Abstract

Patterns of secretion of mucins and non-mucin glycoproteins in human submandibular/sublingual secretion

Description
Patterns of secretion of mucins and non-mucin glycoproteins in human submandibular/sublingual secretion
Categories
Published
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Patterns of secretion of mucins and non-mucinglycoproteins in human submandibular/sublingualsecretion Laura Becerra a , Rodrigo V. Soares b , Lucila S. Bruno a ,Camille C. Siqueira a , Frank G. Oppenheim a,b ,Gwynneth D. Offner a,c , Robert F. Troxler a,b, * a Department of Periodontology and Oral Biology, Goldman School of Dental Medicine,700 Albany Street, Boston, MA 02118, USA b Department of Biochemistry, Boston University School of Medicine,80 East Concord Street K-312, Boston, MA 02118, USA c Department of Medicine, Boston University School of Medicine, 650 Albany Street,Boston, MA 02118, USA Accepted 14 October 2002Archives of Oral Biology (2003)  48 , 147—154 KEYWORDS MG1; MG2;Glycoprotein;Secretion pattern;Submandibular/sublingual gland Summary  The present investigation has characterised the influence of gustatorystimulation and duration of stimulation on the secretion pattern of salivary mucinsMG1 and MG2 and non-mucin glycoproteins in submandibular/sublingual secretion(SMSL). Resting SMSL was collected for three 2 min intervals and stimulated SMSL wascollected for ten 1 min intervals fromsix healthy subjects. Flow rates and total proteinwere significantly different under the two conditions. The secretion patterns of theseproteins under resting and stimulated conditions was examined on periodic acid-Schiffreagent(PAS)-stainedpolyacrylamidegelsusingaKodakDigital-ScienceImageStation.Image analysesrevealed that the levelof MG1increased andthelevel of MG2remainednearlythesameafterstimulation.Sixothermajorglycoproteins(designatedBand1—6)were identified on the basis of their electrophoretic mobilities and immuno-reactivityonWesternblots. After stimulation the intensity of Band 1(lactoferrin and peroxidase)and Band 2 (amylase) decreased whereas the intensity of Band 3 (carbonic anhydrase),Band 4 (proline-rich glycoprotein) and Bands 5 and 6 (basic glycosylated proline-richproteins)increased.Thesepatternsprobablyreflectsecretionfrompreformedvesiclessince de novo synthesis would be unexpected within the time frame of these experi-ments. The variable patterns observed suggest that mucins and non-mucin glycopro-teins in SMSL derive from different subsets of secretory vesicles, some of which maysrcinate in mucous and others in serous acini, as well as in ductal cells. Quantificationof mucins was performed by image analysis technology using purified MG1 and MG2standards. Finally, the present investigation has shown that the secretory patterns ofmucins and non-mucin glycoproteins from submandibular/sublingual glands are com-plex and represent an important aspect of salivary gland physiology.   2003 Elsevier Science Ltd. All rights reserved.  Corresponding author. Tel.:  þ 1-617-638-4048; fax:  þ 1-617-638-5339. E-mail address:  btrox@bu.edu (R.F. Troxler).0003–9969/03/$ — see front matter   2003 Elsevier Science Ltd. All rights reserved.doi:10.1016/S0003-9969(02)00171-1  Introduction Proteins in salivary secretions and whole salivaaffect hard and soft tissues in the oral environmentin a variety of beneficial ways. The physiologicalimportance of the organic and inorganic compo-nents in saliva is illustrated dramatically in situa-tions where saliva production is impaired as inautoimmune diseases such Sjogrens’s Syndrome,systemic lupus erythematosus and scleroderma. 1—3 A reduction in saliva production also occurs inpatients taking some prescription medications 4 and in patients receiving radiation treatment forhead and neck cancer. 5 It is also well recognisedthat diminished salivary flow leads to difficulty inphysiological processes such as mastication, swal-lowing and speech 6 and can affect the quality of lifecausing problems related to retention of dentures,infectionwithoralbacteriaandyeast, developmentand progression of dental caries and alterations intaste. 1,2 Whole saliva is derived primarily from secretionsof three pairs of major salivary glands, the parotid,submandibular, sublingual glands, with a muchsmaller contribution from minor salivary glandsand gingival crevicular fluid. Under resting condi-tions parotid and submandibular glands contribute20 and 65% and under stimulated conditions con-tribute 50 and 35%, respectively, to the volume ofwhole saliva. 7 Sublingual glands, minor salivaryglands and gingival crevicular fluid each togethercontribute less than 15% to the volume of wholesaliva under both resting and stimulated conditions.Since resting conditions prevail in the oral cavity formost of a 24 h period we were interested in changesin secretion patterns of salivary proteins underresting and stimulated conditions.A variety of factors important for oral health aremediated by the protective and antimicrobialeffects of mucins and non-mucin glycoproteins insalivary secretions. 8 The carbohydrate compositionof glycoproteins can be as high as 80% as in the caseof the salivary mucins MG1 and MG2 9 or less than 5% as in the case of salivary peroxidase, 10 amylase 11 orcarbonic anhydrase. 12 Since submandibular/sublin-gual secretions (SMSL) contain both mucin and non-mucin glycoproteins, whereas parotid secretioncontains only non-mucin glycoproteins, the aimof the present investigation was to compare secre-tion patterns of mucins with those of non-mucinglycoproteins in SMSL under resting and stimulatedconditions.The secretion patterns of MG1, MG2 and of non-mucin glycoproteins were examined on periodicacid-Schiff reagent (PAS) stained polyacrylamidegels and analysed on a Kodak Digital-Science ImageStation. A complex pattern of secretion wasobserved where levels of mucin and non-mucinglycoproteins, with one exception, changed duringstimulation. The levels of some of these proteinsincreased while others decreased suggesting regu-lated release from subsets of vesicles in subman-dibular and sublingual glands. In addition, a simpleprocedure for quantification of salivary mucinsbased on image analysis of PAS-stained gels isdescribed. Materials and methods Saliva collection Resting and stimulated SMSLwas collected from sixsubjects (two males and four females; 25—35 yearsof age) between 10:00 and 11:00 a.m. Subjectswere asked not to eat for 2 h prior to collection.SMSL was collected with a custom-fitted device asdescribed previously. 13 Resting SMSL was collectedfor10 minandthefirst4 minofthecollectionswerediscarded. Subsequently, three consecutive 2 minsamples were collected in separate graduated1.5 ml centrifuge tubes followed by 10 consecutive1 min samples that were collected in 15 ml Falconscrew cap tubes. Gustatory stimulation wasinduced by placing fruit flavoured candies on thetongue (Jolly Rancher, Hershey,PA). Tubes ofrest-ing SMSL were labelled 5, 7 and 9 to indicatesamples collected during minutes 5 þ 6, 7 þ 8and9 þ 10. Tubes ofstimulatedSMSLwere labelled11—20 to indicate collections for 1 min intervals.All samples were kept on ice during the collectionprocedure. Flow rate and total protein The flow rate from each subject was recorded andsamples were frozen at   20  8 C until used. Prior tofreezing, 5  m l was removed from each tube andplaced in 995  m l of distilled water for quantificationof total protein. The total protein concentrationwas determined spectrophotometrically at 215 nm(A 215 method)usingaspecificabsorptioncoefficientof 15.0 for a 1 mg/ml solution in a 1 cm cuvette. 14 Gel electrophoresis Samples of resting and stimulated SMSL werethawed on ice and made 5 mM with respect to EDTAin order to promote solubilization of mucins andnon-mucin glycoproteins. 15 For each subject equalvolumes of resting and stimulated SMSL werelyophilised, taken up in 10  m l of sample buffer and148 L. Becerra et al.  subjected to SDS—PAGE. 16 The stacking and separ-ating gels contained 5 and 10% acrylamide, respec-tively. Electrophoresis was carried out untilthe tracking dye reached the bottom of the gel(approximately 2.5 h at 70 V) and gels were stainedwith periodic acid-Schiff reagent. 17 Gels weredestained in 40% ethanol, 5% acetic acid and storedin 5% acetic acid. Western analyses Inseveralexperiments,proteins inSMSLweretrans-ferred to nitrocellulose membrane (Protran, Schlei-cher and Schuell, Keene, NH) and blots were eitherstained with PAS or probed with rabbit polyclonalantibodies against MG1, MG2, lactoferrin, amylaseandcarbonicanhydrase.Theanti-MG1andanti-MG2antibodies were both diluted 1:1000 as described. 18 Antibodiesagainstlactoferrin(Sigma,St.Louis,MO)were diluted 1:2000, against amylase (ZymedLaboratories, South San Francisco, CA) were diluted1:250 and against carbonic anhydrase II (US Biolo-gical, Swampscott, MA) were diluted 1:1000. Thesecond antibody for all blots was goat anti-rabbitIgG coupled to alkaline phosphatase (Promega,Madison, WI) diluted 1:7500. Colour developmentwas carried out with NBT/BCIP according to manu-facturer’s instructions. Image analysis AnalysisofbandsonPAS-stainedgelswasperformedwith a Kodak Digital-Science Image Station (model440CF; subsequently referred to as Image Station).Mucin and non-mucin glycoprotein bands on digitalphoto-images were framed and the intensity (pix-els) of the band in the frame was estimated usingID3 software. The size of the frame used for bandswasthesameforthatparticularbandintheseriesofresting and stimulated SMSL samples. The numberof pixels within a frame of a mucin or non-mucinglycoprotein band was corrected for background bysubtracting the pixels in a frame of the same size ina portion of the gel not containing PAS-reactivematerial. Mucin quantification Purified MG1 and MG2 standards were prepared inour laboratory as described. 19,20 Mucin standardswere electrophoresed, gels were stained with PASand the intensity of the bands was measured in theImage Station. The standard curve for MG1 waslinear from 0.5 to 6.0  m g and the standard curvefor MG2 was linear between 3 and 12  m g. Gels con-taining samplesofresting andstimulated SMSLwererun together with two different quantities of MG1and MG2 internal standards. MG1 and MG2 in thesesamples were quantified based on the intensity ofthe signal obtained from the standards. Statistics Due to the considerable person to person variationin the quantity of mucins and non-mucin glycopro-teins, the intensity (pixels) of these compoundsin each subject’s secretion was expressed aspercentage of their own initial value (i.e. minute5 þ 6 ¼ 100%; see in the Section ‘‘Saliva collec-tion’’). Mean values for each band from the sixsubjects over time was calculated and the datawere plotted to show the pattern of secretion overthe experimental period. Statistical analysis of theeffects of resting and stimulated conditions on theflow rate, total protein, MG1, MG2 and Bands 1—6were performed by SAS software licensed to BostonUniversity. The effects of resting and stimulatedconditions were tested for statistical significance( a  0 : 05)byatwo-wayrepeated-measuresANOVA.The effects of time within resting or stimulatedconditions were tested for statistical significance( a  0 : 05) by a one-way repeated-measures ANOVA.The  R 2 -values for the linear regression of MG1 andMG2standardswereanalysedwithHarvardGraphicssoftware. Results Flow rate and total protein The effect of stimulation and duration of stimula-tion on the flow rate of SMSL in six subjects showedthe expected pattern in which the flow rateincreased after stimulation to a constant level forthe remainder of the experimental period. Theaverage mean value of flow rate under resting con-ditions was 0 : 36  0 : 15 ml/min and under stimu-lated conditions was 2 : 7  0 : 96 ml/min. Statisticalanalyses showed that flow rates under resting andstimulated conditions were significantly different( P   <  0 : 0001). However, no significant time effectwas observed during resting ( P  ¼ 0 : 95) or stimu-lated ( P  ¼ 0 : 50) conditions. The average meanvalue of total protein exhibited a similar patternin which the mean resting value was 151 : 1  33 : 7 mg% and the mean stimulated value was322 : 5  158 : 8 mg%. Statistical analysis revealed asignificant difference in total protein under restingversus stimulated conditions ( P   <  0 : 0001) and nosignificant time effect during resting ( P  ¼ 0 : 92)and stimulated ( P  ¼ 0 : 44) conditions.Patterns of secretion of mucins and non-mucin glycoproteins 149  Electrophoretic pattern The electrophoretic pattern of mucin and non-mucin glycoproteins from a single subject on aPAS-stained gel is shown in Fig. 1. The arrow atthe bottom indicates the onset of gustatory stimu-lation. Lanes labelled 5, 7 and 9 contain restingSMSLcollectedover2-minintervalsandlanes11—20show stimulated SMSL collected over 1 min inter-vals. Salivary mucins were identified on the basis oftheir characteristic electrophoretic behaviour andreactivity with anti-MG1 and anti-MG2 antibodies.MG1 has an apparent molecular weight of >1000 kDaand appears as a tight PAS-reactive band at the topof the stacking gel (Fig. 1). MG2 has a molecularweight ranging from 170 to 190 kDa and comprisesthe largest PAS-reactive band in the separating gel.The molecular weights of PAS-reactive mucin bandscorresponded exactly to immunoreactive bands onWestern blots probed with antibodies against MG1and MG2 (Fig. 2).Non-mucin glycoproteins ranging from approxi-mately 90 to 10 kDa were designated Band 1—6(Fig. 1) and were tentatively identified on the basisof the molecular weights reported for non-mucinglycoproteins in SMSL (Table 1). Band 1 is likely a Figure 1  Representative gel showing the secretion pattern of PAS-reactive glycoproteins in SMSL from one subject.Arrows at the left indicate the electrophoretic mobility of protein standards and the position of MG1, MG2 and theother glycoproteins (Bands 1—6) is shown at the right. Labels at the top of the panel refer to the collection times. Thearrow at the bottom of the gel marks the onset of stimulation. Table 1  Identification of salivary glycoproteins based on molecular weight.Band Protein Molecular weight (kDa)Observed Literature a 1 Lactoferrin 75 76.51 Peroxidase 75 72—782 Amylase 60 55—603 Carbonic anhydrase 38 424 Proline rich glycoprotein 35 38.95 Glycosylated PRPs 28 20—306 Glycosylated PRPs 20 10—20 a Literature values from references. 10—12,21,31 150 L. Becerra et al.  mixture of lactoferrin and peroxidase, both ofwhich have reported molecular weights in the75 kDa range. 10,21 Band 2 most likely representsthe glycolyslated form of amylase and has an appar-ent molecular weight in the low 60 kDa range. 11 Band 3 has a molecular weight of approximately38 kDa and has been designated as carbonic anhy-drase VI. 12 Band 4 has a molecular weight ofapproximately 35 kDa, consistent with that of theproline-rich glycoprotein. Bands 5 and 6 have mole-cular weights of 28 and 10—20 kDa, respectively,and are probably glycosylated proline-rich proteins(Stubbs et al., 1998). Note that Band 6 actuallycontains two closely separated bands. To bettercharacterise non-mucin glycoproteins, Westernblots ofstimulated SMSLwereprobedwith commer-cially available antibodies against lactoferrin, amy-lase and carbonic anhydrase and the position ofimmunoreactive bands was compared to that ofPAS-reactive bands on nitrocellulose membrane.These experiments confirmed that the 75 kDa PASband contained a protein that reacted with anti-lactoferrin antibodies, the   60 kDa PAS band con-tained a protein that reacted with anti-amylaseantibodies and the 38 kDa PAS-reactive band con-tained a protein that reacted with anti-carbonicanhydrase antibodies (Fig. 2). To our knowledge,antibodies against Band 4, designated as proline-rich glycoprotein, and Bands 5 and 6, designated asglycosylated PRPs are not available.The electrophoretic pattern shown in Fig. 1 istypical of that observed for the six subjects whoparticipated in this study. The secretion pattern ofmucins and non-mucin glycoproteins is interestingbecause the relative intensities of some bandsincrease, some bands decrease and one remainsnearly the same. This is in contrast to the situationfor total protein which increases upon stimulation.Visual inspection of the gel in Fig. 1 reveals that theintensity of the MG1 band appears to increaseslightly whereas the intensity of the MG2 bandappears to remain nearly the same. In contrast,Band 1 decreases dramatically after stimulationand Band 2 follows a similar pattern. Band 3 andBand 4 were barely detectable in resting SMSL andincreased after stimulation. Bands 5 and 6 areclearly recognisable in resting SMSL and rise to amaximum level upon stimulation. Image analysis The intensity of PAS-reactive bands on gels of rest-ing and stimulated SMSL from six subjects wasanalysed on an Image Station. Since there was con-siderable variability among subjects, the intensityassociated with a particular band in each subject’ssecretion was expressed as percentage of its owninitial value (i.e. minute5 þ 6 ¼ 100%).Values fromall subjects were then averaged and the meanpercentage was plotted versus time to show thesecretion pattern over the experimental period.After stimulation, MG1 increased significantly( P   ¼ 0 : 0026) although no significant difference wasobservedwithintherestingtimeperiod( P  ¼ 0 : 42)orduring stimulation ( P  ¼ 0 : 75). In this study, MG2 wasthe only component for which stimulation had noapparent effect. After stimulation, the level of MG2was not significantly altered ( P   ¼ 0 : 47) and no sig-nificant change was observed within the resting( P   ¼ 0 : 31) or stimulated ( P  ¼ 0 : 99) time period.Analysis of Bands 1—6 revealed a more complexpattern of secretion in which the intensity of allbands changed as a consequence of stimulation.Band 1 decreased significantly ( P   <  0 : 0001) afterstimulation and a significant increase ( P  ¼ 0 : 0026)was observed during the resting time period and asignificant decrease ( P  ¼ 0 : 039) was observed dur-ing the stimulated time period. Band 2 decreasedsignificantly ( P   <  0 : 0001) after stimulation and a Figure 2  Comparison of PAS staining and immunologicalreactivity of mucin and non-mucin glycoproteins onnitrocellulose membranes. Stimulated SMSL was electro-phoresed in multiple lanes (20  m l per lane) and proteinswere transferred to a nitrocellulose membrane. Themembrane was stained with PAS or probed with specificantibodies against MG1, MG2, lactoferrin, amylase andcarbonic anhydrase. Patterns of secretion of mucins and non-mucin glycoproteins 151
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks