Cell Growth Activity of Growth-Blocking Peptide

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS 250, 194–199 (1998)ARTICLE NO. RC988959

Cell Growth Activity of Growth-Blocking Peptide

Yoichi Hayakawa1,2 and Atsushi OhnishiInstitute of Low Temperature Science, Hokkaido University, Sapporo 060, Japan

Received June 10, 1998

peptide are much higher in the early larval stages ofGrowth-blocking peptide (GBP) is an insect biogenic P. separata than in last instar larvae (5). Parasitization

peptide that retards the development of lepidopteran by C. kariyai induces a four fold elevation of plasmalarvae. cDNAs encoding GBP were isolated from three GBP in last instar larvae, which appears to disturblepidopteran species: Pseudaletia separata, Mamestra normal development from larvae to pupae (6). Isolationbrassicae and Spodoptera litura. Comparison of these of a GBP cDNA from P. separata cDNA revealed thatmolecules revealed that the GBP coding regions were

GBP is first synthesized as a polyprotein precursor70% homologous. In contrast, the upstream regions of(7,8). GBP itself is posttranslationally processed by athe deduced propeptides were only 33 % identical. Se-currently unknown protease(s).quence analysis further suggested that GBP shares

While it is clear that injection of GBP retards larvalsome structural similarities with human epidermaldevelopment of P. separata, whether other lepidopterangrowth factors. Bioassay data revealed that severalinsects produce similar precursor proteins, and thepmol/ml of GBP stimulated DNA synthesis of a humanmechanism underlying the growth blocking activity ofkeratinocyte cell line and of SF-9 insect cells. However,GBP is unclear. Here we describe the primary structuresseveral nmol/ml of GBP did not stimulate cell prolifer-of GBP precursor proteins from two other species of lepi-ation at all. In vivo studies similarly indicated that low

concentrations of GBP stimulated larval growth while doptera. GBP cDNAs cloned from the cabbage army-high concentrations of GBP retarded growth. These worm Mamestra brassicae and common cutwormdata suggest that GBP acts as a growth factor that Spodoptera litura shared little identity with P. separataregulates insect larval development. q 1998 Academic Press pro-GBP outside of the region encoding GBP itself. How-

ever, the GBP peptide regions of the three species werehighly homologous and shared some feature with humanepidermal growth factor (EGF). We found that GBP af-Endoparasitic wasps often disrupt metamorphosis offected proliferation of both human foreskin keratinocytestheir holometabolous hosts by arresting developmentand insect SF-9 cells in a dose-dependent manner.in the larval stage (1). Previous studies revealed the

presence of growth-blocking peptide (GBP) in plasmaMATERIALS AND METHODSof armyworm, Pseudaletia separata, larvae parasitized

by the wasp Cotesia kariyai (2,3). Injection of 20 pmol Animals. The armyworm P. separata, cabbage armyworm M.of GBP into nonparasitized P. separata larvae induced brassicae and common cutworm S. litura were reared on an artificial

diet at 25 { 17C with a photoperiod of 16 h light: 8 h dark (2).alteration in host growth that were similar to thosePenultimate instar larvae undergoing ecdysis between 2 and 2.5 hobserved in parasitized hosts (4). Most importantly,after lights on were designated as Day 0 last instar larvae.larval weight gain was significantly reduced and pupa-

Chemicals. Radioactively labeled reagents were obtained fromtion was delayed in GBP-injected larvae relative toNEN Research Products (Dupont). Human EGF was purchased fromsham injected larvae. We have since determined that Seikagaku Co. (Japan). GBP was synthesized using an Applied Bio-

GBP is a host gene product, and that titers of this system Model 430A peptide synthesizer as previously described (9).Oligonucleotides were synthesized on a model 392 DNA synthesizer(Perkin-Elmer). The GBP cDNA previously cloned from P. separatawas radiolabeled by random priming using [a-32P] dCTP (10).

1 Author for correspondence. Fax: 011-706-7142. E-mail: [email protected].

RNA isolation. Total RNAs were isolated by homogenizing gut-2 Present address: Department of Entomology, 237 Russell Labs,eviscerated penultimate instar larvae of M. brassicae and S. litura1630 Linden Drive, University of Wisconsin-Madison, Madison, WIaccording to the method of Chomczynski and Sacchi (11).53706-1598.

Abbreviations: DMEM, Dulbecco’s modified Eagle’s medium; EGF, cDNA synthesis and cloning. Polyadenylated mRNAs from M.brassicae and S. litura were purified from total RNA using the Quickepidermal growth factor; FBS, fetal bovine serum; GBP, growth-

blocking peptide; MGM, modified Grace’s medium; RT-PCR, reverse Prep mRNA purification system (Pharmacia). Synthesis of cDNAfrom 5 mg of poly (A)/ RNA was performed by the method of Gublertranscriptase-polymerase chain reaction.

0006-291X/98 $25.00Copyright q 1998 by Academic PressAll rights of reproduction in any form reserved.

194

AID BBRC 8959 / 6958$$$801 08-20-98 15:46:08 bbrcg AP: BBRC

Vol. 250, No. 2, 1998 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 1. Comparison of nucleotide sequences of the M. brassicae and S. litura GBP cDNAs with P. separata GBP cDNA. Sequences arenumbered at the left. * Identical residues are indicated. The sequences used for designing S1 and AS1 primers are underlined. The sequenceof GBP is boxed.

and Hoffman (12) using a cDNA synthesis kit (Stratagene). Direc- Reverse transcriptase-polymerase chain reaction (RT-PCR). Oligo-nucleotide primers were designed from the consensus sequences oftional cDNA libraries were made using the Uni-ZAP XR vector (Stra-

tagene). The library was screened with 32P-labeled P. separata GBP the P. separata and M. brassicae GBP cDNAs as follows: Primer S1,5*-AAGAGCTCCTTCAGTCAAGAGAAAATTATCTG; Primer AS1,cDNA (7).

195



FIG. 2. Comparison of deduced amino acid sequences of the three lepidopteran GBPs (A) and with human EGF (B). In (A), the percentageidentities of regions encoding GBP between two species are indicated. Numbers in parentheses indicate the percentage identities of the upstreampolypeptide regions between two species. In (B), (*) Same grouped amino acid residues are indicated. Other explanations as in Fig. 1.

5*-AAGAGCTCG(AC)TATT(AT)AGTA(AG)TTAATTGCTA. RT-PCR Cell culture and growth factor assay. Human foreskin keratino-was performed with the Superscript One-step RT-PCR kit (Gib- cytes were purchased from Morinaga co. (Japan) and maintained incoBRL) according to the manufacturer’s instructions. One hundred Dulbecco’s modified Eagle’s medium (DMEM) containing 0.8 % bo-nanograms of total RNA isolated from S. litura larvae was reverse- vine pituitary extract (Morinaga co.) and 0.8 pmol/ml EGF (completetranscribed, followed by 35 cycles of PCR amplification with primers medium A). DNA synthesis was assayed as follows. After cultivatingS1 and AS1 using a temperature program 947C for 15 sec, 607C for the cells in DMEM without any growth factor for 2 days, either GBP30 sec and 727C for 1 min. or EGF was added to the medium along with 1 mCi of [3H]-thymidine.

Cells were washed three times with 0.15 M NaCl, lysed with 200 mlDNA sequencing. cDNAs and PCR-amplified DNA fragments0.3 N NaOH and then counted using a liquid scintillation counterwere sequenced using the Taq Dye Primer Cycle sequencing kit and(Aloka LSC-5100). Cells were in their exponential growth phase atDye Terminator Cycle sequencing kit (Perkin-Elmer), respectively.time of labeling.Each segment of DNA was sequenced at least twice in both directions

SF9 cells were generously provided by Dr. Sato (Iwate Biotechnol-using an ABI 377 DNA Sequencer. Computer assisted sequence anal-ogy Research Center) and maintained in modified Grace’s mediumysis was done with GENETYX-MAC Ver 6.2.0 (Software Develop-

ment Co., Tokyo). (MGM, Grace’s medium containing 0.33 % TC lactoalbumine hydrol-

196



FIG. 3. Rates of DNA synthesis during EGF or GBP treatment of human foreskin keratinocytes (A,B) and insect SF9 cells (C,D).Synthesis of DNA was measured by the incorporation of [3H]-thymidine. Keratinocyte cells were cultivated for 3 days at 377C in completemedium A ( 0.8% bovine pituitary extract and 0.8 pmol/ml EGF in DMEM) and, after cultivating in DMEM for 2 days, EGF (l), GBP (s)or BSA (h, dotted line) was added into the culture medium along with [3H]-thymidine. SF9 cells were cultivated for 3 days at 257C incomplete medium B (10 % FBS in MGM) and then in MGM for 2 days before adding EGF or GBP into the culture medium along with theradiolabeled thymidine. A,C: effect of various concentrations of EGF or GBP on DNA synthesis (both cell populations were exposed continu-ously for 40 h to one of each peptide and radiolabeled thymidine) ; B,D: effect of time from adding 0.8 pmol/ml EGF or GBP (both cellpopulations were exposed continuously for indicated periods to one of each peptide and radiolabeled thymidine). Results are expressed asradioactivity incorporated into cells growing in each medium. Data are given as means { S.D. for 7 separate measurements.

ysate and 0.33 % TC yeastolate) with 10 % fetal bovine serum (FBS) about 900 bp. Sequence analysis of three of these in-(complete medium B). After cultivating the cells in MGM unsupple- serts revealed a single open reading frame beginningmented for 2 days, either EGF or GBP was added to the medium at residue 59 and ending at residue 514 (Fig. 1). Thealong with [3H]-thymidine. Thymidine incorporation was measured

deduced precursor polypeptide was 152 amino acids.by the same procedures as described above. Under these conditions,cells were in their exponential growth phase at time of labeling. The twenty three amino acid residues homologous to

GBP were found at a carboxyl proximal domain of thededuced precursor polyprotein. The amino acid se-RESULTS AND DISCUSSIONquence of GBP and the preceding polypeptide portions

Screening of the Mamestra brassicae Zap cDNA li- were 82 % and 69% identical respectively to those ofbrary using 32P-labeled GBP cDNA of Pseudaletia sep- the P. separata GBP precursor protein (Fig. 2,A). Based

on the sequence of the M. brassicae and P. separataarata identified multiple clones containing an insert of

197



concentrations of GBP and EGF. To assess the persis-tence of this response, time course studies were con-ducted for a 40 h period after exposure to 0.8 pmol/mlGBP or EGF. DNA synthesis remained elevated for atleast 36 hr following addition of GBP or EGF (Fig. 3,B)and, afterward, gradually decreased (data not shown).From these results, it is apparent that GBP and EGFare more or less equivalent in cell growth activity onkeratinocyte cells.

The same measurements were performed on SF9 cellsderived from the moth, Spodoptera frugiperda. GBP in-creased DNA synthesis in SF9 cells about 10 times morestrongly than EGF. Both factors stimulated DNA syn-thesis in these cells at concentrations of 1002-102 pmol/ml. However, higher concentrations of these factors didnot increase DNA synthesis above levels observed in con-

FIG. 4. The effect GBP on weight gain of P. separata larvae. Day trol cells (Fig. 3,C). Time course studies indicated that0 last instar larvae (0.17-0.18 g) of the armyworm were injected with GBP induced a higher and more persistent increase inindicated amounts of GBP early in the last larval instar and weighed

SF-9 cell proliferation than EGF (Fig. 3,D).15 hrs after injection. Control (0) and BSA indicate that larvae wereinjected by just saline and 10 pmol/larva BSA, respectively. Each To determine whether growth of P. separata larvaebar indicates the mean percentage { S.D. of weight gain of the is also affected by GBP, we injected various concentra-treated larva (n Å 36). *2, *3, significant different from control value tions of GBP solubilized in saline into Day 0 last instar(* 1) (Põ 0.01). *4, no significant difference from control value (*1)

larvae. Larvae were then weighed 24 h later. Larvae(Pú0.05).injected with 10 pmol of GBP gained significantly lessweight than saline or BSA injected control larvae. Incontrast, larvae injected with 0.1 or 1 pmol of peptide

GBP cDNAs, two primers (S1 and AS1) flanking the gained more weight than control larvae (Fig. 4).coding region were designed to amplify a GBP cDNA In the present work we have characterized two newfrom the common cutworm Spodoptera litura by RT- cDNAs from M. brassicae and S. litura that encodePCR. A product of about 640 bp was isolated and bidi-

homologous of P. separata GBP. Analysis of the de-rectionally sequenced (Fig.1). The open reading frameduced proteins encoded by these cDNAs revealed noof the S. litura GBP cDNA encoded a deduced proteinsignificant homologies with other known peptides orof 144 amino acids. Again, the twenty three amino acidsproteins. We therefore conclude that these precursorcorresponding to GBP were located at the carboxyl ter-molecules only encode GBP. However, the primaryminal region of the PCR product. The GBP portions ofstructure of each of these GBP peptides suggested tothe three insects are likely endoproteolyzed at the sameus that they shared some similarity with EGF espe-cleavage site, GlyArg-GBP. Although the amino acidcially in regard to the location of two Cys residues, andsequence of the S. litura GBP portion was highly ho-associated formation of disulfide bonds. This motif ismologous to those of the P. separata and M. brassicaealso found in the primary structures of insect paralytic(74% and 78%, respectively), the upstream region ofpeptides (14) and plasmatocyte-spreading peptide (15).the protein was only 35 % identical to P. separata pro-Using human foreskin keratinocytes and insect SF9GBP and 47 % identical to M. brassicae (Fig. 2,A).cells, we found that GBP induces DNA synthesis. TheSearch of current data bases revealed no significantfact that treatment of human keratinocyte cells withhomology of three proteins to any other known proteinGBP had a similar effect as EGF suggests that GBPor peptide. However, comparison of motifs present inmay stimulate EGF receptors on the cells. On the otherGBP to other molecules suggested some similarities tohand, the proliferative effect of EGF on SF9 cell growthhuman epidermal growth factor (EGF) (Fig. 2,B) (13).is much weaker than that of GBP, indicating that theIn particular, two Cys residues are similarly locatedaffinity of EGF for a putative GBP receptor on SF9and linked through disulfide bonds in both GBP andcells may be much weaker.EGF. This suggested to us the possibility that GBP

We suggest that the cell proliferation and growthmay act as an insect growth factor.blocking effects of GBP are concentration dependent.To test this hypothesis, human foreskin keratino-Concentrations of GBP, ú 100 pmol/ml, did not inducecytes were cultivated in DMEM without serum for 2significant DNA synthesis in either cell line we exam-days and then cultured in medium plus GBP or EGF.ined. Similarly, injection of a similar concentration ofBoth GBP and EGF at a concentration of 1001-102 pmol/GBP into P. separata larvae retards growth (16,17).ml induced an increase in DNA synthesis (Fig.3,A).

DNA synthesis was not increased at lower or higher In contrast, low concentrations of GBP induced DNA

198



4. Hayakawa, Y., and Yasuhara, Y. (1993) Insect Biochem. Molec.synthesis in vitro, and, as shown in Fig.4, caused larvaeBiol. 23, 225–231.to increase in size relative to controls.

5. Hayakawa, Y. (1995) J. Insect Physiol. 41, 1–6.As far as we know, this is the first report concerning6. Ohnishi, A., Hayakawa, Y., Matsuda, Y., Kwon, K. W., Taka-

a natural peptidergic growth factor in insects. Al- hashi, T. A., and Sekiguchi, S. (1995) Insect Biochem. Molec. Biol.though the current study does not fully characterize 25, 1121–1127.potential role of GBP as a cell growth factor, further 7. Hayakawa, Y., Ohnishi, A., Yamanaka, A., Izumi, S., and Tom-

ino, S. (1995) FEBS Lett. 376, 185–189.studies of this type will undoubtedly improve our un-8. Hayakawa, Y., and Noguchi, H. (1998) Eur. J. Biochem. in press.derstanding of this factor in insect development.9. Hayakawa, Y. (1994) J. Biochem. Tokyo 115, 15–17.

10. Feinberg, A. P., and Vogelstein, B. (1983) Anal. Biochem. 132,6–13.ACKNOWLEDGMENTS

11. Chomczynski, P., and Sacchi, N. (1987) Anal. Biochem. 162, 156–159.

We thank Professor Michael R. Strand (University of Wisconsin-12. Gubler, U., and Hoffman, B. I. (1983) Gene 25, 263–269.Madison) and Dr. Vladimir Kostal (Academy of Sciences of the Czech13. Gregory, H., and Preston, B. M. (1977) Int. J. Pept. Protein Res.Republic) for their critical reading. This study was partly supported

9, 107–118.by Grant-in-Aid for Research on Priority Areas (08276101) from the14. Skinner, W. S., Dennis, P. A., Li, J. P., Summerfelt, R. M., Car-Ministry of Science, Education and Culture of Japan.

ney, R. L., and Quistad, G. B. (1991) J. Biol. Chem. 266, 12873–12877.

15. Clark, K. D., Pech, L. L., and Strand, M. R. (1997) J. Biol. Chem.REFERENCES272, 23440–23447.

16. Noguchi, H., Hayakawa, Y., and Downer, R. G. H. (1995) Insect1. Beckage, N. E. (1985) Annu. Rev. Entomol. 30, 317–413. Biochem. Molec. Biol. 25, 197–201.2. Hayakawa, Y. (1990) J. Biol. Chem. 265, 10813–10816. 17. Noguchi, H., and Hayakawa, Y. (1996) Insect Biochem., Molec.

Biol. 26, 659–665.3. Hayakawa, Y. (1991) J. Biol. Chem. 266, 7982–7984.

199


Documents

Cell Growth Activity of Growth-Blocking Peptide