Immunoglobulin switch μ sequence causes RNA polymerase II accumulation and reduces dA hypermutation

Abstract

Repetitive DNA sequences in the immunoglobulin switch μ region form RNA-containing secondary structures and undergo hypermutation by activation-induced deaminase (AID). To examine how DNA structure affects transcription and hypermutation, we mapped the position of RNA polymerase II molecules and mutations across a 5-kb region spanning the intronic enhancer to the constant μ gene. For RNA polymerase II, the distribution was determined by nuclear run-on and chromatin immunoprecipitation assays in B cells from uracil-DNA glycosylase (UNG)–deficient mice stimulated ex vivo. RNA polymerases were found at a high density in DNA flanking both sides of a 1-kb repetitive sequence that forms the core of the switch region. The pileup of polymerases was similar in unstimulated and stimulated cells from Ung−/− and Aid−/−Ung−/− mice but was absent in cells from mice with a deletion of the switch region. For mutations, DNA was sequenced from Ung−/− B cells stimulated in vivo. Surprisingly, mutations of A nucleotides, which are incorporated by DNA polymerase η, decreased 10-fold before the repetitive sequence, suggesting that the polymerase was less active in this region. We propose that altered DNA structure in the switch region pauses RNA polymerase II and limits access of DNA polymerase η during hypermutation.

AID and Somatic Hypermutation

“…It has been shown in vitro that R-loops block transcription (Canugovi et al ., 2009; Tornaletti et al ., 2008), because the polymerases may have difficulty unwinding the stable RNA-DNA hybrid. A recent study (Rajagopal et al ., 2009) measured the density of polymerases in vivo by nuclear run-on, and found a 5–10 fold increase in polymerases located in close proximity to the Sμ repetitive core. These polymerases appeared to be piling up because they encountered a road-block ahead caused by the RNA-DNA hybrids.…”

XRCC1 suppresses somatic hypermutation and promotes alternative nonhomologous end joining inIghgenes

“…also had an altered spectrum of substitutions, showing that UNG participates in mutagenesis (Rada et al, 2002;Saribasak et al, 2006;Rajagopal et al, 2009). Thus, UNG is required for both repairing mutations and generating them, which may explain why the mutation frequency was not dramatically increased in UNG-deficient mice.…”

“…In the S  region, breaks are frequent as a result of multiple targeting sites for AID activity and prolonged RNA pausing in R-loops (Rajagopal et al, 2009;Wang et al, 2009). UNG and APE1 then produce single-strand breaks, which can be repaired by BER.…”

AID and partners: for better and (not) for worse

“…Secondly, RNA polymerase II (RNA Pol II) and its co-factors recruit AID at stalling sites as demonstrated by the striking genome-wide and local correlation between AID occupancy and the density of RNA Pol II and Spt5, a factor preferentially associated with stalled Pol II which binds AID in vitro [7 ,14 ]. Intrinsic properties of S region sequences would cause transcription stalling even at unusual distance from the transcription start (TSS), enabling subsequent recruitment of AID all along S regions via physical interaction with Spt5 [14 ,15,16] (Figure 1a). Of importance, although V regions were excluded from genome-wide analysis because of their high diversity, these results may apply to SHM as well since Spt5 occupancy also correlated with somatic mutations of off-targets [14 ].…”