AZADIRACHTIN EFFECTS ON PROTEINS SYNTHESIZED IN FAT BODY, HAEMOLYMPH, OVARY AND MID-GUT OF LOCUSTS, SCHISTOCERCA GREGABIA.

: 3H-glycine was used as a tracer to examine the effect of azadirachtin on the incorporation of the amino acid into protein in various tissues, both in vivo and in uitro. In locusts, the tissues examined were fat body, ovary and mid-gut. The injected tracer rapidly moved from the haemolymph into many of the tissues. A large fraction of the tracer was present in the fat body as well as in the gut and ovary. It was found that the terpenoid eliminated the stimulatory effects of crude neurohormonal extracts from corpus cardiacum on protein synthesis in fat bodies and ovary. Azadirachtin appeared to have a direct inhibitory effect on protein biosynthesis in the mid-gut and this could be partly accounted for by secondary antifeedant effects.


INTRODUCTION
Azadirachtin, a tetranotriterpenoid, the major component in neem, Azadirachta indica, was first isolated in pure form by Butterworth and Morgan in 1971. The primary effect of azadirachtin on insects is its strong antifeedant activity. The desert locust, Schistocercagregaria, is particularly sensitive to azadirachtin. Many biological effects of azadirachtin, such as growth disruption, fecundity, fitness reduction and oviposition, are well documented, although its effects at the cellular level are still unknown.l,23. 4,5 The mode of action of azadirachtin was studied using radiolabelled dihydroazadirachtin which exerts the same biological effect as azadirachtin. It was reported that the compound is taken up apparently with great specificity into nearly all tissues examined. A large fraction of the radiolabelled compound was absorbed into the gut, Malpighian tubules and nervous tissues. The uptake of [22,23-3H,1 dihydroazadirachtin was inhibited by a large excess of injected dihydroazadirachtin, suggesting that the uptake was not due to diffu~ion.~ The aim of the present study was to examine the incorporation of radiolabelled amino acid into the proteins of a number of identified tissues to determine whether, in fact, azadirachtin can be shown to have a general effect on protein biosynthesis in the locust. The tissues examined for this purpose were fat body, haemolymph and ovary. These tissues are linked in the formation, transport and uptake ofvitellogenin and possibly other proteins. The effects of azadirachtin on the biosynthesis of proteins by the mid-gut was also examined, as any reduction of activity of proteolytic and other digestive enzymes could be expected to have a profound effect on the growth of the insects. The tracer used was tritiated glycine, as this had been previously found to be a suitable amino acid for this p~r p o s e .~

METHODS AND MATERIALS
Insects: Schistocerca gregaria was purchased from Blades Biologcal, Edenbridge, Kent, and maintained under laboratory conditions in metal cages at a temperature of 28-30 "C, 40% r.h. and a 12h light -12h dark cycle. The insects were fed on freshly washed spring cabbage leaves. Fresh tap water was also supplied. Cages were cleaned daily. Seven day old female locusts (after the final ecdysis) were used for all experiments. Each experiment was replicated four times.
Azadirachtir~ treatment: Azadirachtin was extracted and purified from Sri Lankan neem seeds by established method^.^,^,^ Confirmation of the purity of azadirachtin was carried out by melting point (150 -152 "C), TLC, reversed phase HPLC and NMR a n a l y~i s .~ Azadirachtin was dissolved in 10 % ethanol to give a final concentration of 1 mglml of azadirachtin. The experimental locusts were injected with a dose of 3 pg azadirachtinlg body weight and the control insects starved for 12 h before the experiment, received the same amount of 10% ethanollg body weight as the azadirachtin-treated locusts.
Application of 3H-glycine: 3H-glycine was obtained from Amersham International, Aylesbury, Bucks. The specific activity of the 3H-glycine was 21 Cilmmol. The radiolabelled amino acid was dissolved in insect saline to give a final concentration of 1 pCiJlO pl. The amino acid (2 pCi/g) was injected into each locust 12 h after injection of azadirachtin. Injections were made through the abdominal intersegmental membrane using a Hamilton 25 pl syringe.
Loss of radioactivity from the haemolymph: Loss of radioactivity from the haemolymph was measured over an 8 h period by taking 1p1 blood samples from the base of the hind coxa of the locust using 5p1 micropipettes a t different times after injections of the label and the radioactivity within each sample was measured by scintillation counting. Duplicate samples were taken for scintillation counting. Each experiment was replicated four times.
Incorporation of radioactivity into total proteins of whole insects: Initially, the rate of incorporation over 6 h was measured in both control and azadirachtin-treated locusts. The radioactivity in the whole locust was analyzed in the following manner. Individual locusts were anaesthetized in an atmosphere of CO, at different times after injection of the labelled amino acid and were then frozen in liquid N,, before grinding to a uniform powder in a mortar pre-cooled with liquid N,. The pooled samples were homogenized for 30 s in 0.5 rnl of 5 % trichloroacetic acid (TCA) using an ultrasonic microprobe. Samples were kept deep frozen in 5 % TCA until processing.
The homogenate was centrifuged for 5 min a t 10,000 g and the pellet washed three times in 5 % TCA. This process removed the entire radioactivity not incorporated into the proteins. The protein pellet was redissolved in 200 pl of 0.1M NaOH containing 0.1 %w/w SDS overnight a t 30 "C, and 100 pl of the sample was used to estimate the radioactivity by scintillation counting. The rest of the sample was used for protein estimation by the method of Lowry et al.,1° against a standard of bovine serum albumin, and the specific activity of the tissue was estimated with reference to the protein content.Each experiment was replicated four times.
Incorporation of 3H-glycine into specific tissues: The locusts were anaesthetized under CO, at different times after injection of the radioactive amino acid and the tissues were quickly dissected out under ice-cold insect saline. They were washed in fresh insect saline before being placed in 0.5ml of 5 % TCA. The samples were homogenized as described before. Tissues examined were the fat body, the haemolymph, the ovary and the mid-gut. Each experiment was replicated four times.

Effect of extracts from neuroendocrine tissue on the protein turnover
Preparation of corpus cardiacum extraction: Adult female locusts (7 days after final ecdysis) were anaesthetized under CO, and the heads removed. The corpus cardiacum was dissected out, free of the fat body, under saline, and placed in 100 p1 of insect saline. The gland was homogenized for 30 s using an ultrasonic microprobe.

Effect of extracts of corpus cardiacum on the rate of incorporation of 3H-glycine into the specific tissues of the locusts
The locusts were head-ligatured 12 h before injection of the 3H-glycine and only the experimental locusts were injected with azadirachtin along head-ligation. The incorporation levels of the tissues were measured 2 h after injection of the 3H-glycine in both control and azadirachtin treated locusts. The above experiment was extended as follows to observe the effect precisely. The crude extract of equivalent single corpus cardiacum was injected into each ligatured locust 3h after injection of 3H-glycine, and then the rate of incorporation was measured.
To determine whether there is any direct physical effect on the uptake of amino acids due to ligation, the clearance of 3H-glycine from the haemolymph was measured in both control and azadirachtin treated locusts.
The in vitro work was carried out with all the tissues examined in vivo separately, in an attempt to distinguish between the direct and indirect effects of azadirachtin.

In vitro experiments
Incorporation of 3H-glycine into specific tissues of the locust: The locusts were anaesthetized under CO, and dissected in insect saline. The fat body was removed from the locusts and freed from the ovary and tracheae. The sheets of the fat body were pre-incubated separately in lml of insect saline at 30 "C for 30 min, and then 1 mCi of 3H-glycine was added to the medium. Incorporation of the radiolabel into protein was measured after 1 , 2 and 4 h.
Subsequently, the experiments were done using ovary and mid-gut and the incorporation rate was measured after application of 3Hglycine into the medium.
The control experiment was carried out with the locust tissues which were pre-incubated with 12 yg of azadirachtin in the same way as in the medium was same as the in vivo experiment. The insect saline was maintained at a constant temperature of 30 "C and aerated throughout the experiment. Each experiment was repeated four times.
Effect of neurohormonal extract on incorporation of 3Hglycine into the fat body: Insect saline (lml) containing the extract of a single corpus cardiacum was used as the medium for incubation of tissues and then the rate of incorporation was measured as above. The experiments were carried out for both control and azadirachtin-containing media.
Effect of boiling on the corpus cardiacum extraction: In an attempt to determine the nature of the neurosecretory proteins, the extract of corpus cardiacum was boiled for 5 min and 10 min and this added to 1 ml of insect saline and the experiment conducted as above.

Loss of radioactivity from the haemolymph
The clearance of 3H-glycine from the haemolymph was measured over 8h. The percentage of radioactivity remaining in the haemolymph is shown in Figure 1. More than 60 % of the radioactivity was lost from the haemolymph within the first 30 rnin of injection and 8h of injection and only 30 % of the total radioactivity was detectable. In the azadirachtin-treated locusts, the clearance of labelled amino acid from the haemolymph showed the same rate as control locusts. The results ( Figure 1) show that the uptake of label is exponential, but there appears to be more than one mechanism of uptake.

Incorporation of radiolabelled amino acid into total protein in whole locusts
The rate of incorporation over 6h of 3H-glycine into the proteins of the 7 day old female adult locusts is shown in Figure 2. This preliminary experiment demonstrated a linear incorporation rate of radiolabelled amino acids into the locust tissue proteins during the first hour after injection, followed by the attainment of an equilibrium level. The incorporation rate was significantly greater in control locusts than in azadirachtin-treated locusts and it was found that 50 % inhibition of the level of incorporation was due to azadirachtin.

Incorporation of SH-glycine into specific tissues of the locusts
Following the initial experiment, the rate of incorporation of 3H-glycine into the fat body, the haemolymph, the ovary and the mid-gut were measured over 6h to investigate precisely the effect of azadirachtin on individual tissues. PA. Parunagarnu et al. 4 C o n t r o l (a) Incorporation of 3H-glycine i n t o the proteins of f a t body Figure 3 shows the time course of incorporation of 3H-glycine into the fat body of the 7 day old female adult locust. Both groups showed maximum incorporation level at l h after injection. In the azadirachtin-injected locusts, the incorporation of radiolabelled amino acid into the protein was not as efficient as in the control locusts. At the end of the l h period, the incorporation levels of azadirachtin-treated locusts was 43 % of that of controls. These values were significantly different from each other.

(b) Incorporation of 3H-glycine i n t o t h e haemolymph proteins
The incorporation of 3H-glycine into the haemolymph proteins was measured over 6h. The results of the time course of incorporation of 3H-glycine into the haemolymph proteins of the locust are shown in Figure 4. The level of incorporation was very slow compared to that of the fat body. Both control and azadirachtin-treated locusts showed a level of incorporation, which increased gradually with time. The incorporation level did not reach equilibrium l h after injection as in the fat body.
The control locust showed a higher level of incorporation than the azadirachtintreated locusts but these differences did not show any statistical significance. (c) Incorporation of 3H-glycine into the proteins of ovary Figure 5 shows the incorporation of 3H-glycine into the proteins of ovary of adult female locusts 7 days after final ecdysis. The level of incorporation of the ovary was higher than that of the fat body and the haemolymph. The incorporation rates of control locusts were faster than those of the azadirachtin-treated locusts. At the end of one hour period, the maximum incorporation level was reached and the incorporation level of the azadirachtin-injected locusts was 27 % that of the controls. Consequently, azadirachtin strongly inhibited the incorporation of amino acids into the ovary.

(d) Incorporation of 3H-glycine into the proteins of mid-gut
The level of incorporation into the mid-gut protein is the same as in the ovary. The results of the incorporation of 3H-glycine into the mid-gut is shown in Figure 6. The equilibrium level of incorporation was seen l h after injection of radiolabelled glycine. Azadirachtin inhibited the incorporation of 3H-glycine into the mid-gut proteins compared to the control locusts and at the end the incorporation level of azadirachtin-treated locusts was 40 % of that of the controls.  Figure 7a shows the effect of neurohormones and azadirachtin on the level of incorporation of 3H-glycine 2h after injection. The tissues taken for the analysis were the fat body, the haemolymph, the ovary and the mid-gut. Both the fat body and the ovary showed different response. In the preliminary experiment, the control locusts showed a higher level of incorporation than in the azadirachtin treated locusts as described above. In order to determine the neurohormonal effect on the level of incorporation, the locusts were ligatured 12 h before injection of radiolabelled glycine, subsequently, the effect of azadirachtin was also examined by injecting it into the haemolymph of the ligatured locusts (Figure 7b). In both groups, the levels of incorporation into the fat body, the haemolymph and the ovary were very similar to each other and the uptake of glycine into protein of these locusts was dramatically reduced to below 50 % of that of the control. The mid-gut showed high levels of incorporation in both ligatured locusts and azadirachtin treated locusts but this level was lower than in the control locusts.

Effect of extracts of corpus cardiacum and azadirachtin on the rate of incorporation of 3H-glycine into specific tissues
To establish the effect of an extract of corpus cardiacum on the uptake of glycine into the proteins more precisely, the extract of corpus cardiacum was injected into both control and azadirachtin treated ligatured locusts, 3h before injection of radiolabelled glycine (Figure 7c). The results show that the neurohormonal extract increased the uptake of glycine into proteins of the tissues in ligatured insects but the azadirachtin treated locusts did not show the stimulatory effect on the corpus cardiacum extract. This was true for all tissues examined.

Experiments in vitro
Incorporation of 3H-glycine into the protein of isolated tissues Table 1 shows the effects of corpus cardiacum extracts and azadirachtin both separately and together, on the incorporation of labelled glycine into proteins of the fat body, the ovary and the mid-gut. The effect of the neurohormonal extracts was to double the incorporation of the label in the fat body, the ovary and to increase it by 60 % in the gut. These results were statistically highly significant.

Tissue
Specific activity (dpm. mgl of protein)

Conditions
Fat body 8850f 721a 17550f1032b 6202f653a 7673f580a Ovary 8919f 943a 19096f565b 7546f718a 8431f467a (1) control; (2) pre-incubated with the extract of one corpus cardiacum for 30 min; (3) pre-incubated with both corpus cardiacum extract and azadirachtin for 30min; (4) pre-incubated with azadirachtin only for 30 min. The results are the means (+SD) of 4 locusts. The letters indicate statistical significance for each tissue, i.e, those means with the same letters do not differ significantly. Those with different letters are significantly different a t p<0.01.
The addition of azadirachtin into the incubates had the effect of eliminating completely the effect of the crude hormonal extracts. In the case of the fat body and the ovary, although azadirachtin reduced the incorporation to a level below that of the control, the reductions were not statistically significant (p>0.05). However, in the case of the gut, azadirachtin alone, as well as in the presence of the tissue extract, reduced the incorporation of label to an average of 50 % of that of the control value.

Heat lability of crude corpus cardiacum extract
The simple experiment to determine the heat lability of the extracts of corpus cardiacum showed that a 10 min boiling eliminated the stimulatory effect on protein metabolism, suggesting that the active components are heat inactivated ( Table 2).

The experiment
Specific activity (dpm. mgl of protein)

Control
Unboiled extract 5 min boiled extract 10 min boiled extract The results are the means of duplicate experiments which did not differ from each other by > 10%.

DISCUSSION
The experiments reported were performed in order to examine the effects of azadirachtin on protein synthesis in various tissues of the adult female locust by following the incorporation of amino acid into proteins. The age of the locusts was 7 days after the final ecdysis because these represent the peak of protein synthesis in maturing females. The amino acid, glycine was selected as appropriate for following protein synthesis in the different tissues based on previous work.7 Measurement of haemolymph levels of radiolabel shows that approximately 40 % of the radioactivity was lost from the haemolymph 30 minutes after the injection. The rate of uptake of the amino acid is exponential, but does not follow a simple pattern. At least two separate exponentials with different rate constants may be distinguished. Within the first 30 minutes, the uptake is very fast, suggesting that most of the radiolabel can be taken up into the widely distributed fat body where much protein metabolism takes place. These results are consistent with the results published by Mordue et al., for untreated locusts.11 When a physiologically effective dose of azadirachtin (3 pg azadirachtinlg body weight) was injected, it induced an inhibitory effect on the incorporation of the radiolabelled amino acid into the protein of the whole locusts suggesting a general effect.
In untreated locusts, the sequence of incorporation of labelled glycine into the proteins of the fat body, the haemolymph and ovary differs slightly from what was previously reported by Hi11, (1965).7 The previous authors found a clear progression of label first into the fat body, then haemolymph, and finally into the ovary, which was consistent with the idea that the source of the bulk of the ovarian proteins was the fat body. The results here suggest a much higher and earlier incorporation into the ovary, compared to the haemolymph, indicating that the ovary is more capable of synthesizing its own proteins than was earlier believed.
The profound effect of azadirachtin on reducing the incorporation of label into the ovarian protein is likely to be mainly due to a direct effect on that tissue. This would help to explain the fact that the clear-cut reduction of protein synthesis by azadirachtin shown by fat body and ovary is not apparent in haemolymph.
Although it is clear that the neuroendocrine system is involved in the control of protein turnover in the fat body of the female desert locust, the effect of the neuroendocrine system is complicated. Following elimination of neurohormone release by head ligation, the level of incorporation of both the azadirachtin-treated and the control locusts was significantly reduced, but only in the control head-ligatured female locusts the incorporation level could be restored to normal by injection of a n extract of corpus cardiacum 3 hours before injection of the radiolabelled amino acid. This suggests that azadirachtin has a specific effect on the action of corpus cardiacum extract. These results are comparable with the in vitro results which demonstrate that the tissues incubated with a corpus cardiacum extract showed significantly higher (p<0.05) incorporation rates in the control locusts but not for azadirachtin-treated locust tissues. It is probable that the low level of incorporation following cautery of the cerebral neurosecretory cells is the result of such a direct effect upon protein synthesis. Either cautery of the cerebral neurosecretory cells or allactectomy in the desert locust results in the inhibition of oocyte growth.12J3 In the immature female locusts with both these operations, the incorporation rate of 14C-glycine into the fat body protein does not increase, as it does in control animals, but slowly decreases. The decrease of incorporation is more rapid after neurosecretory cell cautery than after allactectomy, and the incorporation rate reaches a lower level.7 The direct effect of azadirachtin on the fat body, the ovary and the haemolymph is consistent with the prevention of stimulatory action of neurohormones on protein synthesis, but the mid-gut shows a different effect. The level of incorporation of amino acids into the gut protein of the head-ligatured control locusts is higher than the ligatured azadirachtin-treated locusts suggesting an additive effect. The incorporation level of both these groups is lower than in control insects. Azadirachtin apparently has a direct effect on protein synthesis of the mid-gut which is independent of neurohormones. The direct effect of azadirachtin upon gut muscle contraction both in vivo and in vitro has been recorded and it reduces the activity of most of the digestive enzymes which are secreted by the m i d -g~t . '~J~ The inhibition of gut contraction by azadirachtin has a marked effect on its passage through the gut. This results not only in lower faecal production but in a lower rate of absorption of food. Studies on mid-gut tissue show that the epithelial cells are greatly disrupted by the action of azadirachtin.14J6 There is also a suppression level of feeding, because feeding is not initiated in the locust until the foregut and hindgut are relatively empty.17J8 Although these are chemo-receptors on the mouthparts, an antifeedant effect is still produced by its effect on the passage of food through the gut.lS The effect of neurohormones on the mid-gut of other insects has also been studied. It is clear that the neurosecretory system can affect the synthesis of digestive enzymes of the rnid-g~t.~O In Manduca sexta, mid-gut wall and in Spodoptera litura, azadirachtin significantly affects the digestive enzymes such as protease, amylase and invertase. Therefore, the low level of incorporation of amino acid into the gut proteins in the azadirachtin-treated locusts could be due to the secondary antifeedant effect of azadirachtin. Thus a direct effect of azadirachtin on gut protein synthesis can account for some part of secondary antifeedant effect.
It becomes clear that although there is a very general effect of azadirachtin on protein biosynthesis, its mode of action varies from tissue to tissue, and is selective even within one tissue. Thus, while the effect on the gut appear to be general, the effect on the fat body and the ovary is only to negate the stirnulatory effects of neurohormonal extracts. It is clear from the lack of effect on the protein biosynthesis induced by juvenile hormone, and from the previously published data, that only some proteins are affected.21 In the light of these differences, it is difficult to propose a single common mechanism underlying an apparently general effect on protein synthesis.