Advertisement

Cartilage Destruction in Granulomatosis with Polyangiitis (Wegener's Granulomatosis) Is Mediated by Human Fibroblasts after Transplantation into Immunodeficient Mice

      A key feature of granulomatosis with polyangiitis (GPA; or Wegener's granulomatosis) is the granulomatous inflammation of the upper respiratory tract, which leads to the subsequent destruction of adjacent tissues. The aim of our work was to study the histopathological and cellular components of tissue destruction of human GPA tissue transplanted into immunodeficient mice. Biopsy specimens from patients with active GPA (n = 10) or sinusitis (controls, n = 6) were s.c. co-implanted with healthy allogeneic human nasal cartilage into immunodeficient pfp/rag2−/− mice. Transplants were examined for their destructive capability of the allografted human cartilage. In addition, nasal fibroblasts from patients with GPA (n = 8) and control healthy nasal fibroblasts (n = 5) were cultured, and cell proliferation and apoptosis were quantified. mRNA and protein levels of matrix metalloproteinases and cytokines were evaluated at baseline and after proinflammatory stimulation. GPA implants showed massive destruction of the co-implanted human cartilage, whereas cartilage destruction was only marginal in control samples. Destruction was mediated by human fibroblasts and could be inhibited by corticoid treatment. The up-regulated production of matrix metalloproteinases 1, 3, and 13 and cytokines IL-6 and IL-8 was found in vivo and in vitro. Although proliferation of isolated fibroblasts was comparable between GPA and controls, GPA samples showed a significant delay of apoptosis. The destruction of nasal cartilage in GPA is mainly mediated by fibroblasts that can be blocked by corticosteroids, and this tissue destruction is not dependent on the influx of leukocytes.
      Granulomatosis with polyangiitis (GPA; or Wegener's granulomatosis) is an autoimmune disease characterized by necrotizing granulomatous inflammation and vasculitis, associated with proteinase 3–directed anti-neutrophil cytoplasmic autoantibodies.
      • Bacon P.A.
      The spectrum of Wegener's granulomatosis and disease relapse.
      • Mueller A.
      • Holl-Ulrich K.
      • Lamprecht P.
      • Gross W.L.
      Germinal centre-like structures in Wegener's granuloma: the morphological basis for autoimmunity.
      One of the main features of GPA is the destructive necrotizing granulomatous inflammation of the upper respiratory tract. In contrast to vasculitis, this disease-specific histopathological feature separates GPA from other vasculitides and remains enigmatic in its pathophysiological features.
      • Bacon P.A.
      The spectrum of Wegener's granulomatosis and disease relapse.
      • Kradin R.L.
      • Mark E.J.
      Case records of the Massachusetts General Hospital: weekly clinicopathological exercises: case 18–2002: a 48-year-old man with a cough and bloody sputum.
      • Lamprecht P.
      • Gross W.L.
      Antineutrophil cytoplasmic antibody-associated vasculitis: autoinflammation, autodestruction and autoimmunity: key to new therapies.
      Previous concepts of the pathogenesis of GPA consider granulomatous inflammation simply as a precursor of generalized GPA. However, recent findings suggest that GPA can broadly be subdivided into two distinct clinical entities (namely, a vasculitic one and a granulomatous one). Isolated granulomatous inflammation of the upper respiratory tract, without signs of systemic involvement, may represent a long-standing disease course, named localized GPA, not necessarily resulting in progress of the disease toward generalization. In these patients, the treatment of granuloma-mediated tissue destruction remains the primary goal during the long run of the chronic disease.
      • Holle J.U.
      • Gross W.L.
      • Holl-Ulrich K.
      • Ambrosch P.
      • Noelle B.
      • Both M.
      • Csernok E.
      • Moosig F.
      • Schinke S.
      • Reinhold-Keller E.
      Prospective long-term follow-up of patients with localised Wegener's granulomatosis: does it occur as persistent disease stage.
      Nasal mucosa specimens of patients with GPA show a broad morphological spectrum of nodular inflammation with organized accumulations of macrophages and other leukocytes; neutrophilic microabcesses, necrotizing palisading granulomas, and epitheloid cell granulomas are observed. The cells within these lesions consist of fibroblasts, epitheloid cells, giant cells, surrounding scattered infiltrates of lymphocytes, monocytes, macrophages, plasma cells, neutrophils, eosinophils, and areas of necrosis.
      • Kradin R.L.
      • Mark E.J.
      Case records of the Massachusetts General Hospital: weekly clinicopathological exercises: case 18–2002: a 48-year-old man with a cough and bloody sputum.
      • Devaney K.O.
      • Travis W.D.
      • Hoffman G.
      • Leavitt R.
      • Lebovics R.
      • Fauci A.S.
      Interpretation of head and neck biopsies in Wegener's granulomatosis: a pathologic study of 126 biopsies in 70 patients.
      • Del Buono E.A.
      • Flint A.
      Diagnostic usefulness of nasal biopsy in Wegener's granulomatosis.
      Mechanisms producing the severe tissue autodestruction typical of GPA (eg, saddle nose deformity and orbita destruction) have remained obscure. In some rare cases, nasal biopsy specimens of patients with GPA contain small bone fragments. Areas that show resorption of bone tissue are surrounded by human fibroblasts and show inflammatory infiltrates.
      • Mueller A.
      • Holl-Ulrich K.
      • Lamprecht P.
      • Gross W.L.
      Germinal centre-like structures in Wegener's granuloma: the morphological basis for autoimmunity.
      In the present study, we focus on these fibroblasts, because they secrete enzymes for matrix degradation, a physiological function during wound healing.
      • Xie J.
      • Bian H.
      • Qi S.
      • Xu Y.
      • Tang J.
      • Li T.
      • Liu X.
      Effects of basic fibroblast growth factor on the expression of extracellular matrix and matrix metalloproteinase-1 in wound healing.
      In rheumatoid arthritis (RA), two major fibroblast-mediated mechanisms lead to matrix degradation: first, antigen presentation, leading to recruitment of inflammatory cells into the tissue; and second, the secretion of matrix metalloproteinases (MMPs).
      • Muller-Ladner U.
      • Ospelt C.
      • Gay S.
      • Distler O.
      • Pap T.
      Cells of the synovium in rheumatoid arthritis: synovial fibroblasts.
      Therefore, we adopted the RA–severe combined immunodeficiency mouse transfer model of inflamed human synovia and articular cartilage, in which fibroblasts were identified as the key mediators of the tissue destruction.
      • Muller-Ladner U.
      • Ospelt C.
      • Gay S.
      • Distler O.
      • Pap T.
      Cells of the synovium in rheumatoid arthritis: synovial fibroblasts.
      Herein, we present the first data of the xenograft model using nasal mucosa and human allogeneic cartilage in pfp/rag2−/− mice to establish a similar model of the destructive inflammation seen in GPA and to analyze the role of fibroblasts and inflammation.

      Materials and Methods

      pfp/rag2−/− Mouse Co-Implantation Experiments

      Human GPA Samples

      Human nasal cartilage was obtained from healthy individuals undergoing plastic surgery at the Department of Otorhinolaryngology, Head and Neck Surgery, University Hospital Kiel, Kiel, Germany. Fragments of nasal mucosa from patients with active GPA lesions of the nasal cavity and a history of GPA-related tissue damage or from active sinusitis (controls) were obtained from selective biopsy specimens or routine surgery, respectively. Figure 1 demonstrates the spectrum of morphological changes in the nose from donor patients of this study.
      Figure thumbnail gr1
      Figure 1Morphological spectrum of destructive lesions of the nose related to GPA from donor patients of this study: patient 9 nasal mucosa transfer (A and B) and nasal fibroblast donor 3 (C and D). A: Early saddle nose deformity (arrow) with severe ulcerative inflammation of the endonasal mucosa. B: The asterisk demonstrates destructive ulcerations of a nasal turbinate. C: Severe saddle nose deformity (arrow) with perforation of the endonasal septum (pound sign; D). Patients provided consent for photographs.
      The diagnosis of GPA was made in accordance with the American College of Rheumatology classification criteria and Chapel Hill definitions for GPA, as recommended by The European League against Rheumatism.
      • Hellmich B.
      • Flossmann O.
      • Gross W.L.
      • Bacon P.
      • Cohen-Tervaert J.W.
      • Guillevin L.
      • Jayne D.
      • Mahr A.
      • Merkel P.A.
      • Raspe H.
      • Scott D.G.
      • Witter J.
      • Yazici H.
      • Luqmani R.A.
      EULAR recommendations for conducting clinical studies and/or clinical trials in systemic vasculitis: focus on anti-neutrophil cytoplasm antibody-associated vasculitis.
      GPA activity was measured at biopsy using the Birmingham Vasculitis Activity Score 2003.
      • Mukhtyar C.
      • Lee R.
      • Brown D.
      • Carruthers D.
      • Dasgupta B.
      • Dubey S.
      • Flossmann O.
      • Hall C.
      • Hollywood J.
      • Jayne D.
      • Jones R.
      • Lanyon P.
      • Muir A.
      • Scott D.
      • Young L.
      • Luqmani R.A.
      Modification and validation of the Birmingham Vasculitis Activity Score (version 3).
      Donor patients for control tissues experienced chronic sinusitis, with persistence of symptoms for >12 weeks.
      • Meltzer E.O.
      • Hamilos D.L.
      Rhinosinusitis diagnosis and management for the clinician: a synopsis of recent consensus guidelines.
      All patients had to be free from local steroid application at surgery. The exact time points of biopsies from patients with GPA had to be scheduled on short notice to avoid a delay of treatment. The numbers of xenografts from patients with GPA and sinusitis controls were matched for each experiment using one source of cartilage. The clinical data of all consecutive patients with GPA were recorded. The numbers of patients with GPA, transplants, and patient characteristics are summarized in Table 1. Implantations of nasal mucosa and cartilage into the pfp/rag2−/− mice were performed under sterile conditions, similar to existing models.
      • Muller-Ladner U.
      • Kriegsmann J.
      • Franklin B.N.
      • Matsumoto S.
      • Geiler T.
      • Gay R.E.
      • Gay S.
      Synovial fibroblasts of patients with rheumatoid arthritis attach to and invade normal human cartilage when engrafted into SCID mice.
      • Lefevre S.
      • Knedla A.
      • Tennie C.
      • Kampmann A.
      • Wunrau C.
      • Dinser R.
      • Korb A.
      • Schnaker E.M.
      • Tarner I.H.
      • Robbins P.D.
      • Evans C.H.
      • Sturz H.
      • Steinmeyer J.
      • Gay S.
      • Scholmerich J.
      • Pap T.
      • Muller-Ladner U.
      • Neumann E.
      Synovial fibroblasts spread rheumatoid arthritis to unaffected joints.
      Table 1Characteristics of Patients with GPA and Fibroblasts Used for Mouse Co-Implantation Experiments
      GPA experiment no.No. of TX/experiment
      Baseline TX experiments of seven patients with GPA (1 to 7), resulting in a total of 25 aggregates, and TX of three patients with GPA (8 to 10) with dexamethasone treatment of animals, resulting in a total of 11 aggregates.
      Animals treatedMOD (months)Age (years)SexInflammatory activity before TX
      Density score of inflammatory infiltrates: +, mild; ++, moderate; and +++, severe.
      Cartilage destruction
      +, mild, up to one third of the estimated cartilage area being destructed; ++, moderate, up to two thirds of the estimated cartilage area being destructed; and +++, severe, more than two thirds of the estimated cartilage area being destructed.
      ANCA
      Findings in ELISA and immunofluorescence testing: +, positive; -, negative.
      Destructive lesions by GPABVASOrgan involvement (ELK classification)
      Indicates involvement of ear, nose, and throat (E); lung (L); kidney (K); eye (Ey); arthritis (A); skin (S); gastrointestinal tract (GI); heart (H); peripheral nervous system (P); muscle (M); or constitutional symptoms (B).
      Current medications
      GCPBSGCPBS
      138439M+++Septum perforation6E, LCYC, GC
      24265M++++Necrotizing sinusitis27E, L, K, BCYC, GC
      341772F++++++Necrotizing lung granuloma3E, L, P, M, BMTX, GC
      448855M++++++Necrotizing parotitis7E, L, K, P, M, A, BLEF, MTX, GC
      52269M+++++Ulcerative rhinitis43E, K, P, S, GICYC, GC
      644164M+++++Ulcerative rhinitis28E, L, K, P, S, M, A, BCYC, GC
      74145M++++++Ulcerative rhinitis22E, K, Ey, A, BCYC, GC
      8422357M+++++Ulcerative rhinitis33E, L, K, P, H, A, BCYC, GC
      9422925M++++Saddle nose6ETMP
      1031218067M+++++++Ulcerative rhinitis7E, L, A, BLEF, MTX, GC
      All data were collected at nasal biopsy.
      F, female; M, male; ANCA, anti-neutrophil cytoplasmic autoantibody; BVAS, Birmingham Vasculitis Activity Score (2003); CYC, cyclophosphamide; GC, glucocorticoid; LEF, leflunomide; MOD, disease duration; MTX, methotrexate; TMP, trimethoprim-sulfamethoxazole; TX, transplant.
      low asterisk Baseline TX experiments of seven patients with GPA (1 to 7), resulting in a total of 25 aggregates, and TX of three patients with GPA (8 to 10) with dexamethasone treatment of animals, resulting in a total of 11 aggregates.
      Density score of inflammatory infiltrates: +, mild; ++, moderate; and +++, severe.
      +, mild, up to one third of the estimated cartilage area being destructed; ++, moderate, up to two thirds of the estimated cartilage area being destructed; and +++, severe, more than two thirds of the estimated cartilage area being destructed.
      § Findings in ELISA and immunofluorescence testing: +, positive; -, negative.
      Indicates involvement of ear, nose, and throat (E); lung (L); kidney (K); eye (Ey); arthritis (A); skin (S); gastrointestinal tract (GI); heart (H); peripheral nervous system (P); muscle (M); or constitutional symptoms (B).

      pfp/rag2−/− Mice

      The pfp/rag2−/− mice were provided by the central animal facilities, University Medical Center Hamburg, Hamburg, Germany, and kept under germ-free conditions in individually ventilated cages. Mice were anesthetized with an i.p. injection containing ketamine- rhompun (0.1 μL/g). A 1-cm-long incision was made in the skin on both the left and right abdominal flank of each animal, respectively, for generating a s.c. pocket to place the aggregates. The skin was then closed using suture clips (Precise Vista Skin Stapler; 3M, St. Paul, MN). Healthy human nasal cartilage and nasal mucosa fragments from seven patients with active GPA were engrafted, depending on the total number of transplants available, bilaterally or unilaterally, as a total of 25 aggregates into 14 mice. Forty-eight mucosal fragments from six patients with sinusitis were equally cotransplanted with nasal cartilage into 24 mice, as described for GPA as controls. For comparison of the tissue cellularity and composition, a fragment of each mucosal biopsy specimen was processed for histological analysis. After 21 days, mice were sacrificed with CO2 and implants were removed. Tissues were fixed in 4% buffered formalin, embedded into wax, and evaluated for cartilage destruction (H&E staining).

      Dexamethasone Treatment

      Normal human nasal cartilage and nasal mucosa fragments of three patients with active GPA of the nasal cavity were engrafted as a total of 11 aggregates into 11 mice. Transplanted mice were divided into two subgroups. The first subgroup (five aggregates in five mice) was treated with dexamethasone (4 μg/g in saline), and the second subgroup (six aggregates in overall six mice) was treated with saline only, as previously described.
      • Brack A.
      • Rittner H.L.
      • Younge B.R.
      • Kaltschmidt C.
      • Weyand C.M.
      • Goronzy J.J.
      Glucocorticoid-mediated repression of cytokine gene transcription in human arteritis: SCID chimeras.
      Dexamethasone was injected i.p., starting at day 3 after transplantation, every second day. Mice were sacrificed 21 days after transplantation.

      Histological Analysis

      Paraffin-embedded tissue was serially sectioned, and standard H&E staining was used. An evaluation of the sections was performed by two blinded experienced pathologists (U.S. and K.H.-U.).
      The inflammatory activity of tissue samples before transplantation was graded as mild, moderate, or severe. To consider the heterogeneity of tissue destructions in different layers of one transplanted aggregate, the maximum area of destruction of each transplant was scored as mild (up to one third of the estimated cartilage area being destructed), moderate (up to two thirds of the estimated cartilage area being destructed), or severe (more than two thirds of the estimated cartilage area being destructed) (Table 1). Histological findings were documented by photomicroscopy (Mirax Midi; Zeiss, Jena, Germany).

      IHC Data

      Immunohistochemistry (IHC) was performed using the avidin/biotinylated enzyme complex-alkaline phosphatase (ABC-AP) method using mouse monoclonal antibodies directed against the intermediate filament vimentin present in fibroblasts (monoclonal mouse anti-vimentin antibody, Vim 3B4; Dako, Hamburg, Germany), human T lymphocytes (monoclonal mouse anti-human CD3, F7.2.38; Dako), human B lymphocytes (monoclonal mouse anti-human CD20cy, L26; Dako), human macrophages (monoclonal mouse anti-human CD68, PG-M1; Dako), and human mitochondria (monoclonal mouse anti-human mitochondria; Millipore, Schwalbach, Germany). Human MMPs were stained using a Catalyzed Signal Amplification System [monoclonal mouse anti-human MMP-1, 41-1E5 (Research Diagnostics Inc, Flanders, NJ); monoclonal antibodies to human MMP-3, 55-2A4 (Acris, Herford, Germany); and anti-MMP-13 (collagenase-3), SPM292 (AnaSpec, San Jose, CA)], according to the manufacturers' manual.
      Sections (5 μm thick) were cut, deparaffinized, and pretreated in a 60°C to 85°C water bath overnight using citrate buffer (pH 6.0) or antigen retrieval solution (pH 6.1; Dako). Sections were rinsed with Tris-buffered saline, treated with normal rabbit serum for 30 minutes to reduce nonspecific antibody binding, and then incubated with the primary antibodies diluted 1:10 to 1:150 overnight at 4°C in a humidified chamber. Negative controls were performed without the primary antibodies, and nonspecific staining was excluded using appropriate isotype controls. The slides were rinsed in Tris-buffered saline and incubated for 30 minutes with biotinylated secondary rabbit anti-mouse IgG antibodies (Dako) using a 1:200 dilution. After 30 minutes of incubation with the ABC-AP complex (ABC-AP-Kit, Vectastain; Vector Laboratories, Burlingame, CA) and addition of substrate, color development was discontinued after 30 minutes and sections were counterstained with hematoxylin. Slides for staining of MMPs were pretreated in a 60°C water bath overnight using citrate buffer (pH 6.0) or EDTA buffer (pH 8.0), primary antibodies were diluted 1:500, and for the remaining steps, the Catalyzed Signal Amplification System (Dako) was used as indicated by the manufacturer.

      Profiling of Nasal Fibroblasts

      Patient Samples and Tissue Preparation

      Nasal fibroblasts were obtained from nasal mucosa biopsy specimens from patients with GPA (n = 8) with active and destructive disease in the nasal cavity and paranasal sinus. Characteristics of GPA fibroblast donor patients are summarized in Table 2. Controls were taken from healthy donors undergoing plastic surgery (n = 5). For cell culture, tissues were minced and digested for 1 hour at 37°C in 150 mg/mL of Dispase II (Roche, Mannheim, Germany). Nasal fibroblasts were grown in RPMI 1640 medium (Gibco, Invitrogen, Darmstadt, Germany), supplemented with 10% fetal calf serum and 50 IU/mL of penicillin-streptomycin, and maintained at 37°C in a humidified incubator with an atmosphere of 5% CO2. To ascertain the purity of the population, fibroblasts were used only from passage 3 to 9 for experiments, as previously described.
      • Ospelt C.
      • Brentano F.
      • Rengel Y.
      • Stanczyk J.
      • Kolling C.
      • Tak P.P.
      • Gay R.E.
      • Gay S.
      • Kyburz D.
      Overexpression of toll-like receptors 3 and 4 in synovial tissue from patients with early rheumatoid arthritis: toll-like receptor expression in early and longstanding arthritis.
      Table 2Characteristics of GPA Fibroblast Donor Patients for In Vitro Profiling Experiments
      Patient no.MOD (months)Age (years)SexInflammatory activity before TX
      Density score of inflammatory infiltrates: +, mild; ++, moderate; and +++, severe.
      ANCA
      Findings in ELISA and immunofluorescence testing: +, positive; −, negative.
      Destructive lesions by GPABVASOrgan involvement (ELK classification)
      Indicates involvement of ear, nose, and throat (E); lung (L); kidney (K); arthritis (A); skin (S); peripheral nervous system (P); muscle (M); or constitutional symptoms (B).
      Current medications
      1944F+++Saddle nose6E, LCYC, GC, TMP
      24163F++++Granulations28E, K, P, S, B, A, MAZA, GC
      311673M+++Granulations6EMTX, GC
      4477F+++Saddle nose6EGC
      56979M++Saddle nose18E, KGC
      6924M++++Saddle nose10E, KAZA, GC
      79070F+++Septum perforation6EMTX, GC
      8124F+++Saddle nose6EGC
      All data were collected at nasal biopsy.
      F, female; M, male; ANCA, anti-neutrophil cytoplasmic autoantibody; AZA, azathioprine; BVAS, Birmingham Vasculitis Activity Score (2003); CYC, cyclophosphamide; GC, glucocorticoid; LEF, leflunomide; MOD, disease duration; MTX, methotrexate; TMP, trimethoprim-sulfamethoxazole; TX, transplant.
      low asterisk Density score of inflammatory infiltrates: +, mild; ++, moderate; and +++, severe.
      Findings in ELISA and immunofluorescence testing: +, positive; −, negative.
      Indicates involvement of ear, nose, and throat (E); lung (L); kidney (K); arthritis (A); skin (S); peripheral nervous system (P); muscle (M); or constitutional symptoms (B).

      Cell Proliferation Assay

      For cell proliferation experiments, fibroblasts were plated in 96-well plates at a density of 105 cells/well and allowed to adhere overnight. Cells were either left unstimulated or stimulated with 10 ng/mL of tumor necrosis factor (TNF)-α and 1 ng/mL IL-1β (R&D Systems, Minneapolis, MN) for 24 hours. Cell proliferation was measured by using a colorimetric assay with 4-[3-(4-Iodophenyl)-2-(4-nitro-phenyl)-2H-5-tetrazolio]-1,3-benzene disulfonate (Cell Proliferation Reagent WST-1; Roche, Mannheim).

      Apoptosis Assessment

      To assess apoptosis, fibroblasts were plated in 96-well plates at a density of 105 cells/well and allowed to adhere overnight. Cells were either left unstimulated or stimulated with 10 ng/mL of TNF-α and 1 ng/mL of IL-1β (R&D Systems) for 24 hours. Apoptosis was induced by incubation in 20 μmol/L camptothecin (BioVision, Milpitas, CA) in RPMI 1640 medium (Gibco, Invitrogen) for 24 hours, and cell death was determined using a histone fragmentation assay (Cell Death Detection ELISAPlus; Roche, Minneapolis). This apoptosis assay is based on a quantitative sandwich enzyme immunoassay using mouse monoclonal antibodies directed against histones and DNA that enable the specific, quantitative determination of histone-associated DNA fragments (mononucleosomes and oligonucleosomes) in the cytoplasmic fraction of cell lysates and has been previously described.
      • Meinecke I.
      • Pap G.
      • Mendoza H.
      • Drange S.
      • Ender S.
      • Strietholt S.
      • Gay R.E.
      • Seyfert C.
      • Ink B.
      • Gay S.
      • Pap T.
      • Peters M.A.
      Small ubiquitin-like modifier 1 [corrected] mediates the resistance of prosthesis-loosening fibroblast-like synoviocytes against Fas-induced apoptosis.
      After induction of apoptosis, cell culture supernatants were withdrawn to remove necrotic cells from fragmented DNA. Cells were directly lysed within the plate wells for 30 minutes at room temperature. Afterward, nuclei were pelleted at 200 × g for 10 minutes and 20 μL of the supernatant (cytoplasmic fraction) was used in the enzyme-linked immunosorbent assay (ELISA), according to the manufacturers' instructions. After incubation with a peroxidase substrate for 15 minutes, the ELISA plates were read at 405 and 490 nm (reference wavelength). The rate of apoptosis is deliberated by the amount of nucleosomes that accumulated in the cytoplasm and was calculated following the manufacturers' protocol with the following formula: OD of sample cells/OD of control cells. Results are expressed as the OD of nucleosomes and shown as mean ± SD values.

      Real-Time RT-PCR Analysis

      To quantify mRNA levels of nasal fibroblasts, cells were seeded in six-well plates at a density of 3 × 105/well and allowed to adhere overnight. Cells were either left unstimulated or stimulated with 10 ng/mL of TNF-α and 1 ng/mL of IL-1β (R&D Systems) for 24 hours. Supernatants were centrifuged and collected, and cells were lysed with RNeasy Lysis Buffer buffer. Total RNA was isolated using an RNeasy Mini Prep Kit (Qiagen, Basel, Switzerland) with DNase treatment. First-strand cDNA was generated by reverse transcription of total RNA using the RT2 PCR Array First Strand Kit (SA Biosciences, Frederick, MD).
      The mRNA levels of disease-related genes (Table 3) were measured using a self-designed Custom RT2 PCR Array (SA Biosciences). Real-time PCR was performed using an ABI Prism 7700 Sequence Detection System (Applied Biosystems, Rotkreuz, Switzerland). Data were analyzed with the PCR Array Data Analysis Web Portal (SABiosciences, a QUIAGEN company, http://www.SABiosciences.com/pcrarraydataanalysis.php, last accessed November 18, 2010) using the ΔΔCT method, with the housekeeping genes for 18S ribosomal RNA, β-actin, and glyceraldehyde-3-phosphate dehydrogenase used for normalization.
      Table 3In Vitro Profiling of Nasal Fibroblasts
      GeneFold regulation (95% CI)
      ControlsGPA fibroblasts
      MMP127.8576 (0.00001–73.78)22.7651 (0.00001–60.85)
      MMP20.9151 (0.52–1.31)0.813 (0.33–1.30)
      MMP3126.0629 (0.00001–262.14)191.8387 (0.00001–538.03)
      MMP70.6105 (0.23–0.99)1.5223 (0.42–2.63)
      MMP84.8703 (0.00001–13.24)4.0174 (0.00001–9.71)
      MMP92.1916 (0.00001–5.67)4.0947 (0.00001–12.74)
      MMP1012.4839 (4.58–20.39)12.0629 (2.53–21.60)
      MMP110.1194 (0.09–0.15)0.1047 (0.05–0.16)
      MMP120.6653 (0.40–0.93)1.5409 (0.95–2.13)
      MMP1320.6776 (5.58–35.78)30.0127 (0.00001–64.76)
      TLR40.7879 (0.49–1.08)0.6998 (0.51–0.89)
      TLR64.494 (2.63–6.36)3.649 (2.75–4.55)
      TLR80.7955 (0.52–1.07)0.7969 (0.50–1.09)
      IL326.1817 (0.00001–13.03)6.0892 (3.01–9.16)
      MMP141.7339 (0.18–3.29)1.8025 (1.24–2.37)
      MMP200.9579 (0.80–1.12)0.9854 (0.87–1.10)
      HDAC11.0882 (0.89–1.28)1.1019 (0.91–1.29)
      HDAC21.0295 (0.69–1.36)0.9355 (0.68–1.20)
      HDAC30.7485 (0.67–0.82)0.6456 (0.57–0.72)
      HDAC40.6543 (0.46–0.85)0.7232 (0.56–0.89)
      HDAC50.668 (0.47–0.87)0.9044 (0.77–1.04)
      HDAC60.6792 (0.49–0.87)0.8215 (0.65–1.00)
      HDAC70.6139 (0.35–0.88)0.7928 (0.68–0.91)
      HDAC80.8213 (0.71–0.93)0.8101 (0.72–0.90)
      HDAC90.354 (0.24–0.47)0.3724 (0.25–0.49)
      HDAC100.3575 (0.12–0.59)0.6546 (0.46–0.85)
      HDAC110.2736 (0.11–0.43)0.5674 (0.42–0.72)
      CXCL120.249 (0.01–0.49)0.238 (0.10–0.37)
      VEGFA0.9075 (0.46–1.35)0.712 (0.49–0.93)
      VEGFB0.7749 (0.61–0.94)0.6736 (0.57–0.78)
      VEGFC3.4678 (2.77–4.17)3.4224 (2.42–4.42)
      IL6179.7689 (105.81–253.73)162.0168 (88.67–235.36)
      IL85525.924 (2973.38–8078.47)4943.27 (2037.19–7849.35)
      HIF1A1.1188 (0.91–1.33)0.933 (0.73–1.14)
      CSF21074.9099 (165.14–1984.68)682.6489 (55.73–1309.56)
      CSF338,060.3181 (14,356.21–61,764.43)8393.1695 (0.00001–18,117.22)
      PRTN30.9395 (0.79–1.09)1.1567 (0.86–1.46)
      CTSK0.905 (0.39–1.42)0.7276 (0.47–0.98)
      FAS0.9579 (0.77–1.14)0.913 (0.78–1.05)
      TP530.4263 (0.34–0.51)0.4957 (0.40–0.59)
      PTEN0.9619 (0.80–1.12)3.8073 (0.00001–13.61)
      IL160.6276 (0.24–1.02)1.0263 (0.26–1.80)
      Results of mRNA expression analysis of GPA fibroblasts versus healthy controls showing fold regulation after stimulation with TNF-α and IL-1β.

      ELISA Data

      To analyze the expression of MMP-1, MMP-3, IL-6, and IL-8 proteins by nasal fibroblasts, cells were seeded in six-well plates at a density of 3 × 105/well and allowed to adhere overnight. Cells were either left unstimulated or stimulated with 10 ng/mL of TNF-α and 1 ng/mL of IL-1β (R&D Systems) for 24 hours. Supernatants were collected, and levels of MMP-1, MMP-3, IL-6, and IL-8 were determined by ELISA with the use of DuoSet ELISA kits (R&D Systems) and OptEIA kits from BD PharMingen (San Diego, CA), according to the manufacturers' recommendations.

      Statistics

      For statistical analysis, SPSS 17.0.0 software and GraphPad Prism software version 5.0 (GraphPad Software, La Jolla, CA) were used. Values are presented as the mean ± SD or the mean fold change and 95% CI. mRNA expression data were analyzed using an unpaired Student's t-test, two-tailed. Values of protein expression experiments were logarithmized. A three-factor analysis of variance model with backwards selection was applied to investigate the effect of different genes, cell lines from patients with GPA and healthy individuals, treatment, and their interactions. For values from proliferation and apoptosis experiments, a mixed-model analysis was performed; values of proliferation were logarithmized; cell lines from patients with GPA and healthy individuals; treatment; and their interactions were included as fixed effects. Patients and experimental setting were included as random effects. For all analyses, a pairwise comparison was performed using a least significant difference post test. The level of significance was set to P < 0.05, two sided.

      Results

      Histological Features Comparing GPA with Controls Within Animal Experiments

      Donor patients with GPA experienced active endonasal disease at biopsy; patients' characteristics are summarized in Table 1. Overall, 36 fragments from 10 patients with GPA and 48 fragments from 6 patients with sinusitis, as controls, were successfully transplanted, as previously described (Figure 2, A and B, and Table 1). No animal fatalities or infections were noted. After 21 days, all transplanted tissues could easily be identified and harvested after sacrifice of the animals (Figure 2C). Transplanted fragments of human nasal mucosa showed no signs of necrosis. Capillaries in the mucosal tissue were perfused by mouse erythrocytes and lined by human endothelium, indicated by positive staining for A1 lectin from Ulex europaeus (Figure 3A). The ciliae of the nasal epithelium were still present, whereas submucosal glands showed cystic degeneration because of occlusion of the excretory ducts, indicating ongoing mucosal secretion (Figure 3, B and C).
      Figure thumbnail gr2
      Figure 2Principle experimental procedures. A: Fragments of nasal mucosa (asterisk) and human nasal cartilage (pound sign) ready for transplantation in a Petri dish. B: Schematic of placing both fragments into the collagen sponge. C: Transplant of cartilage (pound sign) and human nasal mucosa (asterisk) coated by collagen sponge (now appearing transparent) located in the subcutis at day 21 after transplantation.
      Figure thumbnail gr3
      Figure 3Histological features of transplants. A: Transplant of sinusitis control nasal mucosa at day 21 after transplantation. Arrow, perfused capillaries with positive staining for UEA-1 as a human-specific endothelial cell marker. B: Transplant of sinusitis control nasal mucosa, with arrows indicating cysts due to intact mucosal secretion, because excretory ducts were occluded after transplantation. C: Epithelial lining cells of mucosal glands (arrows) in transplant of sinusitis control nasal mucosa. D: Nasal mucosa (asterisk) from sinusitis control patients and human nasal cartilage (rhombus-shaped portion of slide) 21 days after transplantation, showing normal tissue architecture with inflated mucosal glands (arrows), due to occlusion of the excretory ducts after transplantation, coated by collagen sponge (sp) (H&E staining). E: Subsequent section of D, staining against human vimentin (asterisk), demonstrating transplanted tissues containing human fibroblasts. The surrounding mouse connective tissue does not show any immunoreactivity (arrows). F: Serial section of E, staining against human mitochondria, demonstrating that the transplanted tissue is of human origin (asterisk). Pound sign, human nasal cartilage.
      Nasal mucosa from patients with sinusitis showed either no destruction/invasion of the cotransplanted human nasal cartilage or, occasionally, a slightly uneven surface of the cartilage adjacent to the human nasal mucosa. Scattered infiltrates of human lymphocytes were detectable between the human fibroblasts of the connective tissue (Figure 3, D–F). Transplanted nasal mucosa from patients with GPA remained in an equally good condition as samples from patients with sinusitis (Figure 4J). In contrast to the controls, 71% (20/28) of all untreated GPA transplants showed areas of cartilage destruction, mediated by tissue arising from the transplanted nasal mucosal biopsy specimens (Figure 4A). The severity of cartilage destruction ranged from only small areas of invasion to total destruction of the cotransplanted cartilage (Figure 4, B–I and K). The cartilage-invading cells were human fibroblasts, as demonstrated by their positivity for human- specific anti-mitochondrial and anti-vimentin antibodies (Figure 3, E and F). Because nearly all cells of the transplanted tissues in areas of cartilage destruction were positive for these human markers, a significant infiltration of mouse cells could be excluded.
      Figure thumbnail gr4
      Figure 4Cartilage destruction mediated by GPA fibroblasts. A: Percentage of the number of transplants (without steroid treatment) showing cartilage destruction by cotransplanted nasal mucosa mediated by tissues derived from either patients with GPA or sinusitis controls. Numbers of transplantations are in parentheses. B and C: Human GPA nasal mucosa (asterisk) and healthy human nasal cartilage (pound sign) coated by collagen sponge (sp), 21 days after transplantation, showing an invasion of human cells into the cartilage (arrows) (H&E staining). D and E:Serial sections of B and C, staining against human vimentin (red), demonstrating transplanted tissues containing human fibroblasts invading the cartilage (arrows). Asterisk, human nasal mucosa; pound sign, human nasal cartilage. FI: Human GPA nasal mucosa (asterisk) and healthy human cartilage (pound sign) coated by collagen sp, 21 days after transplantation, showing a progressive destruction of cartilage by invading human cells (arrows) (H&E staining). The transplanted tissue 21 days after transplantation (K) closely resembles the tissue of origin (H&E staining) (J). Asterisk, human nasal mucosa; pound sign, human nasal cartilage. L–O: Human GPA nasal mucosa (asterisk) and healthy human cartilage (pound sign), 21 days after transplantation, showing a bone fragment (b) of the patients with GPA transplanted with the nasal mucosa by accident (L and N, arrows). H&E staining (L) and staining against human vimentin (MO), showing areas of proliferating cells expanding from the original mucosal transplant destructing the cartilage (arrows) and in multinucleated giant cells (red arrows) at sites of invasion (black arrows) (O).
      The morphological features of GPA tissues transplanted into mice reflected different morphological stages, from samples showing acute inflammation to samples showing areas of scar formation. Destruction was only observed when areas of active inflammation were present. Destructions were evident by the partial replacement of the cartilage tissue by dense infiltrates of connective tissue cells, mainly fibroblasts (Figure 4G). The results of the quantification of destructive lesions are summarized in Table 1.
      In some cases, invasive human fibroblasts had partially or completely engulfed the cartilage transplant. The surrounding collagen sponge was generally not infiltrated by fibroblasts. Only single human fibroblasts were detected adjacent to the human mucosal transplants. Two of the human nasal mucosal transplants from patients with GPA contained bone fragments that were transplanted incidentally. These bone fragments were also infiltrated by fibroblasts (Figure 4, L–N). IHC revealed inflammatory infiltrates consisting of human B lymphocytes, T lymphocytes, and macrophages, in both patient samples and the transplants (Figure 5, A–C). Human granulocytes were absent in the transplants at 21 days after transplantation. Some tissue samples showed the presence of multinucleated giant cells at sites of cartilage invasion (Figure 4O). IHC analysis for MMPs showed considerably higher expression levels of MMPs 1, 3, and 13 at sites of invasion within the transplants of patients with GPA compared with control transplants and to the areas of human fibroblasts, which surrounded the cartilage but did not infiltrate it (Figure 5, D–I). Comparable expression of these MMPs
      • Bacon P.A.
      The spectrum of Wegener's granulomatosis and disease relapse.
      • Kradin R.L.
      • Mark E.J.
      Case records of the Massachusetts General Hospital: weekly clinicopathological exercises: case 18–2002: a 48-year-old man with a cough and bloody sputum.
      • Muller-Ladner U.
      • Kriegsmann J.
      • Franklin B.N.
      • Matsumoto S.
      • Geiler T.
      • Gay R.E.
      • Gay S.
      Synovial fibroblasts of patients with rheumatoid arthritis attach to and invade normal human cartilage when engrafted into SCID mice.
      at contact sites of control tissues and transplanted cartilage was not detectable.
      Figure thumbnail gr5
      Figure 5Cellular composition and expression of MMPs in GPA transplants. AC: Transfer of human GPA nasal mucosa (asterisk) and healthy human cartilage (pound sign) in pfp/rag2−/− mice 21 days after transplantation. Sections of formalin-fixed tissues were stained with antibodies directed against human CD3 (brown) (A), CD20 (brown) (B), and CD68 (brown) (C). Arrows, sites of invasion into the cartilage. DI: Expression of MMPs 1, 3, and 13 (all in brown) in transplants containing nasal mucosa (asterisk) from patients with GPA (D, F, and H) or sinusitis as controls (E, G, and I). Arrows, sites of invasion into the cartilage (pound sign), 21 days after transplantation.

      Dexamethasone Treatment

      Transplants from animals that received corticosteroid treatment (40 mg/kg weight) showed only minimal areas of cartilage affection, comparable to those observed in the sinusitis control mice from the baseline experiments (Figure 6, A–E). Cartilage destruction appeared in a pattern typical for GPA transplants in mice treated with the vehicle solution.
      Figure thumbnail gr6
      Figure 6Dexamethasone treatment experiments. AE: Human GPA nasal mucosa (asterisk) and healthy human cartilage (pound sign) coated by collagen sponge (sp), 21 days after transplantation. Mice were treated with either dexamethasone [H&E staining (A) and anti-vimentin staining (B)] or saline (CE, anti-vimentin staining) as control, showing destruction of cartilage by invading human cells (arrows).

      In Vitro Profiling of Nasal Fibroblasts

      mRNA Expression Array Analysis and Protein Expression

      mRNA expression array analyses showed that nonstimulated GPA fibroblasts (n = 8) and controls (n = 5) expressed all transcripts at a similar level. In both controls and GPA fibroblasts, mRNA levels for transcripts were increased after stimulation with TNF-α and IL-1β, although no significant differences between these groups were observed (results are summarized in Table 3 and Figure 7A). Protein expression in cell culture supernatants from GPA fibroblasts (n = 8) and controls (n = 5) was measured for MMPs 1 and 3 and ILs 6 and 8, respectively. In supernatants from nonstimulated cell cultures, only baseline levels of all proteins were detectable: for controls, MMP-1 (6.95 ± 7.41), MMP-3 (1.63 ± 1.20), IL-6 (1.03 ± 0.14), and IL-8 (0.06 ± 0.16); and for GPA fibroblasts, MMP-1 (19.87 ± 40.60), MMP-3 (1.11 ± 0.94), IL-6 (2.62 ± 1.51), and IL-8 (0.11 ± 0.08). After stimulation with TNF-α and IL-1β, protein levels for GPA fibroblasts were as follows: MMP-1 (43.73 ± 48.03), MMP-3 (50.76 ± 25.34), IL-6 (368.89 ± 82.04), and IL-8 (277.15 ± 83.98); and for controls, MMP-1 (36.00 ± 28.10), MMP-3 (78.60 ± 58.70), IL-6 (284.54 ± 74.56), and IL-8 (327.00 ± 99.14). MMP-13 was not detectable. Differences between the protein expression of GPA and control fibroblasts were not significant at baseline or after stimulation. Within the group of GPA and control fibroblasts, protein levels were substantially up-regulated after stimulation (MMP-1, P < 0.001; MMP-3, P < 0.001; IL-6, P < 0.001; IL-8, P < 0.001), thus correlating with an increase in the corresponding mRNA levels. Results demonstrating the protein levels in the culture supernatants of unstimulated and stimulated cells are shown in Figure 7, B and C.
      Figure thumbnail gr7
      Figure 7In vitro analysis of GPA fibroblasts. A: mRNA expression of GPA fibroblasts. Induction of mRNA for the genes MMP-1, MMP-3, MMP-8, MMP-9, MMP-10, and MMP-13; TLR6; IL-6, IL-8, and IL-32; VEGFC; CSF2 and CSF3; and PTEN in controls (C; n = 5) and GPA fibroblasts (GPA; n = 8) by proinflammatory cytokines TNF-α and IL-1 β. Values are the mean fold induction of mRNA and the 95% CI relative to the unstimulated condition. B: Expression of MMP-1, MMP-3, IL-6, and IL-8 protein by unstimulated GPA fibroblasts (GPA; n = 8) and controls (n = 5). C: Stimulated with tumor TNF- α and IL-1β. Values are the mean ± SD. D: Proliferation of GPA fibroblasts (GPA; n = 8) and controls (n = 5) that were left unstimulated or stimulated with proinflammatory cytokines TNF-α and IL-1β. Values are the mean ± SD. E: Apoptosis of GPA fibroblasts (GPA; n = 5) and controls (n = 5) that were left unstimulated or stimulated with proinflammatory cytokines TNF-α and IL-1β. Apoptosis then was induced by camptothecin. Values are the mean ± SD.

      Proliferation and Apoptosis

      The differences between the mean rates of apoptosis were significantly lower in GPA fibroblasts (n = 5) at baseline and after stimulation with TNF-α and IL-1β compared with controls (n = 5). Observed values were as follows: GPA baseline (12.70 ± 13.25) versus control baseline (35.72 ± 33.34) and GPA after stimulation (9.08 ± 9.52) versus controls after stimulation (24.20 ± 14.46).
      In contrast, proliferation of GPA fibroblasts (n = 8) and controls (n = 5) gave the following results: GPA baseline (0.05 ± 0.01), control baseline (0.06 ± 0.03), and GPA and controls after stimulation with TNF-α and IL-1β (0.08 ± 0.01) and (0.09 ± 0.03), respectively. Both groups showed a significant increase of cell proliferation after stimulation, whereas differences between GPA and controls were not significant. The results of proliferation and apoptosis analysis are summarized in Figure 7, D and E.

      Discussion

      Inflammation, granuloma, necrosis, and vasculitis are leading signs in the histopathological diagnosis of GPA. Because GPA is an autoimmune disease, this granulomatous inflammation might represent an important mediator of tissue destruction and form the basis of the autoimmune process.
      • Falk R.J.
      • Gross W.L.
      • Guillevin L.
      • Hoffman G.
      • Jayne D.R.
      • Jennette J.C.
      • Kallenberg C.G.
      • Luqmani R.
      • Mahr A.D.
      • Matteson E.L.
      • Merkel P.A.
      • Specks U.
      • Watts R.
      Granulomatosis with polyangiitis (Wegener's): an alternative name for Wegener's granulomatosis.
      The peculiar predilection of GPA to the upper respiratory tract is probably a result of a dysfunctional barrier of the nasal mucosa, as the first results postulate.
      • Ullrich S.
      • Gustke H.
      • Lamprecht P.
      • Gross W.L.
      • Schumacher U.
      • Ambrosch P.
      • Laudien M.
      Severe impaired respiratory ciliary function in Wegener granulomatosis.
      • Laudien M.
      • Hasler R.
      • Wohlers J.
      • Bock J.
      • Lipinski S.
      • Bremer L.
      • Podschun R.
      • Ambrosch P.
      • Lamprecht P.
      • Rosenstiel P.
      • Till A.
      Molecular signatures of a disturbed nasal barrier function in the primary tissue of Wegener's granulomatosis.
      Although remission of vasculitis-related manifestations can successfully be induced by traditional and targeted immunosuppressive therapy, persisting and relapsing destructive lesions causing substantial local damage to the adjacent organs (eg, eye and brain) remain a poorly understood feature and are still a therapeutic challenge.
      • Holle J.U.
      • Gross W.L.
      • Holl-Ulrich K.
      • Ambrosch P.
      • Noelle B.
      • Both M.
      • Csernok E.
      • Moosig F.
      • Schinke S.
      • Reinhold-Keller E.
      Prospective long-term follow-up of patients with localised Wegener's granulomatosis: does it occur as persistent disease stage.
      • Stone J.H.
      • Merkel P.A.
      • Spiera R.
      • Seo P.
      • Langford C.A.
      • Hoffman G.S.
      • Kallenberg C.G.
      • St Clair E.W.
      • Turkiewicz A.
      • Tchao N.K.
      • Webber L.
      • Ding L.
      • Sejismundo L.P.
      • Mieras K.
      • Weitzenkamp D.
      • Ikle D.
      • Seyfert-Margolis V.
      • Mueller M.
      • Brunetta P.
      • Allen N.B.
      • Fervenza F.C.
      • Geetha D.
      • Keogh K.A.
      • Kissin E.Y.
      • Monach P.A.
      • Peikert T.
      • Stegeman C.
      • Ytterberg S.R.
      • Specks U.
      Rituximab versus cyclophosphamide for ANCA-associated vasculitis.
      • Jones R.B.
      • Tervaert J.W.
      • Hauser T.
      • Luqmani R.
      • Morgan M.D.
      • Peh C.A.
      • Savage C.O.
      • Segelmark M.
      • Tesar V.
      • van Paassen P.
      • Walsh D.
      • Walsh M.
      • Westman K.
      • Jayne D.R.
      Rituximab versus cyclophosphamide in ANCA-associated renal vasculitis.
      • Tony H.P.
      • Burmester G.
      • Schulze-Koops H.
      • Grunke M.
      • Henes J.
      • Kotter I.
      GRAID Investigators
      Safety and clinical outcomes of rituximab therapy in patients with different autoimmune diseases: experience from a national registry (GRAID).
      Pathophysiological mechanisms producing tissue destruction, resulting in saddle nose deformity or orbita destruction, have not attracted the same attention as signs and symptoms of life-threatening, full-blown GPA.
      • Bacon P.A.
      The spectrum of Wegener's granulomatosis and disease relapse.
      • Mueller A.
      • Holl-Ulrich K.
      • Lamprecht P.
      • Gross W.L.
      Germinal centre-like structures in Wegener's granuloma: the morphological basis for autoimmunity.
      In this report, we describe the first in vivo xenograft animal model of chronic inflammation typical of GPA featuring cartilage destruction mediated by invasive growth of connective tissue. This work extends earlier observations in the histological composition of biopsy specimens of patients with GPA.
      • Mueller A.
      • Holl-Ulrich K.
      • Lamprecht P.
      • Gross W.L.
      Germinal centre-like structures in Wegener's granuloma: the morphological basis for autoimmunity.
      It also stands in line with findings in other diseases, such as RA or polyposis nasi, which used immunodeficient mice for the establishment of their xenograft models.
      • Muller-Ladner U.
      • Kriegsmann J.
      • Franklin B.N.
      • Matsumoto S.
      • Geiler T.
      • Gay R.E.
      • Gay S.
      Synovial fibroblasts of patients with rheumatoid arthritis attach to and invade normal human cartilage when engrafted into SCID mice.
      • Bernstein J.M.
      • Broderick L.
      • Parsons R.R.
      • Bankert R.B.
      Human nasal polyp microenvironment maintained in viable and functional states as xenografts in SCID mice.
      Although nasal mucosa from patients with chronic sinusitis showed only marginal signs of destruction of the cotransplanted human nasal cartilage, 72% of GPA transplants, taken from patients with active disease of the nasal cavity, showed areas of cartilage destruction, mediated by cells arising from the transplanted nasal mucosa. This invasive tissue mainly consisted of human fibroblasts and few human inflammatory cells (T and B lymphocytes and macrophages), as shown by IHC using human-specific antibodies, which do not cross-react with mouse antigens. Therefore, human cells from patients with GPA are still detectable after transplantation and may govern the fibroblast-mediated degradation of the extracellular matrix, characteristics that were not observed in tissues from sinusitis controls. In addition, the pattern of cartilage destruction shown in our model closely resembles biopsy findings in patients with GPA.
      • Mueller A.
      • Holl-Ulrich K.
      • Lamprecht P.
      • Gross W.L.
      Germinal centre-like structures in Wegener's granuloma: the morphological basis for autoimmunity.
      A review of 20 nasal mucosa biopsy specimens from patients with GPA showed eight cases in which the biopsy specimens contained small bone fragments. These bone fragments showed vimentin-positive fibroblasts that invaded the bone fragments in a fishbone tissue texture (K.H.-U. and A. Müller, personal oral communication) (Figure 8); thus, our model resembles the situation in the patient closely. All patients with GPA included in this study had a history of destructive lesions related to GPA. Because all patients received immunosuppressive treatment at biopsy, we speculate that GPA fibroblasts reactivate their destructive potential after xenotransplantation in mice in which immuosuppressive treatment is absent. Therefore, a further study should investigate if experiments with tissues from patients with subdued disease of the nasal cavity, who never developed destructive lesions, will also be nondestructive after xenotransplantation in mice. This may help to define new subsets of destructive versus nondestructive courses of GPA. An additional destructive effect of multinucleated giant cells cannot be completely excluded because they were also found in some transplants in areas of cartilage invasion (Figure 4M).
      Figure thumbnail gr8
      Figure 8Bone destruction mediated by GPA fibroblasts. Human GPA nasal mucosa (asterisk) with a bone fragment (b) of a patient with GPA showing an invasion of human cells into the cartilage (arrows), staining against human vimentin (red).
      In general, cartilage destruction in the nose, as seen in cocaine abusers, has been attributed to be cell mediated and/or resulting from an ischemic environment of cartilage because of reduced perfusion of the covering mucosa.
      • Trimarchi M.
      • Gregorini G.
      • Facchetti F.
      • Morassi M.L.
      • Manfredini C.
      • Maroldi R.
      • Nicolai P.
      • Russell K.A.
      • McDonald T.J.
      • Specks U.
      Cocaine-induced midline destructive lesions: clinical, radiographic, histopathologic, and serologic features and their differentiation from Wegener granulomatosis.
      Our results clearly demonstrate that GPA-related destruction is mediated by fibroblasts, a process that can be completely inhibited by systemic steroid treatment. It is most likely that remaining human lymphocytes and macrophages orchestrate the fibroblast-mediated tissue destruction. These findings stand in line with the role of fibroblasts as activated effector cells of destruction in other autoimmune diseases, such as RA and Crohn disease.
      • Muller-Ladner U.
      • Kriegsmann J.
      • Franklin B.N.
      • Matsumoto S.
      • Geiler T.
      • Gay R.E.
      • Gay S.
      Synovial fibroblasts of patients with rheumatoid arthritis attach to and invade normal human cartilage when engrafted into SCID mice.
      • Bernstein J.M.
      • Broderick L.
      • Parsons R.R.
      • Bankert R.B.
      Human nasal polyp microenvironment maintained in viable and functional states as xenografts in SCID mice.
      • Reilkoff R.A.
      • Bucala R.
      • Herzog E.L.
      Fibrocytes: emerging effector cells in chronic inflammation.
      In RA, it was shown that fibroblasts isolated from inflamed synovia keep their unique aggressive character, even in vitro, for several passages.
      • Muller-Ladner U.
      • Ospelt C.
      • Gay S.
      • Distler O.
      • Pap T.
      Cells of the synovium in rheumatoid arthritis: synovial fibroblasts.
      Similar to the findings in RA, in IHC, we could demonstrate that MMPs 1, 3, and 13 are expressed at sites of invasion and, therefore, act as mediators of extracellular matrix breakdown in the GPA transplants. In accordance with earlier reports, our findings have shown nasal fibroblasts (GPA and controls) to be capable of producing high levels of proinflammatory factors (eg, ILs 6 and 8) and destructive enzymes (eg, MMPs 1, 3, and 13)
      • Pawankar R.
      Inflammatory mechanisms in allergic rhinitis.
      in vitro. In RA, synovial fibroblasts show an up-regulated production of destructive MMPs that characterized them as a distinct fibroblast subtype. In our in vitro experiments, both fibroblasts from GPA patients and controls produced high levels of destructive MMPs (MMPs 1 and 3) after stimulation. This finding suggests that the destructive growth of inflammatory tissue in patients with GPA results from the significant delay of apoptosis found in GPA fibroblasts. The underlying mechanisms of the delayed apoptosis of GPA fibroblasts are not yet clear, because the expression of p53 and apoptosis antigen 1 mRNA showed no significant differences between GPA and controls. Further analysis of apoptotic pathways (eg, hypoxic modulation and epigenetic modulation, such as SUMOylation of sirtuin 1)
      • Jungel A.
      • Ospelt C.
      • Gay S.
      What can we learn from epigenetics in the year 2009.
      is needed to explain the apoptotic resistance of GPA fibroblasts.
      The growth pattern of invasive GPA tissue frequently resembles growth patterns of malignant tumors. The high levels of MMP-13 expression in GPA tissue at sites of invasion, such as the up-regulation of MMP-13 mRNA levels after stimulation with proinflammatory cytokines in culture, may contribute to this similarity, because MMP-13 is related to invasive growth of malignant tumors.
      • Leivonen S.K.
      • Ala-Aho R.
      • Koli K.
      • Grenman R.
      • Peltonen J.
      • Kahari V.M.
      Activation of Smad signaling enhances collagenase-3 (MMP-13) expression and invasion of head and neck squamous carcinoma cells.
      The continued activation of multiple proinflammatory pathways and the resulting aggressive character of the inflammatory process found in GPA may explain why local destruction remains a therapeutic challenge. Glucocorticoids are an established immunosuppressive treatment in both localized and generalized GPA.
      • Mukhtyar C.
      • Flossmann O.
      • Hellmich B.
      • Bacon P.
      • Cid M.
      • Cohen-Tervaert J.W.
      • Gross W.L.
      • Guillevin L.
      • Jayne D.
      • Mahr A.
      • Merkel P.A.
      • Raspe H.
      • Scott D.
      • Witter J.
      • Yazici H.
      • Luqmani R.A.
      European Vasculitis Study Group (EUVAS)
      Outcomes from studies of antineutrophil cytoplasm antibody associated vasculitis: a systematic review by the European League Against Rheumatism systemic vasculitis task force.
      Earlier xenograft mouse models of giant cell arteritis, for example, have demonstrated that steroid treatment of the mice prevents disease progression, as it does in humans.
      • Brack A.
      • Rittner H.L.
      • Younge B.R.
      • Kaltschmidt C.
      • Weyand C.M.
      • Goronzy J.J.
      Glucocorticoid-mediated repression of cytokine gene transcription in human arteritis: SCID chimeras.
      In our model, corticosteroid treatment was efficient in suppressing cartilage destruction, validating our model to be highly relevant in recapitulating the processes of tissue destruction in humans. Recent findings in the severe combined immunodeficiency mouse model for RA suggest that fibroblasts actively migrate through the mouse to penetrate cartilage at distant sites and initiate the inflammatory process there as well.
      • Lefevre S.
      • Knedla A.
      • Tennie C.
      • Kampmann A.
      • Wunrau C.
      • Dinser R.
      • Korb A.
      • Schnaker E.M.
      • Tarner I.H.
      • Robbins P.D.
      • Evans C.H.
      • Sturz H.
      • Steinmeyer J.
      • Gay S.
      • Scholmerich J.
      • Pap T.
      • Muller-Ladner U.
      • Neumann E.
      Synovial fibroblasts spread rheumatoid arthritis to unaffected joints.
      We regularly observed human fibroblasts around the human cartilage. Thus, it may be speculated that in those with GPA, fibroblasts also may act as circulating mediators of inflammation.

      Acknowledgments

      We thank Dr. Antje Müller for editing assistance, Dr. Mary Connolly for reviewing the manuscript, and Janet Wohlers for collecting patients' tissues for the profiling of nasal fibroblasts.

      References

        • Bacon P.A.
        The spectrum of Wegener's granulomatosis and disease relapse.
        N Engl J Med. 2005; 352: 330-332
        • Mueller A.
        • Holl-Ulrich K.
        • Lamprecht P.
        • Gross W.L.
        Germinal centre-like structures in Wegener's granuloma: the morphological basis for autoimmunity.
        Rheumatology (Oxford). 2008; 47: 1111-1113
        • Kradin R.L.
        • Mark E.J.
        Case records of the Massachusetts General Hospital: weekly clinicopathological exercises: case 18–2002: a 48-year-old man with a cough and bloody sputum.
        N Engl J Med. 2002; 346: 1892-1899
        • Lamprecht P.
        • Gross W.L.
        Antineutrophil cytoplasmic antibody-associated vasculitis: autoinflammation, autodestruction and autoimmunity: key to new therapies.
        Trends Immunol. 2008; 29: 587-588
        • Holle J.U.
        • Gross W.L.
        • Holl-Ulrich K.
        • Ambrosch P.
        • Noelle B.
        • Both M.
        • Csernok E.
        • Moosig F.
        • Schinke S.
        • Reinhold-Keller E.
        Prospective long-term follow-up of patients with localised Wegener's granulomatosis: does it occur as persistent disease stage.
        Ann Rheum Dis. 2010; 69: 1934-1939
        • Devaney K.O.
        • Travis W.D.
        • Hoffman G.
        • Leavitt R.
        • Lebovics R.
        • Fauci A.S.
        Interpretation of head and neck biopsies in Wegener's granulomatosis: a pathologic study of 126 biopsies in 70 patients.
        Am J Surg Pathol. 1990; 14: 555-564
        • Del Buono E.A.
        • Flint A.
        Diagnostic usefulness of nasal biopsy in Wegener's granulomatosis.
        Hum Pathol. 1991; 22: 107-110
        • Xie J.
        • Bian H.
        • Qi S.
        • Xu Y.
        • Tang J.
        • Li T.
        • Liu X.
        Effects of basic fibroblast growth factor on the expression of extracellular matrix and matrix metalloproteinase-1 in wound healing.
        Clin Exp Dermatol. 2008; 33: 176-182
        • Muller-Ladner U.
        • Ospelt C.
        • Gay S.
        • Distler O.
        • Pap T.
        Cells of the synovium in rheumatoid arthritis: synovial fibroblasts.
        Arthritis Res Ther. 2007; 9: 223
        • Hellmich B.
        • Flossmann O.
        • Gross W.L.
        • Bacon P.
        • Cohen-Tervaert J.W.
        • Guillevin L.
        • Jayne D.
        • Mahr A.
        • Merkel P.A.
        • Raspe H.
        • Scott D.G.
        • Witter J.
        • Yazici H.
        • Luqmani R.A.
        EULAR recommendations for conducting clinical studies and/or clinical trials in systemic vasculitis: focus on anti-neutrophil cytoplasm antibody-associated vasculitis.
        Ann Rheum Dis. 2007; 66: 605-617
        • Mukhtyar C.
        • Lee R.
        • Brown D.
        • Carruthers D.
        • Dasgupta B.
        • Dubey S.
        • Flossmann O.
        • Hall C.
        • Hollywood J.
        • Jayne D.
        • Jones R.
        • Lanyon P.
        • Muir A.
        • Scott D.
        • Young L.
        • Luqmani R.A.
        Modification and validation of the Birmingham Vasculitis Activity Score (version 3).
        Ann Rheum Dis. 2009; 68: 1827-1832
        • Meltzer E.O.
        • Hamilos D.L.
        Rhinosinusitis diagnosis and management for the clinician: a synopsis of recent consensus guidelines.
        Mayo Clin Proc. 2011; 86: 427-443
        • Muller-Ladner U.
        • Kriegsmann J.
        • Franklin B.N.
        • Matsumoto S.
        • Geiler T.
        • Gay R.E.
        • Gay S.
        Synovial fibroblasts of patients with rheumatoid arthritis attach to and invade normal human cartilage when engrafted into SCID mice.
        Am J Pathol. 1996; 149: 1607-1615
        • Lefevre S.
        • Knedla A.
        • Tennie C.
        • Kampmann A.
        • Wunrau C.
        • Dinser R.
        • Korb A.
        • Schnaker E.M.
        • Tarner I.H.
        • Robbins P.D.
        • Evans C.H.
        • Sturz H.
        • Steinmeyer J.
        • Gay S.
        • Scholmerich J.
        • Pap T.
        • Muller-Ladner U.
        • Neumann E.
        Synovial fibroblasts spread rheumatoid arthritis to unaffected joints.
        Nat Med. 2009; 15: 1414-1420
        • Brack A.
        • Rittner H.L.
        • Younge B.R.
        • Kaltschmidt C.
        • Weyand C.M.
        • Goronzy J.J.
        Glucocorticoid-mediated repression of cytokine gene transcription in human arteritis: SCID chimeras.
        J Clin Invest. 1997; 99: 2842-2850
        • Ospelt C.
        • Brentano F.
        • Rengel Y.
        • Stanczyk J.
        • Kolling C.
        • Tak P.P.
        • Gay R.E.
        • Gay S.
        • Kyburz D.
        Overexpression of toll-like receptors 3 and 4 in synovial tissue from patients with early rheumatoid arthritis: toll-like receptor expression in early and longstanding arthritis.
        Arthritis Rheum. 2008; 58: 3684-3692
        • Meinecke I.
        • Pap G.
        • Mendoza H.
        • Drange S.
        • Ender S.
        • Strietholt S.
        • Gay R.E.
        • Seyfert C.
        • Ink B.
        • Gay S.
        • Pap T.
        • Peters M.A.
        Small ubiquitin-like modifier 1 [corrected] mediates the resistance of prosthesis-loosening fibroblast-like synoviocytes against Fas-induced apoptosis.
        Arthritis Rheum. 2009; 60: 2065-2070
        • Falk R.J.
        • Gross W.L.
        • Guillevin L.
        • Hoffman G.
        • Jayne D.R.
        • Jennette J.C.
        • Kallenberg C.G.
        • Luqmani R.
        • Mahr A.D.
        • Matteson E.L.
        • Merkel P.A.
        • Specks U.
        • Watts R.
        Granulomatosis with polyangiitis (Wegener's): an alternative name for Wegener's granulomatosis.
        Ann Rheum Dis. 2011; 70: 704
        • Ullrich S.
        • Gustke H.
        • Lamprecht P.
        • Gross W.L.
        • Schumacher U.
        • Ambrosch P.
        • Laudien M.
        Severe impaired respiratory ciliary function in Wegener granulomatosis.
        Ann Rheum Dis. 2009; 68: 1067-1071
        • Laudien M.
        • Hasler R.
        • Wohlers J.
        • Bock J.
        • Lipinski S.
        • Bremer L.
        • Podschun R.
        • Ambrosch P.
        • Lamprecht P.
        • Rosenstiel P.
        • Till A.
        Molecular signatures of a disturbed nasal barrier function in the primary tissue of Wegener's granulomatosis.
        Mucosal Immunol. 2011; 4: 564-573
        • Stone J.H.
        • Merkel P.A.
        • Spiera R.
        • Seo P.
        • Langford C.A.
        • Hoffman G.S.
        • Kallenberg C.G.
        • St Clair E.W.
        • Turkiewicz A.
        • Tchao N.K.
        • Webber L.
        • Ding L.
        • Sejismundo L.P.
        • Mieras K.
        • Weitzenkamp D.
        • Ikle D.
        • Seyfert-Margolis V.
        • Mueller M.
        • Brunetta P.
        • Allen N.B.
        • Fervenza F.C.
        • Geetha D.
        • Keogh K.A.
        • Kissin E.Y.
        • Monach P.A.
        • Peikert T.
        • Stegeman C.
        • Ytterberg S.R.
        • Specks U.
        Rituximab versus cyclophosphamide for ANCA-associated vasculitis.
        N Engl J Med. 2010; 363: 221-232
        • Jones R.B.
        • Tervaert J.W.
        • Hauser T.
        • Luqmani R.
        • Morgan M.D.
        • Peh C.A.
        • Savage C.O.
        • Segelmark M.
        • Tesar V.
        • van Paassen P.
        • Walsh D.
        • Walsh M.
        • Westman K.
        • Jayne D.R.
        Rituximab versus cyclophosphamide in ANCA-associated renal vasculitis.
        N Engl J Med. 2010; 363: 211-220
        • Tony H.P.
        • Burmester G.
        • Schulze-Koops H.
        • Grunke M.
        • Henes J.
        • Kotter I.
        • GRAID Investigators
        Safety and clinical outcomes of rituximab therapy in patients with different autoimmune diseases: experience from a national registry (GRAID).
        Arthritis Res Ther. 2011; 13: R75
        • Bernstein J.M.
        • Broderick L.
        • Parsons R.R.
        • Bankert R.B.
        Human nasal polyp microenvironment maintained in viable and functional states as xenografts in SCID mice.
        Ann Otol Rhinol Laryngol. 2006; 115: 65-73
        • Trimarchi M.
        • Gregorini G.
        • Facchetti F.
        • Morassi M.L.
        • Manfredini C.
        • Maroldi R.
        • Nicolai P.
        • Russell K.A.
        • McDonald T.J.
        • Specks U.
        Cocaine-induced midline destructive lesions: clinical, radiographic, histopathologic, and serologic features and their differentiation from Wegener granulomatosis.
        Medicine (Baltimore). 2001; 80: 391-404
        • Reilkoff R.A.
        • Bucala R.
        • Herzog E.L.
        Fibrocytes: emerging effector cells in chronic inflammation.
        Nat Rev Immunol. 2011; 11: 427-435
        • Pawankar R.
        Inflammatory mechanisms in allergic rhinitis.
        Curr Opin Allergy Clin Immunol. 2007; 7: 1-4
        • Jungel A.
        • Ospelt C.
        • Gay S.
        What can we learn from epigenetics in the year 2009.
        Curr Opin Rheumatol. 2010; 22: 284-292
        • Leivonen S.K.
        • Ala-Aho R.
        • Koli K.
        • Grenman R.
        • Peltonen J.
        • Kahari V.M.
        Activation of Smad signaling enhances collagenase-3 (MMP-13) expression and invasion of head and neck squamous carcinoma cells.
        Oncogene. 2006; 25: 2588-2600
        • Mukhtyar C.
        • Flossmann O.
        • Hellmich B.
        • Bacon P.
        • Cid M.
        • Cohen-Tervaert J.W.
        • Gross W.L.
        • Guillevin L.
        • Jayne D.
        • Mahr A.
        • Merkel P.A.
        • Raspe H.
        • Scott D.
        • Witter J.
        • Yazici H.
        • Luqmani R.A.
        • European Vasculitis Study Group (EUVAS)
        Outcomes from studies of antineutrophil cytoplasm antibody associated vasculitis: a systematic review by the European League Against Rheumatism systemic vasculitis task force.
        Ann Rheum Dis. 2008; 67: 1004-1010