Development of Specific Enzyme-Linked Immunosorbent Antibody Assay Systems for the Detection of Chicken Immunoglobulins G, M, and A Using Monoclonal Antibodies1 M. H. ERHARD, I. VON QUISTORP, I. SCHRANNER, A. jtlNGLING, B. KASPERS, P. SCHMIDT, and R. K O H L M A N N
Institut fur Physiologie, Physiologische Chemie, und Ernahrungsphysiologie, Tierarztliche Fakultat, Universitat Munchen, Veterinarstrasse 13, 8000 Munchen 22, Germany (Received for publication June 3, 1991)
1992 Poultry Science 71:302-310
INTRODUCTION Yolk Ig of the chicken play an increasing role as an alternative to conventional mammal antisera for the production of polyclonal antibodies (Losch et al, 1986). These yolk antibodies, which mainly consist of IgG, can be used for diagnostic purposes, for prophylaxis, and for therapy in virology, bacteriology, and parasitology as well as in biochemistry (Gottstein and Hemmeler, 1985; Kuhlmann et al, 1988; Ricke et al, 1988; Yolken et al., 1988; Schmidt et al, 1989; Jungling et al, 1991; Wiedemann et al, 1991). Although there is
Dedicated to Uli Losch on the occasion of his 60th birthday.
a wide practical use of these antibodies, previously the chicken Ig assay has always been carried out with diffusion methods. Lebacq-Verheyden et al (1974) managed to identify IgM and IgA with an assay limit of 10 u g / m L and IgG with a n assay limit of 30 u g / m L in radial immunodiffusion by using isotype-specific antisera. The use of rocket immune electrophoresis (Schranner et al, 1987) lowered the assay minima to 10 u g / m L for IgG and IgM, and to 5 u g / m L for IgA. To detect very low concentrations as in immunodeficient animals (Fiedler et al, 1980; L6sch et al., 1981), in phylogenetical studies, in cell culture residue testing, or diagnostic, therapeutic, and prophylactic use of specific yolk Ig for infectious intestinal diseases (Wiedemann et al, 1991), the sensitivity of
302
Downloaded from http://ps.oxfordjournals.org/ by guest on April 13, 2015
ABSTRACT The aim of the present study was the development of a sensitive and specific ELISA system for the quantitative and qualitative assay of chicken Ig Isotypes G, M, and A using monoclonal antibodies. Five hybridoma cell lines were developed that synthesized specific antibodies against chicken IgG and three lines each producing specific antibodies against IgM or IgA. Using an immunodiffusion test, the subclasses were determined. Isolation of monoclonal antibodies from ascites was carried out by way of affinity chromatography with protein G sepharose. The purity of the eluates were determined by both SDSPAGE and HPLC. A Sandwich ELISA was found to be the most suitable technique for the assay. Specificity testing was carried out by Western blotting. An epitope analysis was also carried out. By variation of the single steps concerning incubation times, quantities, and concentrations of the substances to be applied, the whole procedure was optimized. Assay limits for individual Ig isotypes were determined. The limits were 20 n g / m L for IgG, 80 n g / m L for IgM, and 160 n g / m L for IgA. (Key words: detection system, enzyme-linked immunosorbent assay, chicken immunoglobulin, immunoglobulin isotypes, monoclonal antibodies)
ELISA SYSTEM FOR CHICKEN IMMUNOGLOBINS
303
Production and Isolation of Monoclonal Antibodies
MATERIALS AND METHODS Chemicals Lipopolysaccharide from Salmonella tninnesota RE 595 (LPS) and o-phenylenediamine (OPD) were purchased from Sigma. 2 Complete Freund's adjuvant (FCA) was obtained from Difco Laboratories, 3 and polyethylene glycol 1500 (PEG) and horseradish peroxidase (HRP) were purchased from Boehringer. 4 All other reagents were obtained from Merck. 5
Antibodies and Immunoglobulins Pure chicken IgG was obtained from Sigma. 2 Chicken IgA with .55 m g / m L was purified from bile and chicken IgM with .51 m g / m L from the serum of dysgammaglobulineamic chickens of Line UM-B19 (Schranner et ah, 1982). Furthermore, a standard serum (Sta 87) with 8.01 m g / m L IgG, 2.19 m g / m L IgM, and .11 m g / m L IgA was used. The protein content of pure
^igma Chemical Co., St. Louis, MO 63178-9916. 3 Difco Laboratories, Detroit, MI 48201. 4 Boehringer, Mannheim, D 6800, Germany, ^ e r c k , Darmstadt, D 6100, Germany. Pharmacia LKB, Uppsala 75182, Sweden. 7 Amicon, Beverly, MA 01915. 'Nunc, Wiesbaden, D 6200, Germany.
For each fusion a BALB/C mouse was immunized with the corresponding antigen (chicken IgG, IgM, or IgA, respectively). The initial immunization was carried out i.p. with 100 ug antigen in 100 uL FCA (Stahli et ah, 1980; Halk, 1986). After about 4 wk every mouse received 20 ug antigen on 3 consecutive days. On Day 1,25 ug LPS were added as a mitogen (Andersson et al., 1978). The fusion took place on the day after the last immunization. Cell fusion was carried out according to Kohler and Milstein (1975), modified as described by Peters et al. (1985) using PEG (Galfre et al., 1977). Per assay, 4 to 6 x 10 7 spleen cells were fused with 3.5 to 4.6 x 10 7 myeloma cells. Then, the cells were added into 10 mL HAT (hypoxantinaminopterin-thymidin) selection medium [81% RPMI-1640 medium, 15% fetal calf serum (FCS), 1% glutamin, 1% penicillinstreptomycin, and 2% HAT] and then applied into 96-microwell plates 8 with 100 |iL per well. Two weeks after the fusion, the media was changed to HT selection medium (2%), and after another 2 wk, the culture medium was changed to RPMI. With the help of indirect ELISA, the culture supernatants were tested for their specific antibody content 10 to 14 days after the fusion as described by Halk (1986) and Kelley et al. (1988). For this purpose, the ELISA plates were coated with the corresponding antigen. For the examination of H-
Downloaded from http://ps.oxfordjournals.org/ by guest on April 13, 2015
such material- a n d time-consuming substances was assayed according to Lowry methods is not satisfactory. et al. (1951). For the determination of extremely low concentrations of chicken Ig, a highly Purification of Chicken specific, sensitive, and simple ELISA Immunoglobulin A from Bile method required monoclonal antibodies Bile fluid was dialysed against .1 m o l / L (MAb), which were able to determine chicken Isotypes G, M, and A. The estab- sodium acetate buffer p H 5.0. After centrilishment of ELISA systems with poly- fugation (12,000 x g), the supernatant was clonal antibodies has been attempted. diluted with saturated ammonium-sulfate However, it was not possible to carry out solution 30 min at 4 C. After centrifugation a systematic classification of the various (3,000 x g), the sediment was resolved in .1 chicken Ig. Uses of MAb to date include a mol/L phosphate-buffered saline (PBS), p H monoclonal anti-chicken antibody that has 8.0, and chromatographed on a Ultrogel already been used by Mockett (1986) for AcA 22 column 6 (1.6 x 85 cm; fluid rate 5 the assay of BSA in ELISA. The purpose of mL/h). The pure IgA was eluated in the the present study was the development of first peak concentrated (.55 mg/mL) by a a sensitive and specific ELISA system for microconcentrater (100,000 Da). 7 the quantitative and qualitative assay of chicken IgG, IgM, and IgA.
304
ERHARD ET AL.
Determination of Specificity The specificity of each individual MAb for the corresponding antigen, particularly for its heavy chain, was determined with a Western blot (Towbin et al, 1979). Chicken serum IgG, IgM, and IgA were used as samples. One sample of each Ig isotype was
reduced by adding mercaptoethanol. Tor reduction, .5 mg of the sample was added to 60 |xL of sample buffer (SDS-PAGE) to which 5% 2-mercaptoethanol had been added. Then incubation was carried out 5 min at 100 C. After electrophoresis, the ensuing transfer onto nitrocellulose membrane (Nitrocellulose BA 83, pore size .2 nm 1 0 ) took 90 min at .8 mA/cm* and was made with a semidry electroblotter (SM17556) 11 as described by Kyhse-Andersen (1984). For this purpose, a transfer buffer according to Bjerrum and Schafner-Nielsen (1986) was used. In order to improve the layout, the same samples were applied to the right and left half of the gel. The nitrocellulose membrane was cut lengthwise after blotting. Staining of the left half was done specifically by using the HRP-labeled isotype-specific MAb. The protein bands on the right half were made visible by a protein stain with Indian ink (Hancock and Tsang, 1983). For later use in an ELISA, a MAb against each isotype was labeled to peroxidase using the periodate reaction technique (Nakane and Kawaoi, 1974).
Epitope Analysis Whether two MAb could recognize the same epitopes was investigated by way of direct competitive ELISA (van Zaane and Hulst, 1987). The method is based on a twostep procedure. First, a nonconjugated MAb (2 ng/mL) in dilution (log2) is applied onto an ELISA plate coated with a chicken Ig isotype. In the second step a competitive peroxidase conjugated MAb in saturated concentration (2 ug/mL) was added. A decrease in absorbance would mean that both antibodies were competing for the same or adjacent epitopes.
Development of a Sandwich Enzyme-Linked Immunosorbent Assay for the Quantitative Assay of Chicken Immunoglobulins
For the assay method to be developed, the Sandwich ELISA proved to be the most suitable. For optimization, various coating concentrations (2, 5, and 10 fig), different ^io-Rad, Munich, D 8000, Germany. 10 incubation times for blocking (30,60, and 90 Schleicher und Schuell, Dassel, D 3354, Germany. 11 min), for sample and conjugate (60,90, and Sartorius, Gottingen, D 3400, Germany.
Downloaded from http://ps.oxfordjournals.org/ by guest on April 13, 2015
chain specificity, the supernatants were tested against two isotypes each (e.g., IgG and IgM). Only those whose extinction differences exceeded .5 and which were specific for the isotype were classified as positive. The cultures that were judged positive in the screening test were cloned according to the "limiting dilution" method (Galfre and Milstein, 1981). The number of cloning steps required was computed with the Poisson probability distribution formula (Coller and Coller, 1986). Cultures with a probability rate above .99 were classified as monoclonal. Immunoglobulin class and subclass determination of MAb in concentrated cell culture supernatants were carried out by immunodiffusion according to Ouchterlony (1962) using subclass specific goatantimouse antisera. Large quantities of serum-free MAb were produced by administering 5 x 10 6 hybridoma cells to BALB/C mice i.p. following pristan treatment (.2 mL per mouse) 10 days before inoculation. The isolation of IgG antibodies from ascites was carried out by way of affinity chromatography using Protein G sepharose (Akerstr6m et al, 1985). After dialysis against PBS, p H 7.2, the protein content was measured spectrophotometrically at 280 nm the following day (Leslie and Clem, 1969). The purity of the antibody solution was determined by SDS-PAGE (Laemmli, 1970) using a linear pore gradient with an acrylamide concentration of 3 to 15% (Goerg et al., 1980). A further purity test was carried out with H P L C A Bio-Sil TSK-250 column 9 (300 x 7.5 m m at 280 nm) was used for this purpose. Twenty microliters of the purified MAb (1 m g / m L ) were added each time.
ELBA SYSTEM FOR CHICKEN IMMUNOGLOBINS
contents. For this purpose, Sta 87 served as a standard. Subsequently, mean values, standard deviations, and coefficients of variation were computed. RESULTS Production of Monoclonal Antibodies Against Chicken Immunoglobulin Isotypes G, M, and A
One fusion per Ig isotype was carried out in order to obtain antibodies against chicken Ig. To produce IgG-specific antibodies, 4 x 107 lymphocytes from the spleen were fused with 3.5 x 10' myeloma cells and cultured in two 96-well cell culture plates. Growth was recorded after 2 wk in 167 of 192 wells (87%). Nine cultures thereof reacted positively (5.4%) in screening testing. Five of these positive cultures (Gl, G2, G3, G4, and G5) were selected and cloned. The fusion to produce anti-IgM antibodies was achieved with 6 x 10' lymphocytes from the spleen and with 4 x 107 myeloma cells added onto four plates. Signs of growth appeared in 372 of 384 cavities (96.8%). Fourteen cultures thereof reacted positively in screening testing (3.8%). Three cultures (Ml, M2, and M3) were cloned. For anti-IgA antibody production, 5 x 107 Sensitivity Testing spleen cells were fused with 4.6 x 107 The standard serum, Sta 87, was applied myeloma cells and applied to four plates. In as a sample in log2 dilution steps. The assay 351 of 384 cavities (91.4%), growth could be limit of each individual Ig ELISA served as shown microscopically. Thirteen cultures showed a positive reaction in screening a dilution step of the standard curve that testing (3.7%). Three (Al, A2, and A3) were could still be distinguished significantly selected and cloned. Between 3 and 29.8 mL (P
Institut fur Physiologie, Physiologische Chemie, und Ernahrungsphysiologie, Tierarztliche Fakultat, Universitat Munchen, Veterinarstrasse 13, 8000 Munchen 22, Germany (Received for publication June 3, 1991)
1992 Poultry Science 71:302-310
INTRODUCTION Yolk Ig of the chicken play an increasing role as an alternative to conventional mammal antisera for the production of polyclonal antibodies (Losch et al, 1986). These yolk antibodies, which mainly consist of IgG, can be used for diagnostic purposes, for prophylaxis, and for therapy in virology, bacteriology, and parasitology as well as in biochemistry (Gottstein and Hemmeler, 1985; Kuhlmann et al, 1988; Ricke et al, 1988; Yolken et al., 1988; Schmidt et al, 1989; Jungling et al, 1991; Wiedemann et al, 1991). Although there is
Dedicated to Uli Losch on the occasion of his 60th birthday.
a wide practical use of these antibodies, previously the chicken Ig assay has always been carried out with diffusion methods. Lebacq-Verheyden et al (1974) managed to identify IgM and IgA with an assay limit of 10 u g / m L and IgG with a n assay limit of 30 u g / m L in radial immunodiffusion by using isotype-specific antisera. The use of rocket immune electrophoresis (Schranner et al, 1987) lowered the assay minima to 10 u g / m L for IgG and IgM, and to 5 u g / m L for IgA. To detect very low concentrations as in immunodeficient animals (Fiedler et al, 1980; L6sch et al., 1981), in phylogenetical studies, in cell culture residue testing, or diagnostic, therapeutic, and prophylactic use of specific yolk Ig for infectious intestinal diseases (Wiedemann et al, 1991), the sensitivity of
302
Downloaded from http://ps.oxfordjournals.org/ by guest on April 13, 2015
ABSTRACT The aim of the present study was the development of a sensitive and specific ELISA system for the quantitative and qualitative assay of chicken Ig Isotypes G, M, and A using monoclonal antibodies. Five hybridoma cell lines were developed that synthesized specific antibodies against chicken IgG and three lines each producing specific antibodies against IgM or IgA. Using an immunodiffusion test, the subclasses were determined. Isolation of monoclonal antibodies from ascites was carried out by way of affinity chromatography with protein G sepharose. The purity of the eluates were determined by both SDSPAGE and HPLC. A Sandwich ELISA was found to be the most suitable technique for the assay. Specificity testing was carried out by Western blotting. An epitope analysis was also carried out. By variation of the single steps concerning incubation times, quantities, and concentrations of the substances to be applied, the whole procedure was optimized. Assay limits for individual Ig isotypes were determined. The limits were 20 n g / m L for IgG, 80 n g / m L for IgM, and 160 n g / m L for IgA. (Key words: detection system, enzyme-linked immunosorbent assay, chicken immunoglobulin, immunoglobulin isotypes, monoclonal antibodies)
ELISA SYSTEM FOR CHICKEN IMMUNOGLOBINS
303
Production and Isolation of Monoclonal Antibodies
MATERIALS AND METHODS Chemicals Lipopolysaccharide from Salmonella tninnesota RE 595 (LPS) and o-phenylenediamine (OPD) were purchased from Sigma. 2 Complete Freund's adjuvant (FCA) was obtained from Difco Laboratories, 3 and polyethylene glycol 1500 (PEG) and horseradish peroxidase (HRP) were purchased from Boehringer. 4 All other reagents were obtained from Merck. 5
Antibodies and Immunoglobulins Pure chicken IgG was obtained from Sigma. 2 Chicken IgA with .55 m g / m L was purified from bile and chicken IgM with .51 m g / m L from the serum of dysgammaglobulineamic chickens of Line UM-B19 (Schranner et ah, 1982). Furthermore, a standard serum (Sta 87) with 8.01 m g / m L IgG, 2.19 m g / m L IgM, and .11 m g / m L IgA was used. The protein content of pure
^igma Chemical Co., St. Louis, MO 63178-9916. 3 Difco Laboratories, Detroit, MI 48201. 4 Boehringer, Mannheim, D 6800, Germany, ^ e r c k , Darmstadt, D 6100, Germany. Pharmacia LKB, Uppsala 75182, Sweden. 7 Amicon, Beverly, MA 01915. 'Nunc, Wiesbaden, D 6200, Germany.
For each fusion a BALB/C mouse was immunized with the corresponding antigen (chicken IgG, IgM, or IgA, respectively). The initial immunization was carried out i.p. with 100 ug antigen in 100 uL FCA (Stahli et ah, 1980; Halk, 1986). After about 4 wk every mouse received 20 ug antigen on 3 consecutive days. On Day 1,25 ug LPS were added as a mitogen (Andersson et al., 1978). The fusion took place on the day after the last immunization. Cell fusion was carried out according to Kohler and Milstein (1975), modified as described by Peters et al. (1985) using PEG (Galfre et al., 1977). Per assay, 4 to 6 x 10 7 spleen cells were fused with 3.5 to 4.6 x 10 7 myeloma cells. Then, the cells were added into 10 mL HAT (hypoxantinaminopterin-thymidin) selection medium [81% RPMI-1640 medium, 15% fetal calf serum (FCS), 1% glutamin, 1% penicillinstreptomycin, and 2% HAT] and then applied into 96-microwell plates 8 with 100 |iL per well. Two weeks after the fusion, the media was changed to HT selection medium (2%), and after another 2 wk, the culture medium was changed to RPMI. With the help of indirect ELISA, the culture supernatants were tested for their specific antibody content 10 to 14 days after the fusion as described by Halk (1986) and Kelley et al. (1988). For this purpose, the ELISA plates were coated with the corresponding antigen. For the examination of H-
Downloaded from http://ps.oxfordjournals.org/ by guest on April 13, 2015
such material- a n d time-consuming substances was assayed according to Lowry methods is not satisfactory. et al. (1951). For the determination of extremely low concentrations of chicken Ig, a highly Purification of Chicken specific, sensitive, and simple ELISA Immunoglobulin A from Bile method required monoclonal antibodies Bile fluid was dialysed against .1 m o l / L (MAb), which were able to determine chicken Isotypes G, M, and A. The estab- sodium acetate buffer p H 5.0. After centrilishment of ELISA systems with poly- fugation (12,000 x g), the supernatant was clonal antibodies has been attempted. diluted with saturated ammonium-sulfate However, it was not possible to carry out solution 30 min at 4 C. After centrifugation a systematic classification of the various (3,000 x g), the sediment was resolved in .1 chicken Ig. Uses of MAb to date include a mol/L phosphate-buffered saline (PBS), p H monoclonal anti-chicken antibody that has 8.0, and chromatographed on a Ultrogel already been used by Mockett (1986) for AcA 22 column 6 (1.6 x 85 cm; fluid rate 5 the assay of BSA in ELISA. The purpose of mL/h). The pure IgA was eluated in the the present study was the development of first peak concentrated (.55 mg/mL) by a a sensitive and specific ELISA system for microconcentrater (100,000 Da). 7 the quantitative and qualitative assay of chicken IgG, IgM, and IgA.
304
ERHARD ET AL.
Determination of Specificity The specificity of each individual MAb for the corresponding antigen, particularly for its heavy chain, was determined with a Western blot (Towbin et al, 1979). Chicken serum IgG, IgM, and IgA were used as samples. One sample of each Ig isotype was
reduced by adding mercaptoethanol. Tor reduction, .5 mg of the sample was added to 60 |xL of sample buffer (SDS-PAGE) to which 5% 2-mercaptoethanol had been added. Then incubation was carried out 5 min at 100 C. After electrophoresis, the ensuing transfer onto nitrocellulose membrane (Nitrocellulose BA 83, pore size .2 nm 1 0 ) took 90 min at .8 mA/cm* and was made with a semidry electroblotter (SM17556) 11 as described by Kyhse-Andersen (1984). For this purpose, a transfer buffer according to Bjerrum and Schafner-Nielsen (1986) was used. In order to improve the layout, the same samples were applied to the right and left half of the gel. The nitrocellulose membrane was cut lengthwise after blotting. Staining of the left half was done specifically by using the HRP-labeled isotype-specific MAb. The protein bands on the right half were made visible by a protein stain with Indian ink (Hancock and Tsang, 1983). For later use in an ELISA, a MAb against each isotype was labeled to peroxidase using the periodate reaction technique (Nakane and Kawaoi, 1974).
Epitope Analysis Whether two MAb could recognize the same epitopes was investigated by way of direct competitive ELISA (van Zaane and Hulst, 1987). The method is based on a twostep procedure. First, a nonconjugated MAb (2 ng/mL) in dilution (log2) is applied onto an ELISA plate coated with a chicken Ig isotype. In the second step a competitive peroxidase conjugated MAb in saturated concentration (2 ug/mL) was added. A decrease in absorbance would mean that both antibodies were competing for the same or adjacent epitopes.
Development of a Sandwich Enzyme-Linked Immunosorbent Assay for the Quantitative Assay of Chicken Immunoglobulins
For the assay method to be developed, the Sandwich ELISA proved to be the most suitable. For optimization, various coating concentrations (2, 5, and 10 fig), different ^io-Rad, Munich, D 8000, Germany. 10 incubation times for blocking (30,60, and 90 Schleicher und Schuell, Dassel, D 3354, Germany. 11 min), for sample and conjugate (60,90, and Sartorius, Gottingen, D 3400, Germany.
Downloaded from http://ps.oxfordjournals.org/ by guest on April 13, 2015
chain specificity, the supernatants were tested against two isotypes each (e.g., IgG and IgM). Only those whose extinction differences exceeded .5 and which were specific for the isotype were classified as positive. The cultures that were judged positive in the screening test were cloned according to the "limiting dilution" method (Galfre and Milstein, 1981). The number of cloning steps required was computed with the Poisson probability distribution formula (Coller and Coller, 1986). Cultures with a probability rate above .99 were classified as monoclonal. Immunoglobulin class and subclass determination of MAb in concentrated cell culture supernatants were carried out by immunodiffusion according to Ouchterlony (1962) using subclass specific goatantimouse antisera. Large quantities of serum-free MAb were produced by administering 5 x 10 6 hybridoma cells to BALB/C mice i.p. following pristan treatment (.2 mL per mouse) 10 days before inoculation. The isolation of IgG antibodies from ascites was carried out by way of affinity chromatography using Protein G sepharose (Akerstr6m et al, 1985). After dialysis against PBS, p H 7.2, the protein content was measured spectrophotometrically at 280 nm the following day (Leslie and Clem, 1969). The purity of the antibody solution was determined by SDS-PAGE (Laemmli, 1970) using a linear pore gradient with an acrylamide concentration of 3 to 15% (Goerg et al., 1980). A further purity test was carried out with H P L C A Bio-Sil TSK-250 column 9 (300 x 7.5 m m at 280 nm) was used for this purpose. Twenty microliters of the purified MAb (1 m g / m L ) were added each time.
ELBA SYSTEM FOR CHICKEN IMMUNOGLOBINS
contents. For this purpose, Sta 87 served as a standard. Subsequently, mean values, standard deviations, and coefficients of variation were computed. RESULTS Production of Monoclonal Antibodies Against Chicken Immunoglobulin Isotypes G, M, and A
One fusion per Ig isotype was carried out in order to obtain antibodies against chicken Ig. To produce IgG-specific antibodies, 4 x 107 lymphocytes from the spleen were fused with 3.5 x 10' myeloma cells and cultured in two 96-well cell culture plates. Growth was recorded after 2 wk in 167 of 192 wells (87%). Nine cultures thereof reacted positively (5.4%) in screening testing. Five of these positive cultures (Gl, G2, G3, G4, and G5) were selected and cloned. The fusion to produce anti-IgM antibodies was achieved with 6 x 10' lymphocytes from the spleen and with 4 x 107 myeloma cells added onto four plates. Signs of growth appeared in 372 of 384 cavities (96.8%). Fourteen cultures thereof reacted positively in screening testing (3.8%). Three cultures (Ml, M2, and M3) were cloned. For anti-IgA antibody production, 5 x 107 Sensitivity Testing spleen cells were fused with 4.6 x 107 The standard serum, Sta 87, was applied myeloma cells and applied to four plates. In as a sample in log2 dilution steps. The assay 351 of 384 cavities (91.4%), growth could be limit of each individual Ig ELISA served as shown microscopically. Thirteen cultures showed a positive reaction in screening a dilution step of the standard curve that testing (3.7%). Three (Al, A2, and A3) were could still be distinguished significantly selected and cloned. Between 3 and 29.8 mL (P
