The use of protein A in solid-phase binding assays: a comparison of four radioiodination techniques.

Preparations of protein A radioiodinated by 4 different methods have been compared in indirect radioimmunoassays. The oxidative methods (chloramine-T and iodogen) for direct iodination of tyrosyl and histidyl residues were applied with high efficiency and gave a suitable product, provided the substitution ratio was kept low (1 iodine atom/molecule of protein A). Higher levels of modification tended to perturb the Fc-binding characteristics of the protein, especially with the use of iodogen. Introduction of the isotope via substitution of lysyl residues (Bolton-Hunter and Wood reagents) was also examined. The Bolton-Hunter modification of protein A gave an unsuitably low labeling efficiency; in contrast, the Wood reagent gave efficiencies approaching 50%. Protein A could be extensively substituted with the latter reagent (greater than 5 diiodinated benzimidate molecules per protein molecule). Thus, the use of the Wood-labeled protein A could raise the sensitivity of the binding assay at least an order of magnitude compared to using protein A iodinated by the oxidative methods. The effects on the biological activity of protein A exerted by the different labeling procedures are rationalized on the basis of the amino acid composition and tertiary structure of the protein.

Comparison of methyl-para-hydroxybenzimidate and 1,3,5-trichlorotriazine in producing sensitive target cells for use in the hemolytic spot assay.

Two bifunctional reagents were shown to be useful in the coupling of immunogens to the surface of either nucleated or non-nucleated cells. The heterobifunctional reagent methyl- para -hydroxybenzimidate was used to couple aromatic amino-haptens to the surface of SRBC which resulted in a stable and sensitive target cell capable of detecting as little as 20 pg of purified anti-hapten antibody in the hemolytic spot assay. The multifunctional reagent 1, 3, 5-trichlorotriazine yielded similar results when coupling amino-haptens to the surface of SRBC for use in the hemolytic spot assay. This reagent was also used to couple protein to the surface of SRBC which were able to detect as low as 1 ng of purified anti-protein antibody in the hemolytic spot assay. The sensitized SRBC produced using either of these coupling reagents were shown to remain stable for several months giving reproducible results from one test to another. Lastly, 1, 3, 5-trichlorotriazine was used to couple the hapten 4-aminophthalate to the surface of nucleated cells with retention of greater than 90% cell viability with continued growth and cellular division in culture.

Detection of single-stranded DNA damage using monoclonal anti-thymidine antibody.

A method to detect single-stranded DNA damage from individual cells has been developed using a monoclonal anti-thymidine antibody (MoAb20B7). Initially, HL-60 cells were incubated with daunomycin at different concentrations, and processed by MoAb20B7. While 73.5% of the cells incubated with 5 micrograms/ml of daunomycin for 24 h reacted positively with MoAb20B7, 83.5% cells at 10 micrograms/ml daunomycin dose were positive. Next, this method was combined with unscheduled DNA synthesis to simultaneously measure repair and damage from individual cells. Finally, patients with acute myeloid leukemias were studied before and 24 h after therapy with a daunomycin containing regimen. In vivo damage could be determined in a prompt fashion.

Hybridomas producing hemolytic plaques used to study the relationship between monoclonal antibody affinity and the efficiency of plaque inhibition with increasing concentrations of antigen.

Hybridoma clones producing anti-hapten antibodies were established by fusing spleen cells from mice immunized with 4-azophthalate keyhole limpet hemocyanin to drug-resistant non-producing myelomas. The affinity and homogeneity of the immunochemically purified anti-phthalate antibodies were determined by equilibrium dialysis. We report here a broad distribution of the monoclonal antibody affinities ranging from a low of 4.0 X 10(4) to a high of 4.0 X 10(7) (L/M). The binding data for eleven anti-phthalate antibodies described a straight line in a Scatchard plot as would be expected for homogeneous antibodies. We determined that all hybridomas, except those secreting very low affinity antibody, produced hemolytic plaques. The hybridomas made it possible to test several assumptions regarding the association of plaque morphology and plaque inhibition with the affinity of the antibody secreted by the plaque-forming cells. Our studies indicate that the affinity of the antibody secreted by the hybridoma clones does not correlate with either the size of the plaque or with the efficiency with which the hybridoma-produced plaques are inhibited by free hapten. By comparing hybridoma-produced plaques to plaques produced by phthalate-immune spleen cells it has been demonstrated that the plaque size within each hybridoma clone was substantially less heterogeneous than that observed for the immune spleen cells. Our results do support the assumption that the range of free hapten concentration over which PFC are inhibited is a reflection of PFC heterogeneity. The analysis of the PFC inhibition of hybridomas produced an interesting and unexpected result which is reported here. It was observed that subinhibitory doses of free hapten caused an enhancement in the number of hybridoma-produced plaques. The relevance of this finding to the recent observation and interpretation of plaque enhancement (i.e., the displacement of anti-idiotype antibodies) observed for immune spleen cells is discussed.

Mechanisms of B cell tolerance. I. Tolerance to dextran B1355 induced with the oxidized dextran.

The bacterial dextran B1355, which is normally a potent thymus-independent immunogen, was made tolerogenic by oxidation. The injection of the oxidized dextran into BALB/c mice before, at the same time, or up to 4 days after the injection of the immunogenic form of the dextran resulted in a marked immunologically specific suppression of the number of anti-dextran antibody-forming cells found in the spleen. This suppression resulted from a direct inactivation of antibody-forming cell precursors rather than from either inhibition of antibody secretion or the exhaustive utilization of precursor B cells that have been observed in other tolerance systems. A substantial degree of tolerance was achieved after only a 1-hr in vivo exposure of the spleen cells to the tolerogen. At a dose of 1 mg of oxidized dextran per mouse, tolerance persised for at least 3 weeks. A complete recovery was apparent by 10 weeks. The stability of the tolerance was demonstrated by transferring tolerant spleen cells to irradiated recipients. The response in the recipient animals to an immunogenic dextran challenge remained suppressed. It appears that the tolerogenicity of the oxidized dextran is due to its ability to couple covalently with free amino groups in or near the receptor site of the cell membrane via the reactive dialdehyde groups of the dextran.

Failure of PFC inhibition assays to distinguish idiotypically between clonotypes that are readily distinguishable by RIA analysis.

Two different hemolytic plaque assay protocols that are commonly used to quantitate idiotype-positive antibody-secreting cells have been compared to a standard radioimmunoassay (RIA) to test for their ability to discriminate between related, but idiotypically distinct, clonotypes. The idiotype proband used in this analysis is the individual specific idiotype associated with the dextran-binding myeloma protein M104E (M104E IdI). Antibodies specific for this private idiotype (anti-M104E IdI) were purified by a combination of adsorption and affinity chromatography of the immunoglobulin fraction isolated from the sera of rabbits repeatedly immunized with M104E. The same affinity-purified anti-M104E IdI antibodies were used in the hemolytic plaque assays and in the RIA. In one of the plaque assays, the putative idiotype-positive antibody-forming cells were scored by lysis of target erythrocytes to which the anti-idiotype had been covalently coupled. In the other plaque assay, the idiotype-positive cells were determined indirectly by anti-idiotype inhibition of PFC produced on dextran-coupled target erythrocytes. The fidelity of these two assays to quantitate the M104E private idiotype expression in individual BALB/c mice after a single immunization with dextran B-1355S was determined by comparing the plaque assay data to the data generated by a double-antibody, post-precipitation RIA of either the antibodies in the serum or of monoclonal antibodies produced by hybridomas. Our data indicate that both plaque assay protocols reflect an overestimate of the actual expression of M104E private idiotype. By using a library of dextran-specific hybridomas (that have been characterized in an RIA with respect to their M104E IdI cross-reactivity), we have shown that the PFC overestimate of the M104E expression observed in dextran-immune mice is due to the inability of both plaque assay protocols to distinguish between dextran-specific clonotypes that express idiotypes cross-reactive with, but not identical to, the 104E IdI. We conclude that the plaque assay should be used only in conjunction with an RIA to estimate the idiotype expression. This is especially true in situations where closely related cross-reactive idiotype families exist.

Comparison of the homogeneous, primary anti-dextran B1355 antibody raised in BALB/c mice with protein 104E.

The primary antibody response to dextran B1355 in BALB/c mice is largely a homogenous IgM antibody. This antibody appears to be similar if not identical to the myeloma protein, protein 104E, for the following reasons: a) isoelectric focusing of the 7S monomers, separately and together in co-isoelectric focusing, give the same pattern, both in the presence and absence of 4M urea; b) the inhibition of lysis of the dextran-coated SRBC by specifically purified anti-dextran antibody and by protein 104E required essentially the same concentration of dextran B1355. This similarity was further demonstrated by plaque assays with dextran-coated SRBC, in which the formation of plaques was inhibited by free dextran. Inhibition of plaques produced by both types of cells required essentially the same concentration of dextran B1355. On these bases, there appears to be no difference in properties (or in structure) between the myeloma protein and the induced antibody.

Selective elimination of idiotype-binding cells in vivo by a drug-idiotype conjugate demonstrates the functional significance of these cells in immune regulation.

A receptor-specific cytotoxic drug delivery system has been used to eliminate idiotype-binding cells in vivo to ascertain the possible functional significance of these cells in regulating the humoral immune response to dextran. Protein M104E, a mouse myeloma protein that binds dextran, expresses a private idiotope that is present on a significant proportion of the normal dextran-specific antibody repertoire. Immunocompetent cells that bind and internalize M104E idiotype-bearing molecules were eliminated by the intravenous administration of a single dose of cytosine arabinonucleoside conjugated to purified M104E protein. The administration of this cytotoxic drug-idiotype conjugate had a profound effect upon the expression of the M104E idiotype in euthymic but not in athymic BALB/c mice following immunization with dextran. In euthymic mice, the depletion of the idiotype-binding cells resulted in a marked elevation in the level of M104E idiotype present in the immune sera. Moreover, treated but not control mice developed idiotype-positive molecules that did not bind dextran. These results demonstrate the functional significance of idiotype-binding cells in the regulation of individual clonotypes during an immune response.

Nucleotide sequence of messenger RNA encoding VHDJH and VKJK of a highly conserved idiotype-defined primary response anti-hapten antibody.

The primary humoral immune response of mice to the hapten phthalate (Xmp) is focused upon two adjacent immunodominant negatively charged carboxyl groups on a benzene ring that are in positions meta and para to the azolinkage (i.e., Xmp) to the protein carrier keyhole limpet hemocyanin. A significant fraction of the anti-Xmp antibodies raised in several different inbred mouse strains (BALB/c, DBA/2, A/HeHa; C3H, and SM/J), and many wild mouse populations express a cross-reactive Id, CRIXmp-1. This CRIXmp-1 is conspicuously absent in C57BL/6 mice. In order to obtain a better understanding of the events and parameters that influence the selection and regulation of the primary response B cell repertoire, and to explore the structural basis of Ag binding, we have determined the nucleotide sequence of the entire V region gene complexes, which encode the H and L chains of these highly conserved and dominant CRIXmp-1+ antibodies. Our data establish that the H chain gene complex consists of a single VH germ-line gene that is identical to VH Oxazolone-1, encoding the H chain of another highly conserved and dominant cross-reactive Id family associated with the primary response to Oxazolone. In CRIXmp-1+ Xmp-specific hybridomas this gene is joined to a limited set of D region sequences that express a conserved amino acid motif-GLR. At least three of the five D regions examined are coded for by DFL16.2. This VHD complex can be utilized with one of three different JH region genes (JH1, JH2, and JH4) without any significant effect upon antibody fine specificity or Id. In spite of this lack of JH fidelity all of the CRIXmp-1+ hybridomas have precisely maintained the same length in the H chain CDR3 and FRW4 by altering either the length of the D segment or the length of JH. Nucleotide sequence analysis of the VL gene complex of CRIXmp-1+ anti-Xmp antibodies indicates that the L chain V region is also encoded by a single germ-line gene. The amino acid sequence predicted from the nucleotide sequence of the VKJK from Xmp-specific CRIXmp-1+ hybridomas is identical to the sequence of the anti-arsonate antibody 1210.7, which is the prototype of another Id family (CRI) that is conserved and dominant in BALB/c mice.

Antigen-specific drug-targeting used to manipulate an immune response in vivo.

The administration of dextran-conjugated cytosine arabinonucleoside (araC) to BALB/c mice at various times prior to but not subsequent to immunization with native dextran renders mice unresponsive to this thymic-independent antigen. These results demonstrate that the primary immune response to an antigen can be selectively and efficiently suppressed or eliminated in vivo by the delivery of a single dose of an appropriate antigen-cytotoxic drug conjugate. Evidence presented here indicates that the dextran-araC conjugate (toxogen) acts directly and selectively upon unprimed dextran-specific antibody-forming cell precursors, presumably by binding to their receptors and subsequent internalization of the resultant receptor-toxogen complexes. The resistance of antigen-primed mice to the cytotoxic effect of the toxogen could result from the failure of dextran-primed cells to reexpress antigen-specific receptors, from an alternative processing of the toxogen, or from the inability of the antigen-primed cells to internalize a second round of receptor-ligand complexes. We also determined that B cells responding to thymic-dependent antigens were not affected by the prior exposure to a toxogen. The inability to eliminate or suppress the primary response to a thymic-dependent antigen via the administration of a cytotoxic drug-antigen conjugate distinguishes the thymic-independent set of B cells from the thymic-dependent B-cell repertoire. The difference between these two B-cell compartments could be due either to differences in the amount of ligand bound to receptors or to differences in the trafficking patterns of receptor-ligand complexes within each cell type.

Antibody response of BALB/c mice to dextran B1355S: alterations in the expression of an idiotype associated with the depletion of idiotype-binding cells.

In this study, we established that BALB/c mice recognize and respond to the idiotype (M104E IdI) of a major dextran-specific clonotype within the BALB/c mouse repertoire. This idiotype recognition is established by demonstrating the presence of idiotype-binding cells and by the production of antibodies specific for the private M104E idiotype. To determine whether or not the idiotype-recognizing cells play a regulatory role during an immune response to dextran, the idiotype-binding cells were selectively removed either by panning or by radiation-induced killing. Two significant effects are observed when the depleted spleen cells are immunized with dextran. First, there is a substantial increase in the proportion of anti-dextran antibodies that are M104E IdI+. The second effect of the idiotype-specific cell depletion is the production of significant amounts of M104E IdI+ immunoglobulin molecules which do not bind dextran. The depletion experiments produced no alteration in the concentration of anti-dextran antibodies found in the serum or in the number of dextran-specific PFC in the spleen. The data indicate that idiotype-reactive cells can play a role in regulating the level of individual clonotype expression (i.e., the M104E clonotype), but that an alternative mechanism must exist for regulating the absolute amount of anti-dextran antibody produced after immunization.

Hapten-specific B cell repertoire probed by hybridoma technology: selection and characterization of representative clonotypes from the antibody-forming cell pool.

In the course of this study, more than 3000 phthalate-specific antibody-forming cell hybrids were identified using the hybridoma technology. With the aid of a rapid screening assay and an extensive library of phthalate analogs, it was possible to assign selected hapten-specific clones to one of 11 distinct fine-specificity sets. This compartmentalization of the phthalate-specific hybridomas has made it possible to focus attention upon a single manageable portion of the phthalate-specific repertoire. Fourteen clones from a single fine-specificity set were selected for further immunochemical characterization. Five of these clones were found to secrete an antibody that was indistinguishable in isoelectric focusing. Affinity-purified, high-resolution anti-idiotype antibodies were prepared with specificity for the antibodies produced by one of these clones (i.e., 4C7). A major portion of the serologically defined private idiotype (4C7 IdI) was shown to be associated with the ligand-combining site. Our results indicate that the five clones that share a common spectrotype also express the 4C7 IdI. Two other independently derived clones from two distinct fusions also share this idiotype. The 4C7 IdI was also identified in affinity-purified anti-phthalate antibodies derived from a pool of phthalate-immune serum (conventional antibody) and from affinity-purified antibodies derived from a pool of serum from unimmunized BALB/c mice (natural antibody). The 4C7 IdI is thus considered to represent a repeating clonotype in the phthalate-specific repertoire of BALB/c mice, and will serve as one of several useful clonal markers that are being developed for studies of the mechanism regulating idiotype expression.