| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Howard Hughes Medical Institute, Departments of
Medicine and Microbiology/Immunology, and
Biochemistry and Biophysics, University of California, San Francisco, CA 94143; and
Genome Sciences Department, Lawrence Berkeley National Laboratory, Berkeley, CA 94720
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
IL-4 is a cytokine secreted by the Th2 subset of T cells that has been implicated in host protection against mucosal pathogens, such as intestinal worms, but also pathologically in asthma and allergy. IL-4 has diverse roles in immunity, however, suggesting functions in addition to the growth and development of Th2 cells (2). Thus, IL-4 is a growth factor for APCs, including B cells and dendritic cells. IL-4 enhances IL-12 secretion from dendritic cells, a process likely to provoke Th1 development, and IL-4-deficient mice demonstrate defects in diverse aspects of both type 1 and type 2 immunity, including cytotoxic T cell development (3, 4, 5). As such, IL-4 is a true helper cytokine that serves to mediate multiple components of the adaptive immune response.
In Th2 cells, IL-4 is frequently coexpressed with IL-13 and IL-5, closely arrayed genes on orthologous regions of chromosomes 5 and 11 in human and mouse, respectively. Cross-species sequence comparisons were used to identify a conserved noncoding region, designated conserved noncoding sequence 1 (CNS-1),3 between the il4 and il13 genes that was required for optimal expression of IL-4 in vitro and in vivo (6, 7). Transcription of the il4 gene, which occurs within 30 min of in vitro activation through the TCR and the costimulatory receptor, CD28, was delayed in the absence of CNS-1 (7, 8). The rapid transcription of il4 in naive T cells is temporally concordant with the targeting of nucleosome remodeling (ATP-dependent nucleosome remodeling complex (SWI/SNF)-like mammalian SWI/SNF-related complex (BAF)) complexes to chromatin after TCR activation (9). Despite these observations, assays of chromatin modifications associated with the il4 gene suggest a relatively inaccessible configuration in naive CD4 T cells, characterized by low levels of lysine acetylation of histone tails and a highly methylated state at CpG nucleotides (10, 11, 12, 13). Examples of discordant correlations between gene expression and chromatin modifications are known (14, 15), however, suggesting that discrete subdomains may exist in modified states before more extensive decondensation of larger chromatin domains (16). Our prior observations using cells from mice with the CNS-1 intergenic region deleted as well as cells deficient in CD4 and the Tec family kinase, Itk, revealed common deficiencies in IL-4 expression (17, 18), raising the possibility of functionally similar pathways. We used cells from these various gene-deleted mice to examine the relationship of early IL-4 expression to chromatin modifications at the cytokine locus compared with wild-type cells. We provide evidence that the IL-4 locus exists in a modified state in naive CD4 T cells, and that this modifying mark is absent not only in CD8 T cells, but also in cells deficient in CNS-1, CD4, or Itk. Intriguingly, a second robust feature of CNS-1-deficient T cells was the failure to reposition the silenced type 2 cytokine locus in apposition to heterochromatin in Th1 cells, a feature potentially linked to the presence of reiterative Ikaros binding motifs in the element. The findings suggest that effector cytokine genes constitute highly dynamic genetic elements poised to respond to activating and silencing signals emanating from the environment.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Six-week-old female BALB/c mice (The Jackson Laboratory, Bar Harbor, ME), CNS-1-deficient mice (6), Itk-deficient mice (19), CD1d-deficient mice (20), and mice with a conditionally deleted Cd4 allele (21) were maintained in the pathogen-free animal care facility at University of California (San Francisco, CA) in accordance with institutional guidelines.
Cell isolation and cultures
Small, resting, naive CD4+CD62Lhigh cells were sorted from lymph nodes to >98% purity and were activated using mAb to TCR
(H57.597; 1 µg/ml) and CD28 (37N51.1; 5 µg/ml) in the presence of irradiated APCs prepared from spleens of TCR-C
-deficient mice as previously described (8). Th1 conditions included recombinant murine IL-12 (5 ng/ml; R&D Systems, Minneapolis, MN) and neutralizing anti-IL-4 mAb (11B11; 20 µg/ml). Th2 conditions included recombinant murine IL-4 (50 ng/ml; R&D Systems) and neutralizing anti-IFN-
mAb (XMG1.2; 50 µg/ml). Polarized CD4 T cell lines, including a myelin basic protein-specific Th1 line, PJR25, and a conalbumin-specific Th2 line, D10, were maintained by passage on irradiated APCs and cognate peptide.
CD4 lineage cells were isolated from mice that carried a conditional Cd4 allele flanked by loxP sites and a null Cd4 allele, which were crossed onto transgenic mice containing Cre recombinase under control of the distal lck promoter (line 3779) (21). CD4 is lost from the mature TCRhigh cells during the late stage of thymic development. These CD4 lineage cells lack CD4 expression and maintain normal functional responses to Ag, but are deficient in Th2 development and IL-4 secretion (21). Naive CD62Lhigh CD4 lineage cells were sorted from lymph nodes using a negative gating strategy with Abs against CD4, CD8, CD11b, CD11c, CD19, CD24, CD80, CD86, B220, DX5, 
TCR, and granulocytes.
Chromatin immunoprecipitation
Resting cells or cell lines, or cells activated under the designated polarizing conditions (5–10 x 106 cells) were cross-linked in formaldehyde, and the nuclei were isolated and sonicated using conditions resulting in prominent DNA fragmentation between 500 and 750 bp. After purification by centrifugation over a cesium chloride gradient, cross-linked protein/DNA complexes were immunoprecipitated using control, acetylated (affinity-purified rabbit polyclonal lysine 9- and 14-specific; Upstate Biotechnology, Lake Placid, NY) or dimethylated histone 3 (H3) Abs (affinity-purified rabbit polyclonal lysine 9-specific; Upstate Biotechnology) according to the manufacturer’s specifications (chromatin immunoprecipitation assay kit; Upstate Biotechnology). After reversion of cross-links, immunoprecipitated DNA was ethanol-precipitated and used to template PCR. Primers for amplification during the PCR reaction were: IL-4 promoter (182-bp product): 5' primer, 5'-TTGGTCTGATTTCACAGG; 3' primer, 5'-AACAATGCAATGCTGGC; IL-13 promoter (140-bp product): 5' primer, 5'-CAGATAATGCCCAACAAAGC; 3' primer, 5'-GCAGTCACCCAGAGCGCCAT; CNS-1 element (150-bp product): 5' primer, 5'-ATCACTTTCCACTCGAGATGG; 3' primer, 5'-CTTTCCCAGATCAACTGATTT; and CD3
promoter (126-bp product): 5' primer, 5'-TTCATCCTTATGGGAAGGC; 3' primer, 5'-ACACAGGAAGTGTAGAGG. Reactions were run for 35–40 cycles. Paired Student’s t test or Mann-Whitney U test was performed on densitometry of PCR products where indicated.
Transfection assay
The 450-bp CNS-1 sequence was cloned into pCRII-TOPO (TOPO TA Cloning; Invitrogen, Carlsbad, CA) to create the plasmid pCNS-1. Site-directed mutagenesis was used according to the manufacturer’s protocol (Stratagene, La Jolla, CA) to delete the five core nucleotides comprising each of the Ikaros binding sequences to create plasmid pCNS
IK4. Each deletion was verified by direct sequencing. CNS-1 and CNS-1IK4 were excised using EcoRI digestion and inserted 5' of the SV40 promoter in the secreted alkaline phosphatase reporter plasmid, pSEAP (Clontech Laboratories, Palo Alto, CA). The resulting plasmids, designated pCNS and pCNSIK4, respectively, were transfected using 20 µg of plasmid DNA with 20 µg of salmon sperm DNA (Invitrogen) into Jurkat T cells by electroporation (960 µF, 250 mV; Bio-Rad Gene Pulsar; Bio-Rad, Hercules, CA). Transfections were performed in triplicate. Supernatants were collected after 48 h and assayed for alkaline phosphatase activity using chemiluminescence (Clontech Laboratories).
Parasites and infection
BALB/c, 129 x C57BL/6 wild-type, and CNS-1-deficient mice on either of these background strains were infected in the hind footpads with 4 x 105 metacyclic Leishmania major promastigotes (strain WHOM/IR/-/173) as previously described (7). After 8 wk, footpad swelling was measured, and lymphocytes were harvested from the draining popliteal lymph nodes and selected for naive (CD4+/CD44low/CD62Lhigh) or effector/memory (CD4+/CD44high/CD62Llow) phenotypes using appropriate mAb and flow cytometry. Sorted populations were >96% pure. Cytokine expression was analyzed after stimulation with 50 ng/ml PMA and 1 µg/ml ionomycin (Sigma-Aldrich, St. Louis, MO) for 4 h, with 5 µg/ml brefeldin A (Sigma-Aldrich) added for the final 2 h. Intracellular cytokines were quantitated after fixing and light permeabilization, and staining with anti-CD4 mAb and directly conjugated anti-cytokine mAb or isotype control mAb as previously described (8). For cytokine detection by ELISA, cells were stimulated overnight using plate-bound TCR mAb (H57.597; 10 µg/ml) and CD28 mAb (37N51.1; 5 µg/ml). After clearing, supernatants were analyzed for IL-4 and IFN- by ELISA as previously described (7). Serum total IgE from infected mice was determined by ELISA as previously described (7).
Fluorescent in situ hybridization
Simultaneous localization of cytokine genes and centromeric satellite domains was performed as previously described (8, 22). After spotting onto slides, fluorescent in situ hybridization was used to label simultaneously the IL-4 alleles and centromeric satellite repeat domains. Cells were stained with 4',6-diamido-2-phenylindole hydrochloride and analyzed for association of the il4 genes with heterochromatin using multiwavelength wide-field, three-dimensional microscopy as previously described (23). Data stacks of immunofluorescent images were acquired in the FITC and rhodamine channels by moving the stage in successive 0.25-µm focal planes while removing out-of-focus light using a constrained iterative deconvolution algorithm.
EMSA
Sense and antisense oligonucleotides, designated IK1 to IK4, corresponding to each of the core Ikaros binding sites with its eight flanking nucleotides were synthesized, end-labeled with [32P]ATP, and annealed to form double-strand probes. Nuclear extracts were prepared from 293T cells transfected with full-length Ikaros as previously described (24). Extracts (2 µg) were incubated with rabbit anti-Ikaros Ig (25) or whole rabbit IgG (control Ig; Jackson ImmunoResearch Laboratories, West Grove, PA) in assay buffer (20 mM HEPES (pH 7.9), 0.2 mM EDTA, 20% glycerol, 100 mM KCL, 1 mM DTT, 2 µg poly(dI-dC) (BD PharMingen, San Diego, CA), 1 µg BSA, 10 µM ZnCl2, and proteinase inhibitors) for 20 min. One picomole of labeled oligonucleotide probe was added for an additional 30 min at room temperature, and the reaction mix was dispersed in a 4% acrylamide gel as previously described (26).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Protein/DNA complexes were purified from naive CD4 T cells, from cells activated for 1 wk under polarized conditions that establish Th1 and Th2 development, and from terminally polarized Th1 and Th2 lines. Sheared DNA was immunoprecipitated using Abs to lysine 9- and 14-acetylated or lysine 9-dimethylated H3 to concentrate chromatin modifications typically associated with accessible or silenced genes, respectively (27). PCR was used to detect association of the IL-4 and the IL-13 promoters and of the CNS-1 intergenic region with the respective immunoprecipitation fractions.
Under these conditions, the IL-4 and IL-13 promoters as well as the CNS-1 intergenic elements were readily immunoprecipitated using anti-acetylated H3 from Th2 cells, but not Th1 cells (Fig. 1). This is consistent with prior studies demonstrating extended DNA modifications accompanying transcription of these cytokine genes in Th subsets (10). Unexpectedly, the IL-4 promoter was also immunoprecipitated using acetyl-H3 Abs from naive, resting, lymph node CD4 T cells (Fig. 1; levels were 2.7 ± 0.3 times higher than those in naive CD8 T cells, which were undetectable/background; p < 0.001; see Fig. 3). Although levels of detection were lower than in Th2 cells (1.5 ± 0.16-fold increase; p < 0.001), amplification of the IL-4 promoter was consistently greater in naive CD4 T cells than in Th1 cells in >10 experiments (0.8 ± 0.07-fold decrease; p < 0.001). Immunoprecipitation of CNS-1 and the IL-13 promoter were consistently less robust, suggesting little spread of acetylation through the cytokine cluster. Similar findings were obtained using cells from both BALB/c and C57BL/6 mice (data not shown). The CD3 promoter was comparably immunoprecipitated in preparations from all three cell populations (p > 0.1), and failure to incubate with anti-H3 Ab resulted in loss of all signals. Histone modification was not a consequence of purification using CD4 Abs, as reverse purification using magnetic depletion of other cells in the spleen to achieve 97% pure naive CD4 T cell populations demonstrated comparable results (data not shown).
|
|
-/- mice, which are highly enriched for naive T cells in the absence of cognate Ag, displayed similar findings (Fig. 2B). Although results in naive CD4 T cells were robust and reproducible, modifications at the locus were yet more striking in highly polarized Th1 and Th2 lines (Figs. 1 and 2C). Not only were both the IL-4 and IL-13 promoters associated strongly with acetyl-H3 in Th2 lines, but these elements were immunoprecipitated with antidimethyl-H3 in Th1 lines, consistent with active repression of the locus in terminally polarized cells.
|
CD4 T cells deleted of the CNS-1 element located between the il4 and il13 genes demonstrate impaired activation of IL-4 expression (7). To assess whether rapid transcription after TCR activation correlated with the histone modifications of the locus, we compared CNS-1-deficient CD4 T cells with wild-type CD4 T cells (Fig. 3). Under the same experimental conditions, naive CD4 T cells from mice lacking CNS-1 did not immunoprecipitate the IL-4 promoter or, although the basal state was less apparent, the IL-13 promoter using anti-acetyl-H3. CD8 T cells, which poorly activate the il4 gene even when stimulated under Th2-inducing conditions (29), demonstrated no basal modification of the IL-4 or IL-13 promoters in the presence or the absence of CNS-1. Deletion of CNS-1 in mast cells, in contrast to CD4 T cells, did not affect IL-4 expression (7), presumably reflecting requirements for lineage-specific elements in regulation of il4 gene transcription. In support of these findings, basal modification of the IL-4 and IL-13 promoters was established independently of CNS-1 in bone marrow-derived mast cells (Fig. 3).
Basal modification of the type 2 cytokine locus correlates with rapid activation of IL-4 expression
To assess further the relationship of locus accessibility with chromatin modifications, we examined CD4 T cells purified from mice deficient in the Tec family tyrosine kinase, Itk (19), or CD4 lineage T cells lacking CD4 (21); both modifications result in impaired differentiation to Th2 cells in vitro and in vivo (17, 18, 21). Production of IL-4 transcripts after stimulation using anti-TCR/CD28 was impaired, similar to findings in CNS-1-deficient T cells (Fig. 4, A and B). Naive CD4 T cells were purified from the two genetically deficient strains of mice, and chromatin immunoprecipitations were performed using anti-acetyl-H3. In contrast to wild-type CD4 T cells, CD4 lineage T cells lacking CD4 or Itk-deficient CD4 T cells did not contain comparable modification of the locus (Fig. 4C). Taken together, these data suggest a close correlation between rapid expression of IL-4 transcripts after TCR activation and basal modification of the locus.
|
Extensive analysis of CD4- and Itk-deficient mice have confirmed robust functional deficiencies in Th2 immunity in various infectious and inflammatory models (17, 18). Prior analysis of CNS-1-deficient mice demonstrated similarly attenuated Th2 responses in vivo, although the mice were not bred to pure backgrounds (7). After backcrossing 12 generations to BALB/c mice, CNS-1-deficient mice were infected with L. major, a protozoan parasite that induces aberrant and fatal Th2 responses in mice on the BALB background (Fig. 5). After 8 wk, when littermates had footpad lesions averaging 6.8 ± 0.5 mm, BALB/c CNS-1-/- mice had completely resolved the disease, with footpad thicknesses averaging 2.7 ± 0.5 mm. Parasites recovered from the footpads were reduced almost 1000-fold and correlated with the numbers recovered from genetically resistant strains of mice (Fig. 5A). Analysis of IL-4 production by intracellular cytokine analysis and ELISA of T cell-stimulated supernatants revealed a substantial loss of IL-4-producing cells, which was corroborated by the markedly attenuated serum IgE response (Fig. 5, B–D). Thus, in all three of these independent genetic deficiencies, a strong correlation exists among the presence of basal histone modification of the IL-4 promoter in CD4 T cells, the capacity to generate early IL-4 transcripts, and the ability to generate robust Th2 immune responses in vivo.
|
The biochemical modifications of histone that were dependent on CNS-1 suggested that this element may attract proteins required for docking the relevant chromatin-modifying enzymatic machinery. Sequence scanning of the CNS-1 region revealed four Ikaros protein binding sites, designated IK1 to IK4 (Fig. 6A). Ikaros is a hemopoietic cell-specific protein involved in the recruitment of histone-modifying complexes to DNA (30). Each of the putative Ikaros binding sequences was competent to bind Ikaros in gel-shift assays, confirming that each could function as a bona fide Ikaros binding site (Fig. 6B).
|
gene in T cells (33). Further complexity occurs at the TdT gene promoter, where competition between Ikaros and Ets proteins at overlapping DNA binding sites coordinately regulates the developmental expression and silencing of the locus (26). A minigene approach using a region that included CNS-1 was able to localize strong enhancer activity to this intergenic sequence (34). Using CNS-1 linked to a heterologous reporter gene, we confirm that CNS-1 contains enhancer activity (Fig. 6C). Further, this enhancer activity is attenuated in the absence of the four Ikaros binding domains, demonstrating that this genetic element, through binding of either Ikaros or other proteins, has marked effects on linked genes. Enhancer activity was readily apparent in transfected T cells, but not in transfected fibroblasts, confirming the tissue specificity of the element as suggested by prior studies (7). CNS-1-deficient Th1 cells do not reposition the IL-4 gene to heterochromatin
The role of Ikaros in nuclear repositioning of epigenetically silenced genes prompted us to examine nuclear localization of the IL-4 locus in differentiated Th cells. Although the il4 genes are positioned in euchromatic areas of the nucleus in naive CD4 T cells, differentiation to Th1 cells that silence IL-4 expression results in reorganization of the genes to areas of pericentromeric heterochromatin (8). Similar repositioning has been demonstrated in lymphocytes to be mediated by Ikaros family members for a number of developmentally expressed genes (22). To examine the role of CNS-1 in this differentiative scheme in vivo, we analyzed cells from mice infected with L. major, a protozoa that activates robust Th1 development in resistant strains of mice (35). As demonstrated above and previously (7), CNS-1-deficient mice are very resistant to L. major and generate protective Th1 responses. Using conventional cell surface markers, naive and effector/memory CD4 T cells were purified from lymph nodes of infected wild-type or CNS-1-deficient mice and examined by fluorescent in situ hybridization for evidence of IL-4 nuclear repositioning (8). Compared with wild-type effector/memory cells, CNS-1-deficient cells that were polarized to the Th1 phenotype in response to infection failed to reposition the il4 gene in apposition to heterochromatin (Fig. 7). The IFN- genes remained in euchromatin nuclear domains in both groups of cells, consistent with their Th1 phenotype (data not shown). We were unable to immunoprecipitate CNS-1 using anti-Ikaros Abs from Th1 cells, perhaps reflecting the highly compacted nature of Ikaros pericentromeric heterochromatic complexes (30). Despite the failure to reposition, IL-4 was not expressed after activation of CNS-1-deficient Th1 cells in vitro or in vivo (data not shown), suggesting the repositioning mediated by CNS-1 is not required for silencing, in agreement with prior studies (36).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
expression after viral infection in fibroblasts. Infection begins a temporally precise cascade of events mediated by a core group of DNA-binding proteins required to configure the IFN- enhanceosome, a structure required for recruitment of the histone acetyltransferase, GCN5. Acetylation of histone tails, initially on H4 lysine 8 and H3 lysine 9, imprints a histone code with competent docking sites for bromodomain-containing transcription complexes SWI/SNF and transcription factor IID, required for nucleosome repositioning and translation initiation, respectively (37). The temporal procession of these events is crucial; disruption of the ordered process interferes with gene expression (38). The earliest histone acetylation events occur within 2–3 h of infection, with IFN- mRNA accumulating 3–4 h later.
Our studies suggest that naive Th cells demonstrate a modified IL-4 promoter, as evidenced by immunoprecipitation using acetyl-H3 Ab, that is not detected in naive CD8 T cells analyzed under the same conditions. Further, the detection of promoter modifications correlated with the capacity to detect IL-4 transcripts rapidly after TCR activation (8). Intriguingly, deletion of the 8-kb-distant enhancer, the intergenic CNS-1 element, was associated with both loss of IL-4 promoter modification in resting CD4 T cells as well as loss of the capacity to transcribe the IL-4 gene rapidly after activation. This seems unlikely to result from the structural implications in replacing the 489 nucleotide CNS-1 element by 255 nt and the residual loxP site, as two separate genetically modified T cells share these properties despite the presence of wild-type CNS-1. Like CNS-1-deficient T cells, Th cells lacking CD4 or Itk are defective in their capacity to differentiate to IL-4-expressing T cells in vitro and in vivo (17, 18, 21). As shown in this study, both lack basal modification of histones at the IL-4 promoter, and both more slowly begin transcription of the IL-4 gene after activation. The most parsimonious conclusion is that CD4 and Itk contribute to a signaling pathway resulting in the basal locus modification evident in Th cells, as mutations in either of these genes or in CNS-1 yield a similar Th phenotype. However, we cannot exclude the possibility that these genetic modifications independently converge to yield similar phenotypes in CD4 T cells. The delay in IL-4 gene expression by several hours after TCR activation in the various gene-deleted T cells results in a kinetic response more closely resembling transcription of the IFN- gene, an example of a locus otherwise unmodified before viral infection.
Prior studies have suggested a relatively inaccessible chromatin structure surrounding the il4 gene in resting CD4 T cells. Induction of Th2 differentiation begins a process of chromatin modifications, including histone acetylation (12, 13, 39), appearance of novel DNase hypersensitivity sites (10), and DNA demethylation (40), which subsequently spreads across the type 2 cytokine locus, enabling stabilized gene expression. Histone acetylation at IL-4 regulatory elements was the earliest detected change after TCR activation (within 40 h), although the Th2-specific, DNA-binding factor GATA3 was weakly bound to regulatory elements even in resting cells (12). Although basal modification of the IL-4 promoter was not noted using acetyl-H4 immunoprecipitation (12), low levels of resting acetyl-H3 modifications were evident in two prior studies using similar methods (12, 13). In yeast, acetylation of K9 on H3 is believed to occur by the SAGA histone acetyltransferase complex, which primarily targets promoter proximal nucleosomes. Acetylation of histone H4 is primarily due to the NuA4 histone acetyltransferase complex, which targets a broader domain of nucleosomes. Importantly, the attraction of SWI-SNF complexes to transcriptional activating complexes at proximal promoters is greatly enhanced at sites of histone acetylation, providing a mechanistic link between our finding of more rapid transcription from a previously modified promoter (41). Although numerous modifications of histone define the increasingly complex histone code, including acetylation, methylation, and phosphorylation, modifications of histone H3 at K9 by acetylation and methylation have proven to be robust markers for transcribed and silenced genes, respectively (27). In an extensive analysis of a 54-kb segment encompassing the chicken globin locus, acetylations of H3 and H4 were highly coordinated, but acetylation of H3 was quantitatively more robust (42).
We believe that the modifications we describe in resting CD4 T cells are functional, as their presence or absence correlates tightly with rapid transcription of the IL-4 gene in multiple other cell types, including CD8 T cells, mast cells, and CNS-1-, CD4-, and Itk-deficient Th cells. The intensity of immunoprecipitation in naive CD4 T cells as well as similar modification of the adjacent IL-13 promoter were consistently less than those in polarized Th2 cells, in agreement with previous observations regarding stabilization of gene expression across the locus. The lesser intensity in naive cells suggests that immunoprecipitation is probably occurring from fewer available histone acetylation sites, although we cannot exclude that the affinity/avidity of the acetyl-H3 Ab might increase with polarization. We speculate that polarization results in spreading across the locus, with creation of additional sites of acetylation on lysines and with greater numbers of sheared protein/DNA fragments available to coprecipitate the IL-4 promoter.
Alternatively, the IL-4 promoter in Th cells might represent a relatively fluid site for dynamic change in the accessibility of associated nucleosomes (43), creating an equilibrium state allowing the intermittent binding of factors, such as GATA3. Such a process would be consistent with the weak GATA3 binding detected in resting Th cells (12). Of interest, GATA4, with HNF3, was capable of mediating chromatin decondensation of the albumin enhancer, a process important in fetal hepatic development (44). GATA3 shares the DNA-binding zinc finger motifs critical for decompacting chromatin by GATA4, GATA3 sites extend throughout the type 2 cytokine cluster, including the CNS-1 element, and GATA3 binding potentiates IL-4 expression dramatically (34). We speculate that a developmental pathway in Th cells, involving signaling via CD4 and Itk by a mechanism that ultimately requires an intact CNS-1 regulatory region, is capable of establishing a relatively poised state for cytokine expression. The existence of a poised state characterized by hyperacetylated histones at the promoters of nonexpressed genes also occurs at the mouse -globin minor promoter during yolk sac development and at the erythroid-specific carbonic anhydrase II gene, defining sites primed for activation during subsequent tissue-specific differentiation (14, 15). Hyperacetylated histones may facilitate rapid recruitment and stabilization of SWI-SNF remodeling complexes that are targeted to the nucleus at the time of T cell activation (9), thus facilitating early gene transcription (37, 41).
Despite a critical role in supporting modifications establishing the primed state of the il-4 gene in naive Th cells, CNS-1 contains four Ikaros binding sites that, when absent, both attenuate in vitro enhancer activity and the in vivo repositioning of the silenced gene to heterochromatin during Th1 differentiation. Ikaros proteins have been implicated in both enhancing and inhibiting gene expression in various contexts (30). Activation of the CD8 gene in T cells is coincident with Ikaros binding, consistent with a role in active transcription (33). At the TdT gene, Ikaros proteins compete at overlapping sites with an activator protein, Elf-1, an Ets family member transcription factor (26). Conversely, the association of Ikaros with localization of silenced genes to condensed regions of pericentromeric chromatin was been convincingly established (22). In that study Ikaros is believed to play a role in gene silencing via epigenetic repositioning in apposition to centromeric heterochromatin, as demonstrated for other lymphocyte genes (22). If a similar mechanism occurs at the IL-4 locus, T cell activation may allow rapid recruitment of general transcriptional machinery to the previously modified locus, but sustained expression may rely on competition between activators and repressors that are necessary to achieve fully differentiated Th2 or Th1 cell fates, respectively. Stat6 has been suggested to have such a role in the maintenance of chromatin modifications at the IL-4 locus (12). Despite its importance in Th cells, the CNS-1 element is neither sufficient (CD8 T cells) nor necessary (mast cells) for priming the il-4 gene, suggesting a lineage-specific component in the process. Taken together, however, these data identify CNS-1 as an element critical both for the early preparation of the IL-4 locus as well as for the ultimate epigenetic repositioning of the gene, providing an exceptional region for study of factors controlling expression fates associated with lineage differentiation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Richard M. Locksley, University of California, Room C443, 521 Parnassus Avenue, San Francisco, CA 94143-0654. E-mail address: locksley{at}medicine.ucsf.edu
3 Abbreviations used in this paper: CNS-1, conserved noncoding sequence 1; SWI/SNF, ATP-dependent nucleosome remodeling complex; BAF mammalian SWI/SNF-related complex; H3, histone 3.
Received for publication April 28, 2003. Accepted for publication October 15, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
loci accompany Th1/Th2 differentiation. J. Immunol. 169:647. gene locus is regulated by the ikaros family of proteins. Mol. Cell 10:1403.[Medline]
This article has been cited by other articles:
|
|
 |
|
  H. K. Hamalainen-Laanaya, J. J. Kobie, C. Chang, and W.-p. Zeng Temporal and Spatial Changes of Histone 3 K4 Dimethylation at the IFN-{gamma} Gene during Th1 and Th2 Cell Differentiation J. Immunol., November 15, 2007; 179(10): 6410 - 6415. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. W. J. Young, O. W. Williams, D. Chandra, L. K. Bellinghausen, G. Perez, A. Suarez, M. J. Tuvim, M. G. Roy, S. N. Alexander, S. J. Moghaddam, et al. Central Role of Muc5ac Expression in Mucous Metaplasia and Its Regulation by Conserved 5' Elements Am. J. Respir. Cell Mol. Biol., September 1, 2007; 37(3): 273 - 290. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Dong, C. Ivascu, H.-D. Chang, P. Wu, R. Angeli, L. Maggi, F. Eckhardt, L. Tykocinski, C. Haefliger, B. Mowes, et al. IL-10 Is Excluded from the Functional Cytokine Memory of Human CD4+ Memory T Lymphocytes J. Immunol., August 15, 2007; 179(4): 2389 - 2396. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Strempel and D. Vercelli Functional Dissection Identifies a Conserved Noncoding Sequence-1 Core That Mediates IL13 and IL4 Transcriptional Enhancement J. Biol. Chem., February 9, 2007; 282(6): 3738 - 3746. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Hayashida, M. Oda, K. Ohsawa, A. Yamaguchi, T. Hosozawa, R. M. Locksley, M. Giacca, H. Masai, and S. Miyatake Replication Initiation from a Novel Origin Identified in the Th2 Cytokine Cluster Locus Requires a Distant Conserved Noncoding Sequence J. Immunol., May 1, 2006; 176(9): 5446 - 5454. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Koyanagi, A. Baguet, J. Martens, R. Margueron, T. Jenuwein, and M. Bix EZH2 and Histone 3 Trimethyl Lysine 27 Associated with Il4 and Il13 Gene Silencing in TH1 Cells J. Biol. Chem., September 9, 2005; 280(36): 31470 - 31477. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. L. Reiner Epigenetic control in the immune response Hum. Mol. Genet., April 15, 2005; 14(suppl_1): R41 - R46. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Baguet and M. Bix Chromatin landscape dynamics of the Il4-Il13 locus during T helper 1 and 2 development PNAS, August 3, 2004; 101(31): 11410 - 11415. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
) mice were activated with plate-bound anti-TCR
) and conditionally induced CD4-deficient mice (CD4-; 
) (
