placed in the same room as the EX mice. After the 6 weeks of training, EX mice were killed by decapitation 30 min after the last session. The CT mice were also decapitated on the same day. Liver and brain tissue were immediately frozen in liquid nitrogen, and stored at −80°C for later analysis.

Glucose, insulin and lactate levels

Blood glucose levels were measured using a One-Touch Ultra Blood Glucose Monitoring System (LifeScan, Milpitas, CA, USA). Plasma samples were also prepared from tail vein blood that was collected in heparinized microhaematocrit capillary tubes (Fisher Scientific, Pittsburgh, PA, USA). Plasma insulin levels were measured using an insulin ELISA kit according to the manufacturer’s instructions (American Laboratory Products Company, Salem, NH, USA). Blood glucose and insulin were measured during a period of unrestricted food access (non-fasting) and after a 15 h fast (fasting blood glucose). Values for the homeostasis model assessment of insulin resistance (HOMA-IR) were calculated from the product of fasting serum glucose (millimolar) and insulin (in microunits per millilitre) divided by 22.5 (Matthews et al. 1985). Plasma lactate levels were assayed using a commercial L-lactate assay kit (Eton Bioscience Inc., San Diego, CA, USA), and the values were normalized to the mean lactate level of CT mice at rest on the same day in order to minimize day-to-day variation.

Complex IV and citrate synthase (CS) activity

Crude mitochondrial fractions isolated from the left cerebral hemisphere and the right liver lobe were used to measure complex IV (cytochrome oxidase; COX) and citrate synthase Vmax activities. To isolate mitochondria, brain and liver tissue were homogenized in mannose sucrose hepes EDTA (MHSE) buffer (210 mM mannitol, 70 mM sucrose, 5 mM Hepes, 1 mM EDTA and 0.5% fatty-acid-free bovine serum albumin, pH 7.2) using a Teflon homogenizer and centrifuged at 800g for 10 min. The supernatant was centrifuged again at 8000g for 10 min. The resulting pellets containing mitochondria were resuspended in MSHE buffer and frozen at −80°C for later use. The COX and CS Vmax activities were determined as previously described (Ghosh et al. 2007). Cytochrome oxidase Vmax activities were normalized to protein concentration, as well as to citrate synthase activity.

Western blotting

Protein lysates were prepared using the front half of the right cerebral hemisphere and the left lobe of the liver. For total protein lysates, Radioimmunoprecipitation Assay Buffer (Cell Signaling Technology, Danvers, MA, USA) was used. The tissue was homogenized in the buffer, and the homogenates were subsequently sonicated three times, for 5 s each time, at setting four using an F60 Sonic Dismembrator (Fisher Scientific, Pittsburgh, PA, USA). To prepare crude membrane protein fractions, tissue was placed in sucrose-Tris (ST) buffer (250 mM sucrose and 10 mM Tris base, pH 7.5) and homogenized by 10 strokes of a Dounce homogenizer. The homogenates were centrifuged at 1000g for 5 min, and the supernatant was further centrifuged at 100,000g for 1.5 h. The resulting pellets containing membrane proteins were resuspended in ST buffer for further analysis. Protein concentration was measured using a BCA protein assay reagent kit (Thermo Scientific, Rockford, IL, USA).

Several energy-sensitive proteins or proteins that influence energy metabolism were analysed by Western blot. Monocarboxylate transporter 2, a proton-linked plasma membrane transporter that transfers lactate into hepatocytes and neurons, was also analysed. Primary antibodies to the following proteins were used: MCT2 (1:200 dilution; sc-166925; Santa Cruz Biotechnology, Santa Cruz, CA, USA); phospho-Thr172 AMP-activated protein kinase (AMPK; 1:1000 dilution; 2531; Cell Signaling Technology); AMPK (1:1000 dilution; 2603; Cell Signaling Technology); cAMP-response element binding protein (CREB; 1:1000 dilution; sc-25785; Santa Cruz Biotechnology); phospho-Ser133 CREB (1:500 dilution; 9198; Cell Signaling Technology); p38 (1:1000 dilution; 9212; Cell Signaling Technology); phospho-p38 (1:1000 dilution; 4511; Cell Signaling Technology); mammalian target of rapamycin (mTOR; 1:1000 dilution; 2983; Cell Signaling Technology); phospho-Ser2448 mTOR (1:1000 dilution; 2976; Cell Signaling Technology); NAD-dependent deacetylase sirtuin-1 (SIRT1; 1:500 dilution; 2028; Cell Signaling Technology); PGC-1α (1:1000 dilution; 516557; Millipore, Billerica, MA, USA); and cytochrome c oxidase subunit 4 isoform 1 (COX4I1; 1:2000 dilution; A21348; Invitrogen, Carlsbad, CA, USA). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH; 1:2000 dilution; 2118; Cell Signaling Technology) was used as a loading control for total protein lysates. Pancadherin (1:1000 dilution; ab16505; Abcam, Cambridge, MA, USA) was used as a loading control for the membrane protein fraction. Primary antibody binding was detected using horseradish peroxidase-conjugated secondary antibodies (1:2000 dilution; Cell Signaling Technology) and SuperSignal West Femto Maximum Sensitivity Substrate (Thermo Scientific). Densitometry was performed using a ChemiDoc XRS with Quantity One software (Bio-Rad, Hercules, CA, USA).

Quantitative real-time RT-PCR

Total RNA was prepared from frozen brain and liver tissue using the TRI Reagent (Life Technologies, Grand Island, NY, USA). Reverse transcription was performed with total RNA (1 μg) using a High Capacity cDNA Reverse Transcription kit (Applied Biosystems, Foster City, CA, USA). Quantitative real-time RT-PCR (qPCR) was performed using the TaqMan Universal PCR Master Mix (Applied Biosystems) and ready-to-use TaqMan Gene Expression Assays (Applied Biosystems) to quantify the mRNA levels of MCT2, PGC-1α, NRF-1, TFAM, cytochrome c oxidase subunit 2 (COX2), COX4I1, tumour necrosis factor α (TNF-α), CREB and BDNF. Glyceraldehyde-3-phosphate dehydrogenase was used as an internal loading control. Real-time RT-PCR amplification was determined using an Applied Biosystems StepOnePlus Real-Time PCR System (Applied Biosystems).

To quantify liver mtDNA, total DNA was extracted with a DNeasy Blood & Tissue Kit (Qiagen, Valencia, CA, USA). TaqMan Gene Expression Assays (Applied Biosystems) for two mtDNA-encoded genes, NADH dehydrogenase subunit 2 (ND2) and 16s ribosomal RNA (rRNA), and the nuclear 18s rRNA gene were used. The relative mtDNA to nuclear DNA copy number ratio was determined using the comparative ΔΔCt method, in which ND2:18s rRNA and 16s rRNA:18s rRNA ratios were calculated.

Statistical analysis

Data were expressed as means ± SEM. Mean values were compared by Student’s unpaired t test or Student’s paired t test using SPSS 18.0 (SPSS Inc., Chicago, IL, USA). Pearson’s correlation analysis was performed to determine relationships between MCT2 and PGC-1α. Values of P < 0.05 were considered significant.

Results

Effect on weight, glucose and insulin

At the beginning of the study, body weights were comparable between the CT and EX groups. After 6 weeks, the CT mice gained significantly more weight (baseline, 29.28 ± 0.42 g; after 6 weeks, 30.82 ± 0.40 g; P < 0.05; Fig. 1A), while the EX mice maintained their weight over the 6 weeks (baseline, 29.70 ± 0.32 g; after 6 weeks of training, 29.95 ± 0.54 g; P > 0.05; Fig. 1A). After 3 weeks, the non-fasting plasma glucose showed a downwards trend in EX mice (CT, 184.33 ± 5.95 mg dl−1; EX, 167.67 ± 8.33 mg dl−1; P = 0.06; Fig. 1B). At 3.5 weeks, fasting blood glucose levels were significantly lower in the EX mice (CT, 119.33 ± 5.75 mg dl−1; EX, 95.08 ± 5.23 mg dl−1; P < 0.05; Fig. 1B). At the beginning of the study, non-fasting plasma insulin levels and 15 h fasting plasma insulin levels (both at rest) were comparable between the CT and EX groups (data not shown). After 6 weeks of training, we did not observe intergroup differences in either the non-fasting insulin level (CT, 1.16 ± 0.20 ng ml−1; EX, 1.77 ± 0.27 ng ml−1; P > 0.05; Fig. 1C) or the 15 h fasting insulin level (CT, 0.42 ± 0.09 ng ml−1; EX, 0.80 ± 0.35 ng ml−1; P > 0.05; Fig. 1C). The HOMA-IR calculation suggested that 6 weeks of moderate-intensity treadmill training did not increase insulin sensitivity (CT, 3.04 ± 0.79; EX, 4.41 ± 1.72; P > 0.05; Fig. 1D).

Figure 1. Effects of treadmill exercise training on weight, glucose and insulin.

Figure 1

Open in a new tab

A, control, sedentary (CT) mice gained significantly more weight, whereas the exercised (EX) mice maintained their weight over the 6 weeks. B, a trend towards lower plasma glucose levels without fasting was seen in the EX mice after 3 weeks of training (P = 0.06). After 3.5 weeks of training, fasting blood glucose levels were significantly lower in the EX mice. C, after 6 weeks of training, intergroup differences were not observed in the non-fasting insulin level and the 15 h fasting insulin level. D, the homeostasis model assessment of insulin resistance (HOMA-IR) value was unchanged by 6 weeks of moderate-intensity treadmill training.

Plasma lactate and MCT2

After the 6 week training period, resting plasma lactate levels were comparable between the CT and EX groups (P > 0.05; data not shown). At the end of the 6 weeks, we found that the plasma lactate level declined in the EX mice while they were running. The plasma lactate appeared to return gradually to its resting baseline level over the first 2 h of the recovery period (Fig. 2). Protein levels of MCT2, a transporter responsible for cell lactate uptake, were significantly increased in EX mice livers (by ~30%; P < 0.05), although no intergroup difference was seen in mRNA expression (Fig. 3). Other investigators have also observed a similar dissociation between MCT2 protein levels and mRNA expression (Jackson et al. 1997). Exercise training did not change brain MCT2 mRNA expression or protein levels (Fig. 3).

Figure 2. Plasmalactate levels during exercise.

Figure 2

Open in a new tab

At the end of the 6 weeks, plasma lactate levels declined in EX mice during treadmill running. Plasma lactate levels appeared to recover gradually over the first 2 h of the postexercise period. *P < 0.05 relative to the lactate level at rest.

Figure 3. Effect of exercise on monocarboxylate transporter 2 (MCT2).

Figure 3

Open in a new tab

A, MCT2 protein increased (*P < 0.05) in the EX mouse livers, while brain MCT2 protein levels were unchanged. B, no changes in MCT2 mRNA were seen in liver or brain.

Mitochondrial enzyme activities

Liver and brain COX Vmax activities, when normalized to protein concentration, were comparable between the CT and EX groups (Table 1). Normalizing COX activity to CS activity, a parameter sometimes used to control independently for mitochondrial mass, did not change this relationship. Citrate synthase activities in both liver and brain were also comparable between the groups.

Table 1.

Brain and liver cytochrome c oxidase (COX) and citrate synthase (CS) activities

COX (s−1 (mg protein) −1)
CS (nmol min−1 (mg protein)−1)
COX/CS ratio
CT EX CT EX CT EX
Liver 1.11 ± 0.07 1.25 ± 0.07 207.87 ± 5.38 185.17 ± 11.70 0.0054 ± 0.00041 0.0070 ± 0.00080
Brain 1.35 ± 0.03 1.28 ± 0.03 999.79 ± 33.74 988.53 ± 24.50 0.0014 ± 0.00003 0.0013 ± 0.00004
Open in a new tab

Values are given as means ± SEM. Abbreviations: CT, control, sedentary group; and EX, exercise group.

Bioenergetics-related proteins

Liver but not brain PGC-1α mRNA was significantly increased in EX mice (approximately twofold; P < 0.001; Fig. 4A), while the liver and brain PGC-1α protein levels were unchanged (Fig. 4B). Liver PGC-1α mRNA and MCT2 protein levels were positively correlated (r = 0.67, P < 0.05; Fig. 5A). No correlation was seen between liver PGC-1α and MCT2 protein levels, liver PGC-1α and MCT2 mRNA levels, or liver PGC-1α protein and MCT2 mRNA levels (Fig. 5B–D).

Figure 4. Effect of exercise on peroxisomal proliferator-activated receptor-γ coactivator 1α (PGC-1α ).

Figure 4

Open in a new tab

A, PGC-1α mRNA was increased (*P < 0.05) in the EX group livers, but exercise training did not alter brain PGC-1α mRNA. B, exercise did not affect PGC-1α protein levels in either tissue.

Figure 5. Relationships between MCT2 and PGC-1α.

Figure 5

Open in a new tab

A, MCT2 protein and PGC-1α mRNA levels were positively correlated. B–D, no correlation was seen between MCT2 protein levels and PGC-1α protein levels, between MCT2 mRNA levels and PGC-1α mRNA levels, or between MCT2 mRNA levels and PGC-1α protein levels.

We assessed the status of several proteins that are known to regulate PGC-1α. The protein level of SIRT1, which activates PGC-1α by deacetylating it (Cantó et al. 2009), was increased in the livers but not brains of the EX mice (~30% increase; P < 0.05; Fig. 6A). AMP-activated protein kinase, another important metabolic sensor that monitors intracellular AMP/ATP ratios (Hardie, 2007), showed increased Thr172 phosphorylation in the EX mice livers but not in brains. This increase was observed when normalized to either total AMPK or GAPDH (~40% increase; P < 0.05; Fig. 6B and C). Liver and brain total AMPK protein levels were comparable between groups (Fig. 6D).

Figure 6. Effect of exercise on NAD-dependent deacetylase sirtuin-1 (SIRT1) and AMP-activated protein kinase (AMPK).

Figure 6

Open in a new tab

A, SIRT1 protein was increased (*P < 0.05) in the EX group livers, but exercise training did not alter the brain SIRT1 protein level. B–D, AMPK Thr172 phosphorylation was increased (*P < 0.05) in the EX mice livers. This increase was observed when corrected for both total AMPK and glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Exercise training did not alter brain AMPK phosphorylation and liver or brain AMPK protein levels.

We analysed p38 mitogen-activated protein kinase (p38 MAPK) activation by Western blot, because p38 MAPK has been shown to activate PGC-1α (Cao et al. 2005). When normalized to GAPDH, p38 phosphorylation in the EX mouse livers showed a trend towards an increase (~15% increase; P = 0.087; Fig. 7A). When corrected for total p38, however, this trend was no longer apparent (Fig. 7B). The total p38 level, when normalized to GAPDH, was higher in EX mouse livers (~15% increase; P < 0.05; Fig. 7C), which suggests that any apparent increase in liver p38 phosphorylation was a secondary consequence of increased total p38 protein. Brain total p38 levels were comparable between the EX and CT groups.

Figure 7. Effect of exercise on p38 mitogen-activated protein kinase and mammalian target of rapamycin (mTOR).

Figure 7

Open in a new tab

A–C, when normalized for GAPDH, liver p38 phosphorylation showed a trend to be higher in the EX mice, but no intergroup difference was suggested when phospho-p38 was normalized to total p38. Total p38 was significantly increased (*P < 0.05) in EX mouse livers. No intergroup difference was observed in brain phospho-p38 or total p38. D–F, brain and liver phospho-mTOR and total mTOR were comparable between groups.

We assessed the activation status of mTOR, a kinase that regulates cell growth, size and survival and which can form a complex with PGC-1α (Cunningham et al. 2007). There was no difference in mTOR Ser2448 phosphorylation, when normalized to either GAPDH or total mTOR, in either liver or brain (Fig. 7D and E). Liver and brain total mTOR protein levels were also equivalent between the two groups (Fig. 7F).

Cyclic AMP-response element binding protein can reportedly activate PGC-1α expression (Herzig et al. 2001). Liver CREB mRNA expression was lower in the EX mice than it was in control mice (Fig. 8A), although total CREB protein levels were comparable (Fig. 8B). Ser133 CREB phosphorylation was reduced in the EX mouse livers when normalized to GAPDH (~60% decrease; P < 0.05), but when normalized to total CREB this exercise-induced phospho-CREB reduction did not remain statistically significant (Fig. 8C and D). No changes in brain CREB expression, total protein or phosphorylation were observed. In addition, to help address whether a functional reduction in liver CREB activity occurred, we measured the expression of BDNF, because CREB drives BDNF expression (Finkbeiner, 2000). Brain-derived neurotrophic factor mRNA levels were significantly lower in EX mouse livers (~40% decrease; P < 0.05; Fig. 8E). Contrary to what has been reported in other studies of exercised mice (Chen & Russo-Neustadt, 2009; Liu et al. 2009; Rasmussen et al. 2009), but consistent with our observed lack of brain CREB changes, we did not observe changes in brain BDNF expression.

Figure 8. Effect of exercise on cAMP-response element binding protein (CREB) and brain-derived neurotrophic factor (BDNF).

Figure 8

Open in a new tab

A, exercise training decreased (*P < 0.05) CREB mRNA expression in the liver but did not alter its expression in the brain. B–D, when normalized to GAPDH, CREB total protein was equivalent between groups, while phospho-CREB (at Ser133) was significantly lower (*P < 0.05) in the livers of EX mice. There was no difference in the phospho-CREB/CREB ratio. No changes in brain total or phospho-CREB protein were observed. E, although exercise did not alter brain BDNF mRNA expression, liver BDNF mRNA expression was reduced (*P < 0.05) in the EX mice.

Nuclear respiratory factor 1 and TFAM are regulated by PGC-1α (Vinã et al. 2009). Liver and brain NRF-1 mRNA levels did not differ between the groups (Fig. 9A). Liver TFAM expression, but not brain TFAM expression, was significantly reduced in EX mice (~15% decrease; P < 0.05; Fig. 9B). In accordance with this reduction in liver TFAM mRNA, we also observed that mtDNA content was lower in the EX mouse livers (16s rRNA/18s rRNA, ~25% decrease, P = 0.1; ND2/18s rRNA, ~30% decrease, P = 0.05; Fig. 10). The mRNA levels of COX2, an mtDNA-encoded protein that constitutes part of the COX holoenzyme, was decreased in the EX group livers (~10% decrease; P < 0.05; Fig. 11A). Brain levels were statistically equivalent. Interestingly, levels of COX4I1 protein, a nuclear DNA-encoded constituent of the COX holoenzyme, were significantly increased in the EX mouse livers (~50% increase; P < 0.05; Fig. 11B). Protein levels of COX4I1 were similar in brain, and liver and brain COX4I1 mRNA expression between EX and CT mice was comparable (Fig. 11B and C).

Figure 9. Expression of nuclear respiratory factor 1 (NRF-1) and mitochondrial transcription factor A (TFAM).

Figure 9

Open in a new tab

A, exercise training did not alter liver or brain NRF-1 mRNA levels. B, TFAM mRNA expression was lower (*P < 0.05) in the EX group livers, but unchanged in brain.

Figure 10. Quantitative PCR analysis of liver mitochondrial DNA (mtDNA) content.

Figure 10

Open in a new tab

Total DNA was extracted from liver tissue, and mtDNA was analysed by real-time quantitative PCR using primers targeting the 18s RNA nuclear gene and the NADH dehydrogenase subunit 2 (ND2) or 16s RNA mtDNA genes. The 16s:18s ratio showed a trend to be lower in the EX mice (P = 0.1), while the ND2:18s ratio was at the significance cut-off (*P = 0.05) in the EX mice.

Figure 11. Effect of exercise on cytochrome c oxidase subunit 2 (COX2) and cytochrome c oxidase subunit 4 isoform 1 (COX4I1).

Figure 11

Open in a new tab

A, liver COX2 mRNA levels were decreased (*P < 0.05) in the EX group, but were unchanged in the brain. B and C, COX4I1 protein increased (*P < 0.05) in the livers of EX mice, while liver COX4I1 mRNA expression did not change. Exercise did not alter brain COX4I1 expression or protein levels.

Bioenergetic metabolism is known to influence inflammation and inflammatory markers (Salminen et al. 2011). To assess the effects of exercise on a marker of inflammation, we measured liver and brain TNF-α expression. Although exercise did not alter liver TNF-α mRNA levels, TNF-α mRNA levels were reduced in EX mouse brains (~35% decrease; P < 0.05; Fig. 12).

Figure 12. Effect of exercise on tumour necrosis factorα (TNF-α) expression.

Figure 12

Open in a new tab

Tumour necrosis factor α mRNA expression was comparable in the livers but lower (*P < 0.05) in the brains of EX mice.

Discussion

We found that when young male C57BL/6 mice were exercised regularly on a treadmill, at 18 m min−1 for two daily sessions, five days per week for 6 weeks, their ability to clear blood lactate increased. In conjunction with this adaptation, and probably contributing to it, liver MCT2 protein levels rose. Liver but not brain PGC-1α expression increased. The NAD-dependent deacetylase sirtuin-1, AMPK, p38 MAPK and CREB are all reported to activate PGC-1α activity or expression, and our data suggest that SIRT1, AMPK and p38 MAPK, but not CREB, may have contributed to the observed increase in PGC-1α.

Although PGC-1α is known to co-ordinate mitochondrial biogenesis, and protein levels of the nuclear DNA-encoded electron transport chain COX4I1 subunit rose in the liver, the liver TFAM expression, COX2 expression and mtDNA content were reduced. In the liver, therefore, exercise appeared to induce at most a relatively selective mitochondrial biogenesis, in which the respiratory capacity was not enhanced. This is potentially consistent with the fact that from a bioenergetics perspective, a key liver function is glucose homeostasis (Herzig et al. 2001). In exercise conditions, it accomplishes this through gluconeogenesis. In our study, the liver bioenergetic infrastructure may have changed to facilitate gluconeogenesis at the expense of oxidative phosphorylation.

If correct, these data could help to explain why the transfer of mtDNA genes to the nucleus, a process that has played out over the course of evolution, remains incomplete. Activating nuclear genes that promote mitochondrial biogenesis, while downregulating mtDNA and transcription factors that promote mtDNA gene expression, would predictably shift mitochondrial function from respiration and towards other non-respiratory activities, such as gluconeogenesis. Retaining key respiratory chain subunit genes on the mtDNA, therefore, could help facilitate this high degree of flexibility.

Lactate produced during exercise appears to have initiated the liver changes we observed. This is suggested by the apparent correlation between PGC-1α expression and MCT2 protein. It is further supported by the fact that by the end of the training period, blood lactate decreased during exercise. We believe this represents evidence of an enhanced Cori cycle, which would contribute to endurance by increasing the liver’s ability to maintain glucose homeostasis during conditions of sustained exertion. Also, we selected a treadmill speed of 18 m min−1 because this speed reportedly corresponds to the lactate threshold of untrained C57BL/6 mice (Billat et al. 2005). By the end of the 6 week training period, the 18 m min−1 speed was clearly below the lactate threshold of our EX mice, which would correspond at most to a ‘moderate’-intensity human exercise regimen.

Our data do not establish a role for lactate in mediating the effects of exercise on the brain, but the paucity of brain changes we observed perhaps provides indirect support for this possibility. Certainly, if lactate entering the brain from the blood can modify the infrastructure of brain bioenergetics, then an inability to deliver exercise-generated lactate to the brain would blunt the effects of exercise on the brain. The enhanced ability of the liver to clear exercise-generated lactate, a consequence of training, would function in this capacity.

Indeed, the only brain change we observed with exercise was a reduction in TNF-α expression. This suggests that exercise may reduce brain inflammation set points, and that this effect occurs independently of lactate. Interestingly, we did not detect an increase in brain BDNF expression, a change that has been reported in other mouse exercise studies (Chen & Russo-Neustadt, 2009; Liu et al. 2009; Rasmussen et al. 2009). On this point there is a fair amount of equipoise, though, because other treadmill exercise studies have found that brain BDNF does not increase or even decreases (Aguiar et al. 2007, 2008; Wu et al. 2011). We wonder whether differences among these studies are due to methodological factors, such as the age of the mice, the brain regions analysed, or the exercise protocol that was used.

Likewise, while one study did report that exercise increased mouse brain and liver COX activity (Navarro et al. 2004), we did not observe a change in COX activity in either tissue. We suspect that this may also reflect methodological factors, because the mice in the positive study were older than the mice we used. Certainly, it might be expected that the ability of exercise to reverse physiological or biochemical declines is greater than its ability to enhance bioenergetic systems that are already optimally functioning. Finally, Steiner et al. (2011) reported that exercise did increase PGC-1α expression in mouse brain, an effect we did not observe in our mice, but the intensity of the exercise protocol in that study (1 h treadmill sessions at 25 m min−1 and a 5% incline, 6 days per week for 8 weeks) was greater than in our study.

On other points our molecular findings and literature data are essentially consistent. A prior study using male C57BL/6 mice found that after a single session of high-intensity treadmill exercise, liver PGC-1α expression increased (Hoene et al. 2009). Hepatic ATP concentrations significantly decrease, while AMP increases immediately after exercise, which would be expected to activate AMPK (Camacho et al. 2006). We postulate that this is why AMPK Thr172 phosphorylation, which is often used to assess the activation status of AMPK (Hawley et al. 1996), occurred in our mice.

It is unclear to us why CREB phosphorylation was reduced in the livers of our exercised mice. To our knowledge, no previous studies have investigated the effect of exercise training on liver CREB expression, although one study did find that during fasting conditions CREB regulates hepatic gluconeogenesis through effects on PGC-1α (Herzig et al. 2001).

Exercise is known to minimize weight gain, and over the course of our 6 week study exercise was associated with weight stabilization. Exercise is also said to enhance insulin sensitivity, but our HOMA-IR calculation showed no evidence of increased insulin sensitivity. We feel it is possible that an exercise-induced increase in gluconeogenesis capacity may have confounded the HOMA-IR calculation. If the livers of the exercised mice were more efficient at producing glucose, a compensatory increase in insulin secretion could have been triggered. Such a change would tend to obscure shifts in insulin sensitivity.

This study was limited by a relatively small sample size. Our results do show changes in the expression, level or post-translational modification of several liver and brain proteins, but the power of our study to detect changes in either tissue was no doubt limited. Future studies with larger sample sizes could more sensitively define how exercise impacts non-muscle bioenergetics and better reveal the mechanisms that mediate exercise-induced molecular changes. Despite this limitation, our study does have translational implications, because it suggests that even a moderate-intensity exercise programme may enhance liver gluconeogenesis and reduce brain inflammation.

In conclusion, our results demonstrate that training C57BL/6 mice at moderate exercise intensities enhances the ability of the liver to import lactate. This effect is associated with and potentially mediated by an increase in liver MCT2. Moderate exercise shifts the bioenergetic profile of the liver in ways that should promote gluconeogenesis and some aspects of mitochondrial biogenesis, but not oxidative phosphorylation. These effects are possibly lactate mediated. Liver lactate import appears to be favoured over brain lactate import, so exercise-mediated brain effects may be robust only when lactate production reliably exceeds the lactate threshold. Further research is needed to investigate whether exercise training above the lactate threshold increases brain lactate delivery and alters energy-sensitive pathways and proteins in that tissue.

Acknowledgments

This work was supported by the Mitochondrial Genomics and Metabolism Core of the University of Kansas Alzheimer’s Disease Center (P30AG035982), the Morgan Family Foundation, the Frank and Evangeline Thompson Alzheimer’s Treatment Development Fund and the Mabel A. Woodyard Fellowship Fund.

References

  1. Aguiar AS, Jr, Tuon T, Pinho CA, Silva LA, Andreazza AC, Kapczinski F, Quevedo J, Streck EL, Pinho RA. Mitochondrial IV complex and brain neurothrophic derived factor responses of mice brain cortex after downhill training. Neurosci Lett. 2007;426:171–174. doi: 10.1016/j.neulet.2007.08.058. [DOI] [PubMed] [Google Scholar]
  2. Aguiar AS, Jr, Tuon T, Pinho CA, Silva LA, Andreazza AC, Kapczinski F, Quevedo J, Streck EL, Pinho RA. Intense exercise induces mitochondrial dysfunction in mice brain. Neurochem Res. 2008;33:51–58. doi: 10.1007/s11064-007-9406-x. [DOI] [PubMed] [Google Scholar]
  3. Berchtold NC, Castello N, Cotman CW. Exercise and time-dependent benefits to learning and memory. Neuroscience. 2010;167:588–597. doi: 10.1016/j.neuroscience.2010.02.050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Berthet C, Lei H, Thevenet J, Gruetter R, Magistretti PJ, Hirt L. Neuroprotective role of lactate after cerebral ischemia. J Cereb Blood Flow Metab. 2009;29:1780–1789. doi: 10.1038/jcbfm.2009.97. [DOI] [PubMed] [Google Scholar]
  5. Billat VL, Mouisel E, Roblot N, Melki J. Inter- and intrastrain variation in mouse critical running speed. J Appl Physiol. 2005;98:1258–1263. doi: 10.1152/japplphysiol.00991.2004. [DOI] [PubMed] [Google Scholar]
  6. Bonen A, Heynen M, Hatta H. Distribution of monocarboxylate transporters MCT1-MCT8 in rat tissues and human skeletal muscle. Appl Physiol Nutr Metab. 2006;31:31–39. doi: 10.1139/h05-002. [DOI] [PubMed] [Google Scholar]
  7. Camacho RC, Donahue EP, James FD, Berglund ED, Wasserman DH. Energy state of the liver during short-term and exhaustive exercise in C57BL/6J mice. Am J Physiol Endocrinol Metab. 2006;290:E405–E408. doi: 10.1152/ajpendo.00385.2005. [DOI] [PubMed] [Google Scholar]
  8. Cantó C, Gerhart-Hines Z, Feige JN, Lagouge M, Noriega L, Milne JC, Elliott PJ, Puigserver P, Auwerx J. AMPK regulates energy expenditure by modulating NAD+ metabolism and SIRT1 activity. Nature. 2009;458:1056–1060. doi: 10.1038/nature07813. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cao W, Collins QF, Becker TC, Robidoux J, Lupo EG, Jr, Xiong Y, Daniel KW, Floering L, Collins S. p38 Mitogen-activated protein kinase plays a stimulatory role in hepatic gluconeogenesis. J Biol Chem. 2005;280:42731–42737. doi: 10.1074/jbc.M506223200. [DOI] [PubMed] [Google Scholar]
  10. Chen MJ, Russo-Neustadt AA. Running exercise-induced up-regulation of hippocampal brain-derived neurotrophic factor is CREB-dependent. Hippocampus. 2009;19:962–972. doi: 10.1002/hipo.20579. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cunningham JT, Rodgers JT, Arlow DH, Vazquez F, Mootha VK, Puigserver P. mTOR controls mitochondrial oxidative function through a YY1–PGC-1α transcriptional complex. Nature. 2007;450:736–740. doi: 10.1038/nature06322. [DOI] [PubMed] [Google Scholar]
  12. Devi SA, Kiran TR. Regional responses in antioxidant system to exercise training and dietary vitamin E in aging rat brain. Neurobiol Aging. 2004;25:501–508. doi: 10.1016/S0197-4580(03)00112-X. [DOI] [PubMed] [Google Scholar]
  13. Fabel K, Tam B, Kaufer D, Baiker A, Simmons N, Kuo CJ, Palmer TD. VEGF is necessary for exercise-induced adult hippocampal neurogenesis. Eur J Neurosci. 2003;18:2803–2812. doi: 10.1111/j.1460-9568.2003.03041.x. [DOI] [PubMed] [Google Scholar]
  14. Finkbeiner S. Calcium regulation of the brain-derived neurotrophic factor gene. Cell Mol Life Sci. 2000;57:394–401. doi: 10.1007/PL00000701. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Ghosh S, Patel N, Rahn D, McAllister J, Sadeghi S, Horwitz G, Berry D, Wang KX, Swerdlow RH. The thiazolidinedione pioglitazone alters mitochondrial function in human neuron-like cells. Mol Pharmacol. 2007;71:1695–1702. doi: 10.1124/mol.106.033845. [DOI] [PubMed] [Google Scholar]
  16. Hardie DG. AMP-activated/SNF1 protein kinases: conserved guardians of cellular energy. Nat Rev Mol Cell Biol. 2007;8:774–785. doi: 10.1038/nrm2249. [DOI] [PubMed] [Google Scholar]
  17. Hawley SA, Davison M, Woods A, Davies SP, Beri RK, Carling D, Hardie DG. Characterization of the AMP-activated protein kinase kinase from rat liver and identification of threonine 172 as the major site at which it phosphorylates AMP-activated protein kinase. J Biol Chem. 1996;271:27879–27887. doi: 10.1074/jbc.271.44.27879. [DOI] [PubMed] [Google Scholar]
  18. Herzig S, Long F, Jhala US, Hedrick S, Quinn R, Bauer A, Rudolph D, Schutz G, Yoon C, Puigserver P, Spiegelman B, Montminy M. CREB regulates hepatic gluconeogenesis through the coactivator PGC-1. Nature. 2001;413:179–183. doi: 10.1038/35093131. [DOI] [PubMed] [Google Scholar]
  19. Hoene M, Lehmann R, Hennige AM, Pohl AK, Haring HU, Schleicher ED, Weigert C. Acute regulation of metabolic genes and insulin receptor substrates in the liver of mice by one single bout of treadmill exercise. J Physiol. 2009;587:241–252. doi: 10.1113/jphysiol.2008.160275. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hoene M, Weigert C. The stress response of the liver to physical exercise. Exerc Immunol Rev. 2010;16:163–183. [PubMed] [Google Scholar]
  21. Ide K, Schmalbruch IK, Quistorff B, Horn A, Secher NH. Lactate, glucose and O2 uptake in human brain during recovery from maximal exercise. J Physiol. 2000;522:159–164. doi: 10.1111/j.1469-7793.2000.t01-2-00159.xm. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Jackson VN, Price NT, Carpenter L, Halestrap AP. Cloning of the monocarboxylate transporter isoform MCT2 from rat testis provides evidence that expression in tissues is species-specific and may involve post-transcriptional regulation. Biochem J. 1997;324:447–453. doi: 10.1042/bj3240447. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jia B, Wang X, Kang A, Wen Z, Yao W, Xie L. The effects of long term aerobic exercise on the hemorheology in rats fed with high-fat diet. Clin Hemorheol Microcirc. 2012;51:117–127. doi: 10.3233/CH-2011-1519. [DOI] [PubMed] [Google Scholar]
  24. Little JP, Safdar A, Cermak N, Tarnopolsky MA, Gibala MJ. Acute endurance exercise increases the nuclear abundance of PGC-1α in trained human skeletal muscle. Am J Physiol Regul Integr Comp Physiol. 2010;298:R912–R917. doi: 10.1152/ajpregu.00409.2009. [DOI] [PubMed] [Google Scholar]
  25. Liu YF, Chen HI, Wu CL, Kuo YM, Yu L, Huang AM, Wu FS, Chuang JI, Jen CJ. Differential effects of treadmill running and wheel running on spatial or aversive learning and memory: roles of amygdalar brain-derived neurotrophic factor and synaptotagmin I. J Physiol. 2009;587:3221–3231. doi: 10.1113/jphysiol.2009.173088. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28:412–419. doi: 10.1007/BF00280883. [DOI] [PubMed] [Google Scholar]
  27. Navarro A, Gomez C, López-Cepero JM, Boveris A. Beneficial effects of moderate exercise on mice aging: survival, behavior, oxidative stress, and mitochondrial electron transfer. Am J Physiol Regul Integr Comp Physiol. 2004;286:R505–R511. doi: 10.1152/ajpregu.00208.2003. [DOI] [PubMed] [Google Scholar]
  28. Pellerin L, Bouzier-Sore AK, Aubert A, Serres S, Merle M, Costalat R, Magistretti PJ. Activity-dependent regulation of energy metabolism by astrocytes: an update. Glia. 2007;55:1251–1262. doi: 10.1002/glia.20528. [DOI] [PubMed] [Google Scholar]
  29. Rasmussen P, Brassard P, Adser H, Pedersen MV, Leick L, Hart E, Secher NH, Pedersen BK, Pilegaard H. Evidence for a release of brain-derived neurotrophic factor from the brain during exercise. Exp Physiol. 2009;94:1062–1069. doi: 10.1113/expphysiol.2009.048512. [DOI] [PubMed] [Google Scholar]
  30. Richter EA, Turcotte L, Hespel P, Kiens B. Metabolic responses to exercise. Effects of endurance training and implications for diabetes. Diabetes Care. 1992;15:1767–1776. doi: 10.2337/diacare.15.11.1767. [DOI] [PubMed] [Google Scholar]
  31. Rockl KS, Witczak CA, Goodyear LJ. Signaling mechanisms in skeletal muscle: acute responses and chronic adaptations to exercise. IUBMB Life. 2008;60:145–153. doi: 10.1002/iub.21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Salminen A, Hyttinen JM, Kaarniranta K. AMP-activated protein kinase inhibits NF-κB signaling and inflammation: impact on healthspan and lifespan. J Mol Med (Berl) 2011;89:667–676. doi: 10.1007/s00109-011-0748-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Somani SM, Husain K. Exercise training alters kinetics of antioxidant enzymes in rat tissues. Biochem Mol Biol Int. 1996;38:587–595. [PubMed] [Google Scholar]
  34. Steiner JL, Murphy EA, McClellan JL, Carmichael MD, Davis JM. Exercise training increases mitochondrial biogenesis in the brain. J Appl Physiol. 2011;111:1066–1071. doi: 10.1152/japplphysiol.00343.2011. [DOI] [PubMed] [Google Scholar]
  35. Stewart KJ. Role of exercise training on cardiovascular disease in persons who have type 2 diabetes and hypertension. Cardiol Clin. 2004;22:569–586. doi: 10.1016/j.ccl.2004.06.007. [DOI] [PubMed] [Google Scholar]
  36. Suzuki A, Stern SA, Bozdagi O, Huntley GW, Walker RH, Magistretti PJ, Alberini CM. Astrocyte-neuron lactate transport is required for long-term memory formation. Cell. 2011;144:810–823. doi: 10.1016/j.cell.2011.02.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. van Hall G, Strømstad M, Rasmussen P, Jans O, Zaar M, Gam C, Quistorff B, Secher NH, Nielsen HB. Blood lactate is an important energy source for the human brain. J Cereb Blood Flow Metab. 2009;29:1121–1129. doi: 10.1038/jcbfm.2009.35. [DOI] [PubMed] [Google Scholar]
  38. Vaynman S, Ying Z, Gómez-Pinilla F. Exercise induces BDNF and synapsin I to specific hippocampal subfields. J Neurosci Res. 2004;76:356–362. doi: 10.1002/jnr.20077. [DOI] [PubMed] [Google Scholar]
  39. Veneman T, Mitrakou A, Mokan M, Cryer P, Gerich J. Effect of hyperketonemia and hyperlacticacidemia on symptoms, cognitive dysfunction, and counterregulatory hormone responses during hypoglycemia in normal humans. Diabetes. 1994;43:1311–1317. doi: 10.2337/diab.43.11.1311. [DOI] [PubMed] [Google Scholar]
  40. Viña J, Gomez-Cabrera MC, Borras C, Froio T, Sanchis-Gomar F, Martinez-Bello VE, Pallardo FV. Mitochondrial biogenesis in exercise and in ageing. Adv Drug Deliv Rev. 2009;61:1369–1374. doi: 10.1016/j.addr.2009.06.006. [DOI] [PubMed] [Google Scholar]
  41. Wilson DO, Johnson P. Exercise modulates antioxidant enzyme gene expression in rat myocardium and liver. J Appl Physiol. 2000;88:1791–1796. doi: 10.1152/jappl.2000.88.5.1791. [DOI] [PubMed] [Google Scholar]
  42. Wu SY, Wang TF, Yu L, Jen CJ, Chuang JI, Wu FS, Wu CW, Kuo YM. Running exercise protects the substantia nigra dopaminergic neurons against inflammation-induced degeneration via the activation of BDNF signaling pathway. Brain Behav Immun. 2011;25:135–146. doi: 10.1016/j.bbi.2010.09.006. [DOI] [PubMed] [Google Scholar]

RESOURCES