The polypeptide structure of Vitellogenin and Vitellin from the cockroach Leucophaea maderae.код для вставкиСкачать
Archives of Insect Biochemistry and Physiology 3:577-591 (1986) The Polypeptide Structure of Vitellogenin and Vitellin From the Cockroach, Leucophaea maderae Victor J. Brookes Department of Entomology, Oregon State University, Cowallis Vitellogenin (Vg) synthesized by the fat body of Leucophaea maderae is made u p of four polypeptides with molecular weights of 160,000, 105,000, 98,000, and 57,000. Other polypeptides previously reported as part of Vg are associated with other proteins. Vitellin (Vt), the yolk protein (UP) isolated from mature oocytes and from newly formed oothecae, i s a protein with a sedimentation coefficient of 28s and consists of three polypeptides with molecular weights of 105,000, 85,000, and 57,000. During vitellogenesis, the YP of developing oocytes contains both Vt and a 14s component. The 14s component i s made up of four polypeptides with molecular weights of 105,000, 90,000, 85,000, and 57,000. The data suggest that 14s may not be a discrete protein but rather a form in transition between Vg and Vt in which the 98,000 dalton polypeptide i s converted t o the 85,000 dalton polypeptide of Vt through a 90,000 dalton intermediate. The 160,000 dalton peptide of Vg does not appear to be a part of Vt. Under alkaline conditions, both the 14s component and Vt are reduced to a polypeptide with a lower sedimentation rate i n sucrose gradients. When acid conditions are restored, a protein resembling 14s i s obtained. This suggests that the YP is a loosely held aggregate of similar or identical proteins with a molecular weight of about 250,000. Key words: Leucophaea maderae, polypeptide structure, protein processing, yolk proteins INTRODUCTION During the development of the oocytes, the fat body of insects synthesizes a protein, Vg", which is secreted into the hemolymph, recovered by the *Abbreviations used: 2,5,diphenyloxazole, PPO; 2-mercaptoethanol, PME; polyacrylamidegel electrophoresis, PAGE; sodium dodecylsulfate, SDS; tetramethylethylenediamine, temed; trichloroacetic acid, TCA; Tris(hydroxymethy1) aminomethane, Tris; vitellin, Vt; vitellogenin, Vg; yolk protein, YP. Acknowledgments: I wish to thank Dr. Renk Feyereisen for a critical reading of the manuscript. Dr. Barbara Triplett kindly supplied radioactive globin. The research was supported in part by funds from the Research Office, Oregon State University. Received September 20, 1985; accepted March 21,1986. Address reprint requests to Victor J. Brookes, Department of Entomology, Oregon State University, Corvallis, OR 97331. 0 1986 Alan R. Liss, Inc. 578 Brookes oocytes, and deposited by them into membrane-enclosed vesicles [l-41. In many insects, Vg is synthesized as a primary translate of one or two proteins with a molecular weight of about 250,000 and then modified by the attachment of lipids, carbohydrates, and phosphorous, and by cleavage into two or more polypeptides . In the hemolymph, Vg is a protein with a molecular weight of about 220,000 to 650,000, depending upon the species; in some insects, further modification occurs during deposition . In another group, the primary translate of Vg is smaller, about 170,000, and no cleavage occurs . In still a third group, the primary translate consists of three proteins with molecular weights of about 45,000 . In all insects studied, the protein in the hemolymph is an aggregate of the primary translates or their cleavage products. Nothing is known about the mechanisms of cleavage of primary translates that exist in the majority of insects but appear to have been lost in the evolutionary development of some members of this group of animals. Among species in which cleavage is extensive, the number of polypeptides produced is not clear [7-121. As a preliminary to a study of cleavage mechanisms, the polypeptide structure of Vg and Vt of Leucophaea maderae has been reexamined. In this species, processing occurs both in the fat body and in the follicles [11,12]. The results obtained are similar in some respects to those previously reported [ll,13,141, but additional evidence suggests another interpretation. MATERIALS AND METHODS Animals Leucophaea maderae was reared as previously described . Females were mated on the eighth to tenth day after eclosion, and mating was established by the presence of spermatophore in the bursa. Chemicals Reagents used for electrophoresis, including SDS, acrylamide, temed, and molecular weight markers, were purchased from BioRad (Richmond, CA). Tris buffer and ingredients for tissue culture were purchased from Sigma (St. Louis, MO). Routine chemicals, including PME were Baker Analyzed Reagent. Radioactive chemicals were purchased from New England Nuclear (Boston, MA), PPO from Packard Instruments (Downers Grove, IL), and Triton N-101 from Rohm and Haas (Philadelphia, PA). Purification of Yolk Proteins The following definitions describe the various proteins associated with the yolk: 1. YE' represents the proteins extracted from developing oocytes or from newly formed oothecae and purified by dilution with water. Purified YP consists of two components, 14s and 28s, which are obtained by centrifugation through a 10-30% sucrose gradient . 2. Vt is the mature yolk protein and is purified as described above. Yolk Proteins From Leucophaea maderae 579 3. Vg is the protein secreted by the fat body during development of the oocytes and is the precursor of YP. It was obtained from tissue cultures as described below. The concentration of protein was estimated by absorption at 280 nm as described . Preparation of Antisera Purified Vt was mixed with Freund’s adjuvant, and 10 mg of protein was injected intramuscularly into the backs of sheep. One further injection of the same amount without adjuvant was given after two weeks, and the sheep were bled after a total of five weeks. In some experiments, the antiserum was adsorbed with male hemolymph. Cultivation of Fat Body and Analysis of Medium Fat body from one insect was cultured for 4 or 5 h in 3 ml of a medium described by Wyatt and Wyatt  and containing 4 pCi [14C]leucine. The medium was removed and filtered through miracloth, and then centrifuged for 15 min at 29,OOOg in a refrigerated centrifuge. In most experiments, the medium from three to six cultures was pooled. For the analysis of total protein secreted, 0.2 ml of medium was taken in triplicate and diluted with an equal volume of 20% TCA. The samples were refrigerated overnight and then centrifuged. The precipitate was washed three times with 1ml of 5% TCA, and once each with ethanol, ethanol-ether (1/1),and ether, dissolved in 0.2 ml 10% SDS and 0.01 ml58% NH40H  and boiled for 3 min. The sample was made up to lml in sample buffer (0.125 M Tris, pH 8.8, and 0.1% @ME).Samples prepared in this way produced bands on SDS gels identical to those obtained by precipitation of protein at low ionic strength. Samples of 0.2 ml were taken in triplicate and diluted with an equal volume of antiserum, allowed to stand at room temperature for 2 h, and then refrigerated for 24 h. In the earlier experiments, the samples were washed three times with 0.15 M NaC1, but more recently with 0.4 M NaCl and then three times with a solution described by Gellissen and Wyatt . Precipitates were dissolved in sample buffer (0.125 M Tris, pH 8.8, 0.1% @ME,0.2% SDS). For determination of radioactivity, 0.2 ml of the SDS solutions were emulsified in 4 ml of counting solution (2 liters xylene, 1 liter Triton N-101, 14g PPO) and counted in a Packard liquid scintillation counter, efficiency 90 to 95%. Collection of Hemolymph Hemolymph was collected by the method of Bohm , diluted five- to tenfold with 0.4 M NaC1, and centrifuged for 15 min at 3,OOOg. Aliquots were made up to 1ml in sample buffer and boiled for 3 min. Polyacrylamide Gel Electrophoresis SDS-PAGE was performed using the buffer system of Laemmli  at 10°C and a constant voltage of 125 V. Slab gels were 9 cm long and 1 mm thick Brookes 580 with a 4.5% stacking gel (pH 6.8). The gels were stained in 0.15% Coomassie brilliant blue, 30% methanol, 0.6% acetic acid, and destained in 5% methanol, 10% acetic acid. For radiofluorography, the gels were immersed in sodium salicylate , dried, and placed on Kodak X-OMAT AR film. Molecular weights were estimated by the method of Weber and Osborn . Myosin (M, = 200,000) galactosidase (M, = 116,000), phosphorylase b (M, = 92,500), bovine serum albumin (M, = 66,000) and ovalbumin (M, = 45,000) were used as standards. RESULTS The Structure of Vitellogenin Hernolymph polypeptides (Fig. 1A). Samples of hemolymph were denatured in sample buffer and the polypeptides separated on SDS-PAGE. The hemolymph of mated females with developing oocytes contains at least 18-20 polypeptides, many of which are also found in the hemolymph of males and young virgin females with nonvitellogenic oocytes. Particularly prominent are polypeptides in the molecular weight range of 78,000-85,000. Polypeptides specific to oocyte development are found at positions corresponding to molecular weights of 160,000 and 57,000 and between 105,000 and 92,000. Polypeptides of 105,000 and above 200,000 daltons are common to all three types of hemolymph but are intensified in the hemolymph of mated females. Fat body polypeptides (Fig. lB,lC). Fat body was incubated in medium containing [14C]leucine for 4 h. Protein was precipitated from the medium with TCA, and the precipitate was dissolved in sample buffer and analyzed -200 200- -116 - - 92 116105 92- - 66 160 85 80 66 - 45 57 45 HEMOLYMPH PROTEINS FAT BODY PROTEINS Fig. 1. Electrophoretic patterns on S D S polyacrylamide gels of hemolymph proteins and proteins synthesized by fat body. Hernolymph was collected and prepared as described in Materials and Methods. Fat body was incubated in radioactive medium for 4 h, and proteins precipitated with TCA and prepared as described in Materials and Methods. A and B) stained with Coomassie brilliant blue; C) Fluorogram of gel shown in B. M = Mated female; V = Virgin female; FB = Fat body proteins for comparison; Std = Standard. Yolk Proteins From ieucophaea maderae 581 by SDS-PAGE. Polypeptides of proteins released into medium during incubation of fat body from mated females showed a banding pattern similar to that of hemolymph when stained with Coomassie Brilliant Blue. Radiofluorographs of the gels showed that the principal proteins being synthesized by fat bodies of males and virgin females were associated with polypeptides larger than 200,000 and with three or four polypeptides in the 78,000 to 90,000 range. Fat bodies from mated females synthesized a polypeptide with a molecular weight of 57,000, four others in the 94,000 to 190,000 range, and several larger than 200,000, one of which was common to all three types of fat bodies but intensified in the medium of mated females. Sucrose gradient separation of fat body proteins. After a 4 h incubation, samples of medium containing [14C]leucine from fat bodies of males, virgin females, and mated females were fractionated by sucrose density-gradient centrifugation. Samples of the peaks were analyzed by SDS-PAGE and radiofluorographed. The rate of synthesis by fat bodies of males and virgin females was only a fraction of that of mated females. As shown in Figure 2, media from all three stages separated into three or more peaks. Medium from fat body from mated females contained a large peak (IIJ), which, in D.P.M 2000c 10001 5001 250l FRACTION NUMBER Fig. 2. Sucrose density gradient fractionation of fat body incubation medium. Fat body was incubated in radioactive media for 4 h, and 0.5 ml samples were centrifuged through a 1030% sucrose gradient for 16 h at 64,OOOg in a 27.1 Beckman rotor. Fractions of 1.2 ml were collected precipitated with TCA, and prepared and counted as described in Materials and Methods. (0-0,mated female; 0-0,virgin female *-*, male. Top of the gradient to the left. 582 Brookes other runs, sedimented with the 14s component of yolk protein. This peak contained a 57,000 dalton polypeptide, four other well defined polypeptides, M, = 98,000, 105,000, 160,000, and one greater than 200,000 (Fig. 3, lane e). Faint bands were also present in the upper part of the gels and at the 78,000 position. Fat bodies from males and virgin females also produced a protein with a peak (IV) in this region of the gradient. As shown in the radiofluorogram (Fig. 3, lane f), Peak IV from males was predominantly labeled in the 78,000 dalton polypeptide. Peak I1 of mated females contained two polypeptides with molecular weights greater than 200,000 and two of about 80,000 (Fig. 3, lane c; the bands are very faint). Similar bands were found in the peaks from the media from males and virgin females sedimenting near the top of the gradient (not shown in Fig. 3). Peak I1 from virgin females contained a band located between the 160,000 and 116,000 positions (not shown). Fractions from the remaining peaks were not visible in radiofluorograms. Variations in polypeptide synthesis during development of the oocytes. Fat body was taken from females at various times during the development of the oocytes and cultured for 4 h in radioactive medium. Samples with about equal numbers of counts (3,500-4,000; except g, 2,500 cpm) were analyzed by SDS-PAGE. The rate of synthesis at the time of mating was low but rose shortly after mating and reached a peak at about the time the oocytes reached half maximal length. This rate was maintained until just before ovulation, a b c d e f -205 ---160 -116 - -. 1 0 5 92- 98 - 87 68 - - 66 - 57 - 45 - 29 Fig. 3. Electrophoretic analysis of sucrose gradient peaks of Figure 2. Samples were separated on SDS polyacrylamide gels and fluorographed. a) TCA precipitate, and b) precipitate obtained by five-fold dilution with water, are from samples of incubation medium of fat body from mated females taken before centrifugation; c fraction 3 (Peak I) mated female; d) fraction 7 (Peak I l l ) mated female; e) fraction 9 (Peak 111) mated female; f) fraction 11 (Peak IV) male. Yolk Proteins From Leucophaea maderae 583 when it dropped to the premated rate. Figure 4A shows the polypeptide pattern of proteins synthesized by fat body at 6 periods during the 15 days of oocyte maturation. At the time of mating, the principal peptides produced were those in the 78,000 to 84,000 dalton range (lane a). Traces of others, including those associated with Vg, were also present, but the bands were very faint. Two days after mating, when the oocytes were 1.6 mm long, polypeptides associated with Vg were obvious, and those in the 78,000 to 84,000 range were diminished. When the oocytes reached one-half maximal length (2.8 mm), the principal peptides synthesized were those associated with Vg (lane c). The synthesis of a very large polypeptide (M,-250,000) was just visible at this time, but it was pronounced in fat body of females with large oocytes (lanes d,e). When the oocyte reached near maximum size, Vg was apparently the major protein produced, and the large polypeptide was no longer present (lane f). Fat bodies taken from females 24 h later and just before ovulation only synthesized peptides in the 78,000 to 84,000 range (lane g). Antibody precipitation of vitellogenin. Antiserum prepared against purified YP was used to precipitate protein from medium of fat body of mated females. The precipitate was washed in 0.15 M NaCl or more extensively, as described in Materials and Methods. As shown in Figure 4B, the antiserum precipitated proteins that separated into four well defined polypeptides, M, = 160,000, 105,000, 98,000, and 57,000 during SDS-PAGE. When the precipitates were washed with saline alone, traces of other polypeptides, a b c A d e f a g b c d a '.I60- -1 60- -1 05T 9 8- -1 05-98- B b c d e f 9 C Fig. 4. Electrophoretic analysis on SDS polyacrylamide gels of fat body incubation media. Figure is of fluorograms of the gels. Fat body was incubated for 4 h. A) Fat body from females during various stages of oocyte development. Oocyte length: a) 1.1 mm (0 time); b) 1.6 mm; c) 2.4 mm, d) 4.0 mm; e) 4.5 mm; f) 5.0 mm; g) 5.0 mm) (ovulation) 3,500-4,000 cpm each lane except g, which was 2,500. B) Antibody precipitation of media proteins. a) and d) washed three times with 0.15 M NaCI; b) and c) washed with 0.4M NaCl and according to Gellissen and Wyatt [I91 approximately 3,000 cpm each. C) Solubility of proteins synthesized by fat body. Fat body from eight incubations were pooled. Samples were taken for total protein analysis (TCA) and the remainder diluted with five volumes H20 and stored on ice for 48 h. The precipitate was recovered and dissolved in 0.4 M NaCI. The recovery was about 90%. Aliquots were diluted to the concentration shown, and the precipitates that formed were dissolved in sample buffer. Approximately 5,000 cpm placed in each lane except e, 1,500 cpm. a) 0.036 M; b) 0.065 M; c) 0.079 M; d) 0.1 M; e) 0.132 M; f) total protein (TCA); g) H20precipitate, original sample. 584 Brookes both larger and smaller than the four, were also present (lanes a,d). These other polypeptides were removed by more extensive washing (lanes b,c). In Ouchterlony immunodiffusion tests, the antiserum reacted with a single component of hemolymph of mated females but not of males, and with samples of medium incubated with fat body of mated females (data not shown). Effect of ionic strength on protein solubility. Fat bodies were dissected from eight females late in the development of oocytes (4-5 mm long) and cultured individually for 4 h in radioactive medium. The medium was pooled and samples taken to measure total labeled protein by TCA precipitation. The remainder was diluted with five volumes of water. The resulting precipitate was recovered and dissolved in 0.4 M NaC1. Ten ml more water were added to the supernatant, which produced an additional precipitate that was recovered and dissolved with the first. The total recovery of labeled protein by dilution was approximately the same as the recovery by precipitation with TCA, and the polypeptide pattern was similar. To determine if the proteins of the medium could be fractionated on the basis of ionic strength, aliquots of the NaCl solution were diluted with various volumes of water, the precipitates were recovered, and the radioactivity counted. Recovery depended on the ionic strength as shown in Table 1. Samples of about equal counts (with the exception of the 0.132 M solution) were separated by SDSPAGE and radiofluorographed (Fig. 4C). The pattern does not vary with the concentration of NaC1. Each concentration contains polypeptides corresponding to Mr = 57,000, 98,000, 105,000, 160,000, and a major and minor band of higher molecular weight. Several peptides found in the original sample were not present, however, suggesting that a second precipitation by dilution produces some purification. Proteolysis of Vg. The possibility that proteolytic digestion of Vg may be occurring after secretion into the medium was tested in three ways: a) fat body was incubated in labeled medium for 1 h, then removed, and the medium incubated for an additional 4 h; b) fat body was incubated for 4 h, removed, and the medium incubated for 16 h at 30°C; c) globin, synthesized in a cell-free system to obtain a high specific activity, was incubated with TABLE 1. Effect of NaCl Concentrationon the Solubility of Vitellogenin* Final concentration NaCl (M) 0.132 0.100 0.078 0.065 0.036 Amount precipitated (% total activity) 0.07 23.0 64.0 81.0 90.0 *Samples of purified Vg in 0.4 M NaCl were titrated with various amounts of water. Precipitates were recovered by centrifugation, dissolved in sample buffer, and counted. The recovery is expressed as the percentage (%) of total activity precipitated by TCA. Yolk Proteins From feucophaea maderae 585 unlabeled medium after a 4 h incubation with fat body. Aliquots of the medium in all three tests were precipitated with TCA or by dilution with water and analyzed by SDS-PAGE. In none of the experiments was there any evidence of proteolysis. Incubation of the medium after removal of the fat body does not result in the disappearance of polypeptides normally found after such incubations nor in the appearance of new polypeptides. The globin protein remains intact even after 4 h of incubation with medium (data not shown). The Structure of Vitellin The effect of pH. Purified YP in 0.4 M NaCl was dialyzed overnight against 0.15 M Tris buffer, pH 8.8, and 0.4 M NaC1. Half the dialyzate was dialyzed again, this time against 0.15 M Tris, pH 6.8, and 0.4 M NaC1. Samples of the two dialyzates were analyzed by density-gradient centrifugation with samples of the original solution that had not been dialyzed. As shown in Figure 5, undialyzed YP separates into two fractions, 14s and 28s, as has been repeatedly demonstrated [11,16]. In slightly alkaline solution (pH 8.8), only one fraction with a lower sedimentation was obtained. Acidification by redialysis aggregated the protein to a component that sedimented at the same rate as 14s. No fraction comparable to 28s was found. Alkaline conditions did not alter the polypeptide pattern of 14s or 28s. When these components were purified and individually subjected to alkaline and acidic conditions as described above, the polypeptide pattern after SDS-PAGE was still characteristic of the component (data not shown, but patterns are described in Fig. 6). Polypeptide structure of 14s and 28s. The 28s component contains three subunits (M, = 105,000, 85,000, and 57,000). The 14s component contains the same polypeptides and a fourth one that migrated to a position just below that of the 92,500 standard (Fig. 6A). Also shown is a radiofluorogram of a partially purified fraction of Vg. Superposition of the exposed X-ray film over the Coomassie brilliant blue-stained gel showed that 105,000 and 57,000 dalton polypeptides of Vg are identical to those of the 14s and 28s components, but the 98,000 dalton polypeptide is slightly larger than the fourth component of 14s. DISCUSSION During periods of vitellogenesis, the hemolymph of mated females contains proteins whose polypeptides separate into at least 18 to 20 bands of SDS-PAGE. Although at least three of these polypeptides are not obvious in the hemolymph of males and young virgin females with nonvitellogenic oocytes, many others are common to all three types, and some are intensified in the hemolymph of mated females (Fig. 1).Because of the large numbers of peptides, those specific to females are difficult to identify in the hemolymph. Although all these proteins are probably synthesized by fat body at some stage, the rates of synthesis and turnover probably vary, so that many proteins are carried over to the adult from earlier stages. 586 Brookes pH 6.8 pH 8.8 FRACTION NUMBER Fig. 5. The effect of pH on the structure of yolk protein. YP was purified as previously described. Samples were dialyzed overnight against 0.15 M tris buffer pH 8.8, 0.4 M NaCI, and then redialyzed against 0.15 M tris buffer pH 6.8, 0.4 M NaCI, and then analyzed by sucrose density gradient centrifugation as described in Figure 2. Optical density of fractions was measured at 280 nm. The principal proteins synthesized by the fat body of adult males and early virgin females contain high molecular weight polypeptides greater than 160,000 daltons and several in the 78,000 to 90,000-daltonrange. Polypeptides of equivalent molecular weight are synthesized by fat body of mated females, but in addition, four others appear to be specific. All the data presented here strongly suggest that these four polypeptides with molecular weights of 160,000, 105,000, 98,000, and 57,000 are part of the Vg molecule. A large polypeptide (M, - 250,000) is present in many fractions where Vg is expected and is most easily identified in Figure 4C.A polypeptide of this size has been described as a subunit of Vg of L. maderue [ll] that is processed during passage through the hemolymph. My data suggest that it is a polypeptide of another protein and not a subunit of Vg for the following reasons: Yolk Proteins From Leucophaea maderae V g 14s 28s 587 V g 14s 28s -200-1 6 0 -105- - 98 -90 - 85 -57 A B j - _I Fig. 6 . Comparison of the polypeptide structure of vitellogenin and yolk protein by SDSPAGE. Vg was purified by precipitation at low ionic strength as described in Figure 3. The 14s and 28s components were obtained from sucrose gradients . 1. It was present in protein precipitated with antiserum to Vt but was removed by extensive washing in 0.4 M NaCl and detergent (Fig. 4B). In previous studies the precipitate was washed only in saline solutions [HI. 2. Its rate of synthesis was not consistent with that of the four polypeptides I associate with Vg. During vitellogenesis, the relative proportion of proteins synthesized by the fat body changed. At the time of mating, and later when vitellogenesis was complete, the principal proteins synthesized contained polypeptides smaller than 90,000 daltons. Polypeptides larger than 200,000, found during the early stages of vitellogenesis, reached a peak before the oocytes were maximum size and were not present or were present in lesser amounts during the final phase (Fig. 4A). 3. Precipitation at low ionic strength, a method also used in previous work [ll], results in only a partial purification of Vg (Fig. 4C). 4. The high molecular weight polypeptides appear to be stable, and no evidence was obtained for a conversion to polypeptides of Vg. Gellissen and Wyatt  found that the principal proteins synthesized during vitellogenesis in the locust were Vg and a diacylglycerol binding protein, lipophorin. They describe lipophorin as a high molecular weight protein with many properties similar to Vg. The rate of synthesis of Vg in locusts increases during vitellogenesis, but that of lipophorin remains more or less constant. Lipophorin now has been identified in a number of insects 588 Brookes  including Peripluneta umericunu , and characterized as an aggregate of two polypeptides with molecular weights of 240,000 to 250,000 and 78,000 to 85,000. Lipophorin has not been described in L. maderue, but such a protein is likely to be present and represented by at least some of the nonvitellogenic polypeptides described here, particularly those with molecular weights comparable to the ones associated with lipophorin in other species. This is indicated by a recent experiment in which hemolymph from vitellogenic females was subjected to density gradient centrifugation in KBr as described by Ryan et al. . A fraction was obtained with a density expected for lipophorin, which contained two major polypeptides with M, = 250,000. In the present work, polypeptides with comparable molecular weights present something of an enigma, because they appear in several fractions after centrifugation of fat body incubation medium in a sucrose gradient (Fig. 3). Lipophorin, if it exists in L. maderue, may consist of a family of lipoproteins differing in density, depending on the amount of lipid attached. Koeppe and Ofengand [ll]isolated Vg by sucrose density-gradient centrifugation and by immunoprecipitation. They identified seven polypeptides associated with this protein and designated them as: a (M, = 260,000); fi (M, = 179,000); the four polypeptides of YP, A, B, C, and D (see below); and one, E, with a molecular weight intermediate between A and D (M, = 108,000). From the evidence presented in this paper and discussed above, I suggest that the polypeptides a,B, and D are not related to Vg. The polypeptide P is probably equivalent to the one I identlfy as 160,000 daltons, and E to the one I find to be 98,000 daltons. The only other study of the structure of Vg of L. maderue was done by Engelmann . Vg was immunoprecipitated from preparations of microsomes and from hemolymph and found to be made up of three polypeptides (M, = 96,000, 82,000, and 52,000). He also found that the three were present in vitellin. The data are presented in the form of scans, and I cannot explain the discrepancy. Koeppe and Ofengand [ll]and Masler and Ofengand  described the polypeptide structure of YP of L. maderue. They identified four polypeptides designated A (M, = 118,000), B (M, = 87,000), C (M, = agreement in that the 28s component contains three polypeptides, A, B, and C; and four polypeptides, A, B, C, and D, are found in 14s. One difference between my results and the earlier ones concerns the molecular weights of these polypeptides. The difference arises because of a discrepancy in the designation of the molecular weight of the standards, P galactosidase and phosphorylase b. Koeppe and Ofengand used 130,000 and 100,000, whereas the supplier of the standards, BioRad, gives 116,000 and 92,500 as the molecular weight. Koeppe and Ofengand [ll] suggested a ratio of polypeptides to conform to the molecular weights established by Dejmal and Brookes (; 14s = lA:lB:2C:2D; 28s = lA:3B:2C). They further suggested that polypeptide B of 28s is derived from one unit of B and two units of D from the 14s component. Masler and Ofengand  attempted to support this hypothesis by an experiment in which the 28s and 14s components were disaggregated in urea and the polypeptides separated by ion exchange chromatography. They found that urea divided each component into three fractions, one of which eluted with the void volume and was much more pronounced in 28s Yolk Proteins From Leucopbaea maderae 589 than in 14s. Analysis by SDS-PAGE indicated that the fraction eluted with the void volume was principally polypeptide B, whereas the other two fractions contained all the polypeptides. Masler and Ofengand interpret these results as indicating the presence of two structural forms of polypeptide B, thus supporting the hypothesis of Koeppe and Ofengand. Their results (Figure 3 in reference ) clearly show, however, that all three fractions contain several polypeptides, indicating that urea only partially degraded the protein into a complex mixture. What has been overlooked in the earlier studies is that both 14s and 28s are no more than loosely bound aggregates that are easily disrupted under mild alkaline conditions. The polypeptide pattern of the alkaline fraction depends on its source, 14s or 28s. Sedimentation in sucrose gradients indicates that, regardless of the source, the fraction consists of a single protein that reverts to a 14s-like structure when acid conditions are restored. The relative position in sucrose gradients indicates that the molecular weight must be about half of 14s, M, = 260,000 (Fig. 5, and reference [ll]). The structure based on the polypeptide ratio of Koeppe and Ofengand cannot be divided into a smaller unit. The sum of the molecular weights of the three polypeptides of 28s is about 250,000 and when multiplied by six, comes to within 10% of the measuredmolecular weight of this component . The four polypeptides of 14s add up to about 350,000 and when multiplied by two, exceed the molecular weight; however, 14s may not be a discrete protein. During maturation of the oocytes, the yolk may be expected to consist of the final yolk protein, Vt, varying amounts of Vg, and transitional forms produced as the result of processing. During purification of YP,much of the Vg would be lost because it is much more soluble at low ionic strength, but it should be present in the supernatant after YP has been recovered. I suggest a scheme (Fig. 6) for the conversion of the 98,000-dalton polypeptide to the 85,000 polypeptide with the intermediate formation of a 95,000 polypeptide. The 14s is thus a transitional form, so that the molecular weight cannot be estimated on the basis of the polypeptides. The polypeptide pattern of 14s and 28s displayed by SDSPAGE always consist of more bands than A, B, C, and D associated with these proteins, as is clearly seen in the figures of Masler and Ofengand . I have many gels that are identical. These minor bands are represented by dotted lines in the scheme shown in Figure 6. In 14s, the minor band is equivalent to the 98,000- and the 90,000-dalton components. I have seen these bands in the polypeptide pattern of YP isolated from the freshly formed ootheca, in which only the 28s component is present, although only in overloaded gels in which they are very faint. According to my scheme, the evidence of processing is also seen in the 28s component and continues even after the egg is mature and fertilized. I have found no evidence that the 160,000-dalton polypeptide of Vg is a part of Vt, but peptide maps have yet to be made. Harnish and White [lo] have divided insects into three groups based on the structure of Vt. The dictyopteran and orthopteran insects fall into Group I, with at least one large and one small polypeptide in the molecule. Processing of the precursor, Vg, occurs shortly after synthesis in the fat body, and, 590 Brookes in some insects, further processing may occur at the time of deposition. The actual number of polypeptides produced is controversial, because different workers working with the same insect obtain different results. Part of the problem stems from the presence of proteins in the hemolymph and synthesized by the fat body in vitro with characteristics similar to those of Vg and Vt. Some of these proteins precipitate nonspecifically in the presence of antiserum, which is otherwise specific to Vg or Vt. Proteases may be present that cleave the proteins during preparation. Although I could find no evidence for proteolytic activity in the experiments designed to test for this, I cannot exclude the possibility that a domain of Vg is proteolytic as it occurs in the nerve growth factor . Denaturing the protein with SDS may expose susceptible regions to this domain resulting in cleavage. The validity of the model in Figure 6 is being tested by peptide mapping. I am also trying to identify the translation products that are the precursors to Vg and to isolate Vg in pure form. LITERATURE CITED 1. Wyatt GR: Insect plasma proteins. Annu Rev Biochem 47, 779 (1978). 2. Hagedorn HH, Kunkel JG: Vitellogenin and vitellin in insects. Annu Rev EntomolZ4, 475 (1979). 3. Engelmann F: Insect vitellogenin: Identification, biosynthesis and role in vitellogenesis. Adv Insect Physiol 15, 49 (1979). 4. Chen TT, Hillen LJ: Expression of the vitellogenin genes in insects. Gamete Res 7, 179 (1983). 5. Harnish DG, Wyatt GR, White BN: Insect vitellins: Identification of primary products of translation. J Exp Zool 220, 11(1982). 6. Postlethwait JH, Jowett T: Regulation of vitellogenesis in Drosophila. In: The Regulation of Insect Development. Senhal F, Zabza A, eds. Technical University of Wroclaw, Wroclaw, Poland, pp 591-627 (1981). 7. Chinzei Y, Chino H, Wyatt GR: Purification and properties of vitellogenin and vitellin from Locusfa migraforia. Insect Biochem 11, l(1981). 8. Harry P, Pines M, Applebaum SW: Changes in the pattern of secretion of locust female diglyceride-carrying lipoprotein and vitellogenin by the fat body in nitro during oocyte development. Comp Biochem Physiol63B, 287 (1979). 9. Sam GR, Bell WJ, Weavor RF: Vitellogenin: Its structure, synthesis and processing in the cockroach Periplanefu americana. Biochim Biophys Acta 609, 121 (1980). 10. Harnish DG, White BN: Insect vitellins: Identification, purification, and characterization from eight orders. J Exp Zool 220, 1(1982). 11. Koeppe J, Ofengand J: Juvenile hormone-induced biosynthesis of vitellin in Leucophaea maderae. Arch Biochem Biophys 173, 100 (1976). 12. Engelmann F: Juvenile hormone regulated vitellogenin synthesis in an insect. In: Comparative Endocrinology. Gaillard PH, Boer HH, eds. ElsevierlNorth-Holland Biomedical Press, Amsterdam, pp 17-20 (1978). 13. Brookes VJ, Dejmal RK: Yolk protein: Structural changes during vitellogenesis in the cockroach Leucophaea maderae. Science 160,999 (1968). 14. Masler EP, Ofengand J: Biochemical properties of the two forms of Leucophaea maderae vitellin and their subunits. Insect Biochem 12, 341 (1982). 15. Chambers DL, Brookes VJ: Hormonal control of reproduction-I. Initiation of oocyte development in the isolated abdomen of Leucophaeu maderae. J Insect Physiol 13, 99 (1967). 16. Dejmal RK, Brookes VJ: Insect lipovitellin. Chemical and physical characteristics of a yolk protein from the ovaries of Leucophaeu maderae. J Bio Chem 247, 869 (1972). Yolk Proteins From feucophaea maderae 591 17. Wyatt GR, Wyatt SS: The development of an insect tissue culture medium. In: Invertebrate Tissue Culture. Kurstak E, Maramorosch K, eds. Academic press, New York, pp 249-256 (1976). 18. Gottlieb TA, Wallace RA: Intracellular phosphorylation of vitellogenin in the liver of estrogen-stimulated Xenopus Zuevis. J Biol Chem 256, 4116 (1981). 19. Gellissen G, Wyatt GR: Production of lipophorin in the fat body of Locustu rnigrutoriu: Comparison with vitellogenin. Can J Biochem 59, 628 (1981). 20. Bohm H: Differential adhesion of the haemocytes of Leucophueu muderue (Blattaria) to a glass surface. J Insect Physiol23, 185 (1977). 21. Laemmli UK: Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature (Lond) 227, 680 (1970). 22. Chamberlain JP: Fluorographic detection of radioactivity in polyacrylamide gels with water-soluble fluor, sodium salicylate. Anal Biochem 98, 132 (1979). 23. Weber K, Osborn M: The reliability of molecular weight determinations by dodecyl sulfatepolyacrylamide gel electrophoresis. J Biol Chem 244, 4406 (1969). 24. Ryan RO, Schmidt JO, Law JH: Chemical and immunological properties of lipophorins from seven insect orders. Arch Insect Biochem Physiol 1, 375 (1984). 25. Chino H, Katase H, Downer RGH, Takahashi K: Diacylglycerol-carrying lipoprotein of hemolymph of the American cockroach: Purification, characterization, and function. J Lipid Res 22, 7 (1981). 26. Greene LA, Shooter EM, Varon S: Enzymatic activities of mouse nerve growth factor and its subunits. Proc Natl Acad Sci USA 60, 1383 (1969).