The covalent tagging of the cell surface insulin receptor in intact cells with the generation of an insulin-free, functional receptor. A new approach to the study of receptor dynamics

A new method is described in which the cell surface insulin receptor can be radioactively tagged in a specific manner with a small insulin-free probe. After protecting the amino groups of insulin essential for binding and bio-activity, insulin is coupled to the heterobifunctional, cleavable cross-linking reagent SASD (sulfosuccinimidyl 2-(p-azidosalicylamido)-1,3'-dithiopropionate), via displacement of the N-hydroxysuccinimide moiety of SASD. Removal of the protecting groups results in the formation of 2-(p-azidosalicylamido)-1,3'-dithiopropionate (ASD)-insulin with insulin receptor binding activity equivalent to unmodified insulin. Iodination of ASD-insulin results in the incorporation of 125I into both the azidohydroxybenzoyl moiety of SASD and a tyrosine residue of insulin. Following binding of 125I-ASD-insulin to intact monolayers of 3T3-C2 cells, radiolabel is incorporated exclusively into a 135-kDa protein in a manner dependent upon the length of exposure of the cells to short wavelength ultraviolet light. This protein corresponds in molecular weight to the alpha subunit of the insulin receptor. Labeling of this protein can be inhibited by excess unlabeled insulin. Reduction of the disulfide bond of ASD with 10 mM glutathione causes the release of the 125I-insulin portion of the reagent from the receptor complex, with the iodinated photoactivated end of ASD covalently attached to the receptor. Insulin receptor labeled in this manner retains its ability to bind insulin. General metabolic processes of the intact cells do not appear to be perturbed by this labeling procedure, and the cellular processing of the insulin receptor does not appear to be modified by the covalent labeling of the receptor protein. This procedure therefore provides a way to specifically label the cell surface insulin receptor in a manner which does not perturb the normal functioning of the labeled cell and equally importantly, does not perturb the normal cellular processing of the insulin receptor itself.     

Identification of the Ca(2+)-independent endocytic hyaluronan receptor in rat liver sinusoidal endothelial cells using a photoaffinity cross-linking reagent.

The Ca(2+)-independent endocytic hyaluronan (HA) receptor in rat liver sinusoidal endothelial cells (LECs) was identified using a novel cross-linking derivative of HA. The heterobifunctional, photoactivatable, reducible reagent sulfosuccinimidyl 2-(p-azidosalicylamido)ethyl-1,3'-dithiopropionate (SASD) was coupled to the terminal amino group of uniquely modified HA-amine oligosaccharides (M(r) approximately 60,000) and subsequently iodinated. 125I-ASD-HA bound to cultured LECs with similar specificity and affinity as a previously characterized 125I-HA-amine/Bolton-Hunter adduct. Permeabilized LECs were incubated with 125I-ASD-HA with 10 mM EGTA and photolysed with UV light. Detergent extracts were reduced to release the HA oligosaccharides and radiolabeled proteins were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and autoradiography. Two polypeptides were consistently and equally labeled at M(r) = 175,000 and 166,000. Photoaffinity labeling of these two proteins was virtually identical in cultured LECs or membranes and was competed greater than 90% with a 100-fold excess of HA. As with the previously characterized bona fide LEC HA receptor, cross-linking was also competed by chondroitin sulfate and heparin, but less efficiently by chondroitin and not with galacturonan. We conclude that the Ca(2+)-independent LEC HA receptor is composed of at least two polypeptides of M(r) approximately 175,000 and 166,000 and may exist as a heterodimer of M(r) approximately 340,000. We also conclude that the LEC HA receptor is distinct from the CD44 family of HA-binding proteins.     

Molecular interactions between human lactotransferrin and the phytohemagglutinin-activated human lymphocyte lactotransferrin receptor lie in two loop-containing regions of the N-terminal domain I of human lactotransferrin.

Fluorescein isothiocyanate derivatization of the human lactotransferrin on Lys-264 inhibits the binding of the protein of human PHA-activated lymphocytes [Legrand, D., Mazurier, J., Maes, P., Rochard, E., Montreuil, J., & Spik, G. (1991) Biochem. J. 276, 733-738], indicating that part of the receptor-binding site is located in the N-terminal domain I of lactotransferrin. In the present study, a 6-kDa peptide (residues 4-52) was isolated from the N-terminal lobe of human lactotransferrin which inhibited the binding of the protein to its cell receptor. In addition, lactotransferrin was derivatized using sulfosuccinimidyl 2-(p-azidosalicylamido)ethyl-1,3'-dithiopropionate (SASD) and sulfosuccinimidyl 6-((4'-azido-2'-nitrophenyl)amino)hexanoate (sulfo-SANPAH), two heterobifunctional reagents generally used for receptor-ligand cross-linking. The azide group of these two reagents was inactivated by photolysis, and only the succinimidyl ester group was allowed to react with lysine residues of the protein. The binding of the derivatized lactotransferrins to the human lymphocyte receptor was assayed. SASD, which binds to Lys-74, was able to inhibit the binding of lactotransferrin to the cell receptor, in contrast to Lys-281-binding sulfo-SANPAH. Molecular modeling showed the position of SASD, sulfo-SANPAH, and fluorescein molecules at the surface of the protein and suggested that SASD and fluorescein could mask residues 4-6 and two loop-containing regions of human lactotransferrin (residues 28-34 and 38-45). The comparison of the primary and tertiary structures of human lactotransferrin and serotransferrin, which bind to specific cell receptors, shows that the above-mentioned regions, which are likely involved in protein-receptor interactions, possess specific structural features.     

Albumin binding proteins are highly expressed in actively proliferating fetal and adult tissues

The expression of albumin binding proteins (ABP, 31 and 18 kDa peptides) in various organs as a function of their ontogenic development was investigated in fetuses (20 days old), neonates (1 day old) and adult rabbits. At each of these stages, tissue extracts of brain, lung, thymus, heart, skeletal muscle and liver as well as whole embryos (11 days old) were examined by ligand blotting and quantitative immunoblot assays. Blots were either incubated with [125I]albumin followed by autoradiography and radioassay or exposed to a radioiodinated antibody raised against affinity-isolated 31 kDa peptide. Anti-31 kDa IgG cross-reacted with both 31 and 18 kDa peptides. Both methods used revealed that ABP are well expressed in embryos and in all fetal organs investigated. By comparison, in neonates, the ABP expression was diminished (by approximately 2-fold) in brain, heart and skeletal muscle. These changes were even more pronounced in the adult rabbit brain, heart, skeletal muscle and liver; no significant modification was detected in the lung. Prompted by these results, which inferred a high level of ABP in actively proliferating/differentiating tissues, we checked for the presence of ABP in other adult cells and tissues. In bone marrow cells, thymocytes and splenocytes, the 31 and 18 kDa peptides represented the major sodium dodecyl sulfate-urea extracted proteins, whereas in mature circulating white blood cells they were moderately expressed. The results indicate that ABP 1) are present early in embryogenesis, 2) are particularly well expressed in organs (fetal or adult) and cells characterized by active proliferation and differentiation, and 3) are not tissue specific.(ABSTRACT TRUNCATED AT 250 WORDS)     

Affinity isolation of albumin-binding proteins using nitrocellulose-bound albumin

Albumin immobilized on a nitrocellulose membrane was used as an affinity matrix to purify albumin-binding proteins (ABP) from extracts of lung, heart, thymus, and isolated microvascular endothelial cells. Albumin was immobilized onto nitrocellulose either (i) directly (physically adsorbed), (ii) cross-linked by treatment with 0.25% glutaraldehyde, or (iii) covalently coupled to the matrix using NaIO4 and Na-borohydride. The affinity support was incubated with a membrane-enriched fraction (obtained from tissue homogenates) in the presence of protease inhibitors; specific binding of ABP occurred within 30 min of incubation. The adsorbed proteins were eluted with 0.5% sodium dodecyl sulfate (SDS) and analyzed by SDS-polyacrylamide gel electrophoresis and ligand blotting. Analysis of electrophoretic mobility of eluted proteins showed that they consisted exclusively of the two sets of polypeptides of 31 000 Da and 18 000 Da previously identified as ABP (N. Ghinea et al., J. Cell Biol. 107, 231-239 (1988]. As demonstrated by ligand blotting, the ABP purified on nitrocellulose-bound albumin maintain the ability to interact specifically with albumin. Preliminary experiments showed that the method employed may be of a broader use for the isolation of receptor proteins from tissue extracts by incubating the latter with the cognate ligand immobilized on nitrocellulose membranes.     

Albumin-binding proteins of endothelial cells: immunocytochemical detection of the 18 kDa peptide.

Extracts of isolated microvascular endothelial cells (MEC) and cultured bovine aortic endothelial cells (BAEC) were subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), electrotransfer and incubation with albumin either radioiodinated or adsorbed to 5-nm gold particles. Both ligands reacted exclusively with two peptides of 18 and 31 kDa. To the 18 kDa peptide (excised from preparative SDS-PAGE), an antibody was raised in rabbits and purified by affinity on 18 kDa obtained from two-dimensional gel electrophoresis and immobilized on nitrocellulose paper. The specificity of the anti-18 kDa was assessed by immunoblotting and immunoprecipitation of endothelial cell extracts. To check whether the 18 kDa peptide is exposed on the endothelial cell surface and/or its components (uncoated pits, open plasmalemmal vesicles), the apical membrane of BAEC was radioiodinated, the solubilized proteins incubated with the anti-18 kDa, and the immune complexes formed were precipitated with protein A-Sepharose CL-4B. The ensuing SDS-PAGE and autoradiography revealed that from all radioiodinatable surface proteins, the 18 kDa was the only polypeptide immunoprecipitated by the anti-18 kDa antibody. To localize the 18 kDa peptide, we applied indirect immunofluorescence technique on cultured MEC and BAEC and immunoelectron microscopy (EM) on ultrathin cryosections of mouse heart. Nonpermeabilized whole MEC and BAEC incubated with anti-18 kDa followed by rhodamine-conjugated second antibody showed a relatively intense surface fluorescence often appearing as small dots. At the EM level, heart ultrathin cryosections exposed anti-18 kDa followed by gold-conjugated second antibody revealed that 18 kDa was primarily associated with the membrane of plasmalemmal vesicles of capillary endothelia.(ABSTRACT TRUNCATED AT 250 WORDS)     

Affinity cross-linking of atrial natriuretic factor to its receptor in bovine adrenal zona glomerulosa

125I-Labeled atrial natriuretic factor (ANF) was covalently cross-linked to its binding sites in bovine adrenal zona glomerulosa membranes using the hydrophilic cross-linker bis(sulfosuccinimidyl) suberate. Analysis by sodium dodecyl sulfate-polyacrylamide gel electrophoresis in the presence of 2-mercaptoethanol revealed that two protein bands with apparent Mr 68,000 and 114,000 were specifically labeled. The labeling of the two bands was prevented in a dose-dependent fashion by unlabeled ANF with a significant inhibition observed at 10(-10) M. High concentrations of angiotensin II and adrenocorticotropic hormone did not compete with 125I-ANF for binding and cross-linking. The dose-response curve for inhibition of covalent cross-linking of 125I-ANF by unlabeled ANF coincided with the dose-response curve for inhibition of binding to the receptor. No radioactive bands were observed in liver membranes. Experiments in which adrenal membranes were prepared and incubated in the presence of protease inhibitors showed no difference in the labeling pattern. Electrophoresis in the absence of reductant showed that the affinity-labeled species are not derived from larger disulfide-linked components. The apparent molecular weight of the two labeled species was not affected by a 100-fold variation in cross-linker concentration, and the labeling of both species increased in parallel. Possible relationships between the two labeled species are discussed.     

Interactions of bacterial lipopolysaccharide with microtubule proteins.

Bacterial LPS is a potent stimulator of immune cells, but its mechanisms are unknown. A possible role for microtubules in LPS actions has been indicated by previous findings that the microtubule-active agent, taxol, can mimic some effects of LPS in macrophages from normal strains of mice, but not from genetically LPS-hyporesponsive strains. In this report we demonstrate that isolated microtubules from mouse brain can bind LPS in vitro. LPS and tubulin coeluted through a gel filtration column, and LPS was cross-linked to microtubule proteins with an iodinatable, photoreactive agent, sulfosuccinimidyl 2-(p-azidosalicylamido) ethyl-1,3'-dithiopropionate. beta-Tubulin and microtubule-associated protein-2 (MAP), a predominant MAP in the brain, bound LPS specifically. Cross-linking was inhibited by an excess of unlabeled LPS or partially by unlabeled lipid A, but not by 2 M NaCl. Under the same conditions, neither myosin nor soybean trypsin inhibitor was labeled by the photoaffinity LPS probe, nor did these proteins compete for binding of LPS to beta-tubulin. These findings support the hypothesis that the microtubule network could be an intracellular target for LPS, and suggest further that a beta-tubulin-associated MAP could have an important role in LPS actions.     

Characterization of the subunit structure of gonadotropin receptor in luteinized rat ovary

Gonadotropin receptors with specificity, high affinity and low capacity for luteinizing hormone and human chorionic gonadotropin (hCG) have been identified in rat luteal cells. To investigate the nature of the receptor, we have employed disuccinimidyl suberate, a cross-linker noncleavable by reducing agents, and dithiobis(succinimidyl propionate), a cleavable cross-linker, to covalently cross-link the 125I-hCG . receptor complex. The molecular weight of 125I-hCG-linked receptor complex and the receptor subunit structure were determined by electrophoresis in either 10 or 4.5% acrylamide in the presence of 0.1% sodium dodecyl sulfate with or without reducing agents. Autoradiographic analysis of the 125I-hCG-linked receptor separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis under nonreducing condition revealed a single labeled band corresponding to Mr = 305,000 +/- 15,000. However, electrophoresis performed in the presence of 50 mM dithiothreitol and 2% beta-mercaptoethanol resulted in the appearance of four labeled bands corresponding to Mr = 105,000 +/- 4,000, 96,000 +/- 5,000, 74,000 +/- 4,000, and 62,000 +/- 4,000 concomitant with the loss of the labeled band in the Mr = 305,000 region. Further experiments demonstrated that these four labeled bands were derived from the same molecular species. In addition, the 125I-hCG-linked receptor in the absence of reducing agent was not dissociated into subunits even by treatment with strong denaturing agent (8 M urea). The appearance of the cross-linked 125I-hCG . receptor was effectively inhibited by the unlabeled beta-subunit of hCG, intact hCG, and luteinizing hormone and partially inhibited by the alpha-subunit of hCG but not by choleratoxin, gonadotropin-releasing hormone, insulin or bovine serum albumin. These data suggest that 1) the hCG/luteinizing hormone receptor is an oligomeric complex linked by disulfide bonds and 2) that under reducing conditions, the oligomeric receptor dissociates into four nonidentical subunits.     

The subunit structure of the follitropin (FSH) receptor. Photoaffinity labeling of the membrane-bound receptor follitropin complex in situ

Human follicle-stimulating hormone (hFSH) was acylated with N-hydroxysuccinimidyl-4-azidobenzoate (HSAB) and radioiodinated (55 microCi/micrograms) for use as a photoaffinity probe to investigate the subunit structure of the FSH receptor in calf testis. After incubation with the photoaffinity probe and photolysis with UV light, the cross-linked hormone-receptor complex was solubilized from the membrane and subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis in the presence and absence of the reducing agent dithiothreitol. Autoradiography of the polyacrylamide gels revealed two major bands, 64 kDa and 84 kDa. These were equivalent in molecular mass to those observed in a previous study (Branca, A. A., Sluss, P. M., Smith, A. A., and Reichert, L. E., Jr. (1985) J. Biol. Chem. 260, 9988-9993) in which performed hormone-receptor complexes were solubilized with detergent prior to formation of covalent cross-linkages through the use of homobifunctional cross-linking reagents. Reduction with dithiothreitol resulted in the loss of radioactivity from the 84-kDa band with a concomitant increase in the intensity of the 64-kDa band. Since dithiothreitol increases the dissociation of intact radioiodinated azidobenzoyl-FSH into subunits, it is suggested that the conversion of the 84-kDa band to the 64-kDa band by dithiothreitol is due to the loss of non-cross-linked hFSH subunit from the 84-kDa band and that the two bands observed after photoaffinity labeling arise from covalent bond formation between hFSH and a receptor subunit having a relative molecular weight (Mr) of 48,000. In addition to the predominant photolabeling of the receptor to yield the 64-kDa and 84-kDa bands, several other, less intense bands (54 kDa, 76 kDa, 97 kDa, and 116 kDa) were also consistently observed on autoradiographs. The appearance of all bands, however, was inhibited by the inclusion of unlabeled hFSH in the initial binding incubation mixtures. The results of this study indicate that the calf testis FSH receptor has a multimeric structure containing at least one 48-kDa subunit and suggest the presence of other nonidentical receptor subunit proteins.     

Identification and purification of a transverse tubule coupling protein which binds to the ryanodine receptor of terminal cisternae at the triad junction in skeletal muscle

In fast twitch skeletal muscle, the signal for excitation-contraction coupling is transferred from transverse tubule across the triad junction; calcium is thereby released from the terminal cisternae of sarcoplasmic reticulum triggering muscle contraction. Recently, the feet structures of terminal cisternae, which bridge the gap at the triad junction, have been identified as the ryanodine receptor and in turn with the calcium release channels of sarcoplasmic reticulum. The latter consists of an oligomer of a single high molecular weight polypeptide (Mr 360,000). This study attempts to identify the component in the transverse tubule which ligands with the foot structure to form the triad junction. The purified ryanodine receptor, derivatized with sulfosuccinimidyl-2-(p-azidosalicylimido)-1,3'-dithiopropionate (SASD), a thiol-cleavable, 125I-iodinatable, and photoactive probe, was shown to selectively cross-link to a protein with Mr of 71,000 in isolated transverse tubules. This coupling protein was purified from transverse tubule by solubilization with the detergent 3-[(3-cholamidopropyl)dimethylammonio]-1-propane sulfonate (CHAPS) and then purified by sequential column chromatography. In the absence of sulfhydryl agents, the purified polypeptide has an Mr of 61,000 on sodium dodecyl sulfate-polyacrylamide gel electrophoresis. A complementary approach using SASD was employed to confirm association of the coupling protein with the ryanodine receptor of terminal cisternae. We conclude that the transverse tubule coupling protein together with the ryanodine receptor (foot structure) is involved in the liganding between transverse tubule and terminal cisternae of sacroplasmic reticulum.     

Characterization of ovarian gonadotropin receptor. Monomer and associated form of the receptor

We have purified luteinizing hormone/human choriogonadotropin (hCG) receptor from rat ovary by sequential affinity column on wheat germ lectin-Sepharose and hCG-Sepharose chromatography. The purified receptor, previously identified as a single protein on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (Kusuda, S., and Dufau, M.L. (1986) J. Biol. Chem. 261, 16161-16168), was further characterized by radioiodination with 1,3,4,6-tetrachloro-3 alpha, 6 alpha-diphenylglycouril, and column chromatography on wheat germ lectin-Sepharose. Autoradiography of SDS-PAGE analysis under reducing conditions showed a single radiolabeled band of Mr = 80,000. The radioiodinated receptors treated with peptide:N-glycosidase F migrated at Mr = 54,000. Treatment with neuraminidase alone caused only a minor reduction in molecular weight, and subsequent treatment with endo-alpha-N-acetyl-D-galactosaminidase had little further effect on the receptor. When the radioiodinated receptor was analyzed by fast protein liquid chromatography, a single broad peak was eluted with Mr of approximately 350,000. The higher Mr of radioiodinated receptors than that of native receptors (Mr = 190,000 dimeric form) could be due to the aggregation of labeled molecules. These complexes dissociated into the monomeric form in the presence of SDS. To determine whether the monomers can bind hormone, the purified unlabeled receptors resolved with SDS were electroblotted to nitrocellulose membranes and incubated with 125I-hCG. Autoradiograms of the blots showed a band of monomer (Mr = 78,000) as well as one of dimer (Mr approximately 150,000). These studies have demonstrated that the luteinizing hormone/hCG receptors are predominantly N-linked glycosylated and suggest that the native receptor is a dimer of identical hormone binding subunits associated by noncovalent interactions. Although the individual subunits can bind hormone, it is conceivable that the dimeric form is necessary for signal transduction.     

Identification of albumin-binding proteins of thymocyte plasmalemma

In the present work we examined whether the interaction between albumin molecules and thymocytes involves albumin-binding proteins (ABP). Two plasmalemma-rich fractions obtained by differential centrifugation from rat thymus lymphocytes were characterized biochemically and morphologically. These fractions were examined by ligand-blotting and ligand affinity chromatography techniques. Plasmalemma proteins separated by SDS-PAGE were electrotransferred onto nitrocellulose membranes and incubated with¹²⁵I-albumin, in the presence or absence of excess native albumin. The autoradiogram revealed specific binding to two sets of polypeptides of 16–18 and 29–31 kDa, which could be blocked by native albumin. To elucidate whether albumin-binding proteins are exposed on the cell surface, intact lymphocytes were surface radioiodinated and membrane fractions prepared from them were subjected to affinity chromatography on albumin-agarose beads. The proteins thus purified had, like ABP, M_r of 16 and 31. These data indicate that ABP (i) are components of thymocyte plasma membrane, (ii) have apparent molecular mass of 16–18 and 29–31 kDa, and (iii) are exposed on the outer membrane surface.Abbreviations ABP albumin-binding proteins - Alb bovine serum albumin - Au gold - DAPI 4,6-diamidino-2-phenylindol - EM electron microscopy - NC nitrocellulose - PAGE polyacrylamidegel electrophoresis - PBS phosphate buffered saline - PEG polyethylene glycol - PMSF phenylmethylsulfonyl fluoride - WGA Wheat germ agglutinin     

Auxin-binding protein from coleoptile membranes of corn (Zea mays L.). I. Purification by immunological methods and characterization

The purification of a putative auxin receptor is one possibility to elucidate the first event in the mechanism of auxin action. By affinity chromatography of membrane proteins on 2-OH-3,5-diiodobenzoic acid-Sepharose and gel filtration on Ultrogel a fraction enriched in auxin-binding protein (ABP) was obtained and used for rabbit immunization. From the immunoglobulin G (IgG) fraction of the antisera IgGs against proteins not binding auxin (nonABP) could be obtained which were used to eliminate the nonABP from the eluates of the 2-OH-3,5-diiodobenzoic acid-Sepharose. The remainder fraction was further purified and concentrated on IgG-Sepharose which retained the ABP that could be eluted without loss of binding activity. A 600-fold purification with a yield of 42% was achieved. The ABP could be identified as the site I "receptor" described by Dohrmann et al. (Dohrmann, U., Hertel, R., and Kowalik, H. (1978) Planta (Berl.) 140, 97-106). It is shown that the competitors tested reduce [14C]1-naphthylacetic acid-(NAA) binding in the following order of effectiveness: NAA greater than 2-naphthylacetic acid greater than 1-phenylacetic acid greater than 2,3,5-triiodobenzoic acid greater than 3-indolylacetic acid greater than 2,4-dichlorophenoxyacetic acid. The ABP has a sharp binding optimum at pH 5.5, and the KD was calculated to be 5.7 X 10(-8) M to [14C]NAA. The binding activity of the ABP linearly decreased with increasing temperature but could partially be restored upon chilling in the presence of auxin. The ABP seems to be a 40-kDa dimer in its native form without disulfide bonds between its monomers.     

Covalent cross-linking of vasoactive intestinal polypeptide to its receptors on intact human lymphoblasts

125I-labeled vasoactive intestinal polypeptide (125I-VIP) was covalently cross-linked with its binding sites on intact cultured human lymphoblasts by each of three bifunctional reagents: disuccinimidyl suberate (DSS), ethylene glycol bis(succinimidyl succinate) (EGS), and N-succinimidyl 6-(4'-azido-2'-nitrophenylamino) hexanoate (SANAH). A fourth cross-linking agent with a shorter chain length, N-hydroxysuccinimidyl 4-azidobenzoate (HSAB), was much less effective in cross-linking 125I-VIP to the site. Analysis by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and autoradiography demonstrated a band of Mr approximately equal to 50,000 +/- 3,000, regardless of which cross-linker was used. The labeling of this band was specific in that it was prevented by 10(-6) M unlabeled VIP and was partially blocked by the homologous hormones secretin and glucagon. The relative potencies of these peptides in blocking the cross-linking of 125I-VIP to the Mr approximately equal to 50,000 band of the lymphoblasts (VIP greater than secretin greater than or equal to glucagon) were similar to those previously found for competitive inhibition of 125I-VIP binding to its putative high-affinity receptor on these cells. The covalent cross-linking required a bifunctional reagent; it was dependent on both the number of Molt cells and the concentration of 125I-VIP. The apparent molecular weight of the cross-linked species was unchanged by treatment with dithiothreitol. These observations suggest that the Mr = 50,000 species represents 125I-VIP cross-linked to a specific plasma membrane receptor and that the receptor does not contain interchain disulfide bonds.     

Identification of albumin-binding proteins in capillary endothelial cells

Isolated fat tissue microvessels and lung, whose capillary endothelia express in situ specific binding sites for albumin, were homogenized and subjected to SDS-gel electrophoresis and electroblotting. The nitrocellulose strips were incubated with either albumin-gold (Alb-Au) and directly visualized, or with [125I]albumin (monomeric or polymeric) and autoradiographed. The extracts of both microvascular endothelium and the lung express albumin-binding proteins (ABPs) represented by two pairs of polypeptides with major components of molecular mass 31 and 18 kD. The ABP peptides have pIs 8.05 to 8.75. Rabbit aortic endothelium, used as control, does not express detectable amounts of ABPs. The ABPs subjected to electrophoresis bind specifically and with high affinity (Kd = approximately 60 X 10(-9)M) both monomeric and polymeric albumin: the binding is saturable at approximately 80 nM concentration and 50% inhibition is reached at 5.5 micrograms/ml albumin concentration. Sulfhydryl-reducing agents beta-mercaptoethanol and dithiothreitol do not markedly affect the ABPs electrophoretic mobility and binding properties. As indicated by cell surface iodination of isolated capillary endothelium followed by electroblotting, autoradiography, and incubation with Alb-Au, the bands specifically stained by this ligand are also labeled with radioiodine.     

Fat cell beta 1-adrenergic receptor: structural evidence for existence of disulfide bridges essential for ligand binding

Agents that react chemically with sulfhydryl groups of proteins modify the response of adenylate cyclase to stimulation by beta-adrenergic agonists. N-Ethylmaleimide, an agent that alkylates sulfhydryl groups, inactivates both the catalytic moiety of adenylate cyclase and the stimulatory, regulatory guanine nucleotide binding protein Ns of rat fat cells but fails to affect binding of antagonists to the beta-adrenergic receptor [Malbon, C. C., Graziano, M. P., & Johnson, G. L. (1984) J. Biol. Chem. 259, 3254-3260]. Treating membranes of rat fat cells with dithiothreitol or beta-mercaptoethanol, agents that reduce disulfide bridges of proteins, results in a loss of binding of beta-adrenergic radioligands to the receptor. The specific binding of radioligands to beta-adrenergic receptors that are solubilized in digitonin is affected similarly by treatment with disulfide bridge reducing agents. beta-Adrenergic receptor purified from rat fat cells and treated with beta-mercaptoethanol (10%) and then subjected to gel electrophoresis in the presence of sodium dodecyl sulfate migrates as a Mr 67 000 peptide [Cubero, A., & Malbon, C. C. (1984) J. Biol. Chem. 259, 1344-1350]. In the absence of disulfide bridge reducing agents, however, the purified receptor exhibits greater electrophoretic mobility, migrating as a peptide with Mr 54 000. Treating the native form of the purified receptor with beta-mercaptoethanol (0.1-10%) or dithiothreitol (0.1-10 mM) decreases the ability of the receptor to bind beta-adrenergic ligands, decreases the electrophoretic mobility of the receptor, and results in receptor peptides migrating with molecular weight ranging from 54 000 to 67 000 when subjected to gel electrophoresis in the presence of sodium dodecyl sulfate.(ABSTRACT TRUNCATED AT 250 WORDS)     

Clearance, metabolic fate and tissue distribution of an injected bolus of photoaffinity-labeled rat androgen binding protein

An injected bolus of tritiated photolabeled rat androgen binding protein (ABP) is cleared from the circulation in a biphasic manner. The rapid component of the clearance curve indicates a clearance half time (t1/2) of 4.24 +/- 0.20 h and the slow component indicates a t1/2 of 13.27 +/- 2.92 h. After injection, photolabeled ABP is rapidly degraded as determined by electrophoresis on polyacrylamide gels containing sodium dodecyl sulfate. This method dissociates ABP into subunits of approximately 48,000 daltons. One-half hour after injection of photolabeled ABP, 90% of the radioactivity migrates as a 20,000- rather than as a 48,000-dalton moiety; at the 24-h point only 20,000- and 16,000-dalton species were detectable. Chromatography of plasma on Sephadex G-150 indicated that, under nondenaturing conditions, radioactivity was still associated with proteins having a molecular weight similar to the native ABP dimer. These results suggest that the dimer is substrate for the proteolytic processes affecting ABP. Small (congruent to 7000 dalton) fragments of ABP could be detected in urine by G-150 chromatography 2 h after injection. The control protein [3H] bovine serum albumin (BSA) was cleared from the blood with a time course identical to that of photolabeled ABP (t1/2's 4.69 +/- 0.09 h and 13.56 +/- 1.03 h). [3H] BSA was degraded to peptides that were similar in size to those formed from photolabeled ABP. The uptake of photolabeled ABP and [3H] BSA by potential target tissues (testis, epididymis and ventral prostate) was equivalent, suggesting that the uptake of ABP was a nonspecific phenomenon rather than a receptor-mediated event. The liver and kidneys contained more radioactivity that did the other tissues and there was a differential uptake and clearance of photolabeled ABP and [3H] BSA by these organs. Whether this difference is related to specific uptake processes or to differences in the size and/or structure of the proteins or other factors remains to be determined.     

Affinity labeling of the sialoglycopeptide antimitogen receptor

An 18-kDa 125I-sialoglycopeptide growth inhibitor was covalently cross-linked to its binding site on intact cultured Swiss 3T3 cells by three bifunctional cross-linkers with short (dimethyl adipimate), medium (disuccinimidyl suberate), and long (bis(2-succinimidooxycarbonyloxyethyl)sulfone) chain lengths. Analysis by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and autoradiography demonstrated a band of Mr approximately 168,000 regardless of which cross-linker was used. The labeling of this band was specific in that it was prevented by excess unlabeled inhibitor and the apparent molecular weight of the cross-linked receptor-ligand complex was unchanged by treatment with reducing agent. The efficiency of the cross-linking was increased by increasing pH, and the extent of covalent cross-linking was dependent on the concentration of the bifunctional reagent. Octyl glucoside and sodium dodecyl sulfate were effective in solubilizing the receptor while Triton X-100 did not extract the receptor from the plasma membrane. These observations suggest that the 168-kDa binding species represents the 125I-sialoglycopeptide cross-linked to a specific plasma membrane receptor and that the receptor does not appear to contain interchain disulfide bonds.     

Human microvascular endothelial cells express integrin-related complexes that mediate adhesion to the extracellular matrix

Microvascular endothelial cells (MEC) must use a set of surface receptors to adhere not only to the vascular basement membrane but, during angiogenic stimulation, to the interstitium. We examined how cultured MEC isolated from human foreskin interact with their subendothelial matrix. MEC were able to attach to diverse extracellular matrix proteins, including fibronectin (Fn), vitronectin (Vn), laminin (Ln), type I and IV collagen, as well as to fibrinogen and gelatin. Adhesion to Fn, but not to laminin or collagens, was specifically blocked in the presence of Arg-Gly-Asp (RGD)-containing peptides. When surface radioiodinated MEC were solubilized and subjected to affinity chromatography on Fn-Sepharose columns, two polypeptides of 150 and 125 kD, corresponding to the integrin heterodimer alpha 5 beta 1, were identified. MEC also express a complex of 150 (alpha) and 95 kD (beta 3) that is related to the Vn receptor. Immunofluorescent staining of MEC cultures with antibodies to the integrin beta 1 subunit demonstrated receptors on the basolateral surface at focal adhesion plaques that co-localized with vinculin and with Fn-positive matrix fibers. Occasionally, antibodies to the Vn receptor stained the vinculin-positive focal adhesion plaques that frequently co-localized with the beta 1 complex. However, in cultures of MEC that were attached to substrates coated with alternating strips of Fn and Vn, the beta 1 complex was preferentially localized to the Fn substrate, while the Vn receptor was concentrated on the Vn substrate. The results indicate that MEC express at least two different heterodimer adhesion receptors that belong to the integrin super-family and appear to have distinct ligand specificities: the Fn receptor and the Vn receptor. These receptors mediate cell adhesion to the extracellular matrix and presumably have an important role in hemostasis and neovascularization.