On the Function of Haemoglobin in Chironomus After Oxygen Lack

The function of haemoglobin in the large chironomid larvae of the Chironomus plumosus group has been studied experimentally by Ewer (1942). She compared the oxygen consumption of normal larvae at different pressures of oxygen with that of larvae whose respiratory pigment had been rendered functionless by treatment with carbon monoxide, and from the differences between the two deduced the extent to which the haemoglobin is of functional value to the animal in the transport of oxygen. Her results show that at 17° C. the haemoglobin of Chironomus larvae (previously kept in aerated water) does not function at air saturation of the water, but only at oxygen concentrations below 3 ml./l. (44% saturation). This can be correlated with the common occurrence of Chironomus larvae in stagnant habitats known to be more or less depleted of oxygen from time to time.

The work of Ewer in part repeats, under careful experimental conditions, similar work done by Harnisch (1936), whose experimental technique is open to criticism. Harnisch studied both animals previously kept in oxygen gas and animals which had been subjected to severe oxygen lack by keeping them about 20 hr. in nitrogen gas before experiments. The nitrogen-treated larvae when replaced in aerated water in a Barcroft manometer had an oxygen uptake 160% that of the oxygen-treated larvae. This indication of the repayment of an oxygen debt persisted for about 3 hr. The increased oxygen uptake of the nitrogen-treated animals was observed not only in aerated water but also at lower oxygen pressures, even at 14% air saturation. The oxygen consumption of nitrogen-treated larvae which had also been subjected to carbon monoxide was, however, lower than that of nitrogen-treated larvae whose haemoglobin had not been converted into carboxyhaemoglobin; this was so in water at all oxygen pressures at and below air saturation. Harnisch concluded that when the metabolic rate rises in recovering from oxygen lack the haemoglobin becomes functional in oxygen transport at all pressures of oxygen, and that the increased oxygen consumption of recovering larvae is, thanks to their haemoglobin, maintained irrespective of the oxygen content of the water. That the larvae, which must frequently experience anaerobic conditions in nature, should, thus be able to recover from oxygen lack even in water containing very little oxygen, Harnisch considers to be of paramount importance in assessing the function of the haemoglobin. This work is, however, open to criticisms, some of which have been made by Ewer (1942). The temperature varied from 16 to 230 C. in experiments supposed to be comparable and Harnisch used carbon monoxide pressures high enough to have had an inhibitory effect on cellular oxidations. He dismisses this possibility with the statement that although Warburg (1926) had shown such a concentration to have an inhibitory effect on yeast he himself considers that ‘bei der Hefe der Kontakt zwischen dem Giftgas und den Zellen doch wohl wesentlich inniger war als bei meinen Tieren’! He applies no statistical tests to his data, and two series of estimations of the metabolism in aerated water of animals after nitrogen treatment (Table 2, p. 396 and Table 6, p. 406) give significantly different averages.

For these reasons a further investigation of the metabolism of chironomid larvae after oxygen lack was considered necessary.

The chironomids used were the final instar larvae of Chironomus plumosus L. collected from the mud of Regent’s Park Lake, London. The species was identified by means of Edwards’s key (1929) from adults which emerged in the laboratory, and checked by comparison with the collection at the British Museum. I wish to thank Dr J. Smart for allowing me access to this collection. Larvae were collected 1 or 2 days before experiments and were kept in the laboratory in aerated water in shallow dishes with a thin layer of mud, under which conditions they remained healthy, and successfully pupated and emerged.

The respiration of larvae was measured in an apparatus in which the larvae were subjected to a continual slow stream of water, their oxygen consumption being calculated from the difference in oxygen content of the water before and after passing through the respiratory chamber. A diagram of the apparatus is given in Fig. 1. A 5 1. aspirator (a), containing aerated buffered water (0·004 N solution of NaHC0₃) was connected by way of a water-mixer (b), to the respiratory chamber (c). This consists of an 11 cm. length of wide glass tubing, 1·5 cm. diam., blackened on the outside except for a longitudinal unpainted slit about 5 mm. wide through which the behaviour of the larvae could be observed without exposing them to excessive light, which has an activating effect. Perforated porcelain disks on the rubber bungs prevented the larvae from crawling out of the respiratory chamber. The three-way tap (f) and capillary outflow tube (d) allowed samples of the water flowing into and out of the respiratory chamber being taken for analysis of dissolved oxygen. This was determined by the syringe-pipette micro-Winkler method described by Fox & Wingfield (1938). The nozzle of the syringe-pipette was inserted into Xhe rubber tubing at (f) or (d) and water samples of 1·4 ml. thus removed. The rate of flow through the apparatus was kept steady by maintaining a constant level in the aspirator (a) by means of a continuous inflow from flask (g) and a simple overflow device, and by firmly clamping the whole system of tubing to prevent any shift in position. The rate was determined by the length of time taken by the outflowing drops to fill a given volume. The respiratory chamber, together with the inflow and outflow tubes, were supported in a large thermostatic water bath.

Between thirty and forty larvae were used in each experiment. With a water flow of 1·3 ml./min. these larvae lowered the oxygen content of the water by about 20% at 170 C. From the difference in oxygen content of the water before and after passing through the respiratory chamber and the rate of flow of the water, the oxygen consumption of the larvae was calculated in cu.mm, oxygen/g. (wet weight)/hr.

The 3 1. bottle (h) contained nitrogenated water (oxygen content less than 0·1 ml./l.); this could be driven through the respiratory chamber by the pressure of gas from a nitrogen cylinder, the larvae being in this way subjected to a period of oxygen lack.

Experiments were of about 24 hr. duration. Larvae were put into the respiratory chamber in flowing aerated water and left for about 2 hr. in order to reach a steady state of activity. The oxygen contents of inflowing and outflowing waters were then determined at intervals of approximately 45 min. for the next 3 hr., and from these determinations the average metabolic rate of the animals was determined. The inflow of aerated water was then stopped and nitrogenated water slowly forced over the animals. The larvae were left thus for 16 hr., at the end of which period the nitrogenated water supply was cut off and aerated water once again allowed to flow through the apparatus. Oxygen samples of the outflowing water were taken at intervals of 5 min. for the next hour and at frequent intervals (generally 15 min.) after that until the oxygen uptake became constant and equal to that found before the anaerobic period. A complete record of the metabolism of larvae recovering from anaerobiosis was thus obtained.

The rate of replacement of oxygen-free water present in the respiratory chamber by inflowing aerated water was determined by taking frequent oxygen samples of the outflowing water with no animals present in the chamber. The mixing was slow, nearly an hour elapsing before the water in the chamber was completely replaced by fully aerated water. Larvae therefore commenced recovery from anaerobiosis in water of low oxygen content and were subsequently subjected to progressively higher oxygen concentrations until air saturation was finally reached.

Most experiments were carried out at 17° C. One series, however, was made at 1-20 C. This low temperature was obtained by surrounding the respiratory chamber and tubes with crushed ice. The purpose of these experiments was to reduce the activity of the animals while estimations of the metabolic rate were being made: the larvae spent the overnight anaerobic period, however, at 17° C. and were thus allowed to accumulate as much oxygen debt, if any, as those in previous experiments. The larvae were again chilled at the end of the overnight period before the aerated water was turned on.

A number of experiments were made on the recovery from oxygen lack of larvae whose haemoglobin had been rendered functionless by conversion to carboxyhaemoglobin. The larvae were kept in darkness for 3 hr. previous to experiments in aerated water containing sufficient carbon monoxide-saturated water to make the pressure of dissolved carbon monoxide one-sixth that of the oxygen. This concentration of carbon monoxide was sufficient to convert the haemoglobin entirely into the carboxy-compound, but at this low concentration it was very unlikely to have had any inhibitory effect on cellular oxidations, for which, owing to the low affinity of oxidases for carbon monoxide, relative pressures considerably greater than one are necessary (Keilin, 1929; Wolsky, 1938). Experiments with larvae so treated were carried out in the same manner as those with normal larvae, except that the water contained 0·2 ml./l. carbon monoxide to prevent dissociation of the carboxyhaemoglobin. The small amount of light introduced through the slit in the respiratory chamber did not cause any dissociation. Individual experiments with carbon monoxide-treated animals were alternated with control experiments with normal animals. This was a necessary precaution to ensure that larvae in the same physiological state were being compared, since seasonal differences in metabolic rate were found to occur.

At the end of all experiments the larvae were dried on filter paper and weighed. In the carbon monoxide experiments a drop of the animal’s blood was then examined with a hand spectroscope: the absence of fading of the absorption bands after the addition of sodium hydrosulphite indicated that no dissociation of carboxyhaemoglobin had occurred during the experiment.

The results of a series of seven experiments on normal animals at 17° C. are given in Table 1. Fig. 2 shows graphically the changes in metabolic rate during the course of the experiments. The oxygen uptake of an average basal value of 191·7 ±7·8^* cu.mm./g. (wet weight)/hr. before nitrogen subjection rose within 5 min. of the introduction of aerated water to a significantly higher value, the average metabolic rate 5-15 min. afterwards being 235·0 ±13·8. This marked increase only lasted for about 10 min., subsequent values from 20 min. to 1 hr. after the end of anaerobiosis being only slightly greater than the basal rate of the previous day. The statistical significance of this slight increase is very doubtful: the averages are 204·5 + 9·3 after, as compared with 191·7 + 7·8 before, anaerobiosis, with the large probability value (p) of 0·27. Further, an analysis of variance of these data did not indicate a significant increase in metabolic rate. At the end of 1 hr. the oxygen consumption had returned to the basal rate.

The initial rise in oxygen consumption was accompanied by great activity on the part of the animals: the larvae, which had been completely inactive in nitrogenated water, started violent undulatory movements of the body upon the introduction of aerated water. It seemed likely that the increased oxygen uptake was due to this activity rather than to the repayment of an oxygen debt; to test this experiments were made in which the activity changes were eliminated by chilling the animals. The results of six such experiments are given in Table 2 and Fig. 3. The basal metabolic rate of such larvae was 25·0 + 2·4 cu.mm./g./hr. After the 16 hr. anaerobic period the metabolism rose slowly during the first 25 min. to a value about 160% of the basal rate and maintained this high level for at least 2 hr. The difference in metabolic rate before and 10 min. to 2 hr. after the anaerobic period is highly significant.

In interpreting the results at 17° C., therefore, the initial increase in oxygen consumption must be attributed to the great increase in activity of the larvae at that time, since immobile, chilled larvae do not show it. Larvae at both temperatures, however, showed a subsequent more prolonged increase in metabolism, and although at 17° C. the increase is so slight that its reality is doubtful, at the lower temperature it is greater.

The results of a third series of experiments in which the recovery from anaerobiosis of normal animals was compared with that of carbon monoxide-treated larvae are given in Tables 3 and 4 and Fig. 4. The metabolism curve is of the same general form as that already described for normal larvae at 17° C. except that the basal level is lower, being 140·3+4·0 cu.mm./g./hr. compared with 191·7 + 7·8 cu.mm./g./hr. previously. This must be due to a seasonal change, these last series of experiments having been made in late September on animals which would have over-wintered as larvae, while the first series was made in May and June on larvae of a summer generation.

The basal metabolic rate of larvae with carboxyhaemoglobin is seen to be the same as that of untreated larvae. This confirms Ewer’s conclusion that in aerated water the haemoglobin is not functional, and shows that treatment with carbon monoxide had not affected cell oxidations in the larvae. The metabolism of the treated larvae after the anaerobic period, on the other hand, differed from that of normal larvae. The initial increase in oxygen consumption caused by activity did not occur, although the activity of the treated larvae appeared in no way diminished.

The larvae show this increased activity at a time when, owing to the slow rate of flow through the respiratory chamber, the inflowing aerated water has only partly replaced the nitrogenated water, the oxygen concentration inside the chamber varying from 3 to 5 ml./l. (45-75 % saturation) during the 10 min. that the activity is most apparent. Thus the failure of the carbon monoxide animals to increase their oxygen uptake with their increase in activity indicates that within this range of oxygen pressures the haemoglobin is normally functional, enabling untreated larvae to pick up the extra oxygen demanded by increased activity; in the carbon monoxide larvae the energy for such activity must have been partly derived from anaerobic processes.

Ewer (1942) showed that in Chironomus dorsalis at 17° C. the haemoglobin of larvae only became functional at oxygen concentrations of 3 ml./l. (44% air saturation) or less; the above experiments indicate that the haemoglobin of my larvae was being used in oxygen transport at a slightly higher concentration. This difference is probably accounted for by the fact that with C. plumosus the increased oxygen demands of the tissues during the period of great activity extend the range of external oxygen pressures at which the haemoglobin functions. The larger size of C. plumosus larvae may also intervene through their relatively smaller surface.

After the first 25 min. of recovery, in other words by the time that the larvae are once again in fully aerated water, the oxygen uptake of the carbon monoxide-treated larvae is equal to that of the untreated animals, and, as with them, is maintained for some time at a level slightly higher than the basal value. With the treated larvae this secondary increase was, in fact, greater than with the untreated animals and continued for a longer time, in some cases for 2 hr. after the beginning of recovery.

In all experiments described the larvae showed an increased metabolic rate during recovery from anaerobiosis. To what extent, however, this represents the repayment of an oxygen debt at 17° C. is uncertain, for the initial well-defined increase in oxygen consumption is due to an increase in activity of the larvae, while statistical analysis of the subsequent more prolonged increase renders its reality doubtful. At 1° C., however, the rise in oxygen consumption lasting about 2 hr. is statistically significant and must be interpreted as the repayment of an oxygen debt. This may also be the case at 17° C. (without and with carbon monoxide), the statistical improbability being due to the small size of the rise.

It is possible from the experimental data to calculate the extent of such a debt in terms of cu.mm, oxygen consumed by recovering larvae over and above that which they would normally consume in the same time; thus in Figs. 2 and 3 this extra oxygen consumption is represented by the area between the curves of observed metabolic values and the lines drawn at the level of the basal metabolism of the larvae. The increased oxygen consumption of the winter and summer larvae at 17° C. amounts to 12·0 and 14·0 cu.mm, oxygen/g. respectively. Had these larvae not been subjected to anaerobic conditions the oxygen they would have consumed during the anaerobic period amounts to 2240 cu.mm, for the winter and 3080 cu.mm, for the summer larvae. The extra oxygen consumed during repayment of an oxygen debt is thus only 0·5 % of that missed during the anaerobic period. Such a very small oxygen debt of Chironomus is in striking contrast to the debts accumulated by vertebrate tissues after anaerobic activity, in which the amount of increased oxygen consumption is proportional to the length of the anaerobic period.

The extent of repayment of oxygen debts among invertebrates, however, varies considerably. Well-defined debts have been recorded in Periplaneta and Lumbricus by Davis & Slater (1928), in Planorbis by Borden (1931), in grasshoppers by Bodine (1928) and in Tenebrio, Cryptocercus and Zootermopsis by Gilmour (194002, 6, 1941). There is evidence that in some of these animals a proportion of the organic acids is resynthesized to carbohydrate: in others a greater repayment of oxygen debt indicates their complete removal by oxidation. In yet other invertebrates oxygen debts are clearly not repaid in full; thus the ciliate Tetrahymena gelei only repays 25% (Thomas, 1942, cited in von Brand, 1945), and Eustrongylides ignotus 30% (von Brand, 1942) of the debts incurred. Both these animals are known to excrete organic acids and in this way lessen the need for repayment. Excretion of the waste products formed may also account for the small size and variability of the oxygen debts recorded in Planaria (Lund, 1921), Nereis (Hyman, 1932), Urechis (Hall, 1931) and Tubifex (Dausend, 1931; Harnisch, 1935, 1936). Finally, Harnisch (1942) could find no evidence of any repayment of oxygen debt in the larvae of Chironomus bathophilus.

Studies of the anaerobic glycogen metabolism of invertebrates which successfully withstand periods of anaerobiosis in nature have shown that various higher and lower fatty acids are formed by glycogen fermentation (for literature see von Brand, I945). In intestinal parasites these fatty acids are either converted to non-toxic fat, and stored, or, more commonly, excreted by the animal. In their ability to excrete the products of anaerobic metabolism they are thus adapted to withstand prolonged periods of oxygen lack. That free-living animals which live in oxygen-poor surroundings should also be able to remove the products of carbohydrate fermentation in this way is not surprising, their respiratory problems being of the same nature as those of internal parasites. In aquatic invertebrates the elimination of such waste products might be either by excretory organs, or in the case of small animals by diffusion across the surface of the body, provided that this is permeable to them. The repayment of oxygen debts by animals such as Mya and Anodonta (van Dam, 1938) is possibly made necessary by the shutting of the valves of the shell during anaerobiosis, which would limit the ready diffusion of waste products out of the body. The accumulation of toxic waste products within the body, and the need for oxygen to remove these in the repayment of an oxygen debt, is most apparent in animals least adapted to withstand oxygen lack, such as terrestrial invertebrates and the vertebrates.

It seems probable therefore that the absence of an oxygen debt at all proportional to the length of the anaerobic period in Chironomus larvae is due to the removal, either by diffusion or excretion, of the waste products as they are formed. With this in view the water in which a number of larvae had been kept anaerobically overnight was tested for lactic acid. This was found to be present. Although lactic acid is only one of several organic acids produced by the incomplete breakdown of glycogen, it at any rate was removed from the larvae as such and not retained in the body. The observed slight increase in oxygen consumption after anaerobiosis would thus account only for the transformation of those products which remained within the body owing to their indiffusability. In this case the greater repayment of debt by chilled animals could be accounted for by a greater accumulation of such products within the body due to a reduced rate of diffusion or excretion. The fatty acids formed during anaerobiosis may also to some extent be converted into fat within the body, as in certain endoparasites, and the need for repayment of oxygen debt further reduced in this way. Evidence supporting this is given by Harnisch (1939) who found that towards the end of a long anaerobic period (23-24 hr.) the production of fatty acids decreases and fat is formed in the bodies of the larvae.

It is interesting that Harnisch (1942) could find no evidence of repayment of oxygen debt after anaerobiosis in Chironomus bathophilus, a species closely related to C. plumosus and capable of living in oxygen-poor lakes. Harnisch interprets this absence of oxygen debt as indicating that the larvae respire anaerobically at all times, even when oxygen is present. It may, however, be that in this euroxybiotic species the mechanisms for the rapid elimination of metabolic waste products have been perfected.

Harnisch (1936) has recorded considerably greater repayment of debt in Chironomus thummi than I found in C. plumosus. The average metabolism of his normal animals lay between 260 and 278 cu.mm./g./hr. while that of animals after 15 hr. in nitrogen varied in different experiments from 368 to 444 cu.mm./g./hr. Thus there was a 41-59% increase in metabolic rate, persisting for two or more hours. Correlated with this increased metabolism he noted that the larvae showed regular undulatory movements which continued for many hours. Since larvae previously kept in oxygenated water did not show such activity he regarded the movements as indicating ‘dass im Körper der Tiere Produkte anoxybiotischen Stoffwechsels vorhanden sind und somit gesteigerter Sauerstoffverbrauch (“sekundare Oxybiose “) besteht’. An alternative interpretation, not suggested by Harnisch, is that the increased metabolism is a direct result of the increased activity rather than of an oxygen debt. The fact that he sometimes found larvae which had previously been kept in oxygenated water to be making vigorous respiratory movements, in which case the animals also showed a high oxygen consumption, supports this interpretation. Harnisch’s increased oxygen uptake would then be parallel to that recorded during the short period of intense activity in my experiments. That his larvae remained active much longer than mine may be due to a specific difference. In view of his lack of appreciation of the activity factor, his results should be treated with caution.

Harnisch also conducted a series of experiments on carbon monoxide-treated larvae after oxygen lack. He found these larvae to have a lower oxygen consumption than untreated larvae at all oxygen pressures at and below air saturation of water, and from this he concluded that when their metabolism is raised in paying back an oxygen debt the haemoglobin is functional all the time. During preliminary carbon monoxide treatment, however, he notes a reduction in the activity of the animals, their typical undulatory movements being no longer apparent, which is perhaps not surprising since he used pure carbon monoxide! The differences which he subsequently records between treated and untreated animals may well be caused merely by this activity difference.

In my series of experiments with carbon monoxide-treated larvae the animals, without functional haemoglobin, were capable of maintaining a metabolic rate greater than normal in aerated water, so that the respiratory pigment is of no significance in paying back a small debt in aerated water. That my larvae with carboxyhaemoglobin should actually have a slightly higher oxygen consumption than normal larvae, appearing thus to pay back a greater debt, is curious; it can hardly be attributed to the fact that during their short activity period the movements were partly anaerobic, for any such additional debt incurred in only io min. must have been very small.

Harnisch’s conclusion that after oxygen lack the larvae use their haemoglobin at all oxygen pressures is thus not confirmed by my experiments. The difference between our results may in part be due to the activity difference between the two species of larvae, the increased oxygen demand of the tissues of his active animals extending the range of function of the pigment so that it is used even at high oxygen pressures. During the short time that my larvae were most active the haemoglobin was functional, but this period coincided with a low oxygen pressure in the water, at which in any case the pigment would be used (Ewer, 1943). In both Harnisch’s larvae and mine the possession of haemoglobin enabled the larvae to show an increased degree of activity aerobically.

The functional significance of the immediate violent activity of the larvae after anaerobiosis when water containing oxygen was introduced is only apparent when the animals are considered in their natural habitat. Chironomus larvae live in U-shaped tubes in mud and remain in their tubes during periods of oxygen lack.

Violent undulatory movements upon the reappearance of oxygen in the overlying water will have the effect of producing a current of water through the tube, thus both washing away any accumulated carbon dioxide or organic acids present in the tube and providing a continuous supply of water from which oxygen may be extracted. Since in nature the oxygen content of water previously depleted of oxygen is unlikely to rise as rapidly as in the experiments the maintenance of a ventilation current is important. Moreover, thanks to the haemoglobin, the undulatory activity necessary to set up such a current can be maintained aerobically in water containing little oxygen, enabling the animals to re-establish full aerobic metabolism as quickly as possible after a period of oxygen lack.

In this connexion I have now undertaken a study of the behaviour of larvae in their tubes in as natural conditions as possible. It may well be that their responses to oxygen lack in such circumstances are different from those under the very artificial experimental conditions described above.

I wish to thank Prof. H. Munro Fox, F.R.S., in whose laboratory this work was done, for advice and help.