In vitro investigation of factors influencing IgE synthesis.

Cytokines play an important role in the induction or inhibition of synthesis of the particular isotypes of immunoglobulin after antigenic stimulation. Studies of these effects in vivo require substantial amounts of reagents that are difficult and time-consuming to prepare. The present paper describes methods for immunization and in vitro culture that permit the investigation of effects of cytokines and anti-cytokines on a secondary, antigen-specific IgE response, using much lower quantities of materials than are required in vivo. The results are compared with those reported from in vivo studies.

Prolongation of the responsiveness of newborn mice to syngeneic IgE by inhibition of IgE synthesis.

Between the ages of 2 to approximately 11 days mice respond to a challenge with syngeneic IgE by producing anti-IgE antibodies; by the age of 2 weeks they are unresponsive. Even adult mice, however, produce high titers of anti-IgE antibodies when immunized with a conjugate of syngeneic IgE and a foreign antigen such as keyhole limpet hemocyanin (KLH), indicating that adult tolerance to unconjugated IgE resides in the T-cell compartment. The loss of responsiveness in 2-week-old mice follows closely after the first appearance of IgE-secreting cells and detectable serum IgE. This suggests that the delayed onset of tolerance is attributable to the delay in synthesis of IgE. Data presented here provide support for this hypothesis. A further delay in the initial synthesis of IgE, induced by neonatal administration of anti-IgM antibodies, caused a corresponding extension of the period after birth during which mice remain responsive to unconjugated IgE.

Efficient production of syngeneic anti-IgE monoclonal antibodies with high affinity and diverse specificity.

Syngeneic monoclonal anti-IgE antibodies are of value in studies of the suppression of IgE synthesis. Procedures are described here for the production of high titers of murine anti-IgE antibodies by initiating immunization in the perinatal period, before mice develop tolerance to their autologous IgE. This in turn facilitates the production of monoclonal anti-IgE antibodies. Properties of some of these mAbs are reported, including affinity, fine specificity and ability to bind to IgE on B lymphoma cells or mast cells.

On the nomenclature for V-region serological markers.

A recent article in Immunology Today raised significant questions concerning the appropriate use of the terms 'idiotype' and 'V-region isotype'. An alternative approach to the usage of these terms, which emphasizes their functional aspects, is presented here by Alfred Nisonoff.

Role of antibody and T cells in the long-term inhibition of IgE synthesis.

We have shown that the long-term inhibition of IgE synthesis associated with perinatal inoculation of syngeneic IgE is accompanied by the synthesis of autoantibodies to IgE. Synthesis of IgE can also be inhibited by passive transfer of syngeneic anti-IgE antibodies. In the present investigation we made use of adoptive transfer experiments to assess the relative roles of antibodies and T cells in the inhibitory process. It was found that spleen cells from IgE-suppressed mice (synthesizing anti-IgE antibodies) could adoptively transfer the state of inhibition to syngeneic adult mice. The inhibition occurred only under conditions in which the recipient mice synthesized anti-IgE antibodies. Separated B cells, CD4+ T cells, CD8+ T cells, or a mixture of B and CD8+ T cells were ineffective. However, strong inhibition of IgE synthesis (as indicated by serum levels and numbers of IgE-secreting cells in the spleen) was observed after transfer of a mixture of B cells and CD4+ (helper) T cells. The results indicate that in this experimental model anti-IgE antibodies are the suppressive agent and that T cells do not play a role other than that of providing help to B cells for anti-IgE synthesis.

Effects of syngeneic anti-IgE antibodies on the development of IgE memory and on the secondary IgE response.

The prolonged inhibition of IgE synthesis in mice caused by perinatal inoculation of IgE is attributable, at least in part, to the formation of anti-IgE antibodies. The induction of unresponsiveness with respect to IgE synthesis requires that IgE be administered during a brief interval (2 to approximately 10) days after birth, that corresponds with the time period during which anti-IgE antibodies are induced. Passive administration of syngeneic anti-IgE also inhibits IgE synthesis. We have now investigated the effect of anti-IgE on the induction of memory for IgE production and on secondary IgE responses. Syngeneic anti-IgE antibodies were found to inhibit secondary IgE responses directly during immunization or after adoptive transfer of primed cells. Anti-IgE did not, however, prevent the induction of memory cells for IgE synthesis or cause the loss of memory for IgE synthesis. Inhibition of a secondary IgE response was found to require the presence of anti-IgE and was lost when anti-IgE antibodies were cleared from the mouse. After the transfer of primed cells and secondary challenge anti-IgE was inhibitory only when given during the first 3 to 5 days, after which the primed cells became resistant to inhibition. The failure of anti-IgE to prevent the induction of IgE memory cells is discussed in terms of class switches that occur during the transition from IgM to IgE production.

An analysis of idiotype expression in a high-affinity, somatically mutated variant of a germline-encoded anti-p-azobenzenearsonate antibody.

Using two polyclonal (rabbit) and two monoclonal anti-idiotype (anti-Id) reagents, we investigated structural correlates of the Id of mAb 36-71, a somatically mutated member of the CRIA Id family that has an exceptionally high affinity for the p-azobenzenearsonate (Ars) hapten. The two monoclonal anti-Ids reacted principally with the L chain of 36-71. The polyclonal anti-Ids interacted with both the H and L chain. The amino acid sequences of the VH and VL regions of 36-71 differ in eight and 11 positions respectively from those of the anti-Ars mAb 36-65, an unmutated prototype of the CRIA family. In the presence of 36-71L only three substitutions in 36-65 VH, introduced by mutagenesis, sufficed to restore full expression of the 36-71 Id. The same three substitutions had previously been shown to increase the affinity of 36-65 by a factor of 200, to a level equivalent to that of 36-71. X-ray crystallography had indicated that two of these substitutions introduce conformational changes consistent with the increase in affinity. We propose that these conformational changes may also account for the critical role of the three amino acids in Id expression. We also found that 36-65 is a very poor inhibitor of the interaction of 36-71 with its polyclonal anti-Ids, despite identity of the hapten-contacting residues in the two mAbs and evidence (from hapten inhibition) that the hapten-binding region is part of an important Id. Again, a difference in conformation at the binding site of the two mAbs could account for these observations.

Three-dimensional structure of two crystal forms of FabR19.9 from a monoclonal anti-arsonate antibody.

The three-dimensional structure of FabR19.9 from a well-characterized anti-p-azobenzenearsonate monoclonal antibody has been determined by x-ray diffraction techniques in two crystalline forms (I and II) to a resolution of 2.8 and 2.7 A, respectively. Essentially the same tertiary and quaternary structure of the Fab is observed in the two forms. The major difference resides in the intermolecular contacts, which are interpreted to favor an irreversible transition from the metastable form I to the more stable form II. The third complementarity-determining region of the heavy chain (H3) folds back over the combining site and requires rearrangement for hapten binding. This dynamic requirement on H3 is consistent with its mobility in the structure and can explain hapten binding to an otherwise inaccessible antibody combining site.

IgE-secreting cells in the thymus: correlation with induction of tolerance to IgE.

We have shown previously that normal mice become tolerant to endogenous IgE when they are approximately 2 weeks old and that this corresponds closely with the initial appearance of IgE in serum. Tolerance evidently is restricted to T cells, since B cells responsive to IgE are present in neonatal and adult mice. The present report shows that IgE-secreting cells can be detected in the thymus between days 7 and 11 after birth and that the onset of tolerance to IgE occurs at the age of 11 days. Similar results were obtained in A/J and (BALB/c x A/J)F1 mice. This suggests that tolerance is induced in the thymus, probably by cells bearing peptide fragments of IgE. The order of appearance of IgE-secreting cells is thymus, spleen, and mesenteric lymph nodes.

Mechanisms of induction of tolerance to syngeneic IgE.

We have previously shown that T cell tolerance to syngeneic IgE in mice is delayed after birth for 2-3 weeks and is induced concurrently with the appearance of IgE in serum. Although inoculation of IgE in saline on the day of birth caused T cell tolerance, an apparent inconsistency was our inability to tolerize mice after day 2, despite evidence indicating that endogenously-produced IgE is effective at the age of 2-3 weeks. Data presented here indicate that repeated i.p. inoculations, during this period, of smaller doses of IgE in saline induce tolerance in many individual mice, thus mimicking to a degree the naturally occurring events. Another earlier finding was that normal adult B cells, but not T cells, can tolerize T cells to IgE when administered on the day of birth. We have now observed that B cells bearing surface IgE are much more tolerogenic than the remaining B cells.

Proteolytic digestion of mouse IgE.

Conditions are described for the preparation of F(ab')2 and Fab fragments of mouse IgE. Papain, pepsin or trypsin each produced F(ab')2 fragments with Mr approximately equal to 130,000 which yielded Fab fragments on further digestion. The release of Fab fragments from F(ab')2 resulted from further cleavage of the H chain. Pepsin, and especially trypsin appear more suitable for the preparation of F(ab')2 because of the difficulty of separating a 93 kDa by-product from the F(ab')2 produced by papain. The best yields of purified Fab were obtained with papain. Rates of digestion were in the order, pepsin approximately equal to trypsin much greater than papain.

B cell abnormalities induced by a mu Ig transgene extend to L chain isotype usage.

We have analyzed the phenotype of B cell populations from mice transgenic for a rearranged Ig mu H chain gene. We find a decrease in the number of B cells in the spleens of these mice. Transgenic B cells have decreased surface levels of both IgM and IgD. The circulating IgM in these mice is 3- to 10-fold enriched in lambda L chains, compared with that in non-transgenic mice. Analysis of IgM-producing hybridomas, from transgenic mice that express the transgene at high levels, demonstrates that this higher lambda frequency is observed in transgene-nonexpressing as well as transgene-expressing hybridomas. A partial loss of L chain isotype exclusion is also noted in these hybridomas, and a significant proportion of primary B cells expressing both kappa and lambda L chains on their surface can be demonstrated. These findings suggest an ability of the transgenic Ig H chain to affect events in B cell ontogeny beyond the H chain locus. Our results support a quantitative model of exclusion for both the H chain alleles and the L chain isotypes.

Induction of tolerance to syngeneic IgE in neonatal mice.

We have previously shown that adult A/J mice are tolerant to syngeneic IgE at the level of T cells, but not B cells. T cells of mice are responsive until the age of 2 to 3 wk, which correlates with the time of appearance of serum IgE. Tolerance can be induced earlier by neonatal administration of IgE in saline. We report here that purified nonimmune adult B cells, but not T cells, can transfer the state of tolerance to neonatal mice. As few as 2 x 10(6) B cells are effective. If IgE-bearing or IgE-secreting cells prove to be responsible, the amount of cell-bound IgE that can induce tolerance must be very small. The results also indicate that suppressor T cells do not have a major role in maintenance of self-tolerance to IgE.

Isotype switching of an immunoglobulin heavy chain transgene occurs by DNA recombination between different chromosomes.

Transgenic mice carrying an immunoglobulin mu heavy chain transgene exhibit isotype switching of the transgene. We have now characterized the mechanism of transgene switching in these mice. The site of mu transgene insertion in one transgenic line has been localized to chromosome 5 using a series of polymorphic endogenous retroviruses as genetic markers in backcross mice. The endogenous immunoglobulin heavy chain locus resides on mouse chromosome 12, which shows that transgene isotype switching can occur between two different chromosomes even though normal antibody gene switching has generally been thought to occur within one chromosome. We find that transgene isotype switching involves interchromosomal DNA recombination, and our data suggest that the same enzymatic mechanisms mediate both normal isotype switch recombination and interchromosomal transgene switching. Our findings also support the notion that the isotype switching mechanism can induce chromosomal translocations such as observed for the c-myc gene in some B cell tumors.

Inhibition of IgE synthesis by anti-IgE: role in long-term inhibition of IgE synthesis by neonatally administered soluble IgE.

Inoculation of syngeneic IgE into 2- to 12-day-old mice results in prolonged synthesis of anti-IgE antibodies without further challenge. These anti-IgE antibodies may be largely responsible for the long-term inhibition of synthesis of IgE that is known to result from a perinatal challenge with IgE. This conclusion is supported by the effect of passive inoculation of syngeneic polyclonal anti-IgE antibodies into young mice, which similarly results in selective inhibition of IgE synthesis. Further evidence is the close relationship between the age dependency of IgE-induced inhibition of subsequent IgE synthesis and the ability of IgE to induce anti-IgE antibodies. IgE synthesis was monitored at the level of secretion by B cells as well as serum IgE levels and IgE antibody responses.

Structure of idiotopes associated with antiphenylarsonate antibodies expressing an intrastrain crossreactive idiotype.

We have explored the structural basis of idiotopes associated with the major idiotype (CRIA) of A/J anti-p-azobenzenearsonate antibodies, with emphasis on the regions of contact with anti-idiotypic antibody. The analysis was facilitated by a recent description of the three-demensional structure of the Fab portion of a CRIA-related antibody molecule. Direct binding measurements failed to reveal idiotopes associated exclusively with the L chain. However, the L chain participated in the formation of approximately 80% of the idiotopes recognized by polyclonal anti-Id. This indicates that multiple complementarity-determining regions (CDRs) participate in the formation of idiotopes. The affinity of anti-Id for CDRs on L chains must be appreciable but insufficient to permit direct binding (i.e., less than approximately 10(4) M-1). Approximately 20-35% of polyclonal anti-Id reacted with high affinity with H chains recombined with non-CRIA-related L chains. This interaction was found to involve the D region as well as one or both CDRs in the VH segment, again indicating the contribution of multiple CDRs. It is suggested that a typical idiotope may be similar in size to that of protein epitopes whose three-dimensional structures are known; such epitopes comprise a substantial fraction of the surface area occupied by the CDRs of an antibody. The expression of an idiotope recognized by the mAb AD8, which interacts with the VH segment, was found to be unaffected by major changes in the neighboring D and VL regions. This observation is relevant to efforts to predict three-dimensional structure from the amino acid sequence of CRIA+ molecules.

Quantitative analysis of idiotypic mimicry and allelic exclusion in mice with a mu Ig transgene.

Analysis of C57BL/6 mice (IgM allotype, Igh-6b or mu b) that carry an Ig H chain transgene of a different allotype (mu a) shows that IgM molecules of mixed allotype (mu a mu b) are present among serum antibodies. The finding was extended to hybridomas prepared from nonimmune transgenic mice, many of which also failed to exhibit allelic exclusion. The proportions of mu a and mu b secreted by individual hybridomas varied markedly, and the product of an individual hybridoma was found to be heterogeneous with respect to the allotype content of individual molecules. The ratio of mu a:mu b chains secreted by individual hybridomas was found to correlate with the number of transgene copies remaining in each hybridoma, and several hybridomas that secrete only mu b-positive molecules had apparently lost all but one copy of the transgene. An idiotype characteristic of the transgene was found to be present only in association with the transgenic (mu a) allotype, and indirect evidence strongly suggests that the idiotype was present only on mu a polypeptide chains. Thus, there is no evidence in this system for the induction of idiotypically cross-reactive endogenous molecules.

Isotype switching by a microinjected mu immunoglobulin heavy chain gene in transgenic mice.

Immunization of transgenic mice carrying an immunoglobulin mu heavy chain resulted in a response dominated by expression of the transgene variable region. Unexpectedly, in a large proportion of the antibody produced by immunized mice, the transgene variable region was associated with IgG rather than IgM. This demonstrates that the transgene can undergo an isotype switch. Four transgenic founder lines all exhibited transgene isotype switching despite the likelihood of random chromosomal integration of the transgene. In addition one of the lines was analyzed by breeding studies and the transgene was found to be genetically unlinked to the immunoglobulin heavy chain (Igh) locus. These results indicate that a precise chromosomal location is not required for isotype switching and suggest the possibility that the isotype switching process can occur interchromosomally.

Three-dimensional structure of Fab R19.9, a monoclonal murine antibody specific for the p-azobenzenearsonate group.

The crystal structure of Fab R19.9, derived from an anti-p-azobenzenearsonate monoclonal antibody, has been determined and refined to 2.8-A resolution by x-ray crystallographic techniques. Monoclonal antibody R19.9 (IgG2b kappa) shares some idiotopes with a major idiotype (CRIA) associated with A/J anti-p-azobenzenearsonate antibodies. The amino acid sequences of the variable (V) parts of the heavy (VH) and light (VL) polypeptide chains of monoclonal antibody R19.9 were determined through nucleotide sequencing of their mRNAs. The VL region is very similar to that of CRIA-positive anti-p-azobenzenearsonate antibodies as is VH, except for its third complementarity-determining region, which is three amino acids longer; it makes a loop, unique to R19.9, that protrudes into the solvent. A large number of tyrosine residues in the complementarity-determining region of VH and VL, with their side chains pointing towards the solvent, may have an important function in antigen binding.