Quantitative representation of two germ-line V genes in the early antibody response to 2-phenyloxazolone

Genetic analysis of the maternally induced affinity enhancement in the non-Ox1 idiotypic antibody repertoire of the primary immune response to 2-phenyloxazolone

The early phases of ontogeny are decisive for the development of the B-cell repertoire. Here, we demonstrate that maternal tertiary immunization of BALB/c mice with 2-phenyloxazolone (phOx) caused a drastic alteration of the primary antigen-specific repertoire of the offspring. Maternal tertiary immunization or quaternary antibodies, which exhibited an extremely weak cross-reactivity with the major Ox1 idiotype (IdOx1), induced a change in the proportion of IdOx1/non-IdOx1 antiphOx antibodies in the F1 and F2 primary repertoire. The observed variability in the level of IdOx1 expression (10-90%) exceeded even the seemingly genetically based differences between various mouse strains. In comparison with the non-IdOx1 of control mice, half of the non-IdOx1 antibodies showed a 5-100-fold enhanced affinity. Sixty per cent of these antibodies exhibited an affinity identical to that of IdOx1 antibodies, which are normally of the highest affinity, while the remaining 40% exceeded even that of IdOx1 by a factor of 10. The non-IdOx1 were encoded by VH/VL genes and/or combinations thereof which are either new, hitherto unobserved in the antiphOx response, or typical of memory responses in normal mice. The significance of these data is discussed with respect to the possibility that maternal antibodies, which are acquired through multiple immune maturation processes, may have an epigenetic (non-Mendelian) inheritable potential for the offspring.

Analysis of hapten binding and catalytic determinants in a family of catalytic antibodies

We report here the cloning and kinetic analysis of a family of catalytic antibodies raised against a common transition state (TS) analog hapten, which accelerate a unimolecular oxy-Cope rearrangement. Sequence analysis revealed close homologies among the heavy chains of the catalytically active members of this set of antibodies, which derive mainly from a single germline gene, whereas the light chains can be traced back to several different, but related germline genes. The requirements for hapten binding and catalytic activity were determined by the construction of hybrid antibodies. Characterization of the latter antibodies again indicates a strong conservation of binding site structure among the catalytically active clones. The heavy chain was found to be the determining factor for catalytic efficiency, while the light chain exerted a smaller modulating effect that depended on light chain gene usage and somatic mutations. Within the heavy chain, the catalytic activity of a clone, but not hapten binding affinity, depended on the sequence of the third complementarity determining region (CDR). No correlation between high affinity for the hapten and high rate enhancement was found in the oxy-Cope system, a result that stands in contrast to the expectations from transition state theory. A mechanistic explanation for this observation is provided based on the three-dimensional crystal structure of the most active antibody, AZ-28, in complex with the hapten. This study demonstrates the utility of catalytic antibodies in examining the relationship between binding energy and catalysis in the evolution of biological catalysis, as well as expanding our understanding of the molecular basis of an immune response.

Correlation between immune maturation and idiotypic network recognition

The maturation of T-dependent humoral immune responses is mediated by somatic mutations. Antigen selection is one mechanism for the activation of B cell clones which express antibodies with progressively increased affinity and which are derived as somatic variants from germ-line-encoded genes. However, the emergence of B cell clones secreting rather low-affinity antibodies and the shift to alternative germ-line V region gene combinations during secondary and tertiary responses cannot be explained by antigen selection. It has been considered that idiotypic suppression may favor this clonal shift. Such an involvement would require that idiotypic recognition in the syngeneic host must be highly restricted to private idiotopes of each clone sequentially activated during immune maturation. To test this possibility, we produced 19 syngeneic anti-idiotypic antibodies to the germ-line-encoded major Ox1 idiotype (IgM-IdOx1 H11.5) of the anti-2-phenyl-oxazolone (phOx) immune response in BALB/c mice. The fine specificity of these anti-IdOx1 was tested with a set of anti-phOx monoclonal antibodies, representing the first steps of maturation. About half of the anti-IdOx1 showed almost no reactivity with the IdOx1 after the switch to IgG and none of the anti-IdOx1 reacted with anti-phOx antibodies which carried a glycine or histidine instead of arginine as the middle amino acid of the D region. These observations suggest a strong correlation between immune maturation and the idiotypic network. A model is presented in which idiotypic suppression may function as a driving force for diversification and maturation of the antigen-induced immune response.

Kinetics of somatic mutation in lymph node germinal centres

Somatic mutation activity in immunoglobulin V kappa genes during the response to the hapten 2-phenyl-5-oxazolone was measured in lymph node B-cell populations at various timepoints after footpad immunization. When the V kappa Ox1 genes rearranged to the J kappa 5 segment were amplified from genomic DNA using the polymerase chain reaction and sequenced, somatic mutations could be detected as early as day 4 after immunization. Somatic mutations were also detected after sequencing RNA from oxazolone-specific hybridomas derived from lymph node cells at day 4 after immunization. These early mutations were found mostly in cells with a germinal centre phenotype. No indication of selection at the population level by apoptosis was detected until day 7 after immunization. These results suggest somatic mutations can be induced very early during the immune response in lymph node cells, prior to the peak of clonal expansion and selection with regard to antigen binding.

The same few V genes account for a majority of oxazolone antibodies in most mouse strains

The early primary anti-phenyloxazolone antibodies of 12 mouse strains were studied by determining proportions of two defined subsets id495 (the classical phOx idiotype) and id350. Id495-positive antibodies bear an H chain encoded by VHOx1 gene (family Q52) and an L chain usually coded for by VKOx1 but occasionally by other VK genes. Id350-positive antibodies are encoded by a VK gene VK45.1, and usually by a VH gene of the S107 family. All 12 strains (representing nine H-chain and four kappa-chain haplotypes) produced id350-positive anti-phOx antibodies. While id495 is the predominant major subset in the BALB/c response (originally studied), id350 seems to be the predominant subset of early anti-phOx antibodies in the mouse species. The combined proportion of the two subsets varied from ca. 50 to almost 100% of the total in all strains except C57BL.

Genetic control of early antibody responses

By using a pair of strains that have similar VK haplotypes but different VH haplotypes (e.g. BALB/c and C57BL) it is possible to demonstrate VH-controlled genetic differences in antibodies. By using a pair that have similar VH haplotypes but different VK haplotypes (RF and BALB/c) it is possible to demonstrate VK-controlled genetic differences in antibodies. A plausible explanation for the high frequency of certain V-gene combinations in the primary response is a high affinity of the product without somatic mutations. The products of two such major primary response combinations (VHOx1/VKOX1(H3) and VH186.2/V lambda 1) have an affinity for the immunogen well above 10(6). One combination of V genes, VHOx1/VKOx1(H3) has a major role in the primary anti-phOx response of several mouse strains - the product is idiotype 260. C57BL/10 mice lack the VHOx1 and RF mice the VKOx1(H3) gene. They use the remaining partner of the pair for the response in combination with other genes, but the affinity of the product is lower than the affinity of id. 260. Concordantly, the frequency of these "half-idiotypes" is lower in the primary response than the frequency of the full combination (23% and 16% instead of 50%). When the product of a V-gene combination is very frequent in the primary response, the affinity for the immunogen must be high, but the reverse is not always true. The product of a combination can have an unusually high affinity but the frequency is low. The simplest explanation then is that the frequency of available virgin B cells is low. It can be low because of a low rearrangement frequency of one of the V genes, VH or VK. Another possibility is that only a small proportion of B cells that have the particular combination rearranged can be recruited to the response. We have discussed an example where strict heavy chain CDR3 requirements must strongly limit available B cells.

V genes of the primary antibody response of C57BL/10 mice to the hapten phenyloxazolone

The mRNA of ten monoclonal phenyloxazolone (phOx) antibodies originating from the primary (day 7) response of C57BL/10 mice were partially sequenced. The sequences were analyzed together with those of two previously published antibodies. The C57BL response does not have a predominant subset of antibodies like the BALB/c response has (VH-Ox1/V kappa-Ox1 JK5). Probably, C57BL mice lack the VH-Ox1 gene and, as a consequence, their V kappa-Ox1 gene does not have a main role in the anti-phOx response. Five V kappa and six VH genes were found to participate. All five V kappa genes or their "alleles" had previously been found from the BALB/c response to 2-phenyloxazolone (phOx). On the other hand, the two strains use different VH genes for the anti-phOx response. Most C57BL antibodies were coded by VH genes of group 1 which has only minor role in the BALB/c response. The remaining VH genes were from group 7. Our data show that one V kappa segment (e.g. V kappa-Ox1) can code for anti-phOx antibodies with several, even widely different, VH genes. On the other hand, they emphasize the role of certain VH/VL gene combinations for the anti-phOx specificity. Thus, VH genes of group 7 were found to code for anti-phOx antibodies only together with the V kappa 45.1 gene.

Activation of memory and virgin B cell clones in hyperimmune animals

To study the long-term memory response, BALB/c mice were allowed to rest for over a year after a secondary immunization with the hapten 2-phenyl-oxazol-5-one (phOx). For the tertiary immunization two different protocols were used. In one protocol mice were injected i.v. and 3 days later spleen cells were fused to a nonproducing hybridoma line. PhOx-specific hybridomas were established and the sequence of the heavy and light chain mRNA was determined. This tertiary response resembled the diversity pattern of the secondary response with a further increase both in somatic mutations and in the average dissociation constant. The high number of somatic mutations demonstrates the persistence of memory B cell clones over a long time period. In the second protocol mice were boosted with an i.p. injection of alumprecipitated antigen phOx and 7 or 14 days later spleen cells were fused. Sequence analysis of heavy and light chain mRNA showed that these tertiary response antibody molecules had surprisingly few somatic mutations, indicating an activation of virgin B cell clones in these hyperimmunized animals. The maturation of these newly stimulated B cell clones seems to follow somewhat similar rules to those found for the primary response. It appears therefore that the two immunization protocols reflect the response of memory and virgin B cells, respectively.

Somatic mutation in anti-phosphorylcholine antibodies

A detailed analysis of the genes and proteins that participate in the murine immune response to PC has provided key insights at the structural level into the phenomenon of somatic mutation in B cells. Most anti-PC antibodies are encoded by 1 VH gene of the S107 subfamily, and 3 VK genes, VKT15 of the VK22 subfamily, VKM3 from the VK8 subfamily, and VK167 from the VK24 subfamily. No mutation was detected in these genes until the 2nd wk after immunization, indicating that mutation is under developmental control. The protein sequences of 73 heavy and light chains derived from the secondary response support the concept of developmental activation of mutation after antigen stimulation. No mutation was found in the IgM antibodies, whereas half of the IgG and IgA antibodies had mutation. Most of the mutated antibodies had higher affinity for antigen than their germline counterparts, which suggests that the major role of somatic mutation is to increase affinity rather than to create new specificities. Nucleotide sequencing established two hallmarks of mutation in immunoglobulin genes: mutations are targeted to a 1 kilobase region surrounding and including the rearranged variable gene, and they occur at an extraordinary frequency of 10(-2) nucleotide substitutions. Mutation is probably caused by DNA repair, and may occur during error-prone repair of nicked DNA around the variable gene or during mismatch repair of misaligned structural intermediates. The elucidation of this remarkable mechanism clearly requires studies of a more dynamic character. Two major questions that need to be answered are: what targets mutation to the variable gene, and what enzymes are involved?