Exhausting exercise and tissue-specific expression of monocarboxylate transporters in rainbow trout

Abstract

Transmembrane lactate movements are mediated by monocarboxylate transporters (MCTs), but these proteins have never been characterized in rainbow trout. Our goals were to clone potential trout MCTs, determine tissue distribution, and quantify the effects of exhausting exercise on MCT expression. Such information could prove important to understand the mechanisms underlying the classic “lactate retention” seen in trout white muscle after intense exercise. Four isoforms were identified and partially characterized in rainbow trout: MCT1a, MCT1b, MCT2, and MCT4. MCT1b was the most abundant in heart and red muscle but poorly expressed in the gill and brain where MCT1a and MCT2 were prevalent. MCT expression was strongly stimulated by exhausting exercise in brain (MCT2: +260%) and heart (MCT1a: +90% and MCT1b: +50%), possibly to increase capacity for lactate uptake in these highly oxidative tissues. By contrast, the MCTs of gill, liver, and muscle remained unaffected by exercise. This study provides a possible functional explanation for postexercise “lactate retention” in trout white muscle. Rainbow trout may be unable to release large lactate loads rapidly during recovery because: 1) they only poorly express MCT4, the main lactate exporter found in mammalian glycolytic muscles; 2) the combined expression of all trout MCTs is much lower in white muscle than in any other tissue; and 3) exhausting exercise fails to upregulate white muscle MCT expression. In this tissue, carbohydrates act as an “energy spring” that alternates between explosive power release during intense swimming (glycogen to lactate) and recoil during protracted recovery (slow glycogen resynthesis from local lactate).

lactate plays essential roles as a glycolytic end-product, energy source for oxidative tissues and gluconeogenic substrate (4, 9). Mammals and fish maintain high basal rates of lactate production that are strongly stimulated during strenuous exercise and hypoxia, when aerobic pathways can no longer meet total ATP demand (6, 27, 40, 45). Suitable mechanisms to regulate intertissue lactate trafficking are therefore required to cope with these common stresses and orchestrate recovery. Transmembrane lactate movements are mediated by monocarboxylate transport proteins (MCTs), a family of lactate/proton symporters particularly well characterized in mammals (13, 21). Among the 14 known isoforms, MCT1, MCT2, and MCT4 are the most significant regulators of transmembrane lactate exchange (11, 36). Each isoform has a dominant function and a different distribution: MCT1 is ubiquitous and specializes in lactate uptake by oxidative tissues; MCT2 predominantly controls lactate uptake by liver, brain, and kidney, whereas MCT4 is mainly geared for lactate efflux from glycolytic muscle (11, 12, 21). Partial or complete MCT sequences are now known for a few species of fish (see Fig. 2). However, MCTs have not been characterized in rainbow trout (Oncorhynchus mykiss), even though several studies show that this common model species uses a proton-linked symporter to shuttle lactate between tissues (17, 42).

Fig. 2.

Phylogeny of monocarboxylate transporter proteins (MCT1a and b, MCT2a, b and c, and MCT4a and b) in selected vertebrate species. The tree was obtained through Bayesian inference of phylogeny [MrBayes v.3.2.(5)]. Nodes are supported by >80% posterior probabilities, except where indicated by numerical values or by asterisk (*) when <50%. Arrows indicate the retrieved rainbow trout MCTs in the reconstructed phylogeny. The scale bar (0.1) represents the estimated mean number of amino acid substitutions per site for all species. Protein accession numbers (ENSEMBL) are also indicated.

After exhausting exercise is completed, lactate efflux is ∼10 times slower in trout white muscle than in mammalian muscle (42, 44). This well-known lactate retention capacity of fish white muscle has intrigued biologists for decades (37, 43), mainly because it prolongs recovery from maximal exercise [>12 h in trout versus <1 h in mammals (41)]. Characterizing the expression of (putative) MCT isoforms in rainbow trout and their tissue distribution could prove important to understand the physiological basis for lactate retention. The few mammalian studies reporting the effects of exercise on MCT gene expression show variable responses depending on the type of exercise, tissue, and isoform (1, 3, 7). Knowing how intense swimming alters expression in an athletic species like rainbow trout may explain intertissue lactate movements, especially during recovery. Therefore, the goals of this study were the following: 1) to clone rainbow trout MCTs and to identify which isoforms are expressed in this species, 2) to determine tissue-specific isoform distribution, and 3) to quantify the effects of exhausting exercise on MCT gene expression in key tissues affecting whole animal lactate kinetics: white muscle (main lactate producer), red muscle, heart, brain, and gills (lactate oxidation), as well as liver (gluconeogenesis). We anticipated that MCT expression would not be stimulated by exercise in white muscle because lactate retention implies a limited capacity for boosting lactate efflux. By contrast, MCT expression of oxidative tissues like heart, red muscle, gill and brain should be increased to exploit the higher availability of this convenient fuel in the circulation.

METHODS

Animals.

Male and female rainbow trout, Oncorhynchus mykiss (Walbaum) (323 ± 27 g), were purchased from Linwood Acres Trout Farm (Campbellcroft, Ontario, Canada) and held in a 1,300-liter flow-through tank in dechlorinated, well-oxygenated water at 13°C under a 12:12 h light-dark photoperiod. The animals were acclimated to these conditions for at least 2 wk before experiments. They were fed floating fish pellets (Martin Mills, Elmira, Ontario, Canada) three times a week to satiation. They were randomly assigned to a control group (rest; N = 6) or treatment group (exercise; N = 5). All procedures were approved by the Animal Care Committee of the University of Ottawa and adhered to the guidelines established by the Canadian Council on Animal Care.

Cloning, sequencing, and partial characterization of MCTs.

MCT primers (Zfish-MCT F and Zfish-MCT R) were designed from zebrafish MCT1a, MCT1b, MCT2, and MCT4 consensus sequence (regions 786 to 806 bp and 1332 to 1356 pb) to amplify trout MCTs by PCR. We reasoned that these conserved regions across zebrafish MCTs would likely be conserved in trout MCTs. It was reported previously for mammals that MCT1 is expressed in all tissues, MCT2 mainly in brain and MCT4 mainly in white muscle (11). Therefore, total RNA (from two fish not used in the gene expression experiments) was isolated from white muscle and brain using the TRIzol method (Invitrogen; Carlsbad, CA). The RNA was quantified using NanoDrop 2000 (Thermo Scientific; Wilmington, DE), and the purity and integrity were verified on a 1.5% agarose gel prepared with 1× MOPS buffer. cDNA synthesis was performed using 5 μg of total RNA, 200 U SuperScript II reverse transcriptase, 250 ng random octamers (Invitrogen), and DNase/RNase-free water for a total volume of 20 μl. The thermal profile was as per manufacturer's instructions. After cDNA synthesis, a PCR reaction was performed as follows: 1× PCR buffer, 1.5 mM MgSO₄, 0.2 mM dNTP, 0.2 mM of Zfish-MCT F and Zfish-MCT R each, 2 μl cDNA, 2 U Taq polymerase (Bio Basic; Amherst, NY), and DNase/RNase-free water for a final volume of 50 μl. The thermal profile was started with 2 cycles (94°C/2 min; 63°C/1 min; 72°C/1 min), followed by 35 cycles (94°C/30 s; 63°C/30 s; 72°C/1 min), and a final extension step (72°C/5 min) using either Eppendorf Mastercycler (Hamburg, Germany) or Bio-Rad S1000 thermal cycler (Hercules, CA). The amplicons [565 base pair (bp)] obtained from both brain and white muscle were subcloned into TopoTA vector (Invitrogen) and transformed in chemically competent Escherichia coli cells. Clones (24 from brain and 28 from white muscle) were picked and screened by PCR in a reaction mix containing: 1× PCR buffer, 1.5 mM MgSO₄, 0.2 mM dNTP, 0.2 mM of M13 forward and reverse primers, 2 U Taq polymerase (Bio Basic), and DNase/RNase-free water to a final volume of 50 μl. The thermal profile was as follows: an incubation step (94°C/5 min), followed by 40 cycles (94°C/30 s; 55°C/30 s; 72°C/1 min), and a final extension (72°C/5 min). Plasmids from clones with insert of the expected size (11/24 from brain and 12/28 from white muscle) were extracted using PureLink Miniprep kit (Invitrogen) and sent for sequencing at Genome Quebec (McGill University, Montreal, QC, Canada). The sequences were then identified by Basic Local Alignment Search Tool (BLAST). From the brain samples, 5/11 clones were identified as MCT1b, 4/11 as MCT2, and the remaining 2 were not positive for MCT. From the white muscle samples, 7/12 clones were identified as MCT1a, 4/12 as MCT1b, and 1/12 as MCT2. BLAST identified a sockeye salmon contiguous sequence (accession no. EZ815764) that had 90% nucleotide identity with trout MCT1b. Therefore, a new primer set Sal-MCT (from the salmon sequence) and Trt-MCT (from trout) was designed to amplify the upstream sequence of MCT1b using the white muscle sample. The PCR reaction and thermal profile were as described above for primer set Zfish-MCT F and Zfish-MCT R. A 460-bp amplicon was obtained and subcloned. Plasmids from clones with insert of the expected size (11/25) were extracted and sent for sequencing. Nine clones were identified as MCT1b, and the remaining 2 were not positive for MCT1b.

In an attempt to obtain full MCT sequences, rapid amplification of cDNA ends (RACE) was performed for MCT1a, MCT1b, and MCT2 using FirstChoice RLM-RACE Kit (Ambion) with total RNA extracted from white muscle. Both 5′ and 3′ ends could not be amplified for MCT1a, only the 3′ RACE yielded a positive result for MCT1b and MCT2 (using primer sets 3′Trt-MCT1-outer and -inner, and 3′Trt-MCT2-outer and inner). Nested PCR was performed for RACE according to the manufacturer's instructions with the following thermal profiles: the first run was started with 2 cycles (94°C/2 min, 60°C/1 min, 72°C/1 min), followed by 35 cycles (94°C/30 s, 60°C/30 s, 72°C/1 min). The second run was the same as the first except for the annealing step (65°C/30 s). An array of amplicons (400–800 pb) were subcloned, and 26 clones each for MCT1b and MCT2 were screen by PCR (as described for Zfish-MCT F and Zfish-MCT R). Five clones were identified as MCT1b and 2 clones as MCT2. The 3′ end sequences had 99% nucleotide identity in the overlapping region (82 bp for MCT1b, and 99 bp for MCT2). Given that amplification of complete sequences for MCT1b and MCT2 was unsuccessful, RACE was not repeated for MCT1a because the partial sequences obtained for all 3 MCTs were deemed adequate to design specific primers for MCT expression. We could not clone MCT4 in our experiments; however, we identified a contiguous sequence from rainbow trout (accession no. EZ780500.1; region 405 to 1071 pb) available in GenBank (NCBI) as MCT4 using BLAST.

Sequence alignments and phylogenetic analysis.

The trout partial MCT nucleotide sequences were translated into putative protein sequences and were aligned with known MCT proteins from various species using Clustal X version 2.0 (see Fig. 1). The transmembrane domains depicted for trout MCTs are as previously identified in Halestrap and Price (12). A phylogenetic analysis of the partial trout MCT protein and known MCT proteins from various fish and mammals were generated as previously described (20). Briefly, the phylogeny was generated by Bayesian analysis using a mixed-model approach [MrBayes version 3.2 (5)] (15, 30). The analysis was run for 10⁶ generations on three heated chains and one cold chain for each simultaneous run, sampling at a frequency of 100 and a burn-in of 2500. MCT1 from Saccharomyces cerevisiae was included in the analysis as an outgroup.

Fig. 1.

Comparisons of the amino acid sequences of monocarboxylate transporters (MCT) MCT1a, MCT1b, MCT2, and MCT4 between rainbow trout and other species (zebrafish, cod, rat, mouse and human). Black areas show identical amino acids and grey areas similar amino acids. The transmembrane domains (TMDs) identified in Halestrap and Price (12) are also indicated. Sequences were aligned using ClustalX 2.0 and Multiple Align Show (www.bioinformatics.org/SMS/multi_align.html).

Exhausting exercise.

All the animals were implanted with a single dorsal aorta catheter following the procedure of Haman and Weber (14) to monitor changes in plasma lactate concentration after maximal exercise and for rapid euthanasia (arterial injection of pentobarbital sodium) before tissue sampling for MCT expression. Briefly, fish were fasted for at least 24 h before surgery. They were anesthetized with ethyl-N-aminobenzoate sulfonic acid (MS-222; 60 mg/l) in well-oxygenated water, and their dorsal aorta was cannulated with a PE-50 catheter (Intramedic, Clay-Adams, Sparks, MD). Only animals with a hematocrit >20% after surgery were used in experiments. Catheters were kept patent by flushing with 0.2 ml Cortland saline (46) containing 50 U/ml heparin (Sigma-Aldrich, St. Louis, MO). After surgery, the fish used for exercise experiments were individually transferred to a 110-liter flow-through circular tank supplied with the same quality water as the holding tank. They were first allowed to recover from surgery for at least 24 h and manually chased around the tank until complete exhaustion (when they were no longer responding to touch after ∼5 min of intense exercise). Resting controls and exercised fish were individually placed in a cylindrical respirometry chamber (47 cm long, 19 cm diameter) to monitor changes in blood lactate concentration. Recovering fish were at rest (no water current) as blood lactate levels were monitored. Blood samples were deproteinized with perchloric acid (6% wt/wt) and centrifuged. Supernatants were stored at −20°C, and lactate concentration was measured within 1 wk as previously (27).

The effects of swimming on MCT expression were quantified by comparing the tissues of resting controls with those of exercised fish, 23 h after exhaustion. This delay was selected because a previous study on mammals reported strong effects after this duration (7), and preliminary results on trout showed no effects after 6 h. After the experiments the fish were euthanized by overdose of pentobarbital sodium injected through the catheter (Euthanyl, Abraxis Pharmaceutical Products, Schaumburg, IL). Gills, heart, liver, red muscle, white muscle, and brain were harvested in random order within 5 min (∼2 g of each tissue, except for the heart and brain that were sampled entirely). Tissues were immediately freeze-clamped in liquid N₂. Fifty- to 100-mg subsamples were ground to a fine powder in liquid N₂ using mortar and pestle, homogenized in Qiazol (Qiagen, Hilden, Germany) using a Polytron (Kinematica, Luzern, Switzerland), and stored at −80°C until RNA extraction. Total RNA extraction including a genomic DNA elimination step was performed using RNeasy Plus Universal kit as per manufacturer's instructions (Qiagen). The RNA was quantified, and the purity and integrity were verified as described in the MCT cloning section. cDNA synthesis including a genomic DNA elimination step was performed with QuantiTect Reverse Transcription kit according to manufacturer's instructions using 1 μg of total RNA (Qiagen).

MCT expression by real-time PCR.

Forward and reverse primer sets (Trt-Exp-MCT1a, Trt-Exp-MCT1b, Trt-Exp-MCT2, Trt-Exp-MCT4; Table 1) were designed from rainbow trout MCT sequences. The transcript levels of MCT1a, MCT1b, MCT2, and MCT4 were quantified by real-time PCR using the QuantiFast SYBR Green PCR kit (Qiagen) in a Bio-Rad CFX real-time PCR detection system. The reaction mix contained 1× QuantiFast SYBR Green PCR master mix, 0.2 mM forward and reverse, 1 μl DNase/RNase-free water, and 2 μl of 2.5× diluted cDNA samples (10 μl total reaction mix). All the samples were run in duplicate, and fluorescence was measured at the end of every extension step. Each run included no-RT and no-template controls. Standard curves were constructed for MCT1a, MCT1b, MCT2, MCT4, elongation factor-1α (EF-1α), 18S and β-actin using serial dilutions of stock cDNA to ensure adequate amplification efficiency. Amplification efficiencies were 111% for MCT1a [linear threshold cycle (C_t) range: 23–29 cycles], 113% for MCT1b (21–27 cycles), 109% for MCT2 (24–30 cycles), 102% for MCT4 (19–31 cycles), 103% for EF-1α (15–21 cycles), 111% for 18S (8–14 cycles), and 115% fo β-actin (14–21 cycles). The thermal profile was 40 cycles (95°C/5 min, 95°C/10 s, 60°C/30 s), followed by a final melt curve (65°C to 95°C with 5°C increment/5 s). Expression levels of MCTs were normalized to EF-1α because comparisons of the three housekeeping genes (EF1-α, 18S, and β-actin) using Excel-based software “BestKeeper” (29) showed that EF1-α is the best housekeeping gene and that β-actin is too variable. The C_t for all the MCT amplifications were within the range of the standard curves. Changes in the expression of MCT1a, MCT1b, MCT2, and MCT4 transcript levels were determined using the “ΔΔC_t method” (28).

Table 1.

Primers for cloning rainbow trout monocarboxylate transporter isoforms MCT1a, MCT1b, MCT2, and MCT4 based on zebrafish sequences

Primer	Forward (5′ to 3′)	Reverse (5′ to 3′)	Amplicon Size, bp
Zfish-MCT	cggctggaggggcagcttccn	caccaggtccatcagtgtctcgaan	565
Sal-MCT;Trt-MCT	ccttcatctccatcggcttctcn	aggaagctgcccctccagcn	460
3′Trt-MCT1-outer	cccaggrtrcagtacttcttygc
3′RACE outer*		gcgagcacagaattaatacgact
3′Trt-MCT1-inner	cggtgtgtgycacgtcytgg
3′RACE inner*		cgcggatccgaattaatacgactcactatagg
3′Trt-MCT2-outer	cggcccaggatccagtacttcttc
3′Trt-MCT2-inner	cggggtgtgccacatcctctg
Trt-Exp-MCT1a	tgcagcgtgtcaatgccgtg	ccttactcttagcgtagttggagagg	138
Trt-Exp-MCT1b	acgacgtgcgggcagaagtg	tccctgctcttggcaaagttac	149
Trt-Exp-MCT2	cgaccacttgtttggggactgtc	tatgtccttggcataggccgtc	149
Trt-Exp-MCT4	tcactgggttctcctacgc	cactggccgacacccaaac	179
EF-1α	cattgacaagagaaccattga	ccttcagcttgtccagcac	98

See text for definitions of abbreviations.

^*Primers were taken from the Ambion FirstChoice RLM-RACE Kit. Trt-MCT, 3′Trt-MCT1-outer and -inner, 3′Trt-MCT2-outer and -inner, Trt-Exp-MCT1a, -MCT1b, MCT2, and MCT4 were designed based on rainbow trout MCT sequences. EF-1α was previously published (25).

Calculations and statistics.

Statistical comparisons were performed using one- or two-way repeated measure analysis of variance (RM-ANOVA) with the Bonferroni post hoc test to determine which means were different from controls. In cases where the assumptions of normality or homoscedasticity were not met, the data was normalized by log₁₀ transformation before parametric analysis. Friedman repeated measures ANOVA on ranks was used with Dunn's test when normality or homoscedasticity were not met after transformation. All values presented are means ± SE, and a level of significance of P < 0.05 was used in all tests.

RESULTS

Partial characterization of monocarboxylate transporters in rainbow trout.

Partial sequences of the rainbow trout monocarboxylate transporters MCT1, MCT2, and MCT4 were cloned and characterized. Figure 1 compares the amino acid sequences of MCT1a, MCT1b, MCT2, and MCT4 between rainbow trout and various other vertebrate species. It shows all the conserved regions (shaded in black) and the transmembrane domains (TMD) as previously identified (12). For MCT1, two homologs were detected: MCT1a and MCT1b. The MCT1a sequence was 513 nucleotides long coding for part of the open reading frame (ORF) (171 amino acids). MCT1b was 1,829 nucleotides long with 1,155 coding for part of the ORF (385 amino acids) and 671 nucleotides in the 3′ untranslated region (UTR). The amino acid sequences of trout MCT1a and MCT1b were 75% identical. MCT1a had 79% and 77% amino acids identities with zebrafish and cod, respectively, as well as 64% and 66% identical to human and rat MCT1 amino acids (Fig. 1). The trout MCT1b was 85% and 81% identical to zebrafish and cod, as well as 72% and 75% identical to human and rat MCT1 at the amino acid level (Fig. 1). Trout MCT2 was 1,048 nucleotides long with 666 coding for part of the ORF (222 amino acids) and the remaining sequence in the 3′-UTR. It had 78% and 81% identities to zebrafish and cod, respectively, as well as 68% and 57% identities with its counterpart in human and rat at the amino acid level. The trout contiguous sequence (accession no. EZ780500.1; GenBank) identified as MCT4 was 1,072 nucleotides long with a partial ORF of 666 (region 405 to 1071 pb) coding for 222 amino acids, and the remainder in the 5′-UTR. It had 88% and 87% identities to zebrafish and cod, respectively, as well as 79% and 81% identities with human and rat MCT4. Percent identities of trout MCTs with other vertebrate MCTs was obtained by comparing partial deduced amino acid sequences of trout MCTs. To confirm the identity of rainbow trout MCTs, we performed a Bayesian analysis (phylogenetic tree) of the MCT protein sequences for isoforms 1, 2, and 4 of various fish and mammalian species. This analysis shows that rainbow trout MCT1a, MCT1b, MCT2, and MCT4 (indicated by arrows) cluster with the corresponding isoforms from other fish species (Fig. 2).

Tissue distribution of monocarboxylate transporters.

Figure 3 shows the presence of MCT1a, MCT1b, MCT2, and MCT4 in gill, heart, liver, red muscle, white muscle, and brain of rainbow trout. All the tissues investigated expressed the four isoforms, but mRNA levels varied greatly. Statistics for intertissue comparisons in control fish are not indicated on Fig. 3, but in the text below. Comparing the control levels of each isoform between tissues reveals the following: MCT1a expression was approximately fourfold lower in white muscle than in gill, heart, and brain (P < 0.05; Fig. 3A). MCT1b was expressed at much higher levels in heart and red muscle than in other tissues [>10 times higher than in liver, 23 times higher than in white muscle, and >50 times higher than in gill and brain (P < 0.001; Fig. 3B)]. The expression of MCT2 was ∼40 times higher in gill and brain than in any other tissue (P < 0.001; Fig. 3C). MCT4 was >40 times higher in the heart than in any other tissue (P < 0.001; Fig. 3D).

Fig. 3.

Effects of intense exercise on the expression of MCT1a (A), MCT1b (B), MCT2 (C), and MCT4 (D) in rainbow trout tissues: gills (G), heart (H), liver (L), red muscle (RM), white muscle (WM), and brain (B). Histograms show preexercise, control levels (white bars) and postexercise levels (black bars; 23 h after intense swimming). Values are means ± SE (n = 6 for controls and 5 for postexercise) and were standardized to the expression of elongation factor 1α (EF-1α). *Differences between controls and exercised fish (P < 0.02). Intertissue differences in expression within control animals are not indicated in this figure, but they are detailed in the text.

Changes in MCT expression.

The effects of exhausting exercise on the expression of MCT1a, MCT1b, MCT2, and MCT4 in rainbow trout tissues are presented in Fig. 3 where values were normalized to EF-1α. The same effects were confirmed when using 18S as an alternative housekeeping gene (results not shown). Transcript levels increased by 90% for MCT1a and by 50% for MCT1b in the heart of exercised fish (P < 0.05; Fig. 3, A and B). However, swimming had no effect on MCT1a and MCT1b expression in other tissues (gill, liver, muscle, and brain; P > 0.05). The expression of MCT2 was strongly upregulated in the brain (+260%; P < 0.02) but remained unchanged in all the other tissues (P > 0.05; Fig. 3C). Exhausting exercise had no effect on the expression of MCT4 (P > 0.05; Fig. 3D).

Blood lactate concentration.

Circulating lactate levels of resting controls remained low and steady at 1.3 ± 0.1 mM during the 24 h preceding tissue sampling. For the exercise group, changes in lactate concentration during recovery from maximal swimming are presented in Fig. 4. Blood lactate was 10.3 ± 1.0 mM immediately after exercise (time 0), subsequently reached a maximum of 17.8 ± 2.1 mM after 3 h, and had returned to baseline after 22 h of recovery (0.6 ± 0.2 mM).

Fig. 4.

Changes in blood lactate concentration of rainbow trout after maximal exercise. Values are means ± SE (n = 5). Error bars are not visible when smaller than the symbols. *Differences from baseline values reached 23 h after exercise (P < 0.01).

DISCUSSION

This study demonstrates the presence of four monocarboxylate transporter isoforms in rainbow trout. These proteins were present ubiquitously, but their level of expression varied greatly between tissues and isoforms. The main lactate exporter of glycolytic muscle in mammals, MCT4, was only expressed significantly in trout heart, but barely detectable in all other tissues, including white muscle. More importantly, exhausting exercise had no stimulating effect on the expression of any MCT isoform of white muscle. This study provides a possible functional explanation for the elusive lactate retention behavior observed during recovery. The classic delay in exporting large lactate loads from white muscle to the circulation may be explained by the much lower expression of all MCTs in white muscle than in other tissues and by the total lack of MCT upregulation by exercise in this tissue.

Cloning and identification of rainbow trout MCTs.

This study shows the existence of MCT1a, MCT1b, MCT2, and MCT4 in Oncorhynchus mykiss. Using zebrafish sequences as a reference, we have characterized 37% of MCT1a (171 amino acids), 83% of MCT1b (385 aa), 56% of MCT2 (222 aa), and 44% of MCT4 (222 aa). To ensure that these peptides were properly identified, amino acid sequence alignments of the partial trout MCTs were performed with known MCTs from mammals and other fish species (Fig. 1). The partial rainbow trout isoforms showed higher identity with zebrafish and cod (72–85%) than with mammals (57–75%). The transmembrane domains identified in trout were extremely conserved across species: TMDs 6–10 for MCT1a, TMDs 2–12 for MCT1b, TMDs 6–11 for MCT2, and TMDs 1–6 for MCT4 (see Fig. 1). A Bayesian phylogenetic analysis of MCT1–4 was performed to compare known fish and mammalian sequences. The tree topology shows clades of fish and mammals in which rainbow trout MCTs (indicated by arrows in Fig. 2) cluster adequately with the corresponding fish homologs. MCT1a and MCT1b likely arose from the teleost genome duplication (39), and our phylogenetic analysis supports this idea by showing that the two MCT1 homologs of trout are also closely related (Fig. 2). Fish are known to express multiple copies of MCT2 (39) and the phylogenetic tree shows that the trout gene identified here is closely related to the MCT2a isoform previously characterized in other fish species (Fig. 2).

Tissue distribution of rainbow trout MCTs.

All MCT isoforms were found in every tissue, but white muscle showed the lowest expression of these genes (Fig. 3). The fact that MCT1a, MCT1b, MCT2, and particularly MCT4 are barely expressed in white muscle may explain why lactate export from this tissue is so slow (37, 42). Numerous mammalian studies have shown that MCT1 plays a key role in lactate uptake by oxidative tissues. With a Michaelis constant (K_m) of 4–7 mM, the mammalian MCT1 is particularly important during recovery from intense exercise, when blood lactate levels are elevated (2, 11). A similar lactate affinity was reported for the MCT1b of fugu fish (K_m of 4 mM) (39), suggesting that fish MCT1b has the same function as the unique MCT1 isoform expressed by mammals. High expression of MCT1b in heart and red muscle of trout (Fig. 3B) is consistent with the better lactate clearance seen during active recovery (low-intensity swimming), when both tissues increase their use of oxidative fuels over passive (resting) recovery (24).

MCT2 was poorly expressed in rainbow trout. It was mostly found in the brain and gill (Fig. 3) but barely detectable in other tissues. Similarly, zebrafish MCT2 was dominant in the brain and expressed significantly in the gills (26). Mammalian studies show that MCT2 is specialized for lactate uptake, has high affinity for its substrate (K_m of 0.7 mM), and plays an important role in the brain to shuttle lactate from astrocytes to neurons for oxidation (5, 9, 34). The tissue distributions of MCT2 in trout and zebrafish suggest that this isoform has a similar function in fish. It could allow highly oxidative tissues such as the brain and gill to use lactate as an oxidative fuel under baseline conditions, when blood lactate is ∼1 mM (Fig. 4). More work is needed to determine whether MCT2 has as low a K_m in fish as previously reported for mammals. Because MCT4 is recognized as the main lactate exporter from glycolytic muscles in mammals (11, 12, 21), its extremely low expression in trout white muscle could well be partly responsible for lactate retention after maximal exercise (37, 42–44). A previous analysis of efflux kinetics using sarcolemmal vesicles supports this observation by showing that simple diffusion is the main mechanism for lactate export from white muscle (33).

Exhausting exercise and MCT expression.

As circulating lactate becomes abundant after exercise (Fig. 4), some tissues increase their ability to import this valuable fuel (brain and heart), whereas others do not (muscle, liver, and gills) (Fig. 3). Exhausting exercise stimulates the expression of MCT2 in the brain (+260%) and of MCT1a and MCT1b in the heart (+90% and +50%, respectively). The brain is well known to favor lactate as an oxidative fuel in birds and mammals (16, 19, 31, 32, 34, 35), and the large MCT2 response observed here suggests that it is also true in fish. Perfused in situ preparations have allowed to show that the heart of rainbow trout oxidizes lactate preferentially (18, 22). Our results are consistent with the higher capacity for lactate uptake potentially afforded by the increased expression of cardiac MCT1a and MCT1b (Fig. 3). Lactate clearance from the circulation would also be facilitated by improving the capacity for lactate entry in red muscle and gills for oxidation, as well as in liver for gluconeogenesis.

Immediately after maximal exercise, lactate concentration reaches 30–40 mM in white muscle before a progressive decline to baseline over the next 10 h (38). By contrast, the recovery pattern in blood is completely different. Blood lactate concentration continues to rise for 3–4 h into recovery to reach a maximum of only 15–20 mM and subsequently returns to baseline over several hours (38) (Fig. 4). This study is the first to demonstrate the presence of monocaboxylate transporters in rainbow trout. Surprisingly, exercise only stimulates MCT expression in the brain (MCT2) and heart (MCT1a, b), presumably to increase capacity for lactate uptake and to provide these highly oxidative tissues with better access to an abundant fuel source. Taken together, results provide a possible functional explanation for the classic “lactate retention” phenomenon. Rainbow trout are unable to release large lactate loads rapidly from white muscle because: 1) MCT4, the main lactate exporter characterized in glycolytic muscles of mammals, is very poorly expressed in trout; 2) the combined expression of all the trout MCT isoforms is much lower in white muscle than in other tissues; and 3) exhausting exercise fails to upregulate MCT expression in white muscle.

Perspectives and Significance

This study supports the idea that white muscle operates as a quasiclosed system with regard to carbohydrate metabolism. Intense swimming is fueled by local glycogen stores that are subsequently replenished in situ from lactate during recovery (23, 37). Resynthesis of glycogen from lactate is known to occur at significant rates in the glycolytic muscles of some ectotherms, possibly through a reversal of the pyruvate kinase reaction (8, 10). Mammalian glycolytic muscles behave very differently because they favor the Cori cycle as a strategy for recovery, exporting lactate rapidly to the circulation and replenishing glycogen stores from hepatic glucose. In trout, white muscle glycogen acts as an “energy spring” that alternates between explosive power release during intense swimming and slow recoil during protracted recovery. Future studies on the time course of changes in gene expression and protein levels will determine whether the pattern of white muscle MCT expression reported here makes such a strategy possible.

GRANTS

This work was supported by an NSERC discovery grant to J.-M. Weber.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: T.O. and J.-M.W. conception and design of research; T.O. performed experiments; T.O. analyzed data; T.O. and J.-M.W. interpreted results of experiments; T.O. prepared figures; T.O. drafted manuscript; T.O. and J.-M.W. edited and revised manuscript; T.O. and J.-M.W. approved final version of manuscript.