| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Centre for Inflammation Research, Department of Clinical and Surgical Sciences (Internal Medicine), Royal Infirmary, University of Edinburgh, Edinburgh, United Kingdom;
Department of Immunology, University of Glasgow, Glasgow, United Kingdom; and
Department of Medicine and Therapeutics, University of Aberdeen, Foresterhill, Aberdeen, United Kingdom
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
then further activated in coculture with LPS or TNF-
elicited a
10-fold induction of rat mesangial cell apoptosis and complete
suppression of mitosis, effects inhibitable by the NO synthase
inhibitors L-monomethyl arginine and
L-N6-(1-iminoethyl) lysine
dihydrochloride. Complete dependence upon macrophage-derived NO was
observed in comparable experiments employing activated bone marrow
macrophages from wild-type and NO synthase 2-/- mice.
Nevertheless, when mesangial cells were primed with IFN- plus
TNF-, increased induction by activated macrophages of mesangial
apoptosis exhibited a NO-independent element. The use of
gld/gld macrophages excluded a role for Fas ligand in
this residual kill, despite increased expression of Fas and increased
susceptibility to soluble Fas ligand exhibited by cytokine-primed
mesangial cells. Finally, activated macrophages isolated from the
glomeruli of rats with nephrotoxic nephritis also induced apoptosis and
suppressed mitosis in mesangial cells by an L-monomethyl
arginine-inhibitable mechanism. These data demonstrate that activated
macrophages, via the release of NO and other mediators, regulate
mesangial cell populations in vitro and may therefore control the
mesangial cell complement at inflamed sites. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
However, little is understood of the cellular control mechanisms that
regulate apoptosis of myofibroblast-like resident cells during
resolution of inflammation. Given that repair of tissue injury may
reiterate processes involved in tissue development, we have become
interested in the possibility that similar cellular mechanisms might
regulate remodeling by apoptosis in both developmental and inflammatory
contexts. Seminal work by Lang and colleagues (4) has
demonstrated a key role for macrophage (M
)-directed apoptosis in the
elimination of unwanted capillaries in neonatal development of the
mouse and rat eye (5). M dock on to the microvascular
endothelial cells and trigger their apoptosis in this model of tissue
remodeling, but the cell-killing mechanisms deployed are as yet
unclear. The probable relevance of this observation to regulation of
cell populations at inflamed sites has been underscored recently by
Tesch and colleagues (6), who, using a nephrotoxic serum
model of glomerulonephritis, demonstrated decreased resident tubular
cell apoptosis in the monocyte chemoattractant protein-1 (MCP-1)
knockout animal compared with wild type. During tubular inflammation,
there was an MCP-1-directed influx of M to the interstitium. In the
knockout animal, the influx of M was reduced. It appears, therefore,
that M in this model are inducing tubular cell apoptosis. The
killing mechanisms in both models, however, remain to be defined,
although in the tubular injury model, in vitro experiments suggested
that M can release soluble factors capable of triggering tubular
cell death.
Furthermore, there have been indications that M might have the
capacity to direct apoptosis of glomerular MC, which assume a
myofibroblast-like phenotype in situations in which they threaten to
promote the persistence of glomerular inflammation by secretion of
cytokines, or progression to scarring by excess deposition of abnormal
extracellular matrix. First, Mene and colleagues (7)
showed that U937 promonocytic cells were able to induce cytolysis of
cultured MC, although it was not determined whether this represented
apoptosis or necrosis. Second, Hruby et al. (8) reported
similar findings in cocultures of rat peritoneal M and MC,
implicating NO in the phenomenon by specific inhibition of MC
disintegration with L-NMMA. Further evidence that M
might have the capacity to induce apoptosis in neighboring cells by the
release of NO comes from coculture studies indicating that
cytokine-activated peritoneal M can trigger apoptosis in
proliferating breast tumor cells by a mechanism inhibitable by
L-NMMA (9, 10). However, in no preceding study
has there been a direct demonstration that M can trigger apoptosis
in glomerular MC.
In this study, we have tested the hypothesis that M activated to
express inducible NO synthase (also known as NOS-2) can trigger
apoptosis in cultured MC, which are well known to reiterate the
myofibroblast features displayed in the injured glomerulus in vivo. We
report that in contrast to quiescent cells, M activated in vitro or
isolated from inflamed glomeruli not only inhibited MC mitosis, but did
indeed induce apoptosis in MC. Cytokine stimulation of MC was able to
increase the kill. Furthermore, use of specific NOS-2 inhibitors and
NOS-2 knockout M clearly implicated M-derived NO as a key
mediator inducing apoptosis and inhibiting mitosis in glomerular cells.
The data establish a new facet to the multifunctional role of the M
in inflammation: the capacity to direct remodeling by coordinate
induction of apoptosis and inhibition of mitosis in resident tissue
cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Tissue culture reagents were from Life Technologies (Paisley, Scotland), and tissue culture plastics from Costar (Cambridge, MA). Experimental culture wells were from Becton Dickinson (Falcon; Franklin Lakes, NJ). Cytokines were from Genzyme (Cambridge, MA), and all other reagents were from Sigma-Aldrich (St. Louis, MO), unless otherwise stated.
Preparation of BMD M
Femurs and tibias were removed from 200- to 250-g rats (Wistar).
Bone marrow was isolated from these by standard sterile techniques
(11) and matured for 7 days in Teflon wells in using
DME/F12 medium with 10% FCS, penicillin (100 U/ml), and streptomycin
(100 mg/ml), conditioned with M-CSF from L929 cells. In some
experiments, BMD M were derived from knockout NOS-2 mice
(129/sv-NOS-2-/-)
(12) and wild-type controls (8 wk). In others, BMD M
were obtained from mice bearing an inactivating point mutation of FasL
(C3H/HEJ-gld/gld) and wild-type controls (8 wk) purchased
from The Jackson Laboratory (Bar Harbor, ME).
Establishment of MC cultures
MC were derived by clonal selection from glomerular single cell preparation (gift of Dr. M. Kitamura, Department of Medicine, University College, London, U.K.) and from primary outgrowths from isolated whole glomeruli (13). MC were verified, and purity established, as described (1, 13, 14). Cells were used from passages 4 to 12. Primary cells were maintained in DME/F12 medium with 16% heat-inactivated FCS, penicillin (100 U/ml), and streptomycin (100 mg/ml). Clonal cells were maintained in the same medium with 10% FCS. MC were maintained in the logarithmic growth phase.
M coculture with MC
Matured (7–10 day) M were plated in 96-well plates initially
at a density to cover 60–70% of the well surface; typically, this
required 2 x 105 cells. MC were prelabeled
with CellTracker Green CMFDA (Molecular Probes, Eugene, OR): cell
cultures, 70–80% confluent, were washed with medium lacking serum and
then incubated for 1 h in serum-free medium containing CMFDA at 5
ng/ml. Cells were washed in medium containing 10% FCS to remove any
unbound CMFDA, then trypsinized and added to cultured M in a 1.5
M:1 MC ratio. Experiments were conducted in DME/F12 medium
containing 10% FCS. Once cells had become adherent, typically 2–4 h,
wells were washed to remove nonadherent cells. Preliminary experiments
mixing unlabeled and labeled cells showed no evidence of transfer of
CMFDA from labeled to unlabeled cells (data not shown).
Activation protocols
When desired, M were primed for 12 h with IFN- (100
U/ml) before coculture. After coculture was established, M were
activated using either IFN- (100 U/ml) plus LPS (1 µg/ml), or
IFN- (100 U/ml) plus TNF- (100 U/ml). Unactivated or quiescent
cells were exposed to carriage medium only (PBS, 0.1% BSA).
In some experiments, MC were primed in flasks for 36 h with a
combination of IFN- (300 U/ml) plus TNF- (300 U/ml). Cells were
washed thoroughly and trypsinized before establishing coculture.
Mock-primed cells were grown in flasks containing carriage medium only
(PBS, 0.1% BSA).
Assessment of apoptosis and mitosis
Morphological criteria.
At the end of the experiment, wells were fixed by adding formaldehyde
(4% final concentration), stored at 4°C for 48 h to ensure firm
adherence of the fixed apoptotic cells to the monolayer. Subsequently,
medium was removed and propidium iodide (PI) in PBS (5 µg/ml) was
added for 5 min to stain both M and MC. This reagent was then
discarded and wells were covered with a fluorescent mountant. Using
inverted fluorescent microscopy, five fields were randomly and blindly
selected from each well so that at least 300 MC were counted in each
well. Wells were counted nonconsecutively. MC labeled with CellTracker
Green CMFDA were discernible from M by their green fluorescence
(M appeared red due to PI binding to RNA as well as DNA). Apoptotic
cells were clearly distinguishable by characteristic morphology
(cytoplasmic blebbing, cell shrinkage, nuclear condensation, and
fragmentation); in our previous studies of MC apoptosis (1, 14), these morphological criteria have been extensively
validated against quantitative and qualitative measures of apoptosis.
Mitotic cells were easily discernible by characteristic morphology and
shape change.
Functional criteria.
MC apoptosis was also assayed in live cells by addition of Hoechst
33342 (bisbenzimidazole dye) to wells at a final concentration of 5
µg/ml. This dye is actively excreted by live, but not apoptotic cells
(15). After 15 min at 37°C, wells were counted as
described above for apoptotic cells. PI was also added to live wells (1
µg/ml) to exclude necrotic cells (see Fig. 1
). Furthermore, 99.8% of
cells in the live coculture were able to exclude trypan blue assessed,
as previously described (10).
|
ratio in
six-well plates (Becton Dickinson). After 24 h, the coculture or
controls were washed, followed by the application of 1x trypsin-EDTA.
All washings and eluted cells were collected, and after centrifugation
(300 x g), they were permeabilized with 70% ethanol
at 4°C overnight. After washing with PBS, cells were treated with
RNase (100 µg/ml) and PI (40 µg/ml) for a minimum of 30 min. Cells
were then detected by forward and side scatter, then assessed by
excitation at 488 nm and measuring fluorescence at 639 nm using
FACScalibur (Becton Dickinson). The characteristic distribution of DNA
content enabled accurate assessment of diploid,
G2/M phase, and hypodiploid populations
(14).
Immunohistochemistry.
Further assessment of mitosis was made by specific identification of
proliferating cell nuclear Ag (PCNA) in MC using PC10 Ab (Dako,
Carpenteria, CA) in fixed wells. Briefly, following quenching of
endogenous peroxidases with 3% hydrogen peroxide in methanol (5 min,
20°C), and permeabilization with 0.1% Triton X-100, 0.1% sodium
citrate in PBS (5 min, 20°C), the primary Ab was applied 1:50 in PBS,
10% NBCS (4 h, 20°C). Specific localization was assured by
biotinylated anti-Ig and streptavidin-peroxidase using the
substrate AEC (Vector). MC were distinguished by the subsequent
labeling with FITC-conjugated smooth muscle actin Ab.
All experimental conditions were prepared in triplicate. All
experiments and conditions were observer blinded. Experiments were
performed on at least three separate occasions. M and primary MC
cultures were derived from at least three different animals.
M-conditioned supernatant transfer
A total of 106 mature BMD M was seeded
to six-well plates with medium containing 10% FCS. Cells were
activated with IFN- and LPS, as described in activation protocols.
After 16 h, supernatant was removed and clarified at room
temperature by centrifugation at 350 x g for 5 min,
followed by centrifugation at 10,000 x g for 5 min.
Two hundred microliters of supernatant were transferred directly to MC
established 24 h previously (covering 70–80% of the well
surface) in 96-well plates. After 8, 16, and 24 h, apoptosis,
mitosis, and cell number were assayed morphologically by PI staining of
fixed monolayers.
Nitrite assay
Griess reagent was prepared using 0.5% sulfanilamide in 2.5%
orthophosphoric acid with 0.05% N-ethylenediamine
(16). Samples were mixed with an equal volume of reagent
and absorbancy measured after 10 min at 540 nm. Standard curves were
prepared using dilutions of sodium nitrite. BMD M and glomerular
M were prepared as described. A total of 105
cells were placed in wells with 400 µl of full medium (without phenol
red). After 24 h, medium was collected and mixed with an equal
volume of Griess reagent. After 5 min, absorbance at 540 nm was
measured. Using sodium nitrite standard curves, the total nitrite
accumulated was calculated and expressed as nanomoles per
106 cells per 24 h. To assess MC nitrite
production, subconfluent 75-cm2 flasks were
cultured for 48 h in 5 ml of medium. A 400-µl aliquot was used
in the Griess assay, and cell number was assessed by hemocytometer to
give equivalent results expressed as nanomoles per
106 cells per 48 h.
Western blot analysis
Cytosolic extracts from cells were prepared using a hypotonic lysis buffer (10 mM HEPES, pH 7.9, 1.5 mM MgCl2, 10 mM KCl, 50 µM DTT, 100 µM phenanthroline, 1 µg/ml pepstatin, 100 µM 1-trans-epoxysuccinyl-leucylamide (4-guanido)butane, 100 µM 3,4-dichloroisocoumarin, 10 mM NaF, 100 µM sodium orthovanadate, 25 mM ß-glycerophosphate) containing the detergent (0.2% v/v) Triton X-100, for 10 min on ice, followed by centrifugation at 10,000 x g for 10 min at 4°C.
Samples were analyzed by SDS-PAGE, followed by Western blotting. Equal amounts of protein extract were mixed with Laemmli dissolving buffer (500 mM Tris-HCl pH 6.8, 20% 2-ME, 8% SDS, 30% glycerol, and bromophenol blue), boiled for 5 min, and electrophoresed in gels containing 12.5% acrylamide.
Proteins were electrotransferred to a nitrocellulose membrane in buffer containing 25 mM Tris base, 192 mM glycine, and 20% methanol (v/v). Following blocking using 5% milk powder in buffered saline for 1 h at room temperature, membranes were incubated for 24–48 h with primary anti-Fas (AB-1; Oncogene Research Products, Cambridge, MA) Ab at 4°C. After washing in Tris-buffered saline, membranes were incubated with peroxidase goat anti-rabbit IgG Ab (1:10,000) at room temperature for 1 h. After further washing, labeled proteins were detected using enhanced chemiluminescence (Amersham, Arlington Heights, IL). Equal protein loading was confirmed by staining gels with Coomassie blue. Adjacent lanes were labeled for ß actin to confirm the specificity of AB-1 for Fas, as actin runs at a similar m.w. to Fas by SDS-PAGE.
Sequencing of mouse FasL by PCR
Genomic DNA was prepared isolated from M from wild-type or
gld/gld mice. A 311-bp segment, covering the predicted
mutation site of the gene, was amplified by PCR using the primers
5'-GTGGCCTTGTGATCAACG-3' and 5'-CTCTGGAGTGAAGTATAAG-3' (Oswel,
Research Products, Southampton, U.K.). Briefly, 5 µl of DNA solution
was diluted with 45 µl of reaction buffer containing 200 µM dNTP,
1.5 mM MgCl2, and primers (100 pmol of each). The
reaction mixture was placed in a DNA thermal cycler (Boimetra,
Goettingen, Germany) and the reaction started by adding 1 U of
Taq polymerase (Bioline, London, U.K.). The conditions of
PCR were 95°C, 1.5 min; 55°C, 1.5 min; and 72°C, 1.5 min for 36
cycles. Sequencing, using the fluorescent dideoxy method, confirmed the
gld/gld mutation, resulting in a phenylalanine to leucine
conversion at amino acid position 273 (data not shown).
Induction of accelerated autologous phase of nephrotoxic nephritis
Nephritis was induced using a standardized protocol (17). Briefly, male Sprague Dawley rats (weight 180–200 g) were immunized by injection of 1 mg normal rabbit IgG, in Freund’s complete adjuvant (Difco, Detroit, MI). After 7 days, the rats were injected with 1 ml of nephrotoxic serum i.v. They were sacrificed, and kidneys were prepared as below. Biopsies were taken for histology, which confirmed severe proliferative glomerulonephritis (18). Urine was collected in metabolic cages for 18 h, and albuminuria was confirmed by rocket immunoelectrophoresis.
Isolation of single cells from glomeruli
Glomeruli were isolated using a standard sieving technique.
Isolated glomeruli were enzymatically digested to single cell
suspensions using trypsin, collagenase, DNase, and EDTA, as described
(19, 20). M were isolated by adherence in 24-well
plates. Aliquots of the single suspension containing 1 x
105 M were added to wells in full medium with
serum. After 2 h, wells were thoroughly washed to leave adherent
M only. Purity of the cells isolated was confirmed by ED-1
immunostaining. Isolation of resident M from control kidneys (normal
histology) by the same method required pooling of the single cell
suspension from all control animals due to the scarcity of this cell
type (<5% of glomerular cells).
Statistics
The data were expressed as mean values with the SE of mean, and compared using the paired t test. No multiple comparisons were made.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
coordinately inhibit mitosis and
stimulate apoptosis in MC
Mature rat BMD M were cocultured with rat MC in the presence of
medium containing 10% FCS. Prelabeling of MC with CellTracker Green
CMFDA, and staining of fixed monolayers with PI enabled selective
assessment of MC. Extensive work in our laboratory has shown that
morphological assays of apoptosis on the basis of typical cell
shrinkage and nuclear condensation have advantages over other
techniques, with which this approach has been compared and validated
(1, 14). First, because apoptotic MC remain intact and
closely associated with the monolayer, it is possible to assay
definitively cell death by apoptosis without disturbing the culture and
risking underestimation of apoptosis due to cell damage or loss, which
we have found to be an inherent problem with certain assays of
apoptosis. Second, microscopical assessment of the cell culture allows
simultaneous detection of cells in mitosis. Nevertheless, to verify the
data, we labeled live cells in coculture before fixation with Hoechst
33342 and counted cells, admitting the dye so that nuclei were stained;
this fluorescent dye is excluded by the plasma membrane of healthy but
not apoptotic cells (14). Interestingly, all stained cells
demonstrated typical condensed apoptotic nuclei, indicating
synchronicity of plasma membrane and nuclear changes of apoptosis in
this cell type. No significant difference between the proportion of
cells admitting Hoechst 33342 and those demonstrating nuclear
condensation on PI staining after fixation of coculture was seen in
activated (in one series at 24 h, there were 35.3 ± 6.2
Hoechst-positive cells per field compared with 31.5 ± 9 apoptotic
MC after fixation and counterstaining (representing 16.1 ± 2.6%
MC per field), n = 9, p = NS) or
unstimulated coculture. To verify further our observations, PI was used
to label DNA in fixed cells from the coculture and assess the
proportion of cells in G2/M phase or with
hypodiploid content of DNA characteristic of apoptosis by flow
cytometry (Table I). Furthermore,
confirmatory evidence of suppression of mitosis was obtained by
immunohistochemistry of fixed coculture for PCNA; the percentage of
PCNA-positive MC after 24 h in quiescent coculture was 88.6
± 3.7%, but this was markedly reduced to 12.6 ± 4.4% in MC
cocultured with M under activating conditions.
|
had no effect on MC mitosis or apoptosis over a 24-h period
compared with control (Fig. 1
A, Table I). However, when
M were primed for 12 h with IFN-, and then the coculture was
activated with IFN- plus LPS, suppression of MC mitosis was observed
as early as 8 h (Fig. 2). At 24
h, there was near total cessation of mitosis compared with control
culture of MC growing alone, stimulated with IFN- and LPS. In
addition, there was almost a 10-fold induction of MC apoptosis (Figs. 1B and 2). These results were supported by flow-cytometric
DNA content data (Table I), which showed a 10-fold induction of
hypodiploid cells in the whole coculture when activated, and a 50%
reduction of all cells with double DNA content
(G2/M). The combination of these effects was, as
expected, associated with a decrease in MC number over this time
period, whereas when cultured under control conditions with FCS, MC
exhibited the expected increase in number (Fig. 2). Replacing LPS with
TNF- in the activating regimen in coculture elicited similar results
to those in Fig. 2. Control cultures of MC alone, stimulated with
IFN- and TNF-, did not undergo apoptosis (data not shown).
Primary cultures of MC also exhibited similar responses to coculture
with activated M. After 24 h of coculture, MC apoptosis was
7.9% (coculture) compared with 1.2% in controls
(p < 0.05), and MC mitosis was 0.1% and 2%,
respectively (p < 0.01). Activated M
underwent very low levels of apoptosis after 24 h (<0.5%). This
low level was seen whether activated M were cultured alone or
with MC.
|
to MC in coculture, a progressively
greater proportion of MC was apoptotic (39.3 ± 3.5% at 24
h, ratio 10:1). However, at ratios higher than 4:1, a small proportion
of apoptotic cells in live cocultures did not exclude PI. A 3-fold
reduction in the number of M in coculture still demonstrated
significant induction of apoptosis (6.1 ± 2.4% at 24 h MC
control 1.2 ± 0.4%, p < 0.05) and inhibition of
mitosis (0.83% ± 0.15 at 24 h MC control 2.3 ± 0.4%,
p < 0.05), although this was reduced compared with
1.5:1 ratio.
NOS-2 mediates inhibition of mitosis and induction of apoptosis in
MC cocultured with activated M
The addition of the nonspecific NOS competitive inhibitor
L-NMMA at 100 µM was able largely to abrogate the
M-induced effects on MC proliferation and apoptosis
(IC50 for apoptosis 5.1 ± 3.6 µM). Higher
concentrations (200 and 500 µM) had no additional effect. Its
inactive analogue (D-NMMA) had no discernible effect in
cocultures with activated BMD M at 24 h (Table II).
L-N6-(1-iminoethyl) lysine
dihydrochloride (L-NIL), a specific competitive
inhibitor of NOS-2, was also able largely to abrogate these M
effects at 24 h (IC50 for apoptosis, 0.33
µM ± 0.40) (Fig. 3, Table I).
Transfer of conditioned supernatants directly from M
(106/ml) activated for 16 h with IFN-
(300 U/ml) plus LPS (1 µg/ml) had no effect on apoptosis or mitosis
of MC compared with control (medium containing IFN- and LPS) (data
not shown). These results are in keeping with NO (a short-lived
unstable molecule), signaling apoptosis from the M to the
MC.
|
|
or FasL
Using the morphological assay for assessing MC apoptosis, MC
pretreated with IFN- and TNF- for 36 h were rendered
susceptible to apoptosis induced by soluble recombinant FasL (Fig. 4A). This effect was not
observed following pretreatment with IFN- alone (data not shown). In
keeping with previous work on human MC demonstrating that similar
pretreatment increased the expression of Fas (Ref. 21 and
our unpublished observations), rat MC Triton X lysates were prepared
for Western blot analysis, which confirmed increased expression of Fas
after pretreatment with IFN- and TNF- (Fig. 4B).
|
-induced kill, following pretreatment with
IFN- and TNF- for 36 h, MC were washed, trypsinized, and
cocultured with IFN--primed rat BMD M in the presence of IFN-
plus LPS. After 24 h (Fig. 5), there
was a significant increase in MC apoptosis compared with concurrent
experiments using mock-pretreated MC. This effect was only partially
reversible by adding the NOS-2 competitive inhibitor
L-NMMA, suggesting that mediators other than NO might
contribute to the induction of apoptosis in primed MC. Quiescent M
had no effect on primed MC, and the latter were no more susceptible to
the suppression of mitosis by activated M (see Fig. 6B).
|
|
expression of NOS-2, but not FasL, determines the capacity to
inhibit mitosis and induce apoptosis in MC
Previous work has indicated that both M and MC produce NO upon
cytokine induction of NOS-2 (22, 23). We confirmed that
rodent BMD M produced NO (as assessed by the use of Griess reagent
to measure accumulated nitrite in culture medium) when challenged with
proinflammatory cytokines (Table III). MC
produced much smaller amounts of NO in response to IFN- plus TNF-
(Table III). Such cells stained weakly for NOS-2 by indirect
immunofluorescence compared with the bright staining of all M
treated with cytokines (not shown).
|
NOS-2, or
both. To differentiate between these possibilities, BMD M from NOS-2
knockout mice and wild-type control mice were prepared. The
NOS-2-/- M produced a little NO in response
to proinflammatory stimuli (Table III) such as IFN- plus LPS when
compared with wild type, in keeping with previous work suggesting that
other NOS may be up-regulated by proinflammatory cytokines (12, 24). Activated 129/sv wild-type mouse M behaved identically
to rat M in the coculture assay, suppressing mitosis (8 and 24
h) and inducing apoptosis (24 h) in both cytokine-primed or unprimed MC
(Fig. 6a). However, activated
NOS-2-/- 129/sv BMD M exhibited a
significantly diminished capacity to induce MC apoptosis. However, the
magnitude of this reduction was much smaller when MC were primed. This
residual kill did not appear to be due to priming rendering MC
susceptible to the low levels of NO produced by activated
NOS-2-/- M, in that no further reduction in
apoptosis was brought about by adding L-NMMA to
the coculture (up to 500 µM). The ability to suppress mitosis was
completely absent when activated NOS-2-/- M
were cultured with unprimed MC, but by contrast with apoptosis, there
was no residual effect when NOS-2-/- M were
incubated with primed MC; there was still no suppression of mitosis
(Fig. 6b). These data support the possibility suggested
above (Fig. 5) that M-derived mediators in addition to NOS-2-derived
NO may contribute to the killing of primed rather than unprimed MC.
Furthermore, in separate experiments (see Materials and
Methods), clarified supernatants transferred from activated
rat M directly to primed MC demonstrate that this additional factor
was soluble and partially transferable: 24 h after transfer of
conditioned supernatant to primed MC, apoptosis was 10.4 ± 0.9%,
control (medium alone with activating cytokines) 3.7 ± 1%
(n = 3).
Given the increased susceptibility of cytokine-primed MC to
FasL-induced apoptosis demonstrated above (Fig. 4a), and the
recently reported capacity of activated human M to release FasL
capable of triggering apoptosis in bystander cells (25, 26), we were interested in the possibility that the residual
capacity of NOS-2-/- M
for killing of primed MC was mediated by FasL. To assess this
possibility further, we cultured BMD M from mice homozygous for a
point mutation in the FasL gene, which results in a protein product
unable to ligate Fas such that death signaling ensues
(C3H/HEJ-gld/gld). These M produced similar amounts of NO
compared with their wild-type counterparts (data not shown), and
behaved identically, whether cocultured with unprimed or primed MC
(Fig. 7). Nevertheless, the primed MC
used in these experiments were susceptible to apoptosis by soluble FasL
(100 ng/ml) (13.4% at 24 h), whereas unprimed were not (1% at
24 h). The residual MC apoptosis observed incubating activated
NOS-2-/- M with cytokine-primed MC therefore
appears most unlikely to have been mediated by FasL.
|
isolated from inflamed glomeruli also inhibit mitosis and
induce apoptosis in cultured MC by an NO-mediated mechanism
To assess the likely in vivo relevance of these observations,
telescoped nephrotoxic nephritis was induced in Sprague Dawley rats.
During the autologous phase, there was predominance of M in the
mesangium (21.4 ± 3.1 M per glomerular cross section on day 2,
16.1 ± 1.5 on day 4, 1.4 ± 0.6 in controls by ED-1
immunohistochemistry). Animals from days 2, 4, and 7 after disease
induction were sacrificed, kidneys were removed, and glomeruli isolated
by sieving. Glomeruli were then exposed to an enzymatic digest,
resulting in a single cell suspension. M were isolated from this by
rapid adhesion of these cells to plastic wells (>90% purity by ED-1
immunohistochemistry). M from these diseased glomeruli were
immediately cocultured with CellTracker Green-labeled MC, and the
coculture was assessed at 8, 16, and 24 h. Inflammatory glomerular
M released similar amounts of NO to activated rat BMD M (Table III). Resident M from normal glomeruli produced little NO, but when
challenged with proinflammatory cytokines, released NO in similar
quantities to BMD M treated with the same cytokines.
M from inflamed glomeruli (activated in vivo) from day 2 of the
disease were also able to induce effects comparable with activated BMD
M upon coculture with MC (Fig. 8).
Induction of apoptosis in coculture compared with controls, and
suppression of mitosis was present at 8 and 16 h, but diminished
at 24 h. However, the combination of these effects led to a
decrease in cell number. Results from coculture using M isolated
from days 4 and 7 of the glomerulonephritis (Table IV) showed similar findings to M from
day 2. The addition of L-NMMA was also able to abrogate
partially the M-mediated effects (Table IV). D-NMMA had
no effect. At 16 h, there was no significant difference in
apoptosis and mitosis compared with controls.
|
|
appeared likely to be due to two factors. First, the ex vivo M
were already activated, whereas the BMD M employed above (Fig. 2)
were used under conditions in which full activation would only have
been expected after the coculture had been established. Second, ex vivo
M underwent programmed cell death (>70% at 24 h), which
appeared to be related to the enzymatic digest. To investigate these
issues further, BMD M were activated with IFN- (100 U/ml) and LPS
(1 µg/ml) for 6 h, then washed. They were exposed to the same
enzyme protocol as glomerular cells, then immediately used in
coculture. This approach gave comparable results to the ex vivo
glomerular M on MC (data not shown). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
coordinately induce apoptosis and
suppress mitosis in cultured glomerular MC, which are well known to
exhibit myofibroblast-like features. The use of competitive
inhibitors (L-NMMA and L-NIL) and M from
gene-targeted animals lacking NOS-2 provided definitive evidence that
M-derived NO was critical in both induction of apoptosis and
suppression of mitosis. When MC susceptibility to these M-induced
effects was deliberately increased by pretreating MC with
proinflammatory cytokines, a proportion of the M-induced kill was
independent of NOS-2. However, despite evidence of sensitization to
FasL by treatment of MC with proinflammatory cytokines, the use of
activated gld/gld M demonstrated that this residual kill
was not mediated by FasL from M. Finally, glomerular M activated
in vivo during experimentally induced glomerulonephritis also induced
apoptosis and suppressed mitosis when cultured ex vivo with MC, by an
L-NMMA-inhibitable mechanism. We conclude that by
virtue of release of NO, activated M can regulate MC number by both
induction of apoptosis and suppression of mitosis.
We believe that these data provide evidence of a new concept in
inflammation: that of M-directed regulation of resident cell
populations by the coordinated induction of apoptosis and inhibition of
mitosis. This control may have a pivotal role in the outcome of acute
inflammatory responses, either progression to a hypocellular scar, or
healing by restoration of the normal cell complement and phenotype. The
ex vivo data argue strongly that M-directed regulation of MC
complement is likely to occur in glomerulonephritis in vivo, but it is
clear that future experiments will need to test the
population-regulating role of M in greater detail. Evidence of
M-directed tubular cell apoptosis already exists in nephrotoxic
nephritis induced in MCP-/- and wild-type mice
(6), in which M accumulation in the tubulointerstitium
was much reduced in the chemokine-deficient mice, and was associated
with a marked decrease in tubular cell apoptosis. Furthermore,
consistent with data suggesting that many cell types exhibit
myofibroblast-like features at sites of tissue injury, preliminary data
from Bennett’s group (27) have linked M with apoptosis
of vascular smooth muscle cells in vitro and in atherosclerotic plaques
in vivo. Therefore, we suggest that M-mediated regulation of
resident cell populations should be sought in a wide range of disease
settings.
The competitor and knockout M experiments indicated that
M-derived NO plays a major role in inducing apoptosis and
suppressing mitosis in MC. The paucity of NO production by rat MC
compared with M is in keeping with work on whole glomeruli from rats
with Thy-1.1 nephritis, in which NO release correlated with the
presence of M rather than proliferating MC (28).
However, the only available study of glomerulonephritis in
NOS-2-/- mice was in a model of mild
nephrotoxic nephritis with little M infiltration of glomeruli
(29), and apoptosis was not assessed. Furthermore, the
equivocal data on NO in glomerulonephritis published to date (in which
two studies show amelioration of disease (30, 31): two
show no difference (29, 32), and one shows exacerbation
(33)) have concentrated on conventional parameters of
glomerular injury rather than apoptosis or remodeling. Therefore,
additional experiments will be required to confirm a role for
M-derived NO in regulating MC number in glomerulonephritis models
exhibiting degrees of M infiltration comparable with that observed
in human disease (34, 35), in which M number may
approach MC number (comparable with the in vitro ratio employed in our
experiments). It is, however, difficult to extrapolate directly from in
vitro ratios, in which cell interaction is in one plane, to the in vivo
situation. Nevertheless, expression of NOS-2 by M in human
glomerulonephritis (36, 37), despite notorious
difficulties in achieving NO production in vitro (38, 39),
suggests again that NO-mediated apoptosis may be relevant in
vivo.
However, the competitor and NOS-2-/- M
studies employing MC treated with proinflammatory cytokines indicated
that, in addition to NO, there is at least one other mechanism by which
M can direct MC killing. Despite evidence in human monocyte-derived
M of FasL-dependent killing of neighboring cells (25, 26), the data from gld/gld mice argue strongly
against this subsidiary M-directed killing mechanism being mediated
by rodent M-derived FasL. To what extent the cytokine-treated or
primed in vitro MC represents a MC in vivo during inflammatory
responses is unclear. In this study, we found that primed cells in
vitro were weakly NOS-2 positive, and produced relatively small amounts
of NO. Immunohistochemistry of glomerular sections from nephrotoxic
nephritis suggested that M are the principal cell type expressing
NOS-2 (40), although similar studies of human disease
demonstrated that some MC in inflamed glomeruli were NOS-2 positive
(37). Furthermore, we observed increased rat MC Fas
protein following cytokine priming, in keeping with studies of human MC
(21). This correlates with work showing that some human MC
during acute glomerular disease have increased Fas protein
(41). Together, these data indicate that at least some MC
in vivo share phenotypic features with cultured MC primed with
cytokines in vitro. Therefore, it would appear desirable for future
studies to investigate further whether such cells are rendered
susceptible to ligation of receptors other than Fas that signal into
the caspase (intracellular death signaling) pathway. Preliminary
experiments (J. Duffield, C. Ware, and J. Savill, unpublished data)
employing TNFR1-Fc chimeras to block TNFR1 on MC point to a role for
M-derived TNF-/TNF-ß in mediating NO-independent residual kill
of cytokine-primed MC, but a range of ligands for the expanding TNFR
family will need to be assessed before definitive conclusions can be
drawn.
The major focus of this study was to determine the potential of
activated M to direct apoptosis of myofibroblast-like MC. However,
the capacity to limit the growth of MC populations was clearly enhanced
by concomitant suppression of MC mitosis. Indeed, exogenous NO donors
have been reported to induce DNA fragmentation and inhibit mitosis in
MC (42, 43). It is known that when MC are treated with the
NO donor, S-nitroso-glutathione, the nuclear protein p53,
whose effects on the cell cycle are well recognized, is induced
(42). Furthermore, exogenous NO donors activate caspase 8,
the most upstream member of the caspase pathway, by cleavage, in
lymphoid cell lines (44). Future studies should address
the molecular mechanisms by which NO coordinately suppresses mitosis
and promotes apoptosis in MC.
To conclude, this study implicates NO derived from activated M in
regulation of MC populations by induction of apoptosis and suppression
of mitosis in this resident glomerular cell type. We speculate that
this may be a generally relevant mechanism for regulation of
myofibroblast-like cell populations at sites of tissue injury.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. J. S. Duffield, Centre for Inflammation Research, Department of Clinical and Surgical Sciences (Internal Medicine), Royal Infirmary, University of Edinburgh, Edinburgh, EH3 9YW, U.K. E-mail address:
3 Abbreviations used in this paper: MC, mesangial cell; BMD, bone marrow-derived; D-NMMA, D-monomethyl arginine; FasL, Fas ligand; gld, generalized lymphoproliferative disease; L-NIL, L-N6-(1-iminoethyl) lysine dihydrochloride; L-NMMA, L-monomethyl arginine; M, macrophage; MCP, monocyte chemoattractant protein; NOS, NO synthase; PCNA, proliferating cell nuclear Ag; PI, propidium iodide.
Received for publication May 24, 1999. Accepted for publication December 3, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
expression in 15 cases of antineutrophil cytoplasmic autoantibody-associated renal vasculitis. Am. J. Kidney Dis. 27:48.[Medline]
RII/CD23 in the IL-4-induced nitric oxide production by normal human mononuclear phagocytes. Cell. Immunol. 163:314.[Medline]
This article has been cited by other articles:
|
|
 |
|
  D. Yang, S. G. Elner, L.-R. Lin, V. N. Reddy, H. R. Petty, and V. M. Elner Association of Superoxide Anions with Retinal Pigment Epithelial Cell Apoptosis Induced by Mononuclear Phagocytes Invest. Ophthalmol. Vis. Sci., October 1, 2009; 50(10): 4998 - 5005. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. Davies, A. J. Stempel, and M. R. Powers MCP-1 Deficiency Delays Regression of Pathologic Retinal Neovascularization in a Model of Ischemic Retinopathy Invest. Ophthalmol. Vis. Sci., September 1, 2008; 49(9): 4195 - 4202. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. C. Henderson, A. C. Mackinnon, S. L. Farnworth, T. Kipari, C. Haslett, J. P. Iredale, F.-T. Liu, J. Hughes, and T. Sethi Galectin-3 Expression and Secretion Links Macrophages to the Promotion of Renal Fibrosis Am. J. Pathol., February 1, 2008; 172(2): 288 - 298. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. V. Dandapani, H. Sugimoto, B. D. Matthews, R. J. Kolb, S. Sinha, R. E. Gerszten, J. Zhou, D. E. Ingber, R. Kalluri, and M. R. Pollak {alpha}-Actinin-4 Is Required for Normal Podocyte Adhesion J. Biol. Chem., January 5, 2007; 282(1): 467 - 477. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. S. Duffield, S. Hong, V. S. Vaidya, Y. Lu, G. Fredman, C. N. Serhan, and J. V. Bonventre Resolvin D Series and Protectin D1 Mitigate Acute Kidney Injury J. Immunol., November 1, 2006; 177(9): 5902 - 5911. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Kipari, J.-F. Cailhier, D. Ferenbach, S. Watson, K. Houlberg, D. Walbaum, S. Clay, J. Savill, and J. Hughes Nitric Oxide Is an Important Mediator of Renal Tubular Epithelial Cell Death in Vitro and in Murine Experimental Hydronephrosis Am. J. Pathol., August 1, 2006; 169(2): 388 - 399. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Duffield, P. G. Tipping, T. Kipari, J.-F. Cailhier, S. Clay, R. Lang, J. V. Bonventre, and J. Hughes Conditional Ablation of Macrophages Halts Progression of Crescentic Glomerulonephritis Am. J. Pathol., November 1, 2005; 167(5): 1207 - 1219. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L.-Q. Qiu, R. Sinniah, and S. I-H. Hsu Coupled Induction of iNOS and p53 Upregulation in Renal Resident Cells May Be Linked with Apoptotic Activity in the Pathogenesis of Progressive IgA Nephropathy J. Am. Soc. Nephrol., August 1, 2004; 15(8): 2066 - 2078. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Voskoboynik, B. Rinkevich, A. Weiss, E. Moiseeva, and A. Z. Reznick Macrophage involvement for successful degeneration of apoptotic organs in the colonial urochordate Botryllus schlosseri J. Exp. Biol., June 15, 2004; 207(14): 2409 - 2416. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. W. M. Minto, L.-P. Erwig, and A. J. Rees Heterogeneity of Macrophage Activation in Anti-Thy-1.1 Nephritis Am. J. Pathol., November 1, 2003; 163(5): 2033 - 2041. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-Y. Chen, C.-W. Chi, H.-L. Chen, C.-P. Wan, W.-C. Yang, and A.-H. Yang TNF-{alpha} renders human peritoneal mesothelial cells sensitive to anti-Fas antibody-induced apoptosis Nephrol. Dial. Transplant., September 1, 2003; 18(9): 1741 - 1747. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.-J. Anders, M. Frink, Y. Linde, B. Banas, M. Wornle, C. D. Cohen, V. Vielhauer, P. J. Nelson, H.-J. Grone, and D. Schlondorff CC Chemokine Ligand 5/RANTES Chemokine Antagonists Aggravate Glomerulonephritis Despite Reduction of Glomerular Leukocyte Infiltration J. Immunol., June 1, 2003; 170(11): 5658 - 5666. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Weller, A. Schwentker, T. R. Billiar, and Y. Vodovotz Autologous nitric oxide protects mouse and human keratinocytes from ultraviolet B radiation-induced apoptosis Am J Physiol Cell Physiol, May 1, 2003; 284(5): C1140 - C1148. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Lenda, E. Kikawada, E. R. Stanley, and V. R. Kelley Reduced Macrophage Recruitment, Proliferation, and Activation in Colony-Stimulating Factor-1-Deficient Mice Results in Decreased Tubular Apoptosis During Renal Inflammation J. Immunol., March 15, 2003; 170(6): 3254 - 3262. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Banas, M. Wornle, T. Berger, P. J. Nelson, C. D. Cohen, M. Kretzler, J. Pfirstinger, M. Mack, M. Lipp, H.-J. Grone, et al. Roles of SLC/CCL21 and CCR7 in Human Kidney for Mesangial Proliferation, Migration, Apoptosis, and Tissue Homeostasis J. Immunol., May 1, 2002; 168(9): 4301 - 4307. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Mene, F. Festuccia, R. Polci, F. Pugliese, and G. A. Cinotti Transmembrane signalling in human monocyte/mesangial cell co-cultures: role of cytosolic Ca2+ Nephrol. Dial. Transplant., January 1, 2002; 17(1): 42 - 49. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Mazzali, T. Kipari, V. Ophascharoensuk, J.A. Wesson, R. Johnson, and J. Hughes Osteopontin--a molecule for all seasons QJM, January 1, 2002; 95(1): 3 - 13. [Full Text] [PDF]
|
|
|
|
|
|
M. Saio, S. Radoja, M. Marino, and A. B. Frey Tumor-Infiltrating Macrophages Induce Apoptosis in Activated CD8+ T Cells by a Mechanism Requiring Cell Contact and Mediated by Both the Cell-Associated Form of TNF and Nitric Oxide J. Immunol., November 15, 2001; 167(10): 5583 - 5593. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Freyer, S. R. Johnson, and I. P. Hall Effects of Growth Factors and Extracellular Matrix on Survival of Human Airway Smooth Muscle Cells Am. J. Respir. Cell Mol. Biol., November 1, 2001; 25(5): 569 - 576. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Duffield, C. F. Ware, B. Ryffel, and J. Savill Suppression by Apoptotic Cells Defines Tumor Necrosis Factor-Mediated Induction of Glomerular Mesangial Cell Apoptosis by Activated Macrophages Am. J. Pathol., October 1, 2001; 159(4): 1397 - 1404. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and conventional multiclonal primary cultures (
).
A, Apoptosis assessed at 16 h. Note that unprimed
MC (two sets of bars on the left) are not susceptible to
soluble FasL at 100 ng/ml (NA = no additives). However, note a
significant increase in apoptosis after exposure of cytokine-primed MC
to FasL. Note also similar results using cloned MC or primary cultures.
*, p < 0.01 (n = 3).
B, By immunoblotting, Fas protein is faintly seen at 45
kDa in unprimed MC (left-hand lane). However, after
priming, there is an obvious increase in Fas protein (equal loading of
lanes confirmed by Coomassie staining of the gel). This blot is
representative of three separate experiments.
