- Email: [email protected]
Myoinhibitory neuropeptides in the American cockroach☆
Peptides 22 (2001) 199 –208
Myoinhibitory neuropeptides in the American cockroach夞 Reinhard Predel*, Ju¨rgen Rapus, Manfred Eckert Institut fu¨r Allgemeine Zoologie und Tierphysiologie, Friedrich-Schiller-Universita¨t, Erbertstrae 1, D-07743 Jena, Germany Received 15 January 2000; accepted 20 September 2000
Abstract A large number of myostimulatory neuropeptides from neurohaemal organs of the American cockroach have been described since 1989. These peptides, isolated from the retrocerebral complex and abdominal perisympathetic organs, are thought to be released as hormones. To study the coordinated action of these neuropeptides in the regulation of visceral muscle activity, it might be necessary to include myoinhibitors as well, however, not a single myoinhibitory neuropeptide of the American cockroach has been described so far. To fill this gap, we describe the isolation of LMS (leucomyosuppressin) and Pea-MIP (myoinhibitory peptide) from neurohaemal organs of the American cockroach. LMS was very effective in inhibiting phasic activity of all visceral muscles tested. It was found in the corpora cardiaca of different species of cockroaches, as well as in related insect groups, including mantids and termites. Pea-MIP which is strongly accumulated in the corpora cardiaca was not detected with a muscle bioassay system but when searching for tryptophane-containing peptides using a diode-array detector. This peptide caused only a moderate inhibition in visceral muscle assays. The distribution of Pea-MIP in neurohaemal organs and cells supplying these organs with Pea-MIP immunoreactive material, is described. Additionally to LMS and Pea-MIP, a member of the allatostatin peptide family, known to exhibit inhibitory properties in other insects, was tested in visceral muscle assays. This allatostatin was highly effective in inhibiting spontaneous activity of the foregut, but not of other tested visceral muscles of the American cockroach. © 2001 Elsevier Science Inc. All rights reserved. Keywords: Periplaneta americana; Insect; Neuropeptide; Leucomyosuppressin; Allatostatin; Immunohistochemistry; Photodiode-array detector
1. Introduction The neuropeptide pattern in neurohaemal organs of the American cockroach, Periplaneta americana, was intensively investigated in the last few years. Hereby, we concentrated on myotropic neuropeptides of the retrocerebral complex and abdominal perisympathetic organs which are the major neurohaemal release sites of the brain/subesophageal ganglion and ventral nerve cord, respectively. Two general conclusions can be drawn from these studies. Firstly, members of all myostimulatory neuropeptide families which were initially isolated from head extracts of Leucophaea maderae [12] are also present in considerable amounts in the corpora cardiaca/allata complex (CC/CA) of the American cockroach. This fact supports the idea that these peptides will be released into the circulation. Sec夞 Taken from a paper presented at the Winter Neuropeptide Conference 2000, Invertebrate Division, Hua Hin, Thailand, January 10 –15, 2000. * Corresponding author. Tel.: ⫹49-03641-9-49125; fax: ⫹49-036419-49102. E-mail address: [email protected] (R. Predel).
ondly, the abdominal perisympathetic organs (PSOs) were found to contain a neuropeptide pattern totally different from that of the retrocerebral complex. Altogether, not less than five different neuropeptide families with myostimulatory properties were identified from neurohaemal organs of the American cockroach, namely kinins, sulfakinins, pyrokinins, corazonin, and periviscerokinins [27]. A detailed knowledge of the structures and the distribution of myotropins in neurohaemal organs of P. americana makes it possible to investigate fluctuations in the amount of these peptides during development and certain physiological events. Muscle activity, however, can also be influenced by myoinhibitory neurohormones. For an understanding of the coordinated action of myotropic neurohormones it is therefore necessary to include myostimulatory peptides, as well as myoinhibitors. No myoinhibitory peptides have been reported from the American cockroach thus far. To fill this gap, we isolated and identified myoinhibitors from neurohaemal organs of P. americana. These putative hormones include leucomyosuppressin (LMS), initially isolated from L. maderae [11] and a novel member of myoinhibitory peptides (Pea-MIP). The efficacy of the different
0196-9781/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S 0 1 9 6 - 9 7 8 1 ( 0 0 ) 0 0 3 8 3 - 1
200
R. Predel et al. / Peptides 22 (2001) 199 –208
myoinhibitors of the American cockroach in visceral muscle assays is compared. During these experiments we also tested allatostatins which were known to inhibit spontaneous activity of the foregut of Leucophaea maderae [6].
ml/min). Eluent A was 0.1% aqueous trifluoroacetic acid (TFA) and eluent B was 0.1% TFA in 60% acetonitrile. Fractions, containing multiple tryptophyl residues, were further purified using an Inertsil ODS-2 (250 mm ⫻ 4.6 mm, 100 Å, 5 m; VDS Optilab, Berlin, Germany) under the same conditions (20 – 80% B) as described for HPLC-step 1.
2. Method 2.1. Animals Cockroaches were reared under a 12-h light, 12-h dark photoperiod at a constant temperature of 28°C. Only adult males four to six weeks after their last moult were used throughout the experiments. They were fed rat food pellets and had free access to water. 2.2. Bioassays The assays on the isolated visceral muscles were performed as previously described [20]. The insect saline (pH 7.25) contained NaCl (8.19 g/liter), KCl (0.37 g/liter), CaCl2 (0.56 g/liter), MgCl2 ⫻ 6H2O (0.2 g/liter), glucose (0.9 g/liter) and HEPES (2.4 g/liter). Muscle preparations were viable for many hours and could be used for multiple tests; a desensitization was not observed. Contractions were recorded photoelectrically by an isotonic transducer (hindgut, foregut), isometric transducer (antennal heart); or by an optical method (heart), by measuring the light that was transmitted through the preparation from below. 2.3. Extraction and purification 2.3.1. Leucomyosuppressin Dissection, extraction and initial HPLC-purification of an extract of 800 CC/CA was performed as previously described [21]. The fractions were collected manually and about 5 pmol was used for the hindgut and heart bioassays. Further purification of the myoinhibitory peptide was carried out on a Waters Model 510 HPLC system utilizing a Vydac-Diphenyl column (4.6 ⫻ 250 mm, 300 Å, 5, repacked by Phenomenex, Torrance, CA). Solvent A: 0.11% aqueous trifluoroacetic acid (TFA); solvent B: 60% aqueous acetonitrile adjusted to 0.1% TFA. Conditions: 5 min at 30% B, then linear gradient of 30 –100% B in 35 min (flow rate: 0.8 ml/min). Absorbance was monitored at 214 nm. 2.3.2. Myoinhibitory peptide (Pea-MIP) from the brain/ retrocerebral complex The initial purifications of extracts of 300 CC/CA and 350 brains were performed on a Gynkotek HPLC system (Gynkotek, Germering/Mu¨nchen, Germany) equipped with diode-array detector. The UV signals at 210 nm, 280 nm, and the 3D field were recorded. The first HPLC-separation was on a Hypersil RP-18 column (250 ⫻ 4.6 mm, 120 Å, 5 m) from VDS-Optilab (Montabaur, Germany) with a linear AB gradient of 10 –90% B over 40 min (flow rate: 1
2.3.3. Pea-MIP from abdominal PSOs (PSOs) Dissection and initial purification of an extract of 1000 abdominal PSOs from about 350 male cockroaches was performed as described previously [23]. HPLC-step 2 and 3 were carried out on an ABI 172 Microbore HPLC System (Applied Biosystems, Foster City, CA) equipped with a 140B dual-microsyringe gradient pump, a 785A UV/VIS detector (wavelength at 214 nm) and a Spheri RP-18 column (50 ⫻ 1 mm, 100 Å, 5 m) from Brownlee Labs (Applied Biosystems). Operating conditions of the second purification were as follows: Solvent A: 0.05% TFA in water. Solvent B: 50% acetonitrile, containing 0.043% TFA. Conditions: 20% B for 4 min, then 20 –100% B over 56 min (flow rate: 30 l/min). Operating conditions of the third HPLC-purification: Solvent A: 0.11% TFA in water. Solvent B: 60% acetonitrile, containing 0.1% TFA. Conditions: 25% B for 4 min, then 25– 80% B over 36 min (flow rate: 30 l/min). 2.4. Enzymatic analysis Putative LMS was dissolved in 25 l buffer, containing 100 mM Na2HPO4, 10 mM ethylenediaminetetraacetic acid (EDTA), 5 mM dithiothreitol (DTT), glycerol (5%v/v) and 0.5 units pyroglutamate aminopeptidase (Sigma). Following incubation for 20 min (37°C) 25 l 0.05% TFA in acetonitrile/water (20:80, v/v) was added to stop the reaction. The sample was then repurified under conditions described for HPLC-step 2 (see above). 2.5. Mass spectrometry, sequence analysis and synthesis A linear G2025A matrix-assisted laser desorption ionization (MALDI) time of flight (TOF) mass spectrometer (Hewlett Packard, Wilmington, DE) and a Voyager DE biospectrometry workstation (Perseptive Biosystems, Framingham, MA) were used to measure spectra in the positive mode. The matrix was always a 33 mM solution of ␣-cyano4-hydroxycinnamic acid dissolved in acetonitrile/methanol. The investigation of the neuropeptide inventory of the corpora cardiaca was performed as follows: Methanolic extracts of single organs were mixed with 0.1% TFA (50/50). Subsequently, 0.5 l of the sample was placed between two layers of ␣-cyano-4-hydroxycinnamic acid. All layers were dried separately (sandwich method) and then rinsed with ice-cold water for about 30 s. Sequences of LMS and PeaMIP from abdominal PSOs were analyzed on a G1005A sequencer system (Hewlett Packard, Wilmington, DE) using the Routine 3.5 chemistry software (Hewlett Packard tech-
R. Predel et al. / Peptides 22 (2001) 199 –208
201
Fig. 1. Initial purification of an extract of 800 CC/CA of Periplaneta americana on a RP-C18 column with 0.1% TFA as organic modifier. The strongly myoinhibitory fraction, which was further purified, is marked with an arrow. Fractions with abundant myostimulatory neuropeptides and Pea-MIP are designated. AUFS, Absorption units full scale, at 214 nm. PSK, perisulfakinin. LSK, leucosulfakinin. CAH, cardioaccelerating hormone.
nical note TN 96 –1). Edman-degradation of Pea-MIP from CC/CA extracts was performed on a model 477A sequenator connected to model 120A on-line PTH-analyzer (Applied Biosystems, Weiterstadt, Germany). Synthetic Pea-MIP was obtained from the Agricultural Biotechnology Center of the Institut for Biochemistry and Protein Research (Go`do`llo`, Hungary). Synthetic leucomyosuppressin was a kind gift from G. Mark Holman (College Station, Texas). 2.6. Production of antisera Rabbits were immunized with Pea-MIP-thyroglobulin conjugates [8]. The rabbits were injected six times over a period of nine months. 1 mg conjugate in complete Freund’s adjuvant was initially injected into each rabbit, and for the remaining five injections, 0.5 mg of the applicable conjugate in incomplete adjuvant was administered. Ten days after the last injection, animals were anesthetized and blood was collected using a vein catheder. The specificity of the antiserum was tested by replacing the antiserum with pre-immune rabbit serum, as well as by liquid-phase pre-absorption using neuropeptide-conjugates of Pea-CAH-1 (adipokinetic hormone; pQVNFSPNWamide) and Pea-MIP (50 g and 100 g per ml antiserum). Immunocytochemical procedures were performed as previously described [8]. Photographs were taken with a digital camera and processed by using the software analySIS Doku 3.0 (Soft Imaging System, Muenster).
3. Results 3.1. Isolation and structural elucidation of leucomyosuppressin An extract of 800 CC/CA was chromatographed on an Alphasil C18 column with TFA as organic modifier. The resulting fractions between 10 and 50 min were collected manually and then tested for myotropic activity on different visceral muscles, namely the hyperneural muscle, heart, and hindgut. The purification of peptides with myostimulatory properties has been described elsewhere (see 27). Only one fraction (Fig. 1) clearly inhibited the activity of isolated heart and hindgut. This fraction was further purified on a Diphenyl-column and, again, only one of the resulting fractions turned out to be myoinhibitory. An aliquot of the bioactive material was used for MALDI-TOF mass spectrometry which yielded a peak with the average mass of 1258.8 Da, typical of LMS. The remaining material was incubated with pyroglutamate aminopeptidase and repurified using the same HPLC-conditions as described for the second HPLC-purification. This purification resulted in the final separation of unblocked putative LMS (more hydrophilic) from contaminants. The deblocked peptide was subjected to sequence analysis and yielded the following sequence: Asp-Val-Asp-His-Val-Phe-Leu-Arg-?. The last amino acid was not detected in Edman-degradation but the remaining sequence was identical with LMS. The mass difference between the calculated mass of the sequence revealed in Edman-degradation (⫹ pGlu) and the mass of
202
R. Predel et al. / Peptides 22 (2001) 199 –208
Fig. 3. Spectra of Pea-MIP, Pea-CAH-1, and Pea-AST-7 as revealed by diode-array detector. The distinctive signal ratios of 210/280 nm result from different aromatic amino acids in these peptides. Pea-CAH-1 contains only a single Trp-residue, Pea-MIP contains two Trp-residues. The less distinct absorption maximum of Pea-Ast-7 at about 278 nm results from a Tyr-residue.
3.2. Isolation and structural elucidation of Pea-MIP from the brain/retrocerebral complex
Fig. 2. Comparison of mass spectra obtained from single CC of the cockroach Deropeltis erythrocephala (Blattidae; A), the mantid Epioscopus spec. (B) and the harvester termite Hodotermes mossambicus (C) (monoisotopic masses are given). The occurrence of leucomyosuppressin (as well as corazonin: 1369 Da) is clearly recognizable.
the native sample obtained in MALDI-TOF mass spectrometry suggested an amidated C-terminal Phe, which is also typical of LMS. Using mass spectrometric methods, this peptide was then confirmed to be stored in the storage lobes of the corpora cardiaca of different cockroach species (Blattidae: Blatta orientalis, Deropeltis erythrocephala, Neostylopyga rhombifolia), as well as related insects groups, including mantids (Epioscopus spec.) and termites (Hodotermes mossambicus) (Fig. 2). LMS was not found in abdominal PSOs of P. americana.
Pea-MIP was detected when searching for an AKH-like peptide in the brain/retrocerebral complex of the American cockroach. Based on spectral analysis and determination of signal ratio of 210/280 nm Trp-containing peptides were sorted out after the first HPLC-separation and further purified on an Inertsil RP-18 column. Fractions, co-eluting in both HPLC-runs with synthetic perikinins (FXSWGamides; 22) were excluded from further examination. Adipokinetic hormones (Pea-CAH-1 and 2) were only found in the extract of the CC/CA. Additionally, several substances with spectra typical of peptides containing more than a single Trp (Fig. 3) were obtained after this second HPLC-step. However, only one fraction of brain and CC/CA extracts contained enough material for structural elucidation, respectively. These substances had similar retention times (24.3 min and 21.1 min in HPLC-step 1 and 2, respectively) which suggested the occurrence of the same peptide in both extracts. The identity of both substances was then confirmed using MALDI-TOF mass spectrometry (monoisotopic [M⫹H]⫹: 1045.4 Da). Mass spectra obtained from this putative peptide showed only weak [M⫹H]⫹ signals but
R. Predel et al. / Peptides 22 (2001) 199 –208
203
HPLC-step 1 and 2, respectively) contained a substance with a mass similar to that of Pea-MIP. Only about 5 pmol of this substance were found after the third HPLC-purification and subjected to Edman-degradation. The sequencing procedure revealed first six amino acids (Gly-Trp-Gln-AspLeu-Gln) which were identical to those of Pea-MIP. This result, together with the mass data obtained from the same material, confirmed the occurrence of Pea-MIP in the abdominal PSOs. 3.4. Immunolocalization of Pea-MIP
Fig. 4. MALDI-TOF mass spectrum of purified Pea-MIP showing distinct [M⫹Na] and [M⫹K] ions, not typical of other neuropeptides of the American cockroach (with adipokinetic hormones being the exception).
distinct [M⫹Na]⫹ and [M⫹K]⫹ peaks (Fig. 4) resembling those of adipokinetic peptides which normally consist of [M⫹Na]⫹ and [M⫹K]⫹ ions only. In addition, mass spectral analysis of the purified material (about 5 pmol/l) revealed a relatively low signal intensity which probably makes the detection in unseparated peptide mixtures difficult. About 50 pmol of the peptide (material from CC/CA extract) was subjected to Edman-degradation and gave the following sequence: Gly-Trp-Gln-Asp-Leu-Gln-Gly-GlyTrp. This sequence was in agreement with mass data and supported a C-terminal amidation. According to the name of the first identified member of this peptide family (LomMIP; 29), the novel peptide was named Pea-MIP. Synthetic Pea-MIP had a retention time identical to that of native material from brain extracts. The retention behavior of PeaMIP in comparison with other known myotropins of the American cockroach is shown in Fig. 1. This figure also illustrates that Pea-MIP belongs to the abundant neuropeptides in the CC/CA of the American cockroach. Using the same purification procedure and mass spectrometric methods it was shown that Pea-MIP is also present in the brain/retrocerebral complex of a blaberoid cockroach, namely Nauphoeta cineraea (not shown). 3.3. Isolation and structural elucidation of Pea-MIP from the abdominal PSOs The neuropeptides of the abdominal PSOs of the American cockroach were thoroughly investigated and structures of abundant substances have been described [25]. All collected fractions of the first HPLC-separation were further purified in a second HPLC-step on a microbore column and, subsequently, the masses of all detectable substances were measured by means of MALDI-TOF mass spectrometry. One of these fractions (at about 24.5 min and 41.5 min in
Using the anti-Pea-MIP serum numerous immunopositive intrinsic cells were found throughout the central nervous system. In this study, however, we focused on cells supplying neurohaemal organs and intestine with Pea-MIP immunoreactive (IR) material. A) Brain/retrocerebral complex: Preabsorption of the Pea-MIP antiserum with Pea-CAH-1 did not abolish any immunostaining. In the CC, only the storage lobes were strongly immunostained, which also indicated that the antiPea-MIP serum did not cross-react with Pea-CAH-1 and 2. No immunostaining was detected, however, after preabsorption with Pea-MIP. Pea-MIP IR material in the storage lobes of the CC (Fig. 5A) originated from neurosecretory cells in the protocerebrum, namely the pars intercerebralis (Fig. 5A) and pars lateralis. These neurons innervated the retrocerebral complex via the nervi corporis cardiaci-1 and 2. Immunopositive fibers were found along the nervi corporis allati-1, run through the CA, and left the retrocerebral complex via the postallatal nerves. B) Abdominal ventral nerve cord: Most prominent PeaMIP immunostaining was detected in large neurons (30 – 40 m) which were located in the dorsal midline of the abdominal ganglia 3– 6 (Fig. 5B), and the terminal ganglion (Fig. 5C). Axons of these neurons from abdominal ganglia 3– 6 projected into the terminal ganglion. Immunostained fibers left the terminal ganglion via the anterior proctodeal nerves which innervate the hindgut. Additionally to the neurons in the dorsal midline, a few smaller cells (15–20 m) were stained in the posterior region of each unfused abdominal ganglion. These cells sent fibers into the segmental nerves. A branch of the segmental nerves of the more anterior abdominal ganglion supplied the abdominal PSOs via the link nerve/transverse nerve with immunoreactive material. Pea-MIP IR fibers have also been detected in the posterior median nerve (Fig. 5B). In the terminal ganglion, transverse nerves 7–9 were immunostained (Fig. 5C). The anterior median nerves of all unfused abdominal ganglia did not contain any immunoreactive material. Immunostaining in the abdominal PSOs consisted of only a few fibers passing these neurohaemal organs and no accumulation of Pea-MIP-like material typical of periviscerokinins [8,23] and pyrokinin-5 [26] was detected. C) Stomatogastric nervous system and intestine: All ganglia of the stomatogastric nervous system, including frontal
204
R. Predel et al. / Peptides 22 (2001) 199 –208
Fig. 5. Pea-MIP-like IR in Periplaneta americana. A: Brain and retrocerebral complex. Numerous immunoreactive cells are visible in the pars intercerebralis (PI) and the posterior protocerebrum. In all neuropiles, as alpha lobes (aL) of the mushroom body and calyces, immunopositive fibers were stained. The storage lobes of the CC, NCA-1, and fibers crossing the CA also revealed MIP-like IR. In addition, fibers in circular muscles of the pharynx, which originate in the stomatogastric nervous system were immunoreactive (arrows). B: Unfused abdominal ganglion 4 with associated anterior PSOs. Large MIPimmunoreactive neurons were stained in the dorsal midline. MIP-immunoreactive fibers are visible in the posterior median nerve, abdominal PSOs, and transverse nerves. C: Terminal ganglion. Four cells in the dorsal midline and numerous lateral positioned neurons are visible. MIP-like IR has been found in segmental nerves and all transverse nerves (TN) of this ganglion. D: Ingluvial ganglion of the stomatogastric nervous system with immunoreactive neurons, and immunopositive fibers in all nerves leaving this ganglion. E: Circular muscles of the foregut with immunoreactive fibers. No Pea-MIP-like IR was detected in longitudinal muscles.
R. Predel et al. / Peptides 22 (2001) 199 –208
205
Fig. 6. Response of an antennal heart preparation to different concentrations of Pea-MIP. A concentration 10-fold higher than the threshold concentration usually produced a total inhibition which lasted longer when the peptide concentration was increased. This preparation was rinsed between the applications only.
ganglion, hypocerebral ganglion, and ingluvial ganglion (Fig. 5D) contained a number of Pea-MIP IR cells. Additionally, immunopositive cells were found on the esophageal nerve between the hypocerebral and ingluvial ganglion. Whole-mount preparations of the foregut revealed that only the circular muscles were innervated by Pea-MIP IR fibers (Fig. 5E). The hindgut is innervated by the anterior proctodeal nerve (see above). Both, the hindgut and foregut, did not contain Pea-MIP IR neurons. 3.5. Muscle efficacy 3.5.1. LMS This peptide was highly effective at inhibiting phasic contractions of all tested visceral muscles, influencing both the amplitude and frequency of contractions. Threshold concentrations (concentrations required to produce a visible change in frequency and/or amplitude), estimated from five muscle preparations, were about 1 ⫻ 10⫺10 M on isolated hindgut and foregut preparations, and 5 ⫻ 10⫺10 M on the semi-isolated heart. The inhibitory effect of LMS on antennal heart preparations of the American cockroach was studied more extensively and will be published in a separate paper (Hertel et al., in prep.). A concentration 10-fold higher than the threshold concentration typically resulted in a total inhibition of phasic activity. This inhibition was reversible after replacement with normal saline. 3.5.2. Pea-MIP Application of synthetic Pea-MIP in different visceral muscle assays always revealed myoinhibitory effects. These effects, however, were obtainable only at high peptide con-
Fig. 7. Parallel recording of two foregut preparations of P. americana. Application of different doses of synthetic Pea-MIP resulted in weak and short-time inhibitory effects. Leucomyosuppressin was added as standard at the end of the test series. The preparation was continuously rinsed with saline, the application of each peptide lasted about five minutes.
centrations. Threshold concentrations were as high as 1 ⫻ 10⫺7 M on isolated hindgut and heart preparations and even the application of 1 ⫻ 10⫺5 M Pea-MIP did not result in a total inhibition in these muscle assays. A more distinct inhibition was found in the antennal heart preparations (threshold at about 2 ⫻ 10⫺9 M, total inhibition at 10⫺8 M) (Fig. 6). On the isolated foregut slightly inhibiting effects were already observed at a threshold of about 5 ⫻ 10⫺9 M but nearly no increase of inhibition was detected after application of higher doses (Fig. 7)
206
R. Predel et al. / Peptides 22 (2001) 199 –208
Fig. 8. Application of different doses of synthetic Pea-AST-7 on a foregut preparation of P. americana. Proctolin was added as standard at the end of the test series. The preparation was continuously rinsed with saline the application of each peptide lasted about five minutes.
3.5.3. Allatostatins
4. Discussion
Members of this peptide family have been reported to inhibit visceral muscle activity in different insects [7,10,31, 34]. In the blaberoid cockroach Leucophaea maderae, allatostatins were highly effective in decreasing phasic activity of the foregut but not active on the hindgut [6], whereas the hindgut of the related Diploptera punctata responded with decreasing activity to allatostatin application [13]. In P. americana, fourteen allatostatins are known from cDNA studies [4], most of these putative messenger molecules were also found to be expressed [24]. Testing a single member of the allatostatin peptides, Pea-Ast-7, we could confirm, that allatostatins are highly effective in inhibiting foregut activity in P. americana (Fig. 8). Other visceral muscles, including hindgut, antennal heart, and heart showed no visible response after application of even 5 ⫻ 10⫺7 M Pea-Ast-7. This allatostatin also did not influence the response of these muscles after application of 10⫺8 M proctolin (not shown).
The only myoinhibitor, which was highly effective in all visceral muscle assays used during this study, was LMS. Action, biologically active core sequence, and distribution of this peptide was extensively investigated in blaberoid cockroaches [1,3,5,9,16,17,18]. LMS seems to be conservative not only in its function as myoinhibitor in insects but also in its sequence which is obviously highly conserved in cockroaches and related insect orders. These properties resemble those of the myostimulatory proctolin, which remained unchanged in structure and function throughout the different insect groups [19]. A second group of potent myoinhibitory peptides are the allatostatins, originally isolated as inhibitors of juvenile hormone synthesis [10,32,35]. As it was found for the cockroach Leucophaea maderae [6], allatostatins (in this study: Pea-AST-7) effectively inhibit spontaneous activity of the foregut of the American cockroach but not those of other visceral muscles. Myoinhibitory efficacy of allatostatins is
R. Predel et al. / Peptides 22 (2001) 199 –208
not correlated with the occurrence of these peptides in visceral muscles. These peptides are also abundant in parts of the intestine other than the foregut [28,38], antennal hearts [24,37], and were (immunocytochemically) also detected in the lateral heart nerve [33]. The third group of inhibitory neuropeptides described here (Pea-MIP) belongs to a peptide family which is already known from locusts, crickets and moths [2,14,15,29]. Interestingly, these peptides, initially isolated using a muscle bioassay [29] were later found to inhibit juvenile hormone biosynthesis in crickets [14] but with threshold concentrations 10-fold higher than those of allatostatins. Immunocytochemical studies in locusts [30], crickets [36], and cockroaches (this paper) always revealed an innervation of the CA via the nervi corporis allati-1. It is not yet clear, if this immunostaining is correlated with regulation of CA activity or only caused by immunoreactive fibers passing the CA. The physiological role of Pea-MIP as myoinhibitor in the American cockroach is questionable. Although present in relatively large amounts in the storage lobes of the CC, it is unlikely that this peptide can be released in quantities sufficient for inhibiting visceral muscle activity. On the other hand, our immunocytochemical studies revealed the presence of Pea-MIP like material in circular muscles of parts of the intestine whereas the routine bioassay, which was used in this study, recorded mainly contractions along the longitudinal axis of the gut. Interruptions in activity of ring muscles are nearly undetectable if longitudinal muscles are still active. Pea-MIP was found not only in neurohaemal release sites of the head, but also in abdominal PSOs. This distribution is unusual among bioactive neuropeptides of the American cockroach since peptides are commonly stored only in one of these organs [25]. The different amounts of Pea-MIP in the CC/CA and abdominal PSO, however, indicated that this peptide may be released as a hormone from the retrocerebral complex but not from abdominal PSO. This was then confirmed by our immunocytochemical results which evidenced that only a few Pea-MIP-IR fibers run through the abdominal PSO. These fibers did not originate from those three median cell groups of each unfused abdominal ganglion known to synthesize all abundant myotropic neuropeptides of the abdominal PSO [8,23,26]. The isolation and identification of LMS and Pea-MIP finalized our efforts to describe the putative myoregulatory neuropeptides from major neurohaemal organs of the central nervous system of the American cockroach.
Acknowledgments We thank Mrs. Virginia Johnson (Protein Technologies Laboratory, Texas A & M University, College Station) for excellent sequence analysis; Mrs. Renate Winkler, Mrs. Erika Krause and Mrs. Ch. Raue (Institut fu¨r Allgemeine Zoologie und Tierphysiologie Jena, Germany) for perfect
207
technical assistance. This study was supported with a postdoc fellowship of the German Academy of Science Leopoldina (LPD 1997, to RP) and grants from Deutsche Forschungsgemeinschaft (Predel 595/1–1,2; Eckert 122/4 – 1).
References [1] Bendena WG, Donly BC, Fuse M, Lee E, Lange AB, Orchard I, Tobe SS. Molecular characterization of the inhibitory myotropic peptide leucomyosuppressin. Peptides 1997;18:157– 63. [2] Blackburn MB, Wagner RM, Kochansky JP, Harrison DJ, ThomasLemont P, Raina AK. The identification of two myoinhibitory peptides, with sequence similarities to the galanins, isolated from the ventral nerve cord of Manduca sexta. Regul Peptides 1995;57:213– 19. [3] Cook BJ, Wagner RM. Comparative effects of leucomyosuppressin on the visceral muscle systems of the cockroach Leucophaea maderae. Comp Biochem Physiol C 1991;99:95–9. [4] Ding Q, Donly BC, Tobe SS, Bendena WG. Comparison of the allatostatin neuropeptide precursors in the distantly related cockroaches Periplaneta americana and Diploptera punctata. Eur J Biochem 1995;234:737– 46. [5] Donly BC, Fuse M, Orchard I, Tobe SS, Bendena WG. Characterization of the gene for leucomyosuppressin and its expression in the brain of the cockroach Diploptera punctata. Insect Biochem Molec Biol 1996;26:627–37. [6] Duve H, Wren P, Thorpe A. Innervation of the foregut of the cockroach Leucophaea maderae and inhibition of spontaneous contractile activity by callatostatin neuropeptides. Physiol Entomol 1995;20:33– 44. [7] Duve H, Thorpe A, Johnsen AH, Maestro JL, Scott AG, East PD. The dipteran Leu-callatostatins: structural and functional diversity in an insect neuroendocrine peptide family. In: Coast, G. M.; Webster, S. M., eds. Recent advances in arthropod endocrinology. Cambridge University Press, 1998:229 –247. [8] Eckert M, Predel R, Gundel M. Periviscerokinin-like immunoreactivity in the nervous system of the American cockroach. Cell Tissue Res 1999;295:159 –70. [9] Fuse M, Bendena WG, Donly BC, Tobe SS, Orchard I. In situ hybridization analysis of leucomyosuppressin mRNA expression in the cockroach, Diploptera punctata. J Comp Neurol 1998;395:328 – 41. [10] Hoffmann KH, Meyering-Vos M, Lorenz MW. Allatostatins and allatotropins: is the regulation of corpora allata activity their primary function? Eur J Entomol 1999;96:255– 66. [11] Holman GM, Cook BJ, Nachman, RJ. Isolation, primary structure and synthesis of leucomyosuppressin, an insect neuropeptide that inhibits spontaneous contractions of the cockroach hindgut. Comp Biochem Physiol C 1986;85:329 –33. [12] Holman GM, Nachman RJ, Wright MS, Schoofs L, Hayes TK, DeLoof A. Insect myotropic peptides. Isolation, structural characterization, and biological activities. In: Menn, J. J.; Kelly T. J.; Masler E. P., eds. Insect neuropeptides. Chemistry, biology and action. ACS Symposium Series:1991:40 –50. [13] Lange AB, Chan KK, Stay B. Effect of allatostatin and proctolin on antennal pulsatile organ and hindgut muscle in the cockroach, Diploptera punctata. Arch Insect Biochem Physiol 1993;24:79 –92. [14] Lorenz MW, Kellner R, Hoffmann KH. A family of neuropeptides that inhibit juvenile hormone biosysnthesis in the cricket, Gryllus bimaculatus. J Biol Chem 1995;270:21103– 8. [15] Lorenz MW, Kellner R, Hoffmann KH. Allatostatins in Gryllus bimaculatus (Ensifera: Gryllidae): new structures and physiological properties. Eur J Entomol 1999;96:267–74.
208
R. Predel et al. / Peptides 22 (2001) 199 –208
[16] Meola SM, Wright MS, Holman GM, Thompson JM. Immunocytochemical localization of leucomyosuppressin-like peptides in the CNS of the cockroach, Leucophaea maderae. Neurochem Res 1991;16:543–9. [17] Nachman RJ, Holman GM, Hayes TK, Beier RC. Structure-activity relationships for inhibitory insect myosuppressins: contrast with the stimulatory sulfakinins. Peptides 1993;14:665–70. [18] Nachman RJ, Giard W, Favrel P, Suresh T, Sreekumar S, Holman GM. Insect myosuppressins stimulate release of the digestive enzyme ␣-amylase in two invertebrates: the scallop Pecten maximus and insect Rhynchophorus ferrugineus. In: Strand F, Beckwith W, Sandman C, editors. Neuropeptides in developing and aging. New York: Annals NY Acad. Sci 1997;814:335– 8. [19] Orchard I, Belanger JH, Lange, AB. Proctolin. A review with emphasis on insects. J Neurobiol 1989;20:470 –96. [20] Predel R, Agricola H, Linde D, Wollweber L, Veenstra JA, Penzlin H. The insect neuropeptide corazonin: physiological and immunocytochemical studies in Blattariae. Zoology 1994;9:35– 49. [21] Predel R, Kellner R, Kaufmann R, Penzlin H, Ga¨de G. Isolation and structural elucidation of two pyrokinins from the retrocerebral complex of the American cockroach. Peptides 1997;18:473– 8. [22] Predel R, Kellner R, Rapus J, Penzlin H, Ga¨de G. Isolation and structural elucidation of eight kinins from the retrocerebral complex of the American cockroach. Regul Peptides 1997;71:199 –205. [23] Predel R, Rapus J, Eckert M, Holman GM, Nachman RJ, Wang Y, Penzlin H. Isolation of periviscerokinin-2 from the abdominal perisympathetic organs of the American cockroach, Periplaneta americana. Peptides 1998;19:801–9. [24] Predel R, Kellner R, Rapus J, Ga¨de G. Allatostatins from retrocerebral complex and antennal pulsatile organ of the American cockroach, Periplaneta americana, aided by matrix-assisted laser desorption/ionization-time-of-flight mass spectrometry. Regul Peptides 1999;82:81–9. [25] Predel R, Eckert M, Holman GM. The unique neuropeptide pattern in abdominal perisympathetic organs of insects. Annals NY Acad Sci 1999;897:282–90. [26] Predel R, Eckert M. Tagma-specific distribution of FXPRLamides in the nervous system of the American cockroach. J Comp Neurol 2000;419:352– 63. [27] Predel R, Nachman RJ, Ga¨de G. Myostimulatory Neuropeptides in Cockroaches: Structures, Distribution, Pharmacological Activities, and Mimetic Analogs (Review). J. Insect Physiol., in press.
[28] Reichwald K, Unnithan GC, Davis NT, Agricola H, Feyereisen R. Expression of the allatostatin gene in endocrine cells of the cockroach midgut. Proc Natl Acad Sci USA 1994;91:11894 – 8. [29] Schoofs L, Holman GM, Hayes TK, Nachman RJ, DeLoof A. Isolation, identification and synthesis of locustamyoinhibiting peptide (Lom-MIP), a novel biologically active neuropeptide from Locusta migratoria. Regul Peptides 1991;36:111–19. [30] Schoofs L, Veelaert J, Vanden Broeck J, DeLoof A. Immunocytochemical distribution of locustamyoinhibiting peptide (Lom-MIP) in the nervous system of Locusta migratoria. Regul Peptides 1996;63: 171–9. [31] Schoofs L, Veelaert D, Vanden Broeck J, De Loof A. Peptides in the locusts, Locusta migratoria and Schistocerca gregaria. Peptides 1997;18:145–56. [32] Stay B, Tobe SS, Bendena WG. Allatostatins: identification, primary structure, functions and distribution. Adv Insect Physiol 1994;25: 267–338. [33] Ude J, Agricola H. FMRFamide-like and allatostatin-like immunoreactivity in the lateral heart nerve of Periplaneta americana: Colocalization at the electron-microscopic level. Cell Tissue Res 1995;282: 69 – 80. [34] Vilaplana L, Maestro JL, Piulachs M-D, Belles X. Modulation of cardiac rhythm by allatostatins in the cockroach Blattella germanica (L.) (Dictyoptera, Blattellidae). J Ins Physiol 1999;45:1057– 64. [35] Weaver RJ, Edwards JP, Bendena WG, Tobe SS. Structures, functions and occurrences of insect allatostatic peptides. In: Coast, G. M.; Webster, S. M., eds. Recent advances in arthropod endocrinology. Cambridge University Press, 3–32; 1998. [36] Witek G, Verhaert P, Lorenz MW, Hoffmann KH. Immunolocalization of two types of allatostatins in the central nervous system of the cricket Gryllus bimaculatus (Ensifera: Gryllidae). Eur J Entomol 1999;96:279 – 85. [37] Woodhead AP, Stoltzman CA, Stay B. Allatostatins in the nerves of the antennal pulsatile organ muscle of the cockroach Diploptera punctata. Arch Insect Biochem Physiol 1992;20:253– 63. [38] Yu CG, Stay B, Ding Q, Bendena WG, Tobe SS. Immunochemical identification and expression of allatostatins in the gut of Diploptera puncata. J Insect Physiol 1993;41:1035– 43.
Myoinhibitory neuropeptides in the American cockroach夞 Reinhard Predel*, Ju¨rgen Rapus, Manfred Eckert Institut fu¨r Allgemeine Zoologie und Tierphysiologie, Friedrich-Schiller-Universita¨t, Erbertstrae 1, D-07743 Jena, Germany Received 15 January 2000; accepted 20 September 2000
Abstract A large number of myostimulatory neuropeptides from neurohaemal organs of the American cockroach have been described since 1989. These peptides, isolated from the retrocerebral complex and abdominal perisympathetic organs, are thought to be released as hormones. To study the coordinated action of these neuropeptides in the regulation of visceral muscle activity, it might be necessary to include myoinhibitors as well, however, not a single myoinhibitory neuropeptide of the American cockroach has been described so far. To fill this gap, we describe the isolation of LMS (leucomyosuppressin) and Pea-MIP (myoinhibitory peptide) from neurohaemal organs of the American cockroach. LMS was very effective in inhibiting phasic activity of all visceral muscles tested. It was found in the corpora cardiaca of different species of cockroaches, as well as in related insect groups, including mantids and termites. Pea-MIP which is strongly accumulated in the corpora cardiaca was not detected with a muscle bioassay system but when searching for tryptophane-containing peptides using a diode-array detector. This peptide caused only a moderate inhibition in visceral muscle assays. The distribution of Pea-MIP in neurohaemal organs and cells supplying these organs with Pea-MIP immunoreactive material, is described. Additionally to LMS and Pea-MIP, a member of the allatostatin peptide family, known to exhibit inhibitory properties in other insects, was tested in visceral muscle assays. This allatostatin was highly effective in inhibiting spontaneous activity of the foregut, but not of other tested visceral muscles of the American cockroach. © 2001 Elsevier Science Inc. All rights reserved. Keywords: Periplaneta americana; Insect; Neuropeptide; Leucomyosuppressin; Allatostatin; Immunohistochemistry; Photodiode-array detector
1. Introduction The neuropeptide pattern in neurohaemal organs of the American cockroach, Periplaneta americana, was intensively investigated in the last few years. Hereby, we concentrated on myotropic neuropeptides of the retrocerebral complex and abdominal perisympathetic organs which are the major neurohaemal release sites of the brain/subesophageal ganglion and ventral nerve cord, respectively. Two general conclusions can be drawn from these studies. Firstly, members of all myostimulatory neuropeptide families which were initially isolated from head extracts of Leucophaea maderae [12] are also present in considerable amounts in the corpora cardiaca/allata complex (CC/CA) of the American cockroach. This fact supports the idea that these peptides will be released into the circulation. Sec夞 Taken from a paper presented at the Winter Neuropeptide Conference 2000, Invertebrate Division, Hua Hin, Thailand, January 10 –15, 2000. * Corresponding author. Tel.: ⫹49-03641-9-49125; fax: ⫹49-036419-49102. E-mail address: [email protected] (R. Predel).
ondly, the abdominal perisympathetic organs (PSOs) were found to contain a neuropeptide pattern totally different from that of the retrocerebral complex. Altogether, not less than five different neuropeptide families with myostimulatory properties were identified from neurohaemal organs of the American cockroach, namely kinins, sulfakinins, pyrokinins, corazonin, and periviscerokinins [27]. A detailed knowledge of the structures and the distribution of myotropins in neurohaemal organs of P. americana makes it possible to investigate fluctuations in the amount of these peptides during development and certain physiological events. Muscle activity, however, can also be influenced by myoinhibitory neurohormones. For an understanding of the coordinated action of myotropic neurohormones it is therefore necessary to include myostimulatory peptides, as well as myoinhibitors. No myoinhibitory peptides have been reported from the American cockroach thus far. To fill this gap, we isolated and identified myoinhibitors from neurohaemal organs of P. americana. These putative hormones include leucomyosuppressin (LMS), initially isolated from L. maderae [11] and a novel member of myoinhibitory peptides (Pea-MIP). The efficacy of the different
0196-9781/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S 0 1 9 6 - 9 7 8 1 ( 0 0 ) 0 0 3 8 3 - 1
200
R. Predel et al. / Peptides 22 (2001) 199 –208
myoinhibitors of the American cockroach in visceral muscle assays is compared. During these experiments we also tested allatostatins which were known to inhibit spontaneous activity of the foregut of Leucophaea maderae [6].
ml/min). Eluent A was 0.1% aqueous trifluoroacetic acid (TFA) and eluent B was 0.1% TFA in 60% acetonitrile. Fractions, containing multiple tryptophyl residues, were further purified using an Inertsil ODS-2 (250 mm ⫻ 4.6 mm, 100 Å, 5 m; VDS Optilab, Berlin, Germany) under the same conditions (20 – 80% B) as described for HPLC-step 1.
2. Method 2.1. Animals Cockroaches were reared under a 12-h light, 12-h dark photoperiod at a constant temperature of 28°C. Only adult males four to six weeks after their last moult were used throughout the experiments. They were fed rat food pellets and had free access to water. 2.2. Bioassays The assays on the isolated visceral muscles were performed as previously described [20]. The insect saline (pH 7.25) contained NaCl (8.19 g/liter), KCl (0.37 g/liter), CaCl2 (0.56 g/liter), MgCl2 ⫻ 6H2O (0.2 g/liter), glucose (0.9 g/liter) and HEPES (2.4 g/liter). Muscle preparations were viable for many hours and could be used for multiple tests; a desensitization was not observed. Contractions were recorded photoelectrically by an isotonic transducer (hindgut, foregut), isometric transducer (antennal heart); or by an optical method (heart), by measuring the light that was transmitted through the preparation from below. 2.3. Extraction and purification 2.3.1. Leucomyosuppressin Dissection, extraction and initial HPLC-purification of an extract of 800 CC/CA was performed as previously described [21]. The fractions were collected manually and about 5 pmol was used for the hindgut and heart bioassays. Further purification of the myoinhibitory peptide was carried out on a Waters Model 510 HPLC system utilizing a Vydac-Diphenyl column (4.6 ⫻ 250 mm, 300 Å, 5, repacked by Phenomenex, Torrance, CA). Solvent A: 0.11% aqueous trifluoroacetic acid (TFA); solvent B: 60% aqueous acetonitrile adjusted to 0.1% TFA. Conditions: 5 min at 30% B, then linear gradient of 30 –100% B in 35 min (flow rate: 0.8 ml/min). Absorbance was monitored at 214 nm. 2.3.2. Myoinhibitory peptide (Pea-MIP) from the brain/ retrocerebral complex The initial purifications of extracts of 300 CC/CA and 350 brains were performed on a Gynkotek HPLC system (Gynkotek, Germering/Mu¨nchen, Germany) equipped with diode-array detector. The UV signals at 210 nm, 280 nm, and the 3D field were recorded. The first HPLC-separation was on a Hypersil RP-18 column (250 ⫻ 4.6 mm, 120 Å, 5 m) from VDS-Optilab (Montabaur, Germany) with a linear AB gradient of 10 –90% B over 40 min (flow rate: 1
2.3.3. Pea-MIP from abdominal PSOs (PSOs) Dissection and initial purification of an extract of 1000 abdominal PSOs from about 350 male cockroaches was performed as described previously [23]. HPLC-step 2 and 3 were carried out on an ABI 172 Microbore HPLC System (Applied Biosystems, Foster City, CA) equipped with a 140B dual-microsyringe gradient pump, a 785A UV/VIS detector (wavelength at 214 nm) and a Spheri RP-18 column (50 ⫻ 1 mm, 100 Å, 5 m) from Brownlee Labs (Applied Biosystems). Operating conditions of the second purification were as follows: Solvent A: 0.05% TFA in water. Solvent B: 50% acetonitrile, containing 0.043% TFA. Conditions: 20% B for 4 min, then 20 –100% B over 56 min (flow rate: 30 l/min). Operating conditions of the third HPLC-purification: Solvent A: 0.11% TFA in water. Solvent B: 60% acetonitrile, containing 0.1% TFA. Conditions: 25% B for 4 min, then 25– 80% B over 36 min (flow rate: 30 l/min). 2.4. Enzymatic analysis Putative LMS was dissolved in 25 l buffer, containing 100 mM Na2HPO4, 10 mM ethylenediaminetetraacetic acid (EDTA), 5 mM dithiothreitol (DTT), glycerol (5%v/v) and 0.5 units pyroglutamate aminopeptidase (Sigma). Following incubation for 20 min (37°C) 25 l 0.05% TFA in acetonitrile/water (20:80, v/v) was added to stop the reaction. The sample was then repurified under conditions described for HPLC-step 2 (see above). 2.5. Mass spectrometry, sequence analysis and synthesis A linear G2025A matrix-assisted laser desorption ionization (MALDI) time of flight (TOF) mass spectrometer (Hewlett Packard, Wilmington, DE) and a Voyager DE biospectrometry workstation (Perseptive Biosystems, Framingham, MA) were used to measure spectra in the positive mode. The matrix was always a 33 mM solution of ␣-cyano4-hydroxycinnamic acid dissolved in acetonitrile/methanol. The investigation of the neuropeptide inventory of the corpora cardiaca was performed as follows: Methanolic extracts of single organs were mixed with 0.1% TFA (50/50). Subsequently, 0.5 l of the sample was placed between two layers of ␣-cyano-4-hydroxycinnamic acid. All layers were dried separately (sandwich method) and then rinsed with ice-cold water for about 30 s. Sequences of LMS and PeaMIP from abdominal PSOs were analyzed on a G1005A sequencer system (Hewlett Packard, Wilmington, DE) using the Routine 3.5 chemistry software (Hewlett Packard tech-
R. Predel et al. / Peptides 22 (2001) 199 –208
201
Fig. 1. Initial purification of an extract of 800 CC/CA of Periplaneta americana on a RP-C18 column with 0.1% TFA as organic modifier. The strongly myoinhibitory fraction, which was further purified, is marked with an arrow. Fractions with abundant myostimulatory neuropeptides and Pea-MIP are designated. AUFS, Absorption units full scale, at 214 nm. PSK, perisulfakinin. LSK, leucosulfakinin. CAH, cardioaccelerating hormone.
nical note TN 96 –1). Edman-degradation of Pea-MIP from CC/CA extracts was performed on a model 477A sequenator connected to model 120A on-line PTH-analyzer (Applied Biosystems, Weiterstadt, Germany). Synthetic Pea-MIP was obtained from the Agricultural Biotechnology Center of the Institut for Biochemistry and Protein Research (Go`do`llo`, Hungary). Synthetic leucomyosuppressin was a kind gift from G. Mark Holman (College Station, Texas). 2.6. Production of antisera Rabbits were immunized with Pea-MIP-thyroglobulin conjugates [8]. The rabbits were injected six times over a period of nine months. 1 mg conjugate in complete Freund’s adjuvant was initially injected into each rabbit, and for the remaining five injections, 0.5 mg of the applicable conjugate in incomplete adjuvant was administered. Ten days after the last injection, animals were anesthetized and blood was collected using a vein catheder. The specificity of the antiserum was tested by replacing the antiserum with pre-immune rabbit serum, as well as by liquid-phase pre-absorption using neuropeptide-conjugates of Pea-CAH-1 (adipokinetic hormone; pQVNFSPNWamide) and Pea-MIP (50 g and 100 g per ml antiserum). Immunocytochemical procedures were performed as previously described [8]. Photographs were taken with a digital camera and processed by using the software analySIS Doku 3.0 (Soft Imaging System, Muenster).
3. Results 3.1. Isolation and structural elucidation of leucomyosuppressin An extract of 800 CC/CA was chromatographed on an Alphasil C18 column with TFA as organic modifier. The resulting fractions between 10 and 50 min were collected manually and then tested for myotropic activity on different visceral muscles, namely the hyperneural muscle, heart, and hindgut. The purification of peptides with myostimulatory properties has been described elsewhere (see 27). Only one fraction (Fig. 1) clearly inhibited the activity of isolated heart and hindgut. This fraction was further purified on a Diphenyl-column and, again, only one of the resulting fractions turned out to be myoinhibitory. An aliquot of the bioactive material was used for MALDI-TOF mass spectrometry which yielded a peak with the average mass of 1258.8 Da, typical of LMS. The remaining material was incubated with pyroglutamate aminopeptidase and repurified using the same HPLC-conditions as described for the second HPLC-purification. This purification resulted in the final separation of unblocked putative LMS (more hydrophilic) from contaminants. The deblocked peptide was subjected to sequence analysis and yielded the following sequence: Asp-Val-Asp-His-Val-Phe-Leu-Arg-?. The last amino acid was not detected in Edman-degradation but the remaining sequence was identical with LMS. The mass difference between the calculated mass of the sequence revealed in Edman-degradation (⫹ pGlu) and the mass of
202
R. Predel et al. / Peptides 22 (2001) 199 –208
Fig. 3. Spectra of Pea-MIP, Pea-CAH-1, and Pea-AST-7 as revealed by diode-array detector. The distinctive signal ratios of 210/280 nm result from different aromatic amino acids in these peptides. Pea-CAH-1 contains only a single Trp-residue, Pea-MIP contains two Trp-residues. The less distinct absorption maximum of Pea-Ast-7 at about 278 nm results from a Tyr-residue.
3.2. Isolation and structural elucidation of Pea-MIP from the brain/retrocerebral complex
Fig. 2. Comparison of mass spectra obtained from single CC of the cockroach Deropeltis erythrocephala (Blattidae; A), the mantid Epioscopus spec. (B) and the harvester termite Hodotermes mossambicus (C) (monoisotopic masses are given). The occurrence of leucomyosuppressin (as well as corazonin: 1369 Da) is clearly recognizable.
the native sample obtained in MALDI-TOF mass spectrometry suggested an amidated C-terminal Phe, which is also typical of LMS. Using mass spectrometric methods, this peptide was then confirmed to be stored in the storage lobes of the corpora cardiaca of different cockroach species (Blattidae: Blatta orientalis, Deropeltis erythrocephala, Neostylopyga rhombifolia), as well as related insects groups, including mantids (Epioscopus spec.) and termites (Hodotermes mossambicus) (Fig. 2). LMS was not found in abdominal PSOs of P. americana.
Pea-MIP was detected when searching for an AKH-like peptide in the brain/retrocerebral complex of the American cockroach. Based on spectral analysis and determination of signal ratio of 210/280 nm Trp-containing peptides were sorted out after the first HPLC-separation and further purified on an Inertsil RP-18 column. Fractions, co-eluting in both HPLC-runs with synthetic perikinins (FXSWGamides; 22) were excluded from further examination. Adipokinetic hormones (Pea-CAH-1 and 2) were only found in the extract of the CC/CA. Additionally, several substances with spectra typical of peptides containing more than a single Trp (Fig. 3) were obtained after this second HPLC-step. However, only one fraction of brain and CC/CA extracts contained enough material for structural elucidation, respectively. These substances had similar retention times (24.3 min and 21.1 min in HPLC-step 1 and 2, respectively) which suggested the occurrence of the same peptide in both extracts. The identity of both substances was then confirmed using MALDI-TOF mass spectrometry (monoisotopic [M⫹H]⫹: 1045.4 Da). Mass spectra obtained from this putative peptide showed only weak [M⫹H]⫹ signals but
R. Predel et al. / Peptides 22 (2001) 199 –208
203
HPLC-step 1 and 2, respectively) contained a substance with a mass similar to that of Pea-MIP. Only about 5 pmol of this substance were found after the third HPLC-purification and subjected to Edman-degradation. The sequencing procedure revealed first six amino acids (Gly-Trp-Gln-AspLeu-Gln) which were identical to those of Pea-MIP. This result, together with the mass data obtained from the same material, confirmed the occurrence of Pea-MIP in the abdominal PSOs. 3.4. Immunolocalization of Pea-MIP
Fig. 4. MALDI-TOF mass spectrum of purified Pea-MIP showing distinct [M⫹Na] and [M⫹K] ions, not typical of other neuropeptides of the American cockroach (with adipokinetic hormones being the exception).
distinct [M⫹Na]⫹ and [M⫹K]⫹ peaks (Fig. 4) resembling those of adipokinetic peptides which normally consist of [M⫹Na]⫹ and [M⫹K]⫹ ions only. In addition, mass spectral analysis of the purified material (about 5 pmol/l) revealed a relatively low signal intensity which probably makes the detection in unseparated peptide mixtures difficult. About 50 pmol of the peptide (material from CC/CA extract) was subjected to Edman-degradation and gave the following sequence: Gly-Trp-Gln-Asp-Leu-Gln-Gly-GlyTrp. This sequence was in agreement with mass data and supported a C-terminal amidation. According to the name of the first identified member of this peptide family (LomMIP; 29), the novel peptide was named Pea-MIP. Synthetic Pea-MIP had a retention time identical to that of native material from brain extracts. The retention behavior of PeaMIP in comparison with other known myotropins of the American cockroach is shown in Fig. 1. This figure also illustrates that Pea-MIP belongs to the abundant neuropeptides in the CC/CA of the American cockroach. Using the same purification procedure and mass spectrometric methods it was shown that Pea-MIP is also present in the brain/retrocerebral complex of a blaberoid cockroach, namely Nauphoeta cineraea (not shown). 3.3. Isolation and structural elucidation of Pea-MIP from the abdominal PSOs The neuropeptides of the abdominal PSOs of the American cockroach were thoroughly investigated and structures of abundant substances have been described [25]. All collected fractions of the first HPLC-separation were further purified in a second HPLC-step on a microbore column and, subsequently, the masses of all detectable substances were measured by means of MALDI-TOF mass spectrometry. One of these fractions (at about 24.5 min and 41.5 min in
Using the anti-Pea-MIP serum numerous immunopositive intrinsic cells were found throughout the central nervous system. In this study, however, we focused on cells supplying neurohaemal organs and intestine with Pea-MIP immunoreactive (IR) material. A) Brain/retrocerebral complex: Preabsorption of the Pea-MIP antiserum with Pea-CAH-1 did not abolish any immunostaining. In the CC, only the storage lobes were strongly immunostained, which also indicated that the antiPea-MIP serum did not cross-react with Pea-CAH-1 and 2. No immunostaining was detected, however, after preabsorption with Pea-MIP. Pea-MIP IR material in the storage lobes of the CC (Fig. 5A) originated from neurosecretory cells in the protocerebrum, namely the pars intercerebralis (Fig. 5A) and pars lateralis. These neurons innervated the retrocerebral complex via the nervi corporis cardiaci-1 and 2. Immunopositive fibers were found along the nervi corporis allati-1, run through the CA, and left the retrocerebral complex via the postallatal nerves. B) Abdominal ventral nerve cord: Most prominent PeaMIP immunostaining was detected in large neurons (30 – 40 m) which were located in the dorsal midline of the abdominal ganglia 3– 6 (Fig. 5B), and the terminal ganglion (Fig. 5C). Axons of these neurons from abdominal ganglia 3– 6 projected into the terminal ganglion. Immunostained fibers left the terminal ganglion via the anterior proctodeal nerves which innervate the hindgut. Additionally to the neurons in the dorsal midline, a few smaller cells (15–20 m) were stained in the posterior region of each unfused abdominal ganglion. These cells sent fibers into the segmental nerves. A branch of the segmental nerves of the more anterior abdominal ganglion supplied the abdominal PSOs via the link nerve/transverse nerve with immunoreactive material. Pea-MIP IR fibers have also been detected in the posterior median nerve (Fig. 5B). In the terminal ganglion, transverse nerves 7–9 were immunostained (Fig. 5C). The anterior median nerves of all unfused abdominal ganglia did not contain any immunoreactive material. Immunostaining in the abdominal PSOs consisted of only a few fibers passing these neurohaemal organs and no accumulation of Pea-MIP-like material typical of periviscerokinins [8,23] and pyrokinin-5 [26] was detected. C) Stomatogastric nervous system and intestine: All ganglia of the stomatogastric nervous system, including frontal
204
R. Predel et al. / Peptides 22 (2001) 199 –208
Fig. 5. Pea-MIP-like IR in Periplaneta americana. A: Brain and retrocerebral complex. Numerous immunoreactive cells are visible in the pars intercerebralis (PI) and the posterior protocerebrum. In all neuropiles, as alpha lobes (aL) of the mushroom body and calyces, immunopositive fibers were stained. The storage lobes of the CC, NCA-1, and fibers crossing the CA also revealed MIP-like IR. In addition, fibers in circular muscles of the pharynx, which originate in the stomatogastric nervous system were immunoreactive (arrows). B: Unfused abdominal ganglion 4 with associated anterior PSOs. Large MIPimmunoreactive neurons were stained in the dorsal midline. MIP-immunoreactive fibers are visible in the posterior median nerve, abdominal PSOs, and transverse nerves. C: Terminal ganglion. Four cells in the dorsal midline and numerous lateral positioned neurons are visible. MIP-like IR has been found in segmental nerves and all transverse nerves (TN) of this ganglion. D: Ingluvial ganglion of the stomatogastric nervous system with immunoreactive neurons, and immunopositive fibers in all nerves leaving this ganglion. E: Circular muscles of the foregut with immunoreactive fibers. No Pea-MIP-like IR was detected in longitudinal muscles.
R. Predel et al. / Peptides 22 (2001) 199 –208
205
Fig. 6. Response of an antennal heart preparation to different concentrations of Pea-MIP. A concentration 10-fold higher than the threshold concentration usually produced a total inhibition which lasted longer when the peptide concentration was increased. This preparation was rinsed between the applications only.
ganglion, hypocerebral ganglion, and ingluvial ganglion (Fig. 5D) contained a number of Pea-MIP IR cells. Additionally, immunopositive cells were found on the esophageal nerve between the hypocerebral and ingluvial ganglion. Whole-mount preparations of the foregut revealed that only the circular muscles were innervated by Pea-MIP IR fibers (Fig. 5E). The hindgut is innervated by the anterior proctodeal nerve (see above). Both, the hindgut and foregut, did not contain Pea-MIP IR neurons. 3.5. Muscle efficacy 3.5.1. LMS This peptide was highly effective at inhibiting phasic contractions of all tested visceral muscles, influencing both the amplitude and frequency of contractions. Threshold concentrations (concentrations required to produce a visible change in frequency and/or amplitude), estimated from five muscle preparations, were about 1 ⫻ 10⫺10 M on isolated hindgut and foregut preparations, and 5 ⫻ 10⫺10 M on the semi-isolated heart. The inhibitory effect of LMS on antennal heart preparations of the American cockroach was studied more extensively and will be published in a separate paper (Hertel et al., in prep.). A concentration 10-fold higher than the threshold concentration typically resulted in a total inhibition of phasic activity. This inhibition was reversible after replacement with normal saline. 3.5.2. Pea-MIP Application of synthetic Pea-MIP in different visceral muscle assays always revealed myoinhibitory effects. These effects, however, were obtainable only at high peptide con-
Fig. 7. Parallel recording of two foregut preparations of P. americana. Application of different doses of synthetic Pea-MIP resulted in weak and short-time inhibitory effects. Leucomyosuppressin was added as standard at the end of the test series. The preparation was continuously rinsed with saline, the application of each peptide lasted about five minutes.
centrations. Threshold concentrations were as high as 1 ⫻ 10⫺7 M on isolated hindgut and heart preparations and even the application of 1 ⫻ 10⫺5 M Pea-MIP did not result in a total inhibition in these muscle assays. A more distinct inhibition was found in the antennal heart preparations (threshold at about 2 ⫻ 10⫺9 M, total inhibition at 10⫺8 M) (Fig. 6). On the isolated foregut slightly inhibiting effects were already observed at a threshold of about 5 ⫻ 10⫺9 M but nearly no increase of inhibition was detected after application of higher doses (Fig. 7)
206
R. Predel et al. / Peptides 22 (2001) 199 –208
Fig. 8. Application of different doses of synthetic Pea-AST-7 on a foregut preparation of P. americana. Proctolin was added as standard at the end of the test series. The preparation was continuously rinsed with saline the application of each peptide lasted about five minutes.
3.5.3. Allatostatins
4. Discussion
Members of this peptide family have been reported to inhibit visceral muscle activity in different insects [7,10,31, 34]. In the blaberoid cockroach Leucophaea maderae, allatostatins were highly effective in decreasing phasic activity of the foregut but not active on the hindgut [6], whereas the hindgut of the related Diploptera punctata responded with decreasing activity to allatostatin application [13]. In P. americana, fourteen allatostatins are known from cDNA studies [4], most of these putative messenger molecules were also found to be expressed [24]. Testing a single member of the allatostatin peptides, Pea-Ast-7, we could confirm, that allatostatins are highly effective in inhibiting foregut activity in P. americana (Fig. 8). Other visceral muscles, including hindgut, antennal heart, and heart showed no visible response after application of even 5 ⫻ 10⫺7 M Pea-Ast-7. This allatostatin also did not influence the response of these muscles after application of 10⫺8 M proctolin (not shown).
The only myoinhibitor, which was highly effective in all visceral muscle assays used during this study, was LMS. Action, biologically active core sequence, and distribution of this peptide was extensively investigated in blaberoid cockroaches [1,3,5,9,16,17,18]. LMS seems to be conservative not only in its function as myoinhibitor in insects but also in its sequence which is obviously highly conserved in cockroaches and related insect orders. These properties resemble those of the myostimulatory proctolin, which remained unchanged in structure and function throughout the different insect groups [19]. A second group of potent myoinhibitory peptides are the allatostatins, originally isolated as inhibitors of juvenile hormone synthesis [10,32,35]. As it was found for the cockroach Leucophaea maderae [6], allatostatins (in this study: Pea-AST-7) effectively inhibit spontaneous activity of the foregut of the American cockroach but not those of other visceral muscles. Myoinhibitory efficacy of allatostatins is
R. Predel et al. / Peptides 22 (2001) 199 –208
not correlated with the occurrence of these peptides in visceral muscles. These peptides are also abundant in parts of the intestine other than the foregut [28,38], antennal hearts [24,37], and were (immunocytochemically) also detected in the lateral heart nerve [33]. The third group of inhibitory neuropeptides described here (Pea-MIP) belongs to a peptide family which is already known from locusts, crickets and moths [2,14,15,29]. Interestingly, these peptides, initially isolated using a muscle bioassay [29] were later found to inhibit juvenile hormone biosynthesis in crickets [14] but with threshold concentrations 10-fold higher than those of allatostatins. Immunocytochemical studies in locusts [30], crickets [36], and cockroaches (this paper) always revealed an innervation of the CA via the nervi corporis allati-1. It is not yet clear, if this immunostaining is correlated with regulation of CA activity or only caused by immunoreactive fibers passing the CA. The physiological role of Pea-MIP as myoinhibitor in the American cockroach is questionable. Although present in relatively large amounts in the storage lobes of the CC, it is unlikely that this peptide can be released in quantities sufficient for inhibiting visceral muscle activity. On the other hand, our immunocytochemical studies revealed the presence of Pea-MIP like material in circular muscles of parts of the intestine whereas the routine bioassay, which was used in this study, recorded mainly contractions along the longitudinal axis of the gut. Interruptions in activity of ring muscles are nearly undetectable if longitudinal muscles are still active. Pea-MIP was found not only in neurohaemal release sites of the head, but also in abdominal PSOs. This distribution is unusual among bioactive neuropeptides of the American cockroach since peptides are commonly stored only in one of these organs [25]. The different amounts of Pea-MIP in the CC/CA and abdominal PSO, however, indicated that this peptide may be released as a hormone from the retrocerebral complex but not from abdominal PSO. This was then confirmed by our immunocytochemical results which evidenced that only a few Pea-MIP-IR fibers run through the abdominal PSO. These fibers did not originate from those three median cell groups of each unfused abdominal ganglion known to synthesize all abundant myotropic neuropeptides of the abdominal PSO [8,23,26]. The isolation and identification of LMS and Pea-MIP finalized our efforts to describe the putative myoregulatory neuropeptides from major neurohaemal organs of the central nervous system of the American cockroach.
Acknowledgments We thank Mrs. Virginia Johnson (Protein Technologies Laboratory, Texas A & M University, College Station) for excellent sequence analysis; Mrs. Renate Winkler, Mrs. Erika Krause and Mrs. Ch. Raue (Institut fu¨r Allgemeine Zoologie und Tierphysiologie Jena, Germany) for perfect
207
technical assistance. This study was supported with a postdoc fellowship of the German Academy of Science Leopoldina (LPD 1997, to RP) and grants from Deutsche Forschungsgemeinschaft (Predel 595/1–1,2; Eckert 122/4 – 1).
References [1] Bendena WG, Donly BC, Fuse M, Lee E, Lange AB, Orchard I, Tobe SS. Molecular characterization of the inhibitory myotropic peptide leucomyosuppressin. Peptides 1997;18:157– 63. [2] Blackburn MB, Wagner RM, Kochansky JP, Harrison DJ, ThomasLemont P, Raina AK. The identification of two myoinhibitory peptides, with sequence similarities to the galanins, isolated from the ventral nerve cord of Manduca sexta. Regul Peptides 1995;57:213– 19. [3] Cook BJ, Wagner RM. Comparative effects of leucomyosuppressin on the visceral muscle systems of the cockroach Leucophaea maderae. Comp Biochem Physiol C 1991;99:95–9. [4] Ding Q, Donly BC, Tobe SS, Bendena WG. Comparison of the allatostatin neuropeptide precursors in the distantly related cockroaches Periplaneta americana and Diploptera punctata. Eur J Biochem 1995;234:737– 46. [5] Donly BC, Fuse M, Orchard I, Tobe SS, Bendena WG. Characterization of the gene for leucomyosuppressin and its expression in the brain of the cockroach Diploptera punctata. Insect Biochem Molec Biol 1996;26:627–37. [6] Duve H, Wren P, Thorpe A. Innervation of the foregut of the cockroach Leucophaea maderae and inhibition of spontaneous contractile activity by callatostatin neuropeptides. Physiol Entomol 1995;20:33– 44. [7] Duve H, Thorpe A, Johnsen AH, Maestro JL, Scott AG, East PD. The dipteran Leu-callatostatins: structural and functional diversity in an insect neuroendocrine peptide family. In: Coast, G. M.; Webster, S. M., eds. Recent advances in arthropod endocrinology. Cambridge University Press, 1998:229 –247. [8] Eckert M, Predel R, Gundel M. Periviscerokinin-like immunoreactivity in the nervous system of the American cockroach. Cell Tissue Res 1999;295:159 –70. [9] Fuse M, Bendena WG, Donly BC, Tobe SS, Orchard I. In situ hybridization analysis of leucomyosuppressin mRNA expression in the cockroach, Diploptera punctata. J Comp Neurol 1998;395:328 – 41. [10] Hoffmann KH, Meyering-Vos M, Lorenz MW. Allatostatins and allatotropins: is the regulation of corpora allata activity their primary function? Eur J Entomol 1999;96:255– 66. [11] Holman GM, Cook BJ, Nachman, RJ. Isolation, primary structure and synthesis of leucomyosuppressin, an insect neuropeptide that inhibits spontaneous contractions of the cockroach hindgut. Comp Biochem Physiol C 1986;85:329 –33. [12] Holman GM, Nachman RJ, Wright MS, Schoofs L, Hayes TK, DeLoof A. Insect myotropic peptides. Isolation, structural characterization, and biological activities. In: Menn, J. J.; Kelly T. J.; Masler E. P., eds. Insect neuropeptides. Chemistry, biology and action. ACS Symposium Series:1991:40 –50. [13] Lange AB, Chan KK, Stay B. Effect of allatostatin and proctolin on antennal pulsatile organ and hindgut muscle in the cockroach, Diploptera punctata. Arch Insect Biochem Physiol 1993;24:79 –92. [14] Lorenz MW, Kellner R, Hoffmann KH. A family of neuropeptides that inhibit juvenile hormone biosysnthesis in the cricket, Gryllus bimaculatus. J Biol Chem 1995;270:21103– 8. [15] Lorenz MW, Kellner R, Hoffmann KH. Allatostatins in Gryllus bimaculatus (Ensifera: Gryllidae): new structures and physiological properties. Eur J Entomol 1999;96:267–74.
208
R. Predel et al. / Peptides 22 (2001) 199 –208
[16] Meola SM, Wright MS, Holman GM, Thompson JM. Immunocytochemical localization of leucomyosuppressin-like peptides in the CNS of the cockroach, Leucophaea maderae. Neurochem Res 1991;16:543–9. [17] Nachman RJ, Holman GM, Hayes TK, Beier RC. Structure-activity relationships for inhibitory insect myosuppressins: contrast with the stimulatory sulfakinins. Peptides 1993;14:665–70. [18] Nachman RJ, Giard W, Favrel P, Suresh T, Sreekumar S, Holman GM. Insect myosuppressins stimulate release of the digestive enzyme ␣-amylase in two invertebrates: the scallop Pecten maximus and insect Rhynchophorus ferrugineus. In: Strand F, Beckwith W, Sandman C, editors. Neuropeptides in developing and aging. New York: Annals NY Acad. Sci 1997;814:335– 8. [19] Orchard I, Belanger JH, Lange, AB. Proctolin. A review with emphasis on insects. J Neurobiol 1989;20:470 –96. [20] Predel R, Agricola H, Linde D, Wollweber L, Veenstra JA, Penzlin H. The insect neuropeptide corazonin: physiological and immunocytochemical studies in Blattariae. Zoology 1994;9:35– 49. [21] Predel R, Kellner R, Kaufmann R, Penzlin H, Ga¨de G. Isolation and structural elucidation of two pyrokinins from the retrocerebral complex of the American cockroach. Peptides 1997;18:473– 8. [22] Predel R, Kellner R, Rapus J, Penzlin H, Ga¨de G. Isolation and structural elucidation of eight kinins from the retrocerebral complex of the American cockroach. Regul Peptides 1997;71:199 –205. [23] Predel R, Rapus J, Eckert M, Holman GM, Nachman RJ, Wang Y, Penzlin H. Isolation of periviscerokinin-2 from the abdominal perisympathetic organs of the American cockroach, Periplaneta americana. Peptides 1998;19:801–9. [24] Predel R, Kellner R, Rapus J, Ga¨de G. Allatostatins from retrocerebral complex and antennal pulsatile organ of the American cockroach, Periplaneta americana, aided by matrix-assisted laser desorption/ionization-time-of-flight mass spectrometry. Regul Peptides 1999;82:81–9. [25] Predel R, Eckert M, Holman GM. The unique neuropeptide pattern in abdominal perisympathetic organs of insects. Annals NY Acad Sci 1999;897:282–90. [26] Predel R, Eckert M. Tagma-specific distribution of FXPRLamides in the nervous system of the American cockroach. J Comp Neurol 2000;419:352– 63. [27] Predel R, Nachman RJ, Ga¨de G. Myostimulatory Neuropeptides in Cockroaches: Structures, Distribution, Pharmacological Activities, and Mimetic Analogs (Review). J. Insect Physiol., in press.
[28] Reichwald K, Unnithan GC, Davis NT, Agricola H, Feyereisen R. Expression of the allatostatin gene in endocrine cells of the cockroach midgut. Proc Natl Acad Sci USA 1994;91:11894 – 8. [29] Schoofs L, Holman GM, Hayes TK, Nachman RJ, DeLoof A. Isolation, identification and synthesis of locustamyoinhibiting peptide (Lom-MIP), a novel biologically active neuropeptide from Locusta migratoria. Regul Peptides 1991;36:111–19. [30] Schoofs L, Veelaert J, Vanden Broeck J, DeLoof A. Immunocytochemical distribution of locustamyoinhibiting peptide (Lom-MIP) in the nervous system of Locusta migratoria. Regul Peptides 1996;63: 171–9. [31] Schoofs L, Veelaert D, Vanden Broeck J, De Loof A. Peptides in the locusts, Locusta migratoria and Schistocerca gregaria. Peptides 1997;18:145–56. [32] Stay B, Tobe SS, Bendena WG. Allatostatins: identification, primary structure, functions and distribution. Adv Insect Physiol 1994;25: 267–338. [33] Ude J, Agricola H. FMRFamide-like and allatostatin-like immunoreactivity in the lateral heart nerve of Periplaneta americana: Colocalization at the electron-microscopic level. Cell Tissue Res 1995;282: 69 – 80. [34] Vilaplana L, Maestro JL, Piulachs M-D, Belles X. Modulation of cardiac rhythm by allatostatins in the cockroach Blattella germanica (L.) (Dictyoptera, Blattellidae). J Ins Physiol 1999;45:1057– 64. [35] Weaver RJ, Edwards JP, Bendena WG, Tobe SS. Structures, functions and occurrences of insect allatostatic peptides. In: Coast, G. M.; Webster, S. M., eds. Recent advances in arthropod endocrinology. Cambridge University Press, 3–32; 1998. [36] Witek G, Verhaert P, Lorenz MW, Hoffmann KH. Immunolocalization of two types of allatostatins in the central nervous system of the cricket Gryllus bimaculatus (Ensifera: Gryllidae). Eur J Entomol 1999;96:279 – 85. [37] Woodhead AP, Stoltzman CA, Stay B. Allatostatins in the nerves of the antennal pulsatile organ muscle of the cockroach Diploptera punctata. Arch Insect Biochem Physiol 1992;20:253– 63. [38] Yu CG, Stay B, Ding Q, Bendena WG, Tobe SS. Immunochemical identification and expression of allatostatins in the gut of Diploptera puncata. J Insect Physiol 1993;41:1035– 43.