Extended Data Figure 3. S1PR2 heterozygosity confers a survival advantage to GC B cells and R147C S1PR2 fails to function.

(a, b) Flow cytometry of Fo and GC B cells from mLN and PP of mixed BM chimeras generated as in Fig. 1e. Gating strategy for FoB and GCB in mLN is shown in a and percentages of CD45.2⁺ cells in Fo and GC B cells from PP are shown in b. Data in b are pooled from 4 independent experiments. (c) Gating strategy of Thy1.1 reporter expression in Fo and GC B cells from PP (c) or fold change in Thy1.1⁺ cells in GC relative to Fo B cells of mLN (d) of retrovirally transduced BM chimeras as described in Fig. 1f. Data in d are pooled from 3 independent experiments. *P<0.05, ***P<0.001, unpaired two-tailed Student's t test. There was increased variability in mLN relative to PP when WT S1PR2 was transduced into S1PR2^+/- BM. Nine of 17 animals reconstituted with S1PR2^+/- BM transduced with WT S1PR2 showed a reduction in expression of Thy1.1 in mLN GC relative to follicular B cells, whereas in 6 of 8 animals reconstituted with R147C S1PR2 there was increased reporter expression. (e) The hydrogen bond formed between Y141 in ICL2 and D130 on TM3 has been observed only in the active state of β₂-adrenergic receptor (shown in pink) and not in the inactive state (shown in cyan). (f) Population distribution of the conformational states showing the predicted hydrogen bond network between R147 (TM4), Y140 (ICL2) and E129 (TM3) of the wild type (solid lines) and R147C mutant (dashed lines) of S1PR2. (g) The network of predicted hydrogen bonds mediated by Y140 on ICL2. The hydrogen bond network tightens the interactions between transmembrane helices TM3 and TM4. We hypothesize that this network stabilizes the putative active state conformation of S1PR2. Such a network is broken in the R147C mutant and hence this mutant does not activate the G-protein.