Partial Purification and Characterization of Mold Commonly · Double-sandwich ELISA. The...

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Aug. 1993, p. 2563-25710099-2240/93/082563-09$02.00/0Copyright X) 1993, American Society for Microbiology

Partial Purification and Characterization of Mold AntigensCommonly Found in Foodst

G.-J. TSAIt AND M. A. COUSIN*Department ofFood Science, Purdue University, West Lafayette, Indiana 47907-1160

Received 17 August 1992/Accepted 8 June 1993

Rapid methods are needed for detection of molds in foods; therefore, an enzyme-linked immunosorbent assay

was developed. The extracellular and mycelial antigens for Mucor, Aspergillus, Cladosporium, and Geotrichumspecies were partially purified and characterized. The molecular masses of the mycelial and extracellularantigens, as determined by size exclusion chromatography, ranged from 4.5 x 105 to 6.7 x 105 Da. There wasonly one main antigenic peak separated by Sepharose CL-4B and concanavalin A-Sepharose columns forMucor, Cladosporium, and Geotrichum mycelial and extracellular antigens, but there were two for Aspergilusmycelial antigens and three for Aspergillus extracellular antigens. These antigens contained 10 to 50%o proteinwhich was part of the active site since protease digestion significantly decreased antigenic activity. Neutralsugars, ranging from 13 to 75%, made up the rest of the active site, and <1% phosphate was detected inmycelial antigens. Geotrichum, Cladosporium, and Aspergillus antigens contained mainly glucose, galactose,and mannose. Mucor antigens contained these sugars plus fucose. The percentage of sugars differed betweenthe mycelia and extracellular antigens. Enzymatic digestion and competitive inhibition tests using differentsugar derivatives showed that galactosyl residues with 1 linkages were immunodominant for Aspergilus,Geotrichum, and Cladosporium antigens and mannosyl residues with a linkages were immunodominant forMucor antigens.

Enzyme immunoassays with 1-,ug/g to 10-ng/g detectionlimits have been developed for detection of molds in foods(25, 26, 33-35, 41, 49). Many of these assays were specificfor an individual genus; however, some cross-reactionsbetween Penicillium and Aspergillus (34, 49) and Mucor andRhizopus (25, 34, 49) species occurred because they are inrelated families. Some molds, such as Alternaia alternata,Leptosphaerulina briosiana, and Epicoccum spp., that are

taxonomically nonrelated also showed cross-reaction in an-

tigen-antibody reactions (25). Since the immunological spec-ificities of antigens in the immunoassays reside in theirstructures, it is important to investigate the immunologicalstructures of the fungal antigens to improve the specificitiesof assays when detection of a specific genus or species isdesired.

Polysaccharides found in fungal mycelia are immunologi-cally active (5, 11, 31, 34, 42, 48). Galactose, mannose, andglucose are the dominant sugars in these fungal polysaccha-rides (4, 10, 48); therefore, cross-reactions of the antigen-antibody among mold species might be due to the presenceof these common sugars. Grappel et al. (12-14) isolatedgalactomannans from Trichophyton species that could reactwith the antibody against Microsporum quinckeanum. Schu-macher et al. (44) injected rabbits with polysaccharidesisolated from Alternaria tenuis, and the subsequent antibod-ies reacted with A. fumigatus, Stemphylium spp., and Cur-vularia spp. Suzuki and Takeda (48) isolated galactomann-ans from pathogenic A. fumigatus, A. niger, T. rubrum, andCladosporium wemeckii that reacted with anti-Hormoden-drum pedrosoi serum in the double-diffusion agar gel assay.

* Corresponding author.t Journal paper 13512 of the Purdue University Agricultural

Experiment Station.1: Present address: Department of Marine Food Science, National

Taiwan Ocean University, Keelung 202, Taiwan, Republic of China.

Hessian and Smith (16) found that polysaccharide antigensfrom pathogenic fungi such as A. corymbifera, Mortierellawolfii, Mucor spp., and Rhizopus spp. share carbohydratesin common.Molds can secrete extracellular antigens (EA) into the

environment (19, 22, 28, 37, 40, 43, 44, 46). Lloyd and Bitoon(28) obtained a serologically active antigen, a peptidorham-nomannan, from the cultural medium of Sporothraischenckii. Schumacher et al. (44) found that antigenic deter-minants were shared by cultural filtrate antigens and myce-lial antigens of A. tenuis. Notermans and Heuvelman (34)noted that the cultural filtrate antigens of P. verrucosum var.

cyclopium, M. racemosus, and F. oxysporum were genusspecific, heat stable, and not present in nonmoldy foods.Fungal proteins or peptides may also be important in

antigenic components. Heam and Mackenzie (15) found thatthe immunological activities of antigens isolated from T.rubrum, Petriellidium boydii, M. pusillus, and A. fumigatuswere almost completely lost after digestion with pronase for18 h. Longbottom and Austwick (29) further characterizedfungal antigens as glycopeptides containing polysaccharideswith small amounts of peptides.

Before further development of a rapid immunoassay formold detection can continue, more information on the im-munodominant components of mold antigens is needed.1-D-Galactofuranose has been identified as the immuno-dominant sugar residue for Aspergillus and Penicillium ex-

tracellular antigens (7, 36). Mannose and fucose were theimmunodominant sugars in both Mucor and Rhizopus extra-cellular antigens (36). Kamphuis et al. (18) used this infor-mation on immunodominant residues in Aspergillus andPenicillium extracellular antigens to develop a latex aggluti-nation assay that included a blocking agent to preventfalse-positive reactions. Since there is little information on

the composition of antigens from both the mycelial andextracellular constituents of the same mold, the objectives of

2563

Vol. 59, No. 8

on March 20, 2021 by guest

http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/

2564 TSAI AND COUSIN

this research were to (i) prepare mold antigens from bothmycelial and culture media for Aspergillus, Cladosponium,Geotrichum, and Mucor species, (ii) partially characterizethese antigens for chemical compositions, and (iii) study theimmunologically active sites by selective removal of resi-dues by enzymatic degradation.

MATERIALS AND METHODS

Growth of molds. Mucor circinelloides NRRL 3614 andGeotnchum candidum NRRL Y-552 (from the U.S. Depart-ment of Agriculture Northern Regional Research Labora-tory, Midwest Area Center for Agricultural Utilization Re-search, Peoria, Ill.) and Aspergillus versicolor ATTC 44605and Cladosporium herbarum ATTC 28987 (from the Amer-ican Type Culture Collection, Rockville, Md.) were used toimmunize New Zealand White rabbits and to prepare anti-gens. All strains were grown on slants of potato dextroseagar (Difco Laboratories, Detroit, Mich.) at 22°C for 5 to 7days, and spores were washed from the surface with 10 ml ofsterile deionized water. Flasks of 250 ml of brain heartinfusion broth (Difco) were inoculated with 0.1 ml of thespore suspension and agitated at 140 rpm at 24°C for 7 (M.circinelloides, G. candidum, and A. versicolor) or 14 (C.herbarum) days. Mold mycelia were harvested by filtrationthrough four layers of cheesecloth. Yeast-type cells wereharvested by centrifugation at 3,000 x g for 15 min. Aftermycelia or cells were washed with deionized water fourtimes, the fungi were steamed for 20 min, freeze-dried, andground in a Wig-L-Bug (Crescent Dental ManufacturingCompany, Chicago, Ill.) for 2 min.

Isolation of mycelial antigens. Dried mold mycelia (120 mg)were stirred in 40 ml of phosphate-buffered saline with 0.05%Tween (PBST; 0.2 g of KH2PO4, 1.15 g of Na2HPO4, 0.2 g ofKCl, 8 g of NaCl, and 0.5 g of Tween 20 in 1 liter of deionizedwater) at 37°C for 3 h. This mixture was centrifuged at 9,820x g for 20 min, and the supernatants were collected anddialyzed against three separate changes of 1 liter of deion-ized water at 5°C overnight. The dialysate was concentratedto 2 ml with a rotary evaporator set at 45°C, and the resultingconcentrate was applied to a Sepharose CL-4B-200 column(2.5 by 100 cm; bed volume, 450 ml) that was equilibratedwith deionized water. Elution was done with deionizedwater at a rate of 15 ml/h. Tubes of 6-ml fractions werecollected and analyzed by both UV light at 280 nm andenzyme-linked immunosorbent assay (ELISA). The mainantigenic peak was collected and concentrated and furtherpurified with a concanavalin A (ConA)-Sepharose column(1.0 by 35 cm; bed volume, 25 ml) equilibrated with 0.08 MTris buffer (pH 7.2) containing 0.8 mM (each) CaCl2 andMnCl2. The same buffer was used as the mobile phase witha flow rate of 15 ml/h. Three-milliliter fractions were col-lected in tubes. After collection of 20 to 50 tubes, 0.2 Mmethyl-a-D-mannopyranoside in the above-described buffer(pH 7.2) was used to elute the bound materials. The fractionswere analyzed by ELISA, and the main positive peaks werecollected, dialyzed against three changes of 1 liter of deion-ized water at 5°C overnight, and freeze-dried.

Isolation of EA from culture medium. After molds weregrown in brain heart infusion broth at 24°C for 7 (M.circinelloides, G. candidum, and A. versicolor) or 14 (C.herbarum) days, the culture fluids were separated from themycelia by filtration through no. 2 filter paper (Whatman,Clifton, N.J.) and then through a 0.45-,um-pore-size mem-brane (Gelman Sciences, Ann Arbor, Mich.). This filtrate

was freeze-dried. A 2-g portion of freeze-dried powder wasextracted with 40 ml of 80% saturated ammonium sulfate bystirring for 2 h at room temperature. The mixture was thenfiltered through Whatman no. 2 filter paper. After the filtratewas dialyzed against deionized water at 5°C overnight, it wasconcentrated and applied to a Sepharose CL-4B-200 columnand then a ConA-Sepharose column. The procedure usedwas that described above.

Determination of antigen molecular masses. Molecularmasses of antigens were estimated by the gel filtrationSepharose CL-4B-200 column used for isolation of fungalantigens. A standard solution containing 2 mg of blue dext-ran 2000 (2,000,000 Da), 20 mg of thyroglobulin (669,000 Da),20 mg of ferritin (440,000 Da), 20 mg of catalase (232,000Da), and 30 mg of aldolase (158,000 Da) was applied to thecolumn under the conditions used for the antigen separationdescribed in the previous paragraph. Each tube was ana-lyzed by the phenolic-sulfuric acid method for neutral sugars(8). The molecular masses of these antigens were estimatedfrom the linear relationship between the logarithmic molec-ular mass of the glycoprotein standard versus the elutedvolume (Ve) divided by the void volume (V0).

Antiserum production and immunoglobulin G purification.Antisera to the EA or to the mycelial antigens of the differentmold species were produced individually by injecting NewZealand White female rabbits intramuscularly with 0.45 mgof EA or 12 mg of fungal mycelia as described by Tsai andCousin (49).

Double-sandwich ELISA. The double-sandwich ELISAprocedure of Tsai and Cousin (49) was used to determineantigenic activities in samples. Briefly, 200 RI of rabbitimmunoglobulin G diluted in a carbonated buffer at pH 9.6was added to the wells of microtiter plates (DynatechLaboratories, Chantilly, Va.) and incubated at 37°C for 2 h.After each well was washed (ELISA Washer; Bio-TekInstruments, Burlington, Vt.) four times with PBST, a 200-,lsample was added to each well and the plates were incubatedat 37°C for 4 h. After each well was washed four times, 200RI of antibody-horseradish peroxidase conjugates was addedto each well and the plates were incubated at SOC overnight.After the wells were washed four times with PBST, 100 ml of5'-aminosalicylic acid (0.8 mg/ml [pH 6.0] with 0.05% H202)was added and the plates were incubated at room tempera-ture for 30 min. Fifty microliters of 1 N NaOH was added toeach well, and the A450 of each well was measured by usingan ELISA reader (Bio-Tek).

Neutral sugar measurement. The phenolic-sulfuric acidmethod of Dubois et al. (8) with D-glucose as the standardwas used to measure the neutral sugar contents of thelyophilized antigens. In this procedure, 0.25 ml of 5% phenolwas added to 0.5 ml of either the standard or the sampleand the solution was mixed well. A 1.5-ml volume ofconcentrated sulfuric acid was then added. After 20 min ofincubation, the color was measured at 490 nm with aspectrophotometer (Lambda 3B; Perkin-Elmer, Norwalk,Conn.).

Protein measurement. A protein assay kit (Sigma ChemicalCo., St. Louis, Mo.) was used to analyze the proteincontents of the antigens. Bovine serum albumin (BSA) wasused as the standard. A volume of antigen solution (2 mg/mlin deionized water) or the BSA standard (400 p,g/ml indeionized water) to give the desired concentration wasadded to a test tube and diluted to a volume of 1.0 ml withdeionized water. After addition of 1 ml of Lowry Reagent(Sigma Chemical Co.) solution, the mixture was left at roomtemperature for 20 min. A 0.5-ml volume of Folin and

APPL. ENvIRON. MICROBIOL.


http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/

MOLD ANTIGENS IN FOODS 2565

Ciocalteu phenol reagent was then added to the solution andmixed immediately. The tubes were incubated for 30 min atroom temperature, and the color was measured at 750 nmwith a spectrophotometer (Perkin-Elmer).

Phosphate determination. The antigen sample (1 mg/0.1 mlof distilled deionized water) was ashed by the method ofAmes and Dubin (2). The inorganic phosphate content wasthen determined by a modification of the procedure ofMurphy and Riley (32). One milliliter of Murphy and Rileyreagent (7 parts of 2.88 N H2SO4, 1 part of 4.8% [wt/vol]ammonium molybdate, 1 part of 0.11% [wt/vol] potassiumantimonyl tartrate, and 1 part of 4.22% [wt/vol] ascorbic acidwere combined and prepared fresh daily) was added to thetube containing the ashed antigen sample and diluted withdistilled deionized water to 5 ml. After incubation at roomtemperature for 20 min, the solution was read at 700 nm(Perkin-Elmer spectrophotometer). At the same time, phos-phate standard solutions (prepared from dried KH2PO4heated at 95°C in an oven for 6 h and cooled in a desiccator)of 2 to 20,g/5 ml and a water blank were assayed to developa standard curve.Sugar analysis by gas chromatography. The sugar compo-

sitions of the antigens were analyzed by gas chromatographyas described by Albersheim et al. (1), with the followingmodifications. One milliliter of 2 N trifluoroacetic acid wasused to hydrolyze the antigens containing approximately 0.1mg of sugar, as estimated by the phenolic-sulfuric acidmethod. Hydrolysis was done at 121°C for 1 h with 0.1 mg ofmyo-inositol added as an internal standard. Standard sugarsolutions containing 0.1 mg each of arabinose, fucose, ga-lactose, glucose, mannose, rhamnose, ribose, and xylose perml were also hydrolyzed and analyzed with a gas chromato-graph (model 3400; Varian Associates, Palo Alto, Calif.).The hydrolysate was checked by thin-layer chromatographyon an aluminum silica gel plate with ethyl acetate-aceticacid-water (3:2:1) as the mobile phase to ensure that thesample was completely hydrolyzed. The sample was evapo-rated to dryness at 40°C under a stream of filtered nitrogengas. After consecutive addition of 0.1 ml of 1 N NH40H and0.5 ml of NaBH4 (10 mg/ml in 1 N NH40H) and incubationat room temperature for 2 h, the excess borohydride wasdecomposed by addition of glacial acetic acid until efferves-cence had ceased. The resulting solution was evaporated todryness under N2 at 40°C. After repeating the addition of 1ml of methanol followed by evaporation under N2 five times,0.2 ml of pyridine and 0.2 ml of acetic anhydride were addedand the sample was heated at 121°C for 30 min. Acetylationwas stopped by addition of 0.5 ml of deionized water. Alditolacetates were extracted with 0.5 ml of chloroform. Thechloroform phase was then dried over anhydrous MgSO4and evaporated to dryness under N2. A 50-pl volume ofacetone was added before the gas chromatographic analysis.Samples of 0.5 to 1 p,l were injected into an OV-5 glasscapillary column (0.246 mm by 30 m; P. J. Cobert, Inc., St.Louis, Mo.) with a split ratio of 1:100. The column temper-ature was programmed from 180 to 192°C at 2°C/min, in-creased to 198°C for 1 min, increased to 210°C at 2°C/min,and kept at 210°C for 2 min. The total running time was 20min. The temperatures were 275°C for the flame ionizationdetector and 225°C for the injector. The linear velocity of H2in the column was 45 cm/s. The flow rates of gases to thedetector were 30 cm3/min for hydrogen, 30 cm3/min forcarrier gas, and 300 cm3/min for air.The retention time of each individual peak in the sample

was compared to the peaks of the standard solutions. Whenthe sugar was identified, then a known amount of that sugar

was added to the sample and reinjected into the gas chro-matograph to see whether the peak would increase in area. Ifit did, then the sugar was correctly identified.The concentrations of sugars in the sample were calcu-

lated by using the following equations: A = 0.1-mg/mlstandard sugar peak area in standard solution/0.1-mg/mlmyo-inositol peak area in standard solution, B = x-mg/mlsugar peak area of sample/0.1-mg/ml myo-inositol peak areain sample, x mg/ml = (B/A) x 0.1 mg/ml, and % (weightbasis) sugar (1) = xl/l1 + x2 + x,) (x in the denominatordepends on the number of sugars found in the sample).Enzymatic analysis. Various enzymes were used for selec-

tive cleavage of the immunodominant groups in the antigens.Individual freeze-dried antigenic powders at 5 mg/ml in 50mM phosphate buffer (pH 6.5) plus 10 mM NaCl were addedwith the same volume of a solution of protease (Streptomy-ces gniseus, type XIV; 4.5 U/ml of sample; Sigma ChemicalCo.) or lipase (wheat germ; 4.5 U/ml of sample; MannResearch Laboratories, New York, N.Y.) or buffer. A fewdrops of toluene were added to prevent microbial growth.The samples were incubated at 37°C for 24 h, and then thesolutions were heated for 5 min to inactivate the enzymes.The protease digests were dialyzed against 0.1 M NaClovernight at 5°C and then against the appropriate buffer forthe glycolytic enzyme: 50 mM acetate buffer (pH 5.0) fora-mannosidase (23); 50 mM citrate-phosphate buffer (pH5.0) for L-fucosidase (50), cellulase (21), and ,-mannosidase(47); 50 mM citrate-phosphate buffer (pH 4.0) for a- and,B-galactosidases (3); 50 mM phosphate buffer (pH 6.5) fora-amylase (6), ,B-glucosidase (17), and a-glucosidase (asindicated by Sigma Chemical Co.) overnight at 5°C. Afteraddition of a-amylase (EC 3.2.1.1; Bacillus sp., type II-A; 20U of sample; Sigma Chemical Co.), cellulase (EC 3.2.1.4;Tnichoderma viride, type VI; 3 U/ml of sample; SigmaChemical Co.), or ,B-glucosidase (EDED 3.2.1.21; almonds,type I; 8.8 U/ml of sample; Sigma Chemical Co.) and a fewdrops of toluene, the solutions were incubated at 37°C for 48h. After heating the enzymes for 5 min to inactivate them,the pH was adjusted to 5.0 with 50 mM sodium acetate bufferbefore addition of a-mannosidase (EC 3.2.1.24; almonds; 0.9U/ml of sample; Sigma Chemical Co.) or L-fucosidase (EC3.2.1.51; beef kidney; 0.52 U/ml of sample; BoehringerMannheim Biochemicals, Indianapolis, Ind.) or the pH wasadjusted to 4.0 with 50 mM citrate buffer before addition ofa-galactosidase (EC 3.2.1.22; A. niger; 1.5 U/ml of sample;Sigma Chemical Co.) or 3-galactosidase (EC 3.2.1.23; A.niger, 1.4 U/ml of sample; Sigma Chemical Co.). Again, afew drops of toluene were added. The samples were incu-bated at 37°C for 48 h and then heated for 5 min. After eachindividual enzyme reaction, the double-sandwich ELISAwas used to compare the relative activities of samples withand without that enzyme.

Competitive ELISA. Methyl glycosides (methyl-a-D-gluco-pyranoside, methyl-3-D-glucopyranoside, methyl-a-D-man-nopyranoside, methyl-a-D-galactopyranoside, and methyl-P-D-galactopyranoside [Sigma Chemical Co.]) were used asinhibitors in the following competitive ELISA. Two hundredmicroliters of carbonate buffer containing antibody-produc-ing molds at 0.1 mg/ml was added to the wells of microtiterplates that were incubated overnight at 5°C. After the plateswere washed four times with PBST, 100 ,ug of the optimalantibody concentration (determined for each mold) diluted inPBST plus 1% BSA was added. One-hundred-microlitervolumes of various concentrations of the inhibitors wereadded to the wells immediately. The plates were incubated at37°C for 2 h. After the plates were washed four times with

VOL. 59, 1993


http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/


TABLE 1. Molecular masses of mycelial and extracellular moldantigens estimated by size exclusion chromatography'

Molecular mass(es) (Da, 105)Antigen

Mycelial Extracellular

Aspergillus sp. 6.3 6.5, 5.0, 4.5bCladosponum sp. 6.7 6.5Geotnchum sp. 6.5 6.5Mucor sp. 6.7 6.5

a Standard glycoproteins (thyroglobulin, ferritin, catalase, and aldolase)were used as molecular mass markers.

b Three major peaks were isolated from the extracellular filtrate.

PBST, 200 pl of goat anti-rabbit immunoglobulin G-peroxi-dase conjugate (Boehringer Mannheim) diluted 1:5,000 inPBST plus 1% BSA was added and the plates were incubatedfor another 2 h at room temperature. The subsequent sub-strate reaction was the same as in the double-sandwichELISA. Percent ELISA activity was calculated as 100 x(ELISA reading with inhibitor/ELISA reading without inhib-itor). The inhibitory concentration that caused 50% extinc-tion (I50) was obtained by plotting the percent ELISAactivity versus the logarithmic concentration of the inhibitortested.

RESULTS

Isolation and purification of antigens. The main ELISA-positive peak occurred earlier than the protein peak forfungal extracellular antigens from M. circinelloides, G. can-didum, and C herbarum; however, there were three majorELISA-positive peaks for A. versicolor (data not shown).The molecular mass of each of these antigenic fractions was6.5 x 10 Da (Table 1), except for the two peaks for A.versicolor antigens that had lower molecular masses (Table1). The main ELISA peak for A. versicolor had a molecularmass of 4.5 x 105 Da. Fractions from M. circinelloides, G.candidum, and C. herbarum were easily eluted; however,fractions from A. versicolor were not absorbed by ConA-Sepharose (data not shown).PBST extracts of mycelia from M. circinelloides, G.

candidum, C. herbarum, and A. versicolor were not sepa-rated from the protein peak by Sepharose CL-4B-200 col-umns but were separated by ConA-Sepharose (data not

shown). Mycelial antigens from A. versicolor produced twodifferent peaks, one absorbed by ConA-Sepharose and theother unabsorbed. The molecular masses of the mycelialantigens ranged from 6.3 x 105 to 6.7 x 1i0 Da (Table 1).Chemical analysis of antigens. The gross compositions of

the partially purified antigens are shown in Table 2. Theneutral sugar contents were dominant in all antigens regard-less of whether they were mycelial or extracellular, exceptfor the ConA-unabsorbed mycelial antigens from A. versi-color, in which protein was dominant (Table 2). For Mucor,Geotrichum, and Cladosporium spp., the mycelial and ex-

tracellular antigens had similar neutral sugar and proteincontents; however, small amounts of phosphate were de-tected in the mycelial antigens but not in the extracellularantigens.Mannose, glucose, and galactose were the main neutral

sugars found in all antigens (Table 3). Antigens from M.circinelloides also contained fucose (Table 3). Mannose was

the major sugar in all antigens except the ConA-unabsorbedmycelial antigen ofA. versicolor, in which galactose was themajor sugar. The three antigens from the extracellular filtrateofA. versicolor had sugar compositions similar to that of theConA-absorbed mycelial antigen (Table 3). Both the extra-cellular and mycelial antigens of C. herbarum and G. candi-dum had similar sugar compositions (Table 3). For M.circinelloides, the fucose content was higher and the glucosecontent was lower in the extracellular antigens but not in themycelial antigens (Table 3).

Enzymatic analyses of the immunodominant structures ofthe antigens. Various enzymes were used to evaluate theimmunodominant part of the antigens by determining loss ofactivity after enzymatic action. The relative activities, as

measured by the double-sandwich ELISA, for antigens afterdigestion with protease and lipase at 37°C for 24 h are

presented in Table 4. Relative activities for all antigens weresignificantly decreased after protease digestion but not afterreaction with lipase.The ELISA activities for M. circinelloides mycelial anti-

gens after treatment with glycohydrolytic enzymes werealmost the same as those of the protease-digested controls,except for the samples treated with a-amylase or cellulase(Fig. 1A). a-Mannosidase reduced the ELISA activity onlywhen protease digests had been reacted further with a-amy-lase or cellulase before addition of a-mannosidase, but,-glucosidase had no effect (Fig. 1B to D). L-Fucosidase didnot decrease any ELISA activity. For G. candidum, after

TABLE 2. Percentages of neutral sugar, protein, and phosphate in freeze-dried mycelial and extracellular mold antigens

Mycelial ExtraceliularAntigen

% Neutral sugar % Protein % Phosphate % Neutral sugar % Protein % Phosphate

Aspergillus sp.ConA unabsorbed 13.3 50.0 NA4ConA absorbed 48.3 23.3 0.79Peak I 39.4 12.0 NDbPeak II 38.2 16.6 NDPeak III 35.0 16.6 ND

Cladosponium sp. 62.0 17.0 0.60 59.0 20.0 NDGeotrichum sp. 75.0 16.6 0.28 46.6 10.0 NDMucor sp. 66.6 28.3 0.45 71.0 25.0 ND

a NA, not analyzed.bND, not detectable.



http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/


TABLE 3. Percentages (weight basis) of neutral sugars in freeze-dried mycelial and extracellular mold antigensa

Mycelial ExtracellularAntigen

Man Gic Gal Fuc Man Gic Gal Fuc

Asperillus sp.ConA unabsorbed 12.1 38.7 49.2 0ConA absorbed 54.2 7.6 38.2 0Peak I 62.3 6.8 30.9 0Peak II 59.4 6.3 34.3 0Peak III 63.8 7.1 29.1 0

Cladosporium sp. 72.8 7.6 19.6 0 72.7 7.8 19.5 0Geotrichum sp. 71.5 9.4 19.1 0 62.2 11.5 26.3 0Mucor sp. 53.8 32.4 8.6 5.2 61.6 10.8 13.9 13.7

a Total main neutral sugars are reported on the basis of 100%. Man,mannose; Glc, glucose; Gal, galactose; Fuc, fucose.

the protease digest had been degraded by ,3-glucosidase, theP-galactosyl residues decreased (Fig. 2D). When cellulasewith endo-I3-1-4-glucolytic activity was added to the pro-tease digest from G. candidum, the endo-,-glucosyl residuesthat linked to galactosyl residues were cleaved, resulting in adecrease of antigenic activity (Fig. 2A). In addition, sincecommercial ,3-galactosidase contains some a-D-glucosidase,P-D-glucosidase, and 1,4-3-mannosidase (3), some maskinggroups in the freeze-dried antigens may be removed by theseenzymes. Thus, the relative P-galactosidase activity showedsome increase after ,-galactosidase digestion that could bedue to these and other enzymes (Fig. 2A).For A. versicolor antigens, the ELISA activity was re-

duced after P-galactosidase action on the protease digest(Fig. 3A). However, the reduction of ELISA activity byP-galactosidase was greater after the protease digest hadbeen treated with at-amylase (Fig. 3B). For C. herbarumantigens, the ELISA activity was greatly reduced by thefunction of cellulase or 3-glucosidase (Fig. 4A). Part of thegalactosyl units was directly removed in the protease digestby 3-galactosidase. After treatment of the protease digest bycellulase or 3-glucosidase, more galactosyl units were re-moved by P-galactosidase, which caused a further decreasein ELISA activity.

I50 values for competitive inhibition tests using methylglycosides as inhibitors in the ELISA are presented in Table5. Methyl-,3-D-galactopyranoside gave the smallest '50 val-ues for Aspergillus, Cladosponum, and Geotrichum anti-gens. For Mucor mycelial antigens, methyl-a-D-mannopyr-anoside gave the smallest 50' With Mucor extracellular

TABLE 4. Relative activity of mycelial and extracellular moldantigens after reaction with protease or lipase at 37°C for 24 h

Relative activity'

Antigen Mycelial Extracellular

Protease Lipase Protease Lipase

Aspergillus sp.b 0.55 0.94 0.46 0.96Cladosponum sp. 0.53 0.80 0.64 0.85Geotrichum sp. 0.62 0.90 0.66 0.93Mucorsp. 0.36 1.02 0.41 1.07

a Relative activity = ELISA reading of antigen with addition of enzyme/ELISA reading of antigen without addition of enzyme.

b Mycelial antigen with ConA affinity.

1.2

> 0.8

w 0.6

8 0.4

0.2

G2 MI M2 Gal Ga2 AENZYME

0.8 0.8 0.8

LU 0.6 0.6 0.6

wU 04A 0.4 -0.4-

0.2-0.2-0.2-

PA PAM PAF PC PCM PCF PG PGM PGFENZYME ENZYME ENZYME

FIG. 1. Relative ELISA activity of M. circinelloides mycelialantigen treated with protease (A), protease followed by a-amylase(B), protease followed by cellulase (C), or protease followed byP-glucosidase (D) and then treated with various glycohydrolyticenzymes as listed below. Panel A: P, protease; Gl, a-glucosidase;G2, P-glucosidase; Ml, a-mannosidase; M2, P-mannosidase; Gal,a-galactosidase; Ga2, 3-galactosidase; A, ax-amylase; C, cellulase;F, a-fucosidase. Panel B: PA, protease followed by a-amylase;PAM, protease then a-amylase then a-mannosidase; PAF, proteasethen a-amylase then a-fucosidase. Panel C: PC, protease followedby cellulase; PCM, protease then cellulase then a-mannosidase;PCF, protease then cellulase then ot-fucosidase. Panel D: PG,protease followed by 3-glucosidase; PGM, protease then P-glucosi-dase then ot-mannosidase; PGF, protease then f-glucosidase thena-fucosidase.

antigens, '50 values were lower for methyl-3-D-glucopyrano-side than for methyl-et-D-mannopyranoside.

DISCUSSION

In this study, the antigens produced either myceliumbound or extracellularly by A. versicolor, C. herbarum, G.candidum, and M. circinelloides were partially purified andcharacterized. Since the antigenic fractions produced inresponse to molds are generally genus specific, it is impor-tant to characterize the various mold antigens before devel-oping commercial ELISA kits to detect molds in foods. Themolecular mass of the antigen depends on the separation andextraction methods plus the genus, species, and age of theculture. The molecular masses reported for these moldantigens were within the range reported for Penicilliumcharlesii of 2.5 x 104 to 6.5 x 104 (9) to 104 to 106 Da for M.

VOL. 59, 1993


http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/


1.2

5°0.8

wU 0.6

50.4

0.2

0P Gl G2 Ml M2 Gal Ga2 A C

ENZYME

1.2-1.2-12-

(B) (C) 1.2 (D)

~0.8 0.8 0.8

oL0.6 0.6 0.6

wU UO 0.4 0.4A

0.2 0.2 0.2

~~0 0

PA PAM PAGa PC PCM PCGa PG PGM PGGaENZYME ENZYME ENZYME

FIG. 2. Relative ELISA activity of G. candidum mycelial anti-gen treated with protease (A), protease followed by a-amylase (B),protease followed by cellulase (C), or protease followed by ,B-glu-cosidase (D) and then treated with various glycohydrolytic enzymesas listed below. Panel A: P, protease; Gl, a-glucosidase; G2,P-glucosidase; Ml, a-mannosidase; M2, f-mannosidase; Gal, a-ga-lactosidase; Ga2, ,B-galactosidase; A, a-amylase; C, cellulase. PanelB: PA, protease followed by a-amylase; PAM, protease thena-amylase then a-mannosidase; PAGa, protease then a-amylasethen 3-galactosidase. Panel C: PC, protease followed by cellulase;PCM, protease then cellulase then a-mannosidase; PCGa, proteasethen cellulase then 3-galactosidase. Panel D: PG, protease followedby 3-galactosidase; PGM, protease then P-glucosidase then a-man-nosidase; PGGa, protease then 3-glucosidase then P-galactosi-dase.

racemosus, Fusarium oxysporum, and P. verrucosum (34).A. versicolor produced three antigens with molecular massesof 4.5 x 105, 5.0 x 105, and 6.5 x 105 Da, possibly becauseof cleavage of the antigen by extracellular enzymes, such asproteases, galactofuranosidase, phosphatase, and glycohy-drolases (7, 9, 39).

Mycelial antigens from A. versicolor produced two anti-genic fractions; one bound to ConA-Sepharose and the otherdid not, possibly because the structure of the D-mannose-rich glycopeptide affects the binding ability (20, 38). Thisindicates that there are structural differences between theantigens from A. versicolor that do not bind to ConA andthose from C. herbarum, G. candidum, andM. circinelloidesthat do bind. Furthermore, these binding differences couldbe useful in developing a specific antibody to this species orgenus that does not cross-react with Penicillium antigens(49). Hessian and Smith (16) found thatAdsidia antigens that

5;0.8

w 0.6

0.4

0.2

0P Gl G2 MI M2 Gal Ga2 A C

ENZYME

1.2 1.2 1.2(B) (C) (D)

,°; 1 1:1 1o l0.8 0.8 0.8

Lu 0.6 0.6 0.6

Lu 0.4 0.4-0.4-

0.2-0.2 0.2-


FIG. 3. Relative ELISA activity of A. versicolor mycelial anti-gen treated with protease (A), protease followed by a-amylase (B),protease followed by cellulase (C), or protease followed by 3-glu-cosidase (D) and then treated with various glycohydrolytic enzymesas listed below. Panel A: P, protease; G1, a-glucosidase; G2,3-glucosidase; Ml, a-mannosidase; M2, P-mannosidase; Gal, a-ga-

lactosidase; Ga2, P-galactosidase; A, a-amylase; C, cellulase. PanelB: PA, protease followed by a-amylase; PAM, protease thena-amylase then a-mannosidase; PAGa, protease then a-amylasethen 1-galactosidase. Panel C: PC, protease followed by cellulase;PCM, protease then cellulase then a-mannosidase; PCGa, proteasethen cellulase then 3-galactosidase. Panel D: PG, protease followedby P-glucosidase; PGM, protease then ,B-glucosidase then a-man-nosidase; PGGa, protease then 3-glucosidase then 3-galactosi-dase.

bound to ConA cross-reacted with M. pusillus antiserum butthose that did not bind did not cross-react.

Protein was part of the immunodominant site for allantigens because protease digestion reduced the ELISAactivity by 36 to 64%. The high protein and low mannosecontents in Aspeigillus antigens may also help to explainwhy these antigens did not bind to ConA. However, theneutral sugars were the dominant component of all of theother mycelial and extracellular antigens studied. Theseantigens could be classified as glycoproteins. Protein-richglycopeptides have been reported for other fungal antigens(24, 29). Although mannose was the most abundant sugar inall of these other antigens, it was immunodominant only inthe M. circinelloides antigens. Selective removal of sugarresidues by the glycohydrolytic enzymes suggested thata-1,4-glucose linked to the mannosyl residues is part of theimmunodominant site in the Mucor mycelial antigens be-



http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/


1.2

S;0.8

w 0.6

0.4-

0.2

0P Gl G2 MI M2 Gal Ga2 A C

ENZYME

1.2 1.2 1.2

(B) (C) (D)

°'1 11:F 1.1 10.8 0.8 0.8

wU 0.6 0.6 0.6-

La 0.4-0.4 0.4-

0.2-0.2-0.2-

~~0 0.


FIG. 4. Relative ELISA activity of C. herbarum mycelial antigentreated with protease (A), protease followed by a-amylase (B),protease followed by cellulase (C), or protease followed by ,B-glu-cosidase (D) and then treated with various glycohydrolytic enzymesas listed below. Panel A: P, protease; Gl, a-glucosidase; G2,1-glucosidase; Ml, a-mannosidase; M2, 1-mannosidase; Gal, a-ga-lactosidase; Ga2, 1-galactosidase; A, a-amylase; C, cellulase. PanelB: PA, protease followed by a-amylase; PAM, protease thena-amylase then a-mannosidase; PAGa, protease then a-amylasethen 13-galactosidase. Panel C: PC, protease followed by cellulase;PCM, protease then cellulase then a-mannosidase; PCGa, proteasethen cellulase then 3-galactosidase. Panel D: PG, protease followedby 3-glucosidase; PGM, protease then 3-glucosidase then a-man-nosidase; PGGa, protease then ,B-glucosidase then 3-galactosi-dase.

cause both a-amylase and a-mannosidase reduced theELISA activity. Similar results were observed for the Mucorextracellular antigens; however, their degradation by 1-glu-cosidase suggested that ,B-linked glucose residues were at thenonreducing terminal end. Longbottom and Austwick (29)also reported that mannose was immunodominant for M.hiemalis. Although fucose was present in Mucor antigens,L-fucosidase did not reduce ELISA activity, suggesting thatit is not part of the immunodominant site.For the antigens from A. versicolor, C. herbarum, and G.

candidum, selective removal of the sugar residues suggestedthat galactosyl groups were immunodominant in both myce-

lial and extracellular antigens. For G. candidum, the galac-tosyl residues were probably 1-1,4 linked to glucosyl resi-dues since both ,3-glucosidase and endocellulase decreasedELISA activity. The activity ofA. versicolor was decreasedby 3-galactosidase and a-amylase, suggesting that ax-glucoselinkages were near the galactosyl side chains. Previousresearch has shown that galactofuranosyl groups are immu-nodominant in Aspergillus antigens (5, 7, 36). Galactosylresidues were also immunodominant itl C. herbarum anti-gens; however, selective removal of sugar units with cellu-lase, ,B-glucosidase, and 1-galactosidase suggested that ga-

lactosyl residues were a linked to glucose at the nonreducingends. These differences in the sugar linkages could explainwhy the antigens from one genus did not cross-react with theantibodies from another genus although galactose was theimmunodominant sugar for these three genera.Phosphate was detected in the mycelial antigens but not in

the extracellular antigens, suggesting that the glycoproteinsfrom the mycelium had phosphate linkages (9, 27, 30, 43, 45).From 0.2 to 3% phosphate has been reported for fungalantigens, depending on the culture age and the extractionand purification procedures used (5, 27, 30, 31, 45). Lloyd(27) reported that phosphate was diester linked to peptide-rich mannosyl oligosaccharides in C. werneckii. The pres-ence of ester-linked phosphate groups in these antigens mayhelp to explain the slight decrease in ELISA activity afteraddition of the commercial lipase because it is an esterasethat can cleave phosphodiester bonds.These results have shown that the immunodominant

structures of the mold mycelial and extracellular antigensfrom Aspergillus, Cladosporium, Geotrichum, and Mucorspecies are not similar although they are all glycopro-teins with similar ratios of protein to neutral sugar. Mannosewas immunodominant in Mucor antigens, and galactosewas immunodominant for Aspergillus, Cladosporium, andGeotrichum antigens, but the linkages to galactose differed

TABLE 5. Results of competitive inhibition ELISA using various methyl glycosides to block binding sitesa

I50 (mg/ml) for:

Inhibitor Mucor Mucor Geotrichum Geotrichum Aspergillus Cladosporium Cldospoucircinelloides circinelloides candidum candidum versicolor herbarum Clarummycelium EA mycelium EA EA mycelium herban EA

Referenceb 0.032 0.010 0.005 0.008 0.075 0.014 0.018Methyl-a-D-mannopyranoside 131 398 >2,000 258 >2,000 501 >2,000Methyl-a-D-galactopyranoside >2,000 >2,000 200 125 >2,000 630 794Methyl-3-D-galactopyranoside >2,000 831 6 63 398 199 158Methyl-a-D-glucopyranoside 200 1,000 602 436 >2,000 400 >2,000Methyl-p-D-glucopyranoside 275 72 160 208 >2,000 440 900

a The ELISA was done with antibodies against the species of mold antigens listed across the top of the table.b Antibody corresponding to the mold mycelium.

VOL. 59, 1993


http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/


for these three genera. This research will be useful fordevelopment of specific immunoassays for mold detection infoods.

ACKNOWLEDGMENTS

We thank the Wisconsin Milk Marketing Board and the NationalDairy Promotion and Research Board for supporting this research.The valuable help of S. S. Nielsen, A. Ekanayake, and M. Yadav

with the chemical analyses is greatly appreciated.

REFERENCES

1. Albersheim, P., D. J. Nevins, P. D. English, and A. Karr. 1967.A method for the analysis of sugars in plant cell-wall polysac-charides by gas-liquid chromatography. Carbohydr. Res. 5:340-345.

2. Ames, B. N., and D. T. Dubin. 1960. The role of polyamines inthe neutralization of bacteriophage deoxyribonucleic acid. J.Biol. Chem. 235:769-775.

3. Bahl, 0. P., and K. M. L. Agrawal. 1969. Glycosidases ofAspergillus niger. J. Biol. Chem. 244:2970-2978.

4. Barreto-Bergter, E. M., L. R Travassos, and P. A. J. Gorin.1980. Chemical structure of the D-galacto-D-mannan componentfrom hyphae of Aspergillus niger and other Aspergillus spp.Carbohydr. Res. 86:273-285.

5. Bennett, J. E., A. K. Bhattacharjee, and C. P. J. Glaudemans.1985. Galactofuranosyl groups are immunodominant in As-pergillusfumigatus galactomannan. Mol. Immunol. 22:251-254.

6. Bernfeld, P. 1955. Amylases, a and ,B. Methods Enzymol.1:149-158.

7. Cousin, M. A., S. Notermans, P. Hoogerhout, and J. H. vanBoom. 1989. Detection of 3-galactofuranosidase production byPenicillium and Aspergillus species using 4-nitrophenyl 1-D-galactofuranoside. J. Appl. Bacteriol. 66:311-317.

8. Dubois, M., A. K. Gilles, J. K Hamilton, P. A. Rebers, and F.Smith. 1956. Colorimetric method for determination of sugarsand related substances. Anal. Chem. 28:350-356.

9. Gander, J. E., N. H. Jentoft, L. R. Drewes, and P. D. RicL 1974.The 5-O-13-D-galactofuranosyl-containing exocellular glycopep-tide of Penicillium charlesii: characterization of the phosphog-alactomannan. J. Biol. Chem. 249:2036-2072.

10. Gomez-Miranda, B., and J. A. Leal. 1981. Extracellular and cellwall polysaccharides ofAspergillus alliaceus. Trans. Br. Mycol.Soc. 76:249-253.

11. Gorin, P. A. J., and J. F. T. Spencer. 1968. Structural chemistryof fungal polysaccharides. Adv. Carbohydr. Chem. 23:367-417.

12. Grappel, S. F., F. Blank, and C. T. Bishop. 1967. Immunologicalstudies on dermatophytes. I. Serological reactivities of neutralpolysaccharides with rabbit antiserum to Microsporum quinck-eanum. J. Bacteriol. 93:1001-1008.

13. Grappel, S. F., F. Blank, and C. T. Bishop. 1968. Immunologicalstudies on dermatophytes. II. Serological reactivities of mann-ans prepared from galactomannans I and II of Microsporumquinckeanum, Trichophyton granulosum, Trichophyton inter-digitale, Trichophyton rubrum, and Trichophyton schoenleinii.J. Bacteriol. 95:1238-1242.

14. Grappel, S. F., F. Blank, and C. T. Bishop. 1968. Immunologicalstudies on dermatophytes. III. Further analyses of the reactiv-ities of neutral polysaccharides with rabbit antisera to Micros-porum quinckeanum, Trichophyton schoenleinii, Trichophytonrubrum, Trichophyton interdigitale, and Trichophyton granulo-sum. J. Bacteriol. 96:70-75.

15. Hearn, V. M., and D. W. R. Mackenzie. 1980. The preparationand partial purification of fractions from mycelial fungi withantigenic activity. Mol. Immunol. 17:1097-1103.

16. Hessian, P. A., and J. M. B. Smith. 1982. Antigenic character-ization of some potentially pathogenic mucoraceous fungi. Sa-bouraudia 20:209-216.

17. Hestrin, S., D. S. Feingold, and M. Schramm. 1955. Hexosidehydrolases. Methods Enzymol. 1:231-257.

18. Kamphuis, H. J., S. Notermans, G. H. Veeneman, J. H. Van

Boom, and F. M. Rombouts. 1989. A rapid and reliable methodfor the detection of molds in foods: using the latex agglutinationassay. J. Food Prot. 52:244-247.

19. Kaufman, L., and P. G. Standard. 1987. Specific and rapididentification of medically important fungi by exoantigen detec-tion. Annu. Rev. Microbiol. 41:209-225.

20. Krusius, T., J. Finne, and H. Rauvala. 1976. The structural basisof the different affinities of two types of acidic N-glycosidicglycopeptides for concanavalin A-Sepharose. FEBS Lett. 71:117-120.

21. Kuhn, D. N., and P. K. Stumpf. 1981. Preparation and use ofprotoplasts for studies of lipid metabolism. Methods Enzymol.72:774-783.

22. Leal, J. A., and P. Ruperez. 1978. Extracellular polysaccharideproduction by Aspergillus nidulans. Trans. Br. Mycol. Soc.70:115-120.

23. Lee, Y. C. 1972. a-Mannosidase, 3-glucosidase, and ,B-galacto-sidase from sweet almond emulsion. Methods Enzymol. 28:699-702.

24. Lehmann, P. F., and E. Reiss. 1980. Detection of Candidaalbicans mannan by immunodiffusion, counterimmunoelectro-phoresis and enzyme-linked immunoassay. Mycopathologia 70:83-88.

25. Lin, H. H., and M. A. Cousin. 1987. Evaluation of enzyme-linked immunosorbent assay for detection of molds in foods. J.Food Sci. 52:1089-1094, 1096.

26. Lin, H. H., R. M. Lister, and M. A. Cousin. 1986. Enzyme-linked immunosorbent assay for detection of molds in tomatopuree. J. Food Sci. 51:180-183, 192.

27. Lloyd, K. 0. 1970. Isolation, characterization, and partial struc-ture of peptido galactomannans from the yeast form of Clados-porium wemeckii. Biochemistry 9:3446-3453.

28. Lloyd, K. O., and M. A. Bitoon. 1971. Isolation and purificationof a peptidorhamnomannan from the yeast form of Sporothrixschenckii: structural and immunochemical studies. J. Immunol.107:663-671.

29. Longbottom, J. L., and P. K. C. Austwick. 1986. Fungal anti-gens, p. 7.1-7.11. In D. M. Weir (ed.), Immunochemistry, 4thed. Blackwell Scientific Publications, Boston.

30. McLellan, W. L., Jr., and J. 0. Lampen. 1968. Phosphoman-nanase (PR-factor), an enzyme required for the formation ofyeast protoplasts. J. Bacteriol. 95:967-974.

31. Miyazaki, T., 0. Hayashi, Y. Oshima, and T. Yadomae. 1979.Studies on fungal polysaccharides: the immunological determi-nant of serologically active substances form Absidia cylin-drospora, Mucor hiemalis and Rhizopus nigricans. J. Gen.Microbiol. 111:417-422.

32. Murphy, J., and J. P. Riley. 1962. A modified single solutionmethod for the determination of phosphate in natural water.Anal. Chim. Acta 27:31-36.

33. Notermans, S., J. Dufrenne, and P. S. Soentoro. 1988. Detectionof molds in nuts and spices: the mold colony count versus theenzyme-linked immunosorbent assay (ELISA). J. Food Sci.53:1831-1833, 1843.

34. Notermans, S., and C. J. Heuvelman. 1985. Immunologicaldetection of moulds in food by using the enzyme-linked immu-nosorbent assay (ELISA); preparation of antigens. Int. J. FoodMicrobiol. 2:247-258.

35. Notermans, S., C. J. Heuvelman, H. P. van Egmond, W. E.Paulsch, and J. R. Besling. 1986. Detection of mold in food byenzyme-linked immunosorbent assay. J. Food Prot. 49:786-791.

36. Notermans, S., G. H. Veeneman, C. W. E. M. van Zuylen, P.Hoogerhout, and J. H. van Boom. 1988. (1->5)-linked 3-D-galactofuranosides are immunodominant in extracellularpolysaccharides of Penicillium and Aspergillus species. Mol.Immunol. 25:975-979.

37. Notermans, S., G. Wieten, H. W. B. Engel, F. M. Rombouts, P.Hoogerhout, and J. H. van Boom. 1987. Purification and prop-erties of extracellular polysaccharide (EPS) antigens producedby different mould species. J. Appl. Bacteriol. 62:157-166.

38. Ogata, S.-I., T. Muramatsu, and A. Kobata. 1975. Fractionationof glycopeptides by affinity column chromatography on con-canavalin A-Sepharose. J. Biochem. 78:687-696.

APPL. ENVIRON. MICROBIOL.


http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/


39. Piechura, J. E., C. J. Huang, S. H. Cohen, J. M. Kidd, V. P.Kurup, and N. J. Calvanico. 1982. Antigens of Aspergillusfumigatus II. Electrophoretic and clinical studies. Immunology49:657-665.

40. Preston, J. F., E. Lapis, and J. E. Gander. 1969. Isolation andpartial characterization of the exocellular polysaccharides ofPenicillium charlesii. III. Heterogeneity in size and compositionof high molecular weight exocellular polysaccharides. Arch.Biochem. Biophys. 134:324-334.

41. Robertson, A., D. Upadhyaya, S. Opie, and J. Sargeant. 1986.An immunochemical method for the measurement of mouldcontamination in tomato paste, p. 163-179. In B. A. Morris,M. N. Clifford, and R. Jackman (ed.), Immunoassays forveterinary and food analysis-1. Elsevier Applied Science, NewYork.

42. Sakaguchi, O., K. Yokota, and M. Suzuki. 1969. Immunochem-ical and biochemical studies of fungi. XIII. On the galactoman-nans isolated from mycelia and culture filtrates of severalfilamentous fungi. Jpn. J. Microbiol. 13:1-7.

43. Salt, S. D., and J. E. Gander. 1985. Variations in phosphorylsubstituents in extracellular peptidophosphogalactomannansfrom Penicillium charlesii G. Smith. Exp. Mycol. 9:9-19.

44. Schumacher, M. J., J. K. McClatchy, R. S. Farr, and P. Minden.1975. Primary interaction between antibody and components ofAlternaria. J. Allerg. Clin. Immunol. 56:39-53.

45. Stewart, T. S., and C. E. Ballou. 1968. A comparison of yeastmannans and phosphomannans by acetolysis. Biochemistry7:1855-1863.

46. Sudhakaran, V. K., and J. G. Shewale. 1988. Exopolysaccharideproduction by Nigrospora oryzae var. glucanicum. EnzymeMicrob. Technol. 10:547-551.

47. Sugahara, K., and I. Yamashina. 1972. ,-Mannosidase fromsnail. Methods Enzymol. 28:769-772.

48. Suzuki, S., and N. Takeda. 1975. Serologic cross-reactivity ofthe D-galacto-D-mannans isolated from several pathogenic fungiagainst anti-Hormodendrum pedrosoi serum. Carbohydr. Res.40:193-197.

49. Tsai, G.-J., and M. A. Cousin. 1990. Enzyme-linked immu-nosorbent assay for detection of molds in cheese and yogurt. J.Dairy Sci. 73:3366-3378.

50. Tsay, G. C., and G. Dawson. 1977. A sensitive spectrophoto-metric method for detection of L-fucose. Anal. Biochem. 78:423-427.

VOL. 59, 1993


http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/

Partial Purification and Characterization of Mold Commonly · Double-sandwich ELISA. The...

Documents

Transcript of Partial Purification and Characterization of Mold Commonly · Double-sandwich ELISA. The...