Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 113 (28), 7846-51

Highly Sensitive and Unbiased Approach for Elucidating Antibody Repertoires

Affiliations

Highly Sensitive and Unbiased Approach for Elucidating Antibody Repertoires

Sherry G Lin et al. Proc Natl Acad Sci U S A.

Abstract

Developing B lymphocytes undergo V(D)J recombination to assemble germ-line V, D, and J gene segments into exons that encode the antigen-binding variable region of Ig heavy (H) and light (L) chains. IgH and IgL chains associate to form the B-cell receptor (BCR), which, upon antigen binding, activates B cells to secrete BCR as an antibody. Each of the huge number of clonally independent B cells expresses a unique set of IgH and IgL variable regions. The ability of V(D)J recombination to generate vast primary B-cell repertoires results from a combinatorial assortment of large numbers of different V, D, and J segments, coupled with diversification of the junctions between them to generate the complementary determining region 3 (CDR3) for antigen contact. Approaches to evaluate in depth the content of primary antibody repertoires and, ultimately, to study how they are further molded by secondary mutation and affinity maturation processes are of great importance to the B-cell development, vaccine, and antibody fields. We now describe an unbiased, sensitive, and readily accessible assay, referred to as high-throughput genome-wide translocation sequencing-adapted repertoire sequencing (HTGTS-Rep-seq), to quantify antibody repertoires. HTGTS-Rep-seq quantitatively identifies the vast majority of IgH and IgL V(D)J exons, including their unique CDR3 sequences, from progenitor and mature mouse B lineage cells via the use of specific J primers. HTGTS-Rep-seq also accurately quantifies DJH intermediates and V(D)J exons in either productive or nonproductive configurations. HTGTS-Rep-seq should be useful for studies of human samples, including clonal B-cell expansions, and also for following antibody affinity maturation processes.

Keywords: HTGTS-Rep-seq; V(D)J recombination; antibody repertoires.

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Fig. S1.
Fig. S1.
Schematic for HTGTS-Rep-seq. (A) Schematic of the generation of DJ and VDJ rearrangements via V(D)J recombination showing Vs (green), Ds (purple), and Js (orange). Representative DJ and VDJ joining events are shown. (B) HTGTS-Rep-seq method overview. Briefly, genomic DNA from B-cell populations are sonicated and linearly amplified with a biotinylated primer that anneals downstream of one specific J segment. The biotin-labeled single-stranded DNA products are enriched with streptavidin beads, and 3′ ends are ligated in an unbiased manner with a bridge adaptor containing a 6-nucleotide random nucleotide (highlighted in the rectangular box). Products were then prepared for 2 × 300-bp sequencing on an Illumina MiSeq. Generated reads were analyzed with the Ig/TCR-Repertoire analysis pipeline described in Materials and Methods.
Fig. 1.
Fig. 1.
HTGTS-Rep-seq of VHDJH and DJH repertoire in pro-B cells and splenic B cells of C57BL/6 mice. (A) Schematic of the murine IgH locus showing VHs (green, functional; black, pseudo), DHs (purple), JHs (orange), and CH region (black). The red arrow indicates the JH4 coding end bait primer. (B) VH repertoire with productive and nonproductive information from VHDJH joins in pro-B cells (Upper) and IgM+ splenic B cells (Lower). Some of the most frequently used VHs are highlighted with arrows as indicated. (C) Utilization numbers of functional VHs and pseudo VHs across 16 families in HTGTS-Rep-seq libraries described in B. (D) Pie chart showing the average overall percentage of productive and nonproductive VHDJH joins from libraries described in B. (E) D use in VHDJH and DJH joins in pro-B cells and IgM+ splenic B cells as indicated. (F) DJH:VHDJH ratios in pro-B cells and IgM+ splenic B cells as indicated. All of the data are showed by mean ± SEM, n = 3.
Fig. S2.
Fig. S2.
HTGTS-Rep-seq of VHDJH and DJH repertoire in pro-B cells and IgM+ splenic B cells of 129SVE mice. (A) Schematic of the murine IgH locus showing VHs (green, functional; black, pseudo), DHs (purple), and JHs (orange). The red arrow indicates the JH4 coding end bait primer. (B) VH repertoire with productive and nonproductive information from VHDJH joins in pro-B cells (Upper) and IgM+ splenic B cells (Lower). Some of the most frequently used VHs are highlighted with arrows as indicated. Mean ± SEM, n = 3. (C) Utilization numbers of functional or pseudo VHs across 16 families in the HTGTS-Rep-seq libraries described in B. (D) Pie charts showing the average overall percentage ± SEM of productive and nonproductive VHDJH joins in pro-B cells (Upper) and IgM+ splenic B cells (Lower). (E) D use in DJH joins in pro-B cells and IgM+ splenic B cells as indicated. Mean ± SEM, n = 3. (F) Comparison of DJH:VHDJH ratios in pro-B cells and IgM+ splenic B cells as indicated. Mean ± SEM, n = 3. Details of the analysis are as described for Fig. 1.
Fig. 2.
Fig. 2.
VHDJH and DJH repertoires in IgM+ splenic B cells across four JH baits. (A) VH repertoire with productive and nonproductive information from VHDJH joins (Left) and pie charts showing the average overall percentage of productive and nonproductive VHDJH joins (Right) in IgM+ splenic B cells using each of the JH coding end bait primers as indicated. (B) Comparison of D use in DJH joins in IgM+ splenic B cells using each of the JH coding end bait primers. (C) Comparison of DJH:VHDJH ratios in IgM+ splenic B cells using each of the JH coding end bait primers. Mean ± SEM, n = 3 for all of the data. Other analysis details are as described for Fig. 1.
Fig. S3.
Fig. S3.
Comparison of VHDJH and DJH repertoire in IgM+ splenic B cells of 129SVE mice using four different JH baits. (A) VH repertoire with productive and nonproductive information from VHDJH joins (Left) and pie charts showing the average overall percentage ± SEM of productive and nonproductive VHDJH joins (Right) in IgM+ splenic B cells using individual JH coding end bait primers. (B) Comparison of D use in DJH joins in IgM+ splenic B cells using each of the JH coding end bait primers. (C) Comparison of DJH:VHDJH ratios in IgM+ splenic B cells using each of the JH coding end bait primers. Mean ± SEM, n = 3 for all of the panels. Other analysis details are as described for Fig. 1.
Fig. S4.
Fig. S4.
In-frame VHDJH proportions across JH coding end lengths for JH1-4. (A) Alignment of the germ-line sequences of JH1-4. The sequences were extracted from the mm9 genome and are highly conserved between 129SVE and C57BL/6. The WGXG-encoding sequences are in red. JH length is marked with blue arrowheads, with 1 indicating the nucleotide most proximal to the bait primer. (B) Red line plots show the number per 10,000 total V(D)J joins that retained the indicated JH length for each JH bait. Blue bar graphs show the percentage of in-frame V(D)J exons at each retained JH length. Mean ± SEM, n = 3.
Fig. S5.
Fig. S5.
IgM+ splenic B-cell VHDJH use profiles in a 129SVE mouse using four JH baits combined. (A) Schematic of IgH locus as in Fig. 1. Red arrows indicate mixed primers that bind downstream of each JH. (B) VH use profiles separated by JH segment baits. One representative profile is shown here from two repeats of combined primer HTGTS-Rep-seq libraries. (C) D use in DJH joins in IgM+ splenic B cells using each of the JH coding end bait primers.
Fig. 3.
Fig. 3.
HTGTS-Rep-seq of VJκ repertoire in IgM+ splenic B cells of C57BL/6 mice using Jκ5 bait primer. (A) Schematic of the murine Igκ locus showing Vκs and Jκs. Green and orange bars indicate functional Vκs with convergent and tandem transcriptional orientations, respectively, to the downstream Jκs. Black bars indicate pseudo Vκs. The red arrow indicates the Jκ5 coding end bait primer. (B, Left) Vκ repertoire with productive and nonproductive information from VJκ joins in IgM+ splenic B cells with Jκ5 bait primer either individually (Upper) or from combined Jκ bait primers (Lower). Some differentially used Vκs among four different Jκs are highlighted with arrows as indicated (see also Fig. S6). (Right) Pie chart showing the overall percentage of productive and nonproductive VJκ joins. Representative results from two repeats are shown. (C) Utilization numbers of functional and pseudo Vκs across 20 families in libraries described in B.
Fig. S6.
Fig. S6.
Igκ repertoire in IgM+ splenic B cells of C57BL/6 mice using different Jκ baits. (A) Schematic of Igκ locus, as in Fig. 3. Red arrows indicate the position of used Jκ bait primers. (B) Vκ use profiles and overall productive/nonproductive ratios of VJκ separated by Jκ baits in IgM+ splenic B cells. In each panel, representative Vκ repertoires with productive and nonproductive information from VJκ joins with each Jκ bait primer either individually (Upper) or from combined Jκ primers (Lower) are shown. Some differentially used Vκs among four different Jκs are highlighted with arrows as indicated (see also Fig. 3). Representative results from two repeats are shown.
Fig. S7.
Fig. S7.
CDR3 length distribution and consensus motif of productive VHDJH and VJκ exons. (A) CDR3 length distribution of productive VHDJH exons in C57BL/6 pro-B libraries made with JH4 bait primer. Consensus CDR3 motif plots were made for the subset of 11- to 13-aa-length CDR3 sequences, flanked on either end by the consensus cysteine and tryptophan. (B) As in A, for C57BL/6 splenic B libraries made with JH4 bait primer. (C) As in A, for C57BL/6 splenic B libraries made with the four JH bait primers. Mean ± SEM, n = 3 for AC. (D) As in A, for C57BL/6 splenic B libraries made with Jκ5 primer. Note that we noticed some errors in our CDR3 sequence analyses due to the basal levels of sequencing errors of current high-throughput sequencing methods, including Illumina MiSeq, and the read length (maximum 600 bp) that are not sufficient to cover entire sequences of longer DNA fragments containing V(D)J exons. However, we eliminated such potential ambiguities by including in our analyses only overlapping joined reads and/or by increasing thresholds for read quality.
Fig. 4.
Fig. 4.
Representative VHDJH repertoire can be generated from small amounts of starting genomic DNA. (A) VH repertoire with productive and nonproductive information from VHDJH joins (Left) and pie charts showing the average overall percentage of productive and nonproductive VHDJH joins (Right) in IgM+ splenic B cells cloned from indicated amounts of genomic DNA using JH4 coding end bait primer. Mean ± SEM, n = 3. (B) VH utilization numbers separated by family, organized as in Fig.1C.
Fig. S8.
Fig. S8.
Characterization of unique CDR3 reads. (A) Proportion of unique CDR3 sequences for each technical repeat library from Fig. 4. Mean ± SEM, n = 3. (B) The number of identical CDR3 sequences between technical repeat libraries at varying amounts of starting material.

Similar articles

See all similar articles

Cited by 17 articles

See all "Cited by" articles

Publication types

LinkOut - more resources

Feedback