| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Dermatology, Case Western Reserve University/University Hospitals of Cleveland, Cleveland, OH 44106
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1-producing mononuclear cells. Here we
characterize the origin and types of those cutaneous effector cells,
the cytokine and chemokine environments, and the effects of
anti-TGF- Ab on skin fibrosis, immune cell activation markers,
and collagen and cytokine synthesis. Donor cells infiltrating skin in
Scl GVHD increase significantly at early time points
post-transplantation and are detectable by PCR analysis of Y-chromosome
sequences when female mice are transplanted with male cells. Cutaneous
monocyte/macrophages and T cells are the most numerous cells in Scl
GVHD compared with syngeneic controls. These immune cells up-regulate
activation markers (MHC class II I-Ad molecules and class A
scavenger receptors), suggesting Ag presentation by cutaneous
macrophages in early fibrosing disease. Early elevated cutaneous mRNA
expression of TGF-1, but not TGF-2 or TGF-3, and
elevated C-C chemokines macrophage chemoattractant protein-1,
macrophage inflammatory protein-1
, and RANTES precede subsequent
skin and lung fibrosis. Therefore, TGF-1-producing donor mononuclear
cells may be critical effector cells, and C-C chemokines may play
important roles in the initiation of Scl GVHD. Abs to TGF- prevent
Scl GVHD by effectively blocking the influx of monocyte/macrophages and
T cells into skin and by abrogating up-regulation of TGF-1, thereby
preventing new collagen synthesis. The Scl GVHD model is valuable for
testing new interventions in early fibrosing diseases, and chemokines
may be new potential targets in scleroderma. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 mRNA expression precede up-regulation of collagen mRNA and
protein synthesis, subsequent skin thickening, and lung fibrosis
(1). This form of GVHD shows mainly fibrosis, with minimal
cytotoxic injury to epithelia (2) and without evidence for
vascular injury or autoantibodies at early time points. In contrast,
cytotoxic GVHD shows epithelial injury that predominates over dermal
fibrosis, and effector cells are thought to be cytotoxic T cells and NK
cells (2). The effector cells and their corresponding
activation markers in Scl GVHD have not been previously identified. We
show here that many donor cells are present in skin of mice with Scl
GVHD, but not in syngeneic bone marrow-transplanted controls. During
characterization of our model, we noted that the initial cutaneous
inflammation consisted primarily of CD11b+
monocyte/macrophages and T cells. Although T cells are critical cells
in initiating immune reactions in skin and are probably the initiating
effector cells in Scl GVHD, monocytes are the predominant cell
population infiltrating skin and lungs in patients with early, rapidly
progressive scleroderma (3) and in murine Scl
GVHD.
The migration and recruitment of leukocytes to a specific tissue site
is a multistep process involving the sequential activation of various
adhesion molecules on immune cells and on the vascular endothelium as
well as a vast array of chemokines (4, 5). These
chemokines are capable of attracting and activating various resident
and inflammatory cutaneous immune cells (6). The
involvement of C-C chemokines in inflammation is an area of active
investigation, but the role of chemokines in the progression of
fibrosing diseases is not completely understood. Chemokines may be
involved in accumulation of inflammatory immune cells that induce
matrix synthesis in scleroderma skin lesions (7, 8, 9).
Specifically, macrophage chemoattractant protein-1 (MCP-1) and RANTES
might play important roles in early pathogenesis of scleroderma, both
by chemoattraction of immunocompetent cells and/or by modulation of
collagen production via TGF-1 in skin (10, 11, 12).
To fully characterize the types of cells infiltrating skin during early
Scl GVHD and their activation status, we performed immunostaining of
skin sections, and flow cytometric analysis of single-cell suspensions
from skin of mice on days 14 and 21 post-bone marrow transplantation
(post-BMT), time points when skin thickening is detectable. We also
examined the up-regulation of cutaneous C-C chemokines, MCP-1,
macrophage inflammatory protein-1 (MIP-1), and RANTES, as well as
TGF- isoform mRNAs in murine Scl GVHD at time points after bone
marrow transplantation, and integrated these findings into a dynamic
model of sclerodermatous fibrosis. Our goals were to understand the
early events preceding skin thickening to devise effective therapies to
inhibit fibrosis in scleroderma and Scl GVHD and to understand the
pathophysiology of cutaneous fibrosing disease.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In a typical transplantation experiment, 7- to 8-wk-old male and female B10.D2 (H-2d) and BALB/c (H-2d, The Jackson Laboratory, Bar Harbor, ME) mice are used as donors and recipients, respectively, for BMT to produce Scl GVHD in a standard method using added spleen cells as a source of mature T cells (1, 13, 14). Briefly, female recipient mice were lethally irradiated with 700 cGy from a Gammacel 137Cs source (J. L. Shepherd & Associates, San Fernando, CA). Approximately 6 h later they were injected by tail vein with male donor bone marrow (1 x 106/mouse) and spleen (2 x 106/mouse) cells suspended in RPMI 1640 (BioWhittaker, Walkersville, MD) with 20 U/ml heparin (Fisher Scientific, Pittsburgh, PA). The dose of cells was a standard one used routinely in the generation of GVHD (13, 14). A control group of female BALB/c recipient mice received male BALB/c spleen and bone marrow cells (syngeneic BMT, referred to as control animals). Animals that did not engraft died within 7–10 days. Transplanted animals were maintained in sterile MicroIsolator cages (Lab Products, Seaford, DE) and supplied with autoclaved food and acidified water. Three to five animals per group (Scl GVHD or control) and per time point were studied in each experiment.
Anti-TGF- Ab treatment
Mice were given two doses of 150 µg anti-pan TGF- Abs
(rabbit polyclonal IgG; Sigma, St. Louis, MO) by tail vein injection on
days 1 and 6 post-BMT as previously described (1). The
dose was selected as a standard one used for in vivo inhibition of
fibrosis in other mouse models (15, 16).
Collection of tissue
For these experiments three to five transplanted animals per group were sacrificed via cervical dislocation on days 7, 14, and 21 post-BMT. Days 14 and 21 were chosen because they are the earliest time points at which there are reliable changes in skin thickening and inflammation. Back skin was depilated and harvested for RNA extraction (snap-frozen in liquid nitrogen and stored at -80°C), flow cytometry (see below), immunostaining (frozen on dry ice and stored at -80°C), and routine histologic staining (fixed in 10% buffered formalin (Surgipath Medical Industries, Richmond, IL) and paraffin-embedded). Tongue and lung were also collected for routine histology to confirm the development of Scl GVHD.
Histologic and morphometric analysis
Formalin-fixed, paraffin-embedded sections of tissue were stained by H&E (Surgipath Medical Industries, Richmond, IL). Frozen skin was embedded in OCT embedding medium (Miles, Elkhart, IN) and sectioned by cryostat (Leica CM1800, Nussloch, Germany) for immunostaining (described below). For morphometric analysis, histologic sections of lung or back skin were scanned with a CCD camera (Optronics, Goleta, CA) using an Axiophot photomicroscope system (Zeiss, Oberkochen, Germany), stored as tiff files, and subjected to image analysis (Optimas 6.1, Bothell, WA). Skin thickening was evaluated for each animal as previously described (1). Quantification of immunostaining by image analysis was performed on slides stained with specific Abs and corresponding isotypes. Isotype control Ab staining was always tested on the same slide as specific Ab staining and subtracted in the analysis. Areas were calculated in arbitrary square units by outlining the dermis on a x10 view for each microscopic image. The same threshold settings were used on the set of slides stained with the same Ab. The density of positive immune cell staining within the outlined areas was plotted as the percentage of positive area. A minimum of six measurements was taken from two or more skin sections from each animal, and the variation among animals was expressed as the SE.
Antibodies
Immunostaining. Abs and corresponding isotype controls were as follows. Anti-CD11b (anti-Mac-1, mAb M1/70, rat IgG2b; BD PharMingen, San Diego, CA) was used to identify monocyte/macrophages, and anti-class A scavenger receptor type I and II Ab (2F8, IgG2b, Serotec, Raleigh, NC) was used to identify mature monocyte/macrophages. Anti-CD3 (mAb 17A2, rat IgG2b; BD PharMingen) was used to identify T cells. Anti-I-Ad (mAb AMS-32.1, mouse IgG2b) was used to detect immune cell activation status (MHC II). Mouse collagen I (rabbit anti-mouse polyclonal antiserum, Chemicon, Temecula, CA) was used to identify type I collagen protein in tissue. Goat anti-rat IgG-biotin or goat anti-rabbit IgG-biotin (Vector Laboratories, Burlingame, CA) followed by peroxidase-labeled streptavidin (Kirkegaard & Perry Laboratories, Gaithersburg, MD) and diaminobenzidine (Kirkegaard & Perry Laboratories) was used for detection. Diaminobenzidine orange enhancer (Kirkegaard & Perry Laboratories) was used for type I collagen staining. Isotype control Abs were rat anti-IgG2b (R35-38; BD PharMingen), mouse IgG2b and normal rabbit IgG (Santa Cruz Biotechnology, Santa Cruz, CA).
Flow cytometry. Direct single stains of CD3 were performed for T cells (mAb 17A2, rat IgG2b-PE). Anti-CD45 (mAb 30-F11, rat IgG2b-FITC; BD PharMingen) was used to identify all leukocytes of hemopoietic origin. Indirect two-color stains were performed for anti-CD11b (anti-Mac-1, mAb M1/70, rat IgG2b-PE) and MCP-1 (mAb 2H5, hamster IgG; BD PharMingen) using F(ab')2 anti-rat IgG-FITC (Jackson ImmunoResearch Laboratories, West Grove, PA) to detect bound Ab. The corresponding isotype controls were used for instrument set-up. All specific and isotype control Abs for two-color flow cytometry were obtained from BD PharMingen. Indirect single-color stains were performed for VLA-4 (CD49d, mAb 9C10, rat IgG2a; BD PharMingen), type A I/II scavenger receptors (mAb 2F8), and macrophage receptor with collagenous structure (MARCO) (mAb ED31, rat IgG1; Serotec) using F(ab')2 anti-rat IgG-FITC (Jackson ImmunoResearch Laboratories) to detect bound Ab. NK cells were detected by single stains using an anti-pan NK Ab (mAb DX5, rat IgM; BD PharMingen) and F(ab')2 anti-rat IgM-PE (Jackson ImmunoResearch Laboratories) as the detecting Ab. Direct two-color stains were performed for anti-I-Ad (mAb AMS-32.1, mouse IgG2b-FITC) and anti-CD11b (anti-Mac-1, mAb M1/70, rat IgG2b-PE) and for anti-I-Ad and anti-CD45 (mAb LY-5, rat IgG2b-FITC) using the corresponding isotype controls and dermal cells from a normal mouse for instrument set-up. All specific and isotype Abs for two-color flow cytometry were obtained from BD PharMingen. Blocking steps included purified murine IgG to inhibit nonspecific staining.
Preparation of skin cell suspensions for flow cytometry
As described previously (1), small pieces of depilated skin were digested in RPMI containing 10 mM HEPES (Irvine Scientific, Santa Anna, CA), 0.01% DNase (Sigma), 0.27% collagenase (Sigma), and 1000 U hyaluronidase (Sigma) at 37°C for 2 h. The digested skin was filtered through 100-µm pore size nylon meshes to generate a single-cell suspension of skin cells containing resident cells (keratinocytes, fibroblasts, endothelial cells, and perivascular cells such as mast cells) and infiltrating cells (lymphocytes, monocytes, and NK cells). This is a standard method for analysis of cutaneous immune cells (17). Approximately 4 x 106 cells were typically obtained from a 1 x 2-cm2 piece of skin for control mice, and 8 x 106 cells were obtained from mice with Scl GVHD on day 21. Before specific Ab staining, all isolated skin cells were blocked with purified murine IgG (1 µg/106 cells; Sigma) for 5 min on ice. For MCP-1 staining, permeabilization buffer containing 1% saponin was used as washing and staining buffers. Specific and isotype-matched Abs were then directly applied (1 µg/106 cells). For MCP-1 indirect stains the FITC-labeled secondary Ab was added at 0.5 µg/106 cells. Before fixation with 1% paraformaldehyde in PBS, all samples were washed twice in PBS supplemented with 1% BSA, 1% FCS, and 0.05% sodium azide. Sample data were acquired on a Becton DickinsonFACScan (Franklin Lakes, NJ) and analyzed using Cell Quest software.
Immunohistochemistry
Immunostaining was performed on acetone-fixed frozen sections. Staining was performed at least three times on each specimen by standard methods (18). In all immunostaining experiments, specific Ab staining was always compared with isotype control-stained sections on the same slide (see histologic and morphometric analysis).
Isolation of monocytes and T cells by magnetic bead separation
Single-cell suspensions of skin cells from the backs of experimental animals with Scl GVHD were prepared as described above. For positive selection of monocyte/macrophages and T cells, MACS CD11b microbeads (Miltenyi Biotec, Auburn, CA) or Thy-1.2 beads (Miltenyi Biotec) were incubated with the skin cells and then applied to a MidiMACS separation column (Miltenyi Biotec). The purity of monocytes and T cells was determined by flow cytometry (>90%) for each population. The isolated monocyte/macrophages, T cells, and residual cells (remaining skin cells after monocyte and T cell separation) were then used for RNA purification and RT-PCR analysis.
RNA and genomic DNA purification
Animals were sacrificed by cervical dislocation at each time point. Dissected depilated back skin was snap-frozen in liquid N2 and stored at -80°C until used for RNA or DNA isolation. RNA was extracted using TRIzol reagent (Life Technologies, Gaithersburg, MD) as described previously (1). Genomic DNA was extracted using standard methods (19).
Semiquantitative RT-PCR and PCR analysis
As previously described for analysis of other cutaneous cytokine
mRNAs (20), PCR reactions contained RT reaction products;
specific oligonucleotide primers for TGF-1 (Clontech, Palo Alto,
CA), TGF-2, TGF-3, MCP-1, MIP-1, RANTES,
pro(1)I collagen, and -actin (Table I
); 10x PCR buffer (Perkin-Elmer,
Norwalk, CT); nucleotide mix (Promega, Madison, WI); and 2 U Taq DNA
platinum polymerase (Life Technologies) in a volume of 50 µl.
Y-chromosome sequence analysis was performed on extracted genomic DNA
using a PCR primer set specific for both X- and Y-chromosome copies of
the supernumerary marker chromosome (SMC) gene
(21). Reactions were heated to 94°C for 5 min, followed
by denaturation at 94°C for 1 min, annealing at 60°C for 1 min, and
extension at 72°C for 2 min using GeneAmp 9600 PCR System
(Perkin-Elmer). Reactions were also performed in the absence of reverse
transcriptase and were always negative. The cycle number for each
system was chosen (cytokines and chemokines, 36; SMC, 40), so that all
signals were in the linear range on ethidium bromide-stained gels,
which were photographed and acquired via GelDoc (Bio-Rad, Hercules,
CA). The bands were then analyzed by image analysis using Optimas 6.1
software, and the results were expressed as relative density for each
PCR product following normalization for the DNA loading amount based on
the -actin band.
|
All data were expressed as the mean ± SE, and unpaired t test (two-tailed) was used for statistical significance to determine differences among means of treatment, experimental, and control groups. Differences with p < 0.05 were considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 mRNA up-regulation occurs
as early as day 7, and skin thickening is detectable by days 14–21
(1). Here we report that up-regulation of mRNA for
cutaneous TGF-1 (not TGF-2 or -3) and C-C chemokines precedes
influx of donor immune cells (mainly monocyte/macrophages and T cells)
and skin thickening. Both monocyte/macrophages and T cells, but not
fibroblasts, appear to be the source of TGF-1. We also document
up-regulation of scavenger receptor molecules, VLA-4 and I-A, on immune
cells, suggesting activation and Ag presentation. Donor cells infiltrate skin in early Scl GVHD
We observed numerous mononuclear cells infiltrating thickened skin
at early time points in Scl GVHD, but not in control animals (Fig. 1). By routine histology these were a
mixture of large pale oval monocyte/macrophages (arrowhead) and
smaller, darker, more compact T cells (arrow). Immunostaining and flow
cytometric analysis confirmed the histologic impressions. This
histology can also be seen in human early morphea and early
scleroderma, which can both be highly inflammatory. To determine
whether these cells infiltrating skin are of donor or host origin, we
transplanted bone marrow and spleen cells from male mice into female
recipient mice and detected Y-chromosome sequences by PCR analysis of
total cutaneous cellular DNA. We used a PCR primer pair that amplifies
a gene (SMC) on both X- and Y-chromosomes (Table I
) (21).
On days 14 and 21 post-transplantation, the smaller SMCY band was seen
when DNA from the skin of female experimental animals with Scl GVHD was
analyzed on ethidium bromide-stained gels, but was absent in the DNA
from skin of female control mice (Fig. 2;
the day 21 control is not shown, but was identical with the day 14
control). We did not test day 7 in this experiment. Therefore, male
donor cells infiltrate the skin of female recipient mice with Scl GVHD
by day 14 post-BMT and may play a role in the resulting skin fibrosis.
The analysis of Y-chromosome sequences by PCR analysis and ethidium
bromide staining in a gel is a relatively insensitive assay for donor
cells, because many resident recipient skin cells express SMCX
sequences (keratinocytes, fibroblasts, and endothelial cells).
Therefore, our evaluation of donor cells infiltrating skin may be an
underestimate, and quantification of donor cells by this method was not
useful. Our experiments demonstrate that donor cells are present and
presumably functional.
|
|
To further evaluate cells infiltrating skin in Scl GVHD, we
prepared single-cell preparations from back skin on days 14 and 21
post-BMT when skin thickening is evident (see Materials and
Methods). This technique provides ample numbers of cells for
flow cytometric analysis with staining for several different surface
markers, reliable from experiment to experiment. The total number of
cells in skin is not as critical as the relative numbers of skin cells
of each type (T cell, monocyte/macrophage, NK cell), and we made no
attempt to quantify cells per square millimeter of skin. Therefore, we
expressed the data as a percentage of the total skin cells. On day 14
post-BMT there was a 2-fold increase (8–18%) in
CD45+ cells in the skin of animals with Scl GVHD
(E) that was further increased (22 to 43%) on day 21 post-BMT compared
with control animals (Fig. 3A
plot). The increase in CD45+ cutaneous immune
cells in controls from days 14 to 21 may represent nonspecific effects
of irradiation and transplantation that later disappear by
immunostaining (data not shown) and do not result in skin fibrosis. The
percentage of CD45+ cells in control animals on
day 21 seemed abnormally high compared with the low values for
CD11b+, CD3+, and NK cells
at the same time, but the results were consistent in all control and
experimental animals in each group (n = 3). The
infiltrating cells in Scl GVHD were primarily
CD11b+ or CD3+. By day 21
post-BMT, a small number of NK cells (<10% of total skin cells)
appeared in the inflamed skin (flow histogram not shown). In these flow
experiments, neutrophils, which can also express CD11b, were gated out
by light scatter. However, neutrophils were rarely seen in skin routine
histology (Fig. 1) and were therefore unlikely to be a prominent cell
population by flow cytometry.
|
To confirm that the predominant CD11b-expressing cells in skin on
days 14 and 21 were mature macrophages and not NK cells or neutrophils,
we performed flow cytometric analysis for macrophage scavenger receptor
using Abs that recognize scavenger receptor A (ScR-A) types I
and II (mAb 2F8) (22) and MARCO (Ab ED31)
(23). The percentage of total skin cells expressing 2F8
and MARCO was elevated by day 14 post-BMT, with marked elevation by day
21 post-BMT (Fig. 3B plot). Furthermore, most of the
CD11b-positive cells in Scl GVHD skin on day 21 also expressd ScR-A
when double staining was performed (2F8 or MARCO; Fig. 4), and the proportion of the
CD11b+ infiltrating cells expressing 2F8
increased further by day 21 post-BMT. ScR-A type I and II (2F8)
expression was greater than that of MARCO. Therefore, ScR-A-expressing
macrophages are the predominant cells infiltrating the skin of animals
with Scl GVHD at early time points, and their influx accompanies skin
thickening.
|
Because macrophage scavenger receptors (MARCO and ScR-A types I
and II), VLA-4, and CD11b are up-regulated in activated macrophages, we
examined the level of expression of these surface markers per cell
using flow cytometric analysis, comparing the mean fluorescence
intensity of positive cells from animals with Scl GVHD vs that of cells
from syngeneic BMT control animals (summarized in Table II, depicted in Fig. 3B
overlays). Our results suggest that there may be active Ag presentation
in skin at early time points in Scl GVHD. On day 14 post-BMT, VLA-4,
MARCO, and ScR-A types I and II were up-regulated on macrophages by
34–76% (Table III). The mean
fluorescence intensity of cells expressing MARCO and 2F8 did not
increase further on day 21. The values were similar for MARCO, VLA-4,
and CD11b on day 21, perhaps reflecting an already up-regulated state.
We also evaluated histocompatibility class II molecules using an Ab
specific to I-Ad that recognizes immune cells of
both donor and recipient animals. We found elevated levels on
CD45-positive cells (Table III). In contrast to the up-regulation of
scavenger receptor molecules, I-A expression on total
CD45+ and CD11b+ cells was
increased on day 14 post-BMT and remained increased on day 21
post-BMT.
|
|
, and RANTES mRNA expression is elevated
during early murine Scl GVHD
In experiments reported previously (1) and here we
determined that CD11b+ monocyte/macrophages are
the predominant cells infiltrating skin at early time points in Scl
GVHD. To test whether chemokines might play a role in attracting these
monocyte/macrophages to skin, we performed semiquantitative RT-PCR
analysis of total skin RNA to examine the expression of cutaneous
MCP-1, MIP-1, and RANTES mRNA in Scl GVHD. Messages for all these
C-C chemokines were increased in Scl GVHD skin before skin thickening
and before infiltration of CD45+ cells was
evident (Fig. 5). On day 7 post-BMT,
MCP-1, MIP-1, and RANTES mRNAs were elevated by approximately 1.5-,
2.1, and 3.7-fold, respectively, in experimental animals compared with
controls. These data are plotted in Fig. 5B and compared
with influx of CD45+ cells and skin thickness by
image analysis (summary plot of flow cytometric analysis and image
analysis of immunostaining and skin thickness; original data not
shown). On day 14, MCP-1, MIP-1, and RANTES mRNA levels remained
elevated, and skin thickening was detectable in Scl GVHD, but not
controls. By day 21 post-BMT, when skin was markedly thickened
(>40%), both MCP-1 and MIP-1 mRNA were elevated 2.5-fold,
while RANTES mRNA was increased 1.9-fold (Fig. 5A). On day 21 we
performed flow cytometric analysis of skin cell suspensions
double-stained with CD11b and MCP-1. Neutrophils can also express
CD11b, but they are not present in routine histology and were gated out
by light scatter in the flow experiments. We found that the number of
CD11b+MCP-1+ cells is
elevated in animals with Scl GVHD (10.9% of total dermal cells
compared with 4% in controls). MCP-1+ cells were
predominantly CD11b+ cells, because
CD11b+ cells were 16–30% of the total cells in
skin (Fig. 5C). Although other cells (endothelial cells,
keratinocytes) may also secrete MCP-1, their numbers in skin were very
low compared with the predominant CD11b+
population in Scl GVHD mice. Therefore, activated donor
monocyte/macrophages may produce their own chemoattractant in an
autocrine loop at this time point.
|
1, but not TGF-2 or -3, mRNA is increased
during early murine Scl GVHD
TGF-1 is a potent fibrogenic cytokine that is known to
induce collagen synthesis by fibroblasts in vitro and in vivo. It has
been implicated in the pathophysiology of scleroderma by several
different methods (24), but the other TGF- isoforms
(TGF-2 and TGF-3) may also be involved. We previously reported
that at an early time point (day 7 post-BMT), TGF-1 mRNA levels are
elevated approximately 3- to 5-fold by RT-PCR analysis of the skin of
experimental animals with Scl GVHD vs syngeneic BMT control animals
(1). To characterize the contribution of other TGF-
isoforms to fibrosis in the Scl GVHD model, semiquantitative RT-PCR
analysis of total skin RNA was performed. We repeated the TGF-1 PCR
experiment and tested for TGF-2 and -3. TGF-1 mRNA levels in
this set of experiments were approximately 4- to 5-fold higher on day 7
and 2-fold higher in experimental animals than in controls on day 21
post-BMT (Fig. 6A). At early
time points (days 7 and 14), no changes in cutaneous TGF-2 and -3
mRNAs were seen in experimental animals compared with controls (Fig. 6A). Only very slight increases in TGF-2 and -3 mRNAs
were observed at later time points (day 21 post-BMT) in experimental
animals (statistically insignificant). Therefore, TGF-1 is probably
the critical isoform of TGF- driving the cutaneous fibrosis in Scl
GVHD and in early fibrosis of scleroderma.
|
1
in murine Scl GVHD
We determined that both monocyte/macrophages and T cells in skin
make TGF-1 mRNA (Fig. 6B) by RT-PCR analysis of total RNA
from magnetic bead-separated skin cells on days 14 and 21 post-BMT.
Residual, negatively selected cells (keratinocytes, fibroblasts,
endothelial cells, and other resident skin cells) did not make
significant TGF-1 mRNA. The amount of TGF-1 mRNA per µg RNA
expressed by monocyte/macrophages was approximately 1.9- and 1.4-fold
more than that expressed by T cells on days 14 and day 21 post-BMT.
This suggests that monocyte/macrophages are the main source of TGF-1
in Scl GVHD skin, since CD11b+ cells were
approximately 2- to 3-fold more numerous than T cells. These data also
suggest that fibroblasts and endothelial cells present in the residual
cell preparation do not make significant amounts of TGF-1 mRNA. We
chose mouse Thy-1.2 microbeads for the positive selection of mouse T
cells, a standard method (25). The purity was >90% by
flow analysis when we analyzed the Thy-1.2 bead-selected cells stained
with anti-CD3 Ab. Although Thy-1.2 may be expressed on other cell
populations, such as endothelial cells, the numbers of non-T cells
expressing Thy-1.2 in skin would be very small compared with the number
of cutaneous infiltrating T cells.
Anti-TGF- Ab treatment reduces CD11b+,
2F8+ monocyte/macrophage influx into skin in murine Scl
GVHD
We hypothesize that monocyte activation by host-reactive donor T
cells is an initiating event in Scl GVHD and scleroderma that may lead
to increased cutaneous TGF- production and skin fibrosis. We
previously demonstrated that starting on day 7 and by day 14 post-BMT,
the percentages of CD11b+ monocyte/macrophages
infiltrating skin are markedly increased, and by day 21,
monocyte/macrophages are increased approximately 5- to 6-fold in Scl
GVHD compared with control animals. This early infiltration of
monocyte/macrophages is accompanied by up-regulation of TGF-1 mRNA
synthesis and prominent skin thickening (1).
We have shown that 300 µg anti-TGF- Ab/animal successfully
prevents skin and lung fibrosis in experimental animals with Scl GVHD
(1). An unexpected finding not presented in that report
was the observation that the numbers of immune cells infiltrating skin
in anti-TGF--treated animals were markedly reduced. To further
investigate this observation, we repeated the experiments and examined
the effect of pan-specific anti-TGF- Ab treatment on types of
cells infiltrating skin using flow cytometric analysis (not shown) and
immunostaining for monocyte/macrophages (CD11b and 2F8). We chose the
dose of anti-TGF- Ab (150 µg by tail vein injection on days 1
and 6 post-BMT) and the early time points because of our previously
reported results (1) and published data using
anti-TGF- Ab to prevent fibrosis in other mouse models of
fibrosis (15, 16). By day 21 post-BMT, infiltration of
skin by CD11b+ and 2F8+
mononuclear cells was effectively blocked by anti-TGF- Abs in
Scl GVHD (Fig. 7, A and
B). The percentage of CD11b+ cells in
anti-TGF- Ab-treated experimental animals that did not develop
skin and lung fibrosis (4.8 ± 0.2% of the total skin cells by
flow analysis (not shown), 6.2 ± 2.8% by immunostaining) was
comparable to that in control animals (4.2 ± 0.3% by flow
analysis (not shown) and 1.6 ± 0.3% by immunostaining).
Untreated animals with Scl GVHD had 25.1 ± 6.4%
CD11b+ cells by flow analysis (not shown) and
28.0 ± 2.6% by immunostaining (plotted in Fig. 8). Staining with 2F8, another macrophage
marker, gave similar results, which confirmed CD11b staining data (Fig. 7B, plotted in Fig. 8).
|
|
Ab treatment reduces T cell infiltration into skin in
murine Scl GVHD
In classic GVHD, donor T cells initiate disease by recognizing
recipient Ag as foreign and providing signals to cytotoxic T cells and
NK cells, causing epithelial injury. In Scl GVHD, donor T cells may
activate monocyte/macrophages, thereby initiating fibrosing disease via
a different subset of effector cells (note the absence of
significant epithelial injury in Fig. 1). Therefore, we analyzed
cutaneous T cells as well as monocyte/macrophages in Scl GVHD. By day
14 post-BMT, there was a >4-fold increase in the number of cutaneous
CD3+ T cells (5–8% of total skin cells) by flow
cytometric analysis of Scl GVHD animals compared with syngeneic BMT
animals (Fig. 3A; flow histogram not shown). By day 21
post-BMT, the number of T cells was increased further (17.3 ±
5.9% of total skin cells) in experimental animals. In contrast, in
anti-TGF- Ab-treated experimental animals T cell infiltration of
skin was markedly reduced and comparable to that in controls, as shown
in the immunostaining data (Fig. 7C, plotted in Fig. 8).
Therefore, anti-TGF- Abs significantly block cutaneous influx by
CD3+ T cells as well as monocyte/macrophages.
Anti-TGF- Ab treatment reduces I-A+ cells in skin in
murine Scl GVHD
Anti-TGF- Ab treatment also reduced the numbers of
I-Ad-positive cells in skin of Scl GVHD mice
(Fig. 7D, plotted in Fig. 8).
Anti-TGF- Ab decreases skin type I collagen synthesis in murine
Scl GVHD
By image analysis, skin and lung fibrosis were abrogated by the
administration of blocking Abs to TGF- in murine Scl GVHD
(1). Immunostaining and RT-PCR analysis were performed to
detect type I collagen synthesis. By day 21 post-BMT, type I collagen
protein (Fig. 7E, plotted in Fig. 8) in skin of experimental
animals receiving anti-TGF- Abs was comparable to that in
controls. Pro(1)I collagen mRNA expressed in
skin of anti-TGF- Ab-treated mice with Scl GVHD was actually
less than that in controls (Fig. 9) as
determined by semiquantitative RT-PCR analysis of total RNA.
|
treatment blocks the elevation of cutaneous TGF--1
mRNA in murine Scl GVHD
Because we hypothesized that skin fibrosis is TGF-1 driven,
RT-PCR was performed on RNA from skin of anti-TGF- Ab-treated
experimental animals to determine whether anti-TGF- Ab has any
effect on TGF-1 mRNA expression. By day 21 post-BMT, cutaneous
TGF-1 mRNA expressed by experimental animals receiving
anti-TGF- Abs was comparable to the baseline in controls (Fig. 9), consistent with the marked reduction in cutaneous mononuclear cell
infiltrates and abrogation of skin thickening in treated animals.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 and
MCP-1, MIP-1, and RANTES chemokine mRNA) is temporally related to
increased collagen mRNA synthesis, skin thickening, and lung fibrosis
(summarized in Table IV), a composite of
multiple experiments. We have also shown that macrophages expressing
markers of activation and Ag presentation (CD11b, I-A, SR-A, and VLA-4)
are the predominant cells infiltrating skin during early time points in
murine Scl GVHD, when skin thickening is first apparent. To our
knowledge this is the first description of macrophage SR-A induction in
scleroderma or Scl GVHD. Abs to TGF- prevent not only skin
thickening and lung fibrosis, but also the infiltration and possible
activation of mononuclear cells into skin. If immune cell migration is
prevented, skin thickening does not occur. Our data provide a
foundation for interventions in early scleroderma at several different
points in the immune cascade of Scl GVHD: chemokine production, T cell
and monocyte/macrophage activation and homing to skin, T cell and
monocyte function in skin, and direct inhibition of fibrogenic TGF-1
itself. We also show that presumably functional donor immune cells
infiltrate skin early in disease.
|
Scleroderma is a disease of unknown etiology that occurs most
commonly in women after childbearing years. Recent reports describe
persistent HLA-compatible fetal cells in the skin and blood of women
with scleroderma that occur at a much higher incidence than in healthy
women (26, 27). These data suggest that a state of
microchimerism could lead to a chronic graft-vs-host (or host-vs-graft)
type of reaction in these women with scleroderma (28). In
our studies numerous mononuclear cells infiltrating skin at early time
points in Scl GVHD are of donor origin, since we detected
Y-chromosome-specific sequences from donor male animals in the skin of
female experimental animals with Scl GVHD, but not in controls (Fig. 2). The presence of donor cells in the skin by day 14 post-BMT also
correlates well with these data and with our previously published
results showing increased levels of TGF-1 by day 7, T cell and
monocyte/macrophage infiltration into skin by day 14 by routine
histology and flow cytometric analysis, detectable skin thickening by
day 14, and subsequent collagen mRNA up-regulation by day 21
(1). We are examining the Ag-presenting capability of
cells infiltrating skin in this model in separate experiments. If donor
mononuclear cells are involved in initiating Scl GVHD, can disease be
effectively treated by the administration of specific Abs or
antagonists directly or indirectly against the infiltrating effector
cells and their functions? Possible antagonists include Abs to
activation/homing markers of monocytes (VLA-4, CD11b), integrins on
endothelial cells, and blocking peptides or molecules to abrogate
immune cell signaling. Is fibrosing disease reversible? Can disease be
transmitted and accelerated by adoptive transfer of cells?
Macrophage scavenger receptors and autoimmunity
Macrophage scavenger receptors are a diverse family of proteins
that bind a wide variety of ligands (reviewed in Ref. 29).
ScR-A are pattern-recognition receptors that have an intrinsic ability
to recognize specific elements unique and essential to self-vs-nonself
discrimination (29, 30). ScR-A have been implicated in the
recognition and phagocytosis of apoptotic thymocytes and have been
strongly implicated in contributing to the development of
atherosclerotic plaques in heart disease (31, 32). ScR-A
type I and II (detected with mAb 2F8) expression has been identified in
marginal zone macrophages of spleen, alveolar macrophages, and
macrophages of heart, gut, and cortical and medullary regions of thymus
(29). MARCO expression is far more restricted and has been
localized primarily to macrophages in splenic marginal zone area where
active Ag presentation occurs. MARCO expression can be induced on
tissue macrophages in response to inflammatory stimuli (33, 34). MARCO-expressing cells are highly phagocytic macrophages
that have been implicated in Ag processing and the induction of
anti-self immune responses due to their ability to present modified
self-Ags. Ligation of macrophage ScR-A does not cause costimulatory
molecule up-regulation, which could explain the minimal CD11b
up-regulation in cutaneous macrophages in Scl GVHD (Table II)
(33).
We have not yet determined whether ScR-A up-regulation during Scl GVHD is due to cutaneous influx of already activated immune cells or maturation of newly infiltrating monocytes to macrophages in skin, with ScR-A up-regulation on cutaneous cells during Ag presentation. Activation could occur very early after BMT, before our first time point at day 7 in these experiments. The up-regulation of ScR-A, however, in conjunction with an increase in I-A molecules (an MHC class II Ag) suggests that there may be active Ag presentation at early time points in cutaneous Scl GVHD. The involvement of scavenger receptors in this putative Ag presentation may explain the development of GVHD due to presentation of modified self Ags by activated macrophages. The concomitant early events of macrophage ScR-A activation and extensive T cell and macrophage infiltration into skin is a critical area of investigation to identify steps that trigger autoimmunity in Scl GVHD and scleroderma. Both protein (including 2F8 Ab) and nonpeptide ScR-A inhibitors have been described that will be interesting to test in our murine Scl GVHD model (35, 36).
TGF-1 is the critical cytokine driving cutaneous fibrosis during
Scl GVHD
The TGF- family of closely related peptides includes a number
of isoforms. The major isoforms of TGF- are TGF-1, -2, and
-3. TGF- isoforms have many different effects in vivo, including
stimulation of collagen synthesis, chemoattraction, modulation of
immune cells, inhibition of epithelial proliferation, and
differentiation of hemopoietic precursors to dendritic cells
(37). TGF-1 is a known potent stimulus for fibroblast
collagen synthesis (24). TGF-2, and -3 may also be
involved in various fibrosing diseases (38, 39). However,
our data demonstrate that TGF-1 appears to be the critical isoform
driving cutaneous fibrosis in Scl GVHD because TGF-1 mRNA is
elevated in animals with Scl GVHD compared with controls at the
earliest time point (day 7 post-BMT), while the other isoforms are
approximately equivalent. We have not looked at time points earlier
than 7 days post-BMT. High TGF-1 mRNA levels appear to paradoxically
precede prominent monocyte/macrophage and T cell influx. However, there
could be a few potent TGF--producing cells at the early time points.
Our experiments do not evaluate that possibility. Secondly, TGF-1
itself is a very strong chemoattractant for mononuclear cells,
particularly monocyte/macrophages (40, 41) and
fibroblasts, and may recruit monocyte/macrophages via chemotaxis and/or
increased monocyte/matrix adhesion. In sites of inflammation, several
types of cells, including monocyte/macrophages, T cells, and even
fibroblasts, are able to produce TGF- (42). Our data
demonstrate that both cutaneous macrophages and T cells isolated by
magnetic bead separation produce TGF-1 mRNA in animals with Scl
GVHD. We have not directly examined TGF- mRNA production by
fibroblasts and endothelial cells, but they are represented in the
residual cells, which have no detectable TGF-1 mRNA production by
RT-PCR analysis (Fig. 6B). In early Scl GVHD, TGF-1
produced by activated T cells and/or monocyte/macrophages activated by
donor T cells may coordinate with MCP-1 and RANTES in attracting more
monocyte/macrophages and T cells to skin. By blocking TGF- with Abs,
we may have also affected the chemoattractant function of TGF-.
Those infiltrating cells responding to TGF-1 are also capable of
producing TGF-1, as demonstrated by the increased mRNA expression by
RT-PCR (Fig. 6B). Cutaneous fibroblasts are stimulated to
synthesize collagens, thereby causing increased collagen deposition and
skin fibrosis.
C-C chemokines may be involved in the pathogenesis of early Scl GVHD
Chemokines may be a potential new target for the treatment of fibrosing disease.
Chemokines and recruitment of immune cells.
It is well established that chemokines are produced locally in tissue
and can selectively recruit different subsets of leukocytes to
inflammatory sites (43). MCP-1, MIP-1, and RANTES
belong to the C-C chemokine family and attract mainly monocytes and T
cells, respectively. In our studies (Fig. 5), C-C chemokines MCP-1,
MIP-1, and RANTES mRNA are elevated in experimental animals with Scl
GVHD at a very early time point (day 7 post-BMT) before significant
numbers of CD45+ immune cells, including
monocytes, infiltrate skin. At later time points (days 14 and 21
post-BMT), MCP-1 and MIP-1 mRNA remain elevated and show
further increases, paralleled by a significant increase in the numbers
of monocyte/macrophages infiltrating skin and subsequent skin
thickening. Our data suggest the biologic relevance of these chemokines
in fibrosis. In contrast, the early up-regulation of RANTES mRNA is
followed by a decrease on day 21 post-BMT, in almost a mirror image
pattern compared with monocyte chemokines. This is an intriguing
observation, suggesting interplay between early T cell activation of
monocytes, then possible dampening of the early T cell effect and
replacement by a monocyte effect as the disease progresses. We are
exploring this hypothesis in separate experiments.
Chemokine effects on matrix deposition.
Chemokines may also affect the homeostasis of extracellular matrix
itself. MCP-1 and RANTES may be involved in the fibrotic pathway by
modulating collagen turnover or type I and IV collagen deposition
directly (by sending signals to fibroblasts via macrophages) (7, 11) or indirectly through the stimulation of TGF-
(44) (Fig. 1).
Chemokine interactions with TGF-.
Our data show that mRNA levels of TGF-1, MCP-1, and RANTES are
increased at early time points in Scl GVHD (Fig. 5). TGF-1 can
induce the up-regulation of MCP-1 and RANTES expression at both mRNA
and protein levels in vitro and in vivo. MCP-1 and MIP-1 can
increase the secretion of TGF- from macrophages, which, in turn,
increases the expression of collagen types I and III
(45, 46, 47), suggesting that complex interactions between
these C-C chemokines and TGF-1 may occur in our model, as in other
inflammatory conditions.
In summary, TGF-1 is one of the most important cytokines in
stimulating collagen synthesis and matrix deposition; however, fibrosis
is a complex process that may involve multiple cytokines and
chemokines, for which the interactions are incompletely understood.
Effects of anti-TGF- Ab in Scl GVHD
Polyclonal anti-TGF- Abs that block all TGF- isoforms
appear to have complex and multiple effects in Scl GVHD when
administered early in disease. We have shown that when mononuclear cell
(mainly monocytes and T cells) migration into skin is blocked,
up-regulation of macrophage activation markers (ScR-A and I-A) is
decreased, TGF-1 mRNA levels are not elevated, type I collagen mRNA
and protein synthesis are reduced, and skin thickening does not occur.
This inhibition of fibrosis may occur via chemokine-TGF-
interactions, monocyte/macrophage activation, and/or monocyte homing or
at multiple levels. The Scl GVHD model provides a unique opportunity to
further investigate these pathways in vivo, to better understand
monocyte/macrophage function, and to design novel interventions for
fibrosing diseases, including scleroderma and Scl GVHD.
Other potential inhibitors of fibrosis and treatments for scleroderma
In addition to TGF-, C-C chemokines, particularly MCP-1 and
RANTES, may be actively involved in Scl GVHD by attracting
monocyte/macrophages and T cells into skin and possibly interacting
with TGF-1, thereby affecting collagen deposition and contributing
to the progression of fibrosis. Thus, blocking Abs or peptide
antagonists to MCP-1, RANTES, or C-C chemokines and chemokine receptor
inhibitors may be potential new therapies for mononuclear cell-driven
progressive fibrosing diseases such as Scl GVHD and scleroderma. A
better understanding of the roles of these effector cells (T cells and
monocytes) may be useful in predicting the course of the disease
as well.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Y.Z. and L.L.M. contributed equally to this work.
3 Address correspondence and reprint requests to Dr. Anita C. Gilliam, Department of Dermatology, Case Western Reserve University, 11100 Euclid Avenue, Cleveland, OH 44106-5028. E-mail address: acg{at}po.cwru.edu
4 Abbreviations used in this paper: Scl GVHD, sclerodermatous graft-vs-host disease; BMT, bone marrow transplantation; MARCO, macrophage receptor with collagenous structure; MCP-1, macrophage chemoattractant protein-1; MIP-1, macrophage inflammatory protein-1; SMC, supernumerary marker chromosome; ScR-A, scavenger receptor A.
Received for publication March 7, 2001. Accepted for publication January 7, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
treatment prevents skin and lung fibrosis in murine sclerodermatous graft-versus-host disease: a model for human scleroderma. J. Immunol. 163:5693.1 gene expression by monocyte chemoattractant protein-1 via specific receptors. J. Biol. Chem. 271:17779. loop. J. Immunol. 164:6174.. J. Ex
