research-article2016
JHCXXX10.1369/0022155416668139Sialidase unmasks lubricin mucin domain epitopesSolka et al.
Article Journal of Histochemistry & Cytochemistry 2016, Vol. 64(11) 647–668 © 2016 The Histochemical Society Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1369/0022155416668139 jhc.sagepub.com
Sialidase Unmasks Mucin Domain Epitopes of Lubricin Kathryn A. Solka, Ira J. Miller, and Thomas M. Schmid
Department of Biochemistry (KAS, TMS) and Department of Pathology (IJM), Rush University Medical Center, Chicago, Illinois
Summary Lubricin is a secreted, mucin-like glycoprotein and proteoglycan abundant in synovial fluid that provides boundary lubrication and prevents cell adhesion in synovial joints. The antilubricin S6.79 monoclonal antibody recognizes an O-linked glycopeptide epitope in lubricin’s mucin domain. The central, long mucin domain of lubricin is extensively O-glycosylated with Gal(β1-3)GalNAc-O-Ser/Thr, and about two thirds of the O-glycosylated sites are capped with sialic acid. Our aim was to determine whether removal of sialic acid by sialidase could improve the detection of lubricin in a number of human tissues using the S6.79 monoclonal antibody. Sialidase treatment caused a dramatic increase in antibody reactivity in human pericardium, splenic capsule and trabeculae, plasma, serum, eye sleep extract, and liver sinusoids. Sialidase had minimal effect on S6.79 antibody reactivity with lubricin in synovial fluid and synovial tissue. These observations suggest that the origin of lubricin in blood may be different from that in synovial fluid and that desialylation of lubricin is essential for unmasking epitopes within the mucin domain. (J Histochem Cytochem 64:647–668, 2016) Keywords epitopes, eyelid, liver, lubricin, neuraminidase, pericardium, sialidase, spleen, synovial fluid
Introduction The lubricin gene, proteoglycan 4 (PRG4), encodes for a heterogeneous population of proteins that differ by alternative splicing, glycosylation, function, and tissue distribution. Lubricin was first isolated and characterized as the major lubricating glycoprotein in synovial fluid.1,2 Lubricin generates a friction-reducing, boundary-lubricating antiadhesive coating on the surface of articular cartilage, which prevents adhesion of synovial cells, chondrocytes, and protein, thereby protecting joint tissues from wear and breakdown.3–5 Lubricin also prevents synovial cell hyperplasia.3 Lubricin’s extensive O-linked glycosylation is essential for its function. Desialylation of lubricin does not affect its lubricating activity; however, subsequent removal of 54% of lubricin’s O-linked carbohydrates increases lubricin’s coefficient of friction by nearly 80%, thereby decreasing its ability to lubricate.6,7 We have generated antihuman lubricin monoclonal antibodies.8 S6.79, one of the highest affinity antibodies,
recognizes an O-linked glycopeptide epitope repeated a few times in the mucin domain.9 In a recent immunohistochemical study with this antibody, we detected lubricin in several pathological tissues including tenosynovial giant cell tumors, myxomas, and ganglion cysts.10 In our survey of more than 60 human tissues, S6.79 reactivity was highly restricted to only a few tissues. The antibody did not stain proteoglycan-rich tissues such as umbilical cord and fetal tissues. In competition ELISAs, it did not react with aggrecan,8 decorin, and biglycan (data not shown). The S6.79 antibody did not detect versican by Western blots of partially purified material from bovine aorta, nor did it react with human or bovine aortic media by IHC. The S6.79 antibody is nonreactive with versican Received for publication March 14, 2016; accepted August 11, 2016. Corresponding Author: Thomas M. Schmid, Department of Biochemistry, Rush University Medical Center, 1735 W. Harrison St., Cohn Research Building, Suite 524, Chicago, IL 60612, USA. E-mail: [email protected]
Journal of Histochemistry and Cytochemistry
668139
648 either with or without chondroitinase ABC treatment (data not shown). The S6.79 antibody is highly specific for O-linked glycopeptides within the mucin domain of lubricin.9 The central, long mucin domain of lubricin is extensively O-glycosylated with Gal(β1-3)GalNAc-OSer/Thr, and about two thirds of the O-glycosylated sites are capped with sialic acid as (α2,3)NeuAc-(β1-3)GalGalNAc-O-Ser/Thr.2,6,7,11–14 Our aim was to determine whether removal of sialic acid by sialidase could enhance the detection of lubricin with the S6.79 antibody in human tissues by IHC and Western blot analysis. Our data show that the S6.79 lubricin epitope is highly masked by sialic acid in human pericardium, splenic capsule and trabeculae, plasma, serum, eye sleep extract, and liver sinusoids compared with the epitope in normal synovial tissue and fluid. Therefore, maximal detection of lubricin with S6.79 and potentially other antilubricin mucin domain antibodies15 requires sialidase treatment for optimal antigen retrieval.
Materials and Methods Histochemical and Immunohistochemical Stains Human tissues and fluids from existing pathological specimens were collected with approval from the Institutional Review Board (IRB No. 13052202-IRB01, 2013) of the authors’ institution, with waiver of consent. Representative samples of normal adult pericardium, spleen, eyelid, and esophagus were chosen from cases of surgical resections, and normal adult liver cases were chosen from biopsies. Five-µm-thick sections of formalin-fixed paraffin-embedded tissues were retrieved from the Department of Pathology files (Rush University Medical Center), baked at 65C for 1 hr, deparaffinized in xylene, and rehydrated in a graded ethanol series. Heat-mediated antigen retrieval was performed by incubating the slides at 95C for 20 min in 10-mM Tris, 1-mM EDTA (catalog number: E5134; Sigma, St. Louis, MO), and 0.05% Tween-20 (catalog number: 23336-2500; Acros Organics, NJ), pH 9.0. Moats were drawn around the tissue sections with an ImmEdge pen (catalog number: H-4000; Vector Laboratories, Burlingame, CA). Tissue sections were incubated in 50-mM sodium phosphate buffer, pH 6.0, with or without 0.025 U/ml Sialidase A (Glyko Sialidase A, catalog number: GK80040; ProZyme, Hayward, CA) from Arthrobacter ureafaciens, for 6 hr at 37C in a humidified chamber. Tissue sections were then blocked with 1% (w/v) BSA in 1× PBS, pH 7.4, at 37C for 30 min. Slides were incubated with primary antibody (S6.79 mouse monoclonal; 2 µg/ml) in 0.1% (w/v) BSA in 1× PBS, pH 7.4, for 16 hr at 4C. Each experiment included a positive control of human synovium
Solka et al. (Supplemental Fig. 1) and negative controls without primary antibody for each tissue type (Supplemental Figs. 1–4). Endogenous peroxidase activity was quenched by 0.3% H2O2 in 1X TBS, pH 7.5, for 15 min at room temperature. Slides were incubated with secondary antibody (goat antimouse IgG, 1:1000, catalog number: 31432; Thermo Fisher Scientific, Rockford, IL) in 1× PBS and 0.05% Tween-20, pH 7.4, for 1 hr at room temperature, and developed with 3,3′-diaminobenzidine (DAB) (SigmaFAST DAB tablets, catalog number: D4293; Sigma). To distinguish brown DAB reactivity from hemosiderin when it was present in the tissue, slides were stained using the Perls’ Prussian blue reaction (0.6-M HCl, 0.06-M potassium ferrocyanide) for 20 min at room temperature. Sections were counterstained with hematoxylin (catalog number: 7211; Thermo Fisher Scientific), dehydrated, mounted, and photographed using an Olympus BX41 microscope and camera (model 18.2 Color Mosaic; Diagnostics Instruments, Inc., Olympus, Japan). A combined orcein (Harleco; Hartman-Leddon Co., Philadelphia, PA) and Giemsa (Harleco; HartmanLeddon Co.) staining of elastic fibers was performed according to Pinkus and Hunter.16
Processing of Human Synovial Fluid, Blood, and Eye Sleep Normal synovial fluid was taken from a non-inflamed knee of an adult at autopsy, diluted 1:1 with sterile saline, and centrifuged at 1200 rpm (Rotor Heraeus 7591, Sorvall Biofuge PrimoR) for 15 min at 10C to remove cells. Clarified synovial fluid was treated with 10-mM EDTA, 0.3-mM 1,10-phenanthroline (catalog number: P9375; Sigma), 5-mM benzamidine (catalog number: B6506; Sigma), 10-mM N-ethylmaleimide (NEM; catalog number: E3876; Sigma), 0.5-mM PMSF (catalog number: P7626; Sigma), and 0.02% NaN3 (catalog number: S2002; Sigma) for 16 hr at 4C and stored as aliquots at −80C until analysis. Normal volunteer venous blood was collected by venipuncture into plasma separator tubes (PSTs; BD Vacutainer PST gel and lithium heparin 83 units, reference number: 367962; BD Biosciences, Franklin Lakes, NJ) and serum separator tubes (SSTs; BD Vacutainer SST, reference number: 367986; BD Biosciences). The serum was allowed to clot for 30 min. The PSTs and SSTs were centrifuged at 2000 rpm (Rotor Heraeus 7591, Sorvall Biofuge PrimoR) for 10 min at 10C, and the supernatants (plasma and serum) were collected; brought to 5-mM EDTA, 0.3mM 1,10-phenanthroline, 5-mM benzamidine, 10-mM NEM, 0.5-mM PMSF, and 0.05% NaN3; and stored as aliquots at −80C until analysis.
649
Sialidase unmasks lubricin mucin domain epitopes Normal eye discharge, also known as rheum or sleep, was collected after drying overnight. It was dissolved in 50-mM sodium phosphate buffer, pH 6.0, with 0.01% Tween-20, vortexed, sonicated for 20 min in a water bath sonicator, and centrifuged at 5000 rpm (Beckman Microfuge 12 rotor, Beckman Coulter, Indianapolis, IN, USA) for 10 min at 4C. The supernatant was collected and stored at −80C until analysis.
SDS-PAGE and Western Blotting To determine total protein concentrations, the bicinchoninic acid (BCA) assay (Micro BCA Protein Assay Kit, catalog number: 23235; Thermo Fisher Scientific) was performed according to the manufacturer’s instructions. Human body fluids (synovial fluid, plasma, serum, and eye sleep extract at 0.2-, 5.0-, 5.0-, and 100-µg total protein, respectively) were digested with Sialidase A in 50-mM sodium phosphate buffer, pH 6.0, with 0.01% Tween-20 for 18 hr at 37C, according to the manufacturer’s instructions. Glycine–SDS-PAGE gels for routine protein samples were used, according to Laemmli.17 Samples were boiled at 95C for 5 min in 350-mM dithiothreitol (catalog number: D-9163; Sigma) in 1X Laemmli sample buffer (catalog number: 161-0737; Bio-Rad, Hercules, CA) and electrophoresed on a 4% to 20% acrylamide continuous gradient precast SDS-PAGE gel (catalog number: 456-1094; Bio-Rad). Molecular weight standards (Precision Plus Protein Kaleidoscope Prestained Protein Standards, catalog number: 161-0375; Bio-Rad) were used on each gel. Resolved proteins were transferred to nitrocellulose membranes (0.45-µm; Bio-Rad) in transfer buffer (12-mM Tris, 10-mM sodium acetate, and 15-mM EDTA, at pH 9.5) at 25 V for 16 hr at 4C. Membranes were blocked in 2% (w/v) non-fat dry milk (catalog number: 232100; Becton Dickinson, San Jose, CA) in 1× PBS, pH 7.4, for 30 min at room temperature, incubated with primary antilubricin antibody (S6.79 mouse monoclonal; 2 µg/ml) in 2% (w/v) skim milk in PBS, pH 7.4, for 16 hr at 4C. Membranes were incubated with secondary antibody (goat antimouse IgG, 1:10,000, catalog number: 31432; Thermo Fisher Scientific) in PBS and 0.05% Tween-20, pH 7.4, for 1 hr at room temperature, developed with SuperSignal West Pico chemiluminescent substrate (catalog number: 34077; Thermo Fisher Scientific), and imaged (ChemiDoc XRS+; Bio-Rad). The band volume (pixel density × area) was analyzed by Image Lab software version 4.1 (Bio-Rad), and fold changes were generated by calculating the sialidase-treated lubricin band volume divided by the untreated lubricin band volume. Data presented are representative of three separate sialidase digestions.
Results Desialylation of the Pericardium Enhances Immunohistochemical Detection of Lubricin by the S6.79 Monoclonal Antibody Patients with mutations in the PRG4 gene, which encodes lubricin, lack functional lubricin in their tissues and develop the disorder camptodactyly-arthropathycoxa vara-pericarditis.18 This suggests that lubricin is associated with the pericardium of the heart. We therefore examined the pericardial mesothelium (Fig. 1A and D) for the presence of lubricin. Without sialidase antigen retrieval, lubricin was weakly detected by IHC with S6.79 in the pericardial mesothelium and stroma (Fig. 1B and E). Removal of sialic acid greatly enhanced the detection of lubricin by the S6.79 antibody (Fig. 1C and F). For comparison, human synovium (known for abundant lubricin expression) was also stained. Removal of sialic acid did not enhance the immunohistochemical detection of lubricin in human synovium positive controls (Supplemental Fig. 1). Negative controls omitting primary antibody in all tissues examined showed no staining (Supplemental Figs. 1–4).
Desialylation Reveals Lubricin in the Splenic Capsule and Trabeculae The spleen is surrounded by a capsule, which is composed of a surface lining mesothelial layer and underlying dense fibrous tissue, elastic fibers, and smooth muscle.19 We performed lubricin IHC to evaluate the distribution of lubricin in the spleen. Without sialidase treatment, S6.79 reactivity was barely detectable in the splenic capsule (Fig. 2A) or trabeculae (Fig. 2C). After antigen retrieval with sialidase, lubricin was observed in the splenic capsule (Fig. 2B) and trabeculae (Fig. 2D). Unfortunately, the mesothelial lining of the capsule was not present in our cases, which is not surprising due to the delicate nature of mesothelium. Lubricin was not detected in the red or white splenic pulp (Fig. 2). Thus, the S6.79 antibody epitope of lubricin in the splenic capsule and trabeculae is masked by sialylation.
Desialylation of Lubricin Enhances the Western Blot Reactivity of the S6.79 Monoclonal Antibody Human body fluids containing lubricin (synovial fluid, plasma, serum, and eye sleep extract) were treated with or without sialidase and were assessed for lubricin by Western blot analysis. Removal of sialic acid increased the mobility of lubricin on the blots and enhanced the level of lubricin detected by S6.79 in plasma by 4.4- to 11.3-fold, serum by 3.9- to 13.4-fold,
650
Solka et al.
Figure 1. Lubricin expression in the pericardium. Hematoxylin and eosin staining of a pericardium showing a sloughed (A) or intact (D) monolayer mesothelium lining the pericardium. S6.79 IHC of pericardium without sialidase (B, E) and with sialidase (C, F). Scale bars: A–F = 50 µm.
Figure 2. Lubricin distribution in the splenic capsule and trabeculae. S6.79 IHC without sialidase of the splenic capsule (A) and splenic trabeculae (C), and with sialidase of the splenic capsule (B) and splenic trabeculae (D). Scale bars: A–D = 50 µm.
and in eye sleep extract by 7-fold, but had a minimal effect on levels detected in synovial fluid with a 0.7- to 1.0-fold change (Fig. 3). This strongly suggests that the epitopes recognized by the S6.79 antibody on lubricin molecules in several body fluids have different degrees of sialylation.
Desialylation Reveals Lubricin in Eye Sleep Sialidase treatment of lubricin in eye sleep extracts enhanced the detection of lubricin epitopes by Western blot analysis (Fig. 3). Human eyelid tissue sections contained conjunctiva, meibomian glands, and dermis
651
Sialidase unmasks lubricin mucin domain epitopes
Figure 3. Sialidase enhances the reactivity of the S6.79 monoclonal antibody in human body fluids. S6.79 Western blot analysis of normal human body fluids (SF, plasma, serum, and eye sleep extract) treated with sialidase or buffer control for 18 hr (6.390-sec exposure; ChemiDoc XRS+; Bio-Rad). Abbreviation: SF, synovial fluid.
as evident by hematoxylin and eosin (H&E) staining (Fig. 4A, D, G, and J). Lubricin was weakly detected in some goblet cells of the conjunctiva after sialidase treatment (Fig. 4B vs C). However, even after sialidase treatment, lubricin was not observed in these sections in either meibomian glands or ducts (Fig. 4E and F). In areas of subdermal skin tissue with sun damage– induced elastic degeneration (Fig. 4G, J, M, and N), extracellular lubricin was associated with the elastic fibers (Fig. 4H and K), and the staining was enhanced after sialidase treatment (Fig. 4I and L). Orcein-stained elastic fibers appeared black to dark purple (Fig. 4M and N).16,20 A second case of solar elastosis is shown in Supplemental Fig. 3.
Immunohistochemical Detection of Lubricin in the Blood of Hepatic Sinusoids Is Dependent on Desialylation Lubricin was detected in the connective tissue of liver biopsy sections, but this was not dependent on sialidase (Fig. 5A without sialidase, B with sialidase). Antigen retrieval with sialidase revealed the immunohistochemical detection of lubricin in the plasma of the sinusoids of liver biopsy cases (Fig. 5C without sialidase, D with sialidase). Lubricin was not detected in liver hepatocytes (Fig. 5A–D). Assessment of multiple resected liver specimens demonstrated variable degrees of staining. This staining often appeared to be due to extravasated blood in the specimens associated with tissue damage. To avoid this type of tissue damage artifact, liver biopsies were evaluated as shown in Fig. 5A to D. An example of this strong extracellular reactivity from extravasated blood is shown in esophageal tissue in Fig. 5E and F where the
extravasated red blood cells are obvious in sections after H&E staining (Fig. 5G). Intraluminal lubricin reactivity in blood vessels is shown in Fig. 5E. Tissues shown in Fig. 5E and F were not treated with sialidase. Lubricin was not observed in non-hemorrhagic esophagus (Supplemental Fig. 4).
Discussion The lubricin glycoprotein present in synovial fluid is 50% to 57% carbohydrate.2,7,13,14 Lubricin’s central, long mucin-like domain is extensively O-glycosylated with Gal(β1-3)GalNAc-O-Ser/Thr, and about two thirds of lubricin’s O-glycosylated sites are capped with sialic acid as (α2,3)NeuAc-(β1-3)Gal-GalNAc-O-Ser/ Thr.2,6,7,11–14 Given lubricin’s ~200 O-glycosylated sites,14 there are approximately 133 sialic acid residues per lubricin molecule. Ali et al.14 identified 11 different O-linked glycans attached to lubricin peptides. Our data show that lubricin epitope recognized by the S6.79 monoclonal antibody in the mucin domain is sialylated differently in certain human body fluids and tissues. Furthermore, lubricin molecules in blood express highly sialylated S6.79 epitopes, which suggests that the origin of the majority of lubricin molecules in blood is different from those in synovial fluid. Patients with mutations in the PRG4 gene, which codes for lubricin, lack functional lubricin and develop the disorder camptodactyly-arthropathy-coxa varapericarditis18 and can exhibit pericardial hyperplasia. This suggests that lubricin is also associated with the pericardium. We found that removal of sialic acid greatly enhanced the lubricin reactivity observed within the pericardial mesothelial cells and pericardial stroma. This is the first observation that lubricin is
652
Solka et al.
Editor’s Highlight
(continued)
Sialidase unmasks lubricin mucin domain epitopes
653
Editor’s Highlight
(continued)
654
Solka et al.
Editor’s Highlight
(continued)
Sialidase unmasks lubricin mucin domain epitopes
655
Editor’s Highlight
(continued)
656
Solka et al.
Editor’s Highlight
(continued)
Sialidase unmasks lubricin mucin domain epitopes
657
Editor’s Highlight
(continued)
658
Solka et al.
Editor’s Highlight
(continued)
Sialidase unmasks lubricin mucin domain epitopes
659
Editor’s Highlight
(continued)
660
Solka et al.
Editor’s Highlight
(continued)
Sialidase unmasks lubricin mucin domain epitopes
661
Editor’s Highlight
(continued)
662
Solka et al.
Editor’s Highlight
(continued)
Sialidase unmasks lubricin mucin domain epitopes
663
Editor’s Highlight
(continued)
664
Solka et al.
Editor’s Highlight
(continued)
Sialidase unmasks lubricin mucin domain epitopes
665
Editor’s Highlight
(continued)
666
Solka et al.
Editor’s Highlight Figure 4. Lubricin distribution in the eyelid. Hematoxylin and eosin staining of conjunctiva (A), meibomian glands (D), and solar elastosis of the dermis (G, J) of the eyelid. S6.79 IHC without sialidase of the conjunctiva (B), meibomian glands (E), and the dermis (H, K). S6.79 IHC after sialidase of the conjunctiva (C), meibomian glands (F), and the dermis (I, L). Orcein staining of the elastic fibers in solar elastosis (M, N). Scale bars: A–N = 50 µm.
Figure 5. The distribution of lubricin in the liver sinusoids and an esophageal hemorrhage. S6.79 IHC of liver biopsy cases of connective tissue near a portal tract without sialidase (A) and with sialidase (B) and of liver sinusoids without sialidase (C) and with sialidase (D). S6.79 IHC without sialidase (E, F) and hematoxylin and eosin (G) of esophagus with hemorrhage. Arrows in C and D point to intrasinusoidal space. Arrows in E point to major blood vessels. Scale bars: A–G = 50 µm.
made by mesothelial cells. Because Camptodactylyarthropathy-coxa vara-pericarditis syndrome (CACP) patients who lack functional lubricin sometimes exhibit pericardial hyperplasia, it remains important to investigate lubricin’s expected role in providing boundary lubrication, lowering the coefficient of friction on the heart surface, and preventing pericardial hyperplasia. In addition, removal of sialic acid revealed lubricin in the splenic capsule and trabeculae, which suggests lubricin may reduce the surface friction of the spleen. In plasma, serum, and the blood within hepatic sinusoids, lubricin must be desialylated for the S6.79 antibody to optimally recognize its epitope. Our results have demonstrated that the population of lubricin molecules in blood has highly sialylated S6.79 epitopes. One possible explanation for this apparent higher proportion of sialylated lubricin molecules is that desialylated glycoproteins, known as asialoglycoproteins, can be cleared from the circulating blood by lectin receptors in the liver.21–24 For example, asialoglycoprotein receptors are C-type lectins on the sinusoidal surface of hepatocytes that specifically bind to terminal β-linked galactose or N-acetylgalactosamine (GalNAc) residues on asialoglycoproteins.21–23,25–27 The asialoglycoproteins are then internalized by receptor-mediated endocytosis via the clathrin-coated pit pathway.23,26–33 Thus, sialylation masks the terminal galactose residues, thereby protecting glycoproteins from hepatocyte and
non-parenchymal liver cell internalization and subsequent lysosomal degradation.24,34 Also, lubricin derived from osteoarthritis synovial fluid contains more monosialylated glycans compared with disialylated glycans in rheumatoid arthritis (RA) synovial fluid.13 Given our data in which the S6.79 epitopes on lubricin molecules in normal plasma and serum are more sialylated than those in normal synovial fluid, it is possible that the increased sialylation in RA synovial fluid could be due to an increased presence of blood-derived lubricin. We detected sialidase-dependent enhancement of lubricin in eye sleep extract. In the eyelid, lubricin reactivity was weakly detected in the conjunctiva but was not observed in meibomian glands. The level of lubricin in these human eyelid specimens likely is below the immunohistochemical detection of S6.79 and accumulates in the concentrated eye sleep samples. Also, lubricin reactivity was enhanced after sialidase in areas of subdermal solar elastosis. Schmidt and colleagues35 observed lubricin in human corneal and conjunctival epithelial cells using LPN, an affinity-purified rabbit polyclonal antibody made against a C-terminal peptide corresponding to amino acids #1356–1374 of human lubricin (Thermo Fisher Scientific). We have previously detected the presence of lubricin in occasional cells in sections of rabbit eyelids.36 Because lubricin is of interest as an ocular lubricant,35 further investigation of lubricin in the eye is needed.
667
Sialidase unmasks lubricin mucin domain epitopes Also, lubricin reactivity was enhanced after sialidase in areas of subdermal solar elastosis. Solar elastosis is a degeneration of the elastic tissue of the dermis resulting from long-term exposure to sunlight. Mature elastic fibers are composed of an elastin core surrounded by fibrillin microfibrils, with additional accessory molecules and glycoproteins that facilitate elastic fiber assembly and function.37 Interestingly, neuraminidase 1 (Neu1) is required for elastic fiber assembly; Neu1deficient fibroblasts exhibit impaired elastogenesis that is reversible upon addition of exogenous sialidase. During elastogenesis, the soluble precursor tropoelastin is chaperoned to the cell surface by elastin-binding protein (EBP). Neu1 is thought to facilitate the release of tropoelastin from EBP by exposing galactosugars on adjacent molecules, which interact with the galectin domain of EBP.38 The released, secreted tropoelastin molecules then become insoluble elastin globules after cross-linking by lysyl oxidase and are deposited onto fibrillin microfibrils.37 Our data show that the addition of exogenous Sialidase A enhances the detection of lubricin associated with the elastic fibers in solar elastosis. Lubricin associated with elastic fibers was only seen in solar elastosis; lubricin was not associated with normal elastic fibers, such as those in the aortic media. Two of the antilubricin monoclonal antibodies (S6.79 and S17.109) raised against purified human lubricin and reported by Su and colleagues8 recognize a similar O-linked epitope in synovial fluid that is highly masked by sialic acid in plasma and serum. Furthermore, Ai and colleagues15 recently generated a panel of antihuman lubricin monoclonal antibodies that also recognize an asialo-O-linked epitope in synovium, synovial fluid, and blood. Thus, the asialo-O-linked mucin domain epitopes appear to be prominent epitopes for mouse antibodies raised against human lubricin. The presence of lubricin in blood is a major complication for the assessment of lubricin in different tissues by IHC and requires careful evaluation. Lubricin reactivity was detected in the plasma of blood vessels. Strong extracellular lubricin reactivity in various tissues was often accompanied by extravasated red blood cells and hemorrhage revealed by H&E staining. Unlike liver biopsies, resected livers displayed an artifact of intracellular lubricin reactivity, as the positive cells had compromised membranes and the staining did not obey a logical, reproducible pattern of reactivity (data not shown). While desialylation generally provides optimal S6.79 reactivity, refraining from sialidase antigen retrieval suppresses the detection of lubricin derived from blood using the S6.79 antibody. In summary, sialidase is essential for unmasking lubricin epitopes recognized by several monoclonal antibodies, including S6.79, in some tissues. We recently detected lubricin in several pathological tissues, including tenosynovial giant cell tumors,
myxomas, and ganglion cysts.10 Sialidase did not affect the detection of the S6.79 lubricin epitopes in these tissues or any other tissue investigated previously.10 Our data presented here show that lubricin in human pericardium, splenic capsule and trabeculae, plasma, serum, eye sleep extract, and liver sinusoids required sialidase pretreatment for optimal detection, whereas normal synovial fluid and synovium did not. Acknowledgment K.A.S. thanks Lev Rappoport, MD (Rush University Medical Center), for his technical instruction.
Competing Interests The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: K.A.S. and I.J.M. declare no potential competing interests with respect to the research, authorship, and/or publication of this article. T.M.S. is an inventor on US Patent # 6,720,156, “Superficial zone protein-binding molecules and uses thereof,” April 13, 2004. T.M.S. is the CEO of Cartilube, LLC.
Author Contributions Research concept and design: IJM and TMS. Contribution of essential reagents or tools: IJM and TMS. Collection and/ or assembly of data: KAS, IJM, and TMS. Data analysis and interpretation: KAS, IJM, and TMS. Writing the manuscript: KAS, IJM, and TMS. Design of figures: KAS, IJM, and TMS. Critical revision of manuscript: KAS, IJM, and TMS. Final approval of manuscript: KAS, IJM, and TMS.
Funding The author(s) received no financial support for the research, authorship, and/or publication of this article.
Literature Cited 1. Swann DA, Radin EL. The molecular basis of articular lubrication. I. Purification and properties of a lubricating fraction from bovine synovial fluid. J Biol Chem. 1972;247:8069–73. 2. Swann DA, Sotman S, Dixon M, Brooks C. The isolation and partial characterization of the major glycoprotein (LGP-I) from the articular lubricating fraction from bovine synovial fluid. Biochem J. 1977;161:473–85. 3. Rhee DK, Marcelino J, Baker M, Gong Y, Smits P, Lefebvre V, Jay GD, Stewart M, Wang H, Warman ML, Carpten JD. The secreted glycoprotein lubricin protects cartilage surfaces and inhibits synovial cell overgrowth. J Clin Invest. 2005;115:622–31. 4. Jay GD, Torres JR, Warman ML, Laderer MC, Breuer KS. The role of lubricin in the mechanical behavior of synovial fluid. Proc Natl Acad Sci U S A. 2007;104: 6194–9. 5. Flannery CR, Zollner R, Corcoran C, Jones AR, Root A, Rivera-Bermudez MA, Blanchet T, Gleghorn JP, Bonassar LJ, Bendele AM, Morris EA, Glasson SS. Prevention of cartilage degeneration in a rat model of
668 osteoarthritis by intraarticular treatment with recombinant lubricin. Arthritis Rheum. 2009;60:840–7. 6. Jay GD. Characterization of a bovine synovial fluid lubricating factor. I. Chemical, surface activity and lubricating properties. Connect Tissue Res. 1992;28:71–88. 7. Jay GD, Harris DA, Cha CJ. Boundary lubrication by lubricin is mediated by O-linked beta(1-3)Gal-GalNAc oligosaccharides. Glycoconj J. 2001;18:807–15. 8. Su JL, Schumacher BL, Lindley KM, Soloveychik V, Burkhart W, Triantafillou JA, Kuettner K, Schmid T. Detection of superficial zone protein in human and animal body fluids by cross-species monoclonal antibodies specific to superficial zone protein. Hybridoma. 2001;20:149–57. 9. Solka KA. Abundant lubricin expression suggests a link between synoviocytes, synovial tumors, and myxomas. [dissertation]. Chicago: Rush University. 2015. Available from: ProQuest/UMI, Ann Arbor, MI; AAT 3664085. 10. Solka KA, Schmid TM, Miller IJ. Abundant lubricin expression suggests a link between synoviocytes, synovial tumors, and myxomas. Histol Histopathol. 2016;31(10):1131–41. 11. Swann DA, Slayter HS, Silver FH. The molecular structure of lubricating glycoprotein-I, the boundary lubricant for articular cartilage. J Biol Chem. 1981;256:5921–5. 12. Swann DA, Silver FH, Slayter HS, Stafford W, Shore E. The molecular structure and lubricating activity of lubricin isolated from bovine and human synovial fluids. Biochem J. 1985;225:195–201. 13. Estrella RP, Whitelock JM, Packer NH, Karlsson NG. The glycosylation of human synovial lubricin: implications for its role in inflammation. Biochem J. 2010;429:359–67. 14. Ali L, Flowers SA, Jin C, Bennet EP, Ekwall AK, Karlsson NG. The O-glycomap of lubricin, a novel mucin responsible for joint lubrication, identified by sitespecific glycopeptide analysis. Mol Cell Proteomics. 2014;13:3396–409. 15. Ai M, Cui Y, Sy MS, Lee DM, Zhang LX, Larson KM, Kurek KC, Jay GD, Warman ML. Anti-lubricin monoclonal antibodies created using lubricin-knockout mice immunodetect lubricin in several species and in patients with healthy and diseased joints. PLoS ONE. 2015;10:e0116237. 16. Pinkus H, Hunter R. Simplified acid orcein and Giemsa technique for routine staining of skin sections. Arch Dermatol. 1960;82:699–700. 17. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970;227:680–5. 18. Marcelino J, Carpten JD, Suwairi WM, Gutierrez OM, Schwartz S, Robbins C, Sood R, Makalowska I, Baxevanis A, Johnstone B, Laxer RM, Zemel L, Kim CA, Herd JK, Ihle J, Williams C, Johnson M, Raman V, Alonso LG, Brunoni D, Gerstein A, Papadopoulos N, Bahabri SA, Trent JM, Warman ML. CACP, encoding a secreted proteoglycan, is mutated in camptodactyly-arthropathy-coxa vara-pericarditis syndrome. Nat Genet. 1999;23:319–22. 19. Cesta MF. Normal structure, function, and histology of the spleen. Toxicol Pathol. 2006;34:455–65. 20. Harikumar PE, Selway JL, Chu A, Langlands K. Collagen remodeling and peripheral immune cell recruitment
Solka et al. characterizes the cutaneous Langerhans cell histiocytosis microenvironment. Int J Dermatol. 2015;54:e7–13. 21. Morell AG, Irvine RA, Sternlieb I, Scheinberg IH, Ashwell G. Physical and chemical studies on ceruloplasmin. V. Metabolic studies on sialic acid-free ceruloplasmin in vivo. J Biol Chem. 1968;243:155–9. 22. Morell AG, Gregoriadis G, Scheinberg IH, Hickman J, Ashwell G. The role of sialic acid in determining the survival of glycoproteins in the circulation. J Biol Chem. 1971;246:1461–7. 23. Ashwell G, Harford J. Carbohydrate-specific receptors of the liver. Annu Rev Biochem. 1982;51:531–54. 24. Stefanich EG, Ren S, Danilenko DM, Lim A, Song A, Iyer S, Fielder PJ. Evidence for an asialoglycoprotein receptor on nonparenchymal cells for O-linked glycoproteins. J Pharmacol Exp Ther. 2008;327:308–15. 25. Pricer WE Jr, Hudgin RL, Ashwell G, Stockert RJ, Morell AG. A membrane receptor protein for asialoglycoproteins. Methods Enzymol. 1974;34:688–91. 26. Spiess M. The asialoglycoprotein receptor: a model for endocytic transport receptors. Biochemistry. 1990;29:10009–18. 27. Stockert RJ. The asialoglycoprotein receptor: relationships between structure, function, and expression. Physiol Rev. 1995;75:591–609. 28. Ashwell G, Morell AG. The role of surface carbohy drates in the hepatic recognition and transport of circulating glycoproteins. Adv Enzymol Relat Areas Mol Biol. 1974;41:99–128. 29. Ciechanover A, Schwartz AL, Lodish HF. Sorting and recycling of cell surface receptors and endocytosed ligands: the asialoglycoprotein and transferrin receptors. J Cell Biochem. 1983;23:107–30. 30. Schwartz AL. The hepatic asialoglycoprotein receptor. CRC Crit Rev Biochem. 1984;16:207–233. 31. Breitfeld PP, Simmons CF Jr, Strous GJ, Geuze HJ, Schwartz AL. Cell biology of the asialoglycoprotein receptor system: a model of receptor-mediated endocytosis. Int Rev Cytol. 1985;97:47–95. 32. Geffen I, Spiess M. Asialoglycoprotein receptor. Int Rev Cytol. 1993;137b:181–219. 33. Weigel PH. Endocytosis and function of the hepatic asialoglycoprotein receptor. Subcell Biochem. 1993;19:125–61. 34. Trowbridge IS, Collawn JF, Hopkins CR. Signal dependent membrane protein trafficking in the endocytic pathway. Annu Rev Cell Biol. 1993;9:129–61. 35. Schmidt TA, Sullivan DA, Knop E, Richards SM, Knop N, Liu S, Sahin A, Darabad RR, Morrison S, Kam WR, Sullivan BD. Transcription, translation, and function of lubricin, a boundary lubricant, at the ocular surface. JAMA Ophthalmol. 2013;131:766–76. 36. Cheriyan T, Schmid TM, Spector M. Presence and distribution of the lubricating protein, lubricin, in the meibomian gland in rabbits. Mol Vis. 2011;17:3055–61. 37. Baldwin AK, Simpson A, Steer R, Cain SA, Kielty CM. Elastic fibres in health and disease. Expert Rev Mol Med. 2013;15:e8. 38. Hinek A, Pshezhetsky AV, von Itzstein M, Starcher B. Lysosomal sialidase (neuraminidase-1) is targeted to the cell surface in a multiprotein complex that facilitates elastic fiber assembly. J Biol Chem. 2006;281: 3698–710.
JHCXXX10.1369/0022155416668139Sialidase unmasks lubricin mucin domain epitopesSolka et al.
Article Journal of Histochemistry & Cytochemistry 2016, Vol. 64(11) 647–668 © 2016 The Histochemical Society Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1369/0022155416668139 jhc.sagepub.com
Sialidase Unmasks Mucin Domain Epitopes of Lubricin Kathryn A. Solka, Ira J. Miller, and Thomas M. Schmid
Department of Biochemistry (KAS, TMS) and Department of Pathology (IJM), Rush University Medical Center, Chicago, Illinois
Summary Lubricin is a secreted, mucin-like glycoprotein and proteoglycan abundant in synovial fluid that provides boundary lubrication and prevents cell adhesion in synovial joints. The antilubricin S6.79 monoclonal antibody recognizes an O-linked glycopeptide epitope in lubricin’s mucin domain. The central, long mucin domain of lubricin is extensively O-glycosylated with Gal(β1-3)GalNAc-O-Ser/Thr, and about two thirds of the O-glycosylated sites are capped with sialic acid. Our aim was to determine whether removal of sialic acid by sialidase could improve the detection of lubricin in a number of human tissues using the S6.79 monoclonal antibody. Sialidase treatment caused a dramatic increase in antibody reactivity in human pericardium, splenic capsule and trabeculae, plasma, serum, eye sleep extract, and liver sinusoids. Sialidase had minimal effect on S6.79 antibody reactivity with lubricin in synovial fluid and synovial tissue. These observations suggest that the origin of lubricin in blood may be different from that in synovial fluid and that desialylation of lubricin is essential for unmasking epitopes within the mucin domain. (J Histochem Cytochem 64:647–668, 2016) Keywords epitopes, eyelid, liver, lubricin, neuraminidase, pericardium, sialidase, spleen, synovial fluid
Introduction The lubricin gene, proteoglycan 4 (PRG4), encodes for a heterogeneous population of proteins that differ by alternative splicing, glycosylation, function, and tissue distribution. Lubricin was first isolated and characterized as the major lubricating glycoprotein in synovial fluid.1,2 Lubricin generates a friction-reducing, boundary-lubricating antiadhesive coating on the surface of articular cartilage, which prevents adhesion of synovial cells, chondrocytes, and protein, thereby protecting joint tissues from wear and breakdown.3–5 Lubricin also prevents synovial cell hyperplasia.3 Lubricin’s extensive O-linked glycosylation is essential for its function. Desialylation of lubricin does not affect its lubricating activity; however, subsequent removal of 54% of lubricin’s O-linked carbohydrates increases lubricin’s coefficient of friction by nearly 80%, thereby decreasing its ability to lubricate.6,7 We have generated antihuman lubricin monoclonal antibodies.8 S6.79, one of the highest affinity antibodies,
recognizes an O-linked glycopeptide epitope repeated a few times in the mucin domain.9 In a recent immunohistochemical study with this antibody, we detected lubricin in several pathological tissues including tenosynovial giant cell tumors, myxomas, and ganglion cysts.10 In our survey of more than 60 human tissues, S6.79 reactivity was highly restricted to only a few tissues. The antibody did not stain proteoglycan-rich tissues such as umbilical cord and fetal tissues. In competition ELISAs, it did not react with aggrecan,8 decorin, and biglycan (data not shown). The S6.79 antibody did not detect versican by Western blots of partially purified material from bovine aorta, nor did it react with human or bovine aortic media by IHC. The S6.79 antibody is nonreactive with versican Received for publication March 14, 2016; accepted August 11, 2016. Corresponding Author: Thomas M. Schmid, Department of Biochemistry, Rush University Medical Center, 1735 W. Harrison St., Cohn Research Building, Suite 524, Chicago, IL 60612, USA. E-mail: [email protected]
Journal of Histochemistry and Cytochemistry
668139
648 either with or without chondroitinase ABC treatment (data not shown). The S6.79 antibody is highly specific for O-linked glycopeptides within the mucin domain of lubricin.9 The central, long mucin domain of lubricin is extensively O-glycosylated with Gal(β1-3)GalNAc-OSer/Thr, and about two thirds of the O-glycosylated sites are capped with sialic acid as (α2,3)NeuAc-(β1-3)GalGalNAc-O-Ser/Thr.2,6,7,11–14 Our aim was to determine whether removal of sialic acid by sialidase could enhance the detection of lubricin with the S6.79 antibody in human tissues by IHC and Western blot analysis. Our data show that the S6.79 lubricin epitope is highly masked by sialic acid in human pericardium, splenic capsule and trabeculae, plasma, serum, eye sleep extract, and liver sinusoids compared with the epitope in normal synovial tissue and fluid. Therefore, maximal detection of lubricin with S6.79 and potentially other antilubricin mucin domain antibodies15 requires sialidase treatment for optimal antigen retrieval.
Materials and Methods Histochemical and Immunohistochemical Stains Human tissues and fluids from existing pathological specimens were collected with approval from the Institutional Review Board (IRB No. 13052202-IRB01, 2013) of the authors’ institution, with waiver of consent. Representative samples of normal adult pericardium, spleen, eyelid, and esophagus were chosen from cases of surgical resections, and normal adult liver cases were chosen from biopsies. Five-µm-thick sections of formalin-fixed paraffin-embedded tissues were retrieved from the Department of Pathology files (Rush University Medical Center), baked at 65C for 1 hr, deparaffinized in xylene, and rehydrated in a graded ethanol series. Heat-mediated antigen retrieval was performed by incubating the slides at 95C for 20 min in 10-mM Tris, 1-mM EDTA (catalog number: E5134; Sigma, St. Louis, MO), and 0.05% Tween-20 (catalog number: 23336-2500; Acros Organics, NJ), pH 9.0. Moats were drawn around the tissue sections with an ImmEdge pen (catalog number: H-4000; Vector Laboratories, Burlingame, CA). Tissue sections were incubated in 50-mM sodium phosphate buffer, pH 6.0, with or without 0.025 U/ml Sialidase A (Glyko Sialidase A, catalog number: GK80040; ProZyme, Hayward, CA) from Arthrobacter ureafaciens, for 6 hr at 37C in a humidified chamber. Tissue sections were then blocked with 1% (w/v) BSA in 1× PBS, pH 7.4, at 37C for 30 min. Slides were incubated with primary antibody (S6.79 mouse monoclonal; 2 µg/ml) in 0.1% (w/v) BSA in 1× PBS, pH 7.4, for 16 hr at 4C. Each experiment included a positive control of human synovium
Solka et al. (Supplemental Fig. 1) and negative controls without primary antibody for each tissue type (Supplemental Figs. 1–4). Endogenous peroxidase activity was quenched by 0.3% H2O2 in 1X TBS, pH 7.5, for 15 min at room temperature. Slides were incubated with secondary antibody (goat antimouse IgG, 1:1000, catalog number: 31432; Thermo Fisher Scientific, Rockford, IL) in 1× PBS and 0.05% Tween-20, pH 7.4, for 1 hr at room temperature, and developed with 3,3′-diaminobenzidine (DAB) (SigmaFAST DAB tablets, catalog number: D4293; Sigma). To distinguish brown DAB reactivity from hemosiderin when it was present in the tissue, slides were stained using the Perls’ Prussian blue reaction (0.6-M HCl, 0.06-M potassium ferrocyanide) for 20 min at room temperature. Sections were counterstained with hematoxylin (catalog number: 7211; Thermo Fisher Scientific), dehydrated, mounted, and photographed using an Olympus BX41 microscope and camera (model 18.2 Color Mosaic; Diagnostics Instruments, Inc., Olympus, Japan). A combined orcein (Harleco; Hartman-Leddon Co., Philadelphia, PA) and Giemsa (Harleco; HartmanLeddon Co.) staining of elastic fibers was performed according to Pinkus and Hunter.16
Processing of Human Synovial Fluid, Blood, and Eye Sleep Normal synovial fluid was taken from a non-inflamed knee of an adult at autopsy, diluted 1:1 with sterile saline, and centrifuged at 1200 rpm (Rotor Heraeus 7591, Sorvall Biofuge PrimoR) for 15 min at 10C to remove cells. Clarified synovial fluid was treated with 10-mM EDTA, 0.3-mM 1,10-phenanthroline (catalog number: P9375; Sigma), 5-mM benzamidine (catalog number: B6506; Sigma), 10-mM N-ethylmaleimide (NEM; catalog number: E3876; Sigma), 0.5-mM PMSF (catalog number: P7626; Sigma), and 0.02% NaN3 (catalog number: S2002; Sigma) for 16 hr at 4C and stored as aliquots at −80C until analysis. Normal volunteer venous blood was collected by venipuncture into plasma separator tubes (PSTs; BD Vacutainer PST gel and lithium heparin 83 units, reference number: 367962; BD Biosciences, Franklin Lakes, NJ) and serum separator tubes (SSTs; BD Vacutainer SST, reference number: 367986; BD Biosciences). The serum was allowed to clot for 30 min. The PSTs and SSTs were centrifuged at 2000 rpm (Rotor Heraeus 7591, Sorvall Biofuge PrimoR) for 10 min at 10C, and the supernatants (plasma and serum) were collected; brought to 5-mM EDTA, 0.3mM 1,10-phenanthroline, 5-mM benzamidine, 10-mM NEM, 0.5-mM PMSF, and 0.05% NaN3; and stored as aliquots at −80C until analysis.
649
Sialidase unmasks lubricin mucin domain epitopes Normal eye discharge, also known as rheum or sleep, was collected after drying overnight. It was dissolved in 50-mM sodium phosphate buffer, pH 6.0, with 0.01% Tween-20, vortexed, sonicated for 20 min in a water bath sonicator, and centrifuged at 5000 rpm (Beckman Microfuge 12 rotor, Beckman Coulter, Indianapolis, IN, USA) for 10 min at 4C. The supernatant was collected and stored at −80C until analysis.
SDS-PAGE and Western Blotting To determine total protein concentrations, the bicinchoninic acid (BCA) assay (Micro BCA Protein Assay Kit, catalog number: 23235; Thermo Fisher Scientific) was performed according to the manufacturer’s instructions. Human body fluids (synovial fluid, plasma, serum, and eye sleep extract at 0.2-, 5.0-, 5.0-, and 100-µg total protein, respectively) were digested with Sialidase A in 50-mM sodium phosphate buffer, pH 6.0, with 0.01% Tween-20 for 18 hr at 37C, according to the manufacturer’s instructions. Glycine–SDS-PAGE gels for routine protein samples were used, according to Laemmli.17 Samples were boiled at 95C for 5 min in 350-mM dithiothreitol (catalog number: D-9163; Sigma) in 1X Laemmli sample buffer (catalog number: 161-0737; Bio-Rad, Hercules, CA) and electrophoresed on a 4% to 20% acrylamide continuous gradient precast SDS-PAGE gel (catalog number: 456-1094; Bio-Rad). Molecular weight standards (Precision Plus Protein Kaleidoscope Prestained Protein Standards, catalog number: 161-0375; Bio-Rad) were used on each gel. Resolved proteins were transferred to nitrocellulose membranes (0.45-µm; Bio-Rad) in transfer buffer (12-mM Tris, 10-mM sodium acetate, and 15-mM EDTA, at pH 9.5) at 25 V for 16 hr at 4C. Membranes were blocked in 2% (w/v) non-fat dry milk (catalog number: 232100; Becton Dickinson, San Jose, CA) in 1× PBS, pH 7.4, for 30 min at room temperature, incubated with primary antilubricin antibody (S6.79 mouse monoclonal; 2 µg/ml) in 2% (w/v) skim milk in PBS, pH 7.4, for 16 hr at 4C. Membranes were incubated with secondary antibody (goat antimouse IgG, 1:10,000, catalog number: 31432; Thermo Fisher Scientific) in PBS and 0.05% Tween-20, pH 7.4, for 1 hr at room temperature, developed with SuperSignal West Pico chemiluminescent substrate (catalog number: 34077; Thermo Fisher Scientific), and imaged (ChemiDoc XRS+; Bio-Rad). The band volume (pixel density × area) was analyzed by Image Lab software version 4.1 (Bio-Rad), and fold changes were generated by calculating the sialidase-treated lubricin band volume divided by the untreated lubricin band volume. Data presented are representative of three separate sialidase digestions.
Results Desialylation of the Pericardium Enhances Immunohistochemical Detection of Lubricin by the S6.79 Monoclonal Antibody Patients with mutations in the PRG4 gene, which encodes lubricin, lack functional lubricin in their tissues and develop the disorder camptodactyly-arthropathycoxa vara-pericarditis.18 This suggests that lubricin is associated with the pericardium of the heart. We therefore examined the pericardial mesothelium (Fig. 1A and D) for the presence of lubricin. Without sialidase antigen retrieval, lubricin was weakly detected by IHC with S6.79 in the pericardial mesothelium and stroma (Fig. 1B and E). Removal of sialic acid greatly enhanced the detection of lubricin by the S6.79 antibody (Fig. 1C and F). For comparison, human synovium (known for abundant lubricin expression) was also stained. Removal of sialic acid did not enhance the immunohistochemical detection of lubricin in human synovium positive controls (Supplemental Fig. 1). Negative controls omitting primary antibody in all tissues examined showed no staining (Supplemental Figs. 1–4).
Desialylation Reveals Lubricin in the Splenic Capsule and Trabeculae The spleen is surrounded by a capsule, which is composed of a surface lining mesothelial layer and underlying dense fibrous tissue, elastic fibers, and smooth muscle.19 We performed lubricin IHC to evaluate the distribution of lubricin in the spleen. Without sialidase treatment, S6.79 reactivity was barely detectable in the splenic capsule (Fig. 2A) or trabeculae (Fig. 2C). After antigen retrieval with sialidase, lubricin was observed in the splenic capsule (Fig. 2B) and trabeculae (Fig. 2D). Unfortunately, the mesothelial lining of the capsule was not present in our cases, which is not surprising due to the delicate nature of mesothelium. Lubricin was not detected in the red or white splenic pulp (Fig. 2). Thus, the S6.79 antibody epitope of lubricin in the splenic capsule and trabeculae is masked by sialylation.
Desialylation of Lubricin Enhances the Western Blot Reactivity of the S6.79 Monoclonal Antibody Human body fluids containing lubricin (synovial fluid, plasma, serum, and eye sleep extract) were treated with or without sialidase and were assessed for lubricin by Western blot analysis. Removal of sialic acid increased the mobility of lubricin on the blots and enhanced the level of lubricin detected by S6.79 in plasma by 4.4- to 11.3-fold, serum by 3.9- to 13.4-fold,
650
Solka et al.
Figure 1. Lubricin expression in the pericardium. Hematoxylin and eosin staining of a pericardium showing a sloughed (A) or intact (D) monolayer mesothelium lining the pericardium. S6.79 IHC of pericardium without sialidase (B, E) and with sialidase (C, F). Scale bars: A–F = 50 µm.
Figure 2. Lubricin distribution in the splenic capsule and trabeculae. S6.79 IHC without sialidase of the splenic capsule (A) and splenic trabeculae (C), and with sialidase of the splenic capsule (B) and splenic trabeculae (D). Scale bars: A–D = 50 µm.
and in eye sleep extract by 7-fold, but had a minimal effect on levels detected in synovial fluid with a 0.7- to 1.0-fold change (Fig. 3). This strongly suggests that the epitopes recognized by the S6.79 antibody on lubricin molecules in several body fluids have different degrees of sialylation.
Desialylation Reveals Lubricin in Eye Sleep Sialidase treatment of lubricin in eye sleep extracts enhanced the detection of lubricin epitopes by Western blot analysis (Fig. 3). Human eyelid tissue sections contained conjunctiva, meibomian glands, and dermis
651
Sialidase unmasks lubricin mucin domain epitopes
Figure 3. Sialidase enhances the reactivity of the S6.79 monoclonal antibody in human body fluids. S6.79 Western blot analysis of normal human body fluids (SF, plasma, serum, and eye sleep extract) treated with sialidase or buffer control for 18 hr (6.390-sec exposure; ChemiDoc XRS+; Bio-Rad). Abbreviation: SF, synovial fluid.
as evident by hematoxylin and eosin (H&E) staining (Fig. 4A, D, G, and J). Lubricin was weakly detected in some goblet cells of the conjunctiva after sialidase treatment (Fig. 4B vs C). However, even after sialidase treatment, lubricin was not observed in these sections in either meibomian glands or ducts (Fig. 4E and F). In areas of subdermal skin tissue with sun damage– induced elastic degeneration (Fig. 4G, J, M, and N), extracellular lubricin was associated with the elastic fibers (Fig. 4H and K), and the staining was enhanced after sialidase treatment (Fig. 4I and L). Orcein-stained elastic fibers appeared black to dark purple (Fig. 4M and N).16,20 A second case of solar elastosis is shown in Supplemental Fig. 3.
Immunohistochemical Detection of Lubricin in the Blood of Hepatic Sinusoids Is Dependent on Desialylation Lubricin was detected in the connective tissue of liver biopsy sections, but this was not dependent on sialidase (Fig. 5A without sialidase, B with sialidase). Antigen retrieval with sialidase revealed the immunohistochemical detection of lubricin in the plasma of the sinusoids of liver biopsy cases (Fig. 5C without sialidase, D with sialidase). Lubricin was not detected in liver hepatocytes (Fig. 5A–D). Assessment of multiple resected liver specimens demonstrated variable degrees of staining. This staining often appeared to be due to extravasated blood in the specimens associated with tissue damage. To avoid this type of tissue damage artifact, liver biopsies were evaluated as shown in Fig. 5A to D. An example of this strong extracellular reactivity from extravasated blood is shown in esophageal tissue in Fig. 5E and F where the
extravasated red blood cells are obvious in sections after H&E staining (Fig. 5G). Intraluminal lubricin reactivity in blood vessels is shown in Fig. 5E. Tissues shown in Fig. 5E and F were not treated with sialidase. Lubricin was not observed in non-hemorrhagic esophagus (Supplemental Fig. 4).
Discussion The lubricin glycoprotein present in synovial fluid is 50% to 57% carbohydrate.2,7,13,14 Lubricin’s central, long mucin-like domain is extensively O-glycosylated with Gal(β1-3)GalNAc-O-Ser/Thr, and about two thirds of lubricin’s O-glycosylated sites are capped with sialic acid as (α2,3)NeuAc-(β1-3)Gal-GalNAc-O-Ser/ Thr.2,6,7,11–14 Given lubricin’s ~200 O-glycosylated sites,14 there are approximately 133 sialic acid residues per lubricin molecule. Ali et al.14 identified 11 different O-linked glycans attached to lubricin peptides. Our data show that lubricin epitope recognized by the S6.79 monoclonal antibody in the mucin domain is sialylated differently in certain human body fluids and tissues. Furthermore, lubricin molecules in blood express highly sialylated S6.79 epitopes, which suggests that the origin of the majority of lubricin molecules in blood is different from those in synovial fluid. Patients with mutations in the PRG4 gene, which codes for lubricin, lack functional lubricin and develop the disorder camptodactyly-arthropathy-coxa varapericarditis18 and can exhibit pericardial hyperplasia. This suggests that lubricin is also associated with the pericardium. We found that removal of sialic acid greatly enhanced the lubricin reactivity observed within the pericardial mesothelial cells and pericardial stroma. This is the first observation that lubricin is
652
Solka et al.
Editor’s Highlight
(continued)
Sialidase unmasks lubricin mucin domain epitopes
653
Editor’s Highlight
(continued)
654
Solka et al.
Editor’s Highlight
(continued)
Sialidase unmasks lubricin mucin domain epitopes
655
Editor’s Highlight
(continued)
656
Solka et al.
Editor’s Highlight
(continued)
Sialidase unmasks lubricin mucin domain epitopes
657
Editor’s Highlight
(continued)
658
Solka et al.
Editor’s Highlight
(continued)
Sialidase unmasks lubricin mucin domain epitopes
659
Editor’s Highlight
(continued)
660
Solka et al.
Editor’s Highlight
(continued)
Sialidase unmasks lubricin mucin domain epitopes
661
Editor’s Highlight
(continued)
662
Solka et al.
Editor’s Highlight
(continued)
Sialidase unmasks lubricin mucin domain epitopes
663
Editor’s Highlight
(continued)
664
Solka et al.
Editor’s Highlight
(continued)
Sialidase unmasks lubricin mucin domain epitopes
665
Editor’s Highlight
(continued)
666
Solka et al.
Editor’s Highlight Figure 4. Lubricin distribution in the eyelid. Hematoxylin and eosin staining of conjunctiva (A), meibomian glands (D), and solar elastosis of the dermis (G, J) of the eyelid. S6.79 IHC without sialidase of the conjunctiva (B), meibomian glands (E), and the dermis (H, K). S6.79 IHC after sialidase of the conjunctiva (C), meibomian glands (F), and the dermis (I, L). Orcein staining of the elastic fibers in solar elastosis (M, N). Scale bars: A–N = 50 µm.
Figure 5. The distribution of lubricin in the liver sinusoids and an esophageal hemorrhage. S6.79 IHC of liver biopsy cases of connective tissue near a portal tract without sialidase (A) and with sialidase (B) and of liver sinusoids without sialidase (C) and with sialidase (D). S6.79 IHC without sialidase (E, F) and hematoxylin and eosin (G) of esophagus with hemorrhage. Arrows in C and D point to intrasinusoidal space. Arrows in E point to major blood vessels. Scale bars: A–G = 50 µm.
made by mesothelial cells. Because Camptodactylyarthropathy-coxa vara-pericarditis syndrome (CACP) patients who lack functional lubricin sometimes exhibit pericardial hyperplasia, it remains important to investigate lubricin’s expected role in providing boundary lubrication, lowering the coefficient of friction on the heart surface, and preventing pericardial hyperplasia. In addition, removal of sialic acid revealed lubricin in the splenic capsule and trabeculae, which suggests lubricin may reduce the surface friction of the spleen. In plasma, serum, and the blood within hepatic sinusoids, lubricin must be desialylated for the S6.79 antibody to optimally recognize its epitope. Our results have demonstrated that the population of lubricin molecules in blood has highly sialylated S6.79 epitopes. One possible explanation for this apparent higher proportion of sialylated lubricin molecules is that desialylated glycoproteins, known as asialoglycoproteins, can be cleared from the circulating blood by lectin receptors in the liver.21–24 For example, asialoglycoprotein receptors are C-type lectins on the sinusoidal surface of hepatocytes that specifically bind to terminal β-linked galactose or N-acetylgalactosamine (GalNAc) residues on asialoglycoproteins.21–23,25–27 The asialoglycoproteins are then internalized by receptor-mediated endocytosis via the clathrin-coated pit pathway.23,26–33 Thus, sialylation masks the terminal galactose residues, thereby protecting glycoproteins from hepatocyte and
non-parenchymal liver cell internalization and subsequent lysosomal degradation.24,34 Also, lubricin derived from osteoarthritis synovial fluid contains more monosialylated glycans compared with disialylated glycans in rheumatoid arthritis (RA) synovial fluid.13 Given our data in which the S6.79 epitopes on lubricin molecules in normal plasma and serum are more sialylated than those in normal synovial fluid, it is possible that the increased sialylation in RA synovial fluid could be due to an increased presence of blood-derived lubricin. We detected sialidase-dependent enhancement of lubricin in eye sleep extract. In the eyelid, lubricin reactivity was weakly detected in the conjunctiva but was not observed in meibomian glands. The level of lubricin in these human eyelid specimens likely is below the immunohistochemical detection of S6.79 and accumulates in the concentrated eye sleep samples. Also, lubricin reactivity was enhanced after sialidase in areas of subdermal solar elastosis. Schmidt and colleagues35 observed lubricin in human corneal and conjunctival epithelial cells using LPN, an affinity-purified rabbit polyclonal antibody made against a C-terminal peptide corresponding to amino acids #1356–1374 of human lubricin (Thermo Fisher Scientific). We have previously detected the presence of lubricin in occasional cells in sections of rabbit eyelids.36 Because lubricin is of interest as an ocular lubricant,35 further investigation of lubricin in the eye is needed.
667
Sialidase unmasks lubricin mucin domain epitopes Also, lubricin reactivity was enhanced after sialidase in areas of subdermal solar elastosis. Solar elastosis is a degeneration of the elastic tissue of the dermis resulting from long-term exposure to sunlight. Mature elastic fibers are composed of an elastin core surrounded by fibrillin microfibrils, with additional accessory molecules and glycoproteins that facilitate elastic fiber assembly and function.37 Interestingly, neuraminidase 1 (Neu1) is required for elastic fiber assembly; Neu1deficient fibroblasts exhibit impaired elastogenesis that is reversible upon addition of exogenous sialidase. During elastogenesis, the soluble precursor tropoelastin is chaperoned to the cell surface by elastin-binding protein (EBP). Neu1 is thought to facilitate the release of tropoelastin from EBP by exposing galactosugars on adjacent molecules, which interact with the galectin domain of EBP.38 The released, secreted tropoelastin molecules then become insoluble elastin globules after cross-linking by lysyl oxidase and are deposited onto fibrillin microfibrils.37 Our data show that the addition of exogenous Sialidase A enhances the detection of lubricin associated with the elastic fibers in solar elastosis. Lubricin associated with elastic fibers was only seen in solar elastosis; lubricin was not associated with normal elastic fibers, such as those in the aortic media. Two of the antilubricin monoclonal antibodies (S6.79 and S17.109) raised against purified human lubricin and reported by Su and colleagues8 recognize a similar O-linked epitope in synovial fluid that is highly masked by sialic acid in plasma and serum. Furthermore, Ai and colleagues15 recently generated a panel of antihuman lubricin monoclonal antibodies that also recognize an asialo-O-linked epitope in synovium, synovial fluid, and blood. Thus, the asialo-O-linked mucin domain epitopes appear to be prominent epitopes for mouse antibodies raised against human lubricin. The presence of lubricin in blood is a major complication for the assessment of lubricin in different tissues by IHC and requires careful evaluation. Lubricin reactivity was detected in the plasma of blood vessels. Strong extracellular lubricin reactivity in various tissues was often accompanied by extravasated red blood cells and hemorrhage revealed by H&E staining. Unlike liver biopsies, resected livers displayed an artifact of intracellular lubricin reactivity, as the positive cells had compromised membranes and the staining did not obey a logical, reproducible pattern of reactivity (data not shown). While desialylation generally provides optimal S6.79 reactivity, refraining from sialidase antigen retrieval suppresses the detection of lubricin derived from blood using the S6.79 antibody. In summary, sialidase is essential for unmasking lubricin epitopes recognized by several monoclonal antibodies, including S6.79, in some tissues. We recently detected lubricin in several pathological tissues, including tenosynovial giant cell tumors,
myxomas, and ganglion cysts.10 Sialidase did not affect the detection of the S6.79 lubricin epitopes in these tissues or any other tissue investigated previously.10 Our data presented here show that lubricin in human pericardium, splenic capsule and trabeculae, plasma, serum, eye sleep extract, and liver sinusoids required sialidase pretreatment for optimal detection, whereas normal synovial fluid and synovium did not. Acknowledgment K.A.S. thanks Lev Rappoport, MD (Rush University Medical Center), for his technical instruction.
Competing Interests The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: K.A.S. and I.J.M. declare no potential competing interests with respect to the research, authorship, and/or publication of this article. T.M.S. is an inventor on US Patent # 6,720,156, “Superficial zone protein-binding molecules and uses thereof,” April 13, 2004. T.M.S. is the CEO of Cartilube, LLC.
Author Contributions Research concept and design: IJM and TMS. Contribution of essential reagents or tools: IJM and TMS. Collection and/ or assembly of data: KAS, IJM, and TMS. Data analysis and interpretation: KAS, IJM, and TMS. Writing the manuscript: KAS, IJM, and TMS. Design of figures: KAS, IJM, and TMS. Critical revision of manuscript: KAS, IJM, and TMS. Final approval of manuscript: KAS, IJM, and TMS.
Funding The author(s) received no financial support for the research, authorship, and/or publication of this article.
Literature Cited 1. Swann DA, Radin EL. The molecular basis of articular lubrication. I. Purification and properties of a lubricating fraction from bovine synovial fluid. J Biol Chem. 1972;247:8069–73. 2. Swann DA, Sotman S, Dixon M, Brooks C. The isolation and partial characterization of the major glycoprotein (LGP-I) from the articular lubricating fraction from bovine synovial fluid. Biochem J. 1977;161:473–85. 3. Rhee DK, Marcelino J, Baker M, Gong Y, Smits P, Lefebvre V, Jay GD, Stewart M, Wang H, Warman ML, Carpten JD. The secreted glycoprotein lubricin protects cartilage surfaces and inhibits synovial cell overgrowth. J Clin Invest. 2005;115:622–31. 4. Jay GD, Torres JR, Warman ML, Laderer MC, Breuer KS. The role of lubricin in the mechanical behavior of synovial fluid. Proc Natl Acad Sci U S A. 2007;104: 6194–9. 5. Flannery CR, Zollner R, Corcoran C, Jones AR, Root A, Rivera-Bermudez MA, Blanchet T, Gleghorn JP, Bonassar LJ, Bendele AM, Morris EA, Glasson SS. Prevention of cartilage degeneration in a rat model of
668 osteoarthritis by intraarticular treatment with recombinant lubricin. Arthritis Rheum. 2009;60:840–7. 6. Jay GD. Characterization of a bovine synovial fluid lubricating factor. I. Chemical, surface activity and lubricating properties. Connect Tissue Res. 1992;28:71–88. 7. Jay GD, Harris DA, Cha CJ. Boundary lubrication by lubricin is mediated by O-linked beta(1-3)Gal-GalNAc oligosaccharides. Glycoconj J. 2001;18:807–15. 8. Su JL, Schumacher BL, Lindley KM, Soloveychik V, Burkhart W, Triantafillou JA, Kuettner K, Schmid T. Detection of superficial zone protein in human and animal body fluids by cross-species monoclonal antibodies specific to superficial zone protein. Hybridoma. 2001;20:149–57. 9. Solka KA. Abundant lubricin expression suggests a link between synoviocytes, synovial tumors, and myxomas. [dissertation]. Chicago: Rush University. 2015. Available from: ProQuest/UMI, Ann Arbor, MI; AAT 3664085. 10. Solka KA, Schmid TM, Miller IJ. Abundant lubricin expression suggests a link between synoviocytes, synovial tumors, and myxomas. Histol Histopathol. 2016;31(10):1131–41. 11. Swann DA, Slayter HS, Silver FH. The molecular structure of lubricating glycoprotein-I, the boundary lubricant for articular cartilage. J Biol Chem. 1981;256:5921–5. 12. Swann DA, Silver FH, Slayter HS, Stafford W, Shore E. The molecular structure and lubricating activity of lubricin isolated from bovine and human synovial fluids. Biochem J. 1985;225:195–201. 13. Estrella RP, Whitelock JM, Packer NH, Karlsson NG. The glycosylation of human synovial lubricin: implications for its role in inflammation. Biochem J. 2010;429:359–67. 14. Ali L, Flowers SA, Jin C, Bennet EP, Ekwall AK, Karlsson NG. The O-glycomap of lubricin, a novel mucin responsible for joint lubrication, identified by sitespecific glycopeptide analysis. Mol Cell Proteomics. 2014;13:3396–409. 15. Ai M, Cui Y, Sy MS, Lee DM, Zhang LX, Larson KM, Kurek KC, Jay GD, Warman ML. Anti-lubricin monoclonal antibodies created using lubricin-knockout mice immunodetect lubricin in several species and in patients with healthy and diseased joints. PLoS ONE. 2015;10:e0116237. 16. Pinkus H, Hunter R. Simplified acid orcein and Giemsa technique for routine staining of skin sections. Arch Dermatol. 1960;82:699–700. 17. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970;227:680–5. 18. Marcelino J, Carpten JD, Suwairi WM, Gutierrez OM, Schwartz S, Robbins C, Sood R, Makalowska I, Baxevanis A, Johnstone B, Laxer RM, Zemel L, Kim CA, Herd JK, Ihle J, Williams C, Johnson M, Raman V, Alonso LG, Brunoni D, Gerstein A, Papadopoulos N, Bahabri SA, Trent JM, Warman ML. CACP, encoding a secreted proteoglycan, is mutated in camptodactyly-arthropathy-coxa vara-pericarditis syndrome. Nat Genet. 1999;23:319–22. 19. Cesta MF. Normal structure, function, and histology of the spleen. Toxicol Pathol. 2006;34:455–65. 20. Harikumar PE, Selway JL, Chu A, Langlands K. Collagen remodeling and peripheral immune cell recruitment
Solka et al. characterizes the cutaneous Langerhans cell histiocytosis microenvironment. Int J Dermatol. 2015;54:e7–13. 21. Morell AG, Irvine RA, Sternlieb I, Scheinberg IH, Ashwell G. Physical and chemical studies on ceruloplasmin. V. Metabolic studies on sialic acid-free ceruloplasmin in vivo. J Biol Chem. 1968;243:155–9. 22. Morell AG, Gregoriadis G, Scheinberg IH, Hickman J, Ashwell G. The role of sialic acid in determining the survival of glycoproteins in the circulation. J Biol Chem. 1971;246:1461–7. 23. Ashwell G, Harford J. Carbohydrate-specific receptors of the liver. Annu Rev Biochem. 1982;51:531–54. 24. Stefanich EG, Ren S, Danilenko DM, Lim A, Song A, Iyer S, Fielder PJ. Evidence for an asialoglycoprotein receptor on nonparenchymal cells for O-linked glycoproteins. J Pharmacol Exp Ther. 2008;327:308–15. 25. Pricer WE Jr, Hudgin RL, Ashwell G, Stockert RJ, Morell AG. A membrane receptor protein for asialoglycoproteins. Methods Enzymol. 1974;34:688–91. 26. Spiess M. The asialoglycoprotein receptor: a model for endocytic transport receptors. Biochemistry. 1990;29:10009–18. 27. Stockert RJ. The asialoglycoprotein receptor: relationships between structure, function, and expression. Physiol Rev. 1995;75:591–609. 28. Ashwell G, Morell AG. The role of surface carbohy drates in the hepatic recognition and transport of circulating glycoproteins. Adv Enzymol Relat Areas Mol Biol. 1974;41:99–128. 29. Ciechanover A, Schwartz AL, Lodish HF. Sorting and recycling of cell surface receptors and endocytosed ligands: the asialoglycoprotein and transferrin receptors. J Cell Biochem. 1983;23:107–30. 30. Schwartz AL. The hepatic asialoglycoprotein receptor. CRC Crit Rev Biochem. 1984;16:207–233. 31. Breitfeld PP, Simmons CF Jr, Strous GJ, Geuze HJ, Schwartz AL. Cell biology of the asialoglycoprotein receptor system: a model of receptor-mediated endocytosis. Int Rev Cytol. 1985;97:47–95. 32. Geffen I, Spiess M. Asialoglycoprotein receptor. Int Rev Cytol. 1993;137b:181–219. 33. Weigel PH. Endocytosis and function of the hepatic asialoglycoprotein receptor. Subcell Biochem. 1993;19:125–61. 34. Trowbridge IS, Collawn JF, Hopkins CR. Signal dependent membrane protein trafficking in the endocytic pathway. Annu Rev Cell Biol. 1993;9:129–61. 35. Schmidt TA, Sullivan DA, Knop E, Richards SM, Knop N, Liu S, Sahin A, Darabad RR, Morrison S, Kam WR, Sullivan BD. Transcription, translation, and function of lubricin, a boundary lubricant, at the ocular surface. JAMA Ophthalmol. 2013;131:766–76. 36. Cheriyan T, Schmid TM, Spector M. Presence and distribution of the lubricating protein, lubricin, in the meibomian gland in rabbits. Mol Vis. 2011;17:3055–61. 37. Baldwin AK, Simpson A, Steer R, Cain SA, Kielty CM. Elastic fibres in health and disease. Expert Rev Mol Med. 2013;15:e8. 38. Hinek A, Pshezhetsky AV, von Itzstein M, Starcher B. Lysosomal sialidase (neuraminidase-1) is targeted to the cell surface in a multiprotein complex that facilitates elastic fiber assembly. J Biol Chem. 2006;281: 3698–710.
