HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Cakouros, D. Articles by Shannon, M. F. Search for Related Content PubMed PubMed Citation Articles by Cakouros, D. Articles by Shannon, M. F. Pubmed/NCBI databases Substance via MeSH

A NF-B/Sp1 Region Is Essential for Chromatin Remodeling and Correct Transcription of a Human Granulocyte- Macrophage Colony-Stimulating Factor Transgene

Dimitrios Cakouros^*, Peter N. Cockerill^3,, Andrew G. Bert, Renu Mital^*, Donna C. Roberts and M. Frances Shannon^1,*

^* Division of Biochemistry and Molecular Biology, John Curtin School of Medical Research, Australian National University, Canberra, Australia; and Division of Human Immunology, Hanson Centre for Cancer Research, Adelaide, South Australia

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
The GM-CSF gene is expressed following activation of T cells. The proximal promoter and an upstream enhancer have previously been characterized using transfection and reporter assays in T cell lines in culture. A 10.5-kb transgene containing the entire human GM-CSF gene has also been shown to display inducible, position-independent, copy number-dependent transcription in mouse splenocytes. To determine the role of individual promoter elements in transgene function, mutations were introduced into the proximal promoter and activity assessed following the generation of transgenic mice. Of four mutations introduced into the transgene promoter, only one, in an NF-B/Sp1 region, led to decreased induction of the transgene in splenocytes or bone marrow-derived macrophages. This mutation also affected the activity of reporter gene constructs stably transfected into T cell lines in culture, but not when transiently transfected into the same cell lines. The mutation alters the NF-B family members that bind to the NF-B site as well as reducing the binding of Sp1 to an adjacent element. A DNase I hypersensitive site that is normally generated at the promoter following T cell activation on the wild-type transgene does not appear in the mutant transgene. These results suggest that the NF-B/Sp1 region plays a critical role in chromatin remodeling and transcription on the GM-CSF promoter in primary T cells.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Activation of T cells in response to foreign Ag is an important control point in an immune response. T cells respond to antigenic peptides presented by MHC molecules on the surface of APCs, together with costimulatory signals provided by the APCs. T cells respond by producing autostimulatory cytokines such as IL-2 as well as many cytokines that influence other arms of the immune system. The types of cytokines produced depend on the nature of the initial Ag and the differentiation of naive T cells into specific effector cell types (1, 2).

Expression of cytokine genes is inducible and is controlled to a large extent at the level of gene transcription. There have been extensive studies documenting the regions of these genes and the transcription factors that are important for response to signal transduction pathways in T cells (3, 4, 5, 6, 7, 8). A general consensus has emerged that groups of transcription factors form activation structures on inducible gene promoter/enhancers that have been dubbed enhanceosomes (9, 10). These enhanceosomes appear to form recruiting surfaces for coactivators of transcription or for components of the basal transcriptional machinery (11, 12, 13, 14).

GM-CSF is produced by T cells following activation, and most myeloid lineage cells (reviewed in Refs. 15, 16). It is also produced at sites of inflammation by cells, such as endothelial or epithelial cells (15, 16). GM-CSF has an important role in the production and function of granulocytes and macrophages, and also influences dendritic cell function (16, 17). The first 120 bp upstream from the start of transcription of the GM-CSF gene contain elements that are required for a fully inducible proximal promoter (reviewed in Ref. 6). This region responds to T cell activation and to costimulatory signals such as CD28 activation (6). Among the transcription factors that bind to and activate this promoter region are the inducible NF-B, AP-1, Ets, and NF-AT family members, as well as constitutive proteins such as Sp1 and CBF (reviewed in Ref. 6). Some of these proteins have been shown to act in a cooperative manner and form a functional transcription unit on the promoter region (6, 18, 19, 20, 21, 22, 23). An enhancer located 3 kb upstream from the transcription start site also plays an important role in response to T cell signals and contains a number of AP-1/NF-AT composite sites (24, 25).

One important region of the proximal promoter, known as the CD28 response region (CD28RR),³ is located from -102 to -69 and consists of two NF-B-like sites and an Sp1 site (Fig. 1a). The distal NF-B binding site, known as the CK-1 element, is required for CD28 responsiveness (26, 27). The CK-1 element binds c-Rel-containing complexes as well as RelA homodimers following T cell activation (26, 27). This element cannot, however, function alone, but requires the adjacent NF-B and Sp1 sites (26, 27, 28). The proximal NF-B site binds a classical RelA/p50 complex following activation and is responsive to TCR signals (26, 27, 29). The entire CD28RR appears to act as a unit and may form an enhanceosome-type structure in response to T cell activation.

View larger version (21K): [in this window] [in a new window]   FIGURE 1. a, Schematic representation of the GM-CSF gene showing the position of the CD28RR (-103 to -66) in the proximal promoter. The sequence of the CD28RR in the gene is shown with the two NF-B-like sites (CK-1 and NF-B) and the overlapping Sp1 site marked. The bases that were mutated to generate pGMCK1 M, pGMBM, pGM2CK1, and pGM2B are underlined and shown in bold. b, Effect of mutating the CK-1 or NF-B sites in the 10.5-kb human GM-CSF transgenic construct, pBSXH10.5. Splenocytes from wt or mutant transgenic mice were either left unstimulated or were stimulated for 16 h with PMA/I (20 ng/ml and 1 µM, respectively) and ELISA assays for both human and mouse GM-CSF conducted on the supernatants. The data are expressed as the amount of human GM-CSF produced from the transgene relative to the amount of mouse GM-CSF, corrected for copy number (H/M/C). The value for mouse GM-CSF in each line was normalized to 1. Between four and seven mice were analyzed in duplicate for each line, and the data represent the averages +/- SDs for all of the experiments. c, RT-PCR analysis was conducted on RNA prepared from splenic T cells prepared from wt or mutant transgenic mice. The cells were either unstimulated or stimulated in culture with PMA/I for 6 h. Both GM-CSF and GAPDH were amplified from each sample using the primers described. The gels were scanned using a Fujifilm LAS 3000 (Fuji, Tokyo, Japan). The GAPDH levels were used to normalize the samples and the relative levels of human GM-CSF plotted relative to GAPDH. A single experiment, which was replicated three times with identical results, is shown. d, Mutating the CK-1 or NF-B site in the human GM-CSF transgene has no effect on CD28 responses. Splenocytes from wt or mutant transgenic mice were incubated without stimulation or in plates coated with CD3 alone or CD3 + CD28 Abs for 16 h, and both human and mouse GM-CSF levels were measured in supernatants from these cells. At least four mice were analyzed for each line. The data are presented as fold induction of CD3 + CD28 activation over CD3 activation alone.

Recently, a 10.5-kb fragment of DNA spanning the human GM-CSF gene was introduced as a transgene into the mouse genome, and expression was monitored in splenocytes following in vitro activation with PMA and Ca²⁺ ionophore. The introduced transgenes were uniformly induced in splenocytes by TCR signaling pathways in a copy number-dependent manner at levels indistinguishable from the endogenous mouse GM-CSF gene (36). Deletion of the upstream enhancer led to a reduction in gene transcription to approximately 30% of wild-type (wt) levels (36). These experiments showed that it is possible to examine the contribution of individual control regions in a GM-CSF transgene that supports correctly regulated expression.

In this study, we have examined the role of individual proximal promoter elements in the regulation of the human GM-CSF transgene. Several individual mutations introduced into specific elements of the CD28RR did not reduce expression levels from the transgene in either splenocytes or bone marrow-derived macrophages. However, one specific mutation in the NF-B/Sp1 region dramatically reduced transgene expression. This mutation was only effective when the DNA was integrated into chromosomal DNA of either transgenic mice or stable cell lines in culture, and not in transiently transfected plasmids. The inducible DH site at the proximal promoter that was normally seen in the wt transgene was absent in the mutant transgene. Thus, we conclude that the NF-B/Sp1 region is important for chromatin remodeling associated with activation of the GM-CSF proximal promoter.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Construction of transgenic and luciferase plasmids and mutagenesis

A pGMSelect construct, containing a 4.5-kb BamHI-EcoRI GM-CSF gene fragment spanning the promoter, was used to mutate sequences in the GM-CSF promoter following the Altered Site II protocol (Promega, Madison, WI). The specific mutations that were generated are shown in Fig. 1a. A Blp1 2.15-kb promoter fragment from pGMSelect, containing the promoter mutations, was excised and cloned into the Blp1-digested wt transgenic construct pBSXH10.5 (36) to generate the mutated transgenic constructs. PCR was used to amplify the human GM-CSF promoter (-620 to +37) from the pGMSelect vector containing the promoter mutations. The amplified fragment for each mutation was cloned into the luciferase reporter vector pXp1 (37).

Generation and identification of transgenic mice

Transgenic mice were generated by microinjection of DNA into the pronuclei of fertilized mouse eggs, which then were cultured overnight to the two-cell stage and transferred into the oviducts of pseudopregnant females. Transgenic founders were identified and their copy numbers estimated by Southern blot analysis of transgenic mouse DNA. Founders were then bred to establish hemizygous transgenic lines used for GM-CSF expression analysis.

Cell culture

The Jurkat cell line, a human T-lymphoblastoid cell line (provided by W. Greene, Gladstone Institute of Virology and Immunology, San Francisco, CA), was grown as previously described (38).

Spleens were removed from 6-wk-old transgenic or normal C57BL/6 x CBA mice into 3 ml DMEM supplemented with 10% FCS (Life Technologies), 50 pM 2-ME (Sigma, St. Louis, MO), 1 mM L-glutamine, 100 u/ml penicillin, and 40 µg/ml gentamicin (Life Technologies, Gaithersburg, MD). Splenocytes were obtained by repetitively piercing the spleen with a scalpel and gently mashing through a metal sieve. Cells were then incubated in six-well tissue culture trays at 37°C for 1 h, where adherent cells were allowed to stick to the petri dish. Nonadherent cells were washed in PBS and resuspended in 5 ml 17 mM Tris-Cl, pH 7.65, 140 mM ammonium chloride, and allowed to stand for 2 min to lyse RBCs. The resultant splenocytes were cultured at 1 x 10⁷ cells/ml in DMEM. To obtain a more purified population of splenic T cells for EMSA and RT-PCR, B cells were removed by complement-mediated lysis. The T cell preparation was approximately 80% pure, as determined by FACS staining with an anti-CD3 Ab (PharMingen, San Diego, CA).

PMA (Boehringer Mannheim, Indianapolis, IN) and calcium ionophore (I; Sigma) were used to activate the Jurkat cells or the splenic cell preparations at concentrations of 20 ng/ml and 1 µM, respectively. An activating Ab to the human CD28 receptor (Bristol-Myers Squibb, New York, NY) was used at a 1/10,000 dilution of ascites fluid. Stimulating Ab to the murine CD3 chain (PharMingen) was used at 1 µg/ml to coat tissue culture wells before incubation of cells. Similarly, Ab to the murine CD28 receptor (from A. Strasser, The Walter and Eliza Hall Institute of Medical Research, Melbourne, Australia (clone 37NF1)) and used at 10 µg/ml to coat tissue culture wells.

Human and murine GM-CSF ELISA assays

The ELISA system used for the detection of both mouse and human GM-CSF was the matched pair Ab system from Endogen (Woburn, MA). The assay was conducted according to the manufacturer’s instructions. The concentration of Abs used was determined empirically. The coating Ab (M-500A-E) was used at 1 µg/ml overnight at 22°C. The biotin-labeled detecting Ab (M-501-B) was used at 0.5 µg/ml, and the HRP-conjugated streptavidin was used at 1/8000 dilution. Fifty microliters of sample or standards were used per well of a 96-well plate. The linear range of the assay was from 1000 to 60 pg/ml, and readings within this range were always used to determine GM-CSF concentrations.

rGM-CSF protein variant E21R (39) (kindly provided by T. Hercus, Hanson Center for Cancer Research, Adelaide, Australia) was used as the standard in the human GM-CSF ELISAs. This protein cross-reacts with the Abs used, as does the wt GM-CSF protein (39). The standard range used was from 1000 to 15 pg/ml, and dilutions were made in DMEM media supplemented as described previously.

For the detection of mouse GM-CSF, ELISAs were conducted as described above. The standard used for the murine ELISA was mouse rGM-CSF protein expressed in insect cells kindly provided by Dr. A. Hapel (Research School of Chemistry, Australian National University). The standard range used was from 1000 to 15 pg/ml. The biotin-labeled detecting Ab (MM-500D-B) was used at a concentration of 0.6 µg/ml in assay buffer. The HRP-conjugated streptavidin was used at 1/2000 in assay buffer (or block buffer).

RT-PCR analysis

Splenic T cells were isolated as described above and stimulated for 4 h with PMA/I before RNA isolation. RNA was extracted using the RNazol B method (40). Total RNA (1 µg) was used to make cDNA, using Superscript II (Life Technologies), and 2 µl cDNA was then used in PCR reactions. The primer sequences used were: 5'-GTAGAAGTCATCTCAGAAAT-3' and 5'-CTCATCTGGCCGGTCTCACT-3'. Taq DNA polymerase (0.2 µl) was used with 5 pmol of each primer. PCR cycle conditions were as follows: 94°C for 20 min, 1 cycle; 94°C for 30 s, 30 cycles; 55°C for 30 s; 72°C for 30 s; and 72°C for 5 min. PCR products were then resolved on a 1% TAE (40 mM Tris-acetate and 2 mM EDTA, pH 8.5) agarose gel.

Transfection of Jurkat T cells and luciferase assay

Transfection of Jurkat T cells was conducted by electroporation, as previously described (38). Cells were stimulated 24 h posttransfection for 9 h with PMA (20 ng/ml), I (10 µM), and activating CD28 Ab (1/10,000 dilution). Luciferase assays were conducted as previously described using 80–100 µg protein per assay (27).

Nuclear extract preparation

For Jurkat T cells, nuclei were prepared by a modified method of Schreiber et al. (41). Briefly 1–5 x 10⁷ Jurkat cells were collected by centrifugation at 1500 rpm for 5 min and washed once with ice-cold PBS. Cells were then resuspended in 800 µl ice-cold buffer A (10 mM HEPES, pH 7.9, 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM DTT, 0.5 mM PMSF, 5 µg/ml leupeptin, 5 µg/ml aprotinin) and put on ice for 15 min. Fifty microliters of 10% Nonidet P-40 were added and vortexed for precisely 30 s and then centrifuged at 13,000 rpm for 30 s. The nuclear pellet was then resuspended in 70 µl ice-cold buffer C (500 mM NaCl, 7.5 mM MgCl₂, 0.2 mM EDTA, 0.1 mM EGTA, 1 mM DTT, 0.5 mM PMSF, 100 µg aprotinin, and 50 µg leupeptin) and agitated on ice for 15 min. The nuclear pellet was then centrifuged at 13,000 rpm for 5 min, and the supernatant was snap frozen in liquid nitrogen and stored at -70^OC in 20-µl aliquots. Preparation of splenic T cell extracts was similar to the preparation of Jurkat T cell extracts, except that 25 µl 10% Nonidet P-40 was used in the lysis step and vortexed for 10 s.

Electrophoretic mobility shift assay

Oligonucleotides were synthesized by Geneworks (Adelaide, South Australia). Double-stranded radiolabeled oligonucleotide probes or competitors were prepared as previously described (42). For analysis of NF-B binding, 3–8 µg nuclear extract was combined with 0.2 ng radiolabeled probe in a final volume of 20 µl binding mix (5% Ficoll, 0.5 mM DTT, 25 mM Tris-HCl, 6.25 mM MgCl₂, 0.5 mM EDTA, 0.3 mM PMSF). One microgram poly(dI:dC) was included in the reaction and incubated at room temperature for 20 min. If Ab supershift experiments were to be performed, 1 µl Ab was preincubated with the nuclear extract for 5 min before the addition of probe. Abs used were rabbit c-Rel (SC-70X) polyclonal Ab, p65 (SC109X) polyclonal, and Sp1 (PEP2 SC 59X) polyclonal Ab from Santa Cruz. DNA:protein complexes were separated on a 5% acrylamide gel in 0.5x TGE (190 mM glycine, 25 mM Tris-HCl, pH 8.5, 0.2 mM EDTA) gel, as previously described. When using splenic T cell extracts, 10% BSA was added to the binding mix.

DH site analysis

DH sites were analyzed as previously described (24). Briefly, a DNase I titration was performed, and the samples that had optimum levels of DNase I digestion were selected for Southern blot analysis. The promoter and enhancer DH sites were mapped using a SacI/EcoRI fragment from the GM-CSF gene (Fig. 5).

View larger version (33K): [in this window] [in a new window]   FIGURE 5. DH site analysis on human GM-CSF transgenes in mice. Splenocytes were isolated from wt, 2CK1, 2B, CK1 M, and BM mice with the individual lines used indicated. Nuclei were prepared from Con A-stimulated splenocytes and digested with the concentrations of DNase I shown (0–24 µg/ml). Genomic DNA was prepared, digested with EcoRI, and separated by agarose gel electrophoresis. Following transfer to nitrocellulose, the blot was probed with a radiolabeled SacI/EcoRI fragment of the GM-CSF gene, as indicated. The arrows indicate the DH site that appears at the enhancer (E) and promoter (P) regions of the genes. The positions of the enhancer, the promoter, and the probe used are illustrated.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

We have previously shown that a 10.5-kb fragment of DNA spanning the human GM-CSF gene can confer correctly regulated expression in transgenic mouse splenocytes (36). We have now employed this model to analyze the role of individual transcription factor binding sites within the proximal promoter region of the GM-CSF gene. The CD28RR of the GM-CSF promoter, which consists of a CK-1 site (CD28 response element), a classical NF-B site (B), and an overlapping Sp1 site, was the initial target for mutagenesis. The CK-1 element binds RelA- and c-Rel-containing complexes, the NF-B site binds p50/RelA complexes, and the Sp1 site binds Sp1 and related proteins (26, 42, 43). We have previously analyzed mutations in the CD28RR in transient transfection assays in Jurkat T cells and found that a mutation in any one of these elements reduced promoter activity by 50–90% (20, 26, 38).

At least four independent lines of mice were analyzed for each mutant construct (Fig. 1a), except for the 2B construct, in which only two lines were analyzed. The copy number of the inserted transgene was determined for each line of mice to allow assessment of copy number dependence in expression levels. Copy number ranged from 3 to 20, with one exception (line 1236) at 39. Two wt lines, from the 11 lines previously characterized (36), were used as a comparison with the mutants. These lines were chosen because they possessed 3 and 10 copies of the transgene (36).

To analyze the expression of the human GM-CSF transgene, splenocytes were isolated from 6-wk-old transgenic mice, stimulated in culture with PMA/I or CD3 + CD28, and the supernatants assayed for both human and mouse GM-CSF using ELISA assays. The data are expressed as the level of human GM-CSF relative to mouse GM-CSF divided by the transgene copy number, unless otherwise stated.

Individual mutations that were introduced into the CK-1 (CK-1M) and the NF-B (BM) elements (Fig. 1a) were shown to abolish transcription factor binding to their respective sites using EMSA assays with extracts from splenocytes (data not shown). In the transgenic context, individual mutations in the CK-1 or the B elements did not reduce the level of human GM-CSF produced by PMA/I-stimulated splenocytes (Fig. 1b). GM-CSF was not detected in any of the unstimulated cells from these mice (data not shown). Previous results have shown that in 10 lines of mice containing the wt transgene, copy number-dependent, position-independent expression was observed (36). The introduction of the CK-1 or NF-B mutations appeared to ablate to some extent the position-independent expression of the GM-CSF transgene (Fig. 1b). For the wt lines, the values of human GM-CSF relative to mouse (set at a value of 1) ranged from 0.5 to 1.5, but in the CK-1 M or BM mice the levels of activity ranged from 0.5 to 7 relative to the mouse level of 1.

To confirm that the changes seen in the production of GM-CSF protein were a reflection of changes in mRNA levels, RT-PCR was performed on RNA isolated from splenic T cells stimulated in culture with PMA/I. In all samples, little or no GM-CSF mRNA was detected in unstimulated cells in accordance with the ELISA results. When cells were stimulated with PMA/I, GM-CSF mRNA was detected in cells from the wt, CK-1M, and BM mice (Fig. 1c). When GAPDH levels were used to normalize the results, little difference was observed between the wt and mutant mice. While this assay is not highly quantitative, it supports the quantitative data obtained from the GM-CSF ELISA assays.

Since the CK-1 element has previously been shown to determine the CD28 responsiveness of the promoter in transient transfection assays (26), the response of the wt and mutated transgenes to CD3/CD28 activation of splenic T cells was analyzed. Again individual mutations in the CK-1 or B elements had no effect on either the absolute levels of GM-CSF produced as measured by ELISA assays of CD3/CD28-stimulated splenocyte supernatants (data not shown) nor the fold induction seen with CD3 + CD28 costimulation compared with CD3 alone (Fig. 1d). An approximate 4-fold induction of CD3 + CD28 over CD3 alone was observed for both the wt and mutant lines (Fig. 1d), which is the approximate level previously observed in transient transfection assays.

These results were unexpected since previous analyses in transient transfections in Jurkat T cells had shown that these mutations lead to a significant reduction in promoter activity either in response to PMA/I or CD28 activation (21, 26, 38). The results imply that there is sufficient redundancy in the entire gene to offset the effect of mutation in individual transcription factor binding sites.

A mutation in the NF-B/Sp1 region abolishes transgene activity

As described above, the CD28RR contains two distinct NF-B-type sites. These elements bind different NF-B family members and appear to have distinct functions. To determine the importance of having two distinct NF-B sites in the CD28RR, the GM-CSF transgene was mutated to alter the CK-1 site to a classical NF-B site and the classical NF-B site to a CK-1 site, thus generating constructs that contained two CK-1 (2CK-1) or two NF-B (2B) sites. Surprisingly, we found that splenocytes from three of four 2CK-1 transgenic lines had greatly reduced ability to produce human GM-CSF in response to activation with PMA/I (Fig. 2a). The single 2CK-1 line that produced high levels of activity (0.5 relative to mouse) contained 39 copies of the transgene, whereas all the other lines had between 2 and 20 transgene copies. In contrast, splenocytes from mice containing the 2B construct produced levels of human GM-CSF that were significantly higher (5–8 times) than wt lines (Fig. 2a). The response of these transgenes to CD3 + CD28 activation was also monitored by stimulating mouse splenocytes with CD3 alone or CD3 + CD28 and measuring GM-CSF production by ELISA. Again, the 2CK-1 mutant transgene showed greatly reduced levels of activity compared with wt, whereas the 2B transgene gave levels greater than wt (data not shown).

View larger version (22K): [in this window] [in a new window]   FIGURE 2. Effect of the 2CK-1 and 2B mutations in the human GM-CSF promoter in a transgenic setting. a, Splenocytes from either wt or 2CK-1 and 2B mutant transgenic mice were cultured for 16 h with PMA/I (20 ng/ml and 1 µM, respectively), and human and mouse GM-CSF levels were assayed from the cell supernatants. The levels of mouse GM-CSF were normalized to 1, and the values for human GM-CSF were plotted relative to mouse levels and corrected for copy number (H/M/C). The 2CK-1 and the 2B mutants are plotted on separate scales. At least five mice were analyzed for each line, and the data are the average ± SDs. b, RT-PCR analysis on RNA isolated from splenic T cells prepared from wt or 2CK-1 mutant transgenic mice. Both GAPDH and human GM-CSF were amplified, and GAPDH was used as a normalization control. The gels were scanned using a Fujifilm LAS 3000, and the digital data were plotted as human GM-CSF levels normalized with GAPDH. The copy number for each line is shown.

We also analyzed the effect of these NF-B site mutations on transgene function in a second mouse cell type that produces GM-CSF. Bone marrow-derived macrophage-type cells were derived from mouse femurs, grown in culture in M-CSF and IL-3, and their response to PMA/I stimulation monitored by measuring human GM-CSF in the cell supernatants. The transgenes all behaved in a similar manner to that described for the splenocytes, with only the 2CK-1 mutant lines having a significantly reduced GM-CSF production (data not shown).

Stable integration into chromosomal DNA is required to observe the effect of the 2CK-1 mutation

Because the results seen with the transgenic CK-1M and BM mutants differed from the previously reported transient transfection studies, we compared the activity of the 2CK-1 and 2B mutations in transgenics and transient transfection assays. Both the 2CK-1 and 2B constructs were responsive to PMA/I or PMA/I + CD28 activation in Jurkat T cells, with comparable or higher levels of activation compared with the wt reporter construct (Fig. 3a). The basal activity in unstimulated cells for both the 2CK-1 and 2B mutant constructs was higher than for the wt, leading to an overall higher activity following activation (Fig. 3a). The fold response to stimulation was, however, very similar for both the wt and the 2CK-1 and 2B mutants. Therefore, both the 2B and the 2CK-1 mutants gave very different responses in transient or transgenic settings.

View larger version (16K): [in this window] [in a new window]   FIGURE 3. a, The wt pGMLuc and the mutant pGM2CK-1 and pGM2B luciferase reporter constructs were transiently transfected into Jurkat T cells. Cells were stimulated for 8 h with PMA/I or PMA/I + CD28 or left unstimulated (NS) and cell lysates harvested for luciferase assays. The value for the pGMLuc construct stimulated with PMA/I/CD28 was set at 100%, and values for all the other samples were plotted relative to this value. The plot is a representation of three experiments. b, Stably transfected Jurkat T cell lines were generated with the pGMLuc, pGM2CK1, and pGM2B constructs. Cells were either left unstimulated (NS) or were stimulated for 8 h with PMA/I. The value of pGMLuc stimulated with PMA/I was set at 100%, and all other values were plotted relative to this. c, Stably transfected EL-4 thymoma cells were generated with the pBSXH10.5 wt GM-CSF construct and with the same construct containing mutations in the CD28RR region (pBSXS10.5CK1 M, pBSXS10.5BM, pBSXS2CK1, and pBSXS10.52B). Cells were stimulated with PMA/I for 8 h or left unstimulated. The stimulated value for the wt construct was given a value of 100%, and the mutant values were plotted relative to this value.

These results showed that the effect of the 2CK-1 and 2B mutations is only manifested when the constructs are integrated into chromosomal DNA. This may indicate that these mutations impact on chromatin structure or chromatin remodeling across the GM-CSF promoter.

The 2CK-1 mutation affects both NF-B and Sp1 binding to the GM-CSF promoter

To determine what effect the 2CK-1 mutation had on transcription factor binding to the CD28RR and to ensure that a second CK-1 site had been created, EMSA assays were performed with nuclear extracts from either Jurkat T cells or mouse splenic T cells. The double-stranded oligonucleotides used as probes are shown in Fig. 4a. The proteins that bind to the CK-1 and NF-B elements have previously been characterized in detail (20, 21, 26, 38, 42). The CK-1 element can bind two RelA-containing complexes from Jurkat nuclear extracts following stimulation for 1 h with PMA/I or PMA/I + CD28 (Fig. 4b, lanes 2 and 3). After 6 h of stimulation, the binding pattern changes and the CK-1-bound complexes are mainly composed of c-Rel (Fig. 4b, lanes 4 and 5). In contrast, the classical NF-B site (GM-B) binds a weak inducible RelA/p50-containing complex that is strongest at 1 h following PMA/I + CD28 activation (Fig. 4b, lanes 12–15). Binding of Sp1 to the overlapping Sp1 site can also be seen with the GM-B oligonucleotide (Fig. 4b, lanes 11–15). The identity of all of these complexes has previously been determined by extensive Ab supershift and competition experiments (20, 21, 26).

View larger version (56K): [in this window] [in a new window]   FIGURE 4. Characterization of transcription factor binding to the 2CK-1 mutation. a, Sequences of the oligonucleotides used as radiolabeled probes or as competitors in EMSAs. The relevant sites within the sequence are underlined. b, Nuclear extracts (6 µg) prepared from Jurkat T cells unstimulated (NS) or stimulated with PMA/I (lanes 2, 4, 7, 9, 12, and 14), PMA/I plus CD28 (PMA/I/C) (lanes 3, 5, 8, 10, 13, and 15) for 1 h (lanes 2, 7, 12 and 3, 8, and 13) or 6 h (lanes 4, 9, 14 and 5, 10, and 15) were used in EMSAs with radiolabeled GM-CK-1 (CK-1) (lanes 1–5), 2CK-1 (lanes 6–10), or GM-B (lanes 11–15) probes. The region of the gel containing the inducible NF-B bands and Sp1 is shown. c, Nuclear extracts were prepared from splenic T cells stimulated for 4 h on plate-bound CD3 alone (lanes 2, 5, and 8), CD3 + CD28 (lanes 3, 6, and 9), or with no Ab (NS) (lanes 1, 4, and 7). EMSAs were performed with 6 µg of these nuclear extracts and radiolabeled CK-1 (lanes 1–3), 2CK-1 (lanes 4–6), and GM-B (lanes 7–9) probes. d, EMSAs were performed, as described in c, with the CK-1 (lanes 1–4) and 2CK-1 (lanes 5–8) probes. Abs to RelA (lanes 3 and 7) or c-Rel (lanes 4 and 8) were added to the nuclear extracts and incubated for 5 min before the addition of the probe. e, Nuclear extracts (6 µg) prepared from PMA/I-stimulated or unstimulated splenic T cells were used in EMSAs with radiolabeled GM-B (lanes 1–10) or 2CK-1 (lanes 11–20). Abs against Sp1 (lanes 3 and 13) or Egr1 (lanes 4 and 14) and Sp1 (lanes 5, 6 and 15, and 16) or HIV-B (lanes 7, 8 and 17, and 18) competitors were preincubated with nuclear extracts before the addition of the probe. Ten-nanogram competitor double-stranded oligonucleotides were used in each reaction. Lanes 9 and 19 had both Sp1 and HIV-B competitors added, while lanes 10 and 20 contained a control competitor.

Similar binding patterns for the wt and mutants were also observed when extracts from mouse splenic T cells were used. Splenic T cells were stimulated with CD3 or CD3/CD28 for 4 h, and binding to CK-1, B, and 2CK-1 oligonucleotides was assayed by EMSA. The CK-1 probe bound two inducible complexes, the upper one of which was increased in intensity following CD28 activation (Fig. 4c, lanes 2 and 3). Both of these inducible complexes contain NF-B proteins (data not shown). The GM-B probe formed an inducible complex that contains NF-B proteins (p50/RelA), as shown by Ab supershift experiments (data not shown) and competition assays (see Fig. 4e, lanes 7 and 8). Again, the 2CK-1 probe had reduced p50/RelA binding compared with the GM-B probe, but increased binding of a slower migrating complex that increased with CD28 activation (Fig. 4c, lanes 4–6). A faint band that migrated in the position expected for Sp1 was observed with both the GM-B and the 2CK-1 probes.

The inducible complexes described above that bound to the 2CK-1 mutant probe resembled the complex that bound to the natural CK-1 site at late times following PMA/I activation in Jurkat T cells, or CD3/CD28 activation of splenic T cells. To determine the composition of this complex, Ab supershift experiments were performed with RelA or c-Rel Abs. Extracts from splenocytes stimulated for 4 h with PMA/I were bound to the CK-1 or 2CK-1 probes. Two inducible bands were observed binding to both the CK-1 and the 2CK-1 probes (Fig. 4d, lanes 2 and 6). The lower faint inducible band is reduced in intensity by the addition of the RelA Ab (Fig. 4d, lanes 3 and 7). This band migrates in the same position as the p50/RelA band that binds to the B probe and is likely to be composed of RelA/p50. The slower migrating inducible complex is removed by the addition of c-Rel Abs, and a strong supershift is observed (Fig. 4d, lanes 4 and 8). Thus, the B site has been changed from a classical RelA/p50 binding site to a site with preference for c-Rel. Therefore, in the 2CK-1 constructs, the promoter now has two sites that can preferentially bind c-Rel, whereas the wt promoter has one c-Rel binding site (CK-1) and one RelA/p50 binding site (NF-B).

By changing a C/G to an A/T base pair (position -77), the 2CK-1 mutation alters the second base of the Sp1 site that overlaps the NF-B element (see Fig. 1a). It appeared from our experiments that Sp1 binding was reduced on the 2CK-1 probe compared with the wt NF-B probe (for example, see Fig. 4b). To characterize Sp1 binding, Ab supershift and competition experiments were conducted with extracts from PMA/I-stimulated splenic T cells. Binding of P/I-stimulated splenic T cell extracts to the NF-B site generated three major bands that have been previously characterized as NF-B and Sp1 bands (43) (Fig. 4e, lane 2). The addition of an Sp1 Ab generated a supershifted complex on the GM-B probe, whereas an Ab to the related Egr1 protein did not (Fig. 4e, lanes 5 and 6). As expected from previous work (43), the most slowly migrating band was competed by the addition of an Sp1 competitor oligonucleotide (Fig. 4e, lanes 5 and 6). On the other hand, an NF-B competitor oligonucleotide (HIV-B) competed the middle complex and the bottom half of the lower complex (Fig. 4e, lanes 7 and 8). Thus, the GM-B probe binds two NF-B-like and one Sp1-like bands. The middle inducible band has previously been characterized as the inducible RelA/p50 complex (26, 42). Binding to the 2CK-1 probe showed a pattern of band formation distinct from that seen for the GM-B probe, as described above (Fig. 4e, compare lanes 2 and 12). On the 2CK-1 probe, the Sp1 Ab did not generate a supershifted complex, implying that Sp1 binding had been eliminated or reduced below the level of Ab detection (Fig. 4e, lane 13). None of the 2CK-1 complexes was inhibited by the addition of an Sp1 competitor, whereas the NF-B competitor (HIV-B) eliminated all but a faint slowly migrating smear (Fig. 4e, lanes 15–18). This band may represent some residual Sp1 binding since it migrates at the appropriate position on the gel and is eliminated by the addition of both Sp1 and NF-B competitors (Fig. 4e, lane 19). Similar results were obtained when nuclear extracts from Jurkat T cells were used in EMSAs (data not shown).

These results show that the NF-B family members that bind to the GM-CSF promoter have been altered by the 2CK-1 mutation. In addition, Sp1 binding is reduced, leading to the possibility that this combined effect may be responsible for the lack of activity of the mutant transgene.

The 2CK-1 mutation prevents chromatin remodeling across the GM-CSF promoter

We have previously shown that upon activation of T cells, the human GM-CSF proximal promoter becomes hypersensitive to DNase I digestion, suggesting that chromatin remodeling occurs across this region (24, 36). To determine whether chromatin remodeling was affected by the 2CK-1 mutation, nuclei were prepared from Con A-elicited T lymphoblasts derived from splenocyte cultures and digested with DNase I. The DH sites across the human GM-CSF promoter were monitored by Southern blotting using an EcoRI-SacI probe shown in Fig. 5. In wt mice, DH sites are observed in stimulated, but not unstimulated T cells at positions that correspond to the promoter and upstream enhancer regions (Fig. 5) (24). In the two lines of the 2CK-1 mutant analyzed, the DH site normally observed at the promoter in the wt transgene was absent (Fig. 5). One of these lines (1236) was the line that showed higher activity than any of the other 2CK-1 mutant lines, implying that the mutation is disrupting promoter structure even in this line. The inducible DH site at the enhancer was, however, still present in all mutant lines (Fig. 5). In contrast, the CK-1M and BM individual mutations had no effect on the appearance of either the promoter or enhancer hypersensitive sites (Fig. 5). Hence, the presence of a DH site at the gene promoter correlates with the transcriptional activity observed with these mutants.

The lack of a promoter DH site in the 2CK-1 mutant suggests that chromatin remodeling is disrupted by this mutation. The appearance of the enhancer hypersensitive site shows that global chromatin organization across the transgene is not disrupted, and that the mutant is having a specific effect at the proximal promoter region.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
Regulation of GM-CSF gene transcription has been extensively studied using transient transfection assays in T cell lines in culture (reviewed in Ref. 6). Using this approach, a region of approximately 120 bp upstream from the start of transcription has been identified that can drive inducible transcription in T cell lines (6). This region and the functional elements that have been identified within it are highly conserved between the mouse and human genes (6, 44, 45). Mutation of these individual elements has been shown in many cases to reduce reporter gene transcription when transfected into Jurkat or EL-4 T cells (18, 19, 20, 21, 22, 23, 26, 45). We show in this study, however, that individual mutations in two of these elements in a 10.5-kb transgene context do not reduce the level of GM-CSF produced. This was true for both splenic T cells and bone marrow-derived macrophages stimulated in vitro with PMA plus I or CD3 + CD28 (for T cells).

The two elements tested in this study are both NF-B-type binding sites and form part of a region (the CD28RR) that has been shown to respond to CD28 activation in T cell lines (26, 27). The mutations introduced have been shown to completely eliminate NF-B binding from both Jurkat T cell and mouse splenic T cell extracts (26, 42) (data not shown). In the context of the 10-kb transgene, however, these individual sites appear to be redundant. Interestingly, the response to CD28 activation is identical for the CK-1M, BM, and the wt transgenes. This result may be due to the dominant nature of the upstream enhancer in the 10.5-kb construct that has been shown to increase promoter activity in reporter gene transient transfection assays by 2- to 10-fold (24). There may also be other dominant elements in the 10-kb construct that have not yet been characterized. In the transgenic context, deletion of the entire upstream enhancer reduced transgene expression by an average of approximately 70% (36). It may be necessary to delete the entire CD28RR to observe a reduction in transgene transcription. It is interesting to note, however, that when these transgenes were used to make stable EL-4 cell lines, both the CK-1M and BM constructs had significantly reduced activity (Fig. 3c, to 35% and 22% of wt, respectively). This result implies that the redundancy seen in the transgenics is either a property of primary T cells or that developmental controls experienced by the transgene in the mouse influence its behavior. Since it has been shown that CD28 activation also influences GM-CSF mRNA stability especially in primary cells (46), and it is possible that this effect is dominant in the transgenic setting.

One of the major concerns for the interpretation of transgene activities is the effect of position of integration on the transgene. It has been shown that the 10.5-kb GM-CSF transgene is expressed in an inducible, position-independent manner in 10 of 11 independent lines examined to date (36). We found, however, that by introducing the CK-1 or B mutations, a greater variability in levels of induction was observed. While we have not examined the reason for this increased variability, it is possible that these mutations influence position independence. We have to date only examined transgene expression at the population level, and it is possible that there is variation in the number of cells in the spleen cell population that express the wt or mutated transgenes as opposed to the absolute levels produced per cell. Such effects have been observed in the -globin gene system as well as for the CD2 gene (47, 48, 49).

By mutating a classical NF-B site to a CK-1 site (2CK-1), we generated a transgene that had greatly reduced activity. On the other hand, the presence of two classical NF-B sites (2B) had greatly increased activity, implying that the NF-B site is a more efficient activator of transcription than the CK-1-type sites. Since the 2CK-1 mutation has reduced activity only when the construct is integrated into a chromosomal setting, it implies that the mutation influences a level of control that is not evident in transient assays. Because transiently transfected plasmids assemble into a dysregulated chromatin structure that is less ordered than chromosomal DNA (50), it is possible that the mutation is affecting a level of control exerted by the chromosomal setting. We have previously shown that the promoter of the GM-CSF gene forms an inducible DH site following T cell activation (26). The 2CK-1 mutation prevented the formation of the promoter DH site within the transgene. The inducible DH site on the upstream enhancer is, however, still present in the 2CK-1 lines, showing that the entire transgene is not disregulated. This result also suggests that the formation of the DH sites at the promoter and enhancer are independent events. The lack of the DH site was found in both a line with little residual activity (1200) and the 1236 line that appeared to have high levels of activity. This result argues that the inability to form a DH site is intrinsic to the 2CK-1 mutation, and that the high level of activity in the 1236 2CK-1 line is due to other effects. The very high copy number in this line may also influence its activity. The inability of the 2CK-1 mutant to form a DH site is in contrast to the ability of the CK-1M and BM mutations to generate a wt-like hypersensitive site. This result argues that mutation of either the CK-1 or B sites independently does not influence the ability of the promoter to recruit chromatin-remodeling activities that are necessary for the generation of the altered chromatin configuration observed as a DH site.

It is possible that the 2CK-1 mutation affects the recruitment of chromatin-remodeling activities to the promoter. Chromatin-remodeling activities are thought to be recruited to genes via interaction with specific combinations of transcription factors (54–57). The 2CK-1 mutation alters the binding of two families of transcription factors to the GM-CSF promoter. First, it changes the NF-B-type complexes that bind to the B site, and second, it reduces the binding of Sp1 to an overlapping site. It has previously been shown that mutation of the Sp1 region influences binding to the B site, and it is likely that these elements operate as a functional unit (29, 43). It has also been shown that NF-B and Sp1 cooperatively activate the HIV long terminal repeat (51), and that at least part of this effect is due to the requirement of both proteins for correct nucleosome positioning on the HIV long terminal repeat (52, 53). Although the EMSA assays shown in this study did not detect any cooperative binding of Sp1 and NF-B proteins, we have recently shown in DNA recruitment-type assays that they can influence each other’s binding to the GM-CSF promoter (Holloway and Shannon, unpublished data). It has been shown in mammalian cells that the RelA NF-B protein directly interacts with the CREB-binding-protein/histone acetyl transferase protein (54). It has also recently been shown that Zn finger proteins (of which Sp1 is a member) may play a role in the recruitment of SWI/SNF-like complexes to the -globin gene (55). The 2CK-1 mutation, therefore, alters the binding of two transcription factors that are thought to play a role in the recruitment of chromatin-remodeling activities to genes.

The results described in this work caution against the extrapolation of results obtained in a transient transfection system to the endogenous gene, and suggest that there may be more redundancy for individual elements in cytokine gene promoters than was previously considered to be the case. In addition, the result obtained with the 2CK-1 mutation stresses the role that the chromatin environment may play in controlling inducible cytokine gene transcription.

	Acknowledgments

 
We thank Donna Woltring for extensive assistance in caring for the transgenic mice, Dr. Sudha Rao for critical reading of the manuscript, and Drs. Ellen Rothenberg and Adele Holloway for useful discussions. We thank Dr. Robyn Slattery and Jennifer Kofler for generating the 2CK-1 and 2B mice.

	Footnotes

 
¹ Address correspondence and reprint requests to Dr. M. Frances Shannon, Division of Biochemistry and Molecular Biology, John Curtin School of Medical Research, Australian National University, Mills Road, Acton, Canberra, ACT 2601, Australia. E-mail address: frances.shannon{at}anu.edu.au

² Abbreviations used in this paper: CD28RR, CD28 response region; DH, DNase I hypersensitive; I, calcium ionophore; wt, wild type.

³ Current address: Molecular Medicine Unit, University of Leeds, Clinical Sciences Building, St. James University Hospital, Leeds LS9 7TF, U.K.

Received for publication February 14, 2001. Accepted for publication April 27, 2001.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. O’Garra, A.. 1998. Cytokines induce the development of functionally heterogeneous T helper cell subsets. Immunity 8:275.[Medline]
2. Abbas, A. K., K. M. Murphy, A. Sher. 1996. Functional diversity of helper T lymphocytes. Nature 383:787.[Medline]
3. Brown, M. A., J. Hural. 1997. Functions of IL-4 and control of its expression. Crit. Rev. Immunol. 17:1.[Medline]
4. Avots, A., C. Escher, S. Muller-Deubert, M. Neumann, E. Serfling. 1995. The interplay between lymphoid-specific and ubiquitous transcription factors controls the expression of interleukin 2 gene in T lymphocytes. Immunobiology 193:254.[Medline]
5. Jain, J., C. Loh, A. Rao. 1995. Transcriptional regulation of the IL-2 gene. Curr. Opin. Immunol. 7:333.[Medline]
6. Shannon, M. F., L. S. Coles, M. A. Vadas, P. N. Cockerill. 1997. Signals for activation of the GM-CSF promoter and enhancer in T cells. Crit. Rev. Immunol. 17:301.[Medline]
7. Kuo, C. T., J. M. Leiden. 1999. Transcriptional regulation of T lymphocyte development and function. Annu. Rev. Immunol. 17:149.[Medline]
8. Rao, A.. 1994. NF-ATp: a transcription factor required for the coordinate induction of several cytokine genes. Immunol. Today 15:274.[Medline]
9. Carey, M.. 1998. The enhanceosome and transcriptional synergy. Cell 92:5.[Medline]
10. Rothenberg, E. V., S. B. Ward. 1996. A dynamic assembly of diverse transcription factors integrates activation and cell-type information for interleukin 2 gene regulation. Proc. Natl. Acad. Sci. USA 93:9358.[Abstract/Free Full Text]
11. Merika, M., A. J. Williams, G. Chen, T. Collins, D. Thanos. 1998. Recruitment of CBP/p300 by the IFN enhanceosome is required for synergistic activation of transcription. Mol. Cell 1:277.[Medline]
12. Agalioti, T., S. Lomvardas, B. Parekh, J. Yie, T. Maniatis, D. Thanos. 2000. Ordered recruitment of chromatin modifying and general transcription factors to the IFN- promoter. Cell 103:667.[Medline]
13. Parekh, B. S., T. Maniatis. 1999. Virus infection leads to localized hyperacetylation of histones H3 and H4 at the IFN- promoter. Mol. Cell 3:125.[Medline]
14. Falvo, J. V., A. M. Uglialoro, B. M. Brinkman, M. Merika, B. S. Parekh, E. Y. Tsai, H. C. King, A. D. Morielli, E. G. Peralta, T. Maniatis, et al 2000. Stimulus-specific assembly of enhancer complexes on the tumor necrosis factor gene promoter. Mol. Cell. Biol. 20:2239.[Abstract/Free Full Text]
15. Gasson, J. C., J. K. Fraser, S. D. Nimer. 1990. Human granulocyte-macrophage colony-stimulating factor (GM-CSF): regulation of expression. Prog. Clin. Biol. Res. 338:27.[Medline]
16. Gasson, J. C., G. C. Baldwin, K. M. Sakamoto, J. F. DiPersio. 1990. The biology of human granulocyte-macrophage colony-stimulating factor (GM-CSF). Prog. Clin. Biol. Res. 352:375.[Medline]
17. Lieschke, G. J.. 1997. CSF-deficient mice: what have they taught us?. Ciba Found. Symp. 204:60.[Medline]
18. Thomas, R. S., M. J. Tymms, A. Seth, M. F. Shannon, I. Kola. 1995. ETS1 transactivates the human GM-CSF promoter in Jurkat T cells stimulated with PMA and ionomycin. Oncogene 11:2135.[Medline]
19. Thomas, R. S., M. J. Tymms, L. H. McKinlay, M. F. Shannon, A. Seth, I. Kola. 1997. ETS1, NFB and AP1 synergistically transactivate the human GM-CSF promoter. Oncogene 14:2845.[Medline]
20. Jenkins, F., P. N. Cockerill, D. Bohmann, M. F. Shannon. 1995. Multiple signals are required for function of the human granulocyte-macrophage colony-stimulating factor gene promoter in T cells. J. Immunol. 155:1240.[Abstract]
21. Shang, C., J. Attema, D. Cakouros, P. N. Cockerill, M. F. Shannon. 1999. Nuclear factor of activated T cells contributes to the function of the CD28 response region of the granulocyte macrophage-colony stimulating factor promoter. Int. Immunol. 11:1945.[Abstract/Free Full Text]
22. Masuda, E. S., H. Tokumitsu, A. Tsuboi, J. Shlomai, P. Hung, K. Arai, N. Arai. 1993. The granulocyte-macrophage colony-stimulating factor promoter cis-acting element CLE0 mediates induction signals in T cells and is recognized by factors related to AP1 and NFAT. Mol. Cell. Biol. 13:7399.[Abstract/Free Full Text]
23. Masuda, E. S., Y. Yamaguchi-Iwai, A. Tsuboi, P. Hung, K. Arai, N. Arai. 1994. The transcription factor Sp1 is required for induction of the murine GM-CSF promoter in T cells. Biochem. Biophys. Acta 205:1518.
24. Cockerill, P. N., M. F. Shannon, A. G. Bert, G. R. Ryan, M. A. Vadas. 1993. The granulocyte-macrophage colony-stimulating factor/interleukin 3 locus is regulated by an inducible cyclosporin A-sensitive enhancer. Proc. Natl. Acad. Sci. USA 90:2466.[Abstract/Free Full Text]
25. Cockerill, P. N., A. G. Bert, F. Jenkins, G. R. Ryan, M. F. Shannon, M. A. Vadas. 1995. Human granulocyte-macrophage colony-stimulating factor enhancer function is associated with cooperative interactions between AP-1 and NFATp/c. Mol. Cell. Biol. 15:2071.[Abstract]
26. Himes, S. R., L. S. Coles, R. Reeves, M. F. Shannon. 1996. High mobility group protein I(Y) is required for function and for c-Rel binding to CD28 response elements within the GM-CSF and IL-2 promoters. Immunity 5:479.[Medline]
27. Himes, S. R., R. Reeves, J. Attema, M. Nissen, Y. Li, M. F. Shannon. 2000. The role of high-mobility group I(Y) proteins in expression of IL-2 and T cell proliferation. J. Immunol. 164:3157.[Abstract/Free Full Text]
28. Fraser, J. D., B. A. Irving, G. R. Crabtree, A. Weiss. 1991. Regulation of interleukin-2 gene enhancer activity by the T cell accessory molecule CD28. Science 251:313.[Abstract/Free Full Text]
29. Kaushansky, K., C. O’Rork, S. G. Shoemaker, J. McCarty. 1996. The regulation of GM-CSF is dependent on a complex interplay of multiple nuclear proteins. Mol. Immunol. 33:461.[Medline]
30. Agarwal, S., A. Rao. 1998. Modulation of chromatin structure regulates cytokine gene expression during T cell differentiation. Immunity 9:765.[Medline]
31. Agarwal, S., A. Rao. 1998. Long-range transcriptional regulation of cytokine gene expression. Curr. Opin. Immunol. 10:345.[Medline]
32. Takemoto, N., N. Koyano-Nakagawa, T. Yokota, N. Arai, S. Miyatake, K. Arai. 1998. Th2-specific DNase I-hypersensitive sites in the murine IL-13 and IL-4 intergenic region. Int. Immunol. 10:1981.[Abstract/Free Full Text]
33. Agarwal, S., J. P. Viola, A. Rao. 1999. Chromatin-based regulatory mechanisms governing cytokine gene transcription. J. Allergy Clin. Immunol. 103:990.[Medline]
34. Takemoto, N., Y. Kamogawa, H. Jun Lee, H. Kurata, K. Arai, A. O’Garra, N. Arai, S. Miyatake. 2000. Cutting edge: chromatin remodeling at the IL-4/IL-13 intergenic regulatory region for Th2-specific cytokine gene cluster. J. Immunol. 165:6687.[Abstract/Free Full Text]
35. Ward, S. B., G. Hernandez-Hoyos, F. Chen, M. Waterman, R. Reeves, E. V. Rothenberg. 1998. Chromatin remodeling of the interleukin-2 gene: distinct alterations in the proximal versus distal enhancer regions. Nucleic Acids Res. 26:2923.[Abstract/Free Full Text]
36. Cockerill, P. N., A. G. Bert, D. Roberts, M. A. Vadas. 1999. The human granulocyte-macrophage colony-stimulating factor gene is autonomously regulated in vivo by an inducible tissue-specific enhancer. Proc. Natl. Acad. Sci. USA 96:15097.[Abstract/Free Full Text]
37. Nordeen, S. K.. 1988. Luciferase reporter gene vectors for analysis of promoters and enhancers. BioTechniques 6:454.[Medline]
38. Himes, S. R., L. S. Coles, R. Katsikeros, R. K. Lang, M. F. Shannon. 1993. HTLV-1 tax activation of the GM-CSF and G-CSF promoters requires the interaction of NF-B with other transcription factor families. Oncogene 8:3189.[Medline]
39. Hercus, T. R., C. J. Bagley, B. Cambareri, M. Dottore, J. M. Woodcock, M. A. Vadas, M. F. Shannon, A. F. Lopez. 1994. Specific human granulocyte-macrophage colony-stimulating factor antagonists. Proc. Natl. Acad. Sci. USA 91:5838.[Abstract/Free Full Text]
40. Gruffat, D., C. Piot, D. Durand, D. Bauchart. 1996. Comparison of four methods for isolating large mRNA: apolipoprotein B mRNA in bovine and rat livers. Anal. Biochem. 242:77.[Medline]
41. Schreiber, E., P. Matthias, M. M. Muller, W. Schaffner. 1989. Rapid detection of octamer binding proteins with ‘mini-extracts,’ prepared from a small number of cells. Nucleic Acids Res. 17:6419.[Free Full Text]
42. Himes, S. R., R. Katsikeros, M. F. Shannon. 1996. Costimulation of cytokine gene expression in T cells by the human T leukemia/lymphotropic virus type 1 trans activator Tax. J. Virol. 70:4001.[Abstract]
43. Kochetkova, M., P. O. Iversen, A. F. Lopez, M. F. Shannon. 1997. Deoxyribonucleic acid triplex formation inhibits granulocyte macrophage colony-stimulating factor gene expression and suppresses growth in juvenile myelomonocytic leukemic cells. J. Clin. Invest. 99:3000.[Medline]
44. Miyatake, S., T. Otsuka, T. Yokota, F. Lee, K. Arai. 1985. Structure of the chromosomal gene for granulocyte-macrophage colony stimulating factor: comparison of the mouse and human genes. EMBO J. 4:2561.[Medline]
45. Miyatake, S., J. Shlomai, K. Arai, N. Arai. 1991. Characterization of the mouse granulocyte-macrophage colony-stimulating factor (GM-CSF) gene promoter: nuclear factors that interact with an element shared by three lymphokine genes: those for GM-CSF, interleukin-4 (IL-4), and IL-5. Mol. Cell. Biol. 11:5894.[Abstract/Free Full Text]
46. Lindstein, T., C. H. June, J. A. Ledbetter, G. Stella, C. B. Thompson. 1989. Regulation of lymphokine messenger RNA stability by a surface-mediated T cell activation pathway. Science 244:339.[Abstract/Free Full Text]
47. Lien, L. L., Y. Lee, S. H. Orkin. 1997. Regulation of the myeloid-cell-expressed human gp91-phox gene as studied by transfer of yeast artificial chromosome clones into embryonic stem cells: suppression of a variegated cellular pattern of expression requires a full complement of distant cis elements. Mol. Cell. Biol. 17:2279.[Abstract]
48. Festenstein, R., M. Tolaini, P. Corbella, C. Mamalaki, J. Parrington, M. Fox, A. Miliou, M. Jones, D. Kioussis. 1996. Locus control region function and heterochromatin-induced position effect variegation. Science 271:1123.[Abstract]
49. Sutherland, H. G., D. I. Martin, E. Whitelaw. 1997. A globin enhancer acts by increasing the proportion of erythrocytes expressing a linked transgene. Mol. Cell. Biol. 17:1607.[Abstract]
50. Jeong, S., A. Stein. 1994. Micrococcal nuclease digestion of nuclei reveals extended nucleosome ladders having anomalous DNA lengths for chromatin assembled on non-replicating plasmids in transfected cells. Nucleic Acids Res. 22:370.[Abstract/Free Full Text]
51. Perkins, N. D., N. L. Edwards, C. S. Duckett, A. B. Agranoff, R. M. Schmid, G. J. Nabel. 1993. A cooperative interaction between NF-B and Sp1 is required for HIV-1 enhancer activation. EMBO J. 12:3551.[Medline]
52. Pazin, M. J., P. L. Sheridan, K. Cannon, Z. Cao, J. G. Keck, J. T. Kadonaga, K. A. Jones. 1996. NF-B-mediated chromatin reconfiguration and transcriptional activation of the HIV-1 enhancer in vitro. Genes Dev. 10:37.[Abstract/Free Full Text]
53. Widlak, P., R. B. Gaynor, W. T. Garrard. 1997. In vitro chromatin assembly of the HIV-1 promoter: ATP-dependent polar repositioning of nucleosomes by Sp1 and NFB. J. Biol. Chem. 272:17654.[Abstract/Free Full Text]
54. Zhong, H., R. E. Voll, S. Ghosh. 1998. Phosphorylation of NF-B p65 by PKA stimulates transcriptional activity by promoting a novel bivalent interaction with the coactivator CBP/p300. Mol. Cell 1:661.[Medline]
55. Armstrong, J. A., J. J. Bieker, B. M. Emerson. 1998. A SWI/SNF-related chromatin remodeling complex, E-RC1, is required for tissue-specific transcriptional regulation by EKLF in vitro. Cell 95:93.[Medline]

This article has been cited by other articles: