In this study, we have described the 454 sequence analysis of a large naïve library of human IgM antibodies, and carried out immunogenetic analysis to study the origin, diversity, and maturation of selected known bnAbs against the HIV-1, SARS CoV RBD, and henipaviruses sG proteins. We have found intermediates of antiviral related bnAbs, of which most of those against the HIV-1 were highly diverged from their mature forms of bnAbs as compared to other viral targets, SARS CoV, and henipaviruses.
Although antibodies are generated through various mechanisms involving VDJ recombination, junctional modification, and hypermutations, the V-genes sculpt the most of the antigen-combining sites, CDR1 and CDR2, and support frameworks for the CDR3. We found that antiviral antibodies targeting different Env binding regions of the HIV-1 and other viruses utilized different germline V-genes as the origins (Table ). We noted that, among antiviral-related bnAbs, the V1-69 gene usage was dominated in the heavy chains while V3-20 and V2-14 genes of kappa and lambda were used with the highest frequencies in the light chains of human IgM repertoire (Figure ). Accordingly, four of the VH
genes of bnAbs (4E10, X5, m102, and m396) originated from the V1-69, and three of them paired with the kappa V3-20 gene. One possible reason for dominance in the usage of those germline genes could be reflecting from the relatively higher frequencies of distributions observed in the expressed IgM repertoire (Figures ). The HV3 gene was used in the three of the HIV-1 bnAbs, 2G12, PG9, and CH01. The structural data for most of the bnAbs selected in this analysis were known and the heavy chains of these bnAbs were dominantly used. The increased number of VH
mutations and longer HCDR3s are characteristics for the HVI-1 bnAbs when compared to other antiviral bnAbs (Breden et al., 2011
). We analyzed the distribution of HCDR3 lengths and extent of somatic VH
mutations in the human IgM repertoire to compare with that of antiviral-related bnAbs (Figure ). The results showed that the longer HCDR3s and low level of somatic VH
mutations as compared to the HIV-1 bnAbs existed in the intermediates as found from the 454 sequencing. The somatic diversity through VDJ recombination involving antiviral-related V-genes in the IgM repertoire was found high; the most abundant VDJ combination consisted of the HV1-69 gene with certain D and J genes as depicted in gray and red (Figure ), which might be the reason for the preferential usage of that HV1-69 in many other viral diseases (Sui et al., 2009
Further, bnAbs against the SARS CoV and henipaviruses shared the heavy chain V-gene germline, HV1-69, with two of the HIV-1 bnAbs, 4E10, and X5. All of these four bnAbs were less divergent from their V-germlines and intermediates, when compared to other HIV-1 bnAbs, and formed a single cluster at a mid-point rooted phylogenetic tree (Figure ). The gp41 membrane-proximal epitope region (MPER) binding site bnAbs, 2F5, and m66, were moderately divergent from their V-germlines and intermediates and formed distinct clusters. The V-gene of VRC01 bnAb was the most divergent from its respective germline as well as the closest intermediate, and was placed at a distal branch of HV1 subgroup of bnAbs. For the mid-point rooted phylogenetic analysis, we included the closest intermediates only; however, favored maturation pathways could involve other intermediates too. We created the germline-rooted phylogenetic tree as a use-case for the bnAb b12 (Figure ) and analyzed the maturation pathway along different V-gene intermediates from HV1-3 gene family. The closest b12 intermediate, designated as G3JY1, had three mutations each at HCDR1 and HCDR2 compared to the germline, and were found similar though not identical to that of mature b12 (Figure ). Interestingly, we also identified a HCDR3 with the same length (20 aa) and 50% sequence identity to that of b12 (Figure ), which was found to be the most similar to the HCDR3 of b12 but the IGHV gene associated with that HCDR3 was found to be V4-b. This might suggest for the possible maturation mechanism of bnAbs which could be involving the VH replacement (Chen et al., 1995
). These two mutated residues (N36 from HCDR1 and Y59 from HCDR2) from the V-gene and a Trp residue from the D-gene (W111.1 from HCDR3) contributed to the most of binding interactions with the gp120 (Figure ) (Zhou et al., 2007
In summary, the 454 sequence analysis of a large naïve human antibody repertoire corresponding to the selected antiviral-related bnAbs revealed the germline V-gene usage, VDJ rearrangement, HCDR3 length diversity, and somatic mutations of potential intermediate antibodies of HIV-1 and other viruses such as SARS CoV and henipaviruses. Thus, B cell germline-lineage analysis using the 454 sequence data from different sources could help finding appropriate antibody intermediates, pathways, and mechanisms useful in the development of bnAbs and vaccines against the HIV-1 and other viral diseases.
Conflict of interest statement
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.