The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tam, R. C.
Right arrow Articles by Stoisavljevic, V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tam, R. C.
Right arrow Articles by Stoisavljevic, V.
The Journal of Immunology, 1999, 163: 4292-4299.
Copyright © 1999 by The American Association of Immunologists

Inhibition of CD28 Expression by Oligonucleotide Decoys to the Regulatory Element in Exon 1 of the CD28 Gene

Robert C. Tam1,*, Catherine J. Lin*, Charmaine Lim*, Bharati Pai* and Vesna Stoisavljevic{dagger}

* Immunology Laboratory and {dagger} Chemistry Laboratory, ICN Research Center, Costa Mesa, CA 92626


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Ligation of CD28 provides a costimulatory signal essential for Ag-mediated T cell activation via the TCR. Previously we demonstrated that inhibition of human and murine CD28 expression by a guanosine (G)-rich oligonucleotide (ODN), GR1, led to immunosuppression in vitro and in vivo. The bioactivity of GR1 was dependent on a G-rich DNA sequence motif consisting of two G tetrads separated by four nucleotides, (G4N4G4). We have shown recently that a G-rich region, designated CD28GR, in exon 1 of the CD28 gene is such a motif and is a positive regulatory element that binds the transcription factors Sp1 and EGR-1. Here we showed that the bioactivity of GR1 and the related GR2 correlated with the sequence-specific formation of distinct nuclear protein complexes and a high degree of ODN secondary structure. In addition, these ODN blocked transcription factor binding to CD28GR (also in a sequence-specific manner) and prevented CD28GR from driving transcription of a reporter gene. Interestingly, GR1 potently inhibited CD28, but not the expression of other Sp1- and EGR-1-regulated genes, an effect associated with lower Sp1 protein binding affinity of GR1 and GR2 compared with that of canonical Sp1 sites. These data show that DNA sequences that contain the G-rich sequence motif, G4N4G4, such as GR1 and GR2, can functionally mimic the regulatory protein binding ability of CD28GR. Thus, GR1 and GR2 act as molecular decoys to selectively interfere with transcriptional regulation of the CD28 gene.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Ligation of CD28 on T cells by its counter-receptors, B7-1 or B7-2, expressed on the surface of APCs, has been shown to induce signals that, in synergy with those derived from engagement of the TCR by an Ag bound to MHC molecules, enhance proliferation and cytokine production (1). Manipulation of this interaction can have dramatic effects on the activation (2, 3) and survival (4) of T cells. Recent evidence in vivo (5, 6) has shown that molecular intervention of the CD28 pathway can result in immunosuppression, with implications for the treatment of autoimmune diseases, organ transplantation, and graft-vs-host disease.

We have previously shown that a guanosine (G)-rich,2 phosphorothioate oligodeoxynucleotide (S-ODN), GR1, could inhibit both human and murine CD28 expression. Such interference with CD28 expression blocked Ag- and alloantigen-dependent immune responses in vitro and impaired murine contact hypersensitivity responses in vivo (7). GR1 inhibited activated levels of CD28, but not IL-2R, ICAM-1, CD2, B7-1, or B7-2, and caused no decrease in cell viability. These data suggested that GR1 induced a selective effect rather than acting as a general immunosuppressant or as a toxic agent.

The precise molecular mechanism by which GR1 exerts its effects on CD28 expression has not been previously determined. However we showed recently that inhibition of CD28 expression by the GR1 sequence was dependent on a DNA sequence motif consisting of two G-tetrads separated by four nucleotides (G4N4G4) (7). This suggested that the interaction of GR1 and its putative target, was not via an antisense mechanism (in which ODN hybridization to RNA is sensitive to single base substitutions) but rather dependent on GR1’s precise secondary structure. Such a conformational requirement is typically found in DNA/protein interactions such as the binding of a transcription factor to regulatory elements in gene promoter regions or in the aptameric association of specific ODN sequences to proteins such as thrombin (8).

The human CD28 sequence has been identified (9), characterized (10), and shown to consist of four exons, each defining a functional domain of the predicted protein. Exon 1 encodes the 5' untranslated region and the leader peptide (10). Our recent studies3 have identified a positive regulatory element, designated CD28GR (5'-GGGGAGGAGGGG-3'), which resides not in the promoter or initiation regions, but within a G-rich region in exon 1 of the CD28 gene. This G-rich region, which lies upstream of the coding sequence of the leader peptide, contains an overlapping binding site for the transcription factors Sp1 and EGR-1. Interestingly, CD28GR contains the same DNA sequence motif, G4N4G4, previously shown to be critical to the bioactivity of GR1. This observation implicated that GR1 may act by binding to a regulatory protein such as Sp1 or EGR-1 and thus prevent transcription of CD28 mRNA. In the present study we showed that the inhibitory activity of S-ODNs that contain the G4N4G4 motif, such as GR1 (ICN 16064), and a second S-ODN, GR2 (ICN 16481), was associated with displacement of the regulatory proteins Sp1 and EGR-1 from CD28GR. Using transient expression of a chloramphenicol acetyltransferase (CAT) reporter plasmid containing the CD28 exon 1 region in Jurkat T cells, we showed that both GR1 and GR2 prevented CD28GR from driving transcription of the CAT reporter gene. We also demonstrated sequence specificity as both a high degree of secondary structure and a sequence-dependent nuclear protein binding profile correlated with the increased bioactivity of GR1 and GR2. Furthermore, we showed that although GR1 was a potent inhibitor of CD28 expression, it was a poor inhibitor of other genes known to be regulated by Sp1 or EGR-1. This selectivity of GR1 for CD28 was associated with its inability to strongly compete with Sp1 binding to regulatory elements other than CD28GR (EGR-1 was not tested). These data support the view that S-ODNs with the motif G4N4G4, such as GR1 and GR2, act as molecular decoys to specifically interfere with transcriptional regulation of the CD28 gene.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Oligodeoxynucleotides (ODN)

Phosphorothioate ODN (S-ODN) in Fig. 1Go (top) were synthesized as described previously (7). The sequence of the 30-bp double-stranded (ds) ODN, ds1 (sense, 5'-GGGTTCCTCGGGGAGGAGGGGCTGGAACCC-3') is from the CD28 exon 1 region and contains the G-rich motif known as CD28GR (underlined). The dsODN, ds1, was prepared by annealing the 30-mer sense sequence shown above to its complimentary antisense sequences. This was achieved by heating equal amounts of each single-stranded 30 mer at 80°C for 5 min in 0.5 M NaCl, followed by slow cooling to room temperature. Annealed dsODNs were purified by electrophoresis on a 6% (29/1) nondenaturing acrylamide gel in 1x TBE. After staining the gel in 0.5 g/ml ethidium bromide in 1x TBE for about 30 min, gel pieces containing the annealed 30 mer were cut out, crushed, and placed in elution buffer (0.5 M ammonium acetate, 10 mM magnesium acetate, 1 mM EDTA (pH 8.0), and 0.1% SDS), followed by 3–4 h of incubation at 37°C with shaking. Eluted dsODN were eventually retrieved by spinning down gel pieces and precipitating the supernatant with 2 vol of ethanol.



View larger version (62K):
[in this window]
[in a new window]
  		FIGURE 1. The correlation of the nuclear protein binding profile of G-rich ODN sequences and ODN bioactivity. Comparison of bioactivity and nuclear protein binding profile (top panel) were determined for S-ODN with the G4N4G4 motif, GR1 and GR2 (designated ICN 16064 and ICN 16481, respectively), and for S-ODN with base substitutions within the G4N4G4 motif (changes shown in bold). ICN 16476 is a representative non-G-rich S-ODN. The top panel shows ODN sequences, ICN number, lane assignment for bottom panel, summarized nuclear protein binding profile, and bioactivity data. The bottom panel shows nuclear protein/ODN complex formation evaluated after incubation for 20 min at ambient temperature of [32P]S-ODN with HeLa cell nuclear extract in 1x gel shift binding buffer. Following resolution by electrophoresis on a 4% (80/1) nondenaturing acrylamide gel and autoradiography, nuclear extract protein/[32P]S-ODN complexes were distinguished as bands A, B, and C. [32P]S-ODN were also eluted in the absence of nuclear extract (even lane numbers) and resolved to give band D. Data are from a representative of three separate experiments. 1 ICN numbers are the same as in reference 7, Table I, except for the addition of 16 in front of each number, e.g., ICN 538 = ICN 16538 in present study. 2 Protein binding profile shown following resolution of [32P]S-ODN/nuclear extract protein interaction as predominant band A or B in the bottom panel. 3 Complexation profiles of S-ODN were compared with bioactivity data of the same S-ODN taken from Table I in Ref. 7. Bioactivity is defined as the inhibitory effect of ODN on activated human CD28 expression as determined by FACS. Low represents <40% and High represents 100% of the mean inhibition of activated human CD28 expression induced by 5 µM GR1.

 
The consensus dsODN for the Sp1 binding site (sense 5'-ATTCGATCGGGGCGGGGCGAGC-3') was purchased from Promega (Madison, WI; consensus sequence underlined). 5' labeling of ODNs was achieved using T4 polynucleotide kinase (Promega or Life Technologies, Gaithersburg, MD) and [{gamma}-32P]ATP (ICN, Irvine, CA).

Cell culture and T cell activation studies

The human T cell line Jurkat, clone E6-1 (American Type Culture Collection, Manassas, VA), was maintained as described previously (7). PBMCs were isolated from healthy donors by density gradient centrifugation followed by T cell enrichment using Lymphokwik (One Lambda, Canoga Park, CA). Purified T cells were maintained as described previously (7).

In activation studies, cells (1 x 106/ml) were incubated for 6 h (RNA) or 48 h (FACS) with plate-immobilized anti-CD3 mAb (HIT3A, 0.25 µg/ml; PharMingen, San Diego, CA) and 10 ng/ml PMA (Calbiochem, La Jolla, CA). The in vitro treatment regimen with S-ODN, GR1, and GR2, was conducted as previously described (7).

Measurements of thermodynamic melting temperature

The thermodynamic melting experiments were conducted on a Varian UV spectrophotometer equipped with an electronic temperature controller and Cary hybridization software. Single-stranded, G-rich, phosphorothioate ODNs from Fig. 1Go and a non-G-rich control, ICN 16476, were analyzed. All ODNs (2 µM) were analyzed as previously described (11) and presented as thermodynamic melting temperature curves across a temperature range of 20–90°C.

CD28-CAT reporter plasmids

Total RNA was prepared from 32 x 106 Jurkat cells using Qiagen RNeasy Mini Kit and Qiagen Qiashredder (Qiagen, Valencia, CA). cDNAs for TA cloning were generated following reverse transcription of 1.8 µg of the Jurkat total RNA (60°C, 30 min) and amplification under the following conditions: 1-min denaturation at 94°C, 35 cycles of 15 s at 94°C and 30 s at 60°C, followed by a final extension of 7 min at 60°C using the Perkin-Elmer EZ rTth RNA PCR kit (Perkin-Elmer/Roche Molecular System, Branchburg, NJ). Primers (sense, 5'-CCTGTGTGAAATGCTGCAGT; antisense, 5'-AAGTTGAGAGCCAAGAGCAG) were used to amplify the exon 1 region +26 to +251 of the human CD28 gene. This construct was designated 28b. Following gel purification using the Qiagen Qiaquick Gel Extraction Kit, the cDNAs were subcloned into TA cloning vector pCR2.1 (Invitrogen, San Diego, CA) in the antisense orientation using standard TA cloning techniques. The 28b CAT reporter plasmid, pCAT3e28b, was constructed by inserting the SacI-XhoI fragment of the pCR2.1-antisense construct into the SacI-XhoI site of pCAT3enhancer (pCAT3e) reporter plasmid (Promega).

The insert for the plasmid pCAT3e28h-1 was made using the method of oligonucleotide-directed mutagenesis in which 30 bp of the wild-type CD28 exon 1 sequence (+172 to +201, 5'-GGGTTCCTCGGGGAGGAGGGGCTGGAACCC-3', CD28GR underlined) were deleted and substituted with 5'-TCATCACAGGGTGCT-3'. This was achieved by first generating two fragments, one 5' and one 3'. To generate the 5' fragment, an original 28b 5' amplimer and a 3' primer that was extended with mutated sequences and tagged with a DraIII site was used. A 3' fragment was generated using a DraIII-tagged 5' primer also extended with mutated sequences, and an original 28b 3' amplimer. All DNA PCRs were performed using AmpliTaq DNA polymerase, Stoffel fragment (Perkin-Elmer) and the following amplification conditions: 95°C for 1 min; 35 cycles of 95°C for 15 s, 60°C for 30 s, and 72°C for 15 s, followed by 72°C for 7 min. After PCR, the 5' fragment contained a terminal 3' DraIII site, and the 3' fragment contained a terminal 5' DraIII site. Following an overnight digestion with DraIII (Boehringer Mannheim, Indianapolis, IN), fragments were purified using either the Qiagen Qiaquick Nucleotide Removal Kit or CentriSpin-20 and CentriSpin-10 columns (Princeton Separation, Adelphia, NJ), ligated, and reamplified again using both 5' and 3' original 28b amplimers. The DNA insert, 28h-1, now mutated in the G-rich region, +172 to +201, was then subcloned into the pCR2.1 vector as before. All CD28 reporter constructs were verified by sequencing (Retrogen Sequencing Service, San Diego, CA).

Transient expression and CAT assays

Cells were transfected using Qiagen Superfect transfection reagent, and each transfection was performed in duplicate. Cells to be transfected were split from confluent stocks 1 day before transfection. Transient expression assays were performed by transfection of 4 x 106 cells with 5 µg of the plasmid DNA using Qiagen Superfect transfection reagent (Qiagen) as directed. After 24 h, cells were harvested and lysed for CAT assay (CAT Enzyme Assay System, Promega). CAT activities in whole cell extracts were determined using the scintillation method as described previously (12). Briefly, whole cell extracts were incubated with [14C]chloramphenicol (100 Ci/ml; ICN) and n-butyryl coenzyme A (Promega) at 37°C for 1 h. The reaction mixture was then extracted with xylene (ICN). The xylene phase was back-extracted a second time with 0.25 M Tris-HCl, pH 8.0, mixed with scintillant, and counted in a scintillation counter.

A negative control vector (a promoter-less SV40 enhancer containing the CAT gene, pCAT3e) was also transfected in each experiment. The relative CAT activity of each construct was derived from the ratio of CAT expression by test plasmid over CAT expression of the negative control, pCAT3e.

EMSAs

Protein-DNA interactions were determined by EMSA using 32P end-labeled dsODN, ds1 and Sp1 consensus ODN; S-ODN; or SacI-XhoI-cDNA fragments of the exon 1 DNA fragments, 28b and 28h-1 (both ~300 bp; excised from their respective pCR2.1 constructs). Following incubation for 20 min at ambient temperature, the interaction of end-labeled cDNA, S-ODN, or dsODN (50,000–100,000 cpm) with either 2–4 footprint units of purified Sp1 protein (Promega) or 10 µg of HeLa nuclear extracts (Promega) in 1x gel shift binding buffer (1 mM MgCl2, 0.5 µM EDTA, 0.5 mM DTT, 50 mM NaCl, 10 mM Tris-HCl (pH 7.5), and 50 µg/ml poly(dI-dC)·poly(dI-dC); Promega) was resolved by electrophoresis on a 4% (80/1) nondenaturing acrylamide gel.

Gels were run in 0.5x TBE at 100 V for 2–3 h, dried, and subjected to overnight autoradiography using a PhosphorImager (Bio-Rad, Hercules, CA). Unlabeled consensus dsODN sequence for Sp1 (Promega) and S-ODN from Fig. 1Go were used in competition experiments. Gel-shift grade Ab specific for Sp1 (mouse monoclonal IgG1, clone 1C6) and EGR-1 (rabbit polyclonal IgG, clone 588) were purchased from Santa Cruz Biotech (Santa Cruz, CA). For competition gel shift using purified Sp1 protein, a 3.7- to 100-molar excess of nonlabeled dsODN, ds1 and Sp1 consensus ODN or S-ODN from Fig. 1Go were preincubated with Sp1 protein in 1x binding buffer at room temperature for 1 h before incubation with labeled probe (20–30 min). For competition gel shift using HeLa nuclear extract, a 100-molar excess of nonlabeled S-ODN from Fig. 1Go, was preincubated with HeLa nuclear extract in 1x binding buffer at room temperature for 1 h before incubation with labeled probe (20–30 min). In supershift assays, HeLa nuclear extract was incubated with labeled probe (20–30 min) before 1-h incubation with anti-Sp1 Ab or anti-EGR-1 Ab (2 µg). The anti-EGR-1 Ab clone 588 has been previously shown in EMSA analyses to abolish the binding of Egr-1 to the Egr-1 binding site of the IL-2 promoter (13).

Analysis of c-myc mRNA

Extraction of total cellular RNA from human PBMCs and generation of cDNA by RT was achieved as described previously (7). PCR of cDNA was performed using primers for human CD28 (7) and for c-myc (Stratagene, La Jolla, CA) according to the manufacturer’s instructions.

Separation of PCR products, Southern blotting, and hybridization were conducted as described previously (7). Blots were hybridized with 32P-labeled ODN probes generated as previously reported (CD28) (7) or generated from the original primers (c-myc). Equivalent loading was assessed using the reporter gene, pHE7 as described previously (7). Washed blots were then analyzed using a PhosphorImager (Molecular Dynamics, Sunnyvale, CA). Densitometry was performed on the images to evaluate the ratio of specific mRNA expression relative to mRNA expression of the housekeeping gene, pHE7.

Immunofluorescence analysis

Human T cells were stained with a combination of PE-CD3 (Becton Dickinson, San Jose, CA) and FITC-CD102 (ICAM-2, Alexis Corp., San Diego, CA), PE-CD3 and FITC-CD95 (Fas, PharMingen, San Diego, CA), PE-CD3 and FITC CD44 (Becton Dickinson), PE-CD28 and FITC-CD3 (both from Becton Dickinson), or PE/FITC-labeled isotype-matched control mAbs before analysis with a FACScan flow cytometer (Becton Dickinson).

Ag density was determined following two-color flow cytometric analysis of 10,000 viable cells. Surface expression of specific Ag (CD102, CD95, CD44, or CD28) on gated CD3+ cells was determined using CellQuest software and is expressed as the mean channel fluorescence.

Statistical analysis

Statistical significance, where relevant, was assessed using modified one-way ANOVA (Kruskal-Wallis test). Trend analysis, as appropriate, was assessed using simple regression analysis.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Protein binding profile of G-rich ODN sequences in HeLa nuclear cell extract

Our previous data (7) suggested that secondary structure was important for the in vitro activity of S-ODN with the motif (G4N4G4), such as GR1 and GR2. Bases were selectively added, deleted, or substituted in GR1, and the magnitude of inhibition of CD28 expression by these ODN was assessed relative to that of the parent ODN. Base substitutions or deletions within each of the two G tetrads significantly reduced bioactivity of GR1, whereas ODN with only one G tetrad had no activity. The inhibitory activity of GR1 was unhindered by the number of bases flanking the two G tetrads, but was decreased substantially when the number of residues between the two G tetrads was greater or less than four bases. In addition, substitutions within the four bases that are flanked by the two G tetrads (-AAGA-, -ATAT-, and -CTCT-) did not alter the ability of the motif to inhibit CD28 expression (R. C. Tam, S. Wu-Pong and B. Pai, unpublished observations). Therefore, the biological activity of GR1 was dependent on a specific sequence motif comprising two sets of four contiguous Gs separated by four residues. This finding suggested that the interaction of these S-ODN and the putative target was dependent on the precise secondary structure of GR1 and GR2. Such a conformational requirement is typically found in DNA/nuclear protein interactions.

Here we examined whether the interaction of GR1, GR2, and other G-rich S-ODN (sequences are shown in Fig. 1Go, top panel) with nuclear proteins resulted in 1) distinct sequence-specific nuclear protein-binding profiles and 2) a correlation between specific nuclear protein-binding profiles and ODN bioactivity. Nuclear protein-binding profiles were determined by EMSA following incubation of 32P-labeled S-ODN with HeLa nuclear extract. The interaction of HeLa nuclear extract with [32P]GR1 (ICN 16064) and [32P]GR2 (ICN 16481), resolved as band A (Fig. 1Go, bottom panel). This complex was absent upon coincubation of HeLa extract and 32P-labeled G-rich S-ODN mutated within the G-rich motif (ICN 16538, 16485, and 16480; Fig. 1Go, bottom panel). In contrast, these mutated G-rich sequences and a non-G-rich S-ODN, ICN 16476, formed complexes that resolved as a more electrophoretically retarded band (B; Fig. 1Go, bottom panel).

The bioactivity data of various ODN were taken from a previous study (Table I in reference 7) and used for comparative analysis with the protein binding profiles of the same ODN (Fig. 1Go, top panel). The presence of band A correlated with elevated bioactivity in GR1 and GR2, as determined by the magnitude of inhibition of human CD28 expression (Fig. 1Go, top panel). The presence of band B and possibly a second band, C, correlated with reduced bioactivity. As band B was distinct in all sequences except GR1 and GR2 (including a non-G-rich ODN, ICN 16476), this band represented sequence-independent, nonspecific protein binding, an effect that may be related to the presence of phosphorothioate linkages in these sequences. Collectively, these data showed that the inhibitory activity of S-ODN with the motif G4N4G4, such as GR1 and GR2, was associated with a specific nuclear protein binding profile.

Comparison of secondary structure formation of G-rich ODN

The presence of a distinct band A and a substantially reduced band B in GR1 and GR2 suggests that these S-ODN may have alternate secondary structures from the other G-rich S-ODN, a property that may enable them to form specific protein complexes. Using thermodynamic melting experiments we had previously shown that GR1 had a distinct melting curve across the temperature range 20–70°C (14). This demonstrated that this ssODN had substantial secondary structure. Here we compared the secondary structure profiles of GR1 and GR2 with two G-rich S-ODNs with substitutions within the two G tetrads, ICN 16538 and ICN 16480, and ICN 16485, which only had one G tetrad, as well as a non-G-rich S-ODN, ICN 16476 (Fig. 2Go; all sequences from Fig. 1Go). Although all the G-rich ODN had a high degree of secondary structure, the melting curves of GR1 and GR2 showed the highest amount of hyperchromicity (change in absorbance across the temperature range studied). The non-G-rich S-ODN, ICN 16476, had no melting curve. These data showed that the highest degree of secondary structure was observed with the S-ODN, GR1 and GR2, the same ODN that formed the specific band A protein-DNA complex and showed the greatest bioactivity. In addition the presence of some secondary structure in the other G-rich S-ODN suggests that this physical property may be responsible for the low sequence-independent bioactivity previously observed with the S-ODN that do not have the G4N4G4 motif (7).



View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 2. The impact of sequence motif on secondary structure of various G-rich S-ODN. The thermodynamic melting curves of single-stranded G-rich S-ODN, GR1 (16064), GR2 (16481), ICN 16538 (16538), ICN 16485 (16485), and ICN 16480 (16480) were compared across a temperature range of 20–90°C. The greater the hyperchromicity, as determined by change in absorbance at 260 nm, the greater the amount of secondary structure. A control non-G-rich S-ODN, ICN 16476 (16476), was run in parallel. All S-ODN sequences are from Fig. 1Go. Data are from a representative of three separate experiments.

 
S-ODN with the G4N4G4 motif can compete for the Sp1/EGR-1 overlapping binding site in CD28GR

We had previously shown that the regulatory element in exon 1 of the CD28 gene, CD28GR, had an overlapping Sp1 and EGR-1 binding site.3 In that study we showed that binding of these transcription factors (from nuclear extracts) to CD28GR was abrogated when the G-rich sequence in exon 1 was mutated. Here using EMSA, we examined whether GR1 or GR2, which both have the same G-rich motif as CD28GR, could prevent Sp1 and EGR-1 from interacting with CD28GR. The interaction of a 32P-labeled 30-bp native sequence containing CD28GR (ds1) with HeLa nuclear extract (a known source for Sp1 and EGR-1 transcription factors), resolved as two broad bands (Fig. 3GoA, lane 3), with the upper band aligning to where [32P]ds1 bound to purified Sp1 protein (Fig. 3AGo, band S, lane 2). We have previously3 designated this upper band the ES complex, because both Sp1/[32P]ds1 and the EGR-1/[32P]ds1 complexes in HeLa and Jurkat cells comigrated to this same position, a comigration phenomenon also observed by others with the overlapping Sp1 and EGR-1 binding ZIP site in the IL-2 promoter (13). The identity of the bottom band is not known at present. The ES complex was supershifted following addition of Sp1 Ab (lane 4) and was partially reduced following addition of EGR-1 Ab (Fig. 3GoA, lane 5). This data showed the presence of an overlapping Sp1/EGR-1 binding site in CD28GR (ds1). More importantly, we showed that addition of a 100-fold molar excess of unlabeled GR1 and GR2 could compete off the [32P]ds1/HeLa extract complex (Fig. 3GoA, band S, lane 3) or [32P]ds1/purified Sp1 protein complex (Fig. 3GoB, band S, lanes 3 and 4). In contrast, no similar competition was observed using a 100-fold molar excess of unlabeled ICN 16480 (a G-rich S-ODN with substitutions within the two G tetrads; Fig. 3Go, lane 9) or ICN 16485 (a G-rich S-ODN with only one G-tetrad; Fig. 3Go, lane 10) as well as a non-G-rich S-ODN, ICN 16476 (Fig. 3Go, lane 8; all sequences from Fig. 1Go). The competitive effect of the S-ODN on the Sp1/ds1 complex is not the result of DNA/DNA interaction, because 32P-labeled GR1 did not bind to the unlabeled CD28 exon 1 DNA insert (data not shown). In addition, Sp1 formed a distinct complex with 32P-labeled GR1 or GR2, but not S-ODN, ICN 16538, or ICN 16476 (data not shown). Collectively, these data showed that S-ODN with the motif G4N4G4, such as GR1 and GR2, can disrupt ES complex formation by interacting with the transcription factors, Sp1 and EGR-1, and can thus act as molecular decoys by competing for the overlapping Sp1/EGR-1 binding site in CD28GR.



View larger version (47K):
[in this window]
[in a new window]
  		FIGURE 3. S-ODN with the G4N4G4 motif can compete for the Sp1/EGR-1 overlapping binding site of the regulatory element, CD28GR. A, EMSA studies were performed to determine whether CD28GR is an overlapping Sp1/EGR-1 binding site and to evaluate whether GR1 and GR2 can compete with transcription factor binding to CD28GR in a sequence selective manner. The interaction of [32P]ds1 (dsODN encompassing CD28GR) with purified Sp1 protein (lane 2) or with HeLa cell nuclear extract in the absence (lane 3) or the presence of specific Ab to Sp1 (lane 4) or EGR-1 (lane 5) or a 100-fold molar excess of unlabeled competitor ODN, GR1 (lane 6), GR2 (lane 7), ICN 16476 (lane 8), ICN 16480 (lane 9), or ICN 16485 (lane 10) was determined by EMSA. [32P]ds1/Sp1 resolved as band S (lane 2, Fig. 2GoA). [32P]ds1/HeLa nuclear protein complexes resolved as two main bands. The upper band, which aligns to band S, is the ES complex, previously3 shown to be the position where Sp1/[32P]ds1 and the EGR-1/[32P]ds1 complexes in HeLa and Jurkat nuclear cell extract comigrated. EMSA using Sp1 and EGR-1 Ab were performed by allowing complexation (20 min) of [32P]ds1 with HeLa nuclear extract before incubation with 2 µg of Ab for 1 h at room temperature. EGR-1 Ab does not directly bind to [32P]ds1 (data not shown). B, EMSA studies were performed by incubating [32P]ds1 with purified Sp1 protein alone (2 and 3 footprint units, lanes 3 and 4, respectively) or in the presence of unlabeled ds1 (lane 5), GR2 (lane 6), GR1 (lane 7), or ICN 16476 (lane 8). [32P]ds1/Sp1 complex resolved as band S. [32P]ds1 was also eluted in the absence of Sp1 protein (lane 2). 32P-labeled BMB m.w. marker VIII (Boehringer Mannheim) are shown in lane 1 in both gels (A) and (B). For competition gel shift, a 100-fold molar excess of unlabeled S-ODN from Fig. 1Go, was preincubated with HeLa nuclear extract (A) or purified Sp1 protein (B) in 1x binding buffer at room temperature for 1 h before incubation with labeled probe for 20 min.

 
S-ODN with the G4N4G4 motif inhibit transcription driven by CD28GR

Because GR1 and GR2 appeared to compete with transcription factors to the Sp1/EGR-1 binding site in CD28GR, we next assessed whether this competition by GR1 and GR2 could impact the regulation of transcription driven by CD28GR. In mutagenicity studies, CD28GR (+181 to +192) was deleted from insert 28b (which contains the exon 1 region +26 to +251) and substituted with alternate sequences. One of these mutants, 28h-1, contains an exon 1 construct in which the G-rich region encompassing CD28GR (+172 to +201) was deleted and substituted with an alternate 15-bp sequence. This mutated exon 1 sequence, 28h-1, in contrast to the wild-type exon 1 sequence, 28b, does not bind Sp1 (Fig. 4GoA). The CAT activity of both pCAT3e28b and the plasmid containing the 28h-1 insert, pCAT3e28h-1, was then assessed in the presence or the absence of 5 µM GR1 or GR2, added 45 min following transfection of the plasmid into Jurkat T cells. Unexpectedly, the plasmid pCAT3e28h-1, showed similar transcriptional activity as pCAT3e28b, as assessed by monitoring the CAT activity of each reporter plasmid (Fig. 4GoB). We presumed that this transcriptional activity was the result of new regulatory sites introduced following mutagenesis. Interestingly, the CAT activity of pCAT3e28b (which contains an intact CD28GR) dropped significantly (p = 0.02) in the presence of GR1 (62% inhibition) or GR2 (57% inhibition), compared with transient expression of pCAT3e28b alone (Fig. 4GoB, left panel). In contrast, GR1 (10% inhibition) and GR2 (12% inhibition) had no significant effect on CAT activity of the mutated plasmid, pCAT3e28h-1. The effect on CAT activity following transfection of pCAT3e28b in the presence of control G-rich ODN, ICN 16480, was evaluated and compared with the CAT activity of GR2 (Fig. 4GoB, right panel). While neither GR2 nor ICN 16480 had an inhibitory effect on CAT activity driven by pCAT3e28h-1, GR2 (59%), but not ICN 16480 (-11%), significantly inhibited CAT activity driven by pCAT3e28b. These data showed that transcription of the CAT reporter gene driven by CD28GR could be specifically inhibited by G-rich S-ODN with the motif G4N4G4, such as GR1 and GR2. Transcription of the CAT reporter gene in a CD28 exon 1 reporter plasmid devoid of CD28GR was not inhibited by GR1 and GR2.



View larger version (33K):
[in this window]
[in a new window]
  		FIGURE 4. S-ODN with the G4N4G4 motif inhibit transcription driven by CD28GR. A, EMSA data showing the interaction of 32P-labeled dsDNA 28b (lane 2) and 28h-1 (lane 3) with purified Sp1 protein. Sp1/dsDNA complex formation was evaluated upon incubation of 32P-labeled dsDNA with purified Sp1 protein for 20 min at ambient temperature in 1x gel shift binding buffer. The 32P-labeled 300-bp 28b fragment resolved in the absence of Sp1 is shown in lane 4. 32P-labeled BMB m.w. markers VIII are shown in lane 1. B, Jurkat T cells (4 x 106) were transiently transfected cells with 5 µg of wild-type pCAT3e28b or mutant pCAT3e28h-1 plasmid using Qiagen Superfect transfection reagent. CAT gene expression was determined in two independent experiments for each reporter plasmid in the absence (filled bars) or the presence of 5 µM GR1 (hatched bars) and GR2 (open bars; Expt. 1, left panel) or 5 µM GR2 (open bars) and ICN 16480 (vertical striped bars; Expt. 2, right panel). A negative control vector (a promoter-less but SV40 enhancer-containing CAT reporter plasmid, pCAT3e) was also transfected in each experiment. Data are presented as relative CAT activity (mean ± SD; n = 4), determined as the ratio of counts per minute of pCAT3e with cDNA insert over pCAT3e alone. The mean CAT expression (counts per minute ± SD; n = 4) in Expt. 1 and Expt. 2, respectively, were: for pCAT3e28b alone, 197,759 ± 15,635 and 127,641 ± 7,366; for pCAT3e28h-1 alone, 203,864 ± 4,258 and 136,584 ± 14,569; and pCAT3e (no insert), 21,960 ± 8,161 and 13,410 ± 6,566. *, p = 0.02 vs pCAT3e28b alone.

 
Comparative effect of GR1 on expression of CD28, CD44, FAS, ICAM-2, and c-myc

Overlapping Sp1 and EGR-1 binding sites occur frequently among human gene promoters such as IL-2 (13), TNF-{alpha} (15), superoxide dismutase (16), IL-2R (17), platelet derived growth factor-{alpha} and -ß (18, 19), and CD19 (20). The presence of the overlapping Sp1/EGR-1 binding site in CD28GR supports the need to determine the effect of GR1 on the expression of genes other than CD28 that are controlled by the ubiquitous transcription factors, Sp1 and EGR-1. For Sp1, these include ICAM-2 (21) and c-myc (22), and for EGR-1, these include ICAM-1 (23) IL-2R (17), CD44 (24), and Fas (25). To address this we determined the effect on cell surface expression of CD28, CD44, Fas (CD95), and ICAM-2 (CD102) imparted by treatment with GR1 in activated human T cells. From immunofluorescence studies, we showed that GR1, in the dose range 1.25–20 µM, significantly induced a dose-dependent inhibition of activated CD28 expression (p < 0.0001), with maximal inhibition of 79% at 20 µM (Fig. 5GoA). At 20 µM, GR1 inhibited CD44, Fas, or ICAM-2 expression by 0, 6, and 13% inhibition, respectively. No significant relationship between GR1 level and CD44, Fas, or ICAM-2 expression was observed (Fig. 5GoA). A G-rich control ODN, ICN 16480, had previously been shown to have substantially reduced inhibitory activity against CD28 expression compared with GR1 (7). This G-rich control ODN did not show any inhibitory effect on surface expression of CD44, an EGR-1-regulated gene (data not shown). The expression of both CD28 and c-myc mRNA levels have previously been shown to be substantially elevated following activation with plate-bound anti-CD3 and PMA after 6 h (7, 26). From densitometry data generated following RT-PCR analyses, we showed that GR1 induced a dose-dependent inhibition of both CD28 and c-myc mRNAs from activated human PBMCs. However the effects of 10, 5, and 2 µM GR1 on activated CD28 expression (85, 54, and 34% inhibition, respectively) were more potent than those on activated c-myc expression (32, 2, and 0%, respectively; Fig. 5GoB). Interestingly, we have shown previously that GR1 has little effect on the expression of EGR-1-regulated ICAM-1 and IL-2R genes, but can inhibit another EGR-1-regulated gene, IL-2. The effect of GR1 on IL-2, however, was dependent on the presence of CD28 (7). These effects on gene expression were not influenced by apparent toxicity of GR1 treatment as we have shown previously (7) that cell viability, as assessed by propidium iodide (5 µg/ml) exclusion in untreated and ODN-treated CD4+ T cells from multiple donors, was typically >90% (range, 90–99%) following 48-h incubation with 1–10 µM of each batch of all ODNs. Together, these data support the view that although CD28 expression may be regulated by the ubiquitous transcription factors, Sp1 and EGR-1, GR1 and GR2 selectively affect CD28 expression.



View larger version (28K):
[in this window]
[in a new window]
  		FIGURE 5. GR1 differentially regulates CD28 compared with other Sp1- and EGR-1-regulated genes. A, Human T cells were activated with anti-CD3/PMA in the presence of 0–20 µM GR1. Twenty-four hours later a second dose of 0–20 µM GR1 was added. Forty-eight hours following initiation of activation, cells were harvested and washed in isotonic saline. Cell surface expression of CD28, CD44, CD95 (Fas), or CD102 (ICAM-2) was determined by FACS in these T cells. Resting levels ({diamondsuit}) of each surface Ag are shown in individual panels. Data (mean of quadruplicate determinations ± SD) are shown as mean channel fluorescence and are representative of three separate experiments. B, Total RNA was isolated from human PBMCs that were resting (R), activated with anti-CD3/PMA (6 h; Ac), or activated with anti-CD3/PMA in the presence of 2, 5, and 10 µM GR1 (Ac + GR1). CD28, c-myc, or pHE7 mRNA levels were assessed following RT-PCR and Southern blot analysis with specific 32P-labeled DNA probes. Data are representative of two separate experiments. *Ratio, the ratio of densitometry units (following background subtraction) from CD28 or c-myc over the reporter gene, pHE7.

 
The effectiveness of GR1 and GR2 as molecular decoys is dependent on the affinity of Sp1 to its target DNA

The ODN consensus sequences for Sp1 and EGR-1 are GGGGCGGGG and GCGGGGGCG, respectively, whereas the G-rich motif in CD28 exon 1, CD28GR, is GGGGAGGAGGGG. Based on the sequence differences, the Sp1 or EGR-1 binding ability of CD28GR may also differ. Thus, the effectiveness of GR1 and GR2 on CD28, but not other Sp1- or EGR-1-regulated genes, could be dependent on how well GR1 and GR2 compete with regulatory elements, including CD28GR, for Sp1 or EGR-1 binding. To address this hypothesis we compared the abilities of GR1 and GR2, and a consensus Sp1 dsODN (representing a canonical Sp1 binding site of a Sp1-regulated gene) to compete with the binding of purified Sp1 protein to either [32P]ds1 (containing CD28GR) or [32P]Sp1 consensus dsODN (containing an Sp1-binding regulatory element; Fig. 6Go). We showed that when Sp1 bound to CD28GR ([32P]ds1), GR1 and GR2 competed as well as Sp1 consensus ODN for Sp1 binding (Fig. 6GoA). However, unlabeled GR1 and GR2, at 1.2 pM (11.1-fold molar excess) and 0.4 pM (3.7-fold molar excess), were substantially weaker competitors for Sp1 protein binding to [32P]Sp1 consensus ODN than unlabeled Sp1 consensus ODN at the same concentrations (Fig. 6GoB). These data suggest that GR1 and GR2 are poorer competitors for Sp1 binding to regulatory elements other than CD28GR. We were unable to evaluate the comparative binding affinities of EGR-1 to GR1 and of EGR-1 to Egr consensus ODN, as purified gel shift grade EGR-1 protein was not readily available.



View larger version (65K):
[in this window]
[in a new window]
  		FIGURE 6. Comparison of the displacement of [32P]ds1 (A; dsODN encompassing CD28GR) and 32P-labeled consensus Sp1 ODN (B) from Sp1 protein by unlabeled competitor ODNs. The effects of 100-, 33.3-, 11.1-, and 3.7-fold molar excesses, respectively, of cold Sp1 consensus ODN (lanes 3–6), GR1 (lanes 7–10), and GR2 (lanes 11–14) ODN on [32P]ds1/Sp1 (A, lane 2) and on [32P]Sp1 dsODN/Sp1 (B, lane 2) complex formation was monitored by EMSA. Unlabeled ODN were preincubated with protein in 1x binding buffer at room temperature for 1 h before a 20-min incubation with labeled probe. [32P]ds1 alone and [32P]Sp1 dsODN alone are shown in lane 1 in A and B, respectively. Lane 15 in A and B is [32P]BMB m.w. marker VIII.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Our previous studies had identified ODN with a sequence motif, G4N4G4, which potently inhibited CD28 expression (7). The mechanism by which these ODNs elicited their inhibitory effects was unknown. Recent studies in our laboratory revealed that such a sequence motif exists in the exon 1 region of the CD28 gene. Interestingly, this motif was shown to be critical for positive transcriptional regulation and was an overlapping Sp1/EGR-1 binding site. Mutation of this region resulted in loss of Sp1/EGR-1 binding and almost complete loss of transcriptional activity.3 We thus designated this regulatory element CD28GR. The focus of this study was 1) to determine whether the sequence motif-containing ODNs could act as molecular decoys to CD28GR, 2) to evaluate the specificity of G-rich motif-containing ODN sequences in the postulated mechanism of inhibition of CD28 gene expression, and 3) to determine whether bioactive ODNs could selectively inhibit expression of the specified target gene, CD28, without affecting other similarly regulated genes.

In this study we showed that the bioactivity of S-ODN with the motif G4N4G4, such as GR1 and GR2, correlated with distinct nuclear protein-ODN binding profiles and the presence of a high degree of ODN secondary structure. This suggested that the inhibitory activity of these active ODN was dependent on a specific protein/ODN interaction. The loss of bioactivity, the presence of alternate nuclear protein-ODN binding profiles, and the reduction of a secondary structure in S-ODN that had mutations to the two G tetrads or only contained one G tetrad provided strong evidence that this protein/ODN interaction was dependent on a precise G-rich sequence motif. More importantly, GR1 and GR2 could substantially abolish transcription factor binding to CD28GR and could inhibit CD28 exon 1-driven transcription of CAT reporter gene, but only in the presence of an intact CD28GR region. These observations provide clear evidence that G4N4G4 motif-containing S-ODNs inhibit CD28 expression by acting as molecular decoys to the overlapping Sp1/EGR-1 binding regulatory element, CD28GR.

S-ODN containing a single G tetrad have been previously shown to have sequence-independent antiproliferative effects (27, 28). The presence of such a high G content and a phosphorothioate backbone in the GR1 and GR2 sequences necessitates consideration of whether the nonspecific binding of proteins contributes to the bioactivity of these ODN. Several lines of evidence suggest that nonspecific protein binding is not the predominant mechanism by which GR1 and GR2 exert their biological effects. Firstly, the mere presence of a G tetrad does not necessarily result in nonsequence-specific effects on gene expression, as we and others have shown previously (7, 29, 30, 31). Secondly, G quartet structures have been shown to bind to specific proteins functionally in vivo, such as in telomeres, centromeres (32), and Ig switch regions (33). Thirdly, mutated G-rich sequences (where both G tetrads are disrupted or only one G tetrad is present) did not compete with CD28GR for Sp1 and EGR-1 binding. Fourthly, a non-G-rich phosphorothioate sequence, ICN 16476, showed none of the characteristics of the G-rich S-ODN. Finally, GR1- and GR2-mediated inhibition of CD28 exon 1-driven transcription requires the presence of an intact CD28GR. We did not attempt to boil the ODNs to determine whether such a treatment would disrupt the ability of GR1 and GR2 to compete with the CD28GR-Sp1/EGR-1 (ES) complex. This effect is only temporary and the possibility of reversibility of the G quartet disruption drove us to choose more permanent disruption by mutating the G tetrads. This type of disruption of the G4N4G4 motif led to decreased competition for binding to the ES complex, lowered the degree of secondary structure, and more importantly decreased bioactivity. Collectively, these data provide further evidence that the bioactivity of GR1 and GR2 is not the result of nonspecific effects, but is associated with a defined secondary structure, a property that allows sequence-specific competition of the CD28GR-Sp1/EGR-1 complex.

The use of DNA-based transcription factor decoys has been described for AP-1 (34), E2F (35), and cAMP response element (36) and has been shown to be a functional approach to inhibit specific transcription factor-regulated genes (37, 38). These decoys are designed using the protein-binding DNA consensus sequence. Transcription factors tend to be ubiquitous regulatory proteins; therefore, the action of these DNA decoys may induce unnecessary abrogation of other unrelated cellular processes. As CD28GR binds Sp1 and EGR-1, both zinc finger proteins that regulate a variety of genes through binding to consensus G-rich motifs (21, 22, 23, 24, 25), it was important to determine whether GR1 and GR2 only mediated the inhibition of CD28GR-driven transcription or whether they could nonselectively abrogate expression of other Sp1 and EGR-1-regulated genes. Our studies established that the inhibitory activity of GR1 and GR2 was selective for CD28 expression; both ODN had little influence on the expression of several Sp1 and EGR-1-regulated genes. Furthermore, we have shown that GR1 and GR2 efficiently interfere with Sp1 binding to the Sp1/EGR-1 binding site, CD28GR, but were unable to abrogate Sp1 binding to canonical Sp1 binding sites with the same efficiency. Based on this observation, one could postulate that the selectivity to the target gene, CD28, by these S-ODN sequences is the result of similar binding affinities of GR1, GR2, and CD28GR for Sp1, whereas Sp1-regulated genes that contain canonical Sp1 binding sites have greater binding affinity for Sp1, preventing these ODN decoys from being effective at inhibiting the expression of genes other than CD28.

In conclusion, as CD28 is involved in many critical signaling pathways of T cell activation, GR1- and GR2-mediated inhibition of CD28 expression may provide the means to regulate these cellular processes and thus have therapeutic potential in preventing unwanted T cell activation such as in allergy, autoimmune diseases, and transplant rejection.


   Acknowledgments
 
We thank Dr. S. Wu-Pong for critical evaluation of this manuscript, and J. Avalos for collection of blood from normal donors and for excellent technical assistance.


   Footnotes
 
1 Address correspondence and reprint requests to Dr. Robert C. Tam, ICN Research Department, 3300 Hyland Avenue, Costa Mesa, CA 92626. E-mail address: Back

2 Abbreviations used in this paper: G-rich, guanosine-rich; N, nucleotide; CAT, chloramphenicol acetyltransferase; ODN, oligodeoxynucleotide; S-ODN, phosphorothioate ODN; dsODN, double-stranded ODN. Back

3 C. J. Lin and R. C. Tam. Identification of a novel regulatory element, CD28GR, which resides in exon 1, is a binding site for the transcription factors, Sp1 and EGR-1, and is critical for expression of the human CD28 gene. Submitted for publication. Back

Received for publication June 21, 1999. Accepted for publication August 9, 1999.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Linsley, P. S., J. A. Ledbetter. 1993. The role of the CD28 receptor during T cell responses to antigen. Annu. Rev. Immunol. 11:191.[Medline]
  2. Harding, F. A., J. G. McArthur, J. A. Gross, D. H. Raulet, J. P. Allison. 1992. CD28-mediated signalling co-stimulates murine T cells and prevents induction of anergy in T cell clones. Nature 356:607.[Medline]
  3. Tan, P., C. Anasetti, J. A. Hansen, J. Melrose, M. Brunvand, J. Bradshaw, J. A. Ledbetter, P. S. Linsley. 1993. Induction of alloantigen-specific hyporesponsiveness in human T lymphocytes by blocking interaction of CD28 with its natural ligand B7/BB1. J. Exp. Med. 177:165.[Abstract/Free Full Text]
  4. Boise, L. H., A. J. Minn, P. J. Noel, C. H. June, M. A. Accavitti, T. Lindstein, C. B. Thompson. 1995. CD28 costimulation can promote T cell survival by enhancing the expression of Bcl-xL. Immunity 3:87.[Medline]
  5. Lin, H., S. F. Bolling, P. S. Linsley, R. Q. Wei, D. Gordon, C. B. Thompson, L. A. Turka. 1993. Long term acceptance of major histocompatibility complex mismatched cardiac allografts induced by CTLA-4Ig plus donor-specific transfusion. J. Exp. Med. 178:1801.[Abstract/Free Full Text]
  6. Lenschow, D. J., Y. Zeng, J. R. Thistlethwaite, A. Montag, W. Brady, M. G. Gibson, J. A. Bluestone. 1992. Long term survival of xenogeneic pancreatic islets induced by CTLA-4Ig. Science 257:789.[Abstract/Free Full Text]
  7. Tam, R. C., U. T. Phan, T. Milovanovic, B. Pai, C. Lim, J. Bard, L. He. 1997. Oligonucleotide-mediated inhibition of CD28 expression induces human T cell hyporesponsiveness and manifests impaired contact hypersensitivity in mice. J. Immunol. 158:200.[Abstract]
  8. Bock, L. C., L. C. Griffin, J. A. Latham, E. H. Vermaas, J. J. Toole. 1992. Selection of single-stranded DNA molecules that bind and inhibit human thrombin. Nature 355:564.[Medline]
  9. Aruffo, A., B. Seed. 1987. Molecular cloning of a CD28 cDNA by a high-efficiency COS cell expression system. Proc. Natl. Acad. Sci. USA 84:8573.[Abstract/Free Full Text]
  10. Lee, K. P., C. Taylor, B. Petryniak, L. A. Turka, C. H. June, C. B. Thompson. 1990. The genomic organization of the CD28 gene. Implications for the regulation of CD28 mRNA expression and heterogeneity. J. Immunol. 145:344.[Abstract]
  11. Wang, G., P. J. Middleton, L. He, V. Stoisavljevic, W. E. Seifert. 1997. Synthesis and evaluation of oligodeoxynucleotides containing 3'-C-aminoethyl- and 3'-aminomethylthymidine. Nucleosides Nucleotides 16:445.
  12. Seed, B., J. Y. Sheen. 1988. A simple phase-extraction assay for chloramphenicol acetyltransferase activity. Gene 67:271.[Medline]
  13. Skerka, C., E. L. Decker, P. F. Zipfel. 1995. A regulatory element in the human interleukin 2 gene promoter is a binding site for the zinc finger proteins Sp1 and EGR-1. J. Biol. Chem. 270:22500.[Abstract/Free Full Text]
  14. Tam, R. C., S. Wu-Pong, B. Pai, C. Lim, A. Chan, D. F. Thomas, T. Milovanovic, J. Bard, P. Middleton. 1999. Increased potency of an aptameric G-rich oligonucleotide is associated with novel functional properties of phosphorothioate linkages. Antisense Nucleic Acid Drug Dev. 9:289.[Medline]
  15. Kramer, B., A. Meichle, G. Hensel, P. Charnay, M. Kronke. 1994. Characterization of an Krox-24/EGR-1-responsive element in the human tumor necrosis factor promoter. Biochim. Biophys. Acta 1219:413.[Medline]
  16. Minc, E., P. deCoppet, P. Masson, L. Thiery, S. Dutertre, M. Amor-Gueret, C. Jaulin. 1999. The human copper-zinc superoxide dismutase gene (SOD1) proximal promoter is regulated by Sp1, EGR-1, and WT1 via non-canonical binding sites. J. Biol. Chem. 274:503.[Abstract/Free Full Text]
  17. Lin, J. X., W. J. Leonard. 1997. The immediate-early gene product EGR-1 regulates the human interleukin-2 receptor ß-chain promoter through non-canonical Egr and Sp1 binding sites. Mol. Cell. Biol. 17:3714.[Abstract]
  18. Khachigian, L. M., K. R. Anderson, N. J. Halnon, M. A. Gimbrone, N. Resnick, T. Collins. 1997. EGR-1 is activated in endothelial cells exposed to fluid shear stress and interacts with a novel shear-stress-response element in the PDGF A-chain. Arterioscler. Thromb. Vasc. Biol. 17:2280.[Abstract/Free Full Text]
  19. Rafty, L. A., L. M. Khachigian. 1998. Zinc finger transcription factor mediate high constitutive platelet-derived growth factor-B expression in smooth muscle cells derived from aortae of newborn rats. J. Biol. Chem. 273:5758.[Abstract/Free Full Text]
  20. Riva, A., G. L. Wilson, J. H. Kehrl. 1997. In vivo footprinting and mutational analysis of the proximal CD19 promoter reveal important roles for an Sp1/EGR-1 binding and a novel site termed the PyG box. J. Immunol. 159:1284.[Abstract]
  21. Cowan, P. J., D. Tsang, C. M. Pedic, L. R. Abbott, T. A. Shinkel, A. J. d’Apice, M. J. Pearse. 1998. The human ICAM-2 promoter is endothelial cell-specific in vitro and contains Sp1 and GATA binding sites. J. Biol. Chem. 273:11737.[Abstract/Free Full Text]
  22. Majello, B., P. De Luca, G. Suske, L. Lania. 1995. Differential transcription regulation of c-myc promoter through the same DNA binding sites targeted by Sp1-like proteins. Oncogene 10:1841.[Medline]
  23. Maltzman, J. S., J. A. Carmen, J. G. Monroe. 1996. Transcriptional regulation of the Icam-1 gene in antigen receptor- and phorbol ester-stimulated B lymphocytes: role for transcription factor EGR1. J. Exp. Med. 183:1747.[Abstract/Free Full Text]
  24. Maltzman, J. S., J. A. Carmen, J. G. Monroe. 1996. The role of EGR1 in regulation of stimulus-dependent CD44 transcription in B lymphocytes. Mol. Cell. Biol. 16:2283.[Abstract]
  25. Dinkel, A., W. K. Aicher, C. Haas, P. F. Zipfel, H.-H. Peter, H. Eibel. 1997. Transcription factor EGR-1 activity down-regulates Fas and CD23 expression in B cells. J. Immunol. 159:2678.[Abstract]
  26. Kim, Y. H., M. A. Buchholz, F. J. Chrest, A. A. Nordin. 1994. Upregulation of c-myc induces the gene expression of the murine homologues of p34cdc2 and cyclin-dependent kinase-2 in T lymphocytes. J. Immunol. 52:4328.
  27. Yaswen, P., M. R. Stampfer, K. Ghosh, J. S. Cohen. 1993. Effects of sequence of thioated oligonucleotides on cultured human mammary epithelial cells. Antisense Res. Dev. 3:67.[Medline]
  28. Perez, J. R., Y. Li, C. A. Stein, S. Majumder, A. van Oorschot, R. Narayanan. 1994. Sequence-independent induction of Sp1 transcription factor activity by phosphorothioate oligodeoxy-nucleotides. Proc. Natl. Acad. Sci. USA 91:5957.[Abstract/Free Full Text]
  29. Maltese, J.-Y., H. W. Sharma, L. Vassilev, R. Narayanan. 1995. Sequence context of antisense RelA/NF{kappa}B phosphoro-thioates determines specificity. Nucleic Acids Res. 23:1146.[Abstract/Free Full Text]
  30. Bennett, C. F., M.-Y. Chiang, L. Wilson-Lingardo, J. R. Wyatt. 1994. Sequence specific inhibition of human type II phospholipase A2 enzyme activity by phosphorothioate oligonucleotides. Nucleic Acids Res. 22:3202.[Abstract/Free Full Text]
  31. Higgins, K., J. Perez, T. Coleman, K. Dorshkind, W. McNomas, U. Sarmiento, C. Rosen, R. Narayanan. 1993. Antisense inhibition of the p65 subunit of NK{kappa}B blocks tumorigenicity and causes tumor regression. Proc. Natl. Acad. Sci. USA 90:9901.[Abstract/Free Full Text]
  32. Blackburn, E. H.. 1990. Telomeres: Structure and synthesis. J. Biol. Chem. 265:5919.[Free Full Text]
  33. Shimizu, A., T. Honjo. 1984. Immunoglobulin class switching. Cell 36:801.[Medline]
  34. Morishita, R., G. H. Gibbons, M. Horiuchi, Y. Kaneda, T. Ogihara, V. J. Dzau. 1998. Role of AP-1 complex in angiotensin II-mediated transforming growth factor-ß expression and growth of smooth muscle cells: using decoy approach against AP-1 binding site. Biochem. Biophys. Res. Commun. 243:361.[Medline]
  35. Tomita, N., M. Horiuchi, S. Tomita, G. H. Gibbons, J. Y. Kim, D. Baran, V. J. Dzau. 1998. An oligonucleotide decoy for transcription factor E2F inhibits mesangial cell proliferation in vitro. Am. J. Physiol. 275:F278.
  36. Cho-Chung, Y. S.. 1998. CRE-palindrome oligonucleotide as a transcription factor decoy and an inhibitor of tumor growth. Antisense Nucleic Acid Drug Dev. 8:167.[Medline]
  37. Morishita, R., G. H. Gibbons, M. Horiuchi, K. E. Ellison, M. Nakajima, L. Zhang, Y. Kaneda, T. Ogihara, V. J. Dzau. 1995. A gene therapy strategy using a transcription factor decoy of the E2F-binding site inhibits smooth muscle proliferation in vivo. Proc. Natl. Acad. Sci. USA 92:5855.[Abstract/Free Full Text]
  38. Maeshima, Y., N. Kashihara, T. Yasuda, H. Sugiyama, T. Sekikawa, K. Okamoto, K. Kanao, Y. Watnabe, Y. S. Kanwar, H. Makino. 1998. Inhibition of mesangial cell proliferation by E2F decoy oligodeoxynucleotide in vitro and in vivo. J. Clin. Invest. 101:2589.[Medline]



This article has been cited by other articles:


Home page
 J. Immunol.Home page
C. J. Lin and R. C. Tam
Transcriptional Regulation of CD28 Expression by CD28GR, a Novel Promoter Element Located in Exon 1 of the CD28 Gene
J. Immunol., May 15, 2001; 166(10): 6134 - 6143.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tam, R. C.
Right arrow Articles by Stoisavljevic, V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tam, R. C.
Right arrow Articles by Stoisavljevic, V.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS