ABSTRACT
The finding (Evans & Dethier, 1957) that glucose and trehalose are the normal blood sugars of the blowfly Phormia regina has been confirmed.
It is estimated that most of the flight energy is derived from the oxidation of trehalose and, to a lesser extent, of other sugars found in the blood.
Several lines of evidence indicate that the concentration of blood trehalose normally regulates the rate at which energy is expended by the flight muscles during flight. ‘Exhaustion’ results when trehalose cannot be supplied to the flight muscle at the necessary rate.
Fat body is the chief source of blood trehalose; endogenous and exogenous substrates are used for its synthesis. The rate of blood trehalose synthesis can be very rapid, almost compensating for the rate of utilization by the flight muscle during flight.
It appears, therefore, that the intensity of flight is determined largely by the interaction of two rate processes : the rate of trehalose utilization (flight muscles) and the rate of trehalose synthesis (fat body).
A diagram is presented which accounts for the establishment of glucose and trehalose as the normal blood sugars of this fly and summarizes our findings on the transfer of carbohydrates between various tissues.
The blood volume (6–7 μl.) has been estimated from dilution of injected 14C-inulin. This volume is not changed appreciably by flight
INTRODUCTION
Insects of the order Diptera possess flight muscles capable of contracting at very rapid rates for long periods. The energy for this activity is derived mainly, if not exclusively, from the aerobic oxidation of carbohydrate (Chadwick & Gilmour, 1940; Davis & Fraenkel, 1940; Williams, Barness & Sawyer, 1943; Chadwick, 1947; Wigglesworth, 1949; Hocking, 1953; Clements, 1955; Hudson, 1958). Glycogen has often been implicated as the reserve substrate for flight, mainly on the basis that this substance appeared to be the only carbohydrate present in appreciable amounts and that it decreased in amount as a function of flight duration. However, Wyatt & Kalf (1956, 1957) more recently showed that the non-reducing disaccharide, trehalose, is found in large concentrations in the haemolymph of several insects, and two other studies independently demonstrated trehalose to be an insect blood sugar (Howden & Kilby, 1956; Evans & Dethier, 1957). Since that time the list of insects having trehalose as a blood sugar has been extended, and several studies have been made of the enzymes and pathways involved in the metabolism of this sugar (cf. Wyatt, 1961 ; Sacktor, 1961). Although several suggestions have been made, there is little experimental evidence on the physiological role of this disaccharide in insects. Since practically all of the studies concerning the flight substrate in Diptera were carried out prior to the discovery of trehalose, the main purpose of our work was to examine the relationship between the two parameters, flight and trehalose, in the blowfly. A preliminary account has been published elsewhere (Clegg & Evans, 1961).
MATERIALS AND METHODS
Phormia regina Meigen, the blowfly used in this study, was reared by the method of Hill, Bell & Chadwick (1947). Only males were used to avoid complications arising from sex differences.
Analytical methods
Whole blood was collected from adult flies by the method of Evans & Dethier (1957). These workers found that the very small number of formed elements in adult blood do not contain an appreciable amount of carbohydrate; accordingly, whole blood was used in this study.
The identification and separation of carbohydrates was carried out by paper chromatography (Evans & Dethier, 1957). For quantification the carbohydrates were eluted from the paper with water into small beakers. These eluates were then frozen on dry-ice, and taken to dryness in vacuo over CaCl2. A known volume of water was added, and an appropriate aliquot taken for colorimetric assay of carbohydrate (Evans & Dethier, 1957) or radioactivity determinations. When entire chromatograms were assayed, the glucose and trehalose areas were cut out for elution and the remainder of the paper was eluted serially in 2 cm. strips.
Glucose-U-14C was obtained from Nuclear-Chicago; mannose-U-14C, galactose-r-14C, fructose-U-14C, and inulin-carboxyl-14C were obtained from ‘Calbiochem’, Los Angeles, California. All of these were demonstrated to be chromatographically pure before use. Radioactive samples were plated on stainless steel planchettes and assayed with a thin-window Geiger-Miiller tube in an automatic counter. Corrections for self-absorption were made when necessary, most samples being counted at infinite thinness. All counting errors are given at the 1 % level. Isotopically labelled trehalose, not available commercially, was synthesized by incubating intact fat body of the roach, Leucophaea maderae, with glucose-U-14C of high specific activity. This tissue converts exogenous glucose into trehalose and liberates it into the medium with high efficiency (specific activity of the trehalose synthesized was 40–60 % of the specific activity of the precursor glucose). Isolation and purification of 14C-labelled trehalose from the incubation medium was accomplished by the above-mentioned chromatographic procedures.
Bodenstein’s (1946) saline, buffered to pH 7 with 0·01 M phosphate, was used for tissue incubations and as a carrier for injected sugar solutions.
Injections of sugar solutions were made with a micrometer-driven microinjection apparatus, containing a 50 μl. syringe equipped with a 27-gauge needle. To insure minimum damage with such injections, a very fine-tipped glass needle (made on a micro-electrode puller) was glued over the metal needle. Injections were made into the scutellum after the fly had been mounted on a clay support. The volume and error of delivery of this apparatus was estimated by determining the amount of glucose delivered by unit turn of the micrometer after the syringe had been filled with a known concentration of glucose. In every case, 0·524 μl. ( ± 0·008 S.E.) was the volume injected.
Tissue preparations
The half-thorax preparation was made by anaesthetizing the fly with carbon dioxide (Williams, 1946), severing the abdomen at its junction with the thorax, and then carefully pulling off the head with attached mid-gut. After removal of the wings and legs, this isolated thorax was hemisected along the mid-line with a thin razor blade. The thoracic ganglion and salivary glands were removed under a microscope. If the dissection is done carefully, the flight muscle remains intact, and comprises the bulk of the tissue in such ‘thorax-halves’, although a few fat body cells are also present. These preparations were rinsed in saline three times, gently blotted after each rinse, and then placed in the incubation flask. For weight determinations the muscle mass was dissected from the thorax, blotted gently, and placed on a small pre-weighed piece of aluminum foil. For determination of dry weight, the muscle was placed in an oven at 100° C. for 60 min., at which time water loss was complete. All weighings were carried out with a Roller-Smith balance (sensitive to 0·05 mg.).
Preparations of intact fat body were obtained from the abdomens of male flies which had been reared on 2M. glucose from eclosion. The abdomen was severed at the thoracic junction, and the most posterior abdominal segment was carefully removed under saline. The abdominal viscera were removed and the larger pieces of fat body teased out with glass needles and transferred through three saline washes with small glass pipettes. The pipettes were treated with silicone to prevent the fat body from sticking to the glass. The abdomens of three to five flies usually contained enough fat body for a single incubation (about 1·5 mg. dry weight). This ‘pooled’ fat body was then incubated in 0·1 ml. of the appropriate medium. Tissue dry weight was determined in the same manner as for flight muscle. Incubations were carried out at room temperature (25–27° C.) with gentle shaking.
Flight procedure
Males, 4–5 days after eclosion, were used for all flight studies. They were maintained on a 2M glucose solution sprayed twice daily on filter paper. Such flies can be considered replete with respect to carbohydrate stores. The flies were mounted for flight, using a method similar to the one Chadwick & Gilmour (1940) developed for Drosophila. This consisted of suspending the fly from a thin cardboard strip attached at one end to the dorsal abdomen with a small drop of glue, and at the other end to the substratum. The fly was now suspended in a fixed position. The tarsi were removed to promote flight (Friedman, 1959). Continuous flight could be initiated in about 60 % of such flies by brushing the ventral surface of the body. Flight was often spontaneous under these conditions. Measurements of the wing-beat frequency were made with a ‘Strobotac’ (type 631-B, General Radio Company) by placing the emission tube 6 in. from the suspended fly. Use of the strobotac on ‘flying’ insects has been outlined by Chadwick (1939), Williams & Galambos (1950), and Williams & Chadwick (1943). All flight experiments were carried out under relatively constant conditions of light intensity, humidity and temperature.
RESULTS AND DISCUSSION
Blood sugars and the flight process
Blood carbohydrates as a function of flight duration
To our knowledge the only previous indication that blood trehalose is utilized by the flight muscles of Diptera during flight was the decrease in blood trehalose concentration observed during continuous flight of the blowfly, Phormia regina (Evans & Dethier, 1957). This result has been confirmed in the present study by comparing the blood sugar concentrations of two groups of similar flies, one flown for 3 hr. and the other rested for the same period. An aliquot of blood was then removed and assayed quantitatively as described. The results (Table 1) showed that flight reduced the blood trehalose concentration to about 1/5 of the control value while glucose concentration fell much less sharply, The only other blood carbohydrate fraction, listed as base line carbohydrate (RF = o), was not altered appreciably by flight. This fraction represents any carbohydrates that do not migrate in these particular solvent systems; while it is likely that glycogen or a similar polysaccharide constitutes the bulk of this fraction (Evans & Dethier, 1957), sugar phosphates might also be present.
The time course of wing-beat frequency during flight to exhaustion
Next, the relationship of blood trehalose to the rate of energy expenditure during flight was sought.
Fortunately, it has been shown that the wing-beat frequency (WBF) is a rather good index of the rate of oxygen consumption throughout flight to exhaustion in three species of Drosophila (Chadwick & Gilmour, 1940; Chadwick, 1947). Since flight energy is derived largely, if not solely, from aerobic pathways in the Diptera, the WBF is also a valid measure of the rate of energy expenditure.
We first measured the time course of WBF during flight to exhaustion. The flies were selected and mounted for flight as described. WBF measurements were made every 5 min. and the values then were expressed as a percentage of the initial rate to compensate for individual differences. Individuals having a flight duration of less than 2 hr. were discarded. This was done to standardize the population of flies used since it has been shown that flight duration is a function of the nutritional state of the fly (Williams et al. 1943). The results of these experiments are summarized in Fig. 1. Although individual flies showed some deviation the trend was the same, much like Chadwick (1939) found for Drosophila repleta, and Smyth (1952) for Phormia regina. For Phormia, WBF falls gradually and continuously during flight. Eventually, a WBF is reached below which flight does not occur and the animal may be termed ‘exhausted’. The average initial WBF was about 11,400 cyc./min. and the ‘exhausted* WBF about 7000 cyc./min.
Blood glucose and trehalose as a function of wing-beat frequency
Since substrate depletion has been implicated as a cause of flight exhaustion (Chadwick & Gilmour, 1940; Williams et al. 1943; Wigglesworth, 1949), experiments were carried out to determine what relationships, if any, existed between blood sugar concentrations and the WBF.
Flies were flown until the desired WBF was reached and an aliquot of the blood immediately removed and assayed quantitatively for glucose and trehalose. Determinations were made separately on each fly so that the results (Fig. 2) could be analysed statistically. Blood trehalose concentration fell with WBF over the entire range of WBF, while blood glucose did not change significantly. The variability in trehalose concentration was large but the downward trend is clear. The smaller variability seen in trehalose concentration as the point of exhaustion is reached is noteworthy. Clearly, WBF and the concentration of trehalose in the blood are correlated.
Wing-beat frequency as a function of blood glucose and trehalose administered by injection
Since others have shown that the administration of carbohydrate by feeding (Wigglesworth, 1949; Drosophila) or injection (Hudson, 1958; Phormia) restores flight in exhausted flies, there are strong indications that exhaustion results from an insufficient rate of supply of substrate to the flight muscles. Other less direct evidence supports this view (Chadwick, 1947; Williams et al. 1943). Accordingly, the relation between blood trehalose and the WBF might be more than a correlation: trehalose concentration might determine the WBF. A direct test of this hypothesis was performed.
Flies were flow to exhaustion and the WBF at which they stopped was designated as the exhausted WBF. Each fly was then injected serially with 52, 105 and 210 μg. of trehalose or glucose dissolved in saline. In all cases 0·524 μl. was the volume injected. After the first injection each fly was rested for a 3 min. equilibration period and then flown. The WBF was recorded every 10 sec. for the first minute of flight; then these values were averaged and designated as the WBF after injection. The flies were then exhausted before the next injection, and the process repeated. Following the last injection, each fly was fed 2M. glucose to repletion, rested for 30 min., and again flown. The WBF thus obtained was designated as the WBF after feeding. This value may be considered as the maximum for a given fly. A total of eight flies was used for the trehalose injections and five flies for the glucose injections.
If WBF were determined by the concentration of blood trehalose, then an increase in the concentration of this sugar in the blood of an exhausted fly should restore continuous flight. Moreover, the increase in trehalose concentration should be correlated with the increase in the WBF measured during subsequent flight. Table 2 summarizes the results of these experiments in which both of the predictions were confirmed. The largest amount of trehalose injected (210 μg.) restored the WBF to its maximum value. Since the blood volume of Phormia is about 7 μl. (Table 4), the injection of 210 μg. of trehalose should increase the concentration in blood to about 30 g./l., which is roughly the maximal concentration found in the blood of Phormia (Table 1, and Evans & Dethier, 1957). We, therefore, conclude that, under these conditions, blood trehalose concentration determines the WBF.
Glucose injections produced a similar effect in the few cases studied (Table 2), emphasizing the fact that this substance must not be overlooked as an important blood sugar (Evans & Dethier, 1957). However, at the concentrations of glucose and trehalose normally present in the blood (Table 1, Fig. 2), trehalose will be, quantitatively, the more important metabolite. That desiccation was not the limiting factor was indicated by the observations that exhausted flies either refused to drink water or did not resume flight if any was imbibed, as was also the case with Drosophila melanogaster (Wigglesworth, 1949). When the carrier saline was injected into exhausted flies, flight ability was never restored (eight cases). It should be pointed out that experiments of this type must be carried out with great care. Several flies were in some manner injured by the repeated injections, refusing to fly after the injection had been made. In some cases the fly responded to initial injections but not to subsequent ones. Such flies were fed on 2M. glucose, rested, and stimulated to fly. In no case was continuous flight observed, indicating that these were not exceptions to the pattern seen in Table 2, but rather were unable to fly due to some sort of mechanical trauma associated with the injections.
Wing-beat frequency of exhausted flies following the injection of trehalose
The curve shown in Fig. 3 represents data from ten flies which had been exhausted, injected with 210 μg. of trehalose, and then reflown. The WBF was recorded every minute until exhaustion was reached. Although the duration of flight was obviously much shorter, the general shape of this curve is very similar to the one constructed from ‘normal’ flies (Fig. 1), indicating that similar factors were operating in the two groups of flies. Hudson (1958) estimated that Phormia utilized carbohydrate at a rate of about 15 /μg./min./fly during flight. The introduction of 210 μg. of trehalose should, therefore, maintain flight for about 14 min. The average flight duration of this group of flies was just under 18 min., a figure that is in very good agreement with the calculated one.
Proposed relationship between blood trehalose and flight metabolism
On the basis of the foregoing observations it is proposed that the concentration of blood trehalose is a rate-limiting physiological factor determining the WBF and, therefore, the rate at which energy is expended by the flight muscles of Phormia regina during continuous flight. During prolonged flight blood trehalose continually falls, which obviously means that its utilization slightly exceeds its supply. As a consequence WBF declines, reflecting the declining blood trehalose concentration. Exhaustion would result when the concentration of this blood sugar reached some threshold concentration, below which trehalose could not be supplied to the flight muscles at a rate sufficient to permit continuous flight. This complements the studies of Wigglesworth (1949) and Williams et al. (1943), from which it was concluded that flight exhaustion was reached when glycogen could no longer be mobilized fast enough to meet the demands of active flight muscle. More will be said about this later. However, it must be pointed out now that the proposed relation between blood trehalose and the WBF need not necessarily apply during very brief flights when the fly is replete with carbohydrate. The role played by flight muscle glycogen (in terms of WBF regulation) under these conditions will be difficult to resolve since there appears to be no way to produce sudden alterations in flight muscle glycogen and then measure the effect on the WBF.
Flight muscle in vitro
Even before the discovery by Wyatt & Kalf (1956) that trehalose was an insect blood sugar, Sacktor (1955) had demonstrated that homogenates of housefly thoraces could metabolize this sugar at rates similar to those for glycogen and glucose. Thus, flight muscle possessed the metabolic pathways necessary for the oxidation of trehalose. However, since the structural integrity of the muscle was destroyed, Sacktor’s study does not necessarily show that trehalose can be removed from the environment by intact muscles. The mechanism of this uptake is of some interest since metazoan cells are, in general, not very permeable to disaccharides, at least not in terms of the rates involved here. A respiratory study of intact flight muscle in vitro was undertaken in hopes of gaining information on this problem.
Conventional Warburg studies were carried out on half-thorax preparations. The conditions and results of these studies are summarized in Table 3. The endogenous respiration of preparations from normally fed flies was very high and could not be elevated by addition of substrate to the flask. However, if the flies were flown to exhaustion before the preparations were made, the endogenous rate was decreased,. This rate could then be elevated to a level near that of muscle from fed flies by the addition of either 30 mM. glucose or trehalose. These results on intact muscle are essentially the same as those which Sacktor (1955) obtained using homogenates, and show that intact flight muscle can remove and metabolize exogenous trehalose. The R.Q. of preparations from normally fed flies was found to be 1-09 by the ‘indirect-method’ of Warburg (Umbreit, Burris & Stauffer, 1957), and respiration was linear with time for more than 1 hr. Sacktor (1955) suggested that glycogen stored in the flight muscle was the endogenous respiratory substrate since he could show a correlation between endogenous respiration and glycogen content of the thorax. If trehalose were also stored in the flight muscle it might contribute to this endogenous rate. We therefore analysed the flight muscle for trehalose. Muscle was dissected as quickly as possible from the thorax of well-fed flies, washed in saline, blotted dry, and frozen on dry ice until enough was collected for the determination (about twenty flies were used). The muscle was then weighed quickly, homogenized in 80 % methanol, and centrifuged. The supernatant was decanted and the pellet washed twice with methanol. The combined supernatant was taken to dryness under vacuum, dissolved in a small amount of water, and chromatographed in toto. After development of the chromatogram, the trehalose was eluted and assayed colorimetrically. Another batch of muscle, removed from similar flies, was used to determine the amount of ‘cell water’ per unit wet weight of muscle. The trehalose content amounted to only 0·25 μg./mg. wet weight muscle or 0·31 μg./ μl. cell water. Thus, it is unlikely that trehalose contributes appreciably to the endogenous respiration of these isolated muscles. These results also indicated that a steep concentration gradient exists for trehalose between the blood (10–30 g./l.) and the flight muscle (about 0–3 g./l.). The pellet from the methanol extraction was analysed for glycogen following the procedure of Hudson (1958). Glycogen was found at a concentration of about 28 μg./mg. wet weight muscle (as glucose); therefore, it is present in adequate amounts to subserve the endogenous respiration of these thorax halves. The reduction in the endogenous respiratory rate produced by flying the flies to exhaustion (Table 3) very likely reflects diminution of this thoracic glycogen.
Further studies were now undertaken to determine the rate at which trehalose was removed from the medium by the intact flight muscle. Preliminary results showed that when thorax halves were incubated with trehalose, large amounts of free glucose appeared in the medium. Controls incubated in carbohydrate-free saline showed no glucose accumulation. This effect was probably due to activity of the hydrolytic enzyme, trehalase, recently isolated from Phormia by Friedman (1960). We have interpreted our results as an indication that trehalose hydrolysis might be occurring at or near the muscle surface, and a detailed study of this is in progress. The hydrolysis of trehalose complicated attempts to measure the rate of trehalose uptake since several factors were always involved : conversion of trehalose to glucose, uptake of trehalose, and uptake of glucose. These considerations did not undermine the interpretation of the respiratory studies, because a lag in rate would be expected if trehalose first had to be hydrolysed, allowing glucose to reach an effective concentration in the medium, and none was ever observed. Although direct proof is lacking, it is likely that the flight muscles are permeable to trehalose. The only work dealing with trehalose penetration into intact insect tissue is that of Treherne (1960) which showed that this sugar penetrates the cells of the nervous system of Periplaneta americana. Permeation of trehalose into muscle is the subject of continuing research.
Blood volume
A knowledge of the blood volume was needed for several aspects of this study, but to our knowledge no such information is available in the literature for Phormia regina. Hudson (1958) used 6 μl. as an approximation, derived from information gathered by Beard (1953). We have used dilution of inulin-carboxyl-14C to estimate the blood volume of Phormia. This method has been used previously in insects by Levenbook (1958) and by Bricteux-Grégoire & Florkin (1959). A tracer amount of inulin-carboxyl-14C (3625 cts./min. ± 67) contained in 0·524 μl. of water was introduced into the blood by injection. After an appropriate incubation period an aliquot of the blood was removed and delivered directly on to a planchette, which was then dried and assayed for radioactivity. This result permits the calculation of blood volume. The results of these studies, given in Table 4, indicated a blood volume, or inulin space, of about 6–7 ·l. for Phormia under these conditions. This volume was not altered appreciably by a continuous flight of 30 min., in spite of the probable increase in water influx into the blood from the gut and perhaps other tissues during flight (cf. Hudson, 1958).
Analysis of excretory drops collected up to 1 hr. following the injection showed only traces of radioactivity ( < 1 % of the amount injected) indicating that the inulin was not being excreted during this time. When whole flies were homogenized in 5 % trichloroacetic acid 1 hr. after the injection of 14C-labelled inulin and the supernatant chromatographed on paper and assayed in toto for radioctivity by serial elution, 97 % of the injected dose was recovered from the inulin area of the chromatogram (RF= o). This observation makes it unlikely that inulin was being metabolized or degraded during this period.
Studies on blood glucose and trehalose
Origin of blood trehalose
As mentioned, Phormia utilizes carbohydrate at an estimated rate of about 15 μg./ min./fly during continuous flight (Hudson, 1958). Since the blood volume of Phormia is about 6–7 μ1., the maximal amount of trehalose in the blood under optimal conditions (when trehalose concentration is about 30 g./l.) would be roughly 210 μg. This amount could support continuous flight for about 14 min., while it has been shown that well-fed flies can fly continuously for about 3 hr. (Fig. 1). It is obvious that the amount of trehalose stored in the blood would contribute very little during long flights unless trehalose was continually being secreted into blood. The Origin of blood trehalose became, therefore, a question of considerable importance, particularly the rate at which this sugar could be delivered into the blood. The approach to this problem was aided by two sets of observations. Wigglesworth (1949), in a classical study, showed histochemically that substrate for the flight of Drosophila melanogaster came in large part from glycogen stored in the abdominal fat body. This glycogen must be mobilized into the blood in some form and transported to the flight muscles. Clements (1955) observed a similar situation using the mosquito Culex pipiens. Obviously, blood trehalose could be the transport form of fat body glycogen. A study of fat body was undertaken to determine whether or not this tissue was in fact the site of blood trehalose synthesis. During the course of this study a report by Candy & Kilby (1959) appeared, showing that the fat body of the locust was a site for trehalose biosynthesis, and they have recently described the pathway involved in this synthesis in some detail (1961).
Fat body
Intact fat body was removed from the abdomens of male flies, washed, and incubated in carbohydrate-free saline. Aliquots of the medium were then chromatographed at various times. Development and coloration of such chromatograms showed that large amounts of trehalose appeared in the medium. Experiments were now undertaken to measure the rate of this production. Fat body was prepared and incubated as described, and the medium was assayed quantitatively for trehalose immediately upon placing the tissue in the flask, and again after 1 hr. of incubation. The results of eight experiments showed that trehalose was liberated into the medium at a mean rate of 252 pg. (±18 S.E.)/mg. dry weight fat body/hr. (range, 184–341). The possibility that this trehalose resulted merely from lysing of the cells with the liberation of preformed trehalose stores was ruled out by the following observations. Freshly isolated fat body was analysed for trehalose content by the methods described for flight muscle. In four separate determinations trehalose was never found to be present at levels exceeding 26 pg./mg. dry weight of fat body, showing that the large amounts of trehalose produced into the medium by the fat body in vitro could not be accounted for on the basis of a preformed trehalose store. If these cells were disintegrating, then glycogen, stored in large amounts in this tissue, should also appear in the medium. Since glycogen does not migrate in the solvent systems used, it will be detected by elution of the application zone after the chromatogram is developed. When the base-line of such chromatograms was eluted and carbohydrate determinations carried out, only traces were detected ( < 2·8 μg./mg. dry weight fat body/hr.), showing that little if any cell breakage was occurring. It should be emphasized that the fat body of Phormia is a very delicate tissue and such incubations must be carried out with caution, particularly when transferring fat body from the wash to the incubation flask. If the ‘zero time’ reading showed the presence of carbohydrate, or if the medium became opaque during the course of the experiment, that experiment was discarded.
Thus, these results represent a valid measure of the rate at which fat body can synthesize trehalose from endogenous reserves and release it into the medium. Although the same rates need not necessarily occur in the intact animal, it is evident that the fat body would rapidly deplete its store of glycogen during flight if fat body glycogen was the sole precursor. Likewise, blood trehalose should fall to a low level shortly after the onset of flight, instead of decreasing gradually as we have observed. It has been shown that the gut delivers large amounts of dietary glucose into the blood of this fly during flight (Hudson, 1958). In spite of this, the concentration of blood glucose during flight remains fairly constant (Fig. 2), indicating that this sugar is being removed as quickly as it enters the blood. No doubt the flight muscles are utilizing some of this blood glucose. If the fat body could directly utilize blood glucose to synthesize blood trehalose, fat body glycogen might be spared to some extent without removing the supply of trehalose to the flight muscles.
To test this hypothesis we measured the ability of the intact fat body to synthesize trehalose directly from hexoses found in the environment. Mannose, fructose, glucose, and galactose (all uniformly labelled with 14C except galactose which was labelled in the i-position) were incubated separately with intact fat body (pooled from ten flies) for 30 min., all hexoses being present at a concentration of 200 mg. %. After the incubation period trehalose was isolated from the medium and its specific activity was determined. This result was expressed as a percentage of the specific activity of the hexose added to the medium. This figure, given in Table 5, can be taken as a measure of the ability of fat body to remove these four sugars from the medium, incorporate them into trehalose after the necessary modifications, and liberate this trehalose back into the medium. Clearly, fat body in vitro can convert exogenous glucose, mannose, and fructose into trehalose which is then liberated into the medium.
The relatively high specific activity of the trehalose argues against glycogen as a necessary intermediate in this synthesis. Galactose was utilized to a lesser extent, which could be due to the relative inability of the fat body to take up galactose, to convert it into glucose, or both. It is interesting to compare these results to the findings of Wigglesworth (1949) that mannose, fructose and glucose when fed to exhausted flies restored the ability to fly continuously, whereas galactose feeding produced only brief, intermittent flights. Also it might be added that experiments on the fat body of the woodroach, Leucophaea maderae, have yielded similar results with respect to trehalose synthesis from labelled sugars added to the medium.
If exogenous glucose was sparing the mobilization of fat body glycogen, then the rate of trehalose production should not be increased by addition of glucose to the medium. To test this we incubated intact fat body with 200 mg. % glucose and measured trehalose production. The results of five experiments showed an average trehalose production rate of 274 μg. (±31 S.E.)/mg. dry weight fat body/hr. (range, 197–310). This figure is not much different from the rate of trehalose production from endogenous reserves. Although a more complete analysis is needed, these results, and those shown in Table 5, indicate that the fat body can synthesize blood trehalose directly from blood glucose, and that this spares, to some extent, the mobilization of fat body glycogen for blood trehalose synthesis.
Next we measured the ability of other tissues to synthesize blood trehalose. The blood, gut, and flight muscles appeared to be possibilities.
Blood
To test whether trehalose synthesis could occur in the blood itself, 160 /xg. of glucose-U-14C having an activity of about 20,000 cts./min. was injected into the blood of three groups of six flies each. This amount will produce approximately a 20-fold increase in the concentration of blood glucose and should upset any equilibrium existing between trehalose and glucose towards trehalose synthesis. Aliquots of the blood were removed about 2 min. after the injection had been made and were pooled from each group for incubation. A known aliquot was delivered on to the base of a paper chromatogram after various incubation periods for determination of radioactivity in glucose and trehalose. The results, given in Table 6, showed no significant change in the distribution of activity between these two sugars, even after 60 min. of incubation, making net trehalose synthesis, or even hydrolysis, a very unlikely possibility. The data also showed that some of the injected glucose was converted to blood trehalose in the brief period before the blood was removed, a point to which we will return. The method used in this study permits visual examination of the blood as it is removed from the fly. This is necessary since free-floating fat body cells are occasionally present in the blood. The presence of these cells would obviously alter the results obtained.
Mid-gut
Thes tudies of Treherne (1958 a, b), using the desert locust, showed that dietary 14C-labelled glucose quickly appeared as trehalose in the blood, but no evidence was presented for the site of this conversion. The works of Candy & Kilby (1959,1961) indicated that the gut of the desert locust, a likely site for this conversion, was not active with respect to trehalose synthesis. Nevertheless, it was possible that the gut of Phormia was able to effect the synthesis of blood trehalose, the mid-gut being the most likely region. Accordingly, the mid-gut was removed from ten flies, cut into small pieces, washed, and then incubated in o-i ml. of saline containing 500 mg. % glucose-U-14C, having a specific activity of 160 cts./min./μg. Assay of the incubation medium by the usual methods showed essentially no activity in the trehalose area of the chromatogram even after 120 min. of incubation. In another experiment a starved fly was fed on a solution containing 1 % glucose-U-14C (specific activity, 210 cts./min./ μg.) marked with carmine. Immediately after feeding, the mid-gut was ligatured just below the proventriculus, and again at the junction of the mid-and hind-gut. This intact piece of mid-gut (coloured bright red) was then removed, washed well, and incubated in œi ml. saline for 1 hr. No leakage at the ligatures occurred since the medium remained clear of carmine. The medium was then chromatographed in toto. After development, the entire chromatogram was serially eluted in 2 cm. strips directly on to planchettes which were dried and assayed for radioactivity. The glucose area contained an activity of 34 cts./min. ( ± 6), with essentially all the activity being confined to this area. These results, although rather preliminary, indicate that the mid-gut does not contribute appreciably, if at all, to blood trehalose.
Flight muscle
The trehalase of flight muscle, whose presence was indicated by the in vitro studies, could theoretically synthesize trehalose. Although the kinetic studies on trehalase isolated from Phormia (Friedman, 1960) indicated that net synthesis was unlikely (Km= 6·7×10−4M) experiments were performed to test this possibility. Preparations of intact flight muscle were incubated for 1 hr. in 0·1 ml. saline, and the medium was assayed for trehalose. Small amounts of trehalose were always detected but never exceeded 2 5 μg./mg. dry weight muscle per hour in the eleven cases studied. Likewise, trehalose isolated from the medium in which muscle had been incubated with 500 mg. % glucose-U-14C (specific activity, 210 cts./min./ μg.) always showed low levels of radioactivity. Since small amounts of fat body are found in the thorax (Wigglesworth, 1949, and unpublished observations) this tissue rather than muscle might be responsible for the activity observed. In any case, it is apparent that the flight muscles, like the blood and mid-gut, do not play a significant role in the production of blood trehalose in Phormia. This view is substantiated by the finding of Turbert & Smith (1961) that the fat body of locusts was the only tissue of those analysed which showed high levels of activity with respect to glucoside synthesis, and by the work of Candy & Kilby (1959, 1961) showing that appreciable trehalose synthesis took place only in the fat body of the desert locust, Schistocerca gregaria.
Synthesis of blood trehalose in vivo
There being now good reason to believe that the fat body is the exclusive site of trehalose synthesis, and with the knowledge that this tissue in vitro can synthesize and liberate trehalose into the medium directly from exogenous glucose, it was possible to inject 14C-labelled glucose into the blood and to measure blood trehalose synthesis by the fat body in vivo.
Accordingly, a tracer amount of glucose-U-14C contained in 0·524 μl of saline and having an activity of 1349 cts./min. ( ± 35) was injected into the blood of each individual of five groups of flies which had been starved for 24 hr. Each group consisted of at least ten flies. An aliquot of the blood was removed from individuals of one group at 30 sec. after the injection, and these aliquots were pooled on the base of a paper chromatogram. The other four groups were treated in the same manner except that the blood was sampled at 2, 5, 10 and 20 min. after the injection. After development of the chromatograms, trehalose and glucose were eluted, and aliquots of the eluates were taken for chemical and radioactivity determinations. Table 7 summarizes the results of these experiments. The ‘zero time’ figure was calculated from the known injected dose and the estimated blood volume, and, therefore, is only an approximation. Radioactivity in trehalose was measurable within 30 sec., and increased as the radioactivity in glucose decreased. The percentage of radioactivity appearing in blood trehalose increased very sharply with time, reaching about 50 % in the short period of 2 min., and about 90% at 10 min. following the injection. Serial elution and assay of entire chromatograms showed that most of the activity was confined to glucose and trehalose ( > 96 % of the total recovered activity), but the area between trehalose and the base-line always contained significant levels of activity. These results, coupled with the results of the in vitro studies, demonstrated that blood glucose is indeed an immediate precursor of blood trehalose, that this conversion is very rapid, and that fat body is the primary site of blood trehalose synthesis.
In this connexion Winteringham (1958), using the housefly, found that injected glucose-U-14C rapidly appeared as trehalose in extracts of the thorax, and Treherne (1958b, 1960) has shown that glucose-U-14C injected into the blood of the roach Periplaneta americana and the locust Schistocerca gregaria was rapidly converted to blood trehalose. It appears, therefore, that this conversion by the fat body is a common feature of trehalose metabolism in insects, regardless of whether they use fat or carbohydrate as the main source of energy for flight.
Hydrolysis of blood trehalose in vivo
It has been mentioned that considerable amounts of free glucose appeared in the medium when intact flight muscle preparations were incubated with trehalose. The question arises as to whether this actually takes place in the intact animal, or is merely an in vitro artifact.
To answer this question we injected each of a number of flies with 3·5 μg. 14C-of labelled trehalose, having an activity of about 4574 cts./min. and contained in a volume of 0·524 μl. At 15 min. after the injection an aliquot of blood was removed from each fly, and the radioactivity in trehalose and glucose determined as usual. The results, given in Table 8, showed that appreciable amounts of radioactivity appeared in glucose, representing up to about 30 % of the total radioactivity found in the blood. We concluded that the hydrolysis of blood trehalose did occur. However, the site of this hydrolysis is not at all clear. Locust blood can hydrolyse trehalose (Howden & Kilby, 1956) and Friedman (1960) has reported that the blood of Phormia contains a trehalase, the activity of which was demonstrable only after the blood had been diluted in vitro (1:100). Our results tend to support the finding that undiluted blood in vitro is not active (Table 6), yet we have found that trehalose hydrolysis was occurring in the blood of the intact organism (Table 8). It is unlikely that the small volume injected (0·524 μl.) could dilute the blood enough to activate this trehalase, although such a possibility exists. Noteworthy in this regard is the apparent absence of trehalase activity in the blood of Leucophaea maderae (Zebe & McShan, 1959) and the silkworm, Bombyx mori (Saito, 1960). Our in vitro studies with intact flight muscle indicated that this tissue was responsible for the trehalose hydrolysis observed in the blood under in vivo conditions, perhaps by an enzyme found near, or on, the muscle surface. It would appear therefore that the blood and the flight muscles of Phormia contain a trehalase. The regulation and function of these two sites of trehalase activity are poorly understood at present.
Diagrammatic representation of carbohydrate transfer between the body compartments
Although based on experimental evidence, the diagram given in Fig. 4 is considerably simplified and even incomplete. Nevertheless, it should serve the function of integrating some of our results and, more important, indicating possibilities for future research. Before discussing this diagram it should be pointed out that a good deal is known concerning the intracellular metabolism of carbohydrate in various insect tissues, and those interested in a survey should consult reviews by Wyatt (1961), Sacktor (1961), Gilbert & Schneiderman (1961), Rockstein (1957), and Gilmour (1961). However, there are few studies of the transfer of carbohydrate between the body compartments, and the regulatory factors involved in these transfers. This has been one objective of the present study.
As shown in Fig. 4, dietary carbohydrates after hydrolysis in the gut are absorbed into the blood largely in the form of monosaccharides. These monosaccharides are removed from the blood by various tissues including the fat body. The fat body, and probably only the fat body, can utilize these sugars as well as endogenous reserves for the synthesis of blood trehalose. It will be recalled that this synthesis can occur at a very rapid rate. The factors regulating the partitioning of blood glucose between the synthesis of fat body glycogen and blood trehalose are not known, nor those involved in the mobilization of fat body glycogen for that matter. This channeling of post-absorptive and stored carbohydrates into a single transportable intermediate by the fat body is strikingly similar to the production of blood glucose by the mammalian liver, the difference being of course that insect fat body mobilizes trehalose instead of glucose into the blood. Obviously, the enzymic composition of the fat body will determine which monosaccharides can serve as precursors of blood trehalose. We know that mannose, glucose, fructose, and to some extent even galactose can be converted (Table 5) and there is evidence that the five-carbon sugar fucose cannot be utilized for trehalose synthesis (Evans & Dethier, 1957). It is interesting that fucose has no nutritional value for Phormia, while mannose, fructose, glucose and even galactose have considerable nutritional value (Hassett, Dethier & Gans, 1950). These results would predict that the enzymes necessary for the conversion of the above-mentioned hexoses to glucose, and thus to trehalose, are present in the fat body, while those for the conversion of fucose are not.
The subsequent fate of blood trehalose is poorly understood in biochemical terms, (but the respiratory study of flight muscle in vitro, as well as the wing-beat frequency studies, justify the indication in Fig. 4 that the flight muscle can metabolize this blood sugar to carbon dioxide and water with the concomitant liberation of energy for the flight process. Reserves stored in the flight muscle as well as sugars absorbed from the gut no doubt add to the energy output and we will consider this in more detail later. Blood trehalose can also be hydrolysed to free glucose by enzymes present in the flight muscle and perhaps, under certain conditions, in the blood itself. The question mark in Fig. 4 indicates that the site of trehalase activity might be on or near the muscle surface rather than intracellular. A cyclic process therefore exists in the blood which must result in glucose and trehalose being the normal blood sugars of this organism. In this connexion we have maintained flies from eclosion on daily injections of mannose into the blood, and examined the blood chromatographically on the sixth day, 2 hr. after the last injection of mannose had been made. Glucose and trehalose were the only carbohydrates present in detectable amounts.
All these processes are taking place concurrently. The levels of glucose and trehalose in the blood at any given time will therefore be determined by the integration of these simultaneous events. Metabolic control could operate on any of these levels, but the fat body is certainly a very likely spot in view of the regulatory function of blood trehalose. By far the biggest gap in our knowledge concerns the mechanism by which the flight muscle removes trehalose from the blood, and the initial steps in the oxidation of this sugar. It is avoiding the issue to state merely that trehalose is a reserve glucose pool, or that trehalose plays the same role in insects as does blood glucose in vertebrates.
The relative importance of blood sugars and tissue glycogen as energy sources for flight
Most studies concerned with the nature of the flight substrate in the Diptera have concluded that it was glycogen. Glycogen stored in the flight muscle certainly is utilized during flight, and the glycogen content in whole Drosophila and the thorax of Lucilia have been correlated with the wing-beat frequency (Williams et al. 1943). Glycogen stored in the fat body can be mobilized into the blood as trehalose which in turn can supply the active flight muscle, and in this sense fat body glycogen is a flight substrate. The study of Hudson (1958) shows, however, that the amount of carbohydrate depleted exceeds the amount of glycogen depleted during the continuous flight of Phormia regina, and the question arises as to the source and nature of this non-glycogen carbohydrate. Table 9 is based chiefly on calculations made from the data of Hudson (1958), and can be used to assess the relative importance of various sources of carbohydrate in Phormia. The breakdown of total glycogen in this table has been made on the basis of the relative amounts of glycogen found in the abdomens and thoraces of a large number of well-fed flies after removal of the gut. The average percentage of the total glycogen was found to be about 46 % for the thorax and about 54 % for the abdomen. Since the flies were all males, most of the abdominal glycogen is in the fat body. Thoracic glycogen is distributed between the flight muscle and thoracic fat body, but mostly in the former. The right-hand column of Table 9 is an estimate of the percentage of the total flight time which a given source could maintain, based on a utilization rate of 15 μg./min./fly. It can be seen that total body glycogen accounts for only about 20% of the total flight time, whereas crop sugar under these conditions would maintain about 69 % of the flight duration. Since crop and mid-gut sugar can serve as a flight substrate only after reaching the blood, and since fat body glycogen must be mobilized and transported as trehalose in the blood it can be estimated that about 90 % of the flight duration is maintained by the oxidation of blood sugars, the most abundant of which is trehalose. The quantitative importance of the crop, rarely mentioned in studies of insect flight metabolism, is rather obvious. In this regard there is reason to believe that the crops of blowflies in nature (Evans & Browne, 1960), and even of blood-sucking flies (Hocking, 1953) contain large amounts of concentrated sugar solutions. In fact, Hocking (1953) concluded that crop sugar was the most important source of flight energy for some of these insects. These observations indicate that correlations between glycogen and the flight process must be viewed with caution. For example, a fly which has been flown to exhaustion will fly vigorously for long periods almost immediately after given access to a concentrated sugar solution. In this case, flight energy is obviously derived from the oxidation of blood sugars. Clearly, glycogen is of no consequence here. On the other hand, a fly which is replete with respect to flight muscle glycogen must also have high concentrations of blood trehalose since dietary sugars are common precursors for the synthesis of glycogen and trehalose. As a result it is difficult to evaluate the relative contributions of glycogen and trehalose, under these conditions, during the initial stages of flight.
Although the values given in Table 9 are only estimations, they point out that flight muscle glycogen cannot be considered as the major flight substrate in this blowfly during continuous flight to exhaustion, and this might well be the case in many Diptera. Thus, it would appear that blood trehalose, and certain monosaccharides delivered into the blood from the gut, normally provide the main source of energy for the continuous flight to exhaustion of Phormia regina.
It should be emphasized that the conclusions derived from this study refer only to the Diptera since quite different processes may be operating in the Orthoptera (Krogh & Weis-Fogh, 1951; Weis-Fogh, 1952), Lepidoptera (Beall, 1948; Zebe, 1954), and Homoptera (Cockbain, 1961) where fat appears to be the chief energy source for flight.
ACKNOWLEDGEMENTS
We express our thanks to Drs L. E. Chadwick and C. Kearns for reading the manuscript. The research was aided by research grants from the National Science Foundation (G5927) and the Public Health Service (E2358). One of us (J.S. C.) received support from a Public Health Service predoctoral fellowship, and the Woodrow Wilson National Fellowship Foundation.