Immunoglobulin (Ig) A serves as the first line of humoral defense at all mucosal surfaces and is present in large quantities of blood. In playing its role in humoral immunity, IgA interacts with a variety of effector molecules present both in serum and on the surfaces of immune and inflammatory cells. To study these interactions, we previously established expression of human IgA1 in insect cells using recombinant baculoviruses and showed that the expressed antibody is a structurally and functionally intact polypeptide useful for examining the molecular properties of IgA. Indeed, since the C alpha 2 N-linked glycosylation site lies near the Fab-distal pole of C alpha 2, the inability of a mutant IgA1 lacking C alpha 2 N-glycosylation to bind its cognate receptor suggested that the monocyte Fc alpha receptor (mFcalphaR) recognizes IgA at a hinge-distal site encompassing the boundary between the C alpha 2 and C alpha 3 domains. In this report, we utilize both domain-swapped IgA/IgG and point-mutated IgA chimeras to verify the above hypothesis. Using an antigen-specific rosetting assay and a mFc alpha R-expressing cell line, we show that (a) C alpha 2 and C alpha 3 together are necessary and sufficient for binding; (b) neither the IgA hinge nor the tailpiece is necessary for binding; (c) mutations away from the interdomain boundary do not affect binding; and (d) mutations located near the three-dimensional boundary between C alpha 2 and C alpha 3 completely disrupt binding. Taken together, these results localize the mFc alpha R recognition site on IgA to the boundary region between the second and third constant domains--a site analogous to that recognized by Staphylococcus aureus protein A on IgG. The use of this hinge-distal site is, to date, unique among Fc receptors of the Ig superfamily.

Using a series of phenotypic markers that include immunoglobulin (Ig)D, IgM, IgG, CD23, CD44, Bcl-2, CD38, CD10, CD77, and Ki67, human tonsillar B cells were separated into five fractions representing different stages of B cell differentiation that included sIgD+ (Bm1 and Bm2), germinal center (Bm3 and Bm4), and memory (Bm5) B cells. To establish whether the initiation of somatic mutation correlated with this phenotypic characterization, we performed polymerase chain reaction and subsequent sequence analysis of the Ig heavy chain variable region genes from each of the B cell subsets. We studied the genes from the smallest VH families (VH4, VH5, and VH6) in order to facilitate the mutational analysis. In agreement with previous reports, we found that the somatic mutation machinery is activated only after B cells reach the germinal center and become centroblasts (Bm3). Whereas 47 independently rearranged IgM transcripts from the Bm1 and Bm2 subsets were nearly germline encoded, 57 Bm3-, and Bm4-, and Bm5-derived IgM transcripts had accumulated an average of 5.7 point mutations within the VH gene segment. gamma transcripts corresponding to the same VH gene families were isolated from subsets Bm3, Bm4, and Bm5, and had accumulated an average of 9.5 somatic mutations. We conclude that the molecular events underlying the process of somatic mutation takes place during the transition from IgD+, CD23+ B cells (Bm2) to the IgD-, CD23-, germinal center centroblast (Bm3). Furthermore, the analysis of Ig variable region transcripts from the different subpopulations confirms that the pathway of B cell differentiation from virgin B cell throughout the germinal center up to the memory compartment can be traced with phenotypic markers. The availability of these subpopulations should permit the identification of the functional molecules relevant to each stage of B cell differentiation.

The anti-idiotypic (anti-Id) antibody (Ab) 9G4 binds a cross-reactive idiotope (CRI) present in a select group of human autoantibodies. This Id has been localized to the portion of immunoglobulin (Ig) heavy (H) chains encoded by the VH4-21 gene segment, a member of the human VH4 family. This gene segment is utilized by essentially all cold agglutinin (CA) Abs with I/i specificity isolated from patients with CA disease stemming from chronic lymphoproliferative disorders. In this study, mutational analysis of a CA has been used to determine the structural basis for 9G4 binding to Abs utilizing the VH4-21 gene segment. Recombinant CA H chain mutants were produced and their 9G4 reactivity determined. Mutants were generated by exchanging VH4-21 sequences in the FR1, CDR1, and CDR2 with corresponding sequences from a closely related gene segment V71-2, a VH4 family member that is associated neither with Abs having CA activity nor with Abs that react with 9G4. The results indicate that the motif AVY at amino acid positions 23-25 in FR1 defines the 9G4 idiotope. Reaction of these recombinant Abs with a polyclonal rabbit anti-CA antiserum absorbed to render it specific for a CA CRI also maps predominantly to FR1. These findings indicate that the solvent-exposed FR1 plays an important role in eliciting an immune response to Igs.

IgH rearrangements (VH-D, D-JH) are central to the generation of antibody diversity. The majority of the diversity seen in the third hypervariable region is generated by the D segment and at the joints formed by both junctional and N segment variation during D-JH and VH-D rearrangements. The mechanisms that regulate rearrangement are thought to obey the 12/23 rule, wherein D-D or VH-JH rearrangements are precluded. Here, we present evidence that D-D fusions do in fact occur, either as direct or inverted rearrangements. The fused D segments so generated may be fully capable of proceeding in subsequent D-JH and VH-D rearrangements. The resultant VH-D-D-JH recombinations add another dimension to the potential repertoire of IgH V regions by increasing the level of combinatorial diversity and by providing additional sites for N region variation.

We have found that syngeneic Ab2s in the antiarsonate system are serologically and structurally similar to one another. In contrast, the allogeneic Ab2 response is heterogeneous and derives from a large number of unrelated germline gene segments. The Ab2 response of the BALB/c strain to polyclonal A/J Ars A molecules can probably best be compared with a response to a foreign protein and might have been predicted in a strain that completely lacks the H chain V region gene from which the Ab1 derives. Partial variable region sequences of Ab2s from three other systems in addition to previously reported Ab2 structures indicates that this difference in allogeneic vs. syngeneic Ab2s may be a general phenomena. These data support Jerne's hypothesis of complementary V region genes existing in the germline. However, there is good evidence that these antiidiotypic antibodies are not derived directly from the germline, as somatic processes most likely play an important role in their generation. The D segments of Ab2s in the arsonate system as well as in other systems, are novel in structure and cannot easily be explained by previously described germline D segments. D-D fusion may play a role in the generation of the third hypervariable region in these antibodies.

Evidence derived from the complete amino acid sequences of the variable regions of both the heavy and light chains of two members (BOR and KAS) of the Wa idiotypic family of human rheumatoid factors suggests that not only are the light chains of these molecules derived from possibly one variable region gene segment, but the heavy chain variable regions are all derived from the VHI subgroup of human V region genes. These molecules exhibit a surprising conservation in the size of D region, and all use the JH4 gene element. This restriction in use of VL, VH, D, and JH suggests all of these elements may play a crucial role in either antigen binding and/or expression of the crossreactive idiotype.

Two allelic forms of the T cell antigen receptor alpha chain gene were discerned by restriction fragment length polymorphism (RFLP) employing the T cell antigen receptor alpha chain probe pGA5, and the restriction enzyme Bgl II. Analysis revealed that the polymorphic fragments are detected by a probe specific for the constant region exon of the T cell antigen receptor alpha chain gene. Furthermore, the polymorphic fragments were shown to segregate within families. The two allelic forms yield two homozygous states, 3.2/3.2 and 2.9/2.9, at a frequency of 76.5 and 2.9%, respectively, within the normal population. The heterozygous state was observed in 20.6% of the population. The discovery of allelic forms of both the alpha and beta chains of the T cell antigen receptor genes may provide a unique opportunity to study heritable markers of T cell function in several human diseases.

The primary structure of A/J anti-p-azophenylarsonate (anti-Ars) antibodies expressing the major A-strain cross-reactive idiotype (CRIA) has provided important insights into issues of antibody diversity and the molecular basis of idiotypy in this important model system. Until recently, this idiotype was thought to be rarely, if ever, expressed in BALB/c mice. Indeed, it has been reported that BALB/c mice lack the heavy chain variable segment (VH) gene that is utilized by the entire family of anti-Ars antibodies expressing the A/J CRI. Recently, however, it has been possible to elicit CRIA+, Ars binding antibodies in the BALB/c strain by immunizing first with anti-CRI and then with antigen. Such BALB/c, CRIA+ anti-Ars antibodies can be induced occasionally with antigen alone. VH region amino acid sequences are described for two CRIA+ hybridoma products derived from BALB/c mice. While remarkably similar to each other, their VH segments (1-98) differ from the VH segments of A/J CRIA+, anti-Ars antibodies in over 40 positions. Rather than the usual JH2 gene segment used by most A/J CRIA+ anti-Ars antibodies, one BALB/c CRIA+ hybridoma utilizes a JH1 gene segment, while the other uses a JH4. However, the D segments of both of the BALB/c antibodies are remarkably homologous to the D segments of several A/J CRIA+ antibodies sequenced previously, as are the amino terminal amino acid sequences of their light chains. These data imply that BALB/c mice express the A/J CRIA by producing antibodies with very similar, if not identical, light chain and heavy chain D segments, but in the context of different VH and JH gene segments than their A/J counterparts. The results document that molecules that share serologic specificities can have vastly different primary structures.

The specificities of the monoclonal antibodies I-LR2 and 109d6, which recognize MT2- and MT3-like serologic determinants, respectively, have been confirmed by panel testing. In addition, the relationships of these antibodies to other monoclonal antibodies and alloantisera have been studied by means of cell surface fluorescence, complement-dependent cytotoxicity and immunoprecipitation. Using these monoclonal antibodies, molecules encoded by the HLA-D region have been isolated and characterized by amino acid sequencing and peptide mapping. By these criteria, the major populations of molecules bearing MT2- and MT3-like determinants are indistinguishable from DR molecules.

VH region amino acid sequences are described for five A/J anti-p-azophenylarsonate (anti-Ars) hybridoma antibodies for which the VL region sequences have previously been determined, thus completing the V domain sequences of these molecules. These antibodies all belong to the family designated Ars-A which bears the major anti-arsonate cross-reactive idiotype (CRI) of the A strain mouse. However, they differ in the degree to which they express the CRI in standard competition radioimmunoassays. Although the sequences are closely related, all are different from each other. Replacements are distributed throughout the VH region and occur in positions of the chain encoded by all three gene segments, VH, DH, and JH. It is likely that somatic diversification processes play a dominant role in producing the sequence variability in each of these segments. The number of differences from the sequence encoded by the germline is smallest for antibodies that express the CRI most strongly, suggesting that somatic diversification is responsible for loss of the CRI in members of the Ars-A antibody family. There is an unusual degree of clustering of differences in both CDR2 and CDR3 and many of the substitutions are located in "hot spots" of variation. The large number of differences between the chains prohibits the unambiguous identification of positions at which alterations play a major role in reducing the expression of the CRI. However, the data suggest that the loss of the CRI is associated with a definable repertoire of somatic changes at a restricted number of highly variable sites.

Monoclonal antibody IVD12 was used to isolate and characterize a human Ia molecule present on B cells that generally display DR4 or DR5 phenotypes. The specificity of binding of IVD12 to human peripheral blood B cells from 75 normal individuals and 19 homozygous human lymphoblastoid B cell lines was identical to the supertypic specificity MB3 previously defined. Furthermore, IVD12-reactivity was shown to segregate with HLA in three informative families. In each family, individuals positive for IVD12 binding were also positive for DR4 or DR5. Using IVD12, a molecule has been isolated from the homozygous cell line PRIESS (DR4/4) and has been shown by amino acid sequence analysis to be homologous to the murine I-A and human HLA-DS molecules. These findings suggest that the MB3 specificity is found on a molecule encoded by loci distinct from those loci which encode HLA-DR molecules. This molecule represents the third family of HLA-D region molecules isolated from the cell line PRIESS. Both HLA-DR and HLA-SB molecules from this cell line were previously shown by amino acid sequence analysis to be I-E-like but distinct from one another. Collectively, these data provide evidence that the HLA-D region contains at least six loci encoding distinct alpha and beta chains for the HLA-SB, HLA-DR, and HLA-DS molecules.

A ubiquitous nonimmunoglobulin molecule that binds p-azobenzenearsonate (ABA) has been detected in the cytoplasm of several murine cell lines, including T cell hybridomas as well as in normal liver and spleen. Similar to many recently described antigen-specific T cell factors, this ABA-binding protein has a 62,000 mol wt, and, when analyzed by direct binding, the molecule reacts with several different rabbit anti-idiotypic antisera specific to the ABA system. The presence of this antigen-specific, "idiotype positive" molecule in many different cells indicates that it is not an important immunoregulatory molecule.

T cell hybrids have been constructed between the BW5147 thymoma cell line and A/J splenocytes from mice suppressed with the p-azophenylarsonte hapten. Three independently derived cloned lines have been chracterized. Each secretes or sheds a 62,000-dalton antigen-specific product bound by rabbit antisera directed against the arsonate cross-reactive idiotype. In addition, each of the antien-specific molecules contains determinants encoded within the I region of the murine major histocompatibility complex. Peptide mapping analysis indictes that, whereas these molecules are remarkably similar, each is individually distinct in primary structure. The availability of cloned T cell lines that produce antigen-specific idiotype-positive I region-containing products should facilitate a more thorough dissection of the interrelationships of T-B interctions in the arsonate idiotypic system.

It has been shown that A/J anti-p-azophenylarsonate antibodies that share a major cross-reactive idiotype (CRI) comprise a family of closely related, but nonidentical, molecules. Our results demonstrate that 12 of 14 monoclonal hybridoma products that express the CRI have in common at least one highly conserved idiotypic determinant. It is proposed that this reflects conservation of a portion of the amino acid sequence, presumably in hypervariable regions. That the conserved determinant(s0 are located in the region of the hapten-binding site is indicated by the ability of haptens to inhibit idiotype-anti-idiotype interactions involving the conserved, or public determinants.

Amino terminal amino acid sequence analyses have been performed on the heavy and light chains of induced monoclonal antibodies with specificity for the hapten p-azophenylarsonate. Four of the eight antibodies react with conventional antisera to the previously described A/J anti-arsonate cross-reactive idiotype (CRI). Of the 16 chains analyzed, all but one contain sequence differences in their first framework segment (residues 1-30) that distinguish them from the heavy- and light-chain sequences found in anti-arsonate antibodies isolated from A/J serum or ascites fluid. The presence of such framework differences appears to be independent of whether or not the hybridoma antibodies bear the CRI. In spite of the framework substitutions, all four of the CRI-positive hybridoma antibodies have variable (V)-region frameworks that are very similar to each other and to the CRI-positive molecules found in A/J serum. Two of the four CRI-negative molecules are also structurally similar to the serum antibodies. Two others, however, are strikingly different from any serum anti-arsonate antibody thus far described and appear to reflect a completely separate repertoire of anti-arsonate antibodies in the A/J MOUSE. In addition, serological analyses with an anti-idiotypic antiserum generated against a CRI-positive hybridoma product suggest that each monoclonal antibody may possess individual antigenic specificities different from the determinant(s) detected with the conventional rabbit anti-CRI. The consistent appearance of framework substitutions in what has been thought to be a homogeneous antibody population has important implications for our understanding of the generation of antibody diversity and for the precise chemical definition of an idiotype.

IgA myeloma proteins of kappa- and lambda-types were isolated from two patients. These were used to produce and purify anti-idiotype antibodies of both broad (myeloma-related) and narrow (individual myeloma) specificities. The anti-idiotype antibodies were conjugated with fluorochromes and used as immunofluorescent probes to trace in the patients clonal expansion at different levels of B-cell differentiation. Our results (a) confirm that B lymphocyte precursors in IgA plasma-cell myelomas are involved in the malignant process, (b) show that B lymphocytes of the malignant clone include those expressing each of the major heavy-chain isotypes, mu, delta, gamma, and alpha, and (c) provide strong circumstantial evidence that pre-B-cell members of the malignant clone are also increased in frequency. T cells expressing idiotypic determinants were not detected. These findings argue that the initial oncogenic event may occur in a B-stem cell and is not influenced through stimulation by antigen. An interesting association was the increased frequency of related clones of B lymphocytes as detected by their reactivity with anti-idiotype antibodies of broad specificity. Neither plasma cell nor pre-B-cell members of these related clones were increased in frequency. Anti-idiotype antibodies or helper T cells reactive with myeloma-related idiotypes could be responsible for this phenomenon. We discuss other implications of these findings and speculate that all of the various phenotypes of B-lineage malignancies may result from oncogenic processes affecting stem cell targets.

The E/C alpha- and beta-subunits of intra-I-region recombinants were analyzed for primary structural variation by comparative tryptic peptide mapping. The E/C alpha-polypeptides from B10.A, B10.A (3R) and B10.A (5R) showed complete coincident elution of peptides; the E/C beta-chains from B10.A and 3R (or 5R) were approximately 40% different. This suggests that the structural gene for the E/C beta-polypeptide is within the I-A subregion.

The murine Ss protein has been isolated and purified. Using specific antisera, the radiolabeled protein has a mol wt of 120,000 in sodium dodecyl sulfate polyacrylamide gels. It is composed of two basic subunits of 23,000 and 14,000 daltons. The smaller molecular weight subunit contains a single disulfide bridge, is devoid of carbohydrate, and may represent the murine equivalent of beta2-microglobulin.

A marked homogeneity of the light chains was observed in an analysis of 17 IgM proteins with anti-γ-globulin activity. The V region subgroups of the light chains were determined by both sequence and antigenic analysis. The latter procedure permitted large scale screening for comparisons with control proteins. The two methods showed general agreement in the determination of Kappa III proteins; all proteins positive by antigenic analysis were also positive by sequence but exceptions were noted in the opposite direction. The anti-γ-globulins showed by antigenic analysis a 92% incidence of VK III light chains as compared to an incidence of 27% among 81 control proteins without this activity. A similar selection was observed for an antigen (VK III b) which subdivided the kappa III proteins. The major Wa group of anti-γ-globulins which had been delineated previously on the basis of cross-idiotypic specificity was primarily responsible for the special light-chain selection. All the proteins of this group contain VK III light chains and all were of the VK III b type. Current evidence indicates that additional light-chain similarities were involved in the cross-idiotypic specificity of the Wa group. It thus appears that for the anti-γ-globulins various levels of selection of light chains are manifest ranging from a preponderance of kappa type, of kappa III subgroup, of kappa III b sub-subgroup and perhaps of still further subdivisions to account for the cross-idiotypic specificity.

Through the use of absorbed idiotypic antisera prepared against single isolated monoclonal IgM anti-γ-globulins, partial cross-idiotypic specificity was demonstrated with other IgM anti-γ-globulins. Such antisera classified these proteins into at least three groups. The major group which included 60% of the anti-γ-globulins was particularly homogeneous. The anti-γ-globulin specific antigens were detected best in hemagglutination and hemagglutination inhibition systems. They were not found in monoclonal IgM proteins that lacked anti-γ-globulin activity although related antigens were detected at low concentrations in pooled immunoglobulin preparations as well as in heterogeneous anti-Rh antibodies. Several lines of evidence were obtained indicating that the antibody combining site was involved in the specific determinants. Attempts were made to analyze the fine specificity of each anti-γ-globulin for the Fc fragment of different subclasses of human immunoglobulins as well as those of other species. Differences were observed but these were not readily related to the cross-specificity antigens. The anti-γ-globulin specific antigens were very analogous to those previously described for monoclonal IgM cold agglutinins. Although each protein could be distinguished from all the others on the basis of individual idiotypic antigens, the antigens common to the specific groups of proteins with each of these activities were prominent and readily detected with multiple antisera. The results indicate basic similarities between proteins of a given activity even in unrelated individuals.