Lactate transport across sarcolemmal vesicles isolated from rainbow trout white muscle

The effects of exhaustive exercise in salmonids such as rainbow trout, Onchorynchus mykiss, have been examined extensively over the past four decades. The fate of lactate, which is produced during glycogenolysis in the white muscle fibres, has been the focus of many of these studies (Turner et al., 1983; Milligan and Wood, 1986; Milligan and McDonald, 1988; Walsh, 1989; Pagnotta and Milligan, 1991). Across the vertebrate groups, there exist different strategies for removing lactate from the glycolytic fibres following anaerobic metabolism. Mammals appear to ‘shuttle’ lactate from the glycolytic fibres to more oxidative tissues, with the removal being complete within 1 h (Brooks, 1987). In teleosts, lactate is primarily retained, and full recovery from exhaustive exercise has been found to require upwards of 6 h (Black et al., 1962; Pearson et al., 1990; Scarabello et al., 1991; Pagnotta and Milligan, 1991). Amongst teleosts, the proportion of lactate that is retained by the white muscle varies; pelagic species such as salmonids release approximately 20 %, and the less active benthic species, such as plaice and flounder, retain approximately 99 % (Wardle, 1978; Batty and Wardle, 1979; Milligan and Wood, 1986, 1987). It is widely believed that the lactate is used as a substrate for in situ glycogenesis, although the complete pathway is unknown (Milligan and McDonald, 1988; Girard and Milligan; 1992; Weber and Haman, 1996).

The mechanism for retaining the substantial amount of lactate produced during exercise, which can exceed 50 mmol kg⁻¹, is unknown. In a perfused trout trunk preparation, the anion transport inhibitor 4-acetoamido-4′-isothiocyanstilbene-2,2′-disulphonic acid (SITS) apparently enhanced lactate efflux relative to controls (Turner and Wood, 1983). The simplest interpretation of these data is that lactate leaving the muscle is subsequently taken back up and that this uptake is inhibited by SITS, leading to an apparent stimulation of efflux. This led to the hypothesis that, after its initial efflux, trout muscle is able to transport lactate back into the white muscle fibre along diffusion gradients (Turner and Wood, 1983). Subsequent in vivo studies have confirmed that trout muscle is capable of net lactate uptake from the blood, even against an electrochemical gradient (Girard and Milligan, 1992; Milligan and Girard, 1993).

The characterization of lactate transport by rainbow trout white muscle was recently revisited by Wang et al. (1997) using an isolated perfused trunk preparation. This study found that, following exercise, lactate efflux from the muscle could be inhibited by as much as 40 %, indicating that a considerable proportion of lactate efflux is probably via a transporter. In the perfused trunk prepared from a resting trout and at 16 mmol l⁻¹ external lactate concentration, lactate uptake was estimated to be 30–36 % via a symport and 39–45 % via the anion antiport (Wang et al., 1997). Overall, this study suggests that the passage of lactate into and out of white muscle is at least partially carrier-mediated.

The characterization of lactate transport in mammalian skeletal muscle was enhanced with the development of the sarcolemmal vesicle preparation (Roth and Brooks, 1990; Juel, 1991). With such an isolated membrane system the transport of lactate is focused on the sarcolemma, and it is possible to determine the kinetic and biochemical properties of a potential carrier. In mammalian giant sarcolemmal vesicles, it has been estimated that up to 90 % of the membrane passage of lactate is via a monocarboxylate transporter, which transported lactate in symport with a proton. This carrier was stereoselective for the L-lactate isomer and sensitive to α-cyano-4-hydroxycinnamic acid (CHC; Juel, 1991).

The molecular characterization of the first isoform of the monocarboxylate carrier (MCT1) was reported for a Chinese hamster ovary cell line, the cDNA was cloned and the tissue distribution assessed using northern blot analysis (Garcia et al., 1994). To date, cDNAs for seven putative isoforms of the monocarboxylate carrier (MCT1–MCT7) have been cloned, six of which have unique tissue distributions (Price et al., 1998). The MCT4 isoform (originally published as MCT3) has been identified in mammalian muscle rich in glycolytic fibres and is thought to be responsible for the efflux of lactate from these fibres. MCT1 is the predominant isoform in oxidative fibres and is thought to operate to take up lactate (Wilson et al., 1998).

The transport of lactate in fish has not been studied at the level of the sarcolemma using an isolated membrane system. Our objectives were to develop a sarcolemmal vesicle technique for use with fish muscle, to use these vesicles to test the hypothesis that fish muscle contains a lactate (monocarboxylate) carrier and to characterize the carrier pharmacologically.

Male and female rainbow trout [Oncorhynchus mykiss (Walbaum); 150–450 g] were purchased from Rainbow Springs Trout Farm (Thamesford, Ontario, Canada) throughout the year. The trout were held in a large circular tank (1000 l) provided with a continuous flow of dechlorinated London tapwater kept at 14±1 °C. The photoperiod was kept at 12 h:12 h light:dark. The fish were fed to satiation every other day with commercial trout pellets. Immediately prior to tissue sampling, fish were killed by anaesthetic overdose: 2 g of MS-222 (tricaine methane sulphonate, Syndel) and 2 g of NaHCO₃, to buffer the anaesthetic, in 10 l of dechlorinated tap water.

KCl/Mops buffer solution consisted of 140 mmol l⁻¹ KCl and 5 mmol l⁻¹ Mops, with the pH adjusted to 7.4 at room temperature. All experiments were carried out at room temperature (20–22 °C). The vesicle preparation medium (VPM) consisted of 50 mg of phenylmethyl sulphonyl fluoride (PMSF) dissolved in 200 μl of dimethylsulphoxide (DMSO; 0.25 % w/v) added to 350 ml of KCl/Mops buffer and filtered through 2167Whatman no. 1 paper.

The giant vesicle technique utilizes collagenase treatment to create the vesicles (Juel, 1991). White muscle tissue (30–50 g) was taken from the dorsal epaxial muscle mass, trimmed free of skin and connective tissue, and placed in a large glass Petri dish containing 50 ml of VPM. The muscle was scored extensively along the length of the fibres. The resultant pieces were placed in a flask containing 40 ml of VPM, 10 000 units of Type IV collagenase (Worthington Biochemical) and 1 mmol l⁻¹ CaCl₂. The tissue/collagenase mixture was incubated in a shaking water bath at 34–37 °C for 1 h. To stop the collagenase reaction, 30 ml of 10 mmol l⁻¹ EDTA was added, the flasks were shaken, and the contents were then filtered through one layer of cheesecloth to remove larger pieces of undigested tissue. The filtrate was poured into 15 ml conical centrifuge tubes and spun in a clinical centrifuge at 25 g for 10 min. The supernatant was collected, and the pellet, consisting of partially digested muscle tissue, was discarded. The supernatant was added to a colloidal medium, Percoll and 1.4 mol l⁻¹ KCl (1:0.33:0.0575). The vesicle/Percoll mixture was bottom-loaded into conical 15 ml centrifuge tubes containing 2.5 ml of 5.75 % 5-(N-2,3-dihydroxypropylacetoamido)-2,4,6-triiodo-N,N′-bis(2,3-dihydroxypropyl)-isophthalamide (Nycodenz). The tubes were topped with 2 ml of KCl/Mops buffer and centrifuged at 100 g for 30 min. Vesicles appeared at the interface between the Nycodenz and KCl/Mops layers. The vesicles were harvested and subsequently pelleted by centrifugation at 1300 g for 30 min. The pellets were resuspended in a known volume of KCl/Mops for use in transport studies.

Vesicles were assayed for total protein content using the Bradford method with bovine serum albumin (Fraktion V, Boehringer Mannheim, Germany) as standards (Bradford, 1976). To assess vesicle viability, a 0.4 % solution of Trypan Blue was added to the resuspended vesicles in a 1:25 ratio, and samples were placed on a haematocytometer and viewed under a compound microscope. Vesicles were considered to be non-viable if they took up the Trypan Blue. Five hundred vesicles were counted, and viability is expressed as a percentage of the total number of vesicles that did not stain blue.

The K⁺-stimulated p-nitrophenylphosphatase (K-pNPPase) assay was used to identify the sarcolemma-rich fraction. Under K⁺ stimulation, p-nitrophenylphosphatase, a non-labile component of the Na⁺/K⁺-ATPase, will convert p-nitrophenyl phosphate to p-nitrophenol (Bers, 1979; Roth and Brooks, 1990). The appearance of p-nitrophenol was measured spectrophotometrically at 415 nm. p-Nitrophenylphosphatase activity is normally measured under both K⁺-stimulated and K⁺-free conditions, with the difference representing the K-pNPPase activity specifically. However, in the present study, because of the high background levels of K⁺ in the vesicle preparation medium, the total activity of the enzyme was measured, and this probably includes some non-specific phosphatase activity. Tissue samples taken prior to vesicle creation were homogenized using a Polytron (Brinkman, Rexdale, Ontario, Canada) and were compared to measure any enrichment of p-nitrophenylphosphatase activity in the vesicles. All samples were run at least in duplicate.

A comparison of the activity of the ATPases found in a sample of homogenized tissue and a sample of vesicles gives an indication of the origin of the membrane vesicles (Sadrzadeh et al., 1993). The Na⁺/K⁺-ATPase is found in the sarcolemma, and the Ca²⁺-ATPase is found primarily in the sarcoplasmic reticulum (Sadrzadeh et al., 1993). ATPase activities were measured in the presence and absence of Ca²⁺ (to determine the Ca²⁺-ATPase activity) and ouabain (to determine Na⁺/K⁺-ATPase activity) following the method described by Sadrzadeh et al. (1993), except that all volumes were tripled and the absorbances were read at 810 nm using a quartz microcuvette in a spectrophotometer. These assays were performed on homogenate and vesicle samples in duplicate.

The reaction mixture for measuring lactate transport consisted of KCl/Mops plus enough universally labelled [¹⁴C]lactate (ICN Radiochemicals) to yield 18.5 Bq per reaction tube and known concentrations of cold lactate. Lactate transport experiments were initiated by mixing 25 μl of vesicle suspension with an equal volume of reaction mixture in a 1.7 ml microfuge tube. All transport studies were performed under zero-trans conditions, i.e. all the substrate and label were initially external to the vesicles. The reaction was stopped at the designated time by the addition of 1.45 ml of ice-cold HgCl₂ stop solution. HgCl₂ modifies sulphydryl groups on proteins and has been shown to inhibit lactate uptake in several different cell types (Grimditch et al., 1985; Roth and Brooks, 1990). To test its effectiveness in our system, the HgCl₂ stop solution was added either first to the vesicles, followed by the substrate, or simultaneously with the substrate, and the suspension was centrifuged after 1 min. There was no measurable lactate uptake by the vesicles in either situation. The microfuge tube was then centrifuged at 15 000 g for 3–5 min to pellet the vesicles, and the supernatant was siphoned off to the edge of the pellet. The tubes were snipped close to the pellet using clippers, and the bottom of the tube, containing the pellet, was dropped into 20 ml scintillation vials. A 10 μl sample of the reaction mix was also taken and added to a scintillation vial. ReadySafe (Beckman) scintillation fluid (10 ml) was added to each vial, and the vial was shaken and left overnight in the dark. The vials were counted on a Packard 1900 TR liquid scintillation analyzer to measure radioactivity (disints min⁻¹) in both the reaction mixture samples and the pellets.

The relative time course of lactate uptake at a 1 mmol l⁻¹ external lactate concentration was initially determined using different transport periods ranging from approximately 1 s to 2 h. More rigorous studies of lactate uptake by the trout sarcolemmal vesicles were then conducted over periods of 10, 20, 30 and 45 s. In all cases, the transport was deemed to have begun upon the addition of the reaction mixture to the vesicles and to have stopped upon the addition of the HgCl₂ stop solution. The effects of different external lactate concentrations on the rate of lactate uptake were determined using a 30 s transport period. The concentration of cold lactate in the reaction mixture was adjusted so that the final concentration once combined with the vesicles was 1, 5, 10, 15, 20, 25, 30, 50, 100, 250 or 500 mmol l⁻¹, while the amount of radiolabelled lactate remained constant at 18.5 Bq. When uptake was measured at the higher lactate concentrations (>50 mmol l⁻¹), the vesicles were first equilibrated for 60 min at 20 °C with an appropriate amount of 500 mmol l⁻¹ sucrose solution such that dilution into the transport medium did not cause osmotic shock.

All inhibition studies were conducted over a transport period of 30 s. Controls (i.e. vesicle preparation minus inhibitor) were always run concurrently with experimentals and contained an equivalent amount of any vehicle (e.g. DMSO) that was used as a carrier for the various pharmacological compounds. p-Chloromercuriphenylsulphonic acid (pCMBS) is an organomercurial protein-modifying agent that is impermeable, reacts with thiol groups and has been shown to inhibit monocarboxylate transport in several systems (Deuticke et al., 1978). Vesicles were pre-incubated in 0.5 mmol l⁻¹ pCMBS in DMSO (0.5 % w/v) for 30 min at 20 °C prior to using them for uptake measurements. Following incubation, the vesicles were used in a zero-trans study of lactate uptake at 1 mmol l⁻¹ and 10 mmol l⁻¹ final external lactate concentrations. Isobutylcarbonyl lactyl anhydride (iBCLA) is thought to acylate an essential sulphydryl group on the transporter (Johnson et al., 1980) and has been shown to inhibit lactate uptake in mammalian erythrocytes and cardiac muscle (Donovan and Jennings, 1985). iBCLA (a gift from Dr G. Tibbits, Simon Fraser University, Burnaby, British Columbia, Canada) was dissolved in 5 % DMSO solution and added directly to the reaction mixture to a final concentration of 0.2 mmol l⁻¹. Uptake measurements were made in the presence of iBCLA at final external lactate concentrations of 1 mmol l⁻¹ and 10 mmol l⁻¹.

CHC competitively inhibits the monocarboxylate carrier in mammalian erythrocytes and non-competitively inhibits the Cl⁻-dependent anion exchanger (Halestrap, 1976). Vesicles were incubated in 5 mmol l⁻¹ CHC (final concentration) in 0.5 % DMSO (w/v), neutralized with 5 mmol l⁻¹ NaHCO₃. The control and CHC reaction mixtures contained either 1 or 10 mmol l⁻¹ final concentration of lactate. SITS has been shown to block anion antiport in a variety of cell types (Poole and Halestrap, 1993). A reaction mixture, containing a final concentration of SITS of 0.5 mmol l⁻¹ in 0.5 % (w/v) DMSO, was prepared under low-light conditions and kept in amber microfuge tubes. Control and inhibitor reaction mixtures contained final concentration of 1 mmol l⁻¹ lactate.

cis-inhibition studies by pyruvate, another monocarboxylate, were conducted with the addition of pyruvate to the reaction medium. All concentrations listed in the figure legends represent the final concentration in the reaction mixture.

The values presented are the mean ±1 standard error of the mean (S.E.M.). Significant difference from paired controls was determined using a two-tailed paired Student’s t-test with P⩽0.05. A regression analysis was performed on the concentration data series to determine the correlation coefficient.

A quantification of the average amount of vesicles produced per protocol (one fish) was expressed as the amount of protein as determined with the Bradford protein assay. The mean concentration used in the transport assays was 0.51±0.09 μg protein μl⁻¹. An estimate of vesicle viability of 88.7±3.4 % (N=5) was determined qualitatively using the method of Trypan Blue dye exclusion. Vesicles were typically spherical in shape, ranging from 0.5 to 35 μm in diameter (Fig. 1). Vesicles could easily be distinguished from lipid droplets because the latter appeared refractory when viewed with phase-contrast microscopy.

Purification indices, to determine that the sarcolemma was enriched as opposed to other membranes (sarcoplasmic reticulum, mitochondria, etc.), were estimated using two methods. A comparison of the K⁺-stimulated p-nitrophenylphosphatase (K+-pNPPase) activities in trout vesicles and homogenate is given in Table 1. On average, the sarcolemma was concentrated fourfold compared with the tissue sample. Multiple membrane ATPase assays estimated a sarcolemmal enrichment factor of approximately 13. There was only trace contamination by sarcoplasmic reticulum, as indicated by a purification index for the Ca²⁺-ATPase of less than 1 (Table 1).

Trout sarcolemmal vesicles incubated with 1 mmol l⁻¹ external lactate concentration and 0.5 μCi of [14C]lactate showed an increase in lactate accumulation over time (Fig. 2). The rate of lactate uptake by the vesicles was linear over the first 30 s at a concentration of 1 mmol l⁻¹ (Fig. 2); all subsequent uptake experiments were therefore carried out over 30 s.

The rate of lactate uptake increased and saturated as the external concentration of lactate increased (Fig. 3). A Lineweaver–Burke plot (Fig. 3 inset) of the mean values was used to determine kinetic variables. The transporter has an apparent low affinity for lactate, with a Michaelis–Menten constant (K_m) of 55.6 mmol l⁻¹, but a high transport capacity, with an estimated maximum rate of uptake (V_max) of 44.5 nmol mg⁻¹ protein min⁻¹.

All cis-inhibition series were performed for a period of 30 s, and all control values represent values for vesicles created from the same batch as the treatment and run concurrently. Incubation of the vesicles with pCMBS for 30 min did not inhibit lactate uptake at 1 mmol l⁻¹ or 10 mmol l⁻¹ external lactate concentration (Fig. 4A). The addition of iBCLA to the reaction mixture to give a final concentration of 200 μmol l⁻¹ did not inhibit lactate uptake at either 1 mmol l⁻¹ or 10 mmol l ⁼⁻¹ external lactate concentration (Fig. 4B).

Pyruvate inhibited lactate uptake in a concentration-dependent manner. At lower concentrations of lactate (1 and 5 mmol l⁻¹), the addition of pyruvate produced no significant change in lactate uptake (Table 2). However, when the external concentrations of lactate were set at 10 mmol l⁻¹ and 20 mmol l⁻¹ and the reaction mixture contained pyruvate in high concentrations, significant inhibition of lactate uptake was seen. At 10 mmol l⁻¹ lactate and 20 mmol l⁻¹ pyruvate, the relative inhibition of lactate uptake was 39 %. When the lactate and pyruvate concentrations were 20 and 50 mmol l⁻¹, respectively, lactate uptake was inhibited by 24 %.

The role of anion antiport in lactate uptake was estimated using the antiport inhibitor SITS. In the presence of 0.5 mmol l⁻¹ SITS, lactate uptake at 1 mmol l⁻¹ external lactate concentration was apparently stimulated by approximately twofold (Fig. 5A). In the presence of 5 mmol l⁻¹ CHC, a monocarboxylate and anion antiport inhibitor, and 1 mmol l⁻¹ external lactate concentration, lactate uptake increased nearly fivefold. (Fig. 5B). When the concentrations of lactate and CHC were adjusted to 10 mmol l⁻¹ and 20 mmol l⁻¹, respectively, the apparent stimulatory effect of CHC on lactate uptake was abolished.

This work not only represents the first investigation of lactate transport using sarcolemmal vesicles in fish, but the first use of trout sarcolemmal vesicles to study the transport of any substrate. The trout sarcolemmal vesicle technique represents an advance in the study of fish metabolism, specifically membrane transport. In this case, the use of such an isolated membrane system provides a unique perspective on the transport of lactate by the white muscle of rainbow trout.

The yields and purity obtained with the trout sarcolemmal vesicle technique are sufficient to conduct transport studies, but only approximately 25 % of those reported for vesicles isolated from mammalian muscle (A. Bonen, personal communication). This has meant that fewer transport treatments can be examined in one batch and that consistency is extremely important. The estimates of nearly 90 % viability attest to the technique’s merit with trout muscle. The orientation of the vesicles was not specifically examined in this study, but it has been shown that the collagenase method produces vesicles from mammalian muscle that are entirely right-side-out (Pilegaard et al., 1993). The diameters of the vesicles produced were in the range 0.5–35 μm, which is within the range of diameters produced by the mammalian giant vesicle technique (Juel, 1991). The results of the K⁺-pNPPase assays show that the membrane vesicles produced have a fourfold higher concentration of sarcolemma than the homogenate, which is probably an underestimate because of the high background [K⁺] in the vesicle suspension. The multiple-membrane ATPase assay gives a more reliable estimate of membrane enrichment and suggested that the vesicles were enriched approximately 13-fold with membranes of sarcolemmal origin. This is in good agreement with the 15-fold enrichment reported by Pilegaard et al. (1993) for this method when preparing vesicles from mammalian muscle. The ATPase assay also indicated only trace contamination of the vesicles with membrane of sarcoplasmic reticular origin. Although not measured, it is likely that contamination by mitochondrial membrane is equally low, given that fish white muscle has a low mitochondrial density (Moyes et al., 1989).

The results of this study show that lactate uptake by trout white muscle sarcolemmal vesicles increases over time, indicating that the sarcolemma of rainbow trout white muscle is capable of transporting lactate. It was also shown that this uptake of lactate was via a saturable mechanism. The Michaelis–Menten constant (K_m) was estimated to be 56 mmol l⁻¹, which is indicative of a low-affinity carrier. The estimates of K_m for lactate uptake for mammalian muscle sarcolemmal vesicles ranges from 20 to 40 mmol l⁻¹ (Juel, 1991; McDermott and Bonen, 1993). The high K_m for lactate uptake estimated in the present study may explain why Wang et al. (1997) were unable to detect lactate uptake in a perfused trout trunk at lactate concentrations below 3 mmol l⁻¹. The maximum rate of lactate uptake, or V_max, was 44.5 nmol mg⁻¹ protein min⁻¹, i.e. less than half the rate of the mammalian estimates of 139 nmol mg⁻¹ protein min⁻¹ reported by Roth and Brooks (1990), but still a fast rate of uptake. High physiological values for K_m and V_max imply a large capacity for the transmembrane movement of lactate (Roth and Brooks, 1990). However, because there appears to be lactate efflux from the vesicles (see below), these estimates of K_m are probably overestimated, and those of V_max underestimated, by unknown amounts. Despite this caveat, these results indicate that lactate uptake across the white muscle membrane of rainbow trout is via a low-affinity, high-capacity carrier.

Incubation of trout white muscle sarcolemmal vesicles with 0.5 mmol l⁻¹ pCMBS prior to transport studies at 1 and 10 mmol l⁻¹ yielded surprising results. pCMBS is thought to modify thiol groups on the extracellular domain of membrane proteins, but had no detectable effects on lactate uptake at either 1 or 10 mmol l⁻¹ external lactate. This is in sharp contrast to the results seen with mammalian sarcolemmal vesicles, in which pCMBS inhibited lactate transport by as much as 90 % (Juel, 1991; Roth and Brooks, 1990). The effect of pCMBS on lactate transport appears to be specific for cell type as well as for species (Poole and Halestrap, 1993). For example, in human erythrocytes, pCMBS inhibited lactate uptake by approximately 90 %, whereas in guinea pig cardiac myocytes, the inhibition was only 55 % (Deuticke et al., 1978; Poole et al., 1989; Wang et al., 1993), and in astrocytes in culture, lactate uptake was insensitive to pCMBS (Nedergaard and Goldman, 1993). In carp erythrocytes, pCMBS inhibited lactate uptake by 90 % (Tiihonen and Nikimaa, 1993), but eel erythrocytes were less sensitive to pCMBS, only 50 % of lactate uptake was inhibited (Soengas and Moon, 1995). It appears that inhibition of monocarboxylate transport by pCMBS falls into three categories: none, moderate or essentially complete (>80 %), with the trout sarcolemmal vesicle system falling into the first category. If these systems possess a different isoform of the monocarboxylate transporter, as recent molecular data tend to suggest (Price et al., 1998; Wilson et al., 1998), then the differences in sensitivity to a protein modifier such as pCMBS could be explained by topographical differences between isoforms yielding variable thiol group exposure.

The lack of inhibition of lactate uptake by iBCLA is consistent with the results from mammalian sarcolemmal vesicles studies (Juel, 1991). These observations contrast with those seen using rabbit erythrocytes in which iBCLA inhibited lactate uptake by nearly 80 % (Donovan and Jennings, 1985). In addition, iBCLA inhibited lactate uptake in trout cardiac sarcolemmal vesicles by as much as 85 % (G. Tibbits, personal communication). These results are a further indication that the lactate transporter in trout white skeletal muscle is different from that in mammalian erythrocytes and even in trout cardiac muscle.

The motivation for using supraphysiological concentrations of pyruvate was simply to determine whether lactate uptake was carrier-mediated. The inhibition of lactate uptake at 10 and 20 mmol l⁻¹ external concentrations in the presence of high pyruvate concentrations (39 and 24 % inhibition, respectively) suggests that pyruvate is competing for the same transporter, probably a monocarboxylate transporter. In mammalian sarcolemmal vesicles, 10 mmol l⁻¹ pyruvate inhibits lactate uptake by 71–81 % at 1 mmol l⁻¹ lactate (Roth and Brooks, 1990; McDermott and Bonen, 1993). In eel erythrocytes, the inhibitory effect of pyruvate on lactate uptake varied depending on the concentrations of both substrates (Soengas and Moon, 1995). Inhibition of lactate uptake was greatest (80 %) at 100 mmol l⁻¹ pyruvate and 2 mmol l⁻¹ lactate, but only 20 % when the pyruvate concentration was less than 20 mmol l⁻¹ (Soengas and Moon, 1995). The 24–39 % inhibition of lactate uptake by trout sarcolemmal vesicles by pyruvate is consistent with the results of Wang et al. (1997), in which it was estimated that 30–36 % of lactate uptake is via a monocarboxylate carrier. It is not possible to determine the K_i for pyruvate from the results of the present study, but it is probably well over 50 mmol l⁻¹ on the basis of the relatively low level of inhibition by 50 mmol l⁻¹ pyruvate, which is consistent with the presence of a low-affinity carrier.

SITS, which is known to inhibit anion antiport, caused a significant twofold enhancement of lactate uptake, and CHC, a non-competitive inhibitor of anion antiport and a competitive inhibitor of the monocarboxylate carrier, stimulated lactate uptake by fivefold. The apparent increases in lactate uptake in the presence of SITS or CHC were surprising and in stark contrast to their reported effects in other systems, where they routinely inhibit lactate uptake (Poole and Halestrap, 1993). The simplest interpretation of these data is that SITS and CHC are blocking the subsequent efflux of lactate from the vesicles. These data form the basis of a model, albeit a highly speculative one at this point, describing lactate transport across the trout white muscle sarcolemmal membrane (Fig. 6). The data collected suggest that lactate enters the vesicles by a combination of passive diffusion and a low-affinity carrier that is partially inhibited by pyruvate. The lactate then subsequently leaves the vesicles via an anion antiport, that is inhibited by SITS and CHC, and a monocarboxylate carrier, that is inhibited by CHC, giving the appearance of a stimulation of net lactate uptake (Fig. 6). The fivefold stimulation by CHC compared with the twofold stimulation by SITS probably reflects the ability of CHC to inhibit both anion antiport and the moncarboxylate carrier. This model is consistent with observations from the isolated perfused trout trunk preparation in which 5 mmol l⁻¹ CHC inhibited lactate efflux by 75 % (Wang et al., 1997). It is admittedly speculative, but serves as a starting point for further research into the mechanisms by which lactate moves across the fish muscle membrane.

In conclusion, trout white muscle sarcolemmal vesicles have proved to be an effective tool for studying lactate transport.

The data presented here indicate that the mechanism of lactate transport by the trout white muscle sarcolemma is very different not only from that described for other species but also from that described for other cell types (e.g. erythrocytes, cardiac muscle) within the same species. The uptake of lactate by the sarcolemma is at least partially via a low-affinity transporter that is inhibited by pyruvate at high concentrations of external lactate (10 and 20 mmol l⁻¹), indicative of a monocarboxylate transporter. Known monocarboxylate transporter inhibitors were not effective at blocking this lactate uptake at lower lactate concentrations. The efflux of lactate from the white muscle is apparently via a mechanism that is inhibited by both SITS and CHC.

We would like to thank Drs Glen Tibbits, Department of Kinesiology, Simon Fraser University, and Arend Bonen, Department of Biology, University of Waterloo, for their technical advice. This research was supported by an NSERC research grant to C.L.M.