mPGES-1 heterozygous (Het) male and female mice on a DBA1 lac/J background were provided by Pfizer Inc (Groton, CT, USA) 13. mPGES-1 Het mice were mated to generate mPGES-1 null, Het, and littermate wild-type (WT) mice. All of the experiments were performed under the guidelines of the Institutional Animal Care and Use Committee. Genotypes were identified by polymerase chain reaction (PCR) of tail biopsy DNA extract by using two-primer sets for the mPGES-1 null allele (PGES-N257R, 5'-TGCTACTTCCATTTGTCACGTC-3' and PGES-4407R, 5'-TCCAAGTACTGAGCCAGCTG-3') and the WT allele (PGES-WT-F, 5'-TCCCAGGTGTTGGGATTTAGAC-3' and PGES-WT-R, 5'-TAGGTGGCTGTACTGT TTGTTGC-3') (Invitrogen Corporation, Carlsbad, CA, USA). After initial denaturation at 95°C for 15 minutes, PCR involved 40 cycles of 30 seconds at 95°C, 30 seconds at 56°C, and 45 seconds at 72°C, followed by elongation for 5 minutes at 72°C. DNA from mPGES-1 WT mice showed one band (412 base pairs [bp]), DNA from mPGES-1 null mice showed one band (720 bp), and DNA from mPGES-1 Het mice showed bands of both 412 and 720 bp 22.

Bleomycin treatment was performed as previously reported 2324. Briefly, bleomycin (Sigma-Aldrich, St. Louis, MO, USA), diluted to 0.1 U/mL with phosphate-buffered saline (PBS), was sterilized with filtration. One hundred microliters of bleomycin or PBS was injected subcutaneously into a single location on the shaved back of mPGES-1 WT and null mice once daily for 4 weeks. Mice were killed by CO₂ euthanasia after 4 weeks, and skin samples were collected for histology, immunohistochemistry, hydroxyproline assay, and Western blotting.

Sections (0.5 μm) were cut with a microtome (Leica Microsystems, Wetzlar, Germany) and collected on Superfrost Plus slides (Fisher Scientific, Pittsburgh, PA, USA). Sections were then de-waxed in xylene and rehydrated by successive immersion in descending concentrations of alcohol. To assess the effects of mPGES-1 genetic deletion on collagen synthesis, trichrome collagen stain was employed as previously described 2324. Briefly, collagen content in each section was assessed by three blinded observers who used the following assessment criteria: 0 signifies no collagen fibres, 1 signifies few collagen fibres, 2 signifies a moderate amount of collagen fibres, and 3 signifies an excessive amount of collagen fibres. In addition, Northern Eclipse (Empix Imaging, Inc., Mississauga, ON, Canada) software was used to determine the dermal thickness in each stained section to account for changes in dermal thickness in WT and mPGES-1 null mice with or without bleomycin injection.

To assess inflammation, the presence of macrophages in skin sections was detected by immunofluorescence with MOMA-2 (monocyte + macrophage marker) antibody (Abcam, Cambridge, UK), a marker for macrophage. Immunofluorescence was performed as previously described 25, and the number of macrophages was then counted. In addition, sections were stained with hematoxylin and eosin (H&E) (Fisher Scientific, Ottawa, ON, Canada). H&E staining was performed in accordance with the recommendations of the manufacturer. The effects of mPGES-1 genetic deletion on inflammation were graded on a scale of 0 to 3 by three separate blinded observers: 0 signifies no inflammatory cells, 1 signifies few inflammatory cells, 2 signifies moderate inflammatory cells, and 3 signifies extensive inflammatory cells.

Sections were cut and processed as described above. Immunolabeling of alpha-smooth muscle actin (α-SMA) was performed with the DakoCytomation LSAB+ System-HRP kit (DakoCytomation, Carpinteria, CA, USA). Immunohistochemical procedures were performed in accordance with the recommendations of the manufacturer. Briefly, endogenous peroxide was blocked by using 0.5% H₂O₂ in methanol for 5 minutes. Non-specific IgG binding was blocked by incubating sections with bovine serum albumin (0.1%) in PBS for 1 hour and then was incubated with primary antibody for α-SMA (1:1,000) in a humidified chamber and left overnight at 4°C. Next, sections were incubated with a biotinylated link for 30 minutes followed by incubation with streptavidin for 30 minutes. The chromogen diaminobenzidine tetrahydrochloride (DAB) was then added until sufficient color developed, and sections were counterstained with Harris's hematoxylin.

Hydroxyproline assay was performed as a marker of collagen content in bleomycin-treated/untreated skin with the method previously described 26. Skin tissues were homogenized in saline and hydrolyzed with 2N NaOH for 30 minutes at 120°C, and then we determined hydroxyproline content by modifying the Neumann and Logan's reaction with Chloramine T and Ehrlich's reagent with a hydroxyproline standard curve measuring at 550 nm. Values were expressed as micrograms of hydroxyproline per milligrams of protein.

Dermal mouse fibroblasts were isolated from explants (4- to 6-week-old WT and mPGES-1 null mice) as described 27. Also, dermal fibroblasts were isolated from an explant culture of 4-mm punch biopsies from the forearm of healthy individuals and those with early-onset (between 3 and 18 months after initial diagnosis) diffuse cutaneous scleroderma (6 each) in Dulbecco's modified Eagle's medium and 10% fetal bovine serum (Invitrogen Corporation) as previously described 28. Donors were age-, site-, and sex-matched. No patients were on immunosuppressants. Experimental protocols were approved by the ethics committee of the Royal Free Hospital (UK), where all participants were recruited under informed written consent and human experimentation was conducted. Cells were subjected to indirect immunofluorescence analysis, as previously described 29, by using anti-mPGES-1 antibody (Cayman Chemical Company, Ann Arbor, MI, USA) followed by an appropriate secondary antibody (Jackson ImmunoResearch Laboratories Inc., West Grove, PA, USA) and were photographed with a Zeiss Axiphot camera (Empix Imaging, Inc.). Alternatively, cells were lysed in 2% SDS, and proteins were quantified (Pierce, Rockford, IL, USA) and subjected to Western blot analysis as previously described 30. The following primary antibodies were used for Western blotting: anti-mPGES-1 (Cayman Chemical Company, Charlotte, NC, USA), anti-α-SMA (Sigma-Aldrich), and anti-β-actin (Sigma-Aldrich).

Statistical analysis was performed with a two-tailed analysis-of-variance test in conjunction with a post hoc Mann-Whitney U test. Results are expressed as the mean ± standard error. A P value of less than 0.05 was considered statistically significant (denoted by an asterisk).

To begin to assess whether mPGES-1 plays a role in fibrogenesis in SSc, we first examined whether mPGES-1 protein showed an altered expression pattern in dermal fibroblasts isolated from fibrotic lesions of early-onset diffuse SSc patients compared with those isolated from identical areas of healthy skin (termed normal fibroblast, or NF) (Figure 1a). Our results clearly showed that mPGES-1 protein was significantly upregulated in fibrotic fibroblasts from the skin of SSc patients compared with NFs isolated from healthy skin (Figure 1b). To continue our studies, we then evaluated whether mPGES-1 was induced in vivo in response to bleomycin-induced skin sclerosis. To do this, we injected WT mice subcutaneously for 4 weeks with bleomycin or PBS and skin biopsies were isolated 4 weeks post bleomycin or PBS treatment. From these, protein extracts were prepared and subjected to Western blotting with anti-mPGES-1 antibody (Figure 1c). Results showed that mPGES-1 was significantly induced in the skin in response to bleomycin as compared with PBS. Collectively, these results revealed that mPGES-1 is induced during fibrosis and may play a role in fibrogenesis.

After having demonstrated that mPGES-1 is overexpressed in fibrosis, we sought to assess whether mPGES-1 is required for fibrogenesis. Accordingly, we subjected WT and mPGES-1 null mice to the bleomycin model of skin scleroderma. Mice harboring a deletion of the mPGES-1 gene were detected by PCR analysis of tail DNA as previously described 223031 and by subjecting dermal fibroblasts cultured from skin explants derived from WT and mPGES-1 null mice to Western blot and immunofluorescence analyses using an anti-mPGES-1 antibody (Figure 2a). Since mPGES-1 mediates inflammation in vitro as well as in vivo 223031 and inflammation is involved with the onset of fibrogenesis 3432, we employed indirect immunofluorescence analysis with an anti-MOMA-2 antibody (a marker of macrophages) to examine the effect of loss of mPGES-1 on the ability of bleomycin to induce the appearance of macrophages. As anticipated, we observed a marked increase in the number of macrophages in WT mice exposed to bleomycin compared with WT mice exposed to PBS (Figure 3a). However, compared with WT control mice, mPGES-1 null mice possessed markedly reduced numbers of macrophages in response to bleomycin (Figure 3a). Furthermore, semiquantitative blinded histological analysis of H&E-stained sections showed that bleomycin exposure resulted in a significantly lower inflammation score in mPGES-1 null mice compared with their WT counterparts (Figure 3c). Thus, loss of mPGES-1 resulted in a resistance to bleomycin-induced inflammation.

To probe whether, in mPGES-1 null mice, reduced bleomycin-induced inflammation corresponded with reduced fibrosis, we then investigated whether loss of mPGES-1 resulted in a resistance to bleomycin-induced matrix deposition. To perform this analysis, we subjected bleomcyin-exposed skin of WT and mPGES-1 null mice to histological and biochemical analyses. As anticipated, as visualized by H&E and trichrome staining and hydroxyproline/praline analyses, bleomycin treatment in WT mice resulted in significant increases in extracellular matrix (ECM) deposition, dermal thickness, collagen score, and collagen content (Figure 4a-c and 5a). However, mPGES-1 null mice were relatively resistant to bleomycin-induced dermal thickness, ECM deposition, collagen score, and collagen content (Figure 4a-c and 5a). We did not observe any significant difference in ECM deposition between WT and mPGES-1 null mice in response to PBS. Thus, mirroring the effect observed on bleomycin-induced inflammation, loss of mPGES-1 resulted in a resistance to bleomycin-induced ECM deposition.

As α-SMA-expressing myofibroblasts are a hallmark of both SSc and bleomycin-induced skin fibrosis 2324, we then continued our studies by determining the effect of loss of mPGES-1 on the induction of α-SMA-expressing myofibroblasts in response to bleomycin injection. To begin to perform these analyses, we first subjected skin sections of bleomycin- or PBS-exposed WT or mPGES-1 null mice to immunohistochemical analysis with an anti-α-SMA antibody. Compared with skin of WT mice injected with PBS, skin of WT mice injected with bleomycin possessed markedly elevated numbers of myofibroblasts (Figure 5b, c), and this is consistent with previously published data 2324. Conversely, mPGES-1 null mice were relatively resistant to the ability of bleomycin to induce α-SMA-expressing myofibroblasts (Figure 5b, c). Confirming these data, Western blot analysis on protein samples derived from WT and mPGES-1 null mice treated with PBS or bleomycin showed that bleomycin resulted in elevated α-SMA protein production in WT mice but not in mPGES-1 null mice (Figure 5d). Collectively, our data are consistent with the notion that loss of mPGES-1 expression confers resistance to bleomycin-induced skin fibrosis and that mPGES-1 may play a key role in inflammation-induced fibrogenesis.