Blood Gases of the Tench (Tinca Tinca) in Well Aerated and Oxygen-Deficient Waters

Detailed studies on the role of the blood in respiratory gas exchange have been conducted on the rainbow trout, dogfish and lungfish, and the results have been reviewed by Randall (1970) and Johansen (1970, 1971). Among the teleosts the less active species have received little attention in this respect. Garey (1970) measured the oxygen content of arterial and venous blood in the carp to determine the cardiac output and Itazawa (1967, 1970) determined the blood oxygen content of the same species under various environmental conditions. Other than these there have been few significant studies on blood-gas function in gas exchange of inactive teleost species.

The life and habits of the tench have been described by Day (1884) and by Kennedy & Fitzmaurice (1970) ; it is predominantly a bottom-living fish of slow and still waters. Although aspects of its respiratory physiology have been described in some detail (Hughes & Shelton, 1962; Hughes, 1964; Randall & Shelton, 1963; Randall, 1966) none of these studies involved the measurement of the blood gases. In the present study results are presented concerning breathing rate, ventilation volume, oxygen uptake and other respiratory parameters of the tench; however, the main object of the study was to examine the respiratory role of the blood of tench in both oxygen-rich and oxygen-deficient waters.

Tench weighing between 150 and 400 g were obtained from suppliers at Great Stambridge, Essex. They were kept in de-chlorinated water contained in 50-gallon glass aquaria; the water temperature was maintained between 11–14 °C. They were fed twice weekly on chopped sheep ‘s heart and maggots.

The fish selected for an experiment was anaesthetized with MS 222 (Sandoz), approximately 150 mg/l; it was then transferred to an operating table where a cannula was inserted into the dorsal aorta in a manner similar to that described for salmon by Smith & Bell (1964). A cannula for inspired water samples was implanted in the buccal cavity and a second cannula was placed in the operculum for the determination of the oxygen tension of expired water (Holeton & Randall, 1967). In the present study the design of the opercular cannula was slightly different from designs previously described ; a 1 mm thick rubber washer was placed over the cannula sleeve on both the inside and the outside of the operculum. These had the effect of holding the cannula firmly in place, making a watertight seal between the cannula sleeve and the operculum, and of sampling water a small distance from the inside surface of the operculum. For each fish the cannula was placed as nearly as possible in the same position in the operculum. Hughes & Knights (1968) suggested that water on the immediate inner surface of the operculum may have had least contact with the gills and may have an unrealistically high value. In the present study, data obtained from cannulae suspected of being loose or of showing signs of leakage were discarded. The above procedures were carried out on fourteen tench.

A further three fish had only cardiac cannulae implanted. A 21-gauge needle attached to Portex PP 50 tubing was advanced into the heart from a ventral and anterior position; the point of entry to the body was a small distance anterior to the pectoral girdle and in the midline. The protruding end of the needle and the first part of the cannula tube were sutured on to the ventral surface of the lower jaw. The fish appeared to survive this operation with no ill effects. The blood cannulae were filled with heparinized saline (Wolf, 1963) containing approximately 100 I.U. of heparin/ml.

After implantation of cannulae the fish was placed in a respirometer through which de-chlorinated and aerated water flowed at a rate between 100–300 ml/min. The water was maintained at 13 ± 1 °C and the fish was left to recover for at least 24 h before the start of an experiment. The ⁠, and pH of blood and water were measured using Eschweiler microelectrodes and microelectrode assemblies; these were maintained at the same temperature as the respirometer and after values had been recorded the blood was returned to the fish via the cannula. The oxygen uptake by the fish was calculated from the difference between the water entering and leaving the respirometer, and the water flow.

Blood samples of 50–100μl were also removed from the tench at intervals; the oxygen content of 10 μl subsamples was determined by the method of Tucker (1967). The stated accuracy of the method is 0·03 ml oxygen/100 ml blood, expressed as the mean difference from variations obtained by the van Slyke method (Tucker, 1967). In the present study determinations were made until values differing by less than 0·5 vol % had been obtained. A further part of the blood sample was placed in a 20 ml tonometer and equilibrated with air for a few minutes for the subsequent determination of the oxygen capacity. The remainder of the blood sample was analysed for lactate by an enzymic method (Hohorst, Kreutz & Bücher, 1959) using the reagents and the procedure supplied by Boehringer, Mannheim. The total amount of blood removed from a fish for these analyses rarely exceeded 0·4 ml.

The fish was studied in an unstressed state for the first 3–4 h of an experiment; then nitrogen was bubbled into the respirometer water supply so that the value of the water steadily decreased 20–30 mmHg after about 4 h. Analyses of blood and water were made throughout this period and breathing rate was measured by counting the number of opercular closures per unit time.

The equations given by Holeton & Randall (1967) and by Randall, Holeton & Stevens (1967) were used to calculate percent utilization (%U), ventilation volume and other respiratory parameters.

Symbols. The symbols used are similar to those employed by Rahn (1966) and by Jones, Randall & Jarman (1970).

Ventilation volume (ml/min/kg).

Oxygen uptake (ml/min/kg).

P Partial pressure of gas (mmHg).

C Concentration of gas (ml/100 ml blood).

Q Cardiac output (ml/min/kg).

T Transfer factor (ml/min./mmHg/kg).

Subscripts.

g referring to gills ;

a arterial blood leaving gills ;

v venous blood entering gills ;

i referring to blood in the buccal cavity, i.e. inspired water;

e referring to blood in the opercular cavity, i.e. expired water.

The experimental results are shown in Fig. 1a, b, c, d  and in Fig. 2a, b, c. The greatest number of determinations were carried out on unstressed fish in well-aerated water; for these the mean value, the standard error of the mean and the number of determinations are indicated. When the oxygen in the water was reduced the individual determinations are shown. The individual results for four unstressed tench are shown in Table 1, these indicating the range of results obtained.

Resting tench observed in the stock aquaria generally had a slow breathing rate; on some occasions opercular movements appeared to cease and the only indication of gill ventilation was the movement of the opercular valves which are well developed in the tench. When in the respirometer the fish frequently did not show this quiet behaviour making it impossible to record true resting values. Tench in well-aerated water in the respirometer will be referred to as unstressed tench. These problems are discussed by Heath (1972) who suggests that in this type of experiment there is always some disturbance to the fish unless extremely elaborate precautions are taken.

The respiratory parameters measured varied considerably between individual fish ; this is illustrated in Fig. 1 a where the effect of hypoxia on the breathing rate of 4 fish is shown. In most cases when the environmental had dropped to around 30 mm Hg the breathing rate had approximately doubled and the maximum rate had apparently been reached. The oxygen tensions of inspired and expired water are shown in Fig. 1 b. The of inspired water in the buccal cavity was always slightly lower than that of the water entering the respirometer ; this was probably because of the nature of the water flow through the respirometer which permitted a small amount of expired water to be re-breathed. For unstressed fish the percent utilization of oxygen from the water was 54·9 ± 2·1 % (21); this decreased to a value of about 35 % when the environmental had fallen to approximately 25 mmHg.

The mean ventilation volume for unstressed tench was 132 ml/min/kg (Fig. 1d); and again there was variation between individuals, the values ranging between ca. 0–200 ml/min/kg. It has already been mentioned that the design of the opercular cannula may give variable results for the of expired water and thus cause variability in the calculated values for ventilation volume. However, the results in Fig. 1 d indicate that there is little change in ventilation volume until the of the water falls to about 80 mmHg, and with further decreases in the of the water ventilation volume increased steadily up to approximately eightfold. The mean oxygen uptake of unstressed tench was 0·5 ml/min/kg (Fig. 1 c). As the environmental decreased the oxygen uptake tended to remain steady until the of the water was reduced to 20–30 mmHg. Here the oxygen uptake fell significantly and the tench appeared to be in considerable respiratory distress.

In well aerated water the mean value for arterial blood of unstressed tench was 35·8 mmHg the range being from 10 mm to over 60 mmHg. The oxygen dissociation curves of tench blood (Eddy 1973) indicate that at tensions of less than 5 mmHg the haemoglobin will be more than 80% oxygen saturated when the is about 10 mmHg. As the oxygen tension in the water was decreased the arterial value decreased (Fig. 2a), and when the environmental was 20–40 mm Hg the oxygen tension in arterial blood had fallen to 5–20 mmHg. In several fish when the ambient was as high as 40–50 mmHg the arterial had dropped to around 5 mmHg; these fish also showed the lowest unstressed arterial oxygen tensions (10–15 mmHg), the highest arterial tensions and the lowest pH values. However, their ability to withstand hypoxia appeared to be about the same as the other experimental fish.

The oxygen tension in the ventral aorta of unstressed tench was about 7 mmHg.

In well aerated water the mean of unstressed tench was 3·3 mm Hg the range being approximately 2–4 mmHg. Decreasing the environmental oxygen tension caused a progressive decrease in arterial (Fig. 2 b), probably because increased gill ventilation caused CO₂ to be removed in increasing quantities from the blood perfusing the gills. When the environmental had fallen to about 20 mmHg the arterial tension was reduced to less than 1 mmHg. The venous unstressed tench was 5·0 ± 0·41 mmHg.

The mean arterial pH value of unstressed tench in well-aerated water was 8·18 ± 0·03 (11) ; as with blood and there was variation between individual fish the range being pH 8·0–8·35. Decreasing the environmental tended to have a variable effect on the arterial pH value and the pH response to hypoxia of 5 fish is shown in Fig. 2c. In many experiments there was a tendency for arterial blood pH to increase as the degree of hypoxia increased ; this is in keeping with a simultaneous decrease in arterial (Fig. 2 c) although the situation is complicated by the appearance of lactic acid in the blood.

The mean value for venous pH in unstressed tench was 8·08.

During the experiments blood was withdrawn from the tench for the simultaneous determination of ⁠, oxygen content and oxygen capacity so that the relationship between the oxygen content and the oxygen tension of the blood in vivo could be derived. In Fig. 3 the percentage oxygen saturation of the blood is plotted against blood this represents the in vivo oxygen dissociation curve of tench blood at 13 °C. In the majority of resting tench the arterial blood was 75–95 % oxygen-saturated ; however, in several individuals where the arterial was low (ca. 10 mmHg) the percent oxygen saturation of the blood was less than 75 %. A few measurements of arterial blood oxygen content were made while the tench were under hypoxic conditions and these are also indicated in Fig. 3. It would seem from the above results that tench can accommodate wide variations in arterial blood oxygen content without ill effects.

The average value for venous oxygen saturation in unstressed tench was 40% ranging between 24–45 % The mean oxygen capacity of tench blood was found to be about 8 vol %, a value similar to that reported by Eddy (1973).

The mean lactate concentration in unstressed tench blood was 14·03 ± 2·13 mg/100 ml (range 0·8–36·3 mg/100 ml, N = 19). This increased to 31·4 ±6·14 mg/100 ml (range 12·9–97·3 mg/100 ml, N = 14) when the ambient was 20–35 mmHg. The lactate values of blood from unstressed tench are similar to those found by Black et al. (1959) for rainbow trout. The increase in blood lactate indicates that the stress of hypoxia may cause some anaerobic metabolism in tench.

In Table 2 some respiratory properties of tench blood are compared with those of other fish species. The most complete data exist for the trout and it is convenient to illustrate the respiratory properties of tench blood by comparing them with those of the trout. Blood oxygen affinity tends to influence the of arterial blood; thus in trout the P₅₀ value is 15 mmHg and the arterial is about 100 mmHg while in tench P₅₀ is about 2–3 mmHg resulting in an arterial value of approximately 36 mmHg (Table 2). Carp appear to be similar to tench in this respect except that the arterial is about 8 mmHg (Garey, 1970). In unstressed tench an increase in arterial would result in a small increase in haemoglobin-bound oxygen and more oxygen dissolved in the blood plasma. If the arterial of tench blood was increased from 36 to 100 mmHg it is estimated that only an extra 3 % oxygen would be added to the blood.

The in vivo relationship between blood and blood oxygen content (expressed as percentage oxygen saturation) is shown in Fig. 3 ; this represents the in vivo oxygen dissociation curve at 13 °C for tench. Unlike the in vitro dissociation curve it is not constructed at any particular pH or value and its shape is determined by factors such as the acid-base status of the blood and the respiration rate of the tissues. Thus it represents the physiological blood oxygen dissociation curve which is most suited to the oxygen uptake system of the fish.

The in vitro oxygen dissociation curves for tench blood at 13 °C (Eddy, 1973) at tensions of 2–3 mmHg and 6 mmHg have been drawn in on Fig. 3 for comparison with the in vivo curve. There is a general similarity between the two types of curve but there are differences which deserve comment; these are mainly concerned with blood tensions. Determined with the electrode, the tension of arterial blood in unstressed tench was about 3 mmHg (Fig. 26); this must also be the CO₂ tension of the arterial points in the tn vivo oxygen dissociation curve (Fig. 3). However, from the in vitro dissociation curves shown in Fig. 3, the of most of these points should be more than 2–3 mmHg and in some cases greater than 6 mmHg. Similarly, according to the tn vitro dissociation curves in Fig. 3, the in vivo venous points should have a tension in excess of 6 mmHg but in fact the measured tension was about 5 mmHg (Fig. 2b).

These discrepancies could arise from the difficulties involved in using the Severing-haus type electrode at relatively low temperatures and also using it to measure the low tensions found in fish blood ; under these conditions the determinations are made uncertain by the difficulties of calibrating the electrode and its slow response time. In part they could arise from differences in blood buffering and this point has been discussed by Campbell (1968) in relation to the mammahan CO₂ dissociation curve. Compared to the situation in vitro, the buffering capacity of the haemoglobin in vivo is reduced because it buffers both the plasma and the interstitial fluid. Since the O₂ and CO₂ dissociation curves are functionally interdependent it follows that the greater buffering capacity in vitro ensures that when is increased the haemoglobin will remain more oxygen saturated than in vivo. Thus at similar O₂ and CO₂ tensions the blood in vitro should be more oxygen saturated than blood in vivo and in Fig. 3 this is seen to be the case.

If, therefore, measurements of in vivo and for tench are made and the percentage oxygen saturations are read off from the in vitro oxygen dissociation curves, there will be a tendency for this amount to be overestimated by a small amount particularly at oxygen tensions of < 10 mmHg. The full significance of this cannot be fully appreciated until an accurate method for measurements in fish blood is available. It is not known whether this divergence occurs in other fish species as the necessary data are lacking.

The effect of hypoxia on fish respiration has been frequently investigated and the subject has been reviewed by Hughes (1964), Randall (1970), Shelton (1970) and by Johansen (1971). In tench the responses to hypoxia were similar to those previously described for teleost fish, there being a steady increase in breathing rate, amplitude and ventilation volume (Fig. 1a, d) ; the branchial pumps appeared to be fully extended when the in the water was 20–30 mmHg. Working with the same species, Randall & Shelton (1963) found a similar pattern of events and noticed that the heart rate became increasingly reduced as the environmental fell below about 30 mmHg, slowing even more as the dropped further.

The oxygen tension of tench arterial blood tended to remain steady until the ambient fell to about 80 mmHg when it began to decrease (Fig. 2b). When the environmental was 20–30 mmHg the arterial averaged about 5 mmHg ; under these conditions the oxygen saturation of the blood was low but was prevented from being even lower by the fact that the arterial had dropped from the unstressed arterial value of about 3 mmHg to around 0·5 mmHg (Fig. 2b). Haemoglobin has a greater Oxygen affinity at lower tensions and it is estimated this decrease in could increase the oxygen content of the blood by 10–20 %. A decrease in blood affects the blood pH value and removal of CO₂ from the blood by hyperventilation results in a decrease in the hydrogen ion concentration. However, hydrogen ions are returned to the blood in the form of lactic acid whose rate of appearance in the blood seems to be accelerated by increasing hypoxia. In unstressed fish the greatest concentrations of lactic acid were usually found in those fish which initially had the lowest arterial tensions. As the level of hypoxia increased more lactate was formed and this was probably the cause of a decrease in arterial pH in some fish as shown in Fig. 2 c, where the two fish which show a steady decrease in arterial pH also had the highest lactate values. Other fish which tended to show an increase in arterial pH tended to have the lowest lactate concentrations. Thus it appears that in most cases the blood pH tends to remain fairly steady under moderate hypoxic conditions despite the removal of hydrogen ions as carbonic acid and their addition as lactic acid.

Calculations for cardiac output ⁠, Δ Pg and the oxygen transfer factor (⁠⁠) were carried out for tench under hypoxic conditions ; in making these calculations it was necessary to assume that the unstressed value for oxygen tension and oxygen content of venous blood decreased linearly to zero with decreasing environmental ⁠. Although these calculations yield only approximate results they indicate that in tench, hypoxic conditions (20–30 mmHg in the environmental water) caused a small increase in a linear decrease in Δ Pg to a value of about 20 mmHg and a fourfold increase in Thus the hypoxic response of tench is in principle similar to that of the trout as described by Randall et al. (1967) and Jones et al. (1970). The main difference in the hypoxic response of tench and trout can be related to the respective blood oxygen affinities ; because of the high oxygen affinity, the blood of the tench averages more than 75 % oxygen saturated at low values and this degree of saturation can be achieved with relatively low values of the ventilation to perfusion ratio The low P₅₀ value of tench blood ensures that the arterio-venous oxygen content remains reasonably constant even at quite low environmental oxygen tensions. Thus the tench is able to inhabit water with an oxygen level too low for an active species such as the trout.

As well as being dependent upon blood with a high oxygen affinity, the tench ‘s resistance to hypoxia must in part depend on the ability to maintain a stable acid-base status ; under hypoxic conditions the blood pH remains remarkably constant in spite of the addition of lactic acid. This suggests that the blood and body buffering systems of the fish are adapted to maintain a stable internal environment for oxygen transport under adverse respiratory conditions.

The author thanks Dr G. Shelton and Mr R. Lloyd for their advice and helpful comments. The project was supported by a Natural Environment Research Council grant to Professor G. M. Hughes; this is gratefully acknowledged.