Journal of Autoimmunity
(1992) $363-377
Evidence that a Putative Anti-idiotypic Monoclonal Antibody may Actually be Recognizing Circulating Immune Complexes
Jo& L. Sub&a, Ana Caturla, Luis F. Pereira, Manuel C. Camargo, Ana Bustos, Ram6n Boimorto and Emilio G. de la Concha Department
of Immunology,
Hospital
Universitario
San Carlos,
Madrid,
Spain
(Received 19 September 1991 and accepted 19 December 1991) Murine monoclonal antibodies (mAb) reacting with affinity-purified antihistone antibodies (AHA) from serum of a patient with systemic lupus erythematosus (SLE) were obtained. One of them, 8B3, was initially considered to recognize idiotypic (Id) determinants in AI-IA since (a) it reacted with AI-IA but not with control IgG, (b) this reactivity could be inhibited using affinity-purified AI-IA, but not with control IgG or whole serum; (c) affinity-isolated 8B3+ antibodies showed antihistone activity and not other activities tested so far; (d) antihistone activity due to 8B3+, but not that of 8B3- from the same serum, could be fully inhibited by the presence of 8B3 mAb in the antihistone assay and (e) serum levels of 8B3 reactivity were higher than normal in SLE patients with AHA (56%), in contrast with SLE patients without AI-IA (6%). From these results it was deduced that 8B3 defined a cross-reactive Id shared by a subset of AI-IA in SLE patients. However, the present results suggest that (a) 8B3 mAb did not recognize AHA or Ig, but did recognize a 55 ltDa histone-binding protein, (b) this 55 kDa protein was present free at low concentration in all human sera, but also associated with IgG in 8B3+ SLE sera and (c) these complexes are responsible for the false positive results in the antihistone assay as shown for DNA/anti-DNA complexes. Thus, mAbs recognizing the non-Ig moiety of circulating immune complexes may resemble anti-Id antibodies with features of the so-called epibodies. These immune complexes may be responsible for false positive results and caution should be exercised in the interpretation of results.
Introduction
Antihistone antibodies (AHA) are frequently detected in patients with drug-induced lupus or systemic lupus erythematosus (SLE) although they may also appear in other Correspondence to: Jose L. Subiza, M.D., Department of Immunology, Hospital Universitario San Carlos, E-28040 Madrid, Spain. 363 0896-841
l/92/030363+
15 $03.00/O
0 1992 Academic Press Limited
364
Josh L. Subiza et al.
autoimmune diseases [ 11.Many studies have been performed to analyse the diversity and fine specificity of AHA in SLE patients or in patients with other conditions [2-51. Results in SLE patients, however, may be misleading since AHA can be easily overestimated owing to the presence of deoxyribonucleic acid (DNA)/anti-DNA complexes [6] that bind histones through their DNA moiety. These artifactual crossreactivities with histones have also been observed using monoclonal antibodies (mAbs) reacting with DNA [7,8] or when testing for autoantibodies to other DNAbinding proteins in sera of SLE patients containing DNA/anti-DNA complexes [9-l 11.An alternative way to study diversity and relationships among autoantibodies is to analyse their idiotypes (Id) [12]. This kind of approach has resulted in better definition of different autoantibody systems reviewed in [13]. Therefore we produced anti-Id mAb to AHA obtained from a SLE patient in order to obtain appropriate anti-Id reagents. Here we describe a mAb fulfilling the standard criteria of anti-Id that actually recognizes a 55 kDa histone-binding serum protein. In SLE sera, this protein may be found associated with IgG, and these complexes are detected as a false positive antihistone activity. Materials and methods Serumsamples
Serum was obtained as described [6] from 56 patients that fulfilled the revised American Rheumatism Association criteria, for SLE [14] and from 42 healthy individuals. Measurement of antibodies to histones, DNA, cardiolipin and Candida albicans antigen
As previously described these antibodies were assessed by specific ELISA using histones, double stranded-(ds)-DNA, single stranded-(ss)-DNA [6] or cardiolipin [ 151. C. albicans antigen (Holister-Stier, Spokane, WA, USA) was bound to polystyrene wells (EIA-plates, Costar, Cambridge, MA, USA) at 5 pg/ml in phosphate buffered saline, pH 7.2 (PBS) by an overnight incubation at 4°C. After coating, wells were washed with I?BS-0.05°/0 Tween, neutralized with 1o/0bovine serum albumin (BSA) and test or control samples were distributed (100 ~1) as described [6]. In some experiments, antihistone activity was assessed in the presence of increased concentrations (10 to 270 ug/ml) of a putative anti-Id mAb. Results were expressed as the percentage of inhibition as compared to the initial activity (without mAb added). Puri’cation of AHA
Five ml of serum from M.J., a SLE patient showing a high titer of antihistone activity, was treated with deoxyribonuclease as described [6]. IgGs were affinity purified using commercially available recombinant protein G-agarose (Gammabind; Genex, Gaithersburg, MD, USA). AHA were then purified using a histone alIinity column (type IIS; Sigma, St. Louis, MO, USA) [6]. The sample (5 ml; 10 mg/ml in PBS) was run onto a 15 ml-column, previously equilibrated with PBS, and washed
Putative anti-Id mAb recognizing immune complexes
365
with the same buffer. The IgGs retained were recovered by washing with 2 M NaSCN and tested for antihistone activity once dialysed against PBS. Selected fractions were pooled, aliquoted and frozen until used. The yield was 42:/, of the initial antihistone activity, enriched 46-fold. Preparation of mAbs BALB/c mice were immunized (100 ug ip weekly) with afhnity-purified AHA from patient M.J. using the protocol described by Salomon et al. [16]. The titers were monitored using an ELISA as described below. Test samples were absorbed with control IgG. Elevated reactivity with AHA was noted at 12 weeks of immunization. Three days after a booster iv injection (50 ug), cell fusion was carried out as described by Gil et al. [17]. Culture supernatants were assayed for anti-AHA by ELISA. Briefly, EIA-plates were coated with affinity-purified AHA from patient M.J. or protein G purified IgG (pooled from controls) at a concentration of 5 ug/ml in PBS. Undiluted hybridoma culture supernatants were added to the wells and incubated for 1 h. After washing with PBS containing 0.05% Tween 20, a dilution of peroxidase-coupled monospecific antibodies to mouse IgG (Pel-Freez, Rogers, AR, USA) in PBS/l Y0 BSA was added. After 1 h incubation and a final wash, the substrate solution (0.05% H,O,, 0.02 M 0-phenylenediamine, 0.1 M citrate buffer; pH 5.5) was added. The calorimetric reaction was read at 492 mn. Selected hybridomas were cloned twice by limiting dilution and the clones were expanded. Isotype typing was carried out using Ouchterlony diffusion gels (Serotec, Kidlington, UK). In some experiments, mAb reactivity was measured in the presence of increasing concentrations (11 to 900 us/ml) of AHA, control IgG or whole serum as inhibitors. These assays were performed as described above using an appropriate dilution of mAb to achieve a higher sensitivity. Results were expressed as the percentage value of inhibition compared with control samples. Screening for 8B3 reactivity 8B3 obtained in this study was a mAb with putative anti-idiotypic activity. Its reactivity with control and SLE sera was screened by ELISA as described [ 18-201. Thus, EIA-plates were coated with the appropriate dilution of serum samples. After overnight incubation at 4”C, microwells were washed and the 8B3 mAb was added and incubated for 1 h. After washing, the same procedure described above for hybridoma screening was performed. The serum dilution used for coating was 1: 1,000, which corresponded approximately to the middle point of a curve established with serum M.J. Purification of 8B3-reacting protein The protein recognized by 8B3 mAb was purified by affinity chromatography using Sepharose CL-4B (Pharmacia, Uppsala, Sweden). Human serum samples were applied to these columns after adsorption on mouse IgG columns to avoid nonspecific binding to the mAb. Thus, 0.5 ml of the serum sample were passed onto the first column (polyclonal murine IgG) and the flow-through fraction, washing with PBS, applied to the 8B3-column. After washing with PBS (at least 10 times the
366 10~6L. Subiza et al.
column volume), the protein retained was eluted with 2 M NaSCN, dialysed extensively against PBS and concentrated by vacuum dialysis or lyophilization. Gel filtration Six hundred ul of serum were filtered through a Sephacryl S-300 (Pharmacia) column (100 x 2.6 cm) equilibrated in PBS using a linear flow rate of 5 cm/h and calibrated with Blue Dextran 2000, IgG and BSA. Four ml fractions were collected, monitoring the protein at 280 nm. Each fraction was tested for 8B3 reactivity as above, by coating wells of EIA-plates at 1:20 dilution in PBS. Pooled IgG-containing and albumin-containing fractions, FRl and FR2 respectively, were concentrated by vacuum dialysis, both preserving the same final volume. Polyacrylamide gel electrophoresis (PAGE)
and immunoblotting
Proteins eluted fron 8B3-columns were lyophilized and diluted in 0.1 M Tris-.HCl, 2% sodium dodecyl sulfate, 10% glycerol, O.Ol”/ bromophenol blue, pH 6.8. These samples (2 ug of protein when possible) were run on 4-20% polyacrylamide continuous gradient (10% PAGE when immunoblotting), 1 .O mm-thick slab gels, under non-reducing conditions. Electrophoresis and immunoblotting were carried out as described [21]; proteins were stained with silver (BioRad) and their molecular weight (MW) estimated using commercial protein standards. Bands transferred to nitrocellulose sheets were tested with mAbs or peroxidase labelled anti-human IgG (H + L) (Tago, Burlingame, CA, USA) diluted in PBS containing 0.3% gelatin and 0.05% Tween 20. After incubation for 1 h at room temperature and washing, the corresponding lines were filled with peroxidase labelled anti-mouse IgG (Pel-Freez). After incubation, membranes were washed and soaked in a substrate solution. The enzyme reaction was stopped by rinsing with distilled water. Protein A affinity-columns In some experiments IgG were purified from IgG-containing fraction (FRl from Sephacryl S-300) with protein A-Sepharose (Pharmacia). Binding of IgG to protein A was performed in isotonic medium (PBS) or under conditions of high ionic strength (1.5 M glycine, 3 M NaCl, pH 8.9) as described by the supplier (Pharmacia). In both cases desorption was achieved with 0.1 M citric acid, pH 3.5. Effluents and eluents of the columns were dialysed against PBS and concentrated by vacuum dialysis preserving the same final volume. Afinity of 8B3-reacting protein to histones ELISA-plates were coated with a highly purified histone preparation (CalbiochemBehring, San Diego, CA, USA) and BSA or gelatin at 5 ug/ml in PBS. Once neutralized with 1 o/0BSA, increasing amounts of afBnity-purified 8B3-reacting protein (0.3 to 4 ug/ml) were added to PBS-0.5% Tween 20 and incubated for 1 h. After washing, the adequate dilution of 8B3 mAb in PBS containing 1 y0 BSA and 0.05% Tween 20 was distributed.
Putative anti-Id mAb recognizing immune complexes 367
0 II
33
100 Inhibitor
300
900
tug/ml)
Figure 1. Reactivity of 8B3 mAb with antihistone antibodies (AHA) from patient M. J., as detected by ELISA, in the presence of the following inhibitors: AHA (0), IgG (a) and whole serum (0) from patient M.J., or pooled IgG (A) or serum (W) from control individuals.
Statistical Data were analysed using Microstat,
analysis
an Ecosoft computer program. Results
Characterization
of the 833 mAb
More than four hundred hybrids, from two fusions, were screened and of these six were identified as producers of putative anti-Id antibodies. One of these, 8B3, an IgG2bk, reacted in the ELISA with affinity-purified AHA from patient M.J. (typically 0.8-l .5 OD), but not with IgG from pooled normal human sera (0.0-O. 1 OD). Reactivity of purified 8B3 mAb with AHA could not be inhibited by increasing amounts of whole serum or control IgG (Figure 1). As shown, a low but significant (P=O.O2; hypothesis test for means), inhibition was observed with the highest concentration of IgG from M. J. These IgG once enriched in AHA were clearly inhibitory in the ELISA (Figure 1). These results indicated that 8B3 did not react with IgG constant region determinants but with IgG showing antihistone activity. To support this conclusion, M.J. serum was purified on a 8B3 aflinity column. As shown in Table 1, only IgG showing antihistone activity could be eluted from the column. Other activities including antihistone activity, also present in M. J. serum, were all recovered in the column effluent (Table 1). As AHA were present in both 8B3 column eluent (8B3+ AHA) and effluent (8B3- AHA), antihistone inhibition assays were carried out using purified 8B3 mAb as inhibitor. Figure 2 shows that antihistone activity due to 8B3+ AHA could be fully inhibited by 8B3 mAb, while that due to 8B3- AHA was unaffected. These results suggested that 8B3 mAb was reacting
368 Jest L. Subiza et al. Table 1. Aj?inity-purified
8B3+ antibodiesfrompatient
M.J. binds to histones*
Effluents from Wells coated with
Eluents from
8B3-column
Control-column
8B3-column
>2.5 1.87fO.2 >2.5 >2.5 0.84 + 0.0
> 2.5 1.66kO.l > 2.5 > 2.5 0.73kO.l
1.42k0.1 0.31 fO.O 0.62fO.l 0.04+ 0.0 0.02 + 0.1
Histones DNAds DNAss
C. albicans antigen Cardiolipin
Control-column 0.33kO.O 0.25kO.O 0.73kO.O 0.06 f 0.0 0.06 f 0.0
*Serum from patient M.J. was passed onto an 8B3-column or a control column (unrelated mAb). Effluents and eluents from both columns were tested for IgG reactivity by ELISA, preserving the same linal volumes. Results are expressed in OD at 492 mn (mean k SD of duplicates). Differences between the columns were all NS except for the antihistone activity detected in the column eluents (F2.5
PI
< 0.001
(1992) $363-377
Evidence that a Putative Anti-idiotypic Monoclonal Antibody may Actually be Recognizing Circulating Immune Complexes
Jo& L. Sub&a, Ana Caturla, Luis F. Pereira, Manuel C. Camargo, Ana Bustos, Ram6n Boimorto and Emilio G. de la Concha Department
of Immunology,
Hospital
Universitario
San Carlos,
Madrid,
Spain
(Received 19 September 1991 and accepted 19 December 1991) Murine monoclonal antibodies (mAb) reacting with affinity-purified antihistone antibodies (AHA) from serum of a patient with systemic lupus erythematosus (SLE) were obtained. One of them, 8B3, was initially considered to recognize idiotypic (Id) determinants in AI-IA since (a) it reacted with AI-IA but not with control IgG, (b) this reactivity could be inhibited using affinity-purified AI-IA, but not with control IgG or whole serum; (c) affinity-isolated 8B3+ antibodies showed antihistone activity and not other activities tested so far; (d) antihistone activity due to 8B3+, but not that of 8B3- from the same serum, could be fully inhibited by the presence of 8B3 mAb in the antihistone assay and (e) serum levels of 8B3 reactivity were higher than normal in SLE patients with AHA (56%), in contrast with SLE patients without AI-IA (6%). From these results it was deduced that 8B3 defined a cross-reactive Id shared by a subset of AI-IA in SLE patients. However, the present results suggest that (a) 8B3 mAb did not recognize AHA or Ig, but did recognize a 55 ltDa histone-binding protein, (b) this 55 kDa protein was present free at low concentration in all human sera, but also associated with IgG in 8B3+ SLE sera and (c) these complexes are responsible for the false positive results in the antihistone assay as shown for DNA/anti-DNA complexes. Thus, mAbs recognizing the non-Ig moiety of circulating immune complexes may resemble anti-Id antibodies with features of the so-called epibodies. These immune complexes may be responsible for false positive results and caution should be exercised in the interpretation of results.
Introduction
Antihistone antibodies (AHA) are frequently detected in patients with drug-induced lupus or systemic lupus erythematosus (SLE) although they may also appear in other Correspondence to: Jose L. Subiza, M.D., Department of Immunology, Hospital Universitario San Carlos, E-28040 Madrid, Spain. 363 0896-841
l/92/030363+
15 $03.00/O
0 1992 Academic Press Limited
364
Josh L. Subiza et al.
autoimmune diseases [ 11.Many studies have been performed to analyse the diversity and fine specificity of AHA in SLE patients or in patients with other conditions [2-51. Results in SLE patients, however, may be misleading since AHA can be easily overestimated owing to the presence of deoxyribonucleic acid (DNA)/anti-DNA complexes [6] that bind histones through their DNA moiety. These artifactual crossreactivities with histones have also been observed using monoclonal antibodies (mAbs) reacting with DNA [7,8] or when testing for autoantibodies to other DNAbinding proteins in sera of SLE patients containing DNA/anti-DNA complexes [9-l 11.An alternative way to study diversity and relationships among autoantibodies is to analyse their idiotypes (Id) [12]. This kind of approach has resulted in better definition of different autoantibody systems reviewed in [13]. Therefore we produced anti-Id mAb to AHA obtained from a SLE patient in order to obtain appropriate anti-Id reagents. Here we describe a mAb fulfilling the standard criteria of anti-Id that actually recognizes a 55 kDa histone-binding serum protein. In SLE sera, this protein may be found associated with IgG, and these complexes are detected as a false positive antihistone activity. Materials and methods Serumsamples
Serum was obtained as described [6] from 56 patients that fulfilled the revised American Rheumatism Association criteria, for SLE [14] and from 42 healthy individuals. Measurement of antibodies to histones, DNA, cardiolipin and Candida albicans antigen
As previously described these antibodies were assessed by specific ELISA using histones, double stranded-(ds)-DNA, single stranded-(ss)-DNA [6] or cardiolipin [ 151. C. albicans antigen (Holister-Stier, Spokane, WA, USA) was bound to polystyrene wells (EIA-plates, Costar, Cambridge, MA, USA) at 5 pg/ml in phosphate buffered saline, pH 7.2 (PBS) by an overnight incubation at 4°C. After coating, wells were washed with I?BS-0.05°/0 Tween, neutralized with 1o/0bovine serum albumin (BSA) and test or control samples were distributed (100 ~1) as described [6]. In some experiments, antihistone activity was assessed in the presence of increased concentrations (10 to 270 ug/ml) of a putative anti-Id mAb. Results were expressed as the percentage of inhibition as compared to the initial activity (without mAb added). Puri’cation of AHA
Five ml of serum from M.J., a SLE patient showing a high titer of antihistone activity, was treated with deoxyribonuclease as described [6]. IgGs were affinity purified using commercially available recombinant protein G-agarose (Gammabind; Genex, Gaithersburg, MD, USA). AHA were then purified using a histone alIinity column (type IIS; Sigma, St. Louis, MO, USA) [6]. The sample (5 ml; 10 mg/ml in PBS) was run onto a 15 ml-column, previously equilibrated with PBS, and washed
Putative anti-Id mAb recognizing immune complexes
365
with the same buffer. The IgGs retained were recovered by washing with 2 M NaSCN and tested for antihistone activity once dialysed against PBS. Selected fractions were pooled, aliquoted and frozen until used. The yield was 42:/, of the initial antihistone activity, enriched 46-fold. Preparation of mAbs BALB/c mice were immunized (100 ug ip weekly) with afhnity-purified AHA from patient M.J. using the protocol described by Salomon et al. [16]. The titers were monitored using an ELISA as described below. Test samples were absorbed with control IgG. Elevated reactivity with AHA was noted at 12 weeks of immunization. Three days after a booster iv injection (50 ug), cell fusion was carried out as described by Gil et al. [17]. Culture supernatants were assayed for anti-AHA by ELISA. Briefly, EIA-plates were coated with affinity-purified AHA from patient M.J. or protein G purified IgG (pooled from controls) at a concentration of 5 ug/ml in PBS. Undiluted hybridoma culture supernatants were added to the wells and incubated for 1 h. After washing with PBS containing 0.05% Tween 20, a dilution of peroxidase-coupled monospecific antibodies to mouse IgG (Pel-Freez, Rogers, AR, USA) in PBS/l Y0 BSA was added. After 1 h incubation and a final wash, the substrate solution (0.05% H,O,, 0.02 M 0-phenylenediamine, 0.1 M citrate buffer; pH 5.5) was added. The calorimetric reaction was read at 492 mn. Selected hybridomas were cloned twice by limiting dilution and the clones were expanded. Isotype typing was carried out using Ouchterlony diffusion gels (Serotec, Kidlington, UK). In some experiments, mAb reactivity was measured in the presence of increasing concentrations (11 to 900 us/ml) of AHA, control IgG or whole serum as inhibitors. These assays were performed as described above using an appropriate dilution of mAb to achieve a higher sensitivity. Results were expressed as the percentage value of inhibition compared with control samples. Screening for 8B3 reactivity 8B3 obtained in this study was a mAb with putative anti-idiotypic activity. Its reactivity with control and SLE sera was screened by ELISA as described [ 18-201. Thus, EIA-plates were coated with the appropriate dilution of serum samples. After overnight incubation at 4”C, microwells were washed and the 8B3 mAb was added and incubated for 1 h. After washing, the same procedure described above for hybridoma screening was performed. The serum dilution used for coating was 1: 1,000, which corresponded approximately to the middle point of a curve established with serum M.J. Purification of 8B3-reacting protein The protein recognized by 8B3 mAb was purified by affinity chromatography using Sepharose CL-4B (Pharmacia, Uppsala, Sweden). Human serum samples were applied to these columns after adsorption on mouse IgG columns to avoid nonspecific binding to the mAb. Thus, 0.5 ml of the serum sample were passed onto the first column (polyclonal murine IgG) and the flow-through fraction, washing with PBS, applied to the 8B3-column. After washing with PBS (at least 10 times the
366 10~6L. Subiza et al.
column volume), the protein retained was eluted with 2 M NaSCN, dialysed extensively against PBS and concentrated by vacuum dialysis or lyophilization. Gel filtration Six hundred ul of serum were filtered through a Sephacryl S-300 (Pharmacia) column (100 x 2.6 cm) equilibrated in PBS using a linear flow rate of 5 cm/h and calibrated with Blue Dextran 2000, IgG and BSA. Four ml fractions were collected, monitoring the protein at 280 nm. Each fraction was tested for 8B3 reactivity as above, by coating wells of EIA-plates at 1:20 dilution in PBS. Pooled IgG-containing and albumin-containing fractions, FRl and FR2 respectively, were concentrated by vacuum dialysis, both preserving the same final volume. Polyacrylamide gel electrophoresis (PAGE)
and immunoblotting
Proteins eluted fron 8B3-columns were lyophilized and diluted in 0.1 M Tris-.HCl, 2% sodium dodecyl sulfate, 10% glycerol, O.Ol”/ bromophenol blue, pH 6.8. These samples (2 ug of protein when possible) were run on 4-20% polyacrylamide continuous gradient (10% PAGE when immunoblotting), 1 .O mm-thick slab gels, under non-reducing conditions. Electrophoresis and immunoblotting were carried out as described [21]; proteins were stained with silver (BioRad) and their molecular weight (MW) estimated using commercial protein standards. Bands transferred to nitrocellulose sheets were tested with mAbs or peroxidase labelled anti-human IgG (H + L) (Tago, Burlingame, CA, USA) diluted in PBS containing 0.3% gelatin and 0.05% Tween 20. After incubation for 1 h at room temperature and washing, the corresponding lines were filled with peroxidase labelled anti-mouse IgG (Pel-Freez). After incubation, membranes were washed and soaked in a substrate solution. The enzyme reaction was stopped by rinsing with distilled water. Protein A affinity-columns In some experiments IgG were purified from IgG-containing fraction (FRl from Sephacryl S-300) with protein A-Sepharose (Pharmacia). Binding of IgG to protein A was performed in isotonic medium (PBS) or under conditions of high ionic strength (1.5 M glycine, 3 M NaCl, pH 8.9) as described by the supplier (Pharmacia). In both cases desorption was achieved with 0.1 M citric acid, pH 3.5. Effluents and eluents of the columns were dialysed against PBS and concentrated by vacuum dialysis preserving the same final volume. Afinity of 8B3-reacting protein to histones ELISA-plates were coated with a highly purified histone preparation (CalbiochemBehring, San Diego, CA, USA) and BSA or gelatin at 5 ug/ml in PBS. Once neutralized with 1 o/0BSA, increasing amounts of afBnity-purified 8B3-reacting protein (0.3 to 4 ug/ml) were added to PBS-0.5% Tween 20 and incubated for 1 h. After washing, the adequate dilution of 8B3 mAb in PBS containing 1 y0 BSA and 0.05% Tween 20 was distributed.
Putative anti-Id mAb recognizing immune complexes 367
0 II
33
100 Inhibitor
300
900
tug/ml)
Figure 1. Reactivity of 8B3 mAb with antihistone antibodies (AHA) from patient M. J., as detected by ELISA, in the presence of the following inhibitors: AHA (0), IgG (a) and whole serum (0) from patient M.J., or pooled IgG (A) or serum (W) from control individuals.
Statistical Data were analysed using Microstat,
analysis
an Ecosoft computer program. Results
Characterization
of the 833 mAb
More than four hundred hybrids, from two fusions, were screened and of these six were identified as producers of putative anti-Id antibodies. One of these, 8B3, an IgG2bk, reacted in the ELISA with affinity-purified AHA from patient M.J. (typically 0.8-l .5 OD), but not with IgG from pooled normal human sera (0.0-O. 1 OD). Reactivity of purified 8B3 mAb with AHA could not be inhibited by increasing amounts of whole serum or control IgG (Figure 1). As shown, a low but significant (P=O.O2; hypothesis test for means), inhibition was observed with the highest concentration of IgG from M. J. These IgG once enriched in AHA were clearly inhibitory in the ELISA (Figure 1). These results indicated that 8B3 did not react with IgG constant region determinants but with IgG showing antihistone activity. To support this conclusion, M.J. serum was purified on a 8B3 aflinity column. As shown in Table 1, only IgG showing antihistone activity could be eluted from the column. Other activities including antihistone activity, also present in M. J. serum, were all recovered in the column effluent (Table 1). As AHA were present in both 8B3 column eluent (8B3+ AHA) and effluent (8B3- AHA), antihistone inhibition assays were carried out using purified 8B3 mAb as inhibitor. Figure 2 shows that antihistone activity due to 8B3+ AHA could be fully inhibited by 8B3 mAb, while that due to 8B3- AHA was unaffected. These results suggested that 8B3 mAb was reacting
368 Jest L. Subiza et al. Table 1. Aj?inity-purified
8B3+ antibodiesfrompatient
M.J. binds to histones*
Effluents from Wells coated with
Eluents from
8B3-column
Control-column
8B3-column
>2.5 1.87fO.2 >2.5 >2.5 0.84 + 0.0
> 2.5 1.66kO.l > 2.5 > 2.5 0.73kO.l
1.42k0.1 0.31 fO.O 0.62fO.l 0.04+ 0.0 0.02 + 0.1
Histones DNAds DNAss
C. albicans antigen Cardiolipin
Control-column 0.33kO.O 0.25kO.O 0.73kO.O 0.06 f 0.0 0.06 f 0.0
*Serum from patient M.J. was passed onto an 8B3-column or a control column (unrelated mAb). Effluents and eluents from both columns were tested for IgG reactivity by ELISA, preserving the same linal volumes. Results are expressed in OD at 492 mn (mean k SD of duplicates). Differences between the columns were all NS except for the antihistone activity detected in the column eluents (F2.5
PI
< 0.001
