Acid-Base Regulation Following Acute Acidosis in Seawater-Adapted Rainbow Trout, Salmo Gairdneri: A Possible Role for Catecholamines

Strenuous exercise in fish often results in an extracellular acidosis of metabolic origin (Holeton, Neumann & Heisler, 1983; Turner, Wood & Clark, 1983). Recent studies on rainbow trout have shown that red cell pH is regulated at a constant level during and following periods of extracellular acidosis brought on either by anaerobic exercise (Primmett, Randall, Mazeaud & Boutilier, 1986) or by direct acid infusion (Boutilier, Iwama & Randall, 1986). In both instances catecholamines are released into the blood where, through β adrenergic mechanisms at the red cell membrane, they are responsible for turning on cellular ion exchange processes which ultimately lead to the regulation of the erythrocyte pH (Nikinmaa, 1982, 1983; Cossins & Richardson, 1985) and therefore blood oxygenation (Boutilier et al. 1986).

Several studies have shown that following exhaustive exercise a transient net H ⁺ flux occurs between the animals and the environmental medium and that the transfer takes place mainly across the gills (Holeton et al. 1983; Holeton & Heisler, 1983; Milligan & Wood, 1986). Accordingly, the present experiments set out to examine whether catecholamines play a significant role in proton equivalent flux across the gills. The experiments were carried out on seawater-adapted rainbow trout (Salmo gairdneri) subjected to acute extracellular acidosis by infusion of HC1. The possible role of the β-agonist adrenaline in regulating these bicarbonate equivalent movements was determined by pre-treating a second group of fish with propranolol (β-adrenergic antagonist) before acid infusion.

Seawater-adapted rainbow trout (Salmo gairdneri) of either sex, weighing between 150 and 300g, were obtained from Ocean Farmers Ltd, St Margaret’s Bay, NS. They were kept in one of the Aquatron seawater laboratories in a 4 m³ fibreglass tank supplied with a continuous flow of aerated sea water (7—12°C) for at least 3 weeks before the experiment. Fish were fed daily with commercial trout pellets, but feeding was suspended 2 days prior to experimentation.

Following anaesthetization in a 1:10000 seawater solution (pH 7·0) of MS-222 (Sigma), the dorsal aorta of each fish was chronically cannulated with Clay-Adams polyethylene (PE 50) tubing. Throughout all surgical procedures, the gills of the fish were continuously irrigated with a lower dose of the anaesthetic solution (1:15 000 MS-222). Fish were allowed to recover for at least 48 h in a darkened Plexiglas box of the water recirculation system described below. At all stages prior to experimentation, the animals were supplied with continuously flowing aerated sea water, kept at 10±0·5°C.

Animals were contained within a water recirculation system maintained at 10 ± 0·5°C. The water in the system was pumped (2·51 min⁻¹) through an aeration column (bubbled with air), through a Plexiglas box containing the fish, and then returned to the aeration column.

Water samples from the recirculation system were taken according to the time schedule of the experiment (see below) and used to measure water ammonia and total CO₂ concentrations. Measurement of total CO₂ in water was made after each sample had been equilibrated at a constant (1 % CO₂/N₂) and constant temperature (30±0·1 °C). The equilibration vessels in which the samples were contained were fitted with scintered glass bottoms for the introduction of humidified and temperature-regulated gases (Wosthoff, FRG). The elevated temperature ensured a more rapid equilibration time, reducing sample processing time. After complete equilibration (temperature and ⁠), samples were transferred to a 5 ml Hamilton gastight syringe for subsequent determination of total CO₂ concentration by a chromatography method detailed below. In principle, our system is the same as the ‘delta-bicarbonate’ system described by Heisler (1984, 1987), except that total rather than pH, is determined at a fixed and temperature. In both cases, bicarbonate concentrations can be estimated by known additions of solid bicarbonate to the recirculation system.

Parallel experiments were performed on three separate groups. The first group (N = 8) was subjected to acid infusion. The second group (N= 8) was subjected to both propranolol and acid infusion. The third group (N=6) was subjected to injection of the saline vehicle, and served as a control for handling and sampling effects.

The experiments with the first group focused on post acid-infusion changes in arterial plasma and red cell acid-base status, as well as net H⁺ flux with the environment (see below). One hour prior to acid infusion, the inflow to the fish box was closed, and the volume standardized to 11100 g⁻¹ body mass. Two water samples were taken 1 h apart for total CO₂ (at fixed ⁠) and ^ammonia measurements. Ten minutes before acid infusion, fish were infused through the dorsal aortic cannula with 0·35 ml 100g⁻¹ body mass of saline. At the time of acid infusion, fish were infused with 0·25 ml 100g⁻¹ body mass of 0·1 moll⁻¹ HC1 in 140mmoll⁻¹ NaCl, subsequently washed in with 0·1 ml 100 g⁻¹ body mass of 140mmoll⁻¹ NaCl. Infusion times ranged from 7 to 9 min. The time when the acid infusion finished was defined as time zero for all succeeding measurements. Arterial blood samples (500 μl) were taken before acid infusion and at 5, 30, 60, 120 and 240min post-infusion. The volume of blood sampled was replaced with heparinized (20i.u.ml⁻¹) saline. Portions of each blood sample were analysed for plasma pH (pHe), red cell pH (pHi), total CO₂ (in both whole blood and true plasma), and haematocrit. Water samples (10 ml) from the delta-bicarbonate system were taken at 0, 60, 120, and 240 min for determination of ammonia concentration and total CO₂ content.

The second group of animals was used to examine the effect of pre-treatment with propranolol on the above acid—base parameters after acid infusion. The experimental protocol was the same as in the first group except that fish were infused with 0·25 ml 100 g⁻¹ body mass of 2×l0⁻⁴moll⁻¹ propranolol in saline followed by 0·1 ml 100 g⁻¹ body mass of saline to clear the cannula of propranolol 10 min before acid infusion.

The third group acted as a control. The same experimental protocol (as the first group) was employed except that fish were infused with 0·35 ml 100 g⁻¹ body mass of 140 mmol I⁻¹ NaCl alone at the time of acid infusion.

Water ammonia concentration was determined by a micro-modification of the salicylate-hypochlorite method of Verdouw, van Echteld & Dekkers (1978). The net fluxes of ammonia and bicarbonate (in μmol kg⁻¹ h⁻¹) were calculated from changes in their respective concentrations in water. Net H⁺ flux was determined as the difference between the net ammonia flux and net bicarbonate flux (delta ammonia minus delta bicarbonate, signs considered).

The method used to determine H⁺ movements in the present study, as with other similar methods (reviewed by Heisler, 1984, 1987), cannot separate H⁺ uptake from HCO₃ excretion, or vice versa, or between ammonia movement in the NH3 or NH₄⁺ forms. Rather, it gives a valid measure of net acidic equivalent flux. Various descriptive terms to denote such H⁺ movements are used interchangeably in the current literature (e.g. acidic equivalent flux, net H⁺ flux, bicarbonate equivalent flux, amongst others) and should not be mistaken for statements about the mechanisms involved in transepithelial ion transfers.

Means ± 1 S.E.M. are reported throughout. Student’s two-tailed t-tests (paired design) were used to assess significant differences (P<0·05) within groups, using each fish as its own control. A non-paired /-test was used for between-group comparisons.

Intra-arterial infusion of HC1 resulted in a pronounced arterial blood acidosis, with plasma pH maximally depressed 5 min after infusion (Fig. 1A). The recovery of blood pH following acid infusion was rapid, taking only 60 min to reach pretreatment control levels. When experiments were terminated, after 240 min of recovery, blood pH was slightly greater than the pre-infusion value. The infused acid load led to a mixed respiratory and ‘metabolic’ acidosis, as indicated by the rapid elevation of (Fig. 2) and the marked depression of plasma [HCO₃⁻] (Fig. 3A). The metabolic component dissipated quickly, as plasma [HCO₃⁻] returned to rest levels by 30 min. However, over the same period the respiratory component of the acidosis prevailed, as reflected by the significantly higher than normal levels (Fig. 2). The recovery of blood pH over 60 min could be attributed primarily to the continuing elevation of plasma [HCO₃⁻]. At 60, 120 and 240 min post-infusion, plasma [HCO₃⁻] was significantly elevated above the resting level, causing the observed alkalosis at these times (Figs 1,3), despite the slow branchial washout of CO₂ (Fig. 2). The pH-[HCO₃⁻] diagram (Fig. 4) illustrates the partitioning of the initial acidosis and the subsequent mode of recovery.

Following acid-infusion, the small but associated perturbations in red cell pH and [HCO₃⁻] were not significantly different from pre-treatment values (Figs IB, 3B).

Propranolol injection followed by acid infusion caused an extracellular acidosis similar in magnitude to that seen following acid infusion alone. However, the recovery of pHe in β-blocked animals was significantly slower than in the untreated animals (Fig. 1A), taking 4h to reach pre-treatment levels (3 h more than the untreated animals). This prolonged recovery was also evident in the restoration of plasma bicarbonate concentrations in β-blocked animals (Fig. 3A). Associated with the decline in pHe was a sharp rise in similar to the pattern seen in untreated animals (Fig. 2). The magnitude of the changes in β-blocked animals were slightly higher than in animals infused with acid alone (Fig. 4). The extracellular acidosis in β-blocked animals induced a significant drop in red cell pH (Fig. 1A), falling 0·16 units below that of pre-infusion levels 5 min following acid infusion, and gradually returning to control values within 2h. Associated with the red cell disturbance was a similar change in red cell [HCO₃⁻] (Fig. 3B) which was depressed significantly 5 min following acid infusion and, thereafter, gradually increased over the remainder of the experimental recovery period.

Control experiments were conducted to examine the influence of the saline vehicle on blood and water parameters. These experiments were performed under conditions identical to those for the experimental groups. Of the variables that were measured, none showed significant changes over the entire experimental period (Figs 1, 2, 3).

In both the control group and the pre-infusion period of the experimental group, a net H⁺ excretion of 17–103 μmol kg⁻¹ h⁻¹ occurred as a result of net ammonia release at a rate of 270—327μmol kg⁻¹h⁻¹ and net bicarbonate release (OH⁻ release, or equivalent H⁺ uptake) of 167—262 μmol kg⁻¹ h⁻¹. These rates did not change significantly following the sham injection in the control group (Fig. 5A). Upon infusion of acid, however, net H⁺ excretion rose significantly in the first 2h (Fig. 5B). The response consisted of an immediate significant increase in net bicarbonate influx and a small but non-significant increase in ammonia efflux (Fig. 5B). Acid infusion following propranolol treatment induced similar changes in the above parameters (Fig. 5C). However, as indicated by an unpaired Student’s Z-test of the corresponding values, propranolol-treated fish excreted significantly less H⁺ (as compared with untreated fish) in the first hour following acid infusion (P<0·05). This difference in H⁺ efflux was primarily due to the discrepancy in bicarbonate equivalent flux between the two groups (Fig. 5B,C).

Acute extracellular acidoses, whether of endogenous or exogenous origin, cause an increase in the circulating levels of catecholamines in salmonid fish, with the release of the catecholamine adrenaline being proportional to the fall in blood pH (Primmett et al. 1986; Boutilier et al. 1986). It has also been repeatedly observed that the erythrocyte pH of certain teleosts increases when the cells are incubated in the presence of adrenaline (reviewed by Nikinmaa, 1986). Indeed, red cell pH is regulated at a constant level in vico during periods of acute extracellular acidosis brought on by exercise (Primmett et al. 1986), acid infusion (Boutilier et al. 1986), hypercapnia (Perry, 1986) and hypoxia (Boutilier, Dobson, Hoeger & Randall, 1987). According to the relationship between the change in plasma adrenaline and the change in blood pH reported by Boutilier et al. (1986), the acidosis in the present study (0·25 and 0·27 units drop in mean blood pH) should have promoted an approximately 10-fold increase in circulating adrenaline levels.

The exogenously induced extracellular acidosis of the present experiments was not transmitted to the erythrocyte, whose pH was regulated instead at a constant level in the face of changes in plasma pH (Fig. 1). As indicated by the decline in erythrocyte pH during the acute plasma acidosis of animals treated with propranolol, the regulation of red cell pH appears to be via β-adrenergic mechanisms.

The titration of blood bicarbonate immediately following the acid infusion resulted in a decline in plasma [HCO₃⁻] and an associated rise in (Figs 2, 3, 4). A short-lived respiratory acidosis following strenuous exercise is a phenomenon observed in all fish species examined to date (reviewed by Wood & Perry, 1985). Heming (1984), Wood & Perry (1985) and Perry (1986) reported that β-adrenergic stimulation inhibits bicarbonate flux through trout red cells in vitro, and that this could, under certain circumstances, lead to a transient reduction or inhibition of CO₂ excretion in vivo. Since adrenaline is released to the circulation during acute acidoses (Boutilier et al. 1986), it follows that the acid-infused animals in the present study might be expected to have incurred an even greater respiratory acidosis than their /3-blocked counterparts. On the contrary, our data indicate that the respiratory acidosis in the propranolol-treated group is slightly greater than that of the untreated acid-infused animals (Fig. 2). Indeed, the sustained elevation appears to play an important role in the much longer term acidosis of μ-blocked animals (Fig. 1). Similar results between propranolol versus untreated animals have also been observed in seawater-adapted trout following exhaustive exercise (Y. Tang & R. G. Boutilier, unpublished results). Certainly, these data indicate that adrenaline probably enhances overall CO₂ excretion as demonstrated by the comparatively greater build-up of an the propranolol-treated animals (Fig. 2). Although many studies have shown that elevated levels of adrenaline stimulate O₂ uptake across the gills of fish (Petterson, 1983; Perry, Payan & Girard, 1984; Peyraud-Waitzenegger, 1979), far less is known about CO₂ exchange. Even so, factors such as adrenaline which enhance O₂ uptake should simultaneously favour CO₂ flux in the opposite direction. To attribute such effects to the actions of the red cells themselves would be over simplistic. Clearly, several limitations apply to whole-animal studies of this kind and the conclusions drawn must be rather speculative. For instance, administration of a β-blocker to an intact animal may evoke numerous effects, including changes in ventilation, cardiac output, patterns of blood flow through the gill, or modulation of membrane/epithelial permeabilities and/or transporter functions (Randall & Daxboeck, 1984; Evans, 1986). Any or all of these effects may modify the response of the animal and lead to the differences in an pH observed between β-blocked and unblocked animals (Figs 1—5).

Proton equivalent exchanges with the environment have been examined in freshwater trout following exercise (Holeton et al. 1983; Milligan & Wood, 1986), during continuous adrenaline infusion (Vermette & Perry, 1987) and in seawater-adapted fishes subjected to acid infusion (Evans, 1982; McDonald, Walker, Wilkes & Wood, 1982; Fig. 5) or hypercapnia (Evans, 1982; Toews, Holeton & Heisler, 1983).

All these studies showed that net H⁺ efflux increased in the face of an extracellular acidosis. Following acid infusion in the present study, trout exhibited a minor but non-significant increase in ammonia excretion and a reversal of bicarbonate flux from net excretion to net uptake, the combined effects of which resulted in an increase in net H⁺ efflux (Fig. 5B,C). In freshwater trout, large increases in net H⁺ efflux following exercise could be attributed mostly to a rise in ammonia excretion (Milligan & Wood, 1986). The minor contribution of ammonia to net H⁺ efflux in the present study may be related to the mode of induction of the acidosis. The exhaustive exercise protocol in Milligan & Wood’s study is likely to have induced an increase in nitrogen metabolism in white muscle (Dobson & Hochachka, 1987), resulting in an elevation of ammonia excretion. In the present study, however, the acidosis was exogenously induced, and increased nitrogen metabolism is unlikely under these circumstances. Since excretion of ammonium ions is thought to be performed by carrier-mediated ion transfer mechanisms in exchange for Na⁺ (Maetz, 1973), an elevated NH₄⁺ excretion in a marine fish would add to the osmoregulatory load of the Na⁺ extrusion mechanism. Our results are similar to those of Toews et al. (1983), who found a hypercapnia-induced acidosis in the marine teleost fish Conger conger was partially compensated by a net uptake of bicarbonate from the sea water, with no increase in ammonia excretion.

The relative importance of gills and kidneys in acid-base regulating processes is not yet well established for fishes. It may be related to species differences as well as to the nature of the acid—base disturbance. For freshwater rainbow trout, Vermette & Perry (1987) reported an increased branchial acid efflux and renal acid excretion during continuous adrenaline infusion, accounting for 85 % and 15 % of total acid efflux, respectively. Kobayashi & Wood (1980) found that only a few percent of an infused lactate load was excreted renally. Wood & Caldwell (1978), however, reported that all of an infused HC1 load was excreted renally. Although renal acid efflux was not examined in the present study, considering the relatively small throughput of urine in marine teleosts, a significant renal contribution to acid excretion in our seawater-adapted fish would not be expected. Therefore, the results presented here (Fig. 5) suggest an important role for adrenaline in facilitating branchial extrusion of acid equivalents to the external environment. Lack of a renal response to acid challenge has been reported for the marine teleost Parophrys ve lulus (McDonald et al. 1982) and a marine elasmobranch Scyliorhinus stellaris (Heisler, 1980). Evans (1982) has also reported that all the H⁺ efflux subsequent to mineral acid loading is via branchial pathways in the marine teleost Opsanus beta.

Our results indicate that the elevation in net H⁺ excretion following acid infusion is facilitated by adrenaline, as demonstrated by the 37 % reduction in net H⁺ efflux in the propranolol-treated animals (Fig. 5). To calculate the net clearance of proton equivalents from acidotic fish to environmental water, we assumed the average H⁺ excretion of the sham-treated animals (Fig. 5A) to be representative of basal H ⁺ excretion following infusion. Therefore, for each interval concerned, the ‘sham’ values were subtracted from the corresponding post-infusion measurements in each treatment group (i.e. Fig. 5B,C). When this was done, we found that the fish apparently ‘overcompensate’ for the infused acid load. Thus, whereas 25 μmol 100 g⁻¹ of acid was infused, 86μmol 100 g⁻¹ was excreted in the 0·4 h post-infusion period in the acid-infused fish and 54μmol 100g⁻¹ in those treated with propranolol. Whether this excess H⁺ clearance represents the stimulation of a secondary metabolic acidosis or is related to a bicarbonate uptake mechanism is not yet known. Regardless, the overcompensation of the infused acid load is not observed in all species.

In the freshwater catfish, Ictalurus punczatus, only 20—30% of an infused acid load was subsequently cleared by the gills over 2 h ; after that time there was little further excretion (Cameron, 1980). Such results imply that the acid load can be buffered intracellularly and slowly cleared from the body over a prolonged period. However, the amount of H⁺ excreted over the first 4h following acid infusion in the marine species Opsanus beta and Squalus acanthias was approximately twice that of the infused load (Evans, 1982), similar to the results presented here. McDonald et al. (1982) also found that the marine teleost, Parophrys vetulus, was able to clear an infused acid load by 5 h in one group and by 16 h in another. In the present study, the continuation of acid excretion beyond the point at which the extracellular acidosis is corrected (Fig. 5,versusFig. 1) evidently accounts for the eventual overshoot of arterial pH (Fig. 4) by 2—4h post-infusion. It is noteworthy that the overshoot only occurred in the unblocked acid-infused animals (Fig. 4). Such overcompensation may be aimed at restoring bicarbonate concentrations in all parts of the CO₂ storage system (e.g. Fig. 3).

There are two possible ways that adrenaline could effect branchial H⁺ efflux. First, adrenaline may affect the ionic exchange systems. Beta-adrenergic stimulation of branchial Na⁺/H ⁺ (NH4⁺) exchange with simultaneous inhibition of CI /HCO₃⁻ exchange has been reported for the isolated perfused head preparation of freshwater trout (Payan & Girard, 1978; Perry et al. 1984). In a recent study, Vermette & Perry (1987) found that adrenaline infusion in freshwater trout caused a significant elevation of the arithmetic difference between Na⁺ and Cl⁻ net fluxes, and that a significant correlation existed between this difference and net branchial H⁺ excretion. This kind of effect is more difficult to approach experimentally in seawater-adapted fish. Second, adrenaline may affect the ionic fluxes by modulating gill permeability. For example, Isaia, Maetz & Haywood (1978) demonstrated that there was an increase in gill permeability to water in response to catecholamines. This indicates that there may be a general increase in permeability to ions as well.

Since transepithelial ion exchange mechanisms at the gill are the primary way in which fish regulate acid—base status (Heisler, 1984), their utility in responding to an acidosis may be limited by the concentrations of Na⁺ and Cl⁻ in the external medium. Obviously, marine fishes are not faced with such limitations. Comparisons of pH compensation following environmentally induced hypercapnic acidoses in marine fish (Toews et al. 1983; Heisler, Weitz & Weitz, 1976) and freshwater trout (Eddy, Lomholt, Weber & Johansen, 1977; Janssen & Randall, 1975; Perry, Haswell, Randall & Farrell, 1981) show that pH compensation takes longer and is less complete in fresh water than in water of greater ionic strength or sea water. Recently, Iwama (1986) also demonstrated that water concentrations of Na⁺ and Cl⁻ have a positive effect in correcting the extracellular acidosis in trout acclimated to higher salinities as well as in marine conger eel conditioned to dilute waters. It is evident that this phenomenon also occurs in pH regulation of trout subjected to acid infusion (Fig. 6). In the present study with seawater-adapted trout, pHe recovered by 1 h following acid infusion. However, in a very similar investigation with freshwater-adapted trout (Boutilier et al. 1986), the restoration of pHe took at least 2h. The faster pH compensation in the seawater-adapted animals might also be explained by the difference in concentrations of Na⁺ and Cl⁻ in external water as discussed above. However, for seawater-adapted fish, additional factors may be involved in this faster pH compensation. First, sea water contains higher concentrations of bicarbonate than fresh water. This buffer pool may provide bicarbonate for fish during acidosis via a 180°-rotated C1⁻/HCO₃⁻ exchange system across the gill epithelium of marine fish. This rotated CH/HCO₃⁻ exchange is indicated by the finding that Cl⁻ efflux from the marine toadfish is increased by increasing external [HCO₃⁻] (Kormanik & Evans, 1979). Second, since the branchial epithelium of marine fishes is decidedly more permeable to ions than that of freshwater fishes (Evans, 1979), the relatively higher permeabilities for marine fishes may necessitate that their ionic exchange systems should be more effective than those of their freshwater counterparts. In addition, the hyporegulating marine fish probably faces greater ionoregulatory pressures than the hyper-regulating freshwater fish, since the osmolality of fish plasma is about one-third that of sea water. Therefore, marine fish may be better equipped to regulate those ions which are related to acid-base regulation.

YT was supported by an Izaak Walton Killam Memorial Scholarship, an International Centre for Ocean Development Scholarship, and a Dalhousie University Graduate Scholarship. This study was supported by a Natural Sciences and Engineering Research Council of Canada operating grant to RGB.