Journal of Insect Physiology 44 (1998) 859–866

Mechanism of parasitism-induced elevation of haemolymphgrowth-blocking peptide levels in host insect larvae

(Pseudaletia separata)

Yoichi Hayakawaa,*, Atsushi Ohnishia, Yasuhisa Endob

a Institute of Low Temperature Science, Hokkaido University, Sapporo 060, Japanb Department of Applied Biology, Kyoto Institute of Technology, Matsugasaki, Sakyo, Kyoto 606, Japan

Received 1 October 1996; accepted 25 October 1997

Abstract

Growth-blocking peptide (GBP) has been purified for the first time from the haemolymph of the host armywormPseudaletiaseparatawhose growth is inhibited and shows developmental arrest in the last larval instar stage when parasitized by the parasitoidwaspCotesia kariyai. GBP naturally occurs in the haemolymph of lepidopteran larvae but its concentration is very low during thelast larval instar in comparison with that in the penultimate larval instar. However, by 24 h after parasitization or polydnavirus(PdV)-infection on day 0 of the last larval instar, a four-fold increase in GBP level, compared with synchronous non-parasitizedcontrol larvae, is observed. Although Northern blot analysis indicates that GBP mRNA is transcribed in brain-nerve cord and fatbody, plasma GBP is likely to be secreted mainly from fat body because the GBP mRNA level is approximately 100-fold higherin fat body than that in brain-nerve cord. RT-PCR analysis demonstrates the constant expression of GBP mRNA in both parasitized(or PdV-infected) and non-parasitized larval fat body, which suggests that parasitism does not influence transcriptional level, butmight influence post-transcriptional level to elevate plasma GBP concentration. This interpretation was supported by estimatingGBP precursor levels in fat body of PdV-infected and non-infected larvae. Virus infection appears to elevate the GBP precursorlevels in fat body to about six times greater than that in non-infected last instar larvae by 6 h after PdV-injection. The GBPprocessing enzyme activity that occurs in Golgi body-rich extract of the fat body is increased by about 90% after parasitization orPdV-injection. 1998 Elsevier Science Ltd. All rights reserved.

Keywords:BSA, bovine serum albumin; FPLC, fast protein liquid chromatography; GBP, growth-blocking peptide; HPLC, high performance liquidchromatography; PBS, phosphate-buffered saline; PdV, polydnavirus; RT-PCR, reverse transcriptase-polymerase chain reaction; TFA, trifluo-roacetic acid

1. Introduction

Endoparasitoid wasps often disrupt the developmentof holometabolous host insects. In many cases, the hostsare developmentally arrested in the larval stages and failto pupate (Beckage, 1985; Lawrence, 1986). Last instarlarvae of the armywormPseudaletia separataparasit-ized by the parasitoid waspCotesia kariyai do notinitiate metamorphosis and, ultimately, the wasp larvaeemerge from the host larvae about 11 days after parasit-ization (Tanaka et al., 1987). Previous studies revealedthat parasitism elevates the level of a biogenic peptide,

* Corresponding author.

0022–1910 /98 /$19.00 1998 Elsevier Science Ltd. All rights reserved.PII: S0022-1910 (98)00017-1

growth-blocking peptide (GBP), in the haemolymph ofthe host larvae. Injection of GBP into non-parasitizedarmyworm larvae during the last larval instar retards lar-val growth and causes a delay in pupation for more thana few days through repression of plasma juvenile hor-mone (JH) esterase activity (Hayakawa, 1990, 1991).Based on these results, it was proposed that GBP mightbe a hormone-like peptide that coordinates, along withJH, the regulation of larval characteristics in lepidop-teran insects (Hayakawa, 1992, 1995; Noguchi et al.,1994). This hypothesis has been substantially supportedby quantifying plasma GBP concentrations during thelate stages of larval development of parasitized and non-parasitized armyworms (Ohnishi et al., 1995). PlasmaGBP levels in non-parasitized penultimate instar larvae

860 Y. Hayakawa et al. /Journal of Insect Physiology 44 (1998) 859–866

are much higher than in last instar larvae. The low levelof plasma GBP concentration usually persists throughoutthe last larval instar stage, but parasitization causes anincrease in plasma GBP concentration, even in the lastlarval instar.

Recently, the sequence of the GBP cDNA in the army-worm has been demonstrated (Hayakawa et al., 1995)and that GBP is first synthesized as a polyprotein precur-sor. Northern blot analysis with the use of the clonedcDNA as a probe, indicated that there was no differencein the levels of GBP mRNA in parasitized and non-para-sitized larvae, although an increase of GBP concen-tration was observed in the plasma after parasitization(Ohnishi et al., 1995). The present study was conductedto examine the site and kinetics of GBP gene expressionin order to clarify the mechanism by which parasitizationelevates plasma GBP levels.

2. Materials and methods

2.1. Animals and chemicals

P. separatawere reared on an artificial diet at 25±1°C with a photoperiod of 16 h light:8 h dark. Parasitiz-ation by C. kariyai was carried out by exposing day 0last instar larvae ofP. separatato female wasps. Aftera single oviposition was observed, the parasitized hostlarva was quickly removed in order to avoid super-paras-itization and then reared on the artificial diet. Adultwasps were maintained with honey diluted with water.Penultimate instar larvae undergoing ecdysis between 2and 2.5 h after the lights were on, were designated asday 0 last instar larvae.

Radioactively-labelled reagents were obtained fromNEN Research Products (Dupont, England). Oligonucle-otides were synthesized on a model 392 DNA synthe-sizer (Applied Biosystems, USA). Radiolabelling of oli-gonucleotides were performed by using T4polynucleotide kinase with [g-32P] ATP (Maxmam andGilbert, 1980), and cDNA fragments were labelled usingthe random prime labelling method with [a-32P] dCTP(Feinberg and Vogelstein, 1983).

2.2. Preparation of plasma and fat body extract

Plasma samples were prepared according to a methoddescribed previously (Hayakawa et al., 1989; Ohnishi etal., 1995).

Fat body was dissected from armyworm larvae andgently rinsed with phosphate-buffered saline (PBS,8 mM Na2HPO4, 1.5 mM KH2PO4, 137 mM NaCl,2.7 mM KCl, pH 7.2). The tissues were immediatelyhomogenized in a chilled microcentrifuge tube with200ml of 50 mM acetate buffer (pH 4.0) containing0.5 mM Pefablock SC and 5 mM (p-amidino-

phenyl)methanesulfonyl fluoride hydrochloride (Wako,Japan), and the homogenate was centrifuged at 15,000g for 10 min at 4°C. The infranatant between the top fatlayer and pellet was then centrifuged at 200,000g for60 min at 4°C, and the supernatant was chromatographedon a reversed phase C8 HPLC column (YMC, Japan; 4.6× 250 mm, pore size: 300Å) to eliminate contaminationby large proteins. A linear gradient elution of 5–50%acetonitrile in 0.1% trifluoroacetic acid (TFA)/H2O wascarried out and the fraction was eluted for 10 to 25 min.This was collected and used as the fat body extract forthe measurement of GBP precursor concentration. Pre-liminary experiments indicated that this fraction con-tained the GBP precursor protein.

2.3. Quantification of GBP precursor and GBP

The fat body extract was incubated with a proteolyticenzyme (or Golgi body-rich fraction) at 37°C for 12 hand the mixture was chromatographed on a C8 HPLCcolumn using gradient elution from 5 to 35% acetonitrilein 0.1% TFA/H20 at a flow rate of 0.4 ml/min. Theplasma sample was also chromatographed under thesame conditions for the measurement of GBP. The pep-tide fraction eluting from 13 to 21 min was collected,because this fraction contained GBP but did not containother peptides, which would cross-react with the anti-GBP monoclonal antibody (Ohnishi et al., 1995). There-fore, quantification of GBP was always preceded by thischromatographic fractionation.

GBP was quantified by an ELISA assay (Handley etal., 1982) as described previously (Ohnishi et al., 1995).Anti-GBP monoclonal antibody used for the GBP assaywas also prepared according to a method described pre-viously (Ohnishi et al., 1995).

2.4. Preparation of a Golgi body-rich fraction andprocessing enzyme

Fat body was dissected from the armyworm larvae andgently rinsed with PBS. The tissue was immediatelyhomogenized in 5 mM Tris–HCl buffer (pH 7.6) thatcontained 0.25 M sucrose and 1 mM DTT, and thehomogenate was centrifuged at 300g for 10 min at 4°C.The infranatant between the top fat layer and pellet wasthen centrifuged at 7,000g for 10 min at 4°C and, fol-lowing centrifugation of the supernatant at 100,000g for60 min at 4°C, the resulting pellet was collected. Thepellet was redissolved in 10 mM Tris–HCl buffer (pH7.6) that contained 0.1% Triton X-100, 0.1 mMb-mer-captoethanol and 0.1 mM EDTA. This was used as theGolgi body-rich fraction.

The Golgi body-rich fraction was concentrated bylyophilization and applied to a Superose 6 gel per-meation column (12× 450 mm), equilibrated with10 mM phosphate buffer, pH 7.0. The active fraction

861Y. Hayakawa et al. /Journal of Insect Physiology 44 (1998) 859–866

was rechromatographed with a Mono Q FPLC column(Pharmacia, Sweden), using a linear gradient elution of0–0.5 M NaCl in 10 mM phosphate buffer, pH 8.0. Theactive peak fraction was concentrated by lyophilizationand applied to Diol-200 gel permeation HPLC column(YMC, Japan; 8.0× 500 mm), equilibrated with 50 mMphosphate buffer (pH 7.0) containing 1 M NaCl. Elutionwas performed with the same buffer at a flow rate of0.8 ml/min and the major peak contained the processingenzyme activity (Fig. 6).

The fat body extract (10 mg protein) was reacted witha processing enzyme sample (10–100mg protein) in200ml of 25 mM Tris–HCl (pH 8.0) that contained0.8 M urea at 37°C for 12 h. One unit of the enzymeactivity was defined as that which produced 10 pmol ofGBP under the experimental conditions.

2.5. Polydnavirus (PdV) and venom fluid preparation

PdV and venom fluid were prepared as described pre-viously (Hayakawa and Yasuhara, 1993). PdV andvenom fluid were diluted with PBS to the desired con-centration (0.5 female wasp equivalents in 5ml) andinjected into a test day-0 last instar larva of the army-worm.

2.6. RNA isolation

After collecting haemolymph from the armyworm lar-vae for isolation of haemocytes, other tissues, such asbrain, nerve cord, fat body, midgut and Malpighian tube,were dissected and washed with PBS. The pelletresulting from centrifugation of whole haemolymph at300 g for 5 min at 4°C was used as a haemocyte prep-aration (Hayakawa et al., 1994). Immediately after isol-ating these tissues, they were frozen in liquid nitrogenand total RNA was isolated by the method of Chom-czynski and Sacchi (1987).

2.7. Reverse transcriptase-polymerase chain reaction(RT-PCR)

RT-PCR was performed with a Gene Amp XL RNAPCR kit (Perkin–Elmer, USA) according to the manufac-turer’s instructions. An aliquot of 100 ng of total RNAisolated from various tissues was reverse transcribedusing an oligo (dT) 16 primer. Two primers, 59-AAGAGCTCATGAAATTAACTATTTCC-39 and 59-AAGAATTCAGTAATTAATTGCTAGAAGGTGGG-39, were synthesized based on the base sequence of GBPcDNA coding region and used for the following PCRreaction. Thirty cycles of amplification were performedusing a temperature programme of 94°C for 1 min and65°C for 3 min. Preliminary studies demonstrated thatthe amplified 460 bp DNA was proportional to the con-centration of a template RNA under the same conditions.

2.8. Northern and Southern hybridization

Aliquots of 10mg of total RNA were denatured withglyoxal and dimethyl sulphoxide, and separated by 1%(w/v) agarose gel electrophoresis in 10 mM sodiumphosphate buffer, pH 7.0. The RNAs were transferred tonylon membranes (Hybond-N+, Amersham) andhybridized at 67°C in a hybridization solution containing32P-labelled GBP cDNA for 12 h. The membranes werewashed with 2× SSC containing 0.1% sodium dodecyl-sulfate (SDS) for 30 min, then twice with 0.1× SSC,which contained 0.1% SDS, for 30 min at 60°C accord-ing to the protocols of Sambrook et al. (1989).

PCR products were size-fractionated on a 5% acryla-mide gel, transferred and hybridized at 60°C in ahybridization solution containing32P-labelled syntheticoligonucleotide (41 bases, 59-TGCGTCGCTGGCTAC-ATGCGCACCCCTGACGGAAGATGCAA-39), whichcorresponded to the portion of GBP sequence. Washingconditions were the same as described for the Northernblot procedure.

2.9. In situ hybridization

Fat body from day 1 last instar larvae of the army-worm (1 day after parasitization) was fixed in 4% (w/v)paraformaldehyde and hybridized as described by Tautzand Pfeifle (1989). The hybridization probe was GBPcDNA labelled with digoxygenin.

3. Results

3.1. Plasma GBP levels after parasitization or PdV-injection

Parasitization and PdV-injection on day 0 last instarlarvae of the armyworm elevated the plasma GBP levelto about four times that of the non-parasitized larvae atthe same stage, as shown in Fig. 1. In the larvae infectedby PdV, plasma GBP reached the maximal level atapproximately 6 h prior to that of the parasitized larvae.These results suggest that parasitization causes anincrease in plasma GBP titre through PdV infection.

3.2. GBP expression in host larval tissues

To investigate the expression of GBP in varioustissues, Northern blot studies were carried out using totalcellular RNAs from day 1 last instar larval tissues 24 hafter parasitization.32P-labelled GBP cDNA hybridizedto the expressed 0.8 kb mRNA from fat body and 2.5 kbmRNA from brain and nerve cord, but no detectablelevel of hybridization was observed to RNAs fromhaemocyte, midgut and Malpighian tube (Fig. 2).Although it is possible that both brain-nerve cord and

862 Y. Hayakawa et al. /Journal of Insect Physiology 44 (1998) 859–866

Fig. 1. Plasma GBP titres of polydnavirus-injected last instarP. sepa-rata larvae. Last instar larvae injected by polydnavirus particles onday 0 (I) and parasitized on day 0 (s) as indicated with an arrow.Dotted line shows the titres of non-parasitized last instar larvae (h).The measurements were performed five times using separately pre-pared plasma samples.

fat body synthesize and secrete GBP into plasma, theabsolute amounts of GBP mRNA in the both tissueswere measured by RT-PCR in order to determine whichtissue mainly contributed to GBP secretion in plasma.

A total of 100 ng of total RNA prepared from brain-nerve cord and fat body were used for RT-PCR. DNAfragments with the predicted size (460 bp) were ampli-fied as shown in Fig. 3.

A densitometer tracing of a short exposure of the auto-radiogram indicated that the ratio of the amplified GBPcDNA fragments in brain-nerve cord and fat body wasabout 1:3, respectively. This result, together with esti-mates of the absolute amount of total RNA in bothtissues, suggests that GBP mRNA is about 100 timesmore abundant in fat body than that in brain-nerve cord.Hence, we conclude that plasma GBP would primarilybe secreted by fat body.In situ hybridization analysisusing fat body dissected from day 1 last instar larvae24 h after parasitization showed that the GBP mRNA isstrongly expressed in most of the cells (Fig. 4).

3.3. GBP mRNA levels in fat body

To precisely measure GBP mRNA levels in fat bodyafter parasitization, RT-PCR was performed. As shownin Fig. 5, parasitization had no effect on GBP mRNAlevels in fat body. Furthermore, the same analysis dem-onstrated that PdV-injection also did not affect fat body

Fig. 2. Expression of the gene encoding the GBP precursor. (a) Brain(Br), nerve cord (Nc), haemocytes (Hc), fat body (Fb), midgut (Mg)and Malpighian tube (Mt) were collected from day 1 last instar larvaeof parasitized army-worm (24 h after parasitization), 10mg of totalRNA was electrophoresed on each lane and analysed by Northernhybridization using the GBP cDNA as a probe. The blot was exposedto X-ray film using an intensifying screen at− 80°C for 15 h. (b) Por-tion of the duplicate gels was stained for ribosomal RNA with ethid-ium bromide.

GBP mRNA levels at all (data not shown). These resultswere interpreted to indicate that parasitism or PdV-infec-tion does not influence transcription of GBP.

3.4. GBP precursor levels in fat body

To examine whether the elevation in titre of GBP thatoccurs in parasitized hosts was caused by translationalor post-translational events, we tried to measure the con-centration of GBP precursor in fat body. In a previousstudy (Ohnishi et al., 1995), an anti-GBP monoclonalantibody was used for quantifying plasma GBP concen-tration, but this antibody did not recognize the GBP pre-cursor protein. Therefore, we attempted to measure theGBP precursor concentrations indirectly by quantifyingGBP produced after proteolytic processing of fat bodyextract using the same monoclonal antibody.Achromob-acter Protease I andStaphylococcus aureusV8 Protease

863Y. Hayakawa et al. /Journal of Insect Physiology 44 (1998) 859–866

Fig. 3. RT-PCR analysis of GBP mRNA in brain-nerve cord and fatbody. Right, electropherogram of GBP DNA fragments amplified fromtotal RNAs of brain-nerve cord and fat body. Left, the gel was blottedonto a nylon membrane, and the membrane was hybridized with32P-labelled 41-base oligonucleotide that corresponded to the portion ofGBP sequence.

Fig. 4. Cellular localization of expression of the GBP gene in parasitizedP. separatalarval fat body. (A) Fat body was dissected from parasitizedday 1 last instar larva (24 h after parasitization). Hybridization was found in most of the fat body cells. (B) Control fat body treated with RNaseat 37°C for 2 h before hybridization.

were used for the proteolysis of the GBP precursor pro-tein, because the substrate-specifications of theseenzymes suggested that these enzymes would be able toproteolyse the precursor to produce GBP fragment.Neither enzyme produced a peptide fragment that wasrecognized by the antibody (data not shown). However,the Golgi body-rich fraction prepared from fat body, pro-teolysed the GBP precursor to produce GBP. The Golgibody-rich fractions were prepared from fat bodies of par-asitized, PdV-injected and non-parasitized larvae, andtheir processing enzyme activities were determined. Asshown in Fig. 6, the fractions prepared from parasitized

and PdV-injected larvae indicated about two timeshigher enzyme activity than that of non-parasitized lar-vae.

The Golgi body-rich fraction prepared from PdV-injected larvae was chromatographed using gel exclusionand ion exchange columns. The chromatograms shownin Fig. 7 indicate that the main peak was eluted at 24 minin the Diol-200 HPLC column. This fraction showedenzyme processing activity. After proteolysis of the fatbody extracts using this fraction, the concentrations ofthe GBP precursor were estimated by measuring GBPlevels by ELISA following injection of PdV into lastinstar armyworm larvae. The GBP precursor level inPdV-infected larvae increased within 4 h after injectingPdV particles, and reached a maximum at approximately9 h prior to that of plasma GBP in PdV-injected larvae(Fig. 8). In contrast, the precursor concentrations in con-trol (BSA-injected) larvae remained very low levelthroughout last larval instar.

4. Discussion

The results described here suggest that, although theGBP mRNA is transcribed both in brain-nerve cord and

fat body, the mRNA present in the fat body is about100 times more abundant than that in brain-nerve cord.Therefore, it is reasonable to propose that plasma GBPis substantially synthesized and secreted by fat body.

Within 24 h of parasitization byC. kariyai, plasmaGBP concentration was elevated to about four times thatof non-parasitized armyworm larvae at the same stage(Ohnishi et al., 1995). A similar increase of plasma GBPis reproduced by PdV-injection into non-parasitized lar-vae (Fig. 1), thus indicating that parasitism-induced GBPelevation in plasma is caused by PdV-infection.Although PdV transcripts are detected in all tested larval

864 Y. Hayakawa et al. /Journal of Insect Physiology 44 (1998) 859–866

Fig. 5. RT-PCR analysis of GBP mRNA in parasitizedP. separatalarval fat body. Right, electropherogram of GBP DNA fragments amplifiedfrom total RNAs of parasitized and non-parasitized larvae. Non-para 0–48= total RNA prepared from fat body of non-parasitized last instar larvaeon day 0 at the indicated time. Para 12–48= total RNA prepared from last instar larvae at indicated time after parasitization on day 0 (2 h afterlights on). Left, the gel was blotted onto a nylon membrane, and the membrane was hybridized with32P-labelled probe of the 41-base oligonucleotide.

Fig. 6. Effect of parasitization and polydnavirus-infection on GBPprecursor processing enzyme activity. Golgi body-rich fraction wasprepared from parasitized (Para), polydnavirus-injected (Vir) and non-parasitized (Non-para) larvae and reacted with fat body extract to mea-sure their processing enzyme activities.

tissues including fat body of the parasitized armyworm(Hayakawa et al., 1994), the GBP mRNA levels in fatbody are not changed by parasitization or injection ofvirus. These observations suggest that PdV-infection that

resulted from parasitization should influence trans-lational or post-translational events and elevate plasmaGBP concentration.

In the course of examining the latter possibility, weidentified the GBP processing enzyme activity in fatbody of the armyworm larvae. Although the neuropep-tide processing enzyme was recently purified from thebrain of the tobacco hornworm,Manduca sexta(Stoneet al., 1994), the rough characterization of the GBP pro-cessing enzyme did not indicate any particular analogybetween these two enzymes. The GBP precursor pro-cessing enzyme activity that occurred in the Golgi body-rich extract was increased by about 90% after parasitiz-ation or PdV-infection. However, this increase was notmuch enough to induce the four times higher GBP levelin parasitized (or PdV-infected) larval plasma comparedwith that in non-parasitized larval plasma. Furthermore,endopeptidase and exopeptidase activities in parasitizedand non-parasitized larval haemolymph were prelimin-arily examined using various synthetic peptides but nosignificant difference in the activities were observed(data not shown). Therefore, PdV-infection is likely topartly influence the post-translational levels of GBPexpression through the elevation of the processingenzyme activity, but it may also affect the translationallevel. To confirm this hypothesis, we measured the con-centration of GBP precursor in fat body of the PdV-infected larvae. PdV-injection clearly elevates the GBPprecursor levels to 6–7 times those of control non-parasi-tized larvae. The GBP precursor concentration in fatbody increases several hours earlier than that of theplasma GBP concentration, and reaches the maximallevel at about 9 h prior to that of the plasma GBP con-centration (Fig. 8). From these results, we conclude thatthe parasitism-induced elevation of plasma GBP ismostly attributable to an increase in levels of GBP pre-

865Y. Hayakawa et al. /Journal of Insect Physiology 44 (1998) 859–866

Fig. 7. Purification procedure for GBP precursor processing enzymeactivity. (A) Golgi body-rich fraction chromatographed on a Superose6 column; (B) Superose 6 active fraction rechromatographed on aMono Q column; (C) Mono Q active fraction rechromatographed ona Diol-200 HPLC column. The columns ran sequentially. Horizontalbars indicate fractions that show processing enzyme activity.

cursor protein in the fat body of larvae infected withPdV.

References

Beckage, N.E., 1985. Endocrine interactions between endoparasiticinsects and their hosts. Annual Review of Entomology 30, 371–413.

Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA iso-lation by acid guanidinium thiocyanate–phenol–chloroform extrac-tion. Analytical Biochemistry 162, 156–159.

Feinberg, A.P., Vogelstein, B., 1983. A technique for radiolabelingDNA restriction endonuclease fragments to high specific activity.Analytical Biochemistry 132, 6–13.

Handley, H.H., Glassy, M.C., Cleveland, P.H., Royston, I., 1982.Development of a rapid micro ELISA assay for screeninghybridoma supernatants for murine monoclonal antibodies. Journalof Immunological Methods 54, 291–296.

Hayakawa, Y., 1990. Juvenile hormone esterase activity repressive fac-

Fig. 8. Effect of polydnavirus-infection on GBP precursor levels infat body. Fat body was dissected from polydnavirus-injected larvae(s) and control larvae (h, injected BSA instead of polydnavirus). Lastinstar larvae were injected by polydnavirus or BSA on day 0 as indi-cated with an arrow. Dotted line shows the titres of plasma GBP inpolydnavirus-injected larvae. The measurements were performed fourtimes using separately prepared fat body extracts.

tor in the plasma of parasitized insect larvae. Journal of BiologicalChemistry 265, 10813–10816.

Hayakawa, Y., 1991. Structure of a growth-blocking peptide presentin parasitized insect hemolymph. Journal of Biological Chemistry266, 7982–7984.

Hayakawa, Y., 1992. A putative new juvenile peptide hormone in lepi-dopteran insects. Biochemical and Biophysical Research Com-munications 185, 1141–1147.

Hayakawa, Y., 1995. Growth-blocking peptide: an insect biogenic pep-tide that prevents the onset of metamorphosis. Journal of InsectPhysiology 41, 1–6.

Hayakawa, Y., Jahagirdar, A.P., Yaguchi, M., Downer, R.G.H., 1989.Purification and characterization of trehalase inhibitor from hemo-lymph of the American cockroach,Periplaneta americana. Journalof Biological Chemistry 264, 16165–16169.

Hayakawa, Y., Ohnishi, A., Yamanaka, A., Izumi, S., Tomino, S.,1995. Molecular cloning and characterization of cDNA for insectbiogenic peptide, growth-blocking peptide. FEBS Letters 376,185–189.

Hayakawa, Y., Yasuhara, Y., 1993. Growth-blocking peptide or polyd-navirus effects on the last instar larvae of some insect species.Insect Biochemistry and Molecular Biology 23, 225–231.

Hayakawa, Y., Yazaki, K., Yamanaka, A., Tanaka, T., 1994.Expression of polydnavirus genes from the parasitoid wasp,Cote-sia kariyai, in two noctuid hosts. Insect Molecular Biology 3,97–103.

Lawrence, P.O., 1986. Host–parasite hormonal interaction: an over-view. Journal of Insect Physiology 32, 295–298.

Maxmam, A.M., Gilbert, M., 1980. Sequencing end-labeled DNA withbase-specific chemical cleavages. Methods in Enzymology 65,499–560.

Noguchi, H., Hayakawa, Y., Downer, R.G.H., 1994. Elevation of dopa-mine levels in parasitized insect larvae. Insect Biochemistry andMolecular Biology 25, 197–201.

Ohnishi, A., Hayakawa, Y., Matsuda, Y., Kwon, K.W., Takahashi,T.A., Sekiguchi, S., 1995. Growth-blocking peptide titer during lar-

866 Y. Hayakawa et al. /Journal of Insect Physiology 44 (1998) 859–866

val development of parasitized and cold-stressed armyworm. InsectBiochemistry and Molecular Biology 25, 1121–1127.

Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: aLaboratory Manual, 2nd ed. Cold Spring Harbor Laboratory,New York.

Stone, T.E., Li, J.P., Bernsconi, P., 1994. Purification and characteriz-ation of theManduca sextaneuropeptide processing enzyme car-boxypeptidase E. Archives of Insect Biochemistry and Physiology27, 193–203.

Tanaka, T., Agui, N., Hiruma, K., 1987. The parasitoidApanteles kari-yai inhibits pupation of its host,Pseudaletia separata, via disrup-tion of prothoracicotropic hormone release. General and Compara-tive Endocrinology 67, 364–374.

Tautz, D., Pfeifle, C., 1989. A non-radioactivein situ hybridizationmethod for the localization of specific RNAs inDrosophilaembryos reveals translational control of the segmentation genehunchback. Chromosoma 98, 81–85.