The effect of class II transactivator mutations on bleomycin-induced lung inflammation and fibrosis

Abstract

IFN-γ expression increases during the inflammatory response after bleomycin injury in mice. IFN-γ deficiency attenuates lung inflammation and fibrosis. Because IFN-γ stimulates class II transactivator (CIITA) expression, which activates major histocompatibility class (MHC) II and represses collagen expression, it was hypothesized that CIITA mediates IFN-γ action after bleomycin injury. To test this hypothesis, two CIITA mouse lines, one carrying a mutation of the leucine-rich region of CIITA (CIITA C-/-) and one with a deletion extending into the GTP-binding domain (CIITA G-/-), were used. IFN-γ treatment of lung cells isolated from both strains of mice induced mutant CIITA expression, which did not activate MHC II transcription. Collagen expression was similar in both mutant mouse strains and comparable to C57BL/6 (wild-type) mice. When mice were exposed to intratracheal bleomycin, both strains of CIITA mutant mice retained body weight and altered inflammation at 14 days after bleomycin injury compared with bleomycin-treated wild-type mice. However, there was no difference in fibrosis as judged by histology, mRNA, and protein expression of lungs. Bronchoalveolar lavage cells from CIITA C-/- and C57BL/6 lungs were examined at 3, 7, and 14 days after bleomycin injury. CD4 mRNA expression in bronchoalveolar lavage cells was down-regulated, whereas IL-4 and IL-10 expression was up-regulated, in CIITA C-/- mice, indicating a diminished, skewed Th2 response. The expression of IFN-γ was the same in all mice tested. Combined, our data suggest that CIITA mutations altered the immune response without affecting fibrosis.

Figure 1.

Expression of class II transactivator (CIITA), major histocompatibility class (MHC II), and collagen type I mRNA in wild-type (WT) and CIITA hypomorphic mice. (A) Diagram of the CIITA mutation in CIITA C−/− (top) and CIITA G−/− mouse DNA (bottom). The CIITA repression domains of collagen are acidic domain (AD) and proline/serine/threonine (PST) domain (light gray). The leucine-rich region (LRR) (black) is deleted in the CIITA C−/− mouse, and the GDP binding domain (dark gray) is deleted in the CIITA G−/− mouse. (B–E) Lung fibroblast cells isolated from WT C57BL/6 (white bars), CIITA C−/− (light gray bars), or CIITA G−/− (dark gray bars) were treated with or without IFN-γ as indicated. The mRNA levels of CIITA (B), HLA-DRα (MHC II) (C), col1a1 (D), and col1a2 (E) were determined by real-time PCR. Data are presented as relative mRNA levels compared with WT samples performed in triplicate (mean ± SD). Statistical analysis by ANOVA with post hoc comparisons by Scheffé. **P < 0.01; ***P < 0.001. A representative experiment out of three repeats is shown. Cell isolation and tranfection methods as described in online supplement.

Figure 2.

IFN-γ represses a collagen promoter activity but does not activate MHC II promoter activity in CIITA hypomorphic mice. (A) Collagen promoter (col1a2) or (B) an MHC II promoter (DRA300) was transfected in triplicate, along with a GFP plasmid, into lung fibroblast cells isolated from WT C57BL/6 (white bars), CIITA C−/− (light gray bars), or CIITA G−/− (dark gray bars), followed by treatment with IFN-γ for 24 hours. Luciferase activities were normalized for protein concentration and GFP fluorescence. A representative experiment out of three repeats is shown. Cell isolation and transfection methods as described in online supplement.

Figure 3.

CIITA hypomorphic mice have less disease but equivalent fibrosis compared with WT mice. (A) Histological sections from the right caudal, left central, and left hilar regions of lung were coded using a quantitative scale (0–100%) of lung area involved with abnormal cellular infiltration, interstitial thickening, or architectural distortion of native lung architecture. The average disease index (± SE) was graphed for bleomycin (B) and saline (S) lungs from C57BL/6 (white bars), CIITA C−/− (light gray bars), and CIITA G−/− (dark gray bars). N = the number of lungs analyzed. Data were analyzed by nonparametric group data using Mann-Whitney test. (B) Representative histological sections of bleomycin-treated lungs stained with Masson's trichrome (left panels) and picrosirus red (right panels). Masson's trichrome = blue, ECM; red, cells. Picrosirus = red, all types of collagen.

Figure 4.

CIITA hypomorphic mice have fewer T cells and macrophages in bleomycin-treated lungs compared with WT mice. Representative histological sections of bleomycin-treated lungs stained with antibody to (A) F4/80 (1:50), (B) CD3 (1:100), and (C) CD45 (1:250). Red staining material is antibody stain. Slides were counterstained with hematoxylin. Nine pictures covering the entire slide were analyzed for red and total lung area using Image-pro plus. The average percent red ± SE from three or four animals was graphed (right panels). Statistical analysis by ANOVA with post hoc comparisons by Scheffé. **P < 0.05.

Figure 5.

Fibrosis is not altered in CIITA hypomorphic mice. (A) Representative histological sections of bleomycin (Bleo)-treated lungs stained with antibody to collagen α1(I) (1:200) and counterstained with hematoxylin. Red staining is collagen. (B) A graph of percent collagen by antibody staining in bleomycin- (B) and saline-treated (S) lungs at 14 days after bleomycin injury. Six pictures of each lung were analyzed. N = number of lungs. (C) A graph of percent ECM from Masson's stained slides in bleomycin- (B) and saline-treated (S) lungs at 14 days after bleomycin injury. Six pictures of each lung were analyzed. N = the number of lungs analyzed. (B and C) C57BL/6 (white bars); CIITA C−/− (light gray bars); CIITA G−/− (dark gray bars). (D) Proteins were acid extracted from lungs 14 days after bleomycin (B) or saline (S) instillation as described in Materials and Methods. Collagen was measured by Sircol assay, which measures all collagenous proteins. (E) RNA was extracted from lungs 14 days after bleomycin (B) or saline (S) treatment, and COL1A1 mRNA was measured and normalized to 18S RNA. Each sample was measured in triplicate. In D E, data were analyzed by ANOVA with post hoc comparisons by the method of Scheffé.

Figure 6.

CIITA hypomorphic mouse has equivalent amounts of active BALF TGF-β and lung TGF-β expression to WT mice. BALF was collected from WT (white bars) or CIITA C−/− (gray bars) mice at 3, 7, and 15 days and added to mink lung epithelial cells as described in Materials and Methods. Luciferase activity was measured in triplicate and plotted against a standard curve to generate TGF-β concentrations (pg/ml). Triplicate concentrations of TGF-β for WT (white bars) or CIITA C−/− mice (dark gray bars) were combined for each time point (n = 5).

Figure 7.

CIITA hypomorphic mouse BAL cells express higher levels of IL-4 and IL-10 cytokines than in WT mice. BAL cells collected from WT (white bars) or CIITA C−/− (gray bars) mice at 3, 7, and 15 days as described in Materials and Methods. RNA was extracted from BAL cells and analyzed by QRT-PCR in triplicate for (A) CIITA, (B) IFN-γ, (C) IL-10, and (D) IL-4 mRNA levels. One representative experiment out of three is shown. WT (white bars) or CIITA C−/− (gray bars); n = 5 animals per group. Data were analyzed by ANOVA with post hoc comparisons by the method of Scheffé.