Eur. J. Biochem. 254, 90295 (1998) FEBS 1998
Cloning of two cDNAs encoding three small serine protease inhibiting peptides from the desert locust Schistocerca gregaria and analysis of tissue-dependent and stage-dependent expression Jozef VANDEN BROECK 1 , Shean-Jaw CHIOU 1 , Liliane SCHOOFS 1 , Ahmed HAMDAOUI 2 , Frank VANDENBUSSCHE 1, Gert SIMONET 1, Said WATALEB 2 and Arnold DE LOOF 1 1 2
Laboratory for Developmental Physiology and Molecular Biology, Leuven, Belgium Faculte´ des Sciences Semlalia, Universite´ Cadi Ayyad, Marrakech, Morocco
(Received 27 January 1998) 2 EJB 98 0131/1
This study describes the cloning of two cDNAs encoding three serine-protease-inhibiting peptides, SGPI I, II and III, which were recently identified from ovarian extracts of the desert locust, Schistocerca gregaria. The first cDNA codes for the precursor polypeptides of SGPI I and SGPI II; the second encodes only a single inhibitor, SGPI III. Northern-blot analysis revealed an approximate length of 0.8 kb for SGPI-I/II mRNA and 0.6 kb for SGPI-III mRNA. The transcripts are present in several locust tissues, but they could not be detected in the midgut. The gene for SGPI-I/II is abundantly transcribed during all larval and adult stages, whereas SGPI-III mRNA is mainly present in adults. Northern-blot hybridization also revealed important changes in the SGPI-mRNA content during the molting cycle and during the adult reproductive cycle. Moreover, a differential hormonal control was observed in adult females which had been treated with precocene, juvenile hormone or ecdysone. Keywords : chymotrypsin ; elastase; enzyme; inhibitor; protease.
In all metazoan species, proteases play a prominent role in a wide array of physiological processes such as food digestion, blood clotting, embryogenesis, tissue reorganization (e.g. wound healing, regeneration, molting, metamorphosis), defense mechanisms and immune responses. Activation and inactivation of protease cascades have to be closely controlled at different regulatory levels such as protease gene transcription, mRNA translation, zymogen activation, substrate specificity, enzyme kinetics, the presence of enzyme-inhibiting substances. Therefore, most animal species possess a variety of protease inhibitors, most of which have different specificities. Based on structural characteristics, several distinct protease inhibitor families can be defined. Some protease inhibitor families such as serpins, A2-macroglobulins and Kunitz-type peptides are encountered in vertebrates as well as in invertebrates (Laskowski and Kato, 1980; Sasaki, 1984; Hergenhahn et al., 1988; Takagi et al., 1990; Eguchi, 1993). Members of novel families of protease inhibitors have recently been discovered in invertebrates, but most of these are still little investigated. Brehe´lin et al. (1991) and Boigegrain et al. (1992) have reported the identification of two serine protease inhibitors (LMCI I or PMP-D2 and LMCI II or PMP-C) from the migratory Correspondence to J. Vanden Broeck, Laboratory for Developmental Physiology and Molecular Biology, Naamsestraat 59, B-3000 Leuven, Belgium Fax: 132 16 323902. E-mail:
[email protected] Abbreviations. JH, juvenile hormone ; LMCI, Locusta migratoria chymotrypsin inhibitor ; RACE, rapid amplification of cDNA ends ; SGPI, Schistocera gregaria protease inhibitor. Note. The cDNA sequences mentioned in this paper have been submitted to the EMBL database and are available under accession numbers Y09605 and Y09606.
locust, Locusta migratoria. By cDNA cloning, both peptides were shown to be derived from one single precursor polypeptide (Kromer et al., 1994). A possible role for these peptides was suggested in the regulation of the prophenoloxidase cascade. More recently, the identification of a third peptide (HI) was reported (Kellenberger et al., 1995) and the solution structures of PMP-D2 and PMP-C were analyzed by NMR studies (Mer et al., 1994, 1996). Hamdaoui et al. (1998) recently isolated additional members of this new family of serine protease inhibitors from ovarian tissue of the desert locust Schistocerca gregaria. The sequences of two of these peptides appear to be very similar to those of LMCI I and LMCI II. At present, the total number of inhibitor sequences belonging to this peptide family in S. gregaria, is five. These peptides were called SGPI I2V and were shown to be present in the hemolymph and in different locust tissues, including the ovary. The demonstration of specific inhibitory activities towards several endogenous locust serine proteases suggests that the physiological role of these peptides is not restricted to the control of the prophenoloxidase cascade (Hamdaoui et al., 1998). This article presents the cloning and identification of cDNAs encoding the first three serine protease inhibitors (SGPI I2III). Gene expression was investigated via Northern-blot analysis of different locust tissues and stages. Hormonal regulation of transcription of these protease inhibitor genes was studied by treatment of locusts with precocene 1, JH III and 20-OH-ecdysone. MATERIALS AND METHODS Rearing of desert locusts. Schistocerca gregaria (Forsk.) was reared under stable temperature conditions (32 61°C). The
Vanden Broeck et al. (Eur. J. Biochem. 254)
animals were kept in special cages and fed daily with fresh grasses, rolled oats and cabbage leaves. Mature females deposit their eggs in pots filled with slightly humidified sand. After oviposition, these pots are collected at given time intervals in order to obtain, later, synchronized pools of hatched first instar hoppers. RNA isolation and cDNA synthesis. Locusts were dissected under a binocular microscope and tissue samples were collected and directly frozen in liquid nitrogen. Total RNA was extracted using TRI-zol reagent (BRL). Poly(A) RNA was extracted from total RNA or from pooled tissue samples under RNase-free conditions using the QuickPrep Micro mRNA Purification Kit (Pharmacia Biotech) and quantified by spectrophotometry. The resulting mRNA was used as a template (0.5 µg) for cDNA synthesis. Double-stranded cDNA was obtained in two steps following the protocol included in the Marathon cDNA Amplification Kit (Clontech). Polymerase chain reaction. PCR primer sequences were based on amino acid sequences of protease-inhibitor peptides from Schistocerca, SGPI I2III. The primers (Eurogentec) had the following sequences. SGPI I, (a) →5′-GGICA(A/G)ACIAA(A/G)AA(A/G)CA(A/G) GA(T/C)TG-3′; (b) ←5′-CAICC(T/C)TTIC(T/G)IGT(A/G)CA IGCCCA-3′. SGPI II, (a) →5′-CCIGGIACIACITT(T/C)AA(A/G)GA(T/C) AA(A/G)TG-3′; (b) ←5′-TGIGG(A/G)CAIGC(T/C)TTIA(A/G) IGT(A/G)CA-3′. SGPI III, (a) →5′-AA(A/G)TA(T/C)GA(T/C)GGITG(T/C)A A(T/C)TGGTG-3′ ; (b) ←5′-G(A/G)CA(A/G)TA(T/C)TTIA (A/G)IGT(A/G)CA(A/G/T)ATCC-3′. 50 µl PCR reactions were performed containing 5 µl 103 PCR buffer (provided by the manufacturer), 0.2 mM each dNTP, 0.2 µM each primer, 5 µl cDNA template (, 0.1 µg cDNA), 1 µl Advantage PCR 503heat-stable DNA polymerase enzyme mix (Clontech). Hot-start PCR, was run for 35 cycles. One cycle consisted of denaturation for 1 min at 94°C, a primer annealing period of 60 s at 60°C and an extension for 60 s at 68°C. After the last cycle, an additional extension time of 7 min at 68°C was applied. PCR reactions were analyzed by horizontal agarose-gel electrophoresis and bands were visualized by ethidium bromide fluorescence. PCR products were subcloned and sequenced as outlined below. Rapid amplification of cDNA ends (RACE) protocol. In order to obtain a complete cDNA sequence, RACE was employed. This RACE protocol was performed according to the instructions of the Marathon cDNA Amplification kit (Clontech). The adapter primers were provided with the kit, whereas the specific 5′-RACE and 3′-RACE primers (Eurogentec) were derived from the sequence of the original PCR fragments. SGPI I/II, (a) →5′-AATTGTACCCCTACTGGAGTTTGG-3′; (b) ←5′-TTCCGTCACTTCCACAGCGACACG-3′. SGPI III, (a) →5′-TGGTGCACGTGTTCCAGTGGCGGC-3; (b) ←5′-GCCGCCACTGGAACACGTGCACCA-3. 5′-RACE and 3′-RACE fragments were analyzed by horizontal agarose electrophoresis and ethidium bromide fluorescence. Cloning, sequencing and sequence analysis. PCR and RACE products were cloned into pCR2.1 vector using the Original TA Cloning Kit (Invitrogen). Recombinant plasmids were isolated and the inserts were sequenced by following the protocols outlined by the Sequenase version 2 sequencing kit (USB-
91
Amersham). The universal M13 (240) and reverse M13 primers were used to obtain the terminal cDNA sequences. Nucleotide and amino acid sequence analyses and comparisons were performed by using PC-Gene software (Intelligenetics). Sequence alignments were obtained by employing the Clustal and Align programs. Database searches were run with the Blast or Fasta programs. Experiments and hormonal treatment. Female locusts were treated at the time of adult emergence by topical administration of precocene I (500 µg in 15 µl acetone/animal). Control animals were treated with acetone only. Hemolymph was collected at day 4 and day 8 after ecdysis. At day 8, the animals were treated with juvenile hormone (JH) III (per animal 10 µg by topical treatment, 5 µg by injection into the abdomen; JH was dissolved at 1 µg/µl in ethanol), with 20-OH ecdysone (twice daily during three consecutive days, injection of 2 µg dissolved in ethanol at 1 µg/µl) or with solvent (control). Hemolymph, fat body and ovaries were collected at day 12. All hemolymph samples were analyzed by measuring the total protein concentration (Bradford, 1976) and by SDS/PAGE in order to monitor vitellogenin appearance and to evaluate the effects of precocene and hormone treatments. Tissues were pooled (. six animals/pool) and mRNA was extracted for Northern-blot analysis. Northern-blot analysis. Poly(A) RNA was prepared from locust tissues or whole animals corresponding to different developmental or physiological conditions. 2 µg each sample (derived from pooled insects) was loaded for RNA electrophoresis. The agarose gel electrophoresis was run under denaturing conditions by including formaldehyde in the gel (Sambrook et al., 1989). The RNA in the gel was then transferred onto ‘Hybond N’ (Amersham) nylon membranes by capillary blotting with diethylpyrocarbonate-treated 203SSPE buffer (175.3 g NaCl, 27.6 g NaH2PO4 · H2O, 7.4 g EDTA in 1 l, pH adjusted to 7.4 and sterilized by autoclaving; Sambrook et al., 1989) and irreversibly blocked by drying and baking in an oven at 80°C for 2 h. The blot was prehybridized for 4 h at 37°C in a solution (20 ml) containing 0.5 mg denatured heterologous herring sperm DNA, 50% desionized formamide, 53SSPE, 103Denhardt’s solution and 0.5 % SDS. A specific cDNA probe was prepared by labeling the cloned RACE fragments via the Rediprime system (Amersham) with 32P (Redivue dCTP, Amersham). The denatured probe was added to the prehybridization solution and the hybridization reaction was incubated for at least 16 h. The hybridized nylon membrane was rinsed in 23SSPE, 0.1 % SDS at room temperature and washed at room temperature and at 65°C in 13SSPE, 0.1% SDS during 15 min for each condition. High stringency washing was performed at 65°C in 0.13SSPE, 0.1% SDS for 10 min. Then, the filter was removed, wrapped in SaranWrap and positioned in an X-ray cassette (Kodak) with intensifying screens. Autoradiography was performed by exposure of a hyperfilm-MP (Amersham) to the hybridized blot during a period of 24 h. All blots were rehybridized with a control actin probe to check if the mRNA content in each lane was comparable. Therefore, the significant changes in S. gregaria protease inhibitor (SGPI) expression patterns which are discussed in the results section are not caused by differences in loading of the lanes.
RESULTS Cloning and sequencing results. The combined approach of PCR and RACE has led to the identification of two cDNAs encoding three peptide serine protease inhibitors. One of the cDNAs contains the genetic information for two peptides which
92
Vanden Broeck et al. (Eur. J. Biochem. 254)
Fig. 1. Nucleotide sequence (cDNA) and translated amino acid sequence of the SGPI I/II precursor. The nucleotide sequences for the start codon, stop codon and two possible polyadenylation consensus sequences (AATAAA) are underlined. The amino acid sequences of the inhibitor peptides, SGPI I and SGPI II, are underlined and shown in bold. Basic residues which are cleaved off are shown in italic. An arrow (⇑) indicates the computed cleavage site of the signal peptide (according to Von Heijne, 1991).
have been shown to have a distinct specificity for serine proteases (trypsin and chymotrypsin for SGPI I compared to chymotrypsin and elastase for SGPI II ; Hamdaoui et al., 1998). The peptide sequences are separated by two basic amino acid residues, indicating that they are split off from a larger precursor polypeptide, as is the case with many neuronal or endocrine peptides. The presence of a putative signal sequence (von Heijne, 1991) suggests that the peptides are secreted to the extracellular space. In contrast, the second cDNA encodes only one chymotrypsin-inhibiting peptide (SGPI III) which is also preceded by a secretory signal sequence. The cDNA-derived amino acid sequences are the same as those identified earlier by automated Edman degradation of the chromatographically purified peptides (Hamdaoui et al., 1998). A number of C-terminal residues which were missing in the Edman degradation data have now been characterized. Moreover, the cDNA-derived sequence data are in full agreement with the determinations (made by matrix-assisted laser desorption/ ionization time-of-flight mass spectroscopy) of the chromatographically purified peptides (Hamdaoui et al., 1998). The cDNA sequences (Figs 1 and 2) have been employed as hybridization probes to study the tissue distribution and physiological or hormonal regulation of gene expression. Northern-blot analysis. Tissue distribution. In a first experiment, RNAs derived from different desert locust tissues were analyzed via Northern-blot hybridization with the above mentioned cDNA probes. The mRNA for SGPI-I/II has an approximate length of 0.8 kb and is a very abundant transcript in the fat body, but is also present in lower amounts in several other tissues such as the ovary, the testes, the nervous system, the foregut and hindgut, etc. Interestingly, in the midgut, the most important source of digestive enzymes, the transcript was not detectable. A similar distribution was observed when hybridizing with SGPI-III cDNA. As expected, the approximate length (0.6 kb) of this transcript is shorter than that of the previous one. Of all tissues tested, the fat body and, to a lesser extent, the gonads
Fig. 2. Nucleotide sequence (cDNA) and translated amino acid sequence of the SGPI III precursor. The nucleotide sequences for the start codon, stop codon and possible polyadenylation consensus sequence (AATAAA) are underlined. The amino acid sequence of the inhibitor peptide, SGPI III, is underlined and shown in bold. An arrow (⇑) indicates the computed cleavage site of the signal peptide (according to Von Heijne, 1991).
Fig. 3. Sequence comparison of locust (SG, S. gregaria and LM, L. migratoria) peptides belonging to the recently discovered family of serine protease inhibitors. Conserved positions are shown by asterisks. Underlined amino acid residues were not determined by automated Edman degradation of the chromatographically purified peptides, but are derived from the cDNA data. Fucosyl-Thr residues are shown in bold (T). All members of this peptide family contain six Cys residues at conserved positions indicating that the tertiary structure of these small peptides might be of fundamental importance for maintenance of their serine-protease-inhibitory activities.
are the most important sources of transcript. Compared to SGPII/II, the SGPI-III mRNA is much less abundant. Staging. In a second test, mRNAs prepared from whole animals of different stages were analyzed by Northern-blot hybridization. The SGPI-I/II transcript is very abundant in all stages (larval stages I2V and in adult females and males). SGPI-III mRNA is present in all stages, but it is more abundant in the adult stage (males and females) than in the larval stages. Reproductive cycle. To follow the time course of expression of the peptide genes during the reproductive cycle, locust fat body and gonads were microdissected at given times (days 32 27) after last ecdysis and the mRNAs of pooled (. six animals/ pool) tissues corresponding to the same stage were prepared and quantified by spectrophotometry. Northern-blot hybridization (semi-quantitative analysis) was performed on 2 µg poly(A) RNA/sample/lane.
Vanden Broeck et al. (Eur. J. Biochem. 254)
93
Fig. 4. Tissue-dependent expression analyzed via Northern-blot hybridization. Hybridization with SGPI I/II cDNA (a) demonstrates the existence of a transcript of 0.8 kb in locust fat body and gonads. Use of SGPI III cDNA probe (b) indicates a transcript of 0.6 kb in locust fat body and gonads. These transcripts could not be detected in mRNA samples derived from midgut tissue (Mg). Ov, ovary; T, testis ; FbW, female fat body ; FbM, male fat body ; Mg, midgut; kb, kilobases.
Fig. 5. Stage dependent expression analyzed via Northern-blot hybridization. The mRNA samples were derived from whole bodies. Larval stages: I, II, III, IV, V; adult males, M and females, W; kb, kilobases; a, SGPI I/II ; b, SGPI III.
The SGPI-I/II transcript is highly abundant in male and female fat bodies and in the testes. In the ovary, the abundance of this mRNA significantly decreases during the period of oocyte growth and oviposition (days 13225). SGPI-III shows a very different pattern. In the adult female fat body, this transcript is only abundant until a few days after emergence and its presence considerably decreases after day 5. In the males the transcript levels show more temporal fluctuation during adult life. Expression in the ovary and testes is much
Fig. 6. Northern-blot analysis of SPGI I/II and SGPI III transcripts during the adult reproductive cycle. (a) Northern-blot analysis of SGPI I/II transcripts. 1, male fat body; 2, female fat body ; 3, testes; 4, ovary. Samples were prepared from pooled tissue extracts 3227 days (d) after the last ecdysis. (b) Northern-blot analysis of SGPI III transcripts. 1, male fat body ; 2, female fat body; 3, testes ; 4, ovary. Samples were prepared from pooled tissue extracts 3227 days (d) after the last ecdysis.
lower than in the fat bodies, but appears to have a similar sexand time-dependency. Molting cycle. Desert locusts were set apart at the time of fifth-instar emergence. mRNA was prepared from whole animals (. six animals pooled/condition) which were taken at certain
94
Vanden Broeck et al. (Eur. J. Biochem. 254)
strong inhibition of the SGPI I/II transcript in the ovary of precocene-I-treated females. In the fat body, this inhibitory effect was not observed. JH III or 20-OH-ecdysone treatments abolished the inhibitory effect of precocene I and restored (normal) ovarian SGPI I/II gene expression. Precocene I and control conditions do not seem to differ in their SGPI III mRNA contents (ovary and fat body). Surprisingly, 20-OH-ecdysone strongly induced the transcript in precocene-treated females (ovary and fat body) while this was not the case in control animals.
DISCUSSION
Fig. 7. Northern-blot analysis of SGPI transcripts during the last molting cycle (from fifth instar, V, to the adult stage, Ad). Samples were prepared from pooled whole body extracts at 1, 2, 4, 6 days after the fourth molt (beginning of stage V) and a few hours after the last molt. a, SGPI I/II; b, SGPI III ; c, actin control.
Fig. 8. Northern-blot analysis of female ovarian mRNA samples taken after treatment of the animals with precocene and/or hormonal factors. Animals were treated with precocene after the last molt. Samples were prepared from pooled tissues 4, 8 and 12 days after this initial treatment. p, precocene-I treated; c, control ; JH, juvenile hormone treated; Ecd, 20-OH-ecdysone-treated conditions; a, SGPI I/II; b, SGPI III.
time intervals (1 day, 2 days, 4 days, 6 days) after fifth-instar emergence and at a few hours (approximately 3 h) after adult molting. Northern-blot analysis was performed (2 µg mRNA/ lane) by hybridizing with SGPI I/II and SGPI III cDNA probes. The detectable amount of mRNA coding for SGPI I/II was much higher during the intermolt stages than during, or shortly after, ecdysis. In contrast, SGPI III shows a different pattern; the intermolt stages did not contain more of this mRNA than the stages closely before or after ecdysis. Precocene I and hormonal treatment. Female locusts which were treated by precocene I had significantly lower protein concentrations in their hemolymph and the ovaries were clearly underdeveloped compared with the control (acetone) animals. In JH-III-treated or 20-OH-ecdysone-treated females, these effects of precocene I were reduced. SGPI I/II and SGPI III gene expression was analyzed via Northern-blot hybridizations on mRNAs extracted from the experimental and control animal tissues. 4 days after precocene I treatment, no detectable differences in SGPI I/II transcripts were observed between treated and control animals (in ovary and fat body). The conditions on days 8 and 12, however, showed a
The present study describes the cloning of two cDNAs encoding three serine-protease-inhibiting peptides which were recently identified from ovarian tissue of S. gregaria (Hamdaoui et al., 1998). The first cDNA contains the message (0.8 kb) for the precursor of SGPI I and SGPI II, two closely related inhibitors which have distinct protease-inhibiting specificities. Therefore, co-expression of these two inhibitors probably leads to the inhibition of a broad spectrum of serine protease (trypsin-, chymotrypsin- and elastase-like) activities. The second (0.6 kb) codes for the precursor of one single peptide, SGPI III, which is a member of the same peptide family and which preferentially inhibits chymotrypsin-like activities (Hamdaoui et al., 1998). The SGPI I/II transcript is highly abundant in all stages, whereas SGPI III is less abundant and reaches its highest levels in adult locusts. These observations indicate that SGPI I/II and SGPI III gene expression is controlled by different regulation mechanisms and that the resulting peptides are probably involved in distinct serine-protease-dependent physiological processes. SGPI I/II is the more abundant transcript, present in all stages of development. This expression pattern suggests the existence of a very general, stage-independent role for the coexpressed ‘broad spectrum’ combination of SGPI I and SGPI II, such as in the prophenoloxidase cascade as suggested for the related Locusta peptides LMCI I and LMCI II (Brehe´lin et al., 1991). A possible role in defense or immune processes, or in protection mechanisms against damage to the insect cells and tissues which could be caused by serine proteases originating from exogenous (e.g. invasive micro-organisms) or from endogenous (e.g. leakage of midgut enzymes into the hemolymph) sources, cannot be excluded. The main source of both mRNAs is the fat body. Many other tissues probably produce mRNAs in lower amounts, but during the dissection of some organs (such as the ventral nerve cord and the testis) it was very difficult to completely avoid the presence of some residual fat body tissue. The midgut of insects is the tissue where most of the digestive proteases are produced. It is interesting that in this tissue the SGPI mRNAs are absent (or they are present in extremely low, undetectable amounts). This is in agreement with our reverse-phase HPLC analysis data ; SGPIs could not be detected in midgut extracts (Hamdaoui et al., 1998). Another important observation is that, during the vitellogenic cycle, when the ovarian follicles are growing and taking up yolk proteins, the ovary contains lower amounts of SGPI transcripts. Therefore, ovarian SGPI (especially SGPI I/II) expression seems to be inversely linked to this process. Interestingly, the SGPI I/II transcript is also (inversely) regulated during the molting cycle. This suggests that the expression of SGPI genes is regulated by hormones. In the midgut, at the epidermis during molting, and probably also in the growing ovarioles of vitellogenic females, higher serine protease activities are required for normal physiological functioning (food digestion in the midgut, cuticle processing and reorganization of the epidermis and remodeling
Vanden Broeck et al. (Eur. J. Biochem. 254)
of yolk protein precursors and/or of extracellular matrix proteins in the ovary). In this context, a recently published hypothesis has to be mentioned (De Loof et al., 1995); it attempts to explain how follicle growth could be controlled by proteolytical activities present in the vitelline membrane. Time-dependent expression of protease inhibitors in the growing follicles might be part of such a control mechanism. The ovary of the desert locust also contains another, very potent serine protease inhibitor-type (14-kDa trypsin inhibitor) which does not belong to the peptide family discussed in this paper (Hamdaoui et al., 1997). This already indicates that a number of different proteases and (corresponding) inhibitors are probably involved in the control of protease activities in the locust ovary. Although it is not possible to define all functions which are exerted by these serine protease inhibitors, the present results show that the genes for different peptides are differentially expressed in a stage-, time- and hormone-dependent manner. Therefore, it is likely that the resulting peptides are involved in the control of distinct physiological processes, some of which might be rather general while others might be more (temporally or spatially) restricted, and hormonally controlled (e.g. during molting or during the reproductive cycle). The authors thank the FRS (Fund for Scientific Research, Flanders, Belgium) and the European Union (TS3*-CT93-0208) for funding this research. J. Vd. B. is a Senior Research Associate of the FRS-Flanders. R. Jonckers and J. Gijbels are gratefully acknowledged for technical assistance. The authors also thank H. Van Den Bergh, M. Van Der Eeken, J. Puttemans and M. Christiaens for their help with text and figure editing.
REFERENCES Boigegrain, R. A., Mattras, H., Brehe´lin, M., Patsoutand, P. & ColettiPreviero, M. A. (1992) Insect immunity : two proteinase inhibitors from hemolymph of Locusta migratoria, Biochem. Biophys. Res. Commun. 189, 7902793. Bradford, M. M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein, utilizing the principle of protein-dye binding, Anal. Biochem. 72, 2482254. Brehe´lin, M., Boigegrain, R. A., Drif, L. & Coletti-Previero, M. A. (1991) Purification of a protease inhibitor which controls prophenoloxidase activation in hemolymph of Locusta migratoria, Biochem. Biophys. Res. Commun. 179, 8412846. De Loof, A., Bylemans, D., Schoofs, L., Janssen, I., Spitaels, K., Vanden Broeck, J., Huybrechts, R., Borovsky, D., Hua, Y. J., Koolman, J. & Sower, S. (1995) Folliculostatins, gonadotropins and a model for
95
control of growth in the grey fleshfly, Neobellieria bullata, Insect Biochem. Molec. Biol. 25, 6612667. Eguchi, M. (1993) Mini-review. Protein protease inhibitors in insects and comparison with mammalian inhibitors, Comp. Biochem. Physiol. 105B, 4492456. Hamdaoui, A., Schoofs, L., Wataleb, S., Vandenbosch, L., Verhaert, P., Waelkens, E. & De Loof, A. (1997) Purification of a novel, heatstable serine protease inhibitor protein from ovaries of the desert locust, Schistocerca gregaria, Biochem. Biophys. Res. Commun. 238, 3572360. Hamdaoui, A., Wataleb, S., Devreese, B., Chiou, S.-J., Vanden Broeck, J., Van Beeumen, J., De Loof, A. & Schoofs, L. (1998) Purification and characterization of a group of five novel peptide serine protease inhibitors from ovaries of the desert locust, Schistocerca gregaria, FEBS Lett. 422, 74278. Hergenhahn, H. G., Hall, M. & Söderhall, K. (1988) Purification and characterization of an A2-macroglobulin-like proteinase inhibitor from plasma of the crayfish Pacifastacus leniusculus, Biochem. J. 255, 8012806. Kellenberger, C., Boudier, C., Bermudez, I., Bieth, J. G., Luu, B. & Hietter, H. (1995) Serine protease inhibition by insect peptides containing a cysteine knot and a triple beta-sheet, J. Biol. Chem. 270, 25 514225 519. Kromer, E., Nakakura, N. & Lagueux, M. (1994) Cloning of a Locusta cDNA encoding a precursor peptide for two structurally related proteinase inhibitors, Insect Biochem. Molec. Biol. 24, 3292331. Laskowski, M. & Kato, I. (1980) Protein inhibitors of proteinases, Annu. Rev. Biochem. 49, 5932626. Mer, G., Kellenberger, C., Koehl, P., Stote, R., Sorokine, O., Van Dorsselaer, A., Luu, B., Hietter, H. & Lefevre, J. F. (1994) Solution structure of PMP-D2, a 35-residue peptide isolated from the insect Locusta migratoria, Biochemistry 33, 15 397215 407. Mer, G., Hietter, H., Kellenberger, C., Renatus, M., Luu, B. & Lefevre, J. F. (1996) Solution structure of PMP-C : a new fold in the group of small serine proteinase inhibitors, J. Mol. Biol. 258, 1582171. Nakakura, N., Hietter, H., Van Dorsselaer, A. & Luu, B. (1992) Isolation and structural determination of three peptides from the insect Locusta migratoria. Identification of a deoxyhexose-linked peptide, Eur. J. Biochem. 204, 1472153. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989) Molecular cloning: a laboratory manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Sasaki, T. (1984) Amino acid sequence of a novel Kunitz-type chymotrypsin inhibitor from hemolymph of silkworm larvae, Bombyx mori, FEBS Lett. 168, 2272230. Takagi, H., Narumi, H., Nakamura, K. & Sasaki, T. (1990) Amino acid sequence of silkworm (Bombyx mori) hemolymph antitrypsin deduced from its cDNA nucleotide sequence: conformation of its homology with serpins, J. Biochem. 108, 3722378. Von Heijne, G. (1991) Computer analysis of DNA and protein sequences, Eur. J. Biochem. 199, 2532256.
