Acclimation To an Acid Medium in the Brown Trout Salmo Trutta

Long-term exposure of fish to acid media and its physiological effects have been studied by relatively few authors and in the majority of studies attempts to demonstrate acclimation to such media have been restricted to observations of changes in resistance times. Few, if any, of the experiments performed by these authors revealed any significant increase in resistance times of brook trout (Robinson et al. 1976; Falk & Dunson, 1977; Swarts, Dunson & Wright, 1978) or rainbow trout (Lloyd & Jordan, 1964) previously exposed to a non-lethal but low pH. In a few cases, resistance times of exposed groups were shorter than for controls. Reasons for the failure to show acclimation to low pH could be ascribed to short experimental duration, an acclimation pH that was too acidic or even to a build-up of ammonia, which may all contribute to the poor survival of fish in many of these experiments. However, Trojnar (1977) found that the percentage survival of trout fry in acid media was higher for fry incubated in acid media than for those reared in neutral media. This differential survival was attributed to acclimation.

Vaala (1971) demonstrated various haematological changes suggestive of acclimation in brook trout exposed to low pH media, including an increase in haematocrit and haemoglobin levels. It would seem, therefore, that although acclimation to low in fish has not so far been rigorously demonstrated, there is strong evidence to suggest that acclimation does occur.

The sensitivity of fish to acid environments has been studied by a number of researchers and differences in acid sensitivity have been shown between strains or the same species (Dunson & Martin, 1973; Robinson et al. 1976; Falk & Dunson, 1977; Swarts et al. 1978). Although different strains of brook trout, Salvelinus fontinalis, were found by these authors to vary in their time of survival in lethally low pH environments, the reasons for this are not clear and attempts to enhance acid resistance by acid acclimation have so far proved inconclusive.

In a previous paper (McWilliams & Potts, 1978) it was demonstrated that shortterm exposure of brown trout, Salmo trutta, to acid media produced significant changes in sodium fluxes and gill potentials. In acid media, positive changes in gill potentials could largely account for the increased loss of sodium from fish observed in these conditions. These observations were made on fish not previously exposed to acid media. The present paper presents the results of similar experiments performed on brown trout, together with haematological studies, following long-term exposure of fish to an acid medium. The principal aim of this study is to demonstrate low pH acclimation in brown trout in terms of physiological changes taking place during the prolonged period of acid exposure, as opposed to changes in resistance times following acid exposure, which have so far proved an unreliable means of assessing acclimation.

Brown trout weighing 40−80 g were obtained from a North-West Water Authority hatchery at Armathwaite, Cumbria. They were kept in large fibreglass tanks at 10+ 2 °C in water containing approximately 0·25 mm-Na⁺; 0·02 mm-K+ and 0·5 mm-Ca²⁺. The fish were fed daily on commercial trout pellets (Pond Pride) before and during exposure to acid media. All measurements were made at 10+ 1 °C in artificial media containing 0·25 mm NaCl; 0·02 mm K₂SO₄ and 0·5 mm (CaNO₃)₂. The methods used to measure gill potentials and sodium fluxes have been previously described (McWilliams & Potts, 1978). In all the experiments reported here brown trout were transferred from a neutral medium to a similar medium of pH 6·0 for a maximum period of 6 weeks. An acclimation pH of 6·0 was chosen because transfer to a medium of lower pH, e.g. pH 5·5, usually proved to be lethal after a few days. The pH of the tank water in all experiments was maintained by a pH RecorderController (Analytical Measurements Model RCP 75) dispensing 0·1 N-H₂SO₄. The media were not buffered.

Gill potentials were measured during and after a 6-week period of exposure to an acid medium of pH 6·0 (H₂SO₄). One group of 7 fish were anaesthetized with MS 222 and fitted with saline-filled peritoneal catheters (McWilliams & Potts, 1978) before acidification of the tank water so that gill potentials could be measured at time zero. Subsequently, groups of fish were removed at suitable intervals, at least 24 h before gill potentials were to be measured, and fitted with catheters. All fish were returned to the tank immediately and allowed to recover from the anaesthetic. Potentials were also measured in media of a range of pH values from 7·0 to 3·5 in fish which had been exposed for 6 weeks to an acid medium of pH 6·0.

Sodium uptake and loss were measured using ²²Na isotope, in media of a range of pH values from 7·0 to 4·0, in fish which had been exposed for 6 weeks to an acid medium of pH 6.0. Sodium uptake was also measured following transfer from a neutral medium to water of pH 6·0. During influx experiments the loading medium was recirculated.

The following equations were used to calculate sodium fluxes.

Corrections for backflux of isotope were not required for calculation of efflux as the external medium was not recirculated (McWilliams & Potts, 1978).

In order to estimate A_x (which is not attained during the short experimental duration) the total body sodium content of the fish was measured by macerating individual fish, removing a weighed sample, dissolving it in 10 ml cone. HNO₃ and making the volume up to 100 ml with de-ionized water. Sodium concentrations were then measured using an EEL Flame Photometer. The results were pooled and a mean value calculated. Body sodium determinations were carried out on fish before and after a 6-week period of exposure to an acid medium of pH 6·0.

Blood samples of between 0·2 and 0·5 ml were taken by cardiac puncture. The blood was immediately transferred into heparinized capillary tubes (approx, vol. 40 µl), one end was plugged with ‘Cristaseal’ (Hawksley, England), and then centrifuged for 6 min in a Hawksley Microhaematocrit centrifuge. Where possible, 5 replicates of each blood sample were taken. After centrifugation each tube was removed and haematocrit determined on a Hawksley Microhaematocrit Reader.

Sodium and chloride analyses were carried out using the plasma fraction of the sample. For sodium analysis, a 5 µl sample of plasma was removed from the capillary using a Terumo micro-syringe, added to 3 ml de-ionized water and sodium content determined using an EEL Flame Photometer. For chloride determination, 10 µl of plasma were removed and added directly to the titration vessel of a Corning Chloride Meter (Model 920). All results are expressed in m-equiv/1.

All values reported for gill potentials are those of the blood compared to the external medium (V_lnt — V_ext).

Changes in TEP during the course of 6 weeks exposure to an acid medium of pH 6·0 are given in Fig. 1. At time zero the TEP is about −6 mV but becomes more negative until, after 4 days exposure, the TEP is about − 12·5 mV. Following this the TEP shifts in a positive direction until, after 14 days exposure, it has reached a value of +9·5 mV. The potential then remains very nearly constant for the remainder of the exposure time. The change in TEP from the moment of transfer (time o) until day 2 is represented by a dotted line because it is known from previous experiments that transfer to an acid medium initially causes a positive shift in potential (McWilliams & Potts, 1978). The TEP at time o is that recorded in normal fish in an identical medium of pH 7·0 before transfer to the acid medium.

Gill potentials in fish following 6 weeks exposure to an acid medium of pH 6-o were measured in media of a range of pH values, and the results are presented in Fig. 2. Compared to the values recorded in normal fish (pH 7·0−7·4 adapted) in similar media, the TEP is shifted in a positive direction over the pH range tested by up to 10 mV. The pattern of change of potential with pH is similar to that shown by normal fish (Fig. 2).

Changes in sodium uptake rate which occur following transfer from a neutral medium to water of pH 6·0 are shown in Fig. 3. There is an initial substantial reduction in uptake rate during the first few days of exposure which is then followed by an increase in uptake rate. After a period of about 11 days exposure, uptake rate had returned to a value similar to that before transfer. A further measurement was made after 16 days exposure which revealed a slight but insignificant increase in uptake rate (P > 0·05). Sodium efflux rates were not measured during this exposure.

The influence of pH on sodium fluxes in brown trout following exposure the medium of pH 6·0 for 6 weeks is shown in Fig. 4. At near neutral pH influx and efflux are similar, indicating a balance of fluxes, but as the external [H⁺] increases the rate of loss begins to increase, although at pH 5·0 the difference between uptake and loss is still not significant (P > 0·05). At pH 4·0 the overall sodium loss is of the order of 0·6 % of the total body sodium per hour. There is no significant change in sodium uptake rate over the whole pH range tested (Fig. 4).

The mean of 8 determinations of total body sodium content in brown trout exposed to water of pH 6·0 for 6 weeks was 26·1 ± 0·9 µequiv/g wet wt. (+ S.E.). For normal fish the value was 323 ± 1·6 µequiv/g wet wt. (+ S.E., n = 9).

Plasma Na⁺ and Cl^- concentrations were monitored in trout transferred from neutral to acid water for a period of 6 weeks (Fig. 5). On transfer to water of pH 6·0 plasma Na⁺ and Cl^- concentrations decrease from a normal level of 156 m-equiv/1 and 137 m-equiv/1, respectively, to 132 m-equiv Na⁺/1 and 107 m-equiv Cl^-/1 during the first 9 days exposure. Following this, Na⁺ and Cl^- concentrations show an increase and after 4 weeks are at 139 m-equiv Na⁺/1 and 131 m-equiv Cl^-/I (Fig. 5). Plasma Cl^- concentration appears to fall more rapidly than Na⁺ concentration, but after 6 weeks exposure, plasma Na⁺ and Cl^- are both back to near-normal values at 150 m-equiv/1 and 138 m-equiv/1 respectively.

On transfer to water of pH 6·0 haematocrit initially decreases, falling to about 65 % of the normal value over a period of 4 days (Fig. 6). On continued exposure haematocrit increases slowly until after 14 days it is back to a near-normal value. Fish exposed for 6 weeks have an haematocrit slightly lower than that of normal fish, i.e. 29·8% compared to 33·1 %. The value given in Fig. 6 for normal fish compares well with that given by Soivio, Westman & Nyholm (1974) for the same species (33·2 ±4·6%).

In contrast to work by other authors involving changes in resistance times of salmonid fish following exposure to ‘sub-lethal’ levels of acidity (Lloyd & Jordan, 1964; Robinson et al. 1976; Falk & Dunson, 1977; Swarts et al. 1978), which failed to provide consistent evidence of acclimation, the results presented here demonstrate physiological compensation in brown trout exposed to a sub-lethal acid medium for a relatively long period of time. Lloyd & Jordan (1964) used an exposure time of

1. days before testing for increased acid resistance and Falk & Dunson (1977) exposed brook trout for only 2 or 24 h. All these experiments involved comparatively low exposure pH (2·5−3·8). Although Falk & Dunson (1977) did not find a consistent change in resistance time following acid exposure, their data do show a trend towards increased resistance after sub-lethal exposure. Robinson et al. (1976) used test pH levels of 2·5−3·0 and were unable to demonstrate an increase in resistance time in brook trout following 7 days exposure to pH 3·75. They suggest that acclimation to non-lethal pH would provide a more meaningful test of the ability of brook trout to compensate for environmental pH changes than the exposure to pH 3·75 used in their experiments. A recent survey of low pH tolerance in freshwater fish (EIFAC, 1969) established a pH of 5·0 as the approximate lower limit for survival for most species. Exceptional cases are field reports of survival at pH 3·7 in the roach, Rutilus rutilus (EIFAC, 1969), at pH 4·0−4·1 for brook trout, Salvelinus fontinalis (Dunson Martin, 1973), and at pH 3-5 for the Japanese teleost, Tribolodon hakonensis (Mashiko, Jozuka & Asakura, 1973). The paucity of laboratory studies documenting such exceptional tolerances suggests that some wild fish populations are acclimated to pH levels which may be lethal on rapid transfer in the laboratory.

The results presented in this paper demonstrate that following a relatively small, non-lethal step change in pH from 7·0 to 6·0, a period of about 14 days is required for physiological compensation (acclimation) to be completed in the brown trout, Salmo trutta (Figs. 1, 3, 6). It is therefore unlikely that short periods of exposure to large step changes in pH, as used by previous workers, could result in a significant degree of physiological compensation manifested as an increase in resistance time. Changes in resistance time following prolonged exposures at non-lethal levels of acidity have not so far been satisfactorily investigated and were not tested in the experiments reported here.

The process of physiological compensation for the effects of sub-lethal acid exposure is clearly demonstrated by the response of the sodium uptake mechanism in brown trout (Fig. 3). In addition, in contrast to the situation found in normal brown trout with no previous acid exposure, where sodium uptake is strongly inhibited at acidities less than pH 5·5 (McWilliams & Potts, 1978), uptake rate in acid exposed fish appears to be insensitive to external acidity (Fig. 4). Sodium efflux rates, however, still show a tendency to increase as pH falls, but in these fish at pH 4·0, net sodium loss is only 0·6% of the total body sodium per hour. In normal unexposed fish, in a similar medium, net sodium loss is almost 1 % of the total body content per hour (McWilliams & Potts, 1978). The overall sodium turnover rate in neutral media in acid exposed fish (Fig. 4) is approximately 50% of that recorded in normal fish (McWilliams & Potts, 1978).

This acclimation in sodium fluxes is reflected by changes in plasma sodium concentration. Following transfer to the acid medium, plasma sodium concentration falls but as exposure continues the level rises again until after 6 weeks exposure it is approaching a near-normal value (Fig. 5). Plasma Cl^- concentration also changes but more rapidly than Na⁺ concentration and is probably dependent on changes in TEP associated with the period of acid exposure (Fig. 1). This agrees with the data of Leivestad & Muniz (1976) who calculated that in acid exposed norwegian brown trout each Cl^- ion lost was accompanied by the loss of 0·75 Na⁺ ions.

Changes in haematocrit during acid exposure are less easily understood. Changes in haematocrit in acid stressed fish have been well documented (Soivio & Oikari, 1976; Dively et al. 1977; Soivio et al. 1977), but in all cases increasing acidity resulted in an increase in haematocrit. Dively et al. (1977) reported a strong seasonal influence on haematocrit changes in acid exposed brook trout but only recorded increases. However, the test medium was far more acid (pH 4·2) than that used here. Soivio & Oikari (1976) and Soivio et al. (1977) measured haematocrit in response to changes in plasma pH, and O₂ tension respectively, parameters not measured in this study. In the present experiments exposure of brown trout to an acid medium of pH 6·0 produced a decrease in haematocrit of 35 % over 4 days but seasonal influence was not examined. Lowered blood plasma electrolyte levels should presumably cause a swelling of the erythrocytes in response to a lowered blood plasma osmolarity in acid exposed fish. Exposure of brown trout to a more acid medium of pH 5·0 did in fact produce an increase in haematocrit of 17% over 3 days (McWilliams, unpublished observations) but this step change in pH proved to be lethal after 4 days. The differing responses of brown trout to sub-lethal and lethal step changes in pH in this respect e not understood and require further investigation.

The permeability ratio 0Na⁺/0Cl^- for brown trout gills has been calculated from equation (4) for suitable intervals during the period of acid exposure (Table 1). The ratios show that the gills of unexposed fish are normally more permeable to sodium than to chloride (ϕNa⁺/ϕCl-= 1·12), accounting for the negative TEP in neutral media. On transfer to acid water of pH 6·0 there is an initial increase in sodium permeability over the first 4 days tending to make the TEP more negative (Fig. 1). The ratio then decreases again and after 42 days exposure ϕNa⁺/ϕCl^- = 0·76 (Table 1). There appears to be a slight fluctuation in permeability values during the early stages of exposure but this is unlikely to be significant and is probably a reflection of individual variations in plasma electrolyte values (Fig. 5). The mechanisms controlling the changes in sodium permeability are not clear but are of obvious importance with respect to the sodium loss usually associated with exposure of fish to acid media, and require more detailed investigation.

Cuthbert & Maetz (1972) concluded that the goldfish, Carassius auratus, could not easily replace calcium removed from the outside of the gills following treatment with chelating agents. It is likely that media of low pH leach calcium ions from the surface of the gills, increasing the permeability to certain ions. The calcium lost in this way would be slowly replaced and perhaps augmented, to a degree dependent on external pH. This is supported by the fact that ϕNa⁺/ϕ5Cl^- returns to near-normal values after 6−9 days but is then further reduced to values considerably less than 1·0 for the remainder of the exposure time (Table 1). A permeability ratio ϕNa⁺/·Cl^_ of less than 1·0 means a reduced rate of passive loss of sodium while passive Cl^- ion loss should also be reduced in these conditions as the gill potential becomes more positive (Fig. 1). Transfer of brown trout from neutral to acid media of pH 4·0 reduces Cl^- ion loss across the gills by a factor of 2·5 (P. G. McWilliams, unpublished observations).

The success of Salmo trutta in acclimating to an acid environment appears to depend largely on the ability to reduce the permeability of the gills to sodium. Calculated values of hydrogen ion permeability reveals little change over the 6 week exposure time (Table 1). Considering the small size of H+ ions compared to Na⁺ ions, changes in H+ ion permeability would probably require extensive structural reorganization of the gill epithelium and there is no evidence for this in acid exposed fish. Hydrogen ion permeability can be influenced by calcium ions (McWilliams & Potts, 1978), but in natural waters subject to pH fluctuations as a result of acid input, the levels of available calcium are usually very low.

The author is grateful to Professor W. T. W. Potts for helpful criticism of the manuscript. This work was supported by C.E.G.B. Research Grant No. VJ 725 to W. T. W. Potts.