The Fc receptor (FcR) γ-chain has been shown to be up-regulated in T cells when the TCR ζ-chain is decreased. We demonstrate that Elf-1, but not other Ets family transcription factors, bind to a cluster of GGAA sites located within the 200 bp upstream from the transcription initiation site of the FcRγ promoter. Forced expression of Elf-1 results in the suppression of FcRγ expression, whereas silencing its expression with small interfering RNA Elf-1 results in increased FcRγ expression. Elf-1 represents the first transcription factor identified to be involved in the transcriptional regulation of FcRγ, and cells that fail to express Elf-1, as is the case with human systemic lupus erythematosus T cells, will express FcRγ-chain.
The Fc receptor (FcR)4 common γ-chain was first described as a subunit of the high affinity IgE receptor I (FcεRI) tetrameric complex found in mast cells and basophils (1). Subsequently, it was shown to be associated with other types of FcR in a variety of immune cells (2) and to be involved in signal transduction via its ITAM. More recently, it has been shown that FcRγ associates with the TCR in CD4+ T effector cells (3) and T cells from patients with systemic lupus erythematosus (SLE) (4), where it assumes the function of the main TCR signaling molecule, the ζ-chain.
The FcRγ-chain and the TCR ζ-chain show high homology in both their sequences and molecular structures (1, 5). In fact, the FcRγ-chain is a member of the TCR ζ-chain family and it is believed that these two proteins are encoded by genes evolved from duplication. However, the FcRγ-chain displays anatomical and functional features distinct from that of the TCR ζ-chain; unlike the TCR ζ-chain, which has three copies of the ITAM, the FcRγ has only one (6). Furthermore, the FcRγ-chain interacts preferentially with the Syk kinase whereas the TCR ζ-chain associates with ZAP-70 (3). The efficacy of Syk to activate downstream signaling pathways is known to be >100-times more potent than that of ZAP-70 (7, 8). Therefore, the substitution of the ζ-chain by the FcRγ-chain results in hyperresponsiveness of the T cell to TCR/CD3 stimulation as exhibited by heightened calcium influx (3, 4, 9).
The molecular mechanism underlying this replacement of the TCR ζ-chain by the FcRγ-chain remains unknown. Previous studies have shown that forced overexpression of FcRγ leads to the down-regulation of the TCR ζ-chain (9), whereas reconstitution of the TCR ζ-chain expression in SLE T cells results in the decrease of the previously up-regulated FcRγ (10). In contrast to the TCR ζ gene that is known to be regulated by the transcription factors Elf-1 and CREM (cAMP response element modulator) (11, 12), the transcription factor or factors that are involved in the transcription of the FcRγ gene have not been defined.
In this communication, we show that the Ets family transcription factor Elf-1 is a potent suppressor of transcription of FcRγ gene in KU812 cells (a basophil cell line that expresses the FcRγ-chain at high levels) and T cells. Our results suggest that Elf-1 is responsible for the antithetic expression of the TCR ζ-chain and FcRγ-chain in T cells.
Materials and Methods
Cells and reagents
KU812 and Jurkat T cells were purchased from the American Tissue Culture Collection. For the purification of primary T cells, peripheral venous blood was obtained from healthy volunteers in heparin-lithium tubes. The blood was incubated for 30 min with a tetrameric Ab mixture against CD14, CD16, CD19, CD56 and glyA that attaches non-T cells to erythrocytes. The T cells were separated by the non-T cell-erythrocyte complexes using a Ficoll-containing gradient (Lymphoprep; Nycomed). The purified cells were >98% positive for CD3 as measured by flow cytometry. The Ab against FcRγ is a product of Upstate Biotechnology whereas the actin Ab was purchased from Sigma-Aldrich. The Abs against Elf-1, Ets-1, Ets-2, and Fli-1 were purchased from Santa Cruz Biotechnology. The two mAbs against Elf-1, 5A3 and 6G7, were gifts from Dr. J. Leiden (University of Chicago, Chicago, IL). Studies were approved by the human use committees of the involved institutions.
Plasmids and small interfering RNA (siRNA)
To generate DNA of different lengths of the FcRγ promoter 5′ to the transcription initiation site, PCR was conducted by using a plasmid that encodes 2,300 bp of the FcRγ promoter in front of a chloramphenicol acetyltransferase (CAT) reporter (a gift from Dr. J. P. Kinet, Harvard University, Boston, MA) as the template and primers encoding sequences spanning the desired areas on the promoter flanked by the restriction sites for HindIII and XbaI. The sequences of the 5′ primers used for PCR amplification of different length of the FcRγ promoter were as follows: CAACAAAGCAAGAGTCCGTC (designated p200); GATAAATCTCTGAACCAGCC (designated p400); CCTGGGGTAAAGTTGGGAGG (designated p500); and TTCCTCACTGGAAATTGCC (designated p900). The sequence for the 3′ primer to amplify the PCR product was AGAGCAAGACCACTGGAATCA. The PCR-generated DNA fragments were then gel purified and ligated with the pGL3 luciferase constructs (Promega). Plasmids were sequenced to detect any errors introduced by PCR. The siRNA against Elf-1 was purchased from Qiagen. The Elf-1 expression plasmid was a gift from Dr. W.J. Leonard (National Institutes of Health, Bethesda, MD).
The transfection of plasmids and siRNA into KU812 was done as follows: the cells were suspended in RPMI 1640 medium containing 10% FCS. Then the cells were subjected to electroporation (Bio-Rad) at 250 μV and 950 farads. For the transfection of primary T cells and monocytes, the Amaxa nucleoporator (Amaxa Biosystems) was used according to the manufacturer’s instructions.
Protein extraction and Western blots
The cells were treated with 400 μl of a cell lysis buffer (10 mM HEPES (pH 7.9), 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA supplemented with freshly added 1 mM DTT, 0.5 mM 4-(2-aminoethyl)-benzenesulfonyl fluoride, 2 mM aprotinin, 1 mM leupeptin, 10 mM NaF, and 2 mM Na3VO4) on ice for 15 min. At the end of the incubation, Nonidet P-40 was added to the reaction mixture at a concentration of 0.6%. The reaction mixture was vortexed for 10 s and then subjected to centrifugation at 13,000 rpm for 15 s. The supernatant was stored at −70°C as a cytoplasmic protein extract. The pellet was resuspended in 40 μl of buffer (20 mM HEPES (pH 7.9), 0.4 M NaCl, 1 mM EDTA, 1 mM EGTA, 10 mM NaF, 1 mM Na3VO4, 1 mM 4-(2-aminoethyl)-benzenesulfonyl fluoride, 2 mM aprotinin, and 1 mM leupeptin) and then shaken for 15 min at 4°C. After centrifugation for 5 min at 13,000 rpm, the supernatant was stored at −70 °C as nuclear protein extract. To extract total cellular protein, the cells were incubated with radioimmune precipitation assay buffer on ice for 30 min. This was followed by centrifugation at 13,000 rpm for 10 min. The supernatant was stored at −70 °C until use.
Nuclear extracts were incubated with a radiolabeled DNA probe, 1 μg of polydeoxyinosinic-deoxycytidylic acid (poly(dI-dC)) in the binding buffer for 15 min at room temperature. The reaction mixture was then subjected to separation in a 6% nondenaturing gel (Invitrogen Life Technologies). The dried gel was then autoradiographed. For supershift assays, the nuclear proteins were incubated with specific Abs at 4°C for 10 min before the probe, and poly(dI-dC) and binding buffer were added. The reaction was further conducted for another 15 min at room temperature. The sequences of the oligonucleotides used are shown in Table I.
|.||α β .|
|.||α β .|
The four GGAA sites, α , β , γ , and δ , are set in boldface. WT, Wild type.
Twenty micrograms of nuclear proteins were separated on 4–12% bis-Tris gel (Invitrogen Life Technologies) followed by transfer to polyvinylidene difluoride membrane at 30 V for 1 h in the cold room (4°C). Membrane-bound proteins were then renatured by soaking in 1× binding buffer (50 mM Tris (pH 8), 1 mM DTT, 150 mM NaCl, 0.3% Tween 20, and 5 mM MgCl2) for 30 min at room temperature. The blot was further incubated in 1× binding buffer containing 5 μg/ml poly(dI-dC) and radiolabeled probe (2 μl/ml buffer) for 1 h at room temperature. After thorough washing with the same binding buffer, the membrane was air dried followed by exposure for autoradiography.
Chromatin immunoprecipitation assays
Five million cells were used per immunoprecipitating Ab. The cells were fixed with 1% formaldehyde for 10 min (NA and protein were cross-linked), lysed,and sonicated. The DNA-protein complexes were immunoprecipitated with the appropriate Ab and, after a series of washing steps, the cross-linking was reversed and the protein was digested with proteinase K. The amount of the immunoprecipitated DNA was assessed by PCR.
RNA purification and RT-PCR
RNA purification and reverse transcription were performed with the use of the RNeasy Mini kit (Qiagen). cDNA from an equal amount of RNA input were then amplified by using specific primers.
We used Microsoft Excel for the statistical analysis and graphical representation of our data.
Determination of regions on the FcRγ promoter responsible for its activity in KU812 cells
To determine the relative contribution of different segments of the FcRγ promoter to its activity, promoter fragments of various lengths were cloned in front of the luciferase reporter gene. We controlled for transfection efficiency by cotransfecting these constructs with a β-galactosidase expression plasmid, and its activity was used to normalize the data. As shown in Fig. 1, the 400-bp region upstream of the transcription initiation site was sufficient to support the majority of the promoter activity, with the first 200-bp region contributing to more than one-half of the FcRγ promoter activity. Inside this region there are a number of sites encoding sequences with various level of homology to the binding sites of several known transcription factors.
GGAA sites within the FcRγ promoter bind nuclear proteins
Because the first 200 bp upstream of the transcription initiation site of the FcRγ promoter are responsible for >50% of the promoter activity as shown in Fig. 1, we searched for potential nuclear factor binding sites within this region. We found that this region contains four GGAA sites (Table I). To determine protein binding to these sites, we incubated nuclear protein extracts from KU812 cells with two labeled oligonucleotides containing two GGAA sites each, one spanning the −121 to −84 nucleotides upstream of the transcription initiation site (containing GGAA sites α and β; Table I), and one spanning the −83 to −46 nucleotides (containing GGAA sites γ and δ; Table I). Shift assays shown in Fig. 2 demonstrated the formation of two bands (indicated by arrows).
We conducted competition EMSA using unlabeled wild-type oligonucleotides or oligonucleotides mutated at the different GGAA sites. The M1 mutated oligonucleotide (Table I) did not compete with the wild-type labeled oligonucleotide for protein binding, indicating that the mutated bases are critical for the binding of nuclear proteins (Fig. 2,A). Similarly, the mutated oligonucleotides of M3 (Fig. 2 B) competed with the labeled oligonucleotides for binding less efficiently than the wild-type and the M4 for the formation of the lower band (Elf-1 containing; see next paragraph), indicating that the γ site (M3) is also important in binding nuclear proteins. Taken together, these data indicate that at least two of the four GGAA elements located at the proximal region of the FcRγ promoter bind nuclear proteins.
Elf-1 constitutively binds to the FcRγ GGAA site in vitro and in vivo
GGAA elements represent preferential DNA binding sites for the Ets family of transcription factors that includes >20 members. To identify the specific Ets member(s) that bind(s) to the FcRγ promoter, we first conducted Southwestern experiments. As shown in Fig. 3,A, Southwestern analysis detected a specific band with an approximate molecular mass of 98 kDa. Because Elf-1 is the only member of the Ets family of proteins known to have a molecular mass close to 98 kDa, this finding suggested Elf-1 as the most probable candidate. We performed supershift experiments using oligonucleotides −146 to −84 and −83 to −46 of the FcRγ promoter to test this prediction. When Abs against various Ets family members were used in the shift assays, we found that only the Ab against Elf-1 and not Abs against Fli-1, Ets-1 (Fig. 3,B), or Ets-2 (not shown) decreased the intensity of the lower oligonucleotide-protein complex (indicated by arrow in Fig. 3). This finding confirmed the results from the Southwestern analysis and indicated that Elf-1 is the principal Ets member that binds to the promoter of the FcRγ gene. In addition, we showed that the Elf-1/GGAA complexes were specific in that the oligonucleotides spanning the neighboring area (−45 to −8; Fig. 3 B, right panel, far right lane), which does not have a GGAA element, cannot form any detectable complex with nuclear proteins. This finding indicates that the protein(s) that contributed to the formation of the upper band was not one of the tested Ets members.
Because the Southwestern experiments (Fig. 3,A) showed intense binding of labeled probe to nuclear proteins from Jurkat T cells, we incubated nuclear proteins from either Jurkat or primary T cells with the oligonucleotides spanning −121 to −84 bp. EMSA using nuclear protein from Jurkat as well as primary T cells resulted in a nuclear protein/GGAA complex that could be supershifted with the polyclonal and two mAbs against Elf-1 (Fig. 3 C). Taken together, these data indicate that Elf-1 binds to the FcRγ promoter in lymphocytes.
Subsequently, we conducted chromatin immunoprecipitation assay experiments to assess whether Elf-1 binds to the FcRγ promoter in vivo by using anti-Elf-1 and control Abs and the indicated primers. As shown in Fig. 4, anti-Elf-1 Ab precipitated more FcRγ promoter DNA from Jurkat T cells in comparison to control or normal rabbit Ab. These data indicate that Elf-1 binds to the FcRγ promoter both in vitro and in vivo.
Elf-1 suppresses the expression of FcRγ
We next asked whether Elf-1 also suppresses the transcription of the endogenous FcRγ gene. To this end we transfected either control siRNA or Elf-1 siRNA into KU812 cells. The FcRγ mRNA was found to be increased in KU812 cells transfected with 75 or 150 nM Elf-1siRNA (Fig. 5,A; p < 0.05, n = 4). Furthermore, we transfected KU812 cells with control plasmid or plasmid overexpressing Elf-1. The expression of FcRγ protein was found to be decreased in KU812 cells overexpressing Elf-1 (Fig. 5 B; p < 0.05, n = 3). These data indicate that Elf-1 is able to suppress the activity of the endogenous FcRγ promoter.
Because the competition shift assays shown in Fig. 2 suggested that the GGAA site designated γ (M3 oligonucleotide) might represent an important one, we cloned the FcRγ promoter with the γ GGAA site mutated in front of the luciferase gene and compared its activity to that driven by the wild-type promoter in T cells, monocytes, and KU812 cells. As can be seen in Fig. 6, the mutation at the γ site releases the promoter activity in T cells (p < 0.05) but not in monocytes. The activity was moderately increased in the KU812 cells transfected with the mutated promoter. Finally, we incubated a labeled oligonucleotide containing the GGAA Elf-1 binding site (−83 to −46 site of the FcRγ promoter) with nuclear proteins from primary human T cells, monocytes and KU812 cells (Fig. 6,B). The intensity of the shifted band was more intense when nuclear proteins from T cells rather than from monocytes and KU812 cells were used (Fig. 6 B).
To establish that Elf-1 directly regulates the expression of the endogenous FcRγ gene in human peripheral blood T cells, we took two approaches. First, we treated cells with small interfering Elf-1 RNA and recorded the expression of actin and FcRγ mRNA. As shown in Fig. 7, A and B (cumulative data, n = 4), small interfering Elf-1 RNA significantly enhances the expression of Elf-1 mRNA (p < 0.05). Second, we transfected primary T cells with an Elf-1 expression vector or empty vector and determined the expression of FcRγ and actin mRNA at 4 h. When normal T cells were transfected with Elf-1, a significant (p < 0.05) decrease in the expression of FcRγ mRNA after transfection was seen (Fig. 7, C and D, cumulative data, n = 3). The data demonstrate that modulation of the expression of Elf-1 either by silencing or by forced expression results in enhanced or suppressed expression of FcRγ mRNA, respectively.
Our data indicate that the Ets family member Elf-1 binds to the FcRγ promoter at distinct GGAA sites in both basophils and T cells. The binding of Elf-1 to these sites resulted in the suppression of the promoter activity in both cell types. In addition, we found that Elf-1-mediated repression of FcRγ gene expression can be lifted by Elf-1-specific siRNA. The finding that Elf-1 exhibited higher levels of binding to the FcRγ promoter in T cells as compared with basophils could provide an explanation for the differential expression of FcRγ in these cell types.
Incubation of KU812 cells with radiolabeled oligonucleotides containing GGAA sites resulted in the formation of two distinct bands, whereas only one band was detected by Southwestern analysis. This can be explained by the different conditions used in these two analyses. In EMSA, protein complexes are allowed to interact freely with the radioactive probe, whereas in Southwestern analysis protein complexes are disrupted and the proteins are subsequently fixed on distinct positions on the membrane before they are exposed to the probe. The upper band in the EMSA experiments (Fig. 2 and Fig. 3) probably represents proteins that can only bind to DNA as a complex and therefore would not be able to bind to DNA under the conditions used in Southwestern analysis.
Elf-1 is a member of the Ets transcription factor family, which includes >20 members (13). Elf-1, like all other Ets members, contains domains for DNA binding, protein interaction, and, importantly, domains that mediate transcriptional activation. However, Ets family members were shown to play dual roles as both transcriptional activators and suppressors. For instance, while Ets-1 participates in the transcriptional activation of many genes, overexpression of Ets-1 suppresses the transcription of the IL-2 gene (14). Similarly, although Elf-1 has been shown to be responsible for the activation of a number of genes in cells of the hemopoietic system, overexpression of Elf-1 decreased the expression of the IL-2 receptor α-chain and tumor suppressor gene Tsc2 (15). Currently the exact mechanism that determines whether the Ets transcription factor becomes an activator or a suppressor is unknown. Elf-1 frequently forms a complex with other molecules to exert its function (16, 17). It is possible that the effect of Elf-1 on different gene promoters depends on its molecular partners.
Previous promoter functional analysis by serial deletion of the promoter has indicated the existence of multiple factors involved in the negative regulation of FcRγ (18). The current study identified GGAA elements located on the FcRγ promoter that, after binding of Elf-1, mediate a potent suppressive effect on the transcription of FcRγ-chain. This finding represents the first report on the identity of a transcription suppressor of the FcRγ gene. It is possible that other transcription suppressors enhance the suppressive effect of Elf-1 on FcRγ transcription. The molecular characterization of these molecules will further contribute to our understanding of the mechanisms leading to the aberrant expression of FcRγ.
The data reported herein help explain the up-regulation of the expression of the FcRγ-chain in SLE T cells that express limited or no TCR ζ-chain (19). It appears that Elf-1 acts as a molecular valve and that its levels control the expression of the TCR ζ-chain or the FcRγ-chain by virtue of serving as a transcriptional enhancer for the first and a repressor for the second. It is possible also that the levels of Elf-1 may directly dictate the ratio of TCR ζ-chain/FcRγ-chain expression.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported by National Institutes of Health Grants RO1 AI42269 and RO1 AI49954. The opinions expressed herein are those of the authors and do not reflect those of the Department of Defense.
Abbreviations used in this paper: FcR, Fc receptor; poly(dI-dC), polydeoxyinosinic-deoxycytidylic acid; siRNA, small interfering RNA; SLE, systemic lupus erythematous.