- Email: [email protected]
A study of the effects of hypoxia on acid-base status of rainbow trout blood using an extracorporeal blood circulation
371
Respiration Physiology (1982) 49, 371-382 Elsevier Biomedical Press
A S T U D Y OF THE EFFECTS OF HYPOXIA O N A C I D - B A S E STATUS OF RAINBOW TROUT B L O O D USING AN EXTRACORPOREAL B L O O D CIRCULATION
S. T H O M A S A N D G. M. H U G H E S Laboratoire de Physiologie Animale, Universit~ de Bretagne Occidentale, FacultO des Sciences et Techniques, 29283 Brest C~dex, France, and Research Unit for Comparative Animal Respiration, University of Bristol, Bristol BS8 lUG, U.K.
Abstract. (1) The use of an extracorporeal blood circulation method has made it possible to obtain
information on the time course of possible modifications in acid-base status of trout blood during hypoxia. (2) Two series of experiments are described: (a) Continuous measurement of arterial pH, Pco2 and Po2 during 24 h of moderate hypoxia (60 Torr). (b) Continuous measurement of acid-base status during 24 h of deep hypoxia (40 Tort). (3) The results of the first series show that hypoxia provokes a stimulation of ventilatory frequency and amplitude which is associated with an increased output of CO 2 from the blood which induces respiratory alkalosis. This alkalosis is compensated during the following 24 h, and results in a return to the initial pH, a decrease in concentration of HCO 3 + CO23- and an increase in lactate levels and C I - ion concentration. (4) The second set of experiments show that in deep hypoxia the respiratory alkalosis is preceded by a brief stage of deep metabolic acidosis during which lactate levels suddenly increase dramatically. Acidosis Acid-base balance Alkalosis
Hypoxia Lactate Metabolic acidosis
Many different types of hypoxia have been identified in fish and, in all cases, the final effect is on the availability of oxygen at the tissue level for the oxidation of substrates and the liberation of energy (Hughes, 1973, 1981). As in most studies the present investigations were mainly concerned with environmental hypoxia in which oxygen levels in the inspired water were reduced. The extent of this reduction together with its duration are important variables as is also the rate of change in oxygen tension (Butler and Taylor, 1971 ; W o o d and Johansen, 1973). Effects have Accepted for publication 18 May 1982 0034-5687/82/0000-0000/$02.75 © Elsevier Biomedical Press
372
S. T H O M A S A N D G. M. H U G H E S
been observed by a number of workers using trout and other freshwater fish (Holeton and Randall, 1967; Hughes and Saunders, 1970; ltazawa and Takeda, 1978). The responses can affect all parts of the respiratory chain including modifications in the pattern and rhythm of gill ventilation as well as changes on the blood side of the gas exchanger. In most studies the method of investigation has involved sampling of both the water and/or blood at different stages of the experiment (Piiper and Baumgarten-Schumann, 1968). As with all methods of sampling there is the problem of how appropriate and representative such samples may be in relation to the time course of responses made by the fish to changes in oxygen level. With the development of an extracorporeal circulation it is possible to monitor such changes continuously (Thomas et al., 1980) and thus gain a better understanding of the adaptive mechanisms involved. Such a technique is especially advantageous in organisms like teleost fishes in which the blood volume is relatively small. During experiments in which blood and water parameters are monitored continuously it is also possible to take samples at the most appropriate times in relation to the overall changes observed as has been shown for sea bass by Hughes and Thomas (1981). The particular purpose of the present study was to use this method to gain information about the time course of modifications in acid-base equilibrium during periods of environmental hypoxia of different intensities.
Material and methods
Rainbow trout (500-700 g) were obtained from a hatchery (Lesneven Brittany) and held for several days in tanks supplied with well aerated city tap water of a known composition (water carbonate alkalinity = 0 . 4 m E q . L ~; pH = 7.90; Pco_~= 0.3 Torr). Following this period of acclimation the fish were prepared as follows: Fish were anaesthetised (MS 222, 0.1 g/L in water) and arterial and venous catheters chronically implanted in the subclavian artery and vein using the method described by Thomas (1982, in preparation); at a later stage this made it possible to establish an extracorporeal circulatory shunt. Another cannula was inserted for sampling inspired water from the buccal cavity. Chlorided silver electrodes implanted in the operculum made it possible to record ventilatory frequency. The fish were allowed to recover in narrow perspex chambers which prevented them from turning head to tail. Recovery in running water was continued for 24-48 h. Some recent studies (Thomas, unpublished) have indicated that the acid-base status returns to its initial equilibrium level 2-3 h following such surgery. Nevertheless, a much greater recovery period was allowed in order to reduce the dangers of any post-surgical stress. After this delay the extracorporeal circulation was set up and normoxic conditions maintained for one hour before hypoxia was begun, and in some cases continued for 24 h. The extracorporeal circuit permitted the continuous recording of arterial Po2, Pco~ and pH values during the first 6 h of hypoxia. During the following 24 h, this mode of recording could be reestablished at any particular
ACID-BASE STATUS OF TROUT BLOOD
373
time. Dorsal aortic blood pH was measured using a Metrohm E 603 L meter (EA 120 combined electrodes). Pao~ and Paco: were measured and recorded by two Radiometer p r i M 71 analysers (E 5036 and 5046 Radiometer electrodes). The temperature of the water was maintained constant at 15 °C. In addition to the continuous recordings, as the levels of carbon dioxide partial pressure are generally less than 5 Torr in fish, Paco, values were frequently checked on 80 /~1 arterial blood samples using the Astrup method (Astrup, 1956) and a Radiometer BMS2 MK2; there was good agreement between values obtained by the two methods. Lactate concentrations were determined enzymatically by using Boehringer-Mannheim test combination systems on 0.5 ml blood samples: LDH
L-lactate + N A D + ~
pyruvate + N A D H + H +
pyruvate + L-glutamate G_PTL-alanine + c¢-cetoglutamate The decrease in extinction of N A D H at 340 nm wavelength was monitored spectrophotometrically with respect to time. Results were expressed as mEq per ml blood. C1 concentrations were measured by using a Radiometer chloridemeter C M T 10 on 20/~1 blood samples. During the continuous measurement of extracorporeal blood the flow was maintained at a constant rate (0.4 ml/min) through the arterial catheter and passed over the electrodes before returning to the fish via the venous catheter. The total volume of blood contained in the extracorporeal shunt was 0.8 ml, and represents less than 4~o of the total blood volume of the fish. The results concerning Paco2, pHa and bicarbonates plus carbonates ([HCO~] + [CO3]) were plotted on diagrams expressing [HCOT] + [CO~-] as functions of blood pH and drawn with CO2 solubility coefficients taken from the data of Severinghaus et al. (1956 a,b), and Wood and Johansen (1973) (~co2 = 0.054 m m o l . L-1 . Torr-1 at 15 °C) and pK{ and pK2 values (6.24 and 9.80) experimentally determined using the method of Siggaard-Andersen as modified by Truchot (1974).
Results
1. A summary of the results obtained in experiments in which fish were subjected to moderate hypoxia is given in table 1. In these experiments the inspired oxygen tension was reduced to 60 Torr for a total of 24 h. The effect of such treatment on the acid-base status of the blood is indicated in fig. 1 in which the concentration of bicarbonate + carbonate ions in plasma ([HCO3] + [CO~-]) is plotted against pH. The changes that occur following transfer from normoxic to hypoxic water can be divided into two phases. During the first period there is a development of respiratory alkalosis because of hyperventilation which takes place at the very beginning of hypoxia. Thus the ventilatory frequency increases from 62 + 3 to 93 + 5 min ~. In addition to these changes in frequency an increase in opercular amplitude was also
n= 6
t =24 h
60 _+ 3
60+3
Hypoxia t = 30 min n=6
Hypoxia
155 _+ 3
t=0 n= 6
Normoxia
Inspired Pwo2 (Torr)
23 _+ 7
25_+ 6
108 _+ 11
Pao, (Torr)
0.8 _+ 0.1
0.9_+0.1
2.4 _+ 0.2
Paco: (Torr)
7.88 + 0.04
8.19_+0.02
7.90 + 0.03
1.80 _+ 0.55
5.50_+0.50
6.85 + 0.95
[ H C O f ] + [CO 3 ] ( m E q . L l)
23 _+ 2
22_+3
20 _+ 2
Hct
127.8 _+ 2.60
121.2_+2.30
120.7 -+ 2.20
CI( m E q . L I)
95 -+ 5
93_+5
62 + 3
Ventilatory frequency (min - I)
6.05 _+ 0.80
3.15_+0.75
2.25 _+ 0.50
Lactate (mEq. L-I)
TABLE 1 Mean values _+ SE of arterial blood acid base characteristics (Pco~, pH, (carbonate + bicarbonate) concentration : [ H C O f ] + [CO 2 -]), ventilatory frequency, Pao2, C1- and lactate concentrations, as a function o f time during a 24 h stage in moderate hypoxia (PwQ = 60 Torr). Temperature 15 °C.
a:
C~
;> Z
>
©
tad, ".d 4a.
375
A C I D - B A S E S T A T U S OF T R O U T B L O O D
'
'
?
'/
'
?
[co
mEq. I. 10
5
0
I
I
Z8
I
!
8.0
I
I
8.2
I
pH
Fig. 1. [HCO~] + [CO2 ] vs pH changes in arterial blood of trout (Salmo gairdneri R.) during moderate hypoxia (Pwo2 = 60 Torr). Calculated Poe2 isopleths are drawn on the diagram. Hypoxia times in minutes or hours are indicated. Vertical and horizontal bars represent standard error. The oblique lines represent mean buffer lines calculated from two in vitro pH measurements at different Pco~- The arrows indicate the time course of changes observed in acid-base status. Temperature 15 °C.
visible but was not monitored in these particular experiments. The representative point for the acid-base status moves along the buffer line, p H increasing from 7.90 _+ 0.03 to 8.19 _+ 0.02 in a b o u t 30 min whilst at the same time Pco_~ decreases from 2.4 _+ 0.2 to 0.9 _+ 0.1 Torr, the level o f [ H C O f ] + [CO~-] falls f r o m 6.85 -I- 0.95 to 5.50 _+ 0.50 m E q • L - ' and no significant change is observed neither in lactate concentration (2.25 _+ 0.50-~3.15 _+ 0.75 m E q . L -~) nor in chloride level (120.7 +_ 2.2 -* 121.2 _ 2.3 m E q • L-~). During the second phase, which lasts between 12 and 24 h, there is a c o m p e n s a t o r y mechanism which makes it possible for the p H o f the blood to return to its initial level, as a result o f a lowering in H C O f + CO~3concentration from 5.50-I-0.50 to 1.8 _+ 0.5 m E q . L - ' . D u r i n g this period the CO2 tension in the arterial blood continues to decrease slowly f r o m 1.9 + 0.1 to 0.8 _+ 0.1 Torr. Lactate levels increase significantly from 3.15 + 0.75 to 6.05 _+ 0.80 m E q • L - ' and [C1-] rises to 127.8 _+ 2 m E q - L '. Following the transfer o f the trout from the hypoxic water to n o r m o x i c conditions the blood characteristics undergo the converse changes to those observed during the transfer from n o r m o x i a to hypoxia. This results in a return to the initial equilibrium values after a b o u t 24 h recovery. The time course o f some o f the main changes which take place during these treatments are illustrated graphically in fig. 2. It is clear that the changes in arterial oxygen levels are very rapid whereas changes in levels o f CO2 and p H are much slower. The arterial p H rises to a stable
376
S. THOMAS AND G. M. HUGHES
Po2 , PCO2, pH torr torr
_PT_~tH YI~)X IA ....................
eao2
.........
....,..",
100.
\
i 80 2
.." ~ \',
____~.i . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pHa *.i
50.
........... Pao2
..................
.-"
\\
Pac02 ...... ~
",,02
: !." . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,
i"
"~" ""
Pac02 /"
:i
pHo
//
1
0 i
0
7.5
,
,
J -.-//.. i
0
1
2
24
~ 25
j ...//___., time,hours 26
48
Fig. 2. A typical example of changes of arterial blood pH, Pco: and Po2, as a function of time during and after a stage in moderate hypoxia (Pwo2 = 60 Torr) in trout (Salmo gairdneri R.). Temperature 15 °C.
value after about 30 min but Paco2 decreases progressively during the 24 h period of hypoxia. After even a very few minutes of hypoxia the haematocrit value increased significantly from 20 _+ 2 to 22 _+ 3 then 23 _+ 2 (P < 0.05, Student's t-test for paired values). 2. Results obtained in experiments involving deep hypoxia are summarised in table 2. In these experiments the inspired water oxygen level was reduced rapidly to 30-40 Torr and the fish were maintained under these conditions for 24 h. Changes in the acid base status of the blood which occurred during such treatments are summarised in fig. 3 and fig. 4 illustrates the time course of variations during such an experiment. It is apparent that soon after the commencement of hypoxia there is a rise in pH of the arterial blood which is accompanied by a decrease in Pco2 again associated with a marked increase in ventilatory frequency. During this respiratory alkalosis the arterial Po, falls to nearly zero. When the inspired Po.~ has reached 30-40 Torr there is a rapid fall in arterial p H and [HCO£] + [CO~ ] also falls. In consequence the pH decreases from 8.09 -I- 0.05 to 7.55 + 0.07 and HCO¢ + CO~ concentration decreases from 5.30 _+ 0.06 to 1.50 + 0.06 m E q . L - ' . During this period of lowered pH the change in arterial carbon dioxide tension is not affected, the decrease resulting from the increase in ventilatory frequency continues without any disturbance of its time course. Lactate level is five-fold increased (1.28 _+ 0.80 m E q . L -~ --,6.16_+ 0.95 m E q . L-~). On the other hand, the haematocrit value increases significantly from 23 _+ 2 to 29 _+ 3 during this time. Immediately after this
155_+4
30< - <40
30< - <40
30< - <40
30< - <40
Normoxia t=O n=8
Hypoxia t = 12 min _+ 2 min n=6
Hypoxia t = 20 -+ 5 min n=6
Hypoxia t = 120 _+ 30 min n=6
Hypoxia t =24h n=6
Inspired Pwo.~ (Torr)
5+
2
7_+ 3
6_+ 3
12+_ 2
115_+10
Pao., (Torr)
0.6_+0.2
0.7_+0.2
1.1_+0.3
1.3+0.3
2.5_+0.4
Paco 2 (Torr)
7.95+-0.04
8.26_+13.05
7.55_+0.07
8.09_+0.05
7.88_+0.04
pHa
1.46_+0.09
3.50+0.07
1.50+0.06
5.30_+0.06
6.71_+1.1
[HCO3- ] + [CO 2 -] (mEq • L I)
26_+3
27_+3
29_+3
23_+3
22_+4
Hct
128.7_+2.4
122.0_+3.0
121.0_+2.0
122.5_+2.0
121.5_+3.0
C1 ( m E q . L -1)
98_+5
100_+6
102_+6
102-+6
60+_5
Ventilatory frequency (min - l)
6.82_+0.95
7.12_+0.82
6.16_+0.95
1.28-+0.80
1.55_+0.73
Lactate ( m E q . L l)
TAB TABLE 2 Mean values _+ SE of arterial blood acid-base characteristics (PcQ, pH, (carbonate + bicarbonate) concentration: [HCO3- ] + [CO 2 -]), ventilatory frequency, Pao2, C I - and lactate concentrations, as a function of time during a 24 h stage in deep hypoxia (Pwo2 = 3(L40 Torr). Temperature 15 °C
..,,j ",...,I
0 0
7~
©
,.H
©
,..] > ,..]
I
D
378
S. THOMAS A N D G. M. H U G H E S
[HCO:~]
PCO 2 , to rr
[co ] -I-
/
5/
/,/
/ 4/
,,
/
3//
2 /
/
mEq L 10.
............
HYPOXIA . t:2Omln
~
~ ~ 2 2 . . ...... - " _ ~ ........
~. T t=120min ................. e'~i~ . ........ .L -
L 0
~
i
i
i
7.6
t= 24 hrs
i
i
Z8
l
i
8.0
I.
8.2
pH
Fig. 3. [HCO3] + [CO 2-] vs pH changes in arterial blood of trout (Salmo gairdneri R.) during deep hypoxia (Pwo: = 30 Torr). Calculated Pco2 isopleths are drawn on the diagram. Hypoxia times in minutes or hours are indicated. Vertical and horizontal bars represent standard error. The oblique lines represent mean buffer lines calculated from two in vitro pH measurements at different Pco.,. The arrows indicate the time course of changes observed in acid-base status. Temperature 15 °C.
pOT PCO2,pH torr
torl
NORMOXIA
pwo21HYPOXIA
I
....................................
......
Pwo2
~O2 .........
100
°. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
: :
/"/
~x ,
/
~02
:" .
~A PoC02/ I .....
/
\
:
.: ,, z
pHo ":! /
X~ ...-"'"
• .-. ......................................
0
0 o2
....................... ,'
pHo
.:
~
~
a -.-/l___i
0
I
2
24
i
25
j.__y/___J time.hours
26
48
Fig. 4. A typical example of changes of arterial blood pH, Pco_~ and Po: as a function of time during and after a stage in deep hypoxia (Pwo2 = 30 Torr) in trout (Salmo gairdneri R.). Temperature 15 °C.
A C I D BASE S T A T U S O F T R O U T B L O O D
379
short period which lasts between 5 and 15 min, there is a reversal in the direction of change in bicarbonate level and pH value which results in a return to a buffer line close to the initial one within one or two hours, whereas lactate continues to increase slowly. At the end of this period the arterial blood characteristics are as follows: pH 8.26 _+ 0.05; [HCO3] + [CO~-] 3.50 +_ 0.07 mEq- L-~; Pco2 0.70 _+ 0.20 Torr; lactate 7.12 + 0.82 mEq • L -~. It is interesting to notice that during this whole period [C1-] remains quite unchanged but only rises during the following 24 h. Indeed, during the following hours [C1-] increases up to 128.7 _+ 2.4 m E q . L J and a pH compensatory mechanism occurs by decreasing the [ H C O f ] + [CO~-] level as described previously for moderate hypoxia whilst lactate level remains high: 6.82 +_ 0.95 mEq • L - l .
Discussion
Literature data on the effects of hypoxia on acid-base status of fish are somewhat confusing since hypocapnic alkalosis induced by hyperventilation and lactacidemia are frequently intermingled. Results obtained in this investigation show clearly that the effects observed in trout depend both on the level of hypoxia and on the time at which measurements are performed. At the onset of hypoxia, trout show a response which is characterised by an initial rise in ventilatory frequency which is presumably associated with an increase in ventilatory volume. This initial response has been shown to be reflexly induced by a nervous pathway in which oxygen lack is probably the effective stimulus (Eclancher, 1972). The increased ventilation leads to a greater difference in CO2 tension between the blood and water across the gill water-blood barrier, and hence enhances the release of this gas. Consequently, in the blood there is a shift of the equilibria: CO2 ~-~H2CO3 ~-~HCO3- + H + ~ CO~- + H + in such a way as to produce a decrease in [H+]. This would explain the initial respiratory alkalosis recorded in the present experiments. When ventilatory frequency has become stabilised at the higher level, the new CO2 tension difference across the gill leads to a new steady state of acid-base balance at a higher blood pH. Following this initial response which lasts between 0.5 and 2 h there is a decrease in blood pH concomitant with a drop in plasma [HCO3-] + [CO~3-]. This 'pH compensatory' mechanism occurs at constant blood Pco2 values and consequently has no ventilatory origin. It may involve either a change in plasma ionic composition or an appearance of metabolic acids in the blood. Concerning ions the above results show some significant modifications in plasma [C1-], which suggest the intervention of a C1- ~ H C O f exchange process such as described in fish by Maetz (1971). The drop in [ H C O f ] + [CO~3-] can be partially compensated by a change in plasma [C1-], in the same way as observed by Thomas
380
S. THOMAS AND G. M. HUGHES
et al. (1980) in hypocapnic conditions which lead to similar changes in acid-base
balance. The results also show an increase in lactate levels which, because of a simultaneous decrease in the alkali reserve is sufficient to provoke a fall of pH value. With respect to the deep hypoxia experiments, some remarkable observations made during the present investigations need to be explained. First, the observation of the sudden onset of metabolic acidosis which occurs only a few minutes after the beginning of hypoxia. This stage is only found when the inspired water Po~ was lowered to values below 40 Torr. Our measurements show that this phenomenon is associated with a sudden rise in haematocrit value similar to those described during hypoxia of trout by Swift and Lloyd (1974). These authors suggested that the increased haematocrit was due to a period of haemoconcentration because urine flow rate became elevated. Swelling of trout erythrocytes is also known to be associated with hypoxia. Other authors observing increased haematocrit have supposed it was due to a mobilisation of erythrocytes which entered the circulation under the effect of hypoxia (Ostroumova, 1964; Stevens, 1968; Hughes et al., 1981). Johansen and Hanson (1967) considered that this release was due either to splenoconstriction or recruitment of cells from other storage organs such as the liver. The fall of pH appears to be associated with a rise of lactate level to 5 times the initial value. Regarding the origin of this lactate, and in order to account for its appearance, the hypothesis can be suggested that the release of stored erythrocytes is accompanied by a release of metabolic acids including lactate stored previous to the hypoxia. Such a release of metabolic acid would be accompanied by a decrease in pH corresponding to about 1 pH unit which is considerably different from what happens in mammals. This difference is closely related to the low buffer power of fish plasma. It has been possible to show with trout blood that the addition of 5 mEq • L-' of lactic acid in ritro induces a fall of pH from 7.95 _+ 0.04 to 7.26 _+ 0.06 and a decrease in [ H C O f ] + [CO~-] from 5.35 _+ 0.8 to 1.80 _+ 0.9 m E q . L - ' . Without making any assumptions about possible changes in ionic composition of plasma which may occur in the trout during this short period of time, the appearance of 5 m E q - L - ' of lactate at physiological pH (i.e., higher than the pK value of lactic acid) involves, in a closed system, the disappearance of a similar quantity of bicarbonate ions (as confirmed by the experimental results given in table 2). The drop in bicarbonate level on the Pco, isopleth = 1.3 Torr of a ( H C O 3 + CO~-) = f (pH) diagram makes it possible to predict the amplitude of the pH decrease attributable to the release of lactic acid. The fact that this estimated drop corresponds to the values actually measured indicates that the observed pH change may justifiably be attributed to the observed release of lactate. In order to estimate quantitatively the effect on blood pH of the increase from 1.28 to 6.16 m E q . L-' in lactate level, a computing model suggested by Stewart (1978) has been used which, for a plasma-like fluid, gives the evolution of pH as a function of plasma
ACID-BASE STATUS OF TROUT BLOOD
381
composition. In human plasma (Pco2 = 40 Torr, T = 37 °C) such a release of lactic acid could only change the pH value from 7.36 to 7.28. The application of the model on the basis of the known values for fish plasma composition shows that the same addition of acid induces a pH decrease from 8.11 to 7.51 and a drop in [HCO3] + [CO~-] from 5.4 to 1.5 m E q . L -I. During severe hypoxia, after the pH decrease, the following stage is a pH recovery which could have been explained within the framework of the above mechanisms by the disappearance of metabolic acids, but the experimental results show that their level is maintained or even increased. Thus, this stage can only be understood in the framework of ionic changes in plasma. Just as 5 mEq • L -1 of lactic acid was enough to explain the pH decrease, so an ionic change which affects the strong ion concentration by 5 m E q . L-~ would be sufficient to explain the increase. Such an hypothesis involving adjustment in plasma ionic composition could explain the experimental results; such a change does not appear in [C1-] which increases only during the ultimate phase of pH recovery; nevertheless, these results bring no new information about the mechanism of adjustment of hydromineral balance which is still insufficiently known in fish. The same model as above indicates that after lactate addition, a 5 mEq • L-l rise in the difference between positive and negative ions is sufficient for a pH recovery to 7.96 and for a [HCO3] + [CO~ ] rise of 4 mEq - L ~. Another question raised by these experiments concerns the observed increases in blood pH and bicarbonates which do not stop when the blood pH has been restored to its original value as might be expected if pH sensitive receptors were involved, but only do so when a return has been made close to the original buffer line. Consequently, at the end of this stage another compensatory mechanism involving a decrease in HCO7 + CO~- concentration is involved as described for moderate hypoxia.
Acknowledgements We wish to thank Professor Claude Peyraud for providing excellent facilities for carrying out these experiments and for very helpful discussions.
References Astrup, P. (1956). A simple electrometric technique for the determination of carbon dioxide tension in blood and plasma, total content of carbon dioxide in plasma, and bicarbonate content in 'separated" plasma at a fixed carbon dioxide tension (40 mm Hg). Scan. J. Lab. Invest. 8: 33-43. Butler, P.J. and E.W. Taylor (1971). Response of the dogfish (So,liorhinus canicula, L.) to slowly induced and rapidly induced hypoxia. Comp. Biochem. Physiol. 39 A: 307 323. Eclancher, B. (1972). Action des changementsde Po2 de l'eau sur la ventilation de la truite et de la tanche. J. Physiol. (Paris) 65:397 A.
382
s. THOMAS A N D G. M. H U G H E S
Holeton, G . F . and D.J. Randall (1967). The effect of hypoxia upon the partial pressure of gases in the blood and water afferent and efferent to the gills of rainbow trout. J. Exp. Biol. 46: 317-327. Hughes, G. M. and R. L. Saunders (1970). Responses of the respiratory pumps to hypoxia in the rainbow trout (Salmo gairdneri R.). J. Exp. Biol. 53 : 529 545. Hughes, G. M. (1973). Respiratory responses to hypoxia in fish. Am. Zool. 13:475 489. Hughes, G . M . (1981). Effects of low oxygen and pollution on the respiratory systems of fish. In: Stress and Fish, edited by A. D. Pickering. London and New York, Academic Press, pp. 121-146. Hughes, G. M. and S. Thomas (1981). Continuous recording of pH, PO2, and PCO 2 of arterial blood in sea bass (Morone labrax) during changes in environmental PO 2. J. Physiol. (London) 319: 8~87P. Itazawa, Y. and T. Takeda (1978). Gas exchange in the carp gills in normoxic and hypoxic conditions. Respir. Physiol. 35: 263-269. Johansen, K. and D. Hanson (1967). Hepatic vein sphincters in elasmobranchs and their significance in controlling hepatic blood flow. J. Exp. Biol. 46:195 203. Maetz, J. (1971). Fish gills: mechanisms of salt transfer in freshwater and seawater. Philos. Trans. R. Soc. London Set. B 262:209 251. Ostroumova, I. N. (1964). Condition of trout blood at adaptation to different oxygen and salt conditions of water, lzv. uses. nauchno-issled. Inst. ozern, rechn, ryb. Khoz. 58: 27-36. Piiper, J. and D. Baumgarten-Schumann (1968). Effectiveness of 02 and CO 2 exchange in the gills of the dogfish (Scyliorhinus stellaris). Re.~pir. Physiol. 5:338 349. Severinghaus, J. W., M. Stupfel and A. F. Bradley (1956a). Accuracy of blood pH and Pco2 determinations. J. Appl. Physiol. 9: 189-196. Severinghaus, J. W., M. Stupfel and A. F. Bradley (1956b). Variations of serum carbonic acid pK' with pH and temperature. J. Appl. Physiol. 9:197 200. Stevens, E. D (1968). The effect of exercise on the distribution of blood to various organs in rainbow trout. Comp. Biochem. Physiol. 25:615q525. Stewart, P. A. (1978). Independent and dependent variables of acid-base control. Respir. Physiol. 33:9 26. Swift, D.J. and R. Lloyd (1974). Changes in urine flow rate and haematocrit value of rainbow trout Salmo gairdneri (Richardson) exposed to hypoxia. J. Fish. Biol. 6: 379-387. Thomas, S., A. Belaud, L. Barth61~my and C. Peyraud (1980). Acid base status in plasma of trout and eel in hypocapnic and normocapnic conditions. J. Comp. Physiol. 140:249 254. Truchot, J. P. (1974). Le transport de gaz respiratoires (oxyg+ne et dioxyde de carbone) par le sang du crabe Carcinus maenas (L.) Th&e de Doctorat des Sciences, Universit6 de Paris, No. OA 9817. Wood, S. C. and K. Johansen (1973). Adaptation to hypoxia by increased HbO 2 affinity and decreased red cell ATP concentration. Nature (London) 237: 278-279.
Respiration Physiology (1982) 49, 371-382 Elsevier Biomedical Press
A S T U D Y OF THE EFFECTS OF HYPOXIA O N A C I D - B A S E STATUS OF RAINBOW TROUT B L O O D USING AN EXTRACORPOREAL B L O O D CIRCULATION
S. T H O M A S A N D G. M. H U G H E S Laboratoire de Physiologie Animale, Universit~ de Bretagne Occidentale, FacultO des Sciences et Techniques, 29283 Brest C~dex, France, and Research Unit for Comparative Animal Respiration, University of Bristol, Bristol BS8 lUG, U.K.
Abstract. (1) The use of an extracorporeal blood circulation method has made it possible to obtain
information on the time course of possible modifications in acid-base status of trout blood during hypoxia. (2) Two series of experiments are described: (a) Continuous measurement of arterial pH, Pco2 and Po2 during 24 h of moderate hypoxia (60 Torr). (b) Continuous measurement of acid-base status during 24 h of deep hypoxia (40 Tort). (3) The results of the first series show that hypoxia provokes a stimulation of ventilatory frequency and amplitude which is associated with an increased output of CO 2 from the blood which induces respiratory alkalosis. This alkalosis is compensated during the following 24 h, and results in a return to the initial pH, a decrease in concentration of HCO 3 + CO23- and an increase in lactate levels and C I - ion concentration. (4) The second set of experiments show that in deep hypoxia the respiratory alkalosis is preceded by a brief stage of deep metabolic acidosis during which lactate levels suddenly increase dramatically. Acidosis Acid-base balance Alkalosis
Hypoxia Lactate Metabolic acidosis
Many different types of hypoxia have been identified in fish and, in all cases, the final effect is on the availability of oxygen at the tissue level for the oxidation of substrates and the liberation of energy (Hughes, 1973, 1981). As in most studies the present investigations were mainly concerned with environmental hypoxia in which oxygen levels in the inspired water were reduced. The extent of this reduction together with its duration are important variables as is also the rate of change in oxygen tension (Butler and Taylor, 1971 ; W o o d and Johansen, 1973). Effects have Accepted for publication 18 May 1982 0034-5687/82/0000-0000/$02.75 © Elsevier Biomedical Press
372
S. T H O M A S A N D G. M. H U G H E S
been observed by a number of workers using trout and other freshwater fish (Holeton and Randall, 1967; Hughes and Saunders, 1970; ltazawa and Takeda, 1978). The responses can affect all parts of the respiratory chain including modifications in the pattern and rhythm of gill ventilation as well as changes on the blood side of the gas exchanger. In most studies the method of investigation has involved sampling of both the water and/or blood at different stages of the experiment (Piiper and Baumgarten-Schumann, 1968). As with all methods of sampling there is the problem of how appropriate and representative such samples may be in relation to the time course of responses made by the fish to changes in oxygen level. With the development of an extracorporeal circulation it is possible to monitor such changes continuously (Thomas et al., 1980) and thus gain a better understanding of the adaptive mechanisms involved. Such a technique is especially advantageous in organisms like teleost fishes in which the blood volume is relatively small. During experiments in which blood and water parameters are monitored continuously it is also possible to take samples at the most appropriate times in relation to the overall changes observed as has been shown for sea bass by Hughes and Thomas (1981). The particular purpose of the present study was to use this method to gain information about the time course of modifications in acid-base equilibrium during periods of environmental hypoxia of different intensities.
Material and methods
Rainbow trout (500-700 g) were obtained from a hatchery (Lesneven Brittany) and held for several days in tanks supplied with well aerated city tap water of a known composition (water carbonate alkalinity = 0 . 4 m E q . L ~; pH = 7.90; Pco_~= 0.3 Torr). Following this period of acclimation the fish were prepared as follows: Fish were anaesthetised (MS 222, 0.1 g/L in water) and arterial and venous catheters chronically implanted in the subclavian artery and vein using the method described by Thomas (1982, in preparation); at a later stage this made it possible to establish an extracorporeal circulatory shunt. Another cannula was inserted for sampling inspired water from the buccal cavity. Chlorided silver electrodes implanted in the operculum made it possible to record ventilatory frequency. The fish were allowed to recover in narrow perspex chambers which prevented them from turning head to tail. Recovery in running water was continued for 24-48 h. Some recent studies (Thomas, unpublished) have indicated that the acid-base status returns to its initial equilibrium level 2-3 h following such surgery. Nevertheless, a much greater recovery period was allowed in order to reduce the dangers of any post-surgical stress. After this delay the extracorporeal circulation was set up and normoxic conditions maintained for one hour before hypoxia was begun, and in some cases continued for 24 h. The extracorporeal circuit permitted the continuous recording of arterial Po2, Pco~ and pH values during the first 6 h of hypoxia. During the following 24 h, this mode of recording could be reestablished at any particular
ACID-BASE STATUS OF TROUT BLOOD
373
time. Dorsal aortic blood pH was measured using a Metrohm E 603 L meter (EA 120 combined electrodes). Pao~ and Paco: were measured and recorded by two Radiometer p r i M 71 analysers (E 5036 and 5046 Radiometer electrodes). The temperature of the water was maintained constant at 15 °C. In addition to the continuous recordings, as the levels of carbon dioxide partial pressure are generally less than 5 Torr in fish, Paco, values were frequently checked on 80 /~1 arterial blood samples using the Astrup method (Astrup, 1956) and a Radiometer BMS2 MK2; there was good agreement between values obtained by the two methods. Lactate concentrations were determined enzymatically by using Boehringer-Mannheim test combination systems on 0.5 ml blood samples: LDH
L-lactate + N A D + ~
pyruvate + N A D H + H +
pyruvate + L-glutamate G_PTL-alanine + c¢-cetoglutamate The decrease in extinction of N A D H at 340 nm wavelength was monitored spectrophotometrically with respect to time. Results were expressed as mEq per ml blood. C1 concentrations were measured by using a Radiometer chloridemeter C M T 10 on 20/~1 blood samples. During the continuous measurement of extracorporeal blood the flow was maintained at a constant rate (0.4 ml/min) through the arterial catheter and passed over the electrodes before returning to the fish via the venous catheter. The total volume of blood contained in the extracorporeal shunt was 0.8 ml, and represents less than 4~o of the total blood volume of the fish. The results concerning Paco2, pHa and bicarbonates plus carbonates ([HCO~] + [CO3]) were plotted on diagrams expressing [HCOT] + [CO~-] as functions of blood pH and drawn with CO2 solubility coefficients taken from the data of Severinghaus et al. (1956 a,b), and Wood and Johansen (1973) (~co2 = 0.054 m m o l . L-1 . Torr-1 at 15 °C) and pK{ and pK2 values (6.24 and 9.80) experimentally determined using the method of Siggaard-Andersen as modified by Truchot (1974).
Results
1. A summary of the results obtained in experiments in which fish were subjected to moderate hypoxia is given in table 1. In these experiments the inspired oxygen tension was reduced to 60 Torr for a total of 24 h. The effect of such treatment on the acid-base status of the blood is indicated in fig. 1 in which the concentration of bicarbonate + carbonate ions in plasma ([HCO3] + [CO~-]) is plotted against pH. The changes that occur following transfer from normoxic to hypoxic water can be divided into two phases. During the first period there is a development of respiratory alkalosis because of hyperventilation which takes place at the very beginning of hypoxia. Thus the ventilatory frequency increases from 62 + 3 to 93 + 5 min ~. In addition to these changes in frequency an increase in opercular amplitude was also
n= 6
t =24 h
60 _+ 3
60+3
Hypoxia t = 30 min n=6
Hypoxia
155 _+ 3
t=0 n= 6
Normoxia
Inspired Pwo2 (Torr)
23 _+ 7
25_+ 6
108 _+ 11
Pao, (Torr)
0.8 _+ 0.1
0.9_+0.1
2.4 _+ 0.2
Paco: (Torr)
7.88 + 0.04
8.19_+0.02
7.90 + 0.03
1.80 _+ 0.55
5.50_+0.50
6.85 + 0.95
[ H C O f ] + [CO 3 ] ( m E q . L l)
23 _+ 2
22_+3
20 _+ 2
Hct
127.8 _+ 2.60
121.2_+2.30
120.7 -+ 2.20
CI( m E q . L I)
95 -+ 5
93_+5
62 + 3
Ventilatory frequency (min - I)
6.05 _+ 0.80
3.15_+0.75
2.25 _+ 0.50
Lactate (mEq. L-I)
TABLE 1 Mean values _+ SE of arterial blood acid base characteristics (Pco~, pH, (carbonate + bicarbonate) concentration : [ H C O f ] + [CO 2 -]), ventilatory frequency, Pao2, C1- and lactate concentrations, as a function o f time during a 24 h stage in moderate hypoxia (PwQ = 60 Torr). Temperature 15 °C.
a:
C~
;> Z
>
©
tad, ".d 4a.
375
A C I D - B A S E S T A T U S OF T R O U T B L O O D
'
'
?
'/
'
?
[co
mEq. I. 10
5
0
I
I
Z8
I
!
8.0
I
I
8.2
I
pH
Fig. 1. [HCO~] + [CO2 ] vs pH changes in arterial blood of trout (Salmo gairdneri R.) during moderate hypoxia (Pwo2 = 60 Torr). Calculated Poe2 isopleths are drawn on the diagram. Hypoxia times in minutes or hours are indicated. Vertical and horizontal bars represent standard error. The oblique lines represent mean buffer lines calculated from two in vitro pH measurements at different Pco~- The arrows indicate the time course of changes observed in acid-base status. Temperature 15 °C.
visible but was not monitored in these particular experiments. The representative point for the acid-base status moves along the buffer line, p H increasing from 7.90 _+ 0.03 to 8.19 _+ 0.02 in a b o u t 30 min whilst at the same time Pco_~ decreases from 2.4 _+ 0.2 to 0.9 _+ 0.1 Torr, the level o f [ H C O f ] + [CO~-] falls f r o m 6.85 -I- 0.95 to 5.50 _+ 0.50 m E q • L - ' and no significant change is observed neither in lactate concentration (2.25 _+ 0.50-~3.15 _+ 0.75 m E q . L -~) nor in chloride level (120.7 +_ 2.2 -* 121.2 _ 2.3 m E q • L-~). During the second phase, which lasts between 12 and 24 h, there is a c o m p e n s a t o r y mechanism which makes it possible for the p H o f the blood to return to its initial level, as a result o f a lowering in H C O f + CO~3concentration from 5.50-I-0.50 to 1.8 _+ 0.5 m E q . L - ' . D u r i n g this period the CO2 tension in the arterial blood continues to decrease slowly f r o m 1.9 + 0.1 to 0.8 _+ 0.1 Torr. Lactate levels increase significantly from 3.15 + 0.75 to 6.05 _+ 0.80 m E q • L - ' and [C1-] rises to 127.8 _+ 2 m E q - L '. Following the transfer o f the trout from the hypoxic water to n o r m o x i c conditions the blood characteristics undergo the converse changes to those observed during the transfer from n o r m o x i a to hypoxia. This results in a return to the initial equilibrium values after a b o u t 24 h recovery. The time course o f some o f the main changes which take place during these treatments are illustrated graphically in fig. 2. It is clear that the changes in arterial oxygen levels are very rapid whereas changes in levels o f CO2 and p H are much slower. The arterial p H rises to a stable
376
S. THOMAS AND G. M. HUGHES
Po2 , PCO2, pH torr torr
_PT_~tH YI~)X IA ....................
eao2
.........
....,..",
100.
\
i 80 2
.." ~ \',
____~.i . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pHa *.i
50.
........... Pao2
..................
.-"
\\
Pac02 ...... ~
",,02
: !." . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,
i"
"~" ""
Pac02 /"
:i
pHo
//
1
0 i
0
7.5
,
,
J -.-//.. i
0
1
2
24
~ 25
j ...//___., time,hours 26
48
Fig. 2. A typical example of changes of arterial blood pH, Pco: and Po2, as a function of time during and after a stage in moderate hypoxia (Pwo2 = 60 Torr) in trout (Salmo gairdneri R.). Temperature 15 °C.
value after about 30 min but Paco2 decreases progressively during the 24 h period of hypoxia. After even a very few minutes of hypoxia the haematocrit value increased significantly from 20 _+ 2 to 22 _+ 3 then 23 _+ 2 (P < 0.05, Student's t-test for paired values). 2. Results obtained in experiments involving deep hypoxia are summarised in table 2. In these experiments the inspired water oxygen level was reduced rapidly to 30-40 Torr and the fish were maintained under these conditions for 24 h. Changes in the acid base status of the blood which occurred during such treatments are summarised in fig. 3 and fig. 4 illustrates the time course of variations during such an experiment. It is apparent that soon after the commencement of hypoxia there is a rise in pH of the arterial blood which is accompanied by a decrease in Pco2 again associated with a marked increase in ventilatory frequency. During this respiratory alkalosis the arterial Po, falls to nearly zero. When the inspired Po.~ has reached 30-40 Torr there is a rapid fall in arterial p H and [HCO£] + [CO~ ] also falls. In consequence the pH decreases from 8.09 -I- 0.05 to 7.55 + 0.07 and HCO¢ + CO~ concentration decreases from 5.30 _+ 0.06 to 1.50 + 0.06 m E q . L - ' . During this period of lowered pH the change in arterial carbon dioxide tension is not affected, the decrease resulting from the increase in ventilatory frequency continues without any disturbance of its time course. Lactate level is five-fold increased (1.28 _+ 0.80 m E q . L -~ --,6.16_+ 0.95 m E q . L-~). On the other hand, the haematocrit value increases significantly from 23 _+ 2 to 29 _+ 3 during this time. Immediately after this
155_+4
30< - <40
30< - <40
30< - <40
30< - <40
Normoxia t=O n=8
Hypoxia t = 12 min _+ 2 min n=6
Hypoxia t = 20 -+ 5 min n=6
Hypoxia t = 120 _+ 30 min n=6
Hypoxia t =24h n=6
Inspired Pwo.~ (Torr)
5+
2
7_+ 3
6_+ 3
12+_ 2
115_+10
Pao., (Torr)
0.6_+0.2
0.7_+0.2
1.1_+0.3
1.3+0.3
2.5_+0.4
Paco 2 (Torr)
7.95+-0.04
8.26_+13.05
7.55_+0.07
8.09_+0.05
7.88_+0.04
pHa
1.46_+0.09
3.50+0.07
1.50+0.06
5.30_+0.06
6.71_+1.1
[HCO3- ] + [CO 2 -] (mEq • L I)
26_+3
27_+3
29_+3
23_+3
22_+4
Hct
128.7_+2.4
122.0_+3.0
121.0_+2.0
122.5_+2.0
121.5_+3.0
C1 ( m E q . L -1)
98_+5
100_+6
102_+6
102-+6
60+_5
Ventilatory frequency (min - l)
6.82_+0.95
7.12_+0.82
6.16_+0.95
1.28-+0.80
1.55_+0.73
Lactate ( m E q . L l)
TAB TABLE 2 Mean values _+ SE of arterial blood acid-base characteristics (PcQ, pH, (carbonate + bicarbonate) concentration: [HCO3- ] + [CO 2 -]), ventilatory frequency, Pao2, C I - and lactate concentrations, as a function of time during a 24 h stage in deep hypoxia (Pwo2 = 3(L40 Torr). Temperature 15 °C
..,,j ",...,I
0 0
7~
©
,.H
©
,..] > ,..]
I
D
378
S. THOMAS A N D G. M. H U G H E S
[HCO:~]
PCO 2 , to rr
[co ] -I-
/
5/
/,/
/ 4/
,,
/
3//
2 /
/
mEq L 10.
............
HYPOXIA . t:2Omln
~
~ ~ 2 2 . . ...... - " _ ~ ........
~. T t=120min ................. e'~i~ . ........ .L -
L 0
~
i
i
i
7.6
t= 24 hrs
i
i
Z8
l
i
8.0
I.
8.2
pH
Fig. 3. [HCO3] + [CO 2-] vs pH changes in arterial blood of trout (Salmo gairdneri R.) during deep hypoxia (Pwo: = 30 Torr). Calculated Pco2 isopleths are drawn on the diagram. Hypoxia times in minutes or hours are indicated. Vertical and horizontal bars represent standard error. The oblique lines represent mean buffer lines calculated from two in vitro pH measurements at different Pco.,. The arrows indicate the time course of changes observed in acid-base status. Temperature 15 °C.
pOT PCO2,pH torr
torl
NORMOXIA
pwo21HYPOXIA
I
....................................
......
Pwo2
~O2 .........
100
°. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
: :
/"/
~x ,
/
~02
:" .
~A PoC02/ I .....
/
\
:
.: ,, z
pHo ":! /
X~ ...-"'"
• .-. ......................................
0
0 o2
....................... ,'
pHo
.:
~
~
a -.-/l___i
0
I
2
24
i
25
j.__y/___J time.hours
26
48
Fig. 4. A typical example of changes of arterial blood pH, Pco_~ and Po: as a function of time during and after a stage in deep hypoxia (Pwo2 = 30 Torr) in trout (Salmo gairdneri R.). Temperature 15 °C.
A C I D BASE S T A T U S O F T R O U T B L O O D
379
short period which lasts between 5 and 15 min, there is a reversal in the direction of change in bicarbonate level and pH value which results in a return to a buffer line close to the initial one within one or two hours, whereas lactate continues to increase slowly. At the end of this period the arterial blood characteristics are as follows: pH 8.26 _+ 0.05; [HCO3] + [CO~-] 3.50 +_ 0.07 mEq- L-~; Pco2 0.70 _+ 0.20 Torr; lactate 7.12 + 0.82 mEq • L -~. It is interesting to notice that during this whole period [C1-] remains quite unchanged but only rises during the following 24 h. Indeed, during the following hours [C1-] increases up to 128.7 _+ 2.4 m E q . L J and a pH compensatory mechanism occurs by decreasing the [ H C O f ] + [CO~-] level as described previously for moderate hypoxia whilst lactate level remains high: 6.82 +_ 0.95 mEq • L - l .
Discussion
Literature data on the effects of hypoxia on acid-base status of fish are somewhat confusing since hypocapnic alkalosis induced by hyperventilation and lactacidemia are frequently intermingled. Results obtained in this investigation show clearly that the effects observed in trout depend both on the level of hypoxia and on the time at which measurements are performed. At the onset of hypoxia, trout show a response which is characterised by an initial rise in ventilatory frequency which is presumably associated with an increase in ventilatory volume. This initial response has been shown to be reflexly induced by a nervous pathway in which oxygen lack is probably the effective stimulus (Eclancher, 1972). The increased ventilation leads to a greater difference in CO2 tension between the blood and water across the gill water-blood barrier, and hence enhances the release of this gas. Consequently, in the blood there is a shift of the equilibria: CO2 ~-~H2CO3 ~-~HCO3- + H + ~ CO~- + H + in such a way as to produce a decrease in [H+]. This would explain the initial respiratory alkalosis recorded in the present experiments. When ventilatory frequency has become stabilised at the higher level, the new CO2 tension difference across the gill leads to a new steady state of acid-base balance at a higher blood pH. Following this initial response which lasts between 0.5 and 2 h there is a decrease in blood pH concomitant with a drop in plasma [HCO3-] + [CO~3-]. This 'pH compensatory' mechanism occurs at constant blood Pco2 values and consequently has no ventilatory origin. It may involve either a change in plasma ionic composition or an appearance of metabolic acids in the blood. Concerning ions the above results show some significant modifications in plasma [C1-], which suggest the intervention of a C1- ~ H C O f exchange process such as described in fish by Maetz (1971). The drop in [ H C O f ] + [CO~3-] can be partially compensated by a change in plasma [C1-], in the same way as observed by Thomas
380
S. THOMAS AND G. M. HUGHES
et al. (1980) in hypocapnic conditions which lead to similar changes in acid-base
balance. The results also show an increase in lactate levels which, because of a simultaneous decrease in the alkali reserve is sufficient to provoke a fall of pH value. With respect to the deep hypoxia experiments, some remarkable observations made during the present investigations need to be explained. First, the observation of the sudden onset of metabolic acidosis which occurs only a few minutes after the beginning of hypoxia. This stage is only found when the inspired water Po~ was lowered to values below 40 Torr. Our measurements show that this phenomenon is associated with a sudden rise in haematocrit value similar to those described during hypoxia of trout by Swift and Lloyd (1974). These authors suggested that the increased haematocrit was due to a period of haemoconcentration because urine flow rate became elevated. Swelling of trout erythrocytes is also known to be associated with hypoxia. Other authors observing increased haematocrit have supposed it was due to a mobilisation of erythrocytes which entered the circulation under the effect of hypoxia (Ostroumova, 1964; Stevens, 1968; Hughes et al., 1981). Johansen and Hanson (1967) considered that this release was due either to splenoconstriction or recruitment of cells from other storage organs such as the liver. The fall of pH appears to be associated with a rise of lactate level to 5 times the initial value. Regarding the origin of this lactate, and in order to account for its appearance, the hypothesis can be suggested that the release of stored erythrocytes is accompanied by a release of metabolic acids including lactate stored previous to the hypoxia. Such a release of metabolic acid would be accompanied by a decrease in pH corresponding to about 1 pH unit which is considerably different from what happens in mammals. This difference is closely related to the low buffer power of fish plasma. It has been possible to show with trout blood that the addition of 5 mEq • L-' of lactic acid in ritro induces a fall of pH from 7.95 _+ 0.04 to 7.26 _+ 0.06 and a decrease in [ H C O f ] + [CO~-] from 5.35 _+ 0.8 to 1.80 _+ 0.9 m E q . L - ' . Without making any assumptions about possible changes in ionic composition of plasma which may occur in the trout during this short period of time, the appearance of 5 m E q - L - ' of lactate at physiological pH (i.e., higher than the pK value of lactic acid) involves, in a closed system, the disappearance of a similar quantity of bicarbonate ions (as confirmed by the experimental results given in table 2). The drop in bicarbonate level on the Pco, isopleth = 1.3 Torr of a ( H C O 3 + CO~-) = f (pH) diagram makes it possible to predict the amplitude of the pH decrease attributable to the release of lactic acid. The fact that this estimated drop corresponds to the values actually measured indicates that the observed pH change may justifiably be attributed to the observed release of lactate. In order to estimate quantitatively the effect on blood pH of the increase from 1.28 to 6.16 m E q . L-' in lactate level, a computing model suggested by Stewart (1978) has been used which, for a plasma-like fluid, gives the evolution of pH as a function of plasma
ACID-BASE STATUS OF TROUT BLOOD
381
composition. In human plasma (Pco2 = 40 Torr, T = 37 °C) such a release of lactic acid could only change the pH value from 7.36 to 7.28. The application of the model on the basis of the known values for fish plasma composition shows that the same addition of acid induces a pH decrease from 8.11 to 7.51 and a drop in [HCO3] + [CO~-] from 5.4 to 1.5 m E q . L -I. During severe hypoxia, after the pH decrease, the following stage is a pH recovery which could have been explained within the framework of the above mechanisms by the disappearance of metabolic acids, but the experimental results show that their level is maintained or even increased. Thus, this stage can only be understood in the framework of ionic changes in plasma. Just as 5 mEq • L -1 of lactic acid was enough to explain the pH decrease, so an ionic change which affects the strong ion concentration by 5 m E q . L-~ would be sufficient to explain the increase. Such an hypothesis involving adjustment in plasma ionic composition could explain the experimental results; such a change does not appear in [C1-] which increases only during the ultimate phase of pH recovery; nevertheless, these results bring no new information about the mechanism of adjustment of hydromineral balance which is still insufficiently known in fish. The same model as above indicates that after lactate addition, a 5 mEq • L-l rise in the difference between positive and negative ions is sufficient for a pH recovery to 7.96 and for a [HCO3] + [CO~ ] rise of 4 mEq - L ~. Another question raised by these experiments concerns the observed increases in blood pH and bicarbonates which do not stop when the blood pH has been restored to its original value as might be expected if pH sensitive receptors were involved, but only do so when a return has been made close to the original buffer line. Consequently, at the end of this stage another compensatory mechanism involving a decrease in HCO7 + CO~- concentration is involved as described for moderate hypoxia.
Acknowledgements We wish to thank Professor Claude Peyraud for providing excellent facilities for carrying out these experiments and for very helpful discussions.
References Astrup, P. (1956). A simple electrometric technique for the determination of carbon dioxide tension in blood and plasma, total content of carbon dioxide in plasma, and bicarbonate content in 'separated" plasma at a fixed carbon dioxide tension (40 mm Hg). Scan. J. Lab. Invest. 8: 33-43. Butler, P.J. and E.W. Taylor (1971). Response of the dogfish (So,liorhinus canicula, L.) to slowly induced and rapidly induced hypoxia. Comp. Biochem. Physiol. 39 A: 307 323. Eclancher, B. (1972). Action des changementsde Po2 de l'eau sur la ventilation de la truite et de la tanche. J. Physiol. (Paris) 65:397 A.
382
s. THOMAS A N D G. M. H U G H E S
Holeton, G . F . and D.J. Randall (1967). The effect of hypoxia upon the partial pressure of gases in the blood and water afferent and efferent to the gills of rainbow trout. J. Exp. Biol. 46: 317-327. Hughes, G. M. and R. L. Saunders (1970). Responses of the respiratory pumps to hypoxia in the rainbow trout (Salmo gairdneri R.). J. Exp. Biol. 53 : 529 545. Hughes, G. M. (1973). Respiratory responses to hypoxia in fish. Am. Zool. 13:475 489. Hughes, G . M . (1981). Effects of low oxygen and pollution on the respiratory systems of fish. In: Stress and Fish, edited by A. D. Pickering. London and New York, Academic Press, pp. 121-146. Hughes, G. M. and S. Thomas (1981). Continuous recording of pH, PO2, and PCO 2 of arterial blood in sea bass (Morone labrax) during changes in environmental PO 2. J. Physiol. (London) 319: 8~87P. Itazawa, Y. and T. Takeda (1978). Gas exchange in the carp gills in normoxic and hypoxic conditions. Respir. Physiol. 35: 263-269. Johansen, K. and D. Hanson (1967). Hepatic vein sphincters in elasmobranchs and their significance in controlling hepatic blood flow. J. Exp. Biol. 46:195 203. Maetz, J. (1971). Fish gills: mechanisms of salt transfer in freshwater and seawater. Philos. Trans. R. Soc. London Set. B 262:209 251. Ostroumova, I. N. (1964). Condition of trout blood at adaptation to different oxygen and salt conditions of water, lzv. uses. nauchno-issled. Inst. ozern, rechn, ryb. Khoz. 58: 27-36. Piiper, J. and D. Baumgarten-Schumann (1968). Effectiveness of 02 and CO 2 exchange in the gills of the dogfish (Scyliorhinus stellaris). Re.~pir. Physiol. 5:338 349. Severinghaus, J. W., M. Stupfel and A. F. Bradley (1956a). Accuracy of blood pH and Pco2 determinations. J. Appl. Physiol. 9: 189-196. Severinghaus, J. W., M. Stupfel and A. F. Bradley (1956b). Variations of serum carbonic acid pK' with pH and temperature. J. Appl. Physiol. 9:197 200. Stevens, E. D (1968). The effect of exercise on the distribution of blood to various organs in rainbow trout. Comp. Biochem. Physiol. 25:615q525. Stewart, P. A. (1978). Independent and dependent variables of acid-base control. Respir. Physiol. 33:9 26. Swift, D.J. and R. Lloyd (1974). Changes in urine flow rate and haematocrit value of rainbow trout Salmo gairdneri (Richardson) exposed to hypoxia. J. Fish. Biol. 6: 379-387. Thomas, S., A. Belaud, L. Barth61~my and C. Peyraud (1980). Acid base status in plasma of trout and eel in hypocapnic and normocapnic conditions. J. Comp. Physiol. 140:249 254. Truchot, J. P. (1974). Le transport de gaz respiratoires (oxyg+ne et dioxyde de carbone) par le sang du crabe Carcinus maenas (L.) Th&e de Doctorat des Sciences, Universit6 de Paris, No. OA 9817. Wood, S. C. and K. Johansen (1973). Adaptation to hypoxia by increased HbO 2 affinity and decreased red cell ATP concentration. Nature (London) 237: 278-279.