A Monoclonal Antibody which Recognizes a Glycosaminoglycan Epitope in Both Dermatan Sulfate and Chondroitin Sulfate Proteoglycans of Human Skin

Studies have been initiated to identify various cell surface and matrix components of normal human skin through the production and characterization of murine monoclonal antibodies. One such antibody, termed PG-4, identifies both cell surface and matrix antigens in extracts of human foetal and adult skin as the dermatan sulfate proteoglycans, decorin and biglycan, and the chondroitin sulfate proteoglycan versican. Treatment of proteoglycans with chondroitinases completely abolishes immunoreactivity for all of these antigens which suggests that the epitope resides within their glycosaminoglycan chains. Further evidence for the carbohydrate nature of the epitope derives from competition studies where protein-free chondroitin sulfate chains from shark cartilage react strongly; however, chondroitin sulfate chains from bovine tracheal cartilage fail to exhibit a significant reactivity, an indication that the epitope, although present in some chondroitin sulfate chains, does not consist of random chondroitin 4- or 6-sulfate disaccharides. The presence of the epitope on dermatan sulfate chains and on decorin was also demonstrated using competition assays. Thus, PG-4 belongs to a class of antibodies that recognize native epitopes located within glycosaminoglycan chains. It differs from previously described antibodies in this class in that it identifies both chondroitin sulfate and dermatan sulfate proteoglycans. These characteristics make PG-4 a useful monoclonal antibody probe to identify the total population of proteoglycans in human skin.     

Reconstruction of glycosaminoglycan chains in decorin

The glycosaminoglycan chain of decorin from human spinal ligaments was digested using the hydrolysis of bovine testicular hyaluronidase. As a result, decorin with hexasaccharide, octasaccharide, and decasaccharide including the linkage region, GlcA-Gal-Gal-Xyl, was obtained. The obtained decorin as an acceptor and hyaluronic acid as a donor were incubated with bovine testicular hyaluronidase under the condition of transglycosylation reaction. The transglycosylation reaction product had hexasaccharide to triacontasaccharide. Judging from the analysis of glycosaminoglycan chain in the transglycosylation reaction product, it was confirmed that hyaluronic acid chain as a donor was transferred to the retained glycosaminoglycan chain of decorin as an acceptor. Similarly, it was possible to reconstruct the glycosaminoglycan chain in decorin to chondroitin, chondroitin 4-sulfate or chondroitin 6-sulfate. Therefore, we succeeded in synthesizing an artificial family of decorins.     

A human antibody recognizing a conserved epitope of H5 hemagglutinin broadly neutralizes highly pathogenic avian influenza H5N1 viruses

Influenza A virus infection is a persistent threat to public health worldwide due to its ability to evade immune surveillance through rapid genetic drift and shift. Current vaccines against influenza A virus provide immunity to viral isolates that are similar to vaccine strains. High-affinity neutralizing antibodies against conserved epitopes could provide immunity to diverse influenza virus strains and protection against future pandemic viruses. In this study, by using a highly sensitive H5N1 pseudotype-based neutralization assay to screen human monoclonal antibodies produced by memory B cells from an H5N1-infected individual and molecular cloning techniques, we developed three fully human monoclonal antibodies. Among them, antibody 65C6 exhibited potent neutralization activity against all H5 clades and subclades except for subclade 7.2 and prophylactic and therapeutic efficacy against highly pathogenic avian influenza H5N1 viruses in mice. Studies on hemagglutinin (HA)-antibody complexes by electron microscopy and epitope mapping indicate that antibody 65C6 binds to a conformational epitope comprising amino acid residues at positions 118, 121, 161, 164, and 167 (according to mature H5 numbering) on the tip of the membrane-distal globular domain of HA. Thus, we conclude that antibody 65C6 recognizes a neutralization epitope in the globular head of HA that is conserved among almost all divergent H5N1 influenza stains.     

Identification of the region including the epitope for a monoclonal antibody which can neutralize human parvovirus B19.

In this study, we identified a region in the human parvovirus structural protein which involves the neutralization of the virus by a monoclonal antibody and site-specific synthetic peptides. A newly established monoclonal antibody reacted with both viral capsid proteins VP1 and VP2. The epitope was found in six strains of independently isolated human parvovirus B19. The monoclonal antibody could protect colony-forming unit erythroid in human bone marrow cell culture from injury by the virus. The monoclonal antibody reacted with only 1 of 12 peptides that were synthesized according to a predicted amino acid sequence based on nucleotide sequences of the coding region for the structural protein of B19 virus. The sequence recognized by the antibody was a site corresponding to amino acids 328 to 344 from the amino-terminal portion of VP2. This evidence suggests that the epitope of the viral capsid protein is located on the surface of the virus and may be recognized by virus-neutralizing antibodies.     

Intracellular avian type X collagen in situ and determination of its thermal stability using a conformation-dependent monoclonal antibody

The thermal stability of the helical domain of intracellular and matrix-associated type X collagen was examined in situ within the hypertrophic region of embryonic chick vertebral cartilages. For this we employed indirect immunofluorescence histochemistry of unfixed tissue sections reacted at progressively higher temperatures (Linsenmayer et al., J cell biol 99 (1984) 1405) with a conformation-dependent monoclonal antibody (X-AC9) (Schmid & Linsenmayer, J cell biol 100 (1985) 598). The hypertrophic chondrocytes which had most recently initiated synthesis of type X did not immediately secrete it, but instead retained it intracellularly within cytoplasmic organelles. This allowed for clear visualization of the intracellular type X. Within the pool of intracellular type X collagen, the epitope recognized by the antibody was stable up to 55 degrees C, but was destroyed at 60 degrees C. This is 5-10 degrees C higher than the thermal stability of the epitope when the molecule is in neutral solution (as determined by competition ELISA). The matrix-associated type X collagen is stable at least to 65-67.5 degrees C. We conclude that in situ the stability of the collagen helix in its normal intracellular environment is considerably greater than might be predicted from measurements made on molecules in solution.     

Mast Cells Produce a Unique Chondroitin Sulfate Epitope

The granules of mast cells contain a myriad of mediators that are stored and protected by the sulfated glycosaminoglycan (GAG) chains that decorate proteoglycans. Whereas heparin is the GAG predominantly associated with mast cells, mast cell proteoglycans are also decorated with heparan sulfate and chondroitin sulfate (CS). This study investigated a unique CS structure produced by mast cells that was detected with the antibody clone 2B6 in the absence of chondroitinase ABC digestion. Mast cells in rodent tissue sections were characterized using toluidine blue, Leder stain and the presence of mast cell tryptase. The novel CS epitope was identified in rodent tissue sections and localized to cells that were morphologically similar to cells chemically identified as mast cells. The rodent mast cell-like line RBL-2H3 was also shown to express the novel CS epitope. This epitope co-localized with multiple CS proteoglycans in both rodent tissue and RBL-2H3 cultured cells. These findings suggest that the novel CS epitope that decorates mast cell proteoglycans may play a role in the way these chains are structured in mast cells.     

Identification and characterization of a novel interaction between pulmonary surfactant protein D and decorin

Surfactant-associated protein D (SP-D) is a collectin that is present in lung surfactant and mucosal surfaces. Although SP-D regulates diverse functions, only a few proteins are known to bind to this collectin. Here we describe the co-purification of decorin, a novel SP-D-binding protein, from amniotic fluid. The human decorin that co-purified with SP-D is a 130-150-kDa proteoglycan, which has a 46-kDa protein core and approximately 90-kDa dermatan sulfate chain. Both native and recombinant decorin can bind to SP-D that is already bound to maltose-agarose matrix, and these SP-D-decorin complexes are dissociated at high salt (0.5-1.0 m NaCl) conditions, releasing the decorin. We further show that SP-D and decorin interact with each other (kd = 4 nm) by two mechanisms. First, the direct binding and competition experiments show that the carbohydrate recognition domain (CRD) of SP-D binds in a calcium dependent-manner to the sulfated N-acetyl galactosamine moiety of the glycosaminoglycan chain. Second, complement component C1q, a complement protein that is known to interact with decorin core protein via its collagen-like region, partially blocks the interaction between decorin and native SP-D. This protein, however, does not block the interaction between decorin and SP-D(n/CRD), a recombinant fragment that lacks the N-terminal and collagen-like regions. Furthermore, the core protein, obtained by chondroitin ABC lyase treatment of decorin, binds SP-D, but not SP-D(n/CRD). These findings suggest that decorin core protein binds the collagen-like region of the SP-D. Concentrations of decorin and SP-D are negatively correlated to each other, in amniotic fluid, implying a functional relevance for SP-D-decorin interaction, in vivo. Collectively, our results show that carbohydrate recognition domains of SP-D interact with the dermatan sulfate moiety of decorin via lectin activity and that the core protein of decorin binds the collagen-like region of SP-D in vitro, and these interactions may be operative in vivo.     

Cloning and sequence analysis of cDNAs encoding the heavy and light chain variable regions of an Ab2beta anti-idiotypic monoclonal antibody possessing an internal image of cocaine.

We report here the cloning and sequence analysis of cDNAs encoding the variable regions of an Ab2beta anti-idiotypic monoclonal antibody (K1-4c, gamma1kappa) that mimics the configuration of cocaine. The Ab2beta specifically binds to the human dopamine transporter as shown by confocal immunofluorescence microscopy. The sequence of the heavy chain complementarity-determining region 3 of K1-4c is strikingly similar to that of a monoclonal antibody (F11.2.32) specific for HIV-1 protease. Three or four amino acids in the epitope recognized by the anti-HIV-1 protease antibody are also present in the third extracellular loop of the dopamine transporter. This epitope is within the conserved region of the known transporters for dopamine, norepinephrine and serotonin in Homo sapiens, Rattus norvegicus, Caenorhabditis elegans and Drosophila melanogaster.     

Virus-neutralizing monoclonal antibody to a conserved epitope on the duck hepatitis B virus pre-S protein.

In this study we used duck hepatitis B virus (DHBV)-infected Pekin ducks and heron hepatitis B virus (HHBV)-infected heron tissue to search for epitopes responsible for virus neutralization on pre-S proteins. Monoclonal antibodies were produced by immunizing mice with purified DHBV particles. Of 10 anti-DHBV specific hybridomas obtained, 1 was selected for this study. This monoclonal antibody recognized in both DHBV-infected livers and viremic sera a major (36-kilodalton) protein and several minor pre-S proteins in all seven virus strains used. In contrast, pre-S proteins of HHBV-infected tissue or viremic sera did not react. Thus, the monoclonal antibody recognizes a highly conserved DHBV pre-S epitope. For mapping of the epitope, polypeptides from different regions of the DHBV pre-S/S gene were expressed in Escherichia coli and used as the substrate for immunoblotting. The epitope was delimited to a sequence of approximately 23 amino acids within the pre-S region, which is highly conserved in four cloned DHBV isolates and coincides with the main antigenic domain as predicted by computer algorithms. In in vitro neutralization assays performed with primary duck hepatocyte cultures, the antibody reduced DHBV infectivity by approximately 75%. These data demonstrate a conserved epitope of the DHBV pre-S protein which is located on the surface of the viral envelope and is recognized by virus-neutralizing antibodies.     

Paratope and epitope mapping of the antithrombotic antibody 6B4 in complex with platelet glycoprotein Ibalpha

The monoclonal antibody 6B4 has a potent antithrombotic effect in nonhuman primates by binding to the flexible loop, also known as the beta-switch region (amino acids 230-242), of glycoprotein Ibalpha (GPIbalpha). This interaction blocks, in high shear stress conditions, the specific interaction between GPIbalpha and von Willebrand factor suppressing platelet deposition to the damaged vessel wall, a key event in the pathogenesis of arterial thrombosis. To understand the interactions between this antibody and its antigen at the amino acid level, we here report the identification of the paratope and epitope in 6B4 and GPIbalpha, respectively, by using computer modeling and site-directed mutagenesis. The docking programs ZDOCK (rigid body docking) and HADDOCK (flexible docking) were used to model the interaction of 6B4 with GPIbalpha and to delineate the respective paratope and epitope. 6B4 and GPIbalpha mutants were constructed and assayed for their capacity to bind GPIbalpha and 6B4, respectively. From these data, it is found that the paratope of 6B4 is mainly formed by five residues: Tyr(27D), Lys(27E), Asp(28), and Glu(93) located in light chain CDR1 and -3, respectively, and Tyr(100C) of the heavy chain CDR3. These residues form a valley, where the GPIbalpha flexible loop can bind via residues Asp(235) and Lys(237). The experimental results were finally used to build a more accurate docking model. Taken together, this information provides guidelines for the design of new derivatized lead compounds with antithrombotic properties.     

A monoclonal antibody recognizing an icosapeptide sequence in the heavy chain of human factor XII inhibits surface-catalyzed activation

A monoclonal antibody (mAb B7C9) to human factor XII was raised in murine somatic cell using purified factor XII antigen. The purified antibody was subtyped IgG1 kappa and had a KD of 9.8 nM for antigen factor XII. Functional studies indicated that mAb B7C9 blocks surface-mediated coagulant activity of factor XII but not the amidolytic activity of factor XIIa against the small substrate H-D-Pro-Phe-Arg-p-nitroanilide (S-2302), suggesting that the mAb B7C9 epitope is located at or near the surface binding domain of the heavy chain region of factor XII. Western blot analysis indicated that the antibody reacts with factor XII and the heavy chain of factor XIIa. Affinity isolation of factor XII peptides, produced after cleavage by kallikrein, resulted in three factor XII heavy chain domain segments that were identified in the known factor XII sequence by limited N-terminal analysis. The epitope was located to a 20-amino acid sequence of 2.5 kDa in the heavy chain of factor XII which is the putative surface binding region of factor XII. The 2.5-kDa peptide was synthesized and demonstrated to react with mAb B7C9. mAb B7C9 was immobilized on an affinity resin and was successfully utilized to purify functionally active factor XII from plasma.     

Properties of a second epitope of the murine Fc receptor for aggregated IgG

The murine macrophage and lymphocyte Fc receptor for aggregated IgG (Fc gamma R) has previously been characterized by using the anti-Fc gamma R monoclonal antibody (mAb), 2.4G2. In the studies presented here, we describe a new mAb, 6B7C, that defines a second epitope of the Fc gamma R. The tissue distribution of the 6B7C epitope is coincident with the 2.4G2 epitope. However, only the 2.4G2 epitope is accessible to mAb binding on intact primary macrophages or lymphocytes. The 6B7C epitope is not detectable on primary macrophages or lymphocytes but is exposed on a portion of B lymphocyte Fc gamma R after activation by lipopolysaccharide and on some tumor cell lines. The expression of the 6B7C epitope on the surface of B lymphoblasts and tumor cell lines seems to correlate with their ability to release soluble Fc gamma R. The 6B7C mAb has the advantage that it reacts with native as well as denatured receptor and therefore can be used for techniques such as immunoblotting.     

Developmental changes in heparan sulfate expression: in situ detection with mAbs

Two mAbs that are specific for heparan sulfate-related epitopes have been raised and used to analyze the cellular and tissular distribution of this glycosaminoglycan during development. mAb 10E4 reacts with an epitope that occurs in native heparan sulfate chains and that is destroyed by N-desulfation of the glycosaminoglycan. The antibody does not react with hyaluronate, chondroitin sulfate, or DNA, and reacts only poorly with heparin. The reactivity of proteoglycan extracts or tissue sections with the 10E4 antibody is completely abolished by heparitinase, but is only partially affected by heparinase. mAb 3G10, in contrast, reacts only with heparitinase-treated heparan sulfate chains, proteoglycans, or tissue sections. The 3G10 epitope is destroyed by treatment with mercuric acetate, which indicates that the desaturated uronate generated by the lyase is essential for the reactivity of the antibody. The 3G10 epitope is not generated by treating heparan sulfate proteoglycans with heparinase or chondroitin sulfate proteoglycans with chondroitin sulfate lyases, which indicates that the 3G10 antibody recognizes desaturated uronates that occur in specific structural contexts. The antibody 10E4 and, after heparitinase treatment, the antibody 3G10 decorate the surfaces of many cell types and the extracellular matrix in proximity of the cells, in particular, the basement membranes. The analysis of embryonic and adult tissues reveals important temporal and regional differences in the abundance of the 10E4 and 3G10 epitopes at these sites. Moreover, the staining pattern of the two antibodies is not always superimposable, which is indicative of regional differences in the exposure or structure of the tissular heparan sulfates. As a whole the results suggest that heparan sulfate abounds at sites of active morphogenesis and that the expression of this glycosaminoglycan is developmentally regulated.     

Production and characterization of monoclonal antibodies to rabbit lymphocyte subpopulations. I. Tissue immunofluorescence and flow cytometric analysis

Nine monoclonal antibodies to rabbit T cells and B subpopulations have been generated from three separate fusions of spleen cells from mice immunized with fractionated populations of rabbit lymphocytes. These monoclonal antibodies, as well as a previously described rabbit T cell monoclonal antibody, 9AE10, have been analyzed by immunofluorescence staining on frozen tissue sections of rabbit thymus, spleen, and appendix. This screening method permits rapid identification of the lymphocyte subdomains in each tissue which is not possible by other screening methods. Each monoclonal antibody selected has a unique tissue staining pattern. Flow cytometric analysis of these monoclonal antibodies, using indirect immunofluorescence techniques on thymocytes, splenocytes, and PBL, revealed varying percentages of positive cells and individual mean fluorescence intensities indicating different epitope densities for each antigen. These monoclonal antibodies are now being used to characterize normal lymphocyte function and the role of specific lymphocyte subpopulations in experimental disease models in the rabbit.     

The goodpasture autoantigen. Identification of multiple cryptic epitopes on the NC1 domain of the alpha3(IV) collagen chain

Goodpasture (GP) disease is an autoimmune disorder in which autoantibodies against the alpha3(IV) chain of type IV collagen bind to the glomerular and alveolar basement membranes, causing progressive glomerulonephritis and pulmonary hemorrhage. Two major conformational epitope regions have been identified on the noncollagenous domain of type IV collagen (NC1 domain) of the alpha3(IV) chain as residues 17-31 (E(A)) and 127-141 (E(B)) (Netzer, K.-O. et al. (1999) J. Biol. Chem. 274, 11267-11274). To determine whether these regions are two distinct epitopes or form a single epitope, three GP sera were fractionated by affinity chromatography on immobilized NC1 chimeras containing the E(A) and/or the E(B) region. Four subpopulations of GP antibodies with distinct epitope specificity for the alpha3(IV)NC1 domain were thus separated and characterized. They were designated GP(A), GP(B), GP(AB), and GP(X), to reflect their reactivity with E(A) only, E(B) only, both regions, and neither, respectively. Hence, regions E(A) and E(B) encompass critical amino acids that constitute three distinct epitopes for GP(A), GP(B), and GP(AB) antibodies, respectively, whereas the epitope for GP(X) antibodies is located in a different unknown region. The GP(A) antibodies were consistently immunodominant, accounting for 60-65% of the total immunoreactivity to alpha3(IV)NC1; thus, they probably play a major role in pathogenesis. Regions E(A) and E(B) are held in close proximity because they jointly form the epitope for Mab3, a monoclonal antibody that competes for binding with GP autoantibodies. All GP epitopes are sequestered in the hexamer configuration of the NC1 domain found in tissues and are inaccessible for antibody binding unless dissociation of the hexamer occurs, suggesting a possible mechanism for etiology of GP disease. GP antibodies have the capacity to extract alpha3(IV)NC1 monomers, but not dimers, from native human glomerular basement membrane hexamers, a property that may be of fundamental importance for the pathogenesis of the disease.     

High molecular weight kininogen-binding site of prekallikrein probed by monoclonal antibodies.

A panel of monoclonal antibodies against human prekallikrein was raised in mice and characterized with respect to the major antigenic epitopes. Of 18 antibodies, nine were directed against the light chain portion performing the proteolytic function of activated kallikrein, and nine recognized the heavy chain mediating the binding of prekallikrein to high molecular weight (H-)kininogen. Among the anti-heavy chain antibodies, one (PK6) interfered with the procoagulant activity of prekallikrein, and prolonged in a concentration-dependent manner the activated partial thromboplastin time of reconstituted prekallikrein-deficient plasma (Fletcher type). Antibody PK6 was subtyped IgG1,k and had an apparent Kass of 6.8 +/- 0.44.10(8) M-1 for prekallikrein. Functional analyses revealed that PK6 does not interfere with prekallikrein activation by activated Hageman factor (beta-F XIIa), and has no effect on the kininogenase function of activated kallikrein. Monoclonal antibody PK6 but none of the other anti-heavy chain antibodies completely prevented complex formation of prekallikrein with H-kininogen, and readily dissociated preformed complexes of prekallikrein and H-kininogen. Likewise, Fab' and F(ab')2 fragments of PK6 blocked H-kininogen binding to prekallikrein. A synthetic peptide of 31 amino acid residues encompassing the entire prekallikrein binding region of H-kininogen effectively competed with PK6 for prekallikrein binding indicating that the target epitope of PK6 is juxtaposed to, if not incorporated in the H-kininogen-binding site of prekallikrein. Extensive cross-reactivity of PK6 with another H-kininogen-binding protein of human plasma, i.e. factor XI, suggested that the structure of the target epitope of PK6 is well conserved among prekallikrein and factor XI, as would be expected for the kininogen-binding site shared by the two proteins. It is anticipated that monoclonal antibody PK6 will be an important tool for the precise mapping of the hitherto unknown kininogen-binding site of prekallikrein.     

Amino acid residues 56 to 69 of HLA-A2 specify an antigenic determinant shared by HLA-A2 and HLA-B17

The mouse monoclonal antibody MA2.1 was previously used to define an epitope shared by native HLA-A2 and HLA-B17 molecules and amino acid sequence comparison of nine HLA-A,B,C molecules identified residues 62 to 65 as the region most likely to form this epitope. An unabsorbed rabbit antiserum raised against a peptide corresponding to residues 56 to 69 of HLA-A2 gives highly specific reactions with HLA-A2 and HLA-B17 heavy chains in Western blots. No interactions with native HLA-A2 and B17 molecules were detected in a variety of assays. Although the topographic relationship between the epitopes recognized by the rabbit antiserum and the monoclonal antibody could not be determined, the results show that residues 56 to 69 of HLA-A2 can form epitopes with specificity for HLA-A2 and HLA-B17.     

Integrin alpha3beta1 expressed by human colon cancer cells is a major carrier of oncodevelopmental carbohydrate epitopes

Oncodevelopmental carbohydrate epitopes are a common feature of human colorectal carcinoma, yet their biological significance remains unclear. We have shown previously that monoclonal antibody (MAb) 3A7, which recognizes a determinant on type 2 chain blood group A and B oligosaccharides, detects oncodevelopmental changes in azoxymethane-induced rat colon tumors and some human colon cancer cell lines. (Laferté S et al. [19951 J. Cell. Biochem. 57:101-119). In this study, we set out to purify gp140, the major glycoprotein carrier of the 3A7 epitope expressed by human colon cancer cells, as a first step towards elucidating the contribution of the 3A7 epitope and its major glycoprotein carrier to colon cancer progression. To this end, gp140 was purified from HT29 cells and used for the preparation of polypeptide-specific monoclonal antibodies. Five monoclonal antibodies (7A8, 7B11, 8C7, 8H7, and 11D4) immunoprecipitated a 3A7-immunoreactive glycoprotein complex of 140 kDa which was subsequently identified by partial protein sequencing as alpha3beta1 integrin. Flow cytometric analysis of Chinese hamster ovary (CHO) cells expressing different human integrin chains revealed that MAbs 7A8 and 7B11 detect the alpha3 integrin subunit whereas MAbs 8C7 and 8H7 detect the beta1 integrin subunit. MAb 11D4, which did not bind to any of the CHO transfectants, detected type 2 chain blood group A determinant. Flow cytometric analysis of a panel of human colon carcinoma cell lines obtained from blood group A, AB, or B individuals revealed a direct correlation between cell-surface expression of the 3A7 epitope and alpha3 integrin subunit, suggesting that alpha3beta1 integrin is a preferred target of the 3A7 epitope in colon cancer cells. Using lectins and glycosidases to examine further the carbohydrate structure of alpha3beta1 integrin, we demonstrated that it is a sialoglycoprotein containing both N- and O-linked oligosaccharides. In addition, both alpha3 and beta1 integrin subunits express beta1-6 branched Asn-linked oligosaccharides and short poly-N-acetyllactosamine units (Galbeta1-4GlcNAc-R; n < or = 3), glycans previously implicated in cancer metastasis.Thus, alpha3beta1 integrin expressed by human colon carcinoma cells is a major carrier of oncodevelopmental carbohydrate epitopes whose presence may modulate tumor cell adhesion, migration, and/or invasion.     

Purification and characterization of iduronic acid-rich and glucuronic acid-rich proteoglycans implicated in human post-burn keloid scar

Small proteoglycans (PGs), extracted from human keloid scar tissue with 4M guanidinium chloride and fractionated by DEAE-cellulose chromatography, were separated by ethanol precipitation into one L-iduronic acid-rich and one D-glucuronic acid-rich fraction. The size of the L-iduronic acid-rich PG was 102 kDa with a 27 kDa glycosaminoglycan chain, that of the D-glucuronic acid-rich PG was 90 kDa with a 26 kDa glycosaminoglycan chain, and the protein core of both PGs was 14.5 kDa. The two PGs carried sulfate groups mostly attached at C-4 of the 2-amino-2-deoxy-D-galactose units. The N-terminal amino acid sequence of both was similar to human bone PGII (decorin), normal and hypertrophic scar, and human dermal tissue PG.     

Analysis of glycosaminoglycan substitution in decorin by site-directed mutagenesis

Posttranslational glycosaminoglycan attachment to decorin, a chondroitin/dermatan sulfate proteoglycan, was studied by expression of a wild-type decorin cDNA and several mutagenized forms in two types of mammalian cells. Transfection of the wild-type cDNA resulted in the synthesis of an authentic chondroitin/dermatan sulfate proteoglycan similar to the decorin molecule synthesized by cultured human fibroblasts. Conversion of the serine residue that serves as the attachment site for the sole glycosaminoglycan chain in decorin to a threonine residue greatly reduced the efficiency of the glycosaminoglycan substitution. Less than 10% of the threonine-mutated core protein acquired a glycosaminoglycan chain, whereas most of the core protein was secreted without such substitution. Expression of cDNA in which an alanine residue had been introduced into the substituted serine position resulted in the secretion of core protein with no detectable glycosaminoglycan. Conversion to alanine of either one of the glycine residues that are adjacent to the substituted serine yielded the proteoglycan form of decorin. These results show that the xylosyltransferase responsible for the initiation of the glycosaminoglycan chain on the core protein can use a threonine residue for this substitution instead of a serine residue, but that such substitution is only partial, creating a "part-time" proteoglycan. Moreover, variations are possible in the sequence context of a glycosaminoglycan-substituted serine residue without loss of glycosaminoglycan substitution. The conformation of the substitution site may therefore be important for xylosyltransferase recognition.