YY1 Is Required for Germinal Center B Cell Development

YY1 has been implicated as a master regulator of germinal center B cell development as YY1 binding sites are frequently present in promoters of germinal center-expressed genes. YY1 is known to be important for other stages of B cell development including the pro-B and pre-B cells stages. To determine if YY1 plays a critical role in germinal center development, we evaluated YY1 expression during B cell development, and used a YY1 conditional knock-out approach for deletion of YY1 in germinal center B cells (CRE driven by the immunoglobulin heavy chain γ1 switch region promoter; γ1-CRE). We found that YY1 is most highly expressed in germinal center B cells and is increased 3 fold in splenic B cells activated by treatment with anti-IgM and anti-CD40. In addition, deletion of the yy1 gene by action of γ1-CRE recombinase resulted in significant loss of GC cells in both un-immunized and immunized contexts with corresponding loss of serum IgG1. Our results show a crucial role for YY1 in the germinal center reaction.


Introduction
Affinity maturation of immunoglobulins (Ig) in B cells largely occurs during the germinal center (GC) reaction where the processes of somatic hypermutation (SHM) and class switch recombination (CSR) occur [reviewed in references [1][2][3]]. B and T cells and that have been activated by antigen migrate to interfollicular regions in secondary lymphoid organs and interact [4,5]. These cells form long-lived interactions resulting in full B cell activation with increased expression of B Cell Lymphoma 6 (BCL6) protein and activation induced cytidine deaminase (AID) [6]. Activated cells migrate from the interfollicular region to the follicle where the B cells proliferate to begin formation of a germinal center [6,7]. Finally, the dark and light zones of the germinal center develop and B cells transition between these zones with SHM occurring in the dark zone, and affinity selection and CSR in the light zone. Ultimately the B cells that are selected, mature into either memory B cells or plasma cells and exit the germinal center [1,2].
A number of transcription factors regulate the germinal center reaction. BCL6 is critical for germinal center formation as its deletion ablates GC formation [6,8]. A variety of other transcription factors effect either early or late germinal center formation and include Pax5, IRF4, IRF8, NF-κB, E2A, c-Myc, MEF2B, MEF2C, EBF1, and SpiB [1][2][3]. In addition, the histone methyltransferase EZH2 is crucial for GC formation [9]. These factors regulate gene expression profiles needed for germinal center formation and control cell proliferation which approaches the highest rates in mammalian systems [10].
Recently, transcription factor Yin Yang 1 (YY1) was proposed to be a master regulator of germinal center function [11]. Using computational approaches, Green and colleagues [11] characterized promoters of genes that are expressed in germinal center cells. The promoters of these GC signature genes were enriched in binding sites for YY1. In addition, it has been proposed that YY1 binding sites, as well as sites for E2A and C/EBPα are enriched within nonimmunoglobulin regions of the genome where AID binds and generates off-target site mutations, perhaps involved in genesis of B cell malignancies [12]. Consistent with this idea, we showed that YY1 physically interacts with AID, leading to its stabilization and nuclear accumulation [13]. We also found YY1 conditional knock-out in splenic B cells, results in reduction of CSR [13]. Furthermore, YY1 is known to be critical for B cell development at other B cell stages. Using mb1-CRE, the Shi laboratory showed that conditional deletion of the yy1 gene in early pro-B cells results in pro-B cell arrest, reduced IgH locus contraction, and reduced VDJ rearrangement of distal Vh genes [14]. Similarly we showed that deletion of the YY1 REPO domain needed for recruitment of Polycomb Group (PcG) proteins to DNA results in arrest at the pre-B cell stage and highly skewed Vκ gene rearrangement patterns [15]. We also showed that YY1 physically interacts with, and co-localizes with proteins involved in long-distance DNA contacts including condensin, cohesin, and PcG subunits [15]. Thus, YY1 clearly plays a significant role in B cell development.
Here we evaluated YY1 expression during B cell development, and used a γ1-CRE conditional knock-out approach to delete YY1 in germinal center B cells. We found that YY1 is most highly expressed GC B cells. Deletion of the yy1 gene resulted in significant loss of GC cells in both un-immunized and immunized contexts. Our results show a crucial role for YY1 in the germinal center reaction.
By RT-PCR we found that YY1 transcripts were about 2 fold higher in germinal center B cells compared to follicular B cells (Fig 1I) corresponding with a roughly 2 fold increase in YY1 proteins levels as measured by MFI (Fig 1G and 1H). Activation of isolated splenic B cells with anti-IgM and anti-CD40 to mimic the germinal center activation phenotype caused a 3 fold increase in YY1 levels as measured by MFI (Fig 1J).

YY1 deletion by γ1-CRE activity impacts splenic B germinal center populations
To determine the importance of YY1 in germinal center development, we used the yy1 f/f mouse line [14] that contains flox sites flanking the first exon in the yy1 gene crossed with a γ1CRE line which expresses CRE recombinase from the IgH γ1 switch region promoter [16], to generate yy1 f/f γ1CRE mice. The γ1CRE gene is expressed upon B cell activation initiating the germinal center reaction and this results in deletion of floxed genes within the first two days of the germinal center reaction [1].
Analyses of the naïve B cell populations present in spleen from non-immunized mice from each genotype (yy1 f/f , γ1-CRE, and yy1 f/f γ1-CRE) showed that the transitional, marginal zone, follicular, and plasma B cell populations were relatively unchanged in each genotype (S1 Fig). Similarly, as expected, little differences were observed in bone marrow populations upon conditional deletion of YY1 by γ1CRE ( S2 Fig). Thus, non-immunized yy1 f/f , γ1CRE, and yy1 f/f γ1CRE mice showed the same levels of pro-B, pre-B, immature B, and recirculating B cells (S2 Fig). However, germinal center B cells showed a pronounced difference. Whereas yy1 f/f and γ1CRE mice showed germinal center B cell populations of 0.9-0.8% of total cells, this population dropped nearly 10 fold to 0.097% in yy1 f/f γ1CRE mice (Fig 2A). Whereas the percentage and number of total B cells remained unchanged, the percentage and number of germinal center B cells dropped dramatically in the yy1 f/f γ1CRE line compared to yy1 f/f and γ1CRE lines (Fig 2B and 2C). Thus, deletion of YY1 by action of the γ1CRE transgene resulted in loss in germinal center B cells. We observed no difference in the total number of T follicular helper (T FH ) cells in the yy1 f/f γ1CRE mice compared to yy1 f/f and γ1CRE mice, which are essential for germinal center formation and maintenance (S3A and S3B Fig). In addition, we did not observe any difference in follicular B cell proliferation in response to various stimuli suggesting no adverse impact of CRE expression on follicular B cells in yy1 f/f γ1CRE mice (S3C Fig). Together, our results indicate that deletion of YY1 by action of the γ1CRE transgene resulted in loss of germinal center B cells.
We also measured serum Ig isotype levels in each genotype. As YY1 impacts germinal center B cell development where Ig CSR generally occurs, we anticipated that levels of IgM would remain similar, but that IgG1 isotype would be reduced in the yy1 f/f γ1CRE background due to activation of the γ1-CRE gene. As expected, we found that IgM levels were comparable amongst the unimmunized yy1 f/f , γ1CRE, and yy1 f/f γ1CRE lineages as were IgA, IgG2, and IgG3 ( Fig 2D). However, levels of IgG1, and total IgG were greatly reduced in the yy1 f/f γ1CRE line compared to the yy1 f/f and γ1CRE lines (Fig 2D).

Antigen-specific GC cells are lost upon YY1 deletion
To determine the impact of YY1 conditional knock-out on germinal center B cells after initiation of an immune response, we injected mice with NP-chicken gamma globulin (NPP-CGG), a T cell-dependent antigen. After two weeks we collected blood and spleen to determine immune responses. Deletion of YY1 by γ1-CRE action in the yy1 f/f γ1CRE line caused tremendous loss in the number of NP+ germinal center B cells compared to the γ1CRE line (Fig 3A  and 3B). This was confirmed by staining histological sections with fluorescent antibodies that recognize germinal center B cells (anti-GL7), follicular B cells (anti-IgD), and T cells (anti-TCRβ) (Fig 3C). Thus, deletion of YY1 by γ1CRE caused loss of germinal center B cells, but not follicular B cells or T cells (Fig 3B).
We also determined the impact of YY1 deletion on high affinity verses low affinity antibodies against NP-CGG using NP(CGG)4 reactivity as definition of high affinity and NP(CGG)26 as low affinity. Deletion of YY1 caused a drop in both high affinity and low affinity IgG1 antibodies against NP-CGG (Fig 3D). In contrast, IgM antibodies against NP-CGG produced by cells that have not entered the germinal center reaction were unaffected (Fig 3E). These results demonstrate that YY1 is critical for germinal center B cell development, germinal center-mediated immune responses, and loss of YY1 ablates the formation of germinal centers.

Discussion
Our results indicate that deletion of the yy1 gene by action of γ1-driven CRE dramatically reduces the number of germinal center B cells in the spleen, as well as the histological appearance of germinal centers. The γ1 promoter is activated early in the germinal center reaction causing gene deletion within the first two days after antigen stimulation. This indicates that YY1 is critical for early events in germinal center development.
The phenotype with γ1CRE-mediated YY1 deletion is very similar to that observed upon EZH2 deletion using the same γ1CRE transgene [9]. EZH2 is a Polycomb Group (PcG) protein  component of the Polycomb Repressive Complex 2 (PRC2) and is responsible for trimethylation of histone H3 on lysine 27 leading to stable transcriptional repression [17][18][19][20][21][22]. EZH2 is also involved in cell proliferation and germinal center B cells are among the highest proliferating cells in mammalian systems [10]. EZH2 expression increases in germinal center B cells indicating its apparent importance in transcriptional repression or proliferation [23,24]. EZH2 is implicated in a number of malignancies and importantly, is directly involved in development of various lymphomas including diffuse large B cell lymphoma (DLBCL) derived from germinal center B cells [9,25,26].
YY1 now joins a growing list of transcription factors involved in germinal center development including BCL6, IRF4, IRF8, NF-κB, E2A, c-Myc, MEF2B, MEF2C, EBF1, and SpiB [1][2][3]. Each factor appears to control various aspects of germinal center biology by either regulating germinal center-specific gene regulatory networks, or by controlling proliferation. Deletion of the c-myc gene by the same γ1-CRE gene used here also results in loss of germinal center B cells. c-Myc is needed for early germinal center formation as well as for germinal center maintenance [33,34]. It is believed that the importance of c-myc for germinal center formation and maintenance relates to its ability to control cell proliferation. Germinal center B cells proliferate at an extremely high rate and loss of this proliferation likely relates to loss of germinal center initiation as well as collapse of germinal center maintenance.
YY1 was proposed to regulate germinal center biology by regulating gene expression networks, as YY1 DNA binding sites lie within the promoters of genes expressed in germinal center B cells [11]. Consistent with this hypothesis, we found that YY1 protein is expressed at highest levels in germinal center B cells. However, YY1 controls multiple stages of B cell development. For instance, YY1 deletion early in B cell development by action of mb1-driven CRE results in arrest at the pro-B cell stage and loss of immunoglobulin heavy chain (IgH) locus contraction needed for distal Vh gene rearrangement [14]. In addition, the long distance DNA contacts needed for V(D)J rearrangement at the pro-B cell stage are ablated upon YY1 deletion [35,36]. At the pre-B cell stage, we showed that YY1 PcG function is required for generating complete Igκ gene repertoires, again likely due to impacting long-distance DNA loops needed for Vκ-Jκ rearrangement [15]. Finally, we showed that YY1 is important in mature splenic B cells for controlling IgH class switch recombination (CSR) [13]. While germinal center-specific genes may be regulated by YY1, the requirement for YY1 early in the germinal center reaction, and the ability of YY1 to control other stages of B cell development suggest that YY1 either impacts numerous distinct B cell stage-specific functions, or it controls common functions at each stage such as cell survival or proliferation. Indeed YY1 controls numerous housekeeping genes such as the ribosomal protein genes among others [37][38][39], and its complete knock-out results in early embryonic lethality [40]. Therefore, additional work will be necessary to determine the multiple and varied roles of YY1 in B cell development.

Mice and Immunization
We obtained yy1 f/f mice from Dr. Yang Shi (Harvard University) and bred these mice with γ1-CRE mice generated by the Rawjewsky laboratory [16] and supplied by Jackson Laboratories (Stock No: 010611). Work using mice followed recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Institutional Animal Care and Use Committee of the University of Pennsylvania (Protocol 803080). Mice were immunized i.p. with 50μg NP-CGG in alum.

Flow Cytometry
For all analyses, bone marrow (BM) and spleens were disrupted to single cell suspension, and red blood cells were lysed using ACK buffer (Lonza). Equal numbers of cells were incubated with live/dead fixable aqua stain in PBS for 20 min at room temperature. Cells were washed and stained for surface antigens in PBS with 2% bovine serum albumin (BSA) for 30 min at 4°C. Following washing, cells were treated with Cytofix/Cytoperm buffer (eBiosciences) and then stained with antibodies against intracellular Ags for 30 min at 4°C. Data were collected on a BD LSR II flow cytometer and analyzed with FlowJo software (Tree Star). The antibodies used for flow cytometry were CD19 (clone 6D5), B220 (clone RA3-6B2), CD43 (clone S7), AA4.1/CD93 (clone AA4.

ELISA
Ninety-six well plates were coated with 10μg/ml anti-Ig (H + L) (Southern Biotech) overnight at 4°C and blocked with PBS containing 2% BSA for one hour. Sera were incubated at various dilutions for one hour at room temperature. Detection was conducted using HRP-conjugated goat anti-mouse IgM, IgG, IgG1, IgG2 a+c , IgG3 or IgA (Southern Biotechnology) with a TMB substrate kit (BD Biosciences) and color development was quantified using EMax (Molecular Devices).