26 Girardie et al. Archives of Insect Biochemistry and Physiology 46:26–35 (2001) This article originally published in Volume 36 Archives of Insect Biochemistry and Physiology 36:11–23 (1997) Expression of Neuroparsin cDNA in Insect Cells Using Baculovirus Vectors J. Girardie,1* H. Chaabihi,2 B. Fournier,1 M. Lagueux,3 and A. Girardie1 1 Laboratoire de Neuroendocrinologie, URA CNRS 1138, Université Bordeaux I, Talence Cedex, France 2 Protéine Performance, St. Christol-les-Alès, France 3 UPR CNRS 9022 Réponse Immunitaire et Développement chez les Insectes, Strasbourg Cedex, France The cDNA encoding neuroparsin A, a polytropic neurohormone of the locust, Locusta migratoria, was inserted into the genome of Autographa californica nuclear polyhedrosis virus such that transcription was under control of the p10 promoter. A polypeptide having the same charge and the same apparent molecular weight as the authentic neuroparsin A and that was reactive against neuroparsin immune serum was produced in recombinant virus-infected lepidopteran cell lines but not in control virus-infected cells. The baculovirus-expressed polypeptide was purified by two steps of liquid chromatography (anion exchange and reversed phase) which were previously used to purify the natural neuroparsin. The purified baculovirus-expressed polypeptide enhanced fluid reabsorption of everted rectum preparations, as did the natural neuroparsin. Thus, this gene expression system produced a polypeptide identical to authentic neuroparsin. Arch. Insect Biochem. Physiol. 36:11–23, 1997. © 1997 Wiley-Liss, Inc. Key words: insect; Locusta migratoria; neuroparsin; gene expression; baculovirus-insect cells INTRODUCTION Neuroparsin is a polytropic neurohormone with a distinctive amino acid sequence isolated from the neural lobes of the corpora cardiaca of Locusta migratoria (Girardie et al., 1989, 1990). It is produced as an 83 residue polypeptide (neuroparsin A, molecular mass = 8,759 daltons, calculated pl = 6.71) by the A1-type cell bodies of the pars intercerebralis (Bourême et al., 1987) and is partially enzymatically transformed into a less basic form of 78 residues (neuroparsin B, molecular mass = 8,188 daltons, calculated pl = 5.98) dur© 2001 Wiley-Liss, Inc. ing storage at the axonal endings located in the corpora cardiaca (Girardie et al., 1987b). Both Acknowledgments: We are grateful to Mrs. Sophie Daburon and Miss Catherine Cêtre for their technical assistance. The authors gratefully thank Dr. Thomas Durkin for linguistic review of the manuscript. This work was partly supported by the Conseil Régional d’Aquitaine, Pôle Médicament. *Correspondence to: J. Girardie, Laboratoire de Neuroendocrinologie, URA CNRS 1138, Avenue de Facultés, Université Bordeaux I, 33405 Talence Cedex, France. E-mail: firstname.lastname@example.org Received 29 October 1996; Accepted 8 February 1997. Baculovirus-Expressed Neuroparsin forms (A and B) trigger antijuvenile (Girardie et al., 1987a), antidiuretic (Fournier and Girardie, 1988), hypertrehalosemic, and hyperlipemic (Moreau et al., 1988) effects in the locust. Only the antidiuretic effect is well studied at the target organ (Girardie and Fournier, 1993; Fournier et al., 1994; Nogaro et al., 1995). A large quantity of neuroparsin is needed to characterize its receptor(s). The relative simplicity of the neuroparsin cDNA sequence, composed only of the neurohormone A and 22 residue signal peptide (Lagueux et al., 1992), makes the synthesis of neuroparsin in a gene expression system feasible. To our knowledge, only four insect neurohormones — the diuretic hormone of tobacco hornworm, the eclosion hormone and the prothoracotropic hormone of silkworm, and the pheromone biosynthesis activating neuropeptide of corn earworm — have been successfully produced by recombinant DNA techniques. Three expression systems have been used: a vector plasmid expressed in Escherichia coli (Kono et al., 1990), a shuttle plasmid expressed in yeast (Hayashi et al., 1990), and a baculovirus vector expressed either in whole insect (Maeda, 1989; O’Reilly et al., 1995) or in insect cells (Eldridge et al., 1991; Vakharia et al., 1995). Each of the expression systems led to the production of biologically active neurohormones. The levels of expression reported were different and were dependent both on the neurohormone and the particular expression system. In the present study, we have attempted to express a cloned neuroparsin cDNA in insect cells using a baculovirus vector. The expression system produced a protein whose charge, apparent molecular weight, and reactivity to neuroparsin immune serum were similar to those of authentic neuroparsin A. This protein, purified by two steps of liquid chromatography used for the natural neuroparsin, triggered a similar antidiuretic activity as the authentic neuroparsin. MATERIALS AND METHODS Generation of Recombinant Baculoviruses Expressing Neuroparsin cDNA Full-length neuroparsin cDNA was purified from plasmid pTZ-neurocDNA (Lagueux et al., 1992) as a 360 bp BsaH I–Pvu II fragment. The BsaH I site is located 12 nucleotides upstream of 27 the ATG start codon of neuroparsin gene, and the Pvu II site is located 22 nucleotides downstream of the stop codon. Following blunt-ending with Klenow polymerase, the BsaH I–Pvu II fragment was ligated to plasma pGm16 (Blanc et al., 1993) previously digested with Bgl II and blunt-ended. The transfer vector pGm16 bears the p10 promoter followed by a Bgl II site, for cloning of foreign genes, and baculovirus-flanking sequences for recombination at the p10 locus. Spodoptera frugiperda Sf9 cells (CRL 1711, ATCC; Rockville, MD) were cotransfected with DNA of the generated vector carrying the neuroparsin gene under control of the p10 promoter (pGm16-NP) and genomic DNA of the baculovirus AcSLP10 (Chaabihi et al., 1993). For this cotransfection, the lipofection method (Felgner and Ringold, 1989) was followed using the DOTAP reagent (BoehringerMannheim, Germany). Because AcSLP10 is a modified Autographa californica polyhedrosis virus, which possesses only one strong late promoter (the p10) with the polyhedrin coding sequence inserted downstream, unrecombined AcSLP10 has an occlusion body-positive phenotype. Consequently, screening and purification of recombinant baculoviruses were carried out by plaque assay as described by Summers and Smith (1987). Three selected recombinant viruses expressing the neuroparsin gene under p10 promoter control were used in the present study and designated bPP1-08-1111, bPP1-08-2111, and bPP1-08-5111. Production of Neuroparsin in Insect Cells Sf9 cells were grown in 25 cm2 Falcon flasks with 5 ml of TC 100 medium (Gibco BRL, Gaithersburg, MD) supplemented with 5% fetal calf serum. Cell cultures with a density of 4 × 106 cells per flask were infected with either one of the three recombinants (bPP1-08-1111, bPP1-08-2111, bPP108-5111) or with AcSLP10 as control (two flasks for each virus type). Sf9 cells were also grown in 75 cm2 Falcon flasks with 15 ml of medium, and cell cultures with a density of 20 × 106 cells per flask were infected with either the recombinant bPP1-08-1111 (three flasks) or with AcSLP10 as control (one flask). Infected cells were harvested 4 days postinfection by centrifugation at 3,000 rpm (Jouan CR 312) for 10 min at 4°C. Cell pellets were washed twice with phosphate buffered saline. Cell 28 Girardie et al. pellets and the clarified supernatant (culture medium) were stored at –80°C until used. Extraction of the Recombinant Neuroparsin Extraction of the putative recombinant neuroparsin was attempted from both clarified supernatants (culture medium) and pellets of cells infected with the three recombinant viruses. Clarified supernatants and pellets of cells infected with the ACSLP10 control virus were treated in the same way. All supernatants were dialyzed overnight at 4°C against distilled water using 6,000 dalton cutoff tubing since neuroparsin is conserved using this procedure (Fournier and Girardie, 1988). Even after dialysis, the culture medium was sometimes not clear (chiefly for the largest volume), and insoluble material was then separated by centrifugation at 3,000 rpm. Soluble and insoluble fractions were separately freeze-dried and stored at –80°C until extraction. Two extraction media were tested both on cell pellets and on soluble and insoluble freeze-dried dialyzed supernatants: Tris-HCl buffer (sample buffer of PAGE or SDS-PAGE or solvent of anion exchange chromatography) and 70% methanol (extraction medium or natural neuroparsin). Proteins extracted by the buffer were analyzed directly by electrophoresis or chromatography, while proteins extracted by methanol were analyzed following total evaporation of the extraction medium and redissolution in adequate buffers. Electrophoresis Analysis The extracted proteins were analyzed by either 7.5% native PAGE according to Laemmli (1970) or 12% SDS-PAGE according to Schägger and Von Jagow (1987). Electrophoresis was performed with the Bio-Rad Mini-Protean II dual Slab Cell (Bio-Rad, Richmond, CA) at constant voltage according to the manufacturer’s protocols. Gels were either stained in a fixative of 50% methanol solution containing 10% acetic acid and 0.1% Coomassie brilliant blue R250 or directly electrotransferred for immunodetection with antineuroparsin immune serum. Western Blotting Procedures Following native or 12% SDS-PAGE, proteins were transferred onto Immobilon-P (Millipore, St. Quentin, Yvelines, France) using the Bio-Rad Mini Trans-Blot Electrophoretic Transfer Cell according to the manufacturer’s protocols (25 mM Trisglycine buffer, pH 8.3, with 20% v/v methanol). The transfer was carried out at a constant voltage (100 v) and a current of 180 mA at the beginning and 250 mA at the end. The efficiency of sample transfer was evaluated by Coomassie blue staining of the gel. Immunostaining Blots were treated for 30 min at room temperature in buffer A (0.2 M Tris-HCl buffer, pH 7.2, 3% bovine serum albumin [BSA]) to saturate nonspecific reactive sites, followed by a 16 h incubation at 4°C with anti-neuroparsin serum (1/ 2,000) in the same buffer. After three washes, blots were incubated for 2 h at room temperature with a peroxidase-labelled goat anti-rabbit Fab (1/1,000) in buffer A. Following buffer washes, the peroxidase reaction was carried out for 5 min at room temperature with 0.05% diaminobenzidine as chromogen (DAB; Sigma, St. Quentin Fallavier, France) and 0.06% hydrogen peroxidase as substrate. Purification of the Recombinant Neuroparsin Anion-exchange liquid chromatography. The extracted proteins were injected onto a Mono Q HR 5/5 column (Pharmacia, Gaithersburg MD) and a Beckman (Fullerton, CA) gradient system (model 333) with a variable wavelength detector (model 165). A 33 min linear gradient from 0–50% of 1 M NaCl and then a 2 min linear gradient from 50–100% were employed at a flow rate of 1 ml/min at room temperature. Fractions were collected manually. Reversed-phase chromatography, desalting, and purification. Fractions collected from anion-exchange chromatography were loaded onto a reversed-phase C8 Pro RPC 5/10 column (Pharmacia) and eluted with a linear gradient from 28–50% of CH3CN in 25 min at a flow rate of 0.5 ml/min. Peak fractions were collected manually and tested for antidiuretic activity. In Vitro Rectal Test Antidiuretic activity of the putative neuroparsin peak was measured using the everted rectal sac preparation as previously described Baculovirus-Expressed Neuroparsin (Fournier and Girardie, 1988). Following a prior incubation of 1 h in chloride-free standard saline for equilibration, recta were emptied and refilled with 10 µl of standard saline (Hanrahan and Phillips, 1982) containing either natural neuroparsin or putative biosynthetic neuroparsin or no neuroparsin. The biological response (fluid reabsorption) was then measured after a 1 h incubation. RESULTS Electrophoretic Analysis Natural native neuroparsins A and B are unique polypeptide chains having 12 cysteine residues. Neuroparsin A (calculated pl = 6.71) and neuroparsin B (calculated pl = 5.98) can be easily distinguished by their charge using native 7.5% PAGE (Girardie et al., 1987a). In spite of the 12 cysteines forming disulfide bonds, denaturation of neuroparsin using SDS induces self-association of two, three, or four molecules (Girardie et al., 1985; Girardie and Fournier, 1993). In the presence of a reducing agent (β-mercaptoethanol), the monomer is the major form and often the only one visualized when Coomassie blue is used to stain the gels (Fig. 3A). However, when sensitive methods of detection are used, such as immunodetection, the monomeric form is always the major one; however, self-association of molecules is generally observed (Fig. 1A). Extracts of cells and their culture media upon electrophoresis separated into several protein bands following the Coomassie blue staining (Fig. 3). Some of these protein bands had the same electrophoretic mobility as natural neuroparsin. However, only specific detection with the antineuroparsin immune serum was capable of confirming the expression of either neuroparsin or neuroparsin-like molecules by the baculovirus-cell system. Whatever the extraction medium and the type of electrophoretic separation used, protein mixtures from equivalent amounts of either media or cells infected by the three distinct recombinant viruses (1111, 2111, 5111) contained immunoreactive band(s), while no band was detected from the culture medium or the cells infected by the control virus (Fig. 1). Following 7.5% PAGE, only one immunoreactive band, having the same charge as neuroparsin A, was detected both in cell and me- 29 dium extracts but chiefly in cell extracts (Fig. 2). This result indicates that the translation product of the neuroparsin cDNA composed of a 22 residue peptide signal containing two basic arginins and neuroparsin A is rapidly processed into neuroparsin A in the cells. A more acidic band (putative neuroparsin B) was never detected either in the cell or in the medium extracts. Following SDS-PAGE, pure natural neuroparsin A was detected for the most part as a monomer (the lowest molecular weight band) and also as two self-associated forms: a dimer and tetramer (Fig. 1A). In cell extracts, several slightly immunostained bands were sometimes visualized (cross-reactivity), and two bands were strongly immunostained. One band (the lowest molecular weight) had the same apparent molecular weight as the monomeric natural neuroparsin A, and the other appears to comigrate with the tetramer (Fig. 1C). In extracts of the medium, one or two reactive bands were immunodetected (Fig. 1M). The lowest molecular weight band had the same apparent molecular weight as the dimer of natural neuroparsin. The other band, which was always less reactive, had an apparent molecular weight between that of the monomer and the tetramer and could represent self-association of three molecules (trimer) of neuroparsin (Fig. 1M). The quantity of expressed neuroparsin extracted from cells, evaluated by the intensity of its immunoreactivity, appeared to be higher than the quantity of neuroparsin contained in the culture medium of the cells (Figs. 1, 2), indicating that neuroparsin A is either mainly present in cells or is more easily extracted from the cells. SDS-PAGE of pellets, obtained with both extraction solutions from either media or cells and reextracted with the sample buffer, showed numerous bands, but a band having the same apparent molecular weight as neuroparsin was observed only in the pellet of extracted cells (Fig. 3, lanes 1–4). This band was also immunostained with anti-neuroparsin immune serum (data not shown). Thus, it appears that the expressed neuroparsin is localized mostly in cells rather than in the culture medium. Moreover, this result also indicates that the extraction of the expressed neuroparsin was not complete. No significant difference in the quantity of extracted neuroparsin was observed between the 30 Girardie et al. Fig. 1. Immunoblot analysis of natural neuroparsin A (A) purified from neural corpora cardiaca from 50 locusts and of proteins extracted by 70% methanol from 106 cells (C) and from the entire volume of the corresponding culture medium (M) following infection by either 0111 AcSLP10 control virus (lane 0) or 1111 (lane 1), 2111 (lane 2), and 5111 (lane 5) AcSLP10 recombinant viruses. Following 12% SDS-PAGE, proteins were electroblotted and probed with anti-neuroparsin immune serum. Three bands were revealed for the natural neuroparsin (A): the lowest form (major band = true monomeric neuroparsin) and two heavier forms induced by SDS (dimer and tetramer). Bands were present in extracts of cells and of their culture media only when cells were infected with recombinant viruses (C, lanes 1, 2, 5; M, lanes 1, 2, 5). In cells (C, lanes 1, 2, 5), the most immunoreactive band (the lowest molecular weight and the largest) had the same apparent molecular weight as the true monomeric neuroparsin. Bands corresponding to dimeric and tetrameric neuroparsin were not easily distinguishable among several slightly stained bands revealed by cross-reactivity. In media (M, lanes 1, 2, 5) two bands were revealed. The most immunoreactive and the largest band had the same apparent molecular weight as the SDS-induced dimer of neuroparsin. The other band, heavier than the dimer and lighter than the tetramer, probably corresponds to the SDS-induced trimeric neuroparsin. Fig. 2. Immunoblot analysis of natural neuroparsin A (lane A) purified from 20 corpora cardiaca and of proteins extracted by Tris-HCl from 2 × 106 cells infected with the 1111 recombinant virus and from the entire volume of the corresponding culture medium (lane M). Following 7.5% PAGE, proteins were electroblotted and probed with anti-neuroparsin immune serum. Only one band was revealed clearly in cells (lane C) and poorly in medium (lane M). This band had the same charge as natural neuroparsin A. Fig. 3. Effect of the SDS-PAGE sample buffer on the 70% methanol insoluble proteins (pellet) from cells infected with the 5111 recombinant virus and their culture medium. Following 12% SDS-PAGE, proteins were stained with Coomassie blue. The sample buffer extracted numerous proteins from the cell pellet as well as the medium pellet, giving an idea of the number of total protein bands present but not immunodetected on the Western blot of the Fig. 1. A protein (the lowest) having a similar apparent molecular weight to neuroparsin A was present only in the cell pellet. A: Neuroparsin A purified from neural corpora cardiaca from 50 locusts. C: (lanes 4, 2, 1): Three serial half dilutions of protein extracts from cells. M: (lanes 8, 4): Two serial half dilutions of protein extracts from culture medium. Baculovirus-Expressed Neuroparsin three recombinant viruses. Quantitative variations between distinct batches of the same recombinant virus were sometimes more evident than between the three recombinant viruses (data not shown). The level of neuroparsin expression was evaluated on immunoblots by measuring the optical density of the major band obtained with expressed neuroparsin and comparing it to that of monomeric natural neuroparsin. The band intensity of monomeric neuroparsin expressed by 106 infected insect cells was always around half as much as the band intensity of natural monomeric neuroparsin purified from neural corpora cardiaca of 50 locusts (Figs. 1, 2). So, the level of neuroparsin expression was quite low, at most 3 µg and even lower per 106 infected cells since the neural corpora cardiaca of one locust contains about 120 ng of neuroparsin A. Chromatographic Purification Proteins extracted from the culture media from either the experimental recombinant virus or the control virus transfected cells were fractionated by anion-exchange chromatography and were eluted as very poorly defined peaks of very high UV absorbance irrespective of the two solvents used (buffer or 70% methanol). No difference was observed between the chromatograms of proteins extracted from the media of either the experimental or control cells. Because of this result and those of the electrophoretic analysis, no further attempt was made to purify neuroparsin from media. Proteins extracted from the experimental and control infected cells eluted as sharp peaks in anion-exchange chromatography. However, the profile obtained with each of the two extraction media was somewhat different with respect to the number of peaks and the intensity of comparable peaks. The most surprising observation was the presence of a peak having the same elution time (same concentration of NaCl) as neuroparsin A in extracts of cells infected either by recombinant or control viruses (Fig. 4A). Consequently, fractions eluted at the time of putative neuroparsin (7–8 min), 2 min before, and 2 min after were reinjected separately onto the reversed-phase column. Only the fractions originating from cells infected with recombinant virus and eluted just 31 before (data not shown) and chiefly at the time of neuroparsin A gave a sharp major peak. This peak was eluted at the same time (6–7 min) as natural neuroparsin A (Fig. 4B). A comparison of the area of peaks with reversed-phase peaks of natural neuroparsin A indicated that 20 × 106 infected cells were equivalent to about 200 locusts, and consequently 106 infected cells were equivalent to ten locusts. So, the level of expression of purified neuroparsin was approximately 1.2 µg of expressed neuroparsin per 106 infected cells. Biological Activity The HPLC fraction eluted at the time of the putative neuroparsin and originating from cells infected by the control virus, the purified expressed neuroparsin, and natural neuroparsin were tested in parallel for their antidiuretic activity. The fraction eluted from cells infected with control virus was tested without dilution even in the absence of the UV peak in the fraction. Since we anticipated that the biological activity of the expressed neuroparsin would be somewhat lower than that of natural neuroparsin, we used three serial half dilutions, all of which induced maximal effects in the case of natural neuroparsin (plateau of the dose-response curve in Fournier and Girardie, 1988). The fraction from the cells infected by the control virus did not increase the spontaneous rectal fluid reabsorption measured in controls, whereas both neuroparsins increased it to a similar degree (Fig. 5). However, because the conditions of bioassay were not quantitative, we can only deduce that the expressed neuroparsin is biologically active, like authentic neuroparsin, within the limitations of the quantification used. Nevertheless, the existence of a maximal effect with each of the three serial half-dilutions supports our estimation of the quantity of purified expressed neuroparsin. DISCUSSION Using the plasmid pTZ-neuroparsin fulllength cDNA (Lagueux et al., 1992), we generated vectors carrying the neuroparsin gene under the control of the p10 promoter of AcSLP 10 virus for infecting Sf9 insect cells. Three recombinant viruses expressing the neuroparsin gene were selected and were observed to produce a bio- 32 Girardie et al. Fig. 4. A: Anion-exchange liquid chromatography profiles of 25 mM Tris-HCl (pH 8) extracts of 10 × 106 cells infected by either the 0111 control virus (CV) or the 1111 recombinant virus (RV). Extracts were applied onto a Pharmacia Mono Q HR 5/5, 10 mm; buffer A was 25 mM Tris-HCl, pH 8; buffer B was 1 M NaCl in buffer A. Elution was conducted by linear gradient from 0–0.5 M NaCl for 33 min at a flow rate of 1 ml/min. Detection was monitored by absorbance at 225 nm. No striking differences were observed between the two profiles, and in both cases a peak (arrow) was eluted at the elution time of neuroparsin A (7–8 min). B: Reversed-phase liquid chromatography profile of the fraction eluted between 7 and 8 min during anion-exchange liquid chromatography of the extracts of 20 × 106 cells infected either with the recombinant virus (RV) or with the control virus (CV). The fraction was applied onto a ProRPC C1/C8 300 Å column (5 × 10 mm); buffer A was 0.08% TFA in 90% water and 10% CH3CN; buffer B was 0.08% TFA in CH3CN. Elution was conducted by linear gradient from 22–54% B for 25 min at a flow rate of 0.5 ml/min. Detection was monitored by absorbance at 220 nm. A sharp peak (arrow) was eluted at the same time as natural neuroparsin (8–9 min) only in the fraction originating from cells infected with the recombinant virus. Baculovirus-Expressed Neuroparsin 33 Fig. 5. Fluid reabsorption of everted rectum preparations with the incubation medium alone (C) and the same medium complemented with the total 8–9 min HPLC fraction from cells infected with the control virus (CV) or three increasing doses (0.5, 1, and 2 corpora cardiaca equivalents) of either the 8–9 min HPLC fraction from cells infected with the recombinant virus (sharp peak) or purified natural neuroparsin. Data are expressed as the mean +/– SE. The number of recta tested is given in parentheses. No stimulation was obtained with the fraction from cells infected with the control virus. Similar stimulation was obtained with either authentic or baculovirus-expressed neuroparsins. synthetic molecule having the same charge, the same apparent molecular weight, the same denaturation-induced self-associations, and the same reactivity against neuroparsin immune serum as the authentic neuroparsin. The baculovirus-expressed molecule was isolated by the same two steps of liquid chromatography previously used to purify the natural neuroparsin (Girardie et al., 1989). The purified molecule increased in vitro rectal water reabsorption as did the natural neuroparsin (Fournier and Girardie, 1988). Thus, a neuroparsin which was identical in structure and function to natural neuroparsin was successfully expressed in the baculovirus-insect cell system. Baculovirus-expressed neuroparsin was detected mainly in the infected cells but not in the culture medium, in contrast to the eclosion hormone (Eldridge et al., 1991) and the pheromone biosynthesis activating peptide (Vakharia et al., 1995) which, in such expression systems, were rapidly secreted from the cells. However, the expressed neuroparsin appeared to be stable in the medium as well as in the cells, since no immunoreactive molecule with either a charge less basic or an apparent molecular weight lower than the monomeric neuroparsin A was detected. Thus, the cleavage of neuroparsin A into a less basic neuroparsin B, which occurs during storage in the axonal endings of the neurosecretory cells of Locusta, did not occur in the expression system. The signal peptide of neuroparsin has two arginines which normally might induce a more basic net charge of the translation product than neuroparsin A. Since the net charge of the translation product extracted either from the cells or from the medium was the same as neuroparsin A, it appears that the signal peptide of the translation product was correctly cleaved intracellularly. This was also the case for the purified eclosion hormone expressed in yeast (Hayashi et al., 1990) but not for all the forms of the purified Escherichia-expressed eclosion hormone, some of them having molecular weights slightly lower than the natural neurohormone (Kono et al., 1990). In addition, recombinant neuroparsin A seems to be correctly folded since it was biologically active. Although a portion of baculovirus-produced neuroparsin was secreted into the culture medium, cells appeared to be the most abundant source of neuroparsin for purification because of the absorbance of the culture medium and its volume. The quantity of purified expressed neuroparsin per 106 cells was approximately equal to the quantity of natural neuroparsin that can be purified from ten locusts. It corresponds to one of the lower levels (about 1 µg per 106 cells) reported for other heterologous proteins synthesized in insect cells using the baculovirus expression vector system (Luckow and Summers, 1988; Miller, 1988). Concerning the biologically estimated yield of the two unpurified insect neurohormones ex- 34 Girardie et al. pressed in a baculovirus vector, the amount of the purified expressed neuroparsin was low compared to that of the eclosion hormone (Eldridge et al., 1991) but similar to that of the pheromone biosynthesis activating neuropeptide (Vakharia et al., 1995). The observed poor secretion of expressed neuroparsin might be due to a perturbation of a certain stage of the secretory pathway. The use of alternative expression systems in order to obtain a better level of expression will require further investigation. LITERATURE CITED Blanc S, Cerutti M, Chaabihi H, Louis C, Devauchelle G, Hull R (1993): Gene II product of an aphid-nontransmissible isolate of cauliflower mosaic virus expressed in a baculovirus system possesses aphid transmission factor activity. Virus 192:651–654. Bourême D, Tamarelle M, Girardie J (1987): Production and characterization of antibodies to neuroparsin A and B isolated from the corpora cardiaca of the locust. Gen Comp Endocrinol 67:169–177. Chaabihi H, Ogliastro MH, Martin M, Giraud C, Devauchelle G, Cerutti M (1993): Competition between baculovirus polyhedrin and p10 gene expression during infection of insect cells. J Virol 67:2664–2671. Eldridge R, Horodyski FM, Morton DB, O’Reilly DR, Truman JW, Riddiford LM, Miller LK (1991): Expression of an eclosion hormone gene in insect cells using baculovirus vectors. Insect Biochem 21:341–351. Felgner PL, Ringold GM (1989): Cationic liposome-mediated transfection. Nature 337:387–388. Fournier B, Girardie J (1988): A new function for the locust neuroparsins stimulation of water reabsorption. J Insect Physiol 34:309–313. Fournier B, Guérineau N, Mollard P, Girardie J (1994): Effects of two neuronal antidiuretic molecules, neuroparsin and 5-hydroxytryptamine, on cytosolic free calcium monitored with indo-1 in epithelial and muscular cells of the African locust rectum. BBA 1220:181–187. Girardie J, Fournier B (1993): Chemistry and physiology of neuroparsin: an antidiuretic neurohormone of the African locust. In Beyenbach KW (ed): Structure and Function of Primary Messengers in Invertebrates: Insect Diuretic and Antidiuretic Peptides. Vol. 12. Basel: Karger, pp 67–83. Girardie J, Bourême D, Couillaud F, Tamarelle M, Girardie A (1987a): Anti-juvenile effect of neuroparsin A, a neuroprotein isolated from locust corpora cardiaca. Insect Biochem 17:977–983. Girardie J, Bourême D, Girardie A (1987b): Production sites of the three neurosecretory proteins characterized in the corpora cardiaca of the migratory locust. Insect Biochem 17:29–36. Girardie J, Faddoul A, Girardie A (1985): Characterization of three neurosecretory proteins from the A median neurosecretory cells of Locusta migratoria by coupled chromatographic, electrophoretic and isoelectrofocusing methods. Insect Biochem 15:85–92. Girardie J, Girardie A, Huet JC, Pernollet JC (1989): Amino acid sequence of locust neuroparsins. FEBS Lett 245:4–8. Girardie J, Huet JC, Pernollet JC (1990) The locust neuroparsin A: Sequence and similarities with vertebrate and insect polypeptide hormones. Insect Biochem 20:659–666. Hanrahan JW, Phillips JE (1982): Electrogenic, K-dependent chloride transport in locust hindgut. Philos Trans R Soc Lond Biol 299:585–595. Hayashi H, Nakano M, Shibanaka Y, Fujita N (1990): Expression of a silkworm eclosion hormone gene in yeast. Biochem Biophys Res Commun 173:1065–1071. Kono T, Nagasawa H, Kataoka H, Isogai A, Fugo H, Suzuki A (1990): Eclosion hormone of the silkworm Bombyx mori. Expression in Escherichia coli and location of disulfide bonds. FEBS Lett 263:358–360. Laemmli UK (1970): Cleavage of structural protein during the assembly of the head of bacteriophage T4. Nature 227:680–685. Lagueux M, Kromer E, Girardie J (1992): Cloning of a Locusta cDNA encoding neuroparsin A. Insect Biochem Mol Biol 22:511–516. Luckow VA, Summers MD (1988): Trends in the development of baculovirus expression vectors. Biotechnology 6:47–55. Maeda S (1989): Increased insecticidal effect by a recombinant baculovirus carrying a synthetic diuretic hormone gene. Biochem Biophys Res Commun 165:1117–1183. Miller LK (1988): Baculovirus as gene expression vectors. Annu Rev Microbiol 42:177–199. Moreau R, Gourdoux L, Girardie J (1988): Neuroparsin: A new energetic neurohormone in the African locust. Arch Insect Biochem Physiol 8:135–145. Nogaro M, Radallah D, Guillon G, Fournier B (1995): HPLC analysis of inositol phosphate isomers induced by neuroparsin in rectal epithelial cells of the African locust: Role of external calcium. Insect Biochem Mol Biol 25:559–567. O’Reilly DR, Kelly TJ, Masler EP, Thyagaraja BS, Robson RM, Shaw TC, Miller LK (1995): Overexpression of Baculovirus-Expressed Neuroparsin Bombyx mori prothoracicotropic hormone using baculovirus vectors. Insect Biochem Mol Biol 25:475–485. Schägger H, Von Jagow G (1987): Tricine-sodium dodecylsulfate–polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal Biochem 166:368–379. Summers MD, Smith GE (1987): A Manual of Methods for 35 Baculovirus Vectors and Insect Cell Culture Procedures. Texas Agricultural Experiment Station Bulletin 1555. College Station TX: Texas Agricultural Experiment Station. Vakharia VN, Raina AK, Kingan TG, Kempe TG (1995): Synthetic pheromone biosynthesis activating neuropeptide gene expression in a baculovirus expression system. Insect Biochem Mol Biol 25:583–589.