Comparative analysis of lysyl oxidase (like) family members in pulmonary fibrosis

Abstract

Extracellular matrix (ECM) composition and stiffness are major driving forces for the development and persistence of fibrotic diseases. Lysyl oxidase (LOX) and LOX-like (LOXL) proteins play crucial roles in ECM remodeling due to their collagen crosslinking and intracellular functions. Here, we systematically investigated LOX/L expression in primary fibroblasts and epithelial cells under fibrotic conditions, Bleomycin (BLM) induced lung fibrosis and in human IPF tissue. Basal expression of all LOX/L family members was detected in epithelial cells and at higher levels in fibroblasts. Various pro-fibrotic stimuli broadly induced LOX/L expression in fibroblasts, whereas specific induction of LOXL2 and partially LOX was observed in epithelial cells. Immunohistochemical analysis of lung tissue from 14 IPF patients and healthy donors revealed strong induction of LOX and LOXL2 in bronchial and alveolar epithelium as well as fibroblastic foci. Using siRNA experiments we observed that LOXL2 and LOXL3 were crucial for fibroblast-to-myofibroblast transition (FMT). As FMT could only be reconstituted with an enzymatically active LOXL2 variant, we conclude that LOXL2 enzymatic function is crucial for fibroblast transdifferentiation. In summary, our study provides a comprehensive analysis of the LOX/L family in fibrotic lung disease and indicates prominent roles for LOXL2/3 in fibroblast activation and LOX/LOXL2 in IPF.

Figure 1

Basal expression levels of lysyl oxidases in primary human lung cells. LOX/L gene expression in primary human lung fibroblasts (NHLF) (left) or in air-liquid-differentiated primary bronchial epithelial cells (HBEC) (right) was determined using Taqman gene expression assays. Data is normalized to RNA-polymerase 2 expression (Ct value of 25) and presented as mean of ΔCt ± SD of five independent experiments.

Figure 2

LOX/L gene expression level changes induced by various pro-fibrotic stimuli in primary human lung cells. (a) Primary human lung fibroblasts (NHLF) were treated with TGF-β1 (5 ng/ml), FGF (20 ng/ml) or PDGF (50 ng/ml) for 24 hours or kept under hypoxic conditions (0.5% O₂) for 24 h followed by RNA isolation and quantitative PCR for the members of the LOX/L family. Data was normalized to RNA polymerase 2 expression and depicted as fold change relative to respective control treatment. Data is shown as mean ± SD of 3–4 independent experiments. (b) Primary human bronchial epithelial cells (HBECs) were differentiated at the air liquid interface for 21 days followed by basal treatment with TGF-β1 (5 ng/ml) for 48 h or kept under hypoxic conditions for 24 hours. RNA was isolated and quantitative PCR for the members of the LOX/L family was performed. Expression data was normalized to RNA polymerase 2 expression and control treatment and is depicted as mean ± SD of three experiments. *p < 0.05; **p < 0.01; ***p < 0.001, relative to control treatment.

Figure 3

LOX/L expression in experimental lung fibrosis. (a) Masson trichrome stained histological sections of control (NaCl) or Bleomycin treated animals, 14 days after Bleomycin instillation. Nuclei appear in dark red, cytoplasm in light red and collagen in blue. (b) LOX/L gene expression was determined in whole lung homogenates from Bleomycin treated mice at day 14 after treatment using Taqman gene expression assays. Data is normalized to RNA-polymerase 2 expression (Ct value of 25) and presented as mean fold change ± SD of 8–9 animals per group. *p < 0.05; **p < 0.01; ***p < 0.001, relative to control treatment. (c) Representative images of immunohistochemical analyses of LOX, LOXL1, LOXL2, LOXL3 and LOXL4 at day 14 after intratracheal Bleomycin instillation. Positive cells are depicted in brown. (d) Correlation of the mRNA results shown in (b) with computational IHC image analysis-derived fold changes in protein expression (see Supplemental Fig. S2 for details).

Figure 4

LOX/L expression in human IPF lungs. (a) Masson trichrome stained histological sections of donor and IPF patient lung tissue. Nuclei appear in dark red, cytoplasm in light red and collagen in blue (scale bar = 200 $\mathit{\mu }\mathrm{m}$). (b) Expression of LOX, LOXL1, LOXL3 and LOXL4 in IPF and donor lungs. Representative pictures of immunohistochemistry in tissue specimen of patients with IPF and transplant donors are shown. Positive cells are depicted in brown (scale bar = 50 $\mathit{\mu }\mathrm{m}$). (c) Immunohistochemistry of LOXL2 in tissue specimen of patients with IPF and transplant donors. LOXL2 in bronchial epithelium is marked with black arrows. LOXL2 expression in hyplerplastic alveolar epithelial cells in the interstitial region and fibroblast foci are indicated by arrow heads (scale bar = 50 $\mathit{\mu }\mathrm{m}$).

Figure 5

Knockdown efficacy and specificity of siRNA pools. Individual LOX/L siRNA pools were analyzed for their knockdown specificity following transfection of primary human lung fibroblasts (NHLFs) with 16.6 nM of control, LOX, LOXL1, LOXL2, LOXL3 or LOXL4 siRNA. 24 hours after transfection, mRNA levels of the individual LOX/L family members were determined using Taqman gene expression assays. The data is depicted as % expression compared to control siRNA treatment (which was set 100%) and presented as mean ± SD of two to four independent experiments. ***p < 0.001, relative to control.

Figure 6

LOX/L knockdown leads to decreased smooth muscle actin production and stress fibre formation in NHLFs. (a) Primary human lung fibroblasts (NHLFs) were transfected with 16.6 nM of control, LOX, LOXL1, LOXL2, LOXL3 or LOXL4 siRNA. After 24 h, cells were stimulated with TGF-β (5 ng/ml) for 48 hours. Gene expression changes of acta2 were determined using Taqman gene expression assays and normalized to RNA-polymerase 2 expression and control treatment. The data represent the mean ± SD of three independent experiments (**p < 0.01; ***p < 0.001). (b) NHLFs were transfected as described in (a) and stimulated with TGF-β1 for 72 hours. αSMA fibrils per cell were quantified after immunocytochemical staining using high-content cellular imaging of 16 wells of a 384-well plate per condition (see “Image analysis” in the methods section for details). The data represent the mean ± SD of three independent experiments (**p < 0.01). Representative pictures of the cells stained for αSMA (red) and nuclei (DAPI, blue) are shown in (c). (d) acta2 induction by TGFβ1 on RNA level is decreased after FAK signaling pathway depletion. NHLF were transfected with 16.6 nM of control, ITGB3, FAK, AKT or SRF siRNA. After 24 h cells were stimulated with TGF-β1 (5 ng/ml) for 48 hours. Gene expression changes of acta2 were determined using Taqman gene expression assays. Data was normalized to RNA-polymerase 2 expression and control treatment and is presented as mean ± SD of three independent experiments (*p < 0.05).

Figure 7

FMT is dependent on the enzymatic activity of LOXL2. (a) Comparison of collagen I crosslinking capacity of recombinant LOXL2 and LOXL3. Human lung collagen I coated microplates were incubated with LOXL2 or LOXL3, respectively, for 2 hours followed by aldehyde detection using O-(biotinylcarbazoylmethyl) hydroxylamine as a probe. Relative fluorescence units (RFU) indicate the amount of aldehyde intermediate in collagen I produced by LOXL2 and LOXL3, respectively. White bars indicate resulting fluorescence after incubation with 20 nM LOX/L, whereas black bars represent 100 nM of the respective enzyme. Data is shown as mean ± SD of three independent experiments. **p < 0.01; ***p < 0.001. (b) NHLF cells were transfected with 16.6 nM of control or LOXL2 siRNA. After 12 hours, cells were incubated in conditioned medium derived from HEK293 cells transfected with LOXL2 wild type, LOXL2 enzymatic dead or control plasmids (mock). 24 h after transfection, cells were stimulated with TGF-β1 (5 ng/ml) for 48 hours in presence of the respective conditioned medium. Gene expression changes of acta2 mRNA were determined using Taqman gene expression assays. Data are normalized to RNA-polymerase 2 expression and control treatment and presented as mean ± SD of three independent experiments. *p < 0.05, relative to siControl treatment. (c) LOXL2 protein content detected by LOXL2-specific Western Blotting, 72 hours after HEK293 transfection with the respective plasmids (10 $\mathit{\mu }\mathrm{l}$ supernatant per lane).