The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

The Journal of Immunology, 2007, 179, 7663 -7673
Copyright © 2007 by The American Association of Immunologists, Inc.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Related articles in The JI
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Shell, S. A.
Right arrow Articles by Milcarek, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Shell, S. A.
Right arrow Articles by Milcarek, C.

Increased Phosphorylation of the Carboxyl-Terminal Domain of RNA Polymerase II and Loading of Polyadenylation and Cotranscriptional Factors Contribute to Regulation of the Ig Heavy Chain mRNA in Plasma Cells1

Scott A. Shell2, Kathleen Martincic, Joseph Tran and Christine Milcarek3

Department of Immunology, School of Medicine, University of Pittsburgh, Pittsburgh, PA 15261


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
B cells produce Ig H chain (IgH) mRNA and protein, primarily of the membrane-bound specific form. Plasma cells produce 20- to 50-fold higher amounts of IgH mRNA, most processed to the secretory specific form; this shift is mediated by substantial changes in RNA processing but only a small increase in IgH transcription rate. We investigated RNA polymerase II (RNAP-II) loading and phosphorylation of its C-terminal domain (CTD) on the IgG2a H chain gene, comparing two mouse cell lines representing B (A20) and plasma cells (AxJ) that express the identical H chain gene whose RNA is processed in different ways. Using chromatin immunoprecipitation and real-time PCR, we detected increased RNAP-II and Ser-2 and Ser-5 phosphorylation of RNAP-II CTD close to the IgH promoter in plasma cells. We detected increased association of several 3' end-processing factors, ELL2 and PC4, at the 5' end of the IgH gene in AxJ as compared with A20 cells. Polymerase progress and factor associations were inhibited by 5,6-dichlorobenzimidazole riboside, a drug that interferes with the addition of the Ser-2 to the CTD of RNAP-II. Taken together, these data indicate a role for CTD phosphorylation and polyadenylation/ELL2/PC4 factor loading on the polymerase in the choice of the secretory poly(A) site for the IgH gene.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
A crucial step in the development of B cells into plasma cells with B lymphocyte induction of maturation protein 1 (blimp-1)4 induction is the increased production of secreted Ig H chain (IgH); this takes place by differential selection of the promoter-proximal, secretory specific (sec) poly(A) site on the pre-mRNA. In mature and memory B cells, most IgH mRNAs are expressed as the membrane-bound (mb) encoding forms while, in plasma cells, mRNA levels increase dramatically and the expression of the IgH mRNA and protein shifts to predominantly the sec form. The ratio of sec:mb forms may exceed 50:1 in some plasma cells, as reviewed in Ref. 1 . Studies with transfected Ig genes have indicated a major role for a shift in the balance between polyadenylation and splicing at the secretory site (2, 3). Overexpression of the polyadenylation cleavage factor CstF-64 (CstF2) in chicken DT40 B cells artificially tipped the balance of IgH mRNA production to the secretory form (4). However, several studies investigating the levels of CstF-64 in plasma cells or models of differentiation to Ig secretion with blimp-1 have failed to show significant increases in CstF-64 mRNA or protein levels (5, 6, 7, 8, 9). Meanwhile, the small nuclear ribonucleoprotein U1A has been implicated in influencing IgH poly(A) site choice by binding to sequences near the sec poly(A) site (10). The levels of U1A were associated with RNA decrease in plasma cells, commensurate with a potential role in regulation (11, 12); in another study, the U1A-binding site was found to influence basal, but not regulated, Ig mRNA expression (13). Additionally, the heterogeneous nuclear ribonucleoprotein (hnRNP) F can block access to the secretory site by CstF-64; lowered levels of hnRNP F in plasma cells may account for increased polyadenylation at the sec site. Artificial overexpression of hnRNP F in plasma cells lowered sec mRNA levels, thereby making it a likely candidate for regulation (14). Some serine-arginine-rich proteins, which can influence the splicing reaction competing for the use of the sec poly(A) site, have been shown to be lower in plasma cells (15). However, regulation directed by any of these trans-acting factors does not explain the curious observation that transfection of an IgH-coding region, driven by the heterologous SV-40 promoter, into a variety of transformed cells, resulted in production of the sec mRNA form in vast abundance to the mb form (16). Activation to Ig secretion from the native gene has been shown to be stimulated 2- to 5-fold by increasing transcription through histone acetylation and generalized opening up of the chromatin (8, 17). Therefore, the possible link between transcription activation and RNA processing of the IgH pre-mRNA to the sec form posed a heretofore unexplored question.

Covalent modifications to the C-terminal domain (CTD) of RNA polymerase II (RNAP-II) have been shown to vary across the length of a yeast gene and result in the recruitment or release of RNA-processing factors (for example, see Ref. 18). The RNAP-II CTD is composed of a seven amino-acid sequence YSPTSPS in 27–52 tandem repeats, depending on the organism. Although modifications can occur on five of the seven residues, the most commonly studied CTD posttranslational modifications are the phosphorylation of the essential Ser-2 and Ser-5 residues in the heptamers (19). Previous studies in yeast have shown that Ser-5 of RNAP-II is phosphorylated on RNAP-II near the promoter while Ser-2 is preferentially phosphorylated downstream during elongation, suggesting an RNAP-II CTD phosphorylation code (20, 21). This pattern of serine phosphorylation sites has not been as clearly established in mammalian cells; but it is apparent that the coupling of RNA-processing events with transcription in mammalian cells occurs via the CTD of RNAP-II (as reviewed in Ref. 22 and more recently in Ref. 23). Higher eukaryotic promoters influence differential association of splicing factors to RNAP-II at the CTD to induce alternative processing (24), but how this is accomplished and where along the gene it occurs is still under active investigation.

Varying elongation rates of RNAP-II by changing the promoter, adding drugs that alter polymerase or chromatin, or modifying the RNAP-II itself resulted in changes in RNA-processing patterns in mammalian cells (22). Faster elongation rates generally favor exon skipping. The transcriptional elongation factor ELL2 was originally discovered because of its similarity to the 11 lysine-rich leukemia gene, ELL1, a frequent fusion partner in multiple lineage leukemia. Transfection of B cells with blimp-1 increased ELL2 expression and IgH secretion (5, 7); ELL2 is increased in authentic plasma cells as well (6). ELL2 has been shown to promote elongation in vitro by keeping the 3'OH of nascent RNA in alignment with the catalytic site; thus, it prevents Pol II from backtracking (25). The role of the cotranscriptional factor PC4 has been somewhat controversial as it was shown to act positively with OCA-B to aid its function in B cells (26) while acting negatively to impair phosphorylation of transcription factors (27). By microarray analyses, the levels of PC4 were shown to increase in some plasma cells (6).

We investigated the patterns of polymerase loading and serine phosphorylation of CTD in two cell lines representing either B cells (A20) or plasma cells (AxJ) which are noteworthy in that they carry the identical IgG2a H chain gene. Using chromatin immunoprecipitation (ChIP) and real-time PCR across the IgH locus, we found that there is a large increase in both Ser-5 and Ser-2 phosphorylation of CTD near the 5' end in AxJ cells concomitant with high IgH sec mRNA production in AxJ cells. Factors for polyadenylation (CPSF-160, CstF-50, and -64), transcription elongation (ELL2), and cotranscription activation (PC4) can be found in much greater abundance near the 5' start site in the AxJ cells than in A20 cells. We conclude that increased phosphorylation of RNAP-II at the start of transcription and the increased association of polyadenylation and elongation/cotranscription factors are important in influencing alternative mRNA processing of the IgH gene.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Cell culture

The following mouse cell lines were used: A20 (Ig {gamma} 2a+) a B cell line, AxJ (Ig {gamma} 2a+) a plasma cell line produced as a hybridoma of A20, the plasma cell line J558L (J558L lacks a H chain gene but behaves like a plasma cell), and 4T001 (Ig {gamma} 2b+) a plasma cell line. These lines were previously described (28). All were grown at 37°C with 5% CO2 in IMDM (Invitrogen Life Technologies) supplemented with 10% FCS, 10 mM glutamax, and 100 U penicillin/streptomycin. Cells were maintained at ~5 x 105 to 1 x 106/ml and passaged every second day. Cells were verified as plasma cells by an increased expression of blimp-1, J-chain, ELL2, PC4, and the transcription factor bright by PCR or Western blotting (data not shown); they also showed decreased expression of pax-5 by PCR (data not shown).

A20 IgG2a promoter sequencing by 5'-RACE PCR

The Universal GenomeWalker kit from BD Biosciences (K1807-1) was used to obtain the promoter region for the A20 IgG2a H chain gene. Beginning with the published sequence for the V region of the A20 B lymphoma cell line, two gene-specific antisense primers (GSPs) were created that targeted this known sequence. These primers were: GSP1, antisense, GCCAGTGAGGGACAAAGAAAGCATAG and GSP2, antisense, CCAATA GTCGCAGGCGGAATAATCAC. In brief, A20 genomic, DNA, purified using the DNeasy kit (Qiagen), was digested with one of each of four blunt-end DNA restriction enzymes (DraI, EcoRV, PvuII, and StuI) provided with the Universal GenomeWalker kit creating four genomic DNA libraries. After observing a streak of DNA on an ethidium bromide-stained agarose gel from a fraction of the digestion reaction confirming successful DNA digestion, each digestion was purified and ligated to the GenomeWalker adaptors according to the manufacturer’s instructions. Two sequential PCR were subsequently performed using AP1 and GSP1 followed by AP2 and GSP2 (see Fig. 2). Each purified genomic DNA library was combined with 10 pM each of forward AP and reverse GSP, Mg2+, dNTP mix, 10x reaction buffer, and Advantage Genomic Polymerase Mix according to the manufacturer’s instructions. A modified touchdown PCR program suggested by the protocol was used to reduce nonspecific amplification. A second round of touchdown PCR was performed using 5 µl of first-round PCR product stemming from each genomic DNA library, and 10 pM each of forward AP2 and reverse GSP2. A significant PCR product was observed from the two sequential rounds of PCR stemming from the PvuII-created genomic DNA library. This PCR product was gel purified using the QIAEX II gel extraction kit (20021; Qiagen) and ligated to the pCR2.1 plasmid as part of the TA Cloning kit (25-0024; Invitrogen Life Technologies) according to the manufacturer’s instructions. The ligated plasmid was amplified in One Shot DH5{alpha}-T1 competent cells (12297-016; Invitrogen Life Technologies), purified over a Maxi-prep column (Qiagen), and sequenced (School of Medicine, University of Pittsburgh, Biotechnology Center, Pittsburgh, PA) using M13 sense and antisense primers that bind to pCR2.1 regions that flank the inserted PCR amplicon, namely sense 5'-CAGGAAACAGCTATGAC and antisense, 5'-GTAAAACGACGGCCAG.


Figure 2
View larger version (64K):
[in this window]
[in a new window]

  		FIGURE 2. The A20 IgG2a promoter possesses elements common to most Ig promoters. A, Illustrated is the 5'-RACE PCR strategy for obtaining the A20 IgG2a promoter sequence. The Genome Walker Adapter is ligated to the blunt-end of genomic DNA created by enzymatic digestion using PvuII. The first round of PCR was performed using the linker-containing genomic library as template, the upstream-most adapter primer (AP1) and the downstream most gene-specific primer (GSP1). The second round of PCR was performed using the PCR product generated from the first-round PCR as template, the nested adapter primer (AP2), and the nested gene-specific primer (GSP2). B, The sequence of the entire PCR product generated from the nested second round of PCR. The underlined sequences are the H chain leader sequence and the H chain V region, respectively. The sequence obtained from the 5'-RACE PCR for the leader and V region are an exact match to the published A20 Ig sequence. The sequence was deposited at GenBank and assigned Accession No. bankit923194/EF688290. C, The locations of canonical promoter elements are listed relative to the start codon contained in the A20 Ig {gamma} 2a H chain leader region.

 
Chromatin preparation and ChIP

Chromatin preparation and ChIP procedures were done according to the ChIP-it kit instructions (Active Motif). Briefly, ~5 x 107 actively growing cells were treated for 10 min with formaldehyde at a final concentration of 1% in IMDM to cross-link nucleic acids and proteins. Cells were then washed once with 1x PBS (pH 7.4) and treated with 1x Glycine Stop Fix solution to stop the cross-linking reaction. Cells were then pelleted, washed with PBS, and resuspended in 1 ml of lysis buffer. The resuspended cells were homogenized with a B-type pestle with 10 strokes at 4°C, and nuclei were pelleted by centrifugation at 5000 rpm for 12 min. Nuclei were then resuspended in 1 ml of ice-cold shearing buffer and sonicated for 20 pulses of a 15 s on followed by 45 s off pulse regime with a Branson microtip (Misonix Microson; Misonix) at a 7-watt (RMS) power setting. Chromatin was stored in 0.32-ml aliquots at –80°C until use.

All ChIP procedures were conducted at 4°C. Briefly, a sheared chromatin aliquot was precleared by rocking on a Nutator (Clay Adams) for 2 h with PBS-washed protein A/G Sepharose (Amersham Pharmacia) or anti-IgM-cross-linked agarose (ICN Pharmaceuticals). IgM Abs were directly added to the precleared chromatin preparations and the IP procedure was conducted overnight with nutation. IgM Ab/protein complexes were collected by adding anti-IgM agarose and incubating for 2 h. IgG Abs were prebound to protein A/G Sepharose and then added to chromatin preparations and rocked overnight. For most Abs, we determined the amount of optimal IP to be from 4 to 20 µg per ~2 x 106 cell equivalents. All IPs were washed according to the ChIP-it kit instructions using the provided buffers (Active Motif). Formaldehyde cross-link reversal was achieved by incubating at 65°C overnight with RNase A. Proteins were then digested for 2 h with proteinase K and DNA was collected on the DNA minicolumn in the kit. Approximately 2% of the sample was used per quantitative PCR (QPCR) well. Commercially available Abs used in this study included those specific for: TFIIB (ChIP-it kit; Active Motif); NH2 terminus of RNAP-II (N20x, polyclonal rabbit IgG, sc899x; Santa Cruz Biotechnology); Ser-2 phosphorylated RNAP-II (H5, monoclonal mouse IgM, MMS-129R; Covance); Ser-5 phophorylated RNAP-II (H14, monoclonal mouse IgM, MMS-134R; Covance); all CTD of RNAP-II phosphorylated or not (CTD-4H8, affinity purified mouse IgG1, MMS-128P; Covance); histone 3 (polyclonal rabbit IgG, 06-755; Upstate Biotechnology); K9/K14-acetylated histone 3 (polyclonal rabbit, IgG, 06-599; Upstate Biotechnology); and PC4 (D18 and N-17, goat IgG, sc. nos. 9442 and 9441; Santa Cruz Biotechnology). Other Abs included anti-CPSF-160 (R7006, polyclonal rabbit; Dr. C. MacDonald, Texas Tech University, Lubbock, TX) and anti-ELL2 (R4502, polyclonal rabbit, raised to the peptide GCLMNKKARISHLTNRV, present in the middle of ELL2, coupled to keyhole limpet hemocyanin). The rabbit Abs were purified from sera over a column of peptide Ag). Two mouse monoclonal IgG Abs used were provided by Dr. C. MacDonald, namely, anti-CstF-50 (2C1) and anti-CstF-64 (3A7). Purified mouse IgM (20 µl or 25 µg; ICN Pharmaceuticals) or normal rabbit serum (~25 µg) was used as negative controls. Additional Abs were goat anti-mouse IgM (M8644, 1 mg/ml; Sigma-Aldrich) for sandwich with protein A/G Sepharose. All ChIP experiments were performed a minimum of three times per Ab, using at least three different nuclear preparations.

Quantitative PCR

QPCR procedures used SYBR Green 2x PCR Master Mix (Applied Biosystems) with primers specific for regions along the specific gene at 0.25 µM final concentration, on an Applied Biosystems 7900HT. The following PCR program was used for all ChIP DNA samples: Taq activation for 12 min at 95°C, followed by 40 cycles of a 15-s denaturing step at 95°C, and a 1-min annealing step at 60°C. Primers were selected which all had melting temperature (Tm) of between 58.5 and 60°C. Primers were first checked on genomic DNA to ensure that under these conditions the half-maximal cycle threshold (Ct) value was ~22–24; genomic DNA was purified using the DNeasy kit (Qiagen). Each sample was run in duplicate or triplicate for each plate, at least two plates per sample. Unrearranged, genomic DNA from the rag–/– mouse was provided by Dr. L. Borghesi, (University of Pittsburgh, Pittsburgh, PA). Data analysis was done using ABI 7900HT plate analysis software (Applied Biosystems) and Microsoft Excel 2000. Calculations were based on {Delta}Ct from the input chromatin DNA sample from that preparation and 2–{Delta}Ct transformations (times for example, 10, if the non-IP chromatin DNA sample was one-tenth the volume of the IP sample) to get a percentage IP of input. All samples were normalized to an actin control, calculated in the same way, which was run for each plate and for each Ab used in this study. Normalization was done using the formula: (percent IP of input for sample)/(percent IP of input for actin) = percent IP normalized. Error bars in the figures indicate the SDs from multiple normalized determinations. Samples were run from at least three different cell preparations and the actin-normalized results were averaged.

Primer sequences

The mouse gene primers used for QPCR analyses for IgHs and blimp are indicated in Table I. See Fig. 1 for the location of those primers on the rearranged Ig {gamma} 2a H chain (IgG2a) in A20 and AxJ cells. The primers for the 5' transcription start sites of the other genes are also indicated. All primers were synthesized by Integrated DNA Technologies. We used their software for determining Tms, lack of hairpins, and nonhybridization to other genes. The typical region amplified by QPCR with these primers was from 60 to 120 nt of genomic DNA.


View this table:
[in this window]
[in a new window]

  		Table I. Primers used for ChIP

 

Figure 1
View larger version (16K):
[in this window]
[in a new window]

  		FIGURE 1. Map of the {gamma} 2a H chain gene and locations of the primers used for the QPCR analysis. The identical, expressed, rearranged IgH gene in A20 and AxJ cells is shown diagrammatically. The top line shows the names of the primer sets used in subsequent QPCR experiments with the locations spanned indicated by a bar. Sequences of the primer pairs are given in Table I. On the gene, the boxes represent exons and the circle represents the area covered by the transcriptional enhancer. L indicates the protein leader sequence which is removed after translation. J3 and J4 are spliced out of the pre-mRNA transcript. Alternative processing occurs at the 3' end of the transcript with mutually exclusive use of the secretory specific poly(A) site and 3'UTR or the splicing of a 5' splice site near the end of CH3 with M1 and M2. Alternatively processed mRNA products are shown with an inset with the relative amount of each mRNA in the cells. A 1-kb size marker is indicated.

 
Elongation assay

Nuclei from ~10 million actively growing A20 and AxJ cells were isolated using Nonidet P40 lysis and differential centrifugation through a sucrose cushion, as previously described (28). RNA was isolated using Ultraspec (BL-10100; Biotecx) as per the manufacturer’s protocol. The RNA was then treated with DNase (DNA-Free; Ambion) for 30 min at 37°C. RNA was further purified using the RNeasy kit (74104; Qiagen). Reverse transcriptase (RT) reaction (Superscript II; Invitrogen Life Technologies) was conducted using random hexamer nucleotide primers from the kit. The RT product was then analyzed by QPCR using the indicated primers. Experiments were done in triplicate. The RNA only and the RT reactions were subject to QPCR analysis; DNA contamination was shown to be eliminated by the treatments. Results are presented as 2 raised to the power of the negative {Delta}Ct of the A20 or AxJ sample relative to 30 ng of genomic A20 DNA run on the same plate x100 to produce the percentage value shown in the figures. Primers to exon 3:intervening sequence (IVS) of β-actin were run in the same QPCR plates as normalization controls relative to 30 ng of purified A20 genomic DNA. The RNA samples were normalized to the actin values to compensate for yields in the isolation and RT reaction. The exon3:IVS primers for actin were: sense, GATCGATGCCGGTGCTAAGA and antisense, GGAGTCCTTCTGACCCATTCC (set no. 18189347). Primers used for the QPCR analysis of the randomly primed, RT product from nuclear RNA included hinge primers, sense CCTGAGTAAC5AGCCTTCT, and antisense CATGGAGGACAGGGCTTGA (set no. 18903658, and from the table above: D:J2, set no. 189036456; enhancer region (EH), set no. 13623138; and M1, set no. 18903660). These primers include at least one end fully in an intron to eliminate amplification of mature mRNA.

Transcription run-ons

Run-on experiments were performed as previously described (29). Antisense RNA was transcribed using T7 or SP6 from DNA templates as described above but in the presence of high levels of NTPs and a trace of [3H]UTP so that the product yield could be determined. The templates for the RNAs immobilized on the filter were obtained from DNA clones flanked by T7 or SP6 promoters. For the V region, we cloned the A20 V region from the 5'-RACE PCR (see Fig. 2) into pGEM (Promega) which, when transcribed with T7, produced a 938-nt antisense transcript (our no. CM535). The EH was used to produce 1200 nt of antisense transcript from a pGEM3 vector (CM67). The CH3 clone produced a 312-nt antisense copy of the {gamma} 2a cDNA in the CH3 region (CM347). The actin clone produced a 1150 antisense transcript of the 3' end of the actin cDNA (CM281). The reverse RNA was made as an SP6 sense transcript from the M1-M2 region of CM332. All of the antisense RNAs were immobilized on nylon membranes (Genescreen; NEN) in 25 mM NaPO4 (pH 6.5) in a slot blotter; 1 µg of RNA was immobilized per well and the membranes were exposed to 254-nm shortwave UV at 1200 µW/cm2 for 2 min.

The filters were hybridized with [32P]RNA made in nuclei of the indicated cells with [32P]UTP in a 30-min labeling. RNA was purified using Ultraspec (BL-10100; Biotecx) and the PCR cleanup kit from Edge Biosystems to remove unincorporated nucleotides. The labeled RNA was then hybridized for 18 h to the filters in a plastic bag with a solution containing 50% formamide; 0.25 M NaPO4 (pH 7.2); 0.25 M NaCl; 1 mM EDTA; 7% SDS; and 100 µg/ml denatured salmon sperm DNA. The filters were extensively washed, treated with 0.25 µg/ml RNase A for 20 min at 25°C, and washed again several times. Filters were exposed to phosphoimage amplification screens and the image was analyzed on a Molecular Dynamics Storm phosphoimager with ImageQuant software after 1 wk. The actin signal was used as a normalization control between the cell types.

5,6-Dichlorobenzimidazole riboside (DRB) treatment

DRB (D1916; Sigma-Aldrich) inhibits RNA synthesis by blocking elongation and inhibiting pTEFb phosphorylation of CTD Ser-2. The drug was used for 2 h on the indicated cells at 65 µM final concentration. DRB was resuspended in pure DMSO at 65 mM and diluted into cell medium immediately upon use, resulting in a final DMSO concentration of 0.1%. Control cells were treated with DMSO at 0.1%. Chromatin isolation and QPCR were performed on cells as described above.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Characterization of the Ig {gamma} 2a H chain and the cell lines

We compiled the published V-region sequence of the A20 B cell lymphoma line (30), the genomic sequence for the murine IgG2a C regions (GenBank accession no. J100470), and membrane-spanning exons and poly(A) sites (GenBank accession no. J100471) into Fig. 1 in preparation for ChIP experiments. We attempted to design primers that spanned the ~11 kb of genomic DNA but the full sequence of the promoter region and 5'UTR for the A20 IgH mRNA was not available, because only a partial cDNA had been published. The A20 V region is present only once in the haploid IgH tandem V locus (31). The A20 B cell line is a clonal population of memory B cells where the identical VDJ2-recombined IgG2a gene and promoter are active in every cell. Therefore, through a 5'-RACE PCR approach upstream from the known V region of A20 IgG2a, we obtained the 5'UTR and promoter sequence. It was known that the V:D:J recombination event involved the J2 segment of the J locus (30), so we designed the first GSP1 to target the genomic region between J2 and J3 (Fig. 2A). The target sequence for the second GSP (GSP2) spanned the region where the V, D, and J regions have uniquely joined in A20 B cells through recombination.

Shown in Fig. 2B is the sequence obtained from the 5'-RACE PCR protocol for the A20 IgH gene. The sequence was deposited at GenBank and assigned accession no. bankit923194/EF688290. The A20 promoter contains multiple conserved sequence motifs that could bind B cell-specific transcription factors required for Ig production. Octamer-binding proteins typically bind this sequence at an octamer element (ATGCAAAT) lying ~50 nt upstream from the transcription start site of most Ig V regions so it is a strong contributor to Ig gene expression; a common Ig heptamer motif also appears in the A20 V-region promoter. Also located in the A20 V-region promoter are a potential C/EBP transcription factor binding site, a pyrimidine stretch, and a TATA box (Fig. 2C), which have been shown to be functionally important to Ig transcription in other IgH genes (32).

A20 and AxJ cells share the same Ig {gamma} 2a H chain gene because AxJ was created as a fusion of A20 and the J558L plasma cell line, which had deleted its own IgH gene. This Ig {gamma} 2a H chain gene is primarily processed into sec mRNA in AxJ cells while the membrane forms of IgH mRNA and protein predominate in A20 (28) and as indicated in Fig. 1 (inset). Several mRNAs, whose expression is distinctive of plasma cells, have been examined in AxJ cells; the results are summarized in Materials and Methods as part of the characterization of the cells. Foremost in the plasma cell gene expression profile is the up-regulation of blimp-1 (33), the master regulator of plasma cell differentiation (34). Because measuring the transcription of blimp-1 by chromatin IP would reveal whether it was up-regulated in the AxJ cells, and also tell us something about the blimp-1 gene itself, we assayed blimp-1 expression using ChIP with N20, an Ab which recognizes the N terminus of RNAP-II independent of its phosphorylation state. We then used QPCR, with primers that span intron-exon boundaries of the blimp-1 gene, as shown in Table I, to quantify polymerase at various regions on the gene. We normalized our results from that same chromatin preparation to the value obtained with the 5' end of the β-actin gene, which is unaffected by the differentiation to the sec phenotype as judged by microarray (5, 6, 7). Normalizing to actin allowed us to compensate for differences in cell metabolism, cross-linking efficiency, and IP yields across cells and experiments. As shown in Fig. 3, the AxJ cells substantially (>10-fold) up-regulate blimp-1 transcription relative to the A20 cells. Increased blimp-1 expression in AxJ cells was also verified by QPCR and Western blots (data not shown). With the blimp-1 transcription, there is an apparent fall-off in the amount of immunoprecipitated DNA after the promoter. This decrease in RNAP-II has been seen with a number of genes (see Discussion) and is especially apparent in this gene which spans 21 kb of DNA down to exon 8 where we can no longer detect a signal above background (data not shown); the distance between the probes used here for exons 1 and 4 is ~7.5 kb (for sequence of primers see Table I). We conclude from the expression profile for AxJ cells that they produce blimp-1 and represent plasma cells in this regard.


Figure 3
View larger version (20K):
[in this window]
[in a new window]

  		FIGURE 3. Increased Blimp-1 transcription in AxJ cells. The loading of RNAP-II was measured on the Blimp-1 gene by ChIP followed by QPCR analysis of the associated DNA in the two cells. Probes used to detect the various blimp exons are listed in Table I. The amount of immunoprecipitated DNA was measured by QPCR and calculated as the percent input of total chromatin, as described in Materials and Methods, normalized to an actin 5' end control. Error bars indicate the SDs between multiple samples. {square}, The results with A20 cells; Figure 3, the results from AxJ cells.

 
Increased phosphorylation of RNAP-II CTD at Ser-2 and Ser-5 in AxJ plasma cells

We designed multiple primer pairs (Table I) that span the 11-kb IgH genomic locus and intron/exon or unique V:D:J2 Ig rearrangement sites, extending from the promoter region to the membrane-specifying poly(A) site whose locations are indicated in Fig. 1. We conducted ChIP assays using the N20 Ab to determine how much total RNAP-II was present along the IgH gene in the two cell lines. As shown in Fig. 4, the IgH locus is actively transcribed; AxJ cells appear to have only slightly more polymerase on the 5' end of the IgH gene than that seen in the A20 cells (Fig. 4, B vs A) with 9% of input vs 5% immunoprecipitatable by the N20 Ab to RNAP-II. There is an apparent decrease in polymerase loading across the Ig gene into the enhancer region in AxJ cells but the drop-off of polymerase is not so precipitous as that seen with the blimp-1 gene. Some have interpreted the decreased IP of RNAP-II after the 5' end vs downstream regions in other genes as indicating extensive polymerase fall-off (35); but for the IgH gene, this result might alternatively represent unequal transit speeds through the downstream regions or unequal accessibility to Ab. Subsequent experiments will address these issues for the IgH gene in these cells.


Figure 4
View larger version (25K):
[in this window]
[in a new window]

  		FIGURE 4. Ser-2 and Ser-5 phosphorylation of RNAP-II CTD and RNAP-II change differentially across the IgG2a H chain genomic locus between AxJ plasma cells and B cells. A and B, An Ab to the N terminus of RNAP-II CTD (N20) was used; in C and D, Abs to Ser-5 or Ser-2 phosphorylation of RNAP-II CTD were used in ChIP across the IgG2a H chain locus. In all panels, IP was followed by QPCR analysis of the associated DNA. The amount of immunoprecipitated DNA was measured by QPCR for all panels and calculated as the percent input of total chromatin, as described in Materials and Methods, normalized to an actin 5' end control. See Fig. 1 for map of the locations of QPCR probes and Table I for the primers used. ChIP experiments at each locus were performed in duplicate or triplicate with at least three different batches of cells; the SDs are shown by the error bars in this and all subsequent figures. A, A20 cells N20 (RNAP-II) Ab. B, AxJ cells N20 Ab. C, Comparisons of the level of Ser-5 ({blacksquare}) and Ser-2 phosphorylation (Figure 4) of RNAP-II CTD in A20 cells. D, Ser-5 and Ser-2 results in AxJ cells.

 
In A20 cells, RNAP-II density is lowest at the enhancer region (Fig. 4A) but higher right after the secretory poly(A) site. This is a region which has been suggested to be a pausing site for polymerase from mutational analyses of the Ig mu gene (13) and follows a poly(A) site which might itself cause pausing (36). However, we do not see as significant pausing in the AxJ cells where that poly(A) site is actually being used (Fig. 4B). Meanwhile, in A20 cells, polymerase pausing might allow for acquisition of factors to work on the CH3 to M1 and M2 splice sites.

By using ChIP with specific Abs that recognize Ser-5 vs Ser-2 phosphorylated CTD of RNAP-II followed by QPCR, we were able to assess the phosphorylation state of RNAP-II CTD across the IgH locus in the two cell lines. We observed that the level of Ser-2 and Ser-5 CTD phosphorylation on the 5' end (TATA and V:J) on the IgH in AxJ cells was 12–14% of that with input DNA, ~6.5-fold more than that observed in A20 memory B cells (Fig. 4, compare C and D). Therefore, the RNAP-II at the start of the Ig gene in the AxJ plasma cells is more heavily phosphorylated on the CTD, both at both Ser-5 and Ser-2. Downstream in AxJ cells, the apparent phosphorylation of RNAP-II CTD on the IgH gene at the internal H chain enhancer was reduced to lower levels. Although CTD phosphorylated at Ser-2 and Ser-5 in AxJ cells was increased slightly 3' of the enhancer, the signal did not rebound to the level observed at the AxJ promoter and V regions. We saw a small increase in polymerase between the sec poly(A) site and M1 but no dramatic pausing downstream of the sec poly(A) site in AxJ cells as has been suggested for the mu H chain gene (13). In marked contrast, in A20 B cells the phosphorylated Ser-2 and Ser-5 signals were consistently lower across the whole IgH gene (Fig. 4C), but we saw an increase in phosphorylated RNAP-II after the enhancer and the secretory poly(A) site, near the "pause site." This may be a pausing to load splicing factors onto the polymerase that would be used to splice CH3 to M1, an action that is mutually exclusive of selection of the sec poly(A) site.

These dramatic changes in CTD phosphorylation at the 5' end of the gene in AxJ cells might be interpreted in several ways, but before trying to decipher them, we did numerous controls to eliminate some possible explanations for the data. To show that the V:D:J2 probe used in the analyses of the chromatin IP was specific for the uniquely rearranged, expressed gene, we used splenic DNA from a rag knockout mouse (provided by Dr. L. Borghesi) in which no Ig gene rearrangement was possible. Our V:D:J2 primer pairs were not able to be amplified in a QPCR using naked DNA from the rag knockout demonstrating the specificity of the V:D:J2 probe for the rearranged allele (data not shown). Similarly, the primer pair for JH1 (see Table I), a J region deleted from the active A20 allele during the somatic recombination process to join DH to J2, was not able to be amplified following the IP reaction while it could be amplified in naked rag–/– DNA where it was unrearranged (data not shown). Primer pairs for β2-globin were not able to be amplified after the ChIP protocol indicating that inactive genes in A20 and AxJ gave no signals. We also analyzed nuclei from J558L, the cell line lacking an IgH, used in the creation of the AxJ cells (28). Performing ChIP with anti-Ser-5 or Ser-2-phosphorylated CTD or RNAP-II Abs with J558L cells revealed no signal above that obtained with a control nonspecific Ab (data not shown). We conclude that the Ser-5 and Ser-2 CTD phosphorylation signals seen in AxJ and A20 cells arise from the authentic IgG2a H chain gene rearranged and expressed in these cells. If there were transcription from the C region in an unrearranged {gamma} allele in the A20 and AxJ cells, it would be difficult to detect with these experiments, but such transcription should be manifested equally in both cell types. In addition, this hypothetical transcription would contribute to the ChIP signal for the C region but not the V:D:J2 signal. Spurious C-region transcription would, in fact, decrease the real differences between the 5' vs downstream signals, thereby making our determinations a minimal estimate of the increase in the RNAP-II and CTD phosphorylation signals near the IgH promoter.

We also wished to determine whether the high level of Ser-5 and Ser-2 phosphorylation at the start of the IgH gene was a characteristic of another plasma cell line. For this, we used 4T001 cells, a plasmacytoma line which expresses high levels of sec mRNA from its resident IgG2b H chain with a different Ig promoter, a unique V:D:J2 combination (37), and a {gamma} 2b H chain C region (38). The 4T001 cell line is derived from the MPC11 tumor, the IgG2b gene of which has been sequenced. We used probes specific for the 5' end of the {gamma} 2b gene (P2b), the enhancer, and the M1 to perform ChIP with the 4T001 cells. As shown in Fig. 5A, we also observe a large amount of both Ser-5 and Ser-2 phosphorylation of CTD at the beginning of the {gamma} 2b H chain gene in these cells but not at the enhancer and beyond. It therefore appears that the increased Ser-5 and Ser-2 phosphorylation of RNAP-II at the CTD in the plasma cells result from modifications to the polymerase that may de-scribe a phenotypic shift fundamental for an increase in sec Ig production.


Figure 5
View larger version (25K):
[in this window]
[in a new window]

  		FIGURE 5. Controls for ChIP experiments. A, ChIP in 4T001 cells. Serine phosphorylation of RNAP-II CTD was measured by ChIP using Abs specific for either of the two sites of phosphorylation of serine (Ser-5 or Ser-2). The amount of immunoprecipitated DNA was measured by QPCR and calculated as the percent input of total chromatin, as described in Fig. 4. Shown are side-by-side comparisons of the level of Ser-5 ({blacksquare}) and Ser-2 phosphorylation (Figure 5) of RNAP-II CTD at several loci across the IgG2b H chain gene. The sequences for the primers for P2b are shown in Table I. The enhancer, EH, is detected by the same primers as used for the A20 and AxJ cells because these regions are identical in all H chains from the J2 to the switch regions. The primers for M1 are from regions where there is identity between {gamma} 2a and 2b sequences, see M1, Fig. 1, and Table I. β-actin was used as a normalization control between experiments. B, Factors on GAPDH gene. Serine phosphorylation of RNAP-II CTD was measured by ChIP using Abs specific for either of the two sites of phosphorylation of serine followed by QPCR analysis of the associated DNA. The amount of immunoprecipitated DNA was measured by QPCR and calculated as the percent input of total chromatin, as described in Fig. 4 and more fully in Materials and Methods. DNA precipitated by anti-TFIIB Abs was used as a normalization control. {square}, Results with A20 cells; Figure 5, results obtained with AxJ cells. C, Equivalent acetylated histone H3 IP between A20 and AxJ. Immunoprecipitation was conducted with Ab to H3 that recognizes lysine 9 and lysine 14 acetylation. The amount of immunoprecipitated DNA was measured by QPCR and calculated as the percent input of total chromatin, as described in Fig. 4. β-actin was used as a normalization control between experiments and cells. {square}, Results with A20 cells; Figure 5, results obtained with AxJ cells.

 
Next, we wished to determine whether there was an overabundance of Ser-5 and Ser-2 phosphorylation of the RNAP-II CTD on all genes in the AxJ cells. We performed ChIP using Abs against Ser-5 or Ser-2 phosphorylated CTD and TFIIB and analyzed the transcription start-site region of the GAPDH gene (primers in Table I). As shown in Fig. 5B, when normalized to the TFIIB signal, the levels of Ser-5 phosphorylated CTD on the GAPDH gene were almost identical in the two cells. Although there was a slight increase in Ser-2-modified RNAP-II CTD on the GAPDH gene in AxJ cells, it was not the 4- to 6-fold change we had seen with the IgH gene. Excessive, indiscriminate serine phosphorylation of the CTD is therefore not a common feature in AxJ cells.

Because the state of chromatin might alter the ability of the ChIP technique to be equally effective in two different cell types, we wished to determine whether the chromatin structure differed between the A20 and AxJ cells around the IgG2a H chain locus. We used Abs recognizing lysine 9- and 14-acetylated histone 3 for ChIP from A20 or AxJ cells. The QPCR results are shown in Fig. 5C. The amount of immunoprecipitable chromatin with acetylated H3 was highest around the 3' end of the H chain enhancer in the middle of the gene for both cell types. Near the promoter, the chromatin should be open but the presence of polymerase can strip histones from the DNA, which may account for the lowered percentage of DNA that we see as immunoprecipitable especially in AxJ cells (39). The pattern of accessibility seen with Abs to total H3 was also similar between the two cells (data not shown). Because the pattern of acetylated H3 on the IgH locus is very similar between the two cell types, alternative chromatin structure cannot account for the observed differences in the phosphorylated Ser-5 and Ser-2 CTD of RNAP-II between these cell lines. The internal enhancer of the H chain locus is flanked by presumptive chromosomal loop anchorage elements (40), but how this influences the ChIP patterns we see here is not known.

Increased association of factors with the H chain gene in AxJ cells

One proposed role for Ser-5 and Ser-2 phosphorylation of the CTD is to facilitate attachment of processing factors onto the polymerase for delivery to the appropriate transcribed sequences; changes in speed and processing patterns might also be achieved if elongation factors were differentially loaded. We therefore conducted ChIP experiments using Abs to several factors that might be important for altering/attaching to the RNAP-II for processing the downstream RNA elements. As shown in Fig. 6, we could detect higher signals for several factors using a QPCR probe (TATA) to the 5' end of the IgH gene in AxJ than in A20 cells. Cleavage/polyadenylation factor CPSF-160 had been previously shown to associate with another promoter and be transferred to the polymerase during transcription (41). CstF-64 (CstF2) which is required for cleavage of the transcript before polyadenylation had been shown to associate with the 5' end of the gene in some experiments in yeast as described in the Introduction. CstF-50 is generally associated with CstF-64 and we were able to detect it at the 5' end of the gene using a mAb of high specificity and affinity. We could not detect a signal with Abs to CstF-77 (rabbit polyclonal; data not shown) but this may be explained by lack of accessibility of the epitope or poor Ab affinity. We found significant higher association of ELL2 and PC4 with the 5' end of the IgH in AxJ vs in A20 cells. The factor PC4 has been shown to interact with OCA-B (42) and CstF-64 in some experiments (43). Significantly higher association near the start of transcription in AxJ cells of several factors, some of which have been previously described as associating with the RNAP-II, indicates that the RNAP-II is more prepared to deliver those factors to the promoter proximal downstream, sec poly(A) signal. The association of ELL2 with the 5' end of a transcribed gene from the cell had not been previously observed and indicates a role for ELL2 in the expression of the IgH gene.


Figure 6
View larger version (22K):
[in this window]
[in a new window]

  		FIGURE 6. Various coactivators are loaded onto the {gamma} 2a H chain near the promoter. Chromatin preparations were immunoprecipitated with Abs against the indicated factors. The resulting DNA was amplified in a QPCR with the primers for the 5' end of the Ig {gamma} 2a H chain transcript, designated TATA on Fig. 1 and in Table I. {square}, Signals from A20 nuclei; Figure 6, signals from AxJ nuclei. NRS is normal rabbit antisera. Samples were normalized to the IP and QPCR of actin with RNAP-II in each cell type.

 
Ser-2 phosphorylation of CTD at the 5' end of Ig is important for elongation

The drug DRB slows RNAP-II elongation by interfering with pTEFb in its phosphorylation of the Ser-2 of CTD, as recently reviewed (44). We wondered whether blocking Ser-2 phosphorylation of CTD, which occurs near the 5' end of the IgH gene in plasma cells, would influence the association of the factors seen in Fig. 6. To that end, treatment of AxJ cells with DRB was followed by ChIP analyses and QPCR across the IgG2a H chain gene. As expected Ser-2 phosphorylation of the CTD was diminished relative to untreated cells (Fig. 7A). DRB treatment in AxJ cells stalls total RNAP-II at the 5'-most end of the H chain gene, blocking its advancement to the V:D:J2 region and beyond. This inability to advance is true not only for total polymerase but also for Ser-5 CTD-modified polymerase in DRB. The phosphorylation of the Ser-2 on CTD near the beginning of transcription, therefore, seems to play a crucial role in polymerase advancement down the IgH gene; blocking Ser-2 phosphorylation with DRB profoundly halts the downstream progress of the polymerase while blocking Ser-5 modifications as well. This result was a surprise considering the differential phosphorylation "gradients" established from yeast studies; but as previously discussed, these have yet to be firmly established with mammalian cells.


Figure 7
View larger version (28K):
[in this window]
[in a new window]

  		FIGURE 7. DRB stalls RNAP-II and CTD Ser-5 phosphorylated RNAP-II at the 5' end of the transcript and diminishes factor loading. A portion of the AxJ cells were treated with DRB for 2 h and chromatin was isolated from the treated and untreated cells as described in Materials and Methods. The chromatin was immunoprecipitated using Abs either the N terminus of RNAP-II (Pol II), Ser-5-phosphorylated CTD of RNAP-II (Ser-5), or Ser-2-phosphorylated CTD of RNAP-II (Ser-2) in A or to the indicated factors in B. The immunoprecipitated DNA was used in a QPCR analysis with the primers TATA, V:J (V:D:J), or CH1 from Fig. 1 and Table I. {blacksquare}, The data with the TATA primers; {square}, with the V:J (V:D:J) primers; and Figure 7, with CH1 primers.

 
We next looked at the effects of DRB inhibition of the binding of CstF-64, ELL2, and PC4 on the IgH gene in AxJ cells. First, we showed that these three factors have an IP profile similar to RNAP-II without DRB (Fig. 7B). Next, by blocking Ser-2 and, by extension, Ser-5 modifications of the RNAP-II, we severely reduced the binding of CstF-64 and ELL2 and decreased the binding of PC4 at the 5' end of the IgH transcription unit. For CstF-64 and ELL2, binding was completely abolished by V:D:J2 with DRB while PC4 seemed to be able to travel a bit downstream. We conclude that the binding of these factors requires RNAP-II to be phosphorylated at the CTD Ser-2 and perhaps Ser-5 and they most likely track down the gene with the polymerase.

RNAP-II elongates IgH mRNA

We have seen a decrease in polymerase downstream of the promoter/5' end by ChIP in AxJ plasma cells with at least three different Abs: anti-phospho-Ser-5, -Ser-2, anti-N terminus RNAP-II (and CTD-4H8, data not shown); yet, there are no differences between B cells and plasma cells in chromatin, as indicated by the acetylated histone data in Fig. 5C. Therefore, differential polymerase speed/elongation rate may contribute to the apparent unequal pattern of polymerase seen in Fig. 4, assuming that the polymerase transcribes to the end of the gene and does not fall off. Polymerase is known to go slowly at the 5' end during an initiation stage; hence, it would appear to accumulate there. After converting to an elongation phase where it is more processive, RNAP-II may then appear as less densely loaded on DNA downstream, as it passes regions more distant from the promoter. The regions of lowest apparent RNAP-II loading on the IgH gene correspond with regions of highest histone acetylation such as at the 3' end of the enhancer region; a sparse loading pattern may be a reflection of more rapid RNAP-II transit.

We had previously shown that polymerase transits the entire IgG H chain gene in B cells and plasma cells by transcription run-on experiments which measure polymerase synthesis across a region of the gene (29, 45). Therefore, fall-off of polymerase in the absence of DRB seemed an unlikely explanation for the decreased loading of polymerase at the 3' end of the IgG gene in AxJ cells. We repeated those run-on experiments with A20 and AxJ nuclei, and as shown in Fig. 8A, the incorporation of nucleotides by RNAP-II does not decrease across the V to CH3 interval in either cell type. In fact, there is slightly increased incorporation at CH3 in a fixed period of time, potentially consistent with higher polymerase speed through the region. Previous data on equal polymerase transcription of the regions between CH1 and the M2 region showing no RNAP-II signal fall-off in run-ons were extensive (29) and not repeated here.


Figure 8
View larger version (34K):
[in this window]
[in a new window]

  		FIGURE 8. RNA is elongated indicating no fall-off of RNAP-II. A, Transcription run-on assay. Nuclei from the indicated cells were isolated and incubated in vitro with [32P]UTP. The [32P]RNA made was hybridized with unlabeled ssRNA immobilized on a filter. After washing extensively, the filters were exposed to a phosphor screen which was analyzed in the phosphoimager. Intensity signal of the hybridized RNA was normalized for size of immobilized unlabeled RNA and normalized between cells to the actin signal, processed in the same way. {square}, The signal from A20 nuclei; Figure 8, the signal from AxJ cells. The plasmids used to transcribe the immobilized RNA on the filters are indicated in Materials and Methods. See Fig. 1 for areas of the IgH gene; rev is a sense strand used as loading control. B, Elongation assay with QPCR of nuclear RNA. Steady-state RNA was isolated from the nuclei of the indicated cells. A RT reaction was conducted using random hexamer nucleotide primers. The RT product was then analyzed by QPCR using the primers corresponding to the indicated regions of the IgH transcript. QPCR results are presented as 2 raised to the power of the negative {Delta}Ct of the A20 or AxJ sample relative to 30 ng of genomic DNA run on the same plate x100 to produce the percentage value shown in the figure. Samples were normalized to an actin control, see Materials and Methods. Primers span intron-exon boundaries. {square}, The signal from A20 nuclei; Figure 8, the signal from AxJ nuclei. The primer pairs used for the QPCR are described in Materials and Methods.

 
Looked at another way, if polymerase were falling off the gene downstream of the promoter to account for this 5' bias in Ser-phosphorylated RNAP-II distribution in AxJ cells, there should be more nuclear RNA corresponding to the 5' end of the gene as compared with the 3' end of the gene. We conducted an experiment for elongation based on QPCR quantification with nascent unspliced RNA in both cell types, an assay used by others to assess elongation (22); this technique assumes that the rates of degradation are equivalent from the 5' only vs the full-length RNAs. We used nuclear RNA that was primed with random oligonucleotides for the RT reaction and performed real-time PCR on regions of the transcript. These PCR probes spanned exon-intron boundaries and therefore would not detect fully spliced mature mRNA; reactions without RT were performed as a control for possible genomic DNA contamination. As shown in Fig. 8B within the AxJ cell samples, there is equal product for regions corresponding to the V and the Ig {gamma} 2a hinge region (see Fig. 1 for gene map); meanwhile, there is diminished product corresponding to the M1 membrane region, perhaps because this RNA is being turned over rapidly in AxJ cells. A similar pattern of transcription, with only a lower level of RNA product, as revealed when standardized to actin, is seen in A20 cells. This RNA encompasses the whole gene, but if it is not processed appropriately it may not produce mature mRNA and may persist in the nucleus. We had previously shown that A20 cells retain excess pre-mRNA while AxJ cells rapidly export that RNA to the cytoplasm (28). Taken together, the data obtained from the elongation assay and the transcription run-ons suggest that the bulk of the polymerases that are proceeding down the gene are holding on, making RNA, and not falling off prematurely. Therefore, we conclude that the decreased ChIP signal with Ser-phosphorylated RNAP-II seen downstream relative to the 5' end in plasma cells is most likely the result of unequal distribution of the modified polymerases, perhaps due to their slow progress at the 5' end followed by increased speed in the later half of the gene. The polymerase in AxJ cells is phosphorylated at Ser-5 and Ser-2 and thus capable of delivering factors to sites in the nascent RNA it is transcribing.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
The build-up of RNAP-II at the 5' end of the IgH gene in AxJ/plasma cells that we see here spans ~1 kb, from the promoter to the last JH region, curiously the same region subject to somatic hypermutation. These cells are actively transcribing the blimp-1 gene, a major characteristic of plasma cell expression. One could interpret the build-up of polymerase at the 5' end of IgH in plasma cells as making RNAP-II more susceptible to additions and modifications like the RNA-processing factors we have described here. Removing any hypothetical contributions by the unrearranged allele to the transcription/ChIP experiments would only serve to increase the real differences between the 5' end of the gene and the C region. Hypophosphorylated RNAP-II has been shown to be recruited to promoters to initiate transcription while the hyperphosphorylated form was actively engaged in mRNA elongation (reviewed in Ref. 23). Actively transcribed yeast genes exhibit high CTD Ser-5 phosphorylation of RNAP-II at the promoter that decreases into the coding regions and low Ser-2 phosphorylation at the promoter that increases downstream (18). By analogy with the yeast studies, we see RNAP-II with CTD phosphorylated at Ser-5 at the 5' end of the Ig {gamma} H chain gene; yet, some of the aspects of the results we have reported here are unlike the phosphorylation dynamics observed in yeast because Ser-2 is heavily phosphorylated near the beginning of Ig gene transcription. Yeast genes differ from mammalian genes in that they are generally shorter and do not require elongation factors. Other exceptions to the yeast Ser-5 at the 5' end/Ser-2 nearer 3' end rule are emerging, for it has been shown that while Ser-2 phosphorylation of CTD occurs near the end of p21 it is not required for efficient 3' processing of the p21 mRNA in the presence of DRB (44). HSV actively eliminates phosphorylated Ser-2 on CTD yet the HSV pre-mRNAs are still polyadenylated efficiently (46). The RNAP-II on genes that are paused before activation, like c-fos and MPK-1, shows increased Ser-2 phosphorylation of the CTD with increased transcription that is coincident with pTEFb binding and efficient elongation (47). It was recently reported that both Ser-5 and Ser-2 phosphorylation of CTD could release RNAP-II from the mediator complex (48) lending credence to the potential significance of the early Ser-2 phosphorylation we see on the IgH gene. The build-up of Ser-2 and Ser-5 on the RNAP-II at the 5' end of IgH gene that we see here appears to be plasma cell type and Ig gene specific in that both Ser-2 and Ser-5 phosphorylation are similar on the GAPDH gene in the two cell lines yet the profiles on the 5' end of the IgH gene differ between them.

We show that DRB inhibits Ser-2 phosphorylation and causes the unmodified RNAP-II on the IgH in AxJ cells to stall near the 5' end of the Ig gene. Inhibition of Ser-2 phosphorylation by DRB diminishes the signals obtained with anti-Ser-5 CTD Ab immediately downstream of the 5' end, indicating a role for Ser-2 phosphorylation in allowing RNAP-II to advance on the IgH gene. DRB also causes the decreased association of the polyadenylation factor CstF-64, suggesting that its binding to RNAP-II may require the Ser-5 or Ser-2 modifications. The increased CTD phosphorylation, polyadenylation factor, and elongation factor association at the 5' end in plasma cells vs B cells are consistent with an RNAP-II on the IgH gene in plasma cells that is primed for recognition of the proximal (sec) poly(A) site. In contrast, in the A20 B cells, the polymerase on the IgH gene is much less heavily modified with either type of Ser phosphorylation, polyadenylation factors, ELL2 or PC4. These polymerases in A20 seem more likely to pause just after the sec poly(A) site/splice donor site for the alternative CH3 to M1 splice. Splicing factors are thought to associate with the polymerase downstream from the 5' end, not exactly at it, although this is still being investigated.

In view of the fact that the level of sec H chain mRNA increases to at least 20:1 relative to the membrane-specific form in plasma cells, a 2- to 5-fold increase in overall transcription is not enough to solely account for the shift. H chain mRNA has been shown to increase at least 2-fold in half-life in plasma cells but because this applies to both sec and mb species, it alone cannot account for the shifting sec:mb ratio (1). But the loading of polyadenylation factors such as CPSF-160 and CstF-64 could shift the balance to the proximal poly(A) site use. The novel observation here is of the increased loading of the elongation/coactivators such as ELL2 and PC4 at the actively transcribing promoter. How they directly influence pre-mRNA processing is thus far unknown.

Using radioactive nucleotide incorporation and filter hybridization, transcription of the Ig locus in plasma cells had previously been reported to be ~2- to 5-fold increased in plasma cells over B cells (49). We have used several different techniques to approach this same issue. There are ~50% more RNAP-II molecules on the {gamma} 2a H chain in AxJ cells while there are ~6- to 7-fold more highly Ser-phosphorylated polymerases at the start of the transcript. Using the transcription run-on experiment, we see ~4- to 6-fold more transcription in AxJ cells. With the elongation assay, we see both that the polymerase is not falling off the gene and an ~2- to 3-fold increase in pre-mRNA in AxJ vs A20 cells. The elongation assay suffers from the fact that the relative stability of the primary transcript between the cell types may be an issue, and membrane-specific transcript may be low in AxJ cells because of higher instability of the cleaved-off fragment. Nonetheless, by all these measures there is consistently more transcription from the H chain locus in the AxJ cells. Therefore, the factor-laden and CTD Ser-phosphorylation-rich polymerases on the IgH gene in AxJ are more efficient at generating authentic RNA products and using the sec poly(A) site.

We see a stacking up of Ser-5- and Ser-2-modified RNAP-II at the promoter through the JH region followed by decreased signals for polymerases in AxJ cells through the internal H chain enhancer region. Based on our transcription run-on and elongation data, this does not appear to be because of a fall-off in transcription as has been determined previously in actin and GAPDH (35). If the transcriptional machinery were traveling through the enhancer at higher speeds than at the upstream promoter and V region, there would be less RNAP-II at steady state at the enhancer to be captured by ChIP. The increase in ELL2 elongation factor association with the IgH gene 5' end in AxJ plasma cells over A20 B cells supports this hypothesis and indicates a potential role for elongation rate differences.

In our hands, the accessibility/histone acetylation of the H chain locus chromatin is not significantly different between the two cell types. Activation to Ig secretion from the native Ig gene was previously shown in two cases to be stimulated by inhibitors of histone deacetylation: 1) a knockout of histone deacetylase-2, but not other histone deacetylases, in chicken DT-40 cells increased secretion (8) and 2) trichostatin A treatment (an inhibitor of deacetylation) of mammalian B cells partially activated the plasma cell program of gene expression (17). It is possible that in each of these experiments, other genes beside the IgH gene were being influenced by the changes in acetylation; this was not tested. It is possible that blimp-1 expression or its downstream targets might have been altered in those studies with inhibitors of histone deacetylation.

An up-regulation of transcription of the IgH gene may be accompanied by an increase in variety of transcription factors binding specifically at the promoter and enhancer regions (32). By as yet unknown mechanisms, these may facilitate Ser-5 and Ser-2 phosphorylations of the RNAP-II at the 5' end of the gene. Yet, these lymphoid factors would not be expected to be expressed in the large variety of cell lines used in the studies with the SV-40-driven Ig constructs that show high sec-mRNA production (16). Therefore, it may be the very act of active transcription, and RNAP-II pile-up, accompanied by increased CTD phosphorylation, rather than any specific transcription factor per se, that drives the use of the first poly(A) site in the IgH gene. This remains to be explored. A number of the cell lines used for the SV-40-driven Ig transfections are tumor lines which may express high levels of ELL2. It could also be that polymerase pausing at the 5' end of the transcript, phosphorylation of Ser-5 and Ser-2, and then efficient elongating of transcription are, themselves, coupled to increased factor association and efficient processing of the first poly(A) site encountered. A prediction from our observations would be that weak poly(A) sites which reside in the promoter proximal position in other genes will be activated with increased rates of transcription, unless some other blocking mechanism prevails.


   Acknowledgments
 
We thank Dr. B. Nikolajczyk for critically reading the manuscript and Dr. L. Borghesi for the rag–/– DNA and helpful suggestions.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflict of interest.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported by National Institutes of Health Grant CA86433 (to C.M.) and Training Grant T32 CA82084 (to S.A.S.). Back

2 Current address: Targeted Molecular Diagnostics, 610 Oakmont Lane, Westmont, IL 60559. Back

3 Address correspondence and reprint requests to Dr. Christine Milcarek, Department of Immunology, University of Pittsburgh, E1054 Biomedical Science Tower, Pittsburgh, PA 15261. E-mail address: milcarek{at}pitt.edu Back

4 Abbreviations used in this paper: blimp-1, B lymphocyte induction of maturation protein 1; IgH, Ig H chain; sec, secretory specific; mb, membrane bound; hnRNP, heterogeneous nuclear ribonucleoprotein; CTD, C-terminal domain; RNAP-II, RNA polymerase II; IP, immunoprecipitation; ChIP, chromatin IP; GSP, gene-specific antisense primer; QPCR, quantitative PCR; Ct, cycle threshold; RT, reverse transcription; EH, enhancer region; DRB, 5,6-dichlorobenzimidazole riboside; Tm, melting temperature; IVS, intervening sequence. Back

Received for publication August 9, 2007. Accepted for publication September 20, 2007.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Edwalds-Gilbert, G., K. Veraldi, C. Milcarek. 1997. Alternative poly(A) site selection in complex transcription units: means to an end?. Nucleic Acids Res. 25: 2547-2561. [Abstract/Free Full Text]
  2. Matis, S. A., K. Martincic, C. Milcarek. 1996. B-lineage regulated polyadenylation occurs on weak poly(A) sites regardless of sequence composition at the cleavage and downstream regions. Nucleic Acids Res. 24: 4684-4692. [Abstract/Free Full Text]
  3. Peterson, M. L., R. P. Perry. 1989. The regulated production of mu-m and mu-s mRNA is dependent on the relative efficiencies of mu-s poly(A) site usage and the Cmu4 to M1 splice. Mol. Cell. Biol. 9: 726-738. [Abstract/Free Full Text]
  4. Takagaki, Y., J. Manley. 1998. Levels of polyadenylation factor CstF-64 control IgM heavy chain mRNA accumulation and other events associated with B cell differentiation. Mol. Cell 2: 761-771. [Medline]
  5. Sciammas, R., M. M. Davis. 2004. Modular nature of Blimp-1 in the regulation of gene expression during B cell maturation. J. Immunol. 172: 5427-5440. [Abstract/Free Full Text]
  6. Underhill, G. H., D. George, E. G. Bremer, G. S. Kansas. 2003. Gene expression profiling reveals a highly specialized genetic program of plasma cells. Blood 101: 4013-4021. [Abstract/Free Full Text]
  7. Shaffer, A. L., M. Shapiro-Shelef, N. N. Iwakoshi, A.-H. Lee, S.-B. Qian, H. Zhao, X. Yu, L. Yang, B. K. Tan, A. Rosenwald, et al 2004. XBP1, downstream of Blimp-1, expands the secretory apparatus and other organelles, and increases protein synthesis in plasma cell differentiation. Immunity 21: 81-93. [Medline]
  8. Takami, Y., H. Kikuchi, T. Nakayama. 1999. Chicken histone deacetylase-2 controls the amount of the IgM H-chain at the steps of both transcription of its gene and alternative processing of its pre-mRNA in the DT40 cell line. J. Biol. Chem. 34: 23977-23990.
  9. Martincic, K., R. Campbell, G. Edwalds-Gilbert, L. Souan, M. Lotze, C. Milcarek. 1998. Increase in the 64-kDa subunit of the polyadenylation/cleavage stimulatory factor during the Go to S phase transition. Proc. Natl. Acad. Sci. USA 95: 11095-11100. [Abstract/Free Full Text]
  10. Phillips, C., S. Jung, S. I. Gunderson. 2001. Regulation of nuclear poly(A) addition controls the expression of immunoglobulin M secretory mRNA. EMBO J. 20: 6443-6452. [Medline]
  11. Ma, J., S. I. Gunderson, C. Phillips. 2006. Non-snRNP U1A levels decrease during mammalian B-cell differentiation and release the IgM secretory poly(A) site from repression. RNA 12: 122-132. [Abstract/Free Full Text]
  12. Milcarek, C., K. Martincic, L.-H. Chung-Ganster, C. S. Lutz. 2003. The snRNP-associated U1A levels change following IL-6 stimulation of human B-cells. Mol. Immunol. 39: 809-814. [Medline]
  13. Peterson, M. L., G. L. Bingham, C. Cowan. 2006. Multiple features contribute to the use of the immunoglobulin M secretion-specific poly(A) signal but are not required for developmental regulation. Mol. Cell. Biol. 26: 6762-6771. [Abstract/Free Full Text]
  14. Veraldi, K. L., G. K. Arhin, K. Martincic, L. Chung-Ganster, J. Wilusz, C. Milcarek. 2001. hnRNP F influences binding of a 64-kilodalton subunit of cleavage stimulation factor to mRNA precursors in mouse B-cells. Mol. Cell. Biol. 21: 1228-1238. [Abstract/Free Full Text]
  15. Bruce, S. R., R. W. C. Dingle, M. L. Peterson. 2003. B-cell and plasma-cell splicing differences: a potential role in regulated immunoglobulin RNA processing. RNA 9: 1264-1273. [Abstract/Free Full Text]
  16. Peterson, M.. 1994. Regulated immunoglobulin (Ig) RNA processing does not require specific cis-acting sequences: non-Ig RNA can be alternatively processed in B cells and plasma cells. Mol. Cell. Biol. 14: 7891-7898. [Abstract/Free Full Text]
  17. Lee, S. C., A. Bottaro, R. A. Insel. 2003. Activation of terminal B cell differentiation by inhibition of histone deacetylation. Mol. Immunol. 39: 923-932. [Medline]
  18. Komarnitsky, P., E.-J. Cho, S. Buratowski. 2000. Different phosphorylated forms of RNA polymerase II and associated mRNA processing factors during transcription. Genes Dev. 14: 2452-2460. [Abstract/Free Full Text]
  19. Bentley, D. L.. 2005. Rules of engagement: co-transcriptional recruitment of pre-mRNA processing factors. Curr. Opin. Cell Biol. 17: 251-256. [Medline]
  20. Bentley, D.. 2002. The mRNA assembly line: transcription and processing machines in the same factory. Curr. Opin. Cell Biol. 14: 336-342. [Medline]
  21. Meinhart, A., T. Kamenski, S. Hoeppner, S. Baumli, P. Cramer. 2005. A structural perspective of CTD function. Genes Dev. 19: 1401-1415. [Abstract/Free Full Text]
  22. Kornblihtt, A. R., M. De La Mata, J.-P. Fededa, M. J. Munoz, G. Nogues. 2004. Multiple links between transcription and splicing. RNA 10: 1489-1498. [Abstract/Free Full Text]
  23. Hirose, Y., Y. Ohkuma. 2007. Phosphorylation of the C-terminal domain of RNA polymerase II plays central roles in the integrated events of eukaryotic gene expression. J. Biochem. 141: 601-608. [Abstract/Free Full Text]
  24. Auboeuf, D., D. H. Dowhan, M. Dutertre, N. Martin, S. M. Berget, B. W. O’Malley. 2005. A subset of nuclear receptor coregulators act as coupling proteins during synthesis and maturation of RNA transcripts. Mol. Cell. Biol. 25: 5307-5316. [Free Full Text]
  25. Shilatifard, A., D. R. Duan, D. Haque, C. Florence, W. H. Schubach, J. W. Conaway, R. C. Conaway. 1997. ELL2, a new member of an ELL family of RNA polymerase II elongation factors. Proc. Natl. Acad. Sci. USA 94: 3639-3643. [Abstract/Free Full Text]
  26. Ge, H., R. G. Roeder. 1994. Purification, cloning, and characterization of a human coactivator, PC4, that mediates transcriptional activation of class II genes. Cell 78: 513-523. [Medline]
  27. Schang, L. M., G.-J. H. Hwang, B. D. Dynlacht, D. W. Speicher, A. Bantly, P. A. Schaffer, A. Shilatifard, H. Ge, R. Shiekhattar. 2000. Human PC4 is a substrate-specific inhibitor of RNA polymerase II phosphorylation. J. Biol. Chem. 275: 6071-6074. [Abstract/Free Full Text]
  28. Milcarek, C., M. Hartman, S. Croll. 1996. Changes in abundance of IgG 2a mRNA in the nucleus and cytoplasm of a murine B-lymphoma before and after fusion to a myeloma cell. Mol. Immunol. 33: 691-701. [Medline]
  29. Flaspohler, J. A., C. Milcarek. 1990. Myelomas and lymphomas expressing the IgG2a H chain gene have similar transcription termination regions. J. Immunol. 144: 2802-2810. [Abstract]
  30. Doenecke, A., E. L. Winnacker, M. Halleck. 1997. Rapid amplification of cDNA ends (RACE) improves the PCR-based isolation of immunoglobulin variable region genes from murine and human lymphoma cells and cell lines. Leukemia 11: 1787-1792. [Medline]
  31. Johnston, C. M., A. L. Wood, D. J. Bolland, A. E. Corcoran. 2006. Complete sequence assembly and characterization of the C57BL/6 mouse Ig heavy chain V region. J. Immunol. 176: 4221-4234. [Abstract/Free Full Text]
  32. Eaton, S., K. Calame. 1987. Multiple DNA sequence elements are necessary for the function of an immunoglobulin heavy chain promoter. Proc. Natl. Acad. Sci. USA 84: 7634-7638. [Abstract/Free Full Text]
  33. Turner, C., D. Mack, M. Davis. 1994. Blimp-1, a novel zinc finger-containing protein that can drive the maturation of B lymphocytes into immunoglobulin-secreting cells. Cell 77: 297-306. [Medline]
  34. Shapiro-Shelef, M., K.-I. Lin, L. J. McHeyzer-Williams, J. Liao, M. G. McHeyzer-Williams, K. Calame. 2003. Blimp-1 is required for the formation of immunoglobulin secreting plasma cells and pre-plasma memory B cells. Immunity 19: 607-620. [Medline]
  35. Cheng, C., P. A. Sharp. 2003. RNA polymerase II accumulation in the promoter-proximal region of the dihydrofolate reductase and {gamma}-actin genes. Mol. Cell. Biol. 23: 1961-1967. [Abstract/Free Full Text]
  36. Orozco, I. J., S. J. Kim, H. G. Martinson. 2002. The poly(A) signal, without the assistance of any downstream element, directs RNA polymerase II to pause in vivo and then to release stochastically from the template. J. Biol. Chem. 277: 42899-42911. [Abstract/Free Full Text]
  37. Tucker, P. W., K. B. Marcu, N. Newell, J. Richards, F. R. Blattner. 1979. Sequence of the cloned gene for the constant region of murine g2b immunoglobulin heavy chain. Science 206: 1303-1306. [Abstract/Free Full Text]
  38. Monk, R. J., S. L. Morrison, C. Milcarek. 1981. Heavy-chain mutants derived from {gamma}-2b mouse myeloma: characterization of heavy-chain messenger ribonucleic acids, proteins, and secretion in deletion mutants and messenger RNA in {gamma}-2a mutant progeny. Biochemistry 20: 2330-2339. [Medline]
  39. Boeger, H., J. Griesenbeck, J. S. Strattan, R. D. Kornberg. 2003. Nucleosomes unfold completely at a transcriptionally active promoter. Mol. Cell 11: 1587-1598. [Medline]
  40. Cockerill, P., M. Yuen, W. Garrard. 1987. The enhancer of the immunoglobulin heavy chain locus is flanked by presumptive chromosomal loop anchorage elements. J. Biol. Chem. 262: 5394-5397. [Abstract/Free Full Text]
  41. Dantonel, J.-C., K. Murthy, J. Manley, L. Tora. 1997. Transcription factor TFIID recruits factor CPSF for formation of 3' end of mRNA. Nature 389: 399-402. [Medline]
  42. Luo, Y., H. Ge, S. Stevens, H. Xiao, R. G. Roeder. 1998. Coactivation by OCA-B: definition of critical regions and synergism with general cofactors. Mol. Cell. Biol. 18: 3803-3810. [Abstract/Free Full Text]
  43. Calvo, O., J. L. Manley. 2001. Evolutionarily conserved interaction between CstF-64 and PC4 links transcription polyadenylation, and termination. Mol. Cell. 7: 1013-1023. [Medline]
  44. Gomes, N. P., G. Bjerke, B. Llorente, S. A. Szostek, B. M. Emerson, J. M. Espinosa. 2006. Gene-specific requirement for p-TEFb activity and RNA polymerase II phosphorylation within the p53 transcriptional program. Genes Dev. 20: 601-612. [Abstract/Free Full Text]
  45. Flaspohler, J. A., D. Boczkowski, B. L. Hall, C. Milcarek. 1995. The 3'-untranslated region of membrane exon 2 from the {gamma} 2a immunoglobulin gene contributes to efficient transcription termination. J. Biol. Chem. 270: 11903-11911. [Abstract/Free Full Text]
  46. Fraser, K. A., S. A. Rice. 2007. Herpes simplex virus immediate-early protein ICP22 triggers loss of serine 2-phosphorylated RNA polymerase II. J. Virol. 81: 5091-5101. [Abstract/Free Full Text]
  47. Fujita, T., S. Ryser, S. Tortola, I. Piuz, W. Schlegel. 2006. Gene-specific recruitment of positive and negative elongation factors during stimulated transcription of the MPK-1 gene in neuroendocrine cells. Nucleic Acids Res. 35: 1007-1017.
  48. Max, T., M. Sogaard, J. Q. Svejstrup. 2007. Hyperphosphorylation of the C-terminal repeat domain of RNA polymerase II facilitates dissociation of its complex with mediator. J. Biol. Chem. 282: 14113-14120. [Abstract/Free Full Text]
  49. Kelly, D. E., R. P. Perry. 1986. Transcriptional and post-transcriptional control of Ig mRNA production during B lymphocyte development. Nucleic Acids Res. 14: 5431-5441. [Abstract/Free Full Text]

Related articles in The JI:

IN THIS ISSUE

The JI 2007 179: 7191-7192. [Full Text]  



This article has been cited by other articles:


Home page
 BloodHome page
R. Pal, M. Janz, D. L. Galson, M. Gries, S. Li, K. Johrens, I. Anagnostopoulos, B. Dorken, M. Y. Mapara, L. Borghesi, et al.
C/EBP{beta} regulates transcription factors critical for proliferation and survival of multiple myeloma cells
Blood, October 29, 2009; 114(18): 3890 - 3898.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Related articles in The JI
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Shell, S. A.
Right arrow Articles by Milcarek, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Shell, S. A.
Right arrow Articles by Milcarek, C.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS