Intricate targeting of immunoglobulin somatic hypermutation maximizes the efficiency of affinity maturation

Abstract

It is believed that immunoglobulin-variable region gene (IgV) somatic hypermutation (SHM) is initiated by activation-induced cytidine deaminase (AID) upon deamination of cytidine to deoxyuracil. Patch-excision repair of these lesions involving error prone DNA polymerases such as poleta causes mutations at all base positions. If not repaired, the deaminated nucleotides on the coding and noncoding strands result in C-to-T and G-to-A exchanges, respectively. Herein it is reported that IgV gene evolution has been considerably influenced by the need to accommodate extensive C deaminations and the resulting accumulation of C-to-T and G-to-A exchanges. Although seemingly counterintuitive, the precise placement of C and G nucleotides causes most C-to-T and G-to-A mutations to be silent or conservative. We hypothesize that without intricate positioning of C and G nucleotides the efficiency of affinity maturation would be significantly reduced due to a dominance of replacements caused by C and G transition mutations. The complexity of these evolved biases in codon use are compounded by the precise concomitant hotspot/coldspot targeting of AID activity and Poleta errors to maximize SHM in the CDRs and minimize mutations in the FWRs.

Figure 1.

SHM is biased for the introduction of excessive silent C-to-T and G-to-A exchanges. (A) The silent (white areas of bars) and replacement (black areas of bars) base exchange tendencies of 28,307 somatic mutations from 1,919 human IgVH genes. (B) Distribution of C-to-T and G-to-A silent (gray) and replacement (black) mutations to hotspots, coldspots, and neutral positions (x-axis), and the proportion of the various exchanges in AID hotspots or coldspots (y-axis) compared with A and T exchanges. AID hotspots and coldspot definitions are within the text (see introduction). (C) Analysis of the genetic code and non-IgV genes. Predicted occurrence of silent (white), replacement (black), or nonsense (gray) mutations for the 12 nucleotide exchanges. This analysis represents the sum result if every nucleotide was mutated to each of the three other nucleotides, and thus the sum of all possible mutations that could occur based on codon composition. The non-IgVH genes analyzed are listed within the text. CS, cold spot; HS, hot spot; Rep., replacement.

Figure 2.

A high frequency of silent C-to-T exchanges in the CDRs corresponds to mostly replacement mutations by all other exchanges. The distribution of silent (white portion of bars) and replacement (black portion of bars) mutations. Asterisks indicate amino acid positions where most of the C-to-T exchanges occurred at AID hotspot motifs (WRC). Whereas C-to-T exchanges are nearly all silent, G-to-A exchanges are predominantly silent in the FWRs only.

Figure 3.

Codon bias plus AID hotspot and coldspot targeting. Predicted effect of C-to-T (A and B), G-to-A (C and D), or other exchanges (E and F) in the CDRs and FWRs occurring in the germline (unmutated) counterparts of the 1,919 IgVH genes analyzed herein referred to as data set (compiled from the data in Tables S2 through S6). “All VH” genes refers to the 47 functional human germline VH genes to demonstrate that although the “Data set” contained certain VH genes more frequently than others, it is representative of the mutation pattern predicted for all VH genes. “Silent” bars indicate predicted frequency of silent mutations, “Replacement” indicates predicted frequency of amino acid replacement mutations, and “Nonsense” represents mutations where the exchange would result in stop codons. Black portions of bars represent proportion of Cs or Gs found within AID hotspot motifs, gray portions represent proportion of Cs or Gs in AID coldspots, and white represents Cs or Gs unaffected by either hotpots or coldspots (Neutral). White T bars represent the proportion of replacements that would result only in conservative amino acid replacements. The dashed line in B represents the proportion of Cs occurring in silent hotspots that are near the CDRs or in the hypervariable portion of FWR3 described in the text.

Figure 4.

Targeting of SHM. (A) Like C and G nucleotides, A and T nucleotides are also positioned to target SHM to the CDRs (black) and out of the FWRs (white). Polη tends to introduce point mutations at A in WA motifs during coding strand synthesis or at the reverse complement motifs (Ts in TW) during noncoding strand synthesis. Graph represents proportion of all As and Ts in WA or TW motifs from each region. The 17 non-Ig genes analyzed in Fig. 1 C were also analyzed for A and T positioning (gray). *, Significantly different by χ² test. (B) Analysis of silent mutation frequencies for Cs and Gs in AID hotpots, coldspots, and in neutral sequence contexts. The fold effect relative to neutral positions of hotspots and coldspots on mutation frequency is indicated at the base of the respective bars. (C and D) C-to-T and G-to-A mutation frequencies are remarkably predictable. There is no evidence for somatic selection of the CDR mutation frequencies and only moderate selection for increased silent and conservative mutations to the FWRs. The predicted frequency of AID-induced C-to-T (C) or G-to-A (D) in the CDRs or FWRs is based solely on codon composition of the unmutated germline genes compared with the “Actual” accumulation of C-to-T or G-to-A mutations in the 28,307 mutations sequenced. “Predicted” represents all possible mutations to the 1,919 VH genes analyzed (the data set values of Fig. 3, A and B) adjusted for the increased frequency of mutations at AID hotspots (3-fold for C-to-T and 2.7-fold for G-to-A) or decreased frequency at coldspots (0.5- or 0.6-fold for C-to-T and G-to-A).