Journal of Insect Physiology 56 (2010) 893–901

Neural substrate and allatostatin-like innervation of the gut of Locusta migratoria

Lisa Robertson 1,*, Angela B. Lange

Department of Biology, University of Toronto Mississauga, 3359 Mississauga Road North, Mississauga, Ont., Canada L5L 1C6

A R T I C L E I N F O

Article history:

Received 19 March 2010

Received in revised form 30 April 2010

Accepted 3 May 2010

Keywords:

Immunohistochemistry

Allatostatin

Nervous system

Locust

Insect

A B S T R A C T

Allatostatin-like immunoreactivity (ALI) is widely distributed in processes and varicosities on the fore-,

mid-, and hindgut of the locust, and within midgut open-type endocrine-like cells. ALI is also observed in

cells and processes in all ganglia of the central nervous system (CNS) and the stomatogastric nervous

system (SNS). Ventral unpaired median neurons (VUMs) contained ALI within abdominal ganglia IV–VII.

Neurobiotin retrograde fills of the branches of the 11th sternal nerve that innervate the hindgut revealed

2–4 VUMs in abdominal ganglia IV–VIIth, which also contain ALI. The VIIIth abdominal ganglion

contained three ventral medial groups of neurons that filled with neurobiotin and contained ALI. The co-

localization of ALI in the identified neurons suggests that these cells are the source of ALI on the hindgut.

A retrograde fill of the nerves of the ingluvial ganglia that innervate the foregut revealed numerous

neurons within the frontal ganglion and an extensive neuropile in the hypocerebral ganglion, but there

seems to be no apparent co-localization of neurobiotin and ALI in these neurons, indicating the source of

ALI on the foregut comes via the brain, through the SNS.

� 2010 Elsevier Ltd. All rights reserved.

Contents lists available at ScienceDirect

Journal of Insect Physiology

journa l homepage: www.e lsev ier .com/ locate / j insphys

1. Introduction

The locust gut is divided into three regions, based not only onembryonic origin but also on physiological function. The foregut andhindgut originate from ectoderm and are lined with cuticle that isshed during ecdysis, while the midgut is derived from endoderm andis not cuticular in nature (Billingsley and Lehane, 1996; Chapman,1998). The foregut is involved with the ingestion, mechanicalbreakdown, storage and passage of food, while the midgut secretesdigestive enzymes, absorbs nutrients, and propels the remains to thehindgut which is primarily responsible for osmoregulation andexpulsion of faeces and urine (Chapman, 1998).

In insects, the central nervous system (CNS) and the enteric orstomatogastric nervous system (SNS) innervate the gut. Theanterior regions of the gut are innervated by the SNS (Penzlin,1985; Albrecht, 1953); a series of three peripheral gangliaassociated with visceral functioning and feeding (Hartenstein,1997; Ayali and Lange, 2010, this issue). The SNS includes thefrontal ganglion, which is connected to the tritocerebrum of thebrain by the paired frontal connectives. The recurrent nerveextends from the frontal ganglion to the hypocerebral ganglion,which is linked by two esophageal nerves to a pair of ingluvialganglia that are located bilaterally on the foregut wall and nervesfrom each ingluvial ganglion extend over the surface of the foregut

* Corresponding author. Tel.: +1 905 828 3898; fax: +1 905 828 3792.

E-mail address: lisa.clark@utoronto.ca (L. Robertson).1 Author’s previous work is published using her maiden name, Clark.

0022-1910/$ – see front matter � 2010 Elsevier Ltd. All rights reserved.

doi:10.1016/j.jinsphys.2010.05.003

(Albrecht, 1953; Konings et al., 1989; Hartenstein, 1997; Sternet al., 2007).

Recently, Braunig (2008) has shown through neurobiotinretrograde fills of the frontal connective of 4th instar Locusta

migratoria that neurons within the CNS project within the frontalconnectives toward the SNS, suggesting that the innervation of thegut is complex and that the tritocerebrum of the brain may act asan area of communication between the CNS and SNS. Early work byClarke and Grenville (1960) on the nervous control of locustforegut contractions suggested that each ganglion of the SNS haseffects on foregut contractions, but it is the ingluvial ganglion itselfthat controls foregut contractions, since contractions of the foregutare completely abolished upon severing the nerves arising from theingluvial ganglia. Lange and Chan (2008) also suggest that theingluvial ganglia are involved in foregut contraction, sinceremoving the ganglia ceased foregut contractions. Since then ithas been shown that the frontal ganglion plays a key role in feedingbehaviour, whereby removal of the frontal ganglion decreasesfeeding activity and prevents the crop from emptying its contents(Hill et al., 1966; Bignell, 1973). The locust frontal ganglioncontains a central pattern generator that controls foregut motorpatterns that control a portion of the foregut (Ayali and Lange,2010, this issue; Ayali et al., 2002; Zilberstein and Ayali, 2002). Arhythmic motor pattern is recorded from the nerves of the frontalganglion that coordinate with peristaltic movements of the foregutmuscles, and this rhythm increases as the foregut fills (Zilbersteinand Ayali, 2002; Ayali and Lange, 2010, this issue).

The CNS innervates the posterior regions of the locust gutthrough branches of the 11th sternal nerve that originates from the

L. Robertson, A.B. Lange / Journal of Insect Physiology 56 (2010) 893–901894

VIIIth abdominal ganglion (Donini et al., 2002). This neuralinnervation influences the contraction of the gut musculature.For example, Nagai and Brown (1969) reported that neuralstimulation of the longitudinal muscles of the rectum of thecockroach resulted in muscle contraction that was associated withthe discharge of rectal contents.

Adaptability to the environment and proper functioning of thegut is key to an insect’s survival. Not only is this plasticity andfunctionality of the gut controlled by nervous input, but it is alsomodulated by neuroactive chemicals such as biogenic amines andpeptides. The allatostatins (ASTs) are a family of peptides that havebeen shown to have effects on insect visceral muscle but wereoriginally isolated and purified from the cockroach, Diploptera

punctata (Stay et al., 1994; Pratt et al., 1989; Woodhead et al., 1989)based on their ability to inhibit the production and release of juvenilehormone from the corpora allata (Stay et al., 1994; Tobe and Stay,1985). Lange et al. (1993) reported that Dippu-AST7 (also calledallatostatin 1) inhibits myogenic and proctolin-induced contractionsof the hindgut of D. punctata. ASTs have also been shown to modulatecontractions of the foregut (Duve et al., 1995) and the midgut (Fuseet al., 1999). A number of peptides have been localized to midgutendocrine cells, suggesting that these peptides are involved indigestive processes (Lange and Orchard, 1998; Zitnan et al., 1993;Veenstra, 2009). For example, Dippu-AST7 stimulates carbohydraseactivityin the cockroachmidgutlumen(Fuse etal., 1999).Expressionof an allatostatin receptor gene (DAR-2) in the gut of Drosophila

melanogaster and identification of a putative AST receptor in thecockroach midgut suggests that the inhibitory effect of AST-likepeptides is physiologically-relevant and is mediated by a G protein-coupled receptor (Lenz et al., 2001; Bowser and Tobe, 2000).

The ASTs have been detected using immunohistochemistry andAST-like immunoreactivity is widely distributed in the CNS, withinaxons in nerves that innervate visceral muscle (including the gut),as well as within midgut endocrine cells of multiple insect species(Stay, 2000). In particular, AST-like immunoreactivity has beenshown to be associated with the midgut and hindgut of D. punctata

(Yu et al., 1995; Lange et al., 1993). The presence of AST-likeimmunoreactivity within nerves that innervate the viscera, as wellas within extensive neuropile regions in ganglia of the CNS,suggests a role for these peptides as neurotransmitters and/orneuromodulators both centrally and peripherally (Hoffman et al.,1999). A hormonal role for ASTs is also suggested by the presenceof these peptides within the hemolymph (Hoffman et al., 1999).

Understanding the neural control of digestion, the associationof peptides with the gut, and the role that these peptides play in gutfunctioning is important in understanding the physiology ofdigestion. The purpose of this study is to identify the neuronswithin the SNS and CNS that innervate the locust gut, and todetermine which of these identified neurons contain AST-likepeptides and thus may be the source of AST-like immunoreactivity(ALI) associated with the gut.

2. Materials and methods

2.1. Animals

Male and female adult L. migratoria, 2–3 weeks old, were usedfor all experimentation. Locusts were housed in a long-kept colonyat the University of Toronto Mississauga, Canada. The colony wasfed fresh wheat seedlings and bran and raised in crowdedconditions at 30 8C on a 12:12 light cycle.

2.2. Chemicals

Neurobiotin was purchased from Vector Laboratories Inc.(Burlingame, CA, USA). The monoclonal mouse anti-biotin cy3

antibody was purchased from Sigma (Oakville, Ont, Canada). Theallatostatin 1 (also called Dippu-AST 7 APSGAQRLYGFGL-NH2)antibody was a kind gift from Hans-Jurgen Agricola (Jena,Germany). The AST 1 peptide used in the preadsorption controlexperiments for the immunohistochemistry procedure wascustom synthesized by the Insect Biotech Canada Core Facility(Queen’s University, Kingston, Ont, Canada) or by ResearchGenetics (Huntsville, AL, USA). AST 1 was reconstituted in doubledistilled water to yield a stock solution of 10�3 M, which wasdivided into 10 mL aliquots and stored at �20 8C until needed.

2.3. Immunohistochemistry

The CNS, SNS, and gut were dissected in locust physiologicalsaline (150 mM NaCl, 10 mM KCl, 4 mM CaCl2, 2 mM MgCl, 4 mMNaHCO3, 5 mM HEPES, 90 mM sucrose, 5 mM trehalose, pH 7.2)and then fixed in 2% or 4% paraformaldehyde in Millonig’s buffer(0.14 M NaH2PO4�H2O, 0.1 M NaOH, 0.3 mM CaCl2�2H2O, pH 7.2)either for 1 h at room temperature or overnight at 4 8C. Afterfixation, preparations were washed with phosphate-bufferedsaline (PBS; 0.9% NaCl, pH 7.2) for 2–5 h and then incubated for1 h in PBS containing 4% Triton-X, 2% bovine serum albumin (BSA),and 10% normal sheep serum (NSS) at room temperature.Preparations were then washed with PBS and incubated for 2–4days at 4 8C in rabbit anti-AST 1 (Dippu-AST 7) IgG fraction purifiedpolyclonal antibody at a dilution of 1:1000 in PBS that contained0.4% Triton-X, 2% BSA and 2% NSS.

Preparations were then washed with PBS and incubatedovernight in affinity purified goat anti-rabbit antibody conjugatedto cy3 at a dilution of 1:600 in PBS containing 2% NSS. Forpreparations that were double-labeled with neurobiotin (seeprocedure below), affinity purified goat anti-rabbit antibodyconjugated to FITC (1:600) was used. Preparations were thenwashed with PBS, run through an ethanol series, and cleared with aglycerol series. Preparations were mounted in 100% glycerol andwere viewed using an epifluorescence microscope (Nikon Optiphot-2, Nikon Corporation, Tokyo, Japan) and drawings were made with acamera lucida attachment. Images were taken using a confocalmicroscope (Zeiss LSM 510, Carl Zeiss, Jena, Germany). For singlecy3-labeled preparations, images were taken using the 20�/1.0objective. A 543 nm laser line was utilized with a 560–615 nm bandpass filter. The oil immersion objective (63�/1.4) was utilized forhigh-magnification images of the gastric ceacal endocrine-like cells.For double-labeling, the laser lines and filter combinations utilizedwere488 nmwitha505–530 nmbandpass filterfortheFITC-labeledAST-like immunoreactivity and 543 nm laser with a 560–615 nmband pass filter for the cy3-labeled neurobiotin-filled preparations.Double-labeled images were taken using the 20�/1.0 objective.

In total, 31 CNS preparations were examined, as well as 10 SNSpreparations and 30 whole gut preparations. Controls wereperformed in which the AST 1 antiserum was pre-incubated with10�5 M synthetic AST 1 for 24 h. Fourteen pre-absorption controlexperiments were performed: 4 SNS preparations, 6 CNSpreparations, and 4 gut preparations. Pre-absorption of AST 1antibody with synthetic AST 1 abolished all staining within theSNS, CNS, and gut preparations.

A competitive ELISA was used to determine the specificity of theDippu-AST 7 (Dip-AST I) antibody used in this study (Vitzthumet al., 1996). Dip-AST I, II, III, IV, and B2 (Dippu-AST 7, 9, 8, 5, and 2,respectively) were recognized by the antibody, while the antibodywas two orders more sensitive to Dippu-AST 7 than the otherallatostatins tested. A non-competitive ELISA was used to detectcross-reactivity with peptides outside of the allatostatin family.The antibody does not cross-react with crustacean cardioactivepeptide, proctolin, corazonin, FMRFamide, locustatachykinin II,leucomyosuppressin, and perisulfakinin (Vitzthum et al., 1996).

L. Robertson, A.B. Lange / Journal of Insect Physiology 56 (2010) 893–901 895

2.4. Neurobiotin retrograde filling

Male and female locusts were dissected under physiologicalsaline. For the SNS, branches of the nerves extending from oneingluvial ganglion were cut as close to the foregut wall as possible.Branches of the 11th sternal nerve of the VIIIth abdominal ganglionwere cut as close to the hindgut as possible. These cut nerves wereplaced in a well made with petroleum jelly that contained distilledwater and secured in place by a petroleum jelly bridge. The rest ofthe nervous system was placed in an adjacent petroleum jelly wellfilled with physiological saline. Distilled water was placed in thewell containing the cut nerve endings for approximately 5 min toallow the nerve to swell allowing the uptake of neurobiotin tooccur more readily. The distilled water was then removed andreplaced with 5% neurobiotin tracer to immerse the cut nerveendings. Preparations were then left to incubate for 1–2 days at4 8C. During this incubation period the saline in the nervous systemwell was replaced at least once a day to ensure that degradation ofthe preparations was minimal. The dish containing the prepara-tions was covered and paper soaked with distilled water was usedto ensure that moisture was kept high in the dish and that thepreparations did not dehydrate.

After incubation in neurobiotin, the preparations were fixed in2% or 4% paraformaldehyde overnight at 4 8C. Preparations werethen washed with phosphate-buffered saline and incubated for 1 hin detergent (1% Triton-X in PBS). Preparations were then washedin PBS and incubated for 2 h in block (10% normal sheep serum inPBS). After this incubation the preparations were then incubated in1:600 cy3-conjugated monoclonal mouse anti-biotin antibody for1–2 days at 4 8C on a spin wheel wrapped in foil. Preparations werethen washed with PBS and dehydrated with ethanol (70% and 100%for 15 min each). A glycerol series was then performed to clear thepreparations and they were then mounted in 100% glycerol forviewing. In total, 37 preparations were examined: 29 CNS and 8SNS, where 22 of the CNS preparations and 4 of the SNSpreparations were double-labeled with neurobiotin and for ALI.

Preparations were viewed with an epifluorescence microscope(Nikon Optiphot-2, Nikon Corporation, Tokyo, Japan) and imageswere taken using a confocal microscope (Zeiss LSM 510, Carl Zeiss,Jena, Germany). Drawings of the preparations were completed onthe epifluorescence microscope using a camera lucida attachment.

3. Results

3.1. Alimentary canal

All three regions of the locust gut (foregut, midgut, and hindgut)contain AST-like immunoreactive axons and processes in differentpatterns (Fig. 1). AST-like immunoreactive axons can be tracedwithin the nerves of the pair of ingluvial ganglia located on theforegut (Fig. 1B) and can be seen to project to the foregut. Uponreaching the foregut the AST-like immunoreactive axons branch andgive rise to an irregular pattern of AST-like immunoreactiveprocesses and varicosities (Fig. 1B). AST-like immunoreactiveprocesses extend the length of the midgut and form an irregularlatticework pattern with associated varicosities (Fig. 1C). Interest-ingly, midgut endocrine-like cells were found to contain ALI (Fig. 1Cand C1). Hundreds of these endocrine-like cells are scatteredthroughout the midgut, with a higher density in the anterior midgut.These endocrine-like cells are teardrop shaped and have an apicalprocess that extends toward the lumen of the midgut (Fig. 1C1). Inaddition, the gastric caecae also contain AST-like immunoreactiveendocrine-like cells and processes (Fig. 1C2). The endocrine-likecells associated with the gastric caecae are smaller than those in themidgut and increase in abundance toward the tip of the caecae. Theprocesses within the gastric caecae are fine and increasingly branch

toward the tip of the caecae. The hindgut contains ALI withinprocesses and varicosities that arise from AST-like immunoreactiveaxons within the 11th sternal nerve from the VIIIth abdominalganglion. The rectum (posterior hindgut) contains a network ofprocesses and varicosities that contain ALI that extend to the colon(middle portion of hindgut), where six main longitudinal nervetracts containing AST-like immunoreactive processes arise andproject anteriorly into the ileum (anterior hindgut) to the pyloricsphincter, where the Malpighian tubules insert on the hindgut(Fig. 1D). The AST-like immunoreactive axons within theselongitudinal nerve tracts give rise to numerous fine lateral processesand varicosities that contain ALI. At the pyloric sphincter, the AST-like immunoreactive processes within the main nerve tracts branchextensively and these processes project anteriorly to give rise to thelatticework pattern of AST-like immunoreactive processes associat-ed with the midgut.

3.2. Stomatogastric nervous system

3.2.1. AST-like immunoreactivity

Fig. 2A is a schematic representation of the SNS and its locationrelative to the locust gut. Each ganglion of the SNS contains ALIwithin cell bodies, processes, and an extensive neuropile region(Fig. 2B1–B3). The frontal connectives contain 10–15 AST-likeimmunoreactive axons that pass through the neuropile of thefrontal ganglion (Fig. 2B1). The frontal ganglion also contains ALI in15–30 cell bodies of varying sizes (10–40 mm in diameter).Between 5 and 15 AST-like immunoreactive axons are present inthe recurrent nerve and ALI is seen within processes of theneuropile of the hypocerebral ganglion (Fig. 2B2). The hypocer-ebral ganglion contains 10–20 small AST-like immunoreactive cellbodies (15–25 mm in diameter). AST-like immunoreactive axonsproject within each of the esophageal nerves to the paired ingluvialganglia (10–20 axons per nerve, Fig. 2B2 and B3). An extensiveneuropile region is seen within the ingluvial ganglia, and 10–25axons containing ALI project to the foregut within each of the threeingluvial nerves (Fig. 2B3). Each ingluvial ganglion contains 20–25AST-like immunoreactive cell bodies (15–30 mm in diameter)located around the periphery of the ganglion.

3.2.2. Neurons within the SNS that innervate the gut

The nerves of one ingluvial ganglion were backfilled withneurobiotin in order to trace the source of neurons projecting tothe foregut. Due to the brightness of the neurobiotin staining, cellsor processes could not be visualized within the ingluvial ganglionbut it can be postulated that, at the very least, that neurobiotin-filled axons pass through the ingluvial ganglion to fill neuronal cellbodies and processes within the rest of the SNS. Approximately 10–20 axons were seen within each of the esophageal nerves thatextend between the ingluvial ganglia and the hypocerebralganglion (Fig. 2C2). Some of the axons passed directly throughthe hypocerebral ganglion while others have neuronal cell bodiessituated within the hypocerebral ganglion. The hypocerebralganglion contains a smaller number of neuronal cell bodies(approximately 15–30), which range in size from 20 to 25 mm indiameter (Fig. 2C2). Approximately 20 axons were revealed withinthe recurrent nerve (Fig. 2C1 and C2), which form the neuropilewithin the frontal ganglion or continued through the frontalganglion to the frontal connectives where 10–20 axons wererevealed (Fig. 2C1). The frontal ganglion contains 40–50 neuro-biotin-filled neuronal cell bodies that range in size from 25 to50 mm in diameter (Fig. 2C1). In two preparations the neurobiotintravelled within axons of the frontal connectives to produce a smallneuropile within each tritocerebral lobe. These axons extended tothe protocerebrum to fill a group of 6–10 medially located smallneuronal cell bodies (not shown).

Fig. 1. AST-like immunoreactivity associated with the locust gut. (A) Schematic representation of the three regions of the locust gut (foregut, midgut, hindgut). The location of

the gastric caecae denotes the border between the fore- and midgut, while the location of the Malpighian tubules delineate the midgut and hindgut. (B) Processes within the

foregut arise from the nerves from the ingluvial ganglia (open arrowhead). These processes branch extensively and end in varicosities (closed arrowhead). (C) Latticework of

processes and varicosites (closed arrowhead) within the midgut. Endocrine-like cells were also found to contain ALI (open arrowhead). (C1) Close-up of the midgut endocrine-

like cells showing an apical extension (open arrowhead). (C2) A gastric caecum showing ALI within endocrine-like cells (open arrowhead) and fine processes (closed

arrowhead). (D) Main longitudinal axonal tracts with lateral projections and varicosities (closed arrowhead) of the hindgut. Scale bars = 100 mm.

L. Robertson, A.B. Lange / Journal of Insect Physiology 56 (2010) 893–901896

3.2.3. Double-labeling of SNS neurons with ALI

None of the neuronal cell bodies in the SNS that filled withneurobiotin stain for ALI (Fig. 5C). This indicates that the ALI thatis present on the foregut arises from neurons with their cellbodies located in the CNS. In the frontal ganglion, the cell bodiesthat contain ALI are mostly located in the posterior portion of theganglion (Fig. 5A), while the neurons that were backfilled withneurobiotin are mostly located in the anterior portion of theganglion (Fig. 5B). Within the hypocerebral ganglion few cellsconsistently stained for ALI and these did not fill withneurobiotin.

3.3. Central nervous system

3.3.1. AST-like immunoreactivity

AST-like immunoreactivity was found in neurons of the brain(Clark et al., 2008) and in neurons within all ganglia of the ventralnerve cord. The cell bodies and processes that consistently stainedwith ALI in the VIIth and VIIIth abdominal ganglia in all of thepreparations were mapped using camera lucida (Fig. 3A). Any cells

that stained faintly or inconsistently were not mapped here. Thecells within the VIIth and VIIIth abdominal ganglia that stainpositively for ALI were in groups located bilaterally or medially(Figs. 3A and 4A, B). The majority of the cells that stain positivelyfor ALI within each ganglion are located ventrally (Fig. 3A, filled cellbodies). Both ganglia contain an extensive neuropile of processesthat stain for ALI (Figs. 3A and 4A, B).

Within the VIIth abdominal ganglion, two large ventral mediancell bodies stain positively for ALI (Fig. 4A). ALI was also presentwithin a bilateral group of 4 cells that have axons within the sternalnerve and fine AST-like immunoreactive axons are also seen withinthe tergal nerves (Fig. 4A). At least six AST-like immunoreactiveprocesses project to the VIIIth abdominal ganglion through theconnectives. The VIIIth abdominal ganglion contains ALI withintwo clusters of ventrally-located medial cell bodies (Figs. 3A and4B, groups a and b). Two bilaterally located groups of 5–7 cellbodies that stained positively for ALI are also seen within theganglion (Figs. 3 and 4B). Two to four AST-like immunoreactiveprocesses project to visceral targets within the 11th sternal nerveand the 10th sternal nerve, respectively (Fig. 4B).

Fig. 2. AST-like immunoreactivity (B) and neurobiotin retrograde fill (C) of the stomatogastric nervous system (SNS). (A) Diagram showing the location of the SNS relative to

the gut. (B) Immunoreactivity associated with the SNS. All ganglia are oriented so anterior is to the top and posterior toward bottom of each image. (B1) Immunoreactivity

within the frontal ganglion, showing processes within the frontal connectives (double arrow) and cell bodies (closed arrowhead). (B2) Immunoreactivity associated with the

hypocerebral ganglion, where AST-like immunoreactive processes are present within the recurrent nerve (single arrow) and the esophageal nerves (double arrow). (B3) The

ingluvial ganglion contains ALI in an extensive neuropile, axons in the esophageal nerve (single arrow), and the ingluvial nerves (closed arrowheads). ALI is also seen within

cell bodies (open arrowheads). (C) Retrograde neurobiotin fills of the SNS ganglia through the ingluvial nerves of the ingluvial ganglion. (C1) Filled axons are seen within the

frontal connectives (double arrow) and within the recurrent nerve (open arrowhead). Numerous cell bodies were filled with neurobiotin (closed arrowhead). (C2) Axons

within the recurrent nerve (open arrowhead), passing through the hypocerebral ganglion and into the paired esophageal nerves (double arrow) fill with neurobiotin, as well

as several neuronal cell bodies within the hypocerebral ganglion (closed arrowhead).

L. Robertson, A.B. Lange / Journal of Insect Physiology 56 (2010) 893–901 897

3.3.2. Neurons within the CNS that innervate the gut

After backfilling branches of one of the 11th sternal nerves (thenerve branches that innervate the hindgut) of the VIIIth abdominalganglion, neuronal cell bodies were filled within all of theabdominal ganglia. The neurobiotin did not travel anteriorlybeyond the abdominal ganglia to the thoracic ganglia. A compositecamera lucida drawing of these neuronal cell bodies and associatedaxons is shown in Fig. 3B, where the VIIth abdominal ganglion isdrawn as a representative of the neurons filled in abdominalganglia IV–VII. As shown in Figs. 3B and 4D, the VIIIth abdominalganglion contains three ventral medial clusters of neurons withcell body diameters of 40 mm or more (groups a, b and c, Fig. 4D)that filled with neurobiotin. In addition, four ventral neurons arelocated contralaterally to the nerve that was filled and the axonsproject medially toward the neuropile or to the lateral margin ofthe ganglion. Three dorsal neurons located ipsilateral to the fillednerve also filled with neurobiotin and axons from these cell bodiesextend toward the midline of the ganglion into the neuropile

region (Fig. 4D). At least six axons were visualized within the 11thsternal nerve that was filled (right 11th sternal nerve, Fig. 3B).Three axons leave the neuropile to project toward the hindgutwithin the 11th sternal nerve that was contralateral to the nervethat was filled with neurobiotin (left 11th sternal nerve, Fig. 4D).

The VIIth abdominal ganglion contains 2 (occasionally 4) largeventral unpaired median (VUM) neurons (cell body diameter atleast 40 mm), which recur in all anterior abdominal ganglia(Figs. 3B and 4C). At least two axons pass directly from the VIIIthabdominal ganglion, through the VIIth abdominal ganglion,ipsilateral to the nerve that was filled (Fig. 3B). Axons wererevealed within the anterior and posterior connectives contralat-eral to the nerve that was filled (Fig. 4C).

3.3.3. Double-labeling of CNS neurons with ALI

Some neurons within the CNS that fill with neurobiotin andtherefore innervate the hindgut also appear to contain AST-likepeptides (Figs. 4E and 5F, I). In the VIIth abdominal ganglion, the

Fig. 3. Composite camera lucida drawings of the VIIth and VIIIth abdominal ganglia with ALI (A) and neurobiotin fills (B). Black cell bodies are located ventrally and open cell

bodies are located dorsally. (B) Retrograde neurobiotin fill of the right 11th sternal nerve of the VIIIth abdominal ganglion (arrow). Black cell bodies are located on the ventral

surface and open cell bodies are located on the dorsal surface. Scale bars = 100 mm.

L. Robertson, A.B. Lange / Journal of Insect Physiology 56 (2010) 893–901898

VUM neurons that innervate the gut also contain ALI (Fig. 5F).Within the VIIIth abdominal ganglion, two groups of neuronswithin the ventral medial neuron groups that filled withneurobiotin co-localize with ALI (groups a and b, Fig. 4E and 5I).

4. Discussion

This study shows that ALI is associated with all regions of thelocust gut and that ALI is localized to cell bodies and processeswithin the ganglia of the SNS and CNS. Through neurobiotinretrograde fills of the nerves that innervate the foregut andhindgut, the neural substrate of the locust gut was determined.These neurons can release peptide onto the gut, so double-labelingof neurobiotin with ALI was performed.

The distribution of ASTs has been described in the CNS and SNSof several insect species. For instance, ALI is present in the brain,stomatogastric ganglia, and retrocerebral complex of the cock-roach, cricket, locust, termite, and most recently the honeybee(Clark et al., 2008; Kreissl et al., 2010; Maestro et al., 1998; Stayet al., 1992; Neuhauser et al., 1994; Yagi et al., 2005) and ALI ispresent in the frontal ganglion of Lacanobia oleracea (Duve et al.,2000). ALI does not seem to co-localize with neurobiotin inneuronal cell bodies within the SNS that innervate the foregut,indicating that the source of ALI on the foregut comes fromneurons with cell bodies in the brain. Candidate neurons may bethe small group of cells located within the protocerebrum that

were filled with neurobiotin (Braunig, 2008; this study). AST-likeimmunoreactive cell bodies are located within the protocerebrallobes of the locust brain (Clark et al., 2008), in a similar location tothose identified in the neurobiotin backfills in the present study.This suggests that the neuronal cell bodies in the protocerebrummay contain ALI and thus be the source of ALI extending over theforegut. The SNS would then act as a conduit for the immunoreac-tive axons from the brain innervating the foregut. Thus, the AST-like immunoreactive neurons that are within the SNS do notappear to project to the foregut and must be considered essentiallyinterneurons. Perhaps these integrate and coordinate patterngeneration for the SNS.

In L. migratoria and Schistocerca gregaria, Dippu-AST 2-likeimmunoreactivity is associated with the thoracic ganglia but notwith the VIIth and VIIIth abdominal ganglia (Veelaert et al., 1995).Skiebe et al. (2006) used an antibody against Dippu-AST 7 and,unlike Veelaert et al. (1995), found ALI within cells, processes, andneuropile regions within all of the abdominal ganglia, including theVIIth and VIIIth. The location of cells within these ganglia isconsistent with the cells described in this study. Of interest are theVUMs that recur within all of the abdominal ganglia. These VUMsalso contain ALI, indicating these cells may be the source of ALI onthe hindgut. Cells of a similar size and location to the VUMsidentified here have been found to have axons in the oviducalnerve of the VIIth abdominal ganglion (Kalogianni and Pfluger,1992). Some of these efferent VUMs, as well as dorsal unpaired

Fig. 4. Co-localization of CNS neurons that fill with neurobiotin and also stain for ALI. (A) ALI associated with the VIIth abdominal ganglion in ventral unpaired median neurons

(open arrowhead) and bilaterally occurring clusters of cells that send processes within the sternal nerves (closed arrowhead). Processes project to the VIIIth abdominal

ganglion through the posterior connectives (double arrow). (B) The VIIIth abdominal ganglion showing ALI within ventral unpaired medial cell bodies (groups a and b) and

bilateral groups of cell bodies (closed arrowhead). (C) Ventral unpaired median neurons (open arrowhead) filled with neurobiotin after retrograde filling of the 11th sternal

nerve of the VIIIth abdominal ganglion. Axons were seen within the contralateral connectives to the nerve that was filled (closed arrowhead). (D) Neurons identified within

the VIIIth abdominal ganglion occur in three ventral medial groups of neurons (groups a, b and c) and within neurons that are located contralateral to the nerve that was filled

(closed arrowhead). The right 11th sternal nerve was filled (double headed arrow) and axons are seen within the contralateral sternal nerve (single arrow). (E) Ventral

neurons identified within the VIIth and VIIIth abdominal ganglia that filled with neurobiotin also contain AST-like immunoreactivity (black cells). Scale bars = 100 mm.

L. Robertson, A.B. Lange / Journal of Insect Physiology 56 (2010) 893–901 899

median neurons within the abdominal ganglia, are octopaminergic(Stevenson et al., 1994; Lange and Orchard, 1986), suggesting thatAST-like peptides within this class of cells may be released as a co-transmitter with octopamine onto the oviducts, but also onto thehindgut. This is interesting since both octopamine (Huddart andOldfield, 1982) and Dippu-AST (Lange et al., 1993, 1995) have beenshown to be inhibitory on hindgut muscle contractions. These cellsmay also be involved in activities that require repetition withinsuccessive segments, such as peristalsis of the gut.

In addition to the ALI associated with the SNS and CNS, allregions of the locust gut exhibit ALI, suggesting that AST-likepeptides are involved in gut functioning. AST-like immunoreac-tivity has been found to be associated with the gut of other insects,including the cockroach (Maestro et al., 1998), the blood-suckingbug Rhodnius prolixus (Sarkar et al., 2003), and the earwig (Rankinet al., 1998). ASTs have been found to inhibit myogenic andproctolin-induced contractions of the gut (Lange et al., 1993; Duveet al., 1995; Fuse et al., 1999). Thus, by opposing the effect ofmyostimulatory peptides, AST can help regulate gut motility andthe movement of the food bolus through the gut, increasing theefficiency of digestion.

Based on the ALI contained within endocrine-like cells, AST-like peptides may serve other physiological functions. Endocrine-like cells are distributed throughout the midgut epithelium,indicating that the gut is an endocrine organ in its own right(Zitnan et al., 1993). In general there are two types of endocrinecells, open and closed, and it is the open-type that were found tocontain ALI within this study. The open-type cells are in directcontact with the midgut lumen by a narrow extension of the apicalend of the cell (Endo and Nishiitsutsuji-Uwo, 1981; Fujita andKobayashi, 1977). Since these cells have contact with the gut

lumen, they may serve as an interface between the digestive andendocrine systems, by assessing nutrient content (Fujita andKobayashi, 1977). The existence of ALI within midgut endocrine-like cells suggests that these peptides have multiple physiologicaleffects relating to digestion. Not only do these peptides affect gutcontractility, but also regulate enzyme secretion (Fuse et al.,1999). Feeding state alters the peptide content of these endocrine-like cells. For example, FMRFamide-like and tachykinin-likeimmunoreactive content of midgut endocrine-like cells changeswith feeding state in L. migratoria (Lange, 2001). This decrease incontent may be due to release, decreased synthesis, or increasedturnover of the peptide (Lange, 2001). AST-like peptides arereleased into the hemolymph from the cockroach midgut (Yuet al., 1995), indicating that the ASTs are acting as endocrinehormones during feeding.

Whereas the SNS of crustaceans is a well studied model forneural functioning and pattern generation, the insect SNS hasreceived far less attention (Stein, 2009; Marder and Bucher, 2007;Bohm et al., 2001). Most research on the SNS of insects is in the fieldof developmental biology, where it is used as a model forembryological cell migration, due to the resemblance of the insectSNS with the autonomic nervous system of vertebrates (Harten-stein, 1997; Copenhaver, 2007). Connections are made betweenthe CNS and SNS within the protocerebrum and tritocerebrum ofthe brain, where neurons and associated axons originate andproject through the SNS. Braunig (2008) recently completed aretrograde fill of one frontal connective and determined thatapproximately 250 neurons within the CNS project within thefrontal connective toward the SNS, with 70 of these neuronslocated within the brain; more than had been found previously(Aubele and Klemm, 1977).

Fig. 5. Double-labels of ALI and neurobiotin-filled neurons and processes in the SNS and CNS ganglia. (A–C) Frontal ganglion. (A) ALI within the frontal ganglion. (B)

Neurobiotin-filled neuronal cell bodies. (C) Merge showing the lack of co-localization. (D–F) VIIth abdominal ganglion. (D) ALI within VUM cell bodies (hatched circle). (E)

VUM neuronal cell bodies containing neurobiotin (hatched circle). (F) Merge showing co-localization within the VUM neurons (hatched circle). Note that the ALI in this

preparation was weak and therefore only one VUM neuron showed double-label staining. (G–I) Posterior portion of the VIIIth abdominal ganglion. (G) ALI within posterior

VUM cell bodies. (H) VUM neuronal cell bodies stained with neurobiotin. (I) Merge showing co-localization within VUM neurons (hatched circle). Scale bars = 100 mm.

L. Robertson, A.B. Lange / Journal of Insect Physiology 56 (2010) 893–901900

The present study extends these findings by completing aretrograde neurobiotin fill of the SNS to determine the neuralsubstrate of the foregut. In two preparations, the neurobiotin wascapable of filling neurons within the brain that are in a similarlocation to a group of neurons within the protocerebrum filled byBraunig (2008). Braunig (2008) also filled neurons within thesubesophageal ganglion and thoracic ganglia that projected withinthe SNS and that may ultimately innervate the locust foregut. Thisis quite interesting since a central pattern generator that controlsmotor rhythms of the foregut exists within the frontal ganglion,while the SOG is responsible for coordinating the mouthparts, andthe metathoracic ganglion is responsible for the ventilatory motorpattern (Rand et al., 2008; Ayali, 2004; Ayali et al., 2002; Zilbersteinand Ayali, 2002; Braunig, 2008; Rast and Braunig, 2001; Ayali andLange, 2010, this issue).

The central pattern generator controlling foregut movementsituated within the frontal ganglion can be influenced not only bythe ventilatory rhythm that resides within the metathoracicganglion, but also by neuromodulators like ASTs (Braunig, 2008;Zilberstein et al., 2004; Ayali and Lange, 2010, this issue). It has alsorecently been shown that the frontal ganglion motor patternintegrates with the motor pattern from the SOG in the absence of

sensory input to coordinate food passage (Rand et al., 2008). Thus,the neurons filled with neurobiotin within this study and Braunig(2008) may work together to coordinate the aforementionedmotor patterns for feeding and gut motility. Since ALI is associatedwith all regions of the gut and within all ganglia of the SNS, thispeptide family may be involved in neuromodulation of the motorpatterns within the relevant ganglia and in the modulation of thegut contraction itself.

Acknowledgements

The Natural Sciences and Engineering Research Council ofCanada supported this work. We are grateful for the gift of theantiserum from Dr. Hans-Jurgen Agricola. We would also like tothank Zach McLaughlin for the drawings of the locust stomato-gastric nervous system and gut.

References

Albrecht, F.O., 1953. The Anatomy of the Migratory Locust. Athlon Press, London.Aubele, E., Klemm, N., 1977. Origin, destination and mapping of tritocerebral

neurons of the locust. Cell and Tissue Research 178, 199–219.

L. Robertson, A.B. Lange / Journal of Insect Physiology 56 (2010) 893–901 901

Ayali, A., 2004. The insect frontal ganglion and stomatogastric pattern-generatornetworks. Neurosignals 13, 20–36.

Ayali, A., Lange, A.B., 2010. Rhythmic behaviour and pattern-generating circuits inthe locust: key concepts and recent updates. Journal of Insect Physiology 56,834–843.

Ayali, A., Zilberstein, Y., Cohen, N., 2002. The locust frontal ganglion: a centralpattern generator network controlling foregut rhythmic motor patterns. Journalof Experimental Biology 205, 2825–2832.

Bignell, D.E., 1973. The effect of removal of the frontal ganglion on growth andprotein synthesis in young adults of Locusta migratoria. Canadian Journal ofZoology 52, 203–208.

Billingsley, P.F., Lehane, M.J., 1996. Structure and ultrastructure of the insectmidgut. In: Lehane, M.J., Billingsley, P.F. (Eds.), Biology of the Insect Midgut.Chapman & Hall, London, pp. 3–30.

Bohm, H., Dybek, E., Heinzel, H.-G., 2001. Anatomy and in vivo activity of neuronsconnecting the crustacean stomatogastric nervous system to the brain. Journalof Comparative Physiology A 187, 393–403.

Bowser, P.R.F., Tobe, S.S., 2000. Partial characterization of a putative allatostatinreceptor in the midgut of the cockroach Diploptera punctata. General andComparative Endocrinology 119, 1–10.

Braunig, P., 2008. Neuronal connections between central and enteric nervoussystem in the locust, Locusta migratoria. Cell and Tissue Research 333, 159–168.

Chapman, R.F., 1998. Alimentary canal, digestion and absorption. The Insects:Structure and Function, vol. 4. Cambridge University Press, Cambridge, pp.38–69.

Clark, L., Lange, A.B., Zhang, J.R., Tobe, S.S., 2008. The roles of Dippu-allatostatin inthe modulation of hormone release in Locusta migratoria. Journal of InsectPhysiology 54, 949–958.

Clarke, K.U., Grenville, H., 1960. Nervous control of movements in the foregut ofSchistocerca gregaria Forsk. Nature 186, 98–99.

Copenhaver, P.F., 2007. How to innervate a simple gut: familiar themes and uniqueaspects in the formation of the insect enteric nervous system. DevelopmentalDynamics 236, 1841–1864.

Donini, A., Ngo, C., Lange, A.B., 2002. Evidence for crustacean cardioactive peptide-like innervation of the gut in Locusta migratoria. Peptides 23, 1915–1923.

Duve, H., Audsley, N., Weaver, R.J., Thorpe, A., 2000. Triple co-localisation of twotypes of allatostatin and an allatotropin in the frontal ganglion of the lepidop-teran Lacanobia oleracea (Noctuidae): innervation and action on the foregut. Celland Tissue Research 300, 153–163.

Duve, H., Wren, P., Thorpe, A., 1995. Innervation of the foregut of the cockroachLeucophaea maderae and inhibition of spontaneous contractile activity bycallatostatin neuropeptides. Physiological Entomology 20, 33–44.

Endo, Y., Nishiitsutsuji-Uwo, J., 1981. Gut endocrine cells in insects: the ultrastruc-ture of the gut endocrine cells of the lepidopterous species. Biomedical Research2, 270–280.

Fujita, T., Kobayashi, S., 1977. Structure and function of gut endocrine cells.International Review of Cytology Supplement 6, 187–233.

Fuse, M., Zang, J.R., Partridge, E., Nachman, R.J., Orchard, I., Bendena, W.G., Tobe, S.S.,1999. Effects of an allatostatin and a myosuppressin on midgut carbohydrateenzyme activity in the cockroach Diploptera punctata. Peptides 20, 1285–1293.

Hartenstein, V., 1997. Development of the insect stomatogastric nervous system.Trends in Neuroscience 20, 421–427.

Hill, L., Mordue, W., Highnam, K.C., 1966. The endocrine system, frontal ganglion,and feeding during maturation in the female desert locust. Journal of InsectPhysiology 12, 1197–1208.

Hoffman, K.H., Meyering-Vos, M., Lorenz, M.W., 1999. Allatostatins and allatotro-pins: is the regulation of corpora allata activity their primary functions?European Journal of Entomology 96, 255–266.

Huddart, H., Oldfield, A.C., 1982. Spontaneous activity of foregut and hindgutvisceral muscle of the locust, Locusta migratoria. II. The effect of biogenicamines. Comparative Biochemistry and Physiology C 73, 303–311.

Kalogianni, E., Pfluger, H.J., 1992. The identification of motor and unpaired medianneurons innervating the locust oviduct. Journal of Experimental Biology 168,177–198.

Konings, P.N.M., Vullings, H.G.B., Kok, O.J.M., Diederen, J.H.B., Jansen, W.F., 1989. Theinnervation of the corpus cardiacum of Locusta migratoria: a neuroanatomicalstudy with the use of Lucifer yellow. Cell and Tissue Research 258, 301–308.

Kreissl, S., Strasser, C., Galizia, C.G., 2010. Allatostatin immunoreactivity in thehoneybee brain. Journal of Comparative Neurology 518, 1391–1417.

Lange, A.B., 2001. Feeding state influences the content of FMRFamide and tachy-kinin related peptides in endocrine-like cells of the midgut of Locusta migra-toria. Peptides 22, 229–234.

Lange, A.B., Bendena, W.G., Tobe, S.S., 1995. The effect of thirteen Dip-allatostatinson myogenic and induced contractions of the cockroach (Diploptera punctata)hindgut. Journal of Insect Physiology 41, 581–588.

Lange, A.B., Chan, K., 2008. Dopaminergic control of foregut contractions in Locustamigratoria. Journal of Insect Physiology 54, 222–230.

Lange, A.B., Chan, K.K., Stay, B., 1993. Effect of allatostatin and proctolin on antennalpulsatile organ and hindgut muscle in the cockroach, Diploptera punctata.Archives of Insect Biochemistry and Physiology 24, 79–92.

Lange, A.B., Orchard, I., 1998. The effects of SchistoFLRFamide on contractions oflocust midgut. Peptides 19, 459–467.

Lange, A.B., Orchard, I., 1986. Identified octopaminergic neurons modulate con-tractions of locust visceral muscle via adenosine 30 ,50-monophosphate (cyclicAMP). Brain Research 363, 340–349.

Lenz, C., Williamson, M., Hansen, G.N., Grimmelikhuijzen, C.J.P., 2001. Identificationof four Drosophila allatostatins as the cognate ligands for the Drosophila orphanreceptor DAR-2. Biochemical and Biophysical Research Communications 286,1117–1122.

Maestro, J.L., Belles, X., Piulachs, M.D., Thorpe, A., Duve, H., 1998. Localization ofallatostatin-immunoreactive material in the central nervous system, stomato-gastric nervous system, and gut of the cockroach Blattella germanica. Archives ofInsect Biochemistry and Physiology 37, 269–282.

Marder, E., Bucher, D., 2007. Understanding circuit dynamics using the stomato-gastric nervous system of lobsters and crabs. Annual Review of Physiology 69,291–316.

Nagai, T., Brown, B.E., 1969. Insect visceral muscle. Electrical potentials and con-traction in fibres of the cockroach proctodeum. Journal of Insect Physiology 15,2151–2167.

Neuhauser, T., Sorge, D., Stay, B., Hobbman, K.H., 1994. Responsiveness of the adultcricket (Gryllus bimaculatus and Acheta domesticus) retrocerebral complex toallatostatin-1 from a cockroach, Diploptera punctata. Journal of ComparativePhysiology B 164, 23–31.

Penzlin, H., 1985. Stomatogastric nervous system. In: Kerkut, G.A., Gilbert, LI..(Eds.), Comprehensive Insect Physiology, Biochemistry and Pharmacology,vol. 5. Pergamon Press, Oxford, pp. 371–406.

Pratt, G.E., Farnsworth, D.E., Siegel, N.R., Fok, K.F., Feyereisen, R., 1989. Identificationof an allatostatin from adult Diploptera punctata. Biochemical and BiophysicalResearch Communications 163, 1243–1247.

Rand, D., Gueijman, A., Zilberstein, Y., Ayali, A., 2008. Interactions of suboesophagealganglion and frontal ganglion motor patterns in the locust. Journal of InsectPhysiology 54, 854–860.

Rankin, S.M., Stay, B., Chan, K., Jackson, E.S., 1998. Cockroach allatostatin-immuno-reactive neurons and effects of cockroach allatostatin in earwigs. Archives ofInsect Biochemistry and Physiology 38, 155–165.

Rast, G.F., Braunig, P., 2001. Insect mouthpart motor patterns: central circuitsmodified for highly derived appendages? Neuroscience 108, 167–176.

Sarkar, N.R.S., Tobe, S.S., Orchard, I., 2003. The distribution and effects of Dippu-allatostatin-like peptides in the blood-feeding bug, Rhodnius prolixus. Peptides24, 1553–1562.

Skiebe, P., Biserova, N.M., Vedenina, V., Borner, J., Pfluger, H.-J., 2006. Allatostatin-like immunoreactivity in the abdomen of the locust Schistocerca gregaria. Celland Tissue Research 325, 163–174.

Stay, B., 2000. A review of the role of neurosecretion in the control of juvenilehormone synthesis: a tribute to Berta Scharrer. Insect Biochemistry and Mo-lecular Biology 30, 653–662.

Stay, B., Chan, K., Woodhead, A., 1992. Allatostatin-immunoreactive neurons pro-jecting to the corpora allata of adult Diploptera punctata. Cell and TissueResearch 270, 15–23.

Stay, B., Tobe, S.S., Bendena, W.G., 1994. Allatostatins: identification, primarystructures, functions and distribution. Advances in Insect Physiology 25,267–337.

Stein, W., 2009. Modulation of stomatogastric rhythms. Journal of ComparativePhysiology A 195, 989–1009.

Stern, M., Knipp, S., Bicker, G., 2007. Embryonic differentiation of serotonin-contain-ing neurons in the enteric nervous system of the locust (Locusta migratoria).Journal of Comparative Neurology 501, 38–51.

Stevenson, P.A., Pfluger, H.-J., Eckert, M., Rapus, J., 1994. Octopamine-like immuno-reactive neurons in locust genital abdominal ganglia. Cell and Tissue Research275, 299–308.

Tobe, S.S., Stay, B., 1985. Structure and regulation of the corpus allatum. Advances inInsect Physiology 18, 305–432.

Veelaert, D., Schoofs, L., Tobe, S.S., Yu, C.G., Vullings, H.G., Couillaud, F., DeLoof, A.,1995. Immunological evidence for an allatostatin-like neuropeptide in thecentral nervous system of Schistocerca gregaria, Locusta migratoria and Neobel-lieria bullata. Cell and Tissue Research 279, 601–611.

Veenstra, J.A., 2009. Peptidergic paracrine and endocrine cells in the midgut of thefruit fly maggot. Cell and Tissue Research 336, 309–323.

Vitzthum, H., Homberg, U., Agricola, H., 1996. Distribution of Dip-Allatostatin I-likeimmunoreactivity in the brain of the locust Schistocerca gregaria with detailedanalysis of immunostaining in the central complex. The Journal of ComparativeNeurology 369, 419–437.

Woodhead, A.P., Stay, B., Seidel, L., Khan, M.A., Tobe, S.S., 1989. Primary structure offour allatostatins: neuropeptide inhibitors of juvenile hormone synthesis.Proceedings of the National Academy of Sciences of the United States of America86, 5997–6001.

Yagi, K.J., Kwok, R., Chan, K.K., Setter, R.R., Myles, T.G., Tobe, S.S., Stay, B., 2005. Phe-Gly-Leu-amide allatostatin in the termite Reticulitermes flavipes: content inbrain and corpus allatum and effect on juvenile hormone synthesis. Journal ofInsect Physiology 51, 357–365.

Yu, C.G., Stay, B., Ding, Q., Bendena, W.G., Tobe, S.S., 1995. Immunocytochemicalidentification and expression of allatostatins in the gut of Diploptera punctata.Journal of Insect Physiology 41, 1035–1043.

Zilberstein, Y., Ayali, A., 2002. The role of the frontal ganglion in locust feeding andmoulting related behaviours. Journal of Experimental Biology 205, 2833–2841.

Zilberstein, Y., Fuchs, E., Hershtik, L., Ayali, A., 2004. Neuromodulation for behaviourin the locust frontal ganglion. Journal of Comparative Physiology A 190, 301–309.

Zitnan, D., Sauman, I., Sehnal, F., 1993. Peptidergic innervation and endocrine cellsof insect midgut. Archives of Insect Biochemistry and Physiology 22, 113–132.