Scleroderma, a debilitating acquired connective tissue disease, is characterized by fibrosis, particularly of the skin and lungs. Monocyte-produced TGF-β1, a potent stimulus for collagen synthesis, is thought to drive the fibrosis. Here, we thoroughly characterize a murine sclerodermatous graft-vs-host disease (Scl GVHD) model for scleroderma that reproduces important features of scleroderma including skin thickening, lung fibrosis, and up-regulation of cutaneous collagen mRNA, which is preceded by monocyte infiltration and the up-regulation of cutaneous TGF-β1 mRNA. Most importantly, we can prevent fibrosis in both the skin and lungs of mice with Scl GVHD by inhibiting TGF-β with neutralizing Abs. The murine Scl GVHD model provides the unique opportunity to study basic immunologic mechanisms that drive fibrosing diseases and GVHD itself and will be useful for testing new therapies for these diseases.

PATIENTS with the systemic form of scleroderma (diffuse systemic sclerosis) may have involvement of skin, lungs, kidneys, gastrointestinal tract, and heart (1). There is no completely effective treatment for this rheumatic disease, and animal models in which to study the disease process are limited (2, 3). Significant progress in understanding the etiology of scleroderma was made in recent reports describing the presence of persistent HLA-compatible fetal cells in the skin and blood of women with scleroderma but not in healthy women (4, 5), suggesting that a graft-vs-host disease (GVHD)3-like rejection reaction due to microchimerism may be a predisposing factor in women with scleroderma after the child-bearing years. Therefore, we chose to use sclerodermatous GVHD (Scl GVHD) as a model for human scleroderma.

In scleroderma, overproduction of normal collagen at both mRNA and protein levels has been shown by many investigators (6, 7). TGF-β protein and messenger RNA have been localized to sites of mononuclear cell infiltration and fibroblast activation by immunostaining and in situ hybridization studies of lung and skin tissue from scleroderma patients (8, 9). We hypothesize that monocyte activation by host-reactive T cells is an initiating event in scleroderma and Scl GVHD. These monocytes infiltrate skin and produce TGF-β1, which then causes the collagen up-regulation leading to skin fibrosis. If this hypothesis is valid, it follows that TGF-β1 inhibition may prevent fibrosis in early disease (Fig. 1).

FIGURE 1.

Hypothesis to be tested: monocyte activation and homing to skin, followed by differentiation to a TGF-β1-producing effector cell in scleroderma and Scl GVHD. Antagonists to TGF-β (anti-TGF-β Abs) could potentially inhibit critical steps in this model.

FIGURE 1.

Hypothesis to be tested: monocyte activation and homing to skin, followed by differentiation to a TGF-β1-producing effector cell in scleroderma and Scl GVHD. Antagonists to TGF-β (anti-TGF-β Abs) could potentially inhibit critical steps in this model.

Close modal

To test our hypothesis, we first produced Scl GVHD by transplanting lethally irradiated BALB/c (H-2d) mice with B10.D2 (H-2d) bone marrow and spleen cells across minor histocompatibility loci (10, 11, 12). Control irradiated BALB/c animals were transplanted with syngeneic bone marrow and spleen cells. We show here that murine Scl GVHD faithfully reproduces important features of scleroderma, including skin thickening, lung fibrosis, the presence of cutaneous CD11b+ mononuclear cells, and the up-regulation of cutaneous TGF-β1 and collagen. Anti-TGF-β Abs inhibit skin and lung fibrosis. Studies using this model will translate readily into therapy for scleroderma and for GVHD itself, a debilitating consequence of bone marrow transplantation (BMT).

Seven- to eight-week-old female B10.D2 (H-2d) and BALB/c (H-2d, Jackson Laboratory, Bar Harbor, ME) mice were utilized as donors and recipients, respectively, for BMT to produce Scl GVHD (10, 11, 12). Briefly, recipient mice were lethally irradiated with 700 cGy from a Gammacel 137Cs source. Approximately 6 h later they were injected i.v. by tail vein with donor spleen (2 × 106/mouse) and bone marrow cells (1 × 106/mouse) suspended in RPMI 1640 (BioWhittaker, Frederick, MD) with 10 U/ml heparin (Fisher Scientific, Pittsburgh, PA) (12). A control group of BALB/c recipient mice received BALB/c spleen and bone marrow cells (syngeneic BMT, referred to as control animals). Transplanted animals were maintained in sterile Micro-Isolator cages (Lab Products, Seaford, DE) and supplied with autoclaved food and acidified water. The dose of donor cells used in these experiments was determined from the literature (10, 11, 12), and from pilot experiments (data not shown) in which an increasing number of spleen cells produced no additional significant skin thickening.

A total of 150 μg of anti-pan TGF-β Abs (rabbit polyclonal IgG, Sigma, St. Louis, MO) or 150 μg of control rabbit IgG (Sigma) were administered by tail vein injection on day 1 and again on day 6 post-BMT. The dose was selected as a standard one used for other mouse models (13, 14, 15). Mice were sacrificed at day 21 as described below.

Three animals per group (experimental or control BMT) per time point were sacrificed via cervical dislocation at days 7, 14, 21, 38, 49, and 75 post-BMT (36–50 animals per experiment). Back skin was depilated and harvested for RNA extraction (snap-frozen in liquid nitrogen), flow cytometry, immunostaining (frozen on dry ice), and routine histologic staining (fixed in 10% buffered formalin, Surgipath Medical Industries, Richmond, IL). Tongue and lung were also harvested, fixed in formalin, and embedded in paraffin for routine staining.

Formalin fixed, paraffin-embedded sections of tissue were stained by hematoxylin and eosin (Surgipath Medical Industries). 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, histological sections of lung or back skin were scanned with a CCD camera (Optronics, Goleta, CA) using an Axiophot photomicroscope system (C. Zeiss, Oberkochen, Germany), stored as TIFF files, and subjected to image analysis (Optimas 6.1, Bothell, WA). Areas were calculated in arbitrary square units by outlining the dermis on a 10× view for each microscopic image, in which length was fixed and thickness varied, giving an average thickness for a broad area of skin. A minimum of eight measurements were taken from two or more skin sections from each animal. For lung tissue, a minimum of eight measurements from one lobe of each lung were taken and the percentage of alveolar space was calculated.

Immunostaining.

Anti-CD11b (anti-Mac-1, mAb M1/70, rat IgG2b, PharMingen, San Diego, CA) was used to identify monocyte/macrophages. Goat anti-rat IgG-biotin (Vector Laboratories, Burlingame, CA) followed by Streptavidin alkaline phosphatase (Kirkegaard & Perry Laboratories, Gaithersburg, MD) and diaminobenzidine (Kirkegaard & Perry) were used for detection. Isotype control Abs were rat anti-IgG2b (R35-38, PharMingen).

Flow cytometry.

PE-labeled mAb M1/70 (PharMingen) was used to detect mononuclear CD11b-positive cells in skin. FITC-labeled mAb to CD3ε (145-2C11, Armenian Hamster IgG, PharMingen) was used to detect T cells in skin.

Staining on frozen sections of skin was performed at least three times on each specimen by standard methods (16).

Total RNA was isolated from snap-frozen skin by guanidinium HCl method (17) and stored at −80°C until use in RT-PCR analysis and RNase protection assays.

Specific oligonucleotide primers for TGF-β1 or G3PDH (both from Clontech, Palo Alto, CA) were employed in RT-PCR reactions with total RNA from skin as previously described for analysis of other cutaneous cytokine mRNAs using Gene Amp 9600 PCR System (Perkin-Elmer, Norwalk, CT) (18). The cycle number of 30 was chosen so that the TGF-β1 and GP3DH 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 expressed as the relative density for TGF-β1 following normalization for the RNA loading amount based on the GP3DH band.

We prepared riboprobes from cDNAs internally labeled with digoxigenin UTP according to the manufacturer’s instructions (Genius II kit, Boehringer Mannheim, Indianapolis, IN). RNase protection assays with a riboprobe for proα1(I) collagen (cDNA, a kind gift from E. Vuorio, University of Turku, Turku, Finland), (19) were performed according to the manufacturer’s instructions (RPAII kit, Ambion, Austin, TX) to assay and quantify collagen mRNA expression in mouse skin. Briefly, the gel-purified riboprobes were hybridized to mRNA prepared from skin and digested with RNase to remove nonhybridized sequences. The hybridization products were separated on a 5% nondenaturing polyacrylamide gel, electrophoretically transferred to HybondN+ nylon membrane (Amersham, Arlington Heights, IL), and detected with peroxidase-conjugated Abs to digoxigenin (Boehringer Mannheim) by chemiluminescence (Supersignal, Pierce, Rockford, IL). Images were obtained by exposure to x-ray film and the results analyzed by image analysis using Optimas 6.1 software. Collagen mRNA expression was normalized to a β actin or 28S rRNA control (Ambion).

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 of hyaluronidase (Sigma) at 37°C for 2 h (20). The digested dermis was filtered through nylon mesh to generate a single cell suspension of DCs containing resident cells (fibroblasts, endothelial cells, and perivascular cells) and infiltrating cells (lymphocytes and monocytes). The cells were stained with PE-labeled mAb for CD11b (PharMingen). After fixation with 1% paraformaldehyde in PBS, flow cytometry was performed with an Epics Elite Cytometer (Coulter, Hialeah, CA) and the data were analyzed with Coulter Elite software. Analysis of CD11b+ cells was performed with and without gating to exclude polymorphonuclear cells based on light scatter.

Lethally irradiated BALB/c mice transplanted via tail vein injection with bone marrow and spleen cells from B10.D2 mice develop skin thickening in the setting of mild histological and clinical changes of cutaneous GVHD (Scl GVHD). Animals that do not engraft die, and control BALB/c animals receiving syngeneic transplants do not develop GVHD or skin thickening. All engrafted animals show colony formation in the spleen, while the nonengrafted animals that died lacked spleen colony formation at autopsy. GVHD is clearly present in experimental animals, seen histologically as satellitosis and apoptotic keratinocytes in tongue epithelium and skin (data not shown). However, the numbers of dyskeratotic keratinocytes in epithelium, indicating severity of GVHD, are small (21). The form of GVHD is fibrotic, rather than cytotoxic. The experimental animals show no alopecia or diarrhea, but fail to gain weight, as expected, from 8 to 18 wk following BMT.

Sclerodermatous thickening of skin is detectable by day 21 post-BMT by image analysis of routine histopathological sections (Fig. 2,A), with an increase in total dermal area in animals with Scl GVHD of ∼40% over the syngeneic BMT control animals (Fig. 2 B). We have noted skin thickening as early as day 14, however 21 days is usually required to observe significant skin thickening in all transplanted animals. The variability in the earliest detectable skin thickening is most likely due to the success of the BMT and the quality of the spleen cell and bone marrow preparations. The skin thickening remains constant as late as 76 days post-BMT.

FIGURE 2.

Skin thickening in Scl GVHD is detectable by day 21 post-BMT by routine histology. Bone marrow and spleen cells were prepared according to the methods described by Jaffe and Claman (10 ) and Korngold and Sprent (12 ). Lethally irradiated BALB/c mice were transplanted by tail vein injection with bone marrow and spleen cells from either BALB/c (C) or B10.D2 (E) mice. The mice were then sacrificed at various time points after BMT (see Materials and Methods). Paraffin-embedded sections of day 21 post-BMT skin (A) were stained with hematoxylin and eosin. Arrows indicate the dermal/fat junction. Scale bar = 62 μm. The dermal area for each skin sample was determined from a 10× microscopic image using image analysis (see Materials and Methods) and the percentage of increase in experimental animal back skin thickness (compared with control, panel B) for days 6–76 was calculated as follows: thickness=[(E area − mean C area)/ mean C area], where E = animals with Scl GVHD and C = syngeneic BMT control animals. For each time point post-BMT, the mean and SD of skin area were calculated using a minimum of eight area measurements from at least two skin sections from each of three animals (p = 0.001).

FIGURE 2.

Skin thickening in Scl GVHD is detectable by day 21 post-BMT by routine histology. Bone marrow and spleen cells were prepared according to the methods described by Jaffe and Claman (10 ) and Korngold and Sprent (12 ). Lethally irradiated BALB/c mice were transplanted by tail vein injection with bone marrow and spleen cells from either BALB/c (C) or B10.D2 (E) mice. The mice were then sacrificed at various time points after BMT (see Materials and Methods). Paraffin-embedded sections of day 21 post-BMT skin (A) were stained with hematoxylin and eosin. Arrows indicate the dermal/fat junction. Scale bar = 62 μm. The dermal area for each skin sample was determined from a 10× microscopic image using image analysis (see Materials and Methods) and the percentage of increase in experimental animal back skin thickness (compared with control, panel B) for days 6–76 was calculated as follows: thickness=[(E area − mean C area)/ mean C area], where E = animals with Scl GVHD and C = syngeneic BMT control animals. For each time point post-BMT, the mean and SD of skin area were calculated using a minimum of eight area measurements from at least two skin sections from each of three animals (p = 0.001).

Close modal

Increased TGF-β1 mRNA was observed by semiquantitative RT/PCR analysis of total RNA prepared from whole mouse skin (Fig. 3). Approximately 3-fold more TGF-β1 mRNA is seen in the skin of experimental compared with control animals as early as 6 days post-BMT. On day 21 post-BMT, TGF-β1 is at least 2-fold higher in animals with Scl GVHD than in syngeneic BMT control animals. Each lane represents data from one animal and is representative of four experiments with two or three animals per group for each experiment. Therefore, increased TGF-β1 in skin precedes skin thickening, as predicted by the hypothesis.

FIGURE 3.

Up-regulation of cutaneous TGF-β1 mRNA in Scl GVHD on days 6 and 21 post-BMT. Upper panel, plotted values showing the ratio of TGF-β1 to G3PDH RT-PCR products. Lower panel, scanned agarose gels showing ethidium bromide-stained PCR products. E, animals with Scl GVHD; and C, syngeneic BMT control animals. RT-PCR was performed using RNA isolated from back skin as described in Materials and Methods.

FIGURE 3.

Up-regulation of cutaneous TGF-β1 mRNA in Scl GVHD on days 6 and 21 post-BMT. Upper panel, plotted values showing the ratio of TGF-β1 to G3PDH RT-PCR products. Lower panel, scanned agarose gels showing ethidium bromide-stained PCR products. E, animals with Scl GVHD; and C, syngeneic BMT control animals. RT-PCR was performed using RNA isolated from back skin as described in Materials and Methods.

Close modal

We used an RNase protection assay to quantify collagen mRNA production in the skin of mice with Scl GVHD. Up to 15-fold increased collagen mRNA synthesis is seen in the skin of experimental animals with Scl GVHD (E) compared with control (C) animals on day 38 post-BMT (Fig. 4). The increased message is detectable on day 21 and returns to near control levels by day 75. Therefore, increased collagen mRNA synthesis correlates with increased dermal fibrosis. A digoxigenin-labeled β-actin riboprobe added to each hybridization solution served as an internal standard for the amount of RNA in the hybridization reaction and the amount loaded on the gel. All collagen densities were normalized to the β-actin signal. For the representative day 38 ribonuclease protection assay (RPA) gel lanes shown, the β-actin signals could not be equalized due to the extreme up-regulation of the collagen mRNA signal. The calculation of 15-fold up-regulation takes into account the differing β-actin signals. Similar results were obtained using a 28S ribosomal RNA probe.

FIGURE 4.

Up-regulation of cutaneous mRNA for proα1(I) collagen in animals with Scl GVHD determined by RNase protection assays. RNA was isolated from the back skin of control animals and animals with Scl GVHD at days 7, 14, 21, 38, 42, and 76 post-BMT for use in RNase protection assays. The gel lanes show representative data from a single mouse from each experimental group on day 38 post-BMT. Results are representative of three animals per experimental group for four experiments. Fold increase (summarized in the accompanying table) indicates the ratio of E/C proα1(I) collagen mRNA following normalization based on the β-actin band (where E = animals with Scl GVHD and C = control syngeneic BMT animals).

FIGURE 4.

Up-regulation of cutaneous mRNA for proα1(I) collagen in animals with Scl GVHD determined by RNase protection assays. RNA was isolated from the back skin of control animals and animals with Scl GVHD at days 7, 14, 21, 38, 42, and 76 post-BMT for use in RNase protection assays. The gel lanes show representative data from a single mouse from each experimental group on day 38 post-BMT. Results are representative of three animals per experimental group for four experiments. Fold increase (summarized in the accompanying table) indicates the ratio of E/C proα1(I) collagen mRNA following normalization based on the β-actin band (where E = animals with Scl GVHD and C = control syngeneic BMT animals).

Close modal

Because others have shown that up-regulation of collagen synthesis and skin fibrosis in scleroderma is accompanied by mononuclear cell infiltrates (22), we hypothesized that monocytes infiltrating skin at early time points could be the effector cells driving cutaneous fibrosis in our model. Activated by donor T cells, these monocytes could be the source of increased fibrogenic TGF-β1 in skin (Fig. 1).

We performed immunostaining of frozen sections of skin, and showed prominent infiltration of CD11b+ mononuclear cells (brown-staining) by day 21 in the skin of animals with Scl GVHD compared with syngeneic BMT controls (Fig. 5 A). There are a few infiltrating CD11b+ cells by day 7, they are markedly increased by day 14, and prominent by day 21. Isotype control Ab immunostaining is negative. The examples shown are representative of three experiments. The infiltrates are especially dense in deep dermis, where fibrosis typically occurs first in human scleroderma.

FIGURE 5.

Infiltration of skin of animals with Scl GVHD by CD11b+ mononuclear cells (brown-staining cells) by day 14 after BMT. The upper panel of micrographs (A) shows immunostaining results on acetone-fixed frozen skin sections from animals with Scl GVHD at days 7, 14, and 21 post-BMT. The low panel shows immunostaining results on skin from control animals (all at 40× magnification; scale bar = 43 μm). B, Flow cytometric analysis of single cell preparations from whole back skin demonstrates the influx of cutaneous CD11b+ mononuclear cells in Scl GVHD on day 21 post-BMT. Scatter plots show CD11b+ cells as a percentage of total DCs in each experimental condition. C, control syngeneic BMT animals; and E, animals with Scl GVHD.

FIGURE 5.

Infiltration of skin of animals with Scl GVHD by CD11b+ mononuclear cells (brown-staining cells) by day 14 after BMT. The upper panel of micrographs (A) shows immunostaining results on acetone-fixed frozen skin sections from animals with Scl GVHD at days 7, 14, and 21 post-BMT. The low panel shows immunostaining results on skin from control animals (all at 40× magnification; scale bar = 43 μm). B, Flow cytometric analysis of single cell preparations from whole back skin demonstrates the influx of cutaneous CD11b+ mononuclear cells in Scl GVHD on day 21 post-BMT. Scatter plots show CD11b+ cells as a percentage of total DCs in each experimental condition. C, control syngeneic BMT animals; and E, animals with Scl GVHD.

Close modal

To quantify the cells infiltrating skin in early Scl GVHD, we performed flow cytometry on DC suspensions isolated from the skin of animals with Scl GVHD and syngeneic BMT control animals on day 21 post-BMT (Fig. 5 B). The results correlate with immunostaining, showing increased CD11b+ cells infiltrating the skin of experimental animals with Scl GVHD (E) but not control animals (C) on day 21. The scatter plots shown are representative of three experiments. The CD11b+ cells present were mononuclear cells, not polymorphonuclear neutrophils (PMN), which also stain for CD11b. Histological analysis has also confirmed that there are very few PMN infiltrating the skin of Scl GVHD mice.

The remainder of cells in the dermal suspensions used for flow cytometry studies are T cells, resident dendritic cells, fibroblasts, endothelial cells, follicular and sebaceous epithelial cells. To evaluate the T cell component in Scl GVHD, we stained the DC suspensions with labeled Abs to CD3e. CD3e+ mononuclear cells make up approximately 5–15% of the total cell population in control animals on day 21 and are approximately 1.5- to 2-fold higher in animals with Scl GVHD (data not shown). CD3e, similar to CD11b+, can also be expressed on NK cells. Therefore, the roles of T and NK cells infiltrating the skin of Scl GVHD animals will require further investigation with specific mAbs.

Because TGF-β1, a fibrogenic cytokine (23), appears to play a central role in the development of scleroderma and Scl GVHD, we next asked if Abs to TGF-β could prevent skin fibrosis. Anti-pan-TGF-β Ab (polyclonal rabbit IgG) was administered to animals at days 1 and 6 post-BMT followed by sacrifice on day 21. Day 21 post-BMT was chosen for sacrifice because it is the time point by which all parameters of Scl GVHD are demonstrable, including skin thickening and up-regulated TGF-β1 and collagen mRNA. This dose is also within the range of normal for Ab inhibition of TGF-β in other mouse models (13, 14, 15). The anti-TGF-β1 Ab treatment did not prevent successful BMT, as evidenced by colonization of the spleen in both control and experimental animals, however, the blocking Ab prevented the skin thickening seen in murine Scl GVHD (Fig. 6, A–D). Untreated syngeneic BMT control animal skin is shown in Fig. 6,A. Skin thickening accompanied by prominent mononuclear cell infiltrates in untreated animals with Scl GVHD is shown in Fig. 6,B. Anti-TGF-β Ab treatment prevented skin thickening in these experimental animals (Fig. 6,C and Fig. 7). Anti-TGF-β treatment of syngeneic BMT control animals had no effect on skin thickness (Fig. 6 D). Therefore, the administration of an antagonist to the fibrogenic cytokine TGF-β can prevent the cutaneous fibrosing process in early Scl GVHD, presumably by blocking TGF-β.

FIGURE 6.

Routine histology of skin (A–D) and lung (E–H) on day 21 post-BMT showing inhibition of fibrosis by Abs to TGF-β in animals with Scl GVHD. Anti-TGF-β was administered on days 1 and 6 post-BMT at a concentration of 150 μg/mouse. A and E, control BALB/c animals (transplanted with syngeneic BALB/c bone marrow and spleen cells). B and F, untreated experimental BALB/c animals with Scl GVHD (transplanted with B10.D2 bone marrow and spleen cells). C and G, experimental animals with Scl GVHD that were treated with anti-TGF-β Abs. D and H, control animals that were treated with anti-TGF-β Abs. (Scale bars: skin = 167 μm; lung = 83.5 μm).

FIGURE 6.

Routine histology of skin (A–D) and lung (E–H) on day 21 post-BMT showing inhibition of fibrosis by Abs to TGF-β in animals with Scl GVHD. Anti-TGF-β was administered on days 1 and 6 post-BMT at a concentration of 150 μg/mouse. A and E, control BALB/c animals (transplanted with syngeneic BALB/c bone marrow and spleen cells). B and F, untreated experimental BALB/c animals with Scl GVHD (transplanted with B10.D2 bone marrow and spleen cells). C and G, experimental animals with Scl GVHD that were treated with anti-TGF-β Abs. D and H, control animals that were treated with anti-TGF-β Abs. (Scale bars: skin = 167 μm; lung = 83.5 μm).

Close modal
FIGURE 7.

Inhibition of skin thickening by anti-TGF-β Abs. Bars represent the mean and SD of the skin thickness taken from (n) animals in one representative experiment (of three). (p = 0.001 when using a paired t test to compare Scl GVHD animal (E) back skin thickness to control animal (C) back skin thickness; p = 0.0006 when experimental Scl GVHD animal back skin thickness and the back skin thickness of anti-TGF-β-treated experimental animals are compared). Skin thickness was determined by image analysis of hematoxylin and eosin-stained tissue sections using Optimas software as described in Materials and Methods.

FIGURE 7.

Inhibition of skin thickening by anti-TGF-β Abs. Bars represent the mean and SD of the skin thickness taken from (n) animals in one representative experiment (of three). (p = 0.001 when using a paired t test to compare Scl GVHD animal (E) back skin thickness to control animal (C) back skin thickness; p = 0.0006 when experimental Scl GVHD animal back skin thickness and the back skin thickness of anti-TGF-β-treated experimental animals are compared). Skin thickness was determined by image analysis of hematoxylin and eosin-stained tissue sections using Optimas software as described in Materials and Methods.

Close modal

Because pulmonary fibrosis is a major cause of morbidity and mortality in scleroderma, we next asked if our model would be useful for the study of lung as well as skin fibrosis. Animals with Scl GVHD (Fig. 6,F), but not control animals (Fig. 6,E), show loss of the normal lacy alveolar pattern of lungs by day 21 post-BMT, with the decrease in alveolar space in Scl GVHD animals being approximately 30% (Fig. 8). Lung fibrosis has not been previously reported for this murine model. In the inhibition experiment, Abs to TGF-β prevented lung fibrosis as well (Fig. 6,G, plotted in Fig. 8). Anti-TGF-β treatment had no effect on the percentage of alveolar space in syngeneic BMT control animals (Fig. 6 H). The treatment of Scl GVHD and control animals with nonspecific rabbit IgG also had no effect on skin or lung fibrosis (data not shown).

FIGURE 8.

Inhibition of lung fibrosis by anti-TGF-β Abs. Bars represent the mean and SD of the percentage of decrease in alveolar space of (n) animals with Scl GVHD (untreated or anti-TGF-β treated) as compared with control animals (three animals) from one representative experiment (of three). p < 0.0003 when using a paired t test to compare Scl GVHD animal (E) alveolar space with control (C) animal alveolar space. Paired t test analysis revealed that there was no significant difference between the Scl GVHD animals treated with anti-TGF-β Ab and the syngeneic control BMT animals indicating that the Ab treatment prevented lung fibrosis. The percentage of alveolar space was determined by image analysis of hematoxylin and eosin-stained tissue sections using Optimas software as described in Materials and Methods.

FIGURE 8.

Inhibition of lung fibrosis by anti-TGF-β Abs. Bars represent the mean and SD of the percentage of decrease in alveolar space of (n) animals with Scl GVHD (untreated or anti-TGF-β treated) as compared with control animals (three animals) from one representative experiment (of three). p < 0.0003 when using a paired t test to compare Scl GVHD animal (E) alveolar space with control (C) animal alveolar space. Paired t test analysis revealed that there was no significant difference between the Scl GVHD animals treated with anti-TGF-β Ab and the syngeneic control BMT animals indicating that the Ab treatment prevented lung fibrosis. The percentage of alveolar space was determined by image analysis of hematoxylin and eosin-stained tissue sections using Optimas software as described in Materials and Methods.

Close modal

We have demonstrated that the Scl GVHD model faithfully reproduces the skin fibrosis as well as the up-regulation of TGF-β1 and collagen mRNA synthesis that occurs in human scleroderma. Furthermore, we have established the following temporal sequence of these events: TGF-β1 mRNA up-regulation and CD11b+ mononuclear cell infiltration are seen by day 7 post-BMT and precede detectable increases in collagen mRNA and skin thickening (summarized in Table I, a composite of data from several experiments).

Table I.

Time course of events in Scl GVHDa

AnalysisDays After Bone Marrow Transplantation
714213875
CD11b+ cells in skin ++ +++ ++ ++ 
TGF-β1 mRNA ++ ND − − 
Collagen mRNA − ND ++ +++ 
Dermal fibrosis − +/− ++ ++ ++ 
AnalysisDays After Bone Marrow Transplantation
714213875
CD11b+ cells in skin ++ +++ ++ ++ 
TGF-β1 mRNA ++ ND − − 
Collagen mRNA − ND ++ +++ 
Dermal fibrosis − +/− ++ ++ ++ 
a

−, Low or absent; +, increased; +++, markedly increased.

Scl GVHD was first presented as a model for scleroderma by Jaffe and Claman (10), Claman et al., (11), and Korngold and Sprent (12), but until now, the cellular changes and the molecular events occurring in the model have not been thoroughly characterized. We have established a time course for skin fibrosis in this murine Scl GVHD model and characterized molecular events that make this model extremely useful for testing therapies for autoimmune fibrosing disorders such as scleroderma and GVHD. Until the present, our understanding of the pathophysiology of scleroderma and consequently our ability to treat this chronic progressive disease have been limited by the lack of a suitable animal model. Scleroderma-like disease occurs in the University of California at Davis autoimmune chicken (reviewed by Van De Water et al., Ref. 3), and the existing murine models (Tight Skin Mouse, Tsk; and Tsk2 mice) show some, but not all features of human scleroderma (3, 24, 25, 26). Tsk animals have mutations in the regulation of DNA-binding proteins for collagen and matrix protein expression (27, 28, 29), and do not exhibit the vascular or pulmonary changes of human scleroderma, although they do exhibit skin thickening (3) and autoantibodies (30). The lung pathology is described as emphysematous, rather than fibrotic. Tsk2 animals display cutaneous fibrosis and a mononuclear cell infiltration in skin; however, TGF-β expression is not related to fibrosis as it is in scleroderma (26, 31). These models are useful for studying fibrosis, but their value for analyzing the complex immunologic dysregulation in scleroderma is limited.

In contrast, the murine Scl GVHD model (B10.D2 > BALB/c) that we have characterized in the present study faithfully reproduces the most important features of human scleroderma, including skin thickening, mononuclear cell infiltrates, lung fibrosis, and up-regulation of cutaneous collagen and TGF-β1 mRNA. Furthermore, the similarities between scleroderma and Scl GVHD, as well as the recent implication of microchimerism as a trigger for scleroderma, make the murine Scl GVHD model particularly appropriate for study. This model provides a system that can be manipulated to determine important variables in disease progression, and will allow us to test newly emerging therapeutic modalities at all stages of the disease, including the early stages when they may be more effective.

mRNA for the cytokine TGF-β1 is clearly up-regulated at early time points in animals with Scl GVHD. TGF-β1, a potent stimulus for increased collagen synthesis, is thought to be critical in the cutaneous and pulmonary fibrosis of scleroderma (32, 33), as well as in other models of fibrosis (34, 35, 36, 37, 38). TGF-β is made by multiple cell types, including monocytes, fibroblasts, and endothelial cells (39, 40), and can be converted to the active form by monocyte/macrophages (41). The three isoforms of human TGF-β (TGF-β1, TGF-β2, and TGF-β3) can be readily distinguished (42). These isoforms have pleuripotential effects on not only extracellular matrix homeostasis (32), but also on immune regulation (40, 43) and on epithelial growth (44). In human scleroderma, TGF-β1 is thought to be the critical isoform implicated in cutaneous and pulmonary fibrosis (45, 46, 47).

We have shown that anti-TGF-β Abs prevent the progression of skin and lung fibrosis in Scl GVHD at day 21, presumably by blocking TGF-β; however, we have not tested the effects of the anti-TGF-β Ab on other cytokines and the effect may be an indirect one. Because our Ab is polyclonal and inhibits all three isoforms of TGF-β, we cannot make any conclusions about the role of the individual isoforms in murine Scl GVHD. The striking results of the Ab inhibition studies suggest the need for further study of these TGF-β isoforms. Furthermore, anti-TGF-β therapy with a humanized mAb may be useful in preventing disease progression in human Scl GVHD and scleroderma. In addition to anti-TGF-β, experiments with Abs to other cytokines or to macrophage and T cell surface markers will be useful to further characterize the disease process in Scl GVHD and to identify other candidate molecules for immunotherapy.

Asai et al. (48) describe successful prevention of cytotoxic GVHD due to transplantation across major histocompatibility loci using blocking Abs to TGF-β. The paradoxical results (TGF-β produced by transplanted donor NK cells protected against GVHD in this model) point out the importance of the genetic backgrounds of donor and recipient individuals in such transplantations, and the diversity of GVHD-like reactions. Several quite different models of murine GVHD exist (reviewed in Refs. 49 and 50). They include transplantation from parental to nonirradiated F1 hybrid offspring (P > F1), in which autoimmune features like those in lupus erythematosus are commonly seen; transplantation across major histocompatibility loci, which is often rapidly lethal; and transplantation across minor histocompatibility loci, which most closely approximates in severity and course allogeneic sibling bone marrow transplantation in humans. In this last type of GVHD, somewhat different forms of disease can be produced experimentally in mice in carefully controlled depletion experiments by selecting for different cell types including the following: cytotoxic CD8 T cells, CD4 T cells, and NK cells (49, 51, 52, 53, 54). Therefore, the identity of effector cells may vary with the type of GVHD. Murine Scl GVHD is a subset of the last type of GVHD, exhibited by only a few transplantation pairs. It is not surprising, therefore, that different effector cells (presumably monocytes synthesizing TGF-β1, rather than cytotoxic T cells attacking epithelium) operate in this distinctive fibrosing form of GVHD. The activation of host or donor monocytes by donor T cells is incompletely understood in this model, and will be an important parameter for examination in future experiments. Recently, Schlomchik et al. (55) determined that host APCs were required for the initiation of GVHD in a murine MHC-mismatch BMT model. Our model is also ideal for testing the involvement of host and donor APCs and T cells and the effects of chimerism in the development of Scl GVHD in an MHC-compatible BMT. Analysis of the diversity of murine GVHD-like reactions may be helpful in understanding the variability in type, development, and progression of human GVHD, and in understanding autoimmune disease itself.

In summary, we have shown that murine Scl GVHD faithfully reproduces the skin and lung fibrosis and up-regulation of cutaneous collagen and TGF-β1 mRNA that occurs in human scleroderma. Cutaneous CD11b+ mononuclear cell infiltrates and increased TGF-β1 and collagen mRNA precede dermal fibrosis and thickening, and the progression of early skin and lung disease can be inhibited with Abs to TGF-β. Analysis of the diversity of murine GVHD-like reactions may be helpful in understanding the variability in type, development, and progression of human GVHD, and in understanding a complex autoimmune disease such as scleroderma. Most importantly, murine Scl GVHD is a useful model for testing potential interventions for scleroderma and GVHD.

Note.

After submission of our manuscript it was reported that Abs to TGF-β (the same R&D Ab (R&D Systems, Minneapolis, MN) used in our experiment) could reduce cutaneous sclerosis in a mouse model of bleomycin-induced scleroderma (56).

We thank K. D. Cooper and S. L. Gerson for helpful discussions and the Case Western Reserve University Department of Dermatology Immunobiology Group (K. Kang, T. S. McCormick, S. Stevens, I. Kremer, and Y. Yoshida) for critical reading of the manuscript. We thank Virginia Erbahr for secretarial assistance.

1

This work was supported by a fellowship from the RGK Foundation (L. L. M.), and grants from the Northeastern Ohio Multipurpose Arthritis Center (AR-20618019D), a Dermatology Foundation Novartis Career Development Award, the Case Western Reserve University Skin Disease Research Center (AR-39750-10), and a National Institutes of Health Shannon Award (AR-45033-01) (A. C. G).

3

Abbreviations used in this paper: GVHD, graft-vs-host disease; Scl GVHD, sclerodermatous GVHD; BMT, bone marrow transplantation; DC, dermal cells; PMN, polymorphonuclear neutrophils; E, experimental animals; C, control animals.

1
Mitchell, H., M. B. Bolster, E. C. LeRoy.
1997
. Scleroderma and related conditions.
Adv. Rheumatol.
81
:
129
2
Rose, N. R., N. Leskovsek.
1998
. Scleroderma: immunopathogenesis and treatment.
Immunol. Today
19
:
499
3
Van De Water, J., S. A. Jimenez, M. E. Gershwin.
1995
. Animal models of scleroderma: contrasts and comparisons.
Int. Rev. Immunol.
12
:
201
4
Nelson, J. L., D. E. Furst, S. Maloney, T. Gooley, P. C. Evans, A. Smith, M. A. Bean, C. Ober, D. W. Bianchi.
1997
. Microchimerism and HLA-compatible relationships of pregnancy in scleroderma.
Lancet
351
:
559
5
Arlett, C. M., J. B. Smith, S. A. Jimenez.
1998
. Identification of fetal DNA and cells in skin lesions from women with systemic sclerosis.
N. Engl. J. Med.
338
:
1186
6
Varga, J., L. Rudnicka, J. Uitto.
1994
. Connective tissue alterations in systemic sclerosis.
Clin. Dermatol.
12
:
387
7
Varga, J., S. A. Jimenez.
1995
. Modulation of collagen gene expression: its relation to fibrosis in systemic sclerosis and other disorders.
Ann. Intern. Med.
122
:
60
8
Peltonen, J., L. Kahari, S. Jaakkola, V.-M. Kahari, J. Varga, J. Uitto, S. A. Jimenez.
1990
. Evaluation of transforming growth factor β and type I procollagen gene expression in fibrotic skin diseases by in situ hybridization.
J. Invest. Dermatol.
94
:
365
9
Broekelmann, T. J., A. H. Limper, T. V. Colby, J. A. McDonald.
1991
. Transforming growth factor β1 is present at sites of extracellular matrix gene expression in human pulmonary fibrosis.
Proc. Natl. Acad. Sci. USA
88
:
6642
10
Jaffee, B. D., H. N. Claman.
1983
. Chronic graft-versus-host disease (GVHD) as a model for scleroderma.
Cell. Immunol.
77
:
1
11
Claman, H. N., B. D. Jaffee, J. C. Huff, R. A. F. Clark.
1985
. Chronic graft-versus-host disease as a model for scleroderma.
Cell. Immunol.
94
:
73
12
Korngold, R., J. Sprent.
1978
. Lethal graft-versus-host disease after bone marrow transplantation across minor histocompatibility barriers in mice.
J. Exp. Med.
148
:
1687
13
Delgado-Rizo, V. A., A. Panduro Salazar, J. Armendariz-Borunda.
1998
. Treatment with anti-transforming growth factor β antibodies influences an altered pattern of cytokine gene expression in injured rat liver.
Biochim. Biophys. Acta
1442
:
20
14
Arteaga, C. L., S. D. Hurd, A. R. Winnier, M. D. Johnson, B. M. Gendly, J. T. Forbes.
1993
. Anti-transforming growth factor (TGF)-β antibodies inhibit breast cancer cell tumoriginicity and increase mouse spleen natural killer cell activity.
J. Clin. Invest.
92
:
2569
15
Wahl, S. M., J. B. Allen, G. L. Costa, H. L. Wong, J. R. Dasch.
1993
. Reversal of acute and chronic synovial inflammation by anti-transforming growth factor β.
J. Exp. Med.
177
:
225
16
Gilliam, A. C., I. B. Kremer, Y. Yoshida, S. R. Stevens, E. Tootell, M. B. M. Teunissen, C. Hammerberg, K. D. Cooper.
1998
. The human hair follicle: a reservoir of CD40+ B7-deficient Langerhans cells that repopulate epidermis after UVB exposure.
J. Invest. Dermatol.
110
:
422
17
Sambrook, J., E. F. Fritsch, T. Maniatis.
1989
.
Molecular Cloning
2nd Ed.
719
-722. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
18
Kang, K., C. Hammerberg, L. Meunier, K. D. Cooper.
1994
. CD11b+ macrophages that infiltrate human epidermis after in vivo ultraviolet exposure potentially produce IL-10 and represent the major secretory source of epidermal IL-10 protein.
J. Immunol.
153
:
5256
19
Metsaranta, M., D. Toman, B. de Crombrugghe, E. Vuorio.
1991
. Specific hybridization probes for mouse type I, II, III and IX collagen mRNAs.
Biochim. Biophys. Acta
1089
:
241
20
Hammerberg, C., N. Duraiswamy, K. D. Cooper.
1996
. Reversal of immunosuppression inducible through ultraviolet-exposed skin by in vivo anti-CD11b treatment.
J. Immunol.
157
:
5254
21
Gilliam, A. C., G. F. Murphy.
1996
. Cellular pathology of cutaneous graft-versus-host disease. J. L. M. Ferrara, and H. J. Deeg, and S. J. Burakoff, eds.
Graft-vs.-Host Disease
291
-336. Marcel Dekker, New York.
22
Kraling, B. M., G. G. Maul, S. A. Jimenez.
1995
. Mononuclear cellular infiltrates in clinically involved skin from patients with systemic sclerosis of recent onset predominantly consist of monocytes/macrophages.
Pathobiology
63
:
48
23
Roberts, A. B., M. B. Sporn, R. K. Assoian, J. M. Smith, N. S. Roche, L. M. Wakefield, U. I. Heine, L. A. Liotta, V. Falanga, J. H. Kehrl, A. S. Fauci.
1986
. Transforming growth factor type β: rapid induction of fibrosis and angiogenesis in vivo and stimulation of collagen formation in vitro.
Proc. Natl. Acad. Sci. USA
83
:
4167
24
Christner, P. J., J. Peters, D. Hawkins, L. D. Siracusa, S. A. Jimenez.
1995
. The tight skin 2 mouse: an animal model of scleroderma displaying cutaneous fibrosis and mononuclear cell infiltration.
Arthritis Rheum.
38
:
1791
25
Christner, P. J., L. D. Siracusa, D. F. Hawkins, R. McGrath, J. K. Betz, S. T. Ball, S. A. Jimenez, J. Peters.
1996
. A high-resolution linkage map of the tight skin 2 (Tsk2) locus: a mouse model for scleroderma (SSc) and other cutaneous fibrotic diseases.
Mamm. Genome
7
:
610
26
Christner, P. J., E. G. Hitraya, J. Peters, R. McGrath, S. A. Jimenez.
1998
. Transcriptional activation of the α1(I) procollagen gene and up-regulation of α1(I) and α1(III) procollagen messenger RNA in dermal fibroblasts from tight skin 2 mice.
Arthritis Rheum.
41
:
2132
27
Philips, N., R. I. Bashey, S. A. Jimenez.
1995
. Increased α1(I) procollagen gene expression in tight skin (TSK) mice myocardial fibroblasts is due to a reduced interaction of a negative regulatory sequence with AP-1 transcription factor.
J. Biol. Chem.
270
:
9313
28
Siracusa, L. D., R. McGrath, Q. Ma, J. J. Moskow, J. Manne, P. J. Christner, A. M. Buchberg, S. A. Jimenez.
1996
. A tandem duplication within the fibrillin I gene is associated with the mouse tight skin mutation.
Genome Res.
6
:
300
29
Kielty, C. M., M. Raghanath, L. D. Siracusa, M. J. Sherratt, R. Peters, C. A. Shuttleworth, S. A. Jimenez.
1998
. The tight skin mouse: demonstration of mutant fibrillin-1 production and assembly into abnormal microfibrils.
J. Cell Biol.
140
:
1159
30
Kasturi, K., T. Muryoi, S. Shibata, A. Hatakeyama, C. Murai, Y. Simakoshi, C. Bona.
1997
. Functional properties and molecular characteristics of autoantibodies associated with tight skin syndrome.
Ann. NY Acad. Sci.
815
:
253
31
Pablos, J. L., E. T. Everett, R. Harley, E. C. LeRoy, J. S. Norris.
1995
. Transforming growth factor-β1 and collagen gene expression during postnatal skin development and fibrosis in the tight-skin mouse.
Lab. Invest.
72
:
670
32
Kikuchi, K., C. W. Hartl, E. A. Smith, E. C. LeRoy, M. Trojanowska.
1992
. Direct demonstration of transcriptional activation of collagen gene expression in systemic sclerosis fibroblasts: insensitivity to TGF-β1 stimulation.
Biochem. Biophys. Res. Commun.
187
:
45
33
Border, W. A., N. A. Noble.
1994
. Transforming growth factor β in tissue fibrosis.
N. Engl. J. Med.
331
:
1286
34
Sime, P. J., Z. Xing, F. L. Graham, K. G. Csaky, J. Gauldie.
1997
. Adenovector-mediated gene transfer of active transforming growth factor-β1 induces prolonged severe fibrosis in rat lung.
J. Clin. Invest.
100
:
768
35
Sanderson, N., V. Factor, P. Nagy, J. Kopp, P. Kondaiah, L. Wakefield, A. B. Roberts, M. B. Sporn, S. S. Thorgeirsson.
1995
. Hepatic expression of mature transforming growth factor β1 in transgenic mice results in multiple tissue lesions.
Proc. Natl. Acad. Sci. USA
92
:
2572
36
Kopp, J. B., V. M. Factor, M. Mozes, P. Nagy, N. Sanderson, E. P. Bottinger.
1996
. Transgenic mice with increased plasma levels of TGF-β1 develop progressive renal disease.
Lab. Invest.
74
:
991
37
Bottinger, E. P., J. J. Letterio, A. B. Roberts.
1997
. Biology of TGF-β in knockout and transgenic mouse models.
Kidney Int.
51
:
1355
38
Anscher, M. S., W. P. Peters, H. Reisenbichler, W. P. Petros, R. L. Jirtle.
1993
. Transforming growth factor β as a predictor of liver and lung fibrosis after autologous bone marrow transplantation for advanced breast cancer.
N. Engl. J. Med.
328
:
1592
39
Kingsley, D. M..
1994
. The TGF-β superfamily: new members, new receptors, and new genetic tests of function in different organisms.
Genes Dev.
8
:
133
40
Roberts, A. B..
1998
. Molecular and cell biology of TGF-β.
Miner. Electrolyte Metab.
24
:
111
41
Massague, J., S. Cheifetz, M. Laiho, D. A. Ralph, F. M. B. Weis, A. Zentella.
1992
. Transforming growth factor-β.
Cancer Surv.
12
:
81
42
Roberts, A. B., S.-J. Kim, T. Noma, A. B. Glick, R. Lafyatis, R. Lechleider, S. B. Jakowlew, A. Geiser, M. A. O’Reilly, D. Danielpour, M. B. Sporn.
1991
. Multiple forms of TGF-β: distinct promoters and differential expression.
Ciba Found. Symp.
157
:
7
43
Letterio, J. J., A. B. Roberts.
1997
. TGF-β: a critical modulator of immune cell function.
Clin. Immunol. Immunopathol.
84
:
244
44
Moses, H. L..
1992
. TGF-β regulation of epithelial cell proliferation.
Mol. Reprod. Dev.
32
:
179
45
Ludwicka, A., T. Ohba, M. Trojanowska, A. Yamakage, C. Strange, E. A. Smith, E. C. LeRoy, S. Sutherland, R. M. Silver.
1995
. Elevated levels of platelet derived growth factor and transforming growth factor-β1 in bronchoalveolar lavage fluid from patients with scleroderma.
J. Rheumatol
22
:
1876
46
Higley, H., K. Persichitte, S. Chu, W. Waegell, R. Vancheeswaran, C. Black.
1994
. Immunocytochemical localization and serologic detection of transforming growth factor β1.
Arthritis Rheum.
37
:
278
47
Corrin, B., D. Butcher, B. J. McAnulty, R. M. Dubois, C. M. Black, G. J. Laurent, N. K. Harrison.
1994
. Immunohistochemical localization of transforming growth factor-β1 in the lungs of patients with systemic sclerosis, cryptogenic fibrosing alveolitis and other lung disorders.
Histopathology
24
:
145
48
Asai, O., D. L. Longo, Z. G. Tian, R. L. Hornung, D. D. Taub, F. W. Ruscetti, W. J. Murphy.
1998
. Suppression of graft-versus-host disease and amplification of graft-versus-tumor effects by activated natural killer cells after allogeneic bone marrow transplantation.
J. Clin. Invest.
101
:
1835
49
Korngold, R..
1993
. Biology of graft-vs-host disease.
Am. J. Ped. Hematol. Oncol.
15
:
18
50
Korngold, R., J. Sprent.
1991
. Graft-versus-host disease in experimental allogeneic bone marrow transplantation.
Proc. Soc. Exp. Biol. Med.
197
:
12
51
Murphy, G. F., D. Whitaker, J. Sprent, R. Korngold.
1991
. Characterization of target injury of murine acute graft-vs-host disease directed to multiple minor histocompatibility antigens elicited by either CD4+ or CD8+ effector cells.
Am. J. Pathol.
138
:
983
52
Wettstein, P. J., R. Korngold.
1992
. T cell subsets required for in vivo and in vitro responses to single and multiple minor histocompatibility antigens.
Transplantation
54
:
296
53
Korngold, R..
1992
. Lethal graft-versus-host disease in mice directed to multiple minor histocompatibility antigens: features of CD8+ and CD4+ T cell responses.
Bone Marrow Transplant.
9
:
355
54
Zeng, D., D. Lewis, S. Dejbakhsh-Jones, F. Lan, M. Garcia-Ojeda, R. Sibley, S. Strober.
1999
. Bone marrow NK1.1- and NK1.1+ T cells reciprocally regulate acute graft versus host disease.
J. Exp. Med.
189
:
1073
55
Shlomchik, W. D., M. S. Couzens, C. B. Tang, J. McNiff, M. E. Robert, J. Liu, M. J. Shlomchik, S. G. Emerson.
1999
. Prevention of graft versus host disease by inactivation of host antigen-presenting cells.
Science
285
:
412
56
Yamamoto, T., S. Takagawa, I. Katayama, K. Nishioka.
1999
. Anti-sclerotic effect of transforming growth factor-β antibody in a mouse model of bleomycin-induced scleroderma.
Clin. Immunol.
92
:
6