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

J. Biol. Chem., Vol. 277, Issue 49, 47898-47906, December 6, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/49/47898    most recent M201025200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ji, H.-b. Articles by Terhorst, C. Search for Related Content PubMed PubMed Citation Articles by Ji, H.-b. Articles by Terhorst, C. Social Bookmarking                      What's this?

T Cell-specific Expression of the Murine CD3 Promoter^*

Hong-bin Ji^§, Anita Gupta, Susumu Okamoto^¶, Michael D. Blum, Lujian Tan^**, Mary B. Goldring^**, Elizabeth Lacy, Ananda L. Roy^§§, and Cox Terhorst

From the Divisions of  Immunology and ^** Rheumatology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts 02215, the  New England Baptist Bone and Joint Institute, Harvard Institutes of Medicine, Boston, Massachusetts 02115, the  Molecular Biology Program, Memorial Sloan-Kettering Cancer Center, Weill Graduate School of Medical Sciences of Cornell University, New York 10021, and the ^§§ Department of Pathology, Tufts University School of Medicine, Boston, Massachusetts 02111

Received for publication, January 30, 2002, and in revised form, August 9, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

T cell-specific expression of human and mouse CD3 is known to be governed by an enhancer element immediately downstream from the gene. Here we demonstrate by transgenic and in vitro studies that the murine CD3 (mCD3) promoter prefers to be expressed in cells of the T lineage. Deletion analyses of a promoter segment (401/+48 bp) followed by transient transfections indicate that upstream elements between 149 and 112 bp contribute to full expression of the gene. Furthermore, a core promoter region 37/+29 appears to contribute to a T cell specificity. Using substitution mutant scanning, four positive and one negative regulatory elements were found within the mCD3 core promoter. The first two positive elements comprise two classical initiator-like sites, which recruit TFII-I, whereas a third contains a functional Ets binding site. Immediately adjacent to the observed transcription start site is a negative element that utilizes the transcription regulator YY1. The last positive regulatory element contains a potentially functional CREB binding site and the minor transcriptional start site. Because NERF-2, Elf-1, and Ets-1 are expressed preferentially in lymphocytes and because, in addition, YY1 represses the promoter activity strongly in non-T cells, we conclude that the combination of these transcription factors contributes to the T cell-specific expression pattern of mouse CD3.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Recent studies have shown that B and T lymphocytes and natural killer cells originate from a common stem progenitor (1, 2). One of the earliest events in commitment to development of the T cell lineage is expression of the CD3 genes, which even precedes TCR¹- rearrangement (3). Functionally the CD3, , and  proteins form the scaffold upon which the TCR·CD3 complex or the pre-TCR·CD3 complex is assembled. The genes encoding the CD3, , and  membrane proteins are tightly linked on chromosome 9 in the mouse and on chromosome 11 in human (4). The CD3 and  promoters are oriented head to head as a divergently transcribed gene pair, their transcription start sites being around 1.5 kb apart (4). Remarkably, no regulatory principles that govern all three CD3 genes have been recognized so far. This is particularly striking because CD3 and CD3 appear to have evolved from one common ancestor gene relatively recently (5).

Both the mouse and human CD3 gene comprise five exons that are organized in a similar fashion. A T cell-specific enhancer element A is found immediately downstream from the 3'-untranslated region of mouse and human CD3 and is thought to govern the T cell-specific expression of the gene (3). A second element B, which is found in the mouse only, increases the activity of the A enhancer but has no activity by itself (3).

Different lines of evidence indicate that expression of most T cell-specific genes is regulated by assembly of promoter and enhancer elements resulting in large complexes that comprise multiple transcription factors (6). Little is known about the CD3 promoter and its importance for continued expression of the gene from the earliest recognizable thymocytes to mature functional T lymphocytes. Both the mouse and human CD3 promoter lack classical TATA and CAAT boxes. Two transcription initiation sites have been described previously for the mouse CD3 gene (7). In the promoter region of human and mouse CD3, three highly conserved regions (CR1, CR2, CR3) are found (Fig. 1A). In the CR3 region are two adjacent classic initiator-like (Inr-L) sites. In human, the CR3 region includes one start site for CD3, whereas in mouse the CR3 region is very close to the first major transcription start site (Fig. 1B).

View larger version (32K): [in this window] [in a new window]   Fig. 1.   Comparison of the human and mouse CD3 genes. A, both human and mouse CD3 genes comprise five exons. Although a T cell-specific enhancer element A is located 0.3 kb downstream from the human and mouse genes, the B sequence is present only in the mouse gene (3). The CD3 promoter regions contain two transcription start sites as judged by primer extension and RNase protection assays (7). Three highly conserved regions (CR1, CR2, and CR3) that contain putative regulatory elements are located adjacent to the two transcription start sites. B, comparison of the mouse and human CR3 regions. The CR3 region contains two adjacent classic Inr-L sites. In human, the CR3 region includes one start site for CD3, whereas in mouse the CR3 region does not include but is very close to the first major transcription start site.

Here we use transgenic and in vitro approaches to show that the mouse CD3 promoter (401/+48) (3) is expressed in a T cell-specific fashion. Dissection of a short core promoter region by mutation analyses revealed the presence of several positive and one negative regulatory element. Our results indicate that TFII-I, the protein binding to two Inr-L sites, the Ets family members, and the potentially negative regulator YY1 govern expression of the mouse CD3 gene in a coordinated fashion.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Plasmids-- The mouse CD3 promoter fragment 401/+48 bp was amplified from the vector p113 (4) by PCR using a 5'-mCD3 primer (5'-GGGGTACCTGATCAGAAACAAGAGGATCT-3') and a 3'-mCD3 primer (5'-TCATCTCGAGTGATCAGCCAGGGTTAGCAC-3') (Invitrogen). This fragment was then cloned into the KpnI and XhoI sites of a promoterless luciferase vector pGL3-Basic (Promega, Madison, WI) and used for the generation of nested deletions. The 5'-deletions beginning at 178, 149, 112, 76, or 37 to +48 from the start site (7) used the 3'-mCD3 primer and one of the following 5'-primers: 178 (5'-ACGGTACCTCATAGTCTCTGCTTTGTA-3'), 149 (5'-CAGGTACCCCCAAACTCTTTGCTTCAGA-3'), 112 (5'-TTGGTACCGTTTGTGTCTTACCCCACAC-3'), 76 (5'-CAGGTACCTTGACAGTCTTACACCATCA-3'), 37 (5'-GGGGTACCGCAGATTTCTTCTAGTTCCT-3').

The 3'-end deletions of the 401/+48 promoter, which extended from 401 to +37, +29, or +10, used 5'-mCD3 primer plus one of the following 3'-primers: +37 (5'-TCATCTCGAGAGCACTATCCTCCGCCAACA-3'), +29 (5'-AATTCTCGAGCCTCCGCCAACAGACTCATG-3'), +10 (5'-CAGACTCGAGGGGTATTCAGAAAAAGGAAG-3').

The core promoter 37/+29 was amplified using the 37, 5'-primer and the +29, 3-' primer.

A panel of linker scanning mutations in the mCD3 promoter 401/+48 was generated by oligonucleotide-directed mutagenesis (Stratagene, La Jolla, CA). Every pair of oligonucleotides was designed to alter a specific 3-4-bp sequence within the promoter region; all other sequences within the region were unaltered, as verified by subsequent DNA sequencing. The mutated sequences are indicated in Fig. 4A. An additional mutant construct containing the M6 mutation of the core promoter (37/+29_M6) was also prepared (Stratagene). The 5'-primers used for the generation of mutations M1-M13 are as follows: M1 (5'-GAAGGTAGAGAGGCAGAACTATTCTAGTTCCTCCCCC-3'), M2 (5'-GTAGAGAGGCAGATTTCTGGTCGTTCCTCCCCCACTCT-3'), M3 (5'-GGCAGATTTCTTCTAGCGCGTCCCCCACTCTTCCTTT-3'), M4 (5'-ATTTCTTCTAGTTCCTCCGTCGCTCTTCCTTTTTCTGAA-3'), M5 (5'-CTAGTTCCTCCCCCACGCGGCCTTTTTCTGAATACCC-3'), M6 (5'-TTCCTCCCCCACTCTTAATTTTTCTGAATACCCA-3'), M7 (5'-CCTCCCCCACTCTTCCGCGTTCTGAATACCCATGAG-3'), M8 (5'-CCCCCACTCTTCCTTTTAGAGAATACCCATGAGTCTG-3'), M9 (5'-ACTCTTCCTTTTTCTGAAGATGCATGAGTCTGTTGGC-3'), M10 (5'-CCTTTTTCTGAATACCCTACAGTCTGTTGGCGGAGG-3'), M11 (5'-TTTCTGAATACCCATGAACGTGTTGGCGGAGGATAG-3'), M12 (5'-GAATACCCATGAGTCTGACGTCGGAGGATAGTGCTAAC-3'), M13 (5'-CCATGAGTCTGTTGGCTCACGATAGTGCTAACCCTG-3'.

The plasmids for expression of Ets factors, including pCI-NERF-1, pCI-NERF-2, pCI-Ets-1, pCI-Ets-2, and pCI-Elf-1, were kindly provided by Dr. Towia A. Libermann (Beth Israel Deaconess Medical Center).

Cell Culture and Reporter Gene Assays-- The murine B cell line A20 and human T cell line Jurkat were grown in RPMI 1640 medium supplemented with 10% fetal bovine serum, L-glutamine, sodium pyruvate, and antibiotics (Invitrogen). The murine T cell line EL4 and fibroblast cell line L929 were maintained in Dulbecco's modified Eagle's medium containing the same additives (Invitrogen).

All cells were transiently transfected using FuGENE 6 transfection reagent (Roche Molecular Biochemicals). In brief, 0.5 µg of each luciferase construct was transfected into cells by FuGENE 6 together with 0.05 µg of an internal control plasmid containing the Renilla luciferase gene under control of the herpes simplex virus-1 thymidine kinase promoter pRL-TK (Promega). In the cotransfection experiments, an additional 0.1 µg of the indicated Ets expression plasmid together with the mCD3 promoter luciferase construct were transfected into L929 cells. Transfected cells were harvested after 48 h, and extracts were then subjected to analysis employing the Dual Luciferase assay as recommended by the manufacturer (Promega). Promoterless luciferase construct pGL3-Basic and pGL3-SV40, which encodes the luciferase gene under the control of SV40 promoter/enhancer, were used as controls. All transfections were performed at least three times to ensure reproducibility.

Electrophoretic Mobility Shift Assay (EMSA)-- All of the nuclear extracts from Jurkat or EL4 cells were prepared using the NE-PER^TM Nuclear and Cytoplasmic Extraction Reagents kit (Pierce). Extraction buffers were supplemented with 1 mM phenylmethylsulfonyl fluoride, 0.5 µg/ml leupeptin, 1 µg/ml aprotinin, and 1 µM pepstatin (Roche). Probes for EMSAs were made by annealing single-stranded oligonucleotides. 50 ng of each probe was labeled by filling in with [-³²P]dCTP and other dNTPs using the Klenow fragment of DNA polymerase (Stratagene). Labeled probes were purified with a NucTrap purification column (Stratagene).

Approximately 4 × 10⁴ cpm of each probe was added to 10 µg of nuclear protein plus 1 µg of poly(dI-dC) in binding buffer (10 mM Tris, pH 7.5, 50 mM NaCl, 1 mM dithiothreitol, 1 mM EDTA, 5% glycerol), and the binding reactions were performed on ice for 30 min (Promega). EMSA products were separated on 4% polyacrylamide, 0.5 × Tris boratem, EDTA gels running at 4 °C for 3 h at 200 V. For competition experiments, a 100-fold molar excess of unlabeled oligonucleotide competitor was added to the binding reaction 10 min before the addition of the radiolabeled probes. The sequence of the probe containing two Inr-L sites was ²⁹TTCTAGTTCCTCCCCCACTCTTCCT⁵]. The TFII-I binding sequence derived from the TCR-V promoter (V-TFII-I) probe was 5'-AGGAAAGAGAAAGGAGGAGCCGACTCTCA⁺¹CTTTCTCACCA-3' (8). Probes YY1 and MY were purchased from Santa Cruz Biotechnology, Santa Cruz, CA. The sequences of other competitors are shown in Fig. 7A.

In experiments with antibodies directed at specific nuclear factors, nuclear extracts were preincubated with anti-TFII-I (8) for 10 min before setting up the binding reaction. In the in vitro binding assay, the radiolabeled probe was incubated with recombinant TFII-I (8) or YY1 (9) at room temperature for 20 min before running the gel.

Construction of mCD3 Promoter/Human CD4 Transgene-- A human CD4 reporter gene driven by the 401/+48 mCD3 promoter was used to generate transgenic mice as described before (10). In brief, a human CD4 (hCD4) minigene was created by fusing an EcoRI-SacI cDNA fragment encoding CD4 exons 2-4 and part of exon 5 to a genomic SacI-BamHI fragment encoding the remainder of exon 5, exons 6-9, and a part of exon 10. A 0.24-kb BclI-BamHI fragment containing the SV40 poly(A) site provides the transcription termination signal. The hCD4 minigene itself does not express a functional protein, nor does it contain any known T cell regulatory element (10).

Northern Blot-- Total RNA was isolated from various organs of mice using TRIZOL (Invitrogen) according to the manufacturer's instructions. 20 µg of total RNA was fractionated by electrophoresis through a 1.5% agarose gel and then transferred to a nitrocellulose membrane. RNA was cross-linked by UV irradiation. The cDNA probe was labeled by [³²P]dCTP incorporation using a random primer labeling kit (Stratagene). Prehybridization (1-2 h) and hybridization (overnight) were carried out at 56 °C in Expresshyb hybridization solution (Clontech, Palo Alto, CA). Hybridized blots were washed once with 0.1 × SSC and 1% SDS at room temperature for 15 min and twice at 54 °C for 15 min. The membranes were then exposed overnight at 80 °C using intensifying screens.

Flow Cytometry-- Approximately 0.5-1 × 10⁶ transfected cells and splenocytes were washed with phosphate-buffered saline and incubated with the indicated antibodies for 20 min in phosphate-buffered saline with 2% bovine serum albumin at 4 °C. Peritoneal cells were stained with biotin-conjugated antibodies first for 20 min on ice and then stained with fluorescein isothiocyanate-conjugated human CD4 antibody (BD Biosciences). The cells were then washed twice with phosphate-buffered saline and fixed with 1% paraformaldehyde in phosphate-buffered saline. Flow cytometry was performed with a FAXStarPlus (BD Biosciences). All of the antibodies (human CD4-fluorescein isothiocyanate, mouse CD3-PE, mouse B220-PE, mouse F4/80-biotin, and mouse Ly6G (Gr-1)-biotin) were purchased from BD Biosciences.

Antibody and Recombinant Protein Purification-- Anti-TFII-I was purified as described before (11). Recombinant TFII-I was purified upon ectopic expression in COS cells (8, 12). A prokaryotic expression vector for the production of YY1 as a polyhistidine fusion protein was a gift from Timothy Osborne (University of California, Irvine) (13). Recombinant YY1 was expressed and purified as described (9, 14).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

T Cell-specific Expression of the mCD3 Promoter in Vitro and in Transgenic Mice-- To test the specificity of the mCD3 promoter, a DNA segment encompassing the mCD3 promoter (401/+48) was fused to a hCD4 minigene and an SV40 poly(A) site (Fig. 2A) (10). Upon transient transfection of the minigene into murine cell lines, the ectopic protein hCD4 was expressed only in the thymoma line EL4 but not in the B cell line A20 or in L929 fibroblast cells as judged by staining with anti-hCD4 for flow cytometry assay (Fig. 2B). These results suggest that the mCD3 promoter governed expression in a T cell-specific manner in cell lines because the CD4 minigene itself does not contain any regulatory T cell-specific element (6, 10).

View larger version (27K): [in this window] [in a new window]   Fig. 2.   T cell-specific expression of the mCD3 promoter in vitro and in transgenic mice. A, schematic representation of the human CD4 minigene (10) located downstream from the 401/+48 bp mCD3 promoter segment. B, flow cytometric analysis of expression of the mCD3 driving a hCD4 minigene in EL4, A20, and L929 cells. Transfected cells were stained with mouse anti-human CD4 conjugated with PE and subjected to flow cytometry. C, Northern blot analysis of hCD4 expression in tissue from two transgenic mice LINE1283L-mCD3-hCD4 and LINE1277-mCD3-hCD4. 20 µg of total RNA was loaded in each lane. Inspection of the ethidium bromide-stained gel revealed the presence of equal amounts of 28 S rRNA in each sample. The blot was probed with a fragment of human CD4 cDNA which does not cross-hybridize to the endogenous mouse CD4 transcript. D, flow cytometric analysis of splenocytes and peritoneal cells from the non-transgenic and two transgenic mouse lines LINE1277-mCD3-hCD4 and LINE1283L-mCD3-hCD4 after staining with various antibodies directed at human CD4, mouse CD3, B220, F4/80, and Ly6G (Gr-1).

To test tissue-specific expression in vivo, the mCD3-hCD4 construct was then used to generate two transgenic mouse lines, LINE1277-mCD3-hCD4 and LINE1283L-mCD3-hCD4. Northern blotting of total RNA isolated from various organs of the two transgenic mouse lines indicated that the hCD4 was expressed only in the thymus and spleen of the transgenic mice, but not in brain, kidney, or liver (Fig. 2C).

To determine further whether expression of the hCD4 transgene was restricted to T cell lineage, fluorescence-activated cell sorter analysis was performed after staining splenocytes with anti-hCD4, anti-mCD3 and anti-mB220. As shown from Fig. 2D, almost all the T cells (mCD3-positive) in the two transgenic lines, LINE1277-mCD3-hCD4 and LINE1283L-mCD3-hCD4, expressed the transgene human CD4 but no expression in non-transgenic T cells. However, only 1-3% spleen B cells (mB220-positive) in these two transgenic lines express the human CD4 transgene (Fig. 2D). These data confirmed the hypothesis that the mouse CD3 promoter is preferentially expressed in T lineage but not B cells. To check whether non-lymphoid cells also express the human CD4 transgene, we isolated peritoneal cells and checked the expression of the human CD4 transgene in macrophages and granulocytes (Fig. 2D). Mouse F4/80 and Ly6G (Gr-1) antibodies were used, respectively, as the maker for macrophages and granulocytes. As shown from Fig. 2D, no transgene expression was observed in macrophages and granulocytes from these two transgenic lines. Taken together, our Northern blot and flow cytometry analysis data support the notion that the mCD3 promoter (401/+48) contributes significantly to T cell-specific expression of the gene in vivo.

Deletion Analysis of the mCD3 Promoter Region-- To investigate which part of the mCD3 promoter (401/+48) directs expression of the mCD3 gene, wild-type (wt) and various deletion mutants of the pGL3-mCD3 construct were transfected into four indicator cell lines. A plasmid encoding the Renilla luciferase under the control of the TK promoter (pRL-TK) was cotransfected as an internal control (15). The pGL3-SV40 construct was used as a positive control (16). The mCD3 promoter activity was found only in the murine T lymphoma cell line EL4 and the human T cell line Jurkat (Fig. 3A). Transiently transfected Jurkat cells displayed much higher CD3 promoter activity than EL-4. This could be explained by the higher transfection efficiency in Jurkat than that in EL4 cells. In contrast, no expression was observed in the murine B cell line A20 and the fibroblast cell line L929. These results support the view that like the transgenic lines, the mCD3 promoter (401/+48) itself contributes to the T cell-specific expression of the gene under transient assay conditions.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Analysis of expression of the mCD3 core promoter. A, T cell-specific expression of the 401/+48 bp mCD3 promoter/luciferase construct. Luciferase activity was determined after transfection of the pGL3-mCD3 401/+48 plasmid and the TK-promoter-Renilla luciferase control plasmid (pRL-TK) into each cell line (see "Materials and Methods"). The pGL3-Basic promoterless vector and pGL3-SV40 containing the SV40 promoter and enhancer served as negative and positive controls, respectively. After 48 h of transfection, cells were subjected to the Dual Luciferase assay ("Materials and Methods"). B and C, identification of the mCD3 core promoter 37/+29. Deletion mutants of pGL3-mCD3 described under "Materials and Methods" were transfected in Jurkat cells, and luciferase activity was analyzed by the dual reporter gene assay. D, the 37/+29 mCD3 core promoter is specifically expressed in Jurkat cells. The mCD3 core promoter 38/+29 luciferase construct was transfected into three cell lines, A20, Jurkat, and L929, and subjected to the Dual Luciferase assay. All data shown in Fig. 3 represent three independent transfection experiments.

For functional dissection of the 401/+48 mCD3 promoter region, a series of 5'- and 3'-deletions was subsequently inserted in front of the luciferase gene. Successive deletion of the 5'-sequences to position 149 bp did not affect the promoter activity in Jurkat cells (Fig. 3B). Two shorter segments, 112/+48 bp and 76/+48 bp, displayed 65 and 50% of the 401/+48 mCD3 promoter activity, respectively. This suggested the presence of positive regulatory elements in the region spanning positions 149 to 76 bp. Most importantly, the minimal promoter segment observed (37/+48 bp) retained 65% of the activity of the 401/+48 segment in T cells. Given the importance of this region in T cell-specific expression of the mCD3 promoter, we carried out various deletions to characterize this region further (Fig. 3B).

Deletion of 19 bp from +48 to +29 at the 3'-end of the 401/+48 bp promoter had no obvious effect upon its activity (Fig. 3C). However, deletion of an additional 19 bp to +10 bp resulted in a dramatic decrease of promoter activity. This suggested the presence of an essential element between +10 and +29 bp which positively regulated mCD3 gene expression (Fig. 3C). This could be caused by disruption of the minor transcription start site and a potentially functional CREB binding site (see below). Taken together, the deletion studies demonstrate that the mouse CD3 core promoter is located between positions 37 and +29. Most significantly, this core promoter 37/+29 bp was expressed in the human T cell line Jurkat at greater than 60% of the activity of the 401/+48 bp segment, but not in A20 and L929 cell lines (Fig. 3D).

Mutation Analysis of the mCD3 Core Promoter Region-- To identify critical functional elements responsible for T cell-specific expression of the 66-bp core promoter further, substitution mutations covering the 37/+29 bp segment were made systematically by oligonucleotide-directed mutagenesis in the full-length 401/+48 bp mCD3 promoter construct. Sequence analyses of the 13 substitution mutants confirmed that the mutations were distributed throughout the 66-bp segment, targeting two potential Inr-L sites and several hypothetical critical transcription factor binding sites as shown along the top bar in Fig. 4A. Upon transient transfection into Jurkat cells, five sites determined by mutations M2/M3, M4/M5, M6, M7, and M9-M12 were found to be potentially involved in regulation of the mCD3 promoter. These delineated two protential Inr-L sites and TFII-I, Ets, YY1, and CREB binding sites, respectively (Fig. 4B).

View larger version (34K): [in this window] [in a new window]   Fig. 4.   Identification of four positive and one negative element in the 37/+29 mCD3 core promoter. A, schematic depiction of various substitution mutants throughout the 33/+28 region in the mCD3 promoter 401/+48. The mutated 3-4-bp sequences are boxed, and the substituted nucleotides are indicated in the corresponding positions. Two Inr-L sites and the binding sites for TFII-I, Ets, YY1, and CREB are indicated. B, relative luciferase activities of mutants in the 401/+48 mCD3 promoter. The five critical elements within the core promoter region 37/+29 are indicated by numbers 1-5. The mutants M1-M13 were transfected into Jurkat cells and subjected to the Dual Luciferase assay. The average of three independent assays is shown.

Mutations M2 and M3 are located in the first Inr-L site and displayed only 15 and 40% of the 401/+48 promoter activity. Both mutations M4 and M5, which disrupted the second Inr-L site containing the putative TFII-I binding site, led to a decrease of ~40% of promoter activity (Fig. 4B). Mutation M6 adjacent to Inr-L site 2, in which the core binding site for Ets family members was mutated, reduced 401/+48 bp promoter activity by more than 90% (Fig. 4B). Taken together, these data support a model in which both the Inr-L sequences and certain Ets family members play an important role in directing transcription of the mCD3 promoter in T cells.

Mutation M7, which disrupted a hypothetical YY1 binding site while keeping the Ets site intact, increased promoter activity by 2-fold. This suggests that YY1s play a negative role in mCD3 regulation. A critical downstream element was disrupted by the four successive mutations M9, M10, M11, and M12, each of which reduced mCD3 promoter activity by 80% (Fig. 4B). This result confirms the earlier observation made in the 3'-deletion analysis (Fig. 3C). In conclusion, the substitution mutation analyses revealed the presence of at least four positive and one negative regulatory element within the mCD3 core promoter region.

TFII-I Binds to the Two Inr-L Sequences of mCD3 Promoter-- Because TFII-I interacts physically and functionally with Initiator and Inr-L sequences (17), we tested the role of TFII-I in regulation of the mCD3 promoter. Using the radiolabeled probe 29/5 containing the two Inr-L sites, EMSA with nuclear extracts derived from EL4 cells gave three bands designated A1, A2, and A3 (see Fig. 5A). Of these, A1 gave the strongest binding. However, all three bands were competed away by 100-fold excess of the wt oligonucleotide. Interestingly, only A1 and A3 were competed by an oligonucleotide corresponding to the TFII-I binding sequence derived from the TCR-V promoter (V-TFII-I), indicating that the A2 complex recruits accessory proteins other than TFII-I (11). Consistent with this observation, an anti-TFII-I antibody completely abrogated A3 and predominantly A1, but A2 remained unaffected to some extent. Therefore, A1 and A3 represent TFII-I-specific binding.

View larger version (25K): [in this window] [in a new window]   Fig. 5.   TFII-I binds to the two critical Inr-L sites in the 29/5 bp region. A, TFII-I is involved in the interaction with probe 29/5. Radiolabeled probe 29/5 derived from the mCD3 promoter was incubated with EL4 nuclear extract for 20 min at room temperature and then separated by 4% PAGE. Three complexes were formed, A1, A2, and A3. In a competition assay, a 100-fold excess of unlabeled wt probe 29/5 or the V-TFII-I probe was used (see "Materials and Methods"). Anti-TFII-I antibody was added 10 min before the binding reaction was initiated. B, recombinant TFII-I binds to the probe 29/5 in vitro. Radiolabeled probe 29/5 was incubated with recombinant TFII-I or nuclear extract for 20 min at room temperature before loading the gel. A 100-fold excess of unlabeled wt probe 29/5 was used for competition assay.

In agreement with the binding observed in nuclear extract, recombinant TFII-I protein bound to the same radiolabeled 25-bp probe 29/5 as shown in Fig. 5B, which showed mobility similar to that of the complex A3. This band disappeared in the presence of a cold 100-fold excess of either wt probe (Fig. 5B). Taken together, our data imply that TFII-I associates with the two Inr-L sites and possibly plays an important role in initiation of transcription by the mCD3 promoter that lacks the classical TATA box.

Ets-1, Elf-1, and NERF-2 but Not Ets-2 or NERF-1 Activate the mCD3 Promoter-- Transcriptional effects mediated by mutation M6 strongly suggest that one or more Ets family members are involved in mCD3 core promoter activity. To test for the functional involvement of Ets factors in mCD3 gene regulation, the mCD3 promoter (401/+48 bp) was transfected into L929 cells together with expression vectors for NERF-1, NERF-2, Ets-1, Ets-2, and Elf-1. The empty vector pCI was used as a negative control. As shown in Fig. 6A, cotransfection of NERF-2, Ets-1, or Elf-1 increased the mCD3 promoter activity by 2-3-fold. By contrast, NERF-1 and Ets-2 had no obvious effect on the mCD3 promoter activity. Thus, certain Ets factors are capable of conferring the ability to express mCD3 in cells that do not normally express.

View larger version (30K): [in this window] [in a new window]   Fig. 6.   NERF-2, Ets-1, and Elf-1 up-regulate the mCD3 promoter activity. A, NERF-2, Ets-1, and Elf-1 but not NERF-1 or Ets-2 up-regulate the mCD3 promoter activity in L929 cells. Expression vectors encoding Ets family members and control vector pCI were cotransfected with the mCD3 promoter 401/+48 luciferase construct into L929 cells and then subjected to the reporter gene assay. B, induction of mCD3 promoter activity by the Ets family members depends on the Ets binding site. Expression vectors encoding NERF-2, Ets-1, Elf-1 and pCI were cotransfected with either wt core promoter 37/+29 or 37/+29_M6 luciferase construct into L929 for reporter gene assay (see "Materials and Methods"). The data are representative of three independent experiments.

To test whether this induction involved the Ets recognition site in the mCD3 core promoter, NERF-2, Ets-1, and Elf-1 were transfected into L929 cells together with the 37/+29 bp promoter segment. As shown in Fig. 6B, the mCD3 core promoter activity was increased by cotransfection of NERF-2, Ets-1, or Elf-1. However, no induction was observed in 37/+29_M6 where the Ets binding site was disrupted in the mCD3 core promoter. Taken together, our data indicate that the Ets family members that are expressed in T lymphocytes, i.e. NERF-2, Ets-1, and Elf-1, play a critical functional role in the mCD3 gene regulation and may contribute in part to the T cell-specific expression pattern of mCD3.

YY1 Negatively Regulates mCD3 Gene Transcription-- YY1 can act both as a positive and negative regulator (9, 18). Although mutation M7, which disrupts the YY1 binding site, increased the activity of the 401/+48 promoter in T cells, it was possible that this mutation eliminated a negative signal in non-T cells to a greater extent than in T cells. We therefore compared the relative activity of the M7 luciferase construct with that of the wt 401/+48 luciferase construct (Fig. 4A), in both L929 and Jurkat cells. As predicted, the ratio of M7 to wt luciferase activity was 7.68 ± 0.5 in L929 cells and 2.44 ± 0.2 in Jurkat cells. This result confirmed that the M7 mutation eliminated a negative effect, possibly because of YY1 binding, upon promoter activity and showed that the strength of the effect was much greater in non-T cells.

Direct YY1 binding was examined next by EMSA using a 15/+9 probe (EY), which included the YY1 binding site partially overlapping the Ets binding site (Fig. 7A). Three specific protein-DNA complexes B1, B2, and B3 were observed when the radiolabeled EY probe was incubated with Jurkat nuclear extracts (Fig. 7B). To confirm that YY1 was involved in binding to the EY probe, an oligonucleotide containing the YY1 consensus site was used as competitor. An excess of wt YY1 consensus probe competed with band B1, but a mutated YY1 (MY) probe did not interfere with the B1 binding event (Fig. 7B). Thus, our data indicated that YY1 was potentially involved in forming the complex B1 but not B2 or B3. Conversely, when a radiolabeled probe containing the YY1 consensus site was used, one specific band appeared using the Jurkat nuclear extracts. Again, this interaction could be competed by addition of an excess of the cold wt YY1 or EY probes, but not by either of the mutant MY or YM probes (Fig. 7, A and C).

View larger version (48K): [in this window] [in a new window]   Fig. 7.   YY1 binds to the 19/+5 segment of the mCD3 core promoter. A, the sequences for probes EY (19/+5), YM, YY1, and MY are listed. Probe EY contains the critical negative element. YM contains a mutation of the YY1 binding site in the EY probe. YY1 contains the YY1 consensus binding site. Probe MY is derived from YY1 with the consensus site mutated. B, the YY1 probe competes with EY (19/+5) in binding to Jurkat nuclear extracts. The radiolabeled probe EY (19/+5) was incubated with Jurkat nuclear extracts for 20 min at room temperature before loading the gel. A total of three specific bands, B1, B2, and B3, appeared. Probes EY, YY1, and MY were used as competitors. C, the EY (19/+5) probe competes with YY1 probe in binding to Jurkat nuclear extracts. The radiolabeled probe YY1 was incubated with Jurkat nuclear extracts for 20 min at room temperature before loading the gel. Probes YY1, MY, EY, and YM were used as competitors. D, recombinant YY1 binds to the EY (19/+5) probe in vitro. Radiolabeled probe EY (19/+5) was incubated with recombinant YY1 protein or Jurkat nuclear extract for 20 min at room temperature before loading the gel. The wt probe EY is used as competitor.

To confirm the interaction between YY1 protein and the EY probe, recombinant YY1 protein was used for in vitro binding assays. The specific band, which was observed by incubation of the radiolabeled EY probe and recombinant YY1 protein, disappeared in the presence of a 50-fold excess of cold EY probe (Fig. 7D). Compared with the B1 complex, this specific band showed higher mobility possibly because of the involvement of other accessory proteins from Jurkat nuclear extract in B1 complex. Collectively, the data show that YY1 binds specifically to the 15/+9 segment of the CD3 promoter and that YY1 binding may have a negative effect on mCD3 expression.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Both our in vivo transgenic experiments and in vitro studies show that T cell-specific expression of the mouse CD3 gene is not only governed by the downstream A enhancer, but also by the TATA-less mCD3 promoter (3). Within the 401/+48 bp promoter segment, a mCD3 core promoter region located in 37/+29 bp retains a T cell-specific pattern of expression. Substitution mutant scanning of the mCD3 core promoter region was employed to localize the critical elements responsible for T cell-specific promoter activity. Thus, several regulatory elements essential for the mCD3 core promoter activity have been identified.

TFII-I is a critical initiator protein that binds to the initiator sites of human immunodeficiency virus type 1, adenovirus major late promoter, terminal deoxynucleotidyl transferase (TdT), and TCR-V 5.2 (8, 19). TFII-I is required for initiator-dependent transcription from the adenovirus major late promoter in both in vitro and in vivo assays (20). Moreover, TFII-I binds and functions through Inr-L sequences in a variant of promoters (17). The two adjacent consensus Inr-L recognition sequences in the mCD3 core promoter have central adenines at positions 25 and 13, respectively instead of at the customary +1. By contrast, in the hCD3 promoter, the central adenine of the second Inr-L site is located in transcription start site +1. Previous analyses have indicated that an initiator can be functional even though the recognition is shifted slightly from the observed start site (17). The importance of both mCD3 Inr-L sites for initiation of gene transcription is clear from the functional mutations. Like other T cell-specific TATA-less promoters (8, 11), the CD3 Inr-L sites might nucleate assembly of transcription factors and RNA polymerase II into a functional preinitiation complex, a process that could be mediated by TFII-I.

Mutation M6 in the Ets binding site dramatically decreased the 401/+48 mCD3 promoter activity, indicating an involvement of Ets family members. Cotransfection of the mCD3 promoter with the Ets family members NERF-2, Ets-1, and Elf-1 increased its activity in non-T cells. Because of its tissue-restricted distribution (21), Elf-1 is a likely partner in the concerted regulation of the mCD3 gene. Moreover, Elf-1 has been shown to play a critical role in regulation of other T cell-specific genes, including CD3, CD4, and CD5 (22-25). Studies with Ets-1-deficient mice showed defects in mature T cells but no defect in T cell development. This could be explained by the partially overlapping function among Ets family members. Because NERF-2, Ets-1, and Elf-1 are all expressed in lymphocytes, it is plausible that the involvement of NERF-2, Ets-1, or Elf-1 also contributes to the T cell-specific expression of mCD3.

A negative element is located immediately downstream from the Inr-L-2. Notably, a zinc finger transcription factor YY1 is involved in binding to this negative element. It would appear from our data that YY1 preferentially represses the mCD3 promoter activity in non-T cells. This is in agreement with the observation that YY1 has been identified as a potential repressor in other genes involved in the immune response including the interferon-, interleukin-3 and the granulocyte-macrophage colony-stimulating factor genes (18, 26, 27). YY1 repression is likely to reflect an ability to interfere with the communication between transcription activators and their targets with the general transcription machinery. The simplest mechanism of repression by YY1 is through the transcription activator displacement from their cis-acting elements within the promoter region and the recruitment of corepressor molecules. Comparison of expression of the wt mCD3 promoter with that of the M7 mutation showed that the YY1 site displayed much higher activity in L929 cells than in Jurkat cells. This possibly reflects the differential strength of function of YY1 in different cell types. Our data suggest that YY1 play an important role in repressing the non-T cell expression of mCD3 gene.

The positive element in between +6/+24 is indispensable for the mCD3 promoter activity as judged by mutations and deletions analyses. Although M11, which disrupts the minor transcription start site, dramatically decreased the mCD3 promoter activity, M9, M10, and M12 displayed very similar effects, indicating that not only the transcription start site but also other transcription factors are possibly involved in this regulation. Consistently, preliminary results indicated that CREB is involved in the regulation of mCD3 through this element, although the importance of the minor transcription start site cannot be ruled out (data not shown). Furthermore, this reminiscent of the previous discovery about the presence of CREB elements in promoter and enhancer region of many T cell-specific genes, including the CD3 enhancer, the TCR- enhancer, the TCR-V promoter, and the CD8 promoter (28-31). Certain activators such as CREB and Ets-1 have been shown to function surprisingly well even through the low affinity binding sites at core promoters in the vicinity of the start site of transcription. Interestingly, the sequence that governs tissue-specific expression of mCD3 appears to be conserved in the TCR-V8.1 promoter, the CD4 promoter, and in part also the TCR-V promoter (30, 31). Taken together, it is reasonable to envision that the interaction and cooperation among these transcription factors contributes to the T cell-specific expression pattern.

In conclusion, our data have shown that the mCD3 promoter largely contributes to a T cell-specific expression pattern both in vitro and in transgenic mouse studies. One of the most interesting aspects of this study is that the core promoter 37/+29 largely confers T cell specificity. Interestingly, we have found that several important elements located in the core promoter region, which may recruit TFII-I, Ets, YY1, and CREB, respectively. Although only NERF-2/Ets-1/Elf-1 is T cell- or lymphocyte-specific, the complex and combinative interaction of these transcription factors may partially explain the preference of expression of the mCD3 promoter in cells of T lineage.

	ACKNOWLEDGEMENTS

We thank Dr. Towia A. Libermann for the plasmids of Ets family members. We also thank Drs. Charles Gullo, Duncan Howie, and William Faubion for helpful discussion and critical reading of the manuscript and Drs. Ana Abadia-Molina, Willie Mark, and Kareem Clarke for technical support.

	FOOTNOTES

^* This work was supported in part by National Institutes of Health Grants AI-17651 (to C. T.), AI-45150 (to A. L. R.), and AR-45378 (to M. B. G).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ To whom correspondence should be addressed: Division of Immunology, RE-206, Beth Israel Deaconess Medical Center, Harvard Medical School, 330 Brookline Ave., Boston, MA 02215. E-mail: hji@caregroup.harvard.edu.

^¶ Recipient of a fellowship from the Crohn's and Colitis Foundation Association.

Published, JBC Papers in Press, September 24, 2002, DOI 10.1074/jbc.M201025200

	ABBREVIATIONS

The abbreviations used are: TCR, T cell receptor; CR, conserved region; CREB, cAMP-response element-binding protein; EMSA, electrophoretic mobility shift assay; Inr-L, initiator-like; TK, thymidine kinase; wt, wild type; YY1, yin-yang 1; PE, phycoerythrin.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				K. Tharakaraman, O. Bodenreider, D. Landsman, J. L. Spouge, and L. Marino-Ramirez The biological function of some human transcription factor binding motifs varies with position relative to the transcription start site Nucleic Acids Res., May 1, 2008; 36(8): 2777 - 2786. [Abstract] [Full Text] [PDF]

				A. D. Smith, P. Sumazin, Z. Xuan, and M. Q. Zhang DNA motifs in human and mouse proximal promoters predict tissue-specific expression PNAS, April 18, 2006; 103(16): 6275 - 6280. [Abstract] [Full Text] [PDF]

B. M. Badran, K. Kunstman, J. Stanton, M. Moschitta, A. Zerghe, H. Akl, A. Burny, S. M. Wolins