Skip to main content
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Aug 6;44(5):e70227. doi: 10.1002/hon.70227

DNA Replication Stress‐Induced Transcriptome of Human Burkitt's Lymphoma Identifies Reciprocal Regulation Between MBD1 and BCL6 During Germinal Center‐Derived B‐Lymphomagenesis

Santosh Kumar Gothwal 1,2,✉, Kyoko Oichai 3, Hongtae Kim 4, Kyungjae Myung 2,✉, Jacqueline H Barlow 5
PMCID: PMC13447708  PMID: 42563211

ABSTRACT

BCL6 is a master transcriptional regulator of germinal center (GC) B cells. BCL6 is frequently translocated at the major translocation cluster (MTC) within intron 1 of the BCL6 locus, a hotspot commonly rearranged in diffuse large B cell lymphomas (DLBCLs). BCL6 amplifications are associated with therapeutic resistance and poor survival outcomes in hematological and solid cancers. However the mechanisms suppressing genome instability at the BCL6‐MTC preventing BCL6 rearragements remain unclear. Here, transcriptome analysis and genome‐wide mapping of histone H3 lysine 4 trimethylation (H3K4me3) in hydroxyurea (HU)‐treated Raji cells (a Burkitt's lymphoma model) revealed the induced expression of MBD1, encoding the DNA CpG methylation‐binding protein. Functional studies using shRNA silencing and ectopic overexpression demonstrated that MBD1 suppresses BCL6 transcription whose promoter harbors conserved CpG methylation sites, suggesting a DNA methylation‐dependent regulation of BCL6 trasncription by MBD1. Conversely, BCL6 repressed MBD1 expression by binding to its promoter. MBD1‐depleted Raji cells exhibited increased genomic instability at the BCL6‐MTC upon HU treatment, heightened sensitivity to DNA replication inhibitors (HU, gemcitabine, and etoposide), and reduced tumorigenicity in xenograft mouse models. We propose that MBD1 prevents genomic instability at the BCL6‐MTC to suppress DLBCL formation. Moreover, MBD1 promotes genomic stability and cell viability during DNA replication stress. MBD1 thus represents a potential therapeutic target for cancers exhibiting resistance to chemotherapies targeting DNA replication.

Keywords: B‐lymphomagenesis, BCL6, DLBCL, IRF4, MBD1, multiple myeloma, waldenström macroglobulinemia

1. Introduction

Dark zones (DZs) and Light zones (LZs) are two main anatomical compartments (Phan, 2005 #36) within GCs [1]. DZs accommodate GC B cells undergoing activation induced cytosine deaminase (AID)‐induced somatic hypermutation (SHM) [2], and LZs are proposed anatomical sites favoring AID induced class switch recombination (CSR) [3, 4]. GC B cells remain at high risk of genetic rearrangement in both DZs and LZs because dysregulated AID activity during SHM and CSR can initiate oncogenic chromosomal translocations including those fusing the Immunoglobulin locus (IG) with trans loci such as MYC and BCL6 [5, 6]. GC B cells with these translocations develop into Burkitt's lymphoma (IG‐MYC) and DLBCLs (IG‐BCL6) [5]. The hijacked regulation of differentiation pathways toward plasma and memory B cells in Germinal center derived B‐lymphomas (GCDBL) can lead to Multiple Myeloma (MM) and B cell chronic lymphocytic leukemia (B‐CLL), respectively [7]. Additionally, DLBCLs overexpressing CXCR4, which encodes G‐protein coupled receptors (GPCRs), can develop into Waldenström macroglobulinemia, another DLBCL subtype marked by BCL6 and CXCR4 co‐amplification [1, 8, 9, 10, 11]. Thus, controlling genetic rearrangements in GC B cells, along with regulated differentiation and affinity maturation, is important for GCDBL elimination.

The BCL6 is a master transcriptional regulator of GC B cells belonging to the zinc finger and BTB (ZBTB) family and is comprised of an N‐terminal BTB/POZ (Broad‐complex, Tramtrack, Bric‐a‐brac/Poxvirus and zinc fingers) domain and Krüppel‐type zinc fingers at the C terminus [12, 13]. The zinc finger domain of BCL6 is required for DNA binding to recruit the SMRT/mSIN3A/histone deacetylase for transcriptional suppression of genes in cell cycle arrest, DNA damage repair and p53 response [12, 13]. Thus, BCL6 drives the transcription program promoting GC B cell activation and survival by inhibiting the DNA damage response, apoptosis in GC B cells undergoing the SHM and CSR [14, 15].

BCL6 overexpression is frequently observed in DLBCL [5, 16]. The breakpoint analysis of hematological malignancies suggests that 31% of DLBCLs harbor BCL6 translocations on intron 1, a 10,502 base pair sequence also known as major translocation cluster (MTC) [5, 16]. The start codon of BCL6 is located on exon 3, thus translocations involving intron 1 encodes for full length BCL6 when juxtaposed with an active locus in trans. Fifty‐two percent of BCL6 translocations involve the Immunoglobulin heavy chain (IGH) locus, 10% with immunoglobulin light chain (IGL), and 38% with non‐IG loci [5, 16, 17], suggesting random nature of BCL6‐translocation to sites other than IG locus. BCL6 also acts as a proto‐oncogene in the pathogenesis of breakpoint cluster region (BCLR)–v‐abl Abelson murine leukemia viral oncogene homolog 1(ABL1)‐driven acute lymphoblastic leukemia (ALL). BCL6 translocations have been also observed in glioblastoma and elevated BCL6 activity—along with its corepressor NcoR—is associated with AXL signaling and therapy resistance in glioblastoma [18]. BCL6 overexpression enables survival of Philadelphia chromosome‐positive ALL (pH+ ALL) cells upon BCLR–ABL1–kinase inhibition through repression of p53 expression [19]. BCL6 also drives therapy escape in non‐small cell lung cancers, breast cancer, colorectal and gall bladder cancer [20, 21].

GC B cell differentiation is driven by dynamic changes in chromatin state including the DNA methylation and EZH2 driven histone H3K27me3 of the genes regulating the plasma B cell differentiation such as IRF4 and PRDM1 [22, 23]. The DNA CpG methylation binding protein 1 (MBD1) exhibits binding to methylated and unmethylated DNA via its CXXC domains ({Fujita, 2003 #78) and regulates gene expression by recruiting transcriptional suppressors such as SETDB2, HP‐1α, SUV39H, histone‐deacetylases‐3 and PML‐RARA {Fujita, 2003 #78; Clouaire, 2010 #79 [24]. However, it is not yet clear if MBD1 regulates BCL6 transcription through DNA methylation‐dependent or ‐independent manner. During the plasma B cell differentiation, DNA hypomethylation of the promoters of PRDM1, XBP1, IRF8, SPIB genes assists plasma B cell differentiation [22, 25]. DNA methyltransferase activity was reported to be decreased in GC B cells than naïve B cells and Dnmt1 hypomorphic mice had a defective GC reaction [22], suggesting a role for DNA methylation axis in GC B cell activation and differentiation.

Intron 1 of BCL6 locus harbors four major CpG rich domains [5, 16], suggesting a role of DNA methylation in BCL6 transcription. CTCF, an insulator which binds to unmethylated DNA, also binds on intron 1 and suppresses BCL6 transcription, suggesting this transcriptional suppression is methylation‐independent [26]. CpG methylation is catalyzed by the 10 11 translocation (TET) enzymes and is recognized by methylation binding proteins (MBDs), suggesting a complex regulation involving DNA demethylation, TET enzymes, MBD and CTCF binding in the regulation of BCL6 transcription [4, 27, 28, 29, 30]. The mechanistic link between DNA methylation‐dependent suppression of BCL6 transcription and whether MBD proteins are involved in BCL6 transcription and remain unknown.

In the current study, we have characterized the roles of MBD1 in regulation of key genes involved in GC B cell fate determination utilizing a recently described approach for identification of new GC regulators. Functional characterization of MBD1 revealed its role in suppressing BCL6 expression during DNA replication stress. Moreover, MBD1 knockdown resulted in increased DNA damage at the BCL6‐MTC region while MBD1‐depelted Raji cells exhibited higher sensitivity to DNA replication inhibitors. These results place MBD1 as a novel regulator of BCL6 translocation and expression of BCL6. Thus MBD1 could be essential for allowing the proper differentiation of GC B cells undergoing dynamic chromatin modification and differentiation ensuring a healthy immune response and suppression of GCDBLs.

2. Results

2.1. Inverse Correlation Between MBD1 and BCL6 Expression in GC B Cells and Cancers

We recently defined a novel approach using Raji cells to define the molecular mechanisms regulating the fate of GC B cells and GC DLBCL cells [31]. Briefly, we exposed Raji cells to 4 mM HU for 12 hours to induce the genotoxic‐stress as experienced by GC B cells experience during rapid proliferation coupled with SHM and CSR [31]. We hypothesized that factors induced during this treatment could be important regulators on GC B cells. Based on this screening, we centered our investigation on MBD1 due to its significant upregulation as one the top 50 genes identified in our study (Figure 1A). We first confirmed whether MBD1 expression is induced in cancer cells treated with DNA replication inhibitors and the DNA damage inducers Hydroxyurea (HU), Gemcitabine, and Etoposide (Supporting Information S1). We performed short hairpin RNA (shRNA) mediated knockdown of MBD1 in Raji cells (shMBD1‐Raji) and in breast cancer cells (shMBD1‐SUM149PT), and control cells with scrambled shRNA (shSCR‐Raji and shSCR‐SUM149PT) (Supporting Information S1). We confimed that MBD1 expression was significantly induced in scramble‐knockdown cells, shSCR‐Raji cells treated with HU, Gemcitabine or Etoposide (Supporting Information S1). While we confirmed that shMBD1 cells reduced MBD1 expression by ∼75% compared to shSCR‐Raji cells (Supporting Information S1), shMBD1‐Raji cells treated with these agents also exhibited a trend of induced MBD1 expression (Supporting Information S1). Similarly, breast cancer origin cells, shSCR‐SUM149PT also induced MBD1 expression with HU‐treatment (Supporting Information S1). These results suggest that MBD1 expression is induced in DNA replication stress‐dependent manner. In addition, we observed histone H3K4me3 enrichment within −2 kilobase pairs of its TSS in HU‐treated cells (Figure 1B), suggesting MBD1 expression is partially dependent on histone H3K4me3 under HU stress (Figure 1B).

FIGURE 1.

FIGURE 1

Identification of the DNA‐replication stress‐induced transcriptome in Raji cells. (A) Differentially expressed genes in control and HU‐treated Raji cells (n = 3) x axis indicated log2FC for mRNA fold change and y axis indicated ‐log10 of adjusted p values. Data is from three independent samples (B) Analysis of histone H3K4me3 trimethylation ChIP‐seq on the MBD1 locus in vehicle‐treated or HU‐treated Raji cells (C) Real‐time kinetics of MBD1 and BCL6 expression in activated germinal center B cells (AGCBs) from human tonsil. (D) BCL6‐diploid diploid tumors were classified into MBD1‐amplification (n = 6), MBD1‐deep deletion (n = 19), MBD1: diploid (n = 4230), MBD1:gain (n = 298) and MBD1‐shallow deletion (n = 1397) exhibiting mean BCL6‐mRNA values of 1.110, 1.375,1.103, 0.8686 and 1.309 in each group respectively. p=< 0.0001 MBD1:diploids versus MBD1:gain and MBD1‐diploid versus MBD1‐shallow deletion respectively. Dunnett's multiple comparison test One way ANOVA. Data are presented as mean ± SEM. (E) Comparison of BCL6 mRNA levels in the DLBCL samples (TCGA database) exhibiting either MBD1:diploid or MBD1:gain status. Normalization of BCL6 mRNA in MBD1 altered groups was calculated by normalizing the RSEM values of BCL6 mRNA by RSEM values of MBD1 mRNA in each sample. P = 0.0.0307 for MBD1:diploid versus MBD1:gain. t‐test, Mann‐Whitney test (F) MBD1‐diploid tumors were classified into BCL6‐amplification, BCL6‐deep deletion, BCL6‐diploid, and BCL6‐gain and BCL6‐shallow deletion groups. Values on y‐axis indicates normalized RSEM values of MBD1 mRNA with the RSEM values of BCL6 mRNA. The mean of normalized MBD1 mRNA was 0.8332 in BCL6‐amplification (n = 182), 2.201 for BCL6‐deep deletion (n = 15), 1.57 in BCL6‐diploid (n = 4230), 1.171 in BCL6‐gain (n = 972), and 2.027 in BCL6‐shallow deletion (n = 390) group. P < 0.0001 for BCL6‐amplification versus BCL6‐diploid, p < 0.0001 for BCL6‐diploid versus BCL6‐gain. Dunnett's multiple comparison test, One way ANOVA.

To examine the expression of MBD1 in GC B cells, we analyzed the real‐time transcriptome data from activated human tonsil GC B cells (AGCBs) (Figure 1C) [32]. We simultaneously plotted the log2 fold change (Log2FC) values of MBD1 transcripts against BCL6 transcripts across subpopulations, including DZa, DZb, DZc, and intermediate (INT) populations (INTa, INTb, INTc, INTd, INTe), as well as LZa, LZb, LZc, pre‐memory (preM), and plasmablasts A and B (Figure 1C). These subpopulations were clustered based on genes specifically enriched in each compartment [32]. Log2FC values for MBD1 were relatively low in DZa, DZb, INTa, INTb, INTc, INTd, INTe, LZa, and LZb, with values ranging from −0.34 to 0.163 (Figure 1C). However, log2FC values of MBD1 were higher in pre‐memory (0.625) and plasmablast a cells (1.320; Figure 1C). In contrast, BCL6 expression sharply declined in pre‐memory and plasmablast stages (Figure 1C), suggesting a reciprocal relationship between MBD1 and BCL6 expression in GC B cells.

Given the reciprocal expression of BCL6 and MBD1 in human AGCBs, we checked the possibility of mututal suppression of MBD1 and BCL6. Among the tumor samples available in the cancer genome atlas (TCGA), we first categorized BCL6 and MBD1 altered tumors based on their genetic status and quantified their mRNA level in each group (Supporting Information S1). We found that MBD1 mRNA levels positively correlate with genetic alteration (Supporting Information S1). We also observed a similar relationhip of BCL6 mRNA expression with its copy number aleration in tumors (Supporting Information S1).

We then examined MBD1‐mutated tumors for alterations in BCL6 expression. Tumors with MBD1‐shallow deletions exhibited higher BCL6 expression than MBD1‐diploid tumors (Figure 1D). On the other hand, BCL6 mRNA levels were significantly lower in MBD1‐gain tumors (Figure 1D). Similarly, MBD1‐amplified tumors exhibited reduced trend of BCL6 mRNA expression while tumors with MBD1‐deep‐deletion status exhibited higher BCL6 mRNA levels than MBD1‐diploid tumors, although this trend was not significant in these groups (Figure 1D). Further, DLBCL samples with MBD1‐gain status exhibited lower BCL6 expression than MBD1‐diploid tumors (Figure 1E). Taken together, these results indicate an inverse relationship between MBD1 and BCL6 expression. To determine if BCL6 levels also inversely correlate with MBD1 mRNA levels, we sorted tumors with MBD1‐diploid samples and classified them as BCL6‐diploid, BCL6‐deep‐deletion, BCL6‐shallow‐deletion, BCL6‐gain and BCL6‐amplification (Figure 1F). We found that MBD1 mRNA was significantly lower in BCL6‐gain and BCL6‐amplified cancers, while MBD1‐mRNA was higher in BCL6‐deleted samples (Figure 1F). These results indiciate that BCL6 levels are inversely corealted with the MBD1 mRNA levels, sugesting the mutual suppresion of BCL6 and MBD1 by each others.

2.2. MBD1 Suppresses BCL6 Expression During DNA Replication Stress

Given the abnormal expression of BCL6 in breast cancer cells, we investigated whether MBD1 knockdown also impacts BCL6 levels in breast cancer cell lines MCF7, MDA‐MB‐231, SUM‐149PT, and MDA‐MB‐436 (Figure 2A). Notably, BCL6 expression was induced in shMBD1‐MCF7 and shMBD1‐MDA‐MB‐231 cells compared to control cells (Figure 2A). These results confirm that MBD1 can suppress BCL6 expression (Figure 2A). However, MBD1 depletion in SUM149PT and MDA‐MB‐436 cells did not induce the BCL6 suppression, suggesting additional mechanisms may regulate BCL6 transcription in these cells. Overall, these results confirm that MBD1 loss is associated with the induced BCL6 expression not only in B‐lymphoma cells but also in breast cancer.

FIGURE 2.

FIGURE 2

MBD1 suppresses BCL6 expression under DNA replication stress. (A) shMBD1‐MCF7, and shMBD1‐MDA‐MB‐231 cells exhibit higher levels of BCL6 expression than shSCR‐MCF7 and shSCR‐MDA‐MB‐231 cells. For BCL6 group: p < 0.0001 for shSCR‐MCF7 versus shMBD1‐MCF7 cells., p < 0.0001 for shSCR‐MDA‐MB‐231 versus shMBD1‐MDA‐MB‐231 cells, p = 0.9927 for shSCR‐MDAMB436 versus shMBD1‐MDAMB436, p > 0.9999 for shSCR‐SUM149PT versus shMBD1‐SUM149PT. For MBD1: p = 0.0051 for shSCR‐MCF7 versus shMBD1‐MCF7, p < 0.0001 for shSCR‐MDAMB231 versus shSCR‐MDAMB231, p = 0.0339 for shSCR‐MDAMB436 versus shMBD1‐MDAMB436, p = 0.0001 for shSCR‐SUM149PT versus shMBD1‐SUM149PT. Two‐way ANOVA. Tukey's multiple comparison test (B) pcDNA (1 μg) or pcDNA‐MBD1 (1 and 5 μg) were transfected into 0.5 million Raji cells seeded in 12‐well plates. RNA was extracted 24 hours post‐transfection, and cDNA preparation and qPCR were conducted for MBD1. P = 0.0070 for pcDNA (1 μg) versus pcDNA‐MBD1 (1 μg) and p < 0.0001 for pcDNA (1 μg) versus pcDNA‐MBD1 (5 μg). Sidak's multiple comparison test (ordinary one‐way Anova). Data are presented as mean ± SEM from three independent experiments (C) MBD1 overexpression was performed in Raji cells and cells were treated with HU and gemcitabine for 24 and 48 hours followed by estimation of BCL6 mRNA levels by qRT‐PCR. For 24 hour‐treated samples: p = 0.9990 for pcDNA (DMSO) versus pcDNA‐MBD1 (DMSO), p = 0.6897 for pcDNA (HU) versus pcDNA‐MBD1 (HU), and p = 0.0009 for pcDNA (Gemcitabine) versus pcDNA‐MBD1 (Gemcitabine). For 48 hour‐treated samples: p = 0.9748 for pcDNA (DMSO) versus pcDNA‐MBD1 (DMSO), p = 0.2108 for pcDNA (HU) versus pcDNA‐MBD1 (HU), and p = 0.0394 for pcDNA (Gemcitabine) versus pcDNA‐MBD1 (Gemcitabine). Two‐way ANOVA. Tukey's multiple comparison test (D) Relative mRNA levels of BCL6 in RIVA (DLBCL) cells. RIVA cells were transfected with 1 μg of empty vector or MBD1 plasmids for 24 hours and BCL6 mRNA levels were tested. Data are mean ± SEM. p = 0.0354, paired t‐test. (E) BCL6 promoter harbors CpG Methylation sites. Encode datasets of CpG methylation on BCL6 gene among several cells lines is shown using the UCSD datasets (F) MBD1 binds to the BCL6 promoters in HCT116 cells. Region highlighted in the above panel indicates the CpG methylation‐rich regions exhibiting MBD1 peaks in HCT116. The lower panel shows BCL6 genetic locus with MBD1 peaks through the gene body. MDB1 binding on BCL6 locus was examined using the dataset #DRX021118 from the ChIP‐atlas (https://chip‐atlas.org/view?id=DRX021118).

To further confirm that MBD1 suppresses BCL6 expression in B cells, we transiently transfected Raji cells with pcDNA‐MBD1 (Figure 2B, C). Compared to pcDNA3.1 (empty vector: EV) transfected cells, pcDNA‐MBD1 cells induced MBD1 expression in Raji cells (Figure 2B). While pcDNA‐MBD1‐transfected cells did not have significantly reduced BCL6 mRNA, treatment with gemcitabine caused a significant reduction in BCL6 levels in pcDNA‐MBD1 transfected cells (Figure 2C). In addition, 24 and 48 hours of gemcitabine treatment yielded significant reduction in BCL6 mRNA since EV‐transfected Raji cells exhibited 86% and 71% of BCL6 levels after 24 and 48 hours of gemcitabine treatment, respectively, while pcDNA‐MBD1‐transfected Raji cells showed 71% and 61% of BCL6 levels (Figure 2C). Exposure to HU induced a similar trend: after 24 hours of HU treatment, BCL6 levels were 90% in EV‐transfected Raji cells and 86% in pcDNA‐MBD1‐transfected cells (Figure 2C). Though not significant, this trend persisted after 48 hours of HU treatment, with pcDNA‐MBD1 showing 43% BCL6 mRNA levels compared to 51% in EV‐transfected cells (Figure 2C). Moreover, MBD1 overexpression in RIVA cells, a second DLBCL line [33], and U20S cells (osteosarcoma) also exhibited reduced BCL6 expression (Figure 2D, Supporting Information S1).

Since MBD1 recognizes DNA methylation on CpG sequences [24, 34, 35], we investigated whether MBD1 and its other family members might bind to methylated DNA sequences on the BCL6 locus and suppress BCL6 transcription (Figure 2D, E, Supporting Information S1). Using the ENCODE database, we identified four dispersed CpG methylation sites within intron 1 of the BCL6 locus, where the BCL6‐MTC resides (Figure 2D). In addition, by analyzing chromatin immunoprecipitation atlas (ChIP‐atlas) and MBD1 ChIP‐seq results from HCT116 cells, we confirmed MBD1 binding along the BCL6 gene body including intron 1 (Figure 2E). In addition, Cistrome data revealed Mbd2 binding on BCL6 locus in HeLa, MCF7, and HMEC cell lines while MBD3 binding was observed on BCL6 locus on intron 1 in MDA‐MB‐231and HeLa cells (Supporting Information S1). These results suggest that MBD1, MBD2 and MBD3 bind to the BCL6 locus and regulate BCL6 transcription which could be influenced by DNA methylation and replication stress.

2.3. MBD1 Suppresses Genomic Instability at the BCL6‐MTC in Raji Cells

Mutations increasing MBD1 expression negatively correlate with BCL6 transcription in B cells (Figure 1D), however the relationship between MBD1 and BCL6 mutations in tumors is unclear. We asked if MBD1 and BCL6 mutations co‐occur, we analyzed their co‐occurrence in 27 different tumor types using TCGA (Figure 3A). We found that MBD1 deep‐deletion or MBD1‐shallow deletion samples (commonly referred as MBD1‐deleted hereafter) exhibited a higher percentage of BCL6 alterations, suggesting that genetic alterations in MBD1 and BCL6 are positively associated (Figure 3A). Further, DLBCL tumors with MBD1‐deleted status exhibited a significantly higher frequency of BCL6 alterations, suggesting a strong correlation between MBD1 and BCL6‐mutated B lymphomas (Figure 3B). Based on these results, we hypothesized that MBD1 may also influence BCL6 translocation by suppressing DNA breaks on BCL6‐MTC preventing its rearrangement. To investigate this hypothesis, we analyzed γH2AX signal by chromatin immunoprecipitation (ChIP) on BCL6‐MTC after HU‐induced DNA replication stress in shSCR‐Raji and shSCR‐MBD1 cells (Figure 3C). We analyzed γH2AX abundance at four primer locations G, J, M and P on BCL6 intron 1, which spans 10.5 kb (Figure 3C) [26]. The level of γH2AX signal was higher in HU‐treated shMBD1‐Raji cells than HU‐treated shSCR‐Raji cells (Figure 3D), suggesting that MBD1 prevents HU‐induced DNA damage at BCL6‐MTC (Figure 3D). This role of MBD1 in suppressing DNA damage at BCL6‐MTC suggests that it could suppress BCL6 rearrangements in GC B cells and prevent DLBCL formation. Since CTCF binds to intron 1 of BCL6‐MTC and suppresses BCL6 transcription in MM cells [26], we hypothesized that MBD1 may regulate BCL6 transcription by collaborating with CTCF. We compared CTCF ChIP signal in HU‐treated shMBD1‐Raji and shSCR‐Raji cells on the BCL6‐MTC (Figure 3E). Compared to HU‐treated shSCR‐Raji cells, HU‐treated shMBD1‐Raji cells exhibited decreased CTCF occupancy at BCL6‐MTC compared to HU‐treated shSCR‐Raji cells (Figure 3E). This result suggests that MBD1 promotes CTCF recruitment to BCL6‐MTC during replication stress (Figure 3E).

FIGURE 3.

FIGURE 3

Increased genomic instability at BCL6‐MTC in MBD1‐depleted cells (A) TCGA dataset of 27 tumor types exhibiting the MBD1‐diploid of MBD1‐deletion status. Deletion and shallow deletion samples are counted as deletion samples. Percentage of BCL6 alteration is compared in each tumor type between MBD1:diploid or MBD1:deletion status. BCL6 alteration are the samples exhibiting with BCL6:deep deletion, BCL6:shallow deletion, BCL6:gain or BCL6:amplification. Tumor samples exhibiting BCL6:diploid status are not included as altered BCL6 status (B) The percentage of BCL6 alterations in DLBCL samples from TCGA database was quantified in MBD1‐diploid and MBD1‐deleted tumors and presented in the graph. MBD1‐deleted tumors include samples with MBD1 deep and shallow‐deletions. BCL6 alterations encompass tumor samples with BCL6 deep‐deletion, shallow‐deletion, gain, and amplification, while BCL6 diploid samples were not considered altered (C) Schematic representation of the major translocation cluster of the human BCL6 gene indicating exons and introns. Primer pair G, J, M, P location analyzed for ChIP‐qPCR is indicated (D) γH2AX ChIP analysis in shSCR‐Raji cells and shMBD1‐Raji cells treated with 10 mM HU. qPCR analysis was performed using four primer pairs, G, J, M, and P, targeting intron 1 of the BCL6 locus. Primer pair locations are not to scale. For primer pair G, p = 0.1563 for shSCR‐Raji (10 mM HU) versus shMBD1‐Raji (10 mM HU); for J, p = 0.0006 for shSCR‐Raji (10 mM HU) versus shMBD1‐Raji (10 mM HU); for M, p = 0.0259 for shSCR‐Raji (10 mM HU) versus shMBD1‐Raji (10 mM HU); for P, p = 0.1209 for shSCR‐Raji (10 mM HU) versus shMBD1‐Raji (10 mM HU). Unpaired t‐test. Data are presented as mean ± SEM (n = 3) (E) CTCF ChIP analysis in vehicle‐ or HU‐treated shSCR‐Raji cells and HU‐treated Raji cells. For primer pair G, p = 0.0002 for shSCR‐Raji (10 mM HU) versus shMBD1‐Raji (10 mM HU); for J, p = 0.0093 for shSCR‐Raji (10 mM HU) versus shMBD1‐Raji (10 mM HU); for M, p = 0.0022 for shSCR‐Raji (10 mM HU) versus shMBD1‐Raji (10 mM HU); for P, p < 0.0001 for shSCR‐Raji (10 mM HU) versus shMBD1‐Raji (10 mM HU) Unpaired t‐test. Data are presented as mean ± SEM (n = 3).

2.4. BCL6 Negatively Regulates MBD1 Transcription

BCL6 suppresses plasma B cell differentiation by inhibiting expression of IRF4 and PRDM1 [36]. Given the inverse correlation of BCL6 and MBD1 expression in AGCBs (Figure 1C, D, F), we hypothesized that BCL6 may suppress MBD1 expression in GC B cells undergoing plasma B cell differentiation. To test the effect of BCL6 inhibition on MBD1 expression, we treated Raji cells with FX1, a potent BCL6 inhibitor which binds to the BCL6 lateral groove and inhibits the formation of the BCL6 repression complex [37]. We found that FX1 treatment significantly induced MBD1 expression in Raji cells (Figure 4A), suggesting that inhibition of BCL6 binding to its DNA target sequences suppresses MBD1 transcription (Figure 4A). These results suggest that MBD1 and BCL6 can regulate each other's expression in GB B cells and is suggestive of mutual negative feedback. In addition, shBCL6‐Raji cells exhibited a trend of increased MBD1 expression compared to shSCR‐Raji cells, however this difference was not significant (Figure 4B). We confirmed that BCL6 knockdown was significantly achieved in shBCL6‐Raji cells (Figure 4B), and IRF4 levels were enhanced in shBCL6‐Raji cells, similar to previous reports [38]. Furthermore, BCL6 inhibition by the inhibitor FX1 induced MBD1 expression in AML37, a primary acute myeloid leukemia cell line while the inhibitor BI3812 or the degrader BI3802 also increased MBD1 expression in SUDHL4 and Farage cells, two DLBCL lines (Figure 4C, D). Moreover, BCL6 overexpression in RIVA cells also reduced MBD1 expression.

FIGURE 4.

FIGURE 4

BCL6 and AID inactivate MBD1 at transcriptional and genetic levels. (A) Treatment of Raji cells with the BCL6 inhibitor, FX1 (10 and 50 μM), induces MBD1 transcript levels in a dose‐dependent manner. Data are presented as mean ± SEM from three independent experiments. Dunnett's multiple comparison test: p = 0.0223 for Raji (DMSO) versus Raji (FX1‐50 μM) treatment groups (B) qPCR analysis of BCL6, MBD1, and IRF4 transcript levels in shSCR‐Raji and shBCL6‐Raji cells. Sidak's multiple comparison test: two‐way ANNOVA, For MBD1, p = 08,412 for shSCR‐Raji versus shBCL6‐Raji groups; For BCL6, p = 0.0030 for shSCR‐Raji versus shBCL6‐Raji. For IRF4 group, p < 0.0001 for shSCR‐Raji versus shBCL6‐Raji groups (C) MBD1 mRNA levels in AML37 primary leukemic cells under untreated, DMSO‐treated, and FX1‐treated conditions for 12 h. Data are presented as mean ± SEM from three independent samples. Tukey's multiple‐comparison test was used. ****p < 0.0001 for untreated versus FX1‐treated AML37 cells and DMSO versus FX1‐treated AML37 cells. (GSE133122) (D) MBD1 mRNA levels in SUDHL4 and Farage (FBXO11‐defective) cells treated with DMSO, BI‐8302 (BCL6 inhibitor, 500 nM), or BI‐8312 (BCL6 degrader, 500 nM) for 168 h. Data are presented as mean ± SEM from three independent samples. Dunnett's multiple‐comparison test was used. In SUDHL4 cells, p = 0.0024 for DMSO versus BI‐8302 and ****p < 0.0001 for DMSO versus BI‐8312 treatment. In Farage cells, p = 0.0639 for DMSO versus BI‐8302 and p = 0.0294 for DMSO versus BI‐8312 treatment (GSE94099) (E) In silico prediction of the BCL6 binding motif at the MBD1 promoter using JASPAR CORE 2018 Vertebrates, available in the Eukaryotic Promoter Database (EPD). The sequences of two BCL6 binding motif located on −235 and −875 base pair upstream of MBD1 TSS are indicated. The given sequence represents the regions deleted in the pGL4‐pr‐MBD1 plasmids (F) Schematic representation of the MBD1 promoter cloned upstream of the luciferase gene in pGL4‐luc2 and the BCL6 binding motif. Location of BCL6 binding motif and deletion are shown. HEK293 T cells were transfected with pGL4‐luc2, pGL4‐luc2 + pcDNA‐BCL6, pGL4‐luc2‐pr‐MBD1, pGL4‐pr‐MBD1 + pcDNA‐BCL6, pGL4‐luc2‐pr‐MBD1Δ235 + pcDNA‐BCL6, and pGL4‐luc2‐pr‐MBD1Δ900 + pcDNA‐BCL6 plasmids and examined for luciferase activity after 24 hours. Data are presented as mean ± SEM from three independent experiments. Tukey's multiple comparison test: p < 0.0001 for pGL4‐luc2 versus pGL4‐luc2‐pr‐MBD1; p < 0.0001 for pGL4‐luc2‐pr‐MBD1 versus pGL4‐luc2‐pr‐MBD1 + pcDNA‐BCL6 groups; p < 0.0001 for pGL4‐luc2‐pr‐MBD1 versus pGL4‐luc2‐pr‐MBD1Δ235 + pcDNA‐BCL6; p = 0.0416 for pGL4‐luc2‐pr‐MBD1 versus pGL4‐luc2‐pr‐MBD1Δ900 + pcDNA‐BCL6 (G) BCL6 binds to MBD1 locus in SUDHL4, OCI‐LY1 (DLBCL), B‐ALL and HepG2 cells cells. The data are analyzed from ChIP‐atlas database. Datasets SUDHL4:SRX4609168; datasets for, OCI‐LY1:SRX689470, B‐ALL:SRX18259603, HepG2:2,636,278 (https://chip‐atlas.dbcls.jp/data/hg38/target/BCL6.10.html) (H) The COSMIC mutation landscape of the MBD1 gene, showing the frequencies of A > C, A > G, A > T, C > A, C > T, C > G, G > A, G > C, G > T, T > A, T > C, and T > G mutations.

To explore how BCL6 suppresses MBD1 transcription, we performed in‐silico binding analysis of BCL6 on the MBD1 promoter (Figure 4E). We observed two BCL6 binding motifs on the MBD1 promoter present at −235 and −900 bp upstream of MBD1‐TSS (Figure 4E, F). To determine if BCL6 binding occurs at these sites, we cloned 1 kb of the MBD1 promoter region, and inserted into the luciferase reporter pGL4 vector then performed site‐directed mutagenesis on DNA sequences flanking these two binding sites. 293T cells transfected with pGL4‐pr‐MBD1 induced luciferase signal nearly 38‐fold compared to 293T cells transfected with empty pGL4 (Figure 4F). Interestingly, co‐transfection of pGL4‐pr‐MBD1 and pcDNA‐BCL6 reduced the luciferase signal to around 15%, suggesting that BCL6 suppresses MBD1 expression by binding its promoter (Figure 4F). Deletion of either BCL6 binding motif significantly increased luciferase signal, suggesting that BCL6 binding to the predicted binding motifs negatively regulate transcription (Figure 4E, F). Of note, pcDNA‐BCL6 expression had a smaller impact on luciferase signal from the pGL4‐pr‐MBD1 Δ235‐BCL6 or pGL4‐pr‐MBD1 Δ900‐BCL6 reporters than the wild‐type promoter sequence (Figure 4F). Particularly, deletion of the site located −900 upstream of MBD1‐TSS (Figure 4F) had a greater rescue from BCL6‐mediated suppression, suggesting that BCL6 binding to this motif plays a crucial role in regulating MBD1 expression. We next examined whether BCL6 binding is present on the endogenous MBD1 promoter. Using BCL6 ChIP‐seq datasets from ChIP‐atlas, we confirmed BCL6 binding on the MBD1 promoter in multiple cell lines including SUDHL4 (DLBCL), OCI‐LY1 (DLBCL), OCI‐LY3 (DLBCL) and in B‐cell acute lymphoblastic leukemia (B‐ALL) cell lines (Figure 4G). These results suggest that BCL6 binds to the MBD1 promoter and suppresses its transcription.

To investigate whether BCL6 regulation is associated with Mbd1 regulation in mouse GC B cells, we examined the DNA methylation status of the Mbd1 promoter in wild‐type and Sca1‐Bcl6 Δ mice (Supporting Information S1 [39]. We noted that the Mbd1 promoter in activated GC B cells of wild type mice showed two distinct DNA methylation peaks, and one peak was lost in Sca1‐Bcl6 Δ mice (Supporting Information S1). These results suggest that BCL6 suppression of Mbd1 could be associated with DNA methylation on the Mbd1 promoter in mouse GC B cells. To investigate if DNA‐methylation impacts MBD1 expression, we analyzed human B lymphomas treated with 5‐Azacytidine (AZA), a DNA methyltransferase inhibitor (Supporting Information S1), and observed that AZA treatment increased transcripts per million (TPM) counts of MBD1 in OCI‐LY1, SUDHL2, and OCI‐LY19 cells. Together these results suggest that MBD1 expression in GC B cells is regulated by BCL6 and DNA methylation (Supporting Information S1).

GC B cells are regulated by BCL6 and AID, both of which also act as oncoproteins. Since BCL6 suppresses MBD1 (Figure 4A, D), it is possible that MBD1 may function as a tumor suppressor in GC B cells. AID is also involved in B‐lymphomagenesis due to its mutagenic roles in GC B cells [10, 40], therefore we explored if AID is associated with MBD1 mutagenesis. We looked for AID signatures in MBD1‐mutant cancers in the Catalog of somatic mutation in cancer cells (COSMIC) (Figure 4H). Interestingly, AID mutation signatures (C > T), which are initiated by cysteine deamination to uracil in non‐replicating B cells, were frequent (38.9%) along the MBD1 gene in MBD1 mutant tumors (Figure 4H). Furthermore, C > G and C > A signatures, which are not as strongly associated with AID activity and could arise due to translesion synthesis and mismatch repair, were 3.4% and 5.5%, respectively (Figure 4H). Among non‐AID signature mutations in the MBD1 gene, G > A were the most frequent (27.9%) among all samples (Figure 4H). These results suggest that MBD1 mutation harbors AID signatures and underscore MBD1 as a pivotal target for both BCL6 and AID in promoting B lymphomagenesis.

2.5. Increased Sensitivity of MBD1‐Depleted Cells to Chemotherapeutics Targeting DNA Replication and Reduced Tumorigenicity in Mouse Xenograft Model

The increased levels of γH2AX signal at the BCL6‐MTC in HU‐treated shMBD1 cells indicates increased susceptibility of MBD1‐depleted cells to chemotherapeutic agents inducing replication stress (Figure 3D). To determine if shMBD1‐Raji cells exhibit higher susceptibility to other chemotherapeutic agents, we treated shSCR‐Raji and shMBD1‐Raji cells with gemcitabine or etoposide (Figure 5A). Control ShSCR‐Raji cells showed modest sensitivity to HU and Gemcitabine, and greater sensitivity to etoposide (Figure 5A). However, depletion of MBD1 led to significantly greater sensitivity to all drugs tested: the viability in shMBD1‐Raji cells was further reduced to 60%, 81.5% and 40.5% compared to shSCR‐Raji cells treated with HU, gemcitabine or etoposide (Figure 5A). These results suggest that MBD1‐deficient cells are more sensitive to genotoxic agents than shSCR‐Raji and indicate a role of MBD1 in promoting cell viability in response to DNA replication stress/genotoxic stress.

FIGURE 5.

FIGURE 5

Reduced tumorigenicity of MBD1 depleted Raji cells in mouse xenograft model‐ (A) Cell viability of shSCR‐Raji and shMBD1‐Raji cells treated with DMSO, HU (10 mM), Gemcitabine (20 mM), and Etoposide (50 μM) for 24 hours. Sidak's multiple comparison test: p < 0.001 for shSCR‐Raji (HU) versus shMBD1‐Raji (HU); p < 0.0001 for shSCR‐Raji (Gemcitabine) versus shMBD1‐Raji (Gemcitabine); p = 0.0005 for shSCR‐Raji (Etoposide) versus shMBD1‐Raji (Etoposide) (B) Tumor xenografts were established by subcutaneously injecting 1 million shSCR‐Raji or shMBD1‐Raji cells mixed with 200 μL of Matrigel above the right front limb of nude mice to reach the subcutaneous pocket. Mice were monitored for tumor growth over a period of 55 days. The sample size consisted of n = 4 for the shSCR‐Raji group (Upper panel) and n = 5 for the shMBD1‐Raji group (lower panel) (C) Tumor xenografts from shSCR‐Raji and shMBD1‐Raji cells injected into nude mice (n = 4 for shSCR‐Raji; n = 5 for shMBD1‐Raji) (D) Violin plot showing the weight range of tumors in shSCR‐Raji and shMBD1‐Raji groups. Unpaired t‐test: p = 0.0086. (E) Proposed model of MBD1 regulation during germinal center B‐cell development, lymphomagenesis. MBD1 and BCL6 expression exhibit an inverse relationship during GC reaction. In GC B cells, where BCL6 expression reaches its peak, MBD1 expression is maintained at levels sufficient to restrain excessive BCL6 transcription, thereby fine‐tuning the GC program. This regulatory balance may permit BCL6‐dependent survival of GC B cells while limiting aberrant transcriptional activity, potentially limiting the development of GC‐derived DLBCL subtypes. In parallel, MBD1 may suppresses genomic‐instabilities thus reducing BCL6‐translocations. As B cells differentiate into memory and plasma cells, BCL6 expression declines whereas MBD1 expression increases. Elevated MBD1 may regulate terminal differentiation through regulation of key transcription factors, including IRF4 and PRDM1 (BLIMP1) (F) Proposed therapeutic model of synthetic lethality following MBD1 inhibition. Targeting MBD1 sensitizes lymphoma cells or cancer cells to DNA replication stress and DNA‐damaging chemotherapeutic agents, resulting in synthetic lethality. Loss of MBD1 compromises the cellular capacity to tolerate replication‐associated genomic stress, thereby enhancing the cytotoxic effects of DNA replication inhibitors, making MBD1 a potential novel chemotherapeutic target.

Finally, to test whether tumorigenicity is altered in MBD1‐silenced cells, we performed xenograft experiments by injecting shSCR‐Raji and shMBD1‐Raji cells into nude mice (Figure 5B–D). Mice injected with shSCR‐Raji cells exhibited mean tumor weight of 0.62 gm (Figure 5D). In contrast, the mean weight of shMBD1‐Raji tumors was 0.15 gm (Figure 5D). This reduction in tumor weight of shMBD1‐Raji xenograft was significantly lower than those in shSCR‐Raji cells (Figure 5C, D). This result suggests that MBD1 is required for the tumorgenicity of Raji cells in vivo (Figure 5E, C, D). It is possible that accumulative genotoxic stress may surpass the threshold cells can tolerate in shMBD1‐Raji cells therefore perturbing the cellular proliferation, tumorgenicity and tumor growth (Figure 5B,C, D).

2.6. Correlation of MBD1 and IRF4 Expression in B‐Lymphoma

Human AGCBs induce MBD1 expression in GC B cells undergoing differentiation into plasma and memory B cells while BCL6 expression was reduced at similar stages, suggesting that the upregulation of MBD1 in differentiating GC B cells is correlated with BCL6 downregulation (Figure 1C). We hypothesized that MBD1 could regulate the expression of genes in GC B cells undergoing terminal differentiation. To define the relationship between MBD1 mutation status and IRF4 expression, we explored their correlation in samples from TCGA datasets and the Gene Expression Profiling Interactive Analysis (GEPIA) (Supporting Information S1). IRF4 expression was highest in DLBCL samples exhibiting MDB1:gain status, suggesting a positive correlation between MBD1 and IRF4 expression (Supporting Information S1). Moreover, DLBCL tumors from GEPIA also exhibited a positive correlation in MBD1 and IRF4 (Supporting Information S1). Since IRF4 expression is induced in plasma B cells, we examined whether mouse plasma B cells exhibit higher Mbd1 expression than activated GC B cells (Supporting Information S1). Interestingly, Mbd1 expression was increased in mouse plasma B cells, when IRF4 is high (Supporting Information S1) [41], suggesting that Mbd1 induction in plasma B cells is coordinated with Irf4 expression. We also measured the effect of MBD1 knockdown on the expression of two key genes determining GC B cell fate, IRF4 and PRDM1 (Supporting Information S1). MBD1 depletion in Raji cells resulted in increased expression of IRF4 and PRDM1, indicating a negative regulatory role of MBD1 in their transcription (Supporting Information S1). Conversely, overexpression of MBD1 led to a reduction in IRF4 expression (Supporting Information S1), suggesting that MBD1 plays a critical role in IRF4 regulation.

Finally, we analyzed if MBD1 depletion affects the expression of key factors regulating GC B cell activation and survival including essential homing receptors (CXCR4, CXCR5, CCR7, CXCR4, CXCR5, S1PR2, P2RY8), genes encoding the components of migration machinery (RAC1, RHOA), BCL6 targets, (IRF7, IFNGR1, STAT1), or genes encoding factors regulating BCL6 turnover, (FBXO11, XBP1) (Supporting Information S1). Among the GC B cell receptors tested, shMBD1‐Raji cells exhibited increased expression of CCR7, P2RY8 and STAT1 but reduced expression of RAC1 compared to shSCR‐Raji cells (Supporting Information S1). The expression of BCL6 targets or regulators S1PR2, RHOA, FBXO11, XBP1, IRF7 and IGNGR1 was not altered upon in the shMBD1‐Raji cells (Supporting Information S1). Overall, these observations suggest that MBD1 can affect the position of B cells within the GC compartment by regulating the expression levels of CCR7, CXCR5, and CXCR4 in cells belonging to T‐B border (T‐B), light zone (LZ), and dark zone (DZ), respectively. MBD1's regulation of RAC1 and P2RY8 may also affect the dynamic migration of GC B cells, given a role of RAC1 and P2RY8 in GC B cell LZ‐migration [42, 43].

3. Discussion

3.1. BCL6 Translocations and Reduced Tumorigenicity in MBD1‐Depleted Cancer Cells

The BCL6‐MTC contains highly conserved CpG methylated sites on it intron 1 (Figure 2D) [26], therefore MBD1 and other DNA methylation binding proteins MBD2 and MBD3 may bind to BCL6‐MTC in a DNA methylation‐dependent manner (Figure 2E, F and Supporting Information S1). MBD1 binding can help recruit histone deacetylases like SUV39H for transcriptional repression at the BCL6 locus [5, 24, 34, 44]. MBD1‐dependent BCL6 suppression may take place in GC B cells undergoing terminal differentiation, correlating with reduced BCL6 levels in plasma B cells [45]. MBD1 binding to the BCL6 promoter is observed in HCT116 cells (ChIP atlas:DRX021118; https://chip‐atlas.org/view?id = DRX021118; Figure 2F). MBD1 binding may contribute to reduced genomic instability at the BCL6 locus preventing BCL6 rearregements as well as global rearrangements in GC B cells. On the other hand, shMBD1‐Raji cells exhibited higher sensitivity to HU, gemcitabine, and etoposide as well as reduced tumorogencity in xenograft models than shSCR‐Raji cells (Figure 5A). These findings align with a recent independent report showing MBD1 regulates survival during DNA replication stress [46]. The reduced tumorigenicity observed in shMBD1‐Raji xenografts could be due to increased accumulation of genomic instability compared to shSCR‐Raji cells (Figure 5C, D, F). Together, these results suggest that MBD1 promotes cancer cell survival, making MBD1 an attractive therpautic target. The detailed mechanims regulating MBD1‐mediated cell survival should be explored in future studies to rationally design effective combination therapy for MBD1 targeting in cancers. MBD1 alteration may also be an effective biomarker for tumor exhibiting higher BCL6 and IRF4 expression which also correlates with increased chemotherapeutic resistance [19, 47].

3.2. AID Mutation Signatures in MBD1‐Mutant Somatic Cancers

AID‐mediated mutagenesis drives B lymphomagenesis by mutating tumor suppresor genes and superenhancers, altering the binding of transcription factors and chromatin remodelers [10]. Interestingly, about 39% of single base‐pair alterations in MBD1 were C > T transition mutations (Figure 4F), a classic signature of AID‐dependent mutagenesis (Figure 4H). Thus AID‐dependent MBD1 mutagenesis could serve as an important adapation for B lymphoma selection during the GC reaction.

3.3. Mechanism of MBD1 Induction in GC B Cells Undergoing Plasma B Cell Differentiation

The increased MBD1 expression we observed in HU‐treated Raji cells correlates with elevated MBD1 levels in human AGCBs classified as pre‐memory and plasma B cells [32] (Figure 1C). Similarly, mouse GC B cells undergoing PC differentiation also exhibit increased Mbd1 levels (Supporting Information S1). This evidence highlights MBD1's role in GC B cell differentiation in both humans and mice (Figure 1C, Supporting Information S1). The transition from GC B cells to plasma cells necessitates dynamic DNA hypomethylation in genes involved in GC B differentiation [25, 48], suggesting that MBD1 expression could be induced by hypomethylation of its promoter. This is supported by our observations of DNA methylation loss in the Mbd1 promoter in Sca1‐BCL6 Δ mice (Supporting Information S1), and the induction of MBD1 expression in the presence of the methylation inhibitor AZA in DLBCLs (Supporting Information S1) [39]. MBD1 induction during plasma B cell differentiation could also be related to BCL6 downregulation since we provide evidence that BCL6 suppresses MBD1 expression by binding to its promoter (Figure 4A, D). The increased MBD1 levels in plasma B cells might also result from reduced EZH2‐dependent silencing of the MBD1 promoter, as this suppresion is removed in GC B cell undergoing differentiation [23]. In contrast, induction of MBD1 in cancer cells undergoing chemotherapy with DNA replication inbibitors suggests a programmed activation of MBD1 transcription which potentially regulates cell survival (Supporting Information S1).

3.4. MBD1 and IRF4 Expression in Differentiating GC B Cells and B‐Lymphomas

We found that MBD1 and IRF4 expression is positively correlated in DLBCL (Supporting Information S1). On the other hand, Mbd1 and Irf4 were induced in mouse and human GC B cells undergoing plasma B differentiation (Figure 1C, Supporting Information S1), indicating a synchronous upregulation of MBD1 and IRF4 in GC B cells undergoing differentiation. The mechanims of MBD1 and IRF4 co‐expression in GC B cells is not yet clear. It is possible that MBD1 and IRF4 could be induced by decreased BCL6 levels in differetiating GC B cells (Figure 5E). Also, MBD1 in differentiating GC B cells could suppreses BCL6, allowing for higher IRF4 expression and plasma B cell differentiation. Importantly, IRF4 suppression is mediated by EZH2‐dependent H3K27me3 [23]. Is is possible that MBD1 may counteract EZH2 by recruiting histone deacetylases and transcriptional repressors to the EZH2 promoter [5, 24, 34, 44]. This positions MBD1 as a novel regulator of GC B cell differentiation regulating the plasma B cell differentiation [49, 50]. Another possibility is that MBD1 may work alongside BCL6 and BACH2, both of which negatively regulate IRF4 and PRDM1 [45], allowing for homestasis of IRF4 expression. This is consistent with our finding where MBD1 depletion in Raji cells induced IRF4 expression (Supporting Information S1). However MBD1‐mediated suppression of IRF4 in Raji lymphoma cells may be distinct from WT B cells residing in the GC compartment, therefore we can not rule out the possibility that MBD1 promotes plasma B cell differentiation.

MBD1‐mediated regulation of IRF4 and PRDM1 and its significance in the suppression and/or survival of B‐lymphomas remains to be elucidated. We propose two potential regulatory scenarios for MBD1‐dependent IRF4 regulation depending on the stage of tumorigenesis. In the first scenario, GC B cells with high‐affinity B cell receptors (BCR) might experience higher MBD1 expression, facilitating normal GC B cell differentiation. In the second, B lymphoma precursors and unproductive GC B cells with damaged BCRs may experience reduced MBD1 expression blocking the expression of plasma B cell differentiation genes including IRF4 and mitigating the risk of malignant transformation into multiple myeloma (MM). Thus, MBD1 may differentially regulate plasma B cell differentiation in normal plasma B cells versus B lymphoma precursors.

In summary, our findings suggest that MBD1 is a key regulator of GC B cell differentiation, acting independently of both BCL6 and BACH2. Additionally, MBD1 may prevent BCL6 rearrangement in GC B cells, thereby inhibiting the formation of GCDBL. Finally, considering MBD1's role in cell viability in cells treated with DNA‐replication inhibitors, it could serve as a novel therapeutic target for relapsed and refractory B cell lymphomas that are resistant to chemotherapy.

4. Material and Methods

4.1. Human B Lymphoma Cultures, Treatment With Inhibitors and Plasmid Overexpression

Human B lymphoma Cell lines Raji were obtained from Department of Hematology, Kyoto university hospital, RIKEN cell bioresources, Japan. MCF7, MDA‐MB‐231, SUM149PT, MDA‐MB‐436 were available at the Cancer Research Institute, Kanazawa University. Raji and HEK293 T cells were cultured in RPMI medium (30,264–85, Nakalai) including the 10% fetal bovine serum (FBS), 1% Penicillin‐Streptomycin & Amphotericin B solution (Waken# 161–23181) at 37°C maintaining the 5% CO2 concentration. Breast cancer cell lines MDA‐MB‐231, MDA‐MB‐436, MCF‐7, U2OS cells were cultured DMEM, supplemented with 10% fetal bovine serum (FBS), 2 mM L‐glutamine and 1 × Penicillin‐Streptomycin (Life Technologies). SUM149PT cell were cultivated in Ham's F‐12 medium containing 5% FBS, 10 μg/mL insulin, 1% Penicillin‐Streptomycin & Amphotericin B solution (Waken# 161–23181), and supplemented with 0.5 μg/mL hydrocortisone. Cells were propagated at 37°C in a 5% CO2 atmosphere. FX1 (S8591, Selleck Japan), Gemcitabine (S1714, Selleck, Japan), Hydroxyurea (085–06653, Wako, Japan) and etoposide (055–0843, Wako, Japan) treatments were performed as described in figure legends. The preparation of these chemical was performed as per the manufacturer's recommendation. For the overexpression of pcDNA and pcDNA‐BCL6 plasmids in Raji cells, the cells were cultured for 2 hours in RPMI medium containing 1% FBS, followed by transfection with Lipofectamine 3000 according to the manufacturer's instructions (Thermo Fisher, Japan). Twelve hours post‐transfection, the total FBS concentration was adjusted to 10% by adding additional FBS. Cells were collected for analysis at the indicated time points.

4.2. shRNA Preparation and Lentivirus Induced Knockdown

HEK293 T cells were cultured at the 70% confluency. For transfection mixture and lentivirus preparation, viral vector PAX2, pVSVG and pLKO.1 plasmids (1 μg of each) cloned with respective shRNA sequences were mixed in 200 μL of warm Optimem medium followed by addition of 3 μg of PEI reagent. The mix incubated for 20 minutes at room temperature follwed by addition to the H3K293 T cells. Fresh medium DMEM medium was replaced to transfected cells after 24 hours. The H3K293 T cells were then cultured for additional 48 hours. The culture medium containing the lentivirus particles was then filtered using the 0.45‐micron filters and the filtrate was aliquoted and stocked in −80°C until used. For stable transfection of Raji cells, 1 million cells were transduced with 400 μL of filtered lentiviral supernatant follwed by selection with puromycin (2000 ng/mL) for 2 days. The 2 days selected cells were again washed and selected for additional 2 days with similar dose of Puromycin.

4.3. RNA Isolation, cDNA Preparation, qPCR Analysis and RNA‐Seq Analysis

Total RNA was isolated with NucleoSpin RNA Plus (#740984.50, Takara) and cDNA was prepared with the RT‐PCR mix (FSQ#201, Toyobo Japan). RNA was isolated from RIVA and U2OS cells using TRIzol reagent (Invitrogen, #15596026), followed by DNase treatment (NEB, #M0570). cDNA was then synthesized using the HiScript cDNA Synthesis Kit (Vazyme, R312). qPCR was performed using the SYBR mix as per manufacturer's instructions (Thunderbolt QPS‐201, Toyobo Japan). Data were analyzed using the delta‐delta Ct method indicating relative transcript levels to the respective control samples. ACTB gene signals were used as a housekeeping gene. Computational analyses of RNA‐seq were performed as described [31]. The RNA‐seq and ChIP data set are under GEO accession numbers GSE242375 and GSE242936 respectively. MBD1 mRNA expression in the Acute myeloid patient sample and DLBCL line treated with BCL6 inhibitors were obtained from GSE133112 [51] and GSE94099 [52] respectively.

4.4. Chromatin Immunoprecipitation (ChIP), ChIP‐Seq Library Preparation, Computational Analysis of ChIP‐Seq Data, and MBD1 Binding to BCL6 Locus withChIP‐Atlas

ChIP was performed as previously described [6]. The ChIP‐seq library was prepared as per the manufacturer's instruction (Thruplex DNA‐seq Kit, R400675, Takarabio, Japan). ChIP‐seq was performed as described [31]. The ChIP‐seq files were deposited to NCBI under accession number GSE242936. DNA CpG methylation status of mouse Mbd1 promoter in B cells were reanalyzed using the wild type and Sca1‐Bcl6 Δ mice data [25, 48]. Binding of MDB1 on BCL6 promoter was examined using the dataset # DRX021118 available on ChIP‐atlas (https://chip‐atlas.org/view?id=DRX021118).

4.5. BCL6 Binding Motif, Luciferase Assay and COSMIC Datasets

The luciferase plasmids depleting the BCL6 binding motifs on the MBD1 promoter were constructed using the Q5 site‐directed mutagenesis kit (NEB# E0554S) and as per the manufacturer's instructions. Primers used for the site‐directed mutagenesis are listed in Table 1. The plasmids were transfected in the HEK293 T cells for 24 hours using lipofectamine‐3000 (ThermoFisher# L3000001). After 24 hours, cell lysates were prepared using Promega luciferase assay kit. The Catalog of somatic mutation in cancers (COSMIC) datasets were searched for cancers with MBD1 mutation and the mutation signatures were classified and filtered for the AID signatures.

4.6. Tumor Xenograft in Nude Mouse and Estimation of Cell Viability

For tumor xenograft, and injection of Raji into nude mice, 0.6 million cells were suspended in 200 μL of Matrigel and medium mix, suspended well and subcutaneously injected into the right arm of the nude mice. The tumor growth was observed for 55 days and mice were sacrificed, followed by isolation of tumor and weighing. Treatment with chemotherapeutics and cell viability was tested by the cell viability kit (Dojindo Japan) as per the manufacturer's instructions.

Author Contributions

S.K.G designed the original hypothesis, performed experiments, and analyzed data. K.O analyzed Mbd1 transcripts in mouse plasma B cells and provided constructive feedback. S.K.G wrote the manuscript. JHB critically read the manuscript and provided constructive feedback. H.K provided constructive feedback and discussed the MS. KJM provided constructive feedback and conceptual insights for the revision of the MS. All authors read and agreed to manuscript.

Funding

This work is supported Grant No. 21K16142 from Japan Society for the Promotion sciences (JSPS) to SKG. This research was also supported by the Institute for Basic Science (IBS‐R022‐D1).

Ethics Statement

The use of human/animal samples (wherever applicable) was approved as per Institutional ethical board.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting Information S1

HON-44-e70227-s001.docx (21.7KB, docx)

Contributor Information

Santosh Kumar Gothwal, Email: gothwal@ibs.re.kr, Email: santoshgothwalbio@gmail.com.

Kyungjae Myung, Email: kmyung@ibs.re.kr.

Data Availability Statement

Research data are not shared.

References

  • 1. Allen C. D., Ansel K. M., Low C., et al., “Germinal Center Dark and Light Zone Organization Is Mediated by CXCR4 and CXCR5,” Nature Immunology 5, no. 9 (2004): 943–952, 10.1038/ni1100. [DOI] [PubMed] [Google Scholar]
  • 2. Victora G. D. and Nussenzweig M. C., “Germinal Centers,” Annual Review of Immunology 40, no. 1 (2022): 413–442, 10.1146/annurev-immunol-120419-022408. [DOI] [PubMed] [Google Scholar]
  • 3. Muramatsu M., Kinoshita K., Fagarasan S., Yamada S., Shinkai Y., and Honjo T., “Class Switch Recombination and Hypermutation Require Activation‐Induced Cytidine Deaminase (AID), A Potential RNA Editing Enzyme,” Cell 102, no. 5 (2000): 553–563, 10.1016/s0092-8674(00)00078-7. [DOI] [PubMed] [Google Scholar]
  • 4. Roco J. A., Mesin L., Binder S. C., et al., “Class‐Switch Recombination Occurs Infrequently in Germinal Centers,” Immunity 51, no. 2 (2019): 337–350.e7, 10.1016/j.immuni.2019.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Lu Z., Tsai A. G., Akasaka T., et al., “BCL6 Breaks Occur at Different AID Sequence Motifs in Ig‐BCL6 and non‐Ig‐BCL6 Rearrangements,” Blood 121, no. 22 (2013): 4551–4554, 10.1182/blood-2012-10-464958. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Gothwal S. K., “BRD2 Promotes Antibody Class Switch Recombination by Facilitating DNA Repair in Collaboration With NIPBL,” Nucleic Acids Research (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Shaffer A. L., Emre N. C. T., Lamy L., et al., “IRF4 Addiction in Multiple Myeloma,” Nature 454, no. 7201 (2008): 226–231, 10.1038/nature07064. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Kaiser L. M., Hunter Z. R., Treon S. P., and Buske C., “CXCR4 in Waldenström’s Macroglobulinema: Chances and Challenges,” Leukemia 35, no. 2 (2021): 333–345, 10.1038/s41375-020-01102-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Kurtova A. V., Tamayo A. T., Ford R. J., and Burger J. A., “Mantle Cell Lymphoma Cells Express High Levels of CXCR4, CXCR5, and VLA‐4 (Cd49D): Importance for Interactions With the Stromal Microenvironment and Specific Targeting,” Blood 113, no. 19 (2009): 4604–4613, 10.1182/blood-2008-10-185827. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Bal E., Kumar R., Hadigol M., et al., “Super‐Enhancer Hypermutation Alters Oncogene Expression in B Cell Lymphoma,” Nature 607, no. 7920 (2022): 808–815, 10.1038/s41586-022-04906-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Calpe E., Purroy N., Carpio C., et al., “ZAP‐70 Promotes the Infiltration of Malignant B‐Lymphocytes Into the Bone Marrow by Enhancing Signaling and Migration After CXCR4 Stimulation,” PLoS One 8, no. 12 (2013): e81221, 10.1371/journal.pone.0081221. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Dhordain P., Quief S., Lantoine D., et al., “The LAZ3(BCL‐6) Oncoprotein Recruits a SMRT/mSIN3A/Histone Deacetylase Containing Complex to Mediate Transcriptional Repression,” Nucleic Acids Research 26, no. 20 (1998): 4645–4651, 10.1093/nar/26.20.4645. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Chang C. C., Ye B. H., Chaganti R. S., and Dalla‐Favera R., “BCL‐6, A POZ/Zinc‐Finger Protein, Is a Sequence‐Specific Transcriptional Repressor,” Proceedings of the National Academy of Sciences of the United States of America 93, no. 14 (1996): 6947–6952, 10.1073/pnas.93.14.6947. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Phan R. T., Saito M., Basso K., Niu H., and Dalla‐Favera R., “BCL6 Interacts With the Transcription Factor Miz‐1 to Suppress the Cyclin‐Dependent Kinase Inhibitor p21 and Cell Cycle Arrest in Germinal Center B Cells,” Nature Immunology 6, no. 10 (2005): 1054–1060, 10.1038/ni1245. [DOI] [PubMed] [Google Scholar]
  • 15. Akkaya M., Kwak K., and Pierce S. K., “B Cell Memory: Building Two Walls of Protection Against Pathogens,” Nature Reviews Immunology 20, no. 4 (2020): 229–238, 10.1038/s41577-019-0244-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Ye B. H., Lista F., Lo Coco F., et al., “Alterations of a Zinc Finger‐Encoding Gene, BCL‐6, in Diffuse Large‐Cell Lymphoma,” Science 262, no. 5134 (1993): 747–750, 10.1126/science.8235596. [DOI] [PubMed] [Google Scholar]
  • 17. Ci W., Polo J. M., and Melnick A., “B‐cell Lymphoma 6 and the Molecular Pathogenesis of Diffuse Large B‐cell Lymphoma,” Current Opinion in Hematology 15, no. 4 (2008): 381–390, 10.1097/moh.0b013e328302c7df. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Ruggieri S., Tamma R., Marzullo A., et al., “Translocation of the Proto‐Oncogene Bcl‐6 in Human Glioblastoma Multiforme,” Cancer Letters 353, no. 1 (2014): 41–51, 10.1016/j.canlet.2014.06.017. [DOI] [PubMed] [Google Scholar]
  • 19. Duy C., Hurtz C., Shojaee S., et al., “BCL6 Enables Ph+ Acute Lymphoblastic Leukaemia Cells to Survive BCR‐ABL1 Kinase Inhibition,” Nature 473, no. 7347 (2011): 384–388, 10.1038/nature09883. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Walker S. R., Liu S., Xiang M., et al., “The Transcriptional Modulator BCL6 as a Molecular Target for Breast Cancer Therapy,” Oncogene 34, no. 9 (2015): 1073–1082, 10.1038/onc.2014.61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Xu L., Chen Y., Dutra‐Clarke M., et al., “BCL6 Promotes Glioma and Serves as a Therapeutic Target,” Proceedings of the National Academy of Sciences of the United States of America 114, no. 15 (2017): 3981–3986, 10.1073/pnas.1609758114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Shaknovich R., Cerchietti L., Tsikitas L., et al., “DNA Methyltransferase 1 and DNA Methylation Patterning Contribute to Germinal Center B‐Cell Differentiation,” Blood 118, no. 13 (2011): 3559–3569, 10.1182/blood-2011-06-357996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Béguelin W., Popovic R., Teater M., et al., “EZH2 Is Required for Germinal Center Formation and Somatic EZH2 Mutations Promote Lymphoid Transformation,” Cancer Cell 23, no. 5 (2013): 677–692, 10.1016/j.ccr.2013.04.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Villa R., Morey L., Raker V. A., et al., “The methyl‐CpG Binding Protein MBD1 Is Required for PML‐RARalpha Function,” Proceedings of the National Academy of Sciences of the United States of America 103, no. 5 (2006): 1400–1405, 10.1073/pnas.0509343103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Caron G., Hussein M., Kulis M., et al., “Cell‐Cycle‐Dependent Reconfiguration of the DNA Methylome During Terminal Differentiation of Human B Cells Into Plasma Cells,” Cell Reports 13, no. 5 (2015): 1059–1071, 10.1016/j.celrep.2015.09.051. [DOI] [PubMed] [Google Scholar]
  • 26. Lai A. Y., Fatemi M., Dhasarathy A., et al., “DNA Methylation Prevents CTCF‐Mediated Silencing of the Oncogene BCL6 in B Cell Lymphomas,” Journal of Experimental Medicine 207, no. 9 (2010): 1939–1950, 10.1084/jem.20100204. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Bell A. C. and Felsenfeld G., “Methylation of a CTCF‐dependent Boundary Controls Imprinted Expression of the Igf2 Gene,” Nature 405, no. 6785 (2000): 482–485, 10.1038/35013100. [DOI] [PubMed] [Google Scholar]
  • 28. Hark A. T., Schoenherr C. J., Katz D. J., Ingram R. S., Levorse J. M., and Tilghman S. M., “CTCF Mediates methylation‐Sensitive Enhancer‐blocking Activity at the H19/Igf2 Locus,” Nature 405, no. 6785 (2000): 486–489, 10.1038/35013106. [DOI] [PubMed] [Google Scholar]
  • 29. Kosak S. T., Skok J. A., Medina K. L., et al., “Subnuclear Compartmentalization of Immunoglobulin Loci During Lymphocyte Development,” Science 296, no. 5565 (2002): 158–162, 10.1126/science.1068768. [DOI] [PubMed] [Google Scholar]
  • 30. Jhunjhunwala S., van Zelm M. C., Peak M. M., et al., “The 3D Structure of the Immunoglobulin Heavy‐Chain Locus: Implications for Long‐Range Genomic Interactions,” Cell 133, no. 2 (2008): 265–279, 10.1016/j.cell.2008.03.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Gothwal S. K. and Barlow J. H., “Stress Induced Signaling Pathways in Burkitt Lymphoma Play Novel Mechanisms in Fate Determination and Pathogenesis of Germinal Center‐Derived B‐Lymphomas,” bioRxiv (2024): 2024.12.19.628635. [Google Scholar]
  • 32. Holmes A. B., Corinaldesi C., Shen Q., et al., “Single‐Cell Analysis of germinal‐center B Cells Informs on Lymphoma Cell of Origin and Outcome,” Journal of Experimental Medicine 217, no. 10 (2020): e20200483, 10.1084/jem.20200483. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Quentmeier H., Pommerenke C., Dirks W. G., et al., “The LL‐100 Panel: 100 Cell Lines for Blood Cancer Studies,” Scientific Reports 9, no. 1 (2019): 8218, 10.1038/s41598-019-44491-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Fujita N., Watanabe S., Ichimura T., et al., “Methyl‐CpG Binding Domain 1 (MBD1) Interacts With the Suv39h1‐HP1 Heterochromatic Complex for DNA Methylation‐Based Transcriptional Repression,” Journal of Biological Chemistry 278, no. 26 (2003): 24132–24138, 10.1074/jbc.m302283200. [DOI] [PubMed] [Google Scholar]
  • 35. Clouaire T., de las Heras J. I., Merusi C., and Stancheva I., “Recruitment of MBD1 to Target Genes Requires Sequence‐specific Interaction of the MBD Domain With Methylated DNA,” Nucleic Acids Research 38, no. 14 (2010): 4620–4634, 10.1093/nar/gkq228. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Ci W., Polo J. M., Cerchietti L., et al., “The BCL6 Transcriptional Program Features Repression of Multiple Oncogenes in Primary B Cells and Is Deregulated in DLBCL,” Blood 113, no. 22 (2009): 5536–5548, 10.1182/blood-2008-12-193037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Cardenas M. G., Yu W., Beguelin W., et al., “Rationally Designed BCL6 Inhibitors Target Activated B Cell Diffuse Large B Cell Lymphoma,” Journal of Clinical Investigation 126, no. 9 (2016): 3351–3362, 10.1172/jci85795. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Saito M., Gao J., Basso K., et al., “A Signaling Pathway Mediating Downregulation of BCL6 in Germinal Center B Cells Is Blocked by BCL6 Gene Alterations in B Cell Lymphoma,” Cancer Cell 12, no. 3 (2007): 280–292, 10.1016/j.ccr.2007.08.011. [DOI] [PubMed] [Google Scholar]
  • 39. Green M. R., Vicente‐Dueñas C., Romero‐Camarero I., et al., “Transient Expression of Bcl6 Is Sufficient for Oncogenic Function and Induction of Mature B‐cell Lymphoma,” Nature Communications 5, no. 1 (2014): 3904, 10.1038/ncomms4904. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Oprea M. and Perelson A. S., “Somatic Mutation Leads to Efficient Affinity Maturation When Centrocytes Recycle Back to Centroblasts,” Journal of Immunology 158, no. 11 (1997): 5155–5162, 10.1371/journal.pcbi.1000800. [DOI] [PubMed] [Google Scholar]
  • 41. Ochiai K., Shima H., Tamahara T., et al., “Accelerated plasma‐cell Differentiation in Bach2‐Deficient Mouse B Cells Is Caused by Altered IRF4 Functions,” EMBO Journal 43, no. 10 (2024): 1947–1964, 10.1038/s44318-024-00077-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Margiotta A. and Bucci C., “Coordination Between Rac1 and Rab Proteins: Functional Implications in Health and Disease,” Cells 8, no. 5 (2019): 396, 10.3390/cells8050396. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Green J. A. and Cyster J. G., “S1PR2 Links Germinal Center Confinement and Growth Regulation,” Immunological Reviews 247, no. 1 (2012): 36–51, 10.1111/j.1600-065x.2012.01114.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44. Fujita N., Shimotake N., Ohki I., et al., “Mechanism of Transcriptional Regulation by Methyl‐CpG Binding Protein MBD1,” Molecular and Cellular Biology 20, no. 14 (2000): 5107–5118, 10.1128/mcb.20.14.5107-5118.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Huang C., Geng H., Boss I., Wang L., and Melnick A., “Cooperative Transcriptional Repression by BCL6 and BACH2 in Germinal Center B‐Cell Differentiation,” Blood 123, no. 7 (2014): 1012–1020, 10.1182/blood-2013-07-518605. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Yu G., Xiong Y., Xu Z., et al., “MBD1 Protects Replication Fork Stability by Recruiting PARP1 and Controlling Transcription‐Replication Conflicts,” Cancer Gene Therapy 31, no. 1 (2024): 94–107, 10.1038/s41417-023-00685-0. [DOI] [PubMed] [Google Scholar]
  • 47. Amanda S., Tan T. K., Ong J. Z. L., et al., “IRF4 Drives Clonal Evolution and Lineage Choice in a Zebrafish Model of T‐Cell Lymphoma,” Nature Communications 13, no. 1 (2022): 2420, 10.1038/s41467-022-30053-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48. Kulis M., Merkel A., Heath S., et al., “Whole‐Genome Fingerprint of the DNA Methylome During Human B Cell Differentiation,” Nature Genetics 47, no. 7 (2015): 746–756, 10.1038/ng.3291. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Kuo T. C., Shaffer A. L., Haddad J. Jr., Choi Y. S., Staudt L. M., and Calame K., “Repression of BCL‐6 Is Required for the Formation of Human Memory B Cells in Vitro,” Journal of Experimental Medicine 204, no. 4 (2007): 819–830, 10.1084/jem.20062104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Lin K. I., Tunyaplin C., and Calame K., “Transcriptional Regulatory Cascades Controlling Plasma Cell Differentiation,” Immunological Reviews 194, no. 1 (2003): 19–28, 10.1034/j.1600-065x.2003.00040.x. [DOI] [PubMed] [Google Scholar]
  • 51. Kawabata K. C., Zong H., Meydan C., et al., “BCL6 Maintains Survival and Self‐Renewal of Primary Human Acute Myeloid Leukemia Cells,” Blood 137, no. 6 (2021): 812–825, 10.1182/blood.2019001745. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52. Kerres N., Steurer S., Schlager S., et al., “Chemically Induced Degradation of the Oncogenic Transcription Factor BCL6,” Cell Reports 20, no. 12 (2017): 2860–2875, 10.1016/j.celrep.2017.08.081. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information S1

HON-44-e70227-s001.docx (21.7KB, docx)

Data Availability Statement

Research data are not shared.


Articles from Hematological Oncology are provided here courtesy of Wiley

RESOURCES