HYBRIDOMAVolume 10, Number 2, 1991\lar\ Ann Lieber!, Inc., Publishers

Production and Characterization of MonoclonalAntibodies to Purified Deglycosylated

Cystic Fibrosis Respiratory Mucin:Evidence for the Presence of FourImmunologically Distinct Epitopes

VINAY C. DESAI,' BASHOO NAZIRUDDIN,1 DONALD C. GRAVES,2SANTIAGO REYES De La ROCHAR and GOVERDHAN P. SACHDEV

'College of Pharmacy, 2College of Medicine and sCystic Fibrosis Center,University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma 73190

ABSTRACT

Respiratory mucus glycoproteins (mucins) were purified from the tracheobronchialsecretions of three Cystic Fibrosis (CF) patients. The mucins were completelydeglycosylated by treatment with trifluoromethanesulfonic acid and subsequent treatmentwith a-N-acetylgalactosaminidase. Over thirty hybrid clones secreting antibodies against thedeglycosylated mucin (DGM) were obtained using standard hybridoma techniques. Hybridswith positive identification for CF-DGM were cloned twice using limiting dilution methodto ensure the monoclonal nature of the antibodies. Eight stable clones (la, lb, 10a, 10c,lOd, lOe, 29d, and 30e) secreting monoclonal antibodies (MAbs) showing specificity ofreaction to CF-DGM were obtained. Two clones, 29d and 30e, secreted antibodies of theIgM class while the other six clones secreted antibodies of the IgGj subclass. Denaturationand reduction experiments suggested that MAbs lb, lOe, 29d and 30e were directed againsta given sequence of amino acids in the DGM while the other four MAbs, in addition tobeing sequence specific, were also conformation dependent. Further, competitive bindingradioimmunoassays suggested that MAbs lb, lOe, 29d and 30e recognize four distinctepitopes in the peptidic core of CF respiratory mucin. In summary, the MAbs may providea promising approach to elucidate the structure of the polypeptide backbone of humanrespiratory mucins as well as for the screening of cDNA libraries for clones secretingmucin(s).

INTRODUCTION

Mucins are considered primarily responsible for the viscoelastic properties ofrespiratory mucus (1,2). Mucins are polydisperse, polyanionic macromolecules withmolecular weights greater than one million (3,4). The carbohydrate portion oftracheobronchial mucins is comprised mainly of five sugars, namely, fucose, galactose, sialicacid, N-acetylglucosamine and N-acetylgalactosamine through which the oligosaccharide sidechains are linked via O-glycosidic linkage to a serine/threonine residue of the polypeptidebackbone. The structure of mucins involve heavily glycosylated protein core interspersedwith "naked" stretches that are held together by disulfide and noncovalent linkages (5).While some information is available on the oligosaccharide side chains of these mucins(6,7), little is known, to date, about the primary structure of the mucin polypeptide core.Woodward et al (8) have reported that the human DGM has an apparent Mr ~ 100,000.

In recent years, MAbs directed against native mucus components have been used tostudy cell specific products (9) and structural relationship between several mucin species(10,11). For example, Podolsky et al (10) generated a library of MAbs directed against

285

purified normal human colonie mucin. Some of the antibodies showed specificity towardsix different mucin species. However, all such MAbs were raised against the intact mucinmolecule and not the core protein. Hence, these MAbs were mostly directed towards thecarbohydrate moieties of the mucins. Thus, lack of structural information about thepolypeptide core of mucins prompted us to develop and characterize MAbs directed againstthe DGM. In the present study, we report the production of MAbs developed againstcompletely deglycosylated CF respiratory mucin antigen. Competitive binding assays usingthe MAbs revealed the presence of four immunologically distinct epitopes in the CF-DGM.

EXPERIMENTAL PROCEDURES

Purification and Deglycosylation of Respiratory Mucins

Mucus secretions collected from three CF patients (CF-6,CF-4 and CF-7) were usedas the starting material. The purification of mucins from human respiratory mucussecretions was carried out using a protocol established in our laboratory (3,12,13).Deglycosylation of respiratory mucins was accomplished by the method of Woodward et al(8) with slight modification. In brief, asialo mucin, prepared by removal of terminal sialicacid residues using neuraminidase (Vibrio cholerae) (Calbiochem-Behring), was treated twicewith trifluoromethanesulfonic acid-anisole (2:1 v/v) for 8 h at 8 °C to remove theoligosaccharide chains. Removal of remaining O-linked N-acetylgalactosamine residues wasachieved by treatment with a-N-acetylgalactosaminidase (Limpet, Patella vulgata; V-Labs,LA) for 24 h at 37 °C. Identical experimental conditions were used for deglycosylation ofall CF mucins studied in this work.

Analytical Methods

Chemical and amino acid compositions were determined for the purified native anddeglycosylated respiratory mucins as described earlier (3,12,13). In brief, sialic acid was

determined by the resorcinol method using N-acetylneuraminic acid as the standard (14).Fucose, galactose, deoxyribose, N-acetylgalactosamine and N-acetylglucosamine weredetermined by GLC (15). In addition, after hydrolysis (4 N HC1, 4 h at 100 °C), the aminosugars were also analyzed using an amino acid analyzer (sensitivity of detection = 0.1 figof amino sugar). Amino acid analyses were performed essentially as described earlier(3,12,13) using a Durrum D-500 amino acid analyzer following hydrolysis of the sample withconstant-boiling hydrochloric acid at 110 °C for 22 h in evacuated sealed tubes. The purityof native mucins was examined by SDS composite gel electrophoresis (2% acrylamide, 0.5%agarose containing 0.1% SDS) (16). The deglycosylated mucins were examined using bothcomposite gel (2% acrylamide & 0.5% agarose) and 6% polyacrylamide gel containing 0.1%SDS. The protein bands in the gel were visualized by the silver stain method (17). Aminoterminal analyses of the mucins was done with an automated gas phase sequenator usingEdman chemistry (18).FPLC Analysis of Native and Deglycosylated Mucins

In order to further test the purity of native and deglycosylated mucins, each mucin(—100 fig) dissolved in 0.1 M Tris-HCl buffer, pH 7.5, containing 0.22 M potassiumthiocyanate, 4 M guanidine-HCl and 0.02% sodium azide was injected on a FPLC Superóse6 (10/30) (Pharmacia-LKB) column pre-equilibrated with the sample buffer and eluted ata flow rate of 0.4 ml/min. The eluant was monitored at 280 nm and 1 ml fractions werecollected.

Production. Screening and Characterization of MAbs

Immunization

Five female BALB/c ByJ mice were immunized at three week intervals using purifieddeglycosylated CF respiratory mucin (CF-6). For the first injection, the purified CF

286

apomucin (100 ¡ig) was emulsified in Freund's complete adjuvant (1:1, v/v) and injectedintraperitoneally and subcutaneously. The second injection (50 ¿ig of the antigen) was givenin a similar fashion with Freund's incomplete adjuvant. The third injection (50 fig ofapomucin) was given intravenously without the adjuvant. Four days after the finalimmunization, mice having the highest antibody response as determined by the indirectenzyme-linked immunosorbent assay (ELISA)(see below) were sacrificed and their spleencells used for fusion experiments.Fusion and Cloning

Cell fusion was done essentially as described by Cunningham et aj (19) using Sp 2/0myeloma cells. Hybridomas were grown in RPMI-1640 medium containing 10% fetalbovine serum. Following detection of hybridoma growth, culture fluids were assayed forantibody activity using ELISA. Antibody positive cells were then transferred to 24-wellplates and sub-cloned by the limiting dilution method. The subcloning process was repeatedonce more to ensure monoclonality of these hybrid cells. The hybrid clones were eitherexpanded to larger cultures, frozen at -70°C or inoculated into mice for ascites production.Ascites Fluid Production

To obtain sufficient amounts of antibody, clones of hybrid cells were injectedintraperitoneally (1 x 107 cells/ml in IMDM serum free medium) into BALB/c mice whichhad been primed 2 or more weeks earlier with 0.5 ml of pristane (2,6,10,14-tetramethylpentadecane)(Sigma Chemical Co.). Animals were observed daily and ascitesfluid collected when formed (usually 8 to 12 days after injection).Purification of IgG Fraction from Ascites Fluid

The antibody was purified from the ascitic fluid by using a FPLC Protein A-Superosecolumn (HR 5/5)(Pharmacia-LKB). Five ml of ascitic fluid was diluted 1:5 (v/v) in bufferA (3 M NaCl, 1.5 M Glycine buffer, pH 8.9) and loaded onto the preequilibrated column.The column was eluted with buffer A at a flow rate of 1 ml/min until all the unboundfractions of the ascitic fluid were eluted from the column. The bound IgG fraction wasthen eluted by using buffer B (0.1 M citric acid buffer, pH 3.0). The IgG fraction was thenimmediately dialyzed against deionized water at 4°C and lyophilized.Screening for Antibodies in Culture Supernates

The culture fluids isolated from hybridomas were screened for antibody activityusing ELISA CF-DGM (50 ¡A, 2.5 fig/ml) dissolved in carbonate-bicarbonate coatingbuffer, pH 9.6, was plated in 96-well microtiter plates (Nunc) at 37°C for 1 h. After properwashings and blocking of the unbound sites on the plates (2% Tween 20 in PBS solution),50 fi\ of culture fluids were added to the wells and incubated at 37°C for 1 h. Followingadditional washings with PBS-Tween 20 buffer, goat antimouse immunoglobulin conjugatedwith alkaline phosphatase (Sigma Chemical Co.)(50 fi\, 1:1000 dilution) was added to thewells and allowed to react for 1 h at 37°C. Subsequently the wells were developed with p-nitrophenyl phosphate substrate (50 fi\, 2 mg/ml solution in diethanolamine buffer, pH 8.1).Positive and negative mouse serum control and conjugate control (antigen coated on platebut no primary antibody) were tested in parallel in all ELISA determinations. Hybrid cellswere considered positive for antibody production only when the observed absorbancereading at 405 nm (using CF-DGM as test antigen) was at least three times over thatobtained with preimmune mouse sera.

Dot Blot AssaysDot blot assays were carried out using a BIO-RAD blotting apparatus according to

the method of Hawkes et al (20). 125 ng of antigen was applied per well and the primaryantibody was used at a dilution of 1:500 followed by addition of F(ab)2 fragment of anti-mouse IgG conjugated to alkaline phosphatase (Sigma Chemical Co.).

287

Characterization of MAbs

The class and subclass of the antibodies were determined by the ELISA methodusing a alkaline phosphatase conjugated isotyping kit (Zymed Laboratories, CA). Also,conformational or sequence specificities of these antibodies were determined by the ELISAand dot blot methods in the presence or absence of dénaturant, SDS.

Denaturation/Reduction of Native and Deglycosylated Mucins

Mucins (native and deglycosylated) were denatured or reduced using SDS or SDScontaining 2-mercaptoethanol, respectively. The effects of these treatments on the bindingproperties of the MAbs were examined. For denaturation experiments, the mucin sample(25 fig/ml) was boiled with 1% SDS (w/v) for 5 min. Prior to coating of the sample on

ELISA plates or nitrocellulose membrane (dot blot assays), the denatured apomucin wasdiluted ten fold with coating buffer (carbonate buffer, pH 9.6 in the case of ELISA or Trisbuffered saline in the case of dot blots). For reduction experiments, the deglycosylatedmucins were first denatured with SDS as described above and subsequently boiled with 2-mercaptoethanol (1.25%). The reduced mucins were also diluted ten fold before carryingout ELISA or dot-blot experiments.

Competitive Binding AssaysThree MAbs (lOe, 29d and 30e) were radioiodinated with 125I (ICN-Biochemicals)

using the iodobead method (Pierce Chemical Co.). The radioiodination was accomplishedby following the protocol provided by the manufacturer. The radioiodinated antibody wasfurther purified using an analytical FPLC desalting column (Pharmacia-LKB). The purifiedlabeled antibodies were examined for their binding property to the CF-DGM antigen usingthe ELISA method. Three different competitive binding experiments (21) were carried outto assess the epitope specificity of MAbs lb, lOe, 29d and 30e. For a given experiment alabeled antibody was allowed to competitively bind to the CF-DGM antigen in the presenceof increasing amounts of unlabeled lb, lOe, 29d and 30e. For these experiments, 125 ngof CF-DGM in 50 ¡i\ of carbonate buffer, pH 9.6, was coated per well of a 96 well polyvinylchloride plate, and incubated at 37°C for 1 h. The samples were then aspirated andwashed with the washing solution (PBS containing 0.05% Tween-20). Subsequently, thewells were incubated at 37°C with 50 fi\ of blocking solution (PBS containing 2% Tween-20). After 1 h, the solutions were aspirated and the wells were washed with the washingsolution. To each well, a 125I-MAb («10,000 cpm) was added in the absence or presenceof unlabeled MAbs lb, lOe, 29d or 30e (10-100 ng). After 2 hours at 37°C, the wells werewashed four times with the washing solution. The individual wells were placed into vialsand counted for radioactivity in a 7-counter.

RESULTS

Purification of Respiratory Mucins

Three CF respiratory mucins were purified usirg a protocol established in our

laboratory (3,12,13). The purified mucins (50 fig each) gave a single broad band on a SDS-composite gel stained with PAS reagent (Fig. 1).No low molecular weight component(s) bands were detected when duplicate gels werestained with either Coomassie Brilliant Blue or the silver stain method (sensitivity ofdetection = 2-5 ng/protein band (17)), indicating the absence of contaminants in the purifiedpreparations. Furthermore, Superóse 6 chromatography of native mucins in buffercontaining 4 M guanidine HC1 gave a single peak in the void volume (13). The absenceof other absorbance peak(s) in the elution profile, in strongly dissociative conditions, furthersupports the purity of mucins used in this work.

288

*

ABCFIGURE 1. Gel electrophoresis of purified cystic fibrosis mucins (50 fig each; Lanes A:CF-6; B:CF-4; GCF-7). Composite gels containing 2% polyacrylamide and 0.5% agarosewere used. Gels were stained with the PAS reagent. In separate experiments duplicategels were stained with coomassie blue or silver staining to detect protein/glycoprotein bands.

Deglycosylation of Purified Respiratory Mucins

Cystic Fibrosis respiratory mucins (CF-4, CF-6 and CF-7) were successfullydeglycosylated by treatment with TFMS and anisóle and subsequent enzymatic treatmentwith a-N-acetylgalactosaminidase. The recovery of protein was usually greater than 90%.A similar observation has been reported by Woodward et aj (8). The deglycosylated mucinswhen examined by gel-permeation chromatography using Superóse 6 (FPLC) gave a broadpeak corresponding to a Mr = 100,000. SDS-gel electrophoresis of the DGM gave a broaddiffuse band when stained by the silver stain method (data not shown).

Chemical Characterization of Purified Native and Deglycosylated Respiratory Mucins

The chemical compositions of CF mucins were determined. In all cases,carbohydrate analyses revealed the presence of fucose, galactose, N-acetylgalactosamine andN-acetylglucosamine. In addition, the mucins contained sialic acid and sulfate. Theabsence of mannose, deoxyribose and uronic acids in all the mucins examined showed thatthe mucins were not contaminated with serum glycoprotein(s), DNA and proteoglycans,respectively. The overall chemical compositions were similar to those reported earlier(3,12,13). In all the deglycosylated mucins, the carbohydrate analyses revealed nodetectable amounts of fucose, galactose, N-acetylglucosamine, N-acetylgalactosamine andsialic acid. Clearly, the deglycosylation protocol used in this work completely removed thecarbohydrate moieties of the mucins. Amino terminal analysis of the deglycosylated mucins

289

showed the absence of free -NH2 terminal amino acid(s), indicating that the -NH2 terminalwas blocked. These data also support that no peptide bond cleavage had occurred duringdeglycosylation.

Amino Acid Composition of Native and Deglycosylated Respiratory Mucins

Amino acid composition of the purified CF mucins are shown in Table I.Comparison of the three CF mucins showed no significant differences in their amino acidcompositions. For the native mucins, threonine, serine, proline, glycine, and alaninecomprised about 72% of the total amino acid residues. Amino acid compositions of thethree deglycosylated mucins (Table I) clearly indicated no notable degradation of thepolypeptide cores of the mucins.

TABLE I: Amino Acid Compositions3 of Native and Deglycosylated Respiratory Mucins

Amino acid Native Mucins Deglycosylated Mucins

CF-6 CF-4 CF-7 CF-6 CF-4 CF-7

Aspartate 30 30 29 28 25 29Threonine 301 289 288 307 310 286Serine 155 143 145 175 158 157Glutamate 37 55 55 57 45 55Proline 137 139 139 115 117 122Glycine 63 63 62 79 76 71Alanine 93 92 86 76 92 90Valine 40 40 40 36 41 41Melhionine 6 7 6 3 6 5Isoleucine 20 19 19 17 20 21Leucine 41 42 41 35 40 40Tyrosine 5 6 8 6 9 8Phenylalanine 10 13 9 9 5 8Hislidine 24 22 21 17 14 20Lysine 12 15 31 17 17 15Arginine 26 25 21 23 25 32

"Data expressed as residues/1000 residues of amino acids.

Production and Characterization of MAbs

When splenic lymphocytes from an immunized mouse were fused with Sp2/0myeloma cells, we obtained thirty hybridoma clones secreting antibodies reactive againstpurified CF-6 DGM. Eight stable hybridomas (la, lb, 10a, 10c, lOd, 10e, 29d, and 30e)were obtained following cloning by the limiting dilution method. To obtain large quantitiesof MAbs (in the form of ascites fluid), each cloned cell line was injected into a separategroup of pristane primed mice. The resulting antibodies were then purified from ascitesfluid using protein-A Superóse column (FPLC, Pharmacia-LKB). The class and subclassof each clone was determined. Six antibodies (la, lb, 10a, 10c, lOd and 10e) belonged tothe IgG, subclass while two antibodies 29d and 30e belonged to the IgM class.

Reaction of MAbs with CF Deglycosylated Mucins

All the eight MAbs la, lb, 10a, 10c, lOd, 10e, 29d and 30e showed a similar reactionwith the three CF deglycosylated mucins. These results are summarized in Table II. Theresults were further confirmed by the dot blot assays (Fig. 2). Interestingly, all eight MAbsfailed to react with native respiratory mucins (Table II).

The effects of SDS-treatment of the respiratory CF mucin antigens on the reactivitywith the MAbs were investigated. Antibodies la, 10a, 10c and lOd failed to react withSDS-treated deglycosylated CF respiratory mucin antigens (Table II). On the other hand,

290

+ + ++ + + + + +

+ + ++ + ++ . + . +

ISs oz n

ilZ Q

tu niS Sm g+ +

ai S+2OC

+ + ++ + + + + +

+ + ++ + + + + +

++ ++ . +

+ ++ ++ ++ , +

Í

a w

+ +

+ + + +1 a j is o s oZ Q Z Q

a «m g+ +

S5 ai ss mQ D Q D

| 23 OZ Q Z D

+++

0\On

++

0\on

oomd

on

«7 &ci ucj 'ST3 "O

ca S21fa oU m

vef t/j¿S a

00a•5

2 D

u nO cö*¿ o

ta oO ai

S <6.-

(N g• • SS -o '

S 2 '

m ^

S«

c(D

.£Ps3

•aniU-4- *-*W3

hr1 Si- "

3 ¿¿5 U

n!CO a

oo

3 DO W5

ilo ZM 1300 cB O

S £« V S.

O u

U Où a aiai 3B § S 3

S

Pu SU E•a ..

u a)IG ,>S 5

3 »

e sT3 '5U Oja Ë>*> ai

o %o ?

oo o<U BQ S

Ü 3

c°*ca>- O O

++

B BO OÜ U

+ g £

m "s >oo -sin U o

-H Z CL,

291

#1

JflBl *3

4

5

6

~Tä ïb 10a 10c 10d Ï0ë^29d 30e" - +

FIGURE 2. Dot-blot assay for the reactivity of MAbs against deglycosylated CF (rows 1:CF-6, 2:CF-4 and 3:CF-7) and native CF (rows 4:CF-6, 5:CF-4 and 6:CF-7) respiratorymucins. For negative (-) control , 100 ¡i\ of Tris buffered saline was added in place of theantigen. For positive (+) control, mouse polyclonal antiserum against CF-6 DGM wasused. Details are provided in Experimental Procedures.

antibodies lb, lOe, 29d and 30e reacted with SDS-treated deglycosylated CF mucins (TableII). These results suggested that MAbs lb, lOe, 29d and 30e appeared to be directedagainst a given sequence of amino acids of the deglycosylated CF mucins while the otherfour MAbs la, 10a, 10c and lOd were, in addition, conformation dependent.

The effects of disulfide bond reduction of the CF mucin antigens on the reactivityof the antigens with the antibodies were also investigated. Antibodies la, 10a, 10c and lOdfailed to react with reduced CF deglycosylated mucin (Table II). In contrast, antibodies lb,10e, 29d and 30e retained their reactivity for the reduced CF antigens (Table II).Interestingly, MAb la showed reactivity towards native mucin antigens upon reduction oftheir disulfide bonds but not with SDS denatured native antigens (Table II).

In order to determine whether MAbs lb, 10e, 29d and 30e (which were notconformation dependent) recognized distinct epitopes on the CF deglycosylated mucin, asolid phase competitive antibody-binding assay was performed. Three different competitivebinding experiments were carried out. As shown in Fig. 3A binding of 125I-10e antibodyto CF deglycosylated mucin coated on polyvinyl chloride plates was inhibited by unlabeledMAb lOe but not by unlabeled MAbs lb, 29d and 30e. Also, binding of 125I-29d antibodyto the coated antigen on the plates was inhibited by unlabeled MAb 29d and not by MAbslb, 10e and 30e (Fig. 3B). Similar results were obtained for the competitive binding assayusing 125I-30e antibody (Fig. 3C). These data indicated that the MAbs lb, 10e, 29d and 30erecognize four distinct epitopes in the peptidic core of CF respiratory mucins.

DISCUSSION

The purification protocol used in this work provided highly purified mucins whichwere totally free of contaminants as judged by gel electrophoresis and FPLC. Absence ofamino sugars in the purified DGMs indicated that the deglycosylation protocol used in thiswork successfully and effectively removed all of the carbohydrate moieties in the mucins.The visualization of deglycosylated mucins as broad diffuse bands upon electrophoresis andprotein staining has been observed earlier (8,22). Since amino acid composition and -NH2terminal analyses of the native and deglycosylated mucins clearly showed no degradation

.

.: '

JÊËt :h:

292

om

Om

20 40 60 eo 100Concentration (ng)

20 40 60 80 100 120

Concentration (ng)

20 40 60 80 100Concentration (ng)

FIGURE 3. Analysis of peptidic epitopes in the CF-DGM using competitive binding RIA.125 ng of deglycosylated mucin was coated on each well of a polyvinyl chloride plate andcoated with ^I-labeled monoclonal antibody along with serially diluted unlabeledcompetitor MAb as described in Experimental Procedures. 125I-labeled MAbs 10e (A), 29d(B) and 30e (C) were used in three independent experiments. Each point represents themean value of duplicate determinations.

of the latter, it is likely that the observed characteristic property of the apomucins isperhaps due to the aggregation of DGM molecules.

Although the amino acid and carbohydrate compositions of CF mucins have beendescribed (3,12,13), the structure of polypeptide core is not well understood. Therefore,in order to learn more about the structures of CF apomucins, we have used eight MAbsas immunological probes for structural studies. It is now agreed that mucin peptide corecontains heavily glycosylated regions (protected from proteolytic attack) and non-

glycosylated (or "naked" regions) that are susceptible to protease digestion. It is interestingto point out that none of the eight MAbs studied in this work showed reactivity towards thenative purified CF mucins. Thus, these antibodies were perhaps not directed against thenaked regions of the mucin structure. In contrast, the polyclonal antibodies raised againstCF-6 deglycosylated mucin (Table II) reacted with the native CF mucins.

Another interesting observation of this work was the finding that SDS-treatment anddisulfide bond reduction of the CF deglycosylated mucins results in the loss of reactivity ofthe four antibodies, suggesting that the antibodies (la, 10a, 10c and lOd) were conformation

293

dependent. It is also possible, however, that SDS blocked the antigenic sites for theseantibodies. Since the MAbs (lb, lOe, 29d, and 30e) did not lose their reactivity for theSDS-treated antigens, these were considered sequence specific but not conformationdependent antibodies. The competitive binding assays demonstrated that these antibodies(lb, lOe, 29d, and 30e) recognize four distinct antigenic determinants on the CF mucinantigen. Earlier work on human intestinal mucins showed that reduction of the disulfidebonds of the mucin antigen resulted in the total loss of the reactivity of the antigen withthe polyclonal antisera (23). In contrast, both SDS and disulfide bond reduction treatmentsdid not result in the loss of reactivity of the treated CF antigens with the polyclonalantisera.

MAbs made against mucus and serous components of rhesus monkey trachea havebeen used to localize secretory products in the respiratory epithelium byimmunohistochemical methods (24). Similarly, MAbs have been used as probes for uniqueantigens in secretory cells of mixed exocrine organs (9). We are currently developingimmunofluorescent procedures for cytochemical localization of the mucins in thetracheobronchial epithelial tissues. The MAbs (lb, lOe, 29d, and 30e) that appeared toreact with different epitopes may also provide a valuable tool for structural probing. Whiledeglycosylated mucins could be used as substrates for investigating mucin carbohydratebiosynthesis, some of these MAbs may also be useful in screening cDNA libraries for clonesproducing different mucin species.

ACKNOWLEDGMENT

This work was supported in part, by a grant from the National Heart, Lung and BloodInstitute (HL34012)

REFERENCES

1. Yeager, H. (1971). Tracheobronchial secretions. Am. J. Med. 50, 493-509.

2. Litt, M., Khan, M. A., Chakrin, L. W., Wardell, J. R. and Christian, P. (1974). Theviscoelasticity of fractionated canine trachéal mucus. Biorheology 11, 111-117.

3. Chace, K. V., Flux, M. and Sachdev, G. P. (1985). Comparison of physicochemicalproperties of purified mucus glycoproteins isolated from respiratory secretions ofcystic fibrosis and asthmatic patients. Biochemistry 24, 7334-7341.

4. Feldhoff, P. A, Bhavanandan, V. P. and Davidson, E. A. (1979). Purification,properties, and analysis of human asthmatic bronchial mucin. Biochemistry 18, 2430-2436.

5. Carlstedt, I., Sheehan, J. K., Corfield, A. P. and Gallagher, J. T. (1985). Mucousglycoproteins: a gel of a problem. Essays in Biochemistry 20, 41-76.

6. Gerken, T. A., Butenhof, K. J. and Shogren R. (1989). Effects of glycosylation onthe conformation and dynamics of O-linked glycoproteins: carbon-13 NMR studiesof ovine submaxillary mucin. Biochemistry 28, 5536-5543.

7. Breg, J., Van Halbeek, H., Vliegenthart, J. F. G., Lamblin, G., Houvenaghel, M-C.and Roussel, P., (1987). Structure of sialyl-oligosaccharides isolated from bronchialmucus glycoproteins of patients (blood group O) suffering from cystic fibrosis. Eur.J. Biochem. 168, 57-68.

8. Woodward, D. H., Ringler, N. J., Selvakumar, R., Simet, I. M., Bhavanandan, V. P.and Davidson, E. A. (1987). Deglycosylation studies on trachéal mucin glycoproteins.Biochemistry 26, 5315-5322.

294

9. Basbaum, C. B., Mann, J. K., Chow, A. W. and Finkbeiner, W. E. (1984).Monoclonal antibodies as probes for unique antigens in secretory cells of mixedexocrine organs. Proc. Nati. Acad. Sei. USA 81, 4419-4423.

10. Podolsky, D. K., Fournier, D. A. and Lynch, K. E. (1986). Development of anti-human colonie mucin monoclonal antibodies. Characterization of multiple coloniemucin species. /. Clin. Invest. 11, 1251-1262.

11. Podolsky, D. K., Fournier, D. A. and Lynch, K. E. (1986). Demonstration of distinctsubpopulations defined by mucin-specific monoclonal antibodies. J. Clin. Invest. 11,1263-1271.

12. Chace, K. V., Naziruddin, B., Desai, V. C, Flux, M. and Sachdev, G. P. (1989).Physical properties of purified human respiratory mucus glycoproteins: effects ofsodium chloride concentration on the aggregation properties and shape. Exptl. LungRes. 15, 721-737.

13. Shankar, V., Naziruddin, B., Reyes de la Rocha, S. and Sachdev, G. P. (1990).Evidence of hydrophobic domains in human respiratory mucins. Effect of sodiumchloride on hydrophobic binding properties. Biochemistry 29, 5856-5864.

14. Svennerholm, L. (1975). Quantitative estimation of sialic acids II. A colorimetricresorcinol hydrochloric acid method. Biochim Biophys Acta 24, 604-611.

15. Wang, C. S., Burns, R.K., Alaupovic, P. (1974). Isolation and composition ofoligosaccharide cores from endotoxins of two Serratia marcescens strains. J. Bacteriol.120, 990-993

16. Holden, K. G., Yim, N. C. F., Griggs, L. J. and Weisbach, J. D. (1971). Gelelectrophoresis of mucous glycoproteins. II. Effect of physical deaggregation anddisulfide-bond cleavage. Biochemistry 10, 3110-3113.

17. Merril, C.R., Goldman, D. and Van Keuren, M.L. (1984). Gel protein stains: Silverstain. Meth. Enzymol. 104, 441-447.

18. Walsh, K.A., Ericsson,L.H., Parmelee, D.C. and Titani, K. (1981). Advances inprotein sequencing. Annu. Rev. Biochem. 50, 261-284.

19. Cunningham, M. W., Krisher, K. and Graves, D. C. (1984). Murine monoclonalantibodies reactive with human heart and group A streptococcal membrane antigens.Infect. Immun. 46, 34-41.

20. Hawkes, R., Niday, E. and Gordon, J. (1982). A Dot-immunoblotting assay formonoclonal and other antibodies. Anal. Biochem. 119, 142-147.

21. Shen, R. F., and Tai, H. H. (1986). Immunoaffinity purification and characterizationof thromboxane synthase from porcine lung. J. Biol. Chem. 261, 11585-11591.

22. Marianne, T., Perini, J.M., Houvenaghel, M.C., Tramu, G., Lamblin, G. and Roussel,P. (1986). Action of trifluoromethanesulfonic acid on highly glycosylated regions ofhuman bronchial mucins. Carbohydr. Res. 151, 7-19.

23. Mantle, M., Forstner, G. G. and Forstner, J. F. (1984). Antigenic and structuralfeatures of goblet-cell mucin of human small intestine. Biochem. J. 217, 159-167.

24. St. George, J. A, Cranz, D. L., Zicker, S. G, Etchison, J. R., Dungworth, D. L. andPlopper, C. G. (1985). An immunohistochemical characterization of rhesus monkey

295

respiratory secretions using monoclonal antibodies. Am. Rev. Respir. Dis. 132, 556-563.

Address reprint requests to:Dr. G. P. SachdevCollege of PharmacyUniversity of OklahomaHealth Sciences CenterP.O.Box 26901Oklahoma City, OK 73190

Received for publication August 24, 1990Accepted November 8, 1990

296