Transcription‐dependent R‐loop formation at mammalian class switch sequences

(and Cδ gene, by alternative splicing) associated with the Robert B.Tracy1 and Michael R.Lieber1,2,3,4 V(D)J complex to be replaced by any one of the downDepartments of 1Pathology, 2Biochemistry and Molecular Biology, and stream CH isotypes (either Cγ, Cε or Cα). This results in 3Molecular Microbiology and Immunology, Norris Comprehensive a deletion of the intervening genomic DNA, which includes Cancer Center, University of Southern California Keck School of the Cμ gene. The switch sequences, which range from 1 Medicine, 1441 Eastlake Avenue, Los Angeles, CA 90089-9176, USA to 10 kb in length, are all highly repetitive and G-rich on 4Corresponding author the non-template DNA strand and each contains its own e-mail: lieber@hsc.usc.edu unique set of short repeating units. The repeat unit lengths range from 20 to 80 nucleotides in length (for reviews, Immunoglobulin class switching is mediated by see Gritzmaker, 1989; Coffman et al., 1993; Lansford recombination between switch sequences located immeet al., 1996). Analysis of immunoglobulin switch region diately upstream of the immunoglobulin constant heavy recombination junctions is consistent with joining proceedchain genes. Targeting of recombination to particular ing via a non-homologous end-joining mechanism (Dunswitch sequences is associated temporally with trannick et al., 1993; Kinoshita et al., 1998). In fact, two scription through these regions. We recently have proteins, Ku and DNA-PK, which are involved in the provided evidence for inducible and stable RNA–DNA general process of non-homologous end joining, have hybrid formation at switch sequences in the mouse been shown to play a significant role in CSR (Rolink genome that are mechanistically important for class et al., 1996; Casellas et al., 1998; Manis et al., 1998). switching in vivo. Here, we define in vitro the precise While this provides some insight into the joining step, the configuration of the DNA and RNA strands within this synapsis, cutting and ligating phases of CSR still remain hybrid structure at the Sμ, Sγ3 and Sγ2b mouse switch entirely undefined. sequences. We find that the G-rich (non-template) DNA In combination with activators, particular cytokines strand of each switch sequence is hypersensitive to have been shown to induce or suppress germline transcripprobes throughout much of its length, while the C-rich tion from upstream activation regions (intron promoters), (template) DNA strand is essentially resistant. These which are located directly upstream of each mammalian results demonstrate formation of an R-loop, whereby switch sequence (Stavnezer, 1996; Snapper and Finkelman, the G-rich RNA strand forms a stable heteroduplex 1999). Upon activation of these promoters, germline or with its C-rich DNA strand counterpart, and the G-rich sterile transcripts are produced, which are directed into DNA strand exists primarily in a single-stranded state. the switch and constant regions (Snapper and Finkelman, We propose that the organized structure of the R-loop 1999). A direct correlation has now been established is essential for targeting the class switch recombination between activation of a specific promoter and subsequent machinery to these sequences. targeting of class switching to the respective CH isotype