Mouse Rankl Expression Is Regulated in T Cells by c-Fos through a Cluster of Distal Regulatory Enhancers Designated the T Cell Control Region^*

Reagents

General biochemicals were purchased from ThermoFisher Scientific (Waltham, MA), Sigma-Aldrich, or as previously described (21). Phorbol 12-myristate 13-acetate (P8139) and ionomycin (I0634) were purchased from Sigma-Aldrich. Anti-c-Fos (sc-7202), NF-κB p50 (sc-1190), and Nfat-pan (sc-7294) antibodies were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Anti-acetyl histone H4 (06-866) and anti-acetyl histone H3 Lys9 (06-942) antibodies were purchased from Millipore (Billerica, MA). Monomethyl histone H3 Lys-4 (ab8895–50) and trimethyl histone H3 Lys-4 (ab1012–100) antibodies were obtained from Abcam (Cambridge, MA). The mice were obtained from Harlan Laboratories (Madison, WI), and anti-RNA polymerase II 8WG16 (MMS-126R) was purchased from Covance (Princeton, NJ). U0126 (662005) was purchased from Calbiochem (San Diego, CA), cyclosporin A (BML-A195–0100) was obtained from Enzo Life Sciences (Farmingdale, NY), and EasySep mouse CD4⁺ T cell enrichment kits (19752) were purchased from Stem Cell Technologies (Vancouver, Canada). RNeasy Plus mini kits (74134) were purchased from Qiagen, and RP1640 (15-040-CV) was obtained from Mediatech (Manassas, VA). CD3e (553057) and CD28 (553294) antibodies were obtained from BD Biosciences (San Jose, CA). RPMI 1640 (11875) and RiboMinus eukaryote kit for RNA-Seq (A1083708) were purchased from Invitrogen. Small RNA sample prep kit v1.5 (FC-102-1009), mRNA-Seq prep kit (RS-100–1801), cBot Single Read Cluster Generation Kit (GD-300-1001), and the Illumina Sequencing Kit v4 (FC-104-4001) were purchased from Illumina (San Diego, CA).

Cell Culture and Isolation of Primary T Cells

2b4.11 cells were cultured in RPMI 1640 (Invitrogen) with 10% fetal bovine serum (heat-inactivated), 50 μm β-mercaptoethanol, and 100 μg/ml gentamicin. Jurkat cells were cultured in RPMI 1640 (Mediatech) with 10% fetal bovine serum (heat-inactivated), 10 mm Hepes, 2.5 mm glutamine, 100 units/ml penicillin, and 100 μg/ml streptomycin. Mouse ST2 osteoblastic cells were cultured in α-minimum essential medium supplemented with 10% fetal bovine serum (heat-inactivated), 100 units/ml penicillin, and 100 μg/ml streptomycin (21). Primary mouse CD4⁺ T cells were isolated from the spleen or whole blood of C57BL6/NHsd mice using the EasySep® mouse CD4⁺ T cell enrichment kit according to the manufacturer's protocol. The cells were cultured in RPMI 1640 (Invitrogen) with 10% fetal bovine serum (heat-inactivated).

In Vitro and ex Vivo T Cell Activation

T cells were activated using 500 ng/ml ionomycin and 10 ng/ml phorbol 12-myristate 13-acetate (PMA) or through CD3/CD28 antibody activation. In the latter condition, the antibodies were first adsorbed overnight to ELISA plates using a solution of 10 μg/ml mouse CD3e antibody (clone 145–2C11) and 2.5 μg/ml CD28 antibody (clone 37.51). The wells were then blocked with 1 mg/ml BSA, and isolated cells were seeded at 0.2 million cells/well.

RNA Isolation and Analysis

RNA was isolated using two methods: 1) total RNA was isolated from cells using Tri-Reagent and DNaseI-treated, or 2) RNA was isolated using the RNeasy Plus mini kit according to the manufacturer's protocol. RNA was reverse-transcribed using the high capacity cDNA reverse transcription kit. The resulting cDNA was then subjected to quantitative PCR analysis.

Quantitative PCR Analysis

Real time PCR was performed on either an Eppendorf Realplex or ABI StepOnePlus using Power SYBR Master Mix with standard cycling conditions. The Mastercycler® ep Realplex software or StepOne Software was used for analysis. The primers used in the RNA analyses include: mouse Rankl mRNA, 5′-TGTACTTTCGAGCGCAGATG-3′ (forward) and 5′-AGGCTTGTTTCATCCTCCTG-3′ (reverse); mouse β-actin mRNA, 5′-TGTTTGAGACCTTCAACACCC-3′ (forward) and 5′-CGTTGCCAATAGTGATGACCT-3′ (reverse); mouse IL-2 mRNA, 5′-GCAGGATGGAGAATTACAGGA-3′ (forward) and 5′-TTGGCACTCAAATGTGTTGTC-3′ (reverse); mouse c-Fos mRNA, 5′-CGAAGGGAACGGAATAAGATG-3′ (forward) and 5′-GCTGCCAAAATAAACTCCAG-3′ (reverse).

RNA-Seq Analysis

Total RNA was isolated from cells using Tri-Reagent and then treated with DNaseI. Ribosomal RNA was depleted using RiboMinus eukaryote kit for RNA-Seq. rRNA depleted RNA (2 μg) was fragmented and a cDNA library prepared according to the Illumina Directional mRNA-Seq Library Prep protocol using the small RNA v1.5 and mRNA-Seq prep kits. Quality of cDNA library was assessed using the Agilent 2100 Bioanalyzer and high sensitivity DNA kit. DNA clusters were generated using a cBot Single Read Cluster Generation kit on an Illumina cBot and included the smRNA/GEX-DpnII sequencing primer in lieu of the typical genomic sequencing primer as described in the manufacturer's instructions, to obtain an average of 2.0 × 10⁷ clusters for each flow cell lane. 75-bp reads of each library were performed on an Illumina Genome Analyzer IIX using the Illumina Sequencing kit (version 4). Fluorescent images were analyzed using the Illumina base-calling pipeline 1.6.0 to obtain FASTQ formatted sequence data. The sequences were mapped to the human genome (March 2006 assembly NCBI36/hg18) using Bowtie (Bowtie 0.12.5, mismatch = 3, solexa1.3-quals on) (28). Intron gaps and splice junctions were predicted and mapped by Tophat (Tophat 1.0.14, A spliced read mapper for RNA-Seq) (29). The data were further processed by ArrayStar 4.0 with Q-seq module (DNA-STAR Madison, WI) and entered into the UCSC Genome Browser.

Chromatin Immunoprecipitation-Microarray Analysis

ChIP analysis was preformed as previously described (23, 24, 30, 31). Immunoprecipitated DNA was blunt-ended by T4 DNA Polymerase, ligated to linkers with the sequence 5′-GAATTCAGATC-3′ and 5′-GCGGTGACCCGGGAGATCTGAATTC-3′ using T4 DNA ligase, and amplified by ligation-mediated PCR with Go Taq Flexi and linear PCR amplifications. PCR purification was performed with the QIAquick PCR purification kit. DNA samples were labeled with Cy3 or Cy5 9-mer wobble primers using Klenow fragment, and the reaction was then stopped with the addition of EDTA, precipitated with NaCl and isopropanol, washed with 80% EtOH, and then resuspended in water. Cy3- and Cy5-labeled DNA samples as indicated were co-hybridized to a custom oligonucleotide microarray using a NimbleGen hybridization kit and a MAUI hybridization system. Microarrays were washed using NimbleGen Wash Buffer and scanned using an Axon 4000B scanner with GenepixPro version 4.1 software at the appropriate wavelengths. Custom oligonucleotide microarrays were synthesized by Roche-NimbleGen Systems (Madison, WI). The microarray oligonucleotide probes were 50- to 70-mer in length with 65–70 bp resolution and synthesized using a mask-less array system tiled from 200 kb upstream of the mouse Rankl TSS to 200 kb downstream of the final 3′ coding exon. This custom array contained additional tiled genes not considered in this study. DNA samples were obtained by immunoprecipitation with antibodies specific for c-Fos, Nfat, p50, CBP, p300, H4K(5,8,12,16)ac, H3K9ac, H3K4me1, H3K4me3, or RNA polymerase II. A series of comparisons were used: 1) untreated and input control and 2) PI-treated and input control. The aforementioned co-hybridizations were analyzed and presented as previously described (24). NimbleScan version 1.9.0.05 was used for statistical analysis of data; a sliding window of 700 bp containing at least four of eight probes statistically above background was deemed significant at a false discovery rate (FDR) of <0.05.

Plasmids

The reporter constructs pTK mRL-D1, -D2, -D3, -D4, -D5a, -D5b, and -D6 were previously described (21, 24, 32). Similar methods were used to prepared pTK mRL-T1A (−122986 to −124293 nucleotides), mRL-T1B (−126338 to −127714 nucleotides), mRL-T1C (−129438 to −130831 nucleotides), mRL-T2 (−139923 to −141306 nucleotides), mRL-T3A (−150408 to −152161 nucleotides), and mRL-T3B (−152870 to −154315 nucleotides) using primers containing HindIII/BamHI/SalI restriction sites. Digested fragments were cloned into the corresponding sites within the pTK-luc vector. The QuikChange mutagenesis kit was used to introduce mutations into the pTK-mRL-D5b and mRL-T1A reporter constructs.

Reporter Transfections

ST2 cells were transfected using Lipofectamine as previous described (24). Jurkat cells (50,000 cells/well in 24-well plates) were transfected with 250 ng of reporter construct and 200 ng of RSV-βgal in serum-free medium using Lipofectamine and Plus reagent. After a 3-h incubation, the wells were supplemented with RPMI 1640 containing 20% FBS to a final concentration of 10% FBS and treated with 500 ng/ml ionomycin and 10 ng/ml PMA or Me₂SO vehicle. The cells were harvested 20 h later for reporter assays using lysis buffer, and both luciferase and β-galactosidase activities were determined as previously described (23, 24). For inhibitor studies, the cells were treated with transfection reagent as above followed by a 30-min exposure to either U0126 or cyclosporin A (CSA) in medium containing 10% serum, and then treatment with 500 ng/ml ionomycin and 10 ng/ml PMA or Me₂SO vehicle for an additional 20 h.

Chromosome Conformation Capture (3C) Assay

3C was performed as previously described (33–35). Briefly, 2b4.11 cells were grown to a density of 500,000 cells/ml and then treated with vehicle or 500 ng/ml ionomycin and 10 ng/ml PMA for 3 h. The cells were cross-linked for 10 min with 1% formaldehyde and then quenched with 0.125 m glycine. The nuclei were collected and disrupted using 20 strokes of a Dounce homogenizer. The nuclear suspension was divided into two tubes, washed, and resuspended in NEBuffer 3. SDS was added to 0.1% final concentration, the nuclei were incubated at 65 °C for 10 min, and Triton X-100 was added to 1% final concentration. Digestion was performed with 400 units of BtgI overnight at 37 °C. Enzyme was inactivated by addition of 2% final concentration SDS and further incubated at 65 °C for 30 min. The samples were diluted 10-fold (to 8 ml) and incubated with or without T4 DNA ligase for 4 h at 16 °C and an additional 30 min at room temperature. Proteinase K (100 μg) was added, and the samples were incubated overnight at 65 °C. The DNA was purified through two rounds of phenol:chloroform:isoamyl alcohol (25:24:1) extractions and a final ethanol precipitation. Final DNA was redissolved in TE buffer (10 mm Tris-HCl, pH 8.1, 1 mm EDTA) and treated with 4 μg of RNase A. As PCR product size control, DNA from BAC clones RP23-68C18 and RP23-52A3 spanning the entire Rankl gene locus were digested overnight with BtgI, religated, and then purified. This sample contained all possible Rankl PCR product combinations. PCR was performed using Go-Taq Flexi Green Master Mix for 35 cycles. The primers were designed on the forward genomic strand in proximity to the BtgI cut site. Analysis was performed from BtgI segments located between the distal elements and proximal promoter.

Statistical Analyses

All of the values were expressed as the means ± S.E. of the mean. All of the statistical calculations were performed using GraphPad PRISM version 4 statistical software package (GraphPad Software Inc., San Diego, CA) using either one-way nonparametric analysis of variance analysis with a Bonferroni post-test or nonparametric T tests as indicated. Activated samples were contrasted with vehicle-treated controls or other indicated samples.

Rankl mRNA Levels Increase with T Cell Activation

It is well established that Rankl gene expression is increased upon T cell activation both ex vivo and in vitro (1, 36, 37). To confirm these findings and to establish a model for further molecular exploration, we treated splenic CD4⁺ T cells with antibodies to CD3/CD28 from 0 to 72 h and then examined the levels of Rankl, IL-2, and β-actin mRNA by quantitative RT-PCR analysis. As shown in Fig. 1A, Rankl transcripts were rapidly and transiently induced with maximum expression occurring at 3 h followed by a return to base line at 24 h. This profile was similar to that for the IL-2 transcripts with respect to the time course of induction, but IL-2 expression remained highly elevated for the duration of the experiment (Fig. 1B).

We also analyzed the mouse 2b4.11 T cell hybridoma for its utility as an in vitro model to study T cell activation of Rankl expression (36). 2b4.11 cells were seeded into plates with or without plate-bound CD3/CD28 antibodies, and the concentrations of Rankl and β-actin mRNA were examined up to 48 h later. As in CD4⁺ T cells, Rankl transcripts were transiently induced as documented in Fig. 1C, with maximum expression at 6 h. Finally, we asked whether Rankl was induced with PMA/ionomycin, a classic inducer of T cell activation. 2b4.11 cells were treated with either Me₂SO vehicle or both ionomycin (500 ng/ml) and PMA (10 ng/ml) (PI) for 6 h, and the isolated RNA was then evaluated. As seen in Fig. 1D, Rankl transcription was induced to a level comparable with that observed with CD3/CD28 activation, thereby providing an additional in vitro model system for studying Rankl gene expression.

ChIP-chip Analysis of Acetylated Histone H4/H3 and RNA pol II Is Used to Identify Putative T Cell Enhancer Regions in Primary Mouse CD4⁺ T Cells and the 2b4.11 Cell Line

Osteoblasts control Rankl expression via a series of distal upstream enhancers (21, 23). Many of these enhancers are marked under basal conditions by elevated levels of H4ac and increased RNA pol II density; these activities were increased following induction by 1,25(OH)₂D₃, PTH, or IL-6 (21, 23, 24). These genetic features are now considered generally typical of functional enhancers (38). In an effort to identify enhancers responsible for Rankl expression in T cells, we conducted a ChIP-chip analysis of the levels of H4ac and histone H3 lysine 9 acetylation (H3K9ac) and of RNA pol II density across the Rankl gene locus in unstimulated 2b4.11 and primary CD4⁺ T cells and contrasted the results with those found in unstimulated osteoblastic ST2 cells. Chromatin was isolated and immunoprecipitated using antibodies to H4ac, H3K9ac, or RNA pol II, and the resulting DNA was co-hybridized with control input DNA overnight to custom microarrays. Fig. 2A depicts the Rankl gene locus on chromosome 14 (February 2006 assembly) with the nucleotide positions (megabases) indicated. Fig. 2B depicts the data tracks representing the log₂ ratio of fluorescence obtained from a vehicle-treated sample precipitated with an antibody to tetra-acetylated histone H4 and co-hybridized with input DNA. Fig. 2 (C and D) depicts similar tracks from DNA precipitated with an antibody to either H3K9ac or RNA pol II. As shown in Fig. 2B, elevated levels of basal H4ac were found upstream of the Rankl TSS in ST2, 2b4.11, and mouse CD4⁺ T cells. Statistically significant elevation of this epigenetic mark was also found in unstimulated ST2 cells near the TSS and across many of the enhancer regions that were described previously (21), although these activities are relatively low in the unstimulated state. Elevated levels of H4ac were also found at significant levels in both untreated 2b4.11 and CD4⁺ T cell populations. Accordingly, although several of the enhancers discovered in the ST2 cells were also marked with H4ac in 2b4.11 cells, an extended region of particularly high H4ac was observed even further upstream in 2b4.11 cells between −123 and −155 kb that was largely absent in ST2 cells. Elevated levels of H3K9ac were also found similarly in both 2b4.11 and primary CD4+ cells (Fig. 2C); increased RNA pol II density was also observed in all three of the cell types (Fig. 2D). The activities of both H3K9ac and RNA pol II were most strikingly observed in the region upstream of −123 kb and restricted to the 2b4.11 and CD4⁺ T cells, suggesting strong similarity between the two cell types within this locus. Because of its unique nature in T cells, we tentatively designated this region the TCCR. These data also suggest that osteoblasts and T cells both may utilize a common enhancer such as mRL-D5 and cell type-specific enhancers such as the TCCR to regulate Rankl gene expression. Surprisingly, PI activation of 2b4.11 cells only modestly altered H4ac and H3K9ac levels and RNA pol II density profiles, suggesting that these cells show some degree of residual activation (data not shown).

Enhanced levels of H4ac and H3K9ac, as well as increased RNA pol II density across the TCCR in T cells, suggest the presence of at least three distinct regulatory regions, as depicted in Fig. 2E. To explore these regions in more detail, we conducted a 17-way vertebrate alignment and generated a mouse versus human Vista plot of the TCCR region using the UCSC Genome Browser to assess the degree of conservation located within this cluster of enhancers. As can be seen, each of the three regions within the TCCR contained components that were highly conserved both across species as well as between mouse and human genomes. These three regions within the TCCR were designated T1, T2, and T3.

Putative Rankl T Cell Enhancers Are Not Alternative Exons of the Rankl Gene

The mouse Rankl transcriptional unit is comprised of a single promoter that directs the synthesis in osteoblasts of five highly conserved and spliced exons; alternative or upstream promoters have not been identified in mouse osteoblasts. To confirm this in T cells, we employed high throughput RNA sequencing (RNA-Seq) analysis to characterize the Rankl RNA transcript and to assess whether components upstream could be identified that were spliced to the Rankl gene itself. 2b4.11 cells were treated with either vehicle or PI, and the isolated RNA was then fragmented, reversed transcribed, and analyzed using RNA-Seq analysis. Read counts for the interval 77.000–77.275 megabases that contain both the Rankl and adjacent Akap11 genes are shown in Fig. 3A. Spliced RNA was detected across each of the five exons of the Rankl gene under both basal and PI-induced conditions; induction was ∼11-fold, a level consistent with that observed via RT-PCR analysis (Fig. 1D). Interestingly, very low levels of RNA transcripts were also detected upstream of the Rankl promoter, including one spliced transcript that appeared to originate from the T1 enhancer within the TCCR; none of these, however, were found to be either spliced to the downstream Rankl transcript or regulated by PI. Inspection of the data more broadly also revealed that none of the genes located within 1 megabase of the Rankl transcriptional unit were similarly regulated by PI as well. These data suggest that the epigenetic and RNA pol II activities observed within the TCCR region do not represent alternative Rankl promoters, although the possibility that an unannotated and unregulated gene(s) may be present within the extended TCCR region cannot be ruled out.

MEK1/2 and Calcineurin Activation Are Necessary for Activation-induced Rankl Expression

To gain further evidence of a role for the TCCR in Rankl expression, we next examined the effect of T cell activation on signaling pathways and transcription factor activity at the Rankl locus. T cell activation is known to stimulate calcium signaling, PKC, and MAPK pathways that ultimately activate specific regulatory factors that include AP-1, NF-κB, and Nfat. Previous studies suggest that Rankl expression in activated T cells and the 2b4.11 cell line is suppressed by inhibitors of the calcineurin/Nfat activation pathway such as FK506 and CSA (1, 39, 40). To confirm this finding and to explore the impact of the MAPK pathway on Rankl expression, 2b4.11 cells were pretreated with either Me₂SO vehicle, CSA, or the MEK1/2 inhibitor U0126 and then treated with vehicle or PI for an additional 4 h. Total RNA was isolated, and Rankl and IL-2 mRNA levels were examined using quantitative RT-PCR. PI-induced Rankl mRNA levels were suppressed with both CSA (Fig. 4A) and U0126 (Fig. 4C) pretreatment. As was observed in osteoblasts, U0126 treatment also suppressed basal Rankl mRNA levels as well (41). T cell activation-induced IL-2 gene expression was also suppressed with both CSA and U0126 treatments (Fig. 4, B and D), a finding consistent with previous studies (42, 43).

IL-2 gene up-regulation in activated T cells is modulated by AP-1, NF-κB, and Nfat (42, 43). Given the similarities between both Rankl and IL-2 induction, we next conducted a ChIP-chip analysis of 2b4.11 cells following either vehicle or PI treatment using antibodies to c-Fos, Nfat, or the p50 subunit of NF-κB. Fig. 5 depicts data tracks representing the log₂ ratios of fluorescence at the mRL-TSS, mRL-D5, and TCCR regions for c-Fos or NF-κB p50 binding activity. As can be seen, both c-Fos and NF-κB localized to the putative enhancer within the Rankl gene locus explored in this study as well as other upstream regions that are associated with increased H4ac levels and RNA pol II density (supplemental Fig. S1); however, c-Fos occupancy was particularly increased across the TCCR. Nfat occupancy was not detected, despite the appearance of this factor at several other gene loci present on the tiled array. The presence of c-Fos and NF-κB at these sites was validated using direct ChIP analysis (data not shown). The appearance of NF-κB near the Rankl promoter confirms a previous observation made using a transfected Rankl promoter-reporter gene constructs (36). These studies suggest that T cell activation prompts the appearance of both c-Fos and NF-κB at sites within the Rankl locus and with respect to c-Fos, across the TCCR that are likely responsible for inducing the Rankl gene. Similar studies revealed the presence of both CBP and p300 at these sites as well, but the recruitment of these acetyltransferases was not induced at the Rankl putative enhancers regions during T cell activation (supplemental Fig. S2). Selectively enhanced levels of H3K4me1 and H3K4me3 activity that mark enhancer and promoter regions, respectively, are also seen (supplemental Fig. S2).

The mRL-D5 and TCCR Enhancer Regions Show Strong Transcriptional Activity in Reporter Assays

Enhancers that modulate transcription of the Rankl gene in osteoblasts are highly conserved across multiple species and capable of transcriptionally regulating reporter activity when cloned upstream of heterologous promoters (21–24). We therefore utilized this approach to assess whether the previously identified osteoblast enhancers or individual segments of the TCCR cluster (T1A-C,T2, T3A, and B) could mediate transcriptional regulation of reporter gene activity and whether these effects were precipitated by c-Fos or NF-κB following T cell activation. Accordingly, each of these individual constructs were transfected into human Jurkat T cells and reporter activity was assessed following T cell activation (these cells also express Rankl, and the locus is similarly responsive to PI activation; data not shown). Although different basal activities were evident, mRL-D5b, the mRL-T1A, mRL-T1C, and mRL-T3A reporter constructs all showed inducible reporter activity in response to PI treatment when compared with vehicle-treated samples, as documented in Fig. 6 (A and B). In contrast, none of the TCCR constructs themselves mediated transcriptional response to 1,25(OH)₂D₃, PTH, or IL-6 (data not shown), strong inducers of the Rankl enhancers mRL-D1 through mRL-D6 in ST2 cells. These data provide support for the idea that mRL-D5 together with enhancer elements of the TCCR may play a central role in Rankl induction in T cells but not in osteoblasts. Further exploration of the mRL-D5b, mRL-T1A, and mRL-T1C enhancers revealed that U0126 inhibited the PI inducible activity of all the constructs, whereas inhibition by CSA was restricted to the mRL-T1A and mRL-T1C constructs (Fig. 6C). This latter observation suggests that certain activities of the mRL-D5b enhancer may be unique relative to those seen at the mRL-T1A and mRL-T1C enhancers.

To assess whether c-Fos and NF-κB binding to the mRL-D5b region and to enhancers within the TCCR play a functional role in regulating transcription, we used an in silico approach to search for potential AP-1, NF-κB, and Nfat binding sites within the mRL-D5b and mRL-T1A regions (44, 45). Two highly conserved AP-1 and several Nfat consensus binding sites were identified in each region, as illustrated in Fig. 7; no NF-κB sites were uncovered. Interestingly, one of the sites within the T1A region was reminiscent of a composite AP-1 and Nfat site found within the IL-2 promoter (Fig. 7A) (46). To determine whether these sites were functional, we mutated each individually or in combination and then assessed the consequence of this activity on both basal and PI-inducible mRL-D5b or mRL-T1A enhancer activity following transfection into Jurkat cells. As documented in Fig. 7B, only mutation of the AP-1B site in mRL-D5b suppressed PI-inducible reporter activity. Interestingly, this AP-1 binding site was located immediately adjacent to two cAMP response elements and one Runx2 binding element that were shown previously to be important for PTH regulation of Rankl expression in osteoblasts (23). Mutation of AP-1A and the single Nfat sites in mRL-D5b was without effect and therefore consistent with the observation made in Fig. 6C that CSA did not prevent activation of mRL-D5b and was therefore unlikely to involve Nfat activation. Interestingly, mutation of either AP-1C or the adjacent Nfat site within the mRL-T1A enhancer reduced basal activity and prevented induction by PI. These results recapitulate Rankl mRNA data suggesting that AP-1 mediates the inducible reporter activity of both mRL-D5b and the mRL-T1A enhancers. They also indicate that despite our inability to detect its presence in the ChIP-chip analysis, Nfat or related transcription factor activation may be involved in transcriptional regulation through the putative mRL-T1A enhancer.

Both mRL-D5 and the TCCR Enhancers Are Located in Close Proximity to the Rankl Promoter

Long distance regulatory enhancers are thought to regulate their target genes via complex chromatin looping mechanisms. To determine whether looping might provide a mechanism for Rankl activation via the mRL-D5 and the TCCR regions, we treated 2b4.11 cells with either vehicle or PI and used 3C analysis to assess potential interactions between the promoter and enhancers as described under “Experimental Procedures.” The analysis was performed using the BtgI restriction enzyme as illustrated in Fig. 8A. As can be seen in Fig. 8B, BtgI-digested fragments located in or near the TSS, mRL-D5, mRL-T1, mRL-T2, and mRL-T3 regions region were all shown to undergo an enhanced frequency of ligation with a DNA fragment that spanned the Rankl promoter, suggesting that each of these segments had been linked via the initial cross-linking step. None of these interactions were sensitive to PI treatment. Control regions located between these enhancers did not show enhanced ligation. These data suggest that the mRL-D5 region and segments within the TCCR are located in close proximity to the Rankl promoter as depicted in the cartoon in Fig. 8C and are thus potentially able to regulate transcription at the TSS.