The Relationship between Nutrition, Hormones and Reproduction in the Blowfly Calliphora Erythrocephala (Meig.): III. The Corpus Allatum in Relation To Nutrition, The Ovaries, Innervation And The Corpus Cardiacum

The corpora allata (c.a.) volumes of reproducing Calliphora erythrocephala females fluctuate cyclically. The c.a. increase in volume during the early stages of egg development and decrease in size during yolk formation (Strangways-Dixon, 1959a, b, 1961 b). The question inevitably arose as to whether the cyclical fluctuations represented variations in activity or in storage and release of hormone, and also as to the means by which the cyclical changes were induced. It was thought that the control of the C.a. might be effected through ovarian or nutritional influences either by way of the nervous system or by humoral means. The experiments to be described are an attempt to elucidate these problems.

Ovariectomy, allatectomy (Thomsen, 1942; Strangways-Dixon, 1961 b) and selective feeding (Strangways-Dixon, 1961a) are described elsewhere—as cited.

The corpus cardiacum (c.c.) was pulled away from the heart with forceps (see Thomsen, 1942, for diagram). A cut was made between these two organs with a microscalpel and the anteriorly directed cardiac-recurrent nerve was loosened from the heart. The c.a. was thus left in isolation yet was held in position by connective tissue and so could be easily located for measurement later. The c.c. on the other hand was pulled away from its normal position by the nervi oesophagi and remained away from the heart, in a position dorsal to the oesophagus. This reduced the probability of the c.a. becoming reinnervated from the normally closely associated c.c.

Preliminary experiments (unrecorded) indicated that in ovariectomized females the c.a. increased in volume cumulatively when constant, small, daily rations of ‘protein’ (mannite in milk) solution were consumed for 16 days. The c.a. of normal females, however, were more erratic and this was thought to be due to the influence of the ovaries. The cumulative effect was, at the time, interpreted as an indication of hormone storage ; for if volume was an indication of activity, then one would expect a constantly applied stimulus to result in an initial fairly rapid change in c.a. volume as activity adjusted itself to the stimulus, followed by a constant state as long as the stimulus remained unchanged. This, however, did not occur. If, on the other hand, these changes were accompanied by storage of hormone, then provided the c.a. remained active its volume would increase cumulatively as hormone storage increased within it. As the latter interpretation fitted the facts, it was temporarily accepted as a working hypothesis around which to plan future experiments.

Assuming that nerves controlled fluctuations of c.a. volume, then denervation would be expected to prevent either storage or release of hormone. In the first instance, the c.a. would remain small and in the second, it would hypertrophy. Experiments were therefore designed to investigate the effects upon the c.a. of: (i) total denervation of the c.a., (2) cutting the cardiac-recurrent nerve, and (3) cutting the nervi oesophagi.

Forty-four females were removed from a sugar culture 2 days after emergence and their c.a. were denervated. Fifteen died during the experiment. On the third day of sugar diet, the culture was divided into two groups. The larger culture of thirty-two flies was provided with meat and sugar, and the twelve controls with sugar. Daily dissections, in which c.a. volumes and egg lengths were measured (see StrangwaysDixon, 1961 b), continued for 8 days. Opposition began on the fourth day of meat diet.

The conclusions drawn from these results (Fig. 1) are: (1) Eggs matured in females with a ‘protein’ diet. Therefore (2) the flies presumably ingested the normal quantity of ‘protein’. (3) The c.a. volumes of flies which were fed on meat, although small, are larger than those of the sugar controls. This suggests that denervated c.a. of ‘protein’-ingesting females remain active. (4) Since the c.a. are not hypertrophied and eggs are mature, it is assumed that the secretions were released. (5) In no single instance is the c.a. larger than the arbitrary unit of 100, whereas the largest glands of normal females (Strangways-Dixon, 1961b, Fig. 10) measure about 210. In fact denervated c.a. appear to remain at volumes typical of those at the bottom of the normal c.a. cycles. If therefore (it was thought) the c.a. cycle was an indication of storage and release of hormone, denervation of the c.a. did not prevent either c.a. activity or the release of hormone, but the ability to store secretions was dependent upon nerves.

If nerves controlled the storage of hormone, then the effect of cutting either the cardiac-recurrent nerve or the nervi oesophagi should indicate the control pathways.

The cardiac-recurrent nerves of forty-two females were cut on the day after emergence. Fifteen died during the experiment. These flies emerged from the same batch of pupae as, and were cultured with, those described in the preceding section. Cellulose paint marks on the thorax were used for identification. Meat and sugar were given to thirty and sugar alone to twelve females. Daily dissections continued for 8 days of meat diet.

The interpretations 1–4 for Fig. 1 are applicable here (Fig. 2), but since the volumes of the c.a. are of intermediate size, the experiment does not throw any light on the question of hormone storage. It is concluded that cutting the cardiac-recurrent nerve does not prevent c.a. activity or the release of hormone.

The second series of experiments (described below) included twenty-four females with denervated c.a. of which eight fed on meat and five fed on sugar survived. The results (not here reproduced) confirm the original data (Fig. 1) and this allows the two investigations to be compared.

Of the forty-five operated females, eleven fed on meat and five fed on sugar survived. Here also (Fig. 3) the interpretations 1–4 for Fig. 1 are applicable, but it is difficult to reach conclusions in relation to storage. It is concluded that cutting the nervi oesophagi does not prevent activation of the c.a. or the release of hormone.

Forty-one females were operated on in precisely the same manner as described above except that no nerves were cut. Of the twenty-five surviving females, twentyone were allowed a diet of meat and sugar and four were controls on sugar alone. Although results (Fig. 4) indicate that the operation itself is not responsible for the very small size of denervated c.a., it is possible to conclude that neither c.a. activity nor the release of hormone are under nervous control.

Since denervated c.a. remain small, it was considered at the time that storage was no longer occurring and that therefore innervation might be necessary for this function. Clearly the next step was to denervate the c.a. in ovariectomized females ; for in these circumstances the inhibitory influences upon the c.a. known to occur during yolk deposition would no longer be effective. If the c.a. remained small in these flies, and if the c.a. of the operated controls hypertrophied as usual, then a nervous control of storage would seem probable.

A hundred females were ovariectomized on emergence and were fed sugar for 3 days ; fifty-two denervations and forty-five control operations were then performed. Three flies had died by this time and a further eleven denervated and seven control flies died during the experiment. Meat was added to the diet on the fourth day after emergence and dissections were carried out daily for 17 days.

As in the previous experiments, the operation areas were examined when the flies were finally dissected to confirm that the c.a. remained isolated. In only a single instance did the c.a. appear to have become re-innervated and even this was not definite.

The c.a. volumes of the operated controls (Fig. 5) rise to a peak on the eighth day of meat diet and then decrease slightly with age. From the fourth until the fifteenth day all c.a. (of the meat-ingesting operated controls) are hypertrophied and are larger than those normally found in reproducing females. The c.a. of flies on sugar diet remain small in spite of denervation and ovariectomy (Fig. 6) and do not appear to differ from those of the ovariectomized operated controls (Fig. 5).

Denervated c.a. of ovariectomized females (Fig. 6) were examined for 17 days of meat diet as compared with 8 days for those of reproducing flies (Fig. 1). Direct comparisons are therefore only possible over the first 8 days. During this period the c.a. in reproducing females measure less than 100 arbitrary units, yet in castrated flies this is the case only on the first day and in one out of four dissections on the second day. By the sixth and seventh days hypertrophied c.a. are present. Over the complete dissecting period of 17 days, only seven out of a total of thirty-three c.a. measure below the 100 mark. Denervated c.a. are able to hypertrophy but, as before, they tend to be slightly smaller than those of the operated controls.

If the storage-release theory was correct, then (it was thought) there was little doubt that denervation did not prevent the c.a. from storing hormone. It was therefore possible to conclude that the hypothesis of a storage-release mechanism under nervous control was no longer tenable. Investigations were therefore concentrated on the humoral aspects of the problem.

Neither c.a. activity, nor storage nor release appeared to be under nervous control. Fluctuations of c.a. volume therefore seemed to be under humoral control from the ovaries and could have been indications of either: (1) storage and release cycles, or (2) variations in activity. Ovarian control might have been either (a) direct (hormonal) or (b) indirect (?). The investigations were planned in such a way as to throw light on all four possibilities.

It was assumed that the ovaries influenced the c.a. during yolk formation either by secreting some substances into—or by removing some substance from—the blood. This led to two alternative approaches to the problem, both of which would necessitate the use of ovariectomized females. The castrated flies were to be fed ‘protein* which would cause the c.a. to enlarge. The intention was then to induce the c.a. to decrease in volume, either by injecting an ovarian hormone (removed from the ovaries of other flies during yolk formation), or by creating an artificial yolk deposition which would remove the ‘protein’ metabolites from the blood stream. Of the two approaches, the latter was thought to resemble the natural events more accurately and therefore would be the more likely to succeed experimentally.

The artificial ‘sink’ for draining away the yolk precursors was created by starving the females of ‘protein’ for several days. It was hoped that this would build up a ‘protein’ debt in the tissues of the fly which, after allowing sufficient time for ingested ‘protein’ to activate the c.a., would drain the former out of the blood stream. The c.a. was expected to respond in one of two possible ways. Either it would increase in volume and remain large, or it would respond with the normal cycle. The former would be an indication of storage and the latter of an activity-volume relationship.

The investigations were performed in four series, each incorporating seventy ovariectomized females. In each experiment, after 6 days of sugar diet, forty-eight of these flies were placed in selective feeding units and thirty-two were cultured on sugar as controls. The forty-eight isolated insects were then divided into four groups of twelve which were fed 1, 2, 4 and 8 cm. of ‘protein’ (measured and fed from capillary tubes) per day, respectively. The ‘protein’ rations were fed for 1, 2, 3 and 4 days in the first, second, third and fourth series of experiments, respectively. Sugar lumps were placed in each feeding unit and, where the quantity of ‘protein’ solution was small (1 or 2 cm. per day), water was presented in cotton-wool balls attached to the outside of the net cover. The flies were removed from the feeding units 24 hr. after being given their final ‘protein’ ration (or later if the food had not been entirely consumed), were isolated in the respective groups and were cultured on a sugar diet. Thereafter two dissections from each group were performed daily and the mean c.a. measurement was used for comparison. Daily dissections were also carried out on the control flies.

A duplicate control series of experiments, using normal females, was undertaken simultaneously for reasons which will become evident below.

Of the total 384 females placed in feeding tubes, eleven normal and eleven ovariectomized flies died.

Of the graphs available, only those necessary to illustrate the various issues will be shown, and these are confirmed by other data not here reproduced.

Figs. 7 and 8 illustrate the induction of c.a. volume cycles in ovariectomized females. Fig. 7 demonstrates that when different quantities of ‘protein’ (1, 2, 4, and 8 cm.) are fed for the same number (three) of days, c.a. volumes increase in relation to the amount of ‘protein’ consumed but decrease when this food is removed. Fig. 8 shows that c.a. volumes increase cumulatively when flies are fed the same quantity of ‘protein’ (eight cm.) on four successive days and that the volumes decrease when ‘protein’ is removed from the diet (after 1, 2, 3 and 4 days). This sensitive response of the c.a. to the consumption of ‘protein’ and in particular the ability of the gland to decrease in volume when ‘protein’ is removed from the diet can best be interpreted —it is suggested—as an indication that variations of c.a. volume represent changes in activity (see Strangways-Dixon, 1959 a, b). It would be difficult to explain these responses in terms of storage and release of hormone since (in the absence of the ovaries) the most likely trigger mechanisms for the release of hormone—a threshold level of concentration of either hormone or yolk precursors—are unlikely in Pew of the graded cyclical responses. Furthermore, these trigger mechanisms would not allow the c.a. to hypertrophy. This does not mean that there is no accumulation (as opposed to active storage) of hormone in the c.a., for this would certainly occur if the production of hormone exceeded the rate of diffusion( ?) to the outside. One cannot entirely rule out the interpretation that the decrease in c.a. volume is due to the withdrawal of food material from the c.a. itself to organs with higher priorities. But this seems unlikely since the amount of food obtained from so small an organ would be negligible. Yet even if this were the case, the effect upon the c.a. would probably be that of reduced activity.

Cycles similar to those described above may be induced in normal females (Fig. 9). C.a. volumes increase cumulatively with successive days of ‘protein’ ingestion (8 cm. per day) and decrease when ‘protein’ is removed from the diet (after 1, 2, 3 and 4 days). The cycles after 1, 2, and 3 days of ingestion are not induced by yolk formation, for only the flies fed on ‘protein’ for 4 days produced yolk. The first three cycles are therefore presumably caused by artificial means (described above) and the fourth (heavy line) is induced by yolk formation. The importance of having normal females as controls now becomes apparent. They are a means of determining the stage at which the artificial ‘sinks’ become filled. For at this point, it is suggested, the metabolites are diverted to the ovaries and yolk deposition occurs [this presumably accounts for succeeding reproductive cycles being more rapid than the first in normal reproducing flies (Strangways-Dixon, 1961 a). The first cycle is probably slow because the tissues, which are not yet fully grown, act as ‘sinks’ for ‘protein’ metabolites].

Fig. 8 shows that the c.a. of ovariectomized females behave similarly to those of normal females up to the point at which the ‘sinks’ become filled. After this stage however, the c.a. hypertrophies (heavy fine). It is suggested that the hypertrophied c.a. represents hyper-activation of the c.a. by unusually high concentrations of c.a.-activating substances (yolk precursors) which, in the absence of the ovaries, have no alternative but to remain in the blood. It is concluded therefore that the action of the ovaries upon the c.a. is indirect and that this influence is effected by the ovaries removing c.a.-activating substances from the blood for the purpose of yolk formation (see Strangways-Dixon, 1959 a, b).

During these experiments, and in many others, ovariectomized females were dissected for examination of gut contents. Flies with hypertrophied c.a. were often observed to have little or no ‘protein’ in their guts, and a comparison with reproducing females showed no obvious differences in ‘protein’ content. These observations support the suggestion implied above that the action of nutrition upon the c.a. is humoral rather than from the gut. Which constituents) in the ‘protein’ mixture attracts the fly as a food and which is effective in activating the c.a. is not yet known. A biochemical approach to this problem should be most interesting.

If the above hypothesis were correct, it was reasoned, then c.a. should not show the usual decrease in volume in flies forced to ingest ‘protein’ in abnormally large quantities during yolk formation.

On the sixth day of sugar diet, thirty females were isolated and were fed 8 cm. ‘protein’ (from capillary tubes) daily for 4 days. During this period, five flies were dissected to check ovarian growth which reached the yolk stage on the fourth day. The remaining flies were then divided into two groups of eighteen and seven and the sugar lumps were removed from all feeding units. The group of eighteen females was supplied with two capillary tubes, one containing ‘protein’ and the other a two-to-one mixture of ‘protein’ and carbohydrate. A more concentrated stock solution of carbohydrate (300 g. in 100 ml. water) was made up so that the final mixed solution would contain carbohydrate in the normal concentrations. These insects had to ingest the carbohydrate mixture to avoid starvation and so were obliged to ingest ‘protein ‘as well. The remaining seven flies were given only carbohydrate solution which was of the same concentration as the sugar in the ‘protein’ mixture.

Daily dissections were made of eleven ‘protein’-fed flies but the remaining seven and the seven controls on sugar were not dissected until the eighth day in feeding units. Results (Fig. 10) show that with a single exception, in a fly dissected before the beginning of yolk deposition, the c.a. of flies which were force-fed ‘protein ‘are larger than those which ingested only carbohydrate during yolk formation. It is noticeable that the latter, without exception, developed mature eggs. The quantity of ‘protein’ consumed during the pre-yolk stage was therefore sufficient for yolk formation.

Differences in c.a. volume of greatest significance would naturally occur at the completion or approach to completion of yolk formation—between egg lengths of 60 and 65. At this stage, the greatest amount of ‘protein’ will have been removed by the ovaries. After the size of 65, shell formation and delayed oviposition may have begun and the effects of active deposition would then have ceased. Between the sizes of 60 and 65 the c.a. of flies on the enforced ‘protein’ diet range from between 165 and 214, while those on a carbohydrate diet measure from 48 to 105. There seems to be little doubt that the normal decrease in c.a. volume during yolk formation can be considerably reduced by the ingestion of greater than normal quantities of ‘protein’. It therefore seems unlikely that the c.a. cycle is a response to either nervous or hormonal influences emanating from the ovaries. This result, in fact, confirms the conclusions reached in the preceding sections.

The corpus cardiacum (c.c.) in Calliphora is in very much closer physical association with the c.a. than with the median neurosecretory cells whose secretions are known (Thomsen, 1954) to enter the c.c. It seemed possible therefore that the c.a. secretions might also enter the c.c. This possibility was investigated in a small-scale experiment whose findings indicate, though by no means conclusively, that c.a. secretions are stored in the c.c.

Thirty females were divided into three groups of ten. The c.a. were removed from the females of one group, the c.a. and the c.c. from the second group and the c.c. alone from the last. The operations were performed on the third day after emergence, and meat was added to the sugar diet on the fourth day. All surviving females were dissected on the eighth day of meat diet and egg sizes are shown in Fig. 11.

The removal of the c.c. alone allows eggs to mature normally in two of the three surviving females. Extirpation of the c.a. by itself prevents maturation of the eggs, but several of the flies contain eggs with yolk. When the c.c. and the c.a. are removed however, none of the eggs contain yolk. This suggests that before the operations at least some c.a. hormone was secreted and that it was stored in the c.c.

The contents of this paper, together with the relevant data recorded in the first and second papers of this series (Strangways-Dixon, 1961a, b), will be discussed in the fourth paper of the series.

My sincerest thanks are due to Prof. V. B. Wigglesworth, F.R.S., for encouragement, advice and facilities, to my wife for her help in preparing the manuscript and to the Colonial Office which supported this investigation with a research grant.