Characterization of the Proteins Comprising the Integral Matrix of Strongylocentrotus purpuratus Embryonic Spicules
1996; Elsevier BV; Volume: 271; Issue: 15 Linguagem: Inglês
10.1074/jbc.271.15.9150
ISSN1083-351X
AutoresChristopher E. Killian, Fred H. Wilt,
Tópico(s)Galectins and Cancer Biology
ResumoIn the present study, we enumerate and characterize the proteins that comprise the integral spicule matrix of the Strongylocentrotus purpuratus embryo. Two-dimensional gel electrophoresis of [S]methionine radiolabeled spicule matrix proteins reveals that there are 12 strongly radiolabeled spicule matrix proteins and approximately three dozen less strongly radiolabeled spicule matrix proteins. The majority of the proteins have acidic isoelectric points; however, there are several spicule matrix proteins that have more alkaline isoelectric points. Western blotting analysis indicates that SM50 is the spicule matrix protein with the most alkaline isoelectric point. In addition, two distinct SM30 proteins are identified in embryonic spicules, and they have apparent molecular masses of approximately 43 and 46 kDa. Comparisons between embryonic spicule matrix proteins and adult spine integral matrix proteins suggest that the embryonic 43-kDa SM30 protein is an embryonic isoform of SM30. An adult 49-kDa spine matrix protein is also identified as a possible adult isoform of SM30. Analysis of the SM30 amino acid sequences indicates that a portion of SM30 proteins is very similar to the carbohydrate recognition domain of C-type lectin proteins. In the present study, we enumerate and characterize the proteins that comprise the integral spicule matrix of the Strongylocentrotus purpuratus embryo. Two-dimensional gel electrophoresis of [S]methionine radiolabeled spicule matrix proteins reveals that there are 12 strongly radiolabeled spicule matrix proteins and approximately three dozen less strongly radiolabeled spicule matrix proteins. The majority of the proteins have acidic isoelectric points; however, there are several spicule matrix proteins that have more alkaline isoelectric points. Western blotting analysis indicates that SM50 is the spicule matrix protein with the most alkaline isoelectric point. In addition, two distinct SM30 proteins are identified in embryonic spicules, and they have apparent molecular masses of approximately 43 and 46 kDa. Comparisons between embryonic spicule matrix proteins and adult spine integral matrix proteins suggest that the embryonic 43-kDa SM30 protein is an embryonic isoform of SM30. An adult 49-kDa spine matrix protein is also identified as a possible adult isoform of SM30. Analysis of the SM30 amino acid sequences indicates that a portion of SM30 proteins is very similar to the carbohydrate recognition domain of C-type lectin proteins. INTRODUCTIONDuring the course of its development a sea urchin embryo constructs a pair of calcareous endoskeletal spicules. These spicules are rod shaped, mineralized structures which are calcitic assemblages of calcium carbonate (95%) and magnesium carbonate (5%) with an occluded proteinaceous integral matrix. The spicules are synthesized by a well characterized single tissue type, the primary mesenchyme cells. The calcite of each spicule rod is aligned along a single crystal axis, appearing as if each spicule is composed of one crystal of calcite. However, the spicule has greater flexural strength than a single crystal of calcite and it fractures as if it is made up of many microcrystals(1.Emlet R.B. Biol. Bull. 1982; 163: 264-275Crossref Google Scholar, 2.Okazaki K. Inoué S. Dev. Growth and Differ. 1976; 18: 413-434Crossref Scopus (82) Google Scholar, 3.Okazaki K. McDonald K. Inou é S. Omori M. Watabe N. The Mechanisms of Biomineralization in Animals and Plants. Tokai University Press, Tokyo1980: 159-168Google Scholar). Persuasive biophysical evidence indicates that the proteins embedded within the mineral phase of the spicules, the integral spicule matrix proteins, cause these interesting physical characteristics(1.Emlet R.B. Biol. Bull. 1982; 163: 264-275Crossref Google Scholar, 4.Berman A. Addadi L. Weiner S. Nature. 1988; 331: 546-548Crossref Scopus (433) Google Scholar, 5.Berman A. Addadi L. Kvick Å. Leiserowitz L. Nelson M. Weiner S. Science. 1990; 250: 664-667Crossref PubMed Scopus (264) Google Scholar, 6.Berman A. Hanson J. Leiserowitz L. Koetzle T. Weiner S. Addadi L. Science. 1993; 259: 776-779Crossref PubMed Scopus (336) Google Scholar, 7.Aizenberg J. Hanson J. Ilan M. Leiserowitz L. Koetzle T.F. Addadi L. Weiner S. Faseb J. 1995; 9: 262-268Crossref PubMed Scopus (104) Google Scholar). It is believed that these proteins interact with specific faces of the calcite crystal when occluded within the mineral, and it is through these interactions that control of spicule growth occurs. However, the precise molecular mechanisms underlying interactions with noncollagenous integral matrix proteins that control the formation and the physical properties of mineralized tissues remain unknown. A question basic to our understanding of these mechanisms in mineralizing tissues is what is the nature of the noncollagenous integral matrix proteins that are intimately associated with the mineral portion of these tissues.Many proteins have been identified and characterized as noncollagenous integral matrix proteins of hard tissues from various vertebrate and invertebrate organisms (most numerously from vertebrate bone and teeth). These types of proteins usually share the general properties of being soluble and acidic(8.Weiner S. Am. Zool. 1984; 24: 945-951Crossref Scopus (110) Google Scholar, 9.Weiner S. CRC Crit. Rev. Biochem. 1986; 20: 365-408Crossref PubMed Scopus (236) Google Scholar). However, it has proven difficult to ascribe to these types of proteins precise functions within the cell that synthesize them. In addition, while many noncollagenous integral matrix proteins have been identified, it has also proven difficult to determine with much certainty what particular integral matrix proteins are contained within a given mineral phase at a given stage of development. Many of these difficulties are due to the complex structures and dynamics of the mineralized tissues most widely studied, i.e. vertebrate bone and teeth.Sea urchin spicule formation, on the other hand, is particularly well suited to ask these sorts of basic questions. The spicules are synthesized by a single well characterized tissue type and they are relatively simple mineralized structures that do not have the complex dynamics of vertebrate bones or teeth. The sea urchin embryo is also very amenable to biochemical and molecular experimental analysis. Much is known about the cell and developmental biology of sea urchin embryos and particularly about the differentiation of the cell lineage which synthesizes the spicules (for some reviews, see (10.Okazaki K. Am. Zool. 1975; 15: 567-581Crossref Scopus (244) Google Scholar, 11.Decker G. Lennarz W.J. Development. 1988; 103: 231-247PubMed Google Scholar, 12.Ettensohn C.A. Gerhart J. Cell-Cell Interactions in Early Development, 49th Symposium of the Society for Developmental Biology. Wiley-Liss, Inc., New York1991: 175-201Google Scholar, 13.Benson S.C. Wilt F.H. Bonucci E. Calcification in Biological Systems. CRC Press, Ann Arbor, MI1992: 157-178Google Scholar, 14.Wilt F.H. Killian C.E. Livingston B.T. Slavkin H. Price P. Chemistry and Biology of Mineralized Tissues. Excerpta Medica, New York1993: 85-92Google Scholar)). Benson et al.(15.Benson S.C. Jones E.M. Benson N.C. Wilt F. Exp. Cell Res. 1983; 148: 249-253Crossref Scopus (44) Google Scholar) reported that the spicule matrix within the mineralized spicules is arranged in concentric sleeves of irregular fibrillar proteinaceous material with some interconnections between the layers of matrix material. The concentric layers of the spicule matrix are also reflected in the concentric layered architecture of the mineralized sea urchin spicules(2.Okazaki K. Inoué S. Dev. Growth and Differ. 1976; 18: 413-434Crossref Scopus (82) Google Scholar). Benson et al.(16.Benson S.C. Benson N.C. Wilt F.H. J. Cell Biol. 1986; 102: 1878-1886Crossref PubMed Scopus (97) Google Scholar) and Venkatesan and Simpson (17.Venkatesan M. Simpson R.T. Exp. Cell Res. 1986; 166: 259-264Crossref Scopus (13) Google Scholar) also identified 8-10 different proteins as comprising the integral spicule matrix. In these studies, one-dimensional SDS-PAGE 1The abbreviations used are: PAGEpolyacrylamide gel electrophoresisCRDcarbohydrate recognition domainBCIP/NBT5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium chloride.was used to resolve the spicule matrix proteins. In addition, both reports demonstrated that most of these detected proteins are N-linked glycoproteins. Total amino acid analysis revealed Strongylocentrotus purpuratus spicule matrix proteins are rich in acidic amino acids(16.Benson S.C. Benson N.C. Wilt F.H. J. Cell Biol. 1986; 102: 1878-1886Crossref PubMed Scopus (97) Google Scholar). This amino acid composition is similar to that of other integral matrix proteins closely associated with the mineral portion of other mineralized tissues(8.Weiner S. Am. Zool. 1984; 24: 945-951Crossref Scopus (110) Google Scholar, 9.Weiner S. CRC Crit. Rev. Biochem. 1986; 20: 365-408Crossref PubMed Scopus (236) Google Scholar).Two different cDNAs that encode two different spicule matrix proteins have also been isolated from S. purpuratus cDNA expression libraries. The first cDNA isolated encodes a protein designated SM50 which has a deduced amino acid sequence with a molecular mass of approximately 50 kDa(18.Sucov H.M. Benson S. Robinson J.J. Britten R.J. Wilt F. Davidson E.H. Dev. Biol. 1987; 120: 507-519Crossref PubMed Scopus (81) Google Scholar, 19.Benson S. Sucov H. Stephens L. Davidson E. Wilt F. Dev. Biol. 1987; 120: 499-506Crossref PubMed Scopus (103) Google Scholar, 20.Killian C.E. Wilt F.H. Dev. Biol. 1989; 133: 148-156Crossref PubMed Scopus (29) Google Scholar, 21.Katoh-Fukui Y. Noce T. Ueda T. Fujiwara Y. Hashimoto N. Higashinakagawa T. Killian C. Livingston B. Wilt F. Benson S. Sucov H. Davidson E.H. Dev. Biol. 1991; 145: 201-202Crossref PubMed Scopus (63) Google Scholar). The second cDNA cloned encodes a protein designated SM30 which has a derived amino acid sequence with a molecular mass of approximately 30 kDa(22.George N.C. Killian C.E. Wilt F.H. Dev. Biol. 1991; 147: 334-342Crossref PubMed Scopus (82) Google Scholar). The predicted chemical characteristics of the deduced amino acid sequences of SM50 and SM30 cDNAs are somewhat different. The cloned SM50 cDNA encodes a protein with an alkaline pI without any consensus N-linked glycosylation site(21.Katoh-Fukui Y. Noce T. Ueda T. Fujiwara Y. Hashimoto N. Higashinakagawa T. Killian C. Livingston B. Wilt F. Benson S. Sucov H. Davidson E.H. Dev. Biol. 1991; 145: 201-202Crossref PubMed Scopus (63) Google Scholar). The cloned SM30 cDNA encodes an acidic protein that contains a consensus N-linked glycosylation site(22.George N.C. Killian C.E. Wilt F.H. Dev. Biol. 1991; 147: 334-342Crossref PubMed Scopus (82) Google Scholar). Both SM50 and SM30 transcripts are expressed exclusively in the primary mesenchyme cells(19.Benson S. Sucov H. Stephens L. Davidson E. Wilt F. Dev. Biol. 1987; 120: 499-506Crossref PubMed Scopus (103) Google Scholar, 22.George N.C. Killian C.E. Wilt F.H. Dev. Biol. 1991; 147: 334-342Crossref PubMed Scopus (82) Google Scholar). Sucov et al.(18.Sucov H.M. Benson S. Robinson J.J. Britten R.J. Wilt F. Davidson E.H. Dev. Biol. 1987; 120: 507-519Crossref PubMed Scopus (81) Google Scholar) have shown that SM50 is a single copy gene in S. purpuratus. Alternatively, there is experimental evidence that there is a small family of SM30 protein genes. Akasaka et al.(23.Akasaka K. Frudakis T.N. Killian C.E. George N.C. Yamasu K. Khaner O. Wilt F. J. Biol. Chem. 1994; 269: 20592-20598Abstract Full Text PDF PubMed Google Scholar) presented Southern blotting analysis indicating that there are between two and four copies of SM30 genes present in the S. purpuratus haploid genome. Akasaka et al.(23.Akasaka K. Frudakis T.N. Killian C.E. George N.C. Yamasu K. Khaner O. Wilt F. J. Biol. Chem. 1994; 269: 20592-20598Abstract Full Text PDF PubMed Google Scholar) further demonstrated that an isolated S. purpuratus genomic clone contains two different SM30 genes that are arranged tandemly. These two SM30 genes were designated SM30-α and SM30-β. Initial characterization of the genomic regulatory regions of the SM50 gene and the SM30-α gene have also been done(23.Akasaka K. Frudakis T.N. Killian C.E. George N.C. Yamasu K. Khaner O. Wilt F. J. Biol. Chem. 1994; 269: 20592-20598Abstract Full Text PDF PubMed Google Scholar, 24.Sucov H.M. Hough-Evans B.R. Franks R.R. Britten R.J. Davidson E.H. Genes and Develop. 1988; 2: 1238-1250Crossref PubMed Scopus (33) Google Scholar, 25.Makabe K.W. Kirchhamer C.V. Britten R.J. Davidson E.H. Development. 1995; 121: 1957-1970Google Scholar, 26.Frudakis T.N. Wilt F.H. Dev. Biol. 1995; 172: 230-241Crossref Scopus (16) Google Scholar).In addition to these two genes that have been shown directly to encode two spicule matrix proteins, a recent report by Harkey et al.(27.Harkey M.A. Klueg K. Sheppard P. Raff R.A. Dev. Biol. 1995; 168: 549-566Crossref PubMed Scopus (71) Google Scholar) characterizes a gene encoding a nonglycosylated 27-kDa protein, designated PM27, that is closely associated with growing sea urchin spicules. While they did not show directly that the PM27 protein is an integral spicule matrix protein, they do show PM27 expression and biochemistry are similar to what one might expect from a spicule matrix protein. In addition they show that PM27 has some sequence similarity to SM50; Harkey et al.(27.Harkey M.A. Klueg K. Sheppard P. Raff R.A. Dev. Biol. 1995; 168: 549-566Crossref PubMed Scopus (71) Google Scholar) point out that portions of PM27, SM50, and the Lytechinus pictus homologue of SM50 (designated LSM34 by Livingston et al.(28.Livingston B.T. Shaw R. Bailey A. Wilt F. Dev. Biol. 1991; 148: 473-480Crossref Scopus (31) Google Scholar)) also have some similarity to the carbohydrate recognition domain (CRD) of a number of C-type lectin proteins.The present paper enumerates more accurately the complexity of the S. purpuratus spicule matrix proteins and more fully characterizes these proteins. These studies provide a biochemical foundation important for the study of the noncollagenous integral matrices of mineralized tissues. Our findings also complement the previously mentioned biophysical studies that examined occluded matrix proteins of sea urchin embryonic and adult mineralized structures and their roles in regulating mineralized tissue formation and structure (1.Emlet R.B. Biol. Bull. 1982; 163: 264-275Crossref Google Scholar, 4.Berman A. Addadi L. Weiner S. Nature. 1988; 331: 546-548Crossref Scopus (433) Google Scholar, 5.Berman A. Addadi L. Kvick Å. Leiserowitz L. Nelson M. Weiner S. Science. 1990; 250: 664-667Crossref PubMed Scopus (264) Google Scholar, 6.Berman A. Hanson J. Leiserowitz L. Koetzle T. Weiner S. Addadi L. Science. 1993; 259: 776-779Crossref PubMed Scopus (336) Google Scholar, 7.Aizenberg J. Hanson J. Ilan M. Leiserowitz L. Koetzle T.F. Addadi L. Weiner S. Faseb J. 1995; 9: 262-268Crossref PubMed Scopus (104) Google Scholar). The studies in the present paper reveal that there are 12 spicule matrix proteins that radiolabel intensely with [S]methionine and approximately three dozen other spicule matrix proteins that are less highly radiolabeled. The majority of the spicule matrix proteins have an acidic pI, while several other moderately radiolabeled to less radiolabeled spicule matrix proteins have a more alkaline pI. Polyclonal antisera that react specifically with the proteins encoded by the previously cloned SM50 and SM30 spicule matrix cDNAs were generated. Western blotting analysis using these antisera identify the SM50 and SM30 proteins. In addition, comparisons are made between the embryonic spicule matrix proteins and the adult spine integral matrix proteins. Further analysis of the protein encoded by SM30-α reveals that a portion of the SM30 proteins is similar to the CRD of the C-type lectin family of proteins.EXPERIMENTAL PROCEDURESCulturing of Sea Urchin EmbryoS. purpuratus gametes were collected, eggs were fertilized, and embryos cultured as described by George et al.(22.George N.C. Killian C.E. Wilt F.H. Dev. Biol. 1991; 147: 334-342Crossref PubMed Scopus (82) Google Scholar).Isolation of Spicule Matrix and Spine Matrix ProteinUnlabeled S. purpuratus embryonic spicule matrix proteins were isolated essentially as described by Venkatesan and Simpson (17.Venkatesan M. Simpson R.T. Exp. Cell Res. 1986; 166: 259-264Crossref Scopus (13) Google Scholar) except that, as a final step, spicules were incubated in 3.5% sodium hypochlorite and then washed extensively with water before they were demineralized with 0.5 N acetic acid. After the calcite was dissolved, the acetic acid was neutralized with Tris base and the spicule matrix proteins were extensively dialyzed against dHO. Proteins were then concentrated by lyophilization. Adult S. purpuratus spine integral matrix proteins were isolated as described by Richardson et al.(29.Richardson W. Kitajima T. Wilt F. Benson S. Dev. Biol. 1989; 132: 266-269Crossref Scopus (31) Google Scholar).Radiolabeled S. purpuratus spicule matrix protein was isolated from micromeres cultured in seawater with 4% horse serum and [S]methionine. The isolation and culture of micromeres was done essentially as described by Benson et al.(30.Benson S. Smith L. Wilt F. Shaw R. Exp. Cell Res. 1990; 188: 141-146Crossref PubMed Scopus (32) Google Scholar). The micromeres isolated from about 2 × 106 16-cell embryos were cultured in four 100-mm Petri plates, each containing 10 ml of seawater containing 4% horse serum that had been dialyzed against seawater. Two hundred μCi of [S]methionine (1000 Ci/mmol; Amersham) were added to each plate just prior to the onset of spiculogenesis and left in the medium until the time of harvest (24-72 h). At the conclusion of labeling, carrier spicules from whole embryos were added to the cultures, and the adherent spicules of the culture were scraped from the Petri plates. The spicules were washed with and then placed into 3.5% sodium hypochlorite overnight at room temperature. They were then washed with 5-7 changes of dHO. After the final wash, the spicules were suspended in 1 ml of dHO. An aliquot was removed to quantitate the amount of radioactivity incorporated into spicule matrix. To prepare the radiolabeled spicule matrix protein for each two-dimensional gel, 5.0 × 104 dpm of the radiolabeled spicule sample was trichloroacetic acid precipitated with 10 μg of cytochrome c carrier; this procedure dissolves the calcite and precipitates the spicule matrix protein. The pellet was then washed with acetone to remove residual trichloroacetic acid. The pellet was dried and dissolved in the appropriate sample buffer.Isolated spicule matrix proteins were labeled in vitro with biotin following the protocol described by Meier et al.(31.Meier T. Arni S. Malarkannan S. Poincelet M. Hoessli D. Anal. Biochem. 1992; 204: 220-226Crossref PubMed Scopus (95) Google Scholar) using the biotinylation agent NHS-CC-biotin purchased from Pierce. The labeled proteins were localized using an enhanced chemiluminescence protocol described by Nesbitt and Horton (32.Nesbitt S. Horton M.A. Anal. Biochem. 1992; 206: 267-272Crossref PubMed Scopus (59) Google Scholar) with the exception that the blocking agent used was 0.1% fish gelatin (Amersham), 0.8% bovine serum albumin, 0.02% Tween 80, 10 mM Tris, pH 8.0, 100 mM NaCl, and the dilution of strepavidin-horseradish peroxidase (Amersham) used was 1:3000.Glycosidase treatment of spicule matrix protein was carried out at 37°C overnight using endoglycosidase F/N-glycosidase F purchased from Boehringer Mannheim following the protocol provided by the manufacturer.Mild alkaline hydolysis β-elimination of O-linked carbohydrate moieties on spicule matrix proteins was done essentially as described by Florman and Wassarman(33.Florman H.M. Wassarman P.M. Cell. 1985; 41: 313-324Abstract Full Text PDF PubMed Scopus (474) Google Scholar). Spicule matrix proteins were incubated in 5 mM NaOH for 24 h at 37°C, neutralized with HCl, and then analyzed by Western blotting analysis as described below. Serine and threonine O-glycosidic linkages are known to be labile in alkaline conditions(34.Sharon N. Complex Carbohydrates, Their Chemistry, Biosynthesis, and Function. Addison-Wesley Publishing Co., Reading, MA1975: 73-78Google Scholar).Two-dimensional Gel Electrophoretic Separation of Spicule Matrix ProteinsTwo-dimensional gel electrophoresis of spicule matrix proteins was carried out using a Bio-Rad mini-protean II two-dimensional gel apparatus. The protocol followed was essentially that described by the gel apparatus manufacturer which is based on the protocol of O'Farrell(35.O'Farrell P.H. J. Biol. Chem. 1975; 250: 4007-4021Abstract Full Text PDF PubMed Google Scholar). Pharmolyte ampholytes with pH ranges of 2.5-4.5, 4.0-6.5, and 3-10 were used (Pharmacia). The first dimensions of the gels shown in Fig. 2, A and B, and 6A were run in the acidic direction using a blend of equal amounts of pH 2.5-4.5 and 4.0-6.5 ampholytes. The first dimensions of the gels shown in Fig. 2C and Fig. 6B were run in the basic direction using pH 3-10 ampholytes. The nonequilibrium pH gradient gels (first dimensions for Fig. 2, A and C, and 6B) were run at 750 V for 20 min. The equilibrium isoelectric focusing gels (first dimension for Fig. 2B and Fig. 6A) were run at 750 V for 2 h. The second dimensions of all two-dimensional gels and all one-dimensional gels were 10% acrylamide SDS gels as formulated by Laemmli (36.Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (205998) Google Scholar) and modified by Dreyfus et al.(37.Dreyfus G. Adams S.A. Choi Y.D. Mol. Cell. Biol. 1984; 4: 415-423Crossref PubMed Scopus (197) Google Scholar). Two-dimensional gels containing radiolabeled protein were prepared for fluorography as described by Laskey and Mills(38.Laskey R.A. Mills A.D. Eur. J. Biochem. 1975; 56: 335-341Crossref PubMed Scopus (3029) Google Scholar).Figure 6Two-dimensional Western blots of spicule matrix proteins using the anti-SM30 and anti-SM50 antiserum. A, 0.5 μg of unlabeled spicule matrix protein was separated in the first dimension on an isoelectric focusing gel using ampholytes with a pH range of 2.5 to 6.5 as described under "Experimental Procedures." The second dimension was a 10% SDS-PAGE gel. The proteins were subjected to Western blotting using the anti-SM30 antiserum. B, 0.5 μg of unlabeled spicule matrix protein was separated in the first dimension on a nonequilibrium pH gradient gel run in the basic direction. The pH range of ampholytes used was 3 to 10. The second dimension was a 10% SDS-PAGE gel. The proteins were subjected to Western blotting analysis using the anti-SM50 antiserum. The immunoreactive proteins in these two blots were visualized using chemiluminescence. NEPHGE, non-equilibrium pH gradient gel electrophoresis.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Generation of Polyclonal AntiseraFusion proteins were engineered and used as immunogens for the generation of polyclonal antisera specific for the proteins encoded by the previously cloned SM50 and SM30 cDNAs. These fusion proteins were generated by subcloning the cDNAs into the maltose-binding protein expression vector pMal-cRI (New England BioLabs). The 1.3-kilobase gt11 cDNA clone, pHS72, which encodes a truncated SM50 protein (168 amino acids of the carboxyl end of the protein)(18.Sucov H.M. Benson S. Robinson J.J. Britten R.J. Wilt F. Davidson E.H. Dev. Biol. 1987; 120: 507-519Crossref PubMed Scopus (81) Google Scholar, 21.Katoh-Fukui Y. Noce T. Ueda T. Fujiwara Y. Hashimoto N. Higashinakagawa T. Killian C. Livingston B. Wilt F. Benson S. Sucov H. Davidson E.H. Dev. Biol. 1991; 145: 201-202Crossref PubMed Scopus (63) Google Scholar), was subcloned into the EcoRI site of the pMal-cRI vector. In addition, the 1.8-kilobase pNG7 gt11 cDNA clone, pNG7, which encodes a complete SM30 protein(22.George N.C. Killian C.E. Wilt F.H. Dev. Biol. 1991; 147: 334-342Crossref PubMed Scopus (82) Google Scholar), was also subcloned into the EcoRI site of pMal-cRI. These engineered fusion protein plasmids were then used to transform XL-1 Escherichia coli (Stratagene). The induction of these fusion constructs, the lysis of the expressing bacteria, and the enrichment of the fusion proteins by affinity chromatography using amylose resin were done as described by the accompanying protocol provided by New England BioLabs. The only deviation was that the bacteria harboring the SM30 maltose-binding protein fusion were grown at 30°C instead of 37°C. This was done to prevent the SM30 fusion protein from becoming insoluble. The SM50 and SM30 fusion proteins were then used as immunogens in rabbits to generate polyclonal antisera following the protocol described by Harlow and Lane(39.Harlow E. Lane D. Antibodies: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Sping Harbor, NY1988Google Scholar). The anti-SM30 antiserum was treated with ammonium sulfate and the immunoglobin fraction was collected and dialyzed as also described by Harlow and Lane(39.Harlow E. Lane D. Antibodies: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Sping Harbor, NY1988Google Scholar). The anti-SM50 antiserum was used without further treatment.A rabbit polyclonal antiserum raised against all of the spicule matrix proteins was generated following the procedure described by Benson et al.(16.Benson S.C. Benson N.C. Wilt F.H. J. Cell Biol. 1986; 102: 1878-1886Crossref PubMed Scopus (97) Google Scholar) and using spicule matrix protein isolated from embryonic spicules as immunogen.Western blotting of one- and two-dimensional gels was done as described by Towbin et al.(40.Towbin H. Stachelin T. Gordon J. Proc. Natl. Acad. Sci. U. S. A. 1979; 76: 4350-4354Crossref PubMed Scopus (44704) Google Scholar). Chemiluminescent detection of immunoreactive proteins was done following the directions of the manufacturer (Amersham) except that 0.1% fish gelatin (Amersham), 0.8% bovine serum albumin, 0.02% Tween 80, 10 mM Tris, pH 8.0, 100 mM NaCl was used as the blocking solution. The anti-SM30 antiserum was used at a 1:2000 dilution and the anti-SM50 antiserum was used at a 1:1000 dilution for the chemiluminesence blots.Detection of immunoreactive proteins using alkaline phosphatase-conjugated secondary antibody and 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium chloride (BCIP/NBT) as substrate was done as described by Richardson et al.(29.Richardson W. Kitajima T. Wilt F. Benson S. Dev. Biol. 1989; 132: 266-269Crossref Scopus (31) Google Scholar). The anti-total spicule matrix antiserum was used at a 1:1000 dilution, the anti-SM30 antiserum was used at a 1:250 dilution, and the anti-SM50 antiserum was used at a 1:100 dilution for these blots.In Vitro Translation of SM30 RNA by Xenopus Oocytes and Reticulocyte LysateThe 1.8-kilobase full-length pNG7 SM30 cDNA (22.George N.C. Killian C.E. Wilt F.H. Dev. Biol. 1991; 147: 334-342Crossref PubMed Scopus (82) Google Scholar) was subcloned into the EcoRI site of pGEM4Z (Promega). This resulting plasmid was then used as a template to synthesize capped in vitro SM30 RNA using the Ambion Megascript kit by following the protocol provided by Ambion.The microinjection of the synthesized RNA into Xenopus oocytes and the collection of [S]methionine-labeled secreted proteins was done as described in Livingston et al.(28.Livingston B.T. Shaw R. Bailey A. Wilt F. Dev. Biol. 1991; 148: 473-480Crossref Scopus (31) Google Scholar). The immunoprecipitation of [S]methionine-labeled proteins using the anti-SM30 antiserum or its preimmune serum at a dilution of 1:100 was also done as described by Livingston et al.(28.Livingston B.T. Shaw R. Bailey A. Wilt F. Dev. Biol. 1991; 148: 473-480Crossref Scopus (31) Google Scholar). SM30 RNA was translated by reticulocyte lysate using the Ambion Reticulocyte Lysate kit following the protocol provided by Ambion.RESULTSEnumeration of the Spicule Matrix ProteinsTo accurately enumerate the proteins comprising the occluded integral matrix of the S. purpuratus sea urchin spicules, spicule matrix proteins were separated using high resolution two-dimensional gel electrophoresis. Given their very acidic makeup, the spicule matrix proteins do not stain very strongly with conventional protein stains such as Coomassie and silver stains. To get a reasonable silver staining signal of spicule matrix proteins on a one-dimensional SDS-PAGE required loading a relatively large amount of protein. This same amount of protein loaded onto a two-dimensional gel resulted in excessive streaking of many of the spicule proteins. We also saw this excessive streaking on two-dimensional gels loaded with isolated spicule matrix proteins that were labeled in vitro with biotin and then subsequently localized using strepavidin-horseradish peroxidase based chemiluminescence. The streaking, we assume, is caused by overloading of the gel and/or the natural tendency of glycoproteins to streak on two-dimensional gels(41.Dunbar B.S. Kimura H. Timmons T.M. Methods Enzymol. 1990; 182: 441-459Crossref PubMed Scopus (61) Google Scholar). This occurrence makes it hard to enumerate reliably the different spicule matrix proteins (data not shown).We decided, alternatively, that radiolabeling the proteins with [S]methionine was a better way to visualize these proteins. We found culturing isolated micromeres in the presence of [S]methionine to be the most effective method of labeling sea urchin spicule matrix proteins. Harkey and Whiteley (43.Harkey M.A. Whiteley
Referência(s)