FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2036476977> ?p ?o ?g. }

• W2036476977 endingPage "11382" @default.
• W2036476977 startingPage "11377" @default.
• W2036476977 abstract "Insulin-like growth factor binding protein 4 (IGFBP-4) is a 24-kDa protein that binds insulin-like growth factor 1 (IGF-1) and IGF-2 with high affinity and inhibits IGF action in vitro. We recently described a protease produced by the B104 neuronal cell line that cleaves IG-FBP-4, yielding an approximate 16-kDa immunoreactive protein that binds IGFs with reduced affinity. We analyzed fragments produced by exposing pure IGFBP-4 to the protease to determine potential cleavage sites. Electrospray mass spectrometry and amino acid sequencing indicated the 16-kDa fragment spanned the NH2 terminus of native IGFBP-4 through Lys-120. There was evidence for an additional proteolytic fragment beginning at amino acid 132 and continuing to the COOH terminus. Proteolysis could be blocked by a synthetic peptide that spanned amino acids 117–126 but not by peptides that contained flanking sequences 111–120 or 125–135. Mutagenesis was used to alter the basic residue at position 120. The expressed mutant IGFBP-4 (K120A) was relatively resistant to cleavage, strongly suggesting that residues 120–121 represent the cleavage site. This region of IGFBP-4 is not homologous with other IGFBPs, explaining the apparent specificity of the protease for IGFBP-4. The 16-kDa IGFBP-4 fragment no longer inhibited IGF-1-stimulated thymidine uptake in vitro, suggesting that proteolytic processing of IGFBP-4 may have important functional consequences in vivo. Insulin-like growth factor binding protein 4 (IGFBP-4) is a 24-kDa protein that binds insulin-like growth factor 1 (IGF-1) and IGF-2 with high affinity and inhibits IGF action in vitro. We recently described a protease produced by the B104 neuronal cell line that cleaves IG-FBP-4, yielding an approximate 16-kDa immunoreactive protein that binds IGFs with reduced affinity. We analyzed fragments produced by exposing pure IGFBP-4 to the protease to determine potential cleavage sites. Electrospray mass spectrometry and amino acid sequencing indicated the 16-kDa fragment spanned the NH2 terminus of native IGFBP-4 through Lys-120. There was evidence for an additional proteolytic fragment beginning at amino acid 132 and continuing to the COOH terminus. Proteolysis could be blocked by a synthetic peptide that spanned amino acids 117–126 but not by peptides that contained flanking sequences 111–120 or 125–135. Mutagenesis was used to alter the basic residue at position 120. The expressed mutant IGFBP-4 (K120A) was relatively resistant to cleavage, strongly suggesting that residues 120–121 represent the cleavage site. This region of IGFBP-4 is not homologous with other IGFBPs, explaining the apparent specificity of the protease for IGFBP-4. The 16-kDa IGFBP-4 fragment no longer inhibited IGF-1-stimulated thymidine uptake in vitro, suggesting that proteolytic processing of IGFBP-4 may have important functional consequences in vivo. The insulin-like growth factors (IGFs) 1The abbreviations used are: IGF, insulin-like growth factor; IGFBP, IGF binding protein; HPLC, high pressure liquid chromatography; PTH, phenylthiohydantoin; CHO, Chinese hamster ovary; ESMS, electrospray mass spectrometry.1The abbreviations used are: IGF, insulin-like growth factor; IGFBP, IGF binding protein; HPLC, high pressure liquid chromatography; PTH, phenylthiohydantoin; CHO, Chinese hamster ovary; ESMS, electrospray mass spectrometry. are mitogens produced by many tissues and are postulated to act in a paracrine/autocrine manner (1Cohick W.S. Clemmons D.R. Annu. Rev. Physiol. 1993; 55: 131-153Crossref PubMed Scopus (571) Google Scholar). IGF action is thought to be greatly influenced by the IGF binding proteins (IGFBPs), a family of secreted proteins that bind the IGFs with high affinity (2Clemmons D.R. Mol. Reprod. Dev. 1993; 35: 368-375Crossref PubMed Scopus (142) Google Scholar). These proteins are expressed by a wide variety of tissues and may direct transport of the IGFs into organs. Also, in vitro experiments have shown that IGFBPs can either inhibit or enhance IGF action depending upon the experimental design and the form of IGFBP examined (3Elgin R.G. Busby W.H. Clemmons D.R. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 3254-3258Crossref PubMed Scopus (664) Google Scholar, 4Busby Jr., W.H. Klapper D.G. Clemmons D.R. J. Biol. Chem. 1988; 263: 14203-14210Abstract Full Text PDF PubMed Google Scholar, 5Andress D.L. Birnbaum R.S. J. Biol. Chem. 1992; 267: 22467-22472Abstract Full Text PDF PubMed Google Scholar, 6Andress D.L. Loop S.M. Zapf J. Kiefer M.C. Biochem. Biophys. Res. Commun. 1993; 195: 25-30Crossref PubMed Scopus (116) Google Scholar, 7De Mellow J.S.M. Baxter J. Biochem. Biophys. Res. Commun. 1988; 156: 199-204Crossref PubMed Scopus (488) Google Scholar), suggesting that the IGFBPs modulate local IGF bioactivity in vivo. Six IGFBPs have been described (8Shimasaki S. Ling N. Prog. Growth Factor Res. 1991; 3: 243-266Abstract Full Text PDF PubMed Scopus (645) Google Scholar). Although structurally related, they are the products of distinct genes that have differing tissue and developmental regulation of expression. Each IGFBP contains amino- and carboxyl-terminal regions that are homologous among the IGFBPs and rich in cystine or cysteine residues. In contrast, the middle third of the molecules is usually free of disulfides and has little shared sequence between the various IGFBPs. All IGFBPs undergo some form of post-translational modification. IGFBP-3, -4, -5, and -6 manifest variable degrees of glycosylation (8Shimasaki S. Ling N. Prog. Growth Factor Res. 1991; 3: 243-266Abstract Full Text PDF PubMed Scopus (645) Google Scholar). Phosphorylation of IGFBP-1 alters its affinity for IGFs and biologic actions (9Jones J.I. D'Ercole A.J. Camacho-Hubner C. Clemmons D.R. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 7481-7485Crossref PubMed Scopus (274) Google Scholar). Recently, extracellular proteolytic processing has emerged as a frequent and potentially important mechanism governing the biologic properties of the IGFBPs (10Conover C.A. Kiefer M.C. Zapf J. J. Clin. Invest. 1993; 91: 1129-1137Crossref PubMed Scopus (171) Google Scholar, 11Fowlkes J. Freemark M. Endocrinology. 1992; 131: 2071-2076Crossref PubMed Scopus (71) Google Scholar, 12Schmid C. Rutishauser J. Sclapfer I. Froesch E.R. Zapf J. Biochem. Biophys. Res. Commun. 1991; 179: 579-585Crossref PubMed Scopus (114) Google Scholar, 13Davenport M.L. Pucilowska J. Clemmons D.R. Lundblad R. Spencer J.A. Underwood L.E. Endocrinology. 1992; 130: 2505-2512Crossref PubMed Scopus (35) Google Scholar, 14Lassarre C. Binoux M. Endocrinology. 1994; 134: 1254-1262Crossref PubMed Scopus (105) Google Scholar). The responsible proteases have, in general, not been identified, although the IGFBP-3 protease produced by human fibroblasts in culture may be a member of the matrix metalloprotease family and the IGFBP-3 protease in seminal fluid appears to be prostate-specific antigen (15Cohen P. Graves H.C. Peehl D. Kamasai M. Guidice L.C. Rosenfeld R.G. J. Clin. Endocrinol. & Metab. 1992; 75: 1046-1053Crossref PubMed Scopus (0) Google Scholar), a serine protease. A protease(s) that cleaves IGFBP-4 is present in media of cultured fibroblasts (10Conover C.A. Kiefer M.C. Zapf J. J. Clin. Invest. 1993; 91: 1129-1137Crossref PubMed Scopus (171) Google Scholar, 11Fowlkes J. Freemark M. Endocrinology. 1992; 131: 2071-2076Crossref PubMed Scopus (71) Google Scholar), decidual cells (16Myers S.E. Cheung P.T. Handwerger S. Chernausek S.D. Endocrinology. 1993; 133: 1525-1531Crossref PubMed Scopus (36) Google Scholar), osteoblasts (17Okazaki R. Riggs L.B. Conover C.A. Endocrinology. 1994; 134: 126-162Crossref PubMed Scopus (115) Google Scholar, 18Kanzaki S. Hilliker S. Baylink D.J. Mohan S. Endocrinology. 1994; 134: 383-392Crossref PubMed Scopus (109) Google Scholar, 19Durham S.K. Kiefer M.C. Riggs B.L. Conover C.A. J. Bone Miner. Res. 1994; 9: 111-117Crossref PubMed Scopus (85) Google Scholar), and smooth muscle cells (20Cohick W.S. Gockerman A. Clemmons D.R. J. Cell. Physiol. 1993; 157: 52-60Crossref PubMed Scopus (79) Google Scholar). An unusual feature is that the IGFBP-4 ligand, IGF-1 or -2, greatly accelerates the rate of proteolysis when present in the reaction mixture (10Conover C.A. Kiefer M.C. Zapf J. J. Clin. Invest. 1993; 91: 1129-1137Crossref PubMed Scopus (171) Google Scholar, 11Fowlkes J. Freemark M. Endocrinology. 1992; 131: 2071-2076Crossref PubMed Scopus (71) Google Scholar, 16Myers S.E. Cheung P.T. Handwerger S. Chernausek S.D. Endocrinology. 1993; 133: 1525-1531Crossref PubMed Scopus (36) Google Scholar). Although the exact mechanism has not yet been delineated, most data suggest that binding of an IGF by the IG-FBP-4 makes the latter more susceptible to proteolytic cleavage, probably through a conformational change. We recently described a glucocorticoid-inducible protease produced by the rat B104 neuronal cell line (21Cheung P.T. Wu J. Banach W. Chernausek S.D. Endocrinology. 1994; 135: 1328-1335Crossref PubMed Scopus (47) Google Scholar). This protease cleaves intact 24-kDa IGFBP-4 to a smaller, approximate 16-kDa species that binds IGFs with a reduced affinity. In order to better understand the nature of IGFBP-4 proteolysis, experiments were performed to identify proteolytic cleavage sites within the IGFBP-4 molecule and to determine the effect of mutagenesis on IGFBP-4 stability. The rat IGFBP-4 antibody raised against a synthetic 20-amino-acid peptide (residues 81–100) was described previously (22Chernausek S.D. Murray M.A. Cheung P.T. Regul. Pept. 1993; 48: 123-132Crossref PubMed Scopus (22) Google Scholar). An additional rat IGFBP-4 antibody raised against amino acids 111–120 was provided by N. Ling (Whittier Institute, La Jolla, CA), along with the corresponding synthetic peptide (no. 1). Peptides 2 and 3 (see Fig. 3) were synthesized by a commercial vender (Anaspec, San Jose, CA) and were judged to be greater than 90% pure. Rat IGFBP-4 was prepared from media conditioned by the B104 cell line as described previously (21Cheung P.T. Wu J. Banach W. Chernausek S.D. Endocrinology. 1994; 135: 1328-1335Crossref PubMed Scopus (47) Google Scholar). Restriction endonucleases and molecular biology reagents were purchased from New England Biolabs, Beverly, MA. IGF-1 was provided by Lilly. Samples were subjected to 4%/12% discontinuous SDS-polyacrylamide gel electrophoresis and electroblotted to nitrocellulose membranes (Schleicher and Schuell). Radioligand blotting using 125I-IGF-1 was performed by the method of Hossenlopp et al. (23Hossenlopp P. Seurin D. Segovia B. Portolan G. Binoux M. Eur. J. Biochem. 1987; 170: 133-142Crossref PubMed Scopus (76) Google Scholar). Immunoblotting was as described previously (22Chernausek S.D. Murray M.A. Cheung P.T. Regul. Pept. 1993; 48: 123-132Crossref PubMed Scopus (22) Google Scholar). Immunoreactive species were identified by either an avidin-biotin alkaline phosphatase conjugate system (Vectastain, Vector Labs, Inc., Burlingame, CA) or by chemiluminescent detection (Amersham Corp.). Media containing the IGFBP-4 protease were prepared by dexamethasone treatment of a subline of B104 cells (B104ns) that produced little IGFBP-4 (24Cheung P.T. Smith E.P. Shimasaki S. Ling N. Chernausek S.D. Endocrinology. 1991; 129: 1006-1015Crossref PubMed Scopus (107) Google Scholar). Cells were grown to confluency in fetal calf serum-containing media and then switched to serum-free media containing 100 nM dexamethasone (Sigma). After 2–3 days of exposure, harvested media (Dex-media) were centrifuged to remove cellular debris and saved at –70 °C. When required, conditioned media were concentrated using a microconcentrator (Centricon-10, Amicon, Beverly, MA). Twenty-five µg of IGFBP-4 was exposed to the IGFBP-4 protease for 48 h. The pure peptide was incubated with 0.5 ml of Dex-media containing 0.1 volume of 0.5 m Tris buffer, pH 7.4. After 24 h at 37 °C, an additional 0.5 ml of media was added followed by a final 0.22 ml 12 h later. The reaction mixture was stored frozen and subsequently applied to an IGF-1 affinity column (binding capacity, 400–600 µg of IGFBP-4) equilibrated in 0.05 m NaH2PO4, pH 6.6, containing 2 nut EDTA and 0.1 m NaCl. After 16 h of recirculation, retained IGFBP-4 was eluted with 0.5 m acetic acid. Fractions containing the immunoreactive 16-kDa IGFBP-4 fragment (determined by immunoblotting) were loaded onto a reverse phase HPLC column (C4 Vydac, Hespena, CA) equilibrated in 0.04% trifluoroacetic acid and H2O. Elution was accomplished with a 0–60% CH3CN gradient over 1 h. Fractions were screened by immunoblotting, and those containing 16-kDa IGFBP-4 were subjected to further analysis. Reverse-phase liquid chromatography fractions were evaporated to dryness and then dissolved in water:methanol: formic acid (50:50:1, v/v/v). Samples were introduced to a Sciex API III triple-quadrupole mass spectrometer (Sciex, Inc., Toronto, Canada) at 4 µl/min with a syringe pump. The device was equipped with an atmospheric pressure ion source used to sample positive ions produced from a pneumatically assisted electrospray interface. Sample ions generated by the electrospray process were mass-analyzed by scanning the first quadrupole in 0.1-atomic mass unit increments over approximately 10 s and passing mass-selected ions through the second and third quadruples, operated in the radiofrequency-only mode, to a channeltron detector. Mass calibration of the instrument was accomplished just prior to sample analysis by matching ions of polypropylene glycol to known reference masses stored in the mass calibration table of the mass spectrometer. Ten scans were averaged for each analysis. The average molecular weights of the sample peptides were calculated from the multiply protonated ions in the mass spectrum. Automated Edman degradation chemistry was used to determine the NH2-terminal protein sequence. An Applied Biosystems, Inc. model 470A gas phase sequencer (Foster City, CA) was employed for the degradations using the standard sequencer cycle, 03RPTH. The respective PTH-derivatives were identified by reverse phase HPLC analysis in an on-line fashion employing an Applied Biosystems, Inc. model 120A PTH analyzer fitted with a Brownlee 2.1-mm (inner diameter) PTH-C18 column. A full-length IGFBP-4 cDNA in pBluescript SK+ (pRBP4–503) served as a template for mutagenesis and was kindly provided by Dr. S. Shimasaki of the Whittier Institute (La Jolla, CA). In preparing the mutant IGFBP-4, the unique restriction sites MscI (near the lysine 120 codon) and SphI (approximately 200 base pairs 5′ to the lysine 120 codon) were exploited. An oligonucleotide primer pair was synthesized in which the upper primer contained the rat IGFBP-4 sequence that overlapped the SphI site (5′ CTGTGGCTCAGGCAT-GCGCGGC 3′) whereas the lower primer contained the MscI site as well as the introduced mutation (5′ ACTTTGGCCATATGCGCCTGCAGGC 3′). These primers were used to generate the 219-base pair DNA fragment by polymerase chain reaction using Vent DNA polymerase (New England Biolabs, Beverly, MA). The polymerase chain reaction was run for 23 cycles in buffers provided by the manufacturer using a 57 °C annealing temperature. The product was exposed to SphI and MscI and after isolation using agarose gel electrophoresis, ligated back into pRBP4–503 from which the matching sequence had been removed by enzymatic digestion. Integrity of the mutant was verified by DNA sequencing. The mutant IGFBP-4 cDNA was transferred to a pUC18-based mammalian expression vector that utilized a mouse metallothionein promoter (pNUT, provided by A. J. D'Ercole, University of North Carolina). Chinese hamster ovary cells (CHO K-1) were transfected by the following method. The cells were maintained in α-minimal essential medium containing 10% fetal calf serum supplemented with penicillin and streptomycin. Twenty-four hours before transfection the cells were seeded into 6-well tissue culture plates to approximately 15% confluency. DNA was introduced into the cells by a calcium phosphate precipitation procedure. The DNA/calcium precipitate was formed by mixing 0.5 ml of 0.25 m calcium chloride containing 10 µg of plasmid DNA in 0.5 ml of 2 × Hanks’ basic salt solution. After incubation at room temperature for 15 min, 200 µl (2 µg of DNA) were applied to the cells in 2 ml of medium. Plates were incubated at 37 °C for 5 h after which calcium-containing media were removed and medium containing 10% glycerol was applied for 3 min. After extensive rinsing this medium was replaced, and the cells were incubated for 48 h. Treated cells were then trypsinized and plated 1:50 in 100-cm dishes with medium containing 1 µg/ml methotrexate. Fresh methotrexate was reapplied every 3–4 days for 10–12 days when colonies of stably transfected cells began to appear. Colonies were isolated by cloning rings, trypsinized, and transferred to individual wells of a 24-well plate. Medium was assayed for IGFBP-4 secretion by immunoblotting after the cells reached confluency. The methotrexate concentration was gradually increased to 50 µg/ml in the surviving clones. Approximately 1 liter of conditioned medium was collected from CHO cells expressing the mutant IGFBP-4. The medium was clarified by centrifugation at 16,000 × g for 20 min, the pH was adjusted to 6.6 with Tris base, and then the medium was loaded at 1 mlAnin at 4 °C onto a 4.4 × 3-cm phenyl-Sepharose column (Pharmacia Biotech Inc.) equilibrated with 0.05 m sodium phosphate buffer, pH 6.6, containing 0.1 m NaCl and 2 mM EDTA. After washing, a linear gradient of 0–20% CHaCN in 0.05 m sodium phosphate, pH 6.6, 2 mM EDTA was used to elute bound proteins. The active fractions were identified by immunoblotting and partially lyophilized to remove the CH3CN. The active pool was subjected to IGF-1 affinity column chromatography as described above. The partially lyophilized eluate was injected onto a Vydac C-4 reverse phase column (4.6 cm × 15 mm). An isocratic elution with 0.4% trifluoroacetic acid was conducted for 10 min followed by a 0–24% linear gradient of CH3CN over 10 min. For the next 20 min the CH3CN concentration was increased linearly from 24 to 34% and for the final 24 min from 34 to 100%. The mutant rat IGFBP-4 eluted as a single peak at 28–30% CH3CN. The material was shown to be pure by both immunoblotting and ligand blotting. B104ns cells were used as targets for IGF action. Tritiated thymidine uptake in response to added IGF-1 was performed as described previously (24Cheung P.T. Smith E.P. Shimasaki S. Ling N. Chernausek S.D. Endocrinology. 1991; 129: 1006-1015Crossref PubMed Scopus (107) Google Scholar). Cells were exposed to IGF-1 and graded concentrations of intact IGFBP-4 or the 16-kDa fragment in serum-free media for 24 h prior to the addition of the [3H]thymidine. Proteolytic products of IGFBP-4 were initially examined using antibodies directed against specific IGFBP-4 peptide sequences. Antisera produced using amino acids 81–100 and 110–120 of the mature IGFBP-4 molecule both recognize a proteolytic fragment of approximately 16 kDa, indicating the 16-kDa species overlaps, at least in part, amino acids 80–120 (Fig. 1). Further characterization of the 16-kDa species was accomplished by subjecting purified 16-kDa IGFBP-4 (see “Experimental Procedures”) to NH2-terminal amino acid sequence analysis and electrospray mass spectrometry (ESMS). The amino acid sequence obtained was DEAIH(C)PP(C)S (where (C) indicates a possible cysteine), which corresponds to the published NH2 terminus of native rat IGFBP-4. Following this, the exact molecular weight (Mr) of the isolated fragment was determined by ESMS. Analysis of the data indicated two species present in the isolate (Fig. 2); the observed Mr of one matches exactly that expected for an IGFBP-4 protein containing amino acids 1–120 while the other matched with amino acids 1–90. Because the 16-kDa IGFBP-4 seen on immunoblot reacted with the antibody raised against amino acids 110–120, it must represent the 1–120 species. The potential cleavage sites are indicated upon the schematic (Fig. 3).FIG. 2Mass spectrometric analysis of 16-kDa IGFBP-4. The observed and expected Mr of two putative fragments and their potential amino acid sequences are given in the table at the bottom. N refers to the number of mass determinants for each. Charge states corresponding to the species of differing Mr are indicated.View Large Image Figure ViewerDownload Hi-res image Download (PPT) The data described suggest a putative cleavage site between Lys-120 and His-121. To assess the potential for cleavage at this site, the ability of synthetic peptides to inhibit the proteolysis of intact IGFBP-4 was examined. Peptides that either spanned or flanked the putative cleavage site (see Fig. 3) were incubated at high molar excess with pure native IGFBP-4 and Dex-media (Fig. 4). The synthetic peptide containing amino acids 117–126 inhibited proteolysis while the flanking peptides were ineffective. Site-directed mutagenesis was used to alter the IGFBP-4 molecule, changing the basic lysine at position 120 to alanine. The mutant (K120A IGFBP-4) was expressed in CHO cells, isolated from conditioned media, and tested for sensitivity to proteolytic cleavage. FIG. 5, FIG. 6 show the cleavage over time of the K120A IGFBP-4 compared with native IGFBP-4 in the presence and absence of IGF-1. Without added IGF-1, proteolytic degradation is minimal (Fig. 5). The addition of IGF-1 significantly enhanced cleavage of both mutant and native IGFBP-4. Although proteolytic cleavage of the mutant IG-FBP-4 occurred eventually, the appearance of the 16-kDa species was significantly delayed compared with wild type IG-FBP-4. The difference in apparent cleavage rates for native and mutant IGFBP-4 was more accurately quantified by incubating the purified proteins with the protease and assessing the decline in 24-kDa species over time through PhosphorImage analysis of a ligand blot (Fig. 6). This experiment confirmed that the proteolytic degradation of K120A IGFBP-4 is slower than that of the wild type protein. It also appears that 28-kDa (glycosylated) IGFBP-4 is relatively resistant to the protease.FIG. 6Top, comparison of proteolytic degradation of native (WT) and mutant IGFBP-4 (MUT) assessed by ligand blot. Pure peptides were exposed to the protease-containing media and IGF-I and sampled at the times indicated. In this experiment 100 ng of IGFBP-4 were incubated with 50 µl of media on a total reaction volume of 60 µl. Bottom, PhosphorImage analysis of the ligand blot data. The experiment shown in the top panel was analyzed using the PhosphorImaging system with the numeric data plotted as percent of control (value at t = 0). The data for the N-glycosylated form of IGFBP-4 was derived by analysis of the less intense 28-kDa band observed just above the wild type IGFBP-4 band.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Examination of the ligand blot data in Fig. 6 suggests greater resistance of the mutant to proteolytic cleavage than is apparent in Fig. 5. This likely reflects differences in time of exposure to the enzyme (15 versus 24 h) but also may reflect different levels of enzymatic activity found in differing batches of conditioned media. In multiple experiments, analyzed by both immunoblot and ligand blot, degradation of the K120A mutant IGFBP-4 was always attenuated relative to that of the wild type, but the amount of intact 24-kDa IGFBP-4 remaining at the end of each experiment varied. This is interpreted to mean the K120A IGFBP-4 is resistant, but not invulnerable, to cleavage by the B104 protease. In light of the highly concordant immunologic amino acid sequence and ESMS data, substantial cleavage of the mutant IGFBP-4 was surprising due to the significant charge change expected by substituting alanine at position 120. In order to understand this better, the proteolytic fragments from the mutant were analyzed by amino acid sequencing and ESMS. The K120A IGFBP-4 was cleaved by exposure to Dex-media in the presence of 100 ng/ml IGF-1. The mixture was subjected to HPLC as described under “Experimental Procedures,” and a fraction containing the 16-kDa species was mass-analyzed. De-convolution of the mass spectra yielded observed Mr of 13,110.3 ± 0.8 and 11,322.2 ± 0.35 (mean ± S.D.). The Mr of the larger species matches the theoretical sequence from the NH2 terminus through histidine 121 (expected Mr 13,112, accounting for the replacement by alanine at position 120). Scanning of the IGFBP-4 sequence for a string of amino acids matching the Mr of the smaller species yielded a sequence beginning at lysine 132 and continuing through the carboxyl terminus (expected Mr 11,324). In order to verify these data, proteins contained in a fraction adjacent to that mass analyzed (also containing the 16-kDa species by immunoblot) were subjected to NH2-terminal sequence analysis. The data indicated three distinct IG-FBP-4 sequences. One corresponded to the normal NH2 terminus of intact IGFBP-4. The second began at lysine 132 (KWGTPREEPRPVPQ), while the other began with valine 134 (VXTPREEXXXVXQGS, where X indicates inability to assign a residue for that position). The first two are thus compatible with ESMS data for K120A IGFBP-4 peptides of 1–121 and 132 through the COOH terminus. Concentration of the third species may have been too low for detection by ESMS. In a prior publication we showed that cleaved IGFBP-4 is capable of binding IGF-1 but with reduced affinity (21Cheung P.T. Wu J. Banach W. Chernausek S.D. Endocrinology. 1994; 135: 1328-1335Crossref PubMed Scopus (47) Google Scholar). In order to determine the effect of cleavage on biologic properties of IGFBP-4, the ability of intact IGFBP-4 to inhibit IGF-1-stimulated thymidine uptake was compared with that of cleaved IGFBP-4. The fragment was isolated by HPLC as above. Intact native 24-kDa IGFBP-4 produced the expected dose-dependent inhibition of IGF-1, using B104 cells as a target (Fig. 7). However, equivalent concentrations of cleaved IGFBP-4 were ineffective, indicating that proteolytic processing of IGFBP-4 changes its properties such that it no longer inhibits IGF-1 action. Our data indicate that IGFBP-4 is proteolytically cleaved into two major fragments following exposure to a protease from B104 cells. One fragment begins at the NH2 terminus and extends through amino acid 120. The other contains residue 132 through the COOH terminus. The B104 cell-derived protease appears to act at the bond between Lys-120 and His-121 (a trypsin-susceptible site) because proteolysis is blocked by an overlapping synthetic peptide and because mutation of lysine 120 attenuates the rate of cleavage. The residues surrounding this putative cleavage site (amino acids 116–121) are conserved between the human and rat but without a homologous sequence in the other IGFBPs (8Shimasaki S. Ling N. Prog. Growth Factor Res. 1991; 3: 243-266Abstract Full Text PDF PubMed Scopus (645) Google Scholar). This is compatible with the reports of potentially related IGFBP-4 proteases from other species that appear to cleave IG-FBP-4 selectively (10Conover C.A. Kiefer M.C. Zapf J. J. Clin. Invest. 1993; 91: 1129-1137Crossref PubMed Scopus (171) Google Scholar, 11Fowlkes J. Freemark M. Endocrinology. 1992; 131: 2071-2076Crossref PubMed Scopus (71) Google Scholar, 16Myers S.E. Cheung P.T. Handwerger S. Chernausek S.D. Endocrinology. 1993; 133: 1525-1531Crossref PubMed Scopus (36) Google Scholar, 18Kanzaki S. Hilliker S. Baylink D.J. Mohan S. Endocrinology. 1994; 134: 383-392Crossref PubMed Scopus (109) Google Scholar, 19Durham S.K. Kiefer M.C. Riggs B.L. Conover C.A. J. Bone Miner. Res. 1994; 9: 111-117Crossref PubMed Scopus (85) Google Scholar, 25Kamyar A. Pirola C.J. Wang H.-M. Sharifi B. Mohan S. Forrester J.S. Fagin J.A. Circ. Res. 1994; 74: 576-585Crossref PubMed Scopus (52) Google Scholar). An IGFBP-4 fragment spanning the NH2 terminus through amino acid 90 was also detected by ESMS in one experiment. The origin and significance of this species is not clear. It might reflect further processing of the 1–120 species and/or be the result of the action of a distinct protease. Although we could not detect a 1–90 species by immunoblotting, in a prior publication we observed an appropriate sized fragment when radiolabeled IGFBP-4 was exposed to the protease (21Cheung P.T. Wu J. Banach W. Chernausek S.D. Endocrinology. 1994; 135: 1328-1335Crossref PubMed Scopus (47) Google Scholar). The relative amount of that species was very low, and therefore we believe the 1–90 IGFBP-4 is probably a minor component in the proteolytic processing, at least as applied to the B104 system. Fowlkes et al. (26Fowlkes J.L. Enghild J.J. Suzuki K. Nagase H. J. Biol. Chem. 1994; 269: 25742-25746Abstract Full Text PDF PubMed Google Scholar) have recently examined the proteolysis of IGFBP-3 in human dermal fibroblast medium. They attributed the proteolytic activity to one or more matrix metalloproteinases that cleaved IGFBP-3 within its nonhomologous domain. Because cleavage of IGFBP-3 does not occur in B104 conditioned media (21Cheung P.T. Wu J. Banach W. Chernausek S.D. Endocrinology. 1994; 135: 1328-1335Crossref PubMed Scopus (47) Google Scholar), the B104 protease is unlikely to be identical to those described by Fowlkes el al. (26Fowlkes J.L. Enghild J.J. Suzuki K. Nagase H. J. Biol. Chem. 1994; 269: 25742-25746Abstract Full Text PDF PubMed Google Scholar). However, recognition of amino acid sequences within the nonhomologous regions of the IGFBPs by IGFBP proteases could be a general mechanism to explain specificity among the IGFBP proteases. The single amino acid substitution we introduced was insufficient to abolish cleavage. It is possible that the mutation simply did not produce enough of a change to render the IG-FBP-4 completely insensitive to cleavage at or around that site and that the His-121-Met-122 bond served as an alternate site. It is also possible that the primary cleavage site is not between amino acids 120 and 121 but a few residues toward the COOH terminus and that processing by a carboxypeptidase removes additional amino acids to produce the large NH2-terminal fragment. Indeed, there are many basic amino acids between residues 120 and 132, indicating several potential cleavage sites. Although it is possible that initial cleavage occurs as distal as residues 132–134, that seems unlikely since the 125–135 synthetic peptide had no effect on proteolysis. A third possibility is that more than one IGFBP-4 protease in B104 cell media may produce the fragments. Because the major fragments observed by ESMS are not contiguous, proteolysis must occur at least at two points in the molecule. The exact relationship between the protease found in the media of B104 cells and that described in other systems is not yet clear. The enhancing of proteolytic cleavage by IGF-1 is similar to that described for human and sheep fibroblasts (10Conover C.A. Kiefer M.C. Zapf J. J. Clin. Invest. 1993; 91: 1129-1137Crossref PubMed Scopus (171) Google Scholar, 11Fowlkes J. Freemark M. Endocrinology. 1992; 131: 2071-2076Crossref PubMed Scopus (71) Google Scholar), human decidua (16Myers S.E. Cheung P.T. Handwerger S. Chernausek S.D. Endocrinology. 1993; 133: 1525-1531Crossref PubMed Scopus (36) Google Scholar), vascular smooth muscle cells (25Kamyar A. Pirola C.J. Wang H.-M. Sharifi B. Mohan S. Forrester J.S. Fagin J.A. Circ. Res. 1994; 74: 576-585Crossref PubMed Scopus (52) Google Scholar), and osteoblasts (18Kanzaki S. Hilliker S. Baylink D.J. Mohan S. Endocrinology. 1994; 134: 383-392Crossref PubMed Scopus (109) Google Scholar, 19Durham S.K. Kiefer M.C. Riggs B.L. Conover C.A. J. Bone Miner. Res. 1994; 9: 111-117Crossref PubMed Scopus (85) Google Scholar). In our original description of the B104 cell protease, the addition of IGF-I was not required to demonstrate proteolysis (21Cheung P.T. Wu J. Banach W. Chernausek S.D. Endocrinology. 1994; 135: 1328-1335Crossref PubMed Scopus (47) Google Scholar), but we believe that this is because the B104 cells secrete small amounts of IGFs. 2S. D. Chernausek, unpublished data. The response to various protease inhibitors has been similar, but not identical, among the proteases even though proteolytic fragments produced by each appear to be of similar sizes. The protease studied here may be the rat homologue of the IGFBP-4 protease found in other species, but there may be multiple IGFBP-4 proteases as well. In a prior publication (21Cheung P.T. Wu J. Banach W. Chernausek S.D. Endocrinology. 1994; 135: 1328-1335Crossref PubMed Scopus (47) Google Scholar) we found that proteolytically cleaved IGFBP-4 could bind IGFs with an affinity that was reduced approximately 15-fold. Since the affinity of IGFBP-4 for IGF-I is approximately 1 log order higher than the affinity of the receptor for IGF-1 (27Orlowski C.C. Chernausek S.D. Akeson R. J. Cell. Physiol. 1989; 139: 469-476Crossref PubMed Scopus (26) Google Scholar, 28Bach L. Hsieh S. Sakano K. Fujiwara H. Perdue J.F. Rechler M.M. J. Biol. Chem. 1993; 268: 9246-9254Abstract Full Text PDF PubMed Google Scholar), such a change might be predicted to abrogate the inhibitory properties of IGFBP-4, as was found in our experiments. This is in accord with prior reports showing that, in other systems, the growth inhibitory properties of IGFBP-4 are lost following exposure to the protease (10Conover C.A. Kiefer M.C. Zapf J. J. Clin. Invest. 1993; 91: 1129-1137Crossref PubMed Scopus (171) Google Scholar, 20Cohick W.S. Gockerman A. Clemmons D.R. J. Cell. Physiol. 1993; 157: 52-60Crossref PubMed Scopus (79) Google Scholar). Although the proteolytic processing of IGFBP-4 is likely to be an important mechanism regulating IGFBP-4 and IGF action, many questions remain. It is not known whether the 16-kDa IGFBP-4TGF complex retains any activity and whether the complex can be bound by the IGF-I receptor as a functional unit. Whether any of the IGFBP-4 cleavage products have functions independent of the IGFs is also unknown. Nevertheless, a plausible scenario for the physiologic role of the IGFBP-4 protease might be as follows. The IGFBP-4 protease is induced or activated when IGF action is needed. The protease is specific for IGFBP-4 and preferentially attacks IGFBP-4 that is complexed with an IGF. Cleavage of IGFBP-4 releases sequestered IGFs and allows them to act on neighboring responsive cells, thereby providing a mechanism for directing the paracrine/autocrine actions of the IGFs. Subsequent to the submission of this work Conover et al. (Conover, C. A., Durham, S. K., Zapf, J., Masiarz, F. R. & Kiefer, M. C. (1995) J. Biol. Chem. 270, 4395–4400) described the cleavage of human IGFBP-4 by an IGFBP-4 protease. A fragment of IGFBP-4 was isolated that appeared to be homologous to the COOH-terminal fragment we identified. Mutations in the vicinity of the site equivalent to residues 132–134 of rat IGFBP-4 were associated with some reduction in protease sensitivity, suggesting that this region of IGFBP-4 might also be involved in the cleavage process. We thank Jennifer Wu and Walter Banach for technical assistance, Jack Jones for performing CHO cell transfections, and Nicholas Ling and Shunichi Shimasaki for provision of reagents. We also acknowledge the assistance of Vicki Livengood with manuscript preparation." @default.
• W2036476977 created "2016-06-24" @default.
• W2036476977 creator A5000288481 @default.
• W2036476977 creator A5039368198 @default.
• W2036476977 creator A5047854064 @default.
• W2036476977 creator A5055241891 @default.
• W2036476977 creator A5072130761 @default.
• W2036476977 creator A5091418550 @default.
• W2036476977 date "1995-05-01" @default.
• W2036476977 modified "2023-10-15" @default.
• W2036476977 title "Proteolytic Cleavage of Insulin-like Growth Factor Binding Protein 4 (IGFBP-4)" @default.
• W2036476977 cites W1562667342 @default.
• W2036476977 cites W1569741533 @default.
• W2036476977 cites W1579811372 @default.
• W2036476977 cites W1602528313 @default.
• W2036476977 cites W1828467840 @default.
• W2036476977 cites W1978275394 @default.
• W2036476977 cites W1978543003 @default.
• W2036476977 cites W1989567446 @default.
• W2036476977 cites W1997209243 @default.
• W2036476977 cites W2007006137 @default.
• W2036476977 cites W2014264595 @default.
• W2036476977 cites W2017530941 @default.
• W2036476977 cites W2018723236 @default.
• W2036476977 cites W2023841557 @default.
• W2036476977 cites W2031340215 @default.
• W2036476977 cites W2032558447 @default.
• W2036476977 cites W2034321052 @default.
• W2036476977 cites W2046337780 @default.
• W2036476977 cites W2048627616 @default.
• W2036476977 cites W2050824531 @default.
• W2036476977 cites W2060577528 @default.
• W2036476977 cites W2090099477 @default.
• W2036476977 cites W2107160391 @default.
• W2036476977 cites W2112514894 @default.
• W2036476977 cites W2120253940 @default.
• W2036476977 cites W2153045976 @default.
• W2036476977 cites W2176506025 @default.
• W2036476977 doi "https://doi.org/10.1074/jbc.270.19.11377" @default.
• W2036476977 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/7538115" @default.
• W2036476977 hasPublicationYear "1995" @default.
• W2036476977 type Work @default.
• W2036476977 sameAs 2036476977 @default.
• W2036476977 citedByCount "78" @default.
• W2036476977 countsByYear W20364769772015 @default.
• W2036476977 countsByYear W20364769772021 @default.
• W2036476977 crossrefType "journal-article" @default.
• W2036476977 hasAuthorship W2036476977A5000288481 @default.
• W2036476977 hasAuthorship W2036476977A5039368198 @default.
• W2036476977 hasAuthorship W2036476977A5047854064 @default.
• W2036476977 hasAuthorship W2036476977A5055241891 @default.
• W2036476977 hasAuthorship W2036476977A5072130761 @default.
• W2036476977 hasAuthorship W2036476977A5091418550 @default.
• W2036476977 hasBestOaLocation W20364769771 @default.
• W2036476977 hasConcept C151730666 @default.
• W2036476977 hasConcept C170493617 @default.
• W2036476977 hasConcept C175156509 @default.
• W2036476977 hasConcept C181199279 @default.
• W2036476977 hasConcept C185592680 @default.
• W2036476977 hasConcept C2775960820 @default.
• W2036476977 hasConcept C2780378035 @default.
• W2036476977 hasConcept C2780689927 @default.
• W2036476977 hasConcept C2781307694 @default.
• W2036476977 hasConcept C43369102 @default.
• W2036476977 hasConcept C51639874 @default.
• W2036476977 hasConcept C55493867 @default.
• W2036476977 hasConcept C86803240 @default.
• W2036476977 hasConceptScore W2036476977C151730666 @default.
• W2036476977 hasConceptScore W2036476977C170493617 @default.
• W2036476977 hasConceptScore W2036476977C175156509 @default.
• W2036476977 hasConceptScore W2036476977C181199279 @default.
• W2036476977 hasConceptScore W2036476977C185592680 @default.
• W2036476977 hasConceptScore W2036476977C2775960820 @default.
• W2036476977 hasConceptScore W2036476977C2780378035 @default.
• W2036476977 hasConceptScore W2036476977C2780689927 @default.
• W2036476977 hasConceptScore W2036476977C2781307694 @default.
• W2036476977 hasConceptScore W2036476977C43369102 @default.
• W2036476977 hasConceptScore W2036476977C51639874 @default.
• W2036476977 hasConceptScore W2036476977C55493867 @default.
• W2036476977 hasConceptScore W2036476977C86803240 @default.
• W2036476977 hasIssue "19" @default.
• W2036476977 hasLocation W20364769771 @default.
• W2036476977 hasOpenAccess W2036476977 @default.
• W2036476977 hasPrimaryLocation W20364769771 @default.
• W2036476977 hasRelatedWork W1979741190 @default.
• W2036476977 hasRelatedWork W1994750899 @default.
• W2036476977 hasRelatedWork W1999920704 @default.
• W2036476977 hasRelatedWork W2014231481 @default.
• W2036476977 hasRelatedWork W2035731217 @default.
• W2036476977 hasRelatedWork W2049777033 @default.
• W2036476977 hasRelatedWork W2089842111 @default.
• W2036476977 hasRelatedWork W2121592873 @default.
• W2036476977 hasRelatedWork W2130733926 @default.
• W2036476977 hasRelatedWork W2439036626 @default.
• W2036476977 hasVolume "270" @default.
• W2036476977 isParatext "false" @default.
• W2036476977 isRetracted "false" @default.
• W2036476977 magId "2036476977" @default.

• next