Targets of somatic hypermutation within immunoglobulin light chain genes in zebrafish

Abstract

In mammals, somatic hypermutation (SHM) of immunoglobulin (Ig) genes is critical for the generation of high-affinity antibodies and effective immune responses. Knowledge of sequence-specific biases in the targeting of somatic mutations can be useful for studies aimed at understanding antibody repertoires produced in response to infections, B-cell neoplasms, or autoimmune disease. To evaluate potential nucleotide targets of somatic mutation in zebrafish (Danio rerio), an enriched IgL cDNA library was constructed and > 250 randomly selected clones were sequenced and analysed. In total, 55 unique VJ-C sequences were identified encoding a total of 125 mutations. Mutations were most prevalent in V(L) with a bias towards single base transitions and increased mutation in the complementarity-determining regions (CDRs). Overall, mutations were overrepresented at WRCH/DGYW motifs suggestive of activation-induced cytidine deaminase (AID) targeting which is common in mice and humans. In contrast to mammalian models, N and P addition was not observed and mutations at AID hotspots were largely restricted to palindromic WRCH/DGYW motifs. Mutability indexes for di- and trinucleotide combinations confirmed C/G targets within WRCH/DGYW motifs to be statistically significant mutational hotspots and showed trinucleotides ATC and ATG to be mutation coldspots. Additive mutations in VJ-C sequences revealed patterns of clonal expansion consistent with affinity maturation responses seen in higher vertebrates. Taken together, the data reveal specific nucleotide targets of SHM in zebrafish and suggest that AID and affinity maturation contribute to antibody diversification in this emerging immunological model.

Figure 1

Targeted zebrafish immunoglobulin L (IgL) germline reference sequence. The existing genomic annotation of zebrafish IgL was extended by assembling overlapping BAC inserts. BACs with overlapping end reads were prioritized for sequencing at the Sanger Center (http://www.sanger.ac.uk/Projects/D_rerio). The 923-kb contiguous chromosomal tiling path was manually assembled using the Artemis Annotation Software package. The IgL gene segments targeted are clustered in a single region on zebrafish chromosome 19. Immunoglobulin segments at this locus are divergent, enabling alignment of cDNA sequences with germline sequences to determine patterns of somatic hypermutation (SHM). Each BAC is designated by its corresponding NCBI accession number and drawn approximately to scale, while the IgL locus is expanded with exon sizes exaggerated.

Figure 2

Distribution of mutation frequency and activation-induced cytidine deaminase (AID) WRCH/DGYW hotspot motifs in targeted zebrafish immunoglobulin L (IgL). The combined frequency of mutations (primary vertical axis) per base sequenced for 55 unique VJ-C cDNA sequences plotted against position number (horizontal axis) reveals variability in V_L, whereas mutations in C_L were infrequent. The additive number of WRCH/DGYW hotspot motifs (secondary vertical axis) in the targeted germline IgL when plotted against position number shows an apparent trend of increasing mutation at AID hotspots. Within mammals, WRCH/DGYW motifs have been found to be the principal hotspot for AID-induced G:U lesions in rearranged immunoglobulin genes during somatic hypermutation. The bi-directional arrows show locations of palindromic WRCH/DGYW hotspot motifs which coincide with the highest mutation frequencies of the zebrafish IgL. The lower diagram depicts locations of V_L, J_L and C_L regions with respect to overall distribution of the VJ-C sequence and the complementarity-determining regions (CDRs) are depicted in the graph plot area.

Figure 3

Base exchanges are biased to replacement (R) mutations and nucleotide targeting at activation-induced cytidine deaminase (AID) hotspot motifs (WRCH/DGYW). The majority of the base exchanges observed resulted in amino acid changes [replacements (R); black areas of bars]. Silent mutations (S; white areas), while found in all four bases, were proportionally higher outside of WRCH/DGYW hotspot motifs. These data suggest that neutral mutations may be more prone to accumulate in bases outside of hotspots whereas replacement mutations are favoured at AID hotspot motifs.

Figure 4

Mutations at activation-induced cytidine deaminase (AID) hotspots are disproportionately concentrated at complementarity-determining region 3 (CDR3). (a) AID hotspot motifs (n = 37) are distributed across framework regions (FRs) and complementarity-determining regions (CDRs) of the targeted germline V_L. (b) The proportion of bases contained in WRCH/DGYW motifs varies slightly between FRs and CDRs. (c) Despite different densities of WRCH/DGYW motifs (a, b), mutations within WRCH/DGYW motifs are highly biased to occurrence in CDR3.

Figure 5

Somatic mutations are highly concentrated in palindromic activation-induced cytidine deaminase (AID) hotspot motifs. Within mammals, WRCH/DGYW motifs are considered the principal hotspot for AID-induced cytidine deamination in rearranged immunoglobulin during somatic hypermutation (SHM) whereas palindromic WRCY/WRCY are often targeted during class switch recombination (CSR). In the zebrafish V_L targeted, every WRCH/DGYW sequence was present (the number of occurrences is listed in parentheses and the mutation number within each is depicted in bold). Overall, V_L mutations were disproportionately concentrated within palindromic AGCT/AGCT and TGCA/TGCA AID hotspot motifs.

Figure 6

Lineage relationships of VJ-C cDNA clones are consistent with the possibility of clonal expansion and affinity maturation in zebrafish B cells. Individual VJ-C cDNA clones are depicted as circles by the accession number. A potential progenitor clone is depicted at the apex of each diagram. (a) Clones harbouring two or more additive mutations. (b) Clones radiating from a founder mutation in complementarity-determining region 3 (CDR3). In both diagrams, the progenitor VJ-C clone (accession no. EU821516) is identical in sequence to unmutated germline immunoglobulin L (IgL). Mutations listed for each VJ-C clone designate the location in framework regions (FRs) or CDRs and numbers in parentheses indicate the concordant germline position. Replacement mutations are indicated with an R while silent mutations are indicated with an S. Listings in italics are mutations at the C/G positions of WRCH/DGYW hotspot motifs. The directionality of arrow segments depicts sequential mutation accumulation in the radiation of clonal descendants.