ICANCER RESEARCH 58. 556-561.
February I. 1998]
Differentiation Inducers Modulate Cytokine Signaling Pathways in a Murine
Erythroleukemia Cell Line1
Takayuki Yamashita,2 Hiroshi Wakao,3 Atsushi Miyajima, and Shigetaka Asano
Department (if Hematology/Oncology.
Tokyo. Tokyo. Japan
The Institute of Medical Science IT. Y.. S. A.¡.and The Institute iif Molecular and Cellular Bit/science ¡H.W., A. M.]. The University of
spleen-focus forming virus. The virus transforms erythroid cells in a
multiple-step mechanism (reviewed in Refs. 11 and 12). In the first
stage, a gp55 envelope glycoprotein encoded by the virus causes
Epo-independent polyclonal expansion of erythroblasts through con
stitutive activation of the EpoR. During second events such as proviral
integration at the Spi-1 site or at the p53 locus, the cells are immor
talized and differentiation is abrogated. One intriguing hypothesis is
that chemical inducers cause MEL cell differentiation by stimulating
differentiation signals from EpoR. However, little support has been
given for this notion, mainly because differentiation signals have
remained to be clearly identified.
Phosphorylation/dephosphorylation
of tyrosine residues of proteins
plays a key role in intracellular signaling system of various cytokines
and growth/differentiation factors (reviewed in Refs. 13-17). Partic
ularly. JAK family proteins (14), STAT family proteins (17), and
adapter proteins, such as SHC and GRB2 (13, 16), are key compo
nents in intracellular signaling of cytokine receptor family. In the
present study, we obtained evidence that chemical inducers dramati
cally activated JAK2/STAT5 (14, 17) but suppressed SHC/GRB2 in
F5-5, a MEL cell line (16). To our knowledge, this is the first report
that chemical inducers modulate cytokine signaling in MEL cells.
ABSTRACT
Hexamethylenebisacetamide
(HMBA)
is a potent differentiation
in-
ducer of murine erythroleukemia cells. Immunoprecipitation followed by
Western blotting with an anti-phosphotyrosine (P-Tyr) antibody revealed
that HMBA increased P-Tyr levels and/or amounts of several proteins
containing P-Tyr in F5-5, a murine erythroleukemia cell line. Among
these proteins, we identified a ,\/,. 130,000 protein to be Janus-activated
kinase 2 (JAK2). HMBA induced tyrosine phosphorylation of JAK2 and
signal transducers and activators of transcription 5 (STATS) but not other
JAKs or STATs. This phosphorylation was apparent 12 h after treatment,
maximal at 24 h, and persisted for at least 96 h. Consistently, HMBA
increased STATS DNA-binding
activities. Other chemical inducers,
DMSO and butyrate, also induced a sustained activation of JAK2/STATS,
whereas fetal calf serum and erythropoietin induced transient activation
but not differentiation. Furthermore, overexpression of a dominant-neg
ative form of STATS inhibited the chemically induced differentiation.
These results suggest that persistent activation of the signaling pathway
plays a significant role in the inducer-mediated differentiation. Our data
also suggest that molecular mechanisms for the inducer-mediated activa
tion of JAK2 are independent of cytokine receptor-mediated activation
mechanisms. We tentatively conclude that cytokine signaling is an impor
tant target of chemical inducers in these cells.
INTRODUCTION
MATERIALS
Pharmacological induction of terminal differentiation of neoplastic
cells is a promising strategy for cancer treatment. To identify molec
ular targets of differentiation inducers, would be useful for the devel
opment of antitumor drugs. Hybrid polar compounds such as DMSO
and HMBA4 induce differentiation in a variety of transformed cells
Cell Culture. F5-5 cells, originally derived from the T3CI-2 Friend leuke
mia cell line (18). were maintained by diluting the cells at about 5 X IO4
cells/ml in Ham's F-12 medium containing 10% PCS every 3-4 days. Three
to 4 days after splitting, the cells were used for experiments.
Antibodies. Anti-STAT5 Ab was raised against the NH2-terminal region of
ovine STATS (19). Anti-JAK2 Ab, anti-JAKl Ab. anti-Tyk2 Ab, and anti-P-
(reviewed in Refs. 1 and 2). MEL cells show erythroid differentiation
and growth arrest by treatment with HMBA and DMSO. Recent
studies revealed that these chemicals induce expression of p21, a
cyclin-dependent kinase inhibitor, and hypophosphorylation of RB,
resulting in a G, arrest in MEL cells (3-5). However, G, arrest is
inadequate for differentiation (4). Although previous studies sug
gested the involvement of intracellular signaling components such as
protein kinase C (6, 7), intracellular Ca"+ (8), and tyrosine phosphata-
Tyr Ab (4G10) were purchased from United Biomédical.Inc. (Lake Success,
NY). Anti-STATl Ab, anli-SHC Ab and anti-GRB2 Ab were from Transduction Laboratories (Lexington. KY). Anti-STAT3 Ab, anti-STAT4 Ab. and
anti-STAT6 Ab were from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA).
Anti-JAK3 Ab was from PharMingen (San Diego. CA). Anti-EpoR (NH2
terminus) Ab (20) was a kind gift of Dr. Alan D. D'Andréa (Dana-Farber
Cancer Institute. Boston. MA).
IP and WB. Cells (1 X ]07 cells/sample) were washed once with PBS and
ses (9, 10) in inducer-mediated differentiation of MEL cells, the
significance of these findings remains to be established.
MEL cells are derived from erythroleukemia by infection of Friend
lysed for 30 min in 500 /j.1of ice-cold lysis buffer [1% Triton X-100, 50 mM
Tris-HCl (pH 7.5), 150 mM NaCl, 10 mM NaF, 10 mM NaPPi, 5 mM EDTA, 1
mM sodium orthovanadate, 1 mM phenylmethylsulfonyl
fluoride. 10 ju.g/ml
leupeptin, and 2 /¿g/mlaprotinin]. Nuclei and cell debris were removed by
centrifugation for 15 min at 4°Cin a microcentrifuge. IP was performed as
Received 7/31/97; accepted 11/21/97.
The costs of publication of this article were defrayed in pan by the payment of page
charges. This article must therefore be hereby marked oAwrfbOMBJ in accordance with
18 U.S.C. Section 1734 solely to indicate this fact.
1 Supported by a grant-in-aid from the Ministry of Education, Science. Sports and
described (21), using various antibodies. The protein A-bound immunoprecipitates were washed three times with lysis buffer and eluted in 50 ju.1of
SDS-loading buffer by boiling. Samples were separated on SDS-polyacrylamide gels and transferred electrophoretically to nitrocellulose membranes.
The membranes were blocked in TEST [20 mM Tris-HCl (pH 7.5). 150 mM
Culture (Japan).
2 To whom requests for reprints should be addressed, at Department of Hematology/
Oncology. The Institute of Medical Science. 4-6-1 Shirokanedai Minato-ku. Tokyo 108.
Japan. Phone: 81-3-5449-5542, 5543; Fax: 81-3-5449-5429; E-mail: [email protected].
u-tokyo.ac.jp.
' Present address: The Helix Research Institute. 1532-3 Yana. Kisarazu-shi Chiba 292.
Japan.
4 The abbreviations
used are: HMBA. hexamethylenebisacetamide:
AND METHODS
NaCl. and 0.2% Tween 20] containing 5% BSA, incubated with the indicated
primary antibody in TBST. washed in TBST. incubated with horseradish
peroxidase-coupled secondary antibodies, and subjected to a chemiluminescence immunodetection system (Amersham Corp.). To reprobe the immunoblots, the membranes were incubated in 62.5 mM Tris-HCl (pH 6.8). 1% (w/v)
SDS, 100 mM /3-mercaptoethanol at 50°Cfor 1 h, washed in TBST, and blotted
MEL. murine
erythroleukemia; Epo, erythropoietin; EpoR. erythropoietin receptor; SHC. Src homology
and collagen; GRB2, growth receptor binding protein 2: JAK, Janus-activated kinase:
STAT. signal transducers and activators of transcription; Ab. antibody: P-Tyr. phospholyrosine: IP, tmmunuprecipilation: WB. Western blotting: EMSA. electrophoretic mobil
ity shift assay; PRE. prolactin response element: DN. dominant-negative.
with appropriate antibodies.
EMSA. Nuclear extracts were prepared as described elsewhere (21) and
mixed with the radiolabeled double-stranded oligonucleotide corresponding to
556
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1998 American Association for Cancer Research.
CHEMICAL
INDUCERS
ACTIVATE
examined time courses of the HMBA-induced phosphorylation of
JAK2 and STATS (Fig. 2ß).When F5-5 cells were cultured in fresh
medium containing 10% PCS, P-Tyr levels of JAK2 and STATS
the PRE in the bovine ß-caseinpromoter. EMSA was performed using 8 /xg of
nuclear extracts in 20 ^1 of reaction mixture containing 12 mM HEPES (pH
7.9), 10 fmol of radiolabeled PRE (40.000 cpm), 5% glycerol, 75 mM NaCl.
0.1% NP40, 1 mg/ml BSA. 1 mM EDTA, and 1 /xg of poly(dl-dC). The mixture
increased within l h and reverted to basal levels within 24 h. HMBA
had no effects on the initial change but clearly increased P-Tyr levels
of these proteins at 12 h. Increases were maximal at 24 h and
persistent for at least 96 h. Protein levels of JAK2 and STATS were
not significantly affected during this treatment. These results indicate
that HMBA causes a sustained activation of JAK2, which, in turn,
phosphorylates STATS.
We then asked whether other inducers, DMSO and butyrate. have
similar effects on JAK2 and STATS. HMBA and DMSO belong to a
group of hybrid polar compounds. Butyrate is structurally unrelated to
the compounds. Incubation with 280 mM DMSO or 2 IHMbutyrate for
4-5 days induced 70-80% hemoglobin-positive
cells (data not
was incubated at room temperature for 30 min. and 5 ¿M!
were loaded onto 5%
polyacrylamide gels in 0.25X TBE (22.5 mM Tris-borate, 0.5 mM EDTA).
Generation of Stable Transfectants. F5-5 cells were transfected by
electroporation
(960 fxF, 300 V) with pME18S-Neo containing ovine
STAT5(Y694F) (19). Selection with G418 (500 jug/ml) was initiated 48 h
after electroporation. Selected cells were subcloned by limiting dilution.
STAT5(Y694F) expression levels were determined by WB with anti-STAT5
monoclonal Ab (Transduction
JAK2 IN MEL CELLS
Laboratories).
RESULTS
When F5-5 cells were cultured in the presence of 5 mM HMBA for
4-5 days, 70-80% of the cells became positive for hemoglobin by
dianisidine staining, as described previously (22). In control cultures,
the percentage of hemoglobin-positive cells was <1%. We analyzed
the effects of HMBA on cellular proteins containing P-Tyr. IP of
whole-cell lysates with an anti-P-Tyr Ab (4G10), followed by WB
with the same Ab, showed increases of signal intensities in several
proteins (A/r 175,000, 150,000, 130,000, 90,000, 75,000, and 70,000)
and a decrease in Mr 63,000 protein after 24-48 h treatment with
HMBA (Fig. 1A). These results indicate that P-Tyr levels and/or
amounts of these proteins were altered during the treatment. Among
these proteins, the Mr 130,000 protein showed the most remarkable
changes. To identify this protein, we probed membranes with various
antibodies and found that an anti-JAK2 antibody recognized this
shown). These compounds, like HMBA, caused sustained tyrosine
phosphorylation of JAK2 and STATS. Fig. 2C shows representative
data at 24 h. On the other hand, Epo induced a rapid and transient
tyrosine phosphorylation of JAK2 and STATS, detectable from 5 min
to 2-3 h after stimulation, but no differentiation (data not shown).
Thus, the sustained activation of JAK2 correlates with differentiation
in F5-5 cells.
To obtain evidence that the HMBA-induced tyrosine phosphoryl
ation of STATS causes DNA binding activity of this transcriptional
factor, EMSA was performed. Consistent with the time course of
STATS phosphorylation, its DNA-binding activities were stimulated
by HMBA for 24-96 h, whereas the activities were not detectable in
control cells (Fig. 3/4). The shifted band was supershifted with antiSTAT5 Ab but not by anti-STATl or anti-STAT3 Ab (Fig. 35). Thus,
HMBA stimulates the JAK2/STAT5 signaling pathway.
To further clarify the role of STATS in inducer-mediated erythroid
differentiation, we introduced a DN form of STATS with substitution
of Tyr to Phe at 694. This mutation abolishes the DNA binding
activity of STATS, and overexpression of this molecule attenuates
STATS-mediated responses (19). Because this dominant-negative
form cannot be distinguished from wild-type STATS by molecular
protein (Fig. Iß).
In the next experiments, we examined effects of HMBA on P-Tyr
levels of JAK2 and other JAK family members. We also examined
effects on P-Tyr levels of STAT family proteins that are major
substrates of JAK family kinases (17). As shown in Fig. 2A, treatment
of cells with HMBA for 24 h increased P-Tyr levels but not protein
amounts of JAK2 and STATS, whereas this compound had little
effects on P-Tyr or protein levels of JAKs 1 and 3 or STATs 1-3.
Tyk2 or STAT4 or 6 was not detectable (data not shown). Next, we
Exp.2
24h
B
24h
48h
5
Mr
Fig. 1. Alteration of cellular phosphotyrosinecontaining proteins by HMBA. A, effects of
HMBA on cellular P-Tyr-containing proteins. Cells
were cultured in fresh medium, with or without 5
IÕIM
HMBA, for 24 or 48 h. Cell lysates (3 mg of
protein) were subjected to IP with 4G10 and blotted
with 4G10. Representative data obtained in two
separate experiments (Exp.l and Exp.2) are shown.
Estimated molecular weights of P-Tyr-containing
proteins affected by HMBA are shown on the right
of each WB. B, recognition of a M, 130,000 protein
by anti-JAK2 Ab. Cells were cultured in fresh
medium, with or without 5 mM HMBA for 24 h.
Cell lysates (3 mg of protein) were subjected to IP
with 4G10 and blotted with anti-JAK2 Ab.
175
, 150
té 130
u
70
63
130 kD
175
150
130
90
75
70
IP: 4G10
WB: JAK2
63
IP: 4G10
WB: 4G10
557
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1998 American Association for Cancer Research.
CHEMICAL
IP:
JAK1
JAK2
,,. <
,.
<
? CD
c
ÃŒli
WB:4G10
Protein
S
«. *^^—
INDUCERS ACTIVATE
JAK2 IN MEL CELLS
JAK3 STATI STAT3 STATS
<
e
o
m
S
Z
X
C
Cw<5^
l
IP:JAK2
M>K0^*^"~
*
Q «
a
—¿ IP:STAT5
WB:4G10
JAK2
WB:4G10
—¿
_
STAT5
B
Control
01
IP: JAK2
HMBA
3 6 12 24 1 3 6 12 24h
WB:4G10
JAK2
Control
HMBA
487296
487296h
_ _
—¿
—¿
—¿
—¿
«-. —¿
—¿
—¿
-•
—¿
IP: STAT5 WB:4G10
STAT5
Fig. 2. Stimulation of tyrosinc phosphorylation of JAK2 and STAT5 by chemical inducers. A. effects of HMBA on tyrosine phosphorylation of JAKs and STATs. Cells (l X IO7)
were cultured in 20 ml of fresh medium, with or without 5 HIMHMBA. for 24 h. Cell lysates (3 mg of protein) were subjected to IP w ith antibodies against JAK I. JAK2. JAK3. STATI.
STAT3. and STAT5 and blotted with 4G10. Reprobing with the antibodies against each protein showed that protein levels are constant, ß,time courses of HMBA-induced tyrosine
phosphorylation of JAK2 and STAT5. Cells were cultured in fresh medium, with or without 5 IHMHMBA. for the indicated time. Cell lysates (3 mg of proleinl were subjected to IP
with anti-JAK2 Ah or anti-STAT5 Ab and blotted with 4C'il(). Reprohing with anti-JAK2 Ab or anti-STATS Ab confirmed equal loading. C tyrosine phosphorylation of JAK2 and
STATS by DMSO and butyrale. Cells were incubated with 280 mM DMSO or 2 m.wsodium butyrate for 24 h. Cell lysates (3 mg of proteinl were subjected to IP with anti-JAK2 Ab
or anti-STATS Ab and blotted with 4GIO. Reprobing with anli-JAK2 Ab or anti-STAT5 Ab confirmed equal loading.
weight, we estimated total amounts of DN-STAT5 and wild-type
STAT5 by WB using a monoclonal anti-STAT5 Ab. We analyzed
three representative clones expressing different amounts of STATS.
Clones I and 2 expressed high levels of STATS, and clone 3 ex
pressed an intermediate level (Fig. 44). Consistently, the HMBAinduced tyrosine phosphorylation of STATS was severely suppressed
in clones 1 and 2 and moderately suppressed in clone 3 (Fig. 4B). The
HMBA-induced erythroid differentiation was greatly suppressed in
clones 1 and 2 and mildly suppressed in clone 3 (Fig. 4C). Therefore,
STATS has a significant role in the inducer-mediated differentiation
of F5-5 cells.
To analyze mechanisms for chemically induced activation, we first
asked whether HMBA induces production of a cytokine-like factor(s)
stimulating the JAK2/STATS pathway. If so, culture supernatants
from HMBA-treated cells would induce a rapid tyrosine phosphoryl
stimulated with fresh medium containing 10% PCS, the SHC proteins
of Mr 46,000 and Mr 52.000 were rapidly tyrosine phosphorylated,
and the phosphorylation persisted for at least 24 h. Interestingly,
HMBA markedly suppressed the SHC phosphorylation at 12-24 h
(Fig. 5C). This change was not clear in Fig. 1, probably because these
signals showed an overlap by other heavily phosphorylated proteins
and antibody heavy chains. Consistent with the decreased phospho
rylation of SHC. the association of SHC with GRB2 was inhibited
(Fig. 5D). DMSO and butyrate also suppressed SHC phosphorylation
and SHC/GRB2 association (data not shown). Thus, unlike cytokines.
the chemical inducers have opposing effects on JAK2/STAT5 and
SHC/GRB2.
ation of STATS. However, the supernatant had no such activity,
whereas control supernatant containing Epo induced a rapid phospho
rylation of STATS (Fig. 5-4). F5-5 cells overexpress EpoR as a result
of insertion of the spleen focus-forming virus in the promoter region
of the EpoR gene, an event that may contribute to immortalization of
the cells (23). To examine the possibility that HMBA activates JAK2/
STATS through stimulation of EpoR, we measured P-Tyr levels of
EpoR. HMBA did not increase P-Tyr levels of EpoR (Fig. 5ß),
whereas Epo induced a slight tyrosine phosphorylation of EpoR,
suggesting that HMBA does not stimulate EpoR.
Finally, we examined the effects of HMBA on SHC. The SHC
proteins are tyrosine phosphorylated by multiple cytokine and growth
factor receptors and have been implicated in the Ras signaling path
way via association with the GRB2 adapter protein (16). If HMBA
activates JAK2 through stimulating such receptors, it would be ex
pected that SHC is also tyrosine phosphorylated. When the cells were
Our most important finding is that chemical inducers activate
JAK2, which plays an essential role in cytokine signaling (14), in
MEL cells. When F5-5 cells were incubated with HMBA for 24-48
h, there were increases in P-Tyr levels and/or amounts of several
proteins containing P-Tyr (Fig. IA), Subsequent experiments identi
fied a M, 130,000 protein to be JAK2 (Fig. Iß)and revealed that
HMBA increased P-Tyr levels but not protein amounts of JAK2 (Fig.
2A). We also obtained evidence that HMBA stimulated tyrosine
phosphorylation and DNA-binding activities of STATS, a major
JAK2 downstream component (Figs. 2A and 3). Other chemical
inducers also stimulated tyrosine phosphorylation of JAK2 and
STATS (Fig. 2O. Furthermore, overexpression of DN-STAT5 signif
icantly suppressed the HMBA-induced differentiation (Fig. 4). These
results strongly suggest that the cytokine signaling pathway plays an
important role in the inducer-mediated differentiation. Other workers
reported that most cellular proteins with P-Tyr are dephosphorylated
DISCUSSION
558
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1998 American Association for Cancer Research.
CHEMICAL
INDUCERS ACTIVATE
B
Antibody
Control
HMBA
0 24 48 72 9624 48 72 96 h
£31 S2 £
o g g g
C
(O
W
W
u«w ¿i
JAK2 IN MEL CELLS
biological consequences of JAK/STAT activation presumably depend
on its kinetics and cell types. Interactions with other signaling path
ways, cell cycle regulatory proteins, and transcriptional factors are
considered to be important determinants. Some MEL cell lines
showed constitutive activation of STATS, without differentiation
(data not shown). In these cells, other effects of the inducers. includ
ing suppression of the SHC/GRB2 pathway (Fig. 5. C and Dìand G,
arrest (3-5), may play a major role in differentiation.
Signaling mechanisms for erythroid differentiation have been ex
tensively studied using Epo-responsive cells. Several lines of evidence
demonstrate that JAK2 plays an essential role in Epo-induced cell
proliferation (14, 27, 28). On the other hand, recent studies show that
Epo-induced JAK2 activation is associated with both proliferation and
differentiation in J2E cells (29) and differentiation but not prolifera
tion in SKT6 cells ( 19). These observations are in line with the present
findings. The role of STATS is more controversial (19, 30-32).
Several groups ( 19, 30, 31 ) presented evidence that STATS plays a
positive role in erythroid differentiation in Epo-responsive cell lines.
By contrast, Chretien el al. (32) showed that impaired STATS acti-
WB:STAT5
IP: STAT5
P1 2 3
B
WB:4G10
Fig. 3. Induction of STATS DNA-binding activity by HMBA. A. time course of
STATS DNA-binding activities. Cells were cultured in fresh medium, with or without 5
mM HMBA, for the indicated time. Nuclear extracts were subjected to EMSA assay using
"P-labeled PRE. B. supershift with anti-STAT5 Ab. Cells were incubated with 5 mM
HMBA for 24 h. Nuclear extracts were incubated with anti-STATl Ab. anti-STAT3 Ab.
or anti-STAT5 Ab before electrophoresis and subjected to EMSA assay.
through increased expression of protein tyrosine phosphatases in a
MEL cell line (9, 10). One possible explanation for the discrepancy is
the difference in cell lines used. We observed that the JAK2/STAT5
pathway is constitutively activated in some MEL lines (data not
shown). In such cases, stimulation of JAK2 by chemical inducers
would not be clear. Another explanation is the difference in methods
used to measure P-Tyr levels. IP followed by WB with an anti-P-Tyr
Ab used in the present study is much more sensitive than WB of
whole-cell lysates used in the previous studies. On the other hand,
decreases of P-Tyr in Mr 63,000 protein and SHC may be related to
the increased expression of tyrosine phosphatases (9, 10).
Another notable finding in the present study is that chemical
inducers caused delayed but sustained activation of JAK2/STAT5
(Figs. 2 and 3), whereas FCS or Epo induced a rapid but transient
activation but no differentiation. These results suggest that kinetics of
activation of the signaling pathway is important for biological effects
in the present system. On the other hand, constitutive activation of
JAKs is suggested to be associated with transformation in other
systems (24, 25). An intriguing analogy is observed in the functions of
mitogen-activated protein kinase (reviewed in Ref. 26). In neuronal
cells, transient activation of this kinase causes mitogenesis. whereas
sustained activation causes differentiation. However, constitutive ac
tivation of the kinase in fibroblasts causes transformation. Likewise,
P1 2 3
* »W —¿
WB:STAT5
Fig. 4. Effects of overexpression of dominant-negative STATS on the inducer-mediated differentiation. A, STATS protein levels in parental and DN-STAT5 expressing cells.
Lysates (50 /ig) from parental cells (P) or DN-STATS-expressing clones (1-3) were
separated on SDS polyacrylamide gels and blotted with anti-STAT5 monoclonal Ab. H.
HMBA-induced tyrosine phosphorylation of STATS in parental and DN-STATS-expressing cells. Parental cells (P} and DN-STATS-expressing clones ( 1-3) were incubated with
5 mM HMBA for 24 h. Cell lysates were subjected to IP with anti-STATS Ah and blotted
with 4G10. Reprobing with anti-STATS Ab confirmed equal loading. C. induction of
hemoglobin-positive cells by HMBA. Parental cells (P} or DN-STATS-expressing clones
(1-3) were seededat 5 x IO4cells/ml in the medium with S nisi HMBA. On the fifth day.
the percentages of hemoglobin-positive cells were scored by dianisidine staining. The
percentagesof hemoglobin-positive cells in control cultures were <0.l%. The values are
means of triplicate determinations: bars. SE.
559
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1998 American Association for Cancer Research.
CHEMICAL
INDUCERS
ACTIVATE
JAK2 IN MEL CELLS
B
IP:EpoR
IP: STATS
WB:4G10
STATS
WB:4G10
__
•¿Â»â€”—
EpoR
IP: SHC
IP: SHC
Control
WB-.4G10
HMBA
0 3 61224 3 61224
—¿
~ •¿
—¿
—¿
T —¿
h
WB:SHC
~T
GRB2
SHC
Fig. 5. Mechanisms of HMBA-induced JAK2/STAT5 activation. A, tyrosine phosphorylation of STAT5 by conditioned media from control or HMBA-treated cells and Epo. Cells
(1 X IO7) were cultured in 20 ml medium, with or without 5 imi HMBA. for 24 h: then supernatant was taken from each culture. Cells (1 X IO7) were incubated in 20 ml of control
conditioned medium (Control), conditioned medium from HMBA-lreated cells (HMBA-CM), or control-conditioned medium containing Epo ( I unit/ml; £/«>)
for IO min. Cell lysates
(3 mg of protein) were subjected to IP with anti-STAT5 Ab and blotted with 4G10. Reprobing with anli-STAT5 Ab confirmed equal loading. B, tyrosine phosphorylation of EpoR
by HMBA and Epo. Cells were incubated with 5 m\i HMBA for 24 h or with I unit/ml Epo for IO min. Cell lysates (3 mg of protein) were subjected to IP with anti-EpoR Ab and
blotted with 4(¡10. Arrow, phosphorylated EpoR in response to Epo stimulation. Reprobing with anti-EpoR Ab confirmed equal loading. C suppression of SHC tyrosine
phosphorylation by HMBA. Cells were cultured in fresh medium, with or without 5 mM HMBA. for the indicated time. Cell lysates (3 mg of protein) were subjected to IP with anti-SHC
Ab and blotted with 4GIO. Reprobing with anti-SHC Ab confirmed equal loading. D, inhibition of SHC-GRB2 association by HMBA. Cells were incubated in fresh medium, with
or without 5 HIMHMBA. for 24 h, and cell lysates (3 mg of protein) were subjected to IP with anti-SHC Ah and blotted with anti-SHC Ab and anti-GRB2 Ab.
vation correlates with erythroid differentiation in TF-1 human erythroleukemia cells. One possible explanation of this disagreement is that
biological effects of STAT5 activation are affected by a number of
factors, depending on its kinetics and cell types, as discussed above.
Alternatively, other signaling pathways activated concomitantly may
play a critical role in differentiation.
Our data suggest that molecular mechanisms for the inducer-mediated activation of JAK2 are independent of receptor-mediated activa
tion mechanisms. Receptor-mediated
activation of JAK2 occurs
within a few minutes and usually reverts to basal levels in several
hours, whereas the chemically induced activation appears at 12-24 h
and persists for at least 4 days (Fig. 2, 3). This delay in activation does
not support the notion that the inducers directly activate JAK2, such
as phorbol ester-induced activation of protein kinase C (33). Our data
(Fig. 5, A and B) do not give credence to the possibility that chemical
inducers activate JAK2 through an autocrine loop or stimulation of
EpoR. Finally, unlike cytokines or hematopoietic factors, the chemical
inducers suppressed the SHC/GRB2 pathway (Fig. 5. C and D). The
effects of chemical inducers may be mediated by regulation of mod
ulators for each molecule. For instance, decreased activity of a tyro
sine phosphatase SHP-1 (34) could account for sustained JAK2 acti
vation, but we found no change in SHP-1 levels or its association with
JAK2 (data not shown).
A recent study by Gianni et al. (35) showed that n\\-trans retinoic
acid, a potent differentiation inducer of human promyelocytic leuke
mia cells, activates STATI, thereby stimulating expression of IFNresponsive genes in these cells. These findings support our idea that
pharmacological differentiation inducers exert their actions, in part,
by modulating cytokine signaling pathways.
ACKNOWLEDGMENTS
We are grateful to Mariko Ohara for helpful comments.
REFERENCES
1. Marks, P. A., and Rifkind, R. A. Erythroleukemic differentiation. Annu. Rev. Biochem.. 47: 419-448. 1978.
2. Marks. P. A., Richon. V. M.. Kiyokawa. H.. and Rifkind. R. A. Inducing differenti
ation of transformed cells with hybrid polar compounds: a cell cycle-dependent
process. Proc. Nail. Acad. Sci. USA, 91: 10251-10254, 1994.
3. Kiyokawa. H.. Richon. V. M.. Venta-Perez. G.. Rifkind. R. A., and Marks. P. A.
Hexamethylenebisacetamide-induced
erythroleukemia cell differentiation involves
modulation of events required for cell cycle progression through Gì.Proc. Nati.
Acad. Sci. USA. 90: 6746-6750. 1993.
4. Zhuo, S.. Fan. S.. Huang. S.. and Kaufman. S. Study of the role of retinoblastoma
protein in terminal differentiation of murine erythroleukemia cells. Proc. Nati. Acad.
Sci. USA, 92: 4234-4238, 1995.
5. Macleod. K. F.. Sherry. N.. Hannon. G.. Beach. D.. Tokino. T.. Kinzler, K.,
Vogelstein, B., and Jacks. T. p53-dependent and independent expression of p21
during cell growth, differentiation, and DNA damage. Genes Dev., 9: 935-944, 1995.
6. Melloni. E.. Pontremoli. S.. Michetti. M.. Sacco. O., Cakiroglu. A. G., Jackson, J. F.,
Rifkind. R. A., and Marks. P. A. Protein kinase C activity and hexamethylenebisacetamide-induced erythroleukemia cell differentiation. Proc. Nati. Acad. Sci. USA, 84:
5282-5286. 1987.
7. GuptaRoy, B.. and Cohen. C. M. Maturation of murine erythroleukemia cells com
mitted to differentiation requires protein kinase C. J. Biol. Chem.. 267: 15326-15333.
1992.
8. Levenson, R., Housman. D.. and Cantley. L. Amiloride inhibits murine erythroleu
kemia cell differentiation: evidence for a Ca2+ requirement for commitment. Proc.
Nati. Acad. Sci. USA. 77: 5948-5952, 1980.
560
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1998 American Association for Cancer Research.
CHEMICAL
INDUCERS
ACTIVATE
9. Watanabe, T., Kume, T., and Oishi, M. Alteration of phosphotyrosine-containing
proteins at the early stage of erythroid differentiation of mouse erythroleukemia
(MEL) cells. J. Biol. Chem., 267: 17116-17120, 1992.
10. Kume, T.. Tsuneizumi, K., Watanabe. T., Thomas, M. L., and Oishi. M. Induction of
specific protein tyrosine phosphatase transcripts during differentiation of mouse
erythroleukemia cells. J. Biol. Chem., 269: 4709-4712,
1994.
11. Ben-David, Y., and Bernstein. A. Friend virus-induced erythroleukemia and the
multistage nature of cancer. Cell, 66: 831-834, 1991.
12. D'Andréa.A. D.. Moreau. J-F., and Showers. M. O. Molecular mimicry of erythro-
13.
14.
15.
16.
17.
18.
19.
20.
poietin by the spleen focus-forming virus gpSS glycoprotein: the first stage of Friend
virus-induced erythroleukemia. Biochem. Biophys. Acta, 1114: 31-41, 1992.
Pawson, T. Protein modules and signalling networks. Nature (Lond.), 373: 573-580,
1995.
Ihle, J. N. Signaling by the cytokine receptor superfamily in normal and transformed
hematopoietic cells. Adv. Cancer Res., 68: 23-65, 1996.
Tonks, N. K., and Neel, B. G. From form to function: signaling by protein tyrosine
phosphatases. Cell, 87: 365-368, 1996.
Bonfini, L., Migliaccio. E., Pelicci, G., Lanfrancone, L., and Pelicci. P. Not all She's
roads lead to Ras. Trends Biochem. Sci., 21: 257-261. 1996.
Horvath, C. M., and Darnell, J. E., Jr. The state of the STATs: recent developments
in the study of signal transduction to the nucleus. Curr. Opin. Cell Biol., 9: 233-239,
1997.
Ikawa. Y.. Inoue. Y.. Aida, M., Kameji. R.. Shibata, C., and Sugano. H. Phenotypic
variants of differentiation-inducible
Friend leukemia lines: isolation and correlation
between inducibility and virus release. Bib!. Haemal., 43: 37-47, 1975.
Wakao, H., Chida, D.. Damen, J. E., Krystal, G., and Miyajima, A. A possible
involvement of StatS in erythropoietin-dependent
hemoglobin synthesis. Biochem.
Biophys. Res. Commun., 234: 198-205. 1997.
Barber. D. L., and D'Andréa.A. D. Erythropoietin and interleukin-2 activate distinct
JAK kinase family members. Mol. Cell. Biol., 14: 6506-6514, 1994.
21. Mui, A. L-F., Wakao, H., Kinoshita. T., Kitamura. T., and Miyajima, A. Suppression
of interleukin-3-induced gene expression by a C-terminal truncated StatS: role of
StatS in proliferation. EMBO J.. /5: 2425-2433, 1996.
22. Yamashita, T., Eto, Y., Shibai, H., and Ogata, E. Synergistic action of activin A and
hexamethylenebisacetamide
in murine erythroleukemia cells. Cancer Res.. 50: 31823185, 1990.
23. Hiño,M., Tojo, A., Misawa, Y., Morii, H.. Takaku, F., and Shibuya, M. Unregulated
expression of the erythropoietin receptor gene caused by insertion of spleen focusforming virus long terminal repeat in a murine erythroleukemia cell line. Mol. Cell.
Biol., //: 5527-5533, 1991.
JAK2 IN MEL CELLS
24. Meydan. N.. Grunberger. T.. Dadi. H.. Sbafar, M.. Arpaia. E.. Lapido!, Z., Leeder,
J. S.. Freeman, M., Cohen, A.. Gazi!. A., Levitzki. A., and Roifman, C. M. Inhibition
of acute lymphoblastic leukaemia by a Jak-2 inhibitor. Nature (Lond.), 379: 645-648.
1996.
25. Migone, T-S., Lin, J-X., Cereseto, A., Mulloy, J. C., O'Shea, J. J., Franchini, G., and
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
Leonard. W. J. Constitutively activated Jak-STAT pathway in T cells transformed
with HTLV-1. Science (Washington DC), 269: 79-81, 1995.
Marshall, C. J. Specificity of receptor tyrosine kinase signaling: transient versus
sustained extracellular signal-regulated
kinase activation. Cell, 80: 179-185.
1995.
Witthuhn, B., Quelle, F. W.. Silvennoinen. O.. Yi. T.. Tang. B., Miura. O., and Ihle.
J. N. JAK2 associates with the erythropoietin receptor and is tyrosine phosphorylated
and activated following EPO stimulation. Cell, 74: 227-236. 1993.
Zhuang. H., Patei. S. V., He, T-C, Sonsteby. S. K., Niu, Z.. and Wojchowski, D. M.
Inhibition of erythropoietin-induced mitogenesis by a kinase-deficient form of Jak2.
J. Biol. Chem., 269: 21411-21414, 1994.
Tilbrook, P. A.. Bittorf, T.. Callus. B. A.. Busfield, S. J., Ingley, E.. and Klinken, S. P.
Regulation of the erythropoietin receptor and involvement of J AK2 in differentiation
of J2E erythroid cells. Cell Growth Differ., 7: 511-520, 1996.
Jiang, N., Wrentmore. A. L.. Reese. T. T.. Gregory, R. C.. Sharlow. E. R., Todokoro.
K.. and Wojchowski. D. M. Epo-induced hemoglobinization of SKT6 cells is blocked
by a dominant-negative form of STATS. Blood. XX(Suppl. 1): 53a, 1996.
Iwatsuki. K., Endo, T.. Misawa, H.. Yokouchi, M.. Matsumoto, A., Ohtsubo, M..
Mori, K. J., and Yoshimura, A. STAT5 activation correlates with erythropoietin
receptor-mediated erythroid differentiation of an erythroleukemia cell line. J. Biol.
Chem.. 272: 8149-8153, 1997.
Chretien, S.. Varlet. P., Verdier. F., Gobert, S., Carton, J-P., Gisselbrecht. S., Mayeux.
P., and Lacombe. C. Erythropoietin-induced erythroid differentiation of the human
erythroleukemia cell line TF-1 correlates with impaired STAT5 activation. EMBO J..
15: 4174-4181. 1996.
Kikkawa, U., Takai, Y.. Tanaka. Y.. Miyake. R.. and Nishizuka. Y. Protein kinase C
as a possible receptor protein of tumor-promoting phorbol esters. J. Biol. Chem., 25#:
11442-11445, 1983.
Klingmuller, U., Lorenz. U.. Cantley, L. C., Neel, B. G., and Lodish, H. F. Specific
recruitment of SH-PTPI to the erythropoietin receptor causes inactivation of JAK2
and termination of proliferative signals. Cell. 80: 729-738, 1995.
Gianni, M., Terao, M., Fortino. I.. LiCalzi, M., Viggiano. V.. Barbui, T.. Rambaldi.
A., and Garattini. E. Stall is induced and activated by all-r/Yin.yretinoic acid in acute
promyelocytic leukemia cells. Blood, 89: IOOI-10I2. 1997.
561
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1998 American Association for Cancer Research.
Differentiation Inducers Modulate Cytokine Signaling Pathways
in a Murine Erythroleukemia Cell Line
Takayuki Yamashita, Hiroshi Wakao, Atsushi Miyajima, et al.
Cancer Res 1998;58:556-561.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/58/3/556
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1998 American Association for Cancer Research.