Contribution to ligand binding by multiple carbohydrate-recognition domains in the macrophage mannose receptor

The extracellular portion of the macrophage mannose receptor is composed of several cysteine-rich domains, including a fibronectin type II repeat and eight segments related in sequence to Ca(2+)-dependent carbohydrate-recognition domains (CRDs) of animal lectins. Expression of portions of the receptor in vitro, in fibroblasts and in bacteria, has been used to determine which of the extracellular domains are involved in binding and endocytosis of ligand. The NH2-terminal cysteine-rich domain and the fibronectin type II repeat are not necessary for endocytosis of mannose-terminated glycoproteins. CRDs 1-3 have at most very weak affinity for carbohydrate, so the carbohydrate binding activity of the receptor resides in CRDs 4-8. CRD 4 shows the highest affinity binding and has multispecificity for a variety of monosaccharides. However, CRD 4 alone cannot account for the binding of the receptor to glycoproteins. At least 3 CRDs (4, 5, and 7) are required for high affinity binding and endocytosis of multivalent glycoconjugates. In this respect, the mannose receptor is like other carbohydrate-binding proteins, in which several CRDs, each with weak affinity for single sugars, are clustered to achieve high affinity binding to oligosaccharides. In the mannose receptor, these multiple weak interactions are achieved through several active CRDs in a single polypeptide chain rather than by oligomerization of polypeptides each containing a single CRD.

Structure of the calcium-dependent lectin domain from a rat mannose-binding protein determined by MAD phasing

Calcium-dependent (C-type) animal lectins participate in many cell surface recognition events mediated by protein-carbohydrate interactions. The C-type lectin family includes cell adhesion molecules, endocytic receptors, and extracellular matrix proteins. Mammalian mannose-binding proteins are C-type lectins that function in antibody-independent host defense against pathogens. The crystal structure of the carbohydrate-recognition domain of a rat mannose-binding protein, determined as the holmium-substituted complex by multiwavelength anomalous dispersion (MAD) phasing, reveals an unusual fold consisting of two distinct regions, one of which contains extensive nonregular secondary structure stabilized by two holmium ions. The structure explains the conservation of 32 residues in all C-type carbohydrate-recognition domains, suggesting that the fold seen here is common to these domains. The strong anomalous scattering observed at the Ho LIII edge demonstrates that traditional heavy atom complexes will be generally amenable to the MAD phasing method.

Determination of the minimum carbohydrate-recognition domain in two C-type animal lectins

Comparison of the primary structures of numerous Ca(2+)-dependent animal lectins reveals the presence of a common sequence motif which has been suggested to form the carbohydrate-recognition domain in these proteins. The extent of the functional carbohydrate-recognition domains in two rat C-type lectins, mannose-binding protein A and the major subunit of the asialoglycoprotein receptor (rat hepatic lectin 1), has been defined by expressing truncated fragments of the proteins in an in vitro transcription and translation system. The shortest fully functional fragments constitute the COOH-terminal 120 amino acids of mannose-binding protein A and 135 amino acids of rat hepatic lectin 1. These segments correspond closely to protease-resistant protein cores which can be isolated from the native lectins. The NH2-terminal boundary of each minimum carbohydrate-recognition domain falls near the site of an intron in the corresponding gene.

Physical characterization and crystallization of the carbohydrate-recognition domain of a mannose-binding protein from rat

A portion of rat mannose-binding protein A (MBP-A), a Ca(2+)-dependent animal lectin, has been overproduced in a bacterial expression system, biochemically characterized, and crystallized. A fragment corresponding to the COOH-terminal 115 residues of native MBP-A, produced by subtilisin digestion of the bacterially expressed protein, contains the carbohydrate-recognition domain (CRD). Gel filtration, chemical cross-linking, and crystallographic self-rotation function analyses indicate that the subtilisin fragment is a dimer, although the complete bacterially expressed fragment, containing the neck and CRD of MBP-A, is a trimer. Crystals of the minimal CRD, obtained only as a complex with a Man6GlcNAc2Asn glycopeptide, diffract to Bragg spacings of at least 1.7 A. Several trivalent lanthanide ions (Ln3+) can substitute for Ca2+, as assessed by their ability to support carbohydrate binding and to protect the CRD from proteolysis in a manner similar to that observed for Ca2+. These assays indicate that Ln2+ binds about 30 times more tightly than Ca2+ to the CRD, and that two Ca2+ or Ln3+ bind to each monomer, a result confirmed by determination of the Ho3+ positions in a Ho(3+)-containing crystal of the CRD. Crystals grown in the presence of Ln3+ belong to different space groups from those obtained with Ca2+ and are therefore not useable for traditional crystallographic phase determination methods, but are well-suited for high resolution structure determination by multiwavelength anomalous dispersion phasing.

Evolutionary conservation of intron position in a subfamily of genes encoding carbohydrate-recognition domains

The structure of the gene encoding a chicken liver receptor, the chicken hepatic lectin, which mediates endocytosis of glycoproteins has been established. The coding sequence is divided into six exons separated by five introns. The first three exons correspond to separate functional domains of the receptor polypeptide (cytoplasmic tail, transmembrane sequence, and extracellular neck region), while the final three exons encode the Ca(2+)-dependent carbohydrate-recognition domain. These results, as well as computer-assisted multiple sequence comparisons, establish this receptor as the evolutionary homolog of the mammalian asialoglycoprotein receptors. It is interesting that the chicken receptor falls into a subfamily of proteins along with the mammalian asialoglycoprotein receptors, since the saccharide-binding specificity of the chicken receptor resembles more closely that of a different set of calcium-dependent animal lectins, which includes the mannose-binding proteins. The portions of the genes encoding the carbohydrate-recognition domains of these proteins lack introns. The results suggest that divergence of intron-containing and intron-lacking carbohydrate-recognition domains preceded shuffling events in which other functional domains were associated with the carbohydrate-recognition domains. This was followed by further divergence, generating a variety of saccharide-binding specificities.

Biosynthesis of human fibroblast growth factor-5

We have analyzed the biosynthesis of human fibroblast growth factor-5 (FGF-5) at the translational and posttranslational levels. FGF-5 RNA synthesized in vitro can be translated in rabbit reticulocyte lysates to yield a 29,500-Da protein, which is consistent with the molecular weight predicted from the coding sequence. The efficiency of FGF-5 translation is dramatically enhanced if an upstream open reading frame (ORF-1) in the RNA is deleted or if both AUG codons in ORF-1 are destroyed by point mutations, while partial enhancement is achieved by individual mutation of either ORF-1 AUG codon. These data suggest that FGF-5 synthesis requires the scanning of ribosomes past the two ORF-1 AUG codons. The introduction of these ORF-1 mutations into a eukaryotic FGF-5 expression vector increases its capacity to transform mouse NIH 3T3 cells up to 50-fold upon transfection. FGF-5 is secreted from transfected 3T3 cells and from human tumor cells as glycoproteins containing heterogeneous amounts of sialic acid. Glycosidase treatments suggest that the growth factor bears both N-linked and O-linked sugars.

Ligand-binding characteristics of rat serum-type mannose-binding protein (MBP-A). Homology of binding site architecture with mammalian and chicken hepatic lectins

Sugar-binding characteristics of rat serum mannose-binding protein (MBP) were studied using the carbohydrate-recognition domain of this protein expressed from a cloned cDNA. To assess the binding affinity of various test compounds, they were added as inhibitors in a binding assay in which 125I-MBP was incubated with yeast cells and the extent of binding was estimated from the radioactivity associated with the pelleted cells. The results of such inhibition assays suggest that MBP has a small binding site which is probably of the trough-type. The 3- and 4-OH of the target sugar are indispensable, while the 6-OH is not required. These characteristics are shared by the rat hepatic lectin and chicken hepatic lectin, both of which are C-type lectins containing carbohydrate-recognition domains highly homologous to that of MBP. Apparently, the related primary structures of these lectins give rise to similar gross architecture of their binding sites, despite the fact that each exhibits different sugar binding specificities.

Carbohydrate-recognition domains as tools for rapid purification of recombinant eukaryotic proteins

Methods have been developed for expression and purification of eukaryotic proteins by creating fusions with the carbohydrate-recognition domain (CRD) of the galactose-specific rat hepatic lectin. In order to generate the fusion proteins, vectors have been constructed so that cDNAs for passenger proteins can be inserted in any reading frame following a segment of DNA encoding the CRD. The feasibility of using this approach as an aid to protein purification has been demonstrated using human placental alkaline phosphatase. Following expression in either of two different eukaryotic expression systems, the CRD-phosphatase fusion protein can be isolated by one step of affinity chromatography on galactose-Sepharose under mild, non-denaturing conditions. Incorporation of a proteinase-sensitive linker allows cleavage of the CRD from the passenger protein. Immobilised proteinase could be rapidly separated from the cleavage products and the released, active phosphatase was purified away from the CRD by re-chromatography on galactose-Sepharose. These methods provide a means of isolating correctly folded recombinant eukaryotic proteins when cDNAs are available, but the properties of the encoded proteins are unknown.

Differential recognition of core and terminal portions of oligosaccharide ligands by carbohydrate-recognition domains of two mannose-binding proteins

Two different mannose-binding proteins (MBP-A and MBP-C), which show 56% sequence identity, are present in rat serum and liver. It has previously been shown that MBP-A binds to a range of monosaccharide-bovine serum albumin conjugates, and that, among oligosaccharide ligands tested, preferential binding is to terminal nonreducing N-acetylglucosamine residues of complex type N-linked oligosaccharides. In order to compare the binding specificity of MBP-C, an expression system has been developed for production of a fragment of this protein which contains the COOH-terminal carbohydrate-recognition domain. After radioiodination, the domain has been used to probe natural glycoproteins, neoglycoproteins, and neoglycolipids. Like MBP-A, MBP-C binds several different monosaccharides conjugated to bovine serum albumin, including mannose, fucose, and N-acetylglucosamine, although binding to the last of these is relatively weaker than observed for MBP-A. The results of binding to natural glycoproteins and to neoglycolipids containing oligosaccharides derived from these proteins are most compatible with the interpretation that MBP-C interacts primarily with the trimannosyl core of complex N-linked oligosaccharides, with additional ligands being terminal fucose and perhaps also peripheral mannose residues of high mannose type oligosaccharides. This binding specificity is thus quite distinct from that of MBP-A. The presence of multiple MBPs with distinct binding specificities in preparations derived from serum and liver explains conflicting conclusions which have been reached about carbohydrate recognition by these proteins.

Primary structure of the mannose receptor contains multiple motifs resembling carbohydrate-recognition domains

Macrophages express a cell surface receptor which mediates phagocytosis and pinocytosis of particles and solutes containing mannose (fucose and N-acetylglucosamine are also ligands for the receptor). An apparently identical protein has been isolated from human placenta. Proteolytic fragments of the placental receptor were sequenced so that oligonucleotide probes complementary to the receptor cDNA could be generated. These probes were used to isolate cDNA clones covering the entire coding portion of the mRNA for the receptor. Confirmation that these clones encode the mannose receptor was obtained by expression in rat fibroblasts. The expressed protein mediates uptake and degradation of mannose-conjugated serum albumin. The deduced amino acid sequence of the receptor reveals that it is most likely to be a type I transmembrane protein (COOH terminus on the cytoplasmic side of the membrane) since the mature polypeptide is preceded by a signal sequence and a hydrophobic stop transfer sequence is located 45 amino acids from the COOH terminus. The extracellular portion of the receptor polypeptide consists of three types of domains. The first 139 amino acids constitute a cysteine-rich segment which does not resemble other known sequences. There follows a domain which closely resembles fibronectin type II repeats. The remainder of the extracellular portion of the receptor is composed of eight segments homologous with the C-type carbohydrate-recognition domains of the asialoglycoprotein receptor, mannose binding proteins, and other Ca2(+)-dependent animal lectins. This structure suggests that the receptor may contain multiple ligand-binding domains thus accounting for its tight binding to highly multivalent ligands.

Endocytosis via coated pits mediated by glycoprotein receptor in which the cytoplasmic tail is replaced by unrelated sequences

Rat 6 fibroblast cell lines expressing wild-type chicken liver glycoprotein receptor (CHL) or chimeric receptors with alternate cytoplasmic tails were produced to study the role of the cytoplasmic tail in mediating receptor localization in coated pits and endocytosis of ligand. Cells expressing CHL or cells expressing a hybrid receptor that contains the cytoplasmic tail of the asialoglycoprotein receptor display high-efficiency endocytosis of N-acetylglucosamine-conjugated bovine serum albumin in experiments designed to measure an initial internalization step, as well as in studies of continuous uptake and degradation. Substitution of the cytoplasmic tail by the equivalent domain of rat Na,K-ATPase beta subunit or by a stretch of Xenopus laevis globin beta chain does not abolish endocytosis but decreases the endocytosis rate constant from 15%-16%/min to 2.4% and 6.5%/min, respectively. Electron microscopy was used to visualize the glycoprotein binding sites at the surface of Rat 6 cells transfected with the various receptors. The percentage of receptors found in coated areas ranged from 32% for CHL to 9% for the Na,K-ATPase hybrid, indicating that clustering in coated pits correlates with efficiency of endocytosis. We concluded that replacement of the CHL cytoplasmic tail with unrelated sequences does not prevent, but decreases to varying extents, coated-pit localization and endocytosis efficiency. The construct with NH2-terminal globin tail lacks a signal for high-efficiency localization in coated pits but nevertheless is directed to the pits by an alternative mechanism.

Polarized expression of functional rat liver asialoglycoprotein receptor in transfected Madin-Darby canine kidney cells

The rat liver asialoglycoprotein receptor or rat hepatic lectin (RHL) consists of two polypeptide species, a major one designated RHL-1 and a minor one designated RHL-2/3, which exists in two differentially glycosylated forms. We have studied the biosynthesis, targeting, and function of the different forms after transfection of their cDNAs into the polarized Madin-Darby canine kidney cell line. In cells expressing only RHL-1, newly synthesized protein undergoes rapid intracellular degradation and is not detected at the cell surface. In contrast, RHL-2/3 when transfected alone is much more stable and is expressed at the basolateral surface of fiber-grown cells. When both forms are expressed together, newly synthesized RHL-1 escapes rapid degradation and is detected at the basolateral surface. In double transfectants a functional receptor is formed that specifically endocytoses and degrades ligand at the basolateral side.

Polarized endocytosis by Madin-Darby canine kidney cells transfected with functional chicken liver glycoprotein receptor

We have studied the expression of the chicken hepatic glycoprotein receptor (chicken hepatic lectin [CHL]) in Madin-Darby canine kidney (MDCK) cells, by transfection of its cDNA under the control of a retroviral promotor. Transfected cell lines stably express 87,000 surface receptors/cell with a kd = 13 nM. In confluent monolayers, approximately 40% of CHL is localized at the plasma membrane. 98% of the surface CHL is expressed at the basolateral surface where it performs polarized endocytosis and degradation of glycoproteins carrying terminal N-acetylglucosamine at a rate of 50,000 ligand molecules/h. Studies of the half-life of metabolically labeled receptor and of the stability of biotinylated cell surface receptor after internalization indicate that transfected CHL performs several rounds of uptake and recycling before it gets degraded. The successful expression of a functional basolateral receptor in MDCK cells opens the way for the characterization of the mechanisms that control targeting and recycling of proteins to the basolateral membrane of epithelial cells.

Neoglycolipids as probes of oligosaccharide recognition by recombinant and natural mannose-binding proteins of the rat and man

Oligosaccharide recognition by three mammalian mannose-binding proteins was investigated by using as probes a series of structurally characterized neoglycolipids in t.l.c. binding assays. The neoglycolipids were derived from N-linked oligosaccharides of complex, high-mannose and hybrid types and from human milk oligosaccharides and simple di- and tri-saccharides. The three proteins, namely the recombinant carbohydrate-recognition domain of rat mannose-binding Protein A and the multi-subunit forms of rat and human serum mannose-binding proteins, were shown to have in common reactivity with oligosaccharide probes containing one or more non-reducing terminal N-acetylglucosamine residue(s). Substitution with galactose masks reactivity. The three proteins also bound to non-reducing terminal mannose residues in high-mannose-type oligosaccharides, non-reducing terminal fucose residues in the sequence Fuc alpha 1-4(Gal beta 1-3)GlcNAc and non-reducing terminal glucose residues in dextran oligomers; the recombinant binding domain gave consistently weaker binding. The relative reactivities with the various probes differ for each protein. Overall, the reaction patterns of the three mammalian proteins differ from that of the plant lectin concanavalin A, which showed preferential binding to the high-mannose type, weak binding to biantennary complex type and no binding to the fuco-oligosaccharide and simple oligosaccharide probes. As a group, the three mammalian proteins resemble bovine serum conglutinin and behave as lectins with rather broad sugar specificities directed at certain non-reducing terminal N-acetylglucosamine, mannose, glucose and fucose residues, but with subtle differences in fine specificities. These results illustrate the potential of neoglycolipids in studies of oligosaccharide recognition by natural and recombinant proteins of diverse biological systems.

The occurrence of disulphide bonds in purified clathrin light chains

Three forms of clathrin light chain contain two cysteine residues. These are the predominant brain-specific forms of LCa and LCb and the non-brain form of LCb. After purification in the absence of thiols they contain intramolecular disulphide bonds. The reduced and the oxidized forms show differences in electrophoretic mobility, explaining the variable and heterogeneous patterns observed on electrophoresis. Accessibility of the thiol groups in the free light chains is greater than when they are associated with the heavy chain. In contrast the cysteine residues of the clathrin heavy chain are completely inaccessible in the absence of denaturants and are not found in disulphide bonds. The antigenic properties of the oxidized and the reduced forms of the clathrin light chains are similar, as is their capacity to bind to the clathrin heavy chain. After isolation in the presence of 10 mM-iodoacetamide, the light-chain cysteine residues are fully alkylated. The results are consistent with the reduced form being the native state and the light-chain disulphide bonds an artifact of isolation.

Conformational changes in the chicken receptor for endocytosis of glycoproteins. Modulation of ligand-binding activity by Ca2+ and pH

Limited proteolysis, gel filtration, and circular dichroism have been used to identify at least three distinct conformational states of a proteolytic fragment containing the ligand-binding domain of the chicken receptor for endocytosis of glycoproteins. Differences in the ligand-binding activity of intact receptor brought about by changing Ca2+ concentrations and pH values can be correlated with different physical states of the binding domain present under similar conditions. An active, ligand-binding state can be detected at either pH 7.8 or 5.4, but 10-fold higher concentrations of Ca2+ are required to stabilize this state at the lower pH. In all cases, the dependence on Ca2+ concentration is second-order, suggesting that two Ca2+ ions are bound to each domain. These studies demonstrate an interdependence between the effects of Ca2+ concentration and pH on both ligand-binding activity and receptor conformation, which is important to consider when describing the binding and dissociation of ligand during endocytosis.