| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Obstetrics and Gynecology, Faculty of Medicine, Tokyo Medical and Dental University, Bunkyo-ku, Tokyo, Japan; and
Department of Medicine, Rheumatology Division, Howard Hughes Medical Institute, Washington University School of Medicine, St. Louis, MO 63110
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-activating
receptors and gp49B inhibitory receptor. Stimulation of macrophages
with immobilized IgG, but not with anti-FcRII/III, induced a
considerable amount of TNF-
and IL-10 production, suggesting that
the high-affinity receptor for IgG (FcRI) can transmit activating
signals in cytokine production of macrophages. Furthermore, coligation
of gp49B with FcRI resulted in the inhibition of TNF- production.
Thus, our data provide evidence that gp49B is an endogenous negative
regulator of macrophage activation and may regulate the function of
macrophages during pregnancy. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
On NK cells, these receptors fall into two general structural types, including type I membrane-oriented Ig-like molecules such as the human killer inhibitory receptors (KIRs) and type II-oriented C-type lectin-like molecules. Many of these receptors are specific for MHC class I molecules on target cells and the ITIM-bearing receptors prevent NK cell activation. However, recent data indicate that inhibitory receptors are more widely expressed, particularly those receptors related to the KIR family.
In the mouse, Ig-like receptors include the gp49 family of molecules,
gp49A and gp49B. gp49A is 89% identical to gp49B in the extracellular
domain but has a short cytoplasmic tail lacking ITIMs. Whereas the
function of gp49A is not known, gp49B is an inhibitory receptor.
Originally isolated as a murine mast cell receptor, gp49B is now known
to be present also on NK cells and macrophage-differentiated M1 myeloid
leukemia cells (3, 4, 5, 6). This receptor can suppress the
release of secretory granule mediators from mast cells when it is
coligated with the high-affinity Fc receptor for IgE (Fc
RI)
(4). Similarly, gp49B can inhibit NK cell functions
(7). Furthermore, the phosphorylated ITIMs of gp49B can
associate and activate the cytoplasmic tyrosine phosphatase SHP-1 in
IL-2-activated NK cells (8).
We previously reported that gp49 mRNA is expressed in the uterine endometrium just before implantation (6). In a subsequent study, we have found that the expression of this transcript is higher in the uterus at midgestation than around implantation. Coincident with this, it has been well established that pregnancy-related leukocytes accumulate in maternal uterine tissues up to midgestation. Although these maternal leukocytes normally act in defense against certain microbial and parasitic infections, they may play a homeostatic role in the control of cytokine balance and placentation while being suppressed in their effector functions on fetal tissues (9 , 10). The most abundant leukocyte in the midgestation uterus is the NK cell. Indeed, HLA-G-specific inhibitory receptors on human NK cells are postulated to protect fetal trophoblasts from NK cell-mediated lysis since trophoblasts do not express classical MHC class I molecules but do express nonclassical HLA-G molecules (11). We therefore postulated that murine uterine NK cells express gp49B. To our surprise, gp49 is expressed by uterine macrophages and not NK cells. We further determined that gp49B can inhibit macrophage function. Thus, we suggest that this receptor down-regulates macrophage function during pregnancy.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
C57BL/6 mice were obtained from Sankyo Laboratory Service (Tokyo, Japan). Female mice at 10–16 wk of age were mated overnight and checked for vaginal plugs in the morning. The morning a vaginal plug was found was designated as day 1 of pregnancy. Mice were killed by cervical dislocation, and uteri were removed at the time points indicated. One uterine horn was cut open, and embryos, placentas, and uteri were collected for RNA preparation, the other one was processed as individual implantation sites, embedded in Tissue-Tek OCT compound (Sakura, Tokyo, Japan) and stored at -80°C until cryostat sectioning.
Northern blot analysis and RT-PCR
Total RNA was prepared using Isogen (Nippon Gene, Tokyo, Japan). Twenty micrograms of total RNA was separated in a denaturing gel containing 2.2 M formaldehyde and transferred onto a Biodyne A membrane (Biodyne, Pall BioSupport, NY) by vacuum blotting, and then cross-linked by UV irradiation in a Stratalinker (Stratagene, La Jolla, CA). The gp49 probe-represented nt 631-1168 of gp49B1 cDNA was labeled with a random primer labeling kit (Amersham, Arlington Heights, IL), and hybridization was performed in QuikHyb solution (Stratagene) for 1 h at 68°C. To identify which isotype of gp49, gp49A, or gp49B is expressed, 3 µg of total RNA was reverse transcribed using Superscript first-strand cDNA synthesis kit (Life Technologies, Rockville, MD) followed by PCR amplification. Amplification was performed in a GeneAmp PCR System 9700 for 25–28 cycles (30 s at 95°C, 30 s at 58°C, and 30 s at 72°C) using a set of primers as follows: forward, GTTGGAACCCACAAAATGAAGACC; and reverse, TGGGCGTA(C/G)ACAATTCCCTG.
Amplified DNA was separated in 25% acrylamide gel followed by ethidium bromide staining.
In situ hybridization
In situ hybridization was performed as described previously (12). To generate antisense and sense probes, a 1166-bp fragment of gp49B1 cDNA (nt 3–1168) cloned into pSPT18 and pSPT19 plasmids were linearized with EcoRI and HindIII, respectively, and transcribed using Sp6 RNA polymerase, followed by limited alkaline hydrolysis, to reduce their average size into 200 bp.
Cell preparation and flow cytometry
Uterine mononuclear cells were prepared as follows. Uterine
horns were excised from C57BL/6 mice on day 13 of pregnancy. Each horn
was opened along the antimesometrial side and placentas were striped
off from the uterine myometrium with a fine forceps. Metrial glands
remaining on uterine strips were peeled off with a scalpel, minced in
fine pieces, and shaken in 5 mM EDTA-PBS. Dispersed cells were obtained
by filtration through a 38-µm pore size stainless steel mesh. After
hemolysis and centrifugation in 40% Percoll, the uterine
mononuclear cells collected at the bottom were subjected to flow
cytometry. The cells were first incubated with mouse IgG (10 µg/ml)
to block Fc-mediated binding except for staining of Fc receptors and
then stained with other Abs described below. Anti-gp49 (G2.3) was
generated as described previously and labeled with FITC. This Ab reacts
with both gp49A and gp49B (7). Anti-gp49 (B23.1), a gift
from P. A. LeBlanc (University of Alabama, Tuscaloosa, AL), also
recognizes both gp49A and gp49B (4, 7). PE-conjugated
anti-NK1.1 (PK136), PE-streptavidin, and isotype controls were
purchased from PharMingen (San Diego, CA). Biotinylated anti-F4/80
(A3-1) was purchased from Serotec (Oxford, U.K.). APC-conjugated mouse
IgG2a was purchased from Cedarlane Laboratories (Hornby, Ontario,
Canada). A total of 2 x 104 cells was
counted and analyzed by FACScalibur (Becton Dickinson, Mountain View,
CA) after gating out dead cells with propidium iodide staining.
Cell culture and stimulation
Bone marrow cells were aseptically flushed from femurs with PBS.
After hemolysis, cells were cultured for 5–7 days in the conditioned
medium of the metrial gland prepared as follows. Briefly, metrial
glands separated from a mouse on day 13 of pregnancy were minced and
cultured in 20 ml of RPMI 1640 medium supplemented with 10% FCS,
penicillin-streptomycin, and 5 x 10-5 M
2-ME for 24 h, and the supernatant was passed through a 0.8-µm
filter to remove cell debris. Bone marrow-derived macrophages thus
obtained were stimulated with immobilized mouse IgG or 2.4G2
(anti-FcRII/III). Mouse IgG or 2.4G2 was immobilized on 96-well
flat-bottom culture plates (Sumilon, Tokyo, Japan) by incubating for
2 h at 37°C. After stimulation for 24 h, TNF- and IL-10
in the supernatant were measured by ELISA using a Quantikine M kit
(Genzyme, Minneapolis, MN). To examine the inhibitory effect of gp49B,
the cells were stimulated with immobilized IgG plus B23.1 or IgG plus
isotype control rat IgM. In some experiments, mAb G2.3 and its isotype
control hamster IgG were used. All data are presented as means ±
SEM from triplicate cultures.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Starting on day 7 of pregnancy, we examined the expression of gp49
mRNA in the uterus of midgestation by Northern blot analysis (Fig. 1
A). The levels of its
expression increased gradually, with the highest level around days
12–14 of pregnancy. However, the two isoforms of gp49, gp49A and
gp49B, are >90% homologous at the nucleic acid level and they
cannot be distinguished by our Northern blot probe. To identify which
isoform is expressed in the uterus, we performed competitive RT-PCR
using a set of primers that can amplify both gp49A and gp49B fragments
from the 3' end of exon 7 to the 5' end of exon 8. These two fragments
were different in size because of the deletion of two cytosines in
gp49B. Both gp49A and gp49B transcripts were detected in the uterus in
almost equal amounts (Fig. 1B). In addition, coexpression of
these two isoforms was detected in bone marrow-derived mast cells.
Thus, gp49A and gp49B transcripts are expressed in the pregnant
uterus.
|
To determine the cellular location of gp49 mRNA, we examined
sections of uteroplacental units on day 13 of pregnancy by in situ
hybridization. With the antisense probe, a strong signal was detected
in the metrial gland also known as the mesometrial lymphoid aggregate
of pregnancy (Fig. 2). The sense probe
detected no signal (data not shown). These results are consistent with
the data shown in Fig. 1
in which higher expression of gp49 was
detected in the extracted mRNA of uterus than placenta, indicating that
the major portion of the metrial glands remains on the uterus when
individual implants are peeled off.
|
A). Next, we stained metrial
gland cells with G2.3 and anti-F4/80 (a marker of mouse
macrophages). When cells with high autofluorescence were gated out,
84.9% of the uterine mononuclear cells were double positive for gp49
and F4/80. By contrast, gp49 single positive or F4/80 single positive
cells were 2.4 and 7.2%, respectively. These data suggest that gp49 is
present on uterine macrophages (Fig. 3B). Furthermore, the
cells could be stained also with B23.1, another anti-gp49 mAb,
although the intensity of its staining was lower than that with G2.3
(data not shown). Since G2.3 and B23.1 react with both gp49A and gp49B
transfectants by flow cytometry, the levels of surface expression of
each isoform could not be determined. However, it is reasonable to
conclude that the gp49B inhibitory receptor is present on the surface
of uterine macrophages because gp49B transcripts were present.
|
Functional studies of uterine macrophages are technically
difficult because of the small amount of cells. Fortunately, in the
course of our studies, we found that gp49 was constitutively expressed
on myeloid cells in bone marrow. We therefore speculated that
immigrants of these myeloid precursors might differentiate into
macrophages in the presence of factors secreted from the metrial gland.
Previous studies demonstrated that several factors, including M-CSF,
were detected in the conditioned medium of metrial gland explants
(13), supporting that the supernatant can induce
differentiation of macrophages and such cells would be useful for
further functional studies. When bone marrow cells were cultured in
metrial gland conditioned medium for 5 days, the cells increased 3-fold
in number. When gated on the small cells with low side scatter (Fig. 4A), 
80% of them were
gp49+F4/80+ and about 70%
of the cells showed high levels of F4/80 staining (Fig. 4B).
F4/80low gp49+ cells were
detected also in bone marrow before culture. In addition, most of the
large cells with high side scatter were positive for gp49 and all of
them showed high levels of F4/80 staining (Fig. 4C).
Concerning the intensity of staining, the large cells showed slightly
lower levels of gp49 and F4/80 staining than the small cells. On the
other hand, when we stained the cells with APC-labeled mouse IgG2a to
examine the expression of Fc receptors (Fig. 4, D and
E), the large cells showed higher levels of staining than
the small cells, although both of them showed heterogeneous intensity
of staining. Thus, the bone marrow-derived macrophages express gp49 and
Fc receptors, although expression levels are different depending on
their activation.
|
production by macrophages inhibited by gp49B
Activated macrophages can produce various cytokines such as
TNF-, IL-1, IL-6, IL-10, and IL-12. Bone marrow derived-macrophages
were generated as described above and their production of TNF- and
IL-10 was examined after Fc cross-linking. When cells were stimulated
with immobilized IgG (10 µg/ml), they produced a considerable amount
of TNF- (868 ± 12.3 pg/ml) and IL-10 (158 ± 2.2 pg/ml).
In contrast, immobilized 2.4G2 (anti-FcRII/III) had no effect,
suggesting that the high-affinity but not low-affinity Fc receptor for
IgG can transmit the activating signal for cytokine production in
macrophages.
We next tested whether gp49B functions as an inhibitory receptor on
macrophages. Since it is generally accepted that inhibitory receptors
require not only binding to their ligands but also simultaneous
stimulation of nearby activating receptors for an exerting inhibitory
effect, we tested whether coligation of gp49B with FcRI inhibited
TNF- production. To provide continuous coligation, IgG and B23.1
anti-gp49 mAb were immobilized together, and bone marrow-derived
macrophages were cultured on them. Inhibition was evaluated by
comparing TNF- production of the cells stimulated with IgG plus
B23.1 with that of the cells stimulated with IgG plus isotype control
rat IgM. As shown in Fig. 5, the
production of TNF- was reduced in a dose-dependent manner when the
cells were stimulated on the plate immobilized with the fixed
concentration of IgG and incremental concentrations of B23.1. About
20% inhibition of TNF- production was detected when the cells were
stimulated at the ratio of 1:3 with IgG and B23.1. On the other hand,
the production of TNF- was nearly constant even when immobilized
control rat IgM was increased. Although B23.1 recognizes both gp49A and
gp49B (7), the inhibitory effect is most likely mediated
through gp49B since it has an ITIM whereas gp49A does not, and attempts
to cross-link isolated gp49A had no obvious functional effect (data not
shown). Thus, gp49B functions as an inhibitory receptor on
macrophages.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Macrophages constitute another major population of uterine leukocytes.
They increase in number during pregnancy and are localized in the
metrial gland by midgestation (20). The production of
M-CSF in the uterus in midgestation is >1000-fold of that in the
nonpregnant uterus, indicating that the uterine environment during
pregnancy is suitable for differentiation of macrophages
(21). Various roles have been proposed for uterine
macrophages, including immunosuppression, Ag presentation, cytokine
secretion, and phagocytosis, but little has been well elucidated so
far. To determine the significance of uterine macrophages, the
fertility of M-CSF-deficient mice (op/op mice) has been
examined. Although the females show a low rate of pregnancy success,
this is due to failure of ovarian follicular development and ovulation.
In addition, the phenotype of gross deficiency of macrophages in the
uteri of op/op mice is improved by gestation day 7,
suggesting that other factors can substitute for M-CSF (22, 23). On the other hand, in the CBA x DBA/2 model with high
rates of spontaneous abortion, increased infiltration of activated
macrophages expressing class II Ag can be detected in maternal decidua
of 20–30% embryos before the first macroscopically detectable signs
of resorption (24). In addition, the decidual cells at
resorbing sites show increased expression of inducible NO synthase and
TNF-, suggesting that these cytotoxic mediators produced by
activated macrophages may lead to embryo death (25, 26).
Therefore, uterine macrophages may have critical functions for
successful pregnancy and activated uterine macrophages may be
responsible for embryo loss.
Fc cross-linking of macrophages in vitro contributes to the activation
of a wide variety of effector functions, including phagocytosis,
Ab-dependent cellular cytotoxicity, and release of inflammatory
cytokines. Among these functions, we examined cytokine production
because it is likely that the cytokine microenvironment in the uterus
plays a key role in the maintenance of pregnancy. Th2-type cytokines
predominate at the maternal-fetal interface, which contributes to the
fetal survival by inhibiting Th1 response (27). Activation
of macrophages upon cross-linking of Fc receptors may reverse the
balance by producing proinflammatory cytokines, which ultimately may
lead to embryo death. In some cases of recurrent spontaneous abortion,
high titers of self-reactive Abs including antiphospholipid Ab and
antinuclear Ab, are detectable in serum (28). These
pathogenic self-reactive Abs in the form of immune complexes can
cross-link macrophage Fc receptors, which may result in the production
of inflammatory cytokines. In our assay using bone marrow-derived
macrophages, a considerable amount of TNF- was produced by Fc
cross-linking. Although simultaneous production of IL-10 was detected,
it was not sufficient for suppressing the production of TNF-.
It would therefore be important to down-modulate the potential
production of cytokines by uterine macrophages. In the studies
presented here, we examined whether gp49B could function as an
endogenous negative regulator of macrophage production of inflammatory
cytokines. Previous studies by Katz et al. (4) have
demonstrated that the gp49B on the surface of bone marrow-derived mast
cells inhibits the FcRI-induced degranulation response. Consistent
with the inhibitory function on mast cells, coligation of gp49B with
FcRI by immobilized B23.1 and IgG resulted in a B23.1-dependent,
dose-related inhibition of TNF- production of macrophages. This
inhibition was not the result of a decrease in the amount of
immobilized IgG as a result of IgM addition, because in preliminary
studies we confirmed that the amount of immobilized IgG was nearly
constant when fixed in the conditions used here. In fact, the evidence
is that the TNF- production of the cells stimulated with mouse IgG
and various amount of isotype control rat IgM was constant. Instead of
B23.1, another anti-gp49 G2.3 showed no inhibitory effect (data not
shown). We suspected that these discrepant results might be due to the
difference in the isotype of the mAbs. B23.1 is an IgM isotype and
therefore it does not bind FcRI, whereas G2.3 may have associated
also with FcRI through its Fc portion since its isotype is IgG.
Thus, gp49B on macrophages can inhibit proinflammatory cytokine
production and thus may be relevant to control of macrophage activity
during pregnancy.
It seems that both gp49A and gp49B messages are expressed in uterine
macrophages, although this may not necessarily reflect the surface
expression of each molecule. Macrophages have another pair of
receptors, paired Ig-like receptor (PIR) A and PIR-B, which are very
homologous in their extracellular domains but distinct in cytoplasmic
domains. PIR-A may function as an activating receptor in association
with the FcR chain (29) whereas PIR-B may mediate an
inhibitory signal through its own ITIMs. The coexistence of PIR-A and
PIR-B suggests the regulation of macrophage activation through their
interaction with the putative common ligand. Assuming that gp49A is an
activating receptor, paired gp49 receptors may also regulate macrophage
activation by themselves. Although the function of gp49A has not yet
been understood, the functional analysis of gp49A will be crucial to
elucidate whether the relative expression level of gp49A and gp49B on
cell surface contributes to the regulation of cellular activation.
In conclusion, gp49B inhibitory receptor is present on macrophages
accumulating in the uterine metrial gland in midgestation. Coligation
of gp49B with FcRI inhibits FcRI-mediated TNF- production of
macrophages in vitro. This finding suggests that gp49B may be an
endogenous negative regulator of inflammatory cytokine production,
which may contribute to the pathophysiology of inflammatory autoimmune
disorders including recurrent spontaneous abortion caused by immune
complexes.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Yukie Matsumoto at her current address: Human Gene Science Center, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8510, Japan.
3 Abbreviations used in this paper: ITIM, immunoreceptor tyrosine-based inhibitory motif; KIR, killer inhibitory receptor; PIR, paired Ig-like receptor.
Received for publication July 10, 2000. Accepted for publication October 20, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
26 mice. J. Exp. Med. 187:217. chain. J. Exp. Med. 188:991.This article has been cited by other articles:
|
|
 |
|
  S. Iida, T. Kohro, T. Kodama, S. Nagata, and R. Fukunaga Identification of CCR2, flotillin, and gp49B genes as new G-CSF targets during neutrophilic differentiation J. Leukoc. Biol., August 1, 2005; 78(2): 481 - 490. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. S. Zhou, D. S. Friend, A. M. Feldweg, M. Daheshia, L. Li, K. F. Austen, and H. R. Katz Prevention of Lipopolysaccharide-induced Microangiopathy by gp49B1: Evidence for an Important Role for gp49B1 Expression on Neutrophils J. Exp. Med., October 20, 2003; 198(8): 1243 - 1251. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ichiba, T. Teshima, R. Kuick, D. E. Misek, C. Liu, Y. Takada, Y. Maeda, P. Reddy, D. L. Williams, S. M. Hanash, et al. Early changes in gene expression profiles of hepatic GVHD uncovered by oligonucleotide microarrays Blood, July 15, 2003; 102(2): 763 - 771. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Daheshia, D. S. Friend, M. J. Grusby, K. F. Austen, and H. R. Katz Increased Severity of Local and Systemic Anaphylactic Reactions in gp49B1-deficient Mice J. Exp. Med., July 16, 2001; 194(2): 227 - 234. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or with the same concentration of immobilized IgG + isotype
control rat IgM (
) for 24 h. The amount of TNF-