isoforms of the ets transcription factor nerf/elf2 ... · esei dnas were inserted into bamhi and...

1

Isoforms of the Ets transcription factor NERF/ELF2 physically interact with AML1

and mediate opposing effects on AML1 mediated transcription of the B cell-specific

blk gene

Je-Yoel Cho1#, Yasmin Akbarali1, Luiz F. Zerbini1, Xuesong Gu1, Jay Boltax1, YihongWang1, Peter Oettgen1, Dong-Er Zhang2, and Towia A.Libermann1*

1BIDMC Genomics Center and New England Baptist Bone and Joint Institute, Beth Israel

Deaconess Medical Center and Harvard Medical School, 4 Blackfan Circle, Boston,

Massachusetts 02115

2Department of Molecular and Experimental Medicine, MEM-L51, The Scripps Research

Institute, 10550 North Torrey Pines Road, La Jolla, California 92037, USA

Running Title: NERF cooperativity with AML1

# Current address: Department of Oral Biochemistry, School of Dentistry, Kyungpook

National University, Daegu, Korea

*Corresponding Author

1BIDMC Genomics Center and New England Baptist Bone and Joint Institute, Beth Israel

Deaconess Medical Center and Harvard Medical School, Harvard Institutes of Medicine,

4 Blackfan Circle, Boston, Massachusetts 02115, USA

Telephone: 617-6673393

Fax: 617-9755299

E-Mail address: [email protected]

JBC Papers in Press. Published on February 17, 2004 as Manuscript M309074200

Copyright 2004 by The American Society for Biochemistry and Molecular Biology, Inc.

by guest on February 2, 2020http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

2

The abbreviations used are: AML, acute myeloid leukemia; blk, B lymphoid kinase;

BSAP, B cell lineage-specific activator protein; ELF, E74-like Factor; GST, glutathione-

s-transferase; MEF, Myeloid elf-1-like factor; NERF, New Ets-Related Factor; RUNX,

runt box.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

3

ABSTRACT

We previously isolated different isoforms of a new Ets transcription factor family

member, NERF/ELF2, NERF-2, NERF-1a and NERF-1b. In contrast to the inhibitory

isoforms NERF-1a and NERF-1b, NERF-2 acts as a transactivator of the B cell-specific

blk promoter. We now report that NERF-2 and NERF-1 physically interact with AML1

(RUNX1), a frequent target for chromosomal translocations in leukemia. NERF-2 binds

to AML1 via an interaction site located in a basic region upstream of the Ets domain.

This is in contrast to most other Ets factors such as Ets-1 that bind to AML1 via the Ets

domain suggesting that different Ets factors utilize different domains for interaction with

AML1. The interaction between AML1 and NERF-2 leads to cooperative transactivation

of the blk promoter, whereas interaction between AML1 and NERF-1a leads to

repression of AML1 mediated transactivation. To delineate the differences in function of

the different NERF isoforms we determined that the transactivation domain of NERF-2 is

encoded by the amino-terminal 100 amino acids which have been replaced in NERF-1a

by a 19 amino acid transcriptionally inactive sequence. Furthermore, the acidic Domains

A and B that are conserved in NERF-2 and the related Elf-1 and MEF/ELF4, but not

NERF-1a are largely responsible for NERF-2 mediated transactivation. Since

translocation of the Ets factor Tel to AML1 is a frequent event in childhood preB

leukemia, understanding the interaction of Ets factors with AML1 in the context of a B

cell-specific promoter might help to determine the function of Ets factors and AML1 in

leukemia.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

4

INTRODUCTION

Immune system development is regulated by the combined action of cytokines,

cell-cell interactions, and a distinct set of transcription factors that modulate and

coordinate developmental stage-specific and lineage-specific gene expression. Since

expression of a specific set of genes is distinct for each cell lineage and each

developmental stage, analysis of transcription factors involved in the regulation of these

lineage-specific genes is one approach to understand the molecular mechanisms

underlying differentiation. Analysis of regulatory regions of B cell-specific genes has

revealed the presence of DNA motifs that are repeatedly found in variations and different

combinations in most B cell-specific genes. Most B cell-specific genes contain binding

sites for different Ets factors such as Pu.1, Ets-1, ELF-1, NERF (ELF2), ERG and factors

such as Oct-2, Ikaros, E2A, rel/NF-κB factors, BSAP, LEF-1, N-myc and EBF (1).

Cooperativity between these different factors leads to selective stage- and cell-specific

expression of a particular gene. B cell-specificity of a transcription factor does not

always coincides with its exclusive expression in B cells such as the ubiquitously

expressed E2A, but rather depends on formation of B cell-specific protein-protein

complexes due to the combination of a particular set of factors expressed in B cells.

The ETS transcription factor family plays a key role in cellular differentiation,

proliferation, development, apoptosis and immune responses including the growth,

survival, and activation of hematopoietic cells (2). More than 30 Ets family homologues

have been cloned (3), which function as transcription factors under physiological

conditions and transform cells when aberrantly expressed. All Ets factors share a highly

conserved 80-90 amino acid long DNA binding domain, the ETS domain (4-6). This


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

5

domain is sufficient to interact specifically with DNA and, due to the conserved DNA

binding domain, binding sites for Ets factors are similar with a core binding motif

“A/GGAA/T” and slight differences in flanking nucleotides for different Ets factors.

Outside the DNA binding domain, very little homology is common to all members of the

Ets family. Ets related proteins can be grouped into subclasses based on additional

homologous domains unique for particular members of the Ets family (4-6) such as

NERF/ELF2, ELF-1 and MEF/ELF4 which contain several homologous regions outside

the ETS domain not found in other Ets factors. Protein-protein interactions are critical

for the function of Ets related proteins and occur with transcription factors of various

other families. Thus, ERP, SAP-1 and ELK-1 form a ternary complex with the serum

response factor, whereas GABP-α interacts with GABP-β (2,6). Additional regulation of

Ets factors involves phosphorylation by kinases activated via different signal transduction

pathways (6,7).

In an effort to search for novel members of the Ets family which might be

relevant for B-cell gene regulation, we have previously identified and characterized

cDNA clones encoding three alternative splice products of a novel member of the Ets

gene family, new Ets-related factor (NERF/ELF2), NERF-1a, NERF-1b, and NERF-2,

which differ in their amino termini (8). NERF is most closely related to ELF-1 and

MEF/ELF4. We have demonstrated that both NERF and the related ELF-1 are involved

in regulating a set of genes in B cells and myeloid cells and are highly expressed in B

cells and myeloid cells (8,9). We also showed that NERF-2 is expressed in endothelial

cells and transactivates the regulatory regions of the Tie2 gene (10). Interestingly, NERF-


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

6

2 expression is also increased in endothelial cells in response to hypoxia and to

angiopoietin-1 indicating functions for NERF in the immune system and vasculature (11).

AML1 (also known as RUNX1, CBFα2, and PEBP2α2) is a transcription factor

critical for definitive hematopoiesis (12,13). The AML1 recognition sequence is required

for tissue specific expression of several hematopoietic genes including M-CSF receptor,

GM-CSF, IL-3, T cell receptors, immunoglobulin µ heavy chain, defensin NP-3, and

myeloperoxidase (14-22). The AML1 gene is the most frequent target for chromosomal

translocations in human leukemias. It is rearranged in distinct chromosomal translocations

associated with acute myeloid leukemia [AML; t(8;21), t(12;21), t(16;21), t(19;21)] (23-

26), acute lymphatic leukemia [ALL; t(12;21)] (27), and myelodysplastic syndrome

(t[3;21]) (28,29) . AML1 (CBFα2) forms a heterodimer with CBFβ. CBFβ does not bind

DNA directly but enhances the binding of AML1 (30). Multiple α subunit genes,

including CBFα1 (AML3), CBFα2 (AML1), and CBFα3 (AML2), as well as alternatively

spliced isoforms of the α and β subunits have been detected (31,32). All of the CBFα

proteins have a DNA-binding domain (the runt domain), which is similar to the

Drosophila pair-rule gene, runt (33). To understand the function and role of AML1 in

leukemia it is important to study the molecular mechanism of AML1 mediated regulation

of gene expression.

Ets related binding sites are evident in most B cell-specific genes. Hematopoetic

genes containing high affinity NERF/ELF-1 binding sites include among others: IgH,

Terminal deoxynucleotidyltransferase (TdT) (34,35), mb-1, B29 (36), BSAP (37), lck

(38), blk (39), lyn (40,41). Blk is a B cell-specific tyrosine kinase which is expressed in


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

7

preB and mature B, but not in plasma cells; this is similar to the expression of mb-1 and

B29 (42). Blk is associated with the antigen receptor and involved in signal transduction

(39). The blk promoter contains a previously uncharacterized NERF/ELF-1 binding site

adjacent to a BSAP and AML1 site. Not much is known about regulation of blk gene

expression except that the B cell-specific transcription factor BSAP plays an important

role and that the transcription factor NF-κB/p50 interacts with the blk gene during B cell

activation (41). We, furthermore, demonstrated that AML1 binds to the blk promoter and

cooperatively transactivates the blk promoter in the presence of BSAP (43). AML1 has

been previously shown to interact with a variety of Ets factors including the related MEF

indicating that a possible interaction between AML1 and NERF may play a role in blk

gene regulation (44).

We now report that NERF-2 and ELF-1 directly interact with the runt homology

domain of AML1 through a basic region upstream of the Ets domain, and cooperate with

AML1 in activating blk promoter transcription. We also demonstrate that the NERF-1a

isoform lacks the NERF-2 transactivation domain and represses AML1 mediated

transactivation of the blk promoter.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

8

MATERIALS AND METHODS

Plasmid Construction

Full length and different fragments of NERF-2, NERF-1a, and ELF-1 were cloned

into pGEX-5X-3 vector to make GST fusion proteins in BL21 Escherichia coli Strain

(Novagen). The expression vectors of full-length AML1, AML1-208, and AML187-208 are

all derived from AML1B and were prepared as reported previously (45). Lyn and blk

promoter constructs were also prepared as reported previously (8). For expression of

NERF-2 in mammalian cells, NERF-2 was cloned into the Not I site of the pCi vector

which has a CMV promoter and enhancer.

For Gal4 expression constructs, NERF-2 and NERF-1 fragments were cloned in

the reading frame into the BamH I site of a Gal4 (1-147) expression vector pSG424. The

plasmids Gal4-NERF-2 (1-203), Gal4-NERF-2 (1-164), Gal4-NERF-2 (1-141), Gal4-

NERF-2 (1-108), Gal4-NERF-2 (1-103), Gal4-NERF1a (1-155), and Gal4-NERF1b (1-

155) were cloned into the Gal4 vector by deleting C-termini using restriction enzymes,

Bgl II, Eae I, Nco I, EcoN I, Nde I, Bgl II and Bgl II. Gal4-NERF-2 (1-103) Mut A, Mut

B, Mut C, Mut D and Mut A+B were generated by site directed mutagenesis by

substituting Glutamic acid (E) for Alanine (A).

5’ Flag vector was prepared by inserting Flag sequence oligos to NheI and KpnI

site of pCDNA3.1 plasmid. The inserted Flag sequence was ATG

GACTACAAAGACGATGACGACAAG. 3’ Myc vector was prepared by inserting Myc

epitope oligos sequences to XhoI and ApaI site of pCDNA3.1 plasmid. The inserted Myc

epitope sequence was GAA CAA AAA CTC ATC TCA GAA GAG GAT CTG. PCR


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

9

amplified -using Hi-Fidelity Taq polymerase (Invitrogen)- NERF2, MEF, PDEF and

ESEI DNAs were inserted into BamHI and XhoI site of 5’Flag-pCDNA3.1 vector. PCR

amplified AML1 was cloned into BamHI and XhoI sites of 3’Myc pCDNA3.1 plasmid to

construct AML1-3’Myc-pCDNA3.1.

Deletion mutants of NERF-2 (del 108-180) were produced by removing the Nerf-

2 DNA sequences between two EcoN I sites at aa 108 and aa 180 in the NERF2-pCi

vector. Briefly, the plasmid was digested with EcoNI and isolated bands were re-ligated

wi th l inkers (5 ' -TTGAGGGATTCAAGAAGTCCTGA-3 ' and 5 ' -

CTCAGGACTTCTTGAAT CCCTCA-3'), which were previously annealed to fuse the

N-terminal 108 amino acids of NERF-2 in frame to the carboxy-terminus of NERF-2

starting at amino acid 180. For NERF-2-Flag (del 108-180), the mutant NERF-2 (del

108-180)-pCi was used as a template for a PCR amplification with primers (5'-

C G C G G A T C C A T G A C A T C A G C A G T G G T T G A C - 3 ' a n d 5 ' -

CGCGTCGACTTTCTCACATGTCACTAGTCC-3') that contain BamH I and Sal I

restriction enzyme sites. The PCR amplified mutant NERF-2 DNA was inserted in-frame

into the 5’Flag-pCDNA3.1 vector.

Cell Culture and Transfection

CV-1 and Human Embryonic Kidney (HEK) 293T cells were grown in DMEM

(BioWhittaker) containing 10% FBS and penicillin/streptomycin. Co-transfections of

3x105 CV-1 cells were carried out with 2 µg reporter gene construct DNA and 3 µg


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

10

expression vector DNA using 12.5 µl Lipofectamine (Invitrogen) as described previously

(45). The cells were harvested 16 hours after transfection and assayed for luciferase

activity as described previously (43). Transfections for every construct were performed

independently in duplicates or triplicates and repeated 3 to 4 times with two different

plasmid preparations with similar results.

In Vitro Translation

Protein in vitro translation was performed using the TNT T7-coupled reticulocyte

lysate system according to the manufacturer's protocol (Promega). The TNT lysate

contains approximately 150 µg/µl endogenous protein. Each in vitro translation reaction

uses 25 µl of TNT lysate per 50 µl reaction.

GST Pull-down Experiments

The GST-pull down experiments were performed as described previously with

some modification (46). The integrity of the bacterially expessed GST fusion proteins

was examined by SDS-PAGE, followed by Coomassie blue staining. Approximately

equal amounts of the fusion proteins were used for each reaction. Briefly, the GST-fusion

proteins were expressed in BL21 E. coli, expression was confirmed and quantified by

SDS-PAGE, and GST-fusion proteins were immobilized on glutathione Agarose beads

for pull-down assays as described (46). Recombinant [S35]-AML1 and [S35]-RBTN2

were produced by in vitro transcription/translation (TNT Coupled Reticulocyte Lysate

Systems Kit, Promega) from pCi-AML1 and pCi-RBTN2 plasmid templates.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

11

Recombinant proteins were incubated with GST-fusion proteins at 4 oC for 1 hr in NETN

buffer (20mM Tris, pH7.5, 100mM NaCl, 1mM EDTA, 0.25% BSA, 0.5% NP40 and

0.1mM DTT). Beads then were washed 3 times and associated proteins were resolved by

SDS-PAGE and visualized by autoradiography.

Co-immunoprecipitation (Co-IP) and Western blot

HEK 293T cells grown on 100-mm dishes were co-transfected overnight with 6

µg Flag-tagged Ets constructs and 6 µg Myc-tagged AML1 expression vector or the

parental vector using Lipofectamine PLUS (Invitrogen). After overnight, the cells were

changed to fresh growth media for 24-36 hrs. Then the cells were collected in lysis

buffer (150 mM NaCl, 1mM EDTA, 20 mM Tris (pH8.0), 0.5% of Igepal (NP-40

substitute), 0.5% of TritonX-100, 10% glycerol and 1:30 diluted protease inhibitor

cocktail (Roche)). The insoluble cell debris was removed by centrifugation at 14,000xg

for 20 min at 4oC. The supernatants were transferred to a new tube. This total cell

lysates were diluted at a 4:6 ratio with IP dilution buffer (50 mM Tris-Hcl, pH7.4, 150

mM NaCl, 1mM EDTA, 1% Triton X-100 and 1:30 diluted protease inhibitor cocktail

(Roche)). Then, 20 µl of anti-Flag monoclonal antibody-conjugated agarose beads (M2

Agarose from Sigma) were added to the cell lysates. Immunoprecipitations were carried

out at 4oC with slow rotating motion overnight. The IP complex was washed 5 times

with TBS (25mM Tris, 2.7mM KCl, 137mM NaCl, pH 7.4). After final washing, the

bound proteins were eluted in non-reducing SDS sample buffer (63mM Tris-HCl, pH6.8,

2% SDS, 10% Glycerol, 0.005% Bromophenol Blue) or in 3X Flag peptide (150ng/µl

final concentration) (Sigma). The samples were boiled for 3 min and loaded on to the


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

12

10% Tris-glycine gel (Bio-Rad). After electrophoresis gels were transferred to a PVDF

membrane for 1hr. The membranes were blocked in 5% dry milk in TBST (25mM Tris,

2.7mM KCl, 137mM NaCl, pH 7.4 plus 0.1% Tween-20) overnight at 4oC. The

transferred membrane was incubated with an anti-myc polyclonal antibody conjugated

with HRP (Santa Cruz, 1:1,250 dilution) 1.5 hr at room temperature. The signal was

detected by ECL detection reagents (Amersham) on an X-ray film.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

13

RESULTS

NERF-2 Physically Interacts with AML1

Our previous studies established that the promoter region of the B cell-specific blk

gene contains Ets- and AML1-binding sites in close proximity to each other (43). Since

AML1 has been shown to interact and cooperate with other Ets transcription factors

including Ets-1 and MEF, we explored the possibility that NERF-2 which activates the

blk promoter can cooperate with AML1 in regulating blk gene expression (8). We first

tested whether NERF-2 directly interacts with AML1 using a GST-pull down assays, in

which Escherichia coli expressing GST fusion proteins immobilized on glutathione-

agarose beads were incubated with in vitro translated 35S-labelled proteins. As shown in

Fig. 1, in vitro translated full-length AML1 can be specifically retained on agarose beads

containing the fusion protein made from NERF-2 (NERF-2-GST), but not on glutathione

beads containing only GST. In contrast to AML1 Rhombotin 2 (RBTN2) did not bind to

NERF-2 in the GST-pull down assay. This result provided evidence that NERF-2 can

specifically interact with AML1.

A Basic Region Upstream of the Ets Domain of NERF-2 Interacts with AML1

To map the region of NERF-2 responsible for interaction with AML1, 35S-labeled

AML1 was incubated with a series of GST-NERF-2 deletion mutant proteins (Fig. 2). As


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

14

shown in Fig. 2 and summarized in Fig. 4, mutants containing the N-terminal amino acids

from 1-203 (lane 4) still interacted with AML1 proteins as efficiently as full length

NERF-2 1-581 (lane 3). C-terminal mutants containing amino acids from 204-581 (lane

5), however, lost interacting activity with AML1.

The N-terminus of NERF-2 contains three domains, A, B, and C, that are highly

conserved between the three members NERF, ELF-1 and MEF, of this Ets subfamily. No

specific function has been attributed to any of these domains until now. To examine

whether any of these N-terminal homology domains A, B, or C is involved in the

physical interaction with AML1, we generated 1-203ΔA, ΔB, ΔC or ΔA/Β mutants by

replacing acidic amino acids with alanine and performed GST-pull down assays with

AML1. All of the deletion mutants tested still showed physical interactions indicating

that none of those regions are important for the interaction with AML1. Furthermore,

GST fusion proteins with NERF-2 deletion mutants containing residues 1-104 (lane 15),

1-141 (lane 6) or 1-164 (lane 11) did not bind AML1 either, indicating that the NERF-2

interaction domain with AML1 is concentrated between residues 165-203. Indeed,

NERF-2 mutants that contained residues 105-203 (lane 16) and 121-203 (lane 17) still

retained full interaction capacity with AML1. Similarly, the NERF-2 mutant containing

residues 111-180 (lane 14) maintained full interacting activity, whereas 165-203 (lane 12)

exhibited somewhat reduced binding activity. These results demonstrate that the minimal

domain for NERF-2 interaction with AML1 is located between aa 165 and aa 180,

although the region 111-165 appears also to contribute to the interaction with AML1

possibly by stabilizing the principal contact regions. To further confirm the importance

of the basic region of NERF-2 with AML1, we also generated a GST fusion protein


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

15

containing the N-terminus of NERF1A from aa 1-143 (lane 18) which still contains the

basic domain of NERF-2, but lacks the N-terminal part of NERF-2. GST-pull down

showed that the N-terminus of NERF-1a interacts with AML1 as well, further supporting

that the basic region of NERF is involved in binding to AML1.

NERF-2 binds to AML1 in vivo.

To confirm that NERF-2 can bind to AML1 in vivo, we performed a co-

immunoprecipitation experiment. For this purpose, we generated expression vectors for a

fusion protein of NERF-2 containing the Flag tag at the N-terminus and for a fusion

protein of AML1 containing the MYC tag at the C-terminus. These constructs were

either individually transfected or co-transfected into 293T cells. Total cell lysates were

immunoprecipitated using anti-Flag conjugated agarose beads followed by Western blot

analysis with anti-MYC antibody. Vector alone, NERF-2-5Flag or AML1-3myc itself

did not give any signal. AMLl, however, was detected clearly when NERF-2-5Flag and

AML1-3myc were co-transfected into the cells (Fig. 3A). These data most vividly

demonstrate that AML1 protein binds to the NERF-2 protein in vivo.

AML1 interacts with selected other members of the ETS family in vivo.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

16

To test if AML1 interacts with other ETS family members in vivo, the AML1-

3myc construct was co-transfected with either NERF-2-5Flag, MEF-5Flag, PDEF-5Flag,

or ESE1-5Flag into 293T cells (Fig 3B). The expression of each ETS-5Flag construct

was then tested by SDS-PAGE and Western Blot using the Flag antibody.

Immunoprecipitation with the Flag antibody followed by Western blot analysis with the

AML1 antibody revealed that AML1 binds to NERF-2 and MEF as expected. However,

no binding of AML1 to PDEF or ESE1 was detected demonstrating that AML1 interacts

with a selected subset of the Ets family.

NERF-2 cooperates with AML1 in transactivation of the blk promoter, but NERF-1A

represses AML1 mediated transactivation.

To evaluate whether AML1 interaction with NERF leads to cooperativity in the

context of the blk promoter, we performed co-transfection experiments. NERF-2 and

AML1 together with its non-DNA-binding heterodimer partner CBFβ, either alone or in

combination, were co-transfected along with the blk promoter-luciferase construct into

CV-1 cells and luciferase assays were performed 16 hours later. NERF-2 activated the blk

promoter 3.6-fold and AML1c (which is longest form of AML1 splice variants, with

480aa) 12-fold. However, the combination of NERF-2 with AML1c led to a synergistic

increase in blk promoter activity of 42-fold which is significantly more than would be

expected due to an additive effect (Fig 4A). This experiment clearly demonstrates that

NERF-2 cooperatively enhances AML1 mediated transactivation of the blk promoter. In


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

17

contrast to the transactivator NERF-2, the NERF-1A isoform by itself did not

significantly transactivate the blk promoter and in combination with AML1 drastically

inhibited AML1 mediated blk promoter transactivation suggesting that NERF-1A might

work as a transcriptional repressor.

The domain of NERF-2 that retained maximum interaction with AML-1 contains amino

acid residues 111-180, including a basic domain conserved among all three Ets family

members, NERF-2, ELF-1 and MEF. To evaluate whether cooperative stimulation of the

blk promoter requires physical interaction between NERF-2 and AML1, we generated

NERF-2 (del 108-180) deletion mutants that lack the AML1 interaction domain between

amino acids 108 to 180 including the basic domain. We derived 2 different clones of

these NERF-2 (del 108-180) mutants fused to the amino-terminal Flag peptide in the Flag

vector and first confirmed their proper expression after transfection into 293 cells and the

size of the proteins by Western blot analysis (Fig 4B). Figure 4B demonstrates that both

mutant NERF-2 proteins were expressed at similar levels as wild type NERF-2 and with

the expected molecular weights. Then we tested by co-immunoprecipitation and Western

blot analysis whether these NERF-2 deletion mutants had lost their abilities to physically

interact with AML1 in vivo. As shown in Fig 4C, both NERF-2 (del 108-180) mutants

were unable to physically interact with AML1, while the wild type NERF-2 efficiently

interacted with AML1, confirming that deletion of aa 108-180 eliminates the AML1

interaction domain.

To evaluate whether these mutants were able to cooperate with AML1 in transactivation

of the blk promoter, CV-1 cells were transiently co-transfected with wild type or mutant

NERF-2 and AML1 together with CBFβ, either alone or in combination (Fig 4D).

Deletion of amino acids 108-180 resulted in complete loss of cooperativity with AML1


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

18

when compared to the activity induced by wild type NERF-2 (Fig 4D). Transactivation of

the blk promoter by these mutants alone was also slightly reduced and may be the result

of reduced transactivation capacity due to either lack of interaction with endogenous

AML1 or an effect on the basal transactivation capacity of NERF-2, even though neither

the transactivation domain (see below) nor the DNA binding domain were changed.

These data clearly demonstrate that disruption of NERF-2 interaction with AML1 results

in the loss of transcriptional cooperativity of NERF-2 with AML1 and that the AML1

interaction domain is critical for cooperativity.

To determine whether the transactivation domain of AML1 is essential for

cooperativity with NERF-2 AML1 mutants truncated at the carboxy terminus were

transfected in the absence or presence of NERF-2 into CV-1 cells (Fig. 4E). While

termination of AML1 at amino acid 381 and 351 did not affect cooperativity with NERF-

2 or transactivation by AML1 alone, termination at amino acid 289 drastically reduced

cooperativity and correlated with the loss of transactivation capability of AML1 itself.

These results reveal that the C-terminal transactivation domain of AML1 is necessary for

synergy with NERF-2.

NERF-2, but not NERF-1A contains a transactivation domain encoded by the amino-

terminal 100 amino acids.

Although NERF-2 and NERF-1A were able to interact with AML1, only NERF-2

acts as a transcriptional activator and cooperates with AML1. This result suggests that the


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

19

AML1 interaction domain is distinct from the NERF-2 transactivation domain. To define

the transactivation domain of NERF-2 in more detail we generated carboxy-terminal and

amino-terminal deletions of NERF-2, as shown in Fig. 5A. Co-transfection experiments

were performed with expression vectors encoding full length and deletion mutants of

NERF-2 and the lyn promoter Ets site luciferase construct, another B cell target for

NERF that is highly inducible by NERF-2. Deletion of the carboxy-terminus of NERF-2

(NERF-2 1-381 or NERF-2 1-510) decreased transactivation slightly, whereas deletion of

the amino-terminal 103 amino acids (NERF-2 104-581) abolished NERF-2

transactivation capacity completely (Fig. 5A). NERF-1A which lacks the amino-terminus

of NERF-2 and instead has a distinct amino-terminus did not transactivate the lyn

promoter Ets site either and actually slightly decreased promoter activity compared to the

parental pCI expression vector. These data provide strong evidence that the main NERF-

2 transactivation domain is located at the amino-terminus that is absent from NERF-1A.

These data also demonstrate that the transactivation domain is distinct from the AML1

interaction domain.

To confirm that the N-terminal deletion NERF-2 does not affect protein

expression, protein stability or binding to the lyn promoter, we performed EMSAs using

whole cell extracts of CV-1 cells transfected with either pCI-NERF-2 WT or pCI-NERF-

2 (del 1-103) and oligos of the lyn promoter NERF binding site. Deletion of the amino-

terminal 103 amino acids of NERF-2 did not affect protein expression levels or binding

to the lyn promoter as demonstrated by the EMSA (Fig. 6), suggesting that the loss of

transactivation activity of N-terminal deleted NERF-2 was not due to either decreased


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

20

protein amount of mutant NERF-2 or the loss of DNA binding activity of mutant NERF-

2, but rather due to the loss of the transactivation domain of NERF-2.

The conserved amino-terminal domains A and B are part of the NERF-2 transactivation

domain

To analyze in more detail which domains of the amino-terminus of NERF-2 might

be part of the transactivation domain, we mutated 3 domains (Domains A, B and C) in the

N-terminal region of NERF-2 that are not present in NERF-1A, but are conserved

between NERF-2, ELF-1 and MEF (Fig 7A). We deleted 5-6 amino acids each in

domain A (Mut A), domain B (Mut B) and domain C (Mut C) or mutated two glutamic

acid residues in domain B (Mut D) to alanine (Fig. 7A and B). We hypothesized that

evolutionary conservation of these acidic domains may indicate an important function for

these domains. Deletions A, B, and D reduced transactivation to about 50% compared to

wild type NERF-2, and a combined deletion of domains A and B (Mut A+B) further

reduced transactivation, but not to the level observed by the full amino-terminal deletion

(Fig. 5B). Deletion of domain C (Mut C), in contrast, did not affect transactivation

capacity (Fig. 5B). These results demonstrate that the amino-terminal conserved domains

A and B, but not C are integral components of the NERF-2 transactivation domain which

is either absent or truncated in NERF-1A and NERF-1B. To confirm that the differences

in transactivation capability of NERF-2 deletion mutants are not due to reduced levels of

expression in transfected cells or reduced DNA binding, we performed EMSAs using


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

21

aliquots of the cell extracts from transfected CV-1 cells (Fig. 6). All NERF-2 deletion

mutants formed complexes with expected mobility and similar intensity suggesting that

decreases in transactivation are not due to a lack of or reduced expression, or a lack of

DNA binding. The fact that deletion of NERF-2 domains A and B together drastically

impairs, but does not abolish transactivation activity completely suggests that there is an

additional region in the N-terminus (aa 1-103), which contributes to the transactivation of

NERF-2 (Fig. 5B).

To further confirm that the amino-terminus of NERF-2, but not NERF-1A

exhibits transactivation activity, we fused different domains of NERF-2 and NERF-1 to

the Gal4 DNA binding domain and tested their transactivation activities on the luciferase

reporter containing 3 binding sites for GAL4 fused to a minimal promoter. We

cotransfected each of these constructs into COS cells along with the Gal4luc reporter

(pGSE1bluc or pSGluc1b). As predicted by our deletion mutants, the amino-terminus of

NERF-2 strongly transactivated the GAL4 luciferase reporter by up to 900 fold compared

to the GAL4 DNA binding domain alone (Fig. 8). This activation was equally strong as

the amino-terminus of the related ELF-1 and significantly stronger that the ELK

transactivation domain. Carboxy-terminal deletions of the NERF-2 amino-terminus down

to amino acid 103 (NERF-2 (1-103)) did not diminish transactivation capacity indicating

that the transactivation domain is indeed encoded by the amino-terminal 103 amino acids

that contain domains A, B, and C. Consistent with the internal deletion mutants described

in Figure 5B, GAL4 fusions containing the amino-terminal 103 amino acids combined

with the deletions A, B and D showed significant loss of transactivation activity, while

deletion mutant C still maintained full transactivation activity. Mutant A+B showed


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

22

again a further decrease in transactivation. In concordance, the amino-terminus of

NERF-1A or NERF-1B (aa 1-155) which lacks domains A and B did not exhibit any

transactivation activity, further highlighting the importance of domains A and B for the

transactivation activity (Fig. 8) (8).


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

23

DISCUSSION

Translocations of Ets factors and AML1 have been shown to be of critical

importance in leukemias and various other types of cancer, including translocations of the

Ets factor Tel/ETV6 to AML1 itself (27). Furthermore, several members of the Ets

family have been demonstrated to enhance AML1 mediated transactivation of various

genes via direct protein-protein interaction via the Ets DNA binding domain and the

AML1 runt domain (2,44). This finding implies also that Ets factors should affect

transactivation in leukemias mediated by AML1 translocation proteins such as Tel/AML1

or AML1/ETO due to these protein-protein interactions. We have previously

demonstrated that AML1 directly interacts with the B cell specific transcription factor

BSAP/PAX-5 and that this interaction leads to synergistic enhancement of transactivation

of the B cell specific blk gene promoter (43). We also previously showed that the blk

promoter is regulated by the Ets factors NERF-2 and ELF-1 that are expressed in B cells

(8).

In this study, we examined the physical interaction of NERF-2 and NERF-1 with

AML1 and the functional consequences of these interactions in the context of the B cell

specific blk gene promoter. We demonstrate that the two NERF isoforms, NERF-2 and

NERF-1, directly interact with AML1. Using various NERF-2 and NERF-1 GST-fusion

proteins, we identified the basic domain of NERF-2 (aa 111-180) upstream of the Ets

DNA binding domain as the major protein-protein interaction domain with AML1. Both

NERF-2 and NERF-1a interact with AML1 indicating that both isoforms could affect

AML1 activity. Indeed, whereas NERF-2 enhanced AML1 mediated transactivation of

the blk promoter, NERF-1a drastically repressed AML1 mediated transactivation. Part of

the explanation for these opposing activities of NERF-2 and NERF-1a is the lack of a


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

24

transactivation domain in NERF-1a which is located within the N-terminal 103 amino

acids of NERF-2 as shown by deletion and mutation studies as well as heterologous Gal4

fusion proteins. The opposite effects of NERF-2 and NERF-1a on AML1 activity is

highly interesting, since both NERF-2 and NERF-1a isoforms are expressed in B cells

and other cell types, although their relative ratio changes in different cell types and under

different conditions. NERF-2 and NERF-1a are actually regulated by different promoters

suggesting that different physiological settings could determine the relative level of

NERF-2 versus NERF-1a. Since NERF-2 is a positive regulator of transcription and

NERF-1a acts as a transcriptional repressor, regulated changes in the ratio of NERF-2 to

NERF-1a are expected to either enhance or repress expression of target genes. In this

context, the interaction of both NERF isoforms with AML1 would imply that AML1

mediated transactivation could be highly dependent on the ratio of NERF-2 versus

NERF-1a within the leukemic cells. With regard to AML leukemic cells that contain

AML1 translocations crucial for transfomation, NERF isoforms may be able to enhance

or reduce the transforming capacities of AML1 translocation proteins.

The AML1 interaction domain of NERF-2 was mapped to a basic domain

upstream of the Ets domain which differs from Ets-1 binding to AML1. AML1 binds to

the Ets domain of Ets-1 and autoinhibitory domains (NRBD and exon VII domain) (47).

We also show, as has previously been demonstrated in vitro, that MEF, a NERF-2

homologous protein, binds to AML1 in vivo. MEF has also been reported to interact with

AML1 through a region amino-terminal to the Ets domain, although this region was not

further defined (48). Therefore, it is likely that the basic domain D which is conserved

among E74 Ets family members is the region in which NERF-2, MEF and Elf-1 interact

with AML1. Thus, this AML1 interacting domain appears to represent a novel protein-


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

25

protein interaction domain and this finding suggests that AML1 can bind to different

members of the Ets family via different interaction domains.

Since various isoforms of NERF are expressed in B cells, NERF is likely to play a

role in B cell function or differentiation. The NERF-AML1 and BSAP-AML1

interactions and synergistic activations of the blk promoter support the notion that NERF,

AML1 and BSAP regulate blk gene expression. Since BSAP has been demonstrated to

interact with the Ets domain of several Ets factors, we are now also in the process of

evaluating whether NERF interacts with BSAP and forms a NERF-BSAP-AML1

complex that regulates blk gene expression. Blk is a B cell specific tyrosine kinase of the

Src family important for B-cell activation after cross-linking of antigens via the B cell

antigen receptor (BCR). In peripheral lymphoid tissues, cross-linking initiated signaling

activates B cells to enter the G1 phase of the cell cycle, which will direct B cells to

respond to proliferative signals (49). Subsequently, proliferating B cells differentiate into

antibody-producing plasma cells. Expression of constitutively active Blk(Y495F) in the B

lineage induces malignant transformation of early lymphoid progenitors in mice,

suggesting a role for Blk in the control of proliferation during B cell development (50).

Our results show that physical interaction of NERF-2 with AML1 synergistically

activates the blk promoter, whereas NERF-1a inhibits AML1 mediated transactivation.

Previously, we have demonstrated that all NERF isoforms bind with comparable affinity

to the same Ets sites in a variety of B cell-specific genes including blk, although only

NERF-2, but not NERF-1a and NERF-1b, function as transcriptional activators of B cell-

specific promoters (8). NERF-1a may act as a competitive inhibitor of endogenous

NERF-2 or possibly other Ets factors by replacing NERF-2 on the blk promoter and thus

inhibiting AML1 transactivation activity which might be NERF-2 dependent.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

26

Alternatively, NERF-1a may be an active repressor that interacts with a co-repressor and

actively inhibits AML1 mediated transactivation. We show here that the transactivation

domain of the NERF-2 does not overlap with the basic AML1 protein interaction domain,

but is located in the N-terminal 103 amino acids. This also explains why NERF-1a and -

1b which differ at their N-terminus from NERF-2 do no exhibit transactivation activity.

Indeed the amino-terminus of NERF-1a does not contain a transactivation domain as

shown by our GAL4 heterologous transactivation assay. Recently, a potent

transactivation domain of MEF, a homologous protein to NERF-2, has been mapped to

the N-terminal region encompassing amino acids 1-52 (51). There is a significant

sequence homology within the amino-terminal 103 amino acids between NERF-2, ELF-1

and MEF, particularly in domains A, B and C. These conserved domains contain many

acidic amino acids and our point mutations replacing acidic amino acids with alanine or

deleting acidic amino acids provide evidence that acidic residues are involved in

transactivation function. Acidic transactivation domains have been observed in many

transcription factors including other members of the Ets family and appear to interact

with several general transcription initiation factors (52-56).

The synergistic and repressor activity of NERF-2 and NERF-1a, respectively, in

conjunction with AML1 provides support for the notion that different NERF isoforms

and their regulation may modulate AML1 function both during normal B cell

development as well as in leukemic cells with a translocated AML1. Future studies will

focus on determining the effect of NERF isoforms on AML1 translocation proteins in

leukemic cells.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

27

ACKNOWLEDGEMENTSThis study was supported by National Institutes of Health Grant PO1/CA72009 to T. A.

L. and D.-E. Z. and RO1/CA76323 to T. A. L.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

28

REFERENCES

1. Hagman, J., and Grosschedl, R. (1994) Curr Opin Immunol 6, 222-230.

2. Li, R., Pei, H., and Watson, D. K. (2000) Oncogene 19, 6514-6523.

3. Sementchenko, V. I., and Watson, D. K. (2000) Oncogene 19, 6533-6548

4. Wasylyk, B., Hahn, S. L., and Giovane, A. (1993) Eur J Biohcem 211, 7-18

5. Seth, A., Ascione, R., Fisher, R. J., Mavrothalassitis, G. J., Bhat, N. K., and

Papas, T. S. (1992) Cell Growth Differ 3, 327-334

6. Janknecht, R., and Nordheim, A. (1993) Biochim Biophys Acta 1155, 346-356

7. Yordy, J. S., and Muise-Helmericks, R. C. (2000) Oncogene 19, 6503-6513.

8. Oettgen, P., Akbarali, Y., Boltax, J., Best, J., Kunsch, C., and Libermann, T. A.

(1996) Mol Cell Biol 16, 5091-5106.

9. Akbarali, Y., Oettgen, P., Boltax, J., and Libermann, T. A. (1996) J Biol Chem

271, 26007-26012.

10. Dube, A., Akbarali, Y., Sato, T. N., Libermann, T. A., and Oettgen, P. (1999)

Circ Res 84, 1177-1185.

11. Christensen, R. A., Fujikawa, K., Madore, R., Oettgen, P., and Varticovski, L.

(2002) J Cell Biochem 85, 505-515.

12. Wang, Q., Stacy, T., Binder, M., Marin-Padilla, M., Sharpe, A. H., and Speck, N.

A. (1996) Proc Natl Acad Sci U S A. 93, 3444-3449

13. Okuda, T., van Deursen, J., Hiebert, S. W., Grosveld, G., and R., D. J. (1996) Cell

84, 321-330

14. Cameron, S., Taylor, D. S., TePas, E. C., Speck, N. A., and Mathey-Prevot, B.

(1994) Blood 83, 2851-2859.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

29

15. Frank, R., Zhang, J., Uchida, H., Meyers, S., Hiebert, S. W., and Nimer, S. D.

(1995) Oncogene 11, 2667-2674.

16. Giese, K., Kingsley, C., Kirshner, J. R., and Grosschedl, R. (1995) Genes Dev 9,

995-1008.

17. Hsiang, Y. H., Spencer, D., Wang, S., Speck, N. A., and Raulet, D. H. (1993) J

Immunol 150, 3905-3916.

18. Hernandez-Munain, C., and Krangel, M. S. (1994) Mol Cell Biol 14, 473-483.

19. Nuchprayoon, I., Meyers, S., Scott, L. M., Suzow, J., Hiebert, S., and Friedman,

A. D. (1994) Mol Cell Biol 14, 5558-5568.

20. Suzow, J., and Friedman, A. D. (1993) Mol Cell Biol 13, 2141-2151.

21. Westendorf, J. J., Yamamoto, C. M., Lenny, N., Downing, J. R., Selsted, M. E.,

and Hiebert, S. W. (1998) Mol Cell Biol 18, 322-333.

22. Zhang, D. E., Fujioka, K., Hetherington, C. J., Shapiro, L. H., Chen, H. M., Look,

A. T., and Tenen, D. G. (1994) Mol Cell Biol 14, 8085-8095.

23. Ramsey, H., Zhang, D. E., Richkind, K., Burcoglu-O'Ral, A., and Hromas, R.

(2003) Leukemia 17, 1665-1666.

24. Hromas, R., Busse, T., Carroll, A., Mack, D., Shopnick, R., Zhang, D. E.,

Nakshatri, H., and Richkind, K. (2001) Blood 97, 2168-2170.

25. Gamou, T., Kitamura, E., Hosoda, F., Shimizu, K., Shinohara, K., Hayashi, Y.,

Nagase, T., Yokoyama, Y., and Ohki, M. (1998) Blood 91, 4028-4037.

26. Miyoshi, H., Shimizu, K., Kozu, T., Maseki, N., Kaneko, Y., and Ohki, M. (1991)

Proc Natl Acad Sci U S A 88, 10431-10434.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

30

27. Golub, T. R., Barker, G. F., Bohlander, S. K., Hiebert, S. W., Ward, D. C., Bray-

Ward, P., Morgan, E., Raimondi, S. C., Rowley, J. D., and Gilliland, D. G. (1995)

Proc Natl Acad Sci U S A 92, 4917-4921.

28. Look, A. T. (1997) Science 278, 1059-1064

29. Nucifora, G., and Rowley, J. D. (1995) Blood 86, 1-14.

30. Wang, S., Wang, Q., Crute, B. E., Melnikova, I. N., Keller, S. R., and Speck, N.

A. (1993) Mol Cell Biol 13, 3324-3339.

31. Ogawa, E., Maruyama, M., Kagoshima, H., Inuzuka, M., Lu, J., Satake, M.,

Shigesada, K., and Ito, Y. (1993) Proc Natl Acad Sci U S A 90, 6859-6863.

32. Levanon, D., Negreanu, V., Bernstein, Y., Bar-Am, I., Avivi, L., and Groner, Y.

(1994) Genomics 23, 425-432.

33. Kagoshima, H., Shigesada, K., Satake, M., Ito, Y., Miyoshi, H., Ohki, M.,

Pepling, M., and Gergen, P. (1993) Trends Genet 9, 338-341.

34. Ernst, P., Hahm, K., and Smale, S. T. (1993) Mol Cell Biol 13, 2982-2992.

35. Riley, L. K., Morrow, J. K., Danton, M. J., and Coleman, M. S. (1988) Proc Natl

Acad Sci U S A 85, 2489-2493.

36. Lin, J., and Justement, L. B. (1992) J Immunol 149, 1548-1555.

37. Fitzsimmons, D., Hodsdon, W., Wheat, W., Maira, S. M., Wasylyk, B., and

Hagman, J. (1996) Genes Dev 10, 2198-2211.

38. Leung, S., McCracken, S., Ghysdael, J., and Miyamoto, N. G. (1993) Oncogene

8, 989-997.

39. Burckhardt, A. I., Brunswick, M., Bolen, J. B., and Mond, J. J. (1991) Proc Natl

Acad Sci U S A 88, 7410-7414


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

31

40. Burg, D. L., Furlong, M. T., Harrison, M. L., and Geahlen, R. L. (1994) J Biol

Chem 269, 28136-28142.

41. Zwollo, P., and Desiderio, S. (1994) J Biol Chem 269, 15310-15317

42. Dymecki, S. M., Zwollo, P., Zeller, K., Kuhajda, F. P., and Desiderio, S. V.

(1992) J Biol Chem 267, 4815-4823

43. Libermann, T. A., Pan, Z., Akbarali, Y., Hetherington, C. J., Boltax, J., Yergeau,

D. A., and Zhang, D. E. (1999) J Biol Chem 274, 24671-24676.

44. Verger, A., and Duterque-Coquillaud, M. (2002) Bioessays 24, 362-370.

45. Zhang, D. E., Hetherington, C. J., Meyers, S., Rhoades, K. L., Larson, C. J., Chen,

H. M., Hiebert, S. W., and Tenen, D. G. (1996) Mol Cell Biol 16, 1231-1240.

46. Petrovick, M. S., Hiebert, S. W., Friedman, A. D., Hetherington, C. J., Tenen, D.

G., and Zhang, D. E. (1998) Mol Cell Biol 18, 3915-3925.

47. Kim, W. Y., Sieweke, M., Ogawa, E., Wee, H. J., Englmeier, U., Graf, T., and Ito,

Y. (1999) Embo J 18, 1609-1620.

48. Mao, S., Frank, R. C., Zhang, J., Miyazaki, Y., and Nimer, S. D. (1999) Mol Cell

Biol 19, 3635-3644.

49. Klaus, G. G., Bijsterbosch, M. K., O'Garra, A., Harnett, M. M., and Rigley, K. P.

(1987) Immunol Rev 99, 19-38.

50. Malek, S. N., Dordai, D. I., Reim, J., Dintzis, H., and Desiderio, S. (1998) Proc

Natl Acad Sci U S A 95, 7351-7356.

51. Suico, M. A., Koyanagi, T., Ise, S., Lu, Z., Hisatsune, A., Seki, Y., Shuto, T.,

Isohama, Y., Miyata, T., and Kai, H. (2002) Biochim Biophys Acta 1577, 113-

120.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

32

52. Carlsson, R., Persson, C., and Leanderson, T. (2003) Mol Immunol 39, 1035-

1043.

53. Fisher, R. C., Olson, M. C., Pongubala, J. M., Perkel, J. M., Atchison, M. L.,

Scott, E. W., and Simon, M. C. (1998) Mol Cell Biol 18, 4347-4357.

54. Laget, M. P., Defossez, P. A., Albagli, O., Baert, J. L., Dewitte, F., Stehelin, D.,

and de Launoit, Y. (1996) Oncogene 12, 1325-1336.

55. Klemsz, M. J., and Maki, R. A. (1996) Mol Cell Biol 16, 390-397.

56. Albagli, O., Soudant, N., Ferreira, E., Dhordain, P., Dewitte, F., Begue, A.,

Flourens, A., Stehelin, D., and Leprince, D. (1994) Oncogene 9, 3259-3271.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

33

FIGURE LEGENDS

Figure 1. NERF-2 protein binds to AML1 in vitro. GST pull-down experiments

showing that GST-NERF-2 proteins from bacteria interacts with in vitro translated S35

labelled AML1 proteins. AML1 protein was pulled down with GST-NERF-2 fusion

proteins but not with GST protein alone in vitro.

Figure 2. A basic domain of NERF-2 physically interacts with AML1. Panel A .

shows the domain structure of NERF-2 and the fragments used for GST pull-down

experiment shown in Panel B. Panel B shows GST pull-down experiment to demonstrate

AML1 interaction with various fragments of NERF-2 and NERF1a. This data shows that

as little as NERF-2 aa 121-203, which is just N-terminal to Ets domain, can bind

efficiently to AML1 proteins.

Figure 3. NERF-2 interacts with AML1 in vivo. A: Flag-tagged NERF-2 construct

was transiently co-transfected with myc-tagged AML1 into 293T cells. Cell lysates were

co-immunoprecipitated using M2 anti-Flag monoclonal antibody. A western blot was

performed using an anti-myc polyclonal antibody to detect AML1 protein. AML1 was

detected only when NERF-2-Flag and AML1-MYC were co-transfected into the cells,

not by vectors only, NERF-2 or AML1 only plus vector. B: Other Ets factors were also

cloned into the Flag vector and co-transfected with the AML1 construct. These Co-

IP/Western blot experiments showed that AML1 interacts with NERF-2 and MEF

proteins but not with PDEF and ESE1 proteins.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

34

Figure 4. Interaction of NERF-2 and AML1 synergistically activates a blk

promoter.

A. blk promoter luciferase assays were performed along with NERF1a, NERF-2, AML1

(+CBFβ) alone or with NERF-2+AML1(+CBFβ) or NERF1A+AML1(+CBFβ)

expression constructs in CV-1 cells. NERF-2+AML1(+CBFβ) co-expression showed the

luciferase activity more than additive increase combined NERF-2 alone and AML1 alone

activity, demonstrating that physical interaction of NERF-2 and AML1 has cooperative

function in activating the blk promoter. NERF1A interaction with AML1, however,

showed suppressive function of AML1-induced blk promoter activity. B. Two clones of

NERF-2 (del 108-180) deletion mutants fused to the Flag peptide in the Flag vector, Mut

#1 and Mut #17, were generated as described in Materials and Methods. These mutant

DNAs were co-transfected with AML1-myc into 293T cells and their proper expressions

were tested by Western blot analysis using anti-Myc or anti-Flag antibodies for AML1 or

NERF-2 mutants, respectively. As a control, wild type Flag-NERF-2 was also

transfected. C. Flag-tagged wild type and (del 108-180) mutant NERF-2 constructs were

transiently co-transfected with myc-tagged AML1 into 293T cells. Cell lysates were co-

immunoprecipitated using M2 anti-Flag monoclonal antibody. A western blot was

performed using an anti-myc polyclonal antibody to detect AML1 protein. AML1 was

detected only when NERF-2-Flag and AML1-MYC were co-transfected into the cells,

but not by NERF-2 (del 108-180) deletion mutants with AML1, vectors only, NERF-2 or

AML1 only plus vector. D. NERF-2 (del 108-180) deletion mutants were co-transfected

with or without AML1 (+CBFβ) along with the blk promoter construct into CV-1 cells.

The blk promoter activities were measured and presented as fold increase to the control

without AML1. The fold cooperation was calculated by dividing the activation of the


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

35

promoter in the presence of both factors by the expected additive result after background

subtraction. The mutants showed only an additive effect of AML1 and NERF-2

transactivation, while the wild typeNERF-2 showed strong transcriptional cooperativity.

E. The C terminus of AML1 is required for cooperativity with NERF-2. Transient

transfections were performed with NERF-2, CBFβ, and either AML1 or mutants of

AML1 in order to identify the region of AML1 required for functional interaction with

NERF-2. Mutants AML1(1-289), (1-351), and (1-381) are carboxy-terminal truncations

of AML1; the numbers represent the amino acids which are encoded. The results

represent the mean ± standard error. The fold cooperation was calculated by dividing the

activation of the promoter in the presence of both factors by the expected additive result

after background subtraction.

Figure 5. Transcriptional activation of the lyn promoter Ets site. A. CV-1 cells

were co-transfected with the indicated NERF1A, NERF-2 wild type, or NERF-2 deletion

mutants expression constructs and luciferase constructs containing two copies of the lyn

promoter Ets site. When the first 103 amino acids of NERF-2 were deleted the

transactivation activity dropped to a basal level similar to pCi vector transfection. B. lyn

promoter luciferase assay was done with 5-6 amino acid deletions in NERF-2 for Mut A,

B and C or with the mutation two glutamic acid residues to alanine for Mut D, as shown

in Figure 7B. Mutants A+B showed the most reduction in the lyn promoter activity,

suggesting that the Domain A and Domain B are largely responsible for the

transactivation activity induced by NERF-2.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

36

Figure 6. EMSA showing the DNA binding of NERF-2 full length or NERF-2

mutants to blk (A) or lyn (B) promoter Ets site. DNA binding analysis was done

using extracts from cells transfected with full length and mutant NERF-2 expression

vectors in an EMSA using oligonucleotides encompassing the lyn promoter Ets site. This

result shows that the NERF-2 mutants used for the experiments in Figure 5A and 5B does

not affect or has little effect on the DNA binding activity to the blk or lyn promoter.

Figure 7. A. Comparison of the amino acid sequence of NERF with that of ELF-1 and

MEF. The five major homologous regions, the domain A, B, C, D, and E (Ets domain)

are boxed. Shaded amino acids denote amino acid identity with NERF. B. the sequence

of the highly acidic transactivation domain and basic domain which is amino-terminal to

the Ets domain. The sequence shown here is aa 1-203 of NERF-2 and some of the

mutations made for the luciferase experiment shown in Figure 5 and of the Gal4 assay

shown in Figure 8.

Figure 8. Transactivation of N-terminal NERF-2 in Gal4 constructs. COS cells were

co-transfected with the indicated NERF-2 mutant Gal4 fusion constructs and Gal4luc

reporter construct (pGSE1bluc or pSGluc1b), which contains minimal promoter and three

Gal4 DNA binding elements. As shown here, NERF-2 aa 1-103 show still high

transactivation activity. Consistent with the promoter assay shown in Figure 5B, deletions

A, B, D or (A+B) showed significant loss of transactivation activity, while mutant C still

maintained transactivation activity.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Peter Oettgen, Dong-Er Zhang and Towia A. LibermannJe-Yoel Cho, Yasmin Akbarali, Luiz F. Zerbini, Xuesong Gu, Jay Boltax, Yihong Wang,

blk geneand mediate opposing effects on AML1 mediated transcription of the B cell-specific Isoforms of the Ets transcription factor NERF/ELF2 physically interact with AML1

published online February 17, 2004J. Biol. Chem.

10.1074/jbc.M309074200Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M309074200

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;M309074200v1&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/early/2004/02/17/jbc.M309074200.citation

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=early/2004/02/17/jbc&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/early/2004/02/17/jbc.M309074200.citation

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/

isoforms of the ets transcription factor nerf/elf2 ... · esei dnas were inserted into bamhi and...

Documents