| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
*
Molecular and Developmental Biology and
Clinical Oncology, Institute of Medical Science, University of Tokyo; and
Core Research for Engineering, Science, and Technology, Japan Science and Technology Corporation, Tokyo, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, which is specific for each cytokine, and ßc, which is shared by
all. Although the role of ßc in signal transduction has been
extensively studied, the role of the subunit has remained to be
clarified. To analyze the role of the human (h) GM-CSF receptor
subunit, we constructed a chimeric receptor subunit composed of
extracellular and transmembrane regions of fused with the
cytoplasmic region of ßc, designated /ß. In BA/F3 cells,
chimeric receptor composed of /ß,ß can transduce signals for
mitogen-activated protein kinase cascade activation and proliferation
in response to hGM-CSF. Although phosphorylation of Jak1 but not of
Jak2 occurred with stimulation of hGM-CSF, the dominant-negative Jak2
but not the dominant-negative Jak1 suppresses c-fos
promoter activation. To determine whether the chimeric receptor
/ß,ß is functional in vivo, we developed transgenic mice
expressing the chimeric receptor /ß,ß. Bone marrow cells from
the transgenic mice expressing the /ß,ß receptor form not only
GM colonies but also various lineages of colonies in response to
GM-CSF. In addition, mast cells were produced when bone marrow cells of
the transgenic mouse were cultured with hGM-CSF. Thus, it appears that
the cytoplasmic region of the subunit is not required for hGM-CSF
promoting activities, even in bone marrow cells. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and ß
subunits (1). The subunit is specific for the GM-CSF
receptor (GM-CSFR)2
(2), whereas the ß subunit (ßc) is shared by receptors
for IL-3 and IL-5 (3). The finding of a shared receptor
subunit for IL-3/GM-CSF/IL-5 receptors suggested that ßc plays a role
in common signaling events of these receptors, whereas the subunit
is responsible for specific activities of each cytokine. On the other
hand, because ßc has a relatively long cytoplasmic region in
comparison to that of the subunit, ßc is assumed to have a major
role in signal transduction of these cytokines. Extensive ßc mutation
analysis revealed the precise mechanism of signals through ßc
(4, 5). We reported that GM-CSF stimulates multiple signal
transduction pathways through distinct cytoplasmic domains or tyrosine
residues of ßc (4, 6, 7, 8). Experiments using
dominant-negative Jak2 revealed that all known activities of GM-CSF
depend on the activation of Jak2 (8). Because activation
of various signaling events by ßc mutants indicated that the box1
region is essential for activation of all the tested GM-CSF-induced
phenomena (9), GM-CSF may exert activities through Jak2
activation, which binds to the box1 region of ßc (10).
Although there are only 40 amino acids within the cytoplasmic region of
the subunit, deletion of the cytoplasmic region of this subunit
resulted in loss of GM-CSF function (11, 12, 13). To examine
the role of and ßc in more detail, we and other groups took the
approach of using chimeric receptors of ßc and the subunit of the
human (h) GM-CSF or mouse (m) IL-5 receptors (14, 15, 16). We
constructed chimeric subunits consisting of extracellular and
transmembrane regions of hGM-CSF fused with the cytoplasmic region
of ßc, did the same with the opposite combination, and designated
these subunits as /ß and ß/, respectively. Interestingly,
chimeric receptors consisting of /ß and ß, or /ß and
ß/ transduce signals for the activation of c-fos,
c-jun, and c-myc and short-term proliferation in
BA/F3 cells (14, 15). That means that no cytoplasmic
region of the subunit seems to be required for these activities and
is inconsistent with results obtained when the cytoplasmic region of
the subunit was deleted. One explanation of this contradiction is
that the cytoplasmic region of subunit mimics box1 of ßc, a
hypothesis supported by the high homology between the membrane-proximal
region of and ß subunits of the hGM-CSF receptor.
Because of the sharing of ßc, it remained to be determined whether
specific activities of GM-CSF, IL-3, or IL-5 do exist and whether
related activities are transduced through the subunit. Great
efforts have been made to define specific activities of these
cytokines, and most of the results rule out the possibility of the
existence of cytokine-specific activities. For example, reconstituted
IL-3 or GM-CSF receptors can induce chronic B cell differentiation,
which is assumed to be an IL-5-specific activity (17).
More striking results were obtained using transgenic mice expressing
GM-CSFR or IL-3 receptors (18). We developed a transgenic
mouse expressing hGM-CSFR and ßc subunits driven by the class I
promoter and found that hGM-CSF induces a variety of colonies,
including erythroid colonies, in a methyl cellulose culture from bone
marrow cells of the transgenic mouse (19). These results
indicate that, although GM-CSF was initially identified as GM
lineage-inducing factor, GM-CSF is not a differentiation-inducing
factor, rather it is a proliferation-promoting factor, regardless of
cell lineages. A similar observation was made using transgenic mice
expressing IL-5R (20), thereby suggesting that the
activity of GM-CSF and the IL-5 are basically the same and that their
difference is determined by different subunit expression patterns.
Moreover, reconstitution of the prolactin receptor in the fetal liver
of erythropoietin receptor knockout mice resulted in acquisition of
erythroid differentiation in response to prolactin (21).
Taken together, it is likely that the specificity of each cytokine does
not determine cell fate, at least among these cytokines. Thus, the
question is raised regarding the role of the subunit of
IL-3/IL-5/GM-CSF. Although numerous works suggested that the
cytoplasmic region plays an essential role in signal transduction, it
is tempting to speculate that the subunit can be replaced by
another molecule with structural similarity. We analyzed the role of
the cytoplasmic region of the subunit using a chimeric receptor
consisting of hGM-CSFR and ß subunits. In in vivo analysis using
transgenic mice, we obtained results suggesting that the cytoplasmic
region of the subunit can be replaced by ßc, regardless of the
cell lineage of bone marrow cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
FCS was from Biocell Laboratories (Carson, CA) and RPMI 1640 was obtained from Nikken Biomedical Laboratories (Kyoto, Japan). Recombinant mIL-3 expressed in the silkworm Bombyx mori was purified as described elsewhere (22). hGM-CSF and G418 were kind gifts from Schering-Plough (Madison, NJ). Antiphosphotyrosine Ab, 4G10, and anti-Jak2 Ab used for immunoprecipitation were from Upstate Biotechnology (Lake Placid, NY). Anti-Jak1, Jak2, Src homology 2-containing protein tyrosine phosphatase-2 (SHP-2), and STAT5 were from Santa Cruz Biotechnology (Santa Cruz, CA).
Cell lines and culture methods
A mIL-3-dependent pro-B cell line, BA/F3 (23), was
maintained in RPMI 1640 medium containing 5% FCS, 0.25 ng/ml mIL-3,
100 U/ml penicillin, and 100 µg/ml streptomycin. For cell washing or
depletion, the same medium without mIL-3 was used (depletion medium).
Various BA/F3 cell clones expressing wild-type hGM–CSFR and
hGM–CSFRß (BA/F-wild), wild-type hGM–CSFR and chimeric
ß/ subunit (BA/F-,ß/), wild-type hGM–CSFRß and chimeric
/ß (BA/F-/ß,ß), and chimeric /ß and ß/
(BA/F-/ß,ß/) were grown in the same type of medium, and 500
µg/ml G418 was added as a supplement.
Cell proliferation assay
DNA replication was analyzed by incorporation of [3H]thymidine as described (24). Briefly, BA/F3 cells in a flat-bottom 96-well plate with various concentrations of mIL-3 or hGM-CSF (1.2 x 104 cells/200 µl/well) were cultured for 24 h and labeled with [3H]thymidine (1 µCi/well) for 4 h. Cells were transferred to a filter using a cell harvester (Micro96, Skatron, Lier, Norway), and [3H] incorporation was analyzed using a filter counter (1450 Microbeta Plus, Wallac, Turku, Finland) according to the manufacturer’s instruction. For long-term proliferation assay, the viable cell number was determined by trypan blue dye exclusion assay.
Immunoprecipitation and Western blotting
Immunoprecipitation and Western blotting were done as described
elsewhere (8, 9). Briefly, cells (1 x
107) were depleted of mIL-3 for 5 h and then
restimulated with hGM-CSF (10 ng/ml) or mIL-3 (3 ng/ml) for 10 min. To
check the function of dominant-negative Jak1, plasmids encoding
hGM-CSFR and ßc were cotransfected with dominant-negative Jak1 or
vector control to BA/F3 cells (5 x 107),
and cells were cultured overnight. The cells were depleted of mIL-3 for
5 h and restimulated with hGM-CSF (10 ng/ml) for 10 min, and then
immunoprecipitation was performed using anti-Jak1 Ab. The cells
were lysed with 500 µl of lysis buffer (20 mM Tris (pH 7.5), 150 mM
NaCl, 1 mM EDTA, 0.5% Nonidet P-40, 1 mM sodium orthovanadate, and 1
mM PMSF) for 1 h. The immunoprecipitation was done using the
indicated Abs and protein A-Sepharose (Pharmacia Biotech, Uppsala,
Sweden) for 2 h. The precipitate was analyzed by immunoblotting
using appropriate Abs (8, 9). The blots were visualized
using an enhanced chemiluminescence kit (Amersham International,
Buckinghamshire, U.K.).
Transient c-fos promoter assay
BA/F3 cells were transfected with DNA by electroporation, as described (25). Cells were immediately subjected to factor starvation for 5 h and stimulated with 5 ng/ml hGM-CSF or mIL-3 for 5 h. In the experiments utilizing dominant-negative Jak1 or Jak2, these plasmids were cotransfected together with c-fos-luciferase. Luciferase assays were done, as described (24). The amount of protein was estimated using bicinchoninic acid kits (Pierce, Rockford, IL) according to the manufacturer’s instructions. In cotransfection experiments, the total amount of transfected DNA was adjusted by adding control vector plasmids. Luciferase activity was expressed in terms of the luminescence intensity (relative light units) per protein.
Plasmid construction and DNA preparation
Dominant-negative Jak1 was constructed by deletion of the kinase
domain located in the C terminus of Jak1. Jak1 cDNA (pBSK-Jak1) was
kindly provided by Dr. J. Ihle (St. Jude Children’s Research Hospital,
Memphis, TN). Construction of the plasmid-containing dominant-negative
Jak1 under the control of the SR promoter was as follows: the coding
region of JAK1, which lacks the kinase domain, was isolated from
pBSK-JAK1 at EcoRI and SacI sites. The
SacI end was blunt-ended using the Klenow fragment and
subcloned into pME18S containing the SR promoter (26)
at the EcoRI site and blunt-ended NotI
site.
The fragment encoding the chimeric /ß subunit (15)
originally inserted into pME18S was isolated using XhoI and
XbaI sites. The fragment was blunt-ended and inserted into
the blunt-ended EcoRI site of the pLD1 expression vector,
which carries the 1.2-kb mouse MHC L-locus gene
(H2-Ld) promoter and the 0.8-kb ß globin polyadenylation sequence.
The wild-type ß subunit was inserted into the same vector using
blunt-ended EcoRI sites (pLd-ß). The pLd-/ß and -ß
plasmids were double digested with SphI and XhoI.
The resulting fragments of pLd-/ß and -ß were 5.1 and 4.8 kb,
respectively, and these fragments were purified using a low-melting
agarose gel (Sea Plaque, FMC, Rockland, ME) and a Qiagen tip 5 column
(Qiagen, Hilden, Germany) and were resuspended in 10 mM Tris-HCl (pH
7.5) and 0.2 mM EDTA.
Generation and maintenance of transgenic mice
Transgenic mice were produced using the C3H/HeN strain by the
standard oocyte injection method. The mice were maintained under
specific pathogen-free conditions in microisolator cages; all the
equipment and supplies including cage, water bottles, wooden chips for
bedding, and food pellets were sterilized. The cages were in an
environmentally controlled clean room with 12-h light-dark cycles in
the authorized animal facility of our institute. Integration of
receptor DNA was checked by genomic PCR and by Southern blot analysis.
Mouse tail-tip DNA was prepared as described (19). Genomic
PCR was done for primary screening of positive mice using a 5' common
primer annealing with LD promoter: 5'-TCATGTTATATGGAGGGGGC-3'; and
using 3' for GMR; 5'-CTCTGGGCTCAGAGCTTG-3', and GMRß,
5'-TGGATCTCAAATGTGTGGTC-3'. Integrated DNA was further confirmed by
Southern blot analysis using tail DNA as described (19).
Surface expression of the receptor was examined by flow cytometry using
an anti-hGM-CSFR Ab (c-20, Santa Cruz Biotechnology),
anti-hGM-CSFRß Ab (s-16, Santa Cruz Biotechnology), and
FITC-conjugated goat anti-mouse IgG (Cappel, Durham, NC). Flow
cytometry was done using a Becton Dickinson (Mountain View, CA) FACScan
and FACScan software, CellQuest.
Methylcellulose colony assay
Methylcellulose colony assay was done using bone marrow cells of
6-wk-old female transgenic mice and their normal littermates. Bone
marrow cells were prepared in -MEM (Flow Laboratories, Rockville,
MD) by repeated pipetting and then were passed through a 50-µM nylon
cell strainer (Becton Dickinson Labware, Franklin Lakes, NJ). Cells
(5 x 104) in 1 ml of medium containing
1.2% methylcellulose (Shin-etsu, Tokyo, Japan), 30% FCS (HyClone,
Logan, UT), 1% deionized fraction V BSA (Sigma, St. Louis, MO),
10-4 M 2-ME (Eastman Organic Chemicals,
Rochester, NY), and cytokines were plated in 35-mm culture dishes.
Colony types were determined on days 2–14 of incubation by in situ
observation using an inverted microscope, according to reported
criteria (19, 27).
Mast cell culture and staining
Bone marrow cells cultured in -MEM, 30% FCS supplemented
with hGM-CSF (20 ng/ml or mIL-3, 5 ng/ml) for 3 wk were stained by
safranin-alcian blue to determine whether mast cells would be produced.
Cells (5000 cells/slide) were cytocentrifuged on a glass slide and
fixed overnight with carnoy solution. After rinsing the slides in
distilled water, the samples were soaked in 3% acetic acid for 3 min
and stained with alcian blue acetic acid for 4 h at room
temperature. After incubation, the samples were first washed with 3%
acetic acid and then with distilled water. Staining was done by
exposure to safranin O for 1 min. Samples were sealed with cover glass
and examined under a microscope.
hGM-CSF in vivo administration
The injection schedules of hGM-CSF were essentially the same as
described (18). Briefly, wild-type hGM-CSFR transgenic
mice, /ß,ß transgenic mice, and control mice, which are
littermates of /ß,ß transgenic mice, (three mice for each
strain) were injected s.c. with 0.3 ml of PBS containing 500 ng hGM-CSF
twice a day at 12-h intervals for 7 consecutive days. Another three
/ß,ß transgenic mice were given 0.3 ml of PBS as a control,
using the same schedule for injection.
Peripheral blood cell counts, bone marrow cell counts, and flow cytometry
The mice were anesthetized with ether, and peripheral blood was collected from hearts, as described (18). Differential count was determined using a hemocytometer (Sysmex K1000, Sysmex, Kobe, Japan) and smear preparations as described (18). Preparations from bone marrow and spleen cells were cytocentrifuged (Cytospin II, Shandon Southern Instruments, Sewickley, PA).
Thymic cells (1 x 106 cells) were washed, stained with FITC-conjugated anti-CD4 and PE-conjugated anti-CD8 Abs (PharMingen, San Diego, CA) in 200 µl staining solution (PBS, 0.2% BSA), and incubated 20 min on ice. After washing, flow cytometry was done using a Becton Dickinson FACScan with CellQuest software.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To clarify the role of the and ßc subunits in hGM-CSFR
signal transduction, we previously constructed chimeric subunits
/ß and ß/ and analyzed the functions of receptors consisting
of various combinations of chimeric and wild-type subunits in BA/F3
cells (15). The /ß contains extracellular and
transmembrane regions of subunit and the cytoplasmic region of
ßc, and the ß/ contains extracellular and transmembrane regions
of ßc and the cytoplasmic region of the subunit. Several
different combinations of chimeric receptors were reconstituted in
BA/F3 cells; among them, [/ß,ß], [/ß,ß/], and
[,ß/] bind to hGM-CSF with high affinity. We designated these
BA/F3 cells expressing chimeric receptors stably as BA/F-/ß,ß;
BA/F-/ß,ß/; and BA/F-,ß/. By Northern blot analysis
and MTT assay, we found that in response to hGM-CSF, chimeric receptors
[/ß,ß] and [/ß,ß/] induced short-term
proliferation and immediate early genes such as c-fos,
c-jun, and c-myc(11). In this study,
we initially performed a more detailed characterization of these
chimeric receptors, [/ß,ß], [/ß,ß/],
[,ß/], using BA/F3 cells. We examined the proliferation
induced through these chimeric receptors in the presence of hGM-CSF by
[3H]thymidine incorporation as well as
long-term proliferation by trypan blue exclusion assay. As shown in
Fig. 1
A, incorporation of
[3H]thymidine increased with hGM-CSF
stimulation in a dose-dependent manner in the BA/F-/ß,ß and
BA/F-/ß,ß/ cells. The value reached a maximum at 1 ng/ml of
hGM-CSF, which corresponds to the value obtained with the wild-type
receptor (24). In contrast, within BA/F-,ß/ cells,
only a background incorporation was observed with up to 1 ng/ml of
hGM-CSF stimulation, and the value increased slightly at concentrations
over 1 ng/ml of hGM-CSF. This observation is consistent with our
previous finding that [,ß/] is not capable of promoting
short-term proliferation (15). Although saturation
concentrations of hGM-CSF for [/ß,ß] and [/ß,ß/]
appear to be similar, concentration curves for these receptors show a
significant shift. Because the affinity for hGM-CSF and the number of
wild-type, [/ß,ß], and [/ß,ß/] receptors are
comparable (data not shown), a difference in the efficiency to promote
proliferation rather than binding to hGM-CSF was surmised. We next
asked whether chimeric receptors [/ß,ß], [/ß,ß/],
[,ß/] could support long-term proliferation of BA/F3 cells,
as determined by trypan blue exclusion assay. As shown in Fig. 1B, BA/F-/ß,ß and BA/F-/ß,ß/ cells can
proliferate for at least 4 days in response to 10 ng/ml of hGM-CSF, and
increase in cell number is comparable to that observed in response to
mIL-3. In contrast, hGM-CSF cannot maintain the long-term proliferation
of BA/F-,ß/.
|
/ß, ß] and [/ß, ß/] induce
phosphorylation of Jak1 and SHP-2 but not Jak2
To analyze in detail signaling events through chimeric receptors
[/ß,ß], [/ß,ß/], and [,ß/], we next
analyzed the phosphorylation of various signaling molecules activated
by hGM-CSF stimulation in BA/F3 cells expressing wild-type hGM-CSFR.
hGM-CSF stimulates tyrosine phosphorylation of Jak1 and Jak2 in various
cells including BA/F3 cells (8, 28). We first examined
tyrosine phosphorylation of Jak1 and Jak2 by immunoprecipitation and
then by Western blotting, using an antiphosphotyrosine Ab. As shown in
Fig. 2, Jak1 is phosphorylated by
stimulation of hGM-CSF in both BA/F3-/ß,ß and -/ß,ß/.
In contrast, tyrosine phosphorylation of Jak2 is not observed with
hGM-CSF stimulation, although mIL-3 does induce tyrosine
phosphorylation of Jak2 in these cells. As expected, within
BA/F-,ß/ cells, tyrosine phosphorylation of neither Jak1 nor
Jak2 occurred with stimulation of hGM-CSF. A similar observation
regarding the absence of Jak2 tyrosine phosphorylation through a
chimeric receptor of mouse IL-5 receptor and ßc has been made
(29), and an essential role for the cytoplasmic region of
the subunit for Jak2 phosphorylation was considered. Because
tyrosine phosphorylation of Jak2 was not evident in BA/F3 cells
expressing [/ß,ß/], which contains the cytoplasmic region
of the subunit, the absence of this phosphorylation may not be due
to absence of the cytoplasmic region of the subunit. We further
analyzed tyrosine phosphorylation of molecules downstream of Jak2,
SHP-2, and STAT5, which are phosphorylated in response to hGM-CSF in
BA/F-wild cells (9). Both are tyrosine phosphorylated by
hGM-CSF in BA/F-/ß,ß and -/ß,ß/ cells but not in
BA/F-,ß/ cells. The level of phosphorylation is comparable to
that observed with mIL-3 stimulation.
|
Using a dominant-negative Jak2 mutant, we had previously found
that Jak2 plays an essential role in all the examined GM-CSF receptor
signals (8). In addition, reports of knockout mice of Jak2
or Jak1 support the idea that Jak2 but not Jak1 is essential for IL-3
and GM-CSF activities (30, 31, 32). To investigate whether
Jak2 or an undetectable level of Jak2 activation plays a role in the
signal transduction of chimeric receptors [/ß,ß] and
[/ß,ß/], we next analyzed the effects of dominant-negative
Jak2. In addition to the dominant-negative Jak2, we constructed
dominant-negative Jak1. Dominant-negative Jak1 suppressed
autophosphorylation of wild-type Jak1 but not Jak2 when cotransfected
in COS7 cells (data not shown). We also checked whether
dominant-negative Jak1 suppressed Jak1 phosphorylation in BA/F3 cells.
hGM-CSF receptor was transiently reconstituted in BA/F3 cells, and the
effects of coexpressed dominant-negative Jak1 on hGM-CSF-induced
endogenous Jak1 phosphorylation was examined by immunoprecipitation and
then by Western blotting. As shown in Fig. 3A, cotransfected
dominant-negative Jak1 abrogated tyrosine phosphorylation of Jak1
induced by the addition of hGM-CSF. In our previous report, we found
that c-fos mRNA was induced by hGM-CSF in BA/F-/ß,ß
and -/ß,ß/ cells but not in BA/F-,ß/ cells by
Northern blot analysis (11). We examined the effects of
dominant-negative Jak2 as well as those of dominant-negative Jak1 for
chimeric receptor activity using c-fos promoter
transient analysis. As shown in Fig. 3B, luciferase activity
driven by the 0.4-kb fragment of c-fos promoter
(24) is induced by hGM-CSF in BA/F-/ß,ß and
-/ß,ß/ cells but not in BA/F-,ß/ cells. We previously
showed that the dominant-negative Jak2 suppressed mIL-3- or
hGM-CSF-induced c-fos promoter activity in BA/F3-wild cells
(8). As shown in Fig. 3, coexpression of dominant-negative
Jak2 but not of dominant-negative Jak1 suppressed hGM-CSF-induced
c-fos promoter activity in BA/F-wild cells. Interestingly,
dominant-negative Jak2 also suppressed c-fos luciferase
activity induced by hGM-CSF in BA/F-/ß,ß and -/ß,ß/
cells. In addition, dominant-negative Jak1 did not affect
c-fos promoter activation by hGM-CSF in any cell tested.
These results indicate that although no phosphorylation was detected by
immunoprecipitation and Western blotting, Jak2 may play an essential
role and Jak1 may not play a role in chimeric receptor [/ß,ß]
and [/ß,ß/] signal transduction.
|
To analyze the in vivo role of GM-CSF, we previously generated
transgenic mice expressing hGM-CSF receptor and ßc
subunits driven by class I promoter (18, 19).
In this mouse, the hGMR-CSFR was expressed in bone marrow cells as well
as in other tissues and organs but not in the brain. When we did a
methylcellulose colony assay using bone marrow cells of the
transgenic mouse, although mouse GM-CSF induces only granulocyte and
macrophage colonies from bone marrow cells, all the examined lineages
of the colonies, including erythroid colonies, were induced by the
addition of hGM-CSF (19). Therefore, as long as the
hGM-CSFR is expressed, hGM-CSF has strong proliferation-promoting
activity, regardless of cell lineages. Based on these results, we
speculated that cell lineage fate may be determined by an intrinsic
factor rather than by extracellular inducing factors like GM-CSF.
Although chimeric receptor [/ß,ß], which does not contain the
cytoplasmic region of the subunit, exerts full activity in BA/F3
cells, there is the possibility that the cytoplasmic region of the
subunit plays a role in native bone marrow cells. Based on this idea,
we generated transgenic mice expressing the /ß,ß receptor. The
class I promoter was used to express /ß and ßc subunits, and
transgenic mice were produced using the C3H/HeN strain by standard
oocyte injection methods. Fig. 4A shows a schematic diagram
of plasmids that were coinjected into oocytes. Expression of the
/ß,ß subunits of the chimeric receptor in bone marrow cells and
thymocytes of transgenic mice was examined after
staining with anti-hGM-CSFR or ßc subunit Abs (Fig. 4B). The level of expression of both subunits was seen to be
comparable with that in transgenic mice expressing the wild-type
hGM-CSFR.
|
To determine whether the chimeric receptor /ß,ß would
support colony formation, we did a methylcellulose colony assay using
bone marrow cells derived from transgenic and littermate mice.
Transgenic mice expressing wild-type hGMR-CSF (19) were
used as positive controls. Table I shows
a summary of the methylcellular colony assay using mIL-3 or hGM-CSF as
added growth factors. hGM-CSF did not induce any type of colony in bone
marrow cells derived from control mice. The addition of mIL-3 resulted
in granulocyte, macrophage, and granulocyte-macrophage colonies and a
number of megakaryocyte colonies and granulocyte, macrophage,
megakaryocyte colonies, as expected from previous reports. As we
reported earlier, the addition of hGM-CSF induced all the examined
colonies including CFU of erythroid within the bone marrow cells of
transgenic mice expressing wild-type hGM–CSFR. Within the bone marrow
cells of transgenic mice expressing chimeric receptor /ß,ß, the
addition of hGM-CSFR also induced all the colonies examined. Colony
numbers were comparable to those observed using bone marrow cells of
transgenic mice expressing the wild-type receptor.
|
/ß,ß in liquid cultures of bone marrow cells. Bone marrow cells
were cultured in the presence of hGM-CSF for 21 days before alcian blue
staining. As shown in Fig. 5, positive
mast cells were induced in the presence of hGM-CSF within bone marrow
cells of transgenic mice expressing wild-type hGM–CSFR or chimeric
receptor /ß,ß but not from bone marrow cells of control mice
(data not shown). Essentially the same results were obtained in the
presence of stem cell factor, with the addition of GM-CSF. These
results indicate that hGM-CSF is fully active through chimeric
/ß,ß receptor, regardless of the cell lineage.
|
We next evaluated the in vivo effects of hGM-CSF on hematopoietic
and lymphopoietic responses in the transgenic mice. Transgenic mice
expressing wild-type hGM-CSFR or the chimeric receptor /ß,ß and
control mice were given hGM-CSF (500 ng/0.3 ml PBS) or PBS (0.3 ml)
s.c. twice a day for 7 days. Peripheral blood differential count was
determined by hemocytometer and smear preparations (Table II). As previously reported,
administration of hGM-CSF increased the number of white blood cells
(WBC) in transgenic mice expressing wild-type hGM-CSFR
(18). When we administrated hGM-CSF (500 ng/0.3 ml PBS)
into ß/,ß transgenic mice, the number of WBC increased, and
differential counts indicated significant increases of granulocytes and
large granular lymphocytes. This phenotype is the same as that observed
within transgenic mice expressing wild-type hGM-CSFR. In both wild-type
hGM–CSFR and chimeric /ß,ß transgenic mice, numbers of red
blood cells and hematocrit were unchanged. Hemoglobin concentration
and number of platelets showed no significant change in the
presence of hGM-CSF. Even giving 500 ng hGM-CSF, the control mice had
almost the same level of differential counts of peripheral blood
cells.
|
The total cell number of femur bone marrow cells slightly
decreased in the presence of hGM-CSF in both wild-type and chimeric
/ß,ß transgenic mice (Fig. 6A). Morphological analysis
using cytospin samples of bone marrow cells indicated that numbers of
monocytes/macrophages remained relatively constant, whereas numbers of
erythroid and lymphoid cells decreased.
|
/ß,ß receptor transgenic mice. Spleen
weight almost doubled (Fig. 6
B) with the number of spleen
cells increasing in parallel with the weight. Morphological analysis of
cytospin samples showed a significant increase in numbers of erythroid
cells and a modest increase with monocytes/macrophages. Effects of in vivo administration of hGM-CSF into the thymuses of the transgenic mice
The administration of hGM-CSF led to a remarkable shrinkage in the
thymus of transgenic mice expressing wild-type hGM-CSFR
(18). The total number of thymic cells of transgenic mice
expressing wild-type hGM-CSFR decreased significantly, and those of
transgenic mice expressing chimeric /ß,ß receptor decreased
slightly (Fig. 7A). When we
analyzed subpopulations of thymocytes by staining anti-CD4 and -CD8
Abs, CD4+CD8+
double-positive cells had decreased, and double-negative cells
increased in the thymuses of transgenic mice of wild-type hGM-CSFR. In
the thymuses from transgenic mice with the chimeric /ß,ß
receptor, similar phenomena were observed, but the extent was less than
that observed in wild-type receptor transgenic mice.
|
/ß,ß receptor by flow cytometer
using anti- and anti-ß Abs. We found that the levels of
all the subunits were comparable in these transgenic mice. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
/ß,ß
consisting of hGM-CSFR /ß,ß can transduce growth signals in
mouse bone marrow cells, which means that the cytoplasmic region of the
subunit of hGM-CSFR is not required for proliferation of bone
marrow cells. These results are consistent with results obtained in
analyses of chimeric receptor /ß,ß activation in BA/F3
cells. In BA/F3 cells, the endogenous mIL-3 receptor is present, and
the possibility that the chimeric receptor /ß,ß can transduce
signals through interaction with the mIL-3 receptor was discussed. In
contrast, only part of the population of bone marrow cells expresses
the mIL-3 receptor, thereby suggesting that the chimeric receptor
/ß,ß itself can transduce signals without involvement of the
mIL-3 receptor. Numerous works examined the signal transduction of
various GM-CSFR or IL-5R subunit mutants in cell lines showing an
essential role for the cytoplasmic region of the subunit for
function of these cytokines (11, 12, 13). In the case of ßc,
the precise requirement of the ßc subregion for signal transduction,
such as an essential role of tyrosine residues and box1 region, have
been reported (4). In contrast, tyrosine phosphorylation
of the subunit has not been observed, whereas requirement of
membrane-proximal proline residues for proliferation as well as Jak2
and STAT5 phosphorylation was reported (13, 16).
Similarity between the proline-rich sequence of the subunit and the
box1 region of ßc was evident and led to the idea that the
requirement of the membrane-proximal region of the subunit could be
explained by mimicry of the box1 region structure. In other words, the
membrane-proximal region of the subunit can be replaced by the box1
region of ßc. Results obtained in the case of the chimeric receptor
/ß,ß support this idea.
Although the chimeric receptors [/ß,ß] and [/ß,ß/]
can fully activate the c-fos promoter and proliferation,
phosphorylation of Jak2 was not observed by stimulation through
chimeric receptors [/ß,ß] and [/ß,ß/] in BA/F3
cells. Defective or weak activation of Jak2 is not due to lack of the
cytoplasmic region of the subunit because chimeric receptor
/ß,ß/ cannot phosphorylate Jak2. Among Jak1, Jak2, Jak3, and
Tyk2 (data not shown), only Jak1 showed phosphorylation after
stimulation of chimeric receptors [/ß,ß] and
[/ß,ß/], which suggests a role of Jak1 in these receptors.
In the wild-type receptor, the essential role of Jak2 but not Jak1 for
GM-CSF signals was indicated from studies done using dominant-negative
Jaks and knockout mice (8, 30, 31, 32). In addition, we also
examined c-fos promoter activation by hGM-CSF wild-type
receptor in mutant cells that lack either Jak1 or Jak2 (28, 33, 34) and found an essential role for Jak2 but not Jak1
(unpublished data). All the data suggest that Jak1 cannot substitute
for Jak2 and that there is no functional redundancy between Jak1 and
Jak2 for GM-CSF signals. Taken together, it can be speculated that the
undetectable level of phosphorylated Jak2 plays an essential role and
Jak1 is dispensable even though it is significantly phosphorylated. The
observation that dominant-negative Jak2 affected c-fos
promoter activation through chimeric receptors [/ß,ß] and
[/ß,ß/] but that dominant-negative Jak1 did not supports
the idea that Jak2 may be phosphorylated at a lower than detectable
level. As another possibility, it should be noted that there is a
report that dominant-negative Jak2 suppresses gp130 activity, even in
mutant cells that lack Jak2 (35). This phenomenon may be
explained by the strong affinity of dominant-negative Jak2 to the
box1/box2 region of gp130, an event which causes interference with the
binding of molecules that play essential roles in gp130 signals. Thus,
we cannot exclude the possibility that dominant-negative Jak2
suppresses molecules other than Jak2 in our chimeric receptors
[/ß,ß] and [/ß,ß/].
Although GM-CSF, IL-3, and IL-5 exert similar but different biological
functions, we obtained no clear evidence to explain the difference by
subunit specificity for each cytokine. It was clearly shown through
approaches using transgenic mice expressing hGM-CSFR and IL-5R that no
phenotypical difference existed (19, 20), which suggests
that the specificity of biological activities is not determined by
receptor structure or by its signal but mainly by differences in
receptor expression pattern in various precursor cells. In vitro
methylcellulose analysis of bone marrow cells of chimeric /ß,ß
receptor transgenic mice shows that all of the examined lineages of
cells were expanded in the presence of hGM-CSF. Thus, regardless of the
cell lineage, the chimeric /ß,ß receptor can function without
the cytoplasmic region of subunit. These observations strongly
support the idea that the reason why GM-CSF is called CSF of GM
lineages may be explained by the finding that GM-CSFR is exclusively
expressed in precursor cells of GM lineages. Our previous analysis of
hematopoietic and lymphopoietic responses of hGM-CSF wild-type receptor
transgenic mice administrated of hGM-CSF showed a more complicated
phenotype (18). Such increases in numbers of WBC and a
remarkable increase of large granular lymphocytes were observed, but
the total cell number of bone marrow cells and thymocytes decreased.
Our work using transgenic mice with chimeric /ß,ß receptors
shows a phenotype indistinguishable from that of wild-type hGM-CSFR
transgenic mice after administration of hGM-CSF; that is, the chimeric
receptor without the cytoplasmic region of the subunit mimics
wild-type hGM-CSFR. Our current work not only supports this idea but
also supports the notion that ßc is sufficient for signal
transduction and that the role of the subunit is to determine the
binding specificity between receptor and factor.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Abbreviations used in this paper: GM-CSFR, GM-CSF receptor; h, human; m, mouse; SHP-2, Src homology 2-containing protein tyrosine phosphatase-2; BA/F-wild, wild-type hGMR and hGMRß; BA/F-,ß/, wild-type hGMR and chimeric ß/ subunit; BA/F-/ß,ß, wild-type hGMRß and chimeric /ß subunit; BA/F-/ß,ß/, chimeric /ß and ß/ subunits; WBC, white blood cell.
Received for publication October 8, 1999. Accepted for publication January 26, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
subunit of human granulocyte-macrophage colony stimulating factor. J. Allergy Clin. Immunol. 6:1110.
subunit is essential for both GM-CSF-mediated growth and differentiation. J. Biol. Chem. 272:17450. subunit in the activation of Jak2 and Stat5. Blood 92:867./ß-granulocyte-macrophage colony-stimulating factor receptor induces ligand-dependent proliferation in a murine cell line. J. Biol. Chem. 269:30173. subunit transduces growth promoting signals in BA/F3 cells. Biochem. Biophys. Res. Commun. 208:368.[Medline]
chain and its function in IL-5-mediated activation of JAK kinase and STAT5. Int. Immunol. 8:237. subunit of IL-3 and GM-CSF receptor. Int. Arch. Allergy Immunol. 1996:12.
subunit constitutively. J. Exp. Med. 181:889. promoter: an efficient and versatile mammalian cDNA expression system composed of the simian virus 40 early promoter and the R-U5 segment of human T-cell leukemia virus type 1 long terminal repeat. Mol. Cell. Biol. 8:466. chain and its function in IL-5-mediated growth signal transduction. Mol. Cell. Biol. 14:7404.
-inducible gene expression but not an antiviral state. EMBO J. 15:799.[Medline]
This article has been cited by other articles:
|
|
 |
|
  K. M. Sakamoto and D. A. Frank CREB in the Pathophysiology of Cancer: Implications for Targeting Transcription Factors for Cancer Therapy Clin. Cancer Res., April 15, 2009; 15(8): 2583 - 2587. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Ouchi, Y. Tabata, K.-i. Arai, and S. Watanabe Negative regulation of retinal-neurite extension by {beta}-catenin signaling pathway J. Cell Sci., October 1, 2005; 118(19): 4473 - 4483. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Evans, S. Ariffin, A. Pierce, and A. D. Whetton Identification of primary structural features that define the differential actions of IL-3 and GM-CSF receptors Blood, October 16, 2002; 100(9): 3164 - 3174. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), BA/F-
), or BA/F-
) cells. Values are means of triplicate
samples and are expressed as relative values of maximal values obtained
by mIL-3 stimulation. B, Long-term cell proliferation in
BA/F-
