Previously, we analyzed the proximal IL-4 promoter in directing Th2-specific activity. An 800-base pair proximal promoter conferred some Th2-selective expression in transgenic mice. However, this region directed extremely low reporter mRNA levels relative to endogenous IL-4 mRNA, suggesting that full gene activity requires additional enhancer elements. Here, we analyzed large genomic IL-4 regions for enhancer activity and interaction with transcription factors. The proximal IL-4 promoter is only moderately augmented by GATA-3, but certain genomic regions significantly enhanced GATA-3 promoter transactivation. Some enhancing regions contained consensus GATA sites that bound Th2-specific complexes. However, retroviral transduction of GATA-3 into developing T cells induced IL-5 to full Th2 levels, but only partially restored IL-4 production. Thus, we propose that GATA-3 is permissive, but not sufficient, for full IL-4 enhancement and may act through GATA elements surrounding the IL-13/IL-4 gene locus.
Thelper type 2 cells selectively express IL-4, IL-13, and IL-5 (1, 2). The transcriptional basis for Th2 cytokine expression has been proposed to involve two Th2-specific transcription factors, c-Maf and GATA-3 (3, 4). GATA-3 is important for embryonic development and T cell development (5). GATA-3 deficiency is embryonically lethal, and GATA-3-deficient T cells do not complete normal thymocyte maturation (6). Recently GATA-3 was found to be selectively expressed in naive T cells and Th2 cells, but extinguished in Th1 cells (3, 7). Two recent studies analyzed GATA-3 in Th2 cytokine expression (3, 7). The first proposed that GATA-3 is necessary and sufficient for IL-4 production by CD4+ T cells (3). The second suggested, however, that GATA-3 acts directly to induce IL-5 promoter activity via a GATA element upstream of the CLEO element (7).
While previous studies of IL-4 gene regulation in T cells focused on the proximal 800 base pairs (8, 9, 10, 11, 12, 13, 14, 15), this promoter region is not sufficient to direct full IL-4 gene activity (16). The 800-bp promoter was able to direct some degree of Th2-selective expression in several transgenic reporter lines. However, the amount of reporter mRNA was virtually undetectable compared with the levels of endogenous IL-4 mRNA. Thus, other regulatory elements outside of the 800-bp proximal promoter are required to direct full IL-4 gene activity (16).
By scanning the IL-4/IL-13 locus, we identified several genomic regions having enhancer activity that increase IL-4 promoter transactivation by GATA-3. GATA elements from these regions bind Th2-specific complexes. Using retroviral transduction of GATA-3 into naive T cells, we show that GATA-3 induces IL-5 expression to levels equivalent with Th2 controls, but does not similarly restore full IL-4 expression. This suggests that GATA-3 is not sufficient for full IL-4 gene activity in Th2 cells, and in fact, may be one permissive factor required to enhance IL-4 expression through distal sites rather than acting exclusively within the IL-4 proximal promoter.
Materials and Methods
Cell culture and transfection
M12 cells were maintained in Iscove’s modified Dulbecco’s medium and Jurkat in RPMI 1640 supplemented as described (8). M12 cells (107) or 2 × 107 Jurkat cells were electroporated as described (8) with 20 μg/ml of the indicated reporter plasmid, 20 μg/ml of the expression plasmid, and 0.5 μg/ml pRL-CMV (Promega, Madison, WI) to monitor transfection efficiency. M12 cells were transfected as described (8), and Jukat cells were transfected at 300 V, 960 μF. Cells were harvested at 16 h and left unstimulated or stimulated with 50 ng/ml PMA and 1 μM ionomycin. Cells were lyzed at 4 h in 100 μl of 10 mM KH2PO4/500 mM NaCl/1 mM EDTA, pH 8, and lysates were assayed for firefly and renilla luciferase activities.
Cloning and plasmid construction
An 86-kb genomic P1 clone was identified as positive by PCR for IL-4 and IL-13 (Genome Systems, St. Louis, MO). The clone was sequenced (InCyte, Palo Alto, CA) at 12-fold redundancy to yield four contiguous DNA sequences. All BamHI fragments were subcloned upstream of the 800-bp IL-4 luciferase reporter construct (IL-4 Luc, here named −800 Luc).3 One 20-kb fragment was further subcloned as five BamHI/HindIII or HindIII fragments into −800 Luc (see Fig. 3). Restriction digest and Southern mapping confirmed the order and position of all fragments. Trimerized sites containing the IL-2 promoter NF-AT consensus (3×NFAT) and 202-bp IFN-γ promoter (IFN-γ Luc) were placed into the luciferase plasmid pBS-LUC (8).
Electrophoretic mobility shift assays (EMSA)
EMSAs were performed as described (8). Nuclear extract was incubated (6.6 μg) in a 10-μl reaction with 5 × 104 cpm of probe, 0.15 mg/ml BSA, 1 μg poly(dI:dC). After 30 min at 4 C, complexes were electrophoresed through 4.5% polyacrylamide at 4 C in 0.4× TBE for 4 h at 150 V. Fully annealed EMSA probes were as follows: probe 1, GAGAAATGATAAATGATAAGAAAAGTTGAAGAAC; probe 2, GTAACAGAGTGATAGGAGATAGATACAATCAGCC; probe 3, GGTGTAATAGATAATTGGAGCAGGCTGGCC; probe 4, CAACCCTACGCTGATAAGATTAGTCTGAAAG; probe 5, TGTGATAGAAACCCAGGAGGCCCAAAGGAGTGCT; and UTR, TGCATTGTTAGCATCTCTTGATAAACTTAATTGTCT.
The retroviral vector GFP-RV was made by placing the 600-bp EcoRI-NcoI EMCV IRES fragment from pCITE-1 (Novagen, Madison, WI) upstream of the 700-bp NcoI-EcoRI humanized green fluorescent protein (GFP) allele (hGFP-S65T; Clontech, Palo Alto, CA) in a trimolecular ligation into the EcoRI site of pBSSKII(−) (Stratagene, La Jolla, CA). The 1.3-kb XhoI/BamHI IRES-hGFP cassette was cut from pBSSKII(−) ligated into the XhoI/BamHI site of the MSCV2.2 retroviral vector, replacing the 1.3-kb phosphoglycerate kinase-neomycin cassette. A BglII/SalI GATA-3 cDNA PCR fragment was ligated into the unique BglII and XhoI sites of GFP-RV to produce GATA3-RV. Phoenix-Eco packaging cells (Dr. G. Nolan, Stanford, CA) were transfected according to Dr. Nolan’s online protocol (http://www-leland.standford.edu/group/nolan/NL-phnxr.html). DO11.10 T cells were activated as described and infected with retroviral supernatant on day 2. On day 7, infected T cells were purified by sorting for GFP expression to a purity of >95% GFP-positive following postsort analysis.
Western blot analysis
T cells (107) were stimulated under the indicated conditions, lysed in 30 μl of cell lysis buffer (5% SDS, 0.5 M Tris, pH 6.8, 0.5 mM EDTA, 1 mM DTT, 1 mM PMSF, 10 μg leupeptin) for 10 min at room temperature, and centrifuged at 100,000 × g for 10 min. Supernatants were resolved by SDS-PAGE, transferred to nitrocellulose (Bio-Rad, Hercules, CA), and probed with murine monoclonal anti-GATA3 (1:3000; Santa Cruz Biotechnology, Santa Cruz, CA) , Goat anti-mouse (1:2000; Santa Cruz Biotechnology) and developed by enhanced chemiluminescence (ECL; Amersham, Arlington Heights, IL).
Results and Discussion
Dose-dependent regulation of IL-4 promoter by GATA-3
We examined the effects of GATA-3 in both M12 and Jurkat cells for two different IL-4 promoters (−157 Luc and −800 Luc reporters). Using 20 μg of GATA3-RV expression plasmid (GATA3-RV), we observed ∼8- to 12-fold augmentation in M12 cells and only 4- to 6-fold in Jurkat cells for both reporters (Fig. 1). We observed a dose-dependent effect of GATA-3 expression for this augmentation (Fig. 1,B), because when using 5 μg rather than 20 μg of GATA3-RV, the augmentation was not significant. The level of augmentation seen in both M12 and Jurkat cells was significantly lower than previously reported for the IL-4 promoter (3) using the same M12 system. To determine whether the GATA-3 actions were specific to the IL-4 promoter, we examined four other cytokine reporters, as well as an NF-κB and an NF-AT reporter construct, for activation by GATA-3 (Fig. 2,A). The IL-2 reporter (8) was unaffected by GATA-3, while the IFN-β, NF-κB (17), and IFN-γ reporters were augmented twofold. As noted above, the −800 Luc reporter showed ∼6-fold activation by GATA-3, as did the trimerized NF-AT site reporter construct (Fig. 2 A).
A single WGATAR motif resides at −274 to −269 in the IL-4 proximal promoter (Fig. 3,A, probe 4), which binds a Th2-specific complex that is shifted by the anti-GATA3 Ab HG3–31 (Fig. 3,C). To test the role of this element in augmenting the IL-4 promoter, we examined a series of promoter truncations and deletion constructs (Fig. 2,B). Removal of this GATA element, either by 5′ truncation (DM3) or internal deletion (ID5), reduced responsiveness to GATA3-RV only slightly (Fig. 2,C), suggesting that this single GATA element is not the only element required for enhancement of IL-4 reporter activity. A sequence between −112 to −107 which was recently suggested (7) to be a potential GATA site has already been shown by supershift analysis to interact with NF-Y (8) rather than GATA-3. A final GATA sequence is present at +19 to +24 of the 5′ untranslated IL-4 promoter region. However, this sequence does not interact with a complex characteristic of GATA-3, since it is not shifted by HG3-31 and is present in both Th1 and Th2 cells (Fig. 3,C). Also, removal of this GATA element from the context of the −157 Luc reporter only slightly reduced the degree of GATA-3 inducibility (Fig. 2 C).
In summary, while we confirm that GATA-3 augments IL-4 reporter activity, we observed substantially lower levels of transactivation than reported earlier. Our deletion/mutation analysis suggests that the consensus GATA elements present between −741 and +60 of the IL-4 promoter may not represent the relevant or only targets for this transactivation. In fact, a recent study showed that that GATA-3 and GATA-2 exhibit distinct DNA-binding properties from GATA-1 based on their unique N-terminal zinc finger domains, (18), suggesting that perhaps nonconsensus GATA elements may need examination in future studies.
Identification of regions that enhance IL-4 promoter activity
Since the 800-bp promoter was insufficient to direct reporter expression in transgenic mice to the levels of endogenous IL-4 (16), we searched for enhancer activity within a 45-kb region spanning the IL-13 and IL-4 genes (Fig. 3,A). Several regions directly increased reporter activity of the −800 Luc reporter (Fig. 3,B, open bars). However, more significant increases were seen in response to transactivation by GATA-3 (Fig. 3,B, closed bars). The two most active fragments (A and B) contained sequences with consensus WGATAR motifs, both upstream and downstream of the IL-13 gene. Two other active fragments (C and D) however had no GATA motifs, suggesting that other factors might participate in their enhancer activity. Notably, the least active region was that immediately upstream of the IL-4 promoter (E). Several predicted GATA motifs from A and B bind Th2-specific complexes in EMSA (Fig. 3 C). Thus, the Th2-specific expression of GATA-3 could potentially influence IL-4/IL-13 gene expression as an enhancer binding factor that augments gene expression from sites at a distance from the promoter rather than acting exclusively within the promoter region.
GATA-3 does not restore IL-4 gene expression
We wished to test the role of GATA-3 in the context of normal CD4+ T helper development rather than in transformed cells. We used retrovirus to direct GATA-3 expression independently of its normal extinction during Th1 development (Fig. 4,A) to allow us to ascertain whether GATA-3 is necessary and sufficient for IL-4 gene expression in normal T cells. We infected GATA-3-expressing or control retrovirus into T cells and induced either Th1 or Th2 development (Fig. 4,B). GATA-3-transduced or control T cells were analyzed for effects on cytokine expression. Forced GATA-3 expression in Th2 cells did not increase IL-4 production, but nearly doubled IL-5 production (Fig. 4, C and D). However, GATA-3 expression in Th1 cells led to only a 30% restoration of IL-4 relative to the Th2 control level, but led to a full restoration of IL-5 production. Western analysis showed that the level of GATA-3 achieved by retroviral expression is equivalent to that of a Th2 control (Fig. 4 E), indicating that lowered cytokine levels of the transduced Th1 cells is not simply a result of lower GATA-3 levels. Thus, in the context of normally developing Th cells, GATA-3 appears to influence the expression of IL-5 more strongly than that of IL-4. GATA-3 appears to be sufficient to induce full IL-5 levels in T cells, even those treated to undergo Th1 development (by exposure to IL-12 and anti-IL-4 Ab). In contrast, GATA-3 was less potent in restoring IL-4 expression in T cells treated by IL-12.
We suggest that the IL-4 promoter is weakly transactivated by GATA-3 directly, but in the context of certain regions of the IL-13/IL-4 locus, becomes highly GATA-3 responsive. This model is similar to the mechanism whereby β-globin gene expression is controlled by GATA-1 through actions at hypersensitive site 4 (19). However, GATA-3 is unlikely to act alone in controlling induction of the IL-4/IL-13 locus, since retroviral expression of GATA-3 into already differentiated Th1 cells does not lead to IL-4 expression (data not shown). Finally, GATA-3 could augment IL-4 expression as an enhancer-binding factor, but our results show that in T cells, GATA-3 alone does not fully restore IL-4 expression.
We thank Nils Jacobson for IFN-γ Luc, Doug Engel for R/m-GATA3, Richard Flavell for the −157 Luc reporter, and David Baltimore for the IFN-β and NF-κB reporters.
This work was supported by National Institutes of Health Grants AI34580 and HL56419. K.M.M. is an Associate Investigator of the Howard Hughes Medical Institute.
Abbreviations used in this paper: −800 Luc, 800-bp IL-4 luciferase reporter construct; EMSA, electrophoretic mobility shift assay; NF-AT, NF of activated T cells; EMCV, encephalomyocarditis virus; IRES, internal ribosomal entry sequence; GFP, green fluorescent protein; RV, retrovirus.