Notch2 controls developmental fate choices between germinal center and marginal zone B cells upon immunization

Babushku, Tea; Lechner, Markus; Ehrenberg, Stefanie; Rambold, Ursula; Schmidt-Supprian, Marc; Yates, Andrew J.; Rane, Sanket; Zimber-Strobl, Ursula; Strobl, Lothar J.

doi:10.1038/s41467-024-46024-1

Download PDF

Article
Open access
Published: 04 March 2024

Notch2 controls developmental fate choices between germinal center and marginal zone B cells upon immunization

Nature Communications volume 15, Article number: 1960 (2024) Cite this article

1459 Accesses
10 Altmetric
Metrics details

Subjects

Abstract

Sustained Notch2 signals induce trans-differentiation of Follicular B (FoB) cells into Marginal Zone B (MZB) cells in mice, but the physiology underlying this differentiation pathway is still elusive. Here, we demonstrate that most B cells receive a basal Notch signal, which is intensified in pre-MZB and MZB cells. Ablation or constitutive activation of Notch2 upon T-cell-dependent immunization reveals an interplay between antigen-induced activation and Notch2 signaling, in which FoB cells that turn off Notch2 signaling enter germinal centers (GC), while high Notch2 signaling leads to generation of MZB cells or to initiation of plasmablast differentiation. Notch2 signaling is dispensable for GC dynamics but appears to be re-induced in some centrocytes to govern expansion of IgG1⁺ GCB cells. Mathematical modelling suggests that antigen-activated FoB cells make a Notch2 dependent binary fate-decision to differentiate into either GCB or MZB cells. This bifurcation might serve as a mechanism to archive antigen-specific clones into functionally and spatially diverse B cell states to generate robust antibody and memory responses.

The concerted change in the distribution of cell cycle phases and zone composition in germinal centers is regulated by IL-21

Article Open access 09 December 2021

Notch2-mediated plasticity between marginal zone and follicular B cells

Article Open access 17 February 2021

Continually recruited naïve T cells contribute to the follicular helper and regulatory T cell pools in germinal centers

Article Open access 31 October 2023

Introduction

Mature splenic B2 cells are comprised of two distinct cell populations: Follicular B (FoB) and Marginal Zone B (MZB) cells. In mouse, FoB cells recirculate between lymphoid organs, whereas MZB cells are mostly sessile in the spleen, where they migrate extensively between the MZ and the follicle, with approximately 40% of MZB cells exchanging between the MZ and follicle within one hour¹. MZB cells respond to T-cell-independent (TI)-antigens by differentiation into low affinity IgM-producing plasma cells (PC)^2,3. In contrast, FoB cells respond to T-cell-dependent (TD)-antigens, giving rise to germinal centers (GC), where they undergo class switching and somatic hypermutation, resulting in the generation of high-affinity Ig-switched PCs and memory B cells^4,5.

B cell development begins in the bone marrow (BM) and is completed in the spleen. Immature B cells leave the BM upon successful rearrangement of the B cell-receptor (BCR) and pass through short-lived developmental stages, the transitional B cells (T1–T3), before differentiating into either FoB or MZB cells in the spleen^6,7,8. MZB cells may also develop from FoB cells, in particular FoB-II cells, a recirculating long-lived population with an IgM^highIgD^highCD21^intCD23⁺ phenotype^{9,10,11,12,13}. We recently strengthened this hypothesis by showing that the adoptive transfer of highly purified FoB cells into immunocompetent recipients led to the generation of donor-derived MZB cells via an intermediate MZ precursor stage¹⁴. Our data provided strong evidence that FoB cells can act as precursors for MZB cells and we postulated that this could be physiologically mediated via activation of Notch2.

Notch2 signaling has been long known to be a crucial player in MZB cell development. Constitutively active Notch2 signaling in T1 or FoB cells strongly induces MZB cell differentiation^14,15. Correspondingly, homozygous or heterozygous inactivation of Notch2 in B lymphocytes or of its ligand Delta Like Canonical Notch Ligand 1 (Dll-1) in non-hematopoietic cells leads to the total loss or reduction of MZB cells, respectively^16,17, suggesting a dependence of MZB cell development on ligand/receptor dosage. This phenomenon may imply that the bottleneck in MZB cell development is the extent of Notch2 signaling, which is modulated by Notch2 surface expression and its modification^17,18,19, as well as the availability of Dll-1-expressing fibroblasts²⁰. Additionally, BCR-signaling strength was suggested to influence MZB cell fate decisions^6,21. We have shown that Notch2 surface expression is strongly upregulated by BCR-stimulation in vitro¹⁴ and others found that BCR-signaling induces recruitment of Adam10 to the plasma membrane, enhancing the cleavage and subsequent activation of Notch2²². These findings suggest that BCR signals act upstream of Notch2 signaling, thereby increasing the likelihood of B cells to acquire a strong Notch2 signal that drives their commitment to the MZB cell fate. Interestingly, BCR and Notch2 signaling all converge on Akt signaling, which recently has been shown to drive MZB cell development²³.

While we recently demonstrated that prolonged Notch2 signaling can trigger the trans-differentiation of FoB cells into MZB cells, the physiological regulation of this trans-differentiation process remained unclear.

In this study, we demonstrate that MZB cells are routinely derived from FoB cells during TD-immune responses. Our findings suggest that the strength of Notch signaling plays a crucial role in determining the fate of antigen-activated B cells. B cells receiving strong Notch2 signals develop into either MZB cells or PCs, while antigen-activated B cells which downregulate the Notch2 signal differentiate into GCB cells, which subsequently give rise to long-lived PCs and memory B cells. Therefore, the Notch2 signaling pathway emerges as a pivotal switch in generating functionally and spatially diverse B cell clones with distinct roles in early, late, and recall responses.

Results

All B cells receive a Notch signal under steady-state conditions

To identify the B cell populations that receive a Notch signal, we analyzed B lymphocytes in the BM and spleen of CBF:H2B-Venus mice, which express Venus fused to histone 2B (H2B) in all cells that undergo active or have recently experienced Notch signaling²⁴. We found that almost all B cells express H2B-Venus. H2B-Venus expression gradually increased during B cell maturation and reached highest expression in MZB cells (Fig. 1a, b, Supplementary Fig. 1a, b). Analysis of the expression of three classical Notch-target genes Deltex1 (Dtx1), Hes1, and Complement receptor 2 (Cr2) revealed that they are less expressed in FoB cells from Notch2-deficient mice compared with controls (Fig. 1c) and more highly expressed in MZB cells compared with FoB cells, (Supplementary Fig. 1c), confirming that FoB cells receive a basal Notch2 signal that is amplified in MZB cells. The upregulation of H2B-Venus and Hes1 started in pre-MZB cells (CD21^highCD23⁺) (Fig. 1d), corroborating our assumption that enhanced Notch2 signaling in T2 or FoB cells initiates MZB cell differentiation.

**Fig. 1: All B cells are recipients of a Notch signal, which gradually increases during B cell development and is highest in Marginal Zone B cells.**

Although some mature MZB cells exhibited the highest H2B-Venus expression within all B cell subpopulations, approximately half of them expressed less H2B-Venus compared to the majority of B cells in the spleen (Fig. 1d). In contrast, the expression of the Notch-target gene Hes1 remained stable in all MZB cells (Fig. 1d). Hes1 levels were even faintly increased in H2B-Venus^low MZB cells in comparison to H2B-Venus^high MZB cells and were significantly higher than in FoB cells (Fig. 2a). Both H2B-Venus^high and H2B-Venus^low cells were found within the follicle as well as in the MZ (Fig. 2b). In the follicle, some of the H2B-Venus^high cells had a MZB cell shape (bigger in size and less roundish¹, white arrows). To further characterize H2B-Venus^high and H2B-Venus^low MZB cells we subdivided pre-MZB and MZB cells according to their CD23⁺ levels, which are sequentially downregulated during MZB cell maturation¹⁴. Both the H2B-Venus intensity in the H2B-Venus^high cell fraction and the percentage of H2B-Venus^low cells increased sequentially from pre-MZB, CD23^+/low, and CD23^low MZB cells (Fig. 2c). Additional stainings indicate that H2B-Venus^low MZB cells exhibit a more mature MZB cell phenotype (lower CD23 and IgD levels and higher CD1d and IgM levels) and a marginally higher percentage of apoptotic cells compared with H2B-Venus^high MZB cells (Supplementary Fig. 2a–c)^11,14. Based on these data, we conclude that the amount of H2B-Venus increases during MZB cell development, which underlines the importance of Notch2 signaling in MZB cell development. However, a subset of these mature MZB cells appears to undergo a reduction or loss of H2B-Venus during their further maturation or when on the path towards cell death/apoptosis. In contrast to FoB cells, MZB cells exhibit self-renewal capacity and an intrinsic capability for plasma cell differentiation. To understand why H2B-Venus levels are lower in certain MZB cells but remain unchanged in FoB cells, we investigated whether this loss occurs through cell division and differentiation processes. Ex vivo isolated splenic B cells were labeled with the proliferation dye CellTrace Far Red and stimulated with LPS, inducing both proliferation and PC differentiation. Plotting CellTrace Far Red against H2B-Venus revealed that the majority of cells lose H2B-Venus during proliferation, with a small subset maintaining high H2B-Venus expression despite proliferation (Supplementary Fig. 3a). We assume that these cells contain Notch-IC in the nucleus and produce new H2B-Venus upon proliferation. Next, we analyzed if the loss of H2B-Venus occurred not only through proliferation but also during PC differentiation. Gating cells based on the same CellTrace Far Red intensity and dividing them into different levels of CD138, which is progressively upregulated during PC differentiation, revealed a correlation between increasing CD138 levels and higher percentages of H2B-Venus^low cells in cells which have the same division rate (Supplementary Fig. 3b). These findings imply a further loss of H2B-Venus expression during PC differentiation.

**Fig. 2: Some Marginal Zone B cells lose their H2B-Venus signal.**

Based on these findings, we propose that MZB cells receive a strong Notch2 signal during their generation. However, the inherent ability of MZB cells to proliferate and differentiate into PCs results in a gradual loss of H2B-Venus in these cells. Supporting this, H2B-Venus^low MZB cells exhibit a higher percentage of Ki-67⁺ cells and elevated levels of Interferon regulatory factor 4 (Irf4) and B-lymphocyte-induced maturation protein-1 (Blimp1), compared to FoB cells and H2B-Venus^high MZB cells, indicating more robust proliferation and stronger differentiation toward plasma cells (Supplementary Fig 3c, d). These results indicate that in MZB cells H2B-Venus is progressively lost due to the combined effects of proliferation and PC differentiation.

Notch2 signaling is dispensable for GCB cell dynamics but appears to regulate the expansion of IgG1⁺ cells in the GC

The H2B-Venus signal was progressively downregulated in germinal center B (GCB) cells (Fig. 3a, Supplementary Fig. 4a). In immunohistochemistry, germinal center structures were predominantly H2B-Venus⁻ (Fig. 3b), suggesting that Notch signaling is switched off in GCB cells. This observation is consistent with a previous report by Valls and colleagues, who showed that Notch2 and Notch2 pathway genes are transcriptionally silenced by Bcl6 in GCB cells²⁵. By day 14, H2B-Venus^high GCB cells were enriched in the light zone (LZ), while H2B-Venus^mid and H2B-Venus^neg cells were predominantly found in the dark zone (DZ) (Supplementary Fig. 4b). H2B-Venus expression and Notch2 expression were slightly higher in LZ-GCB than in DZ-GCB cells (Supplementary Fig. 4c, d). These data suggest the downregulation of Notch2 signaling in GCB cells, but reactivation in some LZ-B cells.

**Fig. 3: Notch signaling is downregulated in most Germinal Center B cells but is re-induced in some Light Zone-GCB cells.**

To validate this finding, we immunized transgenic mice, in which Notch2 is inactivated (N2KO//CAR) or constitutively activated (N2IC/hCD2) upon TD-immunization by employing the Cγ1-Cre strain²⁶. As controls, we used control/CAR mice, in which the human coxsackie/adenovirus receptor (CAR) is expressed as reporter upon Cre-mediated recombination (Cre-reporter gene)²⁷. The Cre-reporter gene CAR was used to trace B cells that successfully underwent Cre-mediated recombination in N2KO//CAR and control/CAR mice, while hCD2 reported Notch2IC expression in N2IC/hCD2 mice (Supplementary Fig. 5a). Deletion efficiency of the targeted regions was comparable among the three mouse strains and percentages of Cre-reporter⁺ cells peaked at day 14 at approximately 5-6% (Supplementary Fig. 5b). While all mice had similar percentages of Cre-reporter⁺ B cells, N2IC/hCD2 mice failed to generate GCB cells (Fig. 3c). The few GCB cells observed in N2IC/hCD2 mice were mostly Cre-reporter^– (Fig. 3d) and formed small GC structures not located at the T/B cell border (Fig. 3e). These findings align with previous data from us and Valls and colleagues, showing that constitutive Notch2IC expression is incompatible with the GC reaction^15,25.

In contrast, N2KO//CAR mice generated a similar percentage of GCB cells as control/CAR mice and formed normal GC structures correctly localized at the B/T cell boundary (Fig. 3c–e), demonstrating that Notch2 signaling is not required for GC dynamics. Considering that H2B-Venus expression increased again in some LZ-B cells (Supplementary Fig. 4c), we wondered whether Notch signaling has a function there. Indeed, at day 14 after immunization, the proportion of Cre-reporter⁺IgG1⁺ B cells with a GCB phenotype was significantly reduced in N2KO//CAR in comparison to control mice (Fig. 3f, g), suggesting that lack of Notch2 signaling impairs the expansion and/or positive selection of IgG1-switched cells in the LZ.

Notch2 surface expression and signaling is upregulated in activated B cells

To better understand the regulation of Notch signaling during the TD-immune response, we investigated whether Notch signaling is enhanced in activated B cells. Combined CD40- and BCR-stimulation in vitro led to an around 20-fold upregulation of Notch2 surface levels. This upregulation started early and peaked between 24 h and 48 h (Fig. 4a, Supplementary Fig. 6a). We next examined whether during a TD-immune response Notch2 expression and Notch signaling are increased in activated B cells but are downregulated in GCB cells again. Indeed, Cre-reporter⁺CD38⁺CD95^high cells, which contain both activated FoB and memory B cells (Supplementary Fig. 6b), displayed the highest Notch2 expression, while GCB cells exhibited the lowest Notch2 levels (Fig. 4b, Supplementary Fig. 6c), which further decreased over time post-immunization (Fig. 4b). Similarly, Hes1 expression was upregulated in Cre-reporter⁺CD95^highCD38⁺ B cells from control/CAR, but not N2KO//CAR mice. Hes1 expression in N2IC/hCD2+ B cells exhibited similar regulation as in control B cells, but was generally higher (Fig. 4c, Supplementary Fig. 6d). Thus, our in vitro and in vivo data suggest that Notch2 surface expression and -signaling are induced in activated FoB cells.

**Fig. 4: Notch2 surface expression and signaling is enhanced in activated pre-Germinal Center B cells.**

To gain further insights into the effects of sustained Notch2 signaling in activated B cells, we examined the phenotype of Notch2IC-expressing B cells in more detail. N2IC/hCD2⁺ B cells were CD38^high and CD95^high-mid (Fig. 4d, e). Amounts of Irf4 and Bcl6, the key regulators of PC- and GCB cell differentiation, respectively^28,29, revealed an inverse regulation in N2IC/hCD2 and control/CAR mice. Control/CAR⁺ B cells strongly upregulated Bcl6 expression while maintaining low levels of Irf4, consistent with a GCB cell phenotype. In contrast, N2IC/hCD2⁺ B cells failed to upregulate Bcl6 but showed an induction of Irf4 (Fig. 4f). These findings indicate that sustained Notch2 signaling interferes with GC formation by suppressing Bcl6 and inducing Irf4.

Among the N2IC/hCD2⁺ B cells a subset exhibited high Irf4 levels, which were also observed in control/CAR mice, but at a significantly lower frequency. The Irf4^high cells were uniformly B220⁻ in both genotypes (Fig. 4g). Histology revealed that the enriched Irf4^high cells in N2IC/hCD2 mice were predominantly located in the extrafollicular regions, where plasmablastic cells (PB) and short-lived PCs are typically found (Fig. 4h). N2IC/hCD2 mice exhibited increased numbers of 4-hydroxy-3-nitrophenylacetyl (NP)-specific IgM antibody-secreting cells (ASC) in the spleen and elevated IgM titers in the serum 14d after immunization. In contrast, NP-specific IgG1 and IgG3 ASCs, as well as serum titers were decreased (Supplementary Fig. 7a). These findings suggested that Notch2 signaling enhances the production of short-lived IgM PCs.

Notch2 signaling in activated FoB cells initiates plasma cell differentiation but has to be switched off to enable their terminal differentiation

To confirm a potential role of Notch2 signaling in promoting PC differentiation, we analyzed N2KO//CAR//Cγ1-Cre and N2KO//CD19-Cre mice in comparison to control mice at different time points after immunization. Seven days after immunization, the frequencies of splenic PC were slightly reduced in N2KO compared to controls, while there was no difference at later time points (Supplementary Fig. 7b, c). ELISpot-analyses also revealed a slight reduction of NP-specific ASCs (mainly NP-IgG1) at day 7 (Supplementary Fig. 7d, e), suggesting that Notch2 signaling positively influences the early wave of PC generation, in which mainly short-lived PCs are generated.

To further understand the impact of Notch2 signaling on PC differentiation, we examined Blimp1 levels in Irf4^highB220^low cells, which increase during PC differentiation. In N2IC/hCD2 mice, the fraction of Irf4^highB220⁻Blimp1^high cells was significantly lower than in controls, and Ki-67 levels were higher (Fig. 5a, b). Although this regulation was not apparent in N2KO//CAR mice (Fig. 5b, Supplementary Fig. 8a), which might be due to a redundant function of Notch1- and Notch2 signaling in PC differentiation, with the major contribution of Notch1, the data suggest that persistent Notch signaling inhibits PC terminal differentiation, keeping PC precursors in a proliferative stage. This led us to hypothesize that downregulation of Notch signaling is necessary for final PC differentiation. To experimentally confirm this, we analyzed the H2B-Venus expression in PCs derived from CBF:H2B-Venus mice. Most TACI⁺CD138⁺ PCs lost their H2B-Venus expression (Fig. 5c), with remaining H2B-Venus^high cells showing partially a (pre-) plasmablast phenotype (CD19⁺B220⁺), while H2B-Venus^mid+low PCs exhibited full maturation into PCs (CD19^–B220^–) (Fig. 5c, d). We concluded that Notch signaling is active in activated B cells that are committed to PC differentiation and is completely shut down in mature PCs. To support this conclusion, an in-silico analysis of Notch2 and Notch-target gene expression was performed using the immgen.org database, which showed a significant reduction in Notch2, Hes1, Hes5, and Dtx1 mRNA in PB and PCs compared with FoB and MZB cells (Supplementary Fig. 8b). Collectively, these findings indicate that Notch signaling maintains pre-plasmablasts in a proliferative state, promoting their expansion and must be shut-off to enable terminal PC differentiation.

**Fig. 5: Notch2 signaling initiates plasma cell differentiation but must be downregulated to enable terminal differentiation.**

Some antigen-activated B cells differentiate into MZB cells during the TD-immune response

The B220⁺Irf4^mid population contained mainly GCB cells (Bcl6⁺) in control/CAR and non-GCB cells (Bcl6^–) in N2IC/hCD2 mice (Fig. 6a, b). Given the previously established role of constitutively active Notch2 signaling in the differentiation of FoB into MZB cells under steady-state conditions¹⁴, we analyzed whether the Cre-reporter⁺B220⁺Irf4^mid non-GCB cells from N2IC/hCD2 mice acquire a MZB cell phenotype over time. Indeed, by day 7 post-immunization, approximately one-fourth of this population displayed a MZB cell phenotype based on their CD21/CD23 surface expression (Fig. 6c, d). The frequencies of these cells increased over time post-immunization (Fig. 6d, Supplementary Fig. 9). Concurrently, these cells increased their size (FSC) and upregulated IgM, CD1d and CD38 (Fig. 6e), consistent with the characteristic MZB cell phenotype. Moreover, the N2IC-expressing cells exhibited migration towards the marginal zone (Fig. 6f). Taken together, our findings suggest that Notch2 signaling in antigen-activated FoB cells not only promotes their entry into PC differentiation, but also drives their differentiation into MZB cells.

**Fig. 6: N2IC/hCD2⁺ B cells differentiate into Marginal Zone B cells with the correct surface phenotype and splenic localization.**

To investigate whether MZB cell differentiation also occurs physiologically during TD-immune responses in B cells with normal Notch responsiveness, we examined the spleen of control/CAR mice after TD-immunization (Fig. 7a). Intriguingly, we observed the presence of Cre-reporter⁺CD23^lowCD21^high cells, which gradually increased in frequency over time following immunization, ranging between 2–10% throughout the observation period (Fig. 7a and b).

**Fig. 7: Notch2 signaling is necessary for the generation of Cre-reporter⁺ B cells with a Marginal Zone B cell phenotype.**

Notably, these Cre-reporter⁺CD23^lowCD21^high B cells were virtually absent in immunized N2KO//CAR mice, indicating that their generation is dependent on Notch2 (Fig. 7a, b). The newly generated Cre-reporter⁺ MZB cells displayed a characteristic MZB cell phenotype (Supplementary Fig. 10a). The percentage of NP-specific cells was comparable within the Cre-reporter⁺ MZB and memory compartments, while it was higher in FoB cells (Fig. 7c and Supplementary Fig. 10b). From day 14 to day 26, the percentages of NP⁺ cells increased significantly in the MZB cell compartment, whereas they remained relatively constant in FoB and memory B cells (Fig. 7c). These data suggest that with progressive time post-immunization Cre-reporter⁺NP⁺ cells accumulate in the MZB cell compartment. To assess the contribution of CAR⁺ MZB cells in recall immune responses, we performed booster immunizations in N2KO//CAR mice, lacking CAR⁺ MZB cells, 35 days after the first immunization and analyzed their immune response 7 days later in comparison to control mice. NP-specific IgM and IgG1 serum titers were lower in N2KO//CAR compared to control/CAR mice (Fig. 7d), suggesting that the CAR⁺ MZB cells contribute to recall immune responses by de novo PC production. However, further analyses are still required as defects in the memory responses of Notch2-deficient B cells could also contribute to the decreased serum titers.

Quantitatively mapping of B cell differentiation during an immune response reveals bifurcation of FoB cells into GCB and MZB cell fates

To gain insights about the precursor cells of CAR⁺ MZB cells, we produced correlation plots of CAR expression kinetics in the FoB, MZB, and GCB cell subsets in the course of the TD-immune reaction. We employed a sequential gating strategy to determine the frequencies of CAR-expressing B cells among the total GCB cells (CAR⁺B220⁺CD38^lowCD95^high), FoB cells (CAR⁺B220⁺CD38⁺CD95^mid/lowCD23⁺CD21^low) and MZB cells (CAR⁺B220⁺CD38⁺CD95^mid/lowCD23^lowCD21^high). An example of this hierarchical gating is depicted in Supplementary Fig. 11a, b. The correlation analysis revealed that the rise and decline in CAR⁺ GCB cell numbers mapped linearly with the accumulation of CAR-expressing FoB cells (Fig. 8a), underlining that GCB cells are mainly derived by the activation of FoB cells. We did not find any indication of a precursor-progeny relationship between CAR⁺ MZB and either CAR-expressing FoB or GCB cells, as their CAR expression kinetics correlated very weakly (Fig. 8b). Therefore, to uncover the developmental origins of CAR⁺ MZB cells, we next integrated the data from the immunization experiments into a deterministic mathematical modeling strategy. We hypothesized that CAR⁺ MZB cells emerge either from FoB cells (branched pathway, Fig. 8c) or from GCB cells (linear pathway, Fig. 8d), in addition to potential activation of CAR^- MZB cells by the TD-antigen, which results in CAR expression on these MZB cells. We also defined a ‘null model’ in which all CAR⁺ MZB cells are generated only by the activation of CAR^- MZB cells (Fig. 8e), without any influx from FoB and GCB cells.

**Fig. 8: Mathematical modeling of CAR expression dynamics in B cells during an ongoing TD-immune response.**

The simplest forms of branched, linear, and null models assume neutral dynamics for the birth-loss of GCB and MZB cells i.e., the influx of new cells into the compartment and their net loss (defined as the balance between division and true loss by death and differentiation) remain unchanged over time. To model the effect of antigen availability on B cell dynamics, we considered time-dependent variation in the rate of influx of FoB cells into the GCB cell compartment (α(t)) and in the rate of upregulation of CAR expression on MZB cells (β(t)). In this ‘time-varying influx’ model, the differentiation of FoB cells into CAR⁺ MZB cells (μ) remains unchanged with time. Additionally, we considered another extension of the neutral model in which the net rate of loss of GCB cells varies with time (δ(t)), while the loss rate of CAR⁺ MZB cells (λ) remains constant. Model specifics are described in further detail in the Materials and Methods.

We fitted each model simultaneously to the time-courses of counts of CAR⁺ MZB cells and CAR⁺ GCB cells from control/CAR mice (red dots in Fig. 9a) and N2KO//CAR mice (blue dots in Fig. 9a), using the Bayesian statistical approach and assessed the relative support for each model using the leave-one-out (LOO) cross validation, calculated as the model weight ‘W’ (estimates in Table 1a). Model weight calculation is described in detail in Supplementary Note 1. We consider that neither FoB nor GCB cells contribute to the generation of CAR⁺ MZB in N2KO//CAR mice (i.e., μ = 0), since activation induces Notch2 deletion in B cells in these mice. We also assumed different loss rate of CAR⁺ MZB cells (λ') in N2KO//CAR mice. The influxes into the CAR⁺ compartments of GCB and MZB cells were defined using the empirical descriptions of the pool sizes of total FoB cells and CAR⁻ MZB cells (Supplementary Fig. 12 and details in Supplementary Note 2).

**Fig. 9: Dynamics of CAR⁺ Marginal Zone B cells and Germinal Center B cells during a TD-immune response.**

Table 1 a Comparison of models describing population dynamics of Marginal Zone B and Germinal Center B cell populations upon TD-immunization. b Parameter estimates from the best-fitting branched pathway with time-varying influx model

Full size table

We found that the combination of the branched pathway with time-varying influx model received the strongest statistical support from the data (W = 79%, Table 1a). We also explored a version of time-varying influx model in which the rate of differentiation of FoB into CAR⁺ MZB varied with time (μ(t)), which received poorer statistical support (W = 17%), in One-on-one comparison to the best-fitting model (W = 83%).

Our best-fitting model nicely captured the rise and fall in CAR⁺ GCB cell numbers and described the persistence of CAR⁺ MZB cells for over a month post-immunization in control/CAR mice (Fig. 9a). The model estimates that activation of CAR^- MZB cells contributes approximately 25% to the CAR⁺ fraction in the MZB cells at day 4, which further declines steadily to near-zero by week 3 post-immunization (Fig. 9b and Table 1b). The numbers of CAR⁺ MZB cells are primarily sustained by the differentiation of antigen-activated FoB cells (Fig. 9b and Table 1b), highlighting the importance of Notch2 signaling in their development and maintenance. Our model predicts that CAR⁺ MZB cells are lost twice as rapidly in Notch2-deficient mice as compared to the control/CAR mice with functional Notch2 signaling (Fig. 9c and Table 1b). The shorter half-lives of CAR⁺ MZB clones in addition to the absence of influx from CAR⁺ FoB cells, explains their substantially diminished pool sizes in N2KO//CAR mice (Fig.9a, blue lines and dots). Overall, our modeling and experimental results reveal an instructive role of Notch2-mediated signals in determining B cell diversification patterns during immune responses.

Discussion

The conventional view of B cell development suggests that in mice, MZB cells emerge from transitional T2 cells, which bifurcate into mature FoB cells and MZB cells in the spleen, dependent on Notch2 signaling⁶. However, recent studies demonstrated that FoB cells can also differentiate into MZB cells^11,14, but it remains unresolved under what circumstances this occurs during normal physiological processes. In this study, we demonstrate that the transition from FoB to MZB cells can be triggered by B cell activation. Our experimental data and mathematical modeling suggest that integration of Notch2, BCR, and CD40 signaling triggers a binary switch in antigen-activated B cells leading to the development of GCB or MZB and plasma cells. Our findings suggest, that activated B cells that downregulate Notch signaling and Irf4 expression develop into GCB cells, while sustained and enhanced Notch signaling leads to the development of MZB cells or PC cells, likely depending on their Irf4 levels^29,30,31. Our analyses based on Notch-reporter transgenes²⁴ and endogenous Notch-target genes demonstrated that except GCB cells and PCs, all immature and mature B cell populations receive a basal Notch signal. Basal ongoing Notch2 signaling in FoB cells is further corroborated by lower expression of classical Notch-target genes in FoB cells in Notch2-deficient mice (Fig. 1c) and in FoB cells from mice treated with a Notch2-inhibitor³² in comparison to controls. A relatively high stability of H2B-Venus in immature and FoB cells from CBF:H2B-Venus mice may result in its gradual increase during B cell development, if B cells receive recurring Notch signals or if Notch-IC remains in the nucleus for an extended time. Pre-MZB and MZB cells show the highest H2B-Venus expression, supporting the hypothesis that a strong Notch2 signal promotes MZB cell development. Surprisingly, ~50% of MZB cells exhibit downregulation or complete loss of the H2B-Venus signal. We assume this is linked to their intrinsic ability to proliferate and to differentiate into PCs. In support of this, the H2B-Venus^low MZB subset exhibited higher percentages of Ki-67⁺ cells and elevated levels of Irf4 and Blimp1, indicating increased proliferation and a greater tendency to differentiate into PCs in comparison to H2B-Venus^high MZB and FoB cells. Our hypothesis gains further support from our observation that Lipopolysaccharide (LPS)-treatment of B cells in vitro leads to a substantial loss of H2B-Venus, attributed to both proliferation and differentiation. The decline of H2B-Venus during proliferation is likely due to dilution, while its decrease during differentiation may result from extensive genome re-organization processes in plasmablasts³³, potentially releasing H2B-Venus from the chromatin. Additionally, the H2B-Venus reduction may be associated with a faster degradation of H2B-Venus in plasmablasts and plasma cells compared to other mature B cell populations. Earlier data already demonstrated a variable H2B-Venus protein stability in different tissues, with a very short half-life in certain cases²⁴.

It was striking that both H2B-Venus^high and H2B-Venus^low MZB cells had higher Hes1 levels than FoB cells. In contrast to Notch-target genes, the Notch-reporter functions as a direct indicator of Notch signaling, while Notch-target genes are regulated by various pathways. This distinction may explain the faster downregulation of H2B-Venus in comparison to Hes1 in MZB cells, which did not receive a fresh Notch signal within a certain time.

The fate of the H2B-Venus^low cells remains uncertain. They may undergo cell death and be replaced by newly immigrating MZB cells, or they may receive a new Notch2 signal and cycle back to the H2B-Venus^high state. Notch2 signaling is essential for the maintenance of MZB cells^31,32, but the source of this maintenance signal has yet to be determined. Dll-1 is only expressed within the follicle by Follicular Dendritic cells (FDC) or by Marginal Reticular cells that are located on the marginal zone sinus on the inner side²⁰, suggesting that MZB cells have to migrate into the follicle or at least at the marginal zone sinus border to refresh their Notch2 signal^34,35. Alternatively, the maintenance signal may be conveyed by another ligand, such as Dll-4, which is strongly expressed by endothelial cells from the marginal zone sinus¹⁹.

Furthermore, our data re-confirmed the findings from previous studies demonstrating that strong and sustained Notch signaling prevents the formation of germinal centers^15,25. Bcl6, the main regulator of the GC reaction, actively represses the transcription of Notch2 and other Notch pathway genes^25,36. In this study, we show that vice versa sustained Notch2 signaling inhibits the upregulation of Bcl6 and instead results in upregulation of Irf4. Inhibition of Bcl6 expression may be mediated by the Notch-target gene Hes1, which has been shown recently to repress Bcl6 expression and negatively regulate GC formation³⁷. We found that Irf4^high cells initiate PC differentiation, while Irf4^med cells differentiate into MZB cells. Although Notch2 signaling is downregulated in GCB cells²⁵, our H2B-Venus data suggest that some LZ-B cells may receive a Notch2 signal, likely provided by T follicular helper (T_fh) cells and/or FDCs, which are known to express high levels of Notch ligands Dll-1 and Jagged-1^38,39. Notch2 signaling appears to positively regulate the numbers of IgG1-switched cells in the GC. In vitro studies from the Allman group already suggested a role of Notch signaling in the generation of IgG1-switched cells⁴⁰. Switching occurs mainly in pre-GCB cells⁴¹. Both IgM⁺ and IgG1⁺ B cell blasts differentiate into GCB cells, where they proliferate but do not undergo further class-switch recombination. Around 10 days post-immunization IgG1-switched cells start to be positively selected at the expense of IgM⁺ cells⁴². Our combined data showing that the number of IgG1⁺ cells is similar between N2KO and control mice at day 7, but significantly diminished in N2KO mice at day 14 of the GC reaction, together with the stronger H2B-Venus expression in a portion of LZ-B cells, suggest that Notch2 signaling in LZ-GCB cells may regulate the positive selection and expansion of IgG1⁺ B cells.

We also provide evidence that Notch signaling plays a dual role in PC differentiation by initiating the process, while hindering the terminal differentiation. Thereby, Notch signaling may keep plasmablasts in a proliferative stage by inhibiting the further upregulation of Blimp1^{43,44,45,46,47}. Accordingly, the reduced PCs in N2KO mice at day 7 after immunization may be due to an impaired proliferation of plasmablastic cells in the early immune response. Previously, a role of Notch signaling in PC differentiation was mainly attributed to an interaction between Dll-1 and Notch1^38,48,49. Therefore, the relatively slight effect of Notch2 ablation on PCs could be due to a redundancy of Notch1- and Notch2 signaling in PC differentiation, with the main contribution being at Notch1.

Mathematical modeling of data derived from our immunization studies in antigen-inducible Cre-reporter mice identified a Notch2-dependent differentiation route, in which MZB cell generation during an immune response is primarily sustained by differentiation of activated non-GC FoB cells. Earlier studies already hinted at an antigen-dependent generation of MZB cells^50,51,52,53. However, it is still elusive whether these MZB cells originate from GC or pre-GC B cells. Studies showing that FoB and MZB cells are clonally related in rats⁵⁴ and that memory-like antigen-specific CD80⁺/CD21^high cells in the MZ are less mutated than CD80⁺/CD21^low cells⁵⁵ support our modeling data, which estimated that the antigen-specific CAR⁺CD21^high MZB cells are mainly generated from antigen-activated pre-GCB cells. It is possible that these MZB cells generated during immune responses adopt a memory-like phenotype. In support of this hypothesis are previous studies showing that antigen-induced memory B cells are equally distributed between the MZ and the B cell follicles, even at later time points after immunization⁵⁵ and that memory B cells with a MZB cell expression profile are detected in the spleen after viral infection⁵⁶. On the other hand, we cannot exclude that some of the CAR⁺ MZB cells are directly generated from memory B cells. This would lead to a CAR⁺ MZB cell compartment consisting of both somatically mutated and non-mutated ‘antigen-experienced’ MZB cells, similar to what has been described in humans⁵⁷. Further investigation is required to understand the dynamics of memory B cell dispersion between MZB and FoB cell compartments and to examine differentiation trajectories within antigen-experienced B cell lineages. For instance, tracking dynamics of MZB cells derived from purified donor FoB or GCB cells in congenic recipient mice may be able to untangle their individual contributions to MZB generation during immune responses. We will explore this direction in a future study, but in the scope of this work, we show results from a theoretical simulation of this transplantation experiment (Supplementary Fig. 13), generated using the parameters derived from fitting the branched and linear models to the data from control/CAR mice. The details of the simulation strategy are outlined in the Supplementary Note 3.

In addition, it will be interesting to investigate the mutation rate and clonal composition of the CAR⁺ MZB cell compartment. In humans, the MZB compartment appears to change during aging⁵⁸. So, it will be interesting to explore whether the clonal composition and frequencies of hypermutated MZB cells change during aging in mice and whether the ratio of naïve and antigen-experienced MZB cells can be shifted in favor of antigen-experienced MZB cells through frequent antigenic exposures.

Methods

We confirm that the research compiles with all relevant ethical regulations. All animal experiments were performed in compliance with the German Animal Welfare Law and were approved by the Institutional Committee on Animal Experimentation and the Government of Upper Bavaria.

Mouse models

CBF:H2B-Venus reporter mice²⁴ were purchased from The Jackson Laboratory (JAX stock #020942), backcrossed into the Balb/c background and further maintained in house. Previously described R26/CAG-CAR∆1^StopF²⁷ and N2IC^flSTOP mice¹⁵ were crossbred to the Cγ1-Cre strain²⁶ to generate R26/CAG-CAR∆1^StopF//γ1-Cre (control/CAR) and N2IC^flSTOP//γ1-Cre (N2IC/hCD2) mice, respectively. R26/CAG-CAR∆1^StopF mice carry the human coxsackie/adenovirus receptor CAR in the Rosa26 locus, preceded by a loxP-flanked STOP cassette. Cre recombinase activity results in an expression of a truncated version of the CAR Cre-reporter on the cell surface. In the N2IC^flSTOP mice, the conditional Notch2IC allele is directly coupled to the coding sequence for the human CD2 (hCD2) receptor through an internal ribosomal entry (IRES) site, preceded by a loxP-site-flanked STOP cassette in the Rosa26 locus. Cre recombinase-mediated excision ensures expression of both Notch2IC inside the cell and hCD2 on the cell surface. Lastly, Notch2^fl/fl mice¹⁸ were mated to the R26/CAG-CAR∆1^StopF//γ1-Cre animals to obtain the Notch2^fl/fl//R26/CAG-CAR∆1^StopF//Cγ1-Cre (N2KO//CAR) mice. Upon Cre-mediated recombination, exons 28 and 29 of the Notch2 gene, which code for the C-terminal part of RAM23 and the nuclear localization signal (NLS), and the STOP cassette upstream of the Car gene in the Rosa26 locus are excised. Thereby, Cre-recombined B cells express a truncated non-functional Notch2 receptor (because the following exons 30–33 of Notch2 are out of frame and are thus not translated) and the Cre-reporter CAR on the cell surface. All mice were on a pure Balb/c background, except for CBF:H2B-Venus reporter mice, which were backcrossed after purchase for 5 generations on a Balb/c background. The animals were kept in the laboratory animal rooms under specified pathogen-free (SPF) conditions with a 12/12-h light cycle. The animal rooms are fully air-conditioned, the set-points are set to 20–24 °C temperature and 45–65% humidity according to Annex A of the European Convention 2007/526 EC. The rooms are equipped with individually ventilated cage systems (e.g., Tecniplast Greenline GM 500, 501 cm² floor area, or BioZone-IVC type II, 370 cm² floor area). The maximum stocking densities comply with Annex III of Directive 2010/63/EU. The cages are equipped with laboratory animal bedding (wood fiber/wood chips, e.g., Lignocel Select Fine, SAFE). The animals were given sterile filtered water and a standard rodent diet (e.g., Altromin 1314) ad libitum. Control and experimental mice were bred separately but were kept in the same area of the animal facility. The mice were euthanized either by CO₂ or cervical dislocation. The QuietekTM apparatus was used for euthanasia with CO₂ to ensure that the CO₂ was supplied with the correct flow rate up to the correct filling volume.

Both male and female mice were used for the experiments, data were analyzed without consideration of the animals’ sex. To ensure a more balanced distribution of both male and female animals throughout the study, control and analytical mice were sex-matched in experiments, whenever possible. A detailed overview of the sex and age of mice in each experiment is provided in the Source Data file and in the Figure legends.

Mouse immunizations

To trigger TD-immune responses, 10- to 18- week-old mice were injected intraperitoneally with 100 μg of alum-precipitated 4-hydroxy-3-nitrophenylacetyl (NP)-chicken-gamma-globulin (CCG) (Biosearch Technologies, Novato, CA) in 200 µl sterile PBS (Gibco). Mice were analyzed at indicated time points after antigen injection. Respective unimmunized (UI) animals were taken down on the day of each experiment and were used as a day 0 time point. For the re-boost experiments examining secondary immune response, mice were injected with 50 μg NP-CCG solved in sterile PBS at day 35 after the first immunization and analyzed 7 days later.

Antibodies used in this study

All details about the different antibodies used in Flow cytometry (FACS), Enzyme-linked Immunosorbent Assay (ELISA), Enzyme-linked immunosorbent spot (ELISpot) assay, and histology are listed in Supplementary Table 1.

Cell purification and in vitro cultures

Naïve FoB cells of wildtype Balb/c mice were purified from splenic cell suspension using the MZB and FoB Cell Isolation Kit (Miltenyi Biotech) according to the manufacturer’s instructions. Cells were cultured at a density of 5 × 10⁵ cells/well in flat-bottom 96-well plates in RPMI-1640 medium (Gibco), supplemented with 10% fetal calf serum (FCS, PAA Cell Culture Company), 1% L-glutamine, 1% non-essential amino acids, 1% sodium pyruvate and 50 µM β-mercaptoethanol. Except for FCS, all supplements were purchased from Gibco. An agonistic antibody against CD40 (2.5 μg/ml; eBioscience HM40-3) and an antibody specific for IgM (15 μg/ml; AffiniPure F(ab′)2 goat-anti-mouse IgM, μ-chain, Dianova) were used as a combined stimulus for B cells, which were harvested at the indicated time points after addition of the stimulus and analyzed by flow cytometry after staining with an anti-Notch2-PE antibody (BioLegend). Dead cells were discriminated from living cells using a fixable live/dead cell staining kit (Invitrogen). For the plasmablast-differentiation assay, naïve B cells from CBF:H2B-Venus Notch-reporter mice were isolated from splenic cell suspension with the CD43 depletion B cell Isolation Kit (Miltenyi) according to the manufacturer’s instructions. To monitor cell division, B cells were stained with CellTrace Far Red (Invitrogen) according to the manufacturer’s protocol prior to the flow cytometry stainings. Cells were cultured at a density of 5 × 10⁵ cells/well in flat-bottom 96-well plates as described above and were stimulated with 50 µg/ml LPS (Sigma). Cells were harvested at the indicated time points and analyzed by flow cytometry after staining with anti-CD138-BV421 and anti-B220-PerCP antibodies. Dead cells were discriminated from living cells using a fixable blue live/dead cell staining kit (Invitrogen).

Histology

Murine splenic tissues were embedded in O.C.T. compound (VWR Chemicals, USA), snap frozen, and stored at −20 °C. Tissues were sliced with a cryostat with 7 µm thickness, mounted on glass slides, and stored at −80 °C for immunohistochemical (IHC) or immunofluorescence (IF) analysis.

For the chromogenic IHC, splenic sections were fixed with ice-cold acetone for 10 min, washed with PBS and blocked with 1% BSA, 5% goat serum in PBS. Samples were blocked using the Avidin/Biotin blocking kit (Vector), following the manufacturer protocol. Detection of hCD2-expression was done by staining with a biotinylated mouse-anti-hCD2 antibody and a rabbit-anti-Laminin antibody overnight at 4 °C in 1% PBS/BSA. Secondary antibody staining was done for 1 h at room temperature with streptavidin-coupled alkaline phosphatase and peroxidase-coupled anti-rabbit IgG in 1% PBS/BSA. Enzymatic chromogenic development was done with the AEC substrate and Blue AP substrate kits (Vector), following the manufacturer manual. Slides were embedded in Kaiser’s Gelatin (Carl Roth) and analyzed using an Axioscope (Zeiss) microscope with an AxioCam MRc5 digital camera (Carl Zeiss GmbH)¹⁴.

For the IF staining, sections were fixed for 10 min with PBS-diluted 3% Formaldehyde ph7 (Histofix, Carl Roth), rinsed in PBS, and rehydrated for 8 min in PBS + 50 mM NH₄Cl. For the visualization of the MZ, and B- cell zone, sections were blocked with 1% Bovine serum albumin (BSA), 5% rat serum, 5% chicken serum in PBS for 30 min, followed by Avidin/Biotin blocking (Vector) according to the manufacturer’s protocol. Sections were stained with the primary antibodies (goat-anti-mouse IgM, anti-Moma1-Biotin) and the secondary antibody (chicken anti-goat IgG AF647) as well as Streptavidin AlexaFluor594. The antibodies and Streptavidin were incubated for 1 h at room temperature in 0.5% BSA/PBS.

To detect plasmablasts and PCs, sections were permeabilized and blocked for 20 min with 0.3% Triton X, 1% BSA, 5% goat serum in PBS. Primary (rat-anti-mouse Irf4, rabbit-anti-mouse Laminin) antibodies and secondary antibodies (goat-anti-rat AlexaFluor488, goat-anti-rat AlexaFluor647, goat-anti-rabbit Cy3) were incubated in 1% BSA/PBS for 1 h at room temperature. Fluorophore-coupled antibodies (anti-mouse B220-APC) were incubated for 2 h at room temperature, where appropriate. For the detection of germinal center (GC) structures, sections were blocked with 1% BSA, 5% rat serum in PBS for 30 min. Rat-anti-mouse GL7-FITC or Rat-anti-mouse GL7-APC antibody was incubated overnight at 4 °C. Directly coupled antibodies (Thy1.2-Biotin, Moma1-Biotin, and B220-APC) and Streptavidin AlexaFluor594 were incubated for 1 h at room temperature in 1% BSA/PBS. Slides were embedded in ProLong Glass Antifade (Invitrogen). Images were acquired on a TCS SP5 II confocal microscope (Leica), equipped with the Leica Application Suite X (LAS X) Software Version 4, and picture stacks were composed in ImageJ Version 1.50e¹⁴.

ELISA

NP-specific antibody titers in the sera were measured by ELISA, following the procedure described by^59,60. NUNC plates (Nunc) were coated with NP3-BSA or NP14–BSA (10 mg/ml, Biosearch Technologies) in carbonate buffer (0.1 M NaHCO₃, pH = 9.5) and incubated overnight at 4 °C. Plates were washed with PBS, followed by blocking for 2 h with blocking buffer (1% milk powder in PBS for IgG1 ELISA, 5% milk powder in PBS for IgM ELISA). Serum samples were distributed in a 1:2 serial dilution in blocking buffer over 8 wells, starting with 100 µl/well of either 1:10 (IgM) or 1:100 (IgG1) diluted serum in the top row of the plate. The sera were incubated for 1 h at room temperature, followed by washing with PBS. For IgM ELISA, a directly coupled anti-IgM-horseradish peroxidase (HRP) antibody was diluted in blocking buffer and incubated for 1 h at room temperature. Afterwards the HRP signal was detected as described below. For IgG1 ELISA, a biotinylated rat-anti-mouse- IgG1 antibody was diluted in blocking buffer and incubated for 30 min at room temperature. Subsequently, plates were washed with PBS and incubated with streptavidin horseradish peroxidase Avidin D (Vector) diluted in blocking buffer. Following three washing steps with PBS and PBS-T, detection of the HRP signal was done with substrate buffer (1 tablet o-Phenylenediamine (Sigma, P-7288) + 35 mL substrate buffer (0.1 M citric acid, 0.1 M Tris (Sigma)) supplemented with 21 µl H₂O₂ (Sigma)). The absorbance was determined with a microplate ELISA reader (Photometer Sunrise RC, Tecan) at an optical density (OD) at 405 nm and data were acquired with the Infinite F200 PRO i-control software, Version 3.37. To correctly quantify the measured titers and compare different independent assays, internal standards consisting of pooled serums from six to eight NP-CGG immunized mice were used. A calibration curve was produced using the absorbance values measured in the serum standards, which was then used as a guideline to calculate the NP-titers in sera of immunized control/CAR and N2IC/hCD2 animals using Excel 2010. All NP-specific titers are given as log10 relative units.

ELISpot

For antigen-specific ELISpot with splenic or BM cells, 96-well ELISpot plates (Millipore) were coated overnight at 4 °C with NP3– or NP14–BSA (Biosearch Technologies), diluted in carbonate buffer. Plates were washed with PBS and blocked for 3 h at 37 °C with 10% FCS/B cell medium (BCM). 5 × 10⁵ cells were seeded per well and incubated for 24 h at 37 °C. Plates were washed with PBS-T (PBS with 0.025% Tween20). Biotinylated antibodies against the detecting isotype were diluted in 1% BSA/PBS and incubated for 2 h at 37 °C. After washing with PBS-T, streptavidin horseradish peroxidase Avidin D (Vector) in 1% BSA/PBS was added and incubated for 45 min at room temperature. For the development of spots, a developing buffer was prepared as follows: one tablet of each 3,3′- Diaminobenzidin peroxidase-substrate (0.7 mg/ml, Sigma-Aldrich, gold and silver tablets) was dissolved in 5 ml distilled water, the solution was pooled and filtered. 50 µl/well of this buffer was incubated for 8 min (splenocytes) or 12 min (BM cells). The reaction was stopped by washing with distilled water. Spots were visualized and counted with the ImmunoSpot Series 5 UV Analyzer (equipped with software version 5.0, CTL Europe).

Flow cytometry

Single cell suspensions were prepared from the spleen and BM. Surface staining of lymphocytes with the corresponding antibodies in MACS buffer (Miltenyi) was performed on ice for 25 min. To omit dead cells from the analysis, cells were stained for 20 min on ice with LIVE/DEAD Fixable Blue Dead Cell Stain Kit (Invitrogen) prior to surface antibody staining. For intracellular FACS staining, cells were fixed with 2% paraformaldehyde (1:2 PBS-diluted Histofix, Carl Roth) for 10 min at room temperature and permeabilized in ice-cold 100% methanol for 10 min on ice. Cells were incubated for 1 h at room temperature with the corresponding antibodies. Dead cells were excluded by staining for 5 min on ice with LIVE/DEAD Fixable Blue Dead Cell Stain Kit (Invitrogen) before fixation. Cytometry analysis was performed on a LSRII FACS Fortessa (BD Biosciences) coupled to the BD FACS DIVA Software V8.0.1. Results were evaluated using FlowJo (v9 and v10). Representative flow cytometry gating strategies for surface and intracellular cell stainings to distinguish B lymphocytes and their subpopulations, reporter⁺ and reporter⁻ cells, and plasmablasts and plasma cells among total live lymphocytes, can be found in Supplementary Fig. 14 a–e.

Statistics & reproducibility

Model ranking and correlation analysis of CAR expression dynamics between FoB, MZB and GCB cells was performed in R. All other statistical analyses, including testing for distributions and equal variance, calculations of means and standard deviation (SD), determining p-values of 1-way or 2-way ANOVA with multiple comparisons tests and paired value tests were done with GraphPad Prism (version 9.5.1). P-values of less than 0.05 were considered significant. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. Error bars in all figures define means ± SD. Sample sizes were determined based on previous experiments and preliminary data. No statistical methods were used to predetermine samples sizes. Our goal was to analyze at least 6 animals per group. In cases in which differences were pronounced enough to achieve significance with smaller sample sizes, adjustments were made accordingly. In general, we analyzed a minimum of 4 animals per genotype and treatment. Only in a few cases we analyzed only three animals. Most of the experiments with animal numbers ≤3 were performed during the revision phase, where we had only a limited number of mice available and were therefore unable to perform all the experiments requested by the reviewers with higher numbers of mice. Despite the smaller number of mice, our analysis still yielded significant results or, at least, clear trends, as the biological differences between the samples were sufficiently robust. Sample sizes with only two animals were excluded from statistical analyses. Control and analytical mice were always analyzed in parallel and predominantly used in an age-matched manner, when possible. The experiments were repeated with different individual animals in at least two independent immunization or cell preparation rounds. No data were excluded from the analyses. The experiments were not randomized. The investigators were not blinded to allocations during experiments and outcome assessment. Sample sizes and the chosen statistical tests are indicated in each figure legend.

Mathematical models of B cell dynamics during a TD-immune response

To model the emergence of CAR⁺ MZB and GCB cells post-immunization in control/CAR and N2KO//CAR mice, we developed an array of system of ordinary differential equations (ODE) that explored diverse pathways of B cell differentiation and logistics of their maintenance post activation. We observe no significant differences in the counts of CAR⁺ MZB and GCB cells up to day 4 post-immunization and therefore anchor the initial condition ${t}_{0}=4$ days in all our models. Model schematics are described in Fig. 8.

Mechanisms of maintenance of GCB cells

We assumed that all GCB cells are derived from antigen-activated FoB cells with a per capita rate α. We explored modeling possibilities in which either the influx of cells into the GC compartment or their loss from the GC pool are varying with time post-immunization.

Neutral dynamics

This is the simplest model, which assumes constant birth-loss. The rates of influx into the GC (α) and net loss of cells from the GC (δ), remain constant with time. In all our models, the net loss rate δ is the balance between production of new GCB cells by cell division and their ‘true loss’ by death and differentiation.

Time-dependent loss of GCB cells

This model assumes that the net loss rate of GCB cells varies with time post-immunization. We explored the sigmoid form of δ(t), as shown in Eq. 1 below. The influx of antigen-activated FoB cells into the GC compartment α is assumed to be constant in this model. Parameters ν1 and δ₀ are unknowns in this model and are estimated from model fits, along with α.

$$\delta \left(t\right) \,=\,\frac{{\delta }_{0}}{1+{e}^{\nu 1{\left(t-{t}_{0}\right)}^{2}}}$$

(1)

Time-dependent recruitment of FoB cells into GC

In this model, the per capita rate of recruitment of FoB cells into the GC varies with time α(t) and is defined as,

$$\alpha \left(t\right) \,=\,\frac{{\alpha }_{0}}{1+{e}^{\nu 2{\left(t-{t}_{0}\right)}^{2}}}$$

(2)

The net loss rate of GCB cells δ is constant in this model, estimated from the model fits along with ν2 and α₀.

Loss of CAR⁺ MZB cells

We assumed a constant per capita rate of net loss λ for CAR⁺ MZB cells. It has been shown that persistent Notch2 signaling is required for B cells to maintain their MZ status^31,32. Therefore, the net loss rate of MZB cells reflects the balance between their self-renewal (division) and their true loss via death, differentiation and deprivation of Notch2-dependent signals. In N2KO//CAR mice, antigen-activation of B cells leads to deletion of Notch2 and thus increases the propensity of their loss by deprivation of Notch2-mediated signals. Therefore, we assume different loss rates of CAR⁺ MZB cells in control/CAR mice and N2KO//CAR mice viz. λ and λ'.

Mechanisms of generation of MZB cells

We explored three different models — branched, linear, and null — to explain the emergence of CAR⁺ MZB cells after immunization.

Branched model

In this model, the antigen-dependent activation of FoB cells induces a branch point in the B cell differentiation pathway, where some cells participate in GC reactions, while others acquire a MZB cell phenotype (Fig. 8c). Since the dynamics of CAR expression on FoB cells are unclear, we considered that all antigen-activated FoB (F) cells (irrespective of their CAR expression status) differentiate into CAR⁺ MZB cells (M) with a per capita rate μ. The dynamics for CAR⁺ GCB (G) cells are identical between control/CAR and N2KO//CAR mice. The general form ODE system depicting the branched pathway is defined as,

$$\dot{G} \, =\,\alpha \left(t\right)\,F\left(t\right)-\delta \left(t\right)\,G\left(t\right)$$

$$\dot{M} \,=\,\mu \left(t\right)\,F\left(t\right)+\beta \left(t\right)-\lambda \left(t\right)\,M\left(t\right)$$

$$\dot{{M}_{N2}}=\beta \left(t\right)-{\lambda }^{{\prime} }(t)\,{M}_{N2}\left(t\right)$$

(3)

where ${M}_{N2}$ represents CAR⁺ MZB cells in N2KO//CAR mice.

Linear model

This model follows a linear path in which antigen-activated FoB cells differentiate into GCB cells, which then subsequently give rise to CAR-expressing MZB cells (Fig. 8d). The general form of ODE system of the linear pathway is given as,

$$\dot{G}=\alpha \left(t\right)F\left(t\right)-\delta \left(t\right)G\left(t\right)$$

$$\dot{M}=\mu \left(t\right)G\left(t\right)+\beta \left(t\right)-\lambda \left(t\right)M\left(t\right)$$

$$\dot{{M}_{N2}}=\beta \left(t\right)-{\lambda }^{{\prime} }\left(t\right){M}_{N2}\left(t\right)$$

(4)

Null model

This model assumes that the CAR⁺ fraction in MZB cells is purely derived from activation of pre-existing MZB cells and is maintained by activation-induced continuous new deletion of the stop cassette in CAR⁻ MZB cells and by self-renewal of newly generated CAR⁺ MZB cells (Fig. 8e).

$$\dot{G}=\alpha \left(t\right)F\left(t\right)-\delta \left(t\right)G\left(t\right)$$

$$\dot{M}=\beta \left(t\right)-\lambda \left(t\right)M\left(t\right)$$

$$\dot{{M}_{N2}}=\beta \left(t\right)-{\lambda }^{{\prime} }\left(t\right){M}_{N2}\left(t\right)$$

(5)

We assumed that the rate of activation of CAR^- MZB to CAR⁺ MZB cells (β) varies with time as shown below.

$$\beta \left(t\right)=\frac{{\beta }_{0}}{1+{e}^{\nu 3{\left(t-{t}_{0}\right)}^{2}}}$$

(6)

Lastly, we explored the possibility that the rate of influx of antigen-activated FoB cells to CAR⁺ MZB cells (defined as μ) may vary with time, likely due to the dependence on antigen availability in the cellular environment. We used the same form as shown in Eq. 6 to describe the time dependence in μ(t).

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

All source data are provided with this paper as a Source Data file. Previously generated RNA sequencing/microarray data, which were analyzed in this study, have been deposited in ENA under the accession code PRJEB35207 (Supplementary Fig. 1c) or at Zenodo under the https://doi.org/10.5281/zenodo.10491127 (https://zenodo.org/records/10491127) (Fig. 1c)⁶¹. Graphs from Supplementary Fig. 8b were produced with data derived from the ImmGen.org databrowser “Gene Skyline – RNAseq” (http://rstats.immgen.org/Skyline/skyline.html)⁶². Source data are provided with this paper.

Code availability

Code defining all our models and scripts defining our plotting and statistical routines are available at a Github repository integrated in the Zenodo records (https://zenodo.org/records/10475721)⁶³. For high reproducibility, we provide definitions of the prior distributions of the unknown variables and the descriptor functions used to evaluate fixed variables, in our model code.

References

Arnon, T. I., Horton, R. M., Grigorova, I. L. & Cyster, J. G. Visualization of splenic marginal zone B-cell shuttling and follicular B-cell egress. Nature 493, 684–688 (2013).
Article ADS CAS PubMed Google Scholar
Martin, F., Oliver, A. M. & Kearney, J. F. Marginal zone and B1 B cells unite in the early response against T-independent blood-borne particulate antigens. Immunity 14, 617–629 (2001).
Article CAS PubMed Google Scholar
Cerutti, A., Cols, M. & Puga, I. Marginal zone B cells: virtues of innate-like antibody-producing lymphocytes. Nat. Rev. Immunol. 13, 118–132 (2013).
Article CAS PubMed PubMed Central Google Scholar
Shlomchik, M. J., Luo, W. & Weisel, F. Linking signaling and selection in the germinal center. Immunol. Rev. 288, 49–63 (2019).
Article CAS PubMed PubMed Central Google Scholar
Victora, G. D. & Nussenzweig, M. C. Germinal centers. Annu Rev. Immunol. 40, 413–442 (2022).
Article CAS PubMed Google Scholar
Pillai, S. & Cariappa, A. The follicular versus marginal zone B lymphocyte cell fate decision. Nat. Rev. Immunol. 9, 767–777 (2009).
Article CAS PubMed Google Scholar
Allman, D. et al. Resolution of three nonproliferative immature splenic B cell subsets reveals multiple selection points during peripheral B cell maturation. J. Immunol. 167, 6834–6840 (2001).
Article CAS PubMed Google Scholar
Martin, F. & Kearney, J. F. Positive selection from newly formed to marginal zone B cells depends on the rate of clonal production, CD19, and btk. Immunity 12, 39–49 (2000).
Article CAS PubMed Google Scholar
Agenes, F. & Freitas, A. A. Transfer of small resting B cells into immunodeficient hosts results in the selection of a self-renewing activated B cell population. J. Exp. Med. 189, 319–330 (1999).
Article CAS PubMed PubMed Central Google Scholar
Cariappa, A. et al. The recirculating B cell pool contains two functionally distinct, long-lived, posttransitional, follicular B cell populations. J. Immunol. 179, 2270–2281 (2007).
Article CAS PubMed Google Scholar
Gomez Atria, D. et al. Stromal Notch ligands foster lymphopenia-driven functional plasticity and homeostatic proliferation of naive B cells. J. Clin. Invest. 132, e158885 (2022).
Article PubMed PubMed Central Google Scholar
Srivastava, B., Quinn, W. J. 3rd, Hazard, K., Erikson, J. & Allman, D. Characterization of marginal zone B cell precursors. J. Exp. Med. 202, 1225–1234 (2005).
Article CAS PubMed PubMed Central Google Scholar
Vinuesa, C. G. et al. Recirculating and germinal center B cells differentiate into cells responsive to polysaccharide antigens. Eur. J. Immunol. 33, 297–305 (2003).
Article CAS PubMed Google Scholar
Lechner, M. et al. Notch2-mediated plasticity between marginal zone and follicular B cells. Nat. Commun. 12, 1111 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Hampel, F. et al. CD19-independent instruction of murine marginal zone B-cell development by constitutive Notch2 signaling. Blood 118, 6321–6331 (2011).
Article CAS PubMed Google Scholar
Hozumi, K. et al. Delta-like 1 is necessary for the generation of marginal zone B cells but not T cells in vivo. Nat. Immunol. 5, 638–644 (2004).
Article CAS PubMed Google Scholar
Saito, T. et al. Notch2 is preferentially expressed in mature B cells and indispensable for marginal zone B lineage development. Immunity 18, 675–685 (2003).
Article CAS PubMed Google Scholar
Besseyrias, V. et al. Hierarchy of Notch-Delta interactions promoting T cell lineage commitment and maturation. J. Exp. Med. 204, 331–343 (2007).
Article PubMed PubMed Central Google Scholar
Tan, J. B. et al. Lunatic and manic fringe cooperatively enhance marginal zone B cell precursor competition for delta-like 1 in splenic endothelial niches. Immunity 30, 254–263 (2009).
Article PubMed Google Scholar
Fasnacht, N. et al. Specific fibroblastic niches in secondary lymphoid organs orchestrate distinct Notch-regulated immune responses. J. Exp. Med. 211, 2265–2279 (2014).
Article PubMed PubMed Central Google Scholar
Lopes-Carvalho, T. & Kearney, J. F. Development and selection of marginal zone B cells. Immunol. Rev. 197, 192–205 (2004).
Article PubMed Google Scholar
Hammad, H. et al. Transitional B cells commit to marginal zone B cell fate by Taok3-mediated surface expression of ADAM10. Nat. Immunol. 18, 313–320 (2017).
Article CAS PubMed Google Scholar
Cox, E. M. et al. AKT activity orchestrates marginal zone B cell development in mice and humans. Cell Rep. 42, 112378 (2023).
Article CAS PubMed Google Scholar
Nowotschin, S., Xenopoulos, P., Schrode, N. & Hadjantonakis, A. K. A bright single-cell resolution live imaging reporter of Notch signaling in the mouse. BMC Dev. Biol. 13, 15 (2013).
Article PubMed PubMed Central Google Scholar
Valls, E. et al. BCL6 antagonizes NOTCH2 to maintain survival of human follicular lymphoma cells. Cancer Discov. 7, 506–521 (2017).
Article CAS PubMed PubMed Central Google Scholar
Casola, S. et al. Tracking germinal center B cells expressing germ-line immunoglobulin gamma1 transcripts by conditional gene targeting. Proc. Natl Acad. Sci. USA 103, 7396–7401 (2006).
Article ADS CAS PubMed PubMed Central Google Scholar
Heger, K. et al. A novel Cre recombinase reporter mouse strain facilitates selective and efficient infection of primary immune cells with adenoviral vectors. Eur. J. Immunol. 45, 1614–1620 (2015).
Article CAS PubMed Google Scholar
Basso, K. & Dalla-Favera, R. Roles of BCL6 in normal and transformed germinal center B cells. Immunol. Rev. 247, 172–183 (2012).
Article PubMed Google Scholar
Ochiai, K. et al. Transcriptional regulation of germinal center B and plasma cell fates by dynamical control of IRF4. Immunity 38, 918–929 (2013).
Article CAS PubMed PubMed Central Google Scholar
Sciammas, R. et al. An incoherent regulatory network architecture that orchestrates B cell diversification in response to antigen signaling. Mol. Syst. Biol. 7, 495 (2011).
Article PubMed PubMed Central Google Scholar
Simonetti, G. et al. IRF4 controls the positioning of mature B cells in the lymphoid microenvironments by regulating NOTCH2 expression and activity. J. Exp. Med. 210, 2887–2902 (2013).
Article CAS PubMed PubMed Central Google Scholar
Gaudette, B. T. et al. Resting innate-like B cells leverage sustained Notch2/mTORC1 signaling to achieve rapid and mitosis-independent plasma cell differentiation. J. Clin. Invest. 131, e151975 (2021).
Article CAS PubMed PubMed Central Google Scholar
Chan, W. F. et al. Pre-mitotic genome re-organisation bookends the B cell differentiation process. Nat. Commun. 12, 1344 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Cinamon, G. et al. Sphingosine 1-phosphate receptor 1 promotes B cell localization in the splenic marginal zone. Nat. Immunol. 5, 713–720 (2004).
Article CAS PubMed Google Scholar
Cinamon, G., Zachariah, M. A., Lam, O. M., Foss, F. W. Jr. & Cyster, J. G. Follicular shuttling of marginal zone B cells facilitates antigen transport. Nat. Immunol. 9, 54–62 (2008).
Article CAS PubMed Google Scholar
Sakano, D. et al. BCL6 canalizes Notch-dependent transcription, excluding Mastermind-like1 from selected target genes during left-right patterning. Dev. Cell 18, 450–462 (2010).
Article CAS PubMed PubMed Central Google Scholar
Shao, X., Liu, X. & Qi, H. A role for Hes1 in constraining germinal center B cell formation. Cell Insight 2, 100078 (2023).
Article PubMed PubMed Central Google Scholar
Santos, M. A. et al. Notch1 engagement by Delta-like-1 promotes differentiation of B lymphocytes to antibody-secreting cells. Proc. Natl Acad. Sci. USA 104, 15454–15459 (2007).
Article ADS CAS PubMed PubMed Central Google Scholar
Yoon, S. O., Zhang, X., Berner, P., Blom, B. & Choi, Y. S. Notch ligands expressed by follicular dendritic cells protect germinal center B cells from apoptosis. J. Immunol. 183, 352–358 (2009).
Article CAS PubMed Google Scholar
Thomas, M. et al. Notch activity synergizes with B-cell-receptor and CD40 signaling to enhance B-cell activation. Blood 109, 3342–3350 (2007).
Article CAS PubMed Google Scholar
Roco, J. A. et al. Class-switch recombination occurs infrequently in germinal centers. Immunity 51, 337–50.e7 (2019).
Article CAS PubMed PubMed Central Google Scholar
Sundling, C. et al. Positive selection of IgG(+) over IgM(+) B cells in the germinal center reaction. Immunity 54, 988–1001.e5 (2021).
Article CAS PubMed Google Scholar
Kallies, A. et al. Plasma cell ontogeny defined by quantitative changes in blimp-1 expression. J. Exp. Med. 200, 967–977 (2004).
Article CAS PubMed PubMed Central Google Scholar
Shapiro-Shelef, M. et al. Blimp-1 is required for the formation of immunoglobulin secreting plasma cells and pre-plasma memory B cells. Immunity 19, 607–620 (2003).
Article CAS PubMed Google Scholar
Shaffer, A. L. et al. Blimp-1 orchestrates plasma cell differentiation by extinguishing the mature B cell gene expression program. Immunity 17, 51–62 (2002).
Article CAS PubMed Google Scholar
Calado, D. P. et al. Constitutive canonical NF-kappaB activation cooperates with disruption of BLIMP1 in the pathogenesis of activated B cell-like diffuse large cell lymphoma. Cancer Cell 18, 580–589 (2010).
Article CAS PubMed PubMed Central Google Scholar
Mandelbaum, J. et al. BLIMP1 is a tumor suppressor gene frequently disrupted in activated B cell-like diffuse large B cell lymphoma. Cancer Cell 18, 568–579 (2010).
Article CAS PubMed PubMed Central Google Scholar
Kang, J. A., Kim, W. S. & Park, S. G. Notch1 is an important mediator for enhancing of B-cell activation and antibody secretion by Notch ligand. Immunology 143, 550–559 (2014).
Article CAS PubMed PubMed Central Google Scholar
Zhu, G. et al. Both Notch1 and its ligands in B cells promote antibody production. Mol. Immunol. 91, 17–23 (2017).
Article CAS PubMed Google Scholar
Liu, Y. J., Oldfield, S. & MacLennan, I. C. Memory B cells in T cell-dependent antibody responses colonize the splenic marginal zones. Eur. J. Immunol. 18, 355–362 (1988).
Article CAS PubMed Google Scholar
Pape, K. A. et al. Visualization of the genesis and fate of isotype-switched B cells during a primary immune response. J. Exp. Med. 197, 1677–1687 (2003).
Article CAS PubMed PubMed Central Google Scholar
Shih, T. A., Meffre, E., Roederer, M. & Nussenzweig, M. C. Role of BCR affinity in T cell dependent antibody responses in vivo. Nat. Immunol. 3, 570–575 (2002).
Article PubMed Google Scholar
White, H. N. & Meng, Q. H. Recruitment of a distinct but related set of VH sequences into the murine CD21hi/CD23- marginal zone B cell repertoire to that seen in the class-switched antibody response. J. Immunol. 188, 287–293 (2012).
Article CAS PubMed Google Scholar
Hendricks, J. et al. The formation of mutated IgM memory B cells in rat splenic marginal zones is an antigen dependent process. PLoS ONE 14, e0220933 (2019).
Article CAS PubMed PubMed Central Google Scholar
Anderson, S. M., Tomayko, M. M., Ahuja, A., Haberman, A. M. & Shlomchik, M. J. New markers for murine memory B cells that define mutated and unmutated subsets. J. Exp. Med. 204, 2103–2114 (2007).
Article CAS PubMed PubMed Central Google Scholar
Gregoire, C. et al. Viral infection engenders bona fide and bystander subsets of lung-resident memory B cells through a permissive mechanism. Immunity 55, 1216–33.e9 (2022).
Article CAS PubMed PubMed Central Google Scholar
Siu, J. H. Y. et al. Two subsets of human marginal zone B cells resolved by global analysis of lymphoid tissues and blood. Sci. Immunol. 7, eabm9060 (2022).
Article CAS PubMed Google Scholar
Kibler, A. et al. Systematic memory B cell archiving and random display shape the human splenic marginal zone throughout life. J. Exp. Med. 218, e20201952 (2021).
Article CAS PubMed PubMed Central Google Scholar
Scheffler, L. et al. ERK phosphorylation is RAF independent in naive and activated B cells but RAF dependent in plasma cell differentiation. Sci. Signal. 14, eabc1648 (2021).
Article CAS PubMed Google Scholar
Sperling, S. et al. Chronic CD30 signaling in B cells results in lymphomagenesis by driving the expansion of plasmablasts and B1 cells. Blood 133, 2597–2609 (2019).
Article CAS PubMed Google Scholar
Strobl, L. J. Differently expressed genes between Follicular B cells from control (CD19-Cre) and conditional Notch2-KO (Notch2fl/fl x CD19-Cre) mice [Internet]. https://zenodo.org/records/10491127 (2024).
Heng, T. S. & Painter, M. W., Immunological Genome Project C. The Immunological Genome Project: networks of gene expression in immune cells. Nat. Immunol. 9, 1091–1094 (2008).
Article CAS PubMed Google Scholar
Rane, S. Notch2 controls developmental fate choices between germinal center and marginal zone B cells upon immunization [Internet]. https://zenodo.org/records/10475721 (2024).
Ehrenberg, S. Notch2 signalling in B cell activation, immune response and lymphomagenesis [Doctoral thesis] (Ludwig-Maximilians-Universität München, 2014).

Download references

Acknowledgements

This work was supported by the Deutsche Forschungsgemeinschaft (DFG ZI1382/4-1), U.Z.-S. and the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under award number R01AI170965, S.R. We thank the animal facility of the Helmholtz Center and our animal caretaker team for the excellent housing of the mice and Krisztina Zeller for excellent technical help with murine DNA preparation and genotyping. This work benefitted from data assembled by the ImmGen consortium.

Funding

Open Access funding enabled and organized by Projekt DEAL.

Author information

These authors contributed equally: Sanket Rane, Ursula Zimber-Strobl, Lothar J. Strobl.

Authors and Affiliations

Research Unit Gene Vectors, Research Group B Cell Development and Activation, Helmholtz Zentrum München, German Research Center for Environmental Health, Feodor-Lynen-Str. 21, D-81377, Munich, Germany
Tea Babushku, Markus Lechner, Stefanie Ehrenberg, Ursula Zimber-Strobl & Lothar J. Strobl
TranslaTUM, Center for Translational Cancer Research, Technical University of Munich, Einsteinstraße 25, D-81675, Munich, Germany
Tea Babushku & Marc Schmidt-Supprian
Institute of Asthma and Allergy Prevention, Helmholtz Zentrum München, German Research Center for Environmental Health, Feodor-Lynen-Str. 21, D-81377, Munich, Germany
Ursula Rambold
Department of Pathology and Cell Biology, Columbia University Irving Medical Center, 630 West 168th Street, New York, NY, 10032, USA
Andrew J. Yates
Irving Institute for Cancer Dynamics, Columbia University, 1190 Amsterdam Ave, New York, 10027, USA
Sanket Rane
Institute of Lung Health and Immunity (LHI), Helmholtz Munich, Comprehensive Pneumology Center (CPC-M), Member of the German Center for Lung Research (DZL), Ingolstädter Landstraße 1, 85764, Neuherberg, Germany
Ursula Zimber-Strobl & Lothar J. Strobl

Authors

Tea Babushku
View author publications
You can also search for this author in PubMed Google Scholar
Markus Lechner
View author publications
You can also search for this author in PubMed Google Scholar
Stefanie Ehrenberg
View author publications
You can also search for this author in PubMed Google Scholar
Ursula Rambold
View author publications
You can also search for this author in PubMed Google Scholar
Marc Schmidt-Supprian
View author publications
You can also search for this author in PubMed Google Scholar
Andrew J. Yates
View author publications
You can also search for this author in PubMed Google Scholar
Sanket Rane
View author publications
You can also search for this author in PubMed Google Scholar
Ursula Zimber-Strobl
View author publications
You can also search for this author in PubMed Google Scholar
Lothar J. Strobl
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

T.B. performed most experiments and analyzed data. M.L., S.E., UR performed experiments. A.J.Y. contributed to mathematical modeling and manuscript editing. M.S.-S. contributed to data interpretation and provided material. S.R. developed and performed mathematical modeling. S.R., L.J.S. and U.Z.-S. designed research and contributed to data interpretation and manuscript preparation. T.B. and U.Z.-S. wrote the manuscript. L.J.S. and T.B. prepared the figures from the experimental part and calculated the statistics. S.R. wrote the mathematical modeling part and prepared the figures. S.R., L.J.S. and U.Z.-S. contributed equally to supervising the study (L.J.S. and U.Z-S.: experimental part; S.R. mathematical modeling part). All authors read and approved the final paper.

Corresponding author

Correspondence to Ursula Zimber-Strobl.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Babushku, T., Lechner, M., Ehrenberg, S. et al. Notch2 controls developmental fate choices between germinal center and marginal zone B cells upon immunization. Nat Commun 15, 1960 (2024). https://doi.org/10.1038/s41467-024-46024-1

Download citation

Received: 08 November 2022
Accepted: 12 February 2024
Published: 04 March 2024
DOI: https://doi.org/10.1038/s41467-024-46024-1

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.