Đang tải... (xem toàn văn)
Glycoprotein methods protocols - biotechnology
Mucin-Bacterial Binding Assays 42942935Mucin-Bacterial Binding AssaysNancy A. McNamara, Robert A. Sack, and Suzanne M. J. Fleiszig1. IntroductionSurface epithelia throughout the body are covered by mucus, a protective secretionthat serves as a selective physical barrier between the epithelial cell plasma membraneand the extracellular environment. Mucin, the glycoprotein constituent of mucus, hasbeen shown to bind bacteria at mucosal surfaces that line the lung, gut, bladder, oralcavity, and eye (1–7). Since bacterial binding to an epithelial cell surface is generallythought to be an important prerequisite for infection (8), the interaction between bac-teria and mucin, together with normal mucosal clearance mechanisms, is believed toact as a defense against infection by inhibiting bacterial adherence to the underlyingepithelial cell surface.In support of mucin’s role as a nonspecific defense mechanism, malfunctions in theproduction and/or clearance of mucin have been implicated in the etiology of manydiseases. This has led to the development of methods that can be used to study the effectsof disease and various interventions (e.g., drugs and medical devices) on the ability ofmucin to protect the underlying tissue. In this chapter, we present methods that can beused to examine the interaction between bacteria and mucin, as well as the extent towhich this interaction serves to protect the epithelial cell surface from bacterial invasion.We have used these methods to study the interactions of Pseudomonas aeruginosa withthe ocular surface in both human and animal models; however, they can also be used totest bacterial/mucin interactions in other tissues with other organisms.2. Materials2.1. Preparation and Analysisof Human Precorneal Tear Film Components1. Noninvasive corneal irrigation chamber (9,10).2. 0.9% NaCl (autoclaved).3. Spectra/Por®cellulose ester membranes, 500 Da molecular weight (MW) cut-off (FisherScientific, Fair Lawn, NJ).4. BCA Protein assay kit (Pierce, Rockford, IL).From:Methods in Molecular Biology, Vol. 125: Glycoprotein Methods and Protocols: The MucinsEdited by: A. Corfield © Humana Press Inc., Totowa, NJ 430 McNamara et al.5. Precast, 4–20% Tris-HCl Ready Gel (Bio-Rad, Hercules, CA) for sodium dodecyl sul-fate-polyacrylamide gel electrophoresis (SDS-PAGE).6. Silver stain kit (Bio-Rad).7. Coomasie blue R-250 (Sigma, St. Louis, MO).2.2. Isolation and Analysis of Human Ocular Sialoglycoprotein1. Glass microcapillary tubes (calibrated, fire-polished, disposable, 10 µL capacity).2. SW 4000 size-exclusion column.3. 0.5 M NaCl, 0.1 M phosphate buffer, pH 5.0.4. 20% Methanol (v/v).5. 3-kDa Cutoff centrifugal ultrafilters (Filtron, Northborough, MA).6. Precast, 4% and 4-20% Tris-glycine gels for SDS-PAGE (Novex, San Diego, CA).7. Kaleidoscope™ (Bio-Rad) and See-Blue ™ (Novex) molecular weight ladders, 6-250 kDa.8. Human lysozyme, IgG, lactoferrin, albumin and sIgA (Sigma).9. Pre-stained molecular weight markers IgM (990 kDa) and thyroglobulin (669 kDa)(Calbiochem®, La Jolla, CA).10. Coomasie brilliant blue R-250 (Sigma).11. Periodate silver and Alcian blue (AB) stains (Sigma).12. Immobilon P membranes (Millipore, Bedford, MA).13. Sialyl-Lewis epitope (clone 258-11413, O.E.M. Concepts, Toms River, NJ).2.3. Preparation of Bacteria1. P. aeruginosa (strain 6294, serogroup O6).2. Trypticase soy agar (TSA) plates (PML Microbiologicals, Wilsonville, OR).3. Phosphate-buffered saline (PBS), pH 7.4 (Sigma).4. MacConkey agar plates (PML Microbiologicals).5. Spectrophotometer.2.4. Microtiter Plate Assay of Bacterial Adherence1. Linbro/Titertek 96-well microtiter plate (ICN Biomedicals, Aurora, OH).2. PBS, pH 7.4 (Sigma).3. 0.50% Triton X-100 (LabChem, Pittsburgh, PA).4. MacConkey agar plates (PML Microbiologicals).5. Bovine submaxillary gland (BSG) mucin (Sigma).2.5. Assay to Confirm Adherence of Mucin to Microtiter Wells1. Same as Subheading 2.4., step 1 (Linbro/Titertek 96-well microtiter plate [ICNBiomedicals]).2. PBS, pH 7.4 (Sigma).3. 1% Bovine serum albumin (BSA) in PBS (w/v).4. Biotinylated wheat germ agglutinin (WGA) (Vector, Burlington, CA).5. PBS/Tween: 0.25 mL Tween-20 (Sigma) in 500 mL PBS.6. PBS/Tween:Streptavidin conjugated alkaline phosphatase (Jackson Immuno Research,West Grove, PN), 500:1 (v/v).7. Development solution: 5 mMp-nitrophenyl phosphate (NPP) in 0.1 M alkaline buffersolution (Sigma).8. Stop solution: 2 M Na2CO3 (2.12 g Na2CO3/10 mL distilled water). Mucin-Bacterial Binding Assays 4312.6. Assay to Measure Bacterial Invasion of Corneal Epithelial Cells2.6.1. Preparation of Rabbit Corneal Epithelial Cell Cultures1. Cell culture-treated 96-well plates (Fischer).2. Modified supplemental hormone essential medium (SHEM) containing 10 µg/mL ofbovine pituitary extract (11).2.6.2. Gentamicin Survival Assay1. Buffered minimal essential medium (BMEM): 9.53 g of MEM (Cellgro™) plus 2.2 g ofsodium bicarbonate per liter of distilled water (pH to 7.4).2. Gentamicin sulfate (BioWhittaker, Walkersville, MD).3. 0.25% Triton X-100 (LabChem).4. BSG mucin (Sigma).3. Methods3.1. Preparation and Analysisof Human Precorneal Tear Film Components1. Collect precorneal tear film components (TFCs) from the ocular surface of human eyesusing a noninvasive corneal irrigation chamber by irrigating each cornea for 30 s with 10 mL ofsterile saline using a metered pump as previously described (9,10).2. Remove cells and debris by centrifuging the eyewash samples three times at 6000 rpm for15 min.3. Dialyze the final supernatant against several changes of distilled water at 4°C and con-centrate to 200 µL using vacuum centrifugation.4. Determine the protein content of each 200 µL eyewash sample using the BCA proteinassay kit (12).5. Separate tear film proteins using a precast 4–20% Tris-HCl gel, and visualize using asilver staining procedure that is ideal for staining polysaccharides and highly glycosylatedproteins as recommended by the manufacturer (Bio-Rad) (13).6. Counter-stain the silver-stained gels with 200 mL of 0.1% Coomassie brilliant blueR-250 in 25% methanol/7.5% acetic acid for 1 h, destain overnight, and then photographin color as previously described (14) (see Note 1).3.2. Isolation and Analysis of Human Ocular Sialoglycoprotein1. Collect closed-eye tear samples (which are rich in high molecular weight sialoglycoprotein)(15) from four human subjects over a period of several weeks as previously described (16).2. Pool samples and centrifuge at 11,000 rpm in a refrigerated Eppendorf microfuge for30 min, then repeat. Store the resultant supernatants at –70°C until needed.3. Separate the supernatant isocratically in 15-µL aliquots on a SW 4000 size exclusioncolumn in 0.5 M NaCl, 0.1 M phosphate buffer (pH 5.0), at a flow rate of 0.25 mL/min,while monitoring the eluent at 254 nm.4. Concentrate each fraction using a 3-kDa centrifugal ultrafilter. To establish the elutionprofile for all the major tear proteins, run fractions under both reducing and non-reducingconditions at 125 V for one hour on a precast, 4–20% Tris-glycine gel for SDS-PAGE.Use Kaleidoscope™ and See-Blue™ MW ladders, as well as human lysozyme, IgG,lactoferrin, albumin, and sIgA as standards.5. Using this method, the high molecular weight glycoprotein fractions are recovered slightlyafter the void volume in the first peak, which elutes off the HPLC column at approxi- 432 McNamara et al.mately 24 min. To prepare the high molecular weight glycoprotein, collect the initialHPLC fraction, concentrate to 100 µL by centrifugal ultrafiltration, dilute 1:1 with HPLCsolvent, and separate into 15-µL aliquots.6. Run high molecular weight glycoprotein fraction under both reducing and nonreducingconditions on a precast 4% Tris-glycine gel with pre-stained molecular weight markersIgM (990 kDa) and thyroglobulin (669 kDa).7. To detect sialoglycoprotein (SG), periodate treat gel and stain with alcian blue in 3%acetic acid as previously described (17).8. For further characterization, transfer glycoprotein overnight onto Immobilon P in 20%methanol (v/v) at 30 V, followed by 80 V for 1 h.9. Probe with a panel of antibodies to specific known mucins, their core proteins, and com-mon sugar epitopes (see Note 2).3.3. Preparation of Bacteria for Binding Assay1. Grow bacteria overnight at 37°C on TSA plates.2. Wash bacteria three times in PBS by centrifugation at 7000 rpm for 5 min (9).3. Prepare the inoculum by resuspending the washed bacteria into PBS until the opticaldensity at 650 nm reaches 0.1 (equivalent to 1 × 108cfu/mL).4. Quantify the starting inoculum used in each experiment (typically 1 × 106cfu/mL) byserially diluting the sample and plating 10 µL (in duplicate) on MacConkey agar.3.4. Microtiter Plate Assay of Bacterial Adherence3.4.1. To Determine Whether or Not Bacteria Bindto Human Tear Film or Mucin (seeNote 3)1. Coat microtiter wells overnight at 37°C with 100 µL of tear film or mucin sample. Pretreatcontrol wells with PBS which does not promote bacterial adherence to these wells (5,7).2. Wash wells four times with PBS to remove non-adherent material.3. Prepare bacterial inoculum containing 1 × 106cfu/mL in PBS (10 µL of 1 × 108cfu/mL +990 µL PBS).4. Add inoculum containing 30 µL of 1 × 106cfu/ml P. aeruginosa 6294 to all wells.5. Incubate plate at 37°C for 30 min.6. Aspirate bacteria with a sterile pipette and wash wells 20 times with PBS to removenonadherent bacteria.7. Dislodge adherent bacteria from the well surface by adding 300 µL of 0.5% Triton X.8. Incubate plate at 37°C for 30 min.9. Vigorously stir each well with a sterile pipet and perform a viable count by plating 10 µL(in duplicate) on MacConkey agar.3.4.2. To Determine Whether or Not Mucin Blocks Bacterial Adherenceto Known Bacterial Binding Factors (seeNote 4)1. Coat microtiter wells overnight at 37°C with 100 µL of known bacterial binding factor(e.g., TFCs or BSG mucin).2. Wash wells four times with PBS to remove nonadherent material.3. Coat microtiter wells with 100 µL of SG for 18 h at 37°C. Treat control wells with 100 µL ofPBS instead of SG since PBS does not affect bacterial binding to either TFCs or BSG mucin.4. Prepare bacterial inoculum containing 1 × 105 cfu/mL in PBS (1 µL of 1 × 108cfu/mL +999 µL PBS).5. Add inoculum containing 30 µL of 1 × 105cfu/mL P. aeruginosa 6294 to all wells. Mucin-Bacterial Binding Assays 4336. Incubate plate at 37°C for 30 min.7. Wash wells 20 times with PBS to remove nonadherent bacteria.8. Dislodge adherent bacteria from the well surface by adding 300 µL of 0.5% Triton X.9. Incubate plate at 37°C for 30 min.10. Vigorously stir each well with a sterile pipet and perform a viable count by plating 10 µL(in duplicate) on MacConkey agar.3.4.3. To Determine Whether or Not Treating Bacteria with Mucin BlocksTheir Ability to Adhere to Known Bacterial Binding Factors (seeNote 4)1. Coat microtiter wells overnight at 37°C with 100 µL of known bacterial binding factor(e.g., TFCs or BSG mucin).2. Wash wells four times with PBS to remove nonadherent material.3. Prepare bacterial inoculum containing 2 × 105 cfu/mL in PBS (2 µL of 1 × 108cfu/mL +998 µLl PBS).4. Prepare starting inoculum by mixing 100 µL of 2 × 105cfu/mLP. aeruginosa (preparedabove) with 100 µL of SG and incubate at 37°C for 1 h. Prepare inoculum for controls bymixing 100 µL of 2 × 105cfu/mL bacteria with 100 µL of PBS (see Note 5).5. Add 30 µL of the starting inoculum containing P. aeruginosa 6294 (1 × 105cfu/mL) andeither SG or PBS (control) to wells.6. Incubate plate at 37°C for 30 min.7. Wash wells 20 times with PBS to remove nonadherent bacteria.8. Dislodge adherent bacteria from the well surface by adding 300 µL of 0.5% Triton X.9. Incubate plate at 37°C for 30 minu.10. Vigorously stir each well with a sterile pipet and perform a viable count by plating 10 µL(in duplicate) on MacConkey agar.3.5. Assay to Confirm Adherence of Mucin to Microtiter Wells1. Coat 96-well microtiter plate overnight at 37°C with several dilutions of SG, TFC, andBSG mucin.2. Wash wells 24 times with PBS.3. Block for 2 h at room temperature (or overnight at 4°C) by adding 200 µL of 1% BSA/PBS to each well.4. After blocking, wash wells twice with PBS/Tween and incubate with 100 µL of 5 µg/mLbiotinylated wheat germ agglutinin (WGA) for 45 min at room temperature (wrap plate in plas-tic). WGA is a plant lectin that binds specifically to sialic acid residues, and thus, serves as aprobe for quantifying the amount of sialylated glycoprotein that is bound to the microtiter well.5. Wash wells six times with PBS/Tween.6. Detect WGA-bound biotin by adding 100 µL of PBS/Tween:streptavidin-peroxidasesolution to each well and incubating the wrapped plate at 37°C for 45 min.7. Wash wells six times with PBS/Tween.8. Add 100 µL of development solution (NPP in alkaline buffer) and incubate at room tem-perature. Detect the sialoglycoprotein-lectin-enzyme complexes by adding the enzyme sub-strate (NPP) which is converted to a colored product in the presence of enzyme. Immediatelybegin monitoring the intensity at 405 nm using a standard ELISA reader. Readings shouldbe taken every 2 min since development occurs quickly. Controls consist of ovalbumintreated wells incubated with the WGA and/or streptavidin system alone (18,19).9. Stop the reaction by adding 10 µL of stop solution. This step can be omitted if the absorbanceis monitored continuously following the addition of development solution (see Note 6). 434 McNamara et al.3.6. Assay to Measure Bacterial Invasion of Corneal Epithelial Cells(seeNote 7)3.6.1. Preparation of Rabbit Corneal Epithelial Cell Cultures1. Place 100 µL of SHEM suspension containing rabbit corneal epithelial cells in cell cul-ture-treated 96-well plates.2. Maintain cells at 37°C while changing SHEM every 2 d until confluent (typically 4–7 d).3.6.2. Preparation of Bacteria for Gentamicin Survival Assay(see Subheading 2.3. for Materials)1. Grow bacteria overnight at 37°C on TSA.2. Prepare the inoculum by resuspending bacteria in 5 mL of buffered Eagle’s minimalessential medium (BMEM) using a sterile cotton swab until the optical density reaches0.1 (equivalent to 1 × 108cfu/mL at 650 nm).3. The starting inoculum of 1 × 107cfu/mL is prepared by placing 100 µL of the bacterialsuspension into 900 µL of BMEM. Quantify the starting inoculum used in each experi-ment by serially diluting the sample and plating 10 µL (in duplicate) on MacConkey agar.3.6.3. Gentamicin Survival Assay1. Wash confluent cells by adding 200 µL of BMEM to each well and aspirating with asterile pipet.2. Coat cells with 70 µL of BMEM containing several concentrations of BSG mucin, TFC,or SG. Pretreat control cells with 70 µL of BMEM alone.3. Incubate plate at 37°C for 2 h.4. Add 8 µL of an inoculum containing 1 × 107 cfu/mL of P. aeruginosa 6294 to all wells.5. Incubate plate at 37°C for 3 h.6. Perform viable counts on each well after 3 h to estimate the number of bacteria in theextracellular medium (see Note 8).7. Aspirate remaining medium and kill extracellular bacteria by adding 200 µL gentamicinsolution (200 µg/mL) to each well.8. Incubate plate at 37°C for 2 h.9. Wash cells with 200 µL BMEM.10. Lyse cells with 100 µL 0.25% Triton-X solution.11. After 15 min, use a sterile pipet to vigorously stir each well. Perform a viable count toquantify intracellular bacteria by plating 10 µL (in duplicate) on MacConkey agar. Com-pare the susceptibility of cells to invasion by P. aeruginosa in the presence of mucin withthat of control cells incubated with medium alone (see Note 9).4. Notes1. Using the silver/coomassie blue double-staining method described in Subheading 3.1.,all of the silver-stained glycoproteins and lipids stain by silver only and appear dark gray,whereas all of the unglycosylated proteins stain with coomassie blue and appear blue.Thus, polypeptides stained by coomassie brilliant blue and glycosylated proteins stainedby silver can be differentially visualized in the same gel. With this method, whole humantear film samples are shown to contain a prominent high molecular weight silver-stainedband of glycosylated protein (>200 kDa) that does not stain with coomassie brilliant blue.Others have identified a glycoprotein of similar size in human tear samples collected witha microcapillary tube (20–22). Mucin-Bacterial Binding Assays 4352. SDS-PAGE in combination with periodate silver and AB staining (described in Sub-heading 3.2.) reveal that high molecuar weight glycoproteins isolated from the ocularsurface consist primarily of three SG bands, one in the stacking gel and two in the runninggel with apparent molecuar weight ranges of 500–450 kDa. In addition, all of the SGsisolated from human tear film exhibit a common reactivity with a mAb raised against asialyl-Lewis epitope specific to salivary mucin (clone 258–11413, O.E.M. Concepts).Immunofluorescence microscopy reveals intense specific staining of the conjunctivalepithelial plasma membrane but not the goblet cells on probing with the mAb to salivarymucin, suggesting that these SGs arise specifically from the epithelium (15).3. In a previous study we showed that P. aeruginosa binds to factors present in whole humantears (23). To examine whether the bacterial binding factors in precorneal human tearfilm may include mucin, we used the bacterial adherence assay, described in Subheading3.4., and compared bacterial adherence to the SG isolated from the human ocular surfaceto binding elicited with whole human TFCs and BSG mucin.4. Using the bacterial adherence assay, we found that P. aeruginosa adhered to wells coatedwith both whole human tear film and BSG mucin, but did not bind to wells coated witheither ocular SG or PBS alone. This suggested that human ocular SG was not responsiblefor bacterial binding to whole tear film and that the bacterial binding factor(s) in humantears are either other glycoproteins or nonglycoprotein fractions. Having ruled out bacterialbinding to SG, we developed two new methods, which were variations of the original bacte-rial adherence assay, to determining whether SG could inhibit bacterial interaction with knownbacterial binding factors. The first variation of the bacterial adherence assay, described inSubheading 3.4.2., sought to determine whether or not ocular SG could block bacterial bind-ing to whole human tear film or to BSG mucin when microtiter plates were precoated withTFCs or mucin and then subsequently coated with SG. The second variation, described inSubheading 3.4.3., was used to determined whether or not bacterial binding to mucin-coatedmicrotiter plates could be blocked by adding SG to the bacterial inoculum.5. When bacteria are pretreated with any factor prior to performing the bacterial adherenceassay, one must demonstrate that preincubation with that factor does not reduce theviability or the replication rate of the bacteria. This can be accomplished by performingviable counts after 2 and 8 h to determine whether inocula containing treated and untreatedbacteria are equivalent.6. Using the biotinylated WGA assay described in Subheading 3.6., we found that theabsorbance reading for a 10–2dilution of SG was similar to that observed for a 10–2dilu-tion of TFCs and 10 µg/mL of BSG mucin; thus, we used these concentrations in thebacterial binding assay to ensure that an equivalent number of sialic acid residues werebound to the bottom of the well for each preparation.7. Bacterial association with the corneal surface is a crucial step towards bacterial coloniza-tion and invasion of corneal epithelial cells. Since soluble factors in whole human tearsand BSG mucin bind P. aeruginosa, while human ocular SG does not, we used the gen-tamicin survival assay to examine whether these factors could block P. aeruginosa inva-sion of corneal epithelial cells. As was expected, treating cells with several differentdilutions of BSG mucin or TFCs caused a two- to threefold decrease in susceptibility tobacterial invasion by P. aeruginosa strain 6294, whereas ocular SG did not significantlyenhance or inhibit bacterial invasion of corneal epithelial cells compared to controls.8. When studying bacterial invasion of epithelial cells using the gentamicin survival assay,it is important to confirm that an equivalent number of extracellular bacteria remain in thewells following each treatment (i.e., the treatment does not alter the viability or replica- 436 McNamara et al.tion rate of the bacteria). This is best done by performing a viable count on one well ineach different treatment group at the end of the 3-h incubation of cells with bacteria (justprior to adding the gentamicin).9. For all the assays described in this chapter, we recommend that at least six wells are usedfor each sample group in each experiment and that experiments are repeated a minimumof three times. Nonparametric statistical techniques are typically used to make compari-sons since the presence of outlying values may cause distributions to deviate from nor-mality. Two group comparisons can be made with the Wilcoxan 2-group test for unpaireddata and the Kruskal-Wallis test can be used for the comparison of three or more indepen-dent groups of data.References1. Florey, H. (1955) Mucin and the protection of the body. Proc. R. Soc. Lond. 143, 147–158.2. Liljemark, W. and Gibbons, R. (1972) Proportional distribution and relative adherence ofStreptococcus miteor (mitis) on various surfaces in the human oral cavity. Infect. Immun.6, 852–859.3. Parsons, C. and Mulholland, S. (1978) Bladder surface mucin. Its antibacterial effectagainst various bacterial species. Am. J. Pathol. 93, 423–432.4. Cohen, R., Rossoll, R., Cabelli, V., Yang, S., and Laux, D. (1983) Relationship betweenthe mouse colonizing ability of a human fecal Escherichia coli strain and its ability to binda specific mouse colonic mucous gel protein. Infect. Immun. 40, 62–69.5. Vishwanath, S. and Ramphal, R. (1984) Adherence of Pseudomonas aeruginosa to humantracheobronchial mucin. Infect. Immunity. 45, 197–202.6. Chick, S., Harber, M., Mackenzie, R., and Asscher, A. (1981) Modified method for studyingbacterial adhesion to isolated uroepithelial cells and uromucoid. Infect. Immun. 34, 256–261.7. Fleiszig, S., Zaidi, T., Ramphal, R., and Pier, G. (1994) Modulation of Pseudomonasaeruginosa adherence to the corneal surface by mucus. Infect. Immun. 62, 1799–1804.8. Ramphal, R., McNiece, M., and Polack, F. (1981) Adherence of Pseudomonas aeruginosato the injured cornea. A step in the pathogenesis of corneal infections. Ann. Ophthalmol.100, 1956–1958.9. Fleiszig, S., Efron, N., and Pier, G. (1992) Extended contact lens wear enhancesPseudomonas aeruginosa adherence to human corneal epithelium. Invest. Ophthalmol.Vis. Sci. 33, 2908–2916.10. Fullard, R. and Wilson, G. (1986) Investigation of sloughed corneal epithelial cells col-lected by non-invasive irrigation of the corneal surface. Curr. Eye Res. 5, 847–856.11. Jumblatt, M. and Neugeld, A. (1983) Serotonergic responsiveness of rabbit corneal epi-thelial cells in culture. Invest. Ophthalmol. Vis.Sci. 24, 1139–1143.12. Lowry, O., Rosebrough, N., Farr, A., and Randall, R. (1951) Protein measurement withthe folin phenol reagent. J. Biol. Chem. 193, 265–275.13. Merril, C., Goldman, D., Sedman, S., and Ebert, M. (1981) Ultrasensitive stain for pro-teins in polyacrylamide gels show regional variations in cerebrospinal fluid proteins. Sci-ence 211, 1437.14. Dzandu, J., Deh, M., Varratt, D., and Wise. G. (1984) Detection of erythrocyte membraneproteins, sialoglycoproteins and lipids in the same polyacrylamide gel using a double stain-ing technique. Proc. Natl. Acad. Sci. USA 81, 1733–1737.15. Sack, R., Bogart, B., Beaton, A., Sathe, S., and Lew, G. (1997) Diurnal variations in tearglycoproteins. Evidence for an epithelial origin for the major non-reducible ≥ 450 kDasialoglycoprotein(s). Curr. Eye. Res. 16, 577–588. Mucin-Bacterial Binding Assays 43716. Sack, R., Tan, K., and Tan, A. (1992) Diurnal tear cycle: evidence for a nocturnal inflam-matory constitutive tear fluid. Invest. Ophthalmol. Vis. Sci. 33, 626–640.17. Wardi, A. and Michos, G. (1972) Alcian blue staining of glycoproteins in acrylamide discelectrophoresis. Anal Biochem. 49, 607–609.18. Kaku, H., Mori, Y., Goldstein, I., and Shibuya, N. (1993) Monomeric, monovalent deriva-tive of Maackia amurensis leukoagglutinin. J. Biol. Chem. 268, 13,237–13,241.19. Rogerieux, R., Belaise, M., Terzidis-Trabelsi, H., Greffard, A., Pilatte, Y., and Lambre, C.(1993) Determination of the sialic acid linkage specificity of sialidases using lectins in asolid phase assay. Anal. Biochem. 211, 200–204.20. Moore, J. and Tiffany, J. (1979) Human ocular mucus. Origins and preliminary character-ization. Exp. Eye. Res. 29, 291–301.21. Chao, C., Vergnes, J., and Brown, S. (1983) O-glycosidic linkage in glycoprotein isolatesfrom human ocular mucus. Exp. Eye. Res. 37, 533–541.22. Jay, G. and Hong, B. (1992) Characterization of a bovine synovial fluid lubricating factor.II. Comparison with purified ocular and salivary mucin. Conn. Tissue Res. 28, 89–98. . CA) for sodium dodecyl sul-fate-polyacrylamide gel electrophoresis (SDS-PAGE).6. Silver stain kit (Bio-Rad).7. Coomasie blue R-250 (Sigma, St. Louis, MO).2.2.. buffer) and incubate at room tem-perature. Detect the sialoglycoprotein-lectin-enzyme complexes by adding the enzyme sub-strate (NPP) which is converted