Glucose transport in fish erythrocytes: variable cytochalasin-B-sensitive hexose transport activity in the common eel (Anguilla japonica) and transport deficiency in the paddyfield eel (Monopterus albus) and rainbow trout (Salmo gairdneri)

Equilibrative transport of D-glucose has been studied extensively in both mammalian and avian erythrocytes. Human erythrocytes and foetal erythrocytes from other mammalian species are capable of rapid, stereoselective transport of glucose (Widdas, 1980; Kim, 1983). In contrast, erythrocytes from the pig are deficient in glucose transport activity (Kim and McManus, 1971; Young et al. 1985), the cells relying on plasma inosine as their physiological energy substrate (Young et al. 1985, 1986). Mature, adult-type erythrocytes from sheep and other non-primate mammalian species exhibit intermediate rates of stereospecific glucose transport (Mooney and Young, 1978; Kim, 1983; Wagner et al. 1984). In most cases, glucose uptake is inhibited by cytochalasin B. However, in rabbit erythrocytes, carrier-mediated D-glucose transport is insensitive to cytochalasin B (Albert, 1984). At the molecular level, the human erythrocyte glucose transporter has been purified and its amino acid sequence determined (Mueckler et al. 1985). A stereoselective, cytochalasin-B-sensitive glucose transporter has also been characterized in avian erythrocytes (Kregenow, 1975). Transport of glucose in these cells, in contrast to that in mammalian erythrocytes, is stimulated by anoxia and by catecholamines (Bihler et al. 1982; Simons, 1983a).

There is little published information about glucose transport in fish erythrocytes. Ingermann et al. (1984) reported a high level of cytochalasin-B-sensitive glucose transport activity in erythrocytes from the Pacific hagfish (Eptatretus stouti), a representative of the most primitive group of living vertebrates. In contrast, in an earlier study, Bolis et al. (1971) were unable to detect stereospecific glucose uptake in erythrocytes from the brown trout (Salmo trutta). We report here the characteristics of 3-O-methyl-D-glucose (3-OMG) and D- and L-glucose transport in erythrocytes from two fish species native to eastern Asia, the common eel (Anguilla japonica) and the paddyfield eel (Monopterus albus). For comparison, we have also studied 3-OMG transport in erythrocytes from the rainbow trout (Salmo gairdneri).

Common eels (Anguilla japonica) and paddyfield eels (Monopterus albus) were obtained locally in Hong Kong and maintained in freshwater aquaria. M. albus belongs to a different order (Synbranchiformes) from A. japonica, but derives its name from its eel-like appearance and behaviour. Rainbow trout (Salmo gaird-neri) were obtained from a commercial source in Cambridge, UK, and also maintained in freshwater aquaria. Fish were not fed while in captivity.

Blood samples were obtained from the dorsal aorta of decapitated fish and collected into heparin. Erythrocytes were washed free of plasma by three washes in a medium containing 150mmoll⁻¹ NaCl, 5mmoll⁻¹ D-glucose and 15mmoll⁻¹ Mops (pH 7·5 at 20°C). The buffy coat was discarded. The washed cells were resuspended to a haematocrit of approximately 10% and left overnight at 4°C.

This was to ensure that the cells were at a steady state with respect to the incubation medium and not in a catecholamine-stimulated condition (Bourne and Cossins, 1982).

The haemoglobin contents of cell suspensions were determined by absorbance measurements of cyanmethaemoglobin [1:50 dilution with Drabkin’s solution (100 mg 1⁻¹ NaCN and 300 mg 1⁻¹ K3Fe(CN)₆ in distilled water)] at 540 nm. These values were used to estimate the volumes of cells in the original suspensions, calculated from an experimentally determined extinction coefficient of 218 for packed fish erythrocytes. There was no difference in the value of this coefficient among the three fish species studied.

Erythrocyte ATP concentrations were determined by a firefly luciferin-luciferase assay (Brown, 1982).

The experiments were performed at 20°C using an isotonic incubation medium containing 150 mmol 1⁻¹ NaCl and 15 mmol E¹ Mops (pH 7·5). Uptake of 3-O- methyl-D-[U-¹⁴C]glucose (Amersham International, Amersham, Bucks., UK), a non-metabolized D-glucose analogue, and D-[U-¹⁴C]glucose and L-[l-¹⁴C]glucose were initiated by mixing equal volumes (0·15 ml) of prewarmed cell suspension (haematocrit 20%) with incubation medium containing the appropriate concentration of radioactive permeant (0·7 μCi ml⁻¹). For incubations in the presence of cytochalasin B or phloretin, cells were equilibrated with inhibitor for 30 min at 20°C before addition of 3-O-methyl-D-[U-¹⁴C]glucose-containing medium. Uptake was terminated by addition of 1 ml of ice-cold stopping solution (150 mmol 1⁻¹ NaCl, 15 mmol 1⁻¹ Mops, 1.25 mmol 1⁻¹ KI, 1μmoll⁻¹ HgCl₂, 100pmoll⁻¹ phloretin, pH7· 5) (Weiser et al. 1983). Cells were rapidly washed four times with 1ml portions of ice-cold stopping solution using an Eppendorf 5414 microcentrifuge (10 s, 12000g). After washing, the cell pellets were lysed with 0·5 ml of 0·5 % (v/v) Triton X-100 in water and 0.5 ml of 5 % (w/v) trichloroacetic acid was added. The precipitates were removed by centrifugation (12000g for 2 min) and 0.9ml samples of protein-free supernatant counted for radioactivity by scintillation spectroscopy with appropriate quench correction. Initial rates of uptake were determined using incubation periods of between 2.5 and 10 min, depending on the hexose permeability of the cells under study. Kinetic constants were determined by nonlinear regression analysis (Enzfitter, Elsevier-Biosoft).

Representative time courses of 3-OMG uptake (5 mmol 1⁻¹ extracellular con-centration) by erythrocytes from two individuals, measured both in the presence and in the absence of 10μmoll⁻¹ cytochalasin B, are shown in Fig. 1. Cells from both fish exhibited cytochalasin-B-sensitive and cytochalasin-B-insensitive components of 3-OMG uptake. However, there was a considerable difference in the magnitude of the former component of transport in the two animals, cells from fish A exhibiting an initial rate of cytochalasin-B-sensitive 3-OMG uptake of 6 ·9 mmol 1 cells⁻¹ h⁻¹ compared with 1.0 mmol 1 cells⁻¹ h⁻¹ for erythrocytes from fish B. In contrast, cytochalasin-B-insensitive 3-OMG uptake rates were similar in the two fish (0·16 and 0·15 mmol 1 cells⁻¹ h⁻¹ for erythrocytes from eels A and B, respectively). Intracellular [¹⁴C]-3-OMG in cells from fish A reached a level of 2·9 mmol 1 cells⁻¹ after 2 h. This corresponds to 89 % equilibration with intracellular water, calculated on the basis that A. japonica erythrocytes contain 66 % (v/v) water (Fincham et al. 1987).

Fig. 2A shows initial rates of cytochalasin-B-sensitive 3-OMG uptake (5 mmol 1⁻¹) measured in erythrocytes from 50 individuals. One fish (arrowed) exhibited no detectable inhibitor-sensitive sugar uptake and was defined as transport-deficient. The remaining 49 transport-positive type fish gave cytochal-asin-B-sensitive 3-OMG uptake rates ranging between 0·07 and 19·5 mmol 1 cells⁻¹ h⁻¹. Initial rates of cytochalasin-B-insensitive 3-OMG uptake were low and much less variable (0·18–0·6 mmol 1 cells⁻¹ h⁻¹) (Fig. 2B). Storage of cells at 4°C in glucose-containing medium had no measurable effect on cytochalasin-B-sensitive 3-OMG uptake. For example, cells from one fish gave a cytochalasin-B-sensitive uptake rate of 0·86 mmol 1 cells⁻¹ h⁻¹ immediately after blood collection (5 mmol 1⁻¹3-OMG) and rates of 0·86,0·88 and 0·85 mmol 1 cells⁻¹ h⁻¹ after 24, 48 and 72 h of storage, respectively.

Of 20 eels tested, intracellular ATP levels ranged from 1·7 to 5·2 mmol 1 cells⁻¹. Cytochalasin-B-sensitive 3-OMG uptake rates in erythrocytes from these fish varied between 0·12 and 17·2mmol 1 cells⁻¹ h⁻¹. There was no correlation between the two parameters.

Fig. 3 shows a typical concentration-dependence curve for 3-OMG uptake by eel erythrocytes at 20°C measured over the concentration range 0·5–10 mmol 1⁻¹ in the presence and in the absence of 10μmol 1⁻¹ cytochalasin B. The cytochalasin-B-sensitive component of 3-OMG uptake was saturable and conformed to simple Michaelis-Menten kinetics. Kinetic constants (±S.E.) estimated by nonlinear regression analysis were: K_m 1·22±0·32mmoll⁻¹ and V_m 8·41±0·59mmoll cells⁻¹ h⁻¹. Cytochalasin-B-insensitive 3-OMG uptake was linear over the same concentration range.

To investigate the substrate specificity of the cytochalasin-B-sensitive saturable component of 3-OMG uptake in eel erythrocytes, various D- and L-hexoses were Ksted as inhibitors of 3-OMG transport (Table 1). Of the sugars tested, D-glucose, galactose and D-deoxyglucose were found to be effective inhibitors of cytochal-asin-B-sensitive 3-OMG uptake and inhibition was concentration dependent. The degree of inhibition was in the order D-deoxyglucose>D-glucose>D-galactose and is consistent with the relative affinities of these sugars for the human erythrocyte sugar transport system (Naftalin and Holman, 1977). L-Glucose, which is not a substrate for the mammalian hexose transport system, had no detectable effect on 3-OMG uptake in eel erythrocytes.

In addition to cytochalasin B, phloretin also inhibits sugar transport in a number of cell types. Fig. 4 shows concentration dependence curves for these two compounds as inhibitors of saturable (cytochalasin-B-sensitive) 3-OMG uptake in eel erythrocytes. IC50 values were 0·125 and 80 μmol 1⁻¹ (5 mmol 1⁻¹ extracellular 3-OMG) for cytochalasin B and phloretin, respectively. To estimate apparent K,, the results were fitted by least-squares analysis to the equation:

As in avian erythrocytes, 3-OMG uptake by eel erythrocytes was increased by catecholamines. Fig. 5 shows typical time courses of 3-OMG uptake in noradrena-line-treated (100μmoll⁻¹, 2 h preincubation) and control (untreated) erythrocytes measured both in the absence and in the presence of 100 moll⁻¹ cytochalasin B. The cytochalasin-B-sensitive 3-OMG uptake rate increased from 0·7 to 2·2 mmol 1 cells⁻¹ h⁻¹, while cytochalasin-B-insensitive uptake of 3-OMG remained unchanged.

Catecholamine stimulation of cytochalasin-B-sensitive 3-OMG transport depended both upon the period of exposure to catecholamine and upon the catecholamine concentration. Noradrenaline and adrenaline were equally effective, maximum stimulation occurring after 2h of exposure (Fig. 6) and at catecholamine concentrations in the range 50–100μmol 1⁻¹ (Fig. 7). There was no detectable stimulation of 3-OMG transport at concentrations below 0·01 μmoll⁻ and half-maximal stimulation occurred at approximately 1μmoll⁻¹. This latter catecholamine concentration is similar to that required for half-maximal stimulation of 3-OMG uptake into goose erythrocytes (Whitfield el al. 1974). For transport-positive type eels with basal 3-OMG (5 mmol 1⁻¹) permeabilities in the range 0·24–3·13 mmol 1 cells⁻¹ h⁻¹, maximal stimulation of sugar transport activity appeared to be independent of the starting cytochalasin-B-sensitive 3-OMG permeability of the cells (Table 2). Quantitatively similar catecholamine-induced increases in 3-OMG permeability were also detected in cells with higher basal transport activities, but this was not verified statistically. The fish previously identified as 3-OMG transport-deficient (Fig. 2) failed to respond to catecholamine stimulation (see also Table 2).

Fig. 8 compares the concentration dependence of 3-OMG uptake (range 0·05·5 mmol 1⁻¹) in control transport-positive cells and cells preincubated with noradrenaline. Kinetic constants for cytochalasin-B-sensitive 3-OMG uptake were: K_m 1·38±0·27 and 1·47±0.31mmoH⁻¹ and V_m 2·52±0·18 and 1·01±0·08 mmol 1 cells⁻¹ h⁻¹ for noradrenaline-treated and control cells, respectively. In agreement with the results presented in Fig. 5, noradrenaline had no effect on cytochalasin-B-insensitive 3-OMG uptake.

Table 3 shows that, in addition to cytochalasin-B-sensitive 3-OMG uptake, eel erythrocytes exhibited cytochalasin-B-sensitive, stereospecific transport of glucose. The observed cytochalasin-B-sensitive D-glucose uptake rate (5·9±0·1 mmol 1 cells⁻¹ h⁻¹ at 5 mmol 1⁻¹ extracellular D-glucose) was similar to that for 3-OMG uptake in the same cells (6·9±0·1 mmol 1 cells⁻¹ h⁻¹) and was equivalent to the stereospecific glucose uptake rate of 5·9±0·1 mmol 1 cells⁻¹ h⁻¹, defined as the difference in permeability between D- and L-glucose. L-Glucose uptake was not inhibited by cytochalasin B (0·09±0·01 and 0·08±0. ·01 mmoll cells⁻¹ h⁻¹ in the absence and in the presence of cytochalasin B, respectively). These L-glucose uptake rates are approximately half that for cytochalasin-B-insensitive uptake of D-glucose. In other words, cytochalasin-B-insensitive glucose uptake was partially stereospecific. A similar observation has been made with respect to D-and L-glucose uptake in cytochalasin-B-treated pigeon erythrocytes (Simons, 1983a).

In the experiment presented in Table 3, cytochalasin-B-insensitive uptake of 3-OMG was substantially greater than the corresponding fluxes for D-ancL L-glucose. This difference is a consequence of the greater hydrophobicity of 3-OMG, allowing it to diffuse through the lipid bilayer more rapidly (Hillman et al. 1959).

Erythrocytes from individual M. albus and S. gairdneri exhibited no detectable cytochalasin-B-sensitive 3-OMG uptake either in the absence or in the presence of 100μmoll⁻¹ catecholamines. Similarly, D-glucose uptake by erythrocytes from M. albus was not inhibited by cytochalasin B. Mean results for a number of fish of each species are presented in Table 4. In agreement with the data for cytochalasin- B-insensitive uptake of D- and L-glucose in erythrocytes from A. japonica (Table 3), D-glucose uptake rates in cells from M. albus were greater than for L-glucose, although the absolute magnitudes of the fluxes were smaller. This f)bservation suggests that the A. japonica result was not because of incomplete nhibition of saturable D-glucose uptake by cytochalasin B.

Of the three teleost species examined, erythrocytes from both the paddyfield eel (Af. Albus) and the rainbow trout (S. gairdneri) exhibited uniformly low rates of 3-OMG uptake in the range 0·10–0·15 mmol 1 cells⁻¹ h⁻¹ (5 mmol 1⁻¹ extracellular 3-OMG, 20°C). D-Glucose uptake rates in M. albus were lower, with a mean value of approximately 0·05 mmol 1 cells⁻¹ h⁻¹ at 5 mmol 1⁻¹ extracellular D-glucose. Sugar uptake in erythrocytes from these two species was not inhibited by cytochalasin B and was insensitive to catecholamine stimulation. Erythrocytes from one common eel (A. japonica) were also transport-deficient with respect to cytochalasin-B-sensitive 3-OMG uptake. In marked contrast, erythrocytes from the other common eels exhibited significant and highly variable rates of cytochalasin-B-sensitive 3-OMG transport in the range 0·07–19·5 mmol 1 cells⁻¹ h⁻¹ when assayed under identical conditions. This variability was not seasonal, did not depend upon the time the animals had been retained in aquaria before bleeding or upon eel size and was not correlated with intracellular ATP levels.

The stress hormones adrenaline and noradrenaline are released into the fish bloodstream during handling and bleeding and have been shown to stimulate erythrocyte ion transport and cause cell swelling (Bourne and Cossins, 1982; DeVries and Ellory, 1982; Baroin et al. 1984). A similar phenomenon occurs in avian species (Kregenow, 1975) and catecholamines have been shown to increase sytochalasin-B-sensitive sugar transport in both pigeon and goose erythrocytes (Whitfield and Morgan, 1973; Whitfield et al. 1974; Bihler et al. 1982; Simons,1983a). Several factors indicate that catecholamine effects cannot account for the variation in A. japonica 3-OMG uptake rates seen in the present study. First, erythrocytes were washed free of plasma and routinely stored overnight in glucose-containing medium at 4°C before assay of transport activity (see Materials and methods). In addition, although the cells were sensitive to catecholamine stimulation, the magnitude of the catecholamine-induced increase in 3-OMG transport activity appeared to be independent of the basal cytochalasin-B-sensitive 3-OMG permeability of the cells (Table 2). Also, cells from one common eel were totally deficient with respect to cytochalasin-B-sensitive transport activity and did not respond to catecholamine stimulation.

It is therefore likely that glucose transport variation in A. japonica erythrocytes is under genetic control. This species therefore represents an intermediate situation between the hagfish, whose erythrocytes exhibit a high cytochalasin-B-sensitive sugar permeability (Ingermann et al. 1984), and M. albus and rainbow trout, whose erythrocytes are devoid of cytochalasin-B-sensitive transport activity. Erythrocytes from the brown trout are also likely to be transport-deficient (Bolis et al. 1971). Genetically determined transport polymorphisms have been identified in erythrocytes from a number of mammalian species. These include amino acid transport in sheep and horse erythrocytes (Young et al. 1975; Fincham et al. 1985), cation transport in sheep, goat, cattle, buffalo and dog erythrocytes (Ellory and Tucker, 1983) and nucleoside transport in sheep erythrocytes (Tucker and Young, 1988).

Cytochalasin-B-sensitive 3-OMG transport in A. japonica erythrocytes was shown to be saturable (apparent K_m 1·2–1·5mmoll⁻¹) and to be inhibited by other D-hexoses (but not by L-glucose) and by phloretin. In these respects the system resembles sugar transporters in both avian and mammalian erythrocytes. Direct transport experiments with D- and L-glucose provided further evidence of stereospecificity. As shown in Figs 3 and 8, transport variation among erythrocytes from different fish was a consequence of differences in V_m rather than K_m and presumably reflects differences in the cellular density of glucose transporter polypeptides. As in avian erythrocytes, the transporter was stimulated by catecholamines. Kinetically, catecholamines increased the V_m of transport without changing apparent K_m. The mechanism of transport stimulation by catecholamines remains to be investigated. Whitfield et al. (1974) reported that catecholamine stimulation of sugar transport in avian erythrocytes is associated with a lowering of intracellular ATP level. This effect is not mediated through α- or β- adrenergic receptors. Stimulation of sugar transport in avian erythrocytes can also be brought about by metabolic depletion (Whitfield and Morgan, 1973; Bihler et al. 1982, 1985; Simons, 1983a,b). Increased sugar transport activity under these conditions is unlikely to involve changes in cytosolic Ca²⁺ concentration (Simons, 1983b). ATP directly modulates glucose transport in human erythrocytes (Carruthers, 1986a, b).

Physiological catecholamine concentrations in fish plasma may rise ta 0·5-1·5 μmol 1⁻¹ during stress (Mazeaud and Mazeaud, 1981; Bourne and Cossins 1982). Such catecholamine levels would be sufficient to induce a half-maximal stimulation of glucose transport activity in A. japonica erythrocytes (Fig. 7), raising the possibility that erythrocyte glucose permeability may increase in vivo in response to prolonged stress. In contrast to trout cells (Bourne and Cossins, 1982), A. japonica erythrocytes do not swell in the presence of adrenaline or noradrenaline (D. A. Fincham and J. D. Young, unpublished observation). Hormone-induced changes in cytochalasin-B-sensitive glucose transport activity are not, therefore, a secondary consequence of cell volume changes.

Cytochalasin-B-insensitive sugar transport in fish (and avian) erythrocytes is partially stereospecific (Tables 3 and 4). It has been suggested that this phenomenon in avian cells reflects stereoselective diffusion of glucose across the lipid bilayer (Simons, 1983n). However, D- and L-glucose exhibit equivalent permeability coefficients in phospholipid vesicles (C. M. Tse and J. D. Young, unpublished observation). Similarly, pig erythrocytes exhibit equivalent rates for cytochalasin-B-insensitive transport of D- and L-glucose (Young et al. 1985). An alternative possibility is that a component of this flux in fish and avian erythrocytes represents stereospecific mediated transport via a minor cytochalasin-B- insensitive transport system. Cytochalasin-B-insensitive sugar transport has been observed in rabbit erythrocytes (Albert, 1984). Rates of cytochalasin-B-insensitive uptake of 3-OMG and D- and L-glucose in A. japonica erythrocytes were significantly higher than the corresponding fluxes in M. albus and the rainbow trout.

For M. albus, measured rates of D-glucose uptake were similar to those reported previously by us for mature pig erythrocytes (Young et al. 1985, 1986). For pig erythrocytes, this low glucose permeability is insufficient to maintain normal ATP:ADP ratios (Kim and McManus, 1971; Young et al. 1985). In contrast to mammalian erythrocytes, and by analogy with avian erythrocytes (Kregenow, 1975), fish erythrocytes have the potential to metabolize glucose via the tricarboxylic acid cycle (Greaney and Powers, 1978; Tetens and Lykkeboe, 1981). However, additional metabolic substrates might still be required. One potential candidate is inosine, the physiological energy source used by pig erythrocytes (Young et al. 1985, 1986). In preliminary experiments we have established that erythrocytes from both the common eel and M. albus are capable of rapid nitrobenzylthioinosine (NBMPR)-sensitive nucleoside transport (approximately 2×10⁵ nucleoside transporters per cell as determined by [³H]NBMPR binding).

This research was supported by the Cancer Research Campaign, UK, and by the Royal Hong Kong Jockey Club. We are grateful to Dr J. C. Ellory, University Laboratory of Physiology, Oxford, for providing laboratory facilities for the S. gairdneri experiments.