adjacent, cooperative elements form a strong, constitutive ... · adjacent, cooperative elements...

Adjacent, Cooperative Elements Form a Strong, Constitutive Enhancer in the Human Granulocyte-Macrophage Colony-Stimulating Factor Gene

By Stephen D. Nimer, Wei Zhang, Karen Kwan, Young Wang, and Jin Zhang

Both copies of a repeated sequence CAlT(A/T), located between bp -53 and -39 in the upstream region of the human GM-CSF gene, are required for mitogen-inducible promoter activity in T lymphocytes. However, the proteins that recognize this region of the granulocyte-macrophage colony-stimulating factor (GM-CSF) promoter, and are responsible for its transcriptional regulatory activity, have not been clearly identified. Using transient transfection assays, we demonstrate that a 19-bp oligonucleotide containing the CAlT(A/ TI repeats has strong constitutive enhancer activity in both T cell and non-T-cell lines, even though GM-CSF is not nor- mally constitutively expressed by these cells. A 12-bp oligonucleotide, containing only the sequence CATTAATCATIT, lacks enhancer activity indicating that the nucleotides sur- rounding these sequences are critical for this enhancer activity. The sequence TTTCCT, which can bind members of the

--

HE TRANSCRIPTIONAL regulation of lymphokine T gene expression involves shared and distinct DNA regulatory elements, such as recognition sites for the binding of activator protein-1 (AP- I), nuclear factor-KB (NF-KB), nuclear factor activated T cells (NFAT), and other transcription factors.'.' The tissue-specific and mitogen-inducible transcription of cytokine genes like granulocyte-macrophage colony-stimulating factor (GM-CSF) can be dictated by many mechanisms, including the presence of tissue specific transcription factors or the activation-dependent ability of these factors to bind regulatory DNA elements and activate or repress transcription. GM-CSF is expressed by activated, but not resting T cells, and plays an important role in stimulating hematopoiesis and modulating host The GM-CSF 5' flanking sequences contain a variety of transcriptional regulatory elements including an NF-KB motif,5 a CD28 response a GC-rich element,8 an AMLl binding site: and the repeated sequence CATT(A/T). Two copies of the CATT(A/T) motif are located between base pair (bp) -57 to -24, in a region that binds nuclear proteins in DNase I footprinting assays." These sequences are required for GM-CSF promoter activity in T cells," myeloid cells," and fibroblasts,I3 and mutations in either copy of the

From the hboratory of Molecular Hematopoiesis, Sloan-Ketter- ing Institute, New York, NY; the Division of Hematologic Oncology, the Department of Medicine, Memorial Sloan-Kettering Cancer Cen- ter, New York, NY; and UCLA School of Medicine, the Department of Medicine, L o s Angeles, CA.

Submitted October 10, 1995; accepted December 15, 1995. Supported by United States Public Health Service National Insti-

tutes of Health Grant No. RO1 DK43025, the DeWitt Wallace Foun- dation, and the Wally Yonamine Leukemia Research Fund.

Address reprint requests to Stephen D. Nimer, MD, Memorial Sloan- Kettering Cancer Center, 1275 York Ave, Box 575, New York, NY 10021.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. section 1734 solely to indicate this fact. 0 1996 by The American Society of Hematology. 0006-4971/96/8709-001.5$3.00/0

ets family of transcription factors, is located just 3' of these CAlT(A/T) repeats, and mutagenesis of the CCT sequence abolishes (1) the constitutive (and mitogen inducible) enhancer activity of the 19-bp GM-CSF sequences, (2) the responsiveness to transactivation by ets-1, and (3) the ability to specifically bind ets-1 and elf-I in electrophoretic mobility shift assays (EMSA). We demonstrate that although T cells contain nuclear proteins capable of independently recognizing the ets binding site and the CATT(A/T) repeats in EMSAs, both of these regulatory elements are required for enhancer function. The strong constitutive activity of this 19-bp region suggests that negative regulation of the GM-CSF promoter is critical for the restricted expression pattern of GM-CSF mRNA. 0 1996 by The American Society of Hematology.

CATT(A/T) motif eliminates basal and inducible promoter activity.".'2 Similar sequences are found in the 5' flanking sequences of the murine and human interleukin-5 (IL-5) genes,14 and they have recently been shown to regulate IL- 5 promoter activity."

A 26- to 30-bp region of the GM-CSF promoter, which contains the CATT(A/T) repeated sequences, has been shown to function as a mitogen responsive enhancer in T c e l l ~ ' ~ ~ ' ' and to allow mitogen-inducible transcription in vitro." However, we demonstrate that a 19-bp region containing these sequences (from bp -52 to -34, identified by our earlier DNase I footprinting studies) has strong constitutive enhancer activity in both T cell and non-T-cell lines, demonstrating that regulatory sequences outside of this region are also critical for controlling physiologic expression of GM-CSF. We show that the 12 bp CATTAATCAT'TT sequence, which has been reported to represent an AP-I binding site,'".'x is insufficient for constitutive enhancer activity in T cells and demonstrate that a TTCCT sequence located at the 3' end of the 19-bp region (which matches the consensus binding sequence for several members of the ets family of transcription factors) is required for full enhancer

Constitutively expression of GM-CSF has been observed in erythroleukemia cells that constitutively express the ets protein Fli-1 (due to insertional activation by the Friend spleen focus-forming virus SFFV).*' Our results demonstrate that the ets-1 protein can bind to this 19-bp region of the GM-CSF promoter and can upregulate GM-CSF promoter function in T cells.

~~

MATERIALS AND METHODS

Tissue culture and electroporations. Cell lines were maintained in Iscove's modified Dulbecco's medium supplemented with either 10% bovine serum (MLA 144) or 10% fetal bovine serum (GCT, HeLa and S-LB-1) and penicillin, streptomycin, and 1% glutamine. All cells were fed the day before electroporation. MLA 144 T cells were electroporated using a BioRad Gene F'ulser at 250 V and 960 pF. Five or 25 micrograms of plasmid DNA was introduced into 5 x 10' cells according to previously published techniques." For studies of mitogen-responsiveness, MLA 144 cells were either stimulated with phytohemagglutinin (PHA) 1 % and 12-0-tetradecoyl phorbol myristate acetate (TPA) 20 ng/mL following the transfection, or

3694 Blood, Vol 87, No 9 (May 11, 1996: pp 3694-3703

For personal use only.on August 30, 2017. by guest www.bloodjournal.orgFrom

http://www.bloodjournal.org/

http://www.bloodjournal.org/site/subscriptions/ToS.xhtml

CONSTITUTIVE GM-CSF ENHANCER ELEMENT 3095

m - C S F Oliaonucleotlde S

23 bp wild type

23 bp mutant

19 bp wild type

19 bp mutant#l

19 bp mutant#?

19 bp mutant #3

19 bp mutant #4

12 bp wlld type

TGGTCACCATTAATC ATTTCCTC

TGGTCACAAGGGATAGAAGCTTC

TCACCATTAATCATTTCCT

TCACAAGGGATAGAAGCD

TCAWTTAATCATTTCCT

TCACC ATTAATCATTT-

T C A C C A T T A A T w n C C T

CATTAATC ATTT

Fig 1. Oligonucleotides used for the analysis of the enhancer activity of the OM-CSF sequences and for EMSA. are shown. A 46- bp oligonucleotide, containing two copies of the 23-bp wild type or mutant sequences, was used for the experiments in Fig 2.

maintained unstimulated for 16 to 20 hours. GCT and HeLa cells were efficiently transfected using a CaC12 precipitation procedurez4 and incubated at 37°C overnight. All cells were then harvested for CAT activity using standard technique^.^^

A total of 25 pg of plasmid DNA was also introduced into 5 X lo6 cells by electroporation for the cotransfection experiments; 12.5 pg of a target GM-CSF CAT reporter plasmid was cotransfected with either 12.5 pg of an expression vector construct BC1226 containing the human ets-1 cDNA, elf-1 cDNA or vector alone. Eighteen to 24 hours later the cells were harvested and CAT activity deter- mined. (The human ets-1 and elf-1 cDNAs were kindly provided by C. Thompson [University of Chicago, Chicago, E] and cloned into the Hind 111 site of the BC12 vector).

The oligonucleotides used for creating the GM-CSF-Tk promoter- CAT reporter gene plasmids are shown in Fig 1. The pTE2 plasmid contains a multiple cloning site just 5' of the herpes simplex virus thymidine kinase (Tk) promoter. Complementary single stranded oligonucleotides were synthesized with BamHI ends, annealed, ki- nased, and ligated into the Bgl I1 site in the multiple cloning site of the pTE2 plasmid. The orientation of the oligonucleotides and the number of copies were confirmed using a dideoxy DNA sequencing method (Sequenase 11; US Biochemical, Cleveland, OH). Preparation of the -53 plasmid, which contains the -53 to +37 region of the GM-CSF gene upstream of the CAT gene, has been previously reported." Multiple plasmid preparations of each construct were tested in each cell type to eliminate variability due to differences in plasmid preparations.

Nu- clear extracts were prepared from unstimulated and PHA/TPA stimulated MLA 144 cells and from PHA stimulated primary human T cells isolated from peripheral blood of normal volunteer donors,26

Nuclear extracts and electrophoretic mobility shifi assays.

using the modified Dignam method?' In vitro translated ets-1 and elf-1 were prepared using an ets-1 or elf-1 cDNA in the CMV cDNA I neo vector (provided by C. Thompson) linearized at the Nsil I and Xba I restriction sites, respectively. In vitro transcription and translation (using rabbit reticulocytes) was performed according to the manufacturer's instructions (Promega, Madison, WI). One strand of the synthetic oligonucleotides containing GM-CSF promoter sequences (shown in Fig 1) was labeled using polynucleotide kinase and '*P -?-ATP, and then annealed to its complementary oligonucleotide. Next, 2.0 X lo5 cpm of probe was incubated with either 2 or 4 pg of nuclear extract or 3 to 6 pL of in vitro transcribed and translated recombinant protein for 30 minutes at room temperature in 30 pL of buffer containing 2 pg poly dl:dC, 20 mmol/L HEPES pH 7.9, 10% glycerol, 150 mmol/L NaCI, 20 mmol/L MgClz, 0.5 mmol/L EDTA and 1 mmol/L DIT, in the presence or absence of unlabeled competitor oligonucleotides. The IL-3 promoter oligonucleotide, used as cold competitor, contains the sequence 5' CTC- AGTGAGCTGAGTCAGGCTTCCCCTTCCTGCCACAGG 3' (which extends from bp -312 to -273 relative to the IL-3 cap site). [The AP-1 site (-302 to -296) and the ets binding site (- -285 to -281)28 are indicated in bold]. Following incubation, the samples were loaded onto a 5% polyacrylamide gel and run in 0.5X TBE at room temperature. Gels were dried and exposed to XAR Kodak film (Eastman Kodak, Rochester, NY) ovemight, using an enhancing screen at -70°C.

RESULTS

GM-CSF sequences from bp -57 to -35 have enhancer activity in multiple cell types. A 23-bp region of the GM- CSF gene, containing the sequence TGGTCACCAnA- ATCAIITTCCTC, was previously shown to bind nuclear proteins present constitutively in MLA 144 T cells in DNase I footprinting experiments." To evaluate the regulatory activity of these sequences a 46-bp oligonucleotide containing two copies of this 23-bp sequence was synthesized and sub- cloned upstream of the Tk promoter for the initial experiments. As shown in Fig 2A, these sequences demonstrated strong constitutive enhancer activity in MLA 144 cells in either the forward or reverse orientation, generating a 100 to 300-fold greater CAT activity than the original pTE2 plasmid. The CAT activity generated by these plasmids in mitogen stimulated T cells was also 300- to 400-fold higher than the pTE2 plasmid. Multiple copies of the GM-CSF oligonucleotide generated similar levels of CAT activity to that generated by a single 46-bp oligonucleotide.

The CAT activity generated by 25 pg of these reporter plasmids in resting and mitogen stimulated cells was ex- tremely high, so to assess the mitogen inducibility of these constructs, we transfected only 5 pg of plasmid into 5 X lo6 cells, which generated less CAT activity in unstimulated MLA 144 cells (ranging from 4% to 20% acetylation, de- pending on the construct) (data not shown). The construct containing a single copy of the 46-bp wild type oligonucleotide demonstrated 12.0-fold or 4.6-fold mitogen inducibility (see numbers in parentheses in Fig 2A), and a construct containing multiple copies of this oligonucleotide was also mitogen inducible (10.8-fold). The original reporter plasmid, pTE2, was not mitogen inducible at either concentration of DNA.

To evaluate whether the repeated CA ' IT(m) sequences were required for the enhancer activity, a mutant oligonucle-




3696

(TGGTCACc*TTI*rc*mCCTC) X 2

NlMER ET AL

30.2 23.9 79.3 t0.5 2.61

- € ! u t ! A. ep, I, '1,. UNIlTlYULlTEDSTUULlTED-

plE2 CAT %NE 0.2 t 0.1 0.2 f 0.1 1.01 I

1 REVERSE COPY

2 FORWARD COPIES

19.7i19.6 61.7t25.5 3.11 (4 6x)

45.5221.7 84.7t11.2 191

3 FORWARD COPIES

I(TGGTCAC.AgggAT.g..QCtTC) X 21

8.

PTEZ

r 1 REVERSECOPY 1

2 FORWARD COPIES

YUU TYPE 0-

I 2 FORWARDCOPIES I

3 FORWARD COPIES

I 1 FORWARD COPY I

1 REVERSE COPY

60.2 f 3.6 65.0 f 8.4 1.1x I10 8x)

1.3 f 0.3 1.0 f 0.4 0 . 8 ~

1.0 r 0.5 0.9 t0 .8 0.91

2.1t0.6 1.221.0 0.61

XACETYLAM)N

tlelaylls !%LELLs

0.4 0.4

63.4 2.6

36.5 3.3

otide that contained multiple nucleotide substitutions in both of the CATT(AjT) repeats was tested (bottom of Fig 2A). These constructs had no mitogen inducible CAT activity, regardless of whether the mutant oligonucleotide was in the forward or reverse orientation, or in multiple copies. The basal CAT activity of these constructs was greater than that of the pTE2 construct, but this phenomena has been observed in T cells on the introduction of a variety of mutated sequences into this vector.29

We also examined the constitutive promoter activity of these constructs in HeLa cells and GCT cells, which are two nonhematopoietic cell lines. GCT cells have been shown to constitutively express GM-CSF,30 whereas HeLa cells do not (data not shown). Interestingly, the wild type (23-bp X 2) GM-CSF sequences generated - 150-fold greater CAT activity than the enhancerless pTE2 plasmid in HeLa cells, and sixfold to eightfold greater activity in GCT cells (Fig 2B). This activity was dependent on the presence of the CATT(A/ T) repeats; the activity of the constructs containing the mutant oligonucleotide was the same as the enhancerless pTE2 plasmid in these cell lines.

Ident$cation of minimal sequences required for enhancer activity. A smaller footprint, 19-bp in length and extending from bp -52 to -34, was previously observed when mitogen stimulated, rather than unstimulated, MLA 144 nuclear extracts were used for DNase I footprinting experiments." Based on this finding, we examined whether the GM-CSF

0.5 0.4

1 .o 0.4

Fig 2. Enhancer activity of e direct repeat of the 23-bp region of OM-CSF 5' flanking sequences in T cell end non-T-cell lines. (A) The parcont acetylation generated by 25 pg of the wild type plasmids, or by mutant plasmids that lack the CAlT(A/Tl repeats, in resting or PHA/TPA stimulated MLA 144 T cells is shown in bold 5 standard deviations, as is the fold induction (N = 3). The fold induction using 5 pg of the GM-CSF oligonucleotide-Tk promoter CAT constructs is shown in parentheses (N = 3). The fold induction = % acetyl (stiml/% acetyl (unstim). The number of copies of the 46-bp oligonucleotide (that contains a direct repeat of the sequences shown) Is indieated. The range of CAT activity using 25 pg of these plasmids was pTE2 (0.2% to 0.3% unrtim, 0.2% to 0.2% stiml; wild type 1 forwerd copy 19.2% to 38.5% unstim, 74.6% to 88.8% dim), two forward copies (28.5% to 69.3% undm, 74.3%to 90.1% stim); mutant one forward copy (1.0% to 1.8% unstim, 0.6% to 1.3% stiml. IBI Mean percent acetytdon generated in undmulatad nonhamatopoMk cell lines by 25 pg of wild type or mutant constructs (N = 2).

sequences in this smaller region were sufficient for enhancer function. Two or four copies of the 19-bp sequence TCA- CCATTAATCATCCT contained both constitutive and mitogen-inducible enhancer activity in MLA 144 cells, that was similar to the 23-bp GM-CSF sequence (data not shown). Because many recognition sequences for DNA binding proteins consist of palindromes or repeated sequences, we also examined whether the 12-bp region, CATTAATCA- TIT, would contain sufficient sequence information to serve as an enhancer element. Although these sequences have been reported to bind AP- 1 ,I6." pTE2 constructs containing from one to six copies of the 12-bp GM-CSF sequences had no significant basal or mitogen inducible enhancer activity (data not shown).

These results suggested that the sequences flanking the CATT(A/T) repeats were critical for enhancer function, so we examined the enhancer activity of additional 19-bp oligonucleotides (diagrammed in Fig 1) that contain three bp substitutions at either the 3' end (mutant #3, disrupting the TTCCT sequence that represents a potential ets-binding site) or near the 5' end (mutant #2, disrupting the first CATT(A/T repeat). The activity of constructs containing two copies of a mutant oligonucleotide was compared with constructs containing two copies of the wild type sequence (see Fig 3). In this set of experiments, the activity of the wild type 19-bp oligonucleotide was 40 to 110-fold higher than the pTE2 plasmid. A 20-fold lower basal activity, and a




CONSTITUTIVE GM-CSF ENHANCER ELEMENT

1 6 0 0

E 1 4 0 0

1 2 0 0

1 0 0 0 *

3 800 5 600

400

2 0 0

0 0

3697

.. -. -. .. .. .. .. 3 2 % 21 I

2 7 t 1 2 4 7 i 2 4 2 33. 1 8 5 1: 3 5 ..

1 0 . 1 6 t 08 T . I

I800 T 171 62 132 1

CONSTRUCT

Fig 3. Effect of mutations in the 19-bp GM-CSF sequences on enhancer activity. The CAT activity genarated by each of the GM- CSF oligonucleotide Tk-CAT plasmids in resting or PHA/TPA stimu- latad M U 144 cells is reported (2 standard deviations), relative to the CAT activity of the enhancerless pTE2 plasmid (which was as- signed the value of 1.0) (N = 3). For each construct, the first bar (labalad with the nama of the construct) reprerants the CAT activity in unstimulated cells, whereas the second bar (labeled "stim") represents the CAT activity in mitogen stimulated cells. The range of percent acetylation in these experiments wes: for the wild type (4.0% to 82.1% unstim, 67.2% to 86.6% stiml, for mutant #2 (3.4% to 46.4% unstim, 8.7% to 64.1% stim) and for mutant W3 (0.4% to 4.4% unstim, 0.8% to 14.2% stim).

>30-fold decrease in mitogen responsive enhancer activity, was seen when the potential ets recognition sequences were disrupted (Fig 3, mutant #3). The activity of mutant #2 was sixfold to sevenfold higher than mutant #3, but it had only 30% of the wild type activity in unstimulated T cells, and <20% of the wild type activity in stimulated T cells. Mutant constructs #1 and #3 gave slightly higher basal activity than the enhancerless pTE2 plasmid.

To examine whether members of the ets family could transactivate the GM-CSF promoter, cotransfection experiments were performed introducing the human ets-1 cDNA expression vector and the -53 GM-CSF promoter CAT construct into resting MLA 1 4 4 cells. The -53 GM-CSF promoter CAT construct was transactivated 5.2-fold by the introduction of ets-1 (Fig 4) but was not transactivated by elf-1 (data not shown). Because elf-1 has not been shown to transactivate any of the promoter elements that it binds to in vitro, no positive control could be used for these experiments, however, this elf-1 cDNA was used successfully for in vitro transcription experiments (see below).

To evaluate the effects of ets proteins on the GM-CSF promoter, we examined the ability of ets-1 to transactivate the wild type or mutant 19-bp GM-CSF sequences, using the Tk promoter CAT constructs (Fig 4). The 19-bp wild type GM-CSF enhancer Tk CAT plasmid was transactivated 7.7-fold by the introduction of ets-1. Mutations in the TTCCT region (mutant #3) completely abolished ets-responsiveness, whereas mutation in the 5' CAl'T(A/T) repeat (mutant #2) only slightly diminished responsiveness to ets-1 .

Characterization of DNA-protein interactions in the GM- CSF enhancer element. To characterize the binding of T- cell nuclear proteins to the GM-CSF sequences extending from bp -52 to -34, we performed electrophoretic mobility

Ets cotransfection experiments.

shift assays (EMSAs) using MLA 144 T cell nuclear extracts and radiolabeled GM-CSF wild type or mutant 19-bp oligonucleotide probes. Two major specific DNA-protein complexes were observed (Fig 5A); both gel shifted bands were completely eliminated by a 100-molar excess of cold 19- bp wild type competitor oligonucleotide (lane l), whereas addition of a 100-fold excess of mutant #3 oligonucleotide competed for binding to the upper band, but not the lower specific gel retarded band (lane 3). Addition of excess IL-3 promoter oligonucleotide as cold competitor, which binds AP-1 and ets proteins such as elf-1:' competed for the lower, but not the upper, band (lane 2). This suggests that the upper band represents binding of CATT(A/T) repeat binding proteins (other than AP-l), and the lower band represents the binding of an ets family member. The 12-bp wild type oligonucleotide, which lacks the ets consensus binding sequences, competed completely for the upper band and slightly for the lower band (data not shown), providing additional evidence that the upper band represents the binding of CATT(A/T) repeat binding proteins. The mutant #1 oligonucleotide did not compete for any of the gel shift complexes (lane 4). No differences in the gel shift patterns were seen using extracts from unstimulated or P H m A stimulated MLA 144 T cell nuclear extracts (data not shown).

To identify the proteins present in T-cell nuclear extracts that recognize the 19-bp GM-CSF sequences, we performed supershift experiments using an anti-ets- 1 polyclonal antiserum (kindly provided by T. Papas, Medical University of South Carolina, Charleston) and an anti-elf-1 polyclonal antiserum (kindly provided by J. Leiden, University of Chi- cago, Chicago, IL) (see Fig 5B). The position of the gel

-53 CATT(NT) REPEAT EIs R.E +3?

CATTbATCATTT CATGENE 5.211

PTEZ 1.1 t1 .2

Want #1 1.2 f 0.0

Mutant ~3 0.9 *0.6

r -

0 2 4 6 8

FOLD STIMULATION BY ETS-1

Fig 4. Ets-1 cotransfection studies. The results of introducing 12.5 p g of BCl2 or ets-1 BC12 plasmid with 12.5 p g of the indicated reportar plasmid is shown (N = 4). The fold stimulation = % acetylation in presence of ets-1 (et$-1 BC12)/% acetylation in absence of ets- 1 (BC12 vector alone) 12 standard deviations). The -53 construct illustrated at the top contains GM-CSF promoter sequences linked to the CAT gene; the fold stimulation by ets-1 is shown. All constructs indicated in the graph contain two copies of the indicated oligonucleotide and the HSV Tk promoter linked to the CAT gene.




3698 NlMER ET AL

1 2 3 4 5 1 2 3 4 5 6 7 8 9

Fig 5. (A) EMSA using radiolabeled 19-bp wild type GM-CSF oligonucleotide as probe and 4 p g of mitogen stimulated MLA 144 cell nuclear extract. Lane 5 contains probe and nuclear extract. Lane 1 contains a 100-fold excess of wild type competitor, lane 2 contains a 100-fold excess of an oligonucleotide containing the ets and AP-1 binding sites in the IL-3 promoter, lane 3 contains a 100-fold excess of mutant #3 oligonucleotide as competitor, and lane 4 contains a 100-fold excess of mutant #1 oligonucleotide. The arrows mark the position of the two specific gel shifted bands; the free 19-bp wild type oligonucleotide probe is shown at the bottom of the autoradiograph. (BI Electrophoretic mobility shift and supershift assays using radiolabeled 19-bp wild type GM-CSF oligonucleotide as probe (present in all lanes) and either in vitro translated elf-1 or T cell nuclear extracts. Lane 1 contains probe and reticulocyte lysate only. Lanes 4 and 5 contain in vitro translated elf-1, either alone (lane 4) or with 1 p L of anti-elf-1 antisera (lane 5) . The arrows point to the position of the specific elf-1 bands and the asterisk shows the position of the supershifted band. Radiolabeled 19-bp probe was incubated with nuclear extracts from PHA-stimulated primary human T cells, or MIA 144 T cells, either alone (lanes 8 and 9) or with 1 p L of the anti-elf-1 antisera (lanes 2 and 3). The supershifted gel shift complexes have the same mobility as the supershifted band seen using in vitro translated elf-1 (lane 5). The specificity of the supershifted band was demonstrated using 1 p L of irrelevant antiserum (lanes 6 and 7).

retarded complex generated by in vitro translated elf-I is the same as the lower (ets) band (indicated by the arrow) generated by MLA 144 or primary human T-cell nuclear extracts (Fig 5B, compare lane 4 with lanes 6 and 7). The additional band seen in lane 4 is also seen in lanes 1 and 5, and represents binding of proteins contained in the reticulocyte lysate. Addition of anti-elf-I antiserum to either MLA 144 T cell nuclear extracts, or primary human T cell nuclear extracts, plus the 19-bp GM-CSF wild type probe generates a slower, supershifted band (indicated by the asterisk in lanes 2 and

3 of Fig 5B). The position of this supershifted band is the same as that generated by in vitro translated elf-I plus anti- elf-I antiserum (which is only weakly seen on this exposure, in lane 5).

Similar experiments using the anti-ets-1 antiserum showed no supershift using MLA 144 T cell nuclear extract and only a slight decrease in the intensity of the lower gel shift band using PHA stimulated primary T cell nuclear extract (data not shown). This antiserum generated only a weak supershift using an oligonucleotide containing the ets- 1 bind-





ing site from the proximal lck promote? (data not shown). Northern blot analysis of MLA 144, PHA-stimulated primary human T cell, or Jurkat T cell total cellular RNA demonstrated the presence of elf- 1 mRNA in all three T-cell lines. Ets-1 mRNA was found in primary T cells and Jurkat T cells, but not in MLA 144 cells (data not shown), consis- tent with the supershift data. The level of ets-1 and elf-1 mFWA did not vary following mitogen stimulation of MLA 144 or Jurkat T cells.

Given the ability of ets-1 to transactivate the GM-CSF promoter in MLA 144 T cells, we performed EMSAs to examine the ability of ets-1 and elf-1 to recognize sequences in the 19-bp wild type GM-CSF oligonucleotide (using in vitro transcribed and translated [IVT] ets-1 and elf-1, and either wild type or mutant GM-CSF oligonucleotides) (Fig 6A, B, and C). A single band was seen using the wild type GM-CSF oligonucleotide and IVT ets-1 (Fig 6A, lane 2), whereas two bands were seen using IVT elf-1 (lane 3). The reason why two bands are generated by elf-1 is unknown, although this is also seen with the CD4 enhancer3* and the IL-2 enhancer.33 The specific binding of ets-1 is shown in lanes 1 to 4, whereas lanes 5 to 8 demonstrate the specific binding of elf-I. A 100-fold excess of unlabeled wild type oligonucleotide competed completely for binding (lanes 2 and 6), whereas neither excess mutant #1 oligonucleotide (lanes 3 and 7), nor mutant #3 oligonucleotide (which lacks the ets consensus binding sequence) (lanes 4 and 8) competed for protein binding. The proteins in the rabbit reticulocyte lysate used to make both IVT ets-1 and elf-1 do not bind to the 19-bp oligonucleotide (lane 9).

Incubation of IVT ets-1 and elf-1 with the mutant #2 oligonucleotide generated the same pattern of protein binding as the wild type oligonucleotide; a single band was seen using IVT ets-1 (Fig 6B, lane l), and two bands were seen using IVT elf-I (lane 3). These bands represent specific binding, because no bands were seen using the rabbit reticulocytes lysate (r.r. lysate) alone (lane 5) and all bands were eliminated by a 100-fold excess of cold self-competitor (lanes 2 and 4).

Neither ets-1 nor elf-1 binds to the mutant #3 or mutant #4 oligonucleotide. The r. r. lysate itself generated a single specific band using the mutant #4 oligonucleotide as probe (Fig 6C), but no additional bands were generated by either IVT elf-1 or ets-1 protein. The identical pattern was seen using the mutant #3 oligonucleotide as probe (data not shown).

Detection of DNA-protein interactions using GM-CSF mutant oligonucleotides. We examined the binding of T- cell nuclear proteins to several 19-bp mutant oligonucleotides using EMSAs, to determine whether the CATT-1 and CATT-2 elements bind similar or distinct proteins and to define the sequence requirements for the binding of ets.

The mutant #3 oligonucleotide, which lacks the ets binding sequence but contains sequences that could bind YY1 or CATT(A/T) binding protein(s), generated two specific gel shift bands (indicated by the arrows in Fig 7). A 100-fold excess of mutant #1 oligonucleotide (which lacks the CATT(A/T) motifs) does not compete, where the mutant #2 oligonucleotide, which contains the CATT-1 element in

common with mutant #3 but lacks the CATT-2 element (which is thought to be the YY1 binding sequence'') com- petes as effectively for binding as the mutant #3 oligonucleotide itself. These results demonstrate that the CA'IT(A/T) binding protein(s) can bind DNA in the absence of the ets binding site and suggest that similar proteins can recognize the CATT-2 and CATT-1 motifs. Using the mutant #2 oligonucleotide as radiolabeled probe and various wild type and mutant competitors in EMSAs provided further evidence that the upper band represents binding of the CATT(A/T) repeat proteins and the lower band the binding of ets family members (data not shown). The CA'IT-2 sequences (and, therefore, the YY1 binding sequences) were not required for ets binding.

DISCUSSION

Lymphokines such as GM-CSF, IL-2, IL-3, or IL-4 are expressed by T cells following an activation process,' and detailed studies of the promoters for these genes have identified negative regulatory regions upstream of the mRNA initiation site for each of them.'0,26,34-37 3 ' AT-rich sequences at the 3' end of lymphokine "As (the mRNA destabilizing sequences3*) likely also prevent their constitutive expression, as do regulatory sequences located several kb upstream or downstream of lymphokine transcriptional initiation sites.39 The 626-bp of GM-CSF 5' flanking sequences have mitogen- inducible, but not basal promoter activity in T cells, whereas constructs containing only 53-bp of upstream sequences have strong constitutive activity in T cells," endothelial cells,35 fibroblasts,13 and several myeloid leukemia cell lines." This suggests that upstream sequences inhibit GM-CSF expression in the basal state and demonstrates that sequences 3' of bp -53 contain strong transcriptional regulatory activity.

In contrast to others, who have used larger regions of GM-CSF promoter sequences to test enhancer function and identified only mitogen-inducible enhancer activity, we demonstrate that the 19-bp GM-CSF sequences extending from bp -47 to -29 (TCACCATTAATCATTTCCT), can act as a strong constitutive enhancer in T cells and in cell lines that do not express GM-CSF. These 19 nucleotides are likely responsible for the strong constitutive promoter activity of the -53 construct.

Electrophoretic mobility shift assays have shown that proteins capable of recognizing these sequences are ubiquitously expressed,"' which given the limited cellular sources of GM- CSF, also suggests that cell-specific negative acting factors are critical to the appropriate regulation of GM-CSF expression. Some of these sequences may be located immediately 5' of this 19-bp region (and possibly overlapping this re- g i ~ n ) . ' ~ , ' ~ The YY1 protein has been shown to negatively regulate GM-CSF promoter function and bind to sequences upstream of the CATT(A/T) repeats, partially overlapping the CATT-2 repeat.l7 Other proteins also have inhibitory effects on the promoter activity of these sequence^'^.'^,^' and may also contribute to the lack of constitutive enhancer activity of a larger region of the GM-CSF pr~moter . '~

The sequence requirements for enhancer function are slightly different than those required for GM-CSF promoter activity. Previous studies have demonstrated that the CATT-




3700 NlMER ET AL

I I I I

"RI Free mut#2 probe

1$;+- probe

9 8 7 6 5 4 3 2 1 1 2 3 4 5

*.""I

1 2 3 4 5

Fig 6. EMSA using rabbit reticulocyte lysate, in vitro translated ets-1 or in vitro translated elf-1 and radiolabeled 19-bp wild type GM-CSF oligonucleotide (A), mutant #2 GM-CSF oligonucleotide (E), or mutant #4 GM-CSF oligonucleotide as probe (C). In (A) lane 9 contains rabbit reticulocyte lysate, whereas lanes 1 to 4 contain IVT ets-1 and lanes 5 to 8 contain IVT elf-1. The competitor oligonucleotide present in each lane is shown at the top. Lanes 1 to 8 contain either no competitor (lanes 1 and 5) or a 500-fold excess of either unlabeled wild type oligonucleotide (lanes 2 and 61, mutant #1 (lanes 3 and 7). or mutant #3 oligonucleotide (lanes 4 and 8). The arrows indicate the specific bands generated by the binding of IW protein to the radiolabeled probes. In (B), lanes 1 and 2 contain ets-1, lanes 3 and 4 contain elf-1, and lane 5 contains rabbit reticulocyte lysate. Lanes 1 and 3 contain no competitor, and lanes 2 and 4 contain a 500-fold molar excess of unlabeled mutant #2 oligonucleotide. The incubation mixtures in each lane of (C) are identical to (E). No specific bands are generated by IVT ets-1 or elf-1, therefore, no arrows are shown.





these mutations left the ets site and a single CATT(A/T) motif intact. The mutant #2 oligonucleotide competed for both gel shift complexes generated by the wild type 19- bp oligonucleotide, demonstrating that a single CATT(A/T) repeat and an intact ets binding site can bind nuclear proteins - cv c3 E as well as the wild type sequences, even though they do not contain full regulatory activity. c: c : & 0 The 12-bp region, containing only the CATT(A/T) re- I3 peats, is not sufficient for constitutive or mitogen- inducible

E E E enhancer activity, demonstrating the importance of the sequences flanking this central element. Mutagenesis of the 19-bp region of GM-CSF sequences demonstrated that the CCT sequences, located at the 3' end of the stimulated footprint region and the 19-bp oligonucleotides, are required for mitogen responsiveness (see also Wang et aI4'). These 3' sequences are part of a consensus sequence for the binding of members of the ets family of transcription factors (AG- GAA),'"*' and at least four members of the ets family have been shown to be expressed in T cells (ets-I, ets-2, elf-I,

Binding of in vitro translated elf-I to these GM-CSF sequences has been recently demonstrated? and we show that elf- I protein, present in MLA 144 and primary T-cell nuclear extracts, specifically recognizes the sequences at the 3' end of the 19-bp GM-CSF oligonucleotide. In contrast to Wang

+ 8 , +-, a> a .-

* 0

and Mi- I).xx.""6

free + mut. #3

probe Fig 7. Results of EMSA using radiolabeled 19-bp mutant #3 oligo-

nucleotide as probe and 2 p g of stimulated MIA 144 cell nuclear extract with either no competitor (lanes 1 and 5) or a 100-fold molar excess of unlabeled mutant #1 llane 2), mutant #2 (lane 3). or mutant #3 oligonucleotide (lane 4). The two specific gel retarded complexes are indicated by arrows, and the free probe is indicated at the bottom of the gel.

2 and CATT-I motifs, located between bp -47 and bp -29, are critical for mitogen-inducible promoter activity." Al- though mutation or deletion of the C A T - 2 repeat in the context of the GM-CSF promoter completely eliminated all promoter function," the 19-bp oligonucleotide that lacks the CATT-2 motif (mutant #2) has some constitutive and mitogen inducible enhancer activity on the Tk promoter. Pre- viously, we and Miyatake et aI4* demonstrated that the ets binding site (TTCCT) and one intact CATT(A/T) motif are not sufficient for mitogen inducible promoter activity; mutation of the CATT-2 motif (the 5' CATT(A/T) repeat) eliminates all promoter activity," as do mutations in the CATT-I motif (the 3' CATT(A/T) repeat). In each case,

- et al, we have demonstrated that IVT ets- 1 can also recognize these sequences in EMSAs and that the introduction of ets- 1 into resting T cells stimulates transcription via the TTCCT sequences contained in these 19-bp of GM-CSF sequence. Identical sequences in the HTLV-I LTR have been shown to bind ets-l and be transactivated by the introduction of an ets-l expression pla~mid.4~ Although the introduction of elf- 1 by cotransfection did not activate the GM-CSF sequences, cotransfection of elf-I with reporter gene plasmids containing elf-1 binding sequences from several different promoters does not change their reporter gene activity. Elf-I levels do not appear to vary following T-cell activation," but phosphorylation of Rb has been shown to dissociate elf- I from Rb;' which may regulate its DNA binding activity.

Cooperativity between members of the ets family and AP- 1, NF-EMS, serum response factor, and core binding factor have been demonstrated.4X-s' Our studies demonstrate that cooperativity between members of the ets family and the CATT(A/T) binding factors is required for the function of the GM-CSF 19-bp enhancer element, although independent in vitro binding of CATT(A/T) binding proteins and ets proteins to DNA can occur. Binding of AP-I to sequences within this 19-bp region has been recently reported,".Ix but our EMSA studies and those of Ye et all7 demonstrate binding of other nuclear proteins to these GM-CSF sequences. Our EMSA results also suggest that similar proteins recognize each CATT(A/T) repeat.

The inability of the IL-3 oligonucleotide, which contains a consensus AP-I site (and an elf-I binding site), to compete for the uppermost band argues against these proteins being predominantly or exclusively AP-I, as does the ability of the mutant #4 oligonucleotide, (which lacks sequences required for AP-I binding) to efficiently compete away this upper gel shift complex. Our data also argue against a pri-




3702 NlMER ET AL

mary regulatory role for AP- 1 binding proteins in controlling the activity of this region of the GM-CSF promoter. The 12- bp GM-CSF oligonucleotide, which contains the putative AP-I binding sequence, is not sufficient to confer mitogen- responsiveness to the Tk promoter in T cells, in contrast to a classic TPA response element (TRE)52 (and our data not shown). Also, prior experiments performed by us and others, showed that the introduction of mutations into the CATT-2 motif, that leave the ets binding site and the putative AP-1 site in the native GM-CSF promoter intact, eliminates GM- CSF promoter activity.".42

It has been proposed that NFAT (nuclear factor-activated T cells) can recognize GM-CSF sequences in this region, based on the ability of an IL-2 NFAT oligonucleotide to compete for binding to the GM-CSF sequences.'* However, an anti-NFATp antibody, kindly provided by Dr A. Rao (Dana-Farber Cancer Institute, Boston, MA), produces a clear supershift with the IL-2 NFAT oligonucleotide, but minimal, if any, supershift using the GM-CSF 19-bp oligonucleotide and 2 to 4 pg of MLA 144 T cell nuclear extract (data not shown). Thus, NFATp does not appear to be a major component of the DNA binding activity described in this study.

The differences between the in vitro DNA-protein interactions involving these GM-CSF sequences and their functional activity in transcription assays suggests that a complex interplay of regulatory factors controls the expression of GM-CSF. Sequences located just upstream of this enhancer element and sequences located several kb upstream of the GM-CSF gene39 have been shown to interact with this 19- bp region of the GM-CSF promoter. A link between the constitutive expression of GM-CSF (and IL-3) and the constitutive expression of the Hi-1 ets protein in Friend SFFV- induced erythroleukemia was suggested by Shimada et a1." We demonstrate that ets family members (le, ets-I) can transactivate the GM-CSF promoter.

ACKNOWLEDGMENT

We thank Roy Lau for technical assistance, Roger Pearse and Michael Sheffrey for their helpful suggestions, and Darlene Shannon for manuscript preparation.

REFERENCES

I . Crabtree GR: Contingent genetic regulatory events in T lym- phocyte activation. Science 243:355, 1989

2. Mitchell PJ, Tjian R: Transcriptional regulation in mammalian cells by sequence-specific DNA binding proteins. Science 245:371, 1989

3. Clark SC, Kamen R: The human hematopoietic colony-stimulating factors. Science 236: 1229, 1987

4. Gasson JC: Molecular physiology of granulocyte-macrophage colony-stimulating factor. Blood 77:1131, 1991

5. James R, Kazenwadel J: T-cell nuclei contain a protein that binds upstream of the murine granulocyte-macrophage colony-stim- dating factor gene. Proc Natl Acad Sci USA 86:7392, 1989

6. Stanley E, Metcalf D, Sobieszczuk P, Gough NM, Dunn AR: The structure and expression of the murine gene encoding granulocyte-macrophage colony stimulating factor: Evidence for utilisation of alternative promoters. EMBO J 4:2569, 1985

7. Fraser JD, Weiss A: Regulation of T-cell lymphokine gene

transcription by the accessory molecule CD28. Mol Cell Biol 12:4357, 1992

8. Miyatake S , Seiki M, Yoshida M, Arai K-I: T cell activation signals and human T-cell leukemia virus type I-encoded p 40" protein activate the mouse granulocyte-macrophage colony-stimulating factor gene through a common DNA element. Mol Cell Biol 8:5S81, 1988

9. Frank R, Zhang J, Uchida H, Hiebert S, Meyers S , Nimer S: A dominant negative AMLllETO fusion protein blocks the transactivation of the GM-CSF promoter by AMLl . Oncogene I I :2667, 1995

10. Nimer SD, Morita EA, Martis MJ, Wachsman W, Gasson JC: Characterization of the human granulocyte-macrophage colony- stimulating factor promoter region by genetic analy with DNase I footprinting. Mol Cell Biol 8:1979, 1988

1 1 . Nimer S, Fraser J, Richards J, Lynch M, Gasson J: The repeated sequence CATT(AlT) is required for GM-CSF promoter activity. Mol Cell Biol 10:6084, 1990

12. Fraser JK, Gasson JC, Guerra JJ, Nguyen CY, lndes JE, Nimer SD: Characterization of a cell type-specific negative regulatory region of the human GM-CSF gene. Mol Cell Biol 14:2213, I994

13. Nimer SD, Gates MJ, Koeffler HP, Gasson JC: Multiple mechanisms control the expression of granulocyte-macrophage colony-stimulating factor by human fibroblasts. J Immunol 143:2374, 1989

14. Campbell HD, Sanderson GJ, Wang Y, Hort Y, Martinson ME, Tacker WQJ. Stellwagen A, Strath M, Young IG: Isolation. structure and expression of cDNA and genomic clones for murine eosinophil differentiation factor. Comparison with other eosinophilo- poietic lymphokines and identity with interleukin-S. Eur J Biochem 174:345, 1988

15. Naora H, van Leeuwen BH, Bourke PF, Young 1G: Functional role and signal-induced modulation of proteins recognizing the conserved TCATTT-containing promoter elements in the murine IL-5 and GM-CSF genes in T lymphocytes. J Immunol 153:3466, 1994

16. Wang C-Y, Bassuk AG, Boise LH, Thompson CB, Bravo R, Leiden JM: Activation of the granulocyte-macrophage colony- stimulating factor promoter in T cells requires cooperative binding of elf-l and AP-I transcription factors. Mol Cell Biol 14:1153, 1994

17. Ye J. Young HA, Ortaldo JR, Ghosh P: Identification of a DNA binding site for the nuclear factor YYI in the human GM- CSF core promoter. Nucleic Acids Res 225672, 1994

18. Masuda ES, Tokumitsu H, Tsuboi A, Shlomai J, Hung P, Arai K-I, Arai N: The granulocyte-macrophage colony-stimulating factor promoter cis-acting element CLEO mediates induction signals in T cells and is recognized by factors related to API and NFAT. Mol Cell Biol 13:7399, 1993

19. Karim FD, Umess LD, Thummel CS, Klemsz MJ, McKercher SR, Celada A, Van Beveren C, Maki RA, Gunther CV, Nye JA, Graves BJ: The ETS-domain: A new DNA-binding motif that recognizes a purine-rich core DNA sequence. Genes Dev 4:1451, 1990

20. Ascione R, Thompson DM, Thomas R, Panayiotakis A, Rani- say R, Tymms M, Kola I, Seth A: Influence of nucleotides flanking the -GGAA- core sequence on ETSl and ETS2 DNA-binding activity and the mechanism of ETSl autoregulation. Int J Oncol 1:631, 1992

21. Wang C-Y, Petrynick B, Ho I-C, Thompson CB, Leiden JM: Evolutionarily conserved ets family members display distinct DNA binding specificities. J Exp Med 175:1391, 1992

22. Shimada Y, Migliaccio G, Ruscetti S, Adamson JW, Migliac- cio AR: Expression of the interleulun-3 and granulocyte-macrophage colony-stimulating factor genes in Friend spleen focus-forming virus-induced erythroleukemia. Blood 79:2423, 1992

23. Cann AJ, Koyanagi Y, Chen ISY: High efficiency transfection





of primary human lymphocytes and studies of gene expression. On- cogene 3:123, 1988

24. Gorman CM, Moffat LF, Howard BH: Recombinant genomes which express chloramphenicol acetyltransferase in mammalian cells. Mol Cell Biol 2:1044, 1982

25. Seed B, Sheen J-Y: A simple phase-extraction assay for chloramphenicol acyltransferase activity. Gene 67:271, 1988

26. Wolin M, Komuc M, Hong C, Shin S-K, Lee F, Lau R, Nimer S: Differential effect of HTLV infection and HTLV tax on interleukin-3 expression. Oncogene 8: 1905, 1993

27. Dignam JD, Lebovitz RM, Roeder RG: Accurate transcription initiation by RNA polymerase LI in a soluble extract from isolated mammalian nuclei. Nucl Acids Res 11:1475, 1983

28. Gottschalk LR, Giannola DM, Emerson SG: Molecular regulation of the human IL-3 gene: Inducible T cell-restricted expression requires intact AP-1 and Elf-1 nuclear protein binding sites. J Exp Med 178:1681, 1993

29. Sakamoto K, Nimer S, Rosenblatt J, Gasson J: HTLV-I and HTLV-I1 tax trans-activate the human EGR-1 promoter through different cis-acting sequences. Oncogene 7:2125, 1992

30. DiPersio JF, Brennan JK, Lichtman MA, Abboud CN, Kirk- patrick FK: The fractionation, characterization and subcellular local- ization of colony-stimulating activities released by the human mono- cyte-like line, GCT. Blood 56:717, 1980

31. Leung S , McCracken S , Ghysdael J, Miyamoto NG: Require- ment of an ETS-binding element for transcription of the human lck type I promoter. Oncogene 8:989, 1993

32. Wurster AL, Siu G, Leiden JM, Hedrick SM: Elf-1 binds to a critical element in a second CD4 enhancer. Mol Cell Biol 145452, 1994

33. Thompson CB, Wang C-Y, Ho I-C, Bohjanen PR, Petryniak B, June CH, Miesfeldt S , Zhang L, Nabel GJ, Karpinski B, Leiden JM: cis-acting sequences required for inducible interleukin-2 enhancer function bind a novel ets-related protein, elf-I. Mol Cell Biol 12:1043, 1992

34. Nabel GJ, Gorka C, Baltimore D: T-cell-specific expression of interleukin-2: Evidence for a negative regulatory site. Proc Natl Acad Sci USA 85:2934, 1988

35. Kaushansky K Control of granulocyte-macrophage colony- stimulating factor production in normal endothelial cells by positive and negative regulatory elements. J Immunol 143:2525, 1989

36. Mathey-Prevot B, Andrews NC, Murphy HS, Kreissman SG, Nathan DG: Positive and negative elements regulate human interleukin-3 expression. Proc Natl Acad Sci USA 87:5046, 1990

37. Li Weber M, Eder A, Krafft Czepa H, Krammer PH: T cell- specific negative regulation of transcription of the human cytokine IL-4. J Immunol 148:1913, 1992

38. Shaw G, Kamen R: A conserved AU sequence from the 3' untranslated region of GM-CSF mRNA mediates selective mRNA degradation. Cell 46:659, 1986

39. Cockerill PN, Shannon MF, Bert AG, Ryan GR, Vadas MA: The granulocyte-macrophage colony-stimulating factodinterleukin 3

locus is regulated by an inducible cyclosporin A-sensitive enhancer. Proc Natl Acad Sci USA 90:2466, 1993

40. Nimer SD: Tax responsiveness of the GM-CSF promoter is mediated by mitogen inducible sequences. New Biol 3:997, 1991

41. Nomiyama H, Hieshima K, Hirokawa K, Hattori T, Takatsuki K, Miura R: Characterization of cytokine LD78 gene promoters: Positive and negative transcriptional factors bind to a negative regulatory element common to LD78, interleukin-3, and granulocyte- macrophage colony-stimulating factor gene promoters. Mol Cell Biol 13:2787, 1993

42. Miyatake S, Shlomai J, Arai K-I, Arai N: Characterization of the mouse colony-stimulating factor (GM-CSF) gene promoter: Nuclear factors that interact with an element shared by three lymphokine genes-those for GM-CSF, interleukin-4 (IL-4), and IL-5. Mol Cell Biol 11:5894, 1991

43. Wang C-Y, Petryniak B, Thompson CB, Kaelin WG, Leiden JM: Regulation of the ets-related transcription factor Elf-1 by binding to the retinoblastoma protein. Science 260:1330, 1993

44. Ben-David Y, Giddens EB, Letwin K, Bemstein A: Erythro- leukemia induction by friend murine leukemia virus: Insertional activation of a new member of the ets gene family, Fli- 1, closely linked to c-ets-1. Genes Dev 5:908, 1991

45. Bhat NK, Komschlies KL, Fujiwara S , Fisher RJ, Mathieson BJ, Gregorio TA, Young HA, Kasik JW, Ozato K, Papas TS: Expres- sion of ets genes in mouse thymocyte subsets and T cells. J Immunol 142:672, 1989

46. Bhat NK, Thompson CB, Lindsten T, June CH, Fujiwara S , Koizumi S , Fisher RJ, Papas TS: Reciprocal expression of human ETSl and ETS2 genes during T-cell activation: Regulatory role for the protooncogene ETS1. Proc Natl Acad Sci 87:3723, 1990

47. Bosselut R, Duvall JF, Gegonne A, Bailly M, Hemar. A, Brady J, Ghysdael I: The product of the c-ets-1 protooncogene and the related Ets-2 protein act as transcriptional activators of the long terminal repeat of human T cell leukemia virus HTLV- 1. EMBO J 9:3137, 1990

48. Wasylyk B, Wasylyk C, Flores P, Begue A, Leprince D, Stehelin D: The c-ets protooncogenes encode transcription factors that cooperate with c-fos and c-jun for transcriptional activation. Nature 346:191, 1990

49. Hipskind RA, Rao VN, Mueller CGF, Reddy ESP, and Nord- heim A: Ets-related protein elk-1 is homologous to the c-fos regulatory factor ~ 6 2 ~ ' ~ . Nature 354531, 1991

50. Pongubala JMR, Van Beveren C, Nagulapalli S , Klemsz MJ, McKercher SR, Maki RA, Atchison ML: Effect of PU.1 phosphorla- tion on interaction with NF-EM5 and transcriptional activation. Sci- ence 259:1622, 1993

51. Wotton D, Ghysdael J, Wang S , Speck NA, Owen MJ: Coop- erative binding of ets-1 and core binding factor to DNA. Mol Cell Biol 14340, 1994

52. Rincon M, Flavell RA: AP-1 transcriptional activity requires both T-cell receptor-mediated and co-stimulatory signals in primary T lymphocytes. EMBO J 13:4370, 1994




1996 87: 3694-3703

SD Nimer, W Zhang, K Kwan, Y Whang, J Zhang and Y] Wang Y [corrected to Whang gene [published erratum appears in Blood 1996 Oct 1;88(7):2818]in the human granulocyte-macrophage colony-stimulating factor Adjacent, cooperative elements form a strong, constitutive enhancer

http://www.bloodjournal.org/content/87/9/3694.full.htmlUpdated information and services can be found at:

Articles on similar topics can be found in the following Blood collections

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://www.bloodjournal.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://www.bloodjournal.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

Copyright 2011 by The American Society of Hematology; all rights reserved.Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American


http://www.bloodjournal.org/content/87/9/3694.full.html

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests

http://www.bloodjournal.org/site/misc/rights.xhtml#reprints

http://www.bloodjournal.org/site/subscriptions/index.xhtml





adjacent, cooperative elements form a strong, constitutive ... · adjacent, cooperative elements...

Documents