B cell development and Ig rearrangement are governed by cell type– and developmental stage–specific transcription factors. PU.1 and Spi-B are E26-transformation–specific transcription factors that are critical for B cell differentiation. To determine whether PU.1 and Spi-B are required for B cell development in the bone marrow, Spi1 (encoding PU.1) was conditionally deleted in B cells by Cre recombinase under control of the Mb1 gene in Spib (encoding Spi-B)–deficient mice. Combined deletion of Spi1 and Spib resulted in a lack of mature B cells in the spleen and a block in B cell development in the bone marrow at the small pre-B cell stage. To determine target genes of PU.1 that could explain this block, we applied a gain-of-function approach using a PU.1/Spi-B–deficient pro-B cell line in which PU.1 can be induced by doxycycline. PU.1-induced genes were identified by integration of chromatin immunoprecipitation–sequencing and RNA-sequencing data. We found that PU.1 interacted with multiple sites in the Igκ locus, including Vκ promoters and regions located downstream of Vκ second exons. Induction of PU.1 induced Igκ transcription and rearrangement. Upregulation of Igκ transcription was impaired in small pre-B cells from PU.1/Spi-B–deficient bone marrow. These studies reveal an important role for PU.1 in the regulation of Igκ transcription and rearrangement and a requirement for PU.1 and Spi-B in B cell development.

B cell development involves ordered rearrangement of Ig loci encoding H and L chain proteins that assemble into Abs capable of recognizing specific Ags. Stages of B cell development can be resolved using cell surface marker expression, as well as Ig gene rearrangement (13). Progenitor B (pro-B; also known as pre-BI) cells are generated from lymphoid progenitors and initiate D-J segment rearrangement of Igh H chain alleles. Rearrangement of Igh alleles is completed by V-DJ rearrangement to encode H chains that can pair with surrogate L chain proteins and be deposited on the cell surface as a pre-BCR. Pre-BCR signaling promotes proliferation of large pre-B cells (also known as pre-BII cells). Cessation of proliferation induces re-expression of the RAG proteins RAG1 and RAG2, whose interaction is widespread throughout the Ig loci at recombination signal sequences (RSSs) (4). Finally, successful pairing of Ig κ or λ L chain proteins with IgH proteins results in expression of a BCR in immature B cells (5). Immature B cells emigrate from the bone marrow (BM) to the spleen to complete their maturation (6).

The mouse Igκ locus contains 101 functional Vκ genes, 60 Vκ pseudogenes, 4 functional Jκ genes, 1 J pseudogene, and 1 Cκ gene (7, 8). The Igκ locus comprises >3 MB of genomic sequence (8). Recombination occurs in a developmental stage–specific order that is thought to be regulated at the level of chromatin accessibility to RAG proteins (7, 9). Accessibility is likely controlled at the level of transcription, as supported by two lines of evidence. First, deletions of enhancers within the Igκ locus impair transcription and Ig recombination (10, 11). Second, deletion of genes encoding histone-modifying enzymes or transcription factors reduce transcription and Ig recombination (8). Thus, Ig locus accessibility is regulated by transcription factor recruitment of the transcriptional machinery to regulatory regions within the Igκ locus.

B cell development is coordinated by the expression of a number of cell type– and developmental stage–specific transcription factors, including E2A, EBF, Pax5, Ikaros, PU.1, and Spi-B (12). PU.1 (encoded by Spi1) and Spi-B (encoded by Spib) are highly related transcription factors of the E26 transformation–specific (ETS) family (13). Several lines of evidence suggest that PU.1 might play an important role in the control of Igκ transcription. Igκ V region promoters, as well as the intronic and 3′ enhancers in the Igκ locus, contain predicted binding sites for PU.1 (1416). Chromatin immunoprecipitation–sequencing (ChIP-seq) analysis in pro-B cells reveals that PU.1 binding is widespread throughout the Igκ locus (8, 17). However, there has not been a clear demonstration of a role for PU.1 in Ig transcription, accessibility, or rearrangement in vivo. The ability of Spi-B to complement PU.1 function may have impeded a clear demonstration of PU.1 as a regulator of Igκ gene transcription (13, 18). We previously generated mice that delete Spi1 encoding PU.1 under control of CD19-Cre on a Spib−/− background (18). These mice had impaired development of follicular B cells and perturbation of BM B cell development, demonstrating a complementary role for PU.1 and Spi-B in B cell development and function (18, 19). However, CD19-Cre does not delete alleles efficiently in BM (20), precluding an examination of PU.1 and Spi-B function in early B cell development. We hypothesized that deletion of Spi1 and Spib during early B cell development would reveal a role for PU.1 and/or Spi-B in the transcription and rearrangement of Ig genes.

To test this hypothesis, Mb1+/CreSpi1lox/loxSpib−/− (Mb1-CreΔPB) mice were generated by crossing Spi1lox/loxSpib−/− mice to Mb1+/Cre mice. Mb1-Cre deletes alleles with high efficiency in BM (20), resulting in a Spib-deficient mouse that is expected to have a high frequency of Spi1 deletion during early B cell development. Analysis of adult mice showed that deletion of Spi1 and Spib in Mb1-CreΔPB mice resulted in the absence of IgM+ B cells in the spleen. In BM, there was a specific block in B cell development at the small pre-B cell to immature B cell transition that is marked by successful rearrangement of Ig L chain genes. To determine target genes of PU.1 that could explain this block, we applied a gain-of-function approach using a PU.1/Spi-B–deficient pro-B cell line in which PU.1 can be induced by doxycycline. PU.1-regulated genes were identified by integration of ChIP-seq and RNA-sequencing (RNA-seq) data. PU.1 interacted with 23,647 sites located near transcription start sites (TSSs) of genes involved in immune system development. Interestingly, we observed PU.1 interaction with multiple sites in Vκ gene promoters, as well as sites downstream of Vκ second exons located near RSSs. Induction of PU.1 resulted in increased transcription of Igκ V genes and Igκ rearrangement. Finally, we found that upregulation of Igκ transcription was impaired in PU.1 and Spi-B–deficient BM pre-B cells. These studies reveal an important role for PU.1 in Igκ transcription and rearrangement and a requirement for PU.1 and Spi-B in B cell development.

Mb1-Cre mice were described previously (20). Mb1-Cre mice were crossed with Spi1lox/loxSpib−/− to generate Mb1-CreΔPB mice. Mb1+/CreSpi1lox/loxSpib+/+ (Mb1-CreΔP) and Mb1+/CreSpi1+/+Spib−/− or Mb1+/+ Spi1+/+Spib−/− (ΔB) mice were used as experimental controls. C57BL/6 mice were purchased from Charles River Laboratories (Saint-Constant, QC, Canada). All experiments were performed in compliance with the Western University Council on Animal Care.

For spleen and BM analysis, cells were prepared from 6–10-wk-old Mb1-CreΔPB, Mb1-CreΔP, ΔB, and wild-type (WT) mice. RBCs were removed from single-cell suspensions using hypotonic lysis. Flow cytometric analyses were performed using an LSR II instrument (BD Immunocytometry Systems, San Jose, CA). Abs were purchased from eBioscience (San Diego, CA), BioLegend (San Diego, CA), or BD Biosciences (Mississauga, ON, Canada) and included PE–anti-CD19 (1D3), FITC–anti-BP-1 (6C3), allophycocyanin–anti-B220 (RA3-6B2), allophycocyanin–anti-IgM (II/ 41), PE–anti-Igκ (187.1), FITC–anti-IL-7R (A7R34), BV421–anti-B220 (RA3-6B2), PE–anti-BP-1 (6C3), FITC–anti-CD24 (M1/69), biotin–anti-CD43 (S7), and PE/Cy5 streptavidin. BM cell sorting was performed on a FACSAria with FACSDiva software (both from BD). Data were analyzed using FlowJo 9.7.4 software.

The 660BM and i660BM cell lines used in this study were described previously (19). Cells were cultured in IMDM (Wisent, QC, Canada) containing 5% IL-7–conditioned medium from the J558L–IL-7 cell line (21), 10% FBS (Wisent), 1× penicillin/streptomycin/l-glutamine (Lonza, Shawinigan, QC, Canada), and 5 × 10−5 M 2-ME (Sigma-Aldrich, St. Louis, MO). i660BM cells were maintained in 0.5 μg/ml puromycin (BioBasic, Markham, ON, Canada). Cell lines were maintained in 5% CO2 atmosphere at 37°C.

i660BM cells were induced for 48 h with doxycycline (70 ng/μl) in the presence of 5% IL-7–conditioned medium described above, and RNA was extracted using an RNeasy Kit (QIAGEN, ON, Canada). Uninduced cells were used as a control. Paired-end (mRNA-sequencing stranded) libraries were prepared using Illumina TrueSeq Adapters. Libraries were sequenced using an Illumina HiSEquation 2000 sequencer in paired-end mode. Data analysis was performed as described previously (22) using the tools available in Galaxy suite (23). Standard Illumina sequencing adaptors (5′-AGATCGGAAGAGC-3′) and short reads were removed using Trim Galore! V0.2.8.1 in mate-paired mode with trim low quality: 20; maximum allowed error rate: 0.1; discard reads that became shorter than length: 20. Trimmed FASTQ files were aligned to the mouse genome (mm10) using TopHat2 v2.0.9 in mate-paired mode, with mean inner distance between mate pairs: 140 and SD for distance between mate pairs: 30. Assembled transcript abundance and differential gene expression were determined using Cufflinks/Cuffdiff v2.1.1. Reference annotation files were downloaded from UCSC RefSeq Genes (GRCm38/mm10) in gtf format. Cuffdiff output genes with a fold change > 0.5 or < 0.5 (log2) were classified as significantly upregulated or downregulated genes, respectively. Transcripts presenting PU.1 ChIP peaks within 15 kb upstream or downstream of the TSS and exhibiting significant fold change by RNA-seq were considered PU.1-regulated genes. Functional classification analysis was performed on predicted PU.1 target genes from the Database for Annotation, Visualization and Integrated Discovery (DAVID) using GOTERM_BP_FAT (24). Functional protein classification of PU.1-regulated genes was determined using the PANTHER classification system (25).

i660BM cells were induced for 24 h with doxycycline (70 ng/μl) in the presence of 5% IL-7–conditioned medium described above. Chromatin was cross-linked with 1% paraformaldehyde for 10 min, and cross-linking was terminated by the addition of glycine. Chromatin yielding 150–300 μg was immunoprecipitated using Dynabeads Protein G (Life Technologies) conjugated to 6 μg of rabbit polyclonal anti-PU.1 Ab (Santa Cruz Biotechnology, Santa Cruz, CA) or 6 μg of rabbit polyclonal anti-IgG Ab (Abcam, ON, Canada). Immunoprecipitated chromatin was de–cross-linked and DNA was purified using a QIAquick PCR Purification Kit (QIAGEN). PU.1 immunoprecipitation was validated by quantitative RT-PCR (RT-qPCR) using two sets of primers for the E2f1 gene: a positive set targeting the PU.1 binding site on intron 1 of the E2f1 gene and a negative set targeting a region on intron 4 of the E2f1 gene. Illumina TruSeq DNA libraries were prepared from two biological replicates of PU.1-immunoprecipitated chromatin and one sample of input chromatin. Libraries were sequenced using an Illumina HiSEquation 2000 SR100 sequencer (Génome Québec Innovation Centre, QC, Canada). ChIP-seq data analysis was conducted using Galaxy Suite (23). Illumina sequencing adapters were removed using Trim Galore! Trimmed FASTQ files were aligned to the mouse reference genome GRCm38/mm10 with Bowtie, reporting only the best alignment for each fragment (--best) with a maximum number of two mismatches with an average quality score ≥ 70 (26). Experimental replicate BAM aligned files were merged using merge BAM files in Galaxy. Peaks were called using MACS1.4.1 with a mappable genome size of 1.8 × 1010 bp (mm10). Peaks were called with a tag size to 100, bandwidth of 300, and a p value cutoff for peak detection of 1e−07. Sequences of regions with significant PU.1 binding were extracted using extract genomic DNA in Galaxy. Motif discovery was performed with MEME-ChIP version 4.11.1. Functional analysis of cis-regulatory regions bound by PU.1 were identified using CEAS (27). Heat maps of ChIP signals surrounding TSSs were generated using deepTools2 (28).

Transcription of Igκ genes was evaluated from RNA-seq data using Cufflinks and the GENCODE mouse reference that contains a complete annotation for the Ig gene segments (29). Average fold change expression of Igκ genes was determined from three replicates for each group (−DOX and +DOX). Igκ genes were classified as upregulated (fold change > 1.0) or downregulated (fold change < −1.0) as a ratio between induced and control condition. A heat map illustrating the upregulated and downregulated genes was generated using GENE-E (Broad Institute, Cambridge, MA). Regions of PU.1 binding were intersected with the reference vM9 GENCODE to correlate ChIP binding with transcripts using bedtools v2.22.1 (30). Regions overlapping by ≥1 bp were reported as PU.1-bound transcripts and presented as Venn diagrams generated using InteractiVenn (31). RNA-seq and ChIP-seq genome tracks were visualized in the UCSC genome browser.

Genomic DNA was prepared from 660BM or i660BM cells after 48 h of PU.1 induction with doxycycline (70 ng/μl) with low (0.1 ng/μl) or high (7.5 ng/μl) IL-7. RNA was prepared using TRIzol Reagent or an AllPrep DNA/RNA Mini Kit (QIAGEN) from 660BM or i660BM cells after 48 h of PU.1 induction with doxycycline (70 ng/μl) in 5 or 2.5% IL-7–conditioned medium. cDNA was synthesized from purified RNA using an iScript cDNA synthesis kit (Bio-Rad). cDNA was diluted to a concentration of 30 ng/μl in RNase-free water for the RT-qPCR reactions. PCR reactions to detect Igκ rearrangements were performed as previously described (32). RT-qPCR reactions were prepared with specific primers for the genes of interest using SensiFAST SYBR green (Bioline, London, U.K.), and amplification was performed on a Rotor Gene 6000 instrument (Corbett Life Sciences, Valencia, CA). Gene expression was normalized to B2m expression, and fold expression was calculated using the Δ threshold cycle method (33). Primer sequences are shown in Supplemental Table I.

All data are graphed as mean ± SEM. Statistical analysis was performed with Prism 5.0 (GraphPad, La Jolla, CA) using ANOVA or the Student t test, as appropriate. The p values ≤ 0.05 were considered significant.

ChIP-seq and RNA-seq data have been submitted to the Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo/) under accession number GSE87316.

It was shown that mice lacking PU.1 or Spi-B in the B cell lineage have relatively normal B cell development and mild impairment of B cell function (3436). However, mice lacking PU.1 and Spi-B in the B cell lineage (Cd19+CreSpi1lox/loxSpib−/− mice) have reduced numbers of B cells, suggesting that these factors are important during development (18, 19). We sought to determine whether efficient deletion of PU.1 and Spi-B in the early stages of B cell development in BM would result in a block at a specific stage. Mice that delete Spi1 under control of the Cd79a (Mb-1) gene, which deletes alleles efficiently in BM (20), were bred to Spib−/− mice to generate Mb1-CreΔPB mice (lacking PU.1 and Spi-B), Mb1-CreΔP mice (lacking PU.1), and ΔB mice (lacking Spi-B). To assess B cell development, the frequency of mature B cells in the spleen of 6–10 wk old mice was determined. Mb1-CreΔPB mice had few B220+ IgM+ B cells in the spleen compared with WT or ΔB mice (Fig. 1). Therefore, PU.1 and Spi-B are required to generate splenic B cells. To determine the stage at which the block of B cell development was occurring, the frequencies of developing B cells in BM of 6–10-wk-old Mb1-CreΔPB mice were measured using flow cytometry according to the cell surface marker scheme described by Hardy et al. (3). There were no significant differences in the frequencies of B220+ cells among the groups (Fig. 2A, 2B). However, immature BM B cells (B220+ CD43+) were increased in frequency and mature BM B cells (B220+ CD43) were decreased in frequency in Mb1-CreΔPB mouse BM compared with Mb1-CreΔP, ΔB, and WT mouse BM (Fig. 2C–E). As a proportion of total B220+ cells, there were reduced frequencies of cells at the pre–pro B cell stage, or fraction A, in Mb1-CreΔPB, Mb1-CreΔP, and ΔB mice compared with WT mice (Fig. 2F, 2G). There were increased frequencies of fraction B (pro-B/pre-BI) cells and fraction C (pre-B, large pre-B/pre-BII) cells in Mb1-CreΔPB mouse BM compared with controls (Fig. 2F, 2G). Mb1-CreΔPB mice had double the frequency of fraction C BM cells compared with controls (Fig. 2F, 2G). None of the groups analyzed showed significant differences in the frequencies of fraction D (small pre-B) cells. However, fraction E (immature B cells) and fraction F (mature recirculating B cells) were nearly absent in Mb1-CreΔPB BM. Strikingly, there were no B cells expressing high levels of surface Ig in 6–10-wk-old Mb1-CreΔPB mice (Fig. 2F, 2G). These results demonstrate a critical requirement for PU.1 and Spi-B in the transition from small pre-B cells to immature B cells.

FIGURE 1.

Absence of PU.1 and Spi-B during B cell development severely impairs B cell maturation. (A) Flow cytometric analysis of B220 and IgM cell surface expression of spleen cells prepared from WT (left panel), ΔB (center panel), and Mb1-CreΔPB (right panel) mice. (B) Percentage of B220+ IgM+ cells in WT, ΔB, and Mb1-CreΔPB mice (n = 7). Data are mean ± SEM. ***p ≤ 0.001.

FIGURE 1.

Absence of PU.1 and Spi-B during B cell development severely impairs B cell maturation. (A) Flow cytometric analysis of B220 and IgM cell surface expression of spleen cells prepared from WT (left panel), ΔB (center panel), and Mb1-CreΔPB (right panel) mice. (B) Percentage of B220+ IgM+ cells in WT, ΔB, and Mb1-CreΔPB mice (n = 7). Data are mean ± SEM. ***p ≤ 0.001.

Close modal
FIGURE 2.

Deletion of PU.1 and Spi-B blocks B cell development in the BM at the pre-B cell transition. (A) Representative graph showing the percentage of B220+ cells in WT, Mb1-CreΔP, ΔB, and Mb1-CreΔPB mice. (B) Percentage of B220+ cells in BM of WT, Mb1-CreΔP, ΔB, and Mb1-CreΔPB mice. (C) Representative flow cytometric analysis showing the percentage of B220+ CD43 and B220+ CD43+ cells in WT (left panel) and Mb1-CreΔPB (right panel) mice. (D) Percentage of B220+ CD43+ cells in BM of WT, Mb1-CreΔP, ΔB, and Mb1-CreΔPB mice. (E) Percentage of B220+ CD43 cells in BM of WT, Mb1-CreΔP, ΔB, and Mb1-CreΔPB mice. (F) Representative flow cytometric analysis according to the Hardy scheme representing the frequency of developing B cells in fractions A–C in WT, Mb1-CreΔP, ΔB, and Mb1-CreΔPB mice (gated on B220+ CD43+ population) (upper panels). Flow cytometric analysis according to the Hardy scheme representing the frequency of developing B cells in fractions D–F in WT, Mb1-CreΔP, ΔB, and Mb1-CreΔPB mice (gated on B220+ CD43 population) (lower panels). (G) Percentage of B220+ cells in BM of WT, Mb1-CreΔP, ΔB, and Mb1-CreΔPB mice in fractions (A–F). WT, n = 10; Mb1-CreΔP, n = 8; ΔB, n = 9; and Mb1-CreΔPB, n = 12. Data are mean ± SEM. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.

FIGURE 2.

Deletion of PU.1 and Spi-B blocks B cell development in the BM at the pre-B cell transition. (A) Representative graph showing the percentage of B220+ cells in WT, Mb1-CreΔP, ΔB, and Mb1-CreΔPB mice. (B) Percentage of B220+ cells in BM of WT, Mb1-CreΔP, ΔB, and Mb1-CreΔPB mice. (C) Representative flow cytometric analysis showing the percentage of B220+ CD43 and B220+ CD43+ cells in WT (left panel) and Mb1-CreΔPB (right panel) mice. (D) Percentage of B220+ CD43+ cells in BM of WT, Mb1-CreΔP, ΔB, and Mb1-CreΔPB mice. (E) Percentage of B220+ CD43 cells in BM of WT, Mb1-CreΔP, ΔB, and Mb1-CreΔPB mice. (F) Representative flow cytometric analysis according to the Hardy scheme representing the frequency of developing B cells in fractions A–C in WT, Mb1-CreΔP, ΔB, and Mb1-CreΔPB mice (gated on B220+ CD43+ population) (upper panels). Flow cytometric analysis according to the Hardy scheme representing the frequency of developing B cells in fractions D–F in WT, Mb1-CreΔP, ΔB, and Mb1-CreΔPB mice (gated on B220+ CD43 population) (lower panels). (G) Percentage of B220+ cells in BM of WT, Mb1-CreΔP, ΔB, and Mb1-CreΔPB mice in fractions (A–F). WT, n = 10; Mb1-CreΔP, n = 8; ΔB, n = 9; and Mb1-CreΔPB, n = 12. Data are mean ± SEM. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.

Close modal

To identify target genes of PU.1 that could explain the developmental block in Mb1-CreΔPB mouse BM, we applied a gain-of-function approach using a PU.1-inducible pro-B cell line (Fig. 3A). The 660BM cell line is fully deleted for Spi1 and Spib, has a nonproductive rearrangement of the Igh locus, and is germline for Igκ locus rearrangement (19) (data not shown). The i660BM cell line is infected with a two-vector inducible system in which PU.1 is inducible using 70 ng/ml doxycycline (19). To define sites of PU.1 interaction, anti-PU.1 ChIP was performed on chromatin prepared from i660BM cells induced with (+DOX) or without (−DOX) 70 ng/ml doxycycline for 24 h. Illumina sequencing was performed on two biological replicates of PU.1-induced chromatin, as well as input chromatin. An average of 70 million 100-bp single-end reads were obtained. Reads were aligned to the mouse genome (mm10), and data were analyzed using the Galaxy suite of software tools. PU.1 was found to be associated with 49,385 unique genomic regions. The top interaction motif recovered for all PU.1-associated regions was the ETS motif containing the core A/5′-GGAA-3′ sequence (Fig. 3B). PU.1 was found to associate primarily with distal intergenic regions and gene bodies, similar to what was reported previously for PU.1 in splenic B cells (Fig. 3C) (37). Seventeen percent of PU.1-interaction sites were at annotated gene promoters, and 23,647 peaks were located within 15 kb of annotated TSSs (Fig. 3C, 3D).

FIGURE 3.

Identification of PU.1-regulated genes in i660BM pro-B cells. (A) Flow chart illustrating the strategy used to identify PU.1-regulated genes. RNA-seq of i660BM cells in the presence (+DOX) or absence (−DOX) of doxycycline was integrated with PU.1 ChIP-seq data sets to determine PU.1-regulated genes. Differentially expressed genes in the RNA-seq were categorized by the presence or absence of PU.1 binding within 15 kb of the TSS. (B) Motif enrichment analysis of PU.1 ChIP-seq data shows the ETS binding motif as the primary motif of PU.1 binding. (C) Pie chart representing the enrichment of PU.1 binding across genomic regions based on PU.1 ChIP-seq analysis. (D) Clustering of PU.1 ChIP-seq data demonstrates PU.1-enriched regions at the TSS of mouse RefSeq genes. (E) Defining PU.1-regulated genes. Venn diagrams illustrate the number of TSS-associated PU.1 peaks associated with upregulated (left panel; green circle) and downregulated (right panel; red circle) genes from the RNA-seq experiment. (F) Distribution of PU.1-regulated genes according to RNA-seq fold change (log2).

FIGURE 3.

Identification of PU.1-regulated genes in i660BM pro-B cells. (A) Flow chart illustrating the strategy used to identify PU.1-regulated genes. RNA-seq of i660BM cells in the presence (+DOX) or absence (−DOX) of doxycycline was integrated with PU.1 ChIP-seq data sets to determine PU.1-regulated genes. Differentially expressed genes in the RNA-seq were categorized by the presence or absence of PU.1 binding within 15 kb of the TSS. (B) Motif enrichment analysis of PU.1 ChIP-seq data shows the ETS binding motif as the primary motif of PU.1 binding. (C) Pie chart representing the enrichment of PU.1 binding across genomic regions based on PU.1 ChIP-seq analysis. (D) Clustering of PU.1 ChIP-seq data demonstrates PU.1-enriched regions at the TSS of mouse RefSeq genes. (E) Defining PU.1-regulated genes. Venn diagrams illustrate the number of TSS-associated PU.1 peaks associated with upregulated (left panel; green circle) and downregulated (right panel; red circle) genes from the RNA-seq experiment. (F) Distribution of PU.1-regulated genes according to RNA-seq fold change (log2).

Close modal

Next, RNA-seq analysis was performed on three biological replicates of RNA prepared from i660BM cells induced or not with 70 ng/ml doxycycline for 48 h. An average of 60 million 100-bp paired-end reads were obtained per sample. RNA-seq reads were aligned to the mouse genome (mm10) using TopHat, and differential gene expression was determined using Cufflinks suite. Genes presenting a log2-fold change in expression > 0.5 were considered in our analysis. Using these criteria, transcript levels of 811 genes were significantly increased upon PU.1 induction, and transcript levels of 531 genes were significantly decreased upon PU.1 induction (Supplemental Table II).

To identify PU.1-regulated genes, the genomic region that interacts with PU.1 was associated with differentially expressed transcripts. A total of 793 of 811 upregulated genes and 521 of 531 downregulated genes were associated with PU.1 binding sites located within 15 kb of the TSS, suggesting that these genes were regulated directly by PU.1 (Fig. 3E). As shown in Fig. 3F, 670 PU.1-associated genes were upregulated and 465 PU.1-associated genes were downregulated with fold changes in expression < log2 (1.5). However, 141 genes were upregulated > log2 (1.5)-fold, of which 12 were upregulated > log2 (5)-fold. Sixty-six genes were downregulated > log2 (1.5)-fold, of which five were downregulated > log2 (3)-fold (Fig. 3F). In summary, 793 upregulated gene transcripts and 521 downregulated gene transcripts were associated with PU.1 peaks and, therefore, were considered direct targets of PU.1 regulation.

Upregulated and downregulated genes associated with PU.1 binding in i660BM cells were explored further using gene ontology analysis. Protein-classification analysis showed that 11% of PU.1-upregulated or -downregulated genes encoded proteins related to nucleic acid binding, whereas 7% of upregulated or downregulated genes encoded transcription factors (Fig. 4A). PU.1-regulated genes encoded cell signaling molecules, cell receptors, and cellular transporters, among others, illustrating a broad role for PU.1 in regulating diverse cellular processes (Fig. 4A, 4B). Gene ontology analysis using DAVID showed that upregulated genes were classified into biological processes including immune response, cell activation, and lymphocyte activation (Fig. 4C). Downregulated genes were classified into biological pathways including nucleosome assembly, chromatin assembly, and nucleosome organization (Fig. 4C). These genes included a number of genes previously shown to be directly regulated by PU.1, such as Blnk (38), Il7r (39), and E2f1 (40) (Fig. 4B). Interestingly, Rag1/Rag2 was the fifth most upregulated gene and was associated with an upstream peak in PU.1 interaction (Fig. 4D). Id2, an inhibitor of E2A transcription factor activity, was a downregulated gene that was associated with a peak in PU.1 interaction (Fig. 4E). RT-qPCR analysis confirmed upregulation of Rag1, Rag2, and Spi1 transcript levels, as well as downregulation of Id2 mRNA transcript levels (Fig. 4F). These results suggest that PU.1 directly regulates genes involved in BCR signaling, as well as Ig recombination and/or accessibility.

FIGURE 4.

Characterization of PU.1-regulated genes. (A) Gene ontology data analysis showing the PANTHER protein classification of PU.1-regulated genes (left panel, upregulated; right panel, downregulated). (B) Examples of PU.1-upregulated (↑) and downregulated (↓) genes according to the protein classification ontology analysis. Numbers indicate log2 fold change. (C) Gene ontology analysis by DAVID identifying the biological processes related to PU.1-regulated genes (left panel, upregulated; right panel, downregulated). (D) UCSC genome browser tracks of PU.1 ChIP and RNA-seq experiments showing Rag2 and Rag1 genes. (E) UCSC genome browser tracks of PU.1 ChIP and RNA-seq experiments showing Id2 gene. (F) Confirmation of changes in gene expression. RT-qPCR analysis for the indicated genes was performed on four biological replicates of RNA prepared from uninduced (−DOX) or 70 ng/ml doxycycline-induced (+DOX) i660BM cells cultured in 2.5% IL-7–conditioned medium. *p ≤ 0.05, ***p ≤ 0.001, ****p ≤ 0.0001.

FIGURE 4.

Characterization of PU.1-regulated genes. (A) Gene ontology data analysis showing the PANTHER protein classification of PU.1-regulated genes (left panel, upregulated; right panel, downregulated). (B) Examples of PU.1-upregulated (↑) and downregulated (↓) genes according to the protein classification ontology analysis. Numbers indicate log2 fold change. (C) Gene ontology analysis by DAVID identifying the biological processes related to PU.1-regulated genes (left panel, upregulated; right panel, downregulated). (D) UCSC genome browser tracks of PU.1 ChIP and RNA-seq experiments showing Rag2 and Rag1 genes. (E) UCSC genome browser tracks of PU.1 ChIP and RNA-seq experiments showing Id2 gene. (F) Confirmation of changes in gene expression. RT-qPCR analysis for the indicated genes was performed on four biological replicates of RNA prepared from uninduced (−DOX) or 70 ng/ml doxycycline-induced (+DOX) i660BM cells cultured in 2.5% IL-7–conditioned medium. *p ≤ 0.05, ***p ≤ 0.001, ****p ≤ 0.0001.

Close modal

Upregulation of Rag mRNA transcripts and downregulation of Id2 mRNA transcripts suggested that PU.1 may be involved in the regulation of Igκ rearrangement during B cell development. Igκ V-J rearrangement is preceded by increased transcription of V region genes, as well as of sterile transcripts initiating in regulatory regions (9, 41). Analysis of ChIP-seq data identified 179 peaks in PU.1 interaction within the 3-Mb Igκ locus, of which 60 were located in GENCODE-annotated Igκ genes (Fig. 5A). Analysis of RNA-seq data revealed widespread changes in Igκ V region mRNA transcript levels upon PU.1 induction (Fig. 5B). For 166 annotated Igκ genes, 69 were unchanged, 42 were downregulated, and 55 were upregulated in the RNA-seq analysis (Fig. 5B, Supplemental Table III). Thirty peaks in PU.1 interaction were associated with increased upregulated Igκ V transcripts, and 10 peaks in PU.1 interaction were associated with downregulated Igκ V transcripts (Fig. 5A). Igκ V genes that were upregulated upon PU.1 induction included Igkv1-135, Igkv10-96, Igkv4-57, Igkv6-13, and Igkv4-73 (Fig. 5B). Each of these Igκ V genes was associated with at least one peak in PU.1 interaction (Fig. 5C, left panels). Quantitative PCR analysis confirmed that each of these Igκ V genes was inducible by PU.1, and reduced IL-7 concentration resulted in increased induction of transcription by PU.1 (Fig. 5C, right panels). PU.1 also was shown to interact with the 2-4 and 3-1 enhancers in the Igλ locus (42). Interestingly, induction of PU.1 increased Igλ1 mRNA transcript levels (Fig. 5D). These results suggest that PU.1 directly regulates Igκ and Igλ V region transcription in i660BM cells.

FIGURE 5.

PU.1 induction in i660BM pro-B cells induces transcription and rearrangement of the Ig κ locus. (A) Venn diagrams showing the number of PU.1 peaks located on upregulated and downregulated Igκ V genes. (B) Heat map showing the differential expression of Igκ V genes with (+DOX) or without (−DOX) doxycycline using the mouse genome annotation available on GENCODE. (C) RT-qPCR for detection of Igκv 1-135, Igκv10-96, Igκv4-57, Igκv6-13, and Igκv4-73 transcripts with (+DOX) or without (−DOX) doxycycline in 5 and 2.5% of IL-7. UCSC genome tracks (left panels) show the PU.1 binding peak from PU.1 ChIP-seq data and the transcription tracks from the RNA-seq experiment in conditions without (−DOX) and with (+DOX) doxycycline. (D) PU.1 induces Ig λ germline transcription. RT-qPCR analysis of the Igλ1 mRNA transcript was performed on four biological replicates of RNA prepared from uninduced (−DOX) or 70 ng/ml doxycycline-induced (+DOX) i660BM cells cultured in 2.5% IL-7–conditioned medium. (E) Distribution of PU.1 peaks within the Igκ V locus. (F) Distribution of PU.1 peaks separated by Igκ V family. (G) MEME motif analysis of sequences containing PU.1 peaks downstream of Igκ V gene second exons show RSS as a top enriched motif. (H) PCR for detection of Ig κ-chain rearrangement (Igvκ-IgJκ5) in i660BM and 660BM cells without doxycycline (−DOX) and with doxycycline (+DOX) induction and various levels of IL-7. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.

FIGURE 5.

PU.1 induction in i660BM pro-B cells induces transcription and rearrangement of the Ig κ locus. (A) Venn diagrams showing the number of PU.1 peaks located on upregulated and downregulated Igκ V genes. (B) Heat map showing the differential expression of Igκ V genes with (+DOX) or without (−DOX) doxycycline using the mouse genome annotation available on GENCODE. (C) RT-qPCR for detection of Igκv 1-135, Igκv10-96, Igκv4-57, Igκv6-13, and Igκv4-73 transcripts with (+DOX) or without (−DOX) doxycycline in 5 and 2.5% of IL-7. UCSC genome tracks (left panels) show the PU.1 binding peak from PU.1 ChIP-seq data and the transcription tracks from the RNA-seq experiment in conditions without (−DOX) and with (+DOX) doxycycline. (D) PU.1 induces Ig λ germline transcription. RT-qPCR analysis of the Igλ1 mRNA transcript was performed on four biological replicates of RNA prepared from uninduced (−DOX) or 70 ng/ml doxycycline-induced (+DOX) i660BM cells cultured in 2.5% IL-7–conditioned medium. (E) Distribution of PU.1 peaks within the Igκ V locus. (F) Distribution of PU.1 peaks separated by Igκ V family. (G) MEME motif analysis of sequences containing PU.1 peaks downstream of Igκ V gene second exons show RSS as a top enriched motif. (H) PCR for detection of Ig κ-chain rearrangement (Igvκ-IgJκ5) in i660BM and 660BM cells without doxycycline (−DOX) and with doxycycline (+DOX) induction and various levels of IL-7. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.

Close modal

Interestingly, there were two distinct patterns of PU.1 interaction with Ig V genes. PU.1 interacted with the promoters of 11 V genes, with a region downstream of the second exon of 26 V genes, and with both the promoter and a region downstream of the second exon of 14 genes (Fig. 5E, Supplemental Table IV). For the Igkv4 family, PU.1 interacted with 11 sites downstream of the second exon and only one site in a promoter. For the IgkV6 family, PU.1 interacted with both the promoter and a site downstream of the second exon for six members (Fig. 5C, 5F). A total of 32 of 40 PU.1 sites downstream of V region second exons were located an average of 91 bp from the RSS heptamer sequence 5′-CACAGTG-3′. MEME analysis of the 32 genomic regions associated with PU.1 binding sites located downstream of V region second exons revealed a 12-bp RSS as the most frequently discovered motif (Fig. 5G). The PU.1 motif was also enriched in these sequences (data not shown). Taken together, these results indicate that PU.1 interacts with a number of sites downstream of Igκ V region second exons that are located close to RSSs. This suggests that PU.1 might be involved in the regulation of chromatin accessibility near RSSs.

Because PU.1 interacts with Igκ V region genes, induces Igκ V region transcription, and induces Rag transcription, we hypothesized that PU.1 induction in i660BM cells coupled with reduced IL-7 concentration might be sufficient to induce Igκ V-J rearrangement in i660BM cells. To confirm this, Igκ V-J rearrangement was measured using DNA prepared from i660BM cells in which PU.1 was induced with doxycycline in the presence of a high or low concentration of IL-7. DNA prepared from WT spleen cells was used as a control (Fig. 5H, right side). Low IL-7 concentration or PU.1 induction resulted in low levels of Igκ V-J rearrangement in i660BM cells (Fig. 5H, left side). Igκ V-J rearrangement was robustly detected in i660BM cells in which PU.1 was induced with a low concentration of IL-7 (Fig. 5H, left side). In parent 660BM cells that did not express PU.1, reduced IL-7 concentration was not sufficient to induce Igκ V-J rearrangement (Fig. 5H, right side). We conclude that induction of PU.1 expression, coupled with reduced IL-7 concentration, is sufficient to induce Igκ V-J rearrangement in a cultured pro-B cell line.

Igκ V gene transcription is upregulated during the large pre-B (fraction C) to small pre-B (fraction D) transition (43). BM B cell development was blocked at the small pre-B cell stage in the absence of PU.1 and Spi-B (Fig. 1). Because induction of PU.1 resulted in increased Igκ V region transcript levels and Igκ V-J recombination, we hypothesized that Mb1-CreΔPB pre-B cells might have a reduced ability to activate Igκ V-J transcription. To test this, fraction C and fraction D pre-B cells were enriched from the BM of Mb1-CreΔPB mice, or ΔB mice as controls, using cell sorting; the gating strategy is shown in Fig. 2F. Fraction E cells were not enriched, because this fraction was absent in Mb1-CreΔPB mice (Fig. 2G). Quantitative PCR was used to determine changes in transcript levels of Igκ V genes with which PU.1 interacted. We found that, in BM cells from ΔB mice, Igκ V gene transcription was upregulated during the fraction C to fraction D transition for Igkv1-135, Igkv12-98, Igkv12-44, Igkv3-5, and the sterile transcript Glk-1 (Fig. 6A–F). In contrast, transcript levels for Btk and the control gene β-actin did not change (Fig. 6G, 6H). In BM cells enriched from Mb1-CreΔPB mice, mRNA transcript levels for Igkv1-135, Igkv12-98, Igkv12-44, Igkv3-5, and Glk-1 failed to increase during the fraction C to fraction D transition (Fig. 6B–F). Transcript levels for Spi1 and its target gene (Btk) decreased, consistent with deletion of Spi1 at this transition, whereas the control gene β-actin did not change from fraction C to fraction D (Fig. 6G–I). These data demonstrate that PU.1 and Spi-B are important for inducing transcription of certain Igκ V genes during B cell development in the BM and suggest that the absence of PU.1 and Spi-B might lead to reduced Igκ V region accessibility and V-J recombination.

FIGURE 6.

Absence of PU.1 and Spi-B in vivo results in reduced levels of Igκ transcripts in small pre-B cells. (A) Schematic diagram of the mouse Igκ locus showing the variable (V) region, the joining and constant regions (C-V regions), and the gene segments assessed in this study by RT-qPCR. Variable gene segments are highlighted, and the Glk-1o) transcript is also indicated. (BE) RT-qPCR showing the fold induction of the Igkv1-135, Igkv12-98, Igkv12-44, and Igkv3-5 genes in developing B cells from fraction C (large pre-B cells) and fraction D (small pre-B cells) of ΔB and Mb1-CreΔPB mouse BM. (F) RT-qPCR showing the fold induction of the Glk-1 mRNA transcript in developing B cells from fraction C (large pre-B cells) and fraction D (small pre-B cells) of ΔB and Mb1-CreΔPB mouse BM. (GI) RT-qPCR showing the fold induction of βactin, Btk, and Spi1 mRNA transcripts in developing B cells from fraction C (large pre-B cells) and fraction D (small pre-B cells) of ΔB and Mb1-CreΔPB mouse BM. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.

FIGURE 6.

Absence of PU.1 and Spi-B in vivo results in reduced levels of Igκ transcripts in small pre-B cells. (A) Schematic diagram of the mouse Igκ locus showing the variable (V) region, the joining and constant regions (C-V regions), and the gene segments assessed in this study by RT-qPCR. Variable gene segments are highlighted, and the Glk-1o) transcript is also indicated. (BE) RT-qPCR showing the fold induction of the Igkv1-135, Igkv12-98, Igkv12-44, and Igkv3-5 genes in developing B cells from fraction C (large pre-B cells) and fraction D (small pre-B cells) of ΔB and Mb1-CreΔPB mouse BM. (F) RT-qPCR showing the fold induction of the Glk-1 mRNA transcript in developing B cells from fraction C (large pre-B cells) and fraction D (small pre-B cells) of ΔB and Mb1-CreΔPB mouse BM. (GI) RT-qPCR showing the fold induction of βactin, Btk, and Spi1 mRNA transcripts in developing B cells from fraction C (large pre-B cells) and fraction D (small pre-B cells) of ΔB and Mb1-CreΔPB mouse BM. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.

Close modal

In this study, we showed that B cell development is blocked in mice that delete the Spi1 gene encoding PU.1 under control of the Mb1 locus, which are also germline knockout for Spib (Mb1-CreΔPB mice). Adult Mb1-CreΔPB mice did not have splenic B cells, and few surface IgM+ B cells were present in BM, suggesting a block in B cell development starting at the pre-B cell stage. Identifying target genes of PU.1 and/or Spi-B might explain this block in B cell development; therefore a Spi1/Spib-deleted IL-7–dependent pro-B cell line was used in which PU.1 expression can be induced using doxycycline (i660BM cells). RNA-seq and anti-PU.1 ChIP-seq experiments were performed to determine the genome-wide target genes of PU.1 in this model. These experiments confirmed a number of previously identified PU.1 target genes involved in the pro-B to pre-B cell developmental transition. Unexpectedly, the Rag locus was also found to be a direct target of PU.1 induction. Closer examination revealed that PU.1 interacted with 179 sites within the Igκ locus, of which 60 sites were located within V genes, including at V gene promoters and near RSSs. An increase in PU.1 expression, combined with reduced IL-7 concentration, induced Igκ rearrangement. Finally, we found that Igκ V region mRNA transcript levels were not increased at the pro-B to pre-B cell transition in Mb1-CreΔPB mice. These results show that PU.1 directly regulates Igκ locus transcription and suggest that PU.1 is an important regulator of Igκ locus accessibility during B cell development.

Comparison of BM B cell development in Mb1-CreΔPB mice with that in control mice revealed a relative increase in the frequency of pre-BII/large pre-B cells (fraction C), no difference in the frequency of small pre-B cells (fraction D), and a near absence of immature sIgM+ B cells (fraction E). These results are most consistent with a block in B cell development at the stage when Ig L chain recombination is taking place in small pre-B cells. Previous analysis of CD19-CreΔPB mice, which did not delete alleles efficiently in BM, showed a relative increase in immature B cell frequencies in BM but a decrease in mature recirculating B cells (fraction F) and in the frequency of splenic follicular B cells. The differences between these two models suggest that PU.1 and Spi-B play important roles in B cell development at late, as well as early, stages.

B cell–specific deletion of the gene encoding PU.1 in mice (34, 35) or deletion of the gene encoding Spi-B in mice (44) resulted in relatively mild defects in B cell development. Similarly, deletion of the Irf4 or Irf8 gene resulted in mild B cell developmental defects (45, 46). Combined deletion of Spi1 and Irf8 resulted in mild impairment of B cell development, whereas combined deletion of Spi1 and Irf4 resulted in impaired B cell development at the pre-B cell stage (36). Strikingly, combined deletion of Irf4 and Irf8 resulted in a block in B cell development at the pre-B cell stage (47). In this study, we showed that combined deletion of Spi1 and Spib results in a block in B cell development at the pre-B cell stage. Collectively, these studies reveal a critically important PU.1/Spi-B/IRF4/IRF8 regulatory axis for pre-B cell development. PU.1 and Spi-B interact interchangeably with IRF4 or IRF8 to regulate genes containing ETS-IRF composite elements (EICEs) (48). Recently, it was shown that ∼50% of PU.1 binding sites in pro-B cells are at EICEs, suggesting that this regulatory element may control a large number of genes in developing B cells (36). These studies collectively suggest that genes important for early B cell development require activation through EICEs to promote their expression. More work is necessary to identify key EICE-regulated target genes.

PU.1 was implicated in the regulation of Ig transcription in numerous studies. In 1991, PU.1 was recognized to interact with the Igκ 3′ enhancer cooperatively with IRF4 (15, 49). Transgenic studies implicated the PU.1 binding site in the Igκ 3′ enhancer in the regulation of the developmental stage specificity of V-J joining (50). In 1995, PU.1 was identified as an NF capable of interacting with pyrimidine-rich sequences in the Vκ19 promoter (14). PU.1 binding sites were predicted in numerous V region promoters of the Igκ locus (16). Previous studies showed that PU.1 interacts with numerous sites in the Igκ locus in the B cell lineage. In mature splenic B cells, PU.1 was shown to interact with 45 sites in the 3-Mb Igκ locus (37). In pro-B cells, PU.1 was shown to interact with 181 sites in the Igκ locus (17). This study closely agrees with our result of 179 total PU.1 binding sites in the Igκ locus, and 74 of these sites were in common, with 35 of these sites in common in all three studies. Taken together, these studies correspond with our identification of PU.1 binding sites in the Igκ locus.

The 3′ enhancer of the Igκ locus is critical for the cell type and developmental stage specificity of V-J rearrangement (10, 5052). STAT5 binding induced by IL-7 signaling is a critically important repressor of Igκ transcription to prevent accessibility and rearrangement during proliferation of large pre-B/pre-BII cells (53). PU.1 can compete with STAT5 for interaction with the Igκ 3′ enhancer to regulate Igκ transcription (54). Consistent with these studies, our results showed that PU.1 induced Igκ transcription and rearrangement more efficiently when IL-7 concentration was reduced (Fig. 5). Our results are consistent with PU.1 being an important factor for inducing Igκ transcription and accessibility when developing pre-B cells migrate away from high IL-7 concentrations (55).

Igκ V region transcription is closely associated with accessibility of the recombinase apparatus and initiation of Igκ V-J recombination (9, 41). The results presented in Fig. 6 demonstrate that Igκ V region transcripts are not appropriately upregulated in the absence of PU.1 and Spi-B. This suggest that PU.1 and Spi-B are important regulators of Igκ V region transcription, accessibility, and rearrangement. The absence of IgM+ B cells in Mb1-CreΔPB mice suggests that Ig λ transcription may also be impaired in the absence of PU.1 and Spi-B. Consistent with this idea, induction of PU.1 activated Igλ1 transcription in i660BM cells (Fig. 5D). We previously demonstrated that Igλ transcription was reduced in PU.1/Spi-B double-knockout pro-B cell lines (56). In summary, we expect that reduced Igκ V region transcription in fraction D small pre-B cells lacking PU.1 and Spi-B results in impaired Igκ V region accessibility and impaired Igκ recombination, which lead to a block in B cell development at the small pre-B cell stage. These studies reveal an important role for PU.1 in Igκ transcription and rearrangement, as well as a requirement for PU.1 and Spi-B in B cell development.

We thank Michael Reth (Freiburg, Germany) and John Weis (University of Utah) for providing Mb1-Cre mice. We thank scientists and staff of the McGill University and Génome Québec Innovation Centre for performing library construction and next-generation sequencing. We also thank Kristin Chadwick and the London Regional Flow Cytometry Core Facility for assistance with flow cytometric sorting and analysis.

This work was supported by Ontario Trillium (to C.R.B.), the Leukemia and Lymphoma Society of Canada (to R.P.D.), and by Canadian Institutes of Health Research Grants MOP-10651 and MOP-137414 (to R.P.D.).

Chromatin immunoprecipitation–sequencing and RNA-sequencing data presented in this article have been submitted to the Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo/) under accession number GSE87316.

The online version of this article contains supplemental material.

Abbreviations used in this article:

ΔB

Mb1+/+ Spi1+/+Spib−/−

BM

bone marrow

ChIP

chromatin immunoprecipitation

ChIP-seq

chromatin immunoprecipitation–sequencing

DAVID

Database for Annotation, Visualization and Integrated Discovery

EICE

ETS-IRF composite element

ETS

E26 transformation–specific

Mb1-CreΔP

Mb1+/CreSpi1lox/loxSpib+/+

Mb1-CreΔPB

Mb1+/CreSpi1lox/loxSpib−/−

pro-B

progenitor B

RNA-seq

RNA-sequencing

RSS

recombination signal sequence

RT-qPCR

quantitative RT-PCR

TSS

transcription start site

WT

wild-type.

1
Lu
L.
,
Smithson
G.
,
Kincade
P. W.
,
Osmond
D. G.
.
1998
.
Two models of murine B lymphopoiesis: a correlation.
Eur. J. Immunol.
28
:
1755
1761
.
2
Osmond
D. G.
,
Rolink
A.
,
Melchers
F.
.
1998
.
Murine B lymphopoiesis: towards a unified model.
Immunol. Today
19
:
65
68
.
3
Hardy
R. R.
,
Carmack
C. E.
,
Shinton
S. A.
,
Kemp
J. D.
,
Hayakawa
K.
.
1991
.
Resolution and characterization of pro-B and pre-pro-B cell stages in normal mouse bone marrow.
J. Exp. Med.
173
:
1213
1225
.
4
Teng
G.
,
Maman
Y.
,
Resch
W.
,
Kim
M.
,
Yamane
A.
,
Qian
J.
,
Kieffer-Kwon
K. R.
,
Mandal
M.
,
Ji
Y.
,
Meffre
E.
, et al
.
2015
.
RAG represents a widespread threat to the lymphocyte genome.
Cell
162
:
751
765
.
5
Corfe
S. A.
,
Paige
C. J.
.
2009
.
Development of B lymphocytes
. In
Molecular Basis of Hematopoiesis.
Wickrema
A.
,
Kee
B. L.
, eds.
Springer
,
New York
, p.
173
199
.
6
Allman
D.
,
Lindsley
R. C.
,
DeMuth
W.
,
Rudd
K.
,
Shinton
S. A.
,
Hardy
R. R.
.
2001
.
Resolution of three nonproliferative immature splenic B cell subsets reveals multiple selection points during peripheral B cell maturation.
J. Immunol.
167
:
6834
6840
.
7
Aoki-Ota
M.
,
Torkamani
A.
,
Ota
T.
,
Schork
N.
,
Nemazee
D.
.
2012
.
Skewed primary Igκ repertoire and V-J joining in C57BL/6 mice: implications for recombination accessibility and receptor editing.
J. Immunol.
188
:
2305
2315
.
8
Ribeiro de Almeida
C.
,
Hendriks
R. W.
,
Stadhouders
R.
.
2015
.
Dynamic control of long-range genomic interactions at the immunoglobulin κ light-chain locus.
Adv. Immunol.
128
:
183
271
.
9
Yancopoulos
G. D.
,
Alt
F. W.
.
1985
.
Developmentally controlled and tissue-specific expression of unrearranged VH gene segments.
Cell
40
:
271
281
.
10
Inlay
M.
,
Alt
F. W.
,
Baltimore
D.
,
Xu
Y.
.
2002
.
Essential roles of the kappa light chain intronic enhancer and 3′ enhancer in kappa rearrangement and demethylation.
Nat. Immunol.
3
:
463
468
.
11
Sleckman
B. P.
,
Gorman
J. R.
,
Alt
F. W.
.
1996
.
Accessibility control of antigen-receptor variable-region gene assembly: role of cis-acting elements.
Annu. Rev. Immunol.
14
:
459
481
.
12
Pang
S. H.
,
Carotta
S.
,
Nutt
S. L.
.
2014
.
Transcriptional control of pre-B cell development and leukemia prevention.
Curr. Top. Microbiol. Immunol.
381
:
189
213
.
13
Solomon
L. A.
,
Li
S. K.
,
Piskorz
J.
,
Xu
L. S.
,
DeKoter
R. P.
.
2015
.
Genome-wide comparison of PU.1 and Spi-B binding sites in a mouse B lymphoma cell line.
BMC Genomics
16
:
76
.
14
Schwarzenbach
H.
,
Newell
J. W.
,
Matthias
P.
.
1995
.
Involvement of the Ets family factor PU.1 in the activation of immunoglobulin promoters.
J. Biol. Chem.
270
:
898
907
.
15
Pongubala
J. M.
,
Atchison
M. L.
.
1991
.
Functional characterization of the developmentally controlled immunoglobulin kappa 3′ enhancer: regulation by Id, a repressor of helix-loop-helix transcription factors.
Mol. Cell. Biol.
11
:
1040
1047
.
16
Brekke
K. M.
,
Garrard
W. T.
.
2004
.
Assembly and analysis of the mouse immunoglobulin kappa gene sequence.
Immunogenetics
56
:
490
505
.
17
Schwickert
T. A.
,
Tagoh
H.
,
Gültekin
S.
,
Dakic
A.
,
Axelsson
E.
,
Minnich
M.
,
Ebert
A.
,
Werner
B.
,
Roth
M.
,
Cimmino
L.
, et al
.
2014
.
Stage-specific control of early B cell development by the transcription factor Ikaros.
Nat. Immunol.
15
:
23
293
.
18
Sokalski
K. M.
,
Li
S. K.
,
Welch
I.
,
Cadieux-Pitre
H. A.
,
Gruca
M. R.
,
DeKoter
R. P.
.
2011
.
Deletion of genes encoding PU.1 and Spi-B in B cells impairs differentiation and induces pre-B cell acute lymphoblastic leukemia.
Blood
118
:
2801
2808
.
19
Christie
D. A.
,
Xu
L. S.
,
Turkistany
S. A.
,
Solomon
L. A.
,
Li
S. K.
,
Yim
E.
,
Welch
I.
,
Bell
G. I.
,
Hess
D. A.
,
DeKoter
R. P.
.
2015
.
PU.1 opposes IL-7-dependent proliferation of developing B cells with involvement of the direct target gene bruton tyrosine kinase.
J. Immunol.
194
:
595
605
.
20
Hobeika
E.
,
Thiemann
S.
,
Storch
B.
,
Jumaa
H.
,
Nielsen
P. J.
,
Pelanda
R.
,
Reth
M.
.
2006
.
Testing gene function early in the B cell lineage in mb1-cre mice.
Proc. Natl. Acad. Sci. USA
103
:
13789
13794
.
21
Winkler
T. H.
,
Rolink
A.
,
Melchers
F.
,
Karasuyama
H.
.
1995
.
Precursor B cells of mouse bone marrow express two different complexes with the surrogate light chain on the surface.
Eur. J. Immunol.
25
:
446
450
.
22
Trapnell
C.
,
Roberts
A.
,
Goff
L.
,
Pertea
G.
,
Kim
D.
,
Kelley
D. R.
,
Pimentel
H.
,
Salzberg
S. L.
,
Rinn
J. L.
,
Pachter
L.
.
2012
.
Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks.
Nat. Protoc.
7
:
562
578
.
23
Afgan
E.
,
Baker
D.
,
van den Beek
M.
,
Blankenberg
D.
,
Bouvier
D.
,
Čech
M.
,
Chilton
J.
,
Clements
D.
,
Coraor
N.
,
Eberhard
C.
, et al
.
2016
.
The galaxy platform for accessible, reproducible and collaborative biomedical analyses: 2016 update.
Nucleic Acids Res.
44
(
W1
):
W3
W10
.
24
Huang da
W.
,
Sherman
B. T.
,
Lempicki
R. A.
.
2009
.
Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources.
Nat. Protoc.
4
:
44
57
.
25
Mi
H.
,
Muruganujan
A.
,
Casagrande
J. T.
,
Thomas
P. D.
.
2013
.
Large-scale gene function analysis with the PANTHER classification system.
Nat. Protoc.
8
:
1551
1566
.
26
Langmead
B.
,
Trapnell
C.
,
Pop
M.
,
Salzberg
S. L.
.
2009
.
Ultrafast and memory-efficient alignment of short DNA sequences to the human genome.
Genome Biol.
10
:
R25
.
27
Shin
H.
,
Liu
T.
,
Manrai
A. K.
,
Liu
X. S.
.
2009
.
CEAS: cis-regulatory element annotation system.
Bioinformatics
25
:
2605
2606
.
28
Ramírez
F.
,
Dündar
F.
,
Diehl
S.
,
Grüning
B. A.
,
Manke
T.
.
2014
.
Deeptools: a flexible platform for exploring deep-sequencing data.
Nucleic Acids Res.
42
:
W187
W191
.
29
Harrow
J.
,
Denoeud
F.
,
Frankish
A.
,
Reymond
A.
,
Chen
C. K.
,
Chrast
J.
,
Lagarde
J.
,
Gilbert
J. G.
,
Storey
R.
,
Swarbreck
D.
, et al
.
2006
.
GENCODE: producing a reference annotation for ENCODE.
Genome Biol.
7 (Suppl. 1)
:
S4.1
S4.9
.
30
Quinlan
A. R.
,
Hall
I. M.
.
2010
.
BEDTools: a flexible suite of utilities for comparing genomic features.
Bioinformatics
26
:
841
842
.
31
Heberle
H.
,
Meirelles
G. V.
,
da Silva
F. R.
,
Telles
G. P.
,
Minghim
R.
.
2015
.
InteractiVenn: a web-based tool for the analysis of sets through Venn diagrams.
BMC Bioinformatics
16
:
169
.
32
Cobaleda
C.
,
Jochum
W.
,
Busslinger
M.
.
2007
.
Conversion of mature B cells into T cells by dedifferentiation to uncommitted progenitors.
Nature
449
:
473
477
.
33
Livak
K. J.
,
Schmittgen
T. D.
.
2001
.
Analysis of relative gene expression data using real-time quantitative PCR and the 2(−Delta Delta C(T)) method.
Methods
25
:
402
408
.
34
Polli
M.
,
Dakic
A.
,
Light
A.
,
Wu
L.
,
Tarlinton
D. M.
,
Nutt
S. L.
.
2005
.
The development of functional B lymphocytes in conditional PU.1 knock-out mice.
Blood
106
:
2083
2090
.
35
Ye
M.
,
Ermakova
O.
,
Graf
T.
.
2005
.
PU.1 is not strictly required for B cell development and its absence induces a B-2 to B-1 cell switch.
J. Exp. Med.
202
:
1411
1422
.
36
Pang
S. H.
,
Minnich
M.
,
Gangatirkar
P.
,
Zheng
Z.
,
Ebert
A.
,
Song
G.
,
Dickins
R. A.
,
Corcoran
L. M.
,
Mullighan
C. G.
,
Busslinger
M.
, et al
.
2016
.
PU.1 cooperates with IRF4 and IRF8 to suppress pre-B-cell leukemia.
Leukemia
30
:
1375
1387
.
37
Heinz
S.
,
Benner
C.
,
Spann
N.
,
Bertolino
E.
,
Lin
Y. C.
,
Laslo
P.
,
Cheng
J. X.
,
Murre
C.
,
Singh
H.
,
Glass
C. K.
.
2010
.
Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities.
Mol. Cell
38
:
576
589
.
38
Xu
L. S.
,
Sokalski
K. M.
,
Hotke
K.
,
Christie
D. A.
,
Zarnett
O.
,
Piskorz
J.
,
Thillainadesan
G.
,
Torchia
J.
,
DeKoter
R. P.
.
2012
.
Regulation of B cell linker protein transcription by PU.1 and Spi-B in murine B cell acute lymphoblastic leukemia.
J. Immunol.
189
:
3347
3354
.
39
DeKoter
R. P.
,
Schweitzer
B. L.
,
Kamath
M. B.
,
Jones
D.
,
Tagoh
H.
,
Bonifer
C.
,
Hildeman
D. A.
,
Huang
K. J.
.
2007
.
Regulation of the interleukin-7 receptor alpha promoter by the Ets transcription factors PU.1 and GA-binding protein in developing B cells.
J. Biol. Chem.
282
:
14194
14204
.
40
Ziliotto
R.
,
Gruca
M. R.
,
Podder
S.
,
Noel
G.
,
Ogle
C. K.
,
Hess
D. A.
,
DeKoter
R. P.
.
2014
.
PU.1 promotes cell cycle exit in the murine myeloid lineage associated with downregulation of E2F1
.
Exp. Hematol.
42
:
204
217.e1
.
41
Schlissel
M. S.
,
Stanhope-Baker
P.
.
1997
.
Accessibility and the developmental regulation of V(D)J recombination.
Semin. Immunol.
9
:
161
170
.
42
Eisenbeis
C. F.
,
Singh
H.
,
Storb
U.
.
1993
.
PU.1 is a component of a multiprotein complex which binds an essential site in the murine immunoglobulin lambda 2-4 enhancer.
Mol. Cell. Biol.
13
:
6452
6461
.
43
Painter
M. W.
,
Davis
S.
,
Hardy
R. R.
,
Mathis
D.
,
Benoist
C.
Immunological Genome Project Consortium
.
2011
.
Transcriptomes of the B and T lineages compared by multiplatform microarray profiling.
J. Immunol.
186
:
3047
3057
.
44
Su
G. H.
,
Chen
H. M.
,
Muthusamy
N.
,
Garrett-Sinha
L. A.
,
Baunoch
D.
,
Tenen
D. G.
,
Simon
M. C.
.
1997
.
Defective B cell receptor-mediated responses in mice lacking the Ets protein, Spi-B.
EMBO J.
16
:
7118
7129
.
45
Mittrücker
H. W.
,
Matsuyama
T.
,
Grossman
A.
,
Kündig
T. M.
,
Potter
J.
,
Shahinian
A.
,
Wakeham
A.
,
Patterson
B.
,
Ohashi
P. S.
,
Mak
T. W.
.
1997
.
Requirement for the transcription factor LSIRF/IRF4 for mature B and T lymphocyte function.
Science
275
:
540
543
.
46
Holtschke
T.
,
Löhler
J.
,
Kanno
Y.
,
Fehr
T.
,
Giese
N.
,
Rosenbauer
F.
,
Lou
J.
,
Knobeloch
K. P.
,
Gabriele
L.
,
Waring
J. F.
, et al
.
1996
.
Immunodeficiency and chronic myelogenous leukemia-like syndrome in mice with a targeted mutation of the ICSBP gene.
Cell
87
:
307
317
.
47
Lu
R.
,
Medina
K. L.
,
Lancki
D. W.
,
Singh
H.
.
2003
.
IRF-4,8 orchestrate the pre-B-to-B transition in lymphocyte development.
Genes Dev.
17
:
1703
1708
.
48
Singh
H.
,
Glasmacher
E.
,
Chang
A. B.
,
Vander Lugt
B.
.
2013
.
The molecular choreography of IRF4 and IRF8 with immune system partners.
Cold Spring Harb. Symp. Quant. Biol.
78
:
101
104
.
49
Pongubala
J. M.
,
Nagulapalli
S.
,
Klemsz
M. J.
,
McKercher
S. R.
,
Maki
R. A.
,
Atchison
M. L.
.
1992
.
PU.1 recruits a second nuclear factor to a site important for immunoglobulin kappa 3′ enhancer activity.
Mol. Cell. Biol.
12
:
368
378
.
50
Hayashi
R.
,
Takemori
T.
,
Kodama
M.
,
Suzuki
M.
,
Tsuboi
A.
,
Nagawa
F.
,
Sakano
H.
.
1997
.
The PU.1 binding site is a cis-element that regulates pro-B/pre-B specificity of Vkappa-Jkappa joining.
J. Immunol.
159
:
4145
4149
.
51
Shaffer
A. L.
,
Peng
A.
,
Schlissel
M. S.
.
1997
.
In vivo occupancy of the kappa light chain enhancers in primary pro- and pre-B cells: a model for kappa locus activation.
Immunity
6
:
131
143
.
52
Gorman
J. R.
,
van der Stoep
N.
,
Monroe
R.
,
Cogne
M.
,
Davidson
L.
,
Alt
F. W.
.
1996
.
The Ig(kappa) enhancer influences the ratio of Ig(kappa) versus Ig(lambda) B lymphocytes.
Immunity
5
:
241
252
.
53
Mandal
M.
,
Powers
S. E.
,
Maienschein-Cline
M.
,
Bartom
E. T.
,
Hamel
K. M.
,
Kee
B. L.
,
Dinner
A. R.
,
Clark
M. R.
.
2011
.
Epigenetic repression of the Igk locus by STAT5-mediated recruitment of the histone methyltransferase Ezh2.
Nat. Immunol.
12
:
1212
1220
.
54
Hodawadekar
S.
,
Park
K.
,
Farrar
M. A.
,
Atchison
M. L.
.
2012
.
A developmentally controlled competitive STAT5-PU.1 DNA binding mechanism regulates activity of the Ig κ E3′ enhancer.
J. Immunol.
188
:
2276
2284
.
55
Clark
M. R.
,
Mandal
M.
,
Ochiai
K.
,
Singh
H.
.
2014
.
Orchestrating B cell lymphopoiesis through interplay of IL-7 receptor and pre-B cell receptor signalling.
Nat. Rev. Immunol.
14
:
69
80
.
56
Schweitzer
B. L.
,
DeKoter
R. P.
.
2004
.
Analysis of gene expression and Ig transcription in PU.1/Spi-B-deficient progenitor B cell lines.
J. Immunol.
172
:
144
154
.

The authors have no financial conflicts of interest.