MOLECULAR AND CELLULAR BIOLOGY, Mar. 2003, p. 1896–1909
0270-7306/03/$08.00?0 DOI: 10.1128/MCB.23.6.1896–1909.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Vol. 23, No. 6
Interleukin-3 Stimulation of mcl-1 Gene Transcription Involves
Activation of the PU.1 Transcription Factor through a p38
Mitogen-Activated Protein Kinase-Dependent Pathway
Ju-Ming Wang, Ming-Zong Lai, and Hsin-Fang Yang-Yen*
Institute of Molecular Biology, Academia Sinica, Taipei, Taiwan, Republic of China
Received 2 August 2002/Returned for modification 11 September 2002/Accepted 13 December 2002
We have previously demonstrated that the antiapoptotic gene mcl-1 is activated by interleukin-3 (IL-3) in
Ba/F3 pro-B cells through two promoter elements designated the CRE-2 and SIE motifs. While the CRE-2-
binding complex contains the CREB protein and is activated by IL-3 through the phosphatidylinositol 3-ki-
nase/Akt-dependent pathway, the identity and cytokine activation pathway of the SIE-binding complex remains
unclear. In this report, we demonstrated that PU.1 is one component of the SIE-binding complex. A chromatin
immunoprecipitation assay further confirmed that PU.1 binds to the mcl-1 promoter region containing the SIE
motif in vivo. While IL-3 stimulation does not significantly alter the SIE-binding activity of PU.1, it markedly
increases PU.1’s transactivation activity. The latter effect coincides with the increased phosphorylation of PU.1
following IL-3 activation of a p38 mitogen-activated protein kinase (p38MAPK)-dependent pathway. A serine-
to-alanine substitution at position 142 significantly weakens PU.1’s ability to be phosphorylated by the
p38MAPKimmunocomplex. Furthermore, this S142A mutant is impaired in the ability to be further stimulated
by IL-3 to transactivate the mcl-1 reporter through the SIE motif. Taken together, our results demonstrate that
IL-3 stimulation of mcl-1 gene transcription through the SIE motif involves phosphorylation of PU.1 at serine
142 by a p38MAPK-dependent pathway.
The mcl-1 gene was originally identified as an early gene
induced during differentiation of ML-1 myeloid leukemia cells
(22). Its product contains some structural motifs that charac-
terize it as a member of the Bcl-2 family of proteins. The
wild-type Mcl-1 protein has antiapoptotic activity (5, 38, 55),
whereas an alternatively spliced variant harboring only the
BH3 domain is a proapoptotic molecule (1, 3). Mcl-1 expres-
sion is induced by a number of growth factors or cytokines,
including interleukin-3 (IL-3), IL-5, IL-6, granulocyte-macro-
phage colony-stimulating factor (GM-CSF), vascular endothe-
lial growth factor, alpha interferon, and epidermal growth fac-
tor (5, 13, 15, 24). However, the signaling pathway activated by
the individual growth factor/cytokine receptor, which leads to
increased expression of the Mcl-1 protein, is largely uncharac-
We have previously demonstrated that mcl-1 is an immedi-
ate-early gene activated by the GM-CSF and IL-3 signaling
pathways and that the mcl-1 gene product is one component of
the viability response of these two cytokines (5). Cytokine
activation of the mcl-1 gene is regulated at the transcriptional
level and requires the membrane-distal region between amino
acids 573 and 755 of the common ? chain of the GM-CSF and
IL-3 receptors (5). Through cloning and extensive character-
ization of the mcl-1 promoter, we have found that the IL-3
inducibility of this gene in Ba/F3 pro-B cells is mediated mainly
through two upstream DNA motifs located at positions ?70
(the CRE-2 site) and ?87 (the SIE site) (49). Interestingly,
these two promoter elements can each confer IL-3 inducibility
on a heterologous promoter but work additively in mediating
IL-3 response via two different signaling pathways. While the
CRE-2-binding complex (which contains the CREB protein) is
induced and activated by IL-3 via activation of the phosphati-
dylinositol 3-kinase (PI3-K)/Akt-dependent pathway, the iden-
tity and the IL-3 activation pathway of the SIE-binding com-
plex remain to be determined (49).
PU.1 is a member of the Ets family of transcription factors,
and this family of proteins is characterized by the presence of
a DNA-binding domain that recognizes a core DNA element
containing the 5?-GGAA/T-3? motif (16, 28, 31). The expres-
sion of PU.1 is restricted specifically to cells of the hematopoi-
etic lineage. These include B cells, macrophages, mast cells,
neutrophils, and early erythroblasts (6, 10, 12, 20, 32, 37).
Knockout mouse studies have demonstrated that PU.1 defi-
ciency results in the absence of morphologically normal B cells
and macrophages, disrupted granulopoiesis, and aberrant T
lymphopoiesis (29, 41). This phenotype suggests that PU.1 may
directly or indirectly regulate some of the genes required for
the development of either lymphoid or myeloid lineages. Con-
sistent with this finding, many B-cell- and myeloid-specific
genes, including those encoding immunoglobulins, receptors,
and enzymes, have been reported to be directly regulated by
PU.1 or have a potential PU.1-binding site in their promoters
(7, 26, 53).
In this study, we explored the identity and the IL-3 activation
pathway of the transcription factor that binds to the SIE ele-
ment of the mcl-1 gene promoter. By expression library screen-
ing, oligonucleotide pulldown, gel shift, and chromatin im-
munoprecipitation assays, we found that the Ets family of
transcription factor PU.1 is one component of the SIE-binding
complex in IL-3-dependent Ba/F3 cells. While IL-3 treatment
of cells does not significantly alter the SIE-binding activity of
* Corresponding author. Mailing address: Institute of Molecular
Biology, Academia Sinica, 128 Yen-Jiou Yuan Rd., Section 2, Nan-
kang, Taipei 11529, Taiwan, Republic of China. Phone: 886-2-2789-
9228. Fax: 886-2-2782-6085. E-mail: firstname.lastname@example.org.
PU.1, it markedly stimulates the transactivation activity of
PU.1. The latter effect involves phosphorylation of PU.1 at
serine 142 following IL-3 stimulation through a p38 mitogen-
activated protein kinase (p38MAPK)-dependent pathway.
MATERIALS AND METHODS
Plasmid construction. The hemagglutinin (HA) epitope-tagged PU.1 expres-
sion vector (pcDNA3HA/PU.1) was constructed by reverse transcription-PCR
amplification of the total RNA isolated from Ba/F3 cells with the following
primers: sense, 5?-TGGAATTCTGTTACAGGCGTGCAAAATG-3?; antisense,
5?-ATGCTCGAGGATCAGTGGGGCGGGAGG-3?. The PCR product was
then restricted with EcoRI and XhoI and cloned into a pcDNA3 derivative,
pcDNA3-HA (H.-W. Peng and H.-F. Yang-Yen, unpublished results), that
would direct the synthesis of an HA-tagged protein in vivo by transfection into
mammalian cells or in vitro by the use of a transcription-translation-coupled
reticulocyte lysate system (Promega). The HA/PU.1dlN133 expression vector
(pcDNA3 HA/PU.1dlN133), which directs the synthesis of a mutant PU.1 pro-
tein without the N-terminal 133 amino acids, was constructed by PCR amplifi-
cation of the wild-type template with primers 5?-CGGGATCCGATGAGGAG
GAGGGTG-3? and 5?-ATGCTCGAGGATCAGTGGGGCGGGAGG-3?. The
resultant PCR products were then cloned into the pcDNA3-HA vector. pQE30/
PU.1 and pQE30/PU.1dlN133 are expression vectors that would direct the syn-
thesis of the histidine-tagged wild-type or N-terminally truncated PU.1 protein in
Escherichia coli and were constructed by inserting an appropriate DNA fragment
from the PU.1 cDNA into the pQE30 vector (Qiagen). All of the constructs
generated via cloning steps involving PCR were sequenced to confirm their
primary structures. pQE30/PU.1S142A, pQE30/PU.1S148A, and pQE30/
PU.1SS142/148AA are identical to pQE30/PU.1, except that the nucleotide se-
quence encoding the serine residue at position 142, 148, or both was mutated to
generate an alanine codon(s). These three constructs were generated by site-
directed mutagenesis of each individual region, and the mutated nucleotides
PU.1S148A, and pcDNA3HA/PU.1SS142/148AA were derived by subcloning the
cDNA inserts from pQE30/PU.1S142A, pQE30/PU.1S148A, and pQE30/
PU.1SS142/148AA, respectively, into the pcDNA3-HA vector. The dominant
negative mutant forms of p38? and p38? (i.e., p38?[AF] and p38?[AF], respec-
tively) were gifts of Jiahuai Han (14, 50, 54).
Cell culture. Ba/F3 cells (mouse bone marrow-derived, IL-3-dependent pro-B
cells) were maintained in RPMI 1640 medium supplemented with 10% fetal
bovine serum, 50 ?M ?-mercaptoethanol, 2 mM L-glutamine, 100 U of penicillin
G per ml, 100 ?g of streptomycin per ml, and 2% WEHI-3B conditioned medium
as a source of IL-3. Ba/F3-HAPU.1 is a Ba/F3 derivative stably overexpressing
the HA-tagged PU.1 protein. This Ba/F3 derivative was generated by electropo-
ration with the HA-tagged PU.1 expression vector (pcDNA3HA/PU.1) and
selected in growth medium supplemented with 500 ?g of G418 per ml. To
examine whether IL-3 stimulates phosphorylation of PU.1 in vivo, Ba/F3-
HAPU.1 cells (2 ? 107) were first deprived of IL-3 for 4 h and then cultured in
phosphate-free medium (Dulbecco modified Eagle medium without phosphate;
GIBCO-BRL) for 2 h more. Fifteen minutes prior to IL-3 stimulation, cells were
fed with 1 mCi of [32P]orthophosphate (Perkin-Elmer Life Sciences). Eighty
minutes after IL-3 treatment, cells were lysed in radioimmunoprecipitation assay
buffer (10 mM Tris-HCl [pH 7.5], 150 mM NaCl, 5 mM EDTA, 0.1% sodium
dodecyl sulfate [SDS], 1% sodium deoxycholate) containing 1 mM phenylmeth-
ylsulfonyl fluoride (PMSF), 1 ?g of aprotinin per ml, and 1 ?g of leupeptin per
ml and the cell lysates were immunoprecipitated with anti-HA antibody (Roche
Applied Sciences). The immunoprecipitated complexes were then resolved by
SDS-polyacrylamide gel electrophoresis (PAGE), and the specific bands were
revealed by autoradiography or subsequently detected by immunoblotting with
anti-HA antibody. For experiments with chemical inhibitors, the following con-
centrations were used: SB203580, 20 ?M; wortmannin, 0.1 ?M; anisomycin, 50
ng/ml. All inhibitors (purchased from Calbiochem) were added 30 min prior to
IL-3 treatment of cells.
Screening of SIE-binding proteins. A32P-labeled concatemer of the double-
stranded SIE oligonucleotide probe (sense strand sequence, 5?CTTTTACGGG
AAGTCC3?) was used to screen a mouse day 15 embryonic cDNA expression
library (Clontech). Briefly, the nitrocellulose filters were immersed in 10 mM
isopropyl-?-D-thiogalactopyranoside (IPTG) before they were placed onto the
phage plates. After 4 h of incubation at 37°C, these membranes were transferred
into SIE binding buffer (10 mM HEPES [pH 7.9], 75 mM NaCl, 1 mM dithio-
threitol, 1 mM EDTA, 10 mM MgCl2, 10% glycerol) containing 6 M guanidine-
HCl and the proteins on the membranes were denatured for 5 min. To allow the
denatured proteins to be renatured, the membranes were subsequently trans-
ferred into a series of binding buffers containing decreasing concentrations of
guanidine-HCl (5 min for each step). Following the renaturation steps, the
membrane filters were allowed to bind the SIE probe for 18 h at 4°C. After a few
washes in SIE binding buffer, positive clones were revealed by autoradiography.
After the tertiary screening, all positive clones were purified and their inserts
were PCR amplified and sequenced.
Oligonucleotide pulldown assay. Ba/F3 cells with or without IL-3 stimulation
were lysed in NP-40 lysis buffer (50 mM Tris-HCl [pH 8.0], 0.5% NP-40, 150 mM
NaCl, 0.1 mM EDTA, 10 mM NaF, 1 mM PMSF, 1 ?g of aprotinin per ml, 1 ?g
of leupeptin per ml, 1 mM Na3VO4, 10% glycerol, 1 mM dithiothreitol [DTT]),
and 200 ?g of lysates from each group was incubated with 1 ?g of biotinylated
SIE oligonucleotide dimers in the presence of 2 ?g of poly(dI-dC). After 1 h of
incubation at 4°C, 40 ?l of agarose-streptavidin was added to the reaction
mixture and the incubation was continued for 1 h more. The SIE-binding com-
plex was then precipitated by centrifugation and washed twice with NP-40 lysis
buffer and twice with SIE binding buffer before it was resolved by SDS-PAGE
and subsequently analyzed by immunoblotting with anti-PU.1 antibody (Santa
Reporter plasmids and luciferase assay. mcl-1 reporter plasmids p(?203/
?10)mcl-luc, ?203/?10dlC, and ?203/?10mS have been previously described
(49). The p(?203/?10)mcl-luc construct contains a luciferase reporter gene that
is driven by the mcl-1 gene promoter element between ?203 and ?10. Plasmids
?203/?10dlC and ?203/?10mS are identical to p(?203/?10)mcl-luc, except
that the CRE-2 site at position ?70 and the SIE site at position ?87 are mutated,
respectively. Plasmids pGL-8xSIE and pGL-2xCRE-2 were derived by inserting
eight or two copies of the oligonucleotide fragments containing the SIE and
CRE-2 sites, respectively, into the SmaI site of the pGL2-promoter vector (Pro-
mega). In both cases, insertion of an SIE or CRE-2 element conferred IL-3
inducibility on the downstream simian virus 40 minimal promoter (49; see Fig.
5B). The sense strand sequences of these two oligonucleotide fragments are as
follows: SIE, 5?CTTTTACGGGAAGTCC3?; CRE-2, 5?TCGCCTGCGTCAGC
ACG3?. To analyze the promoter activity of these reporter genes under various
conditions, these plasmids, together with an appropriate expression vector, were
transiently introduced into Ba/F3 cells by electroporation with a Bio-Rad Gene
Pulser II RF Module system as previously described (5). Electroporated cells
were seeded in growth medium with or without murine IL-3 (mIL-3) (R&D
Systems). Twelve hours after reseeding, cells were harvested and assayed for
luciferase activity. All luciferase activities were normalized to the amounts of
proteins present in the cell lysates and, unless specified otherwise, plotted as
activities relative to that obtained from cells cotransfected with a control vector
and stimulated with IL-3. The latter luciferase activity was assigned a value of
100%. To analyze the effects of various dominant negative mutant forms in the
reporter gene assays, electroporated cells were recovered in mIL-3-containing
medium for 12 h prior to being deprived of mIL-3 for 6 h and restimulated with
mIL-3 for 3 h more.
Gel shift assays. Gel shift assays were carried out essentially as previously
described (49). Briefly, the32P-labeled oligonucleotide probe (?0.2 to 0.5 ng)
containing the SIE site (the same oligonucleotides as described above for re-
porter gene construction) was incubated with 8 ?g of nuclear extracts or 1.5 ?l
of in vitro-translated PU.1 in SIE binding buffer containing 1.5 ?g of poly(dI-
dC). After 20 min of incubation at room temperature, the reaction mixtures were
resolved in a 5% native polyacrylamide gel (acrylamide/bisacrylamide ratio, 80:1)
at 4°C and the specific protein complexes were visualized by autoradiography.
For antibody supershifting experiments, 1 ?g of anti-PU.1 (Santa Cruz Biotech-
nology) or control rabbit immunoglobulin G (IgG) was included in the binding
reaction mixture. For competition experiments, a 100-fold molar excess of un-
labeled wild-type or mutant SIE (mSIE) oligonucleotides was included in the
binding reaction mixture. The sense strand sequence of an mSIE oligonucleotide
is as follows: 5?CTTTTAgGatccGTCC3? (mutated nucleotides are in lowercase).
Western blotting. Cells to be analyzed were lysed in a buffer containing 50 mM
HEPES (pH 7.4), 100 mM NaCl, 1 mM EGTA, 20 mM NaF, 20 mM Na4P2O7,
1 mM Na3VO4, 1 mM PMSF, 1 ?g of aprotinin per ml, 1 ?g of leupeptin per ml,
and 1% Triton X-100. Following lysis, the lysates were resolved on an SDS-
containing 10% polyacrylamide gel, transferred to polyvinylidene difluoride ny-
lon membrane (Millipore), and probed with antibodies specific to mouse Mcl-1
(raised in a rabbit with recombinant histidine-tagged mouse Mcl-1), ?-tubulin
(Amersham, Buckinghamshire, England), active p38 (phospho-p38/Thr180Tyr182),
or all forms of p38 (both from Cell Signaling Technology). The specific bands
were detected by horseradish peroxidase-conjugated goat antibody to rabbit IgG
and revealed by an ECL (enhanced chemiluminescence) Western blot system
(Amersham Pharmacia Biotech).
VOL. 23, 2003ROLES OF p38MAPKAND PU.1 IN Mcl-1 EXPRESSION 1897
ChIP assay. The chromatin immunoprecipitation (ChIP) assay was carried out
essentially as described by Saccani et al. (40), with a minor modification. Briefly,
Ba/F3 cells, with or without prior stimulation with IL-3, were treated with 1%
formaldehyde for 15 min. The cross-linked chromatin was then prepared and
sonicated to an average size of 300 to 400 bp prior to being immunoprecipitated
with antibody specific to PU.1, CREB, p53 (a gift of Yang-Sun Lin, Institute of
Biomedical Sciences, Academia Sinica), or control rabbit IgG at 4°C overnight.
After reversal of cross-linking, the immunoprecipitated chromatin was PCR
amplified with various sets of primers as indicated below. The amplified DNA
product was resolved by agarose gel electrophoresis, subsequently transferred
onto nylon membrane, and subjected to Southern blot analysis with a32P-labeled
mcl-1 promoter-specific probe spanning the region between ?175 and ?38. For
PCR amplification of specific regions (A to D; see Fig. 2) of the mcl-1 genomic
locus, the following sets of primers were used: region A (nucleotide [nt] ?1329
to nt ?1062), mMcl1-1329 (5?-CAGCTTGTGTCAAGTTGATATTAAGTCTA
ACC-3?) and mMcl1-1062 (5?-CCATTTATGGTGGCTAGGCCTATAGCTC-
3?); region B (nt ?519 to nt ?351), mMcl1-519 (5?CCAGGGTTTAACTC-
CCAGCACCCACC-3?) and mMcl1-351 (5?-AACTCTGCATACCCCAGGCT
GGTCC-3?); region C (nt ?175 to nt ?38), mMcl1-175 (5?-AAGCCGCGAGA
GCGCTCCGGCCGGAAG-3?) and mMcl1?38 (5?-ACGCCGCAGGCTGAG
GGGAAGGAGC-3?); region D (nt ?268 to nt ?502), mMcl1?268 (5?-AAAG
GCGGCTGCATAAGTCGCCCGGC-3?) and mMcl1?502 (5?-TCCTCTTCCT
CCTCGGGCGGCGGCGG-3?). The primers used to amplify a promoter region
of a negative control gene (E4BP4) were E4BP4?1 (5?-CAGAAAGGACCTC
CTCGTCCTACAGAC-3?) and E4BP4-360 (5?-TCTGCTGGACCACATAGTC
In vitro kinase assay. IL-3-stimulated Ba/F3 cells were lysed in p38 lysis buffer
(10 mM Tris-HCl [pH 7.5], 0.15 M NaCl, 2 mM EGTA, 10 mM NaF, 0.2% Triton
X-100, 50 mM ?-glycerophosphate, 2 mM Na3VO4, 1 mM DTT, 1 mM PMSF,
10 ?g of leupeptin per ml, 10 ?g of aprotinin per ml), and the cell lysates were
immunoprecipitated with a p38MAPKantibody (kindly provided by Jiahuai Han).
After three washes with p38 lysis buffer and two with kinase buffer (20 mM
HEPES [pH 7.5], 20 mM MnCl2, 20 mM MgCl2, 2 mM DTT, 0.1 mM Na3VO4,
10 mM NaF, 25 mM ?-glycerophosphate), the p38 immunocomplex was resus-
pended in 40 ?l of kinase buffer containing 10 ?Ci of [?-32P]ATP and 5 ?g of
wild-type or mutant PU.1 purified from bacteria (XL-1-blue) transformed with
pQE30/PU.1, pQE30/PU.1dlN133, pQE30/PU.1S142A, pQE30/PU.1S148A, or
pQE30/PU.1SS142/148AA. After 30 min of incubation at 30°C, the reaction
products were resolved by SDS-PAGE and transferred to polyvinylidene diflu-
oride membrane. Specific bands were first revealed by autoradiography and later
by staining of the membrane with Coomassie brilliant blue.
PU.1 binds to the SIE element of the mcl-1 promoter in
vitro. To identify the transcription factor that is involved in
IL-3 stimulation of murine mcl-1 gene transcription through
the SIE motif, a mouse day 15 embryonic cDNA expression
library was screened with a32P-labeled concatemer of double-
stranded SIE oligonucleotide (see Materials and Methods). Of
5 ? 105clones screened, 23 were positive and sequence anal-
ysis revealed that they represented clones derived from 11
distinct cDNAs. Among these positive clones, the one encod-
ing the Ets family of transcription factor PU.1 was first char-
acterized, as PU.1 cDNA was pulled out four times (all with
different sizes) in our library screen and Western analysis (data
not shown) revealed that PU.1 was expressed in our model cell
line, Ba/F3, for studying mcl-1 gene regulation by the IL-3
signaling pathway (49).
To investigate whether PU.1 indeed binds to the SIE ele-
ment of the mcl-1 gene promoter, the following experiments
were carried out. We first examined whether PU.1 prepared
from the in vitro translation system would recognize the SIE
oligonucleotide probe. As shown in Fig. 1A, the rabbit reticu-
locyte lysate programmed with a vector encoding either the
HA-tagged full-length PU.1 (HA/PU.1) protein or a mutant
protein retaining the DNA-binding domain (HA/PU.1dlN133)
gave rise to a specific protein-DNA complex that was not
observed with the reticulocyte lysate programmed with an
empty control vector (compare lanes 2 and 3 to lane 1). The
identity of the protein-DNA complex formed with HA/PU.1 in
this assay was further confirmed either by its ability to be
supershifted by a PU.1 antibody (lane 4) or by its ability to be
competed out by a 100-fold molar excess of unlabeled wild-
type SIE but not mSIE oligonucleotide (data not shown). We
next examined whether the SIE-binding complex previously
identified with the Ba/F3 cell lysates (49) contained the PU.1
protein. To address this issue, two different approaches were
taken. First, the oligonucleotide pulldown assay (see Materials
and Methods) was employed. In this experiment, the PU.1
protein was found to be present in the protein complex copre-
cipitated with the biotinylated SIE oligonucleotide as revealed
by immunoblotting analysis of this complex with antibody spe-
cific to PU.1 (Fig. 1B). The pulldown of PU.1 in this experi-
ment was specific to the SIE element, as it was specifically
inhibited by oligonucleotides containing the wild-type SIE site
but not by oligonucleotides containing the mSIE site (data not
shown). It is worth noting that the formation of the SIE-PU.1-
binding complex was independent of stimulation by IL-3 (com-
pare lanes 1, 2, and 3 in Fig. 1B). The second approach was to
employ the antibody supershift assay with the SIE oligonucle-
otide probe. Figure 1C shows that, in this approach, in addition
to the major complex (designated B1) identified in our previ-
ous report (49), a minor, lower-molecular-weight protein-
DNA complex (designated B2) was detected in an autoradio-
gram developed after a longer exposure time. Like the B1
complex, the B2 complex bound specifically to the SIE site, as
it was competed out by the wild-type SIE-containing oligonu-
cleotide but not by the mSIE-containing oligonucleotide (Fig.
1C, lanes 1 to 3). Unexpectedly, it was the B2 but not the B1
complex that was specifically supershifted by the PU.1 antibody
(compare lanes 4 and 5 in Fig. 1C). Similar to the results shown
in Fig. 1B, IL-3 did not significantly affect the formation of the
B2 complex (compare lanes 1 to 3 in Fig. 1D).
PU.1 binds to the mcl-1 promoter in vivo. We next examined
whether PU.1 bound to the mcl-1 gene promoter in vivo. To
address this issue, the ChIP assay was performed. For this
experiment, we designed five sets of primers (see Materials and
Methods) that would specifically amplify either a promoter
region of a negative control gene (?360 to ?1 of the E4BP4
gene) or some designated regions (A to D) of the mcl-1 gene
locus, as illustrated in Fig. 2A. The results shown in Fig. 2B
indicated that, while all of the primer sets generated a PCR
fragment of the predicted size, as indicated in Fig. 2A, from the
chromatin DNA prior to the immunoprecipitation step (lanes
2, 4, 6, 8, and 11, top), only region C, containing the SIE motif,
was specifically precipitated by the PU.1 antibody and subse-
quently PCR amplified (lane 7, top). The identity of the am-
plified C fragment was further confirmed by Southern blotting
analysis of the same gel shown in the upper part of Fig. 2B with
a genomic DNA probe spanning the region between ?175 and
?38 (Fig. 2B, bottom, lanes 6 and 7). Next, we examined
whether coprecipitation of the C fragment was indeed due to
specific binding of PU.1 to this region. As shown in Fig. 2C,
neither an isotype-matched rabbit IgG (lanes 2 and 3) nor an
antibody recognizing an irrelevant transcription factor (p53,
lanes 6 and 7) precipitated this C fragment. On the other hand,
1898WANG ET AL.MOL. CELL. BIOL.
an antibody recognizing the active form of the CREB protein
(phospho-CREB) specifically coprecipitated the same C frag-
ment in an IL-3-inducible manner (Fig. 2C, lanes 8 and 9). The
latter result was consistent with our previous finding that
CREB binds to the CRE-2 element right next to the SIE motif
of the mcl-1 gene promoter and its binding activity is stimu-
lated by IL-3 treatment of cells (49). Similar to the results
shown in Fig. 1 B and D, binding of PU.1 to the C fragment
was constitutive, i.e., independent of stimulation by IL-3 (com-
pare lanes 4 and 5 of Fig. 2C). Taken together, these results
indicate that PU.1 binds to the mcl-1 gene promoter in vivo.
PU.1 is involved in IL-3 stimulation of mcl-1 gene transcrip-
tion through binding to the SIE motif. Next, we examined
whether PU.1 plays a role in IL-3 stimulation of mcl-1 gene
transcription. To address this issue, the reporter gene assay
using the Ba/F3 cell system as previously described (49) was
performed. In this experiment, cells were cotransfected with
various mcl-1 reporter genes together with a control vector or
a vector expressing mutant PU.1 (HA/PU.1dlN133) that was
previously shown to have a dominant negative effect (43). As
previously noted (49), IL-3 stimulated (about three- to four-
fold) the luciferase activity of cells transfected with a luciferase
reporter driven by the mcl-1 promoter region between ?203
and ?10, i.e., the p(?203/?10)mcl-luc construct (see refer-
ence 49, and compare lanes 1 and 2 in Fig. 3). Mutation of the
CRE-2 or SIE motif resulted in a reporter that still responded
to IL-3 stimulation but to a lesser extent (about twofold) (see
reference 49, and compare lanes 4 and 5 and lanes 7 and 8 in
Fig. 3). Interestingly, coexpression of the dominant negative
mutant form of PU.1 (HA/PU.1dlN133) significantly attenu-
ated the IL-3 inducibility of the mcl-1 gene reporters contain-
ing the SIE motif, i.e., both the p(?203/?10)mcl-luc and
?203/?10dlC reporters (compare lanes 2 and 3 and lanes 8
and 9 in Fig. 3). In contrast, the IL-3 inducibility of the ?203/
?10mS reporter with the mSIE site was not affected by the
dominant negative mutant form of PU.1 (compare lanes 5 and
6 in Fig. 3). Taken together, these results suggest that PU.1 is
involved in IL-3 stimulation of mcl-1 gene transcription and
that its effect is mediated through binding to the SIE motif.
SIE-mediated IL-3 stimulation of mcl-1 transcription in-
FIG. 1. PU.1 is one component of the SIE-binding complex. (A) Gel shift assay with the32P-labeled SIE probe using in vitro-translated
HA/PU.1 (lane 2) or HA/PU.1dlN133 (lane 3). Lane 4, same as lane 2 except that the anti-PU.1 antibody was included in the assay. Lane 1 is the
result of a control experiment using the reticulocyte lysates programmed with an empty expression vector. SS1, the HA/PU.1-DNA complex
supershifted by the PU.1 antibody. (B) PU.1 is present in the complex pulled down by the SIE-containing oligonucleotide. The oligonucleotide
pulldown assay using lysates of Ba/F3 cells with or without prior stimulation with IL-3 was carried out as described in Materials and Methods. The
protein complex pulled down by this assay was analyzed by Western blotting with an antibody specific for PU.1. (C) Lane 1, same as lane 2 of panel
A, except that nuclear extracts from IL-3-stimulated Ba/F3 cells were used in the gel shift assay. Lanes 2 and 3 are results of the same experiment
with the inclusion of a 100-fold molar excess of unlabeled wild-type SIE- and mSIE (see Materials and Methods for the bases changed in the
nucleotide sequence)-containing oligonucleotides, respectively, as a cold competitor (Comp.). SS denotes the B2 complex supershifted by the PU.1
antibody. (D) Same as lane 1 of panel C, except that the assay was carried out with nuclear extracts from Ba/F3 cells with or without IL-3
VOL. 23, 2003ROLES OF p38MAPKAND PU.1 IN Mcl-1 EXPRESSION1899
volves activation of p38MAPK. We have previously demon-
strated that IL-3 activation of mcl-1 gene transcription through
the SIE motif occurs via an unknown but PI3-K/Akt-indepen-
dent pathway (49). To determine which signaling pathway is
involved in this regulatory process, we first tested whether any
known kinase inhibitors would inhibit the IL-3 inducibility of
an SIE reporter configured in a heterologous promoter (pGL-
8xSIE). Of the various chemical inhibitors tested, SB203580,
which specifically inhibits the ? and ? isoforms of p38MAPK,
manifested the best inhibitory effect (data not shown). We next
examined whether SB203580 would inhibit IL-3 stimulation of
mcl-1 expression in vivo. To address this issue, Northern blot-
ting analysis was carried out. The results shown in Fig. 4A
indicated that, like the PI3-K inhibitor wortmannin (49),
SB203580 also inhibited IL-3 stimulation of mcl-1 mRNA ex-
pression (compare lanes 3 and 4 to lane 2). While either in-
hibitor had an approximately 50% inhibitory effect, both in-
hibitors, when added together to cells, nearly completely
blocked IL-3 stimulation of mcl-1 expression. A similar inhib-
itory effect was also observed at the protein level (Fig. 4B). As
p38MAPKwas indeed activated by IL-3 in Ba/F3 cells, as was
evident from the appearance of a phosphorylated form (at
positions Thr-180 and Tyr-182) of p38 following IL-3 treat-
ment (Fig. 5A), we next examined whether the IL-3 inducibility
of the SIE reporter (pGL-8XSIE) could be blocked by a dom-
inant negative mutant form of p38MAPK. In Ba/F3 cells, the ?
isoform of p38MAPK(p38?) is expressed in a much larger
quantity than the ? isoform (p38?) (?/? ratio, ?20:1; data not
shown). We therefore first tested the blocking ability of the
dominant negative mutant form of p38? (DN-p38?). The re-
sults shown in Fig. 5B indicated that coexpression of DN-p38?,
but not the dominant negative mutant forms of other irrelevant
molecules, as indicated in the figure, attenuated IL-3’s stimu-
latory effect on SIE reporter activity (compare lanes 2 and 3).
This inhibitory effect was specifically mediated through the SIE
motif, as under the same conditions, the IL-3 inducibility of the
FIG. 2. PU.1 binds to the mcl-1 gene promoter in vivo. (A) Sche-
matic representation of the mcl-1 genomic locus spanning the pro-
moter region. Fragments A to D (sizes are indicated in base pairs) are
those predicted to be generated by a specified pair of primers as
described in Materials and Methods. (B) ChIP analysis of PU.1 bind-
ing to the mcl-1 gene locus. Formaldehyde-cross-linked chromatin was
immunoprecipitated with PU.1 antibody and processed for PCR am-
plification as described in Materials and Methods, by using primer sets
that would specifically amplify fragments A to D, as indicated in panel
A. As a positive control, PCR amplification was also carried out with
input DNA from chromatin prior to the immunoprecipitation (I) step
(lanes 2, 4, 6, 8, and 11). Serving as a negative control, a primer set that
would specifically amplify the promoter region of the E4BP4 gene was
included in the assay mixture (lanes 10 and 11). At the bottom is the
result of a Southern blot analysis of the same gel with a probe spanning
the mcl-1 promoter region between ?175 and ?38. All of the chro-
matin used in this assay was isolated from Ba/F3 cells stimulated with
IL-3. (C) ChIP analysis similar to that described for lane 7 of panel B,
except that chromatin was isolated from cells with (lanes 3, 5, 7, and 9)
or without (even-numbered lanes) IL-3 treatments and the immuno-
precipitation (IP) step was carried out with various antibodies (Ab), as
indicated. At the bottom is the result of a Southern blot analysis of the
same gel with the same probe as described for panel B. M stands for
DNA size markers (100-bp ladders).
FIG. 3. Dominant negative mutant form of PU.1 attenuates the
IL-3 inducibility of mcl-1 reporters with an intact SIE motif. Ba/F3
cells were transiently transfected with reporter genes as indicated
along with a control or the HA/PU.1dl133 (DN-PU.1) expression vec-
tor. After transfection, cells were cultivated in medium with or without
IL-3 for 12 h before cell lysates were prepared and analyzed for
luciferase activity. The results shown are averages of three indepen-
dent transfection experiments done in duplicate. For experiments with
each indicated reporter, the luciferase activity of that particular re-
porter in cells cotransfected with a control vector and with IL-3 stim-
ulation was assigned a value of 100% and all other results obtained
with cells transfected with the same reporter were normalized to this
activity. The differences between the results shown in lanes 2 and 3 and
those shown in lanes 8 and 9 are statistically significant (*, P ? 0.001;
**, P ? 0.002).
1900 WANG ET AL.MOL. CELL. BIOL.
CRE-2 reporter (pGL-2XCRE-2) was not affected by DN-
p38? (compare lanes 9 and 10 in Fig. 5B).
Given the fact that both PU.1 and p38? play a role in IL-3
activation of SIE-containing reporters, we next examined
whether these two molecules function in the same pathway to
activate transcription of the mcl-1 gene. To address this issue,
a transfection experiment identical to that shown in Fig. 3,
using the p(?203/?10)mcl-luc reporter but with various com-
binations of dominant negative mutant forms, was carried out.
The p(?203/?10)mcl-luc reporter was first tested in this assay
because it contains both the CRE-2 and SIE motifs and the
IL-3 inducibility through these two elements is mediated via
PI3-K (49)- and PU.1 (Fig. 3)-dependent pathways, respec-
tively. Figure 6A shows that IL-3-activated p(?203/?10)mcl-
luc reporter activity was partially inhibited by coexpression of
DN-p38?, DN-PU.1, or a dominant negative mutant form of
PI3-K (DN-p85) alone (lanes 2 to 5). While DN-p38? and
DN-PU.1 each synergized with DN-p85 to inhibit IL-3 stimu-
lation of this reporter activity (compare lanes 5 to 7), no syn-
ergism was observed when DN-p38? and DN-PU.1 were co-
expressed with the reporter gene (compare lanes 3, 4, and 8).
The lack of a synergistic effect between DN-p38? and DN-
PU.1 was also evident in an experiment that was identical
except for the use of the pGL-8XSIE reporter (Fig. 6B, lanes
3 to 5). Taken together, these results suggested that both p38?
and PU.1 function in the same pathway to mediate IL-3 acti-
vation of mcl-1 gene transcription through the SIE motif. Next,
we examined whether activation of the p38MAPKpathway is
important to the cell survival activity of IL-3. As shown in Fig.
6C, like the case observed with the PI3-K inhibitor wortmannin
(49), blocking of the p38 pathway by treatment of cells with
SB203580 also induced apoptosis of cells cultivated in IL-3-
containing medium. Furthermore, SB203580 worked additively
with wortmannin to inhibit the survival activity of IL-3. This
result, together with that illustrated in Fig. 4, suggests that the
p38MAPKand PI3-K pathways both contribute to the regula-
tion of Mcl-1 expression, as well as to the survival activity of
IL-3-induced phosphorylation of PU.1 is mediated via a
p38MAPK-dependent pathway. PU.1 is a phosphoprotein, and
phosphorylation at serine 148 is required for its interaction
with a B-cell-restricted transcription factor, PIP (NF-EM5),
and activates the immunoglobulin ? 3? enhancer (35). In the
context of IL-3-regulated mcl-1 gene transcription, PU.1 plays
a role in this process, yet its DNA-binding activity is not sig-
nificantly influenced following treatment of cells with IL-3
(Fig. 1 and 2). These results suggest that PU.1 may undergo an
IL-3-dependent posttranslational modification, e.g., phosphor-
ylation, that then directly or indirectly activates mcl-1 gene
transcription (see Discussion). To address this possibility, we
next examined whether IL-3 induces phosphorylation of PU.1
in vivo. For this experiment, Ba/F3 cells stably overexpressing
HA-tagged PU.1 were stimulated with IL-3 in the presence of
[32P]orthophosphate and the cell lysates were analyzed as de-
scribed in Materials and Methods. Figure 7A shows that fol-
lowing IL-3 treatment, HA-tagged PU.1 was indeed phosphor-
ylated, as revealed by the detection of a32P-labeled band with
an approximate size of 38 kDa in the anti-HA immunoprecipi-
tate (lane 2, top) that could be further recognized by an an-
ti-HA antibody in the subsequent immunoblotting analysis
(bottom). Furthermore, IL-3-induced phosphorylation of PU.1
was inhibited by pretreatment of cells with SB203580 (compare
lanes 2 and 3 in Fig. 7A), suggesting that this phosphorylation
event is mediated through a p38MAPK-dependent pathway.
Next, we examined whether PU.1 could be phosphorylated in
vitro by p38MAPKimmunoprecipitated from cells treated with
IL-3 or anisomycin (a known p38MAPKactivator). Figure 7B
FIG. 4. p38MAPKinhibitor (SB203580) attenuates IL-3 stimulation
of mcl-1 mRNA and protein expression. (A) Ba/F3 cells deprived of
IL-3 were pretreated with various inhibitors, as indicated, for 30 min
prior to stimulation with IL-3 for 1 h. After IL-3 stimulation, total
RNA was isolated from these cells and analyzed by Northern blotting
with a probe specific for detection of the murine mcl-1 mRNA. The
same blot was later stripped and hybridized with a glyceraldehyde-3-
phosphophate dehydrogenase (G3PDH)-specific probe. At the bottom
is a quantitative analysis of the mRNA levels shown at the top. The
amount of mcl-1 mRNA in each group was normalized to that of
G3PDH mRNA and plotted as a ratio relative to results obtained with
cells deprived of IL-3 (lane 1). (B) Same conditions as in panel A, ex-
cept that cells were lysed and 100 ?g of protein lysates was analyzed by
immunoblotting with antibodies specific to mouse Mcl-1 and ?-tubulin,
respectively. The asterisk points to an unknown protein that was also
recognized by the mouse Mcl-1 antibody. DMSO, dimethyl sulfoxide.
VOL. 23, 2003ROLES OF p38MAPKAND PU.1 IN Mcl-1 EXPRESSION1901
shows that this was indeed the case. The p38MAPKimmuno-
complex purified in both cases phosphorylated the bacterially
produced His-tagged PU.1 (His/PU.1) protein (lanes 2 and 4).
Furthermore, the in vitro phosphorylation activity of the
p38MAPKimmunocomplex was indeed p38MAPKdependent, as
the immunocomplex isolated from cells pretreated with
SB203580 was devoid of this activity (lane 3). These results,
together with those shown in Fig. 7A, suggest that IL-3 stim-
ulates PU.1 phosphorylation through a p38MAPK-dependent
We next examined whether phosphorylation would affect the
ability of PU.1 to mediate the IL-3 inducibility of the mcl-1
gene. The p38MAPK-dependent phosphorylation site of PU.1
was first mapped. We noticed that the PU.1 mutant without the
N-terminal 133 amino acids (His/PU.1dlN133) was still phos-
phorylated in vitro by the p38MAPKimmunocomplex (Fig. 7C).
We therefore searched the remaining C-terminal region for
motifs like Ser-Pro, Thr-Pro, and Ser-Asp that many p38MAPK
downstream targets (direct or indirect) display (17, 45, 48, 50a,
54). Two candidate sites, one at serine 142 and the other at
serine 148, were identified, and each (or both) was mutated to
an alanine residue by site-directed mutagenesis. Figure 7D
shows that a serine-to-alanine mutation at position 148
(S148A) did not significantly affect PU.1’s ability to be phos-
phorylated by the p38MAPKimmunocomplex (lane 3), whereas
a mutation at position 142 (S142A) markedly attenuated PU.1
phosphorylation in the same in vitro kinase assay (lanes 2 and
4). These results suggest that serine 142 is one site to be
phosphorylated in vitro by the p38MAPKimmunocomplex.
To examine whether p38MAPK-dependent phosphorylation
at serine 142 is important for PU.1 to mediate the IL-3 re-
sponse of the SIE motif, an assay that could specifically address
this issue was first used. In the transient transfection experi-
ments shown in Fig. 8A, we noticed that without IL-3 stimu-
lation, overexpression of PU.1 could slightly transactivate both
p(?203/?10)mcl-luc and ?203/?10dlC (compare lanes 1 and
3 and lanes 5 and 7) but not the ?203/?10mS reporter (com-
pare lanes 9 and 11). With IL-3 stimulation, although the
reporter activities of all three constructs were enhanced as
previously noted (reference 49; compare lanes 1 and 2, lanes 5
and 6, and lanes 9 and 10 in Fig. 8A), a further increase in
reporter activity due to overexpression of PU.1 was observed
again only in cells transfected with p(?203/?10)mcl-luc or
?203/?10dlC (compare lanes 2 and 4 and lanes 6 and 8) but
not in cells transfected with the ?203/?10mS reporter (com-
pare lanes 10 and 12). These results suggest that IL-3-en-
hanced transactivation activity of PU.1 on mcl-1 reporters re-
quires the presence of the SIE motif. We next examined
whether IL-3-enhanced transactivation activity of PU.1 is de-
pendent on activation of p38MAPK. To address this issue, the
same transfection experiment as described in the middle part
of Fig. 8A, except with the inclusion of a DN-p38 expression
vector, was carried out. In this experiment, the ?203/?10dlC
reporter was used because it could eliminate the stimulation
FIG. 5. Dominant negative mutant form of p38? attenuates the IL-3 inducibility of the SIE but not the CRE-2 reporter. (A) p38MAPKis
activated by IL-3. Cell lysates from Ba/F3 cells stimulated with IL-3 for various times were analyzed by Western blotting with an antibody
recognizing the active form (P-p38) (top) or all forms (bottom) of p38MAPK. (B) Ba/F3 cells transfected with the indicated reporter plasmid along
with a control vector (lanes 1, 2, 8, and 9) or a vector expressing the dominant negative mutant form of p38? (lanes 3 and 10) or other proteins,
as indicated (lanes 4 to 7), were stimulated with or without IL-3 for 12 h before the cell lysates were prepared and analyzed for luciferase activity.
The results shown are averages of three independent experiments done in duplicate and are plotted as described in the legend to Fig. 3. The
difference between the results shown in lanes 2 and 3 is statistically significant (*, P ? 0.001).
1902 WANG ET AL.MOL. CELL. BIOL.
effect of IL-3 through the CRE-2 site via the PI3-K/Akt-de-
pendent pathway (49). Figure 8B shows that coexpression of
DN-p38? but not DN-p38? or DN-p85 significantly attenuated
the stimulation effect of IL-3 on the transactivation activity of
PU.1 (compare lanes 4 to 7 on the left side of Fig. 8B and lanes
8 to 11 on the right side). These results suggest that IL-3-
enhanced transactivation activity of PU.1 is mediated through
a p38?-dependent pathway.
FIG. 6. (A and B) p38? and PU.1 work in the same pathway to regulate the IL-3 inducibility of SIE-containing reporters. (A) Ba/F3 cells
transfected with the p(?203/?10)mcl-luc reporter along with a control vector or vectors expressing DN-PU.1, DN-p38?, and DN-p85 or a
combination of these expression vectors were left untreated or stimulated with IL-3 for 12 h before cell lysates were prepared and analyzed for
luciferase activity. The data shown here are representative results of three to five independent experiments performed in duplicate. Luciferase
activity was plotted as described in legend to Fig. 3. The differences between the results shown in lanes 3 and 6 and between those shown in lanes
4 and 7 are statistically significant. (*, P ? 0.0001; **, P ? 0.003). (B) Same conditions as in lanes 1 to 4 and 8 of panel A, except that the
pGL-8XSIE reporter was used in the assay. (C) The p38MAPKpathway is involved in the viability response of IL-3 in Ba/F3 cells. Ba/F3 cells
cultivated in growth medium containing IL-3 were treated with various chemical inhibitors as indicated. Fifteen hours after each treatment, cells
were harvested and the percentage of apoptotic cells in each group was quantified by flow cytometric analysis of cells with a sub-G1DNA content.
The data shown are representative results of two independent experiments performed in duplicate. DMSO, dimethyl sulfoxide.
VOL. 23, 2003 ROLES OF p38MAPKAND PU.1 IN Mcl-1 EXPRESSION1903
We next employed a type of assay similar to that shown in
Fig. 8 to determine whether phosphorylation at serine 142 is
responsible for the IL-3-enhanced transcriptional activity of
PU.1 on SIE-containing reporters. For this experiment, we first
examined whether mutation of ser-142, ser-148, or both resi-
dues would affect their transcriptional activities. Figure 9
shows that none of these mutations had a significant effect on
their abilities to transactivate the SIE-containing reporter
(compare panels A and B). However, the transactivation ac-
tivity of the S142A or the SS142/148AA double mutant form
on the SIE-containing reporters, i.e., p(?203/?10)mcl-luc
(panel C) and ?203/?10dlC (panel D), was enhanced by IL-3
significantly less than that of the wild-type or S148A mutant
form. The nearly equal effects of all four proteins on the
?203/?10mS reporter (panel E) further confirmed the speci-
ficity of this type of experiment. Taken together, these results
suggest that the p38MAPK-dependent phosphorylation at serine
142 of PU.1 contributes, at least partially, to IL-3-stimulated
mcl-1 promoter activity through the SIE motif.
IL-3 activation of mcl-1 gene transcription is mediated
through two promoter elements designated the CRE-2 and
SIE motifs. In our previous reports, we stated that the CRE-
2-binding complex contains the CREB protein and is activated
by IL-3 through a PI3-K/Akt-dependent pathway (49). In the
present study, we demonstrated that the Ets family of tran-
scription factor PU.1 is one component of the SIE-binding
complex. Unlike the case with the CREB protein, which binds
FIG. 7. IL-3-induced phosphorylation of PU.1 occurs via a p38MAPK-dependent pathway. (A) In vivo phosphorylation of PU.1. Ba/F3-HAPU.1
cells with (lane 3) or without (lane 2) prior treatment with SB203580 were stimulated with IL-3 in the presence of [32P]orthophosphate as described
in Materials and Methods. After labeling, the HA-PU.1 protein was immunoprecipitated from cell lysates by using the anti-HA antibody, resolved
by SDS-PAGE, and subsequently revealed by autoradiography (top) or Western blotting with an anti-HA antibody (bottom). DMSO, dimethyl
sulfoxide. (B) In vitro phosphorylation of PU.1 by the p38MAPKimmunocomplex. The bacterium-produced His/PU.1 protein was phosphorylated
in vitro by the p38MAPKimmunocomplex as described in Materials and Methods. Lanes 1 and 2 are results of experiments with p38MAPK
immunocomplexes isolated from cells without and with IL-3 treatment, respectively. Lane 3, same as lane 2, except that the p38MAPKcomplex was
prepared from cells pretreated with SB203580 prior to IL-3 stimulation. Lane 4, same as lane 1, except that the p38MAPKcomplex was isolated from
cells pretreated with anisomycin. (C) Same conditions as described for panel B, except that bacterium-produced His/PU.1dlN133 was used as the
substrate. (D) Same conditions as described for lane 2 of panel B, except that various bacterium-produced PU.1 mutant forms, as indicated, were
used as the substrate.
1904 WANG ET AL.MOL. CELL. BIOL.
to the CRE-2 motif in an IL-3-inducible manner, PU.1 binds to
the SIE element constitutively but its transactivation activity is
increased upon IL-3 stimulation of cells. Furthermore, we
showed that IL-3 stimulation of mcl-1 gene transcription
through the SIE motif involves phosphorylation of PU.1 at
serine 142 by a p38MAPK-dependent pathway. These results,
together with those in our previous report (49), indicate that
the PI3-K/Akt/CREB and p38MAPK/PU.1 pathways cross talk
at the mcl-1 gene locus (Fig. 10). Considering the fact that the
core sequences of the CRE-2 and SIE motifs are only sepa-
rated by 9 bp, there may be some protein-protein interactions
between CREB- and PU.1-containing complexes that, via an
unknown mechanism, stimulate transcription of the mcl-1 gene.
More experiments are required to investigate this possibility.
FIG. 8. IL-3 stimulates the transactivation activity of PU.1 on SIE-containing reporters via a p38?-dependent pathway. (A) Ba/F3 cells
transfected with various mcl-1 reporter genes along with a control or PU.1 expression vector, as indicated, were stimulated with or without IL-3
for 12 h before the cell lysates were prepared and analyzed for luciferase activity. The data shown here are representative results of three or four
independent experiments performed in duplicate. (B) Same conditions as in lanes 5 to 8 of panel A, except with the inclusion of vectors expressing
DN-p38?, DN-p38?, and DN-p85 (lanes 5 to 7, respectively) during the transfection step. The data shown are means ? standard deviations of
duplicates from an experiment that was repeated two to four times with similar results. Lanes 8 to 11 illustrate the average normalized results of
four independent transfection experiments as described for lanes 4 to 7. For this normalization method, the percent increase in the transactivation
activity of PU.1 caused by IL-3 treatment was assigned a value of 100 (i.e., the ratio of activity shown in lane 4 over that shown in lane 3). The
activities shown in lanes 5 to 7 were each transformed by the same method, and the resultant quotient was normalized to that obtained from the
results shown in lanes 3 and 4. The result shown in lane 9 is statistically significantly different from that shown in lane 8 (*, P ? 0.002).
VOL. 23, 2003 ROLES OF p38MAPKAND PU.1 IN Mcl-1 EXPRESSION 1905
PU.1 is a hematopoietic, lineage-specific transcription factor
with some characterized functional domains. These include
multiple N-terminal acidic and glutamine-rich transactivation
domains (11, 19, 20, 21, 43), a central proline-, glutamic-acid-,
serine-, and threonine-rich domain that is important for pro-
tein-protein interactions (34, 35), and a C-terminal Ets DNA-
binding domain (19, 20). The activity of PU.1 is primarily
regulated posttranscriptionally by phosphorylation (reviewed
in reference 26). While phosphorylation of two serine residues
(positions 41 and 45) located at one of the acidic transactiva-
tion domains is necessary for macrophage colony-stimulating
factor-dependent proliferation of bone marrow macrophages
(4), a serine phosphorylation at position 148 in the proline-,
glutamic-acid-, serine-, and threonine-rich domain mediates
the interaction between PU.1 and the B-cell enhancer factor
(NF-EM5). The latter interaction is required for optimal trans-
activation of the 3? enhancer elements located in the immu-
noglobulin ? and ? light-chain genes (35). The serine 148
phosphorylation was also found to be required for the lipo-
polysaccharide-induced transactivation function of PU.1 (27).
In contrast, in this study, we demonstrated that serine phos-
phorylation of PU.1 at position 142 instead of position 148
mediates, at least partially, the SIE-dependent IL-3 stimula-
tion of mcl-1 reporter activity. While casein kinase II has been
suggested to be responsible for phosphorylation of PU.1 at
serine 148 (27, 35), our data indicate that the serine 142 phos-
FIG. 9. Mutation of serine 142 to alanine attenuates IL-3-enhanced transactivation activity of PU.1. (A and B) The S142A, S148A, and
SS142/148AA mutant forms all have transactivation activities similar to that of the wild-type (Wt or wt) protein on the SIE-containing reporter.
Reporter gene assays were carried out with cells transfected with the p(?203/?10)mcl-luc (A) or ?203/?10mS (B) reporter along with a pcDNA3
control or a PU.1 (wild-type or mutant)-expressing vector, as indicated. Both transfection assays were performed in the absence of IL-3. The
luciferase activities obtained from cells transfected with the control vector were assigned a value of 100%. (C to E) Reporter gene assays carried
out with cells transfected with the indicated reporter gene along with a wild-type or mutant PU.1 expression vector, as indicated. After transfection,
cells were split into two groups; one remained untreated, and the other was stimulated with IL-3. Twelve hours after transfection, cells were lysed
and analyzed for luciferase activity. The data shown here are normalized results to illustrate the average increase in the transactivation activity of
wild-type or mutant PU.1 following IL-3 treatment. The normalization method used is described in the legend to Fig. 8B. The differences between
the results shown in lanes 1 and 2 and those between the results shown in lanes 1 and 4 of panels C and D are statistically significant (*, P ? 0.001;
**, P ? 0.01, n ? 6).
1906 WANG ET AL.MOL. CELL. BIOL.
phorylation is induced via activation of a p38MAPK-dependent
pathway. The IL-3-induced in vivo phosphorylation of PU.1 is
inhibited by SB203580, suggesting that p38?, p38?, or both
mediate this phosphorylation event. However, the fact that
IL-3-stimulated transactivation activity of PU.1 is inhibited by
the dominant negative mutant form of p38? but not that of
p38? (Fig. 8B) suggests that serine 142 phosphorylation is
induced via a p38?-dependent pathway. On the other hand,
while our results presented in Fig. 7 cannot distinguish whether
PU.1 is directly phosphorylated by p38MAPKor by another
associated kinase in the p38MAPKimmunocomplex, the lack of
phosphorylation of PU.1 at Ser-142 in an in vitro kinase assay
using bacterium-produced p38? (data not shown) suggests that
PU.1 is an indirect substrate of p38?. More experiments are
required to reveal the identity of the putative p38?-associated
kinase that catalyzes the phosphorylation of PU.1 at Ser-142 in
response to IL-3 stimulation.
PU.1 binds to a cis element with a 5?-GGAA/T-3? core
sequence. Its binding to some gene promoters is constitutive
(44, 46), whereas in some other cases, its binding activity is
stimulated by extracellular stimuli or cytokines (18, 42). PU.1
binds to the SIE motif of the mcl-1 promoter irrespective of
IL-3 stimulation, and IL-3 stimulates its phosphorylation-and-
transactivation function. These results suggest that IL-3 stim-
ulates the phosphorylation of PU.1 at serine 142, which either
directly increases the transactivation activity of PU.1 or in-
creases PU.1’s ability to recruit another transcriptional activa-
tor and activates mcl-1 gene transcription in an indirect man-
ner. PU.1 was shown to interact with many transcription
factors or coactivators, including NF-EM5, Ets-1, NF-IL-6,
HMG-I(Y), c-Jun, IRF-1, ICSBP, AML1, and CBP (2, 8, 25,
33, 35, 36, 51, 52). It remains to be determined what other
factors are associated with PU.1 and mediate the SIE-depen-
dent IL-3 stimulation of mcl-1 gene transcription. In the gel
shift assay with the SIE probe, in addition to the B2 complex
that contains PU.1, another specific B1 complex was consis-
tently detected in the Ba/F3 cell extracts (Fig. 1). The forma-
tion of this B1 complex, like that of B2, is not influenced by the
presence or absence of IL-3. It is not clear whether the B1
complex plays any role in the SIE-mediated IL-3 induction of
mcl-1 gene transcription. If it does play a role in this process,
the inability of the PU.1 antibody to supershift this complex
suggests either that B1 does not contain PU.1 or that the PU.1
protein present in the B1 complex exists in a conformation that
cannot be recognized by the antibody used in the assay. More
experiments are required to address this issue.
An SIE-like element (termed SRE) is present in the human
mcl-1 gene promoter (between nucleotides ?105 and ?92)
(47). This DNA element was shown to be recognized by a
protein complex containing serum response factor and Elk-1.
Both serum response factor and Elk-1 act coordinately to affect
both the basal activity and tetradecanoyl phorbol acetate in-
ducibility of the human mcl-1 gene in human K-562 cells.
Unlike the SIE motif reported in this study, the tetradecanoyl
phorbol acetate inducibility of SRE in the human mcl-1 pro-
moter was shown to be mediated through the extracellular
signal-regulated kinase pathway (47). On the other hand, the
STAT3 protein in the peripheral blood mononuclear cell ex-
tracts isolated from patients with large granular lymphocyte
leukemia has been reported to bind to the murine SIE motif
(9). In contrast, although STAT3 is activated by IL-3 in Ba/F3
cells (30; our unpublished data), no evidence suggests that it
binds to SIE or plays a role in the IL-3 regulation of murine
mcl-1 gene transcription (our unpublished results). Further-
more, Mcl-1 has been found to be expressed in many other cell
types of nonhematopoietic origin (23, 39). Taken together,
these results suggest that SIE-dependent mcl-1 gene transcrip-
tion is likely to be regulated by different transcriptional factors
in a cell type- and stimulation signal-dependent manner.
The function of PU.1 is pivotal to myeloid and lymphoid
development, as many genes essential for these two processes
are regulated by this transcription factor. Our identification of
PU.1 as being involved in IL-3 regulation of mcl-1 gene ex-
pression suggests that Mcl-1 may also contribute to the devel-
opment of myeloid and/or lymphoid cells. Mcl-1-deficient em-
bryos did not survive beyond the peri-implantation stage of
mouse development (39). This early embryo lethality precludes
a direct assessment of Mcl-1 function during myeloid and lym-
phoid development. It would be interesting to investigate how
severe mcl-1 gene expression is affected in PU.1-null embryos
and whether overexpression of Mcl-1 would rescue any defects
of these mutant phenotypes.
We thank Jiahuai Han for the p38 antibody used in the in vitro
kinase assay and the plasmids expressing DN-p38? and DN-p38?;
Koichi Nakajima and Toshio Hirano for the plasmid expressing DN-
STAT3; Masato Kasuga for the ?p85 (DN-p85) expression vector; and
Tzu-Hao Wang for the DN-JNK, DN-ERK1, and DN-ERK2 expres-
This work was supported by intramural funds from Academia Sinica.
J.-M. Wang was supported by a postdoctoral fellowship awarded by
FIG. 10. Schematic representation of molecules involved in the
IL-3 signaling pathways that lead to mcl-1 gene transcription through
the SIE and CRE-2 elements (see text for details).
VOL. 23, 2003ROLES OF p38MAPKAND PU.1 IN Mcl-1 EXPRESSION 1907
1. Bae, J., C. P. Leo, S. Y. Hsu, and A. J. Hsueh. 2000. MCL-1S, a splicing
variant of the antiapoptotic BCL-2 family member MCL-1, encodes a pro-
apoptotic protein possessing only the BH3 domain. J. Biol. Chem. 275:
2. Behre, G., A. J. Whitmarsh, M. P. Coghlan, T. Hoang, C. L. Carpenter, D. E.
Zhang, R. J. Davis, and D. G. Tenen. 1999. c-Jun is a JNK-independent
coactivator of the PU.1 transcription factor. J. Biol. Chem. 274:4939–4946.
3. Bingle, C. D., R. W. Craig, B. M. Swales, V. Singleton, P. Zhou, and M. K.
Whyte. 2000. Exon skipping in Mcl-1 results in a bcl-2 homology domain 3
only gene product that promotes cell death. J. Biol. Chem. 275:22136–22146.
4. Celada, A., F. E. Borras, C. Soler, J. Lloberas, M. Klemsz, C. van Beveren,
S. McKercher, and R. A. Maki. 1996. The transcription factor PU.1 is
involved in macrophage proliferation. J. Exp. Med. 184:61–69.
5. Chao, J.-R., J.-M. Wang, S.-F. Lee, H.-W. Peng, Y.-H. Lin, C.-H. Chou, J.-C.
Li, H.-M. Huang, C.-K. Chou, M.-L. Kuo, J. J.-Y. Yen, and H.-F. Yang-Yen.
1998. mcl-1 is an immediate-early gene activated by the granulocyte-mac-
rophage colony-stimulating factor (GM-CSF) signaling pathway and is one
component of the GM-CSF viability response. Mol. Cell. Biol. 18:4883–4898.
6. Chen, H. M., P. Zhang, M. T. Voso, S. Hohaus, D. A. Gonzalez, C. K. Glass,
D. E. Zhang, and D. G. Tenen. 1995. Neutrophils and monocytes express
high levels of PU.1 (Spi-1) but not Spi-B. Blood 85:2918–2928.
7. DeKoter, R. P., H. J. Lee, and H. Singh. 2002. PU.1 regulates expression of
the interleukin-7 receptor in lymphoid progenitors. Immunity 16:297–309.
8. Eklund, E. A., A. Jalava, and R. Kakar. 1998. PU.1, interferon regulatory
factor 1, and interferon consensus sequence-binding protein cooperate to
increase gp91(phox) expression. J. Biol. Chem. 273:13957–13965.
9. Epling-Burnette, P. K., J. H. Liu, R. Catlett-Falcone, J. Turkson, M. Oshiro,
R. Kothapalli, Y. Li, J.-M. Wang, H.-F. Yang-Yen, J. Karras, R. Jove, and
T. P. Loughran, Jr. 2001. Inhibition of STAT3 signaling leads to apoptosis of
leukemic large granular lymphocytes and decreased Mcl-1 expression.
J. Clin. Investig. 107:351–362.
10. Galson, D. L., J. O. Hensold, T. R. Bishop, M. Schalling, A. D. D’Andrea, C.
Jones, P. E. Auron, and D. E. Housman. 1993. Mouse ?-globin DNA-binding
protein B1 is identical to a proto-oncogene, the transcription factor Spi-1/
PU.1, and is restricted in expression to hematopoietic cells and the testis.
Mol. Cell. Biol. 13:2929–2941.
11. Hagemeier, C., A. J. Bannister, A. Cook, and T. Kouzarides. 1993. The
activation domain of transcription factor PU.1 binds the retinoblastoma
(RB) protein and the transcription factor TFIID in vitro: RB shows sequence
similarity to TFIID and TFIIB. Proc. Natl. Acad. Sci. USA 90:1580–1584.
12. Hromas, R., A. Orazi, R. S. Neiman, R. Maki, C. Van Beveran, J. Moore, and
M. Klemsz. 1993. Hematopoietic lineage- and stage-restricted expression of
the ETS oncogene family member PU.1. Blood 82:2998–3004.
13. Huang, H.-M., C.-J. Huang, and J. J.-Y. Yen. 2000. Mcl-1 is a common target
of stem cell factor and interleukin-5 for apoptosis prevention activity via
MEK/MAPK and PI-3K/Akt pathways. Blood 96:1764–1771.
14. Huang, S., Y. Jiang, Z. Li, E. Nishida, P. Mathias, S. Lin, R. J. Ulevitch,
G. R. Nemerow, and J. Han. 1997. Apoptosis signaling pathway in T cells is
composed of ICE/Ced-3 family proteases and MAP kinase kinase 6b. Im-
15. Jourdan, M., J. D. Vos, N. Mechti, and B. Klein. 2000. Regulation of Bcl-
2-family proteins in myeloma cells by three myeloma survival factors: inter-
leukin-6, interferon-alpha and insulin-like growth factor 1. Cell Death Differ.
16. Karim, F. D., L. D. Urness, C. S. Thummel, M. J. Klemsz, S. R. McKercher,
A. Celada, C. Van Beveren, R. A. Maki, C. V. Gunther, and J. A. Nye. 1990.
The ETS-domain: a new DNA-binding motif that recognizes a purine-rich
core DNA sequence. Genes Dev. 4:1451–1453.
17. Khaled, A. R., A. N. Moor, A. Li, K. Kim, D. K. Ferris, K. Muegge, R. J.
Fisher, L. Fliegel, and S. K. Durum. 2001. Trophic factor withdrawal: p38
mitogen-activated protein kinase activates NHE1, which induces intracellu-
lar alkalinization. Mol. Cell. Biol. 21:7545–7557.
18. Kim, Y. M., H. S. Kang, S. G. Paik, K. H. Pyun, K. L. Anderson, B. E.
Torbett, and I. Choi. 1999. Roles of IFN consensus sequence binding protein
and PU.1 in regulating IL-18 gene expression. J. Immunol. 163:2000–2007.
19. Klemsz, M. J., S. R. McKercher, A. Celada, C. Van Beveren, and R. A. Maki.
1990. The macrophage and B cell-specific transcription factor PU.1 is related
to the ets oncogene. Cell 61:113–124.
20. Klemsz, M. J., and R. A. Maki. 1996. Activation of transcription by PU.1
requires both acidic and glutamine domains. Mol. Cell. Biol. 16:390–397.
21. Kominato, Y., D. Galson, W. R. Waterman, A. C. Webb, and P. E. Auron.
1995. Monocyte expression of the human prointerleukin 1? gene (IL1B) is
dependent on promoter sequences which bind the hematopoietic transcrip-
tion factor Spi-1/PU.1. Mol. Cell. Biol. 15:59–68.
22. Kozopas, K. M., T. Yang, H. L. Buchan, P. Zhou, and R. W. Craig. 1993.
MCL1, a gene expressed in programmed myeloid cell differentiation, has
sequence similarity to BCL2. Proc. Natl. Acad. Sci. USA 90:3516–3520.
23. Krajewski, S., S. Bodrug, M. Krajewska, A. Shabaik, R. Gascoyne, K. Be-
rean, and J. C. Reed. 1995. Immunohistochemical analysis of Mcl-1 protein
in human tissues: differential regulation of Mcl-1 and Bcl-2 protein produc-
tion suggests a unique role for Mcl-1 in control of programmed cell death in
vivo. Am. J. Pathol. 146:1309–1319.
24. Leu, C. M., C. Chang, and C. Hu. 2000. Epidermal growth factor (EGF)
suppresses staurosporine-induced apoptosis by inducing mcl-1 via the mito-
gen-activated protein kinase pathway. Oncogene 19:1665–1675.
25. Lewis, R. T., A. Andreucci, and B. S. Nikolajczyk. 2001. PU.1-mediated
transcription is enhanced by HMG-I(Y)-dependent structural mechanisms.
J. Biol. Chem. 276:9550–9557.
26. Lloberas, J., C. Soler, and A. Celada. 1999. The key role of PU.1/SPI-1 in B
cells, myeloid cells and macrophages. Immunol. Today 20:184–189.
27. Lodie, T. A., R. Savedra, Jr., D. T. Golenbock, C. P. Van Beveren, R. A. Maki,
and M. J. Fenton. 1997. Stimulation of macrophages by lipopolysaccharide
alters the phosphorylation state, conformation, and function of PU.1 via
activation of casein kinase II. J. Immunol. 158:1848–1856.
28. Macleod, K., D. Leprince, and D. Stehelin. 1992. The ets gene family. Trends
Biochem. Sci. 17:251–256.
29. McKercher, S. R., B. E. Torbett, K. L. Anderson, G. W. Henkel, D. J. Vestal,
H. Baribault, M. Klemsz, A. J. Feeney, G. E. Wu, C. J. Paige, and R. A. Maki.
1996. Targeted disruption of the PU.1 gene results in multiple hematopoietic
abnormalities. EMBO J. 15:5647–5658.
30. Nagata, Y., and K. Todokoro. 1996. Interleukin 3 activates not only JAK2
and STAT5, but also Tyk2, STAT1, and STAT3. Biochem. Biophys. Res.
31. Nye, J. A., J. M. Petersen, C. V. Gunther, M. D. Jonsen, and B. J. Graves.
1992. Interaction of murine ets-1 with GGA-binding sites establishes the
ETS domain as a new DNA-binding motif. Genes Dev. 6:975–990.
32. Petersson, M., C. Sundstrom, K. Nilsson, and L. G. Larsson. 1995. The
hematopoietic transcription factor PU.1 is downregulated in human multiple
myeloma cell lines. Blood 86:2747–2753.
33. Petrovick, M. S., S. W. Hiebert, A. D. Friedman, C. J. Hetherington, D. G.
Tenen, and D. E. Zhang. 1998. Multiple functional domains of AML1: PU.1
and C/EBP? synergize with different regions of AML1. Mol. Cell. Biol.
34. Pongubala, J. M., S. Nagulapalli, M. J. Klemsz, S. R. McKercher, R. A.
Maki, and M. L. Atchison. 1992. PU.1 recruits a second nuclear factor to a
site important for immunoglobulin ? 3? enhancer activity. Mol. Cell. Biol.
35. Pongubala, J. M., C. Van Beveren, S. Nagulapalli, M. J. Klemsz, S. R.
McKercher, R. A. Maki, and M. L. Atchison. 1993. Effect of PU.1 phosphor-
ylation on interaction with NF-EM5 and transcriptional activation. Science
36. Rao, E., W. Dang, G. Tian, and R. Sen. 1997. A three-protein-DNA complex
on a B cell-specific domain of the immunoglobulin mu heavy chain gene
enhancer. J. Biol. Chem. 272:6722–6732.
37. Ray, D., R. Bosselut, J. Ghysdael, M. G. Mattei, A. Tavitian, and F. Moreau-
Gachelin. 1992. Characterization of Spi-B, a transcription factor related to
the putative oncoprotein Spi-1/PU.1. Mol. Cell. Biol. 12:4297–4304.
38. Reynolds, J. E., T. Yang, L. Qian, J. D. Jenkinson, P. Zhou, A. Eastman, and
R. W. Craig. 1994. Mcl-1, a member of the Bcl-2 family, delays apoptosis
induced by c-Myc overexpression in Chinese hamster ovary cells. Cancer
39. Rinkenberger, J. L., S. Horning, B. Klocke, K. Roth, and S. J. Korsmeyer.
2000. Mcl-1 deficiency results in peri-implantation embryonic lethality.
Genes Dev. 14:23–27.
40. Saccani, S., S. Pantano, and G. Natoli. 2001. Two waves of nuclear factor ?B
recruitment to target promoters. J. Exp. Med. 193:1351–1359.
41. Scott, E. W., M. C. Simon, J. Anastasi, and H. Singh. 1994. Requirement of
transcription factor PU.1 in the development of multiple hematopoietic
lineages. Science 265:1573–1577.
42. Shackelford, R., D. O. Adams, and S. P. Johnson. 1995. IFN-? and lipopoly-
saccharide induce DNA binding of transcription factor PU.1 in murine tissue
macrophages. J. Immunol. 154:1374–1382.
43. Shin, M. K., and M. E. Koshland. 1993. Ets-related protein PU.1 regulates
expression of the immunoglobulin J-chain gene through a novel Ets-binding
element. Genes Dev. 7:2006–2015.
44. Smith, M. F., Jr., V. S. Carl, T. Lodie, and M. J. Fenton. 1998. Secretory
interleukin-1 receptor antagonist gene expression requires both a PU.1 and
a novel composite NF-?B/PU.1/GA-binding protein binding site. J. Biol.
45. Stephanou, A., T. M. Scarabelli, B. K. Brar, Y. Nakanishi, M. Matsumura,
R. A. Knight, and D. S. Latchman. 2001. Induction of apoptosis and Fas
receptor/Fas ligand expression by ischemia/reperfusion in cardiac myocytes
requires serine 727 of the STAT-1 transcription factor but not tyrosine 701.
J. Biol. Chem. 276:28340–28347.
46. Stutz, A. M., and M. Woisetschlager. 1999. Functional synergism of STAT6
with either NF-?B or PU.1 to mediate IL-4-induced activation of IgE germ-
line gene transcription. J. Immunol. 163:4383–4391.
47. Townsend, K. J., P. Zhou, L. Qian, C. K. Bieszczad, C. H. Lowrey, A. Yen,
and R. W. Craig. 1999. Regulation of MCL1 through a serum response
factor/Elk-1-mediated mechanism links expression of a viability-promoting
member of the BCL2 family to the induction of hematopoietic cell differ-
entiation. J. Biol. Chem. 274:1801–1813.
1908WANG ET AL.MOL. CELL. BIOL.
48. Visconti, R., M. Gadina, M. Chiariello, E. H. Chen, L. F. Stancato, J. S. Download full-text
Gutkind, and J. J. O’Shea. 2000. Importance of the MKK6/p38 pathway for
interleukin-12-induced STAT4 serine phosphorylation and transcriptional
activity. Blood 96:1844–1852.
49. Wang, J.-M., J.-R. Chao, W. Chen, M.-L. Kuo, J. J.-Y. Yen, and H.-F.
Yang-Yen. 1999. The antiapoptotic gene mcl-1 is up-regulated by the phos-
phatidylinositol 3-kinase/Akt signaling pathway through a transcription fac-
tor complex containing CREB. Mol. Cell. Biol. 19:6195–6206.
50. Wang, Y., S. Huang, V. P. Sah, J. Ross, Jr., J. H. Brown, J. Han, and K. R.
Chien. 1998. Cardiac muscle cell hypertrophy and apoptosis induced by
distinct members of the p38 mitogen-activated protein kinase family. J. Biol.
50a.Xing, J., J. M. Kornhauser, Z. Xia, E. A. Thiele, and M. E. Greenberg. 1998.
Nerve growth factor activates extracellular signal-regulated kinase and p38
mitogen-activated protein kinase pathways to stimulate CREB serine 133
phosphorylation. Mol. Cell. Biol. 18:1946–1955.
51. Yamamoto, H., F. Kihara-Negishi, T. Yamada, Y. Hashimoto, and
T. Oikawa. 1999. Physical and functional interactions between the tran-
scription factor PU.1 and the coactivator CBP. Oncogene 18:1495–
52. Yang, Z., N. Wara-Aswapati, C. Chen, J. Tsukada, and P. E. Auron. 2000.
NF-IL6 (C/EBP?) vigorously activates il1b gene expression via a Spi-1
(PU.1) protein-protein tether. J. Biol. Chem. 275:21272–21277.
53. Yordy, J. S., and R. C. Muise-Helmericks. 2000. Signal transduction and the
Ets family of transcription factors. Oncogene 19:6503–6513.
54. Zhao, M., L. New, V. V. Kravchenko, Y. Kato, H. Gram, F. di Padova, E. N.
Olson, R. J. Ulevitch, and J. Han. 1999. Regulation of the MEF2 family of
transcription factors by p38. Mol. Cell. Biol. 19:21–30.
55. Zhou, P., L. Qian, K. M. Kozopas, and R. W. Craig. 1997. Mcl-1, a Bcl-2
family member, delays the death of hematopoietic cells under a variety of
apoptosis-inducing conditions. Blood 89:630–643.
VOL. 23, 2003ROLES OF p38MAPKAND PU.1 IN Mcl-1 EXPRESSION1909