Antibody Repertoire Analysis Of The Immune Response To Recombinant Anthrax Vaccine

Vaccine is a useful tool for preventing and controlling infectious diseases. Vaccination with antigen formulation can activate the adaptive immunity and facilitate the establishment of immune memory, and thus provide a long-lasting or even lifelong protection. Adaptive immunity consists of celluar immunity and humoral immunity, in which the antibody-antigen interaction plays a fundamental role. In humoral immunity, antibodies(including B cell receptors, the membrane-bound form of antibodies expressed on the surface of B cells) are the key molecules that protect the body from the hurt caused by specific pathogens, and are abundant in diversity. Antibody diversity includes combinatorial diversity, junctional diversity, and somatic hypermutation. Combinatorial diversity refer to the diversities in random combination of germline genes of antibody variable regions and the combination of antibody heavy chains and light chains. Junctional diversity results from the inaccurate junction between three(heavy chain) or two(light chain) germline gene segments. Somatic hypermutation means high frequency point mutations in the rearrangement antibody genes following antigenic stimulations, which also greatly increases the antibody diversity. Such an abundant diversity makes antibodies able to recognize almost all pathogens the body may encounter. Accordingly, the analysis of antibodies, antibody secreting cells or memory B cells is an important part of vaccine evaluation.At present, the assessment of humoral immune response mainly depends on enzyme-linked immunosorbent assay(ELISA), enzyme-linked immunospot assay(ELISPOT) and so on. These methods can be used to analyze antibodies, antibody secreting cells and memory B cells in quality or quantity, but they cannot be used to analyze samples in molecular level, such as obtaining the sequences and features of antigen specific antibodies. Hence how to acquire the sequence information of antibodies and how to distinguish antigen specific antibodies from other antibodies become urgent issues. However, the bountiful diversity of antibodies poses great challenge for these analyses.High-throughput sequencing technology has developed over several years and has been applied in an increasing number of research fields. High-throughput sequencing technology could be used to acquire plenteous number of sequences parallelly, and thus is suitable to analyze molecules with abundant diversity such as antibodies. In brief, analyzing antibodies with high-throughput sequencing technology is called antibody repertoire analysis.Currently, antibody repertoire analysis has been utilized in research of infectious and autoimmune diseases, investigation of mechanism for immune tolerance, study of responses to vaccination, generation of specific antibodies and so on. Herein, we tried to study the immune response to a recombinant anthrax vaccine using antibody repertoire analysis, determine the feasibility of whether antibody repertoire analysis could be developed into an alternative method for vaccine evaluation, and provide directions for antibody screening as well as vaccine design based on the analysis results.Sequencing library preparation is the first step of antibody repertoire analysis. Therefore, a sequencing library preparation method using gene specific primers was established at the start of this study. Compared with other types of methods, preparing sequencing library using gene specific primers is easier to perform and could give shorter amplification products which facilitate the following high-throughput sequencing, but there is a possibility of potential amplification bias. To overcome this shortcoming, proper primers were selected. The results showed that the amplification with gene specific primers exhibited high correlation with amplification by the 5’-RACE method and the correlation coefficients were above 0.65. These results showed that the sequencing library preparation method used here largely avoided the amplification bias, which facilitated our following work.Two volunteers were vaccinated with the recombinant anthrax vaccine and received a second booster vaccination 28 days later. Peripheral blood mononuclear cells from the two volunteers were obtained on days 0, 7, 28 and 35. After the effectiveness of the vaccination was confirmed with ELISA and toxin neutralization assay(TNA), sequencing libraries were constructed using the method mentioned above and sequenced in parallel.Once the sequencing completed, antibody diversity reflected in the results of antibody repertoire sequencing was analyzed. The results showed that the germline gene frequency distributions of the two volunteers at different time points were close. The most common V gene segments of heavy chains were from families 1, 3 and 4 IX(>83%); the most common D gene segments of heavy chains were from family 3(>28%); the most common J gene segments of heavy chains were from family 4(>45%); the most common V gene segments of kappa chains were from families 1 and 3(>75%); the most common J gene segments of kappa chains were from families 1, 2 and 4(73%); the most common V gene segments of lambda chains were from families 1, 2 and 3(70%); the most common J gene segments of lambda chains were from family 3(44%). For the two volunteers, the mean length of complementary determining regions(CDR) 3 in heavy chains, kappa chains and lambda chains were 43 nucleotides, 37 nucleotides and 31 nucleotides respectively, and did not show any significant differences at different time points. The frequency of three types of heavy chain D gene segment reading frames and the numbers of P-nucleotides and N-nucleotides were also stable before and after the vaccinations. The nucleotides mutation rates varies slightly among those antibody chains at different times, however no apparent associations with immunization schedule were discovered. Thence we speculated that the complicated influences of antigens from environment might hide the vaccine specific effect. These results showed that the germline gene frequency distribution, the compositions of CDR3 and amino acid frequency distribution remained relatively stable and thus likely to be constant characteristics of antibody repertoire. Even if the somatic hypermutation rates could reflect the antigen-driven selective pressure, they could not reflect the response to vaccine directly due to the sophisticated environmental antigen influences.Considering that CDR3 plays a dominant role in the antigen-antibody interaction, we analyzed the features of CDR3 clones in the antibody repertoires. The results showed that both the CDR1 and CDR2 scores of high frequent CDR3 were significantly lower than those of low frequent CDR3, which reflected the effects of antigen-driven selective pressure on modeling the antibody repertoire. The CDR1 and CDR2 scores of high frequent CDR3 were unstable but did not show apparent associations with immunization schedule. The dynamic changes of CDR3 in heavy chains showed closely association with immunization schedule and clones emerged on days 7 and 35 accounted for more than 60% of clones existing at the two time points. For kappa and lambda chains, there were also new emergent clones, but the clones existed on day 0 accounted for almost half of the total clones throughout the entire observation period. Besides, considering that there was study showed the change in B cell clonality was positively correlated with the fold change in antibody X titer 14 days later, the dynamics of the heavy chain CDR3 convergence were consistent with the dynamics of the antigen specific antibody concentrations. The results showed that the pairing relationships between CDR3 and CDR1 or CDR2 could reflect the antigen-driven selective pressure, but they could not reflect the response to vaccine directly for the complicated environmental antigen influences. While the CDR3 dynamics and convergence features were associated with antigen specific antibodies after the vaccination and booster, and thus could be used to describe the vaccine-induced changes to the antibody repertoire.These results classified these antibody repertoire features into three types, and demonstrated that it was feasible to evaluate vaccine efficacy with antibody repertoire analysis, which added new dimensions for vaccine efficacy description.Based on the analysis of CDR3 dynamics, a new antibody clone screening strategy considering both static CDR3 frequency at a single time point and CDR3 dynamics was developed. Identification with ELISA showed that 52% of putative positive immunoglobulin heavy chains composed antigen-specific antibodies.The epitopes of 28 antibodies were predicted using the software Discovery Studio 4.5 and an algorithm used to analyze the RCF(residue contact frequency). The predicted results were verified by determining the domains where the epitopes located using ELISA. The result showed that the epitope predicted results of 21 antibodies among the 24 antigen-binding antibodies were consistent with the results of ELISA, which confirmed the reliability of the prediction method. Taking into account the results of TNA, we found that only 3 antibodies among the 24 antigen-binding antibodies showed toxin neutralization activity, which means that most high frequent antibodies screened here were non-neutraliztion antibodies and the epitopes recognized by these antibodies were non-neutraliztion epitopes. These results provide directions for the improvement of anthrax vaccine: Weakening the influences of dominant non-neutralization epitopes and increasing the activities of neutralizing epitopes would improve the neutralization antibody production efficiency of vaccine and thus the protective effects.In summary, this study established a sequencing library preparation method for antibody repertoire analysis, then analyzed the immune response to recombinant anthrax vaccine with antibody repertoire analysis, and discovered some new metrics for evaluation of vaccine efficiency. On this basis, we developed a strategy for screening positive clones, and analyzed the epitopes and toxin neutralizing activities XI of some antigen specific antibodies, which provided directions for vaccine improvement. These results demonstrated that antibody repertoire analysis could add new dimensions to the evaluation of vaccine efficiency, facilitate the generation of antigen specific antibodies, and direct the improvement on vaccine efficiency.