Acid-Base Regulation Following Exhaustive Exercise: a Comparison Between Freshwater-and Seawater-Adapted RainbowTrout (Salmo Gairdneri)

Exhaustive exercise in fish leads to a marked extracellular acidosis of respiratory and metabolic origin. The nature, severity and duration of this acid-base disturbance have been well characterized (see reviews by Wood & Perry, 1985; Heisler, 1986). Associated with this extracellular acidosis is net H⁺ equivalent excretion to the environmental water which is thought to occur as a result of combined NH₄⁺ and H⁺ equivalent excretion (or HCO₃⁻ and OH⁻ uptake) (Holeton & Heisler, 1983; Holeton et al. 1983; Milligan & Wood, 1986a,b). Net H⁺ efflux has also been demonstrated in fishes subjected to extracellular acidoses induced by environmental hypercapnia (Toews et al. 1983) and mineral acid infusion (McDonald et al. 1982; Evans, 1982; Tang et al. 1988), and is considered to be achieved primarily by branchial HCO₃⁻/C1⁻ and/or H⁺(NH₄⁺)/Na⁺ exchange mechanisms (Evans, 1986; Heisler, 1986).

Since transepithelial ion exchange mechanisms at the gill are central to the regulation of acid–base status in fish (Heisler, 1984), their utility in responding to an acidosis may be limited by the concentrations of the corresponding counter-ions in the environmental water. Certainly, marine fishes are not faced with such limitations. In addition, the branchial epithelium of marine fishes is decidedly more permeable to ions than that of freshwater fishes (Evans, 1979). Moreover, recent studies indicated that, following acid infusion, seawater-adapted rainbow trout were able to compensate their blood pH faster than freshwater-adapted trout (Boutilier el al. 1986; Tang et al. 1988).

Few studies on acid–base regulation in actively exercising fish have been carried out in which intraspecific comparisons could be made between freshwater and seawater animals. The marine fishes which have been studied are either inactive benthic forms which seldom perform exhaustive exercise (e.g. flounder; Wood & Milligan, 1987), or species for which no freshwater counterpart exists (e.g. dogfish; Holeton & Heisler, 1983). In the present study, dealing with blood acid–base status and branchial ionic fluxes following exhaustive exercise, a comparison was made between freshwater-(FW) and seawater-(SW) adapted rainbow trout of the same genetic stock. Rainbow trout are particularly well suited for such comparison since they are active pelagic fish which can be readily adapted to either sea water or fresh water.

Freshwater rainbow trout (Salmo gairdneri), weighing 170–300 g, were obtained from Merlin Fish Farms, Wentworth, Nova Scotia. After 1 week of acclimation to dechlorinated Halifax City tapwater, they were unselectively divided into two stocks. One stock remained in fresh water ([Na⁺] = 0·3 mequiv l⁻¹; [Cl⁻] = 0·2 mequiv l⁻¹ ; [HCO₃⁻] = 0·5 mmol l⁻¹ ; pH 7·5–7; 6–9°C) as a freshwater-adapted group. The other stock was adapted to filtered sea water (32 %o; [Na⁺] ≈ 470 mequiv l⁻¹ ; [Cl⁻] ≈ 540 mequiv l⁻¹; [HCO₃⁻] = 2·2 mmol l⁻¹; pH 7·9; 8–10°C) supplied by the Aquatron Laboratory of Dalhousie University. To minimize stress, the turnover from fresh water to sea water was performed gradually and took 10 days to reach 100% seawater strength. Fish of both stocks were kept in 4 m³ fibreglass tanks supplied with a continuous flow of aerated water for at least 6 weeks before the experiment. They were fed daily with commercial trout pellets, but feeding was suspended 4 days prior to surgery.

One week before surgical operation, fish were transferred to a 560-1 living stream tank (Frigid Unit Inc., USA). The water in the tank was automatically recirculated, aerated and thermostatted to 10 ± 0·2°C. Following anaesthetization in a 1:10000 seawater solution (pH 7·0) of MS-222 (Sigma) or a 1:10000 freshwater solution of MS-222 buffered to pH 7·0 with NaHCO₃, catheters (Clay-Adams PE50) were surgically implanted into the dorsal aorta using the method of Smith & Bell (1964). Catheters were filled with heparinized (20i.u. ml⁻¹) Cortland saline (Wolf, 1963) for FW fish, or marine teleost saline (Hoar & Hickman, 1983) for SW fish. Following surgery, fish were placed into 2·8-1 darkened Lucite flux boxes supplied with flowing aerated water from a thermostatted living stream tank (10 ± 0·2°C) and allowed to recover for at least 48 h prior to experimentation. The flux boxes were designed so that vigorous aeration of an outer chamber provided convective mixing of water throughout an inner animal chamber (see McDonald & Rogano, 1986, for details).

Post-exercise changes in arterial acid–base status were measured in 11 fish. The experiment began by closing the inflow and outflow of the flux boxes so that the system contained a fish within a standardized volume of water (approximately 121 kg⁻¹ fish). The temperature was maintained by bathing the box with thermostatted water (10 ± 0·2°C). Control blood samples were drawn anaerobically from the dorsal aorta catheter for analysis of true plasma pH (pHe), and whole blood total CO₂ and lactate concentrations (see below for analytical procedures). Each fish was then subjected to exhaustive exercise; manual chasing around a circular tank (90 litres, diameter 70 cm) for 10 min. At this point it was incapable of further burst performance but still able to swim slowly around the tank. Each fish was then immediately returned to its individual box and monitored for the following 4h. Blood samples for pHe, whole blood total CO₂ and lactate were taken at 10, 40, 70,130 and 250 min post-exercise. The volume of blood sampled was replaced with heparinized (20i.u. ml⁻¹) saline.

For measurement of the pre-and post-exercise changes in net H⁺ equivalent flux with the environment, water samples (18 ml) were taken over the 1-h period just before exercise (resting flux) and at 0,1, 2 and 4h post-exercise. These water samples were subsequently measured for titratable alkalinity and ammonia concentrations. These measurements, together with the above sampling schedule, allowed us to calculate the net H⁺ flux, in terms of μmol kg⁻¹ h⁻¹, during the periods before exercise and over the 0–1,1–2 and 2–4 h post-exercise periods (see below for analytical procedures and calculations).

A genetically identical group of seven SW trout was treated to the same experimental protocol and acid–base measurements as the FW trout described above. Net H⁺ equivalent flux was determined using a similar protocol to that of SW trout, except that total CO₂ (at a fixed ⁠) was measured instead of titratable alkalinity.

Arterial pH was determined using a microcapillary pH electrode (Radiometer G279/G2) coupled with a PHM84 pH meter. The total CO₂ of anaerobically obtained whole blood was measured on 50 μl samples using a gas chromatography method (Boutilier et al. 1985) in SW fish, and using a total CO₂ analyser (Corning model 905) for FW fish. Measurements of pHe and whole blood total CO₂ were used to calculate arterial CO₂ tension (⁠ and whole blood bicarbonate concentration ([HCO₃–]_wb) using experimentally determined pK′ and CO₂ solubility coefficients for trout blood (Boutilier et al. 1984). This method (i.e. using whole blood instead of true plasma) will result in slight errors; Paco₂ will be consistently underestimated by 5–10% and [HCO₃⁻]_wb overestimated by less than 0·5%. For the purposes of calculating the metabolic proton load (below), such differences are minor. Whole blood lactate concentrations were analysed using the L-lactate dehydrogenase/NADH method (Loomis, 1961) using Sigma reagents.

Mean values ± 1 S.E.M. are reported throughout. Results were analysed statistically using paired or unpaired Student’s t-tests between sample means, 5 % being taken as the fiducial limit of significance.

Exhaustive exercise in FW trout caused an acute extracellular acidosis (Fig. 1), the magnitude and duration of which were similar to that of other analogous studies (see reviews by Wood & Perry, 1985; Heisler, 1986). In contrast to their FW counterparts, the same exercise protocol in SW trout produced a markedly smaller blood acid-base disturbance. Ten minutes after exercise, changes in blood acid-base variables were approximately half those of FW trout (Fig. 1). Subsequent restoration to pre-exercise levels in SW trout took about half as long as in their FW counterparts; i.e. time to full recovery for pHe, 130min (SW) vs 250 min (FW); [HCO₃⁻]_wb, 130min (SW) vs 250min (FW); ⁠, 70min (SW) vs 130min (FW).

In both FW and SW trout, whole blood lactate concentrations rose rapidly from resting values of about 0·5 mmol l⁻¹ to approximately 8 mmol l⁻¹ at 10 min and about 10 mmol l⁻¹ at 40 min post-exercise (Fig. 1D). Thereafter, lactate levels in SW trout continued to increase, albeit slowly. In contrast, lactate levels in FW trout rose to a peak at 70 min, and slowly declined thereafter. As a result, by 130min following exercise, lactate concentrations in SW trout were significantly higher than those in their FW counterparts (Fig. ID).

The calculated blood metabolic proton load (ΔHm⁺), blood lactate load ((Δ laciate) and the ‘H⁺ deficit’ ((Δ laciate − ΔHm⁺, signs considered) following exercise are illustrated in Fig. 2. At rest, ΔHm⁺ is equal to zero by definition. Differences in the levels of blood (Δ laciate and ΔHm⁺ at any time resulted in a large H⁺ deficit, especially in SW trout (Fig. 2B). In FW trout, a negative H⁺ deficit developed immediately following exercise. Thereafter, the H⁺ deficit increased in a sigmoidal fashion. A negative H⁺ deficit was never observed in the SW animals. Instead, a positive H⁺ deficit developed immediately after exercise, and increased in a sigmoidal fashion to values which were markedly higher than in FW trout (Fig. 2B).

During the period prior to exercise, net ammonia flux and apparent net base flux in FW trout were higher than the corresponding measurements in SW trout; however, the net H⁺ exchange with the environment was not significantly different between the two groups (Fig. 3); i.e. −74 ± 92 μmol kg⁻¹ h⁻¹ (FW) and −31 ± 50 μmol kg⁻¹ h⁻¹ (SW). During the period of exercise, it was not possible to follow H⁺ equivalent excretion owing to the short time (10 min) and the large volume of water required to exercise the fish.

During the first hour after exercise in FW trout, there was a small but significant increase in net H⁺ equivalent excretion (212 μmol kg⁻¹ h⁻¹) due to an approximate doubling of ammonia excretion without much change in apparent base flux.

Ammonia excretion continued at this elevated level for 1–2 h and thereafter declined towards resting levels over the 2–4 h period of recovery. Apparent base excretion was unchanged over the first hour following exercise, but decreased slightly during the 1–2 and 2–4 h periods (Fig. 3). As a result, net H⁺ excretion was maintained at 225–344 μmol kg⁻¹ h⁻¹ during these periods.

In SW trout, net ammonia excretion significantly increased by a factor of 1·5 during the first hour following exercise, and remained at this level over the following experimental periods (Fig. 3). The exercise-induced increase in ammonia excretion in SW trout was smaller than in FW trout. However, the apparent base flux in SW trout was markedly changed from a pre-exercise net excretion to a net uptake. As a result of the significant increase in both ammonia excretion and apparent base uptake, the net H⁺ excretion in SW trout was approximately five times that of FW trout over the 4-h period following exercise (Fig. 3).

The changes in blood acid-base status observed following exhaustive exercise in both FW and SW trout were similar to those reported for other teleosts (see Wood & Perry, 1985; Heisler, 1986). The initial decreases in arterial pH were the result of both ‘respiratory’ and ‘metabolic’ contributions (Fig. 1). In both groups, the increase in can be attributed to a greater aerobic metabolism in addition to the buffering of metabolic protons by bicarbonate. It is clear from these data that SW trout are more proficient than FW trout at overCorning extracellular acid-base disturbances caused by exhaustive exercise (Fig. 1). Since both groups of animals were from the same genetic stock and underwent the same exercise protocol, any differences in the magnitude of the acid-base disturbance and in its time course to recovery must be attributable to modulation of internal structures and/or regulatory mechanisms by the different properties of environmental water.

Different amounts of work and therefore of total lactate production could account for some of the differences between the FW and SW trout. Although the extent of work was not directly quantified, all animals were subjected to exactly 10 min of manual chasing, by which time all had reached a stage where it was impossible to elicit any further (anaerobic) burst performance. Moreover, if the large acidosis in FW animals was simply a reflection of greater lactate production, this was not obvious from the measurements of lactate concentrations in blood (Fig. 1). Another possibility for differences in the magnitudes of the acidoses could arise if there were large differences in the apparent buffering capacity between FW and SW animals. It is unlikely that such differences can be attributed to unique blood buffering capacities, since the resting bicarbonate concentrations and haematocrits were not significantly different between FW and SW animals. Although the intracellular compartments are undoubtedly important in buffering the endogenous acidosis of exhaustive exercise (Milligan & Wood, 1986a,b,Heisler, 1986), there is no reason to suspect that these compartments behave so differently between the two groups of animals. Certainly, further work is needed to establish whether intraspecific differences in scope for activity, buffering capacity, and the like, can arise as a result of acclimation to different environmental salinities.

It is clear from our data that the greater excretion of H⁺ equivalents in SW trout (Fig. 3) is reflected by a more rapid clearance of the metabolic proton load (Fig. 1) as well as a greater H⁺ deficit (Fig. 2). Thus, ‘storage’ of protons in the external environment may play a more important role in facilitating pHe recovery in SW than in FW trout. Assuming that the pHi and tissue buffering capacity measurements of similarly treated FW rainbow trout (Milligan & Wood, 1986a,b) are comparable to those in these experiments (i.e. that there are no large FW to SW differences in total lactate production or internal buffering), it is possible to calculate what fraction of the net whole body H⁺ production (due to glycolysis and ATP degradation) would be ‘buffered’ by cumulative flux of acidic equivalents to the environment. Such analysis shows that less than 10% of the total load is transferred to the environment in FW compared with nearly 40% in SW. This comparatively greater H⁺ excretion in SW animals (Fig. 3) must, at least in part, account for the more rapid recovery of pHe in SW than in FW trout (Fig. 1). The basis for such FW and SW differences in net H⁺ equivalent excretion between animal and environment almost certainly resides with the gills. Although the principal sites for ion and acid-base regulation in fish include both gills and kidneys, the latter usually contribute relatively little to the overall transfer of acid-base relevant ions. Indeed, the maximum contribution of the kidney in FW fishes is usually less than 6 % under various stress situations (see review by Heisler, 1984). Moreover, in SW fishes the contribution is even less (<1 %) owing in large part to the comparatively small output of urine in these animals (Evans, 1982; McDonald et al. 1982; Heisler, 1980).

Differences in branchial H⁺ equivalent transfer between animal and environment could occur for several reasons, including dynamic modulation of active ionic exchange mechanisms, changes in passive permeability and/or the availability of external ions. The gills of FW fishes are important sites of ionic and acid-base regulation owing to the coupling of Na⁺ uptake to acidic equivalent (H⁺ or NH₄⁺) excretion and Cl⁻ uptake to basic equivalent (HCO₃⁻ or OH⁻) excretion (see reviews by Heisler, 1984, 1986; Wood & Perry, 1985). Recent studies have indicated that these same ionic exchange mechanisms may also occur in the branchial epithelium of some SW fishes, even with the obligatory NaCl load that these hyporegulating animals must already endure (Evans, 1986; Heisler, 1986).

Dynamic modulation of Na⁺ and Cl⁻ net fluxes can be achieved not only by branchial ionic exchange mechanisms, but also by passive diffusion of Na⁺ and Cl⁻ as a function of their respective electrochemical gradients and the transepi-thelial permeabilities. Certainly, the branchial epithelium of SW teleosts is regarded as being much more permeable to ions than that of FW teleosts (reviewed by Evans, 1979). For example, the unidirectional efflux of Na⁺ in SW rainbow trout was found to be 10 times that of their FW counterparts (Greenwald et al. 1974), and similar differences in flux rates (for both Na⁺ and Cl⁻) were also found when comparing SW and FW Atlantic salmon, Salmo salar (Potts et al. 1970). Larger efflux rates for Na⁺ and Cl⁻ in sea water may indicate a greater potential for modulation of net acidic equivalent flux, and could serve to explain why proton equivalent excretion in SW trout is five times that of the FW animals (Fig.3).

The greater acidic equivalent flux in SW trout may also be correlated with the greater availability of counter-ions (including HCO₃⁻) in the external medium. For example, recent studies on rainbow trout have shown that net H⁺ excretion can be increased by approximately 50% by raising the external concentrations of NaCl from 0·3 to 1·3 mmol l⁻¹ (D. G. McDonald, Y. Tang & R. G. Boutilier, unpublished results). Furthermore, in hypercapnic-exposed rainbow trout, extracellular pH compensation was facilitated when animals were exposed either to bicarbonate-enriched fresh water (Perry et al. 1981) or to progressive increases in external salinity at constant [HCO₃^-] (G. K. Iwama & N. Heisler, personal communication). In the present study, our sea water contained more bicarbonate than our fresh water (4·4 times that in FW) and could, therefore, have provided the SW animals with a greater capacity for C1⁻/HCO₃⁻ exchange. The greater acidic equivalent net flux in SW animals in the present investigation might also be related in part to the slightly more favourable conditions set up by the pH of sea water (7·9) compared with fresh water (7·6).

In conclusion, the more rapid recovery of acid-base status following exhaustive exercise in SW than in FW trout appears to be facilitated by greater branchial H⁺ equivalent excretion to the environmental water. It is not yet possible to determine whether the basis for these differences resides with changes in gill epithelium function or with the water composition itself.

This study was supported by NSERC operating grants to RGB and DGM and NSERC infrastructure support for the Aquatron Laboratory at Dalhousie University. YT is the recipient of an Izaak Walton Killam Memorial Scholarship, and an International Centre for Ocean Development Scholarship.