Altered expression of the immune system enhances affinity maturation

Node Size Distribution of a Cellular Component During Clonal Blast-up: Evidence for an Alternating pmut Model

To determine how the experimental data compared with the theoretical predictions we plotted the experimental cumulative distribution function (CDF) of node sizes alongside the simulations (Fig. 4a). Notably, the experimental results were closer to the node size distribution predicted by the decreasing pmut model than the constant pmut model. The longer tails that the experimental cohort displayed were not seen in the decreasing pmut model. The assumption that the maximum number of cells per DZ cycle is six may be to blame for the discrepancy. Modestly changing the maximum number of cell divisions from six to eight produced a model that displayed a longer-tailed node size distribution and largely resolved the differences with respect to the experimental results (Fig. 4b, green). This change is consistent with the experimental data that contained more than 64 cells and at least two divisions. Alternatively, keeping the maximum number of divisions at six but changing the pmut model’s rate of change to 0.6 to 0.2 gave a nodes size distribution consistent with the experiments. A distribution of node sizes that were consistent with the experimental data was not generated by dramatically lowering pmut or extending the number of cell divisions.

Our findings raise the hypothesis that GC B cells actively downregulate SHM during clonal bursting. To test this, we sought to measure SHM in the presence or absence of perturbations that promote the inertial proliferation of GC B cells that drives bursting21 (Extended Data Fig. 3a). We first examined a genetic model in which GC B cells carry a Burkitt-lymphoma-associated gain-of-function mutation in the Ccnd3 gene (Ccnd3T283A). The growth of the DZ compartment and the size is increased due to the slowing of the nuclear export and the breakdown of the cell-cycle regulators. The previous method of obtaining the average percentages of wild-type and Ccnd3T283A/+ GC B cells was by using double-nucleotide pulse10 in a 30-minute window. On the basis of these data, we estimated that T283A-mutant GC B cells had undergone 67% more inertial cycles in the DZ by 10 dpi (13.5 versus 8.1) The extended data fig contains more data. 3b The Ccnd3T283A/+ B cells would accumulate more mutations than the WT cells if the DZ B cells had a uniform rate of mutations per division. 3c. However, the observed SHM distributions in WT and Ccnd3T283A/+ GC B cells were indistinguishable (Extended Data Fig. 3c–e), and, accordingly, the estimated mutation rate of Ccnd3T283A/+ GC B cells was significantly lower than that of WT GC B cells (Extended Data Fig. 3f). More precisely, whereas mutations among WT cells occurred at a rate close to the predicted one14,15,16 (0.38 ± 0.09 mutations per daughter, or 1.01 ± 0.15 mutations per 103 bases), mutations in Ccnd3T283A/+ GC B cells were substantially less frequent than expected (0.22 ± 0.05 mutations per daughter, or 0.59 ± 0.08 mutations per 103 bases) (Extended Data Fig. 3f). Therefore, despite increasing the number of DZ cells engaging in inertial cycling, Ccnd3T283A/+ mice did not show a concomitant increase in mutation accrual over time, pointing to a reduced mutation rate per cell division.

The DHB–tdTomato+ GC B cells sorted onto the monolayers were imaged longitudinally to see if they were stimulators of GC B cell proliferation. To follow the behaviour of our reporter during the transition from M phase to the following S phase, we aligned multiple single-cell DHB–tdTomato localization traces in time using anaphase as a reference (Extended Data Fig. 6a). GC B cells were cultured in the same conditions with a blocking anti- CD 40L, but they were not able to observe any cells entering a full CDK2low state. GC B cells do not deactivated CDK2 when stimulation is constant. Instead, immediately after mitosis, all cells exhibited an intermediate level of CDK2 activity, which continued to build and reached a C/N ratio of 1.0 (the reported threshold for the initiation of DNA synthesis43) within a median time of 1.3 h after anaphase, out of an average total cell-cycle duration of 8.7 h (Fig. 3g). GC B cells are able to transition fast from M to S phases by maintaining CDK2 activity close to the threshold needed for S stage entry and avoiding a CDK2low G0 stage.

We tracked GC B cell division in mice that had either mCherry or a DOX promoter under the control of a doxycycline promoter23,28. Lymphocytes from these mice (H2b-mCherry mice) constitutively express the mCherry indicator. Administration of DOX turns off the reporter gene and, upon dividing, cells dilute the indicator in proportion to the number of divisions made, whereas quiescent cells retain the indicator22,23,28 (Extended Data Fig. 1c–c

Despite being derived from the same unmutated common ancestor (UCA), expanded clones are heterogeneous, composed of a collection of diversified somatic variants (nodes) that have accumulated additional point mutations due to SHM. Genotype-collapsed phylogenetic trees were produced to visualize the contribution of individual somatic variants to each of the expanded clones (Fig. 2b and Extended Data Fig. 2f)27:30 UCAs are shown at the roots of the trees and are connected to variants through branching, as indicated by dotted lines. The sub-branching shows the distance between the variants. Division, as measured by mCherry dilution and affinity-enhancing mutations (pink outline), were also mapped onto the tree to annotate cell division status and relative affinity (Fig. 2b and Extended Data Fig. 2f.

To determine whether these effects were adjuvant specific, we profiled GC B cells responding to SARS-CoV-2-mRNA vaccination (Extended Data Fig. 5a). GC B cells were obtained 14 days after mRNA vaccination, and 36 h after DOX exposure and isolated according to mCherry status. Single-cell Sequencing was performed to resolve the IgH and IgL sequence. Genotype-collapsed phylogenetic trees obtained from the expanded clones confirmed the contribution of large nodes to affinity maturation (Extended Data Fig. 5b,c). Thus, expanded cells with identical sequences arise in GCs elicited by different immunogens and adjuvants.

We considered how to account for the large part observed in the experimental data. To do this, we used the stochastic pmut. 7a), and the case where the number of divisions corresponding to a constant level of T cell help was stochastic (Extended Data Fig. 7b. We found that neither scenario had a way to account for the long-tail behavior of nodes sizes. Both the theory and experimental data support the idea that the per division rate of SHM might be regulated and decreases with increasing T cell help.

AID introduces C>U mutations at preferential nucleotide sequence hotspots (WRC, W = A/T, R = A/G, underline indicates residue targeted for mutation)37,38. We wanted to know if there was a previous loss of these motifs that resulted in the absence of SHM in the expanded nodes. Hotspot motifs were equally intact in cells belonging to all three classes of nodes (average 92%, 93% and 94%, respectively; Extended Data Fig. 8d–f. Thus, target motif decay does not account for differences in SHM between nodes. We compared the level of cell death among mCherry compartments and found no differences. 8g,h).

The data is consistent with the idea that strong selection signals decrease the relative time DZ cells spend in G0/G1, which in turn reduces their exposure to AID.