Regulation of fat body glycogen phosphorylase activity during refeeding in Manduca sexta larvae

In 12-h-starved larvae of the tobacco hornworm, Manduca sexta, fat body glycogen phosphorylase was quickly inactivated when insects were refed with normal diet and agar which contained 3% sucrose. Only the first 2 min of refeeding were necessary to induce enzyme inactivation. During this short period, larvae did not ingest enough sucrose to increase the hemolymph glucose concentration. This may indicate that the gut released a hormone(s) which directly or indirectly led to the inactivation of fat body glycogen phosphorylase. Inactivation of the enzyme could also be induced by injection of glucose (30 mg) into the hemolymph of starving M. sexta larvae suggesting that there may be separate control from a neuroendocrine site such as the brain or the corpora cardiaca. Trehalose was less effective. Bovine insulin (2 and 4 μg/starved larva) did not induce phosphorylase inactivation over 20 min or decrease hemolymph carbohydrate or lipid concentrations within 60 min. It is, therefore, necessary to screen insect tissues for substances which could bring about inactivation of fat body glycogen phosphorylase. © 1992 Wiley-Liss, Inc.     

Allosteric effectors and trehalose protect larval Manduca sexta fat body glycogen phosphorylase B against thermal denaturation

In this paper we assessed the ability of modulators of the activity of glycogen phosphorylase b from the fat body of larval Manduca sexta to stabilize the enzyme against thermal denaturation. This approach has allowed us to distinguish between modulators that stabilize the enzyme, presumably through some conformational effect, from those that do not affect thermal stability. For example, 5'-AMP and 5'-IMP are both positive modulators of the enzyme and the K(m)s for AMP and IMP were similar, 0.71 and 1.09 mM, respectively. However, the V(max) for AMP (123 nmol/mg/min) was 10 times higher than the value found for IMP (12.5 nmol/mg/min) and AMP increased the thermal stability of glycogen phosphorylase b, however IMP did not increase the enzyme's thermal stability. Indeed, IMP decreased both the allosteric activation of the enzyme by AMP and the thermal protection conferred by AMP. The allosteric inhibitors ADP and ATP, which in vertebrate phosphorylase bind to the same site as AMP, both increased the thermal stability of the enzyme, however with less efficiency than AMP. Inorganic phosphate increased thermal stability, but glycogen and amylose did not. Glycerol, at 600 mM, protected the enzyme against thermal inactivation, whereas sorbitol at the same concentration did not show any effect. Among the polyols tested, trehalose was the most effective in conferring thermal stability. In fact, in the presence of 20 mM AMP and 600 mM trehalose, 90% of the enzyme activity remained after 20 min at 60 degrees C.     

Sequential structural changes in the fat body of the tobacco hornworm, Manduca sexta, during the fifth larval stadium

Light and electron microscopy revealed a series of structural changes that occur in the fat body of the tobacco hornworm, Manduca sexta, during the fifth, i.e. the final, larval stadium. At each developmental stage studied, the cells of the fat body were homogeneous in structure. We found no evidence suggesting the presence of more than one type of fat body cell. Our structural data are consistent with published observations on biochemical activities of M. sexta fat body at particular developmental stages. Specific points of agreement include: (a) acquisition of Golgi complex (GC) and rough endoplasmic reticulum (RER) concomitant with the time of major protein production; (b) loss of many cellular organelles (such as GC and RER) as protein production drastically decreases; (c) accumulation of protein granules and urate granules after the onset of wandering (i.e. during the pre-pupal period); (d) accumulation of lipid and glycogen throughout the feeding period. In addition we found that (a) the plasma membrane reticular system (PMRS) developed during the period when protein secretion was great; (b) the PMRS was lost abruptly at the onset of wandering; and (c) the nucleus changed in shape from being roughly spherical to elliptoid in the pre-pupal stage. We found that the structure of M. sexta fat body is similar to that published for other Lepidoptera. However, it differs from that of Heliothis zea in that regional differences are not obviously apparent.     

Developmental changes in the response of larval Manduca sexta fat body glycogen phosphorylase to starvation, stress and octopamine

Fasting or starvation of 1(st)- and 2(nd)-day fifth instar Manduca sexta larvae leads to rapid activation of fat body glycogen phosphorylase. Under feeding conditions, 21-29% of the phosphorylase was found in the active form. However, after only one hour of starvation, the active form increased to 55-65%. In larvae on the 3(rd)-day there was a slower increase in the activation, requiring three hours of starvation to reach a maximum of 60-65%. No activation was observed in 4(th)-day larvae after three hours of starvation. When 1(st)- or 2(nd)-day larvae were decapitated, the time-course of activation of glycogen phosphorylase was very similar to that observed in intact insects. However, activation of glycogen phosphorylase following decapitation was only observed in 1(st)- and 2(nd)-day larvae. In 2(nd)-day larvae, octopamine promoted activation of glycogen phosphorylase and 100-pmol of octopamine promoted maximum activation. Higher amounts of injected octopamine caused a decrease in activation. The injection of 100 pmol of octopamine caused a 50-55% activation of phosphorylase within 30 minutes. The simultaneous injection of the alpha-adrenergic receptor antagonist phentolamine with octopamine blocked the octopamine effect in 1(st)- and 2(nd)-day feeding larvae. However, the activation of glycogen phosphorylase observed in ligated/decapitated larvae on the 1(st)- and 2(nd)-day was not abolished by injection of phentolamine. All of these data suggest that factors other than adipokinetic hormone and octopamine may be involved in the activation of glycogen phosphorylase during fasting or starvation in the early part of the fifth larval stage of M. sexta.     

Phospholipase A2 activity in the fat body of the tobacco hornworm Manduca sexta

We report on phospholipase A₂ (PLA₂) activity in homogenates prepared from fat bodies of the tobacco hornworm Manduca sexta. PLA₂ activity is responsible for hydrolyzing fatty acids from the sn-2 position of phospholipids. The rate of hydrolysis increased with increasing homogenate protein concentration up to ～? 320 μg protein/ml reaction volume. Higher protein concentrations did not appreciably increase the rate of PLA₂ activity. As seen in some, but not all PLA₂s from mammalian sources, hydrolyzing activity increased linearly with time. The fat body activity was sensitive to pH (optimal activity at pH 8–9) and temperature (optimal activity at ～?40°C). The activity was associated with fat body rather than hemolymph, because no activity was detected in cell-free serum. The fat body PLA₂ activity differs from the majority of PLA₂s with respect to calcium requirements. Whereas most PLA₂s are calcium-independent. A few others are known to require submicromolar calcium concentrations. The fat body activity appears to be calcium independent. These data show that a PLA₂ activity that can hydrolyze arachidonic acid from the sn-2 position of phospholipids is associated with the tobacco hornworm fat body. The biological significance of this activity relates to biosynthesis of eicosanoids. Pharmacological inhibition of PLA₂ impairs the ability of this insect to respond to bacterial infections. Since the impairment can be reversed by treatment with exogenous arachidonic acid, the PLA₂ activity may be an important step in eicosanoid biosynthesis. © 1993 Wiley-Liss, Inc.     

Isolation and characterization of RNA polymerase B from the larval fat body of the tobacco hornworm, Manduca sexta.

DNA-dependent RNA polymerase B has been extensively purified from the larval fat body of the tobacco hornworm (Manduca sexta) by employing chromatography on ion-exchange columns of DEAE-Sephadex, DEAE-cellulose and phosphocellulose and centrifugation on glycerol gradients. The isolated enzyme after electrophoresis on acrylamide gels shows one main band and one minor band, both having enzyme activity sensitive to alpha-amanitin. The catalytic and physicochemical properties of the enzyme are similar to those of other eucaryotic B-type RNA polymerases. The enzyme has an apparent molecular weight of 530000, is inhibited 50% by alpha-amanitin at 0.04 microgram/ml and shows maximum activity on denatured DNA at 5 mM Mn2+ and 100 mM ammonium sulfate. An antibody was obtained that cross-reacts with the pure enzyme and forms a precipitin line. This antibody does not cross react with either Escherichia coli RNA polymerase or with wheat germ RNA polymerase but does react with one of the B polymerases isolated from wing tissue of the silkmoth, Antheraea pernyi.     

Organization of the larval pre-ecdysis motor pattern in the tobacco hornworm,Manduca sexta

The tobacco hornworm, Manduca sexta, undergoes several larval molts before transforming into a pupa and then an adult moth. Each molt culminates in ecdysis, when the old cuticle is shed. Prior to each larval ecdysis, the old cuticle is loosened by pre-ecdysis behavior, which consists of rhythmic compressions that are synchronous along the abdomen and on both body sides, and rhythmic retractions of the abdominal prolegs. Both pre-ecdysis and ecdysis behaviors are triggered by a peptide, eclosion hormone. The aim of the present study was to investigate the neural circuitry underlying larval preecdysis behavior. The pre-ecdysis motor pattern was recorded in isolated nerve cords from eclosion hormone-treated larvae, and the effects of connective transections and ionic manipulations were tested. Our results suggest that the larval pre-ecdysis compression motor pattern is coordinated and maintained by interneurons in the terminal abdominal ganglion that ascend the nerve cord without chemical synaptic relays; these interneurons make bilateral, probably monosynaptic, excitatory connections with identified pre-ecdysis motor neurons throughout the abdominal nerve cord. This model of the organization of the larval pre-ecdysis motor pattern should facilitate identification of the relevant interneurons, allowing future investigation of the neural basis of the developmental weakening of the pre-ecdysis motor pattern that accompanies the larval-pupal transformation.Abbreviations A3, A4... abdominal ganglia 3, 4... - AT terminal abdominal ganglion - ALE anterior lateral external muscle - DN dorsal nerve - DN_A anterior branch of the dorsal nerve - DN_L lateral branch of the dorsal nerve - DN_P posterior branch of the dorsal nerve - EH eclosion hormone - TP tergopleural muscle - VN ventral nerve - VN_A anterior branch of the ventral nerve - VN_L lateral branch of the ventral nerve - VN_P posterior branch of the ventral nerve     

Effects of plant flavonoids on Manduca sexta (tobacco hornworm) fifth larval instar midgut and fat body mitochondrial transhydrogenase

The reversible, membrane-associated transhydrogenase that catalyzes hydride-ion transfer between NADP(H) and NAD(H) was evaluated and compared to the corresponding NADH oxidase and succinate dehydrogenase activities in midgut and fat body mitochondria from fifth larval instar Manduca sexta. The developmentally significant NADPH-forming transhydrogenation occurs as a nonenergy- or energy-linked activity with energy for the latter derived from either electron transport-dependent NADH or succinate utilization, or ATP hydrolysis by Mg++-dependent ATPase. In general, the plant flavonoids examined (chyrsin, juglone, morine, quercetin, and myricetin) affected all reactions in a dose-dependent fashion. Differences in the responses to the flavonoids were apparent, with the most notable being inhibition of midgut, but stimulation of fat body transhydrogenase by morin, and myricetin as also noted for NADH oxidase and succinate dehydrogenase. Although quercetin inhibited or stimulated transhydrogenase activity depending on the origin of mitochondria, it was without effect on either midgut or fat body NADH oxidase or succinate dehydrogenase. Observed sonication-dependent increases in flavonoid inhibition may well reflect an alteration in membrane configuration, resulting in increased exposure of the enzyme systems to the flavonoids. The effects of flavonoids on the transhydrogenation, NADH oxidase, and succinate dehydrogenase reactions suggest that compounds of this nature may prove valuable in the control of insect populations by affecting these mitochondrial enzyme components.     

Ecdysteroid synthesis and molting by the tobacco hornworm, Manduca sexta, in the absence of prothoracic glands

When a pair of prothoracic glands (PGs) were removed from Manduca sexta pupae on the day of pupation, the hemolymph ecdysteroid titer remained at a low level. When a portion of the gland pair was extirpated from pupae after the critical period for prothoracicotropic hormone release, the maximum hemolymph ecdysteroid titer was reduced in proportion to the mass of the PGs removed. These findings clearly showed that the PGs in intact pupae are responsible for the elevated ecdysteroid titer required to elicit adult development on schedule. When brains were removed on the day of pupation, the initiation of adult development was delayed for weeks or months. In contrast, pupae whose PGs were removed on the day of pupation initiated development only 7 days late, indicating the existence of an additional source of pupal ecdysteroids. Further, abdomens of male M. sexta that were isolated on the day of pupation initiated adult development spontaneously within 70 days. The implantation of day 0 pupal brains into these isolated abdomens accelerated the initiation of adult development and elicited synchronous adult development. The hemolymph ecdysteroid titer of those isolated abdomens receiving implants of brains increased within 5 days and reached a maximum level of 1.5 micrograms/ml. The analysis of hemolymph ecdysteroids by reverse-phase HPLC revealed that ecdysone was the major moiety and that the ecdysteroid composition was similar to that of normal, intact pupae that had just initiated adult development. These results demonstrate that the PGs are not requisite for adult development. An increased hemolymph ecdysteroid titer was also observed in isolated abdomens from which the testes were removed and in abdomens devoid of their digestive tract. Indeed, in the latter case, the ecdysteroid titer attained much higher levels than those observed for abdomens with intact guts. Despite numerous attempts to identify the tissue(s) in the isolated abdomens responsible for the increase in ecdysteroid titer, its identity remains unknown.     

Synthesis of sn-1,2-diacylglycerols by monoacylglycerol acyltransferase from Manduca sexta fat body

The pathway for the synthesis of sn-1,2-diacylglycerol stimulated by the action of adipokinetic hormone (AKH) in the insect fat body is unknown. Previous results from this laboratory suggested that the hydrolysis of stored triacylglycerol to sn-2-monoacylglycerol followed by the stereospecific acylation of sn-2-monoacylglycerol catalyzed by a monoacylglycerol-acyltransferase (MGAT) could be the major route of AKH-stimulated sn-1,2-diacylglycerol synthesis. Thus, MGAT might represent a key enzyme of this pathway. In this study we characterized the MGAT activity from the Manduca sexta fat body. The activity, which was assayed by acylation of 2-monoolein using radioactive labeled palmitoyl-CoA, was found to be primarily a microsomal enzyme. The products of the acylation of 2-monoolein were 1,2-diacylglycerol (40–50%), 1,3-diacylglycerol (20–30%), and triacylglycerol (30–40%). The presence of triacylglycerol as a product revealed the presence of diacylglycerol-acyltransferase activity in the fat body microsomes. The pH optimum of MGAT activity was 7.0, and the dependence of the activity on the concentration of 2-monoolein showed saturation kinetics. An endogenous MGAT activity, which represented 20% of the maximal activity observed with added substrate, was detected. Optimal concentrations of palmitoyl-CoA ranged between 0.10–0.20 mM. The specific activity of MGAT, measured under optimal conditions, was about 0.6 nmol DG formed/min-mg protein. MGAT activity was greatest with 2-monoolein, and lower activity was observed when a saturated 2-monoacylglycerol was employed. The activity observed with sn-1-monoacylglycerol was lower than that observed with sn-2-monoacylglycerol. AKH did not stimulate MGAT activity, suggesting that either the enzyme is not under hormonal regulation or the monoacylglycerol pathway is not involved in the AKH-stimulated production of sn-1,2-diacylglycerol in the M. sexta fat body. © 1996 Wiley-Liss, Inc.     

Regulation of ecdysteroidogenesis in prothoracic glands of the tobacco hornworm Manduca sexta

Ecdysteroidogenesis in Manduca sexta prothoracic glands is regulated by a set of bioregulatory molecules, including prothoracicotropic hormone (PTTH) and a protein factor present in larval hemolymph, and by the competence of the glands to synthesize ecdysteroids in response to those molecules. A larval molting bioassay was used to assess the in vivo activity of Manduca PTTHs. Crude PTTH, big PTTH, and small PTTH each elicited a larval molt in head-ligated larvae. However, big PTTH was approximately 10-fold more potent than crude PTTH, which was, in turn, several orders of magnitude more potent than small PTTH. When big and small PTTH were combined, the molting response was similar to that elicited with crude PTTH. The chemical nature of the hemolymph protein factor was also investigated. Injection of [3H]cholesterol into last-instar larvae and fractionation of the radiolabeled hemolymph by gel filtration chromatography revealed three peaks of radioactivity. One peak eluted in fractions containing the hemolymph protein factor, a result consistent with the notion that the factor transports a sterol substrate. The possibility that the factor is a 3(2)-ketoreductase was investigated by assessing the effect of the factor on the accumulation of RIA-detectable ecdysteroids in prothoracic-gland-conditioned medium. Three of five preparations of the factor significantly enhanced the amount of RIA-detectable ecdysteroids in conditioned medium, indicating that at least some preparations of the factor may contain ketoreductase activity. The above findings are discussed in the context of current hypotheses of how bioregulatory molecules interact with the prothoracic glands to regulate ecdysteroidogenesis in Manduca.     

Regulation of translation during oogenesis and embryogenesis in the tobacco hornworm,Manduca sexta

Summary

Translational activity in the oocyte and early embryo of Manduca sexta may be regulated by a number of mechanisms including availability of ribosomal subunits, the protein complement of the maternal mRNP, and the presence of a functional 5′ cap structure on the maternal mRNA. We present preliminary experiments concerning the role of intracellular pH in the activation of protein synthesis and the nature of the cortical cytoskeleton of the mature oocyte and its possible role in immobilizing maternal mRNA.     

Microstructure of feeding by tobacco hornworm caterpillars, Manduca sexta

The microstructure of the feeding activity of tobacco hornworm caterpillars (Manduca sexta Johansson) on tomato leaf was examined by means of an automated cafeteria. In this device each activity of the caterpillar generates a characteristic slow electrical change which can be recorded. The apparatus is therefore both accurate and sensitive. Examination of the activity records indicated that larger animals ate more than smaller ones by increasing both bite frequency and the lengths of meals. Meal frequency did not increase. Correlations amongst a variety of measures indicated that there was regulation of feeding both between and within meals.     

Studies on vitellogenin from the tobacco hornworm,Manduca sexta

In the tobacco hornworm moth, Manduca sexta, vitellogenin (Vg) is a very high-density (1.29 g/ml) phosphoglycolipoprotein containing 13% lipids, 3% carbohydrates, and 0.6% protein-bound phosphorus. Vitellogenin (M_r～500,000) has two apoproteins designated apoVg-l (M_r 177,000 ± 3,600) and apoVg-ll (M_r45,000 ± 5,000). ApoVg-l and apoVg-II can be dissociated with 6 M guanidine HCI and separated from each other by gel permeation chromatography. Immunoblotting experiments using antibodies against the apoproteins showed that apoVg-l and apoVg-II antigens were immunologically distinct polypeptides. Antibodies against Vg reacted only with apoVg-l. Antibodies against Vg and apoVg-l reacted with Vg in double immunodiffusion experiments, whereas antibodies against apoVg-II did not. These results suggest that in the native Vg molecule, apoVg-II is positioned inside the molecule away from the aqueous environment. Only apoVg-I contained covalently bound carbohydrate as shown by fluorescein isothiocyanateconjugated concanavalin A, periodate-Schiff reagent, and in vivo labeling with ³H-Man. In vivo labeling with ³²P-inorganic phosphate and chemical determination showed that apoproteins of both Vg and vitellin contain covalently bound phosphate groups.     

Proteinases in molting fluid of the tobacco hornworm,Manduca sexta

Manduca sexta pharate pupal molting fluid contains more than 10 proteolytic enzymes that differ in relative mobility during electrophoresis in polyacrylamide gels containing sodium dodecyl sulfate and gelatin. The major gelatin digesting enzyme was an endoprotease with an apparent molecular weight of 100 kDa. Gel filtration on a Sephacryl S-300 column resolved another endoprotease of similar size that digests azocoll and [³H]casein. In addition we found an aminopeptidase-like enzyme (MW_app 500 kDa) and at least three carboxypeptidase-like enzymes (MW_app 10–60 kDa). Use of pseudosubstrates and inhibitors suggested the presence of both trypsin-like and chymotrypsin-like enzymes with the former activity approx. 10-fold greater than the latter. However, none of the proteolytic enzymes were substantially inhibited by diisopropylphosphorofluoridate or phenylmethylsulfonyl fluoride which are poteint inhibitors of trypsin and chymotrypsin. No carboxyl or sulfhydryl proteases were detected. The enzymes were most active in the neutral to alkaline pH range, but they were relatively unstable during storage which precluded their purification to homogeneity. Proteolysis of Manduca cuticular protein appears to involve a rather complex and unique mixture of endo- and exo-cleaving proteolytic enzymes.     

Developmental expression of three genes for larval cuticular proteins of the tobacco hornworm, Manduca sexta

Three cDNA clones coding for the 12.8, 13.3, and 14.6 kDa larval cuticular proteins of the tobacco hornworm, Manduca sexta, were isolated and characterized. Hybridization to abdominal epidermal RNA from different stages showed that the genes for the 12.8 and 13.3 kDa proteins were expressed only during larval life. By contrast, the gene for the 14.6 kDa protein was expressed throughout the segment during the feeding, growing larval stages, then only in the flexible intersegmental regions during the deposition of endocuticle in the pharate pupa and adult. Quantitative RNA dot blot hybridizations showed that the RNA for each protein disappeared during the larval molt when the ecdysteroid titer was high, then reappeared during the preecdysial deposition of endocuticle. All disappeared when the epidermis became pupally committed at the onset of wandering. Exposure of the fourth instar epidermis to 20-hydroxyecdysone (20HE) in vitro under conditions that lead to the formation of a new larval cuticle by 48 hr caused the disappearance of these RNAs by 18 hr. Exposure of Day 2 fifth instar epidermis to 20HE in vitro caused a depression of these RNAs which in the case of the RNAs coding for the 12.8 and 13.3 kDa proteins was partially prevented by simultaneous exposure to methoprene, a juvenile hormone (JH) mimic. By contrast, the RNA for the 14.6 kDa protein was suppressed by exposure to methoprene alone. Thus, each of these larval cuticular genes is turned off by high ecdysteroid; the presence or absence of JH determines whether or not this suppression is permanent in some or all cells.     

Degradation of aberrant proteins by larval tobacco hornworm,Manduca sexta L. (Sphingidae)

The tobacco hornworm Manduca sexta (Sphingidae) readily incorporates L-canavanine, the L-2-amino-4-(guanidinooxy)butyric acid structural analog of L-arginine, into newly synthesized proteins. As a result, the developing fifth-instar larva produces structurally aberrant canavanyl proteins that can exhibit severely impaired function. This situation is exacerbated by canavanine's ability to stimulate de novo protein synthesis. M. sexta larvae can respond to anomalous protein production by degrading canavanyl proteins nearly five times faster than normal proteins. The proteases of this insect can distinguish between normal and anomalous proteins and thereby avoid destruction of essential macromolecules. Aberrant protein degradative activity is not dependent upon de novo protein synthesis induced by canavanyl proteins. The fat body appears to be the source of proteases that degrade aberrant proteins; degradation is curtailed in the presence of sulfhydryl protease inhibitors as well as inhibitors of trypsin-like activity.     

Carbohydrate and lipid metabolism during the last larval moult of the tobacco hornworm, Manduca sexta

Abstract. At 25°C and with a light regime of 17 h light and 7h dark, the last larval moult of the tobacco hornworm, Manduca sexta , lasts approximately 32 h, during which profound changes of metabolism were observed. At the onset of the moult, which coincides with the cessation of feeding, the proportion of active fat body glycogen phosphorylase increased from 10 (-2h) to 25–30% (Oh). A biphasic pattern with peak activities of 45–50% after t – 12 h and again just prior to the shedding of the cuticle (32 h) was subsequently observed. Haemolymph trehalose concentration decreased significantly from c. 35 (Oh) to 20mM (8h), but then recovered to an intermediate level (30mM; 12h). After completion of the moult, the trehalose concentration was 35–40 mM. The haemolymph glucose level in feeding fourth instar larvae was 4–5 mM, but decreased sharply before the onset of the moult to c. 1 mM, followed by a slow 6-fold increase over the next 20h. Prior to the shedding of the cuticle, the glucose level dropped again dramatically. The haemolymph lipid level increased slowly from an initial level of 1.2–1.4mg/ml during the early part of the moult, reaching a maximum of 1.8mg/ml after /= 16 h. Afterwards, a decrease of c. 50% was observed until ecdysis occurred. Oxygen consumption per animal decreased steadily from 30–35 μl/min pre-moult by approximately 70% to c. 10 μl/min but started to increase about 5 h before the animals resumed feeding.     

The influence of larval diet on adult feeding behaviour in the tobacco hornworm moth, Manduca sexta

Lab-reared sphingid and noctuid moths appear to feed less than wild moths, and often are starved to enhance responsiveness in feeding assays. To measure the impact of larval nutrition on adult feeding, we raised a model sphingid species, Manduca sexta, on control or modified diets (reduced sugar, protein or water, supplemented beta-carotene) or cut tobacco leaves, then conducted feeding assays with artificial flowers. Behaviour was scored and analysed in a double-blind manner. Larval diet affected adult eclosion time, size and fat content, the latter of which was inversely proportional to moth approaches to the floral array in a flight cage. In contrast, behaviours refractory to feeding (sitting, escaping) were associated with sex and barometric pressure, but not with diet or fat content. Frequency of floral approaches and probing was not associated with any variable. However, moths reared on beta-carotene-supplemented diet were 2-3 times more likely to feed, and significantly less likely to sit or show "escape" behaviour than were moths from most other treatments. Our results suggest that decreased visual sensitivity, rather than increased fat content, accounts for reduced adult feeding by lab-reared M. sexta.