| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Graduate School of Biotechnology, Institute of Life Science and Resources, Kyung Hee University, Yongin, Gyeonggi-do, Korea;
Department of Biological Science, Sookmyung Women’s University, Seoul, Korea; and
Department of Anatomy, College of Medicine, Seoul National University, Seoul, Korea
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It has been long suspected that a certain other gene in the vicinity of the IgJ gene might be expressed in the pre-B cell stage during B cell development (7) under the control of DNase I-hypersensitive sites (HSS)4 3/4 enhancer, which was shown to play as the IgJ enhancer in the plasma cell stage (8). This speculation has been based on the fact that the chromatin of HSS3/4 enhancer is also opened in the pre-B cell stage, although the chromatin of the HSS1 IgJ promoter is in a closed state and so its expression is not yet allowed in this pre-B cell stage (7).
The charged amino acid-rich leucine zipper 1 (Crlz1) gene, which was cloned as encoding a CBF
2-associated protein by the yeast two-hybrid technique (9), has been mapped on the mouse chromosome 5 at 30 kb upstream of the IgJ gene with the HSS3/4 enhancer positioned between them in both our genomic clones and basic local alignment search tool Contig sequences. Transcriptional direction of the Crlz1 gene was found to be opposite to that of the IgJ gene. The Crlz1 gene was speculated as a potential pre-B cell stage-specific gene that had been searched in the vicinity of the IgJ gene, and actually turned out to be truly the case when we performed a Northern blot experiment using the B cell lines representing various developmental stages and an RT-PCR experiment using the mouse normal pre-B cells sorted out by a flow cytometer.
As mentioned above, the Crlz1 gene was originally cloned because its protein product could bind to the 2 subunit of core-binding factor (CBF) transcription factor (9), which has been reported to play an essential role in both hemopoiesis (10, 11, 12, 13) and osteogenesis (14, 15). The CBF subunit arises in three isoforms that are presumed to arise by alternative RNA splicing and encode the polypeptides consisting of 187, 182, and 155 aa, respectively (16). Only the CBF2 (182 aa) isoform was shown to bind to Crlz1 in a yeast-two-hybrid assay (9). Although the function of Crlz1 is not yet clear, it might be related to the transcriptional activities of CBF or its yeast homolog, Sas10 (17).
The eukaryotic genome is packaged into a chromatin of nucleosomes in the nucleus. For a eukaryotic gene to be expressed, the chromatin around the gene, especially on the regulatory elements such as promoter and enhancer, must be opened. The opening and closing of chromatin is known to be regulated by signals regulating histone modification and/or chromatin remodeling (18, 19). The modification of histones includes acetylation, methylation, phosphorylation, and ubiquitination. Especially, the N-terminal tails of H3 and H4 histones on the opened chromatin are known to be acetylated by a histone acetyltransferases such as CBP/p300 which might be recruited by transcriptional activators, whereas those on the closed chromatin are known to be deacetylated by a histone deacetylase recruited by transcriptional repressors (20, 21, 22, 23). A bromodomain-containing factor could bind on this acetylated histone to initiate a chromatin remodeling (19, 24). Transcription on this opened and/or remodeled chromatin is then initiated by a transcription complex assembly on its promoter, which might be facilitated or inhibited by the proximal and/or distal DNA regulatory elements around the gene such as enhancer and silencer, which contain various sequence motifs for binding their respective transcription factors.
In this report, we show a novel phenomenon of stage-specific gene expression regulation of two neighboring Crlz1 and IgJ genes during B cell development. Their gene expressions during B cell development are shown to be regulated not only by chromatin accessibility as assayed by the DNase I hypersensitivity, but also by histone acetylation as assayed by chromatin immunoprecipitation (ChIP) over the Crlz1-IgJ locus. These dynamic stage-specific expression profiles of two neighboring genes, which are 30 kb apart, during B cell development could suggest the presence of some additional regulatory elements such as boundary element or insulator (25) between them to coordinate chromatin accessibility and histone acetylation over the locus. It remains to be elucidated whether some of the identified DNase I-HSS over the locus might correspond to those regulatory elements. The Crlz1-IgJ locus showing the phenomenon of stage-specific gene expression regulation of two neighboring genes should provide a good model system for studying the regulation of gene expression during B cell development.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cell lines were maintained at 37°C in DMEM or RPMI 1640 medium supplemented with 10% heat-inactivated FBS, 2 mM L-glutamine, 0.1 mM nonessential amino acids (Invitrogen Life Technologies), 1 mM sodium pyruvate, 50 µM 2-ME, 100 U/ml penicillin G, and 100 µg/ml streptomycin in an atmosphere of 5% CO2 saturated with water. PD36 is an Abelson virus-transformed pre-B cell line (26). S194 (TIB-19; American Type Culture Collection (ATCC)) and MOPC315 (TIB-23; ATCC) are plasmacytoma cell lines. WEHI231 (CRL-1702; ATCC) is an immature B cell line. BCL1 is an IL-2-inducible presecretor B cell line (5). EL4 (TIB-39; ATCC) is a T cell line. RAW264.7 (TIB-71; ATCC) and NIH3T3 (CRL-1658; ATCC) are macrophage and fibroblast cell lines, respectively. All the cell lines are mouse originated.
DNase I-hypersensitivity assay and transient transfection luciferase assay
DNase I-hypersensitivity assay and transient transfection luciferase assay were performed essentially as described previously (8).
Northern blot
Total RNA was isolated using the TRIzol reagent (Invitrogen Life Technologies). Then, 25 µg of total RNA was run on a 1% formaldehyde-agarose gel and blotted to a Hybond-N+ nylon membrane (Amersham Biosciences). The blot was probed with the random primer-labeled [32P]Crlz1 cDNA fragment. The probed blot was then stripped and reprobed with the random primer [32P]IgJ cDNA fragment. Finally, the blot was again stripped and reprobed with the random primer-labeled [32P]G3PDH cDNA (Clontech) for a positive loading control.
Flow cytometric sorting of the mouse normal pre-B cells and RT-PCR
Normal leukocytes were isolated from the bone marrow, spleen, and thymus of a 6-wk-old BALB/c mouse (Samtaco) and resuspended in the cold PBS containing 1% BSA and 0.05% NaN3 (PBS-BN1). The isolated leukocytes were filtered through a 70-µm nylon mesh (Falcon). RBC were removed by lysis for 5 min in the RBC lysis buffer (150 mM NH4Cl, 10 mM NaHCO3, 0.1 mM Na2EDTA). After washing with the cold PBS containing 0.2% BSA and 0.05% NaN3 (PBS-BN2), cells were resuspended in a concentration of 2.5 x 107 cells/ml in PBS-BN2. A total of 200 µl of the cell suspension was mixed with 300 µl of PBS-BN2 containing 3 µg of anti-mouse CD16/CD32 (553142; BD Biosciences) and incubated on a rotator for 30 min at 4°C. After washing twice with PBS-BN2, cells were resuspended in 500 µl of PBS-BN2. The resuspended cells were incubated with 3 µg of biotin-conjugated rat anti-mouse pre-BCR Ab (551863; BD Biosciences). After washing with PBS-BN2, cells were stained with streptavidin-PE conjugate (554061; BD Biosciences). The PE-stained cells were washed and then double-stained with 3 µg of FITC-conjugated rat anti-mouse CD45R/B220 Ab (553088; BD Biosciences). The double-stained cells were again washed, resuspended in 500 µl of PBS-BN2, and sorted by a flow cytometer (FACSVantage; BD Biosciences). The sorted cells were collected from the gates of B220+/pre-BCR+ and B220+/pre-BCR–. The same procedure was performed in parallel for the collection of two more cell populations of B220+/IgM+ and B220+/IgM– phenotypes, where PE-conjugated rat anti-mouse IgM Ab (553409; BD Biosciences) was used instead of the biotin-conjugated rat anti-mouse pre-BCR Ab and streptavidin-PE conjugate.
Total RNA was extracted from the collected cells with the TRIzol reagent (Invitrogen Life Technologies), and then treated with DNase I (Promega) to remove any possible contaminating DNA. After the usual phenol-chloroform extraction and ethanol precipitation, the extracted RNA was reverse-transcribed to make cDNA using SuperScript II RT (Invitrogen Life Technologies) with the oligo-dT as a primer. Finally, the cDNA was used for a semiquantitative PCR analysis. The PCR primer pairs for measuring the expression of Crlz1, VpreB, 
5, IL-7R
, and -actin genes are given in Table I (see also, Ref. 27). The amplification was performed using a two-step PCR. The first step PCR with a predenaturation at 94°C for 5 min was performed in two cycles of 94°C for 1 min, 44–54°C for 7.5 min depending on the primers and 72°C for 2 min. The second step PCR was performed in 28–36 cycles of 94°C for 1 min, 48–58°C for 2 min depending on the primers, and 72°C for 0.5 min with a final extension at 72°C for 10 min. The annealing temperature was determined by calculating the melting temperature for each primer using the formula given by Ausubel et al. (28). The PCR products were electrophoresed on a 1.5% agarose gel, stained with ethidium bromide (EtBr), and compared in the linear range of amplification using the -actin as a control.
|
Three genomic DNA clones of the Crlz1-IgJ locus were cloned by screening a phage mouse genomic DNA library (mouse 129/SvJ in FIX II; Stratagene), and subcloned into the pBluescript plasmid (Stratagene) for a further usage and sequencing. One of them was obtained by hybridization with the 165-bp BamHI-DraI fragment, which was located at 9.4 kb upstream of the IgJ transcription start site. This genomic clone covered the –21.4 to –5.4-kb region of DNA from the IgJ transcription start site. Another genomic clone was obtained by hybridization with the Crlz1 cDNA, and covered the +13 to –2-kb region of DNA from the Crlz1 transcription start site. The last genomic clone, which overlaps with both of the above genomic clones, thereby filling the missing genomic sequence between them, was screened out using the 497-bp XbaI-HindIII fragment positioned at about –1 kb from the Crlz1 transcription start site as a probe. The sequences of the three genomic clones were determined after subcloning their genomic DNA fragments from the phage to the pBluescript plasmid and compared with those in the National Center for Biotechnology Information (NCBI) databases.
Chromatin immunoprecipitation
Cells in a logarithmic phase of growth were treated with 1% formaldehyde for 10 min at room temperature to cross-link the DNA-protein and/or protein-protein interactions. The cross-linking reactions were stopped by adding glycine to a final concentration of 125 mM. After centrifugation, cells were washed two times with cold PBS containing 1 mM PMSF and resuspended in a lysis buffer (1% SDS, 10 mM EDTA, 50 mM Tris-HCl (pH 8.0), 1 mM PMSF) and sonicated on ice (amplitude 30%, five pulses with 9 s on and 0.5–2 min off, Ultrasonic VCX400; Sonics and Materials) to obtain an average DNA length of 700 bp, where the DNA was distributed between 400 bp and 1 kb. After microcentrifugation of the sonicated chromatin at 13,000 rpm, the supernatant was diluted 10-fold with a dilution buffer (1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-HCl (pH 8.0), 167 mM NaCl, 1 mM PMSF) and divided into aliquots each corresponding to 1 x 106 cells. An aliquot was precleared by an addition of 20 µl of 50% slurry of protein A-Sepharose (Amersham Biosciences) or protein G-agarose (Calbiochem) with 20 µg of BSA and 20 µg of sonicated salmon sperm DNA. A total of 4 µg of specific Abs was added to these precleared chromatin samples and rotated overnight at 4°C. The Ab-bound chromatin was captured by protein A-Sepharose or protein G-agarose beads (20 µl of 50% slurry) in the presence of 20 µg of BSA and 20 µg of sonicated salmon sperm DNA. The beads were spun down and washed sequentially with low salt buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl (pH 8.0), 150 mM NaCl), with high salt buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl (pH 8.0), 500 mM NaCl), with LiCl buffer (250 mM LiCl, 1% Nonidet P-40, 1% deoxycholate, 1 mM EDTA, 10 mM Tris-HCl (pH 8.0)), and finally with TE (10 mM Tris-HCl (pH 8.0), 1 mM EDTA). After the final wash, the chromatin was eluted from the bead by incubating it in 200 µl of an elution buffer (1% SDS, 100 mM NaHCO3) at room temperature for 15 min. The elution was repeated and combined with the first eluate to get a total volume of 400 µl. After adding 16 µl of 5 M NaCl, the eluted chromatin was incubated at 65°C for 4 h to reverse the formaldehyde cross-links. Each chromatin sample was further treated with proteinase K at 45°C for 1 h after mixing with 8 µl of 0.5 M EDTA, 16 µl of 1 M Tris-HCl (pH 6.8), and 20 µg of proteinase K. The chromatin sample was extracted with phenol-chloroform and the DNA in the aqueous supernatant was precipitated in the presence of 1 µg of yeast transfer RNA using 2 volumes of ethanol, washed with 70% ethanol, and dissolved in 100 µl of H2O per 1 x 106 cells. PCR analyses were performed using aliquots of the ChIP DNA samples (see below for the conditions and primers for PCR). Abs used in the ChIP experiments are anti-acetyl-histone H3 (rabbit polyclonal IgG, 06-599; Upstate Biotechnology) and anti-acetyl-histone H4 (rabbit antiserum, 06-866; Upstate Biotechnology). Nonspecific rabbit IgG Ab and preimmune serum that were used in the ChIP control experiments were obtained from Santa Cruz Biotechnology (sc-2027) and directly from rabbits, respectively.
PCR for the ChIP detection
The PCR programs for amplifying the various regions of the Crlz1-IgJ locus depended on the primers used (see Table I for the sequences of primers). Typically, they consisted of an initial denaturation step at 94°C for 5 min, and then, 32–36 amplification cycles of 1 min at 94°C, 1 min at 53–58°C depending on the primers, and 2 min at 72°C, and a final extension step at 72°C for 10 min. Again as in RT-PCR analyses above, the annealing temperature was determined by calculating the melting temperature for each primer using the formula given by Ausubel et al. (28). The amplified DNA fragments in a linear range of PCR were run on a 1.5% agarose gel and visualized by EtBr staining.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The HSS3/4 enhancer for the IgJ gene expression was mapped at 7.5 kb upstream of the HSS1 IgJ promoter (8). The authenticity of this enhancer was verified in various ways in our previous report (8). One of these verifications was that both the HSS3/4 enhancer and HSS1 promoter were opened simultaneously only in the terminally differentiated plasma cells (7, 8), which was well correlated with the expression pattern of the IgJ gene during B cell development (2). Interestingly, when we extended the DNase I-hypersensitivity assay from the stage of plasma cells to the stage of pre-B cells to correlate the chromatin structure on the HSS3/4 enhancer and the HSS1 promoter with the IgJ gene expression profile, we found that the HSS3/4 enhancer was already opened although the HSS1 IgJ promoter was still in a closed state in the PD31 and PD36 pre-B cell lines, which definitely do not yet express the IgJ gene (Fig. 1A) (7, 8). We hypothesized that a certain other gene in the vicinity of the HSS3/4 enhancer should be regulated and so expressed by this open enhancer in the pre-B cell stage, and thereby embarked to search for this potential pre-B cell-specific gene (7). If this hypothesis was correct, the HSS3/4 enhancer might play as an enhancer for a certain other neighboring gene in the pre-B cell stage, and then as an enhancer for the IgJ gene later in the plasma cell stage.
|
phage cDNA expression library. Alternatively, now that the mouse genomic sequences were publicized, we scrutinized the neighboring genes assigned in the vicinity of the HSS3/4 enhancer and thereby the IgJ gene. We focused on a newly assigned Crlz1 gene (9), and tested whether this gene is expressed in a pre-B cell-specific manner. Indeed as shown in our Northern blot analysis using several B cell lines representing various developmental stages (Fig. 1A), Crlz1 was expressed specifically only in PD36, which is the Abelson virus-transformed pre-B cell line (26), but not or very little in WEHI231 immature B, K46R mature B (data not shown), S194, and MOPC315 plasma cell lines. Crlz1 gene expression is tissue-limited as tested in several lymphoid and nonlymphoid tissues
Spurred by the Northern blot results of a pre-B cell-specific expression of the Crlz1 gene using the B cell lines of various developmental stages, we next investigated whether the results seen in the cell lines are also the case in normal mice. We first decided to determine the gross levels of Crlz1 mRNA transcripts in several lymphoid and nonlymphoid tissues in mice by an RT-PCR analysis. The RT-PCR analysis revealed that the Crlz1 mRNA was expressed in the lymphoid tissues such as bone marrow, spleen, and thymus, whereas it was not detected in the nonlymphoid tissue such as dorsal root ganglion (DRG) (Fig. 1B). These results have definitely demonstrated that the expression of Crlz1 gene is not ubiquitous, but tissue-limited in vivo. This nonexpression of Crlz1 gene in the peripheral neuronal DRG tissue was also intriguing as some neuronal tissues of brain were reported to express the Crlz1 gene by an in situ hybridization experiment (9).
Pre-B cell stage-specific expression of the Crlz1 gene was confirmed in the flow cytometer-sorted normal pre-B cells
Cells were sorted by the surface expression of B220 and pre-BCR using a flow cytometer from the cell populations of mouse lymphoid organs such as bone marrow, spleen, and thymus. From these sorted cells, two populations of cells were collected, one of which was B220+/pre-BCR+ and the other was B220+/pre-BCR–. Cells were also sorted similarly by the surface expression of B220 and IgM. Two cell populations were again collected from these sorted cells, one of which was B220+/IgM+ and the other was B220+/IgM–. Thereby, four normal cell populations from each lymphoid organ were obtained for testing the Crlz1 gene expression.
The flow cytometric diagrams (Fig. 1C) show that both bone marrow and spleen contain many B220+ B cells, but thymus does few. The diagrams also show that most of bone marrow B220+ cells are B220low while most of spleen B220+ cells are B220high, indicating that the bone marrow B cells are in an earlier stage than the spleen B cells (29). Among the B220+ B cells, IgM+ immature or mature B cells are found in a high proportion, but pre-BCR+ pre-B cells are found in a very small amount. These flow cytometric diagrams showing the profiles of B cell distribution in the lymphoid organs are well-matched to our general knowledge of B cell development and distribution, indicating that our flow cytometric sorting of B cells should be successful.
Total RNA was extracted from each cell population and subjected into an RT-PCR analysis. As shown in Fig. 1D, the Crlz1 expression was found in the pre-B cell population of B220+/pre-BCR+ phenotype from all the three lymphoid organs, but not in the non-pre-B cell population of B220+/pre-BCR– phenotype. Interestingly, the pre-B cell population of spleen showed the highest Crlz1 expression among them, indicating that the pre-B cells could be further typed in terms of the Crlz1 gene expression and their locations. This notion was further supported as the bone marrow pre-B cells expressed the pre-B cell marker genes such as VpreB, 5, and IL-7R much higher, and should be in an earlier stage than the spleen pre-B cells (Fig. 1E and also see the paragraph below).
Weak Crlz1 expression was also found in the cell population of B220+/IgM– phenotype from the bone marrow (Fig. 1D). However, Crlz1 expression was not detected in the cell populations of B220+/IgM– phenotype from the spleen or thymus. We guess that a portion of the cell population of B220+/IgM– phenotype from the bone marrow might be pre-B cells, but those from the spleen or thymus might not be the case.
To verify the authenticity of the sorted primary pre-B cells, we further checked those sorted cells for the expression of several pre-B cell marker genes such as VpreB, 5, and IL-7R by the RT-PCR analyses. These RT-PCR control experiments for checking the expression of pre-B cell marker genes have verified that the flow cytometer-sorted primary B220+/pre-BCR+ pre-B cells are indeed pre-B cells (Fig. 1E). Because the pre-B cells from bone marrow expressed those pre-B cell marker genes much higher than the ones from spleen, it appeared that the pre-B cells from bone marrow should be in an earlier stage than the ones from spleen (30).
The B220+/pre-BCR– and B220+/IgM– cells from bone marrow might include pro-B and pro-/pre-B cells, respectively, thereby also showing the expression of those marker genes somehow (Fig. 1E). It is well-known that pro-B cells already start to express VpreB, 5, and IL-7R genes (30, 31).
Despite some phenotypic variations among the pre-B cell populations from the different lymphoid organs, Crlz1 was definitely shown to be expressed specifically in the normal sorted pre-B cells (Fig. 1D) as well as in the pre-B cell lines (Fig. 1A). Thereby, the study of Crlz1 as a pre-B cell-specific gene could be continued with high confidence.
Chromatin openings over the Crlz1-IgJ locus are regulated coordinately with their gene expression profiles during B cell development
As the Crlz1 gene was expressed specifically in the pre-B cells, while its neighboring IgJ gene was expressed specifically in the plasma cells, chromatin openings over the Crlz1-IgJ locus were expected to be regulated coordinately with their stage-specific gene expression profiles. Initially, the pre-B cell-specific expression of the Crlz1 gene had been suspected because the HSS3/4 enhancer was shown to be opened despite the closed HSS1 IgJ promoter in the PD36 pre-B cells (7). Now that the Crlz1 gene was shown definitely to be expressed in the pre-B cells, some additional DNase I-HSS related to the pre-B cell-specific expression of the Crlz1 gene were searched over the Crlz1 gene locus.
Two strong adjacent (HSS9/10) and one weak (HSS8) HSS, which were numbered in series from the previous ones, were found at the Crlz1 promoter and its 1 kb upstream regions in the PD36 pre-B cells, respectively (Fig. 2, A and B). However, the HSS9/10 and HSS8 (Fig. 2B) were barely or not detected in the WEHI231 immature B, K46R mature B (data not shown), S194 plasma, and RAW264.7 macrophage cells. Two strong adjacent HSS9/10 at the Crlz1 promoter region and one weak HSS8 at its 1 kb upstream region were well-correlated with its pre-B cell-specific gene expression profile (Fig. 1, A and D). Surprisingly, a DNA fragment spanning the HSS9/10 turned out to have a very strong promoter activity, which was 
27% that of CMV in the transient transfection luciferase assays (Fig. 2C). The role of HSS8 remains yet to be found.
|
Histone acetylation status over the Crlz1-IgJ locus is regulated coordinately with its gene expression profiles during B cell development
It is well-known that the chromatin of the transcriptionally active gene locus is opened by a hyperacetylation of histones H3 and H4 (21, 32). The locus of two neighboring Crlz1 and IgJ genes on the mouse chromosome 5 is very interesting in that the Crlz1 gene is expressed pre-B cell stage-specifically, whereas its neighboring IgJ gene is expressed plasma cell stage-specifically during B cell development as described above. Therefore, histone acetylation should be regulated dynamically to allow the chromatin opening of the locus and to express those two neighboring genes stage-specifically during B cell development.
To validate these scenarios, the acetylation status of histones H3 and H4 was analyzed over the locus using the ChIP technique (33). The regions for checking the acetylation status of the locus were selected by the information of DNase I-HSS (HSS1–10) over the locus and their functional roles. The acetylation statuses on the enhancer (HSS3/4), IgJ promoter (HSS1) and Crlz1 promoter (HSS9/10) were examined especially to see their coordination with the chromatin openings and gene expression profiles during B cell development. ChIP controls such as nonspecific IgG Ab, preimmune serum, the erythrocyte-specific -globin gene, and the housekeeping -actin gene were included to verify our ChIP results.
In PD36 pre-B cells, the chromatin toward the Crlz1 gene was shown to be hyperacetylated on the histones H3 and H4, especially on H3, as compared with the one toward the IgJ gene (Fig. 3). Contrary to this acetylation status in PD36 pre-B cells, the opposite was shown to be the case in the terminally differentiated S194 plasma cells, i.e., the H3 and H4 on the chromatin toward the IgJ gene were shown to be hyperacetylated as compared with the ones on the chromatin toward the Crlz1 gene (Fig. 4). Because the same Abs were used for the ChIP experiments in both the pre-B and plasma cells, the more preferential ChIP detection of H3 or H4 hyperacetylation should not be due to a higher affinity of the anti-acetylated H3 Ab or anti-acetylated H4 Ab, respectively. Likewise, the preferential ChIP detection of H3 hyperacetylation on the -actin control should not be due to a higher affinity of the anti-aceylated-H3 Ab (Figs. 3–6). These histone acetylation patterns in the pre-B and plasma cells over the locus were well-coordinated with the stage-specific gene expression profiles of the two neighboring Crlz1 and IgJ genes during B cell development as mentioned above.
|
|
|
|
Finally, the coordination of plasma cell stage-specific expression of the IgJ gene with the histone acetylation status over the Crlz1-IgJ locus was further confirmed using the IL-2-inducible presecretor BCL1 cell line. BCL1 cell line has been well-characterized and used as a model system to study the terminal differentiation of B cells into plasma cells, and thereby the IgJ gene expression (2, 5, 8, 34) and Blimp-1 gene expression (6). When BCL1 cells were treated with IL-2 to induce the terminal B cell differentiation, the H3 and H4 were found to be hyperacetylated toward the chromatin of the IgJ gene, but not toward the one of the Crlz1 gene, again well-correlating with their gene expression profiles and chromatin openings during the terminal differentiation of B cells (Fig. 6). It is quite convincing to see that the histone acetylation pattern of the Crlz1-IgJ locus in the IL-2-treated BCL1 cells is similar to that in the S194 plasma cells (compare Figs. 4 and 6).
All these dynamic stage-specific changes of histone acetylation status over the Crlz1-IgJ locus during B cell development were well-correlated not only with its stage-specific HSS chromatin openings, but also with its stage-specific gene expression profiles of two neighboring genes; i.e., Crlz1 gene expression in the pre-B cell stage and then later IgJ gene expression in the terminally differentiated plasma cell stage (summarized in Table II).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
HSS3/4, which played as the IgJ enhancer in the plasma cells (8), were opened in the pre-B cells although the HSS1 IgJ promoter was still in a closed state as shown by the DNase I-hypersensitivity assay (7). The closed IgJ promoter was certainly consistent with the absence of IgJ gene expression in the pre-B stage (Fig. 1A) (1). Then, we wondered why the HSS3/4 was opened in the pre-B cells in the absence of opening of the HSS1 IgJ promoter, which was shown to be opened simultaneously with the HSS3/4 enhancer in the plasma cells (8). To answer this question, another target gene of the opened HSS3/4 enhancer was postulated to be expressed in the pre-B cell stage. Thereby, we started to search for a pre-B cell stage-specific gene in the vicinity of the HSS3/4 enhancer. The strategy for this search was to find any mRNA transcript by performing a Northern blot experiment using the genomic DNA fragments in the vicinity of HSS3/4 enhancer as the hybridization probes. Although we identified one pre-B cell-specific mRNA transcript in such a Northern blot (7), we could not isolate any cDNA clone for such a pre-B cell-specific transcript that originated in the vicinity of the HSS3/4 enhancer, because the genomic DNA probe used in the Northern blot turned out to contain many repetitive sequences, such as MT repetitive sequences (36), resulting in an appearance of an overwhelming number of different positive clones in the hybridization screening of a phage cDNA expression library.
In an alternative method of finding such a pre-B cell-specific gene in the vicinity of the HSS3/4 enhancer, we took advantage of the now publicized mouse genomic sequences. A quite recently cloned novel gene called Crlz1 (9) was found to be assigned at 15 kb upstream of the IgJ gene on the mouse chromosome 5 in the first publicized draft of mouse genomic sequences (Contig NT_039307.1), but then at 200 kb downstream of it in the second modified version of mouse genomic sequences (Contig NT_039308.2). Now, in the newest version (NT_109320.1) of the mouse genomic sequences, the Crlz1 gene was positioned at 30 kb upstream of the IgJ gene with the HSS3/4 enhancer between them. The newest version (Contig NT_109320.1) is still not a perfect one, as evidenced by an incomplete sequence assignment of the Crlz1 gene. However, this most recent version of mouse genomic sequences has almost matched our genomic sequences of three overlapping genomic clones over the Crlz1-IgJ locus that we isolated by a chromosome walking. Our genomic map over the locus shows that the Crlz1 and IgJ genes are 30 kb apart with the HSS3/4 enhancer (7.5 kb from the IgJ gene and 22.5 kb from the Crlz1 gene) between them, and that the two genes are transcribed oppositely (Fig. 2A). Furthermore, a human homolog of the mouse Crlz1 gene is also found in the upstream neighbor of the IgJ gene in the current version (NT_006216.14) of human genomic sequences with their transcriptions directed oppositely, indicating that the relative positioning of the two genes is evolutionarily conserved. Based on all the above results and facts, the Crlz1 gene appears to be correctly positioned in the upstream of the IgJ gene with their potentially shared HSS3/4 enhancer in between. These issues might be geared with a potential regulation of Crlz1 gene expression by the opened HSS3/4 enhancer in the pre-B cells, which remains to be verified with certainty. If this might be the case, the HSS3/4 enhancer should be called as the Crlz1 enhancer in the pre-B cell stage, and then the IgJ enhancer in the plasma cell stage. It is quite convincing to see that the HSS9/10 on the Crlz1 promoter region was shown to be simultaneously opened with the HSS3/4 enhancer in the PD36 pre-B cells (Table II and Fig. 2B). A potential molecular interaction between the HSS3/4 enhancer and the HSS9/10 Crlz1 promoter remains to be elucidated in the pre-B cells.
Crlz1 was found indeed as a pre-B cell stage-specific gene
We suspected that Crlz1 might be the pre-B cell-specific gene that we had eagerly sought in the vicinity of the HSS3/4 enhancer in the last several years. Indeed, as we suspected, the Crlz1 gene was found to be expressed specifically in pre-B cells. In a preliminary experiment, Crlz1 expression was found in all three lymphoid organs, but not in the neuronal DRG in our RT-PCR analyses (Fig. 1B). Our flow cytometric sorting analyses showed that the pre-B cell was found not only in the bone marrow but also in the spleen and thymus, despite some variations of their proportions (Fig. 1C), indicating that all three lymphoid organs might contain pre-B cells more or less. This observation was surprising because pre-B cells have generally been believed to be resident in the bone marrow until they become the immature B cells (31). However, the identity of the pre-B cells isolated from the three lymphoid organs by a flow cytometer was definitely verified by demonstrating their expression of several pre-B cell marker genes such as VpreB, 5, and IL-7R at the mRNA level (Fig. 1E). In these pre-B cell verification experiments, we also noticed that the pre-B cells of bone marrow expressed those marker genes higher than the ones of spleen, indicating that the pre-B cells of bone marrow might be in the earlier stage than the ones of spleen as judged by the higher expression levels of those marker genes in the bone marrow pre-B cells (30). Although we also verified the pre-B cells from thymus by noticing a very weak expression of the VpreB gene, the data were not shown as the expressions of 5 and IL-7R genes were very hard to detect.
Unexpectedly, the Crlz1 gene was found to be expressed higher in the pre-B cells of spleen than in the ones of bone marrow or thymus (Fig. 1D). Taking into account the expression profile of the Crlz1 gene together with those of marker genes above, it might be possible that the pre-B cells could be further typed in terms of their locations, their expression profiles of those marker and Crlz1 genes. As an example of flexible distribution of hemopoietic cells, we have noticed that even the hemopoietic stem cells of bone marrow could be found in the peripheral blood (37). Thereby, it is quite possible that many of the pre-B cells might have already migrated out to the peripheral lymphoid organs, but in this pre-B cell case, with a little differentiation in terms of the expression of those marker and Crlz1 genes.
It is known that pre-B cells proliferate initially and then rearrange their L chain genes (38, 39). Inferred from the pre-B cell-specific Crlz1 gene expression, we speculate a possibility that the expression of the Crlz1 gene might be involved in the regulation of cellular proliferation. A possibility of Crlz1 involvement in cellular proliferation could be even more enforced when it is reminded that CBF, a binding partner of Crlz1, is related to the acute myeloid leukemia (10, 12, 13). Furthermore, it is also quite suggestive that mature B, plasma, or DRG cells, where the Crlz1 gene expression is rarely or not found, are normally the cells of no proliferation. Sakuma et al. (9) reported that the Crlz1 gene expression was mainly detected in the neuronal cells of brain in their in situ hybridization experiments using a developing mouse embryo. It might be interesting to see whether those cells in brain might also proliferate during the development of an embryo.
We imagine that Crlz1 might exert its regulatory roles by playing as a transcription factor and/or cofactor because it was originally cloned by its ability to associate with the CBF transcription factor (9) and its yeast homolog of Sas10 was reported to derepress transcription (17). Furthermore, the HSS3/4 enhancer containing CBF-binding sites was found to be activated by an overexpression of Crlz1 (our unpublished data), possibly via its association with CBF. Taken together, all these reports and results might suggest a possibility that the expression of Crlz1 gene could be autoregulated via the nearby HSS3/4 enhancer, which contains the CBF-binding sites. A further characterization of pre-B cell-specific Crlz1 gene expression and the association of Crlz1 with CBF will certainly provide important information about the roles of Crlz1 in the pre-B cell stage during early B cell development.
Histone acetylation status over the Crlz1-IgJ locus
As stated in the results above, the H3 and H4 toward the Crlz1 gene, especially H3, as compared with those toward the IgJ gene, were shown to be hyperacetylated in the pre-B cells. Contrary to this acetylation status of the histones in the pre-B cells, the H3 and H4 toward the IgJ gene as compared with those toward the Crlz1 gene were shown to be hyperacetylated in the plasma cells. Furthermore, in accordance with the previous IL-2-induced simultaneous openings of the HSS3/4 enhancer and HSS1 promoter for the expression of IgJ gene in the presecretor BCL1 cells (8), the H3 and H4 toward the IgJ gene but not those toward the Crlz1 gene were shown to be hyperacetylated by IL-2 (Fig. 6). This IL-2-induced hyperacetylation toward the IgJ gene was reminiscent of the acetylation status seen in the IgJ-expressing S194 plasma cells (compare Fig. 4 with Fig. 6). At this time, it is hard to explain why both the H3 and H4 are hyperacetylated toward the IgJ gene in the IgJ-expressing plasma cells, whereas a more pronounced hyperacetylation of H3 than H4 is seen toward the Crlz1 gene in the Crlz1-expressing pre-B cells. However, one possible explanation might be that these phenomena could be related to the facts that the IgJ promoter has very weak activity, whereas the Crlz1 promoter has very strong activity (Fig. 2C).
We noticed a more generally decondensed chromatin state over the Crlz1-IgJ locus poised for the dynamic regulation of its regulatory HSS openings and gene expressions in the B cells than in the non-B cells, based on overall easier DNase I accessibility to the locus in the B cells (data not shown). Despite a more generally decondensed chromatin state of the locus in the B cells, we noticed that the Crlz1-IgJ locus is hypoacetylated in the B cells such as WEHI231 in much the same way as in the non-B cells such as EL4, RAW264.7, and NIH3T3 (Fig. 5), indicating that the poised decondensed chromatin over the locus in the B cells should be further regulated in terms of histone acetylation and HSS opening to express the two neighboring genes within the locus stage-specifically during B cell development. Thus, the regulation of gene expression in the locus could be envisaged as a multistep process consisting of a B cell-specific overall chromatin decondensation, a fine-tuning stage-specific histone acetylation and HSS opening, and a final transcription complex assembly leading to the stage-specific transcription of two neighboring Crlz1 and IgJ genes. The stage-specific gene expression regulation of the Crlz1-IgJ locus, which is shown to be coordinated by regulating its chromatin opening and histone acetylation (summarized in Table II), should provide a good model system for studying the gene expression regulation in the context of linear chromatin configuration, not only in the B cells, but also in the other eukaryotic cells.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 This work was supported by the Korea Research Foundation Grant funded by the Korean Government (MOEHRD, R05-2002-000739-0). This study was also supported by a grant of the Korea Health 21 R&D Project, Ministry of Health and Welfare, Republic of Korea (00-PJ1-PG3-21000-0016). 
2 J.-H.L. and S.-J.C. contributed equally to this work.
3 Address correspondence and reprint requests to Dr. Chang-Joong Kang, Graduate School of Biotechnology, Institute of Life Science and Resources, Kyung Hee University, Yongin, Gyeonggi-do 449-701, Korea. E-mail address: cjkang{at}khu.ac.kr
4 Abbreviations used in this paper: HSS, hypersensitive site; Crlz1, charged amino acid-rich leucine zipper 1; CBF, core-binding factor; ChIP, chromatin immunoprecipitation; EtBr, ethidium bromide; DRG, dorsal root ganglion.
Received for publication October 24, 2005. Accepted for publication July 27, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-binding protein (charged amino acid rich leucine zipper-1 (Crl-1)) in the mouse. Mech. Develop. 104: 151-154. [Medline]/CBFA1/AML3) gene in a Japanese patient with cleidocranial dysplasia. J. Bone Miner. Metab. 19: 263-266. [Medline], the heterodimeric partner of a novel Drosophila runt-related DNA binding protein PEBP2. Virology 194: 314-331. [Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME |
