| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Molecular Genetics, Biochemistry, and Microbiology, University of Cincinnati College of Medicine, Cincinnati, OH 45267; and
Department of Cell Biology, Albert Einstein College of Medicine, Bronx, NY 10461
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
RIIb gene. Gel-shift analysis suggests that Spi-C, ectopically expressed in pro-B cells, can bind PU.1 consensus sites in the IgH intronic enhancer and FcRIIb promoter. Transient transfection analysis demonstrated that PU.1 functions to repress the IgH intronic enhancer and activate the FcRIIb promoter, while Spi-C opposes these activities. WT-Spi-C pro-B cells have reduced levels of dimethylation on lysine 9 of histone H3 within the IgH 3' regulatory region, indicating that Spi-C can contribute to removal of repressive features in the IgH locus. Overall, these studies suggest that Spi-C may promote B cell differentiation by modulating the activity of PU.1-dependent genes. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
5, along with the signaling adaptor proteins, Ig
and Ig
, comprising the pre-BCR (2). Cells that express the pre-BCR on their surface are termed pre-B cells. Pre-B cells initiate rearrangement of the Ig L chain loci, either 
or , which results in surface expression of a functional BCR complex and progression to the immature B cell stage. Alternative splicing of the IgH transcript results in the expression of the IgD isoform of IgH protein on the surface, which, in addition to IgM expression, defines the mature B cell stage. Several transcription factors play important roles early in B cell development (reviewed in Ref. 3). In the early stages, these factors promote commitment to the B cell fate by activating B cell-specific genes and suppressing genes of other cell lineages. As the B cell develops, transcriptional regulation of the Ig loci becomes central because rearrangement and expression of the Ig loci are critical for stage progression. Each of the Ig loci contains regulatory enhancers that confer cell type-specific and developmental stage-specific expression. In particular, the intronic enhancer of the IgH locus contains multiple transcription factor-binding sites (reviewed in Ref. 4). Transcriptional activation of the IgH intronic enhancer promotes the rearrangement process and the expression of IgH protein (5). In addition to the intronic enhancer, the IgH locus contains a regulatory region 3' of the C region genes that is made up of multiple enhancers characterized by DNase I hypersensitive sites (HS; reviewed in Ref. 6). These enhancers, termed HS3a, HS1,2, HS3b, and HS4, gain hypersensitivity at various stages of B cell development (7, 8). Two of these enhancers, HS3b and HS4, are required for class switch recombination (9).
Various Ets-family transcription factors are expressed in developing B cells. The factors Ets-1, Elf-1, and PU.1 bind to various sites within the IgH intronic enhancer (10, 11, 12). The 3' regulatory region contains binding sites for both Elf-1 and PU.1 (13, 14). The Ets transcription factor PU.1 is predicted to be important in B cell development based on two observations. First, there are PU.1-binding sites in many B cell-specific genes (15, 16, 17, 18, 19), including all of the Ig loci, such as the IgH intronic enhancer and 3' regulatory regions, the Ig 3' enhancer, and the Ig 2–4 enhancer (20, 21, 22, 23). Second, PU.1 is required for the generation of lymphoid progenitors, and therefore no B cells are detected in PU.1–/– fetal livers (24, 25). However, conditional knockout studies indicate that PU.1 is not required for the continued survival and differentiation of B cells once they are generated (24, 26, 27). One of the consequences of PU.1 deletion in committed B cells is a switch from a B-2 to a B-1 phenotype (27). However, it is not yet clear what the underlying changes in gene expression are upon PU.1 deletion. PU.1–/– or PU.1–/–Spi-B–/– pro-B cells can be generated by retroviral transduction with the IL-7R cDNA and grown in IL-7 (28). We have previously shown that PU.1–/–Spi-B–/– pro-B cells have a phenotype similar to pre-B cells with increased CD25 and decreased c-Kit surface expression relative to wild-type (WT) pro-B cells. In addition, PU.1–/–Spi-B–/– pro-B cells express increased IgH sterile transcripts relative to WT pro-B cells (29). The PU.1–/–Spi-B–/– phenotype could be recapitulated by ectopic expression of a mutant form of PU.1 lacking an activation domain (29). Based on these observations, the inhibition of PU.1 appears to promote a pre-B cell-like phenotype, which is advanced over the pro-B cell stage.
Spi-C/Prf (PU.1-related factor) was identified simultaneously in two independent laboratories through a yeast-one hybrid method using the PU.1-binding sites of the Ig 3' enhancer and SP6 promoter sequences as bait (30, 31). Although the DNA-binding domain of Spi-C shares 59% aa identity to the DNA-binding domain of PU.1, the transactivation domains are divergent (30, 31). Spi-C is predicted to be a 28-kDa protein, which is smaller than the 31-kDa PU.1 protein (32). Northern blot analysis shows high levels of Spi-C in the spleen (30, 31). Specifically, Spi-C is expressed in macrophage cell lines, pre-B cell lines, and mature B cell lines, but is not found in pro-B cell lines or Ig+ immature B cell lines (30). This pattern of expression partially overlaps the other members of the Spi family, PU.1 and Spi-B, which are expressed throughout B cell development (33, 34, 35). Gel mobility shift analysis shows that Spi-C can bind to PU.1-binding sites in the Ig locus (32). However, Spi-C cannot form a ternary complex with IFN regulatory factor (IRF)-4 on DNA (32). This is in contrast to PU.1, which cooperates with IRF-4 to efficiently bind Ets-IRF composite elements (36).
In this study, we investigate a biological function for Spi-C. We show that, in contrast to PU.1, Spi-C is inefficient at restoring B cell development in PU.1–/– fetal liver progenitors. Ectopic expression of Spi-C in WT fetal liver-derived pro-B cells (WT-Spi-C) induces an advanced pre-B cell-like phenotype characterized by increased expression of CD25 and reduced expression of c-Kit. In addition, sterile transcription from the IgH locus is up-regulated in WT-Spi-C pro-B cells relative to WT pro-B cells infected with a control GFP retrovirus (WT-R1). Similar to pro-B cells lacking PU.1, WT-Spi-C pro-B cells have increased IgH intronic enhancer sterile transcription (Iµ) and reduced transcription of the FcRIIb gene. Using gel-shift analysis, we observed Spi-C binding to PU.1 consensus sites within the IgH intronic enhancer and FcRIIb promoter. Transient transfections into 38B9 pro-B cells indicate that PU.1 represses IgH intronic enhancer activity while activating FcRIIb promoter activity. The effect of Spi-C was opposite to that of PU.1, because Spi-C partially activates the IgH intronic enhancer and represses the FcRIIb promoter. Finally, WT-Spi-C pro-B cells have globally reduced methylation on lysine 9 of histone H3 within the IgH 3' regulatory region. These observations suggest that Spi-C has the potential to modulate PU.1 activity to promote B cell differentiation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
Spi-C was amplified from cDNA generated from spleen total RNA using gene-specific primers with NotI and ApaI restriction sites located after the ATG start codon (5'-CGGCGGCCGCACTTGTTGTATTGATCAAGAC-3') and the termination codon (5'-AGGGGCCCGGTCAGCTCTGGTAACTGCCGTGA-3'), respectively. The PCR product was cloned using the pCR-TOPO2.1 cloning kit (Invitrogen Life Technologies). Spi-C cDNA was cloned into the hemagglutinin (HA) tag-containing pcDNA3 vector described previously (36) using the NotI and ApaI restriction sites. MIG-Spi-C retrovirus containing HA-Spi-C cDNA, murine stem cell virus, an internal ribosomal entry site (IRES), and GFP (IRES-GFP) was generated by ligating the HA-Spi-C fragment into the MIGR1 vector (37).
The promoter of the FcRIIb gene was PCR amplified from pro-B cell genomic DNA using a forward primer located 
2 kb upstream of the promoter (5'-GTCTGAGGATATCTTCCTTCTCTCACGCTGT-3') and a reverse primer located just downstream of the translation start site (5'-AGTAGCATATGGCACAAAGTCCGTGAGAAC-3'). The PCR product was cloned using the pCR-TOPO2.1 cloning kit (Invitrogen Life Technologies). A second PCR amplification using nested primers containing the NheI (5'-GCGCTAGCGAGGATATCTTCCTTCTCTCACGCTG-3') and HindIII (5'-ATAAGCTTCAGCTCTGGAGCGAGTCAGCTGCAGG-3') restriction sites was performed on the originally cloned fragment in the TOPO2.1 vector. The 1817-bp promoter fragment was cloned into the pGL3 luciferase reporter vector (Promega) using the HindIII restriction sites located in the amplified fragment and TOPO2.1 vector. A KpnI restriction site within the promoter fragment was used, along with the KpnI restriction site in the multiple cloning site of pGL3, to remove a 1398-bp fragment. The pGL3 vector with a remaining 419-bp fragment was religated to generate a minimal promoter vector.
The pEµSR vector containing the IgH intronic enhancer (Eµ) driving a modified SV40 promoter (SR) has been described previously (38). The luciferase gene was cloned from a HindIII and XbaI fragment of the pGL3 luciferase reporter vector (Promega) and inserted into the pEµSR vector using the HindIII and EcoRV restriction sites downstream of the Sr promoter. The pEµSR-Luciferase vector was mutated using the Quickchange XL site-directed mutagenesis kit (Stratagene), with a GGAA
GGAC mutation at the PU.1-binding site.
Retroviral transduction of PU.1–/–Lin– fetal liver progenitors and pro-B cells
Cell-free viral supernatants encoding PU.1, Spi-B, or Spi-C were made by transfection of the Plat-E packaging cell line (39) with MIGR1-derived vectors. Briefly, Plat-E cells were plated at 106 cells/6-well plate 24 h before transfection. The cells were transfected with 4 µg of plasmid DNA and 12 µl of Lipofectamine 2000 (Invitrogen Life Technologies) overnight. The transfection medium was then replaced by 1 ml of fresh medium. Virus was harvested 36 h after fresh medium was placed on the cells by centrifugation at 2500 rpm for 10 min and the supernatant was frozen at –80°C. PU.1–/– lineage-negative (Lin–) fetal liver progenitors (FLPs) were isolated as previously described (28). A spin-infection procedure was used for the retroviral transduction of PU.1–/–Lin– FLPs. One milliliter of cell-free retroviral supernatant was rapidly thawed at 37°C and placed on 105 PU.1–/–Lin– FLPs/24-well plate with 8 µg of polybrene, followed by centrifugation at 2000 x g at 30°C for 2 h. The polybrene and virus were removed by one wash with complete IMDM, and the cells were cultured in complete IMDM containing IL-3, IL-6, and SCF for 48 h to promote retroviral integration. After 48 h of culture, the cells were extensively washed in complete IMDM, and subjected to flow cytometric analysis to assess transduction efficiency by GFP expression.
Generation of pro-B cell lines and colony assays
WT C57BL/6 mice were mated overnight and the presence of a vaginal plug verified on the morning after matings was taken as 0.5 days postcoitum. Lin– fetal liver progenitors were isolated and plated onto a monolayer of irradiated S17 stromal cells. Cultures were fed with fresh IL-7-containing complete IMDM medium every 4 days and analyzed by flow cytometry to confirm expression of CD19.
Cells retrovirally transduced as described above were placed in methylcellulose containing IL-7 (Stem Cell Technologies) and plated on a monolayer of irradiated S17 stromal cells to assay for B cell colonies. To assay for macrophage colonies, cells were placed in methylcellulose containing M-CSF (PeproTech) at 10 ng/ml. Colony counts were scored after 10 days of proliferation. Cells retrovirally transduced as described above were also placed in liquid culture of complete IMDM containing IL-7 (Stem Cell Technologies) and plated on a monolayer of irradiated S17 stromal cells to assay for B cell/macrophage growth.
Flow cytometric analysis and Western blot
Pro-B cells (0.5 x 106) were washed three times in flow cytometry wash buffer. Following the third wash, cells were suspended in an appropriate dilution of Ab and stained for 20 min on ice. Ab used included (BD Pharmingen): CD19 (ID3), CD11b (M1/70), B220 (RA3-6B2), CD43 (S7), Fc II/III receptor (2.4G2), c-Kit (2B8), CD25 (7D4), IgM (II/41). Cells were pelleted by centrifugation and washed three times before staining with an appropriate dilution of secondary for 20 min on ice. Cells were pelleted by centrifugation, washed three times, and finally diluted into 748 mM propidium iodide solution. Cells were analyzed using a BD Biosciences FACSCalibur system. The final gating procedure involved either staining from PE or allophycocyanin based on the presence of endogenous GFP expression. Western blot was performed using standard methods with rabbit anti-PU.1 peptide Ab (Santa Cruz Biotechnology), goat anti-actin peptide Ab (Santa Cruz Biotechnology), mouse anti-HA peptide Ab (12CA5; Roche), and HRP-conjugated anti-rabbit or anti-goat secondary Abs (Pierce).
PCR analysis
Total RNA was extracted from cultured IL-7-dependent pro-B cell lines using RNA-Bee (Tel-test). Total RNA was used as template for cDNA synthesis using a cDNA synthesis kit (BD Biosciences) according to manufacturer’s protocol. cDNA was analyzed using primers spanning intronic sequence for -actin to verify lack of genomic DNA contamination and equal cDNA amplification from various template RNAs. Primers for target sequences were designed to span introns or splice junctions using Stratagene PCR primer designer. PCR was performed on a Robocycler (Stratagene) using the following reaction cycle. This includes a 5-min melting step at 95°C followed by 25–30 cycles of amplification at the optimum annealing temperature, followed by an extension step of 2 min at 72°C. PCR products were run on agarose gels and visualized by ethidium bromide staining. The following primers were used: Iµ: 5'-ACCTGGGAATGTATGGTTGTGGCTT-3' and 5'-ATGCAGATCTCTGTTTTTGCCTCC-3' (40), FcRIIb: 5'-CCCAAGTCCAGCAGGTCTTTACC-3' and 5'-CCCAATGCCAAGGGAGACTAAAT-3', -actin: 5'-TCCTTCGTTGCCGGTCCACA-3' and 5'-CGTCTCCGGAGTCCATCACA-3', CD25: 5'-AGAACGGCACCATCCTAAACT-3' and 5'-TGCATGTCTGTTGTGGTTTGT-3', c-Kit: 5'-CAACAGCAATGGCCTCACGAGT-3' and 5'-GTGGTACACCTTTGCTCTGCTC-3'. Relative quantitation real-time RT-PCR was performed as previously described (29). The primers used for Iµ were the same as for standard RT-PCR above. Primers for FcRIIb: 5'-AATTGTGGCTGCTGTCACTG-3' and 5'-TTCCCTGTGATCAGGGTTTC-3', PU.1: 5'-CGGATGTGCTTCCCTTATCAAAC-3' and 5'-TGACTTTCTTCACCTCGCCTGTC-3'.
Gel mobility shift assays
PU.1 protein was generated by in vitro translation using the TNT-coupled transcription-translation kit (Promega) with 1 µg of pcDNA3-HA-PU.1 vector. Nuclear lysates were obtained from 1 x 108 cultured cells. Cells were washed in Dulbecco’s PBS following centrifugation. Washed cells were suspended in buffer (10 mM HEPES (pH 7.9), 1.5 mM MgCl2, 10 mM KCl, 0.5 mM DTT) and incubated on ice for 10 min. Nuclei were obtained by lysing membranes with a glass homogenizer on ice and centrifugation. Nuclei were resuspended in buffer (20 mM HEPES (pH 7.9), 25% glycerol, 0.42M NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, 0.5 mM PMSF, 0.5 mM DTT) and lysed with a glass homogenizer on ice. Lysed nuclear extract was stirred for 30 min on ice, separated from nuclear membrane by centrifugation, and dialyzed for 5 h at 4°C using the Slide-A-Lyzer Mini Dialysis Unit (Pierce). Synthetic complementary oligonucleotides were annealed and end-labeled with [-32P]dCTP using the Klenow fragment. DNA-binding reactions were performed for 30 min at 25°C and contained 2.5 x 105 cpm of probe, 2 µl of in vitro-translated PU.1 or 10 µg of nuclear lysate, and 100 ng of poly(dI-dC) in a final volume of 30 µl. Supershifting of binding reactions was performed using 1 µl of anti-HA Ab (Covance), anti-PU.1 Ab (Santa Cruz Biotechnology), or control rabbit serum. Abs were added to nuclear lysates and incubated for 20 min on ice before addition of radiolabeled probe and binding reactions. Binding reactions were resolved on a 5% native polyacrylamide gel using 1x Tris-glycine-EDTA buffer, at 150 V for 2.5 h. Double-stranded oligonucleotide probes contained the PU.1-binding site in the: IgH intronic enhancer (µB): 5'-TCGACAAGGCTATTTGGGGAAG-3', or a mutated PU.1-binding site (mutant µB), 5'-TCGACAAGGCTATTTGGGGACG-3'. FcRIIb promoter, 5'-TCGATTCTTTTCACTTCCCCAT-3', or a mutated PU.1-binding site (mutant FcRIIb), 5'-TCGATTCTTTTCACGTCCCCAT-3'. An E2A consensus site oligonucleotide was used as control, 5'-GATCTACATCTGTGCAGTTCGCTG-3'. All oligonucleotides contained GATC overhangs to facilitate radiolabeling. Italicized letters represent the Ets binding consensus sequence.
Transfection and reporter assays
A total of 2 x 106 38B9 cells were plated 24 h before transfection. Cells were washed three times in serum-free medium and transfected with 15 µg of total DNA containing 5 µg of reporter, 200 ng of Renilla luciferase vector (Promega), 5 µg of pcDNA3-HA expression vector of PU.1 or Spi-C, and empty pcDNA3 vector. Transfections were performed by electroporation using the Gene Pulser II system (Bio-Rad). Aliquots of cells in 400 µl were incubated with DNA at room temperature for 10 min and then transferred to cuvettes and electroporated at 950 µF and 220 V. Electroporated cells were incubated at room temperature for 10 min before returning them to standard serum-containing medium. Cells were lysed 24 h after transfection and analyzed for both firefly and Renilla luciferase activities using a dual luciferase assay kit (Promega) and an analytical Monolight tube luminometer. Transfections were done in duplicate at least four independent times.
Chromatin immunoprecipitation
Chromatin immunoprecipitation was performed as described previously (41). Briefly, 108 cells were grown as described above and collected by centrifugation, resuspended in IL-7-containing IMDM, and subjected to cross-linking with formaldehyde for 10 min at 37°C. The cross-linking reaction was quenched, cells were collected by centrifugation, washed three times with cold PBS, frozen in liquid nitrogen, and stored at –80°C before chromatin preparation and sonication. Immunoprecipitations were done overnight at 4°C using 4 µg of one of the following Abs: normal rabbit IgG (Upstate Biotechnology catalog no. 12-370), anti-acetylated histone H3 recognizing the K9 and K14 diacetylated tail (Upstate Biotechnology catalog no. 06-599), anti-acetylated histone H4 recognizing a K3, K6, K10, and K14 tetra-acetylated tail (Upstate Biotechnology catalog no. 06-866), anti-dimethyl K9 of histone H3 (Upstate Biotechnology catalog no. 07-212). Real-time PCR for regions of the 3' regulatory region, IgH intronic enhancer, and IL-7R was performed as described previously (41). Primer pairs used to amplify these regions were described previously (41).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
PU.1–/– mice fail to produce B cells and macrophages in vivo, and cultured PU.1-deficient progenitors also fail to do so. However, if PU.1 is reintroduced by retroviral transduction, B cell and macrophage development can be rescued (42). This system provides an assay for transcription factor redundancy in vivo. To determine whether the other two Ets subfamily members, Spi-B and Spi-C, could rescue development as efficiently as PU.1, we retrovirally transduced PU.1–/– fetal liver progenitors with GFP-expressing retroviruses that encoded PU.1, Spi-B, or Spi-C (Fig. 1A). Equivalent numbers of infected fetal liver cells were placed in colony-forming assays of methylcellulose containing M-CSF to support the generation of macrophage colonies, or methylcellulose containing IL-7 and stromal cells to support the generation of pro-B cell colonies (Fig. 1B). Colonies were scored based on morphology using microscopy as performed previously (28). As shown in Table I, infection of progenitors with a PU.1 retrovirus generated a high number of macrophage colonies in M-CSF, as well as significant numbers of pro-B cell colonies in IL-7 and stromal cells. Spi-B was less efficient than PU.1 at inducing macrophage and pro-B cell colonies with 38% of macrophage rescue and 56% of B cell rescue, respectively. Infection with Spi-C cDNA was extremely inefficient at rescuing either macrophage or pro-B cell colony formation, producing only 1% of the macrophage colonies and 3% of the pro-B cell colonies relative to PU.1-rescued colony numbers (Table I). In parallel experiments, infected cells were also placed in liquid cultures containing S17 stromal cells and IL-7. Under these culture conditions, CD19+ pro-B cells can proliferate continuously whereas CD11b+ macrophages terminally differentiate and stop proliferation after the first 5–7 days. After 10–11 days, these cultures were analyzed by flow cytometry for CD19 and CD11b (Fig. 1C). Both PU.1 and Spi-B were able to promote macrophage growth in culture, with 22 and 27% of the cells positive for CD11b, respectively (Fig. 1C, bottom panels). However, Spi-C inefficiently rescued macrophage growth with only 4% positive for CD11b (Fig. 1C, bottom panels). In contrast, the relative frequency of pro-B cell growth in Spi-C-rescued progenitors was nearly twice (66%) that of PU.1 (37%) and Spi-B (30%) (Fig. 1C, top panels). In summary, we found from the colony rescue experiments that Spi-C is inefficient at rescuing progenitor development into B cells and macrophages relative to PU.1 and Spi-B. However, once they are generated, Spi-C-rescued pro-B cells proliferate well in response to IL-7 in liquid culture.
|
|
RT-PCR analysis of IL-7-dependent pro-B cells demonstrates minimal to no expression of Spi-C. However, Spi-C is detected in pre-B cell and mature B cell lines (28, 30, 31). When IL-7-dependent pro-B cells are cultured in the absence of IL-7 for 3 days, they differentiate and rearrange the Ig L chain genes (1, 43). When we examined Spi-C transcripts by real-time RT-PCR, we found that IL-7 withdrawal induced Spi-C expression 2-fold compared with pro-B cells maintained in IL-7 (data not shown). These observations suggest that Spi-C might be involved in B cell differentiation.
To determine whether ectopic expression of Spi-C can induce marks of differentiation in pro-B cells, we infected WT fetal-liver derived pro-B cells with a GFP-containing retrovirus expressing the Spi-C cDNA (Fig. 2A). After sorting, flow cytometric analysis confirmed that these cells maintained GFP expression indefinitely, indicating that the control virus and Spi-C are not detrimental to cell growth (Fig. 2A). These cell lines are hereafter referred to as WT-R1 and WT-Spi-C. Real-time RT-PCR analysis confirmed increased expression of Spi-C in the WT-Spi-C pro-B cells relative to WT-R1 pro-B cells (Fig. 2B). When Western blot analysis was performed against the HA tag on the retroviral Spi-C, we detected a protein at the predicted molecular mass for Spi-C only in nuclear lysates obtained from WT-Spi-C pro-B cells (Fig. 2C). Interestingly, analysis by real-time RT-PCR indicated increased transcript expression of PU.1 in WT-Spi-C pro-B cells relative to WT-R1 pro-B cells (Fig. 2D). However, Western blot analysis indicated equal PU.1 protein expression in WT-R1 and WT-Spi-C pro-B cells (Fig. 2E). Also of note, the polyclonal PU.1 Ab also detected a faster migrating band in the WT-Spi-C nuclear lysate that corresponds to the protein detected by the HA Ab (Fig. 2E). Based on the high homology between the PU.1 and Spi-C DNA-binding domains, this result suggests that the polyclonal Ab is cross-reactive with Spi-C.
|
R expression was severely reduced by Spi-C ectopic expression (MF = 81.6 ± 9 and 167.3 ± 7, respectively) compared with WT-R1 pro-B cells (MF = 164.3 ± 10 and 281.5 ± 8, respectively, Fig. 3D). Overall, WT-Spi-C pro-B cells exhibit an intermediate developmental phenotype between pro-B and pre-B based on the increase in CD25 and decrease in c-Kit on the surface and at the transcript level. In addition, the B cell markers B220 and FcR are severely reduced in WT-Spi-C pro-B cells.
|
RIIb transcripts in pro-B cells caused by ectopic expression of Spi-C
To further characterize the phenotype of WT-Spi-C pro-B cells, we examined gene expression by RT-PCR. Similar to the phenotype of PU.1–/–Spi-B–/– pro-B cells (29), we found that steady-state levels of sterile transcripts initiating from the IgH intronic enhancer (Iµ) were increased relative to WT-R1 pro-B cells (Fig. 4A). Increased sterile transcripts are indicative of an active IgH locus, which is observed in total spleen containing mature B cells (Fig. 4A). Quantitation of this induction by real-time PCR showed that Iµ transcripts were 4.5-fold higher in WT-Spi-C pro-B cells relative to WT-R1 pro-B cells (Fig. 4B). We also examined FcRIIb transcripts, because we had previously observed a reduction in these transcripts in PU.1–/–Spi-B–/– pro-B cells (29) and a corresponding reduction in FcRII/III surface expression (Fig. 3C). The FcRIIb gene encodes an inhibitory FcR, which is expressed on the surface of all stages of B cell development and mature B cells. The FcRIIb receptor can block proliferation in B cells through the inactivation of MAPKs. In addition, mice deficient for FcRIIb are susceptible to induced autoimmune disease (reviewed in Ref. 45). WT-Spi-C pro-B cells have a severe reduction in the levels of FcRIIb transcripts relative to WT-R1 pro-B cells (Fig. 4C). Quantitatively, this reduction was 6-fold relative to WT-R1 pro-B cells as determined by real-time PCR (Fig. 4D). In summary, Iµ sterile transcription was significantly increased in WT-Spi-C pro-B cells, but expression of the FcRIIb gene was impaired by the ectopic expression of Spi-C. These changes in Iµ and FcRIIb expression are similar to those detected in PU.1–/–Spi-B–/– pro-B cells (29).
|
RIIb promoter
Previous studies showed that rSpi-C bound to PU.1 consensus sites, such as those in the Ig locus in vitro (32). Because we observed altered patterns of expression from the IgH intronic enhancer and the FcRIIb promoter in WT-Spi-C pro-B cells, we set out to examine whether Spi-C was capable of binding PU.1 consensus sites within these two regulatory elements. Using an oligonucleotide containing a PU.1 consensus site previously shown to bind PU.1 in the IgH intronic enhancer core (µB) (11), we examined the binding potential of nuclear lysates from WT-R1 pro-B cells and WT-Spi-C pro-B cells. An oligonucleotide containing a consensus E2A-binding site, used as a loading control, demonstrated that the nuclear lysates contained equal amounts of E2A protein (Fig. 5A, lanes 16–18, B, lanes 12–14). In vitro-translated PU.1 produced a unique complex that was not present in reactions with a µB probe mutated at the Ets site (Fig. 5A, lanes 4 and 5). In vitro translation reactions charged with only the pcDNA3 vector generated nonspecific complexes present on both the WT and mutated µB probes (Fig. 5A, lanes 2 and 3). When compared with in vitro-translated PU.1 protein, nuclear lysates from the WT-R1 pro-B cells produced a similar complex that could be disrupted by Ab specific to PU.1 (Fig. 5A, lanes 4, 6, and 10). When nuclear lysates from the WT-Spi-C pro-B cells were incubated with the µB probe, additional faster migrating complexes appeared (Fig. 5A, lane 8). The unique complexes were, along with the complex corresponding to PU.1, absent in reactions with a µB probe mutated at the Ets site (Fig. 5A, lanes 5, 7, and 9). Ab specific to PU.1 disrupted the complex containing PU.1 from nuclear lysates of WT-Spi-C pro-B cells, but not the unique complexes (Fig. 5A, lane 11). By incubating with an Ab specific to the HA tag on the retrovirally expressed Spi-C, the unique complexes could be disrupted, and a slower migrating complex appeared (Fig. 5A, lane 13). In addition, the band corresponding to PU.1 increased in intensity (Fig. 5A, compare lane 8 to lane 13). Control antisera did not disrupt the Spi-C or the PU.1 complexes (Fig. 5A, lanes 14 and 15). This indicates that the µB-binding complex generated from nuclear lysates of WT-Spi-C pro-B cells contained functional Spi-C protein. In addition, the amount of PU.1 complex was reduced relative to the WT-R1 nuclear lysates (Fig. 5A, compare lane 6 to lane 8).
|
RIIb promoter. The FcRIIb promoter had not been previously characterized, so sequence alignments were performed using TRAFAC software (46). A region upstream of the translation start site, highly conserved between mouse and human, contains a consensus Ets-binding site (M. B. Kamath and R. P. DeKoter, unpublished results). Nuclear lysates were prepared from WT-R1 pro-B cells or WT-Spi-C pro-B cells. Using gel mobility shift analysis, nuclear lysates were incubated with radiolabeled oligonucleotide probes containing a WT or a point mutant in the Ets site were identified within the FcRIIb promoter (Fig. 5B). In vitro-translated PU.1 produced a unique complex that was not present in reactions with a µB probe mutated at the Ets site (Fig. 5B, lanes 3 and 9). In vitro translation reactions charged with only the pcDNA3 vector generated nonspecific complexes present on both the WT and mutated µB probes (Fig. 5B, lanes 2 and 8). When compared with in vitro-translated PU.1 protein, nuclear lysates from the WT-R1 pro-B cells produced a similar complex that could be disrupted by Ab specific to PU.1 (Fig. 5B, lanes 3 and 4, and data not shown). When nuclear lysates from WT-Spi-C pro-B cells were incubated with the FcRIIb promoter probe, additional faster-migrating complexes appeared (Fig. 5B, lane 5). The unique complexes, as well as the complex corresponding to PU.1, were absent in reactions with an FcRIIb promoter probe mutated at the Ets site (Fig. 5B, lanes 9–11). By incubating with an Ab to the HA tag on the retrovirally expressed Spi-C, the unique complexes could be disrupted, and a slower migrating complex appeared (Fig. 5B, lane 5). This indicated that the complexes generated from nuclear lysates of WT-Spi-C pro-B cells contained functional Spi-C. Control antisera did not disrupt the Spi-C or the PU.1 complexes (Fig. 5B, lane 7). In addition, the amount of PU.1 complex was reduced relative to the WT-R1 pro-B nuclear lysates (Fig. 5B, compare lane 5 to lane 4). However, Western blotting revealed that PU.1 protein levels were equal in WT-R1 and WT-Spi-C pro-B cells (Fig. 2C). In summary, Spi-C ectopically expressed in pro-B cells can bind to PU.1 consensus sites within the IgH intronic enhancer and FcRIIb promoter.
PU.1 and Spi-C exert opposite activities on the Eµ enhancer and FcRIIb promoter
Based on the similar effects of PU.1 deficiency and ectopic Spi-C expression in pro-B cells on Iµ and FcRIIb transcription, we investigated the activity of PU.1 and Spi-C on the IgH intronic enhancer and the FcRIIb promoter using transient transfection assays with luciferase reporter vectors driven by these regulatory elements. Using the 38B9 pro-B cell line, which expresses endogenous PU.1 but not Spi-C, we transiently transfected these reporters together with plasmids encoding PU.1, Spi-C, or both. When PU.1 was cotransfected with the IgH intronic enhancer reporter, we saw a 2.3-fold reduction in activity relative to the enhancer activity assayed in the presence of endogenous levels of PU.1 (Fig. 6A). This difference in activity was no longer detected when we analyzed a mutant form of the PU.1 (R232, 235A; Fig. 6A), which is unable to bind DNA (47). When the PU.1-binding site in the IgH intronic enhancer was mutated, activity of the reporter did not significantly change (Fig. 6B). This result suggests that endogenous PU.1 expression in 38B9 pro-B cells is not required for activation of this enhancer. Results were different with Spi-C. When Spi-C was cotransfected with the intronic enhancer, activity increased (Fig. 6A), an outcome opposite to that observed with PU.1. When both PU.1 and Spi-C were cotransfected with the IgH intronic enhancer, the reporter activity was reduced relative to the reporter transfected alone, but not reduced to the same degree as with PU.1 alone (Fig. 6A).
|
RIIb expression, the FcRIIb promoter was activated 1.7-fold by PU.1 when cotransfected into 38B9 pro-B cells (Fig. 6C). However, Spi-C did not significantly affect the activity of the FcRIIb promoter (Fig. 6C). In summary, PU.1 repressed IgH intronic enhancer activity and activated FcRIIb promoter activity, whereas Spi-C enhanced IgH intronic enhancer activity, but did not alter FcRIIb promoter activity in transient transfection assays. Coexpression of PU.1 and Spi-C with the IgH intronic enhancer reporter produced an intermediate activity between PU.1 and Spi-C expression individually. Ectopic Spi-C expression removes marks of inactive chromatin in the IgH 3' regulatory region
Because we observed that IgH sterile transcription was increased by ectopic expression of Spi-C, we further characterized the IgH locus by chromatin immunoprecipitation assays on histone modifications. The IgH intronic enhancer is typically in an active state in pro-B cells, and the 3' regulatory region appears to be active at various stages of B cell development. For example, the HS4 enhancer is active throughout development (48, 49) and a recently identified segment downstream of HS4 that contains additional DNase I hypersensitive sites is associated with marks of active chromatin beginning in pro-B cells (41). The 3' regulatory region of the IgH locus plays an important role in the accessibility of IgH gene expression in differentiated B cells and has been implicated in sterile IgH µ0 transcription in pro-B cells (50, 51, 52, 53, 54). Sterile transcription within the IgH locus is an indicator of active chromatin. Posttranslational modification of histone N-terminal tails by acetylation provides access to chromatin remodeling factors that open regions to transcriptional machinery (55). Therefore, we analyzed this region by performing chromatin immunoprecipitation for the acetylated histones H3 and H4. The active IL-7R gene locus was used as a positive control for acetylation in our analysis. The 3' regulatory region was associated with both acetylated histone H3 and H4 in WT-R1 pro-B cells, and no significant changes in either H3 or H4 histone acetylation profiles were detected in WT-Spi-C pro-B cells (Fig. 7A). Instead, Spi-C ectopic expression resulted in a significant reduction in dimethylation of lysine 9 (di-Me K9 H3) of histone H3 (Fig. 7B, right panel), a modification associated with repression because it prevents histone acetylation and recruits chromatin silencing factors (reviewed in Ref. 56). Although the entire 3' regulatory region was associated with di-Me K9 H3 in WT-R1 pro-B cells, this mark was essentially absent in WT-Spi-C pro-B cells (Fig. 7B). In summary, these experiments show that ectopic Spi-C expression changed the status of the IgH 3' regulatory region from a repressive to a permissive chromatin environment.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
without a requirement for either PU.1 or Spi-B (25). Spi-B is also less efficient than PU.1 at rescuing both B cells and macrophages. This is in contrast to previous observations that suggest that Spi-B can substitute for PU.1 for myeloid development but not for lymphoid development (57). Our data provides the first quantitative comparison of transactivation potential between PU.1, Spi-B, and Spi-C in a biological setting. These results suggest that PU.1 possesses the strongest transactivation potential among this subfamily of Ets transcription factors. When PU.1 is ectopically expressed in pro-B cells, their proliferation is rapidly arrested (Ref. 42 and our unpublished observations). Successful high-level ectopic expression of Spi-C in pro-B cells indicates that its biological activity differs from PU.1. It is important to note that WT IL-7-dependent pro-B cells express PU.1 whereas Spi-C expression is minimal to undetectable (28). The ectopic expression of Spi-C in WT pro-B cell lines (WT-Spi-C pro-B cells) results in several phenotypic changes including increased CD25 expression as well as reduction or loss of c-Kit and B220 expression. Acquisition of CD25 and loss of c-Kit represents markers for the pro-B to pre-B transition. Therefore, WT-Spi-C pro-B cells have a phenotype intermediate between pro-B cells and pre-B cells. This is similar to our observation that PU.1–/–Spi-B–/– pro-B cells and pro-B cells ectopically expressing a mutant form of PU.1 display an advanced differentiation phenotype (29). Recent data has shown that deletion of PU.1 in B cells can induce a B-2 to B-1 cell switch, accompanied by loss of B220 expression (27). In addition, deletion of a regulatory element upstream of the PU.1 gene eliminates B-2 cells, but enhances the B-1 cell population (58). Our data indicate that pro-B cells ectopically expressing Spi-C have dramatically reduced levels of B220 on their surface, which could suggest that these cells may be switching to a B-1 cell phenotype. The phenotype observed in WT-Spi-C pro-B cells is not due to decreased PU.1, because protein levels remain equal to WT-R1 pro-B cells as observed by Western blot (Fig. 2E). Interestingly, the levels of PU.1 transcripts in WT-Spi-C pro-B cells increase. This may be a result of the pre-B cell transition as noted by other studies using GFP knockin at the PU.1 locus (59). Taken together, these results indicate that Spi-C has the potential to promote B cell differentiation, as well as the possibility of functioning in B-1 cell development.
Previous work suggests that the FcRIIb gene is a target of PU.1 activity (Refs. 29 and 60 and our unpublished observations). We found that FcRIIb transcripts are greatly reduced by the ectopic expression of Spi-C (Fig. 4D). In addition, we demonstrated that PU.1, but not Spi-C, activates transcription from the FcRIIb promoter using a transient transfection assay (Fig. 6C). Although transient transfection of the minimal FcRIIb promoter with Spi-C did not repress activity like that seen with reduced transcripts in WT-Spi-C pro-B cells, we speculate that the repressive activity of Spi-C on the FcRIIb promoter may be mediated through another DNA-binding element not included in the minimal promoter used in transient transfection assays. However, our results indicate that PU.1 is a positive regulator of the FcRIIb gene, while Spi-C functions as a negative regulator of this gene either as an active repressor or an inert inhibitor of PU.1 activity. The regulation of receptor genes plays an important regulatory role during B cell development. In addition, the FcRIIb inhibitory receptor has been linked to various autoimmune diseases (reviewed in Ref. 61). Proliferation in response to functional pre-BCR signaling is necessary for progression to the immature stages, so down-regulation of the FcRIIb inhibitory receptor would promote this proliferation and may play an important role in B cell development. The expression of Spi-C at particular stages requiring proliferation, such as pre-B and mature B cell stages, suggest that it functions to eliminate the antiproliferative effects of PU.1 and the FcRIIb receptor.
In contrast to down-regulation of the FcRIIb gene as a result of Spi-C expression, IgH sterile transcription (Iµ) was increased. This is similar to our findings that IgH sterile transcripts were increased in PU.1–/–Spi-B–/– pro-B cells relative to WT pro-B cells (29). Using transient transfection assays, we were able to demonstrate that PU.1 repressed activity of the Eµ enhancer, while Spi-C partially relieved this repression (Fig. 6A). Repression of Eµ by PU.1 was dependent on DNA-binding interactions (Fig. 6A). Finally, we found that mutation of the PU.1 consensus site in Eµ resulted in a slight, although statistically insignificant, activation of reporter gene activity (Fig. 6B). Taken together, these results suggest that PU.1 functions as a repressor for the Eµ enhancer. Although much of the literature describes PU.1 as a transcriptional activator, we note that PU.1 was recently found to participate in developmentally regulated repression of several regulatory elements, including the tal-1 silencer, the p16Ink4a promoter, and the IgH 3' regulatory region enhancer HS1,2 (14, 62, 63).
When examining the 3' regulatory region of the IgH locus, we observed that acetylation of histones H3 and H4 was generally equivalent between WT-R1 pro-B cells and WT-Spi-C pro-B cells (Fig. 7A). In contrast, di-Me K9 H3 was significantly reduced in WT-Spi-C pro-B cells (Fig. 7B). This histone mark causes a condensation of chromatin and inhibits expression. The observation that this mark is removed by Spi-C ectopic expression suggests that this may help facilitate IgH expression, and is associated, directly or indirectly, with the increased IgH sterile transcription observed in WT-Spi-C pro-B cells. Perhaps, compared with PU.1, Spi-C has unique interactions with either histone methyltransferases or factors that recruit these enzymes, providing a mechanism of derepression at the IgH locus.
Our gel-shift results demonstrate that PU.1 and Spi-C can interact with a defined binding site in the Eµ enhancer and FcRIIb promoter (Fig. 5). Although levels of PU.1 protein were found to be equivalent, gel-shift complexes corresponding to PU.1 are reduced in WT-Spi-C pro-B cell nuclear lysates relative to WT-R1 pro-B cell nuclear lysates. Combined with the results discussed above that PU.1 and Spi-C have opposing activities on the FcRIIb and IgH genes, our data suggest that Spi-C can inhibit PU.1 DNA binding either directly or indirectly. This type of inhibition is also observed with Ets-2 repressor factor, which acts as a repressor to Ets-2 activity (64). Therefore, we propose that when Spi-C expression is induced during normal B cell development, it inhibits PU.1 binding to target genes. We expect that the activation domain of Spi-C functions differently from that of PU.1, such that Spi-C relieves PU.1-mediated repression of the IgH intronic enhancer, while interfering with PU.1-mediated activation of the FcRIIb promoter. Because Spi-C expression is confined to particular stages of B cell development, beginning at the pre-B cell stage and ending in mature B cells, it may function in a developmental stage-specific manner. Taken together, our data lead us to propose that the biological function of Spi-C is to antagonize PU.1 to promote B cell differentiation.
|
Acknowledgments |
|---|
vector; William Miller (University of Cincinnati College of Medicine, Cincinnati, OH) for his gift of the monoclonal anti-HA Ab; Olivia Schneider for initial cloning of the FcRIIb promoter; and Ryan Lorber and Sabrina Volpi for technical assistance. We acknowledge the flow cytometry core of the Cincinnati Children’s Hospital Research Foundation for assistance in cell sorting and the real-time PCR core of Albert Einstein College of Medicine for assistance in ChIP analysis. We thank Isaac Houston for critically reading the manuscript. |
Disclosures |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Footnotes |
|---|
1 This work was supported by National Institutes of Health Grants 1R01AI052175 (to R.P.D.) and AI13509 (to B.K.B.). 
2 Address correspondence and reprint requests to Dr. Rodney P. DeKoter, Department of Molecular Genetics, Biochemistry and Microbiology, University of Cincinnati College of Medicine, Medical Sciences Building 3256, 231 Albert Sabin Way, Cincinnati, OH 45267-0524. E-mail address: dekoterp{at}ucmail.uc.edu
3 Abbreviations used in this paper: pro-B, progenitor B; HS, hypersensitive site; IRF, IFN regulatory factor; WT, wild type; HA, hemagglutinin; IRES, internal ribosomal entry site; FLP, fetal liver progenitor; MF, mean fluorescence.
Received for publication February 23, 2006. Accepted for publication May 27, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
-hs4, a novel Ig heavy chain enhancer element regulated at multiple stages of B cell differentiation. Nucleic Acids Res. 23: 975-981. 3'E and hs3). J. Immunol. 158: 4780-4787. [Abstract] 2–4 enhancer. Mol. Cell. Biol. 13: 6452-6461. 3' enhancer activity. Mol. Cell. Biol. 12: 368-378. 
