| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Integrative Biology, University of California, Berkeley, California, USA;
Center for Comparative Respiratory Biology and Medicine, University of California, Davis, California, USA
1Correspondence: Department of Integrative Biology, 5101 VLSB, University of California, Berkeley, CA 94720-3140 USA. E-mail: gbrooks{at}berkeley.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
coactivator-1
(PGC1) expression and the DNA binding activity of nuclear respiratory factor (NRF)-2. We conclude that the lactate signaling cascade involves ROS production and converges on transcription factors affecting mitochondrial biogenesis.—Hashimoto, T., Hussien, R., Oommen, S., Gohil, K., Brooks, G. A. Lactate sensitive transcription factor network in L6 cells: activation of MCT1 and mitochondrial biogenesis.
Key Words: muscle • lactate shuttle • lactate oxidation complex • exercise • cell signaling
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
[at a physiological pH, lactic acid (pK 3.8) is almost fully dissociated to protons and lactate anions]. High rates of lactate production in working muscle serves to regenerate NAD+, allowing glycolytic energy production to progress rapidly. As part of a shuttle mechanism, lactate is either used within cells of formation or exported to adjacent and anatomically distributed cells, tissues, and organs for utilization (1
, 3)
. Lactate and pyruvate are exchanged across lipid bilayer membranes by facilitated, proton-linked transport (4
, 5)
involving a family of monocarboxylate transport (MCT) proteins (6)
. MCT1 is widely expressed in different tissues (7)
and has been localized in muscle to sarcolemmal and mitochondrial membranes (8
9
10
11
12
13)
. As part of the Lactate Shuttle mechanism, MCT1 facilitates uptake of lactate into working human skeletal muscle from interstitium and plasma (14
, 15)
. Using combinations of immunohistochemistry, confocal laser scanning microscopy, immunoprecipitation, and cell fraction techniques, with adult rat plantaris (11)
and cultured rat-derived cells (10)
we demonstrated colocalization of MCT1, LDH, CD147 (Basigin), and cytochrome c oxidase (COX) and described the presence of a previously unrecognized mitochondrial lactate oxidation complex. We and others have shown that exercise training and muscle contraction increase the expression of lactate transporter MCT1 in mammalian skeletal muscle (14
, 16
17
18)
. However, the precise mechanisms to up-regulate MCT1 expression are unclear.
In preliminary unpublished studies two observations were noteworthy. First the incubation of L6 cells with hydrogen peroxide (H2O2) or Ca2+ increased MCT expression severalfold. Second, time-dependent increases in medium [lactate] and cell MCT1 expression occurred with glucose as the only exogenous substrate in cells not treated with H2O2 or Ca2+. Therefore, based on our analysis of the promoter region of MCT1 gene (Supplemental Fig. S1), we posited that lactate may be a trigger to activate ROS sensing transcription factors such as cAMP-response element-binding protein (CREB), nuclear factor-kappaB (NF-
B), activated protein-1 (AP-1), stimulating protein-1 (SP-1), or nuclear factor erythroid 2 (NF-E2, or Nrf2) (19
, 20)
. Furthermore, it was reasonable to posit that physical activity results in lactate and ROS formation whose signals are transduced to generate changes in MCT1 expression as well as other components of the mitochondrial lactate oxidation complex. Beyond our initial hypothesis, GeneChip analysis revealed the presence of hundreds of lactate-responsive genes that collectively may represent a transcriptional network involved in muscle adaptation to exercise stress.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Myotubes were incubated in DMEM (high glucose) containing 0, 10, or 20 mM sodium lactate (Sigma, St. Louis, MO, USA) and harvested after 1, 6, 24, and 48 h for protein, and 1- and 6-h for mRNA determinations. The medium was changed twice daily to maintain [lactate] low in all groups for two days prior to experimentation. In the ‘background’ group no lactate was added, but cells were exposed to continuously rising endogenously produced lactate. For the GeneChip analysis, L6 cells were incubated in DMEM containing 0 or 20 mM lactate at the start of the experiment, and harvested at 1 or 6 h. The values of 10 and 20 mM incubation [lactate] were decided on as those levels are seen in human blood and muscle after hard exercise (21
, 22)
.
Glucose and lactate concentration measurements
Glucose (23)
and lactate (24)
in spent culture media were measured by spectrophotometry.
Hydrogen peroxide measurement
Cells were preincubated in DMEM containing the glutathione peroxidase (GPx) inhibitor (10 mM mercaptosuccinic acid) and the catalase inhibitor (30 mM 3-amino-1, 2, 4-triazole) for 30 min (25)
followed by additional incubation with 0 or 20 mM lactate at the start, and harvested at 90 s, 3, 5, 10, 15, or 30 min. After harvesting with PBS, cells were homogenized in PBS buffer with 1% triton X-100, centrifuged at 12,000 g for 10 min at 4°C to precipitate particulates, and H2O2 was assayed spectrophotometrically using a kit (BioAssay Systems, Hayward, CA, USA).
Immunoblots
Cell homogenates were separated from nuclear, sarcolemmal, and mitochondrial membrane fractions as previously (9
, 10)
. Protein concentration of cell fractions were determined using a BCA protein assay kit (Pierce Biotechnology, Radford, IL, USA), then MCT1, MCT4, CD147, COX, and LDH in cell subfractions were detected by standard Western blotting techniques as described elsewhere (9
, 13)
. Primary antibodies used in this study were previously described (10)
. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Abcam, Cambridge, MA, USA) was used as a loading control except for mitochondrial fractions. Representative Western blots showing expressions of whole muscle homogenates MCT1 and GAPDH can be seen (Supplemental Fig. S2).
Electrophoretic mobility shift assay (EMSA)
Nuclear and cytoplasmic extracts were prepared using NE-PER extraction reagents (Pierce Biotechnology). Double-strand oligonucleotide containing the sequence binding sites for NRF-1, NRF-2, CREB, NF-E2, NF-B, AP-1, and SP-1 (See Table S4) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA), or Promega (Madison, WI, USA); 100 nM of oligo were labeled by a Biotin Labeling Kit (Pierce). EMSA was carried out using a Light Shift Chemiluminescent EMSA kit (Pierce). Equal amounts of nuclear extracts were incubated with 2 µl of 10x binding buffer, 1 µl of (1 µg/µl) poly (dI.dC) and 2 µl (20 fmol) of biotin labeled probe for 30 min at 37°C. To check whether the observed bands were specific, a competitive reaction was run by adding 4 fmol unlabeled oligo to the mix reaction.
The oligonucleotide-DNA complex was separated by 6% TBE gel (Invitrogen, Carlsbad, CA, USA) in 0.5% TBE buffer (160V) then transferred to a positive charge nylon membrane in 0.5% TBE buffer. After the transfer was completed, the membrane cross-linked for 15 min on a UV transilluminator. Biotin-labeled DNA was detected by using a chemiluminescent detection kit (Pierce).
RNA isolation
Total RNA was isolated from L6 cells (RiboPure kit, Ambion, Austin, TX, USA); the RNA concentration and purity were determined by spectrophotometry and Experion (Bio-Rad, Hercules, CA, USA).
Reverse transcriptase-PCR (RT-PCR)
RNA (2 µg) in 20 µl was reverse-transcribed to single-strand cDNA by using a RETROscript kit (Ambion); 2 µl of cDNA were amplified using GoTag DNA polymerase (Promega) on thermal cycler (PTC-100 Thermal Cycler, MJ Research, Hercules, CA, USA) with specific primers pair (see Supplemental Table S5) for rat MCT1, PGC1, and GAPDH. The samples were amplified for 35–40 cycles after an initial denaturation at 90°C for 2 min, using the following PCR cycle conditions: 94, 55, and 72°C for 30 s each. Linearity of RT-PCR was tested by amplification of 2 µl of RT reaction from 25 to 40 cycles to establish optimal conditions. PCR products were separated by 1% agarose gel stained with ethidium bromide and visualized by UV (Gel Doc EQ, Bio-Rad).
GeneChip analysis
Pooled samples from four or more incubation dishes were used. Total RNA (20 µg) was reverse-transcribed using the Affymetrix protocol. Fragmented, biotin-labeled cRNA samples were hybridized to RAT 230A GeneChips. Hybridizations, washing, labeling, and scanning of microarrays were performed as described in Affymetrix protocols, and data were analyzed using the GeneChip Operating System (GCOS) 1.3. In the GeneChip assay there were a total of 11 pairs of oligonucleotide probes for the detection and quantification of each mRNA. Each pair has an oligonucleotide probe for specific binding and non-specific binding. Net binding intensities for each mRNA are obtained from each pair of probes resulting in a total of 11 intensities for each mRNA. Mean signal intensity for each mRNA, its SD, and P-value were calculated from these data. When the P-value for detection signal was <0.04 the expression of the mRNA was classified as present, and all mRNAs with the P-value for detection >0.06 were considered absent. The upper-limit P-value for "reliable" detection of an mRNA was 0.04. Lists of lactate sensitive genes were obtained by using "batch analysis" function in GCOS and sorted to satisfy three requirements: 1) the mRNA was detectable in at least one of the two samples being compared; 2) the fold-change was 
2; and 3) the mRNA was annotated.
Real-time PCR
Quantitative PCR was performed using iQ5 System (Bio-Rad). Reactions were performed using an iScript One-Step kit with SYBR green. Thermocycling conditions were: 10 min 50°C, 5 min 95°C, followed by 40 cycles of 95°C for 10 s, and 55°C for 30 s and 72°C for 30 s with each reaction in triplicate. Each specific amplification was verified by the presence of a single melting temperature peak. Samples were normalized using 18S rRNA as an internal standard. Primer sequences used are listed in Supplemental Table S5.
Statistical analyses
Data are represented as means ± SEM. Significance of differences among treatment groups were determined by ANOVA and Fisher’s LSD tests (P<0.05). Time-dependent differences were determined by Bonferroni posthoc tests. Relative expression values of real-time PCR were compared using REST 2005 and –384 (Corbett Ltd, Sydney, Australia).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
=
–7.5 mM/h from start to 1 h and –1.3 mM/h from 1 h to 48 h), indicating that glucose consumption was similar across groups (Fig. 1
A). Lactate concentration gradually increased (=0.3 mM/h) to 10 mM when the starting medium contained high glucose, but no lactate (background group, Fig. 1B
). Lactate concentration in the background group became significantly higher than in 0 mM lactate group in which media were changed. Lactate levels in background and 0 mM lactate groups were lower than in the 10 mM lactate group in which [lactate] rose continuously. In the 20 mM lactate group [Lactate] decreased initially (=–2.4 mM/h), suggesting lactate oxidation. After 6 h, medium [lactate] increased in the 20 mM group. Even though aerobic glycolysis is a feature of L6 cell metabolism and [lactate] rose continuously in the 0 mM group, by repeatedly changing the medium we did achieve a condition in which the [lactate] was significantly lower than in other conditions (Fig. 1B
).
|
Lactate activates H2O2 production
To examine whether lactate has the potential to activate H2O2 production, cells were incubated in the presence of mercaptosuccinic acid and 3-amino-1, 2, 4-triazol, GPx and catalase inhibitors. With and without exogenous lactate, peroxide production was evident within 15 min of incubation. But, with the presence of 20 mM exogenous lactate, peroxide production [80 µM] was significantly higher than without added lactate [60 µM] (Fig. 2
), indicating that exogenous lactate rapidly activates H2O2 production in vitro. In previous reports on working muscle (26
, 27)
, investigators attributed ROS generation to mitochondrial electron transport. A limitation to our study is that we were technically unable to measure cell oxygen consumption, and so assumed that as during exercise in vivo (28)
, exogenous lactate did not change tissue oxygen consumption rate. However, we cannot exclude the possibility that mitochondrial respiration as well as lactate accumulation were sources of ROS generation during cell lactate incubation.
|
Elevated lactate concentration increased MCT1 mRNA expression
Exogenous lactate increased MCT1 mRNA expression in L6 cells within 1 h of treatment with 20 mM lactate (Fig 3
A). (At 6 h, MCT1 mRNA expression was greater with either lactate treatment than in the 0 mM lactate group (Fig. 3B
).
|
Electrophoretic mobility shift assays (EMSA) for DNA binding activities
In order to examine whether the increased MCT1 gene expression in L6 cells after lactate incubation involves activation of ROS-responsive transcription factors, EMSA was performed after 10, 30 min, 1 h, 3 h, 6 h, and 24 h of 20 mM lactate treatment. The time course of increased NF-B DNA binding activity varied but was detected from 10 min throughout 3 h of incubation (Fig. 4
A). As well, incubation of L6 cells with exogenous lactate increased NF-E2 DNA binding activity (Fig. 4B
). However, lactate treatment did not increase AP-1 DNA binding activity (Fig. 4C
). Similarly, lactate treatment did not increase SP-1 DNA binding activity during the experiment (data not shown). Hence, it appears that NF-B and NF-E2 DNA binding are responsive to lactate-induced oxidative stresses.
|
Effects of exogenous lactate on components of the mitochondrial lactate oxidation complex
MCT1: After 1 h of incubation in 10 mM or 20 mM lactate, MCT1 protein expression in cell homogenates increased significantly (1.7- and 2.0-fold) as compared to controls (Table 1
A). In mitochondrial fractions, similar significant increases in MCT1 protein expression were seen at 1 h in 10 mM (1.6-fold) and 20 mM (1.9-fold) (Table 1B
). MCT1 protein levels in sarcolemmal fractions were unchanged except at 24 h incubation with 10 mM (Supplemental Table S1). Thus, the cell homogenate response was primarily mitochondrial. No changes in MCT4 protein expression were seen in whole muscle homogenates (Table 1A
).
|
CD147 (Basigin): Cell homogenate CD147 protein levels increased 1.9- and 2.8-fold in cells treated with 10 mM and 20 mM lactate groups (Table 1A
). After 48 h incubation mitochondrial CD147 protein levels increased to a greater extent following incubation with 20 mM than with 10 mM lactate (Table 1B
). In sarcolemmal fractions, CD147 protein levels were increased after 24 h incubation in 20 mM lactate and 48 h incubation in 10 mM lactate (Supplemental Table S1). However, after 48 h incubation in 20 mM lactate, CD147 protein was no longer elevated (Supplemental Table S1).
Cytochrome c oxidase (COX): Cell homogenate COX protein expression, our surrogate for mass of the mitochondrial reticulum, increased 2.0–2.5 fold after 6 h cell incubation in 10 or 20 mM lactate (Table 1A
). After 48 h lactate treatment, cell homogenate COX protein level was elevated by 20, but not 10 mM lactate treatment (Table 1A
). COX protein expression in mitochondrial fractions was increased in both 10 mM and 20 mM lactate groups at 48 h (Table 1B
).
Lactate dehydrogenase (LDH): The LDH protein expression was increased in whole muscle homogenates when cells were incubated in 20 mM lactate for 6 h. At 48 h, homogenate LDH expression was higher after incubation with 20 mM compared with 10 mM lactate (Table 1A
). In mitochondrial fractions, LDH protein was significantly elevated after 24 h incubation with 20 mM lactate (Table 1B
). There were no significant increases in expression of any lactate oxidation complex protein when comparing control and background groups (data not shown).
Mitochondrial biogenesis after the incubation with exogenous lactate
As lactate incubation increased cell COX protein expression, we examined the effects of lactate on known regulators of mitochondrial biogenesis. Peroxisome proliferator activated-receptor coactivator-1 (PGC1), a master coordinator of mitochondrial biogenesis (29)
, has been shown to interact with transcription factors for mitochondrial gene expression including those for COX such as CREB, NRF-1, and NRF-2 (30)
. Cell PGC1 mRNA expression was increased after 6 h incubation with either 10 or 20 mM lactate (Fig. 3D
). These changes corresponded to increments in COX protein expression. Additionally, we determined the effect of cell incubation with 20 mM lactate on NRF-1, NRF-2, and CREB binding to DNA (Fig. 5
). DNA binding activity of NRF-1 showed no change after incubation with 20 mM lactate (Fig. 5A
). On the other hand, the DNA binding activity of NRF-2 was increased after 1 h incubation with 20 mM lactate (Fig. 5B
). Further, CREB binding to DNA increased within 30 min of incubation in 20 mM lactate (Fig. 5C
). To further explore the roles of NRF-1 and NRF-2, we performed RT-PCR and real-time PCR to assess the effects of lactate on NRF-1 and -2 mRNA expression, but we observed no changes in response to either 10 or 20 mM lactate over 6 h (Table 2
). The findings suggest that lactate exposure affects transcriptional regulation of mitochondrial biogenesis by binding of NRF-2 and CREB to DNA.
|
|
GeneChip analysis reveals a lactate transcriptome
Our initial hypothesis was that elevated lactate would stimulate transcriptional regulation of the lactate transporter MCT1 and other components of the mitochondrial lactate oxidation complex. Our results (vide supra) were consistent with the hypothesis, but we were concerned that it might have been too simplistic and that we might have missed a broader range of systems-related responses to lactate anion. Therefore, we performed microarray analysis using Affymetrix RAT 230A GeneChips.
Incubation of L6 cells with 20 mM lactate significantly increased the expression of 33 genes and decreased the expression of 76 genes at 1 h, but incubation of L6 cells in 20 mM lactate for 6 h increased the expression of 673 genes while the expression of only 3 genes decreased. We classified genes into 7 categories (Supplemental Tables S2, S3). At 1 h, genes associated with CHO metabolism and gluconeogenesis (e.g., PEPCK1) were up-regulated, whereas genes associated with lipid metabolism (e.g., VLDL receptor) were down-regulated. At 6 h of incubation, 79 genes involved in metabolism and mitochondria and 31 genes involved in transcriptional activation and signal transduction were up-regulated with only cAMP-dependent protein kinase inhibitor beta down-regulated. Further, large numbers of genes related to transport, oxidative stress, apoptosis, cell growth, and Ca2+ signaling were up-regulated. Among lactate-sensitive genes involved in metabolism and mitochondria were ATP synthase (ATP5g1 and –5o), NADH dehydrogenase (NADH-dh), translocase in the inner membrane (TIM), succinate dehydrogenase (SDH) complex, hydroxyacyl-CoA dehydrogenase (HAD2) as well as (COX4i1, whose expression was validated by real-time RT-PCR together with that of PGC1, NRF1, and NRF2. GeneChip data also showed induction of CREB Ca2+/calmodulin-dependent protein kinase (CaMK), and protein kinase C (PKC) genes. Importantly, in terms of the hypotheses articulated here regarding the involvement of ROS in the response to lactate anion, several genes responsive to oxidant stress were up-regulated, including GPx as well as other genes of interest (MCT1 and CD147) (Table 2
, Supplemental Table S2). Among those genes, only COX and CaMK were significantly up-regulated according to GCOS analysis, and real-time PCR confirmed the increase in the expression of COX and CaMK. Additionally, real-time PCR demonstrated increased expression of MCT1 and CD147 at 1 h and MCT1 as well as PGC1 at 6 h (Table 2)
. The expression of other selected genes is summarized in Supplemental Table S2. Notably, IB, Jun, and GPx1 were significantly increased by lactate incubation at 6 h by GCOS analysis.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B and NF-E2 pathways leading, in turn, to increases in MCT1 expression. And, although incubation with lactate did not increase AP-1 DNA binding activity (Fig. 4)
, microarray analysis showed that lactate incubation increased the expression of Jun, one of the components of AP-1. Mitogen-activated protein kinase (MAPK) pathways are involved in the regulation of Jun, and our analysis of GeneChip data demonstrated that lactate incubation increased the transcription of genes for Ras and CREB that are components of the MAPK pathway (Supplemental Tables S2, S3). As already noted, the DNA binding activity of CREB was increased by lactate within 30 min (Fig. 5C
). Therefore, despite our technical inability to demonstrate increased binding of AP-1 to DNA by EMSA, we should hold open the possibility that this is involved in MCT1 gene transcription.
At present our conclusion about the mechanism of MCT1 induction by lactate is based on increased DNA binding activity of NRF-2 and CREB after lactate treatment, the presence of NRF-2 and CREB binding sites on the MCT1 gene promoter, and the well-known effects of NRF-2 and CREB on mitochondrial biogenesis (vide infra). MCT1 is predominant in slow-twitch oxidative fibers (11)
, as well as is a constituent of the mitochondrial lactate oxidation complex (10)
. Coordination of MCT1 expression and mitochondrial biogenesis by NRF-2 and CREB is likely physiologically relevant for increasing oxidative lactate clearance capacity in skeletal muscle (14
, 15)
.
In addition to increasing MCT1 mRNA expression, we found that incubating L6 cells with high lactate resulted in transient increases in MCT1 protein expression (Table 1A
). Both dose dependence (Fig. 3A
) and time course responses (Fig. 3A
vs. B) were evident. However, counter or alternative regulation was apparent as MCT1 mRNA expression increased after 6 h incubation in both 10 and 20 mM lactate, while protein expression was not changed. Hence, it appears that MCT1 protein expression was not regulated exclusively at the level of transcription. In a previous study (18)
it was observed that MCT1 protein expression was rapidly up-regulated in rat skeletal muscle by a single bout of exercise, but that increase in MCT1 protein was not always accompanied by concomitant changes in transcript level. Similarly, with human colonic epithelial cells it was demonstrated that butyrate, another substrate for MCT1, increased MCT1 mRNA expression by the dual control of MCT1 gene transcription and stability of the MCT1 transcript (31)
. Results of the present investigation are consistent with the idea that the lactate-induced MCT1 protein expression is regulated by both transcriptional and post-transcriptional mechanisms. In contrast to the MCT1, L6 cell homogenate MCT4 protein expression was unchanged after incubation with lactate (Table 1A
). That result was expected because we (14)
, and others (17)
, have shown MCT1 and MCT4 to be regulated independently in humans and rats, respectively.
The single-span transmembrane glycoprotein CD147 (basigin) is considered to be the chaperon protein for MCT1, localizing it to cell surface (32
, 33)
as well as mitochondria inner membrane (10)
, and it has been postulated that CD147 and MCT1 are coregulated (34)
. In the present study, we found increased CD147 and MCT1 protein contents in whole muscle homogenates after 1 h incubation with lactate. Similarly, lactate incubation increased both MCT1 and CD147 transcript levels within an hour. Therefore, at the tissue level, CD147 transcript and protein levels coincided. However, consistent with results of Philip et al. who studied retinas of CD147 null mice, CD147 and MCT1 levels in specific cell domains did not necessarily track levels of mRNA expressed (34)
.
In the mitochondrial fraction of L6 cells, we found a tendency of increased MCT1 insertion at 1 h, but CD147 did not increase, although it was abundant in mitochondria. At 6 h, mitochondrial MCT1 and CD147 levels did not change; however, considering the increased mitochondrial mass represented by increased COX expression in the whole muscle homogenate (Table 1A
), lactate treatment up-regulated the abundance of mitochondrial MCT1 and CD147. Lactate treatment increased mitochondrial CD147 levels 48 h as well, but this was not accompanied by increased insertion of MCT1. Overall, our results are consistent with the idea that CD147 and MCT1 genes are differentially regulated, as are the insertions of scaffold and transporter proteins into specific cell domains. In this regard we note that analysis of GeneChip data revealed that 20 mM lactate increased TIM 13 and TIM 17 at 6 h. Mechanisms of the coordination of TOM/TIM complexes with MCT1 and its accessory protein CD147 remain to be elucidated.
While addressing the expression of other components of the mitochondrial lactate oxidation complex, we found that incubation with high lactate increased COX protein expression in the whole muscle homogenate (Table 1A
). Additionally, microarray analysis revealed that COX-IV gene expression was increased by 20 mM lactate at 6 h, and this was confirmed by real-time PCR. Although the DNA binding activity of NRF-1 was not changed in L6 cells incubation in 20 mM lactate, NRF-2 DNA binding activity increased. NRF-2, as opposed to NRF-1, may be the more important coactivator of COX subunit IV gene because NRF-2 is considered to be a master transcriptional regulator for all subunits of nuclear-encoded COX, with no binding site for NRF-1 in the promoter sequence of COX-IV of rat neurons (35)
. Also, although PGC1 mRNA expression was increased by 20 mM lactate at 6 h, the initial adaptive increase in mitochondrial biogenesis was regulated by activated transcription factors (e.g., NRF-2), which are coactivated by translocation of PGC1 into the nucleus (36)
. Therefore, both increased expression of PGC1 and the increase in the DNA binding activity of NRF-2 by lactate would result in the increase in the expression of COX protein. The demonstration of an effect of lactate incubation on COX expression is important because it is an integral component of the mitochondrial ETC and the lactate oxidation complex (10)
.
Finally, with regard to the effects of lactate incubation on components of the mitochondrial lactate oxidation complex, we found that mitochondrial LDH increased after 24 h incubation in 20 mM lactate (Table 1B
). Our findings can be interpreted to mean that elevated concentrations of lactate as occur in contracting muscle up-regulate the total mitochondrial mass and abundance of the lactate oxidation complex. Such an adaptation would facilitate lactate oxidation in skeletal muscle cells permitting high power outputs and glycolytic fluxes to occur while minimizing acidosis. Previously Mootha et al. (37)
and Taylor et al. (38)
performed proteomic surveys of mitochondria from mouse brain, heart, kidney, and liver, and human heart. These investigators provided evidence for more than 600 mitochondrial or mitochondria-associated proteins including components of mitochondrial lactate oxidation complex (i.e., MCT1, CD147, COX, and LDH). Here we expand their demonstration and provide a significance of how mitochondrial constituents are physiologically coordinated by lactate in the L6 cells, and likely in adult mammalian skeletal muscle (11)
.
Beyond our hypothesis-specific observations related to the expression of genes coding for COX-IV and other components of the mitochondrial lactate oxidation complex, GeneChip analyses demonstrated that cell incubation in high lactate increased the expression of large numbers of other genes related to mitochondrial biogenesis; they encode components of electron transport chain (e.g., NADH-dh and SDH complex, COX-VII, ATP synthase), enzymes of ß-oxidation (e.g., HAD), and import machinery (e.g., TIM). Up to now, it has been reported that exercise increases the expression of PGC1 by two pathways: AMP-activated protein kinase (AMPK) and calcium signaling pathways (30
, 39
40
41)
. However, those pathways cannot explain completely the increased PGC1 expression induced by exercise (41
, 42)
. Here, we demonstrate for the first time that metabolic signals induced by lactate exposure can stimulate the transcription of genes involved in mitochondrial biogenesis. With regard to the MAPK signaling pathway, the transcription factor CREB is considered to be important for PGC1 gene expression (30
, 43)
, and we found 20 mM lactate increased DNA binding activity of CREB. Further, GeneChip analysis showed increased CREB and Ras transcription at 6 h. Both CREB and Ras are involved in the MAPK signaling pathway (44)
. Recently, it was reported that ROS, especially H2O2 was involved in the regulation of gene expression in skeletal muscle cells including PGC1 (45)
. Similarly, it was demonstrated that H2O2 treatment increased PGC1 through increased CREB transcription activity in a mouse fibroblastic cell line (43)
. Therefore, our results support the conclusion that in vivo lactate formation from glycolysis is a physiological mechanism for ROS production and the subsequent changes in mitochondrial biogenesis.
In addition to effects on MCT1 expression and mitochondrial biogenesis via the MAPK pathway, lactate-induced H2O2 production has a potential to increase intracellular calcium concentration and thereby activate CaMK (46)
. In the present study, 6 h incubation of L6 cells with 20 mM lactate increased the expression of CaMKI and PKC potentially increasing the ability of cells to respond to calcium. Additionally, lactate increased gene expression of slow isoform of troponin I, which is mainly expressed in high oxidative, slow-twitch muscle fibers. The transcription of troponin I can be activated by calcium/calmodulin-activated serine-threonine phosphatase calcineurin and myogenin, which are essential for muscle development and may work downstream of calcineurin and may be involved in the regulatory process of increasing muscle oxidative capacity (44)
. Taken together, our results suggesting a possibility that lactate may increase cytosolic calcium signaling, and recent reports of a link between ROS and calcium, lend to a further understanding of the mechanisms involved in mitochondrial biogenesis in response to exercise (47)
.
In terms of recently identified mitochondrial and neighborhood proteins (37)
, we found that genes for 18 mitochondrial neighborhood proteins were up-regulated by lactate (highlighted in Supplemental Table S3). The coordinated up-regulation of large numbers of genes encoding mitochondrial and/or mitochondria-associated proteins by lactate provides a novel explanation of how mitochondrial biogenesis may occur in response to physical exercise.
It is notable also that lactate incubation affected the expression of large numbers of genes, which could be classified as adaptive responses to metabolic stress, for example the response to ROS. Cellular redox is maintained by the glutathione and the thioredoxin systems (48)
. As already noted, cell incubation in 20 mM lactate up-regulated GPx, an event that is considered to be a response to oxidative stress and is mediated through PGC1. Additionally, lactate incubation up-regulated glutaredoxin, peroxiredoxin, and thioredoxin gene expression at 6 h. As well, our microarray analysis also detected the induction of a large number of genes previously shown to be up-regulated after endurance exercise (49)
; they are mitochondrial ribosomal protein, solute carrier family, spermidine N1-acetyl transferase, zinc finger protein, cell division cycle, homeo box, Kruppel-like factor, and peripheral myelin protein 22.
The results reported here showing presence of a lactate-sensitive network were obtained on a memorialized muscle cell line incubated in buffered media. This begs the question of relevance to what happens in vivo. Findings relevant to this issue are that the lactate oxidation complex is evident in both L6 cells (10)
and adult rat skeletal muscle cells (11)
. Additionally, results from our previous studies using primed-continuous [3-13C]lactate infusion, femoral arterial venous concentration difference and blood flow measurements, and muscle biopsies in humans (14
, 15)
clearly show an association between muscle lactate production during exercise and increases in sarcolemmal and mitochondrial MCT1 and LDH expression. Thus, at this time it is reasonable to posit that the lactate-sensitive transcriptome observed on myocytes in vitro is operant in vivo.
Ideas that lactate is produced in muscle as the result of O2 lack (50)
and that lactate accumulation poisons muscles causing fatigue (51)
are traceable to early 20th century studies on noncirculated and inadequately oxygenated frog muscle preparations (52)
. In contrast it is now widely appreciated that the Lactate Shuttle mechanism is an important means for distributing an important energy substrate and the major gluconeogenic precursor (1)
. In fact, intracellular lactate shuttles permit the flux of substrate and reducing equivalents from cytosol to mitochondria (53)
and redox balance in peroxisomal ß-oxidation (13)
. Results from this report show important cell signaling ("lactormone") functions of lactate anion and thus complete the ironic twist in the tale of lactate metabolism. Rather than purely a stress response, a more appropriate view may be that lactate production is part of a stress-strain mechanism during muscle contraction. The stress to maintain ATP homeostasis activates glycolysis and lactate production and accumulation. In turn, the transient rises in [lactate] and mitochondrial O2 consumption induce ROS generation, which activate a transcriptional network signaling adaptive cell responses. Among these are increases in MCT1 protein expression, mitochondrial biogenesis, and up-regulation of antioxidant enzymes and moieties of Ca2+ signaling (Fig. 6
). Taken together, results of the present and previous investigation on the mitochondrial lactate oxidation complex (10)
can be interpreted to mean that elevated lactate flux and concentration signals not only lead to adaptation of pathways of lactate removal, but also signals many of the adaptations in muscle found in response to endurance training.
|
|
ACKNOWLEDGMENTS |
|---|
Received for publication January 22, 2007. Accepted for publication March 1, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  L. Metz, J. Mercier, A. Tremblay, N. Almeras, and D. R. Joanisse Effect of weight loss on lactate transporter expression in skeletal muscle of obese subjects J Appl Physiol, March 1, 2008; 104(3): 633 - 638. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. F. Essop Cardiac metabolic adaptations in response to chronic hypoxia J. Physiol., November 1, 2007; 584(3): 715 - 726. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Thomas, D. Bishop, T. Moore-Morris, and J. Mercier Effects of high-intensity training on MCT1, MCT4, and NBC expressions in rat skeletal muscles: influence of chronic metabolic alkalosis Am J Physiol Endocrinol Metab, October 1, 2007; 293(4): E916 - E922. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||
