Protein components of human tracheobronchial mucin: partial characterization of a closely associated 65-kilodalton protein

Glycosylation in the Fc domain of IgG increases resistance to proteolytic cleavage by papain

IgG antibodies (Abs) and fragments of IgG Abs are becoming major biotherapeutics to treat an assortment of human diseases. Commonly prepared fragments of IgGs include Fc, Fab, and F(ab')2 fragments, all of which can be made using the sulfhydryl protease papain, although prolonged digestion times and/or excessive amounts of papain typically result in further cleavage of the Fc domain into smaller fragments. During our attempts to use papain to isolate Fc fragments from different IgG monoclonal Abs, it was observed that prior removal of Fc glycans resulted in a faster rate of papain-mediated degradation of the Fc domain. Subsequent time-course experiments comparing glycosylated and deglycosylated versions of IgG antibodies showed that the majority of molecules in a deglycosylated IgG sample were converted into Fab, Fc, and smaller Fc fragments in less than one hour, whereas the original glycosylated IgG required more than two hours to convert into a comparable amount of Fab and Fc fragments. Furthermore, whereas papain digestion converted almost all of a deglycosylated Fc fragment into smaller fragments of approximately 10 and approximately 12 kDa within 4 h, more than 40% of a glycosylated Fc fragment remained intact even after 24 h of digestion. These results indicate that the presence of CH(2) domain glycans in either IgGs or purified Fc fragments increases resistance to papain digestion. Increased sensitivity of non-glycosylated Fc domains to papain is consistent with the Fc domains lacking a defined structure, as exemplified by their inability to bind Fcgamma receptors, since misfolded proteins are often degraded by proteases because of increased accessibility of their proteolytic cleavage sites. Based on these observations it is possible to use papain sensitivity as a means of assessing proper Fc structure of IgG molecules.

Identification of a nonmucin glycoprotein (gp-340) from a purified respiratory mucin preparation: evidence for an association involving the MUC5B mucin

Rate-zonal centrifugation of a reduced and alkylated respiratory mucin preparation identified a protein-rich fraction. This was subjected to trypsin treatment and one of the many liberated peptides was purified and its N-terminal sequence determined. The peptide was identical to a 14 amino acid sequence from the scavenger receptor cysteine-rich domain containing glycoprotein gp-340. A polyclonal antiserum, raised against the peptide, stained the serous cells in the submucosal glands of human tracheal tissue. The glycoprotein was purified from respiratory mucus by density-gradient centrifugation, gel chromatography, and anion exchange chromatography. The molecule exhibited a heterogeneous distribution of buoyant density (1.28-1.46 g/ml) that overlapped with the gel-forming mucins, was included on Sepharose CL-2B and was quite highly anionic. SDS-PAGE indicated a mass greater than 208 kDa and measurements performed across the molecular size distribution indicated an average M(r) of 5 x 10(5) with a range of M(r) from 2 x 10(5) to 1 x 10(6). Gel chromatography of respiratory mucus extracts ("associative" and "dissociative") indicated that this glycoprotein forms complexes that may involve the large gel-forming mucins MUC5AC and MUC5B. Rate zonal centrifugation suggested such complexes are more likely to involve MUC5B rather than MUC5AC mucins.

Disulfide-bound proteolytic fragments of gastric mucin are 100- and 140-kDa proteins

Pig gastric mucus was tested for its autodegradative proteolytic degradation at pH 7.0, in the presence or absence of proteinase inhibitors and SDS. Samples of crude mucus were incubated at room temperature for 48 and 96 h in sodium azide stabilized buffer, pH 7. 0, and urea-extracted mucin was purified. Electrophoretically homogenic mucin preparation was reduced and alkylated with iodo[(14)C]acetamide, and analyzed for labeled products. On 7.5% SDS/PAGE protein bands at 80 and 120 kDa were noted, but radioactivity was incorporated into 100- and 140-kDa bands, with increasing intensity from T(0) to T(96), and into high molecular mass mucin subunits. The results confirmed the autodegradative properties of gastric mucin and demonstrated that the 100- and 140-kDa fragments are the main proteolytical products of pig gastric mucin and are disulfide bound with the rest of the molecule.

Binding of Shigella to rat and human intestinal mucin

Invasion of epithelial cells by Shigella is an early step in their pathogenesis. Adherence is generally presumed to be a prerequisite for invasion. This study examined the possibility of intestinal mucins serving as initial binding sites for clinical isolates of S. boydii and S. sonnei. The interactions of Shigella with rat and human small intestinal and colonic mucin were investigated. In solid phase binding assays, [35S] labelled Shigella did not show any preferential binding to rat/human small intestinal mucin or to rat colonic mucin. On the other hand, Shigella bound specifically to human colonic mucin in a concentration-dependent manner. This specific binding to human colonic mucin was not by weak hydrophobic interactions and could not be attributed to the presence of contaminating glycolipids in the mucin preparation. The human colonic mucin receptor was sensitive to periodate treatment suggesting the involvement of the carbohydrate portion of the mucin. Reduction and alkylation of mucin enhanced adherence probably by exposing buried binding sites. The monosaccharides present in mucins were ineffective as hapten inhibitors as was the lectin wheat germ agglutinin suggesting that the mucin receptor is a more complex one. This study identifies, for the first time, the presence of a specific Shigella-binding site on the carbohydrate portion of human colonic mucin, which is not present in rat colonic mucin or in rat/human small intestinal mucin.

Biochemical analysis of a bladder-cancer-associated mucin: structural features and epitope characterization

Three monoclonal antibodies (mAbs), M344, M300 and M75, were shown to define a unique tumour-associated antigen (TAA) of superficial bladder tumours. The antigenic determinants are expressed on a very-high-molecular-mass component and, in about 50% of the positive samples, one determinant is also detected on a 62 kDa molecular species, observed only under reducing conditions. The objectives of the present study were to characterize further this TAA by analysing (1) the biochemical nature of the epitopes recognized by the three mAbs, and (2) the biochemical and structural features of the molecule bearing them. The antigenicity was resistant to heat denaturation, trypsin and alpha-chymotrypsin treatments but highly sensitive to papain and Pronase digestion. NaIO4 oxidation decreased reactivity to mAbs M344 and M300 but enhanced reactivity to mAb M75. The three determinants were insensitive to beta-galactosidase and alpha-L-fucosidase but were sensitive to Vibrio cholerae neuraminidase. None of the three mAbs reacted with ovine, bovine or porcine submaxillary mucins. Deglycosylation with O-glycosidase or trifluoromethanesulphonic acid completely abolished the reactivity of the mAbs whereas N-glycosidase F deglycosylation had no appreciable effect. The presence on the molecule of cryptic Gal(beta(1-3))GalNAc as a major core disaccharide was demonstrated by a heterologous sandwich assay using mAb M75 and peanut agglutinin. Thiol reduction using beta-mercaptoethanol increased mobility of the high-molecular-mass component in polyacrylamide gels. We thus conclude that mAbs M344 and M300 react with sialylated carbohydrate epitopes, and mAb M75 reacts with a partially cryptic and periodate-resistant sialylated epitope expressed on a typical secreted high-molecular-mass oligomeric mucin which we named MAUB for mucin antigen of the urinary bladder.

Promoter of the canine tracheobronchial mucin gene

The mucin gene is up-regulated in diseases such as cystic fibrosis (CF) and asthma. To understand the mechanisms involved in transcriptional regulation of mucin gene expression we have characterized the region of the mucin gene up-stream of the transcriptional start site and analysed the cis-acting elements required for mucin promoter activity. We isolated clones from a dog genomic library containing the promoter region for the tracheobronchial mucin gene (TBM). The authenticity of the promoter was tested by nucleotide sequencing, primer extension analysis, electrophoretic mobility shift assay (EMSA) and reporter gene expression analysis. The canine TBM promoter is different from housekeeping gene promoters (as it is not rich in GC content and contains TATA- and CAAT-like sequences) and different from that of regulatory genes (because it contains many TATA- and CAAT-like sequences and multiple transcriptional initiation sites). Reporter gene analysis using canine TBM promoter-chloramphenicol acetyltransferase (CAT) fusion plasmids established the regions responsible for promoter activity and verified the positions of the major mucin transcriptional initiation sites. Reporter gene analysis also established that a region of the canine TBM promoter and first exon containing all of the transcriptional initiation sites is more active in mucin expressing cells (e.g. CT1 cells-immortalized canine tracheal epithelial cells, human CFT1 cells-immortalized tracheal epithelial cells from a CF subject, or HBE1 cells-immortalized tracheal epithelial cells from non-CF subject) than in mucin non-expressing cells (COS7, 3T3), suggesting cell specificity. The promoter region contained cAMP response element (CRE) sequences, and the TBM gene transcription was enhanced when cAMP analogs were added to transfected cells. EMSA indicated the presence of at least two DNA binding proteins in CT1 cells. This is the first report describing the characterization of a TBM gene promoter. The information obtained in the present studies will be valuable in understanding mucin gene regulation in normal and pathological conditions.

Gelation of fractionated canine submaxillary mucin in a chaotropic solvent

Rheological measurements have been performed on three molecular weight fractions of purified canine submaxillary mucin (CSM) dissolved in the chaotropic solvent 6 M guanidine hydrochloride (GdnHCI). Solutions of the lower molecular weight fractions are viscoelastic sols, and their dynamic moduli can be scaled with respect to molecular weight and concentration according to linear viscoelasticity theory. In contrast, preparations of the highest molecular weight fraction form viscoelastic gels that exhibit an equilibrium shear modulus, Ge', which scales with mucin concentration as Ge' approximately c3. Amino acid and carbohydrate analyses of all three fractions are similar; thus, the differences in rheological behavior are attributed to molecular weight differences, which affect the degree of coil overlap in solutions of a given concentration. These observations demonstrate conclusively that mucin glycoproteins of high molecular weight form gels under conditions in which the mucin chains physically interpenetrate, even when non-covalent intermolecular interactions are extensively disrupted. A comparison of these results with previous studies of purified submaxillary and tracheobronchial mucins indicates that the carbohydrate side-chain length, in addition to molecular weight, is an important determinant of the observed elastic response and the ability to form physical gels.

Viscoelastic properties of human tracheobronchial mucin in aqueous solution

Human tracheobronchial mucin isolated from cystic fibrosis patients (CF HTBM) was purified using a combination of gel filtration and density gradient centrifugation. The resulting mucin was fractionated to reduce polydispersity and to facilitate studies of the molecular weight dependence of mucin viscoelasticity in concentrated solution. The viscoelastic properties of CF HTBM were examined in distilled water, 0.1M salt solutions and chaotropic solvents. In controlled strain experiments (strain > or = 5%) with increasing mucin concentration, a crossover from sol to gel behavior is observed. The gel strength, as measured by the magnitude of the storage modulus at comparable mucin concentrations, is greatest for distilled water, intermediate for 0.1M NaCl, and lowest for 6M GdnHCl. In distilled water, high molecular weight mucin undergoes a sol-gel transition at approximately 12 mg/mL, and shows evidence of a plateau modulus at higher concentrations. The storage and loss moduli of concentrated high molecular weight fractions in 6M GdnHCl exhibit a power law dependence on frequency typical of weak gels near the sol-gel transition at 20 mg/mL. Similar rheology is observed in 0.1M NaCl and 0.091M NaCl/3 mM CaCl2, but with evidence for additional weak associations at low frequency. The power law exponent in these systems is 0.70 +/- 0.02, in good agreement with prediction for networks formed by a percolation mechanism. Low molecular weight fractions in these solvents exhibit a fluid-like viscoelastic response. However, low molecular weight mucin in distilled water shows a strain-dependent increase in elasticity at low frequency indicative of weak intermolecular associations. Comparison of the rheological behavior of CF HTBM with our earlier studies of ovine submaxillary mucin lends support to the idea that carbohydrate side-chain interactions are important in the gelation mechanism of mucins.

Molecular cloning and sequencing of a canine tracheobronchial mucin cDNA containing a cysteine-rich domain

To date the complete sequence of only one mammalian mucin cDNA, MUC1, has been reported, although several mucin proteins have been partially characterized. Here we report the nucleotide sequence of a canine tracheal mucin cDNA containing two potential translation initiation codons, one translation termination codon and a poly(A) tail. A lambda gt11 cDNA library prepared from canine tracheal epithelial cells was screened with polyclonal anti-apo-canine tracheal mucin antibodies with the aim of obtaining the deduced amino acid sequence of the mucin core protein. Antibody-positive clones containing overlapping inserts of various lengths were purified and used for nucleotide sequencing. Based on the sequencing data, synthetic oligonucleotide primers were constructed and both ends (5' and 3') of the cDNA were determined. The complete sequence was 3.7 kb and included an open reading frame with coding capacity for 1118 aa, two translation initiation ATG codons in context with Kozak consensus sequences, one polyadenylylation site, and a poly(A) stretch. The protein was rich in Thr, Pro, Ser, Gly, and Ala and poor in Tyr, Phe, and Trp. Although tandem repeats of amino acids were absent in the deduced canine tracheal mucin sequence, motifs TPTPTP and TTTTPV appeared 13 and 19 times, respectively. The C-terminal region contained a Cys-rich domain (although a few Cys residues were also present in the middle of the protein) as has been reported for bovine submaxillary mucin, porcine submaxillary mucin, rat intestinal mucin, human intestinal mucin, and frog skin mucin. This suggested that a broad group of mucins contain such a Cys-rich domain whose functional significance is yet to be understood. Three potential N-glycosylation sites were present in canine tracheal mucin and the amino acid sequence showed homology with both human tracheal and intestinal mucins. The N-terminal domain showed more flexibility (probably due to a high number of Pro residues in this region) when analyzed by the University of Wisconsin Genetics Computer Group program package to determine the predicted secondary structure. Evaluation of the transcripts using the canine mucin cDNA as a probe indicated a polydisperse message with total RNA.

Biophysical approaches to salivary mucin structure, conformation and dynamics

Our understanding of the origins of the physical and biochemical properties of mucous glycoproteins is incomplete and not with out controversy. Recent molecular biological and biophysical studies revealing the architecture and solution structure and dynamics of a series of salivary mucins, invaluable toward resolving many of these questions, are discussed. Mucins are very large, structurally heterogeneous, and highly expanded molecules with the carbohydrate playing a key role in maintaining the extended mucin conformation.

Purification of mucin glycoproteins by density gradient centrifugation in cesium trifluoroacetate

A procedure for the rapid isolation of mucin glycoprotein by density gradient centrifugation in cesium trifluoroacetate (CsTFA) is described. The separation of mixtures of rat tracheobronchial mucin, DNA, hyaluronic acid, and bovine serum albumin in CsTFA gradients was superior to that in cesium bromide gradients. Inclusion of guanidinium chloride or urea in the gradient had no influence on the separation obtained. The mucins isolated from sputum samples of cystic fibrosis patients by this procedure are largely free of nucleic acid, nonglycosylated proteins, and glycosaminoglycans. The results of the use of CsTFA gradient centrifugation for the isolation of mucin from extracts of bovine submaxillary gland are also presented. The CsTFA method is particularly suitable for the high-yield isolation of mucin from individual samples which are available in limited quantities.

Human tracheobronchial mucin: purification and binding to Pseudomonas aeruginosa

Colonization of the respiratory tract with Pseudomonas aeruginosa is a serious problem in cystic fibrosis and seriously ill hospitalized patients. Human tracheobronchial mucin (HTBM), the major glycoprotein of human tracheobronchial secretions, is known to interact with this pathogen, which may then be cleared by mucociliary action. However, the mechanism of interaction is not known. To understand this process, pure HTBM was isolated from tracheobronchial secretions of a laryngectomee. Following initial fractionation on Sepharose CL-2B, the HTBM-containing fraction was subjected to reductive methylation and then gel filtration. Pure HTBM was employed in an overlay binding assay to identify the bacterial adhesin(s) and mucin receptors that participate in mucin-P. aeruginosa interactions. An approximately 16-kDa nonpilus protein component(s) of P. aeruginosa was found to be the adhesin(s) for HTBM. The mucin receptor for the 16-kDa component(s) was found in the peptide moiety. This study confirms that P. aeruginosa utilizes the nonpilus adhesin(s) to bind to HTBM. Identification of the specificity of the HTBM-P. aeruginosa interactions can lead to a better understanding of the predominance of P. aeruginosa colonization in individuals with cystic fibrosis.

The presence of neutral metalloproteolytic activity and metalloproteinase inhibitors in the interphotoreceptor matrix

Neutral proteolytic activity, having a pH optimum of about 7, was present in the high molecular weight fraction of bovine interphotoreceptor matrix (IPM) separated by gel filtration on Sephacryl S-500. The enzyme(s) was active toward a number of exogenous substrates, including albumin, Azocoll, and gelatin. However, it was inactive toward a synthetic substrate for bacterial collagenase. Proteolytic activity was proportional to protein; however, the time course of the reaction was nonlinear, suggesting that "activation" of a precursor form might be necessary. Of a number of specific inhibitors tested, those directed toward metalloproteinases (1,10-phenanthroline greater than EDTA greater than EGTA) proved most effective. While activity was also inhibited by sulfhydryl reagents and dithiothreitol, inhibitors specific for cysteine proteinases were ineffective. Higher specific activity was present in IPM obtained from retinal pigment epithelium (RPE) than from retina. An endogenous proteinase inhibitor(s) was also found in IPM from both RPE and retina. It was effective against the endogenous metalloproteolytic activity of IPM and also against thermolysin, but not against trypsin or papain. Fractionation of IPM on Sephacryl S-500 revealed a broad peak of inhibitory activity at molecular weights of less than 10(5) daltons. This is the first report of the presence of neutral proteolytic activity and metalloproteinase inhibitor(s) in bovine IPM. These materials may function in concert to maintain the proper level of various components within this matrix.

Rat sublingual gland as a model to study glandular mucous cell secretion

To study the regulation of mucous cell secretion, we have developed an in vitro cell model consisting of enzymatically dispersed mucous acinar structures (cell aggregates) from rat sublingual glands. Histological and ultrastructural evidence demonstrates that the cell aggregates are highly enriched in mucous cells, retain the morphological and ultrastructural features observed in intact glands, and undergo transition to an extensive secretory state when stimulated by 10 microM carbachol. The secretory responsiveness of the cell aggregates was verified in pharmacological studies. Carbachol stimulated secretion in a dose-dependent manner with high affinity (concentration causing half-maximal response = 0.3 microM) and was completely inhibited by atropine. Secretion was also stimulated by vasoactive intestinal peptide and substance P but not by alpha- or beta-adrenergic agonists. Biochemical characterization of secretion during nonstimulated and carbachol-stimulated conditions (after preincubation in [3H]glucosamine) demonstrated that, in response to carbachol, cell aggregates synthesized and secreted mucins which were similar to mucin glycoproteins isolated from whole glands. Collectively, our results establish that the rat sublingual cell aggregate model is a viable and pharmacologically responsive cell system to study the regulation of mucous cell secretion.

Effects of hydrogen peroxide, mild trypsin digestion and partial reduction on rat intestinal mucin and its disulphide-bound 118 kDa glycoprotein

The role of the disulphide-bound 118 kDa glycoprotein of rat intestinal mucin is unknown, although it has been proposed to serve as a 'link' component for the mucin monomers. The present studies investigated release or destruction of the 118 kDa glycoprotein (monitored by gel electrophoresis and Western-blot analysis) during progressive breakdown of the mucin polymer (assessed by Sepharose 2B chromatography). H2O2 gradually destroyed the 118 kDa glycoprotein and dissociated the mucin polymer into components of similar size to the monomers. After 3 h, mucin samples contained almost no 118 kDa glycoprotein or its breakdown products, but 50% of the mucin was still eluted in the void volume of a Sepharose 2B column. Although mild trypsinolysis had little effect on the Sepharose 2B elution profile of the mucin, the 118 kDa glycoprotein was completely cleaved into 54-56 kDa and 60-66 kDa fragments which remained disulphide-bound to the high-molecular-mass mucin. Increasing levels of thiol reduction resulted in progressive loss of disulphide bonds, release of the 118 kDa glycoprotein and depolymerization of the mucin. Although approx. 40% of the mucin in partially reduced samples was recovered in the Sepharose 2B void volume, this material contained no 118 kDa glycoprotein and apparently consisted of disulphide-bound mucin monomers. Thus the 118 kDa glycoprotein may be destroyed by H2O2, extensively cleaved by trypsin or released by reduction without completely dissociating the mucin into monomers. Therefore the 118 kDa glycoprotein may not function as a 'link' component for all of the mucin monomers in the native polymer.

Polymeric structure of human respiratory mucin: studies on two protein components released upon reduction of disulfide bonds

A major mucus glycoprotein (mucin) was purified from the tracheobronchial secretions of an asthmatic patient. Upon SDS-composite gel electrophoresis, the purified native (non-reduced) mucin gave a single band. SDS-gel electrophoresis on 6% polyacrylamide gels showed the absence of low molecular mass protein contaminants. However, SDS-PAGE (6% gels) of the reduced mucin showed the presence of a major high molecular mass mucin component and two low molecular mass components of 118 and 70 kDa, respectively. The 118 and 70 kDa components were purified by preparative electroelution of the reduced mucin. These components were also separated from the reduced mucin by gel-permeation chromatography on a Superose 6 column. Chemical compositional analyses showed that the 118 kDa component was a glycoprotein while the 70 kDa component was non-glycosylated. The effect of disulfide bond reduction on mucin structure and the hydrophobic probe binding properties of native and reduced mucin were studied using the fluorescent probe technique. Mansylphenylalanine was used as the fluorescent probe. The native mucin showed the presence of a large number of low-affinity (KD approximately 10(-5) M) binding sites for the probe. On the other hand, reduced-alkylated mucin containing the 118 and 70 kDa components showed the presence of additional high-affinity (KD approximately 10(-6) M) binding sites as well as low-affinity binding sites for the probe. Reduced alkylated mucin devoid of the 118 and 70 kDa components showed the presence of only low-affinity binding sites. These observations suggest that the availability of high-affinity probe binding sites upon reduction of mucin disulfide bonds may be either due to binding of the probe to the released component(s) and/or due to noncovalent interaction of the released component(s) with the mucin causing a conformational change in the mucin structure. Thus, the 118 and 70 kDa components appear to be an integral part of the total polymeric structure of the human respiratory mucin.

Cloning and cDNA sequence of a bovine submaxillary gland mucin-like protein containing two distinct domains

A lambda gt11 cDNA library prepared from bovine submaxillary gland mRNA was screened with polyclonal anti-apo-bovine submaxillary mucin antibodies with the aim of obtaining the deduced amino acid sequence of the mucin core protein. One of the positive clones had a 1.8 kilobase (kb) cDNA insert and coded for an incomplete protein. A 2.0-kb cDNA clone was isolated by rescreening the library with the 1.8-kb cDNA. Nucleotide sequencing of the full-length 2.0-kb cDNA revealed an open reading frame that coded for a 563-amino acid protein. A striking feature of the cloned protein is the skewed distribution of the amino acids, most notably that of the hydroxy amino acids and cysteine. The amino-terminal domain of 339 residues is very rich in threonine, serine, and glycine and poor in cysteine, aspartic acid, tyrosine, phenylalanine, and tryptophan. In contrast, the carboxyl-terminal domain of 224 residues is rich in cysteine, aspartic acid, tyrosine, lysine, and asparagine and relatively poor in threonine, serine, and glycine. A search of the protein data bank for homologies to the deduced amino acid sequence revealed statistically significant matches to several proteins, including the porcine submaxillary apomucin fragment. The cysteine-rich domain by itself was not statistically homologous with any of the registered polypeptide sequences. RNA blot analysis using DNA probes corresponding to the mucin-like and cysteine-rich regions detected a nearly identical pattern of transcripts, demonstrating that the characterized clones are not artifacts of cDNA library construction. The blots also showed the presence of polydisperse transcripts in bovine submaxillary gland but no detectable hybridization signals in liver or brain RNA.

Structural analysis of purified human tracheobronchial mucins

Light scattering has been used to investigate the structure of human tracheobronchial mucin glycoproteins (HTBM) from the sputum of cystic fibrosis patients. The specimen was extracted using 6M guanidinium hydrochloride solution and fractionated by gel exclusion chromatography on Sephacryl S-1000. The fractionated HTBM was purified by density gradient ultracentrifugation. Purity of the resulting material was confirmed by SDS polyacrylamide gel electrophoresis and uv spectroscopy. Light scattering measurements on the fractionated mucins yield weight-average molecular weights Mw, and z-average radii of gyration Rg,z. The native cystic fibrosis HTBM consisted of a high molecular weight fraction with Mw = 9.3 X 10(6) daltons and a lower molecular weight fraction containing partly degraded mucins. After reduction and carboxymethylation of the high molecular weight native fraction, the resulting material was separated into three pools with Mw values of 5.1 X 10(6), 1.6 X 10(6), and 400,000. The derived molecular weights for the protein cores Mp,w, and the experimental radii of gyration are found to be consistent with the Mp,w -Rg relation established previously for submaxillary, cervical, and gastric mucins. These results imply that HTBM has the same extended-coil conformation reported for other mucins and has a molecular structure consisting of subunits, linked into linear chains via covalent (disulfide) bonds.