Mycopathologia 120: 79-85, 1992. 9 1992 KluwerAcademic Publishers. Printedin the Net#eriands.
The solid phase attachment of fungal hyphae in an ELISA to screen for antifungal antibodies John N. Banks & Sarah J. Cox
MAFF, Central Science Laboratory, London Road, Slough, UK Received 23 December i991; acceptedin revised form 19 May 1992
Key words: ELISA, hyphae, solid phase attachment
Abstract
A method to immobilize fungal hyphae onto the wells of 96-well microplates for use in an in-direct ELISA to screen for antifungal antibodies in sera and cell culture supernatants is described. The hyphae from three genera (Penicillium, Eurotium and Fusarium) were successfully attached by overnight drying onto wells precoated with poly-L-lysine and glutaraldehyde. Microscopy revealed that the hyphae remained attached to the wells throughout the ELISA and antiserum titrations showed that the attached hyphae were uniformly coated and remained reactive. Background absorbances were low and the plates could be stored at - 2 0 ~ without loss of reactivity.
Introduction
There is increasing interest in rapid immunological techniques for the detection and identification of fungi in crops and stored products [1-3]. During the production of polyclonal antibodies (PAbs) or monoclonal antibodies (MAbs) required for such a technique, sera needs to be screened for antibodies in order to ensure the animals are responding to immunization. This is usually carried out by indirect enzyme-linked immunosorbent assay (ELISA) [4] performed during the immunization period. In MAb production such an ELISA is also required for initial screening for antibody during hybridoma production and subsequent cloning of antibody positive secreting cells. At this laboratory we are currently involved in a programme to raise antifungal antibodies against Penicillium aurantiogriseum var. melano-
conidium, Eurotium amstelodarni and Fusarium culmorum. Some of the immunizations have been carried out using an insoluble (cell wall) preparation of fungus, prepared according to Clarke et al. [5] and consisting essentially of hyphal fragments washed with phosphate buffered saline. There was, therefore, a need to be able to immobilize or stick hyphae to the wells of 96-well microplates, and for it to retain its antigenicity, so that sera or tissue culture supernatants could be screened for antifungal antibodies by ELISA. Some success has been reported for the solid phase immobilization of plant, animal, bacterial and viral material [6-9] and spores [10] but there does not appear to be a method described in the literature for effecting the adhesion of hyphal fragments to 96-well microplates. Methods for the solid phase immobilization of non-fungal material to microplates have utilized the crosslinking properties of glutaraldehyde [9, 11, 12] sometimes in.
80
Fig. 1. Penicillium aurantiogriseum var. melanoconidium hyphae dried on overnight on a poly-l-lysineand glutaraldehydepre-
treated well.
combination with bovine serum albumin [6]. Poly-L-lysine has also been used to effect binding [6-8, 10]. A method to attach hyphal fragments t o 96well microplates incorporating precoating the wells with poly-L-lysine and then glutaraldehyde has been developed and the optimized method is described.
Materials and methods
The fungi were grown by a method similar to that detailed by Smith [13] except that Penicillium aurantiogriseum var. melanoconidium was grown on 2% malt agar [14], Eurotium amstelodami on 2% malt with 10% sodium chloride and Fusarium culmorum on potato sucrose agar [15]. A spore suspension was prepared in sterile distilled water with 0.01% Tween 80 (British Drug Houses) and the number of spores counted in an improved
Neubower counting chamber (Weber, England). The spore suspension was adjusted to give approximately 1 x 106 spores m1-1 with sterile distilled water and an inoculum volume of 1.0 ml was added to flasks containing 100 ml Burrell et al. medium [16]. Eurotium amstelodami was grown in Burrell et al. medium [16] supplemented with 10% sodium chloride. The flasks were incubated on a rotary shaker at 140 rpm for 7 d at 25 ~ in the dark to produce pellets after which time the flasks were stored at - 2 0 ~ To harvest the fungus and release as much potentially antigenic material as possible, the flasks were thawed and the pellets removed by vacuumassisted filtration through a sintered glass Buchner funnel onto filter paper. The pellets were washed with 30 ml of sterile distilled water followed by 30 ml of phosphate buffered saline (PBS) [17] per flask, removed from the filter paper and placed at - 2 0 ~ overnight. The fungal material was then freeze-dried, snap-frozen in
81
Fig. 2. Same well as Fig. 1 but just before the addition of enzyme substrate.
liquid nitrogen and homogenized using an Omnimixer (Camlab Ltd) at setting No. 7 for i rain. The fine powder was washed three times by resuspending it in 8 ml PBS g 1 of fungal material, vigorously mixed and centrifuged at 3000 g for 10 min at 4 ~ The pellet was finally taken up in an appropriate volume of PBS and stored at -20 ~ Prior to use in an ELISA, total protein estimations were performed on each preparation using a modified Bradford protein assay [18]. The Costar flat bottomed 96-well microplates (Northumbria Bi01ogicals) were coated with 50 txl per well of 0.005% poly-L-lysine (PLL) (Sigma Chemical Co.) and the plates were then incubated for 45 min at 25 ~ The plates were next washed four times with 250 ixl of PBS with 0.05% Tween 20 (PBST) per well, blotted dry and treated with 501xl per well of 2% glutaraldehyde (Sigma Chemical Co., Grade 1, 25% aqueous solution). After an incubation period of 15 rain at 25 ~ the plates were washed as before. The ELISA buffers for coating and conducting
the assay were as described by Voller et al. [4]. The antigen at an optimum concentration oJ 3.4 Ixg m1-1 (previously determined by chequerboard titration) was added in 50 Ixl amounts into wells pretreated with either PLL and glutaraldehyde or untreated. The microplates were incu. bated open at 25 ~ overnight to allow the antiger to dry onto the bottom of the wells. The microplates were next washed four times with PBS~I and then blocked by adding 250 ~1 of 3% bovine serum albumin (BSA) (Sigma Chemical Co.) i~ PBST per well. After an incubation of 60 rain al 25 ~ the plates were washed and either stored a! -20 ~ or used immediately. If the plates were stored they were washed three times before storage and washed once when retrieved. Following coating of the plates titration assays were carried out with antiserum raised againsl hyphal fragments of Penicilliurn aurantiogriseum var. melanoconidium in mice at this laboratory oi negative mouse serum (Serotec) on both treated and untreated plates. Doubling dilutions in PBS~I
82
Fig. 3. Penicillium aurantiogriseum var. melanoconidium hyphae dried on a well not pre-treated with poly-l-lysineor glutaraldehyde.
covered the range 1:250 to 1:8000. Diluted sera or PBST (negative control) was added in triplicate in 50 Ixl amounts to each well and incubated for 60 min at 25 ~ The plates were washed four times with PBST and 200 ~1 of horseradish-peroxidase conjugated goat anti-mouse immunoglobulins (Nordic) were added to each of the wells. After incubation at 25 ~ for 60 min the plates were washed as before and 200 txl of substrate Ophenylenediamine dihydrochloride (Sigma Chemical Co., 0.4 g in 100 ml phosphate citrate buffer (PCB), diluted a further 1:10 in PCB with 0.012% hydrogen peroxide) were added. The reaction was allowed to progress at 25 ~ for 10 min in the dark before being stopped by the addition of 50 Ixl of 2.5 M sulphuric acid to each of the wells. The plates were read at 490 nm on a Dynatech MR600 microplate reader. Titration assays were carried out on plates coated with Penicillium aurantiogriseum var. rnelanoconidiurn, Eurotiurn amstelodami and Fusar-
ium culmorum on the day of coating (day 0) and after storage at -20 ~ for 67 days. For all three genera the microplate wells were examined for hyphae under an inverted light microscope (Leitz Diavert) after each stage of the ELISA. Photomicrographs were taken of the plates coated with Penicilliurn aurantiogriseurn var. rnelanoconidium after overnight drying, the first wash with PBST and immediately prior to the addition of conjugate.
Results
The procedure for attaching fungal hyphae to 96well microplates with PLL and glutaraldehyde and the subsequent stage s of the indirect ELISA were first developed for Penicillium aurantiogriseum var. melanoconidium. In the development of this procedure the studies included directly,
83
Fig. 4. Samewell as Fig. 3 but just before the additionof enzymesubstrate.
coating hyphae onto modified wells and unmodified wells. Microscopy showed that for the wells pretreated with PLL and glutaraldehyde virtually all the hyphae present after drying (Fig. 1) were retained in the wells throughout the ELISA (Fig 2). For wells not pretreated many of the hyphae coated (Fig. 3) were virtually all lost by the time the enzyme substrate was added (Fig. 4). The procedure involving pretreatment with PLL and glutaraldehyde was then applied to Eurotturn amstelodami and Fusarium culmorurn to determine if the hyphae of other fungal genera could be made to adhere to microplates. The same result as that for Penicilliurn aurantiogriseurn var. melanoconidium was observed by microscopy in that the hyphae remained adhered to the wells throughout the ELISA. Results of the titrations using the optimised ELISA method and antisera raised against Penicillium aurantiogriseum var. rnelanoconidium
using microplates coated with all three genera of fungi are given in Table 1. As expected there was a decrease in absorbance with dilution of antiserum. The highest absorbance reading was given, by the homologous pairing. Similar results were, however, gained with the heterologous antigen Fusarium culrnorum but for Eurotium amstelodarni the absorbances were a little over half those produced for the homologous pairing. The results of the titration (Table 1) also show that nonspecific binding of the enzyme conjugate was low as the absorbances of the PBST reagent blanks were between 0.063 and 0.084 and nonimmune mouse serum absorbances in all assays were less than 0.049. Additionally, absorbance results for plates stored at -20 ~ for 67 days were essentially the same as for those which had not been stored at - 2 0 ~ (day 0) and microscopy confirmed that hyphae remained in the wells throughout the ELISA.
84 Table 1. Absorbance values for serum titrations using plates coated with Penicillium aurantiogriseum var. melanoconidium (PAM) Eurotium amstelodami (EA) and Fusarium culrnorum (FC) unstored (day 0) and after storage (67 days) at -20 ~
Fungus/Antigen Pam Serum dilution
Ea
FC
Serum
Day 0
Day 67
Day 0
Day 67
Day 0
Day 67
1/250
PMS NMS
1.00 0.04
0.93 0.02
0.51 0.05
0.63 0.04
0.95 0.04
0.85 0.03
1/500
PMS NMS
0.63 0.02
0.63 0.00
0.28 0.02
0.37 0.01
0.73 0.01
0.51 0.00
1/1000
PMS NMS
0.26 0.00
0.34 0.00
0.15 0.01
0.22 0.00
0.38 0.00
0.35 0.00
1/2000
PMS NMS
0.17 0.00
0.22 0.00
0.13 0.01
0.12 0.00
0.24 0.00
0.23 0.00
1/4000
PMS NMS
0.13 0.00
0.14 0.00
0.06 0.00
0.07 0.00
0.15 0.00
0.13 0.00
1/8000
PMS NMS
0.12 0.00
0.10 0.00
0.05 0.00
0.04 0.00
0.11 0.00
0.06 0.00
PMS = Positive mouse serum: containing antifungal antibodies. NMS = Negative mouse serum: not containing antifungal antibodies.
Discussion For an E L I S A to screen for specific antibodies it is advisable to use the same form of antigen to coat the plates as used for immunization. A method for attaching fungal hyphae to 96-well microplates was therefore developed and conditions for an indirect E L I S A optimized. Uniform coating of the E L I S A microplates was observed by microscopy and supported by typical titration results even though particulate material was used to coat the microplate. The m e t h o d developed for the homologous antigen Penicillium aurantiogriseum var. m e l a n o c o n i d i u m , worked equally well with two other genera tested and thus could probably be applied to other fungi. Differences in reactivity as shown by the antiserum titrations are attributable to the fact that the antiserum was raised against Penicillium and will not necessarily react to the same extent with other fungi. Microscopy and photomicrography confirmed that very little if any of the fungal material was lost from pretreated wells during the washing stages of the E L I S A . It was interesting to note that the hyphae on the pretreated wells (Fig. 1)
appeared m o r e fragmented than on the untreated wells (Fig. 3). The latter appeared in branch like clumps which look relatively dense c o m p a r e d to pretreated wells even though the amounts of fungal material applied were identical. No explanation can be offered for this difference except that adhesive or cross-linking properties of P L L and glutaraldehyde may have caused the hyphae to fragment. The use of glutaraldehyde is thought to alter some antigenic determinants in m a m m a l i a n cells [8]. Further work in this area for fungi may prove necessary in developing immunoassays for detecting fungi. T h e use of glutaraldehyde in this assay, however, does not appear to seriously affect the antigen and antibody binding reaction which must have occurred to give positive absorbance readings in the titration studies. E L I S A s carried out on microplates not pretreated with P L L and glutaraldehyde gave absorbances almost as high as those on pretreated microplates (unpublished data). This suggests that some soluble antigenic material must remain bound to the wells after the hyphae have been washed off. The antigenic characteristics are
85 likely to be different for the soluble and particulate hyphal antigenic fragments. For the screening of PAbs it may not matter if the antigen is not representative of the whole but for MAbs it may be more critical. The almost identical titration results for ELISAs carried out on plates at day 0 and day 67 after storage at - 2 0 ~ show that the coated plates may be prepared in advance of requirement. This is particularly useful during MAb production when the early stages of production require a large number of screening ELISAs to be carried out. Coating large batches of microplates which may be stored also helps to reduce interplate variability. The method described is simple and fairly quick and although originally developed for Penicillium auratiogriseum var. melanoconidium proved suitable for other fungi and has been successfully used to screen rabbit and mouse sera for antibodies.
3. 4.
5.
6.
7.
8.
9.
10.
11.
Acknowledgements Thanks are due to Dr J.H. Clarke for comments during development of the method described and to Miss R.H. Shamsi, Miss L.A. Cox and Miss K.H. Yau for their excellent technical assistance. This study was funded by the Ministry of Agriculture, Fisheries and Food.
12.
13.
14. 15. 16.
References I. Dewey FM, Munday CJ, Brasier CM. Monoclonal antibodies to specific components of Dutch Elm disease pathogen Ophiostoma ulmi. Plant Path 1989; 38: 9-20. 2. Notermans S, Heuvelman CJ. Immunological detection of moulds in food by using the enzyme-linked immuno-
17. 18.
sorbent assay (ELISA): Preparation of antigens. Int J Food Microbiol 1985; 2: 247-258. Polonelli L, Morace G. Serological potential for fungal identification. Bot J Linnean Soc 1989; 99: 33-38. Voller A, Bidwell DE, Bartlett A. The enzyme-linked immunosorbent assay (ELISA): A guide with abstracts of microplate applications. Guernsey: Dynatech Europe, 1979. Clarke JH, MacNicoll AD, Norman JA. Immunological detection of fungi in plants, including stored cereals. Int Biodet 1986; 22 (Supplement): 123-130. Burbrick P, Goldstein L, Frensdorff A. Immobilised whole algal cells for solid phase binding assays. J Immunol Meth 1986; 86(2): 171-177. Deng M, Cliver DO. A broad-spectrum enzyme-linked immunosorbent assay for the detection of human enteric viruses. J Viro Meth 1984; 8: 87-98. Epstein SL, Lunney JK. A cell surface ELISA in the mouse using only poly-L-lysine as cell fixative. J Immunol Meth 1985: 76: 63-72. Evan GI. A single and rapid solid phase enzyme-linked immunosorbent assay for screening monoclonal antibodies to poorly soluble proteins. J Immunol Meth 1984: 73: 427-435. Banowetz GM, Trione EJ, Krygier BB. Immunological comparisons of teliospores of two wheat bunt fungi, Tilletia species, using monoclonal antibodies and antisera. Mycologia 1984; 76(1): 51-62. Kennedy JH, Kricka LJ, Wilding P. Protein-protein coupling reactions and the applications of protein conjugates. Clin Chem Acta 1976; 70: 1-31. Wold F. Bifunctional reagents. In: Hirs CHW, Reimashelf AN. Methods in enzymology. London: Academic Press, 1972; 25: 623. Smith D. Maintenance of fungi. In: Kirsop BE, Snell JJS, eds. Maintenance of microorganisms: A manual of laboratory methods. London: Academic Press, 1984. Christensen CM, Lopez LC. Pathology of stored seeds. Proc Int Seed Testing Assoc 1963; 28: 701-711. Booth C. The genus Fusarium. Kew (Surrey, UK): Commonwealth Mycological Institute, 1971. Burrell RG, Clayton CW, Gallegly ME, Lilly VG. Factors affecting the antigenicity of the Mycelium of three species of Phytophthora. Phytopathology 1966; 56(1): 422-426. Hudson L, Hay FC. Practical immunology, 2nd ed. Oxford: Blackwell Scientific Publications 1980:336 pp. Banks JN, Cox SJ, Clarke JH, Shamsi RH, Northway BJ. Towards the immunological detection of field and storage fungi. In: Samson RA, Hocking AD, Pitt JI, Kind AD, eds. Modern methods in food mycology Amsterdam: Elsevier Press, 1992 .
The solid phase attachment of fungal hyphae in an ELISA to screen for antifungal antibodies John N. Banks & Sarah J. Cox
MAFF, Central Science Laboratory, London Road, Slough, UK Received 23 December i991; acceptedin revised form 19 May 1992
Key words: ELISA, hyphae, solid phase attachment
Abstract
A method to immobilize fungal hyphae onto the wells of 96-well microplates for use in an in-direct ELISA to screen for antifungal antibodies in sera and cell culture supernatants is described. The hyphae from three genera (Penicillium, Eurotium and Fusarium) were successfully attached by overnight drying onto wells precoated with poly-L-lysine and glutaraldehyde. Microscopy revealed that the hyphae remained attached to the wells throughout the ELISA and antiserum titrations showed that the attached hyphae were uniformly coated and remained reactive. Background absorbances were low and the plates could be stored at - 2 0 ~ without loss of reactivity.
Introduction
There is increasing interest in rapid immunological techniques for the detection and identification of fungi in crops and stored products [1-3]. During the production of polyclonal antibodies (PAbs) or monoclonal antibodies (MAbs) required for such a technique, sera needs to be screened for antibodies in order to ensure the animals are responding to immunization. This is usually carried out by indirect enzyme-linked immunosorbent assay (ELISA) [4] performed during the immunization period. In MAb production such an ELISA is also required for initial screening for antibody during hybridoma production and subsequent cloning of antibody positive secreting cells. At this laboratory we are currently involved in a programme to raise antifungal antibodies against Penicillium aurantiogriseum var. melano-
conidium, Eurotium amstelodarni and Fusarium culmorum. Some of the immunizations have been carried out using an insoluble (cell wall) preparation of fungus, prepared according to Clarke et al. [5] and consisting essentially of hyphal fragments washed with phosphate buffered saline. There was, therefore, a need to be able to immobilize or stick hyphae to the wells of 96-well microplates, and for it to retain its antigenicity, so that sera or tissue culture supernatants could be screened for antifungal antibodies by ELISA. Some success has been reported for the solid phase immobilization of plant, animal, bacterial and viral material [6-9] and spores [10] but there does not appear to be a method described in the literature for effecting the adhesion of hyphal fragments to 96-well microplates. Methods for the solid phase immobilization of non-fungal material to microplates have utilized the crosslinking properties of glutaraldehyde [9, 11, 12] sometimes in.
80
Fig. 1. Penicillium aurantiogriseum var. melanoconidium hyphae dried on overnight on a poly-l-lysineand glutaraldehydepre-
treated well.
combination with bovine serum albumin [6]. Poly-L-lysine has also been used to effect binding [6-8, 10]. A method to attach hyphal fragments t o 96well microplates incorporating precoating the wells with poly-L-lysine and then glutaraldehyde has been developed and the optimized method is described.
Materials and methods
The fungi were grown by a method similar to that detailed by Smith [13] except that Penicillium aurantiogriseum var. melanoconidium was grown on 2% malt agar [14], Eurotium amstelodami on 2% malt with 10% sodium chloride and Fusarium culmorum on potato sucrose agar [15]. A spore suspension was prepared in sterile distilled water with 0.01% Tween 80 (British Drug Houses) and the number of spores counted in an improved
Neubower counting chamber (Weber, England). The spore suspension was adjusted to give approximately 1 x 106 spores m1-1 with sterile distilled water and an inoculum volume of 1.0 ml was added to flasks containing 100 ml Burrell et al. medium [16]. Eurotium amstelodami was grown in Burrell et al. medium [16] supplemented with 10% sodium chloride. The flasks were incubated on a rotary shaker at 140 rpm for 7 d at 25 ~ in the dark to produce pellets after which time the flasks were stored at - 2 0 ~ To harvest the fungus and release as much potentially antigenic material as possible, the flasks were thawed and the pellets removed by vacuumassisted filtration through a sintered glass Buchner funnel onto filter paper. The pellets were washed with 30 ml of sterile distilled water followed by 30 ml of phosphate buffered saline (PBS) [17] per flask, removed from the filter paper and placed at - 2 0 ~ overnight. The fungal material was then freeze-dried, snap-frozen in
81
Fig. 2. Same well as Fig. 1 but just before the addition of enzyme substrate.
liquid nitrogen and homogenized using an Omnimixer (Camlab Ltd) at setting No. 7 for i rain. The fine powder was washed three times by resuspending it in 8 ml PBS g 1 of fungal material, vigorously mixed and centrifuged at 3000 g for 10 min at 4 ~ The pellet was finally taken up in an appropriate volume of PBS and stored at -20 ~ Prior to use in an ELISA, total protein estimations were performed on each preparation using a modified Bradford protein assay [18]. The Costar flat bottomed 96-well microplates (Northumbria Bi01ogicals) were coated with 50 txl per well of 0.005% poly-L-lysine (PLL) (Sigma Chemical Co.) and the plates were then incubated for 45 min at 25 ~ The plates were next washed four times with 250 ixl of PBS with 0.05% Tween 20 (PBST) per well, blotted dry and treated with 501xl per well of 2% glutaraldehyde (Sigma Chemical Co., Grade 1, 25% aqueous solution). After an incubation period of 15 rain at 25 ~ the plates were washed as before. The ELISA buffers for coating and conducting
the assay were as described by Voller et al. [4]. The antigen at an optimum concentration oJ 3.4 Ixg m1-1 (previously determined by chequerboard titration) was added in 50 Ixl amounts into wells pretreated with either PLL and glutaraldehyde or untreated. The microplates were incu. bated open at 25 ~ overnight to allow the antiger to dry onto the bottom of the wells. The microplates were next washed four times with PBS~I and then blocked by adding 250 ~1 of 3% bovine serum albumin (BSA) (Sigma Chemical Co.) i~ PBST per well. After an incubation of 60 rain al 25 ~ the plates were washed and either stored a! -20 ~ or used immediately. If the plates were stored they were washed three times before storage and washed once when retrieved. Following coating of the plates titration assays were carried out with antiserum raised againsl hyphal fragments of Penicilliurn aurantiogriseum var. melanoconidium in mice at this laboratory oi negative mouse serum (Serotec) on both treated and untreated plates. Doubling dilutions in PBS~I
82
Fig. 3. Penicillium aurantiogriseum var. melanoconidium hyphae dried on a well not pre-treated with poly-l-lysineor glutaraldehyde.
covered the range 1:250 to 1:8000. Diluted sera or PBST (negative control) was added in triplicate in 50 Ixl amounts to each well and incubated for 60 min at 25 ~ The plates were washed four times with PBST and 200 ~1 of horseradish-peroxidase conjugated goat anti-mouse immunoglobulins (Nordic) were added to each of the wells. After incubation at 25 ~ for 60 min the plates were washed as before and 200 txl of substrate Ophenylenediamine dihydrochloride (Sigma Chemical Co., 0.4 g in 100 ml phosphate citrate buffer (PCB), diluted a further 1:10 in PCB with 0.012% hydrogen peroxide) were added. The reaction was allowed to progress at 25 ~ for 10 min in the dark before being stopped by the addition of 50 Ixl of 2.5 M sulphuric acid to each of the wells. The plates were read at 490 nm on a Dynatech MR600 microplate reader. Titration assays were carried out on plates coated with Penicillium aurantiogriseum var. rnelanoconidiurn, Eurotiurn amstelodami and Fusar-
ium culmorum on the day of coating (day 0) and after storage at -20 ~ for 67 days. For all three genera the microplate wells were examined for hyphae under an inverted light microscope (Leitz Diavert) after each stage of the ELISA. Photomicrographs were taken of the plates coated with Penicilliurn aurantiogriseurn var. rnelanoconidium after overnight drying, the first wash with PBST and immediately prior to the addition of conjugate.
Results
The procedure for attaching fungal hyphae to 96well microplates with PLL and glutaraldehyde and the subsequent stage s of the indirect ELISA were first developed for Penicillium aurantiogriseum var. melanoconidium. In the development of this procedure the studies included directly,
83
Fig. 4. Samewell as Fig. 3 but just before the additionof enzymesubstrate.
coating hyphae onto modified wells and unmodified wells. Microscopy showed that for the wells pretreated with PLL and glutaraldehyde virtually all the hyphae present after drying (Fig. 1) were retained in the wells throughout the ELISA (Fig 2). For wells not pretreated many of the hyphae coated (Fig. 3) were virtually all lost by the time the enzyme substrate was added (Fig. 4). The procedure involving pretreatment with PLL and glutaraldehyde was then applied to Eurotturn amstelodami and Fusarium culmorurn to determine if the hyphae of other fungal genera could be made to adhere to microplates. The same result as that for Penicilliurn aurantiogriseurn var. melanoconidium was observed by microscopy in that the hyphae remained adhered to the wells throughout the ELISA. Results of the titrations using the optimised ELISA method and antisera raised against Penicillium aurantiogriseum var. rnelanoconidium
using microplates coated with all three genera of fungi are given in Table 1. As expected there was a decrease in absorbance with dilution of antiserum. The highest absorbance reading was given, by the homologous pairing. Similar results were, however, gained with the heterologous antigen Fusarium culrnorum but for Eurotium amstelodarni the absorbances were a little over half those produced for the homologous pairing. The results of the titration (Table 1) also show that nonspecific binding of the enzyme conjugate was low as the absorbances of the PBST reagent blanks were between 0.063 and 0.084 and nonimmune mouse serum absorbances in all assays were less than 0.049. Additionally, absorbance results for plates stored at -20 ~ for 67 days were essentially the same as for those which had not been stored at - 2 0 ~ (day 0) and microscopy confirmed that hyphae remained in the wells throughout the ELISA.
84 Table 1. Absorbance values for serum titrations using plates coated with Penicillium aurantiogriseum var. melanoconidium (PAM) Eurotium amstelodami (EA) and Fusarium culrnorum (FC) unstored (day 0) and after storage (67 days) at -20 ~
Fungus/Antigen Pam Serum dilution
Ea
FC
Serum
Day 0
Day 67
Day 0
Day 67
Day 0
Day 67
1/250
PMS NMS
1.00 0.04
0.93 0.02
0.51 0.05
0.63 0.04
0.95 0.04
0.85 0.03
1/500
PMS NMS
0.63 0.02
0.63 0.00
0.28 0.02
0.37 0.01
0.73 0.01
0.51 0.00
1/1000
PMS NMS
0.26 0.00
0.34 0.00
0.15 0.01
0.22 0.00
0.38 0.00
0.35 0.00
1/2000
PMS NMS
0.17 0.00
0.22 0.00
0.13 0.01
0.12 0.00
0.24 0.00
0.23 0.00
1/4000
PMS NMS
0.13 0.00
0.14 0.00
0.06 0.00
0.07 0.00
0.15 0.00
0.13 0.00
1/8000
PMS NMS
0.12 0.00
0.10 0.00
0.05 0.00
0.04 0.00
0.11 0.00
0.06 0.00
PMS = Positive mouse serum: containing antifungal antibodies. NMS = Negative mouse serum: not containing antifungal antibodies.
Discussion For an E L I S A to screen for specific antibodies it is advisable to use the same form of antigen to coat the plates as used for immunization. A method for attaching fungal hyphae to 96-well microplates was therefore developed and conditions for an indirect E L I S A optimized. Uniform coating of the E L I S A microplates was observed by microscopy and supported by typical titration results even though particulate material was used to coat the microplate. The m e t h o d developed for the homologous antigen Penicillium aurantiogriseum var. m e l a n o c o n i d i u m , worked equally well with two other genera tested and thus could probably be applied to other fungi. Differences in reactivity as shown by the antiserum titrations are attributable to the fact that the antiserum was raised against Penicillium and will not necessarily react to the same extent with other fungi. Microscopy and photomicrography confirmed that very little if any of the fungal material was lost from pretreated wells during the washing stages of the E L I S A . It was interesting to note that the hyphae on the pretreated wells (Fig. 1)
appeared m o r e fragmented than on the untreated wells (Fig. 3). The latter appeared in branch like clumps which look relatively dense c o m p a r e d to pretreated wells even though the amounts of fungal material applied were identical. No explanation can be offered for this difference except that adhesive or cross-linking properties of P L L and glutaraldehyde may have caused the hyphae to fragment. The use of glutaraldehyde is thought to alter some antigenic determinants in m a m m a l i a n cells [8]. Further work in this area for fungi may prove necessary in developing immunoassays for detecting fungi. T h e use of glutaraldehyde in this assay, however, does not appear to seriously affect the antigen and antibody binding reaction which must have occurred to give positive absorbance readings in the titration studies. E L I S A s carried out on microplates not pretreated with P L L and glutaraldehyde gave absorbances almost as high as those on pretreated microplates (unpublished data). This suggests that some soluble antigenic material must remain bound to the wells after the hyphae have been washed off. The antigenic characteristics are
85 likely to be different for the soluble and particulate hyphal antigenic fragments. For the screening of PAbs it may not matter if the antigen is not representative of the whole but for MAbs it may be more critical. The almost identical titration results for ELISAs carried out on plates at day 0 and day 67 after storage at - 2 0 ~ show that the coated plates may be prepared in advance of requirement. This is particularly useful during MAb production when the early stages of production require a large number of screening ELISAs to be carried out. Coating large batches of microplates which may be stored also helps to reduce interplate variability. The method described is simple and fairly quick and although originally developed for Penicillium auratiogriseum var. melanoconidium proved suitable for other fungi and has been successfully used to screen rabbit and mouse sera for antibodies.
3. 4.
5.
6.
7.
8.
9.
10.
11.
Acknowledgements Thanks are due to Dr J.H. Clarke for comments during development of the method described and to Miss R.H. Shamsi, Miss L.A. Cox and Miss K.H. Yau for their excellent technical assistance. This study was funded by the Ministry of Agriculture, Fisheries and Food.
12.
13.
14. 15. 16.
References I. Dewey FM, Munday CJ, Brasier CM. Monoclonal antibodies to specific components of Dutch Elm disease pathogen Ophiostoma ulmi. Plant Path 1989; 38: 9-20. 2. Notermans S, Heuvelman CJ. Immunological detection of moulds in food by using the enzyme-linked immuno-
17. 18.
sorbent assay (ELISA): Preparation of antigens. Int J Food Microbiol 1985; 2: 247-258. Polonelli L, Morace G. Serological potential for fungal identification. Bot J Linnean Soc 1989; 99: 33-38. Voller A, Bidwell DE, Bartlett A. The enzyme-linked immunosorbent assay (ELISA): A guide with abstracts of microplate applications. Guernsey: Dynatech Europe, 1979. Clarke JH, MacNicoll AD, Norman JA. Immunological detection of fungi in plants, including stored cereals. Int Biodet 1986; 22 (Supplement): 123-130. Burbrick P, Goldstein L, Frensdorff A. Immobilised whole algal cells for solid phase binding assays. J Immunol Meth 1986; 86(2): 171-177. Deng M, Cliver DO. A broad-spectrum enzyme-linked immunosorbent assay for the detection of human enteric viruses. J Viro Meth 1984; 8: 87-98. Epstein SL, Lunney JK. A cell surface ELISA in the mouse using only poly-L-lysine as cell fixative. J Immunol Meth 1985: 76: 63-72. Evan GI. A single and rapid solid phase enzyme-linked immunosorbent assay for screening monoclonal antibodies to poorly soluble proteins. J Immunol Meth 1984: 73: 427-435. Banowetz GM, Trione EJ, Krygier BB. Immunological comparisons of teliospores of two wheat bunt fungi, Tilletia species, using monoclonal antibodies and antisera. Mycologia 1984; 76(1): 51-62. Kennedy JH, Kricka LJ, Wilding P. Protein-protein coupling reactions and the applications of protein conjugates. Clin Chem Acta 1976; 70: 1-31. Wold F. Bifunctional reagents. In: Hirs CHW, Reimashelf AN. Methods in enzymology. London: Academic Press, 1972; 25: 623. Smith D. Maintenance of fungi. In: Kirsop BE, Snell JJS, eds. Maintenance of microorganisms: A manual of laboratory methods. London: Academic Press, 1984. Christensen CM, Lopez LC. Pathology of stored seeds. Proc Int Seed Testing Assoc 1963; 28: 701-711. Booth C. The genus Fusarium. Kew (Surrey, UK): Commonwealth Mycological Institute, 1971. Burrell RG, Clayton CW, Gallegly ME, Lilly VG. Factors affecting the antigenicity of the Mycelium of three species of Phytophthora. Phytopathology 1966; 56(1): 422-426. Hudson L, Hay FC. Practical immunology, 2nd ed. Oxford: Blackwell Scientific Publications 1980:336 pp. Banks JN, Cox SJ, Clarke JH, Shamsi RH, Northway BJ. Towards the immunological detection of field and storage fungi. In: Samson RA, Hocking AD, Pitt JI, Kind AD, eds. Modern methods in food mycology Amsterdam: Elsevier Press, 1992 .
