(a) qRT-PCR analysis of S1pr2 mRNA in naïve B cells (B220+CD138−CD38+IgD+), activated B cells (NP+B220+CD138−CD38+) at day 3, IgG1+ LZ GC (NP+B220+CD138−CD38−GL7+IgG1+CXCR4loCD86hi), DZ GC (NP+B220+CD138−CD38−GL7+IgG1+CXCR4hiCD86lo) B cells and plasmablasts/plasma cells (NP+B220loCD138+) at day 10, and IgG1+ memory B cells (NP+B220+CD38+IgG1+) at day 28. Samples are prepared from splenocytes of NP-CGG immunized wild-type mice. (b) Schematic representation of the structure of the S1pr2-ERT2cre BAC transgenic mice. (c) Flow cytometry of tdTomato+ cells among splenic NP-specific plasmablasts/plasma cells (NP+B220loIgG1+CD138+), IgG1+ memory B cells (NP+B220+IgG1+CD38+) and GC B cells (NP+B220+IgG1+CD38−GL7+) in NP-CGG immunized S1pr2-ERT2cre-tdTomato mice at day 10. The mice were treated with tamoxifen for 12 h before analysis (left). Bar graph showing the percentage of tdTomato+ cells in each population (right). (d) Schematic illustration of the experimental protocol for Fig. 1b. (e) Absolute numbers of tdTomato+IgG1+ memory (left) and GC B cells (right) for determining ratio (Memory/GC) in Fig. 1b. (f) Frequency of W33L+ or W33G99 B cells (as in Fig. 1c) among NP-specific tdTomato+IgG1+ LZ GC (CD38−CD86hiCXCR4lo) B cells and memory (CD38+) B cells from S1pr2-ERT2cre-tdTomato mice treated with tamoxifen for 2 d beginning at day 18 after NP-CGG-immunization (upper), and quantification of mutations in sequence encoding VH186.2 in those cells (lower). Each symbol (f) represents an individual cell; small horizontal lines indicate the mean. ***p < 0.001 (unpaired Student’s t-test). Data are pooled from two independent experiments (a; mean and s.d. n=3 samples), representative of two independent experiments (c; mean and s.d. n=3 mice), pooled from three independent experiments (e; mean of n=4 mice (day 8 and 32), n=5 mice (day 12 and 22)) and representative of three independent experiments (f; mean).