Genetic variation in the immunoglobulin heavy chain locus shapes the human antibody repertoire

Abstract

Variation in the antibody response has been linked to differential outcomes in disease, and suboptimal vaccine and therapeutic responsiveness, the determinants of which have not been fully elucidated. Countering models that presume antibodies are generated largely by stochastic processes, we demonstrate that polymorphisms within the immunoglobulin heavy chain locus (IGH) impact the naive and antigen-experienced antibody repertoire, indicating that genetics predisposes individuals to mount qualitatively and quantitatively different antibody responses. We pair recently developed long-read genomic sequencing methods with antibody repertoire profiling to comprehensively resolve IGH genetic variation, including novel structural variants, single nucleotide variants, and genes and alleles. We show that IGH germline variants determine the presence and frequency of antibody genes in the expressed repertoire, including those enriched in functional elements linked to V(D)J recombination, and overlapping disease-associated variants. These results illuminate the power of leveraging IGH genetics to better understand the regulation, function, and dynamics of the antibody response in disease.

Ultrasensitive allele inference from immune repertoire sequencing data with MiXCR

“…To assess the performance of the developed algorithm we utilized publicly available datasets (Rodriguez et al 2023) containing both Rep-Seq data and highly reliable genotyping data of the IGH locus reconstructed using Pacific Biosciences HiFi long-read sequencing. For the sake of comparison we utilized 33 Rep-Seq data sets of sufficient sequencing depth (> 500,000 sequencing reads) and at least 3,000 unique full-length clonotypes.…”

“…For the alignment step and V and J gene annotation we used a custom minimalistic gene set library with only one allelic variant per V and J gene, derived from a custom public genome reference to match the one used for the long-read based genotyping (Rodriguez et al 2020). Then we performed allele variant inference and genotyping with both tools for all datasets containing more than 3,000 unique full-length clonotypes and compared the resulting individualized V and J gene libraries with the accurate genotype inferred with the next generation long-read sequencing (Rodriguez et al 2023), comparing nucleotide sequences of the genes. We also excluded poorly expressed allelic variants as determined by aligning the reads to the individualized gene reference libraries.…”

“…Most immune repertoire research has focused primarily on somatically derived immune receptor diversity, namely V(D)J recombination and somatic hypermutation (SHM) diversity. In recent years, however, the extent of population diversity has begun to be appreciated at both the immunoglobulin (IG) (Gidoni et al, 2019; Mikocziova et al 2021; Corcoran et al 2023; Rodriguez et al 2023; Gibson et al 2022) and T cell receptor (TR) loci (Omer et al, 2022; M. Corcoran et al, 2023; Rodriguez et al 2022). The functional significance of allelic variation in adaptive immune loci has also been recognized in the context of influenza, HIV and COVID-19 immunity and vaccination (Avnir et al 2016; Lee et al 2021; Leggat et al 2022; Pushparaj et al, 2022).…”

“…However, there are two other challenging obstacles for accurate genotyping and haplotyping of TR and IG loci using Rep-Seq data only. Common structural variants (SVs) in IG loci (Rodriguez et al 2023), especially gene duplications, in some cases make it hard to unequivocally map a sequence fromRep-Seq data to a particular germline gene without an additional source of information. Further, some alleles exhibit low usage levels, precluding their detection with Rep-Seq.…”

“…Although being the most recognized source of germline immunoglobulin sequence data, IMGT® lacks information on population allele frequencies and harbors sequences ‘mapped’ to the identified genes at the specific genomic locations. However, structural variation is quite common (Rodriguez et al 2023) in the IG and TR loci, while most new IG and TR sequence data is coming from Rep-Seq experiments, and can be hard to map to a particular gemline locus position. Other databases of immune receptor gene alleles have been introduced, such as pmTR (Dekker, van Dongen, Reinders, & Khatri, 2022, https://pmtrig.lumc.nl/), IgPdb (https://cgi.cse.unsw.edu.au/∼ihmmune/IgPdb) and Karolinska Institutet human T-cell receptor database (M. Corcoran et al, 2023, https://gkhlab.gitlab.io/tcr/).…”