 |
INTRODUCTION |
There is compelling evidence for a primary pathogenic role
of immune complexes (ICs)1 in
many immunological diseases, such as systemic lupus erythematosus, rheumatoid arthritis, Goodpasture's syndrome, vasculitis, and glomerulonephritis (1-4). Although it had been assumed for a long time
that IC diseases are mediated mainly by complement activation, the
mechanisms by which IC formation triggers inflammation are still not
fully understood. Spontaneous and induced mouse models of
IC-dependent glomerular injury reveal an involvement of
various complement components, including C1q, C3, and C5a (5-7).
However, studies in FcR
chain-deficient mice define another
potential pathway responsible for kidney damage in autoimmune
glomerulonephritis requiring activating Fc receptors for IgG,
FcRs (8). The demonstration that both FcRs and complement are
essential in the experimental Arthus reaction suggests that the
initiation of glomerulonephritis may also depend on the combined action
of these two prominent effector systems (9-11).
Three classes of FcR are known on murine leukocytes: the high
affinity receptor, FcRI, and the low affinity receptors, FcRII and FcRIII (12). Activating FcRI and FcRIII are expressed in
association with the FcR chain essential for triggering their various effector functions including antibody-dependent
cell-mediated cytotoxicity, phagocytosis, and the production of
proinflammatory molecules (13-15). In contrast, the two major isoforms
of FcRII, FcRIIb1 and FcRIIb2, apparently lack such activating
capacity but can inhibit FcR-dependent activation
signals when co-expressed on the same effector cell (16). This balance
between activating/inhibitory FcRs has been established in various
murine models and is highlighted by an increased susceptibility of
FcRII (
/) mice to the pathogenic effects of antibodies in
IC-triggered inflammation and autoimmune disease (17-20).
Glomerular mesangial cells (MC) are bone marrow-independent
mesenchymal cells related to vascular smooth muscle cells (21, 22).
Their contractility controls the rate of glomerular filtration and they
are implicated in inflammatory and pathogenic processes of glomerular
injury. When activated, MC are able to produce a variety of mediators,
including nitric oxide, oxygen radicals, chemotactic cytokines, and
several growth factors (23-25). This is accompanied by the induction
of specific receptors, which enable them to form autocrine
proliferative loops (26). Human MC, which are normally FcR-negative,
can also be stimulated to express the activating receptors, FcRI and
FcRIII. This induction of functional FcRs may provide an
important mechanism of initiation and progression of chronic glomerular
inflammation by immune complexes (27, 28).
In the present study, we examined the capacity of mouse glomerular MC
to express individual FcRs by using primary cultures of MC
established from normal and FcR-deficient mice. Moreover, we analyzed
the role of the cytokines IFN-, TNF
, and IL-1
in the induction
and/or down-regulation of the low affinity receptors FcRII and
FcRIII on MC and in the IC/FcRIII-dependent
generation of various proinflammatory mediators known to be involved in
the pathogenesis of glomerulonephritis. By using this strategy, we first identified the b2 isoform of FcRII to be expressed on
resting MC and normal kidney tissue. Second, we showed a requirement of IFN- in changing the expression profile from inhibitory FcRII to
activating FcRIII. Third, the induction of activating FcRIII on
proliferating MC and the subsequent interaction with ICs triggered the
enhanced synthesis of chemoattractants, including CC and CXC subfamily
chemokines RANTES, MCP-1, MCP-5, and KC, respectively. Given the
established function of these chemokines in renal IC inflammation
(29-31), alterations in the levels of locally expressed inhibitory and
activating FcRs might be of pathogenic importance in glomerular
disease in mice. This was verified in an experimental model of acute
anti-GBM nephritis in which early renal mediator production and
recruitment of neutrophils correlated with suppression of FcRII and
induction of glomerular FcRIII. Moreover, the finding of increased
MCP-1, MIP-2, and KC levels and augmented neutrophil influx upon local
FcRII-blockade suggested that the function of the inhibitory
FcRII and its down-regulation on mesangial cells during inflammation
is critical in kidney pathology.
| |
EXPERIMENTAL PROCEDURES |
Mice
The generation of FcRII (/), FcRIII (/) and FcR
(/) mice derived from embryonic stem cells with 129 origin
and back-crossed for 8-12 generations with C57BL/6 mice has been
described previously (13, 14, 16). Mice double deficient for FcRII
and FcR chain (FcRII/FcR/) were purchased from Taconic
(Germantown, NY). C57BL/6 mice were obtained from Charles River
Laboratories (Sulzfeld, Germany). All these mice were used at
14-18 weeks of age for immunohistochemical analysis of kidney tissues
(see below). All animal experiments were conducted in accordance with
the regulations of the local authorities.
Immunohistochemistry of Mouse Kidney Tissue
To analyze the expression of individual FcRs within the
glomerulus, kidney tissue of FcR-deficient and C57BL/6 mice treated or not with increasing doses (50-500 ng) of intraperitoneal
Escherichia coli LPS (Sigma) were processed for
histological examination by immunocytochemical techniques as described
(18, 32). FcR-positive staining was determined by incubating
cryostat sections for 30 min with the anti-FcRII/III mAb 2.4G2
followed by incubation with the bridging antibody (Z0494) and the
rat-APAAP antibody complex (D0488) for 30 min. The last two steps were
repeated for 15 min followed by visualization using a Fast Red
detection kit (Dako). All slides were counterstained with hematoxylin
and mounted in Glycergel (Dako).
Mouse Glomerular Mesangial Cell Preparation and
Characterization
Preparation--
Glomerular mesangial cells from C57BL/6
wild-type, FcRIII-deficient, and FcR-deficient mice were
prepared, characterized, and cultured according to the protocol
described (23, 33). For each of the three independent preparations, 12 mouse kidneys were pooled. The kidneys were dissected from 6-week-old
mice; the capsule and medulla were removed and the cortex cut into
slices and rinsed. The material was sequentially pushed through 125- and 180-µm stainless steel sieves. The glomeruli separate from tubuli
and Bowman capsules were recovered in a sieve with a 60-µm mesh
screen and washed. By this procedure we obtained microscopically 95%
pure, decapsulated glomeruli free of tubular contamination. These steps
were followed by digestion with collagenase (1 mg/ml, type CLS4, 184 units/mg (Worthington)) for 30 min at 37 °C with shaking of the
mixture every 5 min. After two washing cycles the digested glomeruli
were seeded in 25-cm2 plastic culture flasks (Nunc) in a
low volume (2 ml) of culture medium (see below) with 20% FCS to permit
adherence. These conditions resulted in a predominantly mesangial
outgrowth starting at day 6-12. Cells not resembling the typical
growth in interwoven bundles were scraped off and removed together with
the glomerular debris by gentle washing with fresh medium. Cells were
then grown to confluence and switched to 10% FCS culture medium
between the second and third passage.
Characterization--
Morphologically uniform cultures of
stellate mesangial cells were used for characterization between the
second and 12th passage. Light microscopic evaluation showed one layer
of this cell type, with cell-cell contacts at discreet cell regions in
a three-dimensional fashion resembling a net-like structure.
Intercellular junctions were starting points for intracellular fibers,
which by indirect immunofluorescence stained positively for myosin (see
below). In mouse MC cultures extracellular matrix and "hillocks"
(23, 25, 33) were visible by light microscopy. There was no
morphological evidence of the presence of macrophages,
dendrite-, endothelial-, or epithelial-like cells, or
fibroblasts. However, to confirm the presence of exclusively intrinsic
mouse mesangial cells of mesodermal origin we performed a series of
biochemical, immune cytochemical, and functional tests. Briefly, these
cells were resistant to puromycin treatment (10 µg/ml, 24 h),
sensitive to mitomycin (10 µg/ml, 24 h). They stained positive
by an indirect immunofluorescence technique for smooth muscle myosin,
desmin, fibronectin, vimentin, and mouse MHC I antigen; and at
confluence, they stained positive for extracellular deposits of
collagen type IV and negative for factor VIII and cytokeratin. After
the second passage, both dendritic cells and macrophages, which may
represent 3-5% of glomerular cells in vivo, were carefully
excluded as a contaminating subpopulation in MC cultures. Repeated MHC
II antigen determinations with nonstimulated MC adherent on slides
(with and without fixation) or flow cytometry (FACS) of suspended MC with anti-mouse I-Ad antibodies up to passage 12 were
negative. Additionally, we did not detect any interferon activity in
the MC supernatant, which might have been due to dendritic cells.
Culture Conditions and Stimulation for Mouse MC and Control
Cells
Mouse MC used in passages 6-12 were cultured in RPMI 1640 medium supplemented with non-essential amino acids (1%),
L-glutamine (2 mM), sodium pyruvate (2 mM), bovine insulin (5 µg/ml), -mercaptoethanol (45 µM), and endotoxin-free FCS (10%). Serum-free culture of
mouse MC was performed using MCDB-302 medium (Sigma) with the
supplements mentioned above, but without FCS, leading to growth arrest
of MC after 48-72 h. Mouse macrophage J774 cells were used as control cells and cultured under standard conditions. All cell types were free
of any mycoplasma contamination as tested every second week by a highly
sensitive PCR analysis using previously established protocols (27). For
the stimulating conditions for mouse MC, we used incubations for
48 h with the following concentrations of the various stimuli:
500-1000 units/ml mIFN- (Roche Molecular Biochemicals), 500 units/ml mTNF (Phillips, Bissendorf, Germany), 10 ng/ml recombinant
human IL-1 (Hoffmann-La Roche), 10 ng/ml mIL-10 (TEBU GmbH,
Offenbach, Germany), 10 µg/ml E. coli LPS (Sigma).
RNA Isolation, RT-PCR Conditions, and Southern Blot Analysis
Total cellular RNA was isolated from mouse MC and J774 cells
with RNA-Clean (AGS, Heidelberg, Germany) followed by LiCl2
precipitation and finally suspended in diethyl pyrocarbonate-treated
water at a concentration of 1 µg/µl. Single-stranded cDNA was
synthesized from 1.5 µg of RNA using reverse transcriptase and
amplified by a modified Hot Start PCR protocol (27, 28) for a
submaximal number of cycles (40 cycles for FcRI, FcRII,
FcRIII, FcR chain, MCP-5, and KC and 28 cycles for -tubulin)
in a Varius VR thermocycler (Landgraf, Hannover, Germany). The primer
sets used in PCR are given in Table I. PCR products were analyzed on
1.5% agarose gels. In all experiments, gels were blotted onto nylon
membranes for Southern blot hybridization with 32P-labeled
cDNA probes (FcRI, 714-bp fragment covering EC2 to TM/C exons
(34); FcRII, 905-bp fragment covering TM to C3 exons of FcRIIb1
(35); FcRIII, 612-bp fragment covering the TM/C/UT exon (35);
FcR, KC, and MCP-5, all using their complete cDNA sequences
(36-38)) to confirm specificity.
mAbs and FACS Analysis
The following FcR-recognizing antibodies were used:
anti-FcRII/III (clone 2.4G2, rat anti-mouse IgG) (BD PharMingen,
Heidelberg, Germany) and anti-Ly17.2 (clone K9.361, mouse
anti-mouse IgG; kindly provided by Dr. U. Hämmerling,
Sloan-Kettering Institute, New York), the latter specifically binding
the FcRII antigen (18). Isotype control mAbs with irrelevant
specificities were obtained from Immunotech (Hamburg, Germany). Binding
of the mAb to the respective antigens was performed by indirect
immunofluorescence. After mild trypsinization, mouse MC
(106 cells) were incubated with the first mAb, followed by
fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse or goat
anti-rat IgG F(ab')2 fragments (Dianova, Hamburg, Germany),
and were analyzed on a FACScan flow cytometer (BD PharMingen).
Detection of Monomeric IgG2a Binding to FcRI on the
Surface of Mouse MC
Cells were seeded on glass cover slides (8-well; Nunc, Roskilde,
Denmark) and grown to subconfluency by covering the slides with 10%
FCS medium in 10-cm Petri dishes. After stimulation with 500 units/ml
IFN- and 10 µg/ml LPS for 48 h, cells were washed twice with
PBS, 0.05% Tween, 0.1% bovine serum albumin, fixed for 10 min at
20 °C with acetone/methanol (1:2), followed by three
further washes with Tris-buffered saline. Cells were then incubated
with monomeric mouse IgG2a at 37 °C for 30 min and phenotyped by the
alkaline phosphatase anti-alkaline phosphatase technique (Dako).
Activation of Mouse MC by IgG Immune Complexes
In functional experiments, mouse MC induced to express or not
the activating FcRIII were further stimulated with 50 µg/ml heat-aggregated IgG according to the methods of Santiago et
al. (39) and Hora et al. (40). After 12 h of IgG
IC activation, RNA was prepared and analyzed for chemokine mRNA
synthesis by RNase protection assay. The culture supernatants were
examined for the production of MCP-1.
RNase Protection Assay (RPA)
Assays for chemokine mRNA were conducted with the mCK-5
multiprobe template set according to the manufacturer's protocol
(RiboQuant assay kit, BD PharMingen). The assay kit allows for the
simultaneous detection of mRNA for each of the following mouse
chemokines: lymphotactin, RANTES, eotaxin, MIP-1, MIP-1, MIP-2,
IFN--inducible protein-10 (IP-10), MCP-1, TCA3, and mRNA for the
L32 and GAPDH housekeeping genes.
ELISA for MCP-1 Quantitation
Immuno-MaxiSorb 96-well microtiter plates (Nunc, Wiesbaden,
Germany) were coated with 100 µl containing 1 µg/ml goat anti-mouse MCP-1 antibody (R&D Systems, Wiesbaden,
Germany) in 100 mM sodium carbonate-bicarbonate buffer (pH
9.6) at 4 °C. After removing the excess capture Ab, the wells were
filled with 200 µl of 1% bovine serum albumin in PBS and incubated
at room temperature for 2 h to saturate excess binding sites.
After three washes with PBS, serial dilutions of the experimental
samples diluted in 1% bovine serum albumin/PBS were added to the
plates and incubated at 4 °C overnight. After three washes, 100 µl
of biotinylated detector anti-MCP-1 Ab was added to the wells and
incubated for 4 h at room temperature. After three additional
washes, the plates were incubated with peroxidase-conjugated
streptavidin for 2 h. After three final washes, plates were
developed with hydrogen peroxide and tetramethylbenzidine (Sigma) and
stopped by the addition of 1 M
H2SO4. Titrations of recombinant mouse MCP-1
(R&D Systems) were included in each experiment for standardization.
Experimental Anti-GBM Nephritis
Rabbit anti-mouse glomerular basement membrane (GBM) IgG
antiserum was kindly provided by P. Heeringa. After decomplementation by heating at 56 °C for 30 min, 300 µl of anti-GBM antiserum was injected intravenously via the tail vein of ether-anesthetized mice.
This concentration was effective to induce significant glomerular neutrophil influx in the early heterologous phase (2-4 h) of acute anti-GBM nephritis (58). In some experiments, glomerular FcRII was
selectively blocked (30 min before anti-GBM treatment) by intracapsular
injection of the neutralizing anti-FcRII Ly17.2 mAb (40 µg in 20 µl of sterile PBS) in one of the two kidneys per mouse. Mice were
killed at 2 h after initiation of anti-GBM nephritis, and kidneys
were assayed for glomerular expression of FcRs, MCP-1
mRNA/protein production, MIP-2/KC CXC chemokine release, and
neutrophil accumulation. The levels of FcR and MCP-1 mRNA in
kidneys normalized to tubulin were quantitated by TaqMan real-time PCR using the following FcRII-, FcRIII-, MCP-1-, and -tubulin-specific primers and probes (FcRII: sense,
5'-AGTCTCCCTTGGCATTGGG-3', antisense, 5'-AGCATCCCTTGGACCAGGA-3',
probe, 6-FAM-AAAGCAAGCCAGAAAGGCCAGGATCTAGT-TAMRA; FcRIII: sense,
5'-TGCAGCTCTTCCGAAGGCT-3', antisense, 5'-TGTCTTCCTTGAGCACCTGGA-3', probe, 6-FAM-TGGTGAAACTGGACCCCCCATGG-TAMRA; MCP-1: sense,
5'-CCAACTCTCACTGAAGCCAGC-3', antisense, 5'-CAGGCCCAGAAGCATGACA-3',
probe, 6-FAM-CTCTCTTCCTCCACCACCATGCAGGT-TAMRA; -tubulin: sense,
5'-CACCATGAGCGGCGTCA-3', antisense, 5'-TTCGAAGGTCAGCATTAAGCTG-3', probe, 6-FAM-ACCTGCCTCCGTTTCCCGGG-TAMRA). Myeloperoxidase (MPO) serves
as marker of neutrophil infiltration into kidney tissue and was assayed
as described previously (9). In brief, homogenized tissue was suspended
in 50 mM potassium phosphate buffer, pH 6, 0.5%
hexadecyltrimethylammonium bromide, subsequently exposed to three
freeze-thaw cycles, and finally sonicated. A total of 0.167 mg/ml
o-dianisidine dihydrochloride and 0.0005% hydrogen peroxide was added to the supernatant. The change in OD at 
= 450 nm was recorded. A serial dilution of MPO from human
polymorphonuclear leukocytes (Calbiochem) served as a standard. Samples
were run in duplicate. The concentrations of MCP-1, MIP-2, and KC in
supernatants of homogenized kidneys were assayed in duplicate with
MCP-1, MIP-2, and KC-specific ELISA kits (R&D Systems) according to
the manufacturer's instructions.
| |
RESULTS |
Constitutive Expression of the Inhibitory FcRII in Mouse
Kidney--
FcR expression in the glomerulus was first
determined by immunohistochemistry of kidney tissue sections obtained
from FcR-deficient and C57BL/6 control mice. Weak anti-FcRII/III
2.4G2 mAb staining detectable in the kidneys of B6 mice was not
observed in FcR/FcRII (/) mice lacking surface expression of
all known Fc receptors, e.g. FcRI, FcRII, and
FcRIII (Fig. 1A). The
comparison of FcRIII (/) and FcRII (/) mice identified
low constitutive expression of FcRII but not FcRIII in normal
kidney (Fig. 1A). In FcRII (/) mice,
anti-FcRII/III 2.4G2 mAb stainings became evident after
intraperitoneal injection of 500 ng of LPS (Fig. 1B),
indicating de novo renal expression of FcRIII. This
glomerular appearance of FcRIII may be based on either infiltrating
FcRIII-positive cells and/or induction of FcRIII on normally
FcRIII-negative resident kidney cells.
View larger version (76K):
[in this window]
[in a new window]
|
Fig. 1.
Different expression of low affinity
Fc receptors (FcRII
and FcRIII) in kidney as assessed by
immunohistochemistry. Kidney tissue sections from the indicated
mice were stained with the anti-FcRII/III-specific 2.4G2 mAb and
visualized using a Fast Red detection kit. A, staining
intensity was similar in C57BL/6 and FcRIII (/) mice but was
absent in FcRII (/) and FcRII/FcR double-deficient mice
(original magnification ×400). B, renal cortical tissue of
FcRII (/) mice stained positive for 2.4G2 after intraperitoneal
injection of 500 ng LPS (original magnification ×1000).
|
|
FcRII Gene Transcription in Resting Mouse MC under Basal and
Stimulated Conditions--
We previously described a protocol in the
generation of resting human MC (27, 41). This model system was
therefore adopted to similarly induce growth arrest of mouse MC for up
to 7 days in serum-free MCDB-302 medium without showing any
irreversible phenotypic changes. Resting mouse MC were then incubated
with or without IFN-, IL-1, TNF, TNF/IL-1, IL-10, and
LPS for 48 h. To quantitate mRNA expression of Fc
receptors, we performed RT-PCR with various FcRs-specific primer
pairs (Table I) and subsequent Southern
blot hybridization. RT-PCR was controlled by -tubulin, which is not
influenced by activation of MC, and RNA from J774 cells served as a
positive control. Under basal conditions, resting mouse MC did not
express any mRNA for activating FcRI and FcRIII (data not
shown). In contrast, RT-PCR products of the expected size, specific for
FcRIIb2 transcripts, could be amplified. Comparison with both the
-tubulin control on MC and FcRII-specific mRNA in J774 cells
indicated, however, that the abundance of FcRIIb2 was low (Fig.
2). From the stimuli mentioned above, the
combination of IL-1 and TNF up-regulated FcRII gene transcription, whereas IFN- and IL-10 led to a complete
down-regulation (Fig. 2). When given alone, LPS, IL-1, and
TNF had only minor effects (Fig. 2, and data not shown). These
results suggest that MC express basal levels of FcRIIb2 mRNA
under nonproliferating conditions with no mRNA expression for
activating FcRs.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 2.
FcRIIb2 mRNA
expression is adversely regulated by cytokines in resting MC.
After 48 h of serum-free culture, growth-arrested MCs were either
kept in serum-free medium or treated with the indicated stimuli for an
additional 48 h. Per sample, 1.5 µg of total cellular RNA was
subjected to RT-PCR and Southern blot hybridization. RNA from J774
cells served as a positive control. Ratios between the FcRIIb2 and
-tubulin hybridization signals were determined using a Phosphoimager
(Fuji) (the value obtained from the medium sample was set to 1 relative unit).
|
|
Proliferating Mouse MC Express FcRII Protein, Which Is
Differently Regulated by IFN- and IL-1/TNF--
We recently
showed by FACS analysis that antibodies detecting the mouse Ly-17.1/2
alloantigen system are specific for FcRII with no cross-reactivity
to FcRIII (18). Now, we used the FcRII(Ly17.2)-specific mAb,
K9.361 (42), to examine FcRII surface expression of mouse MC kept
under proliferating conditions and stimulated cells with either
IFN- or a combination of TNF and IL-1. As shown in Fig. 3A, proliferating unstimulated
MC revealed surface FcRII protein at a low density that was
substantially up-regulated by IL-1/TNF. In contrast, MC became
FcRII-negative after exposure to IFN- for 48 h (Fig.
3B). These data were consistent with the RT-PCR analysis
(Fig. 2) and demonstrated that the expression of inhibitory FcRII on
glomerular mesangial cells is differently regulated by different
cytokines.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 3.
Surface expression of
FcRII on mouse MC is differently regulated by
TNF/IL1-
(A) and IFN-
(B). Proliferating cells were cultured for
48 h under 10% FCS medium conditions with or without the
indicated stimuli. Surface expression of FcRII was analyzed by
staining with anti-FcRII (Ly17.2)-specific mouse IgG mAb and
FITC-conjugated goat anti-mouse F(ab')2. The fluorescence
intensity of 10,000 cells was determined using a FACScan. The
unfilled lines represent a negative control mAb.
|
|
IFN--dependent Down-regulation of Inhibitory
FcRII Is Accompanied by Induced Expression of Activating FcRs on
Cycling Mouse MC--
IFN- is capable of inducing or increasing
activating FcRI and FcRIII receptors on proliferating human MC
(27, 28). In mouse MC, low amounts of FcRI, FcRIII, and FcR
chain mRNAs were detected upon proliferation (Fig.
4). Both IFN- and LPS, when given
alone, increased mRNA expression (data not shown). The maximal
stimulation with a combination of IFN-/LPS (resulting again in
abrogated FcRIIb2 mRNA) markedly activated FcR chain transcription, suggesting a coupling between
IFN--dependent FcRII down-regulation and induced
expression of FcR chain-associated FcRs, FcRI and FcRIII,
on mouse MC (Fig. 4). IFN--dependent induction of
FcRIII protein was demonstrated by FACS analysis with the
anti-FcRII/III 2.4G2 mAb (Fig. 5). The
appearance of FcRI was examined more indirectly due to the lack of
anti-FcRI-specific mAbs. Immunohistochemical analysis showed
abundant staining of FITC-labeled monomeric IgG2a on IFN- stimulated
mouse MC with no staining on either unstimulated or
TNF/IL-1-treated MC (Fig. 6).
Because FcRI is well established as interacting efficiently with
monomeric IgG2a (12, 34), these findings may indicate that IFN-
induces, in addition to FcRIII, surface membrane expression of
FcRI.
View larger version (63K):
[in this window]
[in a new window]
|
Fig. 4.
IFN-/LPS-dependent
down-regulation of FcRII is accompanied by the
strong induction of mRNA for the FcR chain
in proliferating MC. After 48 h of culture in 10%
FCS-containing medium, proliferating MCs were treated or not with the
indicated stimuli for an additional 48 h. Per sample, 1.5 µg of
total cellular RNA was subjected to RT-PCR and subsequent Southern blot
hybridization using -tubulin- and FcR-specific primer sets and
cDNAs (see Table I) and analyzed using a Phosphoimager (Fuji). RNA
from J774 cells was used as a positive control.
|
|
View larger version (10K):
[in this window]
[in a new window]
|
Fig. 5.
IFN- induced FcRIII surface
expression on proliferating MC. Cells were cultured for 48 h
under 10% FCS medium conditions with or without 500 units/ml IFN-.
Surface expression of FcRIII was analyzed by staining with the
anti-FcRII/III-specific 2.4G2 rat IgG and FITC-conjugated goat
anti-rat F(ab')2. The fluorescence intensity of 10,000 cells was determined using a FACScan. The unfilled lines
represent a negative control mAb.
|
|
View larger version (90K):
[in this window]
[in a new window]
|
Fig. 6.
IFN-/LPS induced
binding of monomeric IgG2a to proliferating MC. Mouse MC were
grown to subconfluency on glass cover slides in 10% FCS culture medium
for 48 h, and cells were stimulated or not with IFN-/LPS or
TNF/IL-1 for an additional 48 h. After fixation, MC were
incubated with monomeric IgG2a, and binding was visualized by the
alkaline phosphatase anti-alkaline phosphatase technique using
Fast Red as the substrate for alkaline phosphatase. Positive and
negative controls produced the expected results (original magnification
×400).
|
|
IgG Immune Complex Activation of Mouse MC--
To
investigate the functional relevance of FcR expression in the
activation of mouse MC, the effects of IgG ICs on the production of
chemokines were measured by RPA, RT-PCR analysis, and ELISA. After
induction of activating FcRs by IFN- or IFN-/LPS (48 h),
proliferating MC were subjected for 12 h to 50 µg/ml
heat-aggregated IgG ICs. IFN- itself or in combination with LPS
induced a substantial accumulation of mRNA for MCP-1 and IP-10 but
not lymphotactin, eotaxin, MIP-1, or MIP-1 in RPA (Fig.
7A, and data not shown), for MCP-5 and KC in RT-PCR (Fig. 7B) but only
slightly for RANTES in RPA (Fig. 7A). The
interaction of activating FcRs with ICs led to the specific
induction or increased synthesis of RANTES or MCP-1, MCP-5, and KC
transcripts, respectively (Fig. 7). Moreover, MCP-1 protein levels of
about 60 ng/ml were detected in the supernatants of MC cultures from
control but not FcR chain (/) mice (Fig. 7C),
suggesting that increased expression of FcR chain is required for
IC-triggered MCP-1 release through activating FcRs, FcRIII and/or
FcRI. In support, IgG IC activation of MC that had been stimulated
by TNF/IL-1 to express mainly FcRII did not result in such an
enhanced chemokine production throughout all of the experiments (Fig.
7, A and C).
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 7.
FcRI/III-dependent chemokine
production by MC. Proliferating MC from C57BL/6 or FcR (/)
mice were first incubated with the indicated stimuli for 48 h to
either increase inhibitory FcRII (by TNF, TNF/IL-1) or
induce activating FcRI/III (by IFN-, IFN-/LPS) on the cell
surface. After activation with heat-aggregated IgG IC, MC were
analyzed for production of MCP-1, IP-10, and RANTES chemokine mRNAs
by RPA (A), KC and MCP-5 mRNAs by RT-PCR (B),
and MCP-1 protein by ELISA (C) (for details see
"Experimental Procedures").
|
|
IgG IC-triggered Synthesis of MCP-1 Is Impaired in
FcRIII-deficient MC--
To determine the role of FcRIII in the
production of MCP-1 observed in mouse MC activated by ICs after IFN-
stimulation, we subjected MC from FcRIII (/) mice to IFN-,
and subsequently to IgG IC activation, and examined mRNA and
protein levels of MCP-1. As shown in Fig.
8, FcRIII-deficient MC failed to
respond to ICs with no induced synthesis of MCP-1. Impaired MCP-1
accumulation was similar to that seen in FcR chain-deficient mice
lacking both FcRI and FcRIII (Fig. 7C). Together,
these data suggest that FcRIII is essential for IC-mediated MCP-1
synthesis by mouse MC in vitro with only a minor additional
role for FcRI.
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 8.
IC-mediated synthesis of MCP-1 is impaired in
FcRIII-deficient MC. Proliferating MC
from C57BL/6 and FcRIII (/) mice were stimulated with 500 units/ml of either TNF or IFN- for 48 h and then activated
with IgG IC for an additional 12 h. MCP-1 synthesis was examined
at both RNA and protein levels by RPA (A) and ELISA
(B). The results of RPA are expressed as the ratio between
MCP-1 and GAPDH mRNA determined by densitometry using a
Phosphoimager (Fuji).
|
|
Inverse Regulation of Glomerular FcRII and FcRIII in Acute
Anti-GBM Nephritis--
Examination of the inflammatory processes in
FcR (/) mice had previously shown that the activating FcRIII
is an essential effector for neutrophil influx in the acute phase of
anti-GBM nephritis (47). Because FcRII but not FcRIII receptors
are normally expressed in MC and kidney tissue (Figs. 1-3), we now
tested for modulation of renal FcRII and FcRIII expression in the
initiation of experimental nephritis. Hereby, anti-GBM
glomerulonephritis was induced by intravenous injection of 300 µl of
anti-GBM IgG antiserum, leading to the formation of immobilized
GBM:anti-GBM ICs in the glomerulus (58) followed by local MCP-1
production and early neutrophil influx that reached maximal levels of
223.9 ± 68.7 of µg MPO in kidney (n = 8, p < 0.001) at 2 h after challenge (Fig.
9, A and B). As
assessed by TaqMan real-time PCR of kidney tissues, the
FcRII/tubulin mRNA ratio strongly decreased from high levels of
147.9 ± 24.5 down to 74.0 ± 13.0 AU (arbitrary units)
(n = 4; p < 0.05) with simultaneously
enhanced amounts of FcRIII mRNA from low levels of 10.4 ± 2.9 up to 29.1 ± 0.8 AU (n = 4; p < 0.001) after the anti-GBM inflammatory insult (Fig. 9C).
This regulatory principle of FcRII suppression and induced synthesis
of FcRIII mRNA was equally seen in control experiments after
intracapsular injections of 500 ng of LPS (Fig. 9D).
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 9.
Inverse regulation of glomerular
FcRII/FcRIII in acute
anti-GBM nephritis. IC nephritis was induced by intravenous
application of 300 µl of anti-GBM IgG antiserum (+ anti-GBM) (A-C). Mice receiving 300 µl of
sterile PBS served as negative controls ( anti-GBM). Mice receiving 500 ng of LPS served as
positive controls (+ LPS) (D). After 2 h,
kidneys were processed and assayed for neutrophil infiltration, as
evaluated by a colorimetric MPO assay of kidney homogenate
(A) for renal MCP-1 mRNA/protein production by
TaqMan PCR and MCP-1-specific ELISA (B) and for
glomerular expression of FcRII and FcRIII mRNA (also assessed
by TaqMan PCR) (C and D). Data are
expressed as mean ± S.E. obtained from 4-8 mice. Significance is
determined by Student's t test (*, p < 0.05; **, p < 0.001).
|
|
Local Blockade of Glomerular FcRII Results in Enhanced Anti-GBM
Inflammation--
The regulatory importance of the inhibitory FcRII
had been established in mouse models of anti-GBM nephritis and
Goodpasture's syndrome (20, 59). However, the issue of FcRII
inhibition mediated by either infiltrating or local cells remained to
be investigated. We thus designed in vivo experiments in
which the inhibitory FcRII was blocked in a single mouse kidney by
intracapsular injection of 20 µl (2 µg/µl) of purified
anti-FcRII (Ly17.2) mAb 30 min before the induction of the anti-GBM
nephritis. The analysis of inflammatory signs in anti-GBM-treated mice
(n = 4) revealed augmented responses of neutrophil
infiltration and chemokine production of MCP-1, MIP-2, and KC in the
left kidney (receiving the anti-FcRII mAb) as compared with the
right kidney (not receiving anti-FcRII mAb) of the same animal (Fig.
10). As a control, anti-GBM treated
mice (n = 4) receiving an irrelevant isotype-matched
antibody instead of the anti-FcRII mAb did not display such
differences.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 10.
Increased anti-GBM nephritis in kidney
selectively treated with blocking anti-FcRII
antibody. 40 µg of anti-FcRII mAb (black bars) or
isotype control mAb (gray bars) was applied in the right
kidney of mice followed by intravenous injections of anti-GBM IgG.
After 2 h, MPO activity (A), MCP-1 (B), and
MIP-2/KC chemokine levels (C) were determined in both
kidneys of anti-GBM-treated mice by differentiating between the
nontreated left kidney (left) and the mAb-treated right
kidney (right). Shown are the mean values ± S.E.
obtained from four mice each. Significance is determined by Student's
t test (*, p < 0.05).
|
|
| |
DISCUSSION |
Previous studies in FcR-deficient mice suggested the importance
of FcRs in promoting IC-mediated inflammation occurring at different
tissue sites, including skin, lung, peritoneum, and kidney (12,
43-48). For example, FcR-deficient mice had an attenuation of
severe nephritis in models of spontaneous lupus nephritis and induced
nephrotoxic serum glomerulonephritis (8, 46). Moreover, the activating
FcRIII was found to be critical for glomerular neutrophil
accumulation in acute GBM-anti-GBM IC nephritis (47). In contrast to
the attenuated phenotype in FcRIII-deficient mice, the genetic
deletion of the inhibitory FcRII augmented anti-GBM nephritis (20).
These results are very similar to that obtained in other models of IC
inflammation (17, 18), indicating that FcRIII-mediated activation
responses are regulated by the inhibitory FcRII when co-expressed on
the same effector cells. In this report, we now tested the concept that
resident kidney cells express and function via activating and
inhibitory FcRs, thus providing a possible regulatory pathway for
renal infiltration in IC-mediated nephritis. Immunohistochemical
analysis of murine renal tissue revealed weak FcRII/III expression
in glomeruli, attributed mainly to the non-activating FcRII in
normal kidney and to FcRIII after challenge with LPS (Fig. 1). The
source of induced FcRIII might be either activated resident kidney
cells or, as recently suggested (47), infiltrating effector cells.
Human mesangial cells can be induced to express functional FcRIIIA
after stimulation with IFN- and LPS (27). In preliminary
experiments, FcR-deficient mice made chimeric by bone marrow
transplantation pointed toward a substantial, although not exclusive,
contribution of locally expressed FcRs in the initiation of
nephritis (data not shown). Thus, it is very likely that
inhibitory/activating FcRII/III receptors on both infiltrating cells
(neutrophils and monocytes) and intrinsic glomerular MC are important
for mediating the glomerular response to IgG immune complexes.
An interesting observation is that the basal constitutive expression of
FcRII observed in normal kidney could similarly be detected in
primary cultures of MC (Figs. 2 and 3). RT-PCR revealed mRNA
specific for FcRIIb2 but not FcRIIb1, which suggests that MC are
related to macrophages, previously established to express FcRIIb2,
which is able to mediate endocytosis of IgG immune complexes (35). In contrast to macrophages (49), however, the expression of
FcRIIb2 on MC is tightly regulated. The combination of TNF and
IL-1 increased FcRIIb2 on MC under both resting and proliferating culture conditions in vitro, whereas IFN- induced
complete down-regulation of FcRIIb2 mRNA and protein.
Importantly, the inhibitory effect of IFN- on FcRIIb2 expression
inversely correlated with a strong accumulation of mRNA for the
FcR chain and increased mRNA levels of the chains for
activating FcRI and FcRIII (Fig. 4). The pattern of induced
activatory FcR gene expression appears to be specifically dependent
on IFN-, because it was not observed with TNF and IL-1. Gene
promoters of FcRI, FcRIII, and the FcR chain contain
IFN--responsive DNA elements (50-52), which further suggests that
the IFN--dependent induction of FcRI/III expression
is mediated by enhancement of gene transcription. The critical
requirement of IFN- for the appearance of FcRIII on the surface
membrane of mouse MC was documented by FACS analysis (Fig. 5). We also
observed strongly enhanced binding of monomeric IgG2a after stimulation
with IFN-/LPS (Fig. 6), which may indicate an induced surface
expression of FcRI. However, because of the lack of specific
anti-mouse FcRI antibodies, more direct proofs were not possible.
This is in contrast to the situation for human MC, where
IFN--dependent induction of functional FcRI (CD64) has clearly been demonstrated by the use of the CD64 mAb 22 (28, 53).
Chemokines play an important role in selectively recruiting certain
subsets of leukocytes to specific sites of inflammation and tissue
injury. Given the oppositely regulated expression of inhibitory
FcRII versus activating FcRIII (and possibly FcRI) by IFN- but not TNF/IL-1, the effect of IgG IC activation for chemokine induction by MC was examined for the different cytokine combinations (Figs. 7 and 8). IFN--induced FcRI/III activated mRNA synthesis for CC and CXC chemokines, including MCP-1, MCP-5, RANTES, and KC. In contrast to IFN-, the combination of
TNF/IL-1 leading to increased expression of the inhibitory
FcRII on MC resulted in only minor responses to IC activation. These
results are similar to recent findings in other models, which
demonstrated that FcRIII is an activating receptor, whereas FcRII
is not (54). The contribution of activating FcRI in relation to
FcRIII for enhanced chemokine production appears rather low, because the profoundly impaired synthesis of MCP-1 observed in MC from FcR-deficient mice (lacking the expression of both FcRI and FcRIII) is equally seen in MC from FcRIII-deficient mice. Whether FcRI is indeed redundant awaits further investigation using the previously established FcRI-deficient mice (55). Taken together, our
present in vitro results demonstrate that IFN- is a
potent stimulus for changing the expression profile from inhibitory
FcRII to activatory FcRIII on MC, thereby enabling them to
respond to IgG immune complexes via induced synthesis of MCP-1 and
other chemokines which, as recently shown in several in vivo
disease models (29, 31, 56, 57), are important mediators for renal infiltration.
In experimentally induced anti-GBM nephritis, it is striking to see
that the ratio of glomerular FcRII and FcRIII expression is
modulated during IC inflammation. The acute response triggered by
GBM:anti-GBM IC formation leads to the production of MCP-1, MIP-2, and
KC and subsequent neutrophil accumulation in the kidney and is
associated with the down-regulation of renal FcRII and induced
FcRIII expression (Fig. 9). Local intracapsular injections with LPS
induce the same pattern of inverse FcR regulation. Considering the
fact that the activating FcRIII is essential for neutrophil influx
in anti-GBM-nephritis (47), the anti-GBM-dependent effect of inverse modulation of glomerular FcRs may represent one of the
initial events in the acute phase of this model of IC-induced glomerulonephritis. The genetic deletion of the inhibitory
FcRII in FcRII (/) mice results in augmented inflammation in
experimental models of anti-GBM nephritis and Goodpasture's syndrome
(20, 59). However, it remained to be investigated whether
FcRII-dependent inhibition is mediated by resident
versus infiltrating effector cells. Our data demonstrate
that functional and selective blockade of FcRII in kidney tissue is
efficient in increasing neutrophil recruitment via enhanced
production of CC/CXC chemokines (MCP-1, MIP-2, and KC) in acute
anti-GBM nephritis in normal mice (Fig. 10). These findings strongly
suggest that IC-dependent activation responses mediated by
anti-GBM-induced FcRIII are normally controlled by the inhibitory
FcRII expressed on kidney cells in vivo.
In summary, we have used primary cultures of glomerular mesangial cells
from normal and FcR-deficient mice as an in vitro model of
renal inflammation to investigate the relationship between cytokines
and Fc receptors in the regulation of IC-mediated chemokine production. This approach enabled us to demonstrate for the first time
that the non-activating b2 isoform of FcRII is constitutively expressed in kidney and MC, thus providing a potential interaction with
IgG immune complexes in the glomerulus. Similar to reports obtained
from other species (27, 39), activating FcRI and FcRIII are
normally not expressed under non-inflammatory conditions in mice.
Interestingly, our observations establish the importance of murine
IFN- as an activator to promote IgG IC-mediated CC and CXC chemokine
synthesis through the down-regulation of FcRII and induction of the
activating FcRIII on MC. Furthermore, a major role of FcRIII over
FcRI is suggested by a comparison between FcR (/) and
FcRIII (/) mice in the activation of the CC chemokine, MCP-1.
Earlier reports have defined a contribution of MCP-1 and FcRIII in
mouse models of nephrotoxic nephritis (47, 56), which may indicate that
FcRIII and MCP-1 are critical components for the initiation of
kidney disease in vivo. Our study adds to this point by
identifying FcRIII-activated MC as an important cellular source of
MCP-1 in vitro. Locally induced chemokine synthesis by
cytokine-dependent expression of inhibitory/activating
FcRs in MC may thus be relevant for the regulation of cellular
infiltration during nephritis. In support of this concept, initiation
of acute anti-GBM nephritis is associated with induction of FcRIII
and down-regulation of glomerular FcRII. Strikingly, the functional blockade of FcRII, when induced selectively in kidney, results in
enhanced chemokine production and neutrophil influx. These findings
establish that opposite regulation of and function through activating and inhibitory FcRs on resident kidney cells, like mesangial cells, contribute to glomerular disease involving immune complexes in mice.
§,
,
TOP
ABSTRACT
INTRODUCTION
