During B cell development in the bone marrow, a tightly controlled series of events ensures that functional B cells are produced that can recognize an almost infinite number of foreign antigens while limiting the escape of autoreactive B cells into the periphery [reviewed in (1, 2)]. The immunoglobulin (Ig) genes are rearranged sequentially and tested during this process, beginning with the Ig heavy chain at the pro-B cell stage. Successful pairing of the expressed heavy chain with the surrogate light chain promotes a proliferative expansion at the large pre-B cell stage. IL-7 also promotes expansion of pro-B and large pre-B cells. At the small pre-B cell stage, cells exit the cell cycle and rearrange the Ig light chain genes. The transition between the large and small pre-B stage is driven by a particular mode of pre-BCR signaling dependent on the adaptor molecule BLNK (also known as SLP-65) (3, 4), loss of IL-7 signaling (3, 5), and CXCL12/CXCR4 responses (5, 6). After light chain rearrangement, cells transition to the immature B stage where they express the BCR on the surface and are tested for autoreactivity by central tolerance mechanisms.

These developmental transitions are coordinated by a complex network of transcription factors (3). Among these, the related factors IRF4 and IRF8 play a critical role. Previous studies have shown these to be redundantly required for early B cell developmental events. Mice lacking both IRF4 and IRF8 have an almost complete block at the large pre-B stage, while single knockouts have relatively normal numbers of peripheral B cells (7). IRF4/IRF8 double deficient pre-B cells proliferate extensively in response to IL-7, fail to rearrange Ig light chain genes due to impaired germline transcription, and have reduced expression of CXCR4 (5, 6, 8). In vitro, reconstitution of IRF4/IRF8 double knockout pre-B cells with either IRF4 or IRF8 allowed the cells to undergo Ig light chain rearrangement upon IL-7 withdrawal (8). This indicates that both IRF4 and IRF8 are both capable of exerting this function and act in a B cell intrinsic manner to do so.

While IRF4 and IRF8 work together in early B lymphopoiesis, IRF4 has a unique role in the periphery in terms of both B cell subset distribution and differentiation. Two major subsets of mature B cells are marginal zone (MZ) B cells, which respond rapidly to T-independent antigens, and follicular B cells, which are activated in the context of T cell help to form germinal centers and produce high affinity, long lived class switched humoral responses. In the absence of IRF4, splenic B cells are present in relatively normal numbers but are skewed towards MZ B cells and away from follicular B cells (9). This is thought to be due to increased Notch2 signaling (9, 10), which is necessary for MZ B cell development and localization (11–13). IRF4 is also required in B cells for germinal center formation and plasma cell differentiation (14–16).

Because of IRF4’s unique role in mature B cells, we hypothesized that it may also have some functions earlier in B cell development that cannot be compensated for by IRF8. This is supported by the observation that loss of several other components of IL-7 and pre-BCR signaling pathways causes defects in pre-B cells and partial blocks in B lymphopoiesis, but still allows the eventual development of relatively normal numbers of mature B cells (17–19). Indeed we find that IRF4 deficient mice have a partial block at the pre-B stage, accompanied by reduced expression of the pre-B cell marker CD25, increased responsiveness to IL-7, and impaired migration towards CXCL12. We also show that IRF4 acts prior to the mature B cell stage to control MZ B cell numbers.

Based on the study of IRF4/IRF8 double knockout mice and the reconstitution of pre-B cells from these animals with either IRF4 or IRF8, it has been thought that IRF4 and IRF8 are redundant for early B cell developmental events. Here, by characterizing IRF4-deficient mice, we find that IRF8 does not compensate, or can only partially compensate, for IRF4. Expression of CD25 in B220+IgM- cells was almost completely dependent on IRF4 alone. IRF4 also has a dominant role in suppressing proliferation in response to IL-7, as demonstrated by the increased expansion of B220+IgM- cells in IL-7 cultures of IRF4-deficient bone marrow. Finally, we observed a reduction in IgM+ immature B cells both ex vivo and in IL-7 cultures of IRF4-deficient mice. However, IgM+ immature B cells were still present in these mice, and splenic B cell numbers are not decreased although they are skewed towards MZ B cells and away from follicular B cells. In contrast, IRF4-/-IRF8-/- mice lack all immature and mature B cells in both the bone marrow and the spleen due to a failure to undergo Ig light chain recombination. Thus, while IRF4 contributes to light chain assembly and subsequent IgM expression, IRF8 can compensate for this activity to allow for the development of peripheral B cells in the absence of IRF4.

One caveat to the interpretation of these results is that the mice used for the bone marrow studies here were whole body knockouts. We attempted to generate mice lacking IRF4 throughout the B lineage using mb1-cre. However, these studies were unsuccessful because the mb1-cre caused frequent deletion of IRF4 in the germline as has been described for some loci (21). In vitro transduction of purified IRF4-/-IRF8-/- pre B cells with IRF4 results in upregulation of CD25 and, upon IL-7 withdrawal, light chain rearrangement and surface IgM expression (5, 8). This, as well as the links to B cell intrinsic signaling pathways discussed below, suggest that IRF4 has a B cell intrinsic role in the progression of B lineage cells beyond the pre-BCR+ CD25-stage. However, we cannot rule out a role for non-B cell expressed IRF4 in regulating proliferation or B lineage cell numbers.

Several of the early B cell developmental defects in IRF4-deficient mice resemble those of mice lacking either of the pre-BCR signaling components Btk or BLNK (also known as SLP-65). B220+IgM- bone marrow cells from both Btk-/- and BLNK-/- mice have reduced CD25 expression, increased pre-BCR expression, and increased expansion in response to IL-7 (17–19). Similarly, Btk and BLNK knockout animals have reduced light chain recombination but are able to develop IgM+ peripheral B cells (17–19). This supports a model in which a pre-BCR/Btk/BLNK signaling module preferentially signals via IRF4, rather than IRF8, to control B cell development in the bone marrow. Indeed, BLNK has been shown to mediate pre-BCR induced IRF4 expression (3, 23). This is due to its attenuation of PI3K/Akt signaling and subsequent upregulation of Foxo1, which in turn drives IRF4 expression (3, 23). Intriguingly, inhibiting PI3K signaling increases IRF4, but not IRF8 expression, in BLNK-deficient pre-B cells (23). In a feed-forward loop, IRF4 but not IRF8 can promote Pax5 expression (23) which, together with Foxo1, upregulates BLNK levels (3, 24). The participation of IRF4 but not IRF8 in these amplification mechanisms may account for the more dominant role of IRF4 in early B cell development.

In the periphery, IRF4-deficient mice have an increase in MZ B cells. This is thought to be a result of increased Notch signaling which is known to drive the MZ B cell fate decision and promote retention in the marginal zone (9). We found that CD21-cre.IRF4f/f mice had normal numbers of MZ B cells. This suggests that IRF4 acts prior to the upregulation of CD21 to drive MZ B cell expansion. It has recently been shown that T1 B cells expressing cell surface Adam10, a metalloproteinase required for the cleavage and activation of Notch2 (13), give rise to MZ B cells upon adoptive transfer (10). This study also reported that IRF4-deficient mice have a greater frequency of T1 cells expressing surface Adam10 compared to their wild type counterparts (10). Here, we find that IRF4-deficient transitional B cells are enriched in T1 cells. Taken together, these data support a model in which loss of IRF4 prior to the mature B stage results in a skewing of transitional B cells towards a MZ B-committed T1 subset. MZ B cells are also increased when Foxo1 is deleted with CD19-cre but not CD21-cre (25). As they do in pre-B cells (3, 23), Foxo1 and IRF4 may act together to limit T1 cell commitment to the MZ B fate. Alternatively, and not mutually exclusively, a low level of residual IRF4 in CD21-cre.IRF4f/f B cells may be sufficient to limit MZ B cell expansion but not to drive plasma cell differentiation, as dosage of IRF4 has been shown to affect both its binding partners and its downstream function (15).

In summary, we have identified a unique role for IRF4 during early B lymphopoeisis that cannot be compensated for by IRF8. We also find that IRF4 acts prior to the expression of CD21 to limit MZ B cell development. Future studies comparing IRF4 and IRF8 targets in pre-B cells and identifying novel IRF4 targets in T1 cells will provide further clarity on the role of this critical transcription factor in B cell development.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

The animal study was reviewed and approved by UT Southwestern Institutional Animal Care and Use Committee (IACUC).

KO performed experiments, analysed data, and edited the manuscript. AS designed experiments, analysed data, and wrote the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by NIH grant AI137746 to AS. AS is a Southwestern Medical Foundation Scholar in Biomedical Research and holds the Peggy Chavellier Professorship in Arthritis Research and Treatment.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.