Storage proteins are present in the hemolymph from larvae and adults of the Colorado potato beetle.код для вставкиСкачать
Archives of Insect Biochemistry and Physiology 20:119-I 33 (1992) Storage Proteins Are Present in the Hemolymph From larvae and Adults of the Colorado Potato Beetle Bertha Koopmanschap, Hans Lammers, and Stan de Kort Department of Entomology, Agricultural University, Wqeningen, Netherlands The protein composition of larval and adult hemolymph from the Colorado potato beetle, Leptinotarsa decernlineata, was investigated and some abundant, high molecular weight proteins were identified and characterized. Diapause protein l , which occurs in the hemolymph of last instar larvae and short-day adults, appeared to be a storage protein. This protein dissociated into two bands due to the high pH used in nondenaturing gels. Its quaternary structure was established by chemical crosslinking. It appeared to be a hexamer. Diapause protein 1 is composed of = 82,000 subunits. The amino acid composition and N-terminal sequence of this protein has been determined. Specific antibodies against diapause protein 1 have been developed. Topical application of 1 kg pyriproxyfen, a juvenile hormone analog, to last instar larvae and short-day adults suppressed the appearance of this protein in the hernolymph. Pyriproxyfen prematurely induced vitellogenin, when applied to last instar larvae. A larval specific protein was also identified i n the hemolymph. Its temporary appearance in the hemolymph of last instar larvae, its subunit composition (M, * 82,000) and its suppression by pyriproxyfen suggests that this protein is a storage protein as well. @1992WiIey-Liss, inc. Key words: diapause proteins, hemolymph proteins, larval specific protein, pyriproxyfen INTRODUCTION The protein compositionof the hernolymphfrom adult Leptinotmsa ~ e c e ~ ~ ~ ~ e u ~ f f varies in beetles reared under different photoregimes. Under long-day condi- Acknowledgments: We thank Wallace Clark (Biotechnology Center, University of Arizona, Tucson) for the amino acid analyses and N-terminal sequencing of the diapause proteins. The critical remarks on the manuscript by Dr. john Law (University of Arizona, Tucson) and two unknown referees are highly appreciated. Received September 16,1991;accepted January31,1992. Address reprint requests to C.A.D. de Kort, Department of Entomology, Agricultural University, P.O. Box 8031,6700 EH Wageningen, The Netherlands. 0 1992 Wiley-Liss, Inc. 120 Koopmanschap et al. tions (18 h L*/6 h D), females start oviposition 5 days after adult emergence, which is reflected by the appearance of vitellogenin in the hemolymph from day 2 onwards. Under short-day conditions (10 h LA4 h D), which lead to diapause 11-12 days after emergence, diapause proteins (short-day proteins) are predominant in the hernolymph, whereas vitellogenin is hardly detectable [l-31. Diapause proteins and vitellogenin are major proteins in the hemolymph of this beetle. Another major protein in the hemolymph is lipophorin, the primary lipid transport protein of insect hemolymph. Lipophorin from the Colorado potato beetle was isolated and its molecular composition has been characterized. It functions not only as a lipid transport vehicle, but in this beetle it also transports JH .Thus the functions of vitellogenin and lipophorin are relatively well known. Less information exists about the function of diapause proteins. De Loof  identified by disk electrophoresis three different diapause proteins in the hemolymph of beetles reared under short-day conditions. One of these proteins (diapause protein 1)also occurs in last instar larvae. It disappears towards the end of the pupal stage, suggesting that it is identical to so-called storage proteins, described for other species [5-71. Dortland  arrived at the same conclusion after he observed that diapause protein 1was stored in the fat body. Insect storage proteins have been well studied in Diptera and Lepidoptera, and different storage protein classes can be distinguished (for review see ). Recently, Telfer and Kunkel  introduced the name hexamerin as a descriptive and generic term that covers all hexamers of arthropods with M, of approximately 500,000, the most prominent being the insect storage hexamers and the hemocyanins. There is no information about the existence of hexamerins in Coleoptera. Peferoen et al.  have reported the molecular weights of the subunits of the most abundant proteins in the hemolymph from the Colorado potato beetle, but their data do not reveal any value typical for insect storage hexamers. Because data about the molecular weights of the native proteins do not exist, we investigated these characteristics for a number of abundant hemolymph proteins from the Colorado potato beetle. Our studies revealed that diapause protein 1from the Colorado potato beetle may be considered as an insect storage hexamer, which appearance is suppressed by JH. MATERIALS AND METHODS Insects A laboratory strain of the Colorado potato beetle, Leptinotarsa decemlineata Say, was reared on fresh potato foliage at 25°C under long-day (18 h L/6 h D) or at 23°C under short-day (10 h L/14 h D) conditions. Fourth instar larvae *Abbreviations used: ApoLp I = apolipophorin I; apoLp I I = apolipophorin I I ; D = darkness; DSP = dithiobis(succinimidy1propionate);JH = juvenile hormone; JHA = juvenile hormone analog; L = light on; LSP = larval specific protein; PBS = phosphate buffered saline; PMSF = phenylmethylsulfonyl fluoride; PTC-AA = phenylthiocarbamyl-amino acid; PTH = phenylthiohydantoin; S D S = sodium dodecyl sulfate. Hemolymph Storage Proteins 121 used in our experiments were taken from long-day cultures only. Under these conditions, fourth instar larvae stop feeding 6 days after the last larval molt and display digging behavior in preparation for pupation. Pupation takes place in the soil and adults emerge 11-12 days after larval digging. Under long-day conditions, newly emerged beetles immediately start feeding, mate between day 3 and 4, and oviposit from day 5 onwards. Under short-day conditions, adults also feed intensively, but do not show reproductive behavior. They stop feeding and display digging behavior between days 11 and 12, indicating the onset of diapause. Diapause lasts for at least 3 months. Hemolymph was collected in capillary pipettes from precisely aged larvae or adults after clipping a leg. The hemolymph samples were immediately diluted to one-half in cold buffer containing 50 mM phosphate (pH 7.4),150 mM NaC1, 10 mM EDTA, 0.1 mM paraoxon, 0.1 mM PMSF, and a few grains of phenylthiourea. The protease inhibitor and paraoxon were kept in stock solutions in ethanol and the solvent was evaporated immediately before dilution of the hemolymph sample. Hemocytes were sedimented by centrifuging at 10,OOOg for 4 min and the hemolymph was used immediately or stored at - 20°C. The JHA, pyriproxyfen,2-[ 1-methyl-2-(4-phenoxyphenoxy)-ethoxy]pyridine, was a 10% emulsifiable concentrate, supplied by Dr. H. Oouchi, Sumitomo Chemical Co., Ltd., Osaka, Japan. Controls were treated with emulsifier solution, without JHA. Both solutions were stored at 4°C. For each application, fresh solutions were prepared in acetone and applied to the insects in a volume of 1~ 1using , a microsyringe, operated with a repeating dispenser. Gel Electrophoresis Nondenaturing PAGE, using 7% continuous or 5.5-20% gradient slab gels, was performed horizontally as described before IS], except that the buffer was changed when indicated. SDS-PAGE was also performed horizontally with 6-20% slab gels as described elsewhere  or vertically with the BIO-RAD Protean I1 slab cell (Richmond, CA), according to the instructions of the manufacturer. The hemolymph samples for SDS-PAGE were further diluted with Laemmli [lo] sample buffer containing 5% P-mercaptoethanol and heated to 100°C for 2 min, before application to the slots. The molecular weights of the proteins were estimated by comparison with standard proteins of high and low molecular weight (Pharmacia, Uppsala, Sweden), using regression analysis. Gel Permeation Chromatography Chromatography was carried out at room temperature in a 2.6 x 40 cm column packed with Biogel A-1.5 m (BIO-RAD). The elution buffer was 50 mM phosphate buffer (pH 7.2), containing 0.15 M NaCl and 0.02% sodium-azide. Samples (ca. 30 mg of protein) were eluted at a flow of 30 mllh, controlled by a LKB 2132 microperpex pump (LKB-Produkter AB, Bromma, Sweden) and the absorbance of the eluant was monitored continuously at 280 nm using a LKB 2138 Uvicord S.Fractions (2.5ml each) were collected with a LKB Ultrorac fraction collector and the appropriate fractions were pooled and concentrated by ultrafiltration, using an Amicon (Grace & Co., Amicon Division, Beverly, MA) stirred cell system. 122 Koopmanschap et al. Chemical Crosslinking To study the quaternary structure of diapause protein 1, purified protein (100 pg) was incubated for 1 h at room temperature in 100 pl of 0.2 M triethanolamine buffer (pH 7.5 or 8.3) containing 0.15 M NaCl and 0.5 or 0.05 mglml DSP. After incubation 30 p1 of the reaction mixtures was diluted with twofold concentrated SDS sample buffer without P-mercaptoethanol, boiled for 2 min, and then applied to a 3-8% SDS gradient gel. Immunology Antisera were prepared against LSP and diapause proteins 1A and lB, which were designated as antisera 311, 291, and 292, respectively, The purified antigens were dissolved in 0.9% NaC1, emulsified with an equal volume of Freund’s complete adjuvant, and injected subcutaneously into rabbits. After 5 weeks, the same amount of antigen was emulsified in Freund’s incomplete adjuvant and again injected. The same injection was repeated after another 3 weeks. Eight days after the second booster injection, 40 ml of blood were collected from each rabbit and serum was separated from the clot after one night in the cold room. It took 5 booster injections to prepare a suitable antiserum against LSP. The protein fractions used for immunization were purified by PAGE and eluted from the gels with a BIO-RAD Mode1 422 electro-eluter, using a volatile ammonium bicarbonate buffer containing 0.1% SDS, according to the instructions of the manufacturer. Twenty slots, each containing 2 p1 of hemolymph, were run in a nondenaturing 5.5-20% gradient slab gel as described above. After staining of the gel for 30 min in 0.1% Coomassie R-250, dissolved in 10% acetic acid, 40% methanol, and destaining of the gel in the same solvent, the bands of interest were cut from the gel, sliced into small pieces, and eluted for 4-5 h with the electro-eluter. The eluted proteins were collected in 500 pl of elution buffer per elution tube, transferred to microcentrifuge tubes, and freezedried. The dried protein fractions were dissolved in a small volume of SDS sample buffer by heating at 100°C for 2 min and subsequently subjected to SDS-PAGE, using a 6-20% gradient gel. After another round of staining and destaining, the = 80,000 protein bands from each fraction were cut from the gel, sliced, and eluted by the same procedure. After freeze-drying, the protein fractions were dissolved by boiling in a small amount of SDS sample buffer and further diluted with 750 ~ 1 0 . 9 % NaCl. This fraction was used for injection after emulsification with Freund’s adjuvant. This protein fraction contained a single band after Coomassie staining of Western blots from an SDS gel. Immunological tests were performed on whole hemolymph protein Samples or on purified protein samples after separation by SDS-PAGE and Western blotting of the proteins to Immobilon transfer membranes (Millipore, Bedford, MA), as described before .After the transfer step, unoccupied binding sites of the transfer membrane were blocked by incubating the membranes for 45 min at 37°C in a solution of 5% ( w h ) of skimmed milk powder in PBS (10 mM phosphate (pH 7.4), 0.8% NaC1, 0.02% KC1, and 10 mM NaN3). After blocking, the membranes were washed 3 times in PBS containing 0.1% skimmed milk for 5 min, and subsequently for 2 h at room temperature in the same Hemolymph Storage Proteins 123 solution containing the primary antibody diluted by a factor of 2,000. After this incubation, the blots were rinsed 3 times for 5 min in 0.1% skimmed milk PBS solution. Detection of the immunologically responsive proteins occurred with gold-labelled goat antirabbit immunoglobulin G (Janssen Life Sciences Products, Beerse, Belgium), essentially as described before [4J. Amino Acid Analysis and N-Terminal Sequencing Samples were analyzed using an Applied BioSystems (Foster City, CA)Model 420A DerivatizerA30A Separation/92OAData Analyzer with automatic hydrolysis (Vapor Phase at 160°Cfor 1h 40 min) using pre-column PTC-AA analysis. Samples were sequenced using an Applied BioSystems 477A ProteidPeptide Sequencer (Edman chemistry) interfaced with a 120A HPLC (C-18 PTH, reversephase chromatography) Analyzer to determine PTH amino acids. RESULTS SDS Gel Electrophoresis Figure 1 illustrates an SDS gel of hemolymph samples from different aged larvae, short-day adults, and long-day males and females. The pattern shows that most protein bands are similar in larvae and adults and in long-day and short-day beetles. A number of these bands can be identified. ApoLp I and apoLp I1 are indicated. Their M,s are 240,000 ? 10,000 and 77,000 ? 4,000 (n = 5), respectively, when calculated by comparison with the mobility of standard proteins. ApoLp I and apoLp I1 are absent from larval samples, because they were frozen before use. In freshly prepared larval hemolymph, these apoproteins are normally present in SDS gels. Two vitellogenin subunits can be identified with M,s of 199,000 k 5,000 and 162,000 h 5,000 (n = 5 ) , respectively. A prorninent band (indicated **) is visible adjacent to apoLp 11. It is present in fullgrown larvae and increases in concentration in short-day beetles, but is absent in hemolymph from long-day beetles and day zero larvae. The M, of this protein band is = 82,000 ? 2,000 (n = 4). This, together with the appearance during certain developmental stages, suggests that this protein might be related to insect storage hexamers. To investigate this, hernolymph samples from different stages were studied by nondenaturing gel electrophoresis. Nondenaturing PAGE Figure 2 illustrates the separation of hemolymph proteins from larvae and short-day and long-day adults by a conventional 7% nondenaturing slab gel. Three bands of relative high M, (M, 2 300,000) are present in the hemolymph from larvae or short-day adults, which do not occur in long-day males or females. These proteins are designated LSP (larval specific protein) and diapause protein 1A and lB, respectively. In addition, two other proteins are present in the hemolymph of diapausing adults, whose M,s are relatively small on the basis of their mobility in the gel. These are designated diapause protein 2 and 3, respectively. Their M,s are too small to conform to the definition of a hexamerin and have not been studied further. Figure 3 illustrates the protein pattern of hemolymph samples from longday males and females, last instar larvae, and short-day adults separated in a 124 Koopmanschap et al. 2. Fig. 1. SDS gradient gel of hemolymph samples from long-day females, day Iand day 6 lastinstar larvae (L4), and from different aged (2,6,10,90 days) short-day (SD) adults. S = standard proteins of M, given in kDa. Apolp I and II = apolipophorin I and II; Vg = vitellogenin; ** = 82,000 subunit. The following amounts of hernolymph were applied: long-day female: 1 FI. Larvae day 1 : 2 pi; day 6: 1 pi. Short-day beetles, day 2: 1 ~ lday ; 6 : 0.75 PI; day 10 and 90:0.5 pl. Fig. 2. Conventional 7% slab gel of hemolymph samples from larvae (L4),diapausing adults (D), and long-day males and females. S = high molecular weight standard proteins. Their M+ are given in kDa. LSP = larval specific protein; 1A and 1B = diapause protein 1A and 16;2 and 3 = diapause protein 2 and 3. In each slot 2 p1. of hemolymph was applied, except in D,where 1 pl was applied. Hernolymph Storage Proteins 125 Fig. 3. Nondenaturing gradient gel (5.5-20%) of hernolymph samples from long-day males and females, last-instar larvae (14) of day 1 or day 6, and different aged (2,6,10,90 days) shortday (SD) adults. S = standard proteins (see Fig. 2). Lp = lipophorin; LSP = larval specific protein; 1Aand 1B = diapause protein 1A and 16. The following amounts of hemolymph were applied: long-day males and females: 1 PI. Day 1 larvae: 2 PIand day 6 larvae: 0.5 PI. Short-day beetles, day 2: 2 pl; day6: 1.5 pl; daylOand90: 1 PI. nondenaturing 5.5-20% gradient gel at pH 8.9. Gradient gels offer a better possibility to determine the M, of native proteins [ l l ] . The M, values of the three high M, proteins were calculated by comparison with standard proteins. LSP has a M, of = 614,000 5 29,000 (n = 6). The values for diapause protein 1A and 1B are = 465,000 4,000 (n 6) and 347,000 2 20,000 (n = 6), respectively. The M, of vitellogenin was also calculated and was = 583,000 -I 18,000 (n = 7). Diapause proteins 1A and 1B show unusual behavior in nondenaturing gels at high pH. Comparison of Figures 2 and 3 indicates differences in staining intensities, with diapause protein 1B more prominent in the gradient gel. Since pH-dependent dissociation of insect storage hexamers is well known , we ran another nondenaturing gradient gel at pH 8.3 instead of 8.9. Hemolymph from different aged short-day beetles was used in this experiment. The results are illustrated in Figure 4.At this pH, diapause protein 1A is much more prominent, which suggests that diapause protein 1B is a dissociation product of diapause protein 1A rather than another protein. To prove this more directly, the high molecular weight protein bands from gradient gels (LSP, diapause protein 1A and 1B) were sliced after rapid staining and destaining of the gel and electroeluted. After electroelution and freezedrying, the protein samples were subjected to SDS gradient gel electrophoresis, transferred to Immobilon, and stained by Coomassie. As shown in Figure 5A, 126 bopmanschap et al. Fig. 4. Nondenaturing gradient gel (5.5-20%) of hemolymph from different aged short-day beetles carried out at pH 8.3. The ages of the beetles and the amount of hemolymph applied are similar to Figure 3. Abbreviations as in Figure 3. the three protein bands all contain a single type of subunit of M, = 82,000. Next, antibodies against the subunits of the three proteins were developed and tested on Western blots from SDS gels, containing the following protein samples: electroeluted bands from native gels representing LSP, diapause protein 1A and lB, whole hemolymph from last instar larvae, diapausing beetles, and long-day females. The results with antidiapause protein 1A serum (serum 291) are illustrated in Figure 5B. It can be seen that it responds with the subunits from diapause protein 1A and lB, but not with the subunit of LSP. Also antidiapause protein 1 B serum (antiserum 292) crossreacted with the subunits of 1A and lB, but not with LSP (not shown). Moreover, both antisera responded with a single band of whole hemolymph from full-grown larvae and short-day adults, but not with hemolymph from long-day females. Anti-LSP serum reacted with the subunit of LSP, with a single band of whole larval hemolymph, but not with hemolymph from day zero larvae or shortday adults (not shown). These studies suggest that diapause protein XA and 1B are composed of immunologicallyidentical subunits and that LSP is another hemolymph protein. Further proof that diapause proteins 1A and 1B contain the same subunit was derived from analysis of the amino acid composition and N-terminal amino acid sequences of the two protein bands. The amino acid compositions of diapause proteins 1A and 1B are very similar (Table l), whereas the amino acid sequence of the first 20 amino acids of both proteins Hernolyrnph Storage Proteins 127 Fig. 5. A: Western blots of SDS gradient gels of protein fractions electroeluted from native gradient gels and stained with Coomassie R-250. 6: The same fractions plus whole hemolymph from last-instar larvae (L4) and diapausing (D) and long-day females were immunologically stained with antidiapause protein 1A serum (291). The following amounts of protein fractions were applied: A: LSP, equivalents of 0.8 pl of hernolymph; I A , equivalents of 2.0 pl; I B , equivalents of 1 .O kl. B: LSP, equivalents of 0.8 pI of hernolymph; I A , equivalents of 0.4 pl; I B , 0.2 PI; L4,0.25 PI of hernolymph; D, 0.25 pl; females, 0.5 pl of hernolymph. Abbreviations as in Figure 3. is exactly the same: Asn-pro-val-ala-asp-thr-asn-tyr-leu-lys-?-glu-gln-~n-ileleu-lys-leu-leu-tyr. The question mark on position 11 is probably cys or a glycosylated amino acid. From these experiments it can be concluded that diapause protein 1 B is a dissociation product of diapause protein 1A, with the dissociation probably due to the high pH used during electrophoresis of nondenaturing gels. Structural Aspects of Diapause Protein 1 Since diapause protein 1 yielded two bands in nondenaturing gels, the question arises whether different forms of the protein occur under physiological conditions. To study this, diapause protein 1was purified without using electrophoresis and subsequently chemically crosslinked. For this purpose hemolymph from diapausing beetles was centrifuged in a KBr gradient . After removal of lipophorin, the rest of the KBr gradient was diluted with phosphate buffered saline and concentrated by ultrafiltration. After another round of dilution and concentration, to remove the bulk of KBr, the fraction was applied to a gel permeation column. The first peak from this column appeared to be pure diapause protein 1, which was verified by PAGE. After SDS-PAGE this fraction yielded one band of M, = 82,000 (Fig. 6, lane 1). If this fraction is incubated with high concentration of crosslinker (0.5 mg/ml), one band of high 128 Koopmanschapet al. TABLE 1. Amino Acid Composition of Diapause Protein 1A and 1B From the Colorado Potato Beetle Amino acid Aspartic acid Glutamic acid Serine Glycine Histidine Arginine Threonine Alanine Proline Tyrosine Valine Methionine Cysteine Isoleucine Leucine Phenylalanine Lysine mol % protein 1A mol % protein 1B 13.77 11.78 5.26 8.33 2.99 6.62 3.40 14.05 11.28 5.84 10.31 2.86 6.18 3.57 5.91 5.45 6.82 4.59 0.94 0.15 3.14 6.31 7.03 5.59 6.03 5.15 7.32 5.02 0.68 0.17 3.56 6.85 7.53 5.53 M, is visible in the SDS gradient gel (Fig. 6, lane 3). At 10 times lower crosslinker concentration, six bands are apparent, which indicates that diapause protein 1 only occurs as a hexamer (Fig. 6, lane 2). Thus dissociation is probably an artefact due to electrophoresis. Effect of a JHA Diapause protein 1occurs in the hemolymph of last instar larvae and shortday adults, two stages which are characterized by low JH titers . To study whether or not JH affects the appearance of diapause protein 1in the hemolymph, a JHA was applied topically to last instar larvae and short-day adults and the effect on the protein pattern of the hemolymph was studied by nondenaturing and SDS-PAGE. Nondenaturing gel electrophoresis was carried out at pH 8.3. Pyriproxyfen was used as JHA, because it had strong juvenilizing effects in larvae and adults of this beetle [13,14]. Larvae and young adults were treated topically with 1 pg of pyriproxyfen on day 1 and day 3. The protein pattern of the hemolymph was studied on day 6. The results with larval hernolymph are illustrated in Figure 7. It can be seen from the native gel (Fig. 7A) that the intensity of LSP and diapause protein 1decreased after JHA treatment. In addition, a new band appeared with mobility very close to LSP. This band is probably vitellogenin. Since LSP and vitellogen migrate very close together in native gels, this region of the gel was sliced and electroeluted. After electroelution, this fraction was subjected to SDS electrophoresistogether with whole hemolymph from controls and pyriproxyfen treated larvae (Fig. 7B). The 82,000 subunit decreased in intensity after pyriproxyfen treatment and two bands similar to the vitellogenin subunits are present. The lane with the eluant from control larvae only contained the 82,000 subunit, whereas in the lane with the sample from JHA treated larvae two additional bands are visible with apparent M, similar to vitellogenin. That these subunits are indeed derived from vitellogenin was shown by blotting and immunostaining using HemolymphStorage Proteins 129 Fig. 6. SDS-PACE of diapause protein 1 after incubation with two concentrations of DSP. Diapause protein l (100 pg) was incubated with 0 (1),0.05(2)and 0.5 (3) mg/ml of crosslinker. After 1 h the reaction was stopped by addition of twofold concentrated SDS sample buffer without p-mercaptoethanol and boiled for 2 min. An aliquot of each sample was electrophoresed at 30 mA in a 3-8% SDS gradient gel. an antiserum developed against adult vitellogenin. The two vitellogenin subunits responded positively with the antiserum (not shown). DISCUSSION This paper describes the identification and characterization of some high molecular weight proteins in the hemolymph of the Colorado potato beetle, Leptinotarsa decemlineata . Diapause protein 1, which has been reported previously by de Loof , Dortland , and Peferoen et al. [ti],occurs as two separate bands on conventional nondenaturing slab gels and 5.5-20% gradient gels used in the current study. With a high pH buffer system, diapause protein 1B stained more strongly than diapause protein 1A (Figs. 2, 3). At lower pH, diapause protein 1A was more prominent (Fig. 4). Subsequent experiments showed that the two bands are probably derived from the same native protein. Following electroelution, the two protein bands appeared to be composed of subunits of similar M, (= 82,000), which also crossreacted immunologically (Fig. 5A,B). Determination of the amino acid composition (Table 1) and the 130 Koopmanschap et al. Fig. 7. Nondenaturing gradient gel with pH 8.3 electrophoresis buffer (A) and a 7% SDS gel (6)of hernolymph samples (H) from control ( - 1 and pyriproxyfen ( ) treated larvae. Larvae were treated on day 1 or day 3 and the hemolymph composition was analysed on day 6. All slots contained 1 pI of hernolymph. Areas of the native gel containing LSP were electroeluted from hernolymph samples of control (E -1 and pyriproxyfen (E +)treated larvae, lyophilized, boiled in SDS sample buffer and subjected to electrophoresis. Lane L D 0 contains 1 PIof hemolymph from long-day females. Apolp I and Apolp II = apolipophorin I and II, respectively; S = standard proteins (see Fig. 2); Vg = vitellogenin; ** = 82,000 subunit. + N-terminal sequence revealed that the two subunits are probably the same. Calculation of the M, of the two protein bands suggests that diapause protein 1A is a hexamer (M, = 465,000) and 1B is a tetramer (M, 347,000). Dissociation, due to high pH, is well known for insect storage hexamers, particularly in Diptera . The combination of high pH (pH = 8.9 at room temperature) and low ionic strength leads to dissociation of subunits, particularly during long runs used for gradient gels (Fig. 3). At lower pH, dissociation of the hexamer is much less (Fig. 4).The occurrence of the protein in different bands is probably an artefact due to nondenaturing electrophoresis. Direct proof that diapause protein l occurs only as a hexamer is derived from chemical crosslinking experiments. Prior to chemical crosslinking, diapause protein 1was purified in two steps without the use of electrophoresis. The purity of the protein was checked by SDS-PAGE. The fraction appeared to contain one protein band after SDS-PAGE (Fig. 5, lane 1).Subsequently, the purified fraction was subjected to chemical crosslinking, using two concentrations of crosslinker. With high concentration (0.5mg/ml), one band was visible after SDS gradient PAGE (Fig. 5, lane 3). If hexamers and tetramers are present together in the sample, chemical crosslinking would result in two bands with mobilities comparable to a hexamer and a tetramer, respectively. With limiting concentrations of crosslinker i= HemolymphStorage Proteins 131 (0.05mg/ml), six bands appeared after SDS-PAGE (Fig. 5, lane 2), which gradually decrease in intensity towards the higher oligomers. In addition, if the oligomer numbers are plotted semilogarithmicallyagainst the distance of migration in the gel, a straight line is obtained, which strongly indicates that the bands are different associations of the same monomer. Determination of the M, of the different oligomers, by comparison with standard proteins, is not possible, because interaction with the crosslinker affects the mobility of proteins during electrophoresis. Thus, this experiment demonstrates that diapause protein 1, purified by a method without electrophoresis, occurs in a hexameric form of = 82,000 subunits. The M, of the protein should be around 500,000. Determination of the M, by native gel electrophoresis is therefore not accurate. The hexameric configuration is direct proof that this protein belongs to the group of hexamerins according to Telfer and Kunkel , and resembles the insect storage hexamers in a number of biological and biochemical characteristics. It accumulates in the hemolymph during the last larval instar and its concentration decreases during metamorphosis . The protein also accumulates in the hemolymph of adults reared under diapause-inducing (short-day) conditions (Figs. 1-3). During diapause it is present in the fat body as well [15,161. The amino acid composition reveals a relatively high tyrosine/phenylalanine content (Table l), although the values are not as high as in storage proteins from Diptera [6,7]. The fact that diapause protein 1 occurs in the hemolymph of larvae and adults was shown previously by de Loof and co-workers [2,8]. Recently, Wyatt  described the existence of a persistent storage protein in larvae and adults of the migratory locust, Locusta migruforia. Apparently, storage proteins may exist in adults, at least under certain physiological conditions. The production of diapause protein 1is suppressed by a JHA, pyriproxyfen, when applied to last instar larvae (Fig. 7) or short-day adults . In a number of insect species, synthesis of storage hexamers is suppressed by JHA application to last instar larvae [18,19]. We recently described the suppression of a larval storage protein by pyriproxyfen in last instar larvae of the migratory locust . Disappearance of the larval protein under the influence of JHA was accompanied by the precocious induction of vitellogenin in larvae. Our results confirmed an earlier report on vitellogenin induction in larvae of Locusfa migruforia . The experiments illustrated in Figure 7 demonstrate that pyriproxyfen precociously induced vitellogenin when applied to last instar larvae of the Colorado potato beetle. This is the first report of vitellogenin induction in larvae of holometabolous insects. The presence of vitellogenin in larval hemolymph was demonstrated by nondenaturing and SDS-PAGEand by imrnunostaining. Vitellogenin normally occurs in the hemolymph of long-day females after day 2. Its M,, determined by nondenaturing gels, is = 580,000, which is similar to values found in other species . After SDS-PAGE, two vitellogenin subunits can be identified, with M,s of 199,000 and 162,000, respectively. The same bands are present in larval hemolymph after application of 1pg pyriproxyfen, but not in hemolymph from controls (Fig. 7). Both vitellogenin subunits responded on Western blots with an antiserum developed against adult vitellogenin. A second storage protein, which is larval specific, is probably LSP. It is also 132 Kooprnanschap et al. composed of -- 82,000 subunits, which were immunologically different from the diapause protein subunits. A specific antiserum (serum 311) against this protein has been developed. It responded with larval hemolymph, but not with adult hemolymph. The apparent M, was determined by nondenaturing gels to be = 614,000, which is too high for a hexamer with 82,000 subunits. However, determination of the M, by native gel electrophoresis can be illusive, because the method measures the Stoke’s radius of the protein and can yield misleading results if applied to proteins that are not spherical [ 6 ] .Moreover, the native protein may contain ligands, which affect the size of the molecule. One of us (H.L.) recently found that LSP contains 15% lipid. On this basis, LSP is probably a hexamer, but further research on this point is required. Lipophorin, the lipid transport and JH carrying protein of the hernolymph, has been extensively discussed elsewhere .Its apoprotein composition has been confirmed here. Larval lipophorin contains the same type of apoproteins as adult lipophorin, but larval lipophorin is very sensitive to freezing and thawing, which damages the protein moiety of the protein. Its mass density (1.098 g/ml), determined by KBr gradient centrifugation , is similar to adult lipophorin, and its intense red color suggests that it contains high concentration of carotenoids. LSP has not been noticed by de Loof and co-workers [2,8], despite the fact that de Loof  claimed that the band which he referred to as diapause protein 1is composed of several proteins. Our results differ from their data with regard to the molecular weights of the subunits (apoproteins) of most of the hemolymph proteins. They reported a molecular weight of 114,000 for the subunit of diapause protein 1. Other data summarized in their Table 1  are also difficult to reconcile with our measurements. For example, they reported that the most apparent chromoprotein (chromoprotein 2) is colored by orange carotenoid pigments, constitutes 10-28% of the total protein concentration, and occurs in eggs and hemolymph. We conclude from this that chromoprotein 2 is identical with lipophorin. They found three subunits (apoproteins)in this fraction, yet the values of the M, of these apoproteins (356,000; 95,500; 90,000) do not agree with our data 141 nor with those from other lipophorins [ 5 ] .We do not have an explanation for these discrepancies. LITERATURE CITED 1. de Loof A, de Wilde J: The relation between haemolymph proteins and vitellogenesis in the Colorado beetle, Leptinotarsa decemlineata. J Insect Physioll6, 157 (1970). 2. de Loof A: Diapause phenomena in non-diapausing last instar larvae, pupae, and pharate adults of the Colorado beetle. J Insect PhysiolZ8, 1039 (1972). 3. Dortland JF: Synthesis of vitellogenin and diapause proteins by the fat body of Leptinotarsa, as a function of photoperiod. Physiol Entomol3,281 (1979). 4. de Kort CAD, Koopmanschap AB: Molecular characteristics of liuouhorin, the iuvenile hormonebinding protein in-the hemoiymph of the Colorado potato beeie.*ArchInseci Biochem Physiol 5, 255 (1987). 5. Kanost MR, Kawooya JK, Law JH, Ryan RO, van Heusden MC, Ziegler R: Insect haemolymph proteins. Adv Insect Physiol22,299 (1990). HemolymphStorage Proteins 133 6. Telfer WH, Kunkel JG: The function and evolution of insect storage hexamers. Annu Rev Entomol36,205 (1991). 7. Levenbook L: Insect storage proteins. In: Comprehensive Insect Physiology. Kerkut GA, Gilbert LI, eds. Pergamon Press, Oxford, vol. 10, pp 307-346 (1985). 8. Peferoen M, Stynen D, de Loof A: A re-examination of the protein pattern of the hemolymph of Leptinotarsa decernlineutu with special reference to vitellogenins and diapause proteins. Comp Biochem Physiol [B] 72,345 (1982). 9. de Bruyn SM, Koopmanschap AB, de Kort CAD: High-molecularweight serum proteins from Locusfa migvaforia:Identification of a protein specificallybinding juvenile hormone 111. Physiol Entomolll, 7(1986). 10. Laemmli UK:Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227,680 (1970). 11. Telfer WH, Keim PS, Law JH: Arylphorin, a new protein from Hyulophoru cecropia: Cornparisons with calliphorin and manducin. Insect Biochem 13,601 (1983). 12. de Kort CAD Thirty-five years of diapause research with the Colorado potato beetle. Entomol Exp Appl56,1(1990). 13. Koopmanschap AB, Oouchi H, de Kort CAD: Effects of a juvenile hormone analogue on the eggs, post-embryonic development, metamorphosis and diapause induction of the Colorado potato beetle, Leptinotarsa decemlineutu. Entomol Exp Appl50,255 (1989). 14. de Kort CAD, Koopmanschap AB: Effects of a juvenile hormone analogue on development, metamorphosis and diapause induction of the Colorado potato beetle. In: Advances in Invertebrate Reproduction, Hoshi M, Yamashita 0, eds. Elsevier Science Publ, Amsterdam, vol. 5, pp 383-386 (1990). 15. de Loof A, Lagasse A: Juvenile hormone and structural properties of the fat body of the adult Colorado beetle, teptinotursa decernlineutu Say. Z Zellforsch 106,439 (1970). 16. Dortland JF, Hogen Esch T: A fine structural survey of the development of the adult fat body of Leptinotarsa decemlineutu. Cell Tissue Res 201,423 (1979). 17. Wyatt GR: Developmental and juvenile hormone control of gene expression in locust fat body. In: Molecular Insect Science. Hagedorn HH, Hildebrand JC, Kidwell MG, Law JH, eds. Plenum Press, New York, pp 143-172 (1990). 18. Tojo S, Kiguchi K, Kimura S: Hormonal control of storage protein synthesis and uptake by the fat body in the silkworm, Bombyx rnori. J Insect Physiol27,491(1981). 19. Jones G, Hiremath ST, Hellmann GM, Wozniak M, Rhoads RE: Juvenile hormone regulation of mRNA levels for a highly abundant hemolymph protein in larval Trichoplusiu ni. J Biol Chem 263,1089 (1988). 20. de Kort CAD, Koopmanschap AB: A juvenile hormone analogue affects the protein pattern of the haemolymph in last instar larvae of Locusfa rnigrutoriu.J Insect Physiol37,87 (1991). 21. Dhadialla TS,Wyatt G R Juvenilehormone-dependent vitellogenin synthesis in Locustu miptoria fat body: Inducibility related to sex and stage. Dw Biol96,436 (1983). 22. Hagedorn HH, Kunkel JG: Vitellogenin and vitellin in insects. Annu Rev Entomol24, 475 (1979).