Inherent properties of somatic hypermutation as revealed by human non‐productive V_H6 immunoglobulin rearrangements

Abstract

To study inherent properties of somatic hypermutation of human immunoglobulin genes in the absence of antigen selection, mutations of human non-productive VH6 rearrangements enriched by subtractive hybridization were characterized. Ten unique clones arising from nine non-productive rearrangements were isolated. The frequency of mutation was 3.0%. Analysis of these mutations showed intrinsic bias for transitions and cytosine (C) to guanine (G) and G to C transversions. Bias for the strand of DNA targeted by mutation was not evident. Replacement mutations in the complementarity-determining region (CDR) occurred more frequently than expected based on the primary DNA sequence. This targeting of replacement mutations to the CDR may explain the conservation of the VH6 sequence in primates.

Altered VH6-D-JH repertoire in human insulin-dependent diabetes mellitus and autoimmune idiopathic thrombocytopenic purpura

“…A likely explanation for this difference might be that the mutational process in IDDM patients is not directed against the VH6 gene, but to functional rearrangements containing other VH gene families. A strong negative selection against highly mutated VH6-D-JH gene rearrangements is operative on functional VH6 containing rearrangements in these patients, in line with previous reports on conservation of the VH6 gene sequence [33,36]. In AITP patients the observed increase in VH6 gene usage in the available repertoire [20] together with increased numbers of mutated VH6 genes indicates a continuous antigendriven expansion of a VH6 gene expressing memory B cell pool, rather than increased numbers of newly formed VH6-expressing B cells.…”

“…Comparison of the mutational frequencies in VH6containing lymphocytes shows that the overall mutation frequency (F + NF) was low in all four groups, as expected for peripheral resting B lymphocytes, namely 2.04 % and 1.52 % in the IDDM and IDDM controls, respectively, and 2.20 % and 2.73 % in AITP and AITP controls, respectively (Table 2). It has recently been shown that non-productive VH6-D-JH rearrangements undergo somatic mutation at a similar frequency as productive rearrangements in PBL [32] as well as in human splenic lymphocytes [33]. The cellular selection process following somatic mutation, operating only on F rearrangements, alters frequency and distribution of mutations in the total lymphocyte population [34].…”

Antigen selection in the IgE response of allergic and nonallergic individuals

“…In some studies sequences of interest have been compared with nonproductive Ig rearrangements, 22,23 and in such circumstances useful comparisons can be made of mutations in both the FR and CDR. We wished to compare productive IgE sequences with productive IgG sequences, and on the reasonable assumption that all such sequences have been subject to the same selection pressures for structural integrity, we were able to simplify the analysis by ignoring FR replacement mutations as a separate category.…”

Genesis of the strand‐biased signature in somatic hypermutation of rearranged immunoglobulin variable genes

“…A strand‐biased mutation signature is a clear prediction of the RT‐model. However, from about the mid‐1990s, confusion on this point reigned in the field as papers began to appear denying the existence of strand bias 30 , 31 , 32 , 33 , 34 .…”

“…We have concluded 36 that this artefact problem has affected many extant V gene data sets generated by PCR 30 , 31 , 32 , 33 , 34 , 37 , 38 , 39 , 40 . It is expected to be exacerbated by larger numbers of diverse but highly similar starting templates 30 , 31 , 32 , 37 , 38 and increasing cycle numbers (e.g. two‐step nested PCR protocols which employ> 50 cycles) 30 , 31 , 32 , 33 , 34 .…”