Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2006 Nov 9.
Published in final edited form as: Nat Immunol. 2004 Jul 18;5(8):853–861. doi: 10.1038/ni1099

B cell–specific loss of histone 3 lysine 9 methylation in the VH locus depends on Pax5

Kristen Johnson 1, David L Pflugh 2, Duonan Yu 3, David G T Hesslein 4, Kuo-I Lin 1, Alfred L M Bothwell 2, Andrei Thomas-Tikhonenko 3, David G Schatz 2,5, Kathryn Calame 1,6
PMCID: PMC1635547  NIHMSID: NIHMS13347  PMID: 15258579

Abstract

Immunoglobulin heavy chain rearrangement (VH-to-DJH) occurs only in B cells, suggesting it is inhibited in other lineages. Here we found that in the mouse VH locus, methylation of lysine 9 on histone H3 (H3-K9), a mark of inactive chromatin, was present in non–B lineage cells but was absent in B cells. As others have shown that H3-K9 methylation can inhibit V(D)J recombination on engineered substrates, our data support the idea that H3-K9 methylation inhibits endogenous VH-to-DJH recombination. We also show that Pax5, a transcription factor required for B cell commitment, is necessary and sufficient for the removal of H3-K9 methylation in the VH locus and provide evidence that one function of Pax5 is to remove this inhibitory modification by a mechanism of histone exchange, thus allowing B cell–specific VH-to-DJH recombination.

During hematopoiesis, transcription factors initiate and maintain lineage-specific commitment1. In B and T lymphocytes, the unique variable (V), diversity (D) and joining (J) (V(D)J) recombination of immunoglobulin and T cell receptor (TCR) gene segments provides critical developmental checkpoints as well as diverse, clonotypic antigen recognition capability2. All immunoglobulin and TCR loci use a common recombinase machinery, including proteins encoded by the recombination activating genes Rag1 and Rag2 that recognize and cut at common DNA recognition elements called recombination signal sequences (RSSs). However, V(D)J recombination proceeds with strict T lineage–B lineage specificity and developmentally determined order. The mechanisms responsible for these aspects of V(D)J regulation remain poorly understood.

It is increasingly apparent that V(D)J recombination requires chromatin structure changes to allow the recombinase machinery access to rearranging gene segments. In vitro, nucleosomes inhibit V(D)J recombination3-5, demonstrating that chromatin can act as an obstacle. In vivo, multiple hallmarks of accessible chromatin occur at V, D and J gene segments before rearrangement: germline transcription, DNase I sensitivity, demethylation of DNA and histone acetylation6. Thus, lineage and developmentally regulated changes in chromatin structure seem to be a prerequisite for V(D)J recombination, although they may not be sufficient7.

Methylation of lysine 9 on histone 3 (H3-K9) provides a stable epi-genetic mark that is sufficient to establish repressed chromatin in multiple settings8. H3-K9 methylation inhibits V(D)J rearrangement when directed to the promoter of an artificial, stably integrated recombination substrate9. However, at present there is no evidence that H3-K9 methylation is involved in V(D)J regulation at endogenous antigen receptor loci in primary lymphocytes.

During early B cell development, the immunoglobulin heavy chain locus (Igh) rearranges first. Indeed, DH-to-JH recombination can occur in progenitors not fully committed to the B lineage10. However, VH-to-DJH recombination occurs only in committed B cells, after DH-to-JH recombination. This suggests that chromatin in the VH locus may be subjected to a unique and stringent level of repression that must be overcome during lineage commitment. Irreversible commitment to the B cell lineage requires the transcription factor Pax5 (also know as BSAP)11, and Pax5-deficient bone marrow cells are halted in their development at the stage of VH-to-DJH rearrangement12,13. Specifically, the DJH-distal VH gene segments remain blocked from undergoing VH-to-DJH rearrangement despite the fact that the VH locus has some hallmarks of accessible chromatin, including germline transcription and histone acetylation14,15. Expression of Pax5 in T cells causes VH-to-DJH recombination for the DJH-proximal VH segments15,16, and Pax5 is involved in the compaction of the VH locus, which was suggested to promote rearrangement of DJH-distal gene segments15. However, the mechanism by which Pax5 regulates VH-to-DJH recombination is unknown.

In this study we provide evidence that methylation of H3-K9 is present in the VH locus in hematopoietic cells outside the B lineage and that removal of H3-K9 methylation accompanies Pax5 expression and B cell commitment. Pax5 was both required and sufficient for removal of H3-K9 methylation in the VH locus. Concomitant accumulation of the histone variant H3.3 at the VH locus suggests that histone exchange is the likely mechanism by which this heterochromatic mark is lost. These data provide strong evidence for a model in which one important function of Pax5 is removal of H3-K9 methylation in the VH locus, making VH genes accessible for VH-to-DJH recombination and allowing further B cell development.

RESULTS

H3-K9 methylation in the VH locus

Low H3-K9 methylation associated with the VH locus does not vary between B and T cell lines17. As our preliminary studies (data not shown) using different cell lines yielded different results from those in this study, we extended our analysis to primary mouse cells. We obtained pro–B cells from Rag1−/− bone marrow and cultured them briefly in interleukin 7 (IL-7). We obtained freshly isolated thymocytes from either wild-type or Rag1−/− mice. We obtained bone marrow–derived macrophages after 9 d of culture in macrophage colony-stimulating factor. The VH locus is accessible to RAG-dependent cutting in pro–B cells from Rag1−/− mice, as measured by DNase I sensitivity, germline transcription, histone acetylation and the ability to undergo recombination after expression of the RAG proteins6,18. VH gene segments are inaccessible in cells committed to either the T lineage19,20 or the macrophage lineage.

We did chromatin immunoprecipitation assays of Rag1−/− Abelson virus–transformed pro–B cell lines (6312 and AR2) and a double-negative T cell line (S49; data not shown), as well as on primary pro–B cells, thymocytes and bone marrow–derived macrophages, using an antibody to dimethylated H3-K9 (anti–dimethylated H3-K9; Fig. 1). We designed primers to amplify the promoters or RSS elements of individual VH gene segments and intergenic sequences with known locations within the VH locus (Fig. 1a and Supplementary Table 1 online), as described21. The promoter for the gene encoding c-Myc (Myc), not associated with H3-K9 methylation22, provided a negative control. Because signals using anti–dimethylated H3-K9 are weak22, only signals fulfilling two criteria were considered to be positive: at least twofold greater than the signal of Myc and an appropriate decrease in signal intensity after dilution of the DNA used in PCR. We quantified signals meeting these criteria and plotted the results as percent of input (Fig. 1b).

Figure 1.

Figure 1

Open in a new tab

H3-K9 methylation associated with the immunoglobulin heavy chain gene segments in Rag1−/− bone marrow pro–B cells, thymocytes and bone marrow–derived macrophages. (a) Chromatin immunoprecipitation assays, with anti–dimethylated H3-K9, of primary Rag1−/− bone marrow pro–B cells after short-term culture in IL-7; primary thymocytes; and bone marrow–derived macrophages. Bottom (c-Myc), the Myc promoter serves as a negative control. Wedges indicate 1:5 serial dilutions of 1% input DNA and 1:2 dilutions of H3-K9 immunoprecipitated fraction DNA; each begins (far left) with an undiluted sample. (b) Immunoprecipitated (IP) DNA from a, quantified with a phosphorimager; the results are presented as percentage of input. Data are from one experiment representative of multiple experiments (four pro–B cell cultures, five thymocyte preparations and three macrophage cultures). BD (below detection), intensities that were not at least twofold greater than that of the Myc promoter or background cannot be accurately quantified. −(far right), negative control. Bottom, the VH locus and relative location of the gene segments. (c) Chromatin immunoprecipitation assays, done as described in a with primers for Xist (X3), the VH gene V10b RSS and the Myc promoter, of embryonic fibroblasts from female mice (Female MEFs). Input dilutions begin at 1:5. Right, band intensities of immunoprecipitated DNA, presented as percentage of input (vertical axis), representative of three experiments.

Methylation of H3-K9 was not associated with VH promoters, VH RSSs or VH intergenic regions in pro–B cells (Fig. 1a) or in Rag1−/− pro–B cell lines (data not shown). A single exception was the promoter of a VH gene S107 member, S107b (also known as V11), which was reproducibly associated with methylated H3-K9 in multiple experiments. This gene segment recombines much less frequently than other S107 family members23. In this and subsequent experiments, we used H3-K9 methylation at the S107b promoter as an internal control to show immunoprecipitation with anti–dimethylated H3-K9 was efficient in B cells.

In the double-negative T cell line S49 (data not shown), primary thymocytes (both Rag1−/− double-negative and wild-type) and bone marrow–derived macrophages we found H3-K9 methylation associated with most but not all VH DNA sequences analyzed (9 of 11 sequences in thymocytes and 7 of 11, in bone marrow macrophages). H3-K9 methylation in the VH locus occurred in promoters, RSSs and intergenic regions and was substantially higher than that of the negative control (Myc) and higher than that in pro–B cells (Fig. 1a,b). The H3-K9 methylation pattern was reproducible in T or myeloid lineage cells regardless of whether they were cell lines or primary cells (data not shown) but was different between the two lineages (for example, 7183a, J558b and the intergenic regions in Fig. 1a,b). We found low or undetectable H3-K9 methylation of DH and JH regions in both B and T cells, consistent with the fact that some DH-to-JH recombination occurs before B lineage–T lineage commitment. Slightly more H3-K9 methylation was associated with DH and JH gene segments in myeloid cells.

We next assessed how H3-K9 methylation in the VH locus compared with H3-K9 methylation in regions where its function had been established before. We analyzed the region 5′ of the inactive X-specific transcript gene (Xist) on the X chromosome (X3), which is H3-K9 methylated and is thought to act as a nucleation center for Xist RNA–dependent X inactivation22. Consistent with published data, in female mouse embryonic fibroblasts, X3 had low but measurable H3-K9 methylation (1.14% of input DNA); we found slightly higher H3-K9 methylation (1.5% of input) for the V10b RSS in these cells (Fig. 1c). In hematopoietic cells, H3-K9 methylation in the VH locus was usually higher (5–10% of input; Fig. 1a,b). We conclude that this amount of H3-K9 methylation is in a range consistent with inactive chromatin.

We also studied H3-K4 methylation in the VH locus, a modification associated with active chromatin9. We analyzed Rag1−/− bone marrow pro–B cells and thymocytes by chromatin immunoprecipitation assay using anti–dimethylated H3-K4. We found reduced H3-K4 methylation associated with the VH locus in thymocytes compared with that of pro–B cells (Supplementary Fig. 1 online). Nevertheless, H3-K4 methylation, substantially greater than that found in thymocytes or in the negative control gene (encoding α-amylase 2), was associated with multiple gene segments (7183a, V10b P and V10bRSS) in pro-B cells. For both B and T cells, H3-K4 methylation was restricted to the gene segments (both promoter and RSS elements) and was not found in intergenic regions, suggesting different mechanisms were responsible for the regulation of H3-K4 and H3-K9 methylation. Thus, we found a general pattern of increased H3-K9 methylation in the VH locus in non–B cell lineages and absence of H3-K9 methylation from the VH locus in B cells (Fig. 1). These data provide an inverse correlation between this heterochromatic mark and VH-to-DJH recombination and suggest the possibility that H3-K9 methylation inhibits VH-to-DJH recombination.

H3-K9 and DNA methylation are not associated in the VH locus

H3-K9 methylation has been linked to methylation of CpG in adjacent DNA8. Although varied results have been reported on the function of DNA methylation in regulation of V(D)J recombination6, to our knowledge no studies have addressed the DNA methylation status of endogenous VH gene segments. We sought to determine if H3-K9 methylation of VH genes in non–B cell lineages was accompanied by increased CpG methylation. We used bisulfite conversion of genomic DNA from the double-negative T cell line S49; the Rag2−/− Abelson virus–transformed pro–B cell line 6312; primary pro–B cells; and primary double-negative thymocytes. We analyzed two VH gene segments that had strong H3-K9 methylation in T cells. We designed primers to amplify coding sequences 5′ of each RSS element and to include all (7183a) or a portion (V10b) of the RSS elements (Fig. 2 and Supplementary Table 1 online). Three of four CpG sites (two in V10b and one in 7183a) in the coding regions were methylated in all clones analyzed, whereas the fourth site (7183a) was partially methylated (about 20% of all clones; Fig. 2). Two CpG sites 3′ of the RSS element in the 7183a gene segment were not methylated (Fig. 2). This DNA methylation pattern was not lineage specific, suggesting that H3-K9 methylation and DNA methylation occur independently at the VH locus. Furthermore, as the VH gene segments analyzed can rearrange in the pro–B cell line24, these data show that recombination can occur at genes in which DNA methylation occurs outside of the RSS.

Figure 2.

Figure 2

Open in a new tab

DNA methylation is not linked to lineage-specific H3-K9 methylation of VH gene segments. DNA sequences of two VH gene segments, 7183a (81X) and V10b. Portions of each gene were amplified by PCR after bisulfite conversion of genomic DNA from a Rag2−/− Abelson virus–transformed pro–B cell line (6312), a double-negative T cell line (S49), Rag1−/− primary pro–B cells and Rag1−/− double-negative T cells. Black underlining, primers; gray boxed shading, RSS elements. After conversion, PCR products were subcloned and six clones per conversion were sequenced. Two to three bisulfite conversions were done on each cell type. Data were consistent between different cell populations and are representative results of experiments. Partial CpG methylation, about 20% of clones.

H3-K9 methylation is inversely correlated with Pax5 expression

Our data (Fig. 1) suggest that methylation of H3-K9 is involved in repressing VH-to-DJH recombination in non–B cells, indicating involvement of Pax5. Pax5 has the unique function of being required for B cell commitment and of blocking alternate hematopoietic fates, including the T cell and myeloid lineages25. Pax5−/− bone marrow cells have normal amounts of DH-to-JH recombination but have defective VH-to-DJH joining12,13, specifically for the DHJH-distal VH gene segments14,15. This suggests that the VH genes are not fully accessible to recombinase, despite having histone acetylation and germline transcription14,15. Although B220+ bone marrow cells from Pax5−/− mice cultured with IL-7 have hallmarks of the B cell lineage, they cannot develop into B cells. In appropriate conditions, these cells can differentiate into non–B cell lineages, including T cells, macrophages, dendritic cells and osteoclasts26,27. Thus, we reasoned that studying H3-K9 methylation in the VH locus in Pax5−/− bone marrow cells would show the methylation status of VH H3-K9 in multipotent progenitors and test a requirement for Pax5 in the regulation of VH H3-K9 methylation.

We did chromatin immunoprecipitation assays of sorted bone marrow pro–B cells isolated from Pax5−/− or Pax5+/+ (wild-type) mice after a short-term culture in IL-7, Fms-like tyrosine kinase 3 ligand and stem cell factor. Because of the potential of Pax5−/− bone marrow cells to differentiate into alternate lineages, we periodically used flow cytometry to ensure the cultures were not contaminated with committed myeloid or T cells (data not shown). We analyzed four independent cultures; in each case H3-K9 methylation of the VH regions was much higher in Pax5−/− cells than in wild-type cells (Fig. 3a). DH and JH segments could not be analyzed because of prior rearrangement. H3-K9 methylation was associated with most of the VH gene segments analyzed in the Pax5−/− cells (including 7183a, S107b, V10b, J558a and J558b). We conclude that loss of H3-K9 methylation at the VH locus accompanies commitment to the B cell fate and that removal of this heterochromatic mark requires the presence of Pax5. Additionally, the data provided evidence that H3-K9 methylation was the ‘default state’ for the VH locus in uncommitted hematopoietic progenitors as well as in cells committed to non–B cell lineages.

Figure 3.

Figure 3

Open in a new tab

H3-K9 methylation is inversely correlated to Pax5 expression. (a) Chromatin immunoprecipitation assay, with anti–dimethylated H3-K9, of sorted primary Pax5−/− bone marrow pro–B cells and wild-type (WT) bone marrow pro–B cells after short-term culture in the presence of IL-7, Fms-like tyrosine kinase 3 ligand and stem cell factor. Data are representative of four experiments. Wedges indicate 1:5 serial dilutions of 1% input and 1:2 dilutions of immunoprecipitated fractions; input samples begin with a 1:5 dilution. (b) Chromatin immunoprecipitation assay, with the anti–dimethylated H3-K9, of primary Rag1−/− bone marrow pro–B and Rag1−/− pre–B cells that express the transmembrane form of the human mu protein, after short-term culture in the presence of IL-7. Data are representative of three experiments. Wedges indicate 1:5 serial dilutions of 1% input and 1:2 dilutions of H3-K9 immunoprecipitated fractions; input samples begin with a 1:5 dilution. (c) Chromatin immunoprecipitation assays, with anti–dimethylated H3-K9, of Rag1−/− primary pro–B cells and primary thymocytes (also shown in Fig. 1). Data are representative of three experiments. Wedges indicate 1:5 serial dilutions of 1% input and 1:2 dilutions of H3-K9 immunoprecipitated fractions; each sample begins with a nondiluted sample.

Removal of H3-K9 methylation in the VH locus required Pax5 (Fig. 3a), and Pax5 is expressed throughout most of B cell development11. However, VH-to-DJH recombination is restricted to pro–B cell development and is inhibited in pre–B cells, in which functional rearrangement of one heavy chain allele inhibits rearrangement of the second allele. Inhibition involves changes in chromatin structure, as demonstrated by decreases in DNase I sensitivity, reduced germline transcription and histone deacetylation19,21,28. To determine whether H3-K9 methylation is involved in allelic exclusion, we analyzed H3-K9 methylation at the VH locus in pre–B cells of mice expressing a transmembrane form of the human mu protein29. There was no detectable change in H3-K9 methylation associated with the VH locus (Fig. 3b) in pro– versus pre–B cells. Also, the H3-K4 methylation pattern did not change during the process of allelic exclusion (Supplementary Fig. 2 online), suggesting that neither H3-K9 methylation nor H3-K4 methylation was involved in regulation of immunoglobulin heavy chain allelic exclusion. Thus, H3-K9 methylation in the VH locus is a default state in progenitors and its absence correlates with Pax5 expression throughout B cell development.

H3-K9 methylation might be a B cell–specific mechanism for the inhibition of VH gene recombination or it might be a common mechanism for the inhibition of B and T cell receptor rearrangement in inappropriate lineages. To distinguish these possibilities, we compared TCR Vγ and Vβ gene segments in pro–B cells and thymocytes (both Rag1−/− double-negative cells and wild-type cells). These loci are inaccessible for V(D)J recombination in B cells20. We found no lineage-specific differences in H3-K9 methylation for five individual TCR V gene segments (two Vγ and three Vβ gene segments) in B cells versus T cells (Fig. 3c) or for two V gene segments (Vγ2 and Vβ12) in T versus myeloid cells (data not shown). Thus, lineage specific regulation of H3-K9 methylation of V gene segments is a B cell–specific mechanism, consistent with its dependence on Pax5.

VH H3-K9 methylation is lost after forced Pax5 expression

Our data establish a requirement for Pax5 in the loss of H3-K9 methylation in the VH locus but do not distinguish between Pax5 itself or other mechanisms associated with B cell commitment and/or identity. To investigate Pax5 more directly, we took advantage of the B cell lymphoma cell line Myc5 (ref. 30), which was generated by ectopic expression of c-Myc in p53-deficient bone marrow cells31. During growth in culture, Myc5 cells reverted to a macrophage phenotype, resulting in loss of B cell surface molecules and the transcription factors Pax5 and early B cell factor. When Pax5 was ectopically expressed in Myc5 cells, macrophage-specific proteins were lost26 but full B cell commitment, determined by surface B220 and CD19 (ref. 30) and transcription factor expression (Fig. 4) did not occur, possibly because of the continued absence of early B cell factor. Thus, Myc5 cells provide a way to separate Pax5 expression from full B cell commitment, thereby allowing us to determine whether Pax5 is sufficient to remove H3-K9 methylation in the VH locus.

Figure 4.

Figure 4

Open in a new tab

Enforced expression of Pax5 in non–B cell lines results in the removal of H3-K9 methylation at the VH locus. (a). Surface expression of the macrophage protein CD11b and the leukocyte marker CD43, analyzed by flow cytometry of Myc5 cells after retroviral infection with a bicistronic retrovirus expressing Pax5 and puromycin resistance (Pax5-Puro) or a control retrovirus expressing only the puromycin-resistance gene (Puro). (b) Semiquantitative RT-PCR of total RNA from Myc5 cells after retroviral infection. This analysis includes mRNA encoding factors known to be important in B-cell development (PU.1, E12, E47, early B cell factor (EBF) and Pax5) and macrophage identity (macrophage colony-stimulating factor receptor (M-CSFR)); proteins encoded are included in parentheses along the left margin. Bottom, Gapd (glyceraldehyde phosphodehydrogenase) serves as the control. (c) Chromatin immunoprecipitation assays, with anti–dimethylated H3-K9, of Myc5 cells after retroviral infection. Wedges indicate 1:5 serial dilutions of 1% input and 1:2 dilutions of H3-K9 immunoprecipitated fractions; input samples begin with a 1:5 dilution. Right, immunoprecipitated DNA, representative of one of three experiments, presented as a percentage of input. Hatch marks across bars indicate a disruption in the linear scale on the vertical axis; numbers above bars indicate the final value. (d) Semiquantitative RT-PCR of total RNA from the double-negative T cell line S49 after retroviral infection. Gapd, control. (e) Chromatin immunoprecipitation assays, with anti–dimethylated H3-K9, of S49 cells after retroviral infection. Wedges indicate 1:5 serial dilutions of 1% input and 1:2 dilutions of H3-K9 immunoprecipitated fractions; input samples begin with a 1:5 dilution. Right, immunoprecipitated DNA band intensities, representative of one of three experiments, presented as a percentage of input.

Myc5 cells did demonstrate H3-K9 methylation in the VH locus (data not shown), consistent with their macrophage phenotype and the absence of Pax5. We infected Myc5 cells with a retrovirus expressing puromycin-resistance gene or a bicistronic virus expressing Pax5 and the puromycin-resistance gene. We selected cells based on puromycin resistance at 2 d after infection and analyzed cells 2 d later. We confirmed Pax5 function by loss of surface CD11b30 (Fig. 4a) and by a reduction in Csflr mRNA, which encodes the receptor for macrophage colony-stimulating factor32 (Fig. 4b). A substantial loss of H3-K9 methylation resulted at all VH gene segments tested, except for the promoter of S107b, in Myc5 cells expressing Pax5 compared with control cells (Fig. 4c). The loss of H3-K9 methylation occurred at both the promoter and RSS elements of VH gene segments and was not dependent on gene location. These data show that expression of Pax5 is sufficient for removal of H3-K9 methylation in the VH locus, despite the fact that these cells have not become fully converted to the B cell lineage.

To ensure this finding was not unique to a cell line or type of lineage commitment, we enforced expression of Pax5 in the double-negative T cell line S49 using the same approach. In agreement with published results25, T cell surface markers were not decreased after Pax5 expression (not shown). However, mRNA encoding the T cell–commitment factor Notch1 was reduced in Pax5-expressing cells compared with control cells25 (Fig. 4d). Within 4 d, Pax5 expression resulted in the loss of H3-K9 methylation at the VH locus in S49 cells (Fig. 4e). Thus, we conclude that expression of Pax5 was sufficient to cause removal of H3-K9 methylation at the VH locus regardless of cell type or commitment status, demonstrating a previously unknown function for Pax5.

Variant histone H3.3 in the VH locus after Pax5 expression

The mechanism by which Pax5 expression causes a loss of H3-K9 methylation at the VH locus (Fig. 4) is not known, and so far no histone demethylases have been described8. Although methylated histones can be removed by deposition of new nucleosomes during replication, cell division does not occur after Pax5 expression in pro–B cells before expression of a functionally recombined Igh and pre–B cell receptor33. However, replication-independent histone exchange involving deposition of the H3 variant H3.3 has been linked to the dynamic decondensation of heterochomatin34-36. Therefore, we explored the possibility that exchange involving H3.3 might contribute to Pax5-dependent loss of H3-K9 methylation at the VH locus.

We did chromatin immunoprecipitation assays using polyclonal anti–histone H3.3, focusing on the VH gene segments. We used S49 cells that had been infected with the Pax5-expressing retrovirus or the control virus expressing the puromycin-resistance gene only, as described above. Because 28S rDNA is associated with the histone variant H3.3 (refs. 34,35), we used it as a positive control. As α-amylase 2 is not associated with H3.3, it served as a negative control. There was a Pax5-dependent increase in H3.3 abundance at both the promoter and RSS elements of gene segments V10a and J558a but not in the 7183a RSS (Fig. 5). There was no increase in H3.3 content at the S107b promoter, where H3-K9 methylation remained high (Figs. 4,5). We did not, however, note a change in steady-state mRNA encoding either H3.3 or its chaperone HIRA (histone cell cycle regulation–defective homolog A)37 after Pax5 expression (data not shown). These data provide evidence that Pax5-dependent removal of H3-K9 methylation at the VH locus may proceed by histone exchange involving H3.3.

Figure 5.

Figure 5

Open in a new tab

Accumulation of the histone variant H3.3 at the VH locus correlates with the loss of H3-K9 methylation after Pax5 expression in a double-negative T cell line. Chromatin immunoprecipitation assays, with polyclonal anti–histone H3.3, of S49 cells after infection with a bicistronic retrovirus expressing Pax5 and the puromycin-resistance gene (Pax5-Puro) or the puromycin-resistance gene alone (Puro). Wedges indicate 1:5 serial dilutions of 1% input and 1:2 dilutions of immunoprecipitated fractions; input samples begin with a 1:5 dilution. Top, 28S rDNA serves as a positive control; bottom, α-amylase 2 serves as a negative control. Data are representative of three experiments.

DISCUSSION

Our data provide support for the idea that methylation of H3-K9 in the VH locus is important in enforcement of the strict B lineage specificity of VH-to-DJH recombination. Increasing evidence that H3-K9 methylation is associated with repressed chromatin in centromeric heterochromatin, X chromosome inactivation and gene-specific repression8 led us to hypothesize that H3-K9 methylation might be involved in repression of VH-to-DJH recombination in non–B cells. Consistent with this idea, H3-K9 methylation in the VH locus was absent from B cells for all but one of the sequences analyzed but was found in all non–B cells studied, including hematopoietic progenitors and committed T lymphocytes and myeloid cells. Although a previous study did not find lineage specific H3-K9 methylation associated with the VH genes17, this is most likely because of the different populations analyzed (transformed cell lines versus primary cells) and is consistent with the fact that transformed cell lines do not always repress V(D)J recombination properly38.

Targeting of H3-K9 methylation to an engineered, integrated recombination substrate provided direct evidence that H3-K9 methylation is sufficient to inhibit V(D)J recombination9. Our data have shown that in primary hematopoietic cells, H3-K9 methylation in the endogenous VH locus is inversely correlated with B cell commitment and VH-to-DJH recombination. These data provide a compelling argument that the ability of H3-K9 methylation to inhibit V(D)J recombination is physiologically relevant for inhibition of VH-to-DJH recombination in non–B lineage hematopoietic cells, thus providing a mechanism for the strict B cell specificity of this recombination event and the ordered recombination of DH-to-JH before VH-to-DJH.

Targeting of the histone methyl transferase G9a to the promoter of a single recombination substrate showed that local H3-K9 methylation can inhibit V(D)J recombination9. Our finding that the gene encoding S107b VH, which rearranges poorly23, remained associated with methylated H3-K9 in B cells provides evidence that local H3-K9 methylation does inhibit VH-to-DJH recombination of endogenous VH genes. In non–B lineage cells, we found H3-K9 methylation frequently associated with VH promoters and RSS elements and, based on the earlier work9, propose that VH-to-DJH recombination of these genes is inhibited by local H3-K9 methylation. However, not every VH gene was associated with H3-K9 methylation, and the pattern of methylated regions varied between T and myeloid lineage cells (Fig. 1). Furthermore, we found H3-K9 methylation of intergenic regions located at least 1 kilobase from any VH gene segment (Fig. 1). Thus, in non–B lineage cells, H3-K9 methylation in the 2-megabase VH locus seems to be generalized and not always localized to VH gene segments. As all VH genes fail to rearrange in non–B lineage cells, we suggest that generalized H3-K9 methylation in this large chromatin domain may inhibit V(D)J recombination throughout the locus, although additional studies will be necessary to establish this point definitively.

Multiple steps, including acquisition of activating modifications and loss of repressive modifications, are probably required for full accessibility of the VH locus. Histone acetylation is an early modification that precedes recombination, and its presence in B cells is found at a developmental time when H3-K9 methylation is lost21,39. Methylation of H3K27 seems to affect V(D)J recombination40, and localization away from the nuclear periphery and compaction between variable and constant regions15,41 have also been linked to regulation of V(D)J recombination. Although our data suggest that H3-K9 methylation inhibits V(D)J recombination, they do not indicate that this modification alone is sufficient in all settings to inhibit recombination. Pax5−/− pro–B cells can recombine DJH-proximal but not DJH-distal VH genes14,15. Histone H3 acetylation is present throughout the VH locus in Pax5−/− pro–B cells14; here we have shown that H3-K9 methylation is also present (Fig. 3). Thus, H3-K9 methylation may not block VH-to-DJH rearrangement of DJH-proximal VH genes in this cell environment where factors such as nuclear localization15 or proximity to regulatory elements may override H3-K9 repression. A differential effect for H3-K9 methylation on DJH-proximal and DJH-distal VH genes would be consistent with results of other studies showing that regulation of VH-to-DJH rearrangement varies between DJH-proximal and DJH-distal regions15,19,21,28,39,40,42,43 and that there is preferential rearrangement of DJH-proximal VH genes44. Alternatively, if the proportion of Pax5−/− pro–B cells that rearrange DJH-proximal VH genes is small, loss of H3-K9 methylation in this subset, if present, might not have been detected in our assay.

DNA methylation is linked with histone methylation in other systems8 and in studies using stably integrated V(D)J recombination substrates9,45. DNA demethylation is also important for endogenous VκJκ recombination46. In our experiments, H3-K9 methylation and DNA methylation did not correlate for the two VH genes studied. As these two VH genes recombine in vivo24, moderate DNA methylation does not inhibit VH-to-DJH recombination. This is in agreement with studies showing that genes methylated outside the RSS heptamer can recombine47 and that inhibitory effects of DNA methylation can be relieved by histone acetylation48,49. Although more VH genes must be analyzed, these data suggest that CpG methylation in the VH locus is not B cell specific or obligatorily linked to H3-K9 methylation.

Pax5 has the unique function in B cell development of inducing genes required for B cells and inhibiting genes required for other lineages11. In the absence of Pax5, bone marrow progenitors retain the ability to develop into T cells, macrophages, dendritic cells and osteoclasts26,27, even though they express B220 and the B cell transcription factors early B cell factor and E2A12. Our data have identified a previously unknown function for Pax5: it is necessary and sufficient for removal of H3-K9 methylation in the VH locus. We propose that Pax5-dependent removal of H3-K9 methylation is required for VH-to-DJH recombination and further B cell development. This provides an explanation for why lineage-specific VH-to-DJH recombination and further B cell development depend on Pax5

Understanding how histone methylation can be altered is advantageous, as no histone demethylase has been described and the half-life of histone methylation is the same as that of the histone protein50. There is increasing evidence that histone exchange with the variant H3.3 replacing H3 is involved in the switch from repressed to active chromatin34,36, and the histone chaperone HIRA has been identified as being responsible for replication-independent nucleosome exchange involving H3.3 (ref. 37). Given that pro–B cells do not divide in response to Pax5 expression, we suspected that replication-independent exchange involving H3.3 might mediate the loss of H3-K9 methylation in the VH region noted after Pax5 expression. Indeed, for four of five VH regions tested we found increased H3.3 in cells after Pax5 expression (Fig. 5). Although this finding does not rule out the possibility of other mechanisms, it provides an important clue that H3.3 histone exchange is involved in the removal of H3-K9 methylation in the VH region. We do not understand the lack of H3.3 in the 7183a gene, but it could indicate a different mechanism for removal of H3-K9 methylation. Alternatively, as S49 cells divide in culture, providing the possibility of histone replacement, there may be differential replacement of H3.3 after insertion by histone exchange.

Deposition of H3.3 may occur at areas of active transcription34. Antisense transcription has been reported to precede VH-to-DJH recombination at the CD19+ stage, when Pax5 is first expressed. The pattern of ‘anti-transcription’ (transcription of the strand complementary to that transcribed for mRNA encoding immunoglobulin variable regions), in both VH genes and intergenic sequences, is similar to that of H3-K9 methylation51. It may be that ‘pioneer polymerases’, which can transcribe through heterochromatin, might initiate ‘anti-transcription’ and cause changes in chromatin structure52. We speculate that anti-transcription in the VH region might precede and allow H3.3 histone exchange, which would then remove H3-K9 methylation and prime the locus for VH-to-DJH recombination. The known activities of Pax5 provide few clues for how Pax5 might drive this process. Ectopic Pax5 caused removal of H3-K9 methylation within a few days and was independent of full B cell commitment or expression of early B cell factor, ‘arguing against’ the possibility of a developmental mechanism. We found no evidence that Pax5 induces transcription of H3.3 or its chaperone HIRA. If our speculation regarding H3.3 is correct, Pax5 or a gene product regulated by Pax5 may bind in the VH locus and recruit pioneer polymerases for anti-transcription. However, many other possibilities remain to be explored. Pax5 expression in T cells was shown to cause rearrangement of DJH-proximal VH genes15,16, and Pax5 was found to be required for compaction of the heavy chain chromatin15. The mechanism of Pax5 action in T cells is also unknown15,16 and may not be related to its effects in B cells. Nonetheless, removal of H3-K9 methylation may promote chromatin compaction that juxtaposes VH and DJH gene segments and allows their recombination.

METHODS

Cell lines and cell cultures

The 6312 cells (Rag2−/− Abelson virus–transformed pro–B cells) and S49 cells (double-negative thymoma; a gift from G. Siu, Amgen) were cultured in RPMI medium plus 10% heat-inactivated FCS (Gemini), 50 μM β-mercaptoethanol and gentamicin. Myc5 culture conditions have been described30. Bone marrow pro–B cells are derived from Rag1−/− mice, and pre–B cells, from mu-transgenic mice on a Rag1−/− background. Both B cell types underwent short-term culture in the presence of IL-7 as described21. Wild-type thymocytes (which are the majority at the double-positive stage) were obtained from C57BL/6 mice (unsorted) and double-negative thymocytes were pooled from Rag1−/− mice, after red blood cell lysis and were used immediately in chromatin immunoprecipitation assays. Primary macrophages were obtained by culture of bone marrow cells in the presence of macrophage colony-stimulating factor for 8 d according to published protocols53. Mature CD3 splenocyte samples were isolated from a wild-type C57BL/6 mouse, subjected to red blood cell lysis and negatively depleted of B220 (Dynabeads) and major histocompatibility complex class II (Miltenyi) as specified by the manufacturer's protocol. All cell populations were checked for purity by flow cytometry (at least 90% pure, using B220-BP1 for pro–B and pre–B cells, CD11b for macrophages and CD4–CD8–Thy-1.2 for thymocytes) and were used immediately in chromatin immunoprecipitation assays. Bone marrow cells were collected from 15-day-old Pax5−/− and wild-type control mice and were cultured according to a published protocol54, with slight modifications: cells were cultured in X-VIVO-15 medium (Fisher) containing 0.5% endotoxin-free BSA (Stem Cell Technologies), 50 μM β-mercaptoethanol, 25 ng/ml of Fms-like tyro-sine kinase 3 ligand, 20 ng/ml of stem cell factor and 2.5 ng/ml of IL-7 (R&D Systems). Fresh media was added every 3 d. At days 10–14 of culture, pro–B cells were purified by sterile flow cytometry sorting of B220+CD43hiGR-1− cells (Pax5−/− and wild-type) or CD19+BP-1+CD11b− cells (wild-type). Cells were then cultured in the X-VIVO media with cytokines as described above and were analyzed by chromatin immunoprecipitation. Animal protocols were approved by the Institutional Animal Care and Use Committee.

Chromatin immunoprecipitation

These assays were done as described21 with the following modifications. Each immunoprecipitation used 5 × 106 to 7.5 × 106 cells, with 5 μl of the following antibodies: polyclonal anti–dimethylhistone H3-K9 (07-212; Upstate Biotechnology), anti–dimethyl-histone H3-K4 (07-030; Upstate Biotechnology) and polyclonal anti–histone H3.3 (4263; Abcam). After ethanol precipitation, DNA was resuspended in 100 μl per 5 × 106 cells and 1–2 μl was used as a template for each PCR. Serial dilutions of were made at a ratio of 1:5 for 1% input or at a ratio of 1:2 for immunoprecipitated fractions. Comparable dilutions between all primer sets are shown. All PCRs (primers, Supplementary Table 1 online) used a volume of 50 μl and the following conditions: 30 s at 94 °C, 30 s at the annealing temperature (Supplementary Table 1 online) and 30 s at 72 °C. PCR products were separated by 1.5% agarose gel electrophoresis, blotted onto Hybond-N membranes (Amersham Biosciences) and probed with specific oligonucleotides (Supplementary Table 1 online) that were end-labeled with [γ-32P]ATP using T4 Polynucleotide Kinase (Amersham Pharmacia). Southern hybridization results were analyzed with a phosphorimager (Molecular Dynamics). The cycle number for each primer set varied from 29 to 37, according to the strength of input signal, and were adjusted so that dilutions fell into the linear range when quantified with the phosphorimager (Molecular Dynamics) after Southern hybridization with an internal probe. No hybridization was required for the H3.3 antibody, as specified by the manufacturer's protocol. In graphs, immunoprecipitated DNA band intensities were expressed relative to the signal of 0.01% input with volume intensity numbers as determined by ImageQuant (Molecular Dynamics) that fell into the linear range, as shown on the vertical axes (Figs. 1, 4). The graphed data are representative of the primary data from a single experiment and represent a trend that was reproduced three to five times from individual preparations. Numbers that were not twofold above background or c-Myc and numbers that did not ‘dilute’ appropriately (because of low signal and sporadic background) could not be quantified. In some cases, additional dilutions were done (but are not shown) to ensure signals were within the linear range.

RT-PCR

RNA was isolated by the TRIzol method (Gibco-BRL). A total of 1–2 μg RNA was reverse transcribed with random hexamers after DNase (Promega) treatment as described24. Primers and annealing temperatures are listed in Supplementary Table 1 online.

Retrovirus preparation and infection

Retroviruses were generated as described52. Myc5 cells and S49 cells were infected at a multiplicity of infection of 1–2 in the presence of 5 μg of polybrene per ml. After cells had ‘recovered’ for 2 d after infection, puromycin (9 μg/ml for Myc5 and 4 μg/ml for S49) was added to the media and after 2 d the cells were collected, dead cells were removed with Histopaque-1077 (Sigma) and the remaining cells were used immediately for chromatin immunoprecipitation. RNA was also obtained at this time and flow cytometry was done.

Plasmids

For the generation of retroviruses containing cDNA encoding Pax5, a NotI fragment of cDNA from Pax5-pEBB (B. Birshtein, Albert Einstein College of Medicine New York, New York) was cloned into a NotI site of the retrovirus vector Vxy-Puro (L. Staudt, National Institutes of Health, Bethesda, Maryland).

Methylation analysis

For methylation analysis, 1.5 μg of genomic DNA was treated for 16 h at 50 °C with sodium bisulfite as described55. Bisulfite-treated DNA was subject to nested PCR and was sequenced, with over 99% of all non-CpG cytosine residues being converted to thymidine residues. At least six bisulfite-converted clones were sequenced and multiple bisulfite conversions were done on individual cell types. PCR (primers, Supplementary Fig. 1 online) used a volume of 100 μl and an annealing temperature of 55 °C.

Flow cytometry

For determination of cell purity and developmental stage, cells were stained on ice for 20 min with 1:100 dilutions of the following antibodies in staining buffer (3% fetal calf serum and 1× PBS): allophycocyanin–anti-CD45R (B220), phycoerythrin–anti-BP-1, phycoerythrin–anti-CD11b (Mac1), fluorescein isothiocyanate–anti-Mac3, phycoerythrin–anti-CD90 (Thy-1.2), phycoerythrin–anti-CD4, fluorescein isothiocyanate–anti-CD8 and phycoerythrin–anti-CD19 (all antibodies were purchased from PharMingen) and were analyzed on a FACSCalibur with CellQuest software (Becton Dickinson).

Supplementary Material

Supplemental Data

ACKNOWLEDGMENTS

We thank M. Busslinger (Reasearch Institute of Molecular Pathology, Vienna, Austria) for the gift of the Pax5−/− mice; P. Kincade (Oklahoma Medical Research Foundation, Oklahoma City, Oklahoma) for direction in establishing the culture conditions used for the Pax5−/− cells; C. Wickens for assistance in maintaining the mouse colonies; C. Tunyaplin and Y. Zou for critical reading of the manuscript; and D. Spector for advice regarding controls for the H3.3 analyses. Supported by National Institutes of Health (R01 AI43576, R01 AI32524, R01 CA102709 and GM40924).

Footnotes

Note: Supplementary information is available on the Nature Immunology website.

COMPETING INTERESTS STATEMENT

The authors declare that they have no competing financial interests.

References

  • 1.Rothenberg EV, Anderson MK. Elements of transcription factor network design for T-lineage specification. Dev. Biol. 2002;246:29–44. doi: 10.1006/dbio.2002.0667. [DOI] [PubMed] [Google Scholar]
  • 2.Bassing CH, Swat W, Alt FW. The mechanism and regulation of chromosomal V(D)J recombination. Cell. 2002;109:S45–55. doi: 10.1016/s0092-8674(02)00675-x. [DOI] [PubMed] [Google Scholar]
  • 3.Kwon J, Imbalzano AN, Matthews A, Oettinger MA. Accessibility of nucleosomal DNA to V(D)J cleavage is modulated by RSS positioning and HMG1. Mol. Cell. 1998;2:829–839. doi: 10.1016/s1097-2765(00)80297-x. [DOI] [PubMed] [Google Scholar]
  • 4.Golding A, Chandler S, Ballestar E, Wolffe AP, Schlissel MS. Nucleosome structure completely inhibits in vitro cleavage by the V(D)J recombinase. EMBO J. 1999;18:3712–3723. doi: 10.1093/emboj/18.13.3712. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.McBlane F, Boyes J. Stimulation of V(D)J recombination by histone acetylation. Curr. Biol. 2000;10:483–486. doi: 10.1016/s0960-9822(00)00449-8. [DOI] [PubMed] [Google Scholar]
  • 6.Krangel MS. Gene segment selection in V(D)J recombination: accessibility and beyond. Nat. Immunol. 2003;4:624–630. doi: 10.1038/ni0703-624. [DOI] [PubMed] [Google Scholar]
  • 7.Sikes ML, Meade A, Tripathi R, Krangel MS, Oltz EM. Regulation of V(D)J recombination: a dominant role for promoter positioning in gene segment accessibility. Proc. Natl. Acad. Sci. USA. 2002;99:12309–12314. doi: 10.1073/pnas.182166699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Sims RJ, 3rd, Nishioka K, Reinberg D. Histone lysine methylation: a signature for chromatin function. Trends Genet. 2003;19:629–639. doi: 10.1016/j.tig.2003.09.007. [DOI] [PubMed] [Google Scholar]
  • 9.Osipovich O, Milley R, Meade A, Tachibana M, Shinkai Y, Krangel MS, Oltz EM. Targeted inhibition of V(D)J recombination by a histone methyltransferase. Nat. Immunol. 2004;5:309–316. doi: 10.1038/ni1042. [DOI] [PubMed] [Google Scholar]
  • 10.Borghesi L, et al. B Lineage-specific regulation of V(D)J recombinase activity Is established in common lymphoid progenitors. J. Exp. Med. 2004;199:491–502. doi: 10.1084/jem.20031800. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Busslinger M. Transcriptional control of early B cell development1. Annu. Rev. Immunol. 2004;22:55–79. doi: 10.1146/annurev.immunol.22.012703.104807. [DOI] [PubMed] [Google Scholar]
  • 12.Nutt SL, Urbanek P, Rolink A, Busslinger M. Essential functions of Pax5 (BSAP) in pro-B cell development: difference between fetal and adult B lymphopoiesis and reduced V-to-DJ recombination at the IgH locus. Genes Dev. 1997;11:476–491. doi: 10.1101/gad.11.4.476. [DOI] [PubMed] [Google Scholar]
  • 13.Urbanek P, Wang ZQ, Fetka I, Wagner EF, Busslinger M. Complete block of early B cell differentiation and altered patterning of the posterior midbrain in mice lacking Pax5/BSAP. Cell. 1994;79:901–912. doi: 10.1016/0092-8674(94)90079-5. [DOI] [PubMed] [Google Scholar]
  • 14.Hesslein DG, Pflugh DL, Chowdhury D, Bothwell AL, Sen R, Schatz DG. Pax5 is required for recombination of transcribed, acetylated, 5′ IgH V gene segments. Genes Dev. 2003;17:37–42. doi: 10.1101/gad.1031403. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Fuxa M, et al. Pax5 induces V-to-DJ rearrangements and locus contraction of the immunoglobulin heavy-chain gene. Genes Dev. 2004;18:411–422. doi: 10.1101/gad.291504. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Hsu LY, Liang HE, Johnson K, Kang C, Schlissel MS. Pax5 activates immunoglobulin heavy chain V to DJ rearrangement in transgenic thymocytes. J. Exp. Med. 2004;199:825–830. doi: 10.1084/jem.20032249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Morshead KB, Ciccone DN, Taverna SD, Allis CD, Oettinger MA. Antigen receptor loci poised for V(D)J rearrangement are broadly associated with BRG1 and flanked by peaks of histone H3 dimethylated at lysine 4. Proc. Natl. Acad. Sci. USA. 2003;100:11577–11582. doi: 10.1073/pnas.1932643100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Hesslein DG, Schatz DG. Factors and forces controlling V(D)J recombination. Adv. Immunol. 2001;78:169–232. doi: 10.1016/s0065-2776(01)78004-2. [DOI] [PubMed] [Google Scholar]
  • 19.Chowdhury D, Sen R. Transient IL-7/IL-7R signaling provides a mechanism for feedback inhibition of immunoglobulin heavy chain gene rearrangements. Immunity. 2003;18:229–241. doi: 10.1016/s1074-7613(03)00030-x. [DOI] [PubMed] [Google Scholar]
  • 20.Stanhope-Baker P, Hudson KM, Shaffer AL, Constantinescu A, Schlissel MS. Cell type-specific chromatin structure determines the targeting of V(D)J recombinase activity in vitro. Cell. 1996;85:887–897. doi: 10.1016/s0092-8674(00)81272-6. [DOI] [PubMed] [Google Scholar]
  • 21.Johnson K, Angelin-Duclos C, Park S, Calame KL. Changes in histone acetylation are associated with differences in accessibility of V(H) gene segments to V-DJ recombination during B-cell ontogeny and development. Mol. Cell. Biol. 2003;23:2438–2450. doi: 10.1128/MCB.23.7.2438-2450.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Heard E, Rougeulle C, Arnaud D, Allis CD, Spector DL. Methylation of his-tone H3 at Lys-9 is an early mark on the X chromosome during X inactivation. Cell. 2001;107:727–738. doi: 10.1016/s0092-8674(01)00598-0. [DOI] [PubMed] [Google Scholar]
  • 23.Love VA, Lugo G, Merz D, Feeney AJ. Individual V(H) promoters vary in strength, but the frequency of rearrangement of those V(H) genes does not correlate with promoter strength nor enhancer-independence. Mol. Immunol. 2000;37:29–39. doi: 10.1016/s0161-5890(00)00023-7. [DOI] [PubMed] [Google Scholar]
  • 24.Angelin-Duclos C, Calame K. Evidence that immunoglobulin VH-DJ recombination does not require germ line transcription of the recombining variable gene segment. Mol. Cell. Biol. 1998;18:6253–6264. doi: 10.1128/mcb.18.11.6253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Souabni A, Cobaleda C, Schebesta M, Busslinger M. Pax5 promotes B lymphopoiesis and blocks T cell development by repressing notch1. Immunity. 2002;17:781–793. doi: 10.1016/s1074-7613(02)00472-7. [DOI] [PubMed] [Google Scholar]
  • 26.Nutt SL, Heavey B, Rolink AG, Busslinger M. Commitment to the B-lymphoid lineage depends on the transcription factor Pax5. Nature. 1999;401:556–562. doi: 10.1038/44076. [DOI] [PubMed] [Google Scholar]
  • 27.Rolink AG, Nutt SL, Melchers F, Busslinger M. Long-term in vivo reconstitution of T-cell development by Pax5-deficient B-cell progenitors. Nature. 1999;401:603–606. doi: 10.1038/44164. [DOI] [PubMed] [Google Scholar]
  • 28.Haines BB, Brodeur PH. Accessibility changes across the mouse Igh-V locus during B cell development. Eur. J. Immunol. 1998;28:4228–4235. doi: 10.1002/(SICI)1521-4141(199812)28:12<4228::AID-IMMU4228>3.0.CO;2-H. [DOI] [PubMed] [Google Scholar]
  • 29.Nussenzweig MC, et al. Allelic exclusion in transgenic mice that express the membrane form of immunoglobulin μ. Science. 1987;236:816–819. doi: 10.1126/science.3107126. [DOI] [PubMed] [Google Scholar]
  • 30.Yu D, Allman D, Goldschmidt MH, Atchison MH, Monroe JG, Thomas-Tikhonenko A. Oscillation between B-lymphoid and myeloid lineages in Myc-induced hematopoietic tumors following spontaneous silencing/reactivation of the EBF/Pax5 pathway. Blood. 2003;101:1950–1955. doi: 10.1182/blood-2002-06-1797. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Yu D, Thomas-Tikhonenko A. A non-transgenic mouse model for B-cell lymphoma: in vivo infection of p53-null bone marrow progenitors by a Myc retrovirus is sufficient for tumorigenesis. Oncogene. 2002;21:1922–1927. doi: 10.1038/sj.onc.1205244. [DOI] [PubMed] [Google Scholar]
  • 32.Heavey B, Charalambous C, Cobaleda C, Busslinger M. Myeloid lineage switch of Pax5 mutant but not wild-type B cell progenitors by C/EBPα and GATA factors. EMBO J. 2003;22:3887–3897. doi: 10.1093/emboj/cdg380. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Rolink AG, Winkler T, Melchers F, Andersson J. Precursor B cell receptor-dependent B cell proliferation and differentiation does not require the bone marrow or fetal liver environment. J. Exp. Med. 2000;191:23–32. doi: 10.1084/jem.191.1.23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Janicki SM, et al. From silencing to gene expression: real-time analysis in single cells. Cell. 2004;116:683–698. doi: 10.1016/s0092-8674(04)00171-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Ahmad K, Henikoff S. The histone variant H3.3 marks active chromatin by replication-independent nucleosome assembly. Mol. Cell. 2002;9:1191–1200. doi: 10.1016/s1097-2765(02)00542-7. [DOI] [PubMed] [Google Scholar]
  • 36.Ahmad K, Henikoff S. Histone H3 variants specify modes of chromatin assembly. Proc. Natl. Acad. Sci. USA. 2002;99:16477–16484. doi: 10.1073/pnas.172403699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Tagami H, Ray-Gallet D, Almouzni G, Nakatani Y. Histone H3.1 and H3.3 complexes mediate nucleosome assembly pathways dependent or independent of DNA synthesis. Cell. 2004;116:51–61. doi: 10.1016/s0092-8674(03)01064-x. [DOI] [PubMed] [Google Scholar]
  • 38.Schlissel M, Corcoran L, Baltimore D. Virus-transformed pre-B cells show ordered activation but not inactivation of immunoglobulin gene rearrangement and transcription. J. Exp. Med. 1991;173:711–720. doi: 10.1084/jem.173.3.711. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Chowdhury D, Sen R. Stepwise activation of the immunoglobulin mu heavy chain gene locus. EMBO J. 2001;20:6394–6403. doi: 10.1093/emboj/20.22.6394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Su IH, et al. Ezh2 controls B cell development through histone H3 methylation and Igh rearrangement. Nat. Immunol. 2002;4:142–131. doi: 10.1038/ni876. [DOI] [PubMed] [Google Scholar]
  • 41.Kosak ST, et al. Subnuclear compartmentalization of immunoglobulin loci during lymphocyte development. Science. 2002;296:158–162. doi: 10.1126/science.1068768. [DOI] [PubMed] [Google Scholar]
  • 42.Corcoran AE, Riddell A, Krooshoop D, Venkitaraman AR. Impaired immunoglobulin gene rearrangement in mice lacking the IL-7 receptor. Nature. 1998;391:904–907. doi: 10.1038/36122. [DOI] [PubMed] [Google Scholar]
  • 43.Costa TE, Suh H, Nussenzweig MC. Chromosomal position of rearranging gene segments influences allelic exclusion in transgenic mice. Proc. Natl. Acad. Sci. USA. 1992;89:2205–2208. doi: 10.1073/pnas.89.6.2205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.ten Boekel E, Melchers F, Rolink AG. Changes in the V(H) gene repertoire of developing precursor B lymphocytes in mouse bone marrow mediated by the pre-B cell receptor. Immunity. 1997;7:357–368. doi: 10.1016/s1074-7613(00)80357-x. [DOI] [PubMed] [Google Scholar]
  • 45.Ji Y, Zhang J, Lee AI, Cedar H, Bergman Y. A multistep mechanism for the activation of rearrangement in the immune system. Proc. Natl. Acad. Sci. USA. 2003;100:7557–7562. doi: 10.1073/pnas.0932635100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Mostoslavsky R, et al. κ chain monoallelic demethylation and the establishment of allelic exclusion. Genes Dev. 1998;12:1801–1811. doi: 10.1101/gad.12.12.1801. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Whitehurst CE, Schlissel MS, Chen J. Deletion of germline promoter PDβ1 from the TCRβ locus causes hypermethylation that impairs Dβ1 recombination by multiple mechanisms. Immunity. 2000;13:703–714. doi: 10.1016/s1074-7613(00)00069-8. [DOI] [PubMed] [Google Scholar]
  • 48.Durum SK, et al. Interleukin 7 receptor control of T cell receptor γ gene rearrangement: role of receptor-associated chains and locus accessibility. J. Exp. Med. 1998;188:2233–2341. doi: 10.1084/jem.188.12.2233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Cherry SR, Baltimore D. Chromatin remodeling directly activates V(D)J recombination. Proc. Natl. Acad. Sci. USA. 1999;96:10788–10793. doi: 10.1073/pnas.96.19.10788. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Bannister AJ, Schneider R, Kouzarides T. Histone methylation: dynamic or static? Cell. 2002;109:801–806. doi: 10.1016/s0092-8674(02)00798-5. [DOI] [PubMed] [Google Scholar]
  • 51.Bolland DJ, et al. Antisense intergenic transcription in V(D)J recombination. Nat. Immunol. 2004;5:630–637. doi: 10.1038/ni1068. [DOI] [PubMed] [Google Scholar]
  • 52.Orphanides G, Reinberg D. RNA polymerase II elongation through chromatin. Nature. 2000;407:471–475. doi: 10.1038/35035000. [DOI] [PubMed] [Google Scholar]
  • 53.Chang D, Angelin-Duclos C, Calame K. BLIMP-1: trigger for differentiation of myeloid lineage. Nat. Immunol. 2000;1:169–176. doi: 10.1038/77861. [DOI] [PubMed] [Google Scholar]
  • 54.Kouro T, Medina KL, Oritani K, Kincade PW. Characteristics of early murine B-lymphocyte precursors and their direct sensitivity to negative regulators. Blood. 2001;97:2708–2715. doi: 10.1182/blood.v97.9.2708. [DOI] [PubMed] [Google Scholar]
  • 55.Grunau C, Clark SJ, Rosenthal A. Bisulfite genomic sequencing: systematic investigation of critical experimental parameters. Nucleic Acids. Res. 2001;29:E65–65. doi: 10.1093/nar/29.13.e65. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supplemental Data
NIHMS13347-supplement-supp.pdf (902.1KB, pdf)

RESOURCES