BLIMP1 shapes germinal center B-cell clonal diversity by gating chromatin accessibility during light-to-dark zone transition

Extensive studies have delineated signaling systems and transcription factors (TFs) that positively regulate GC B cell responses. In the LZ, T follicular helper (Tfh) derived CD40L and ICOS along with B cell receptor (BCR) signals license B cell clones to pass through the G1-S checkpoint; the cells then re-enter the DZ for clonal expansion and AID mediated somatic hypermutation (SHM). Among TFs that promote GC initiation, selection, and output, BCL6 programs GC identity. MYC licenses positively selected LZ cells for re-entry and proliferation in the DZ, whereas FOXO1 controls the program underlying rapid cell divisions. PU.1 and SPIB via interactions with IRF8 help maintain GC identity and selection, acting on ETS–IRF composite elements (EICE). BATF supports robust GC responses and class switching. The POU TF module comprised of OCT1/POU2F1, OCT2/POU2F2, and their co-activator OBF1/POU2AF1 maintains BCR-dependent transcription required for GC expansion. Canonical and alternative NF-κB components c-Rel and p52/RelB also promote GC maintenance. A p52–ETS1 complex has been recently shown to induce OCT1 and OBF1, thereby generating a feedforward regulatory loop. In contrast, few signaling-induced negative regulators of the GC response that restrict clonal dominance have been identified, notably Nr4a1 (NUR77). We posited that BLIMP1 which promotes exit of GC B cells into the plasma cell (PC) pathway, could also act as a signaling-induced transient feedback gate to restrain the GC B cell response, thereby reducing clonal dominance. In keeping with this hypothesis, low and heterogenous expression of BLIMP1 has been reported in GC B cells and its loss has been shown to result in larger GCs. Furthermore, Conter et al. recently demonstrated using bone marrow chimeras that B-cell-intrinsic BLIMP1 loss results in more exuberant GC responses associated with enhanced proliferation of antigen-specific GC B cells. We note that in addition to the feedback gating model for BLIMP1 action in the GC considered by us, two other models could account for the GC phenotype: (i) increased GC seeding due to impaired plasmablast differentiation or (ii) failed GC exit due to blocked PC differentiation. We reasoned that integrated single cell (sc) transcriptional and chromatin analyses of GC B cells along with delineation of their clonal dynamics could distinguish among the three models. Furthermore, such high dimensional analyses of the perturbed and amplified GC response would yield insights into the underlying molecular mechanisms of BLIMP1 action. Using sc transcriptome, V(D)J and chromatin profiling coupled with single nucleotide-resolution accessibility modeling of regulatory DNA sequences, we show that Prdm1-deficient B cells mount an exaggerated GC reaction characterized by larger clone sizes and enhanced affinity maturation culminating in greater clonal dominance. This phenotype is not attributable to increased GC seeding nor can it be merely accounted for by impaired exit of GC B cells into the PC pathway. Rather, the integrated genomic analyses reveal that BLIMP1 constrains expression of genes encoding components of the BCR-signaling cascade. Loss of this feedback, results in enhanced Syk, Lyn and Btk activity in GC B cells and augments chromatin engagement of signaling-inducible TFs at EICE, NF-κB and POU (OCT) motifs, promoting the G1–S transition during LZ selection and fueling DZ expansion.