Transglutaminase 2 knockout mice are protected from bleomycin-induced lung fibrosis with preserved lung function and reduced metabolic derangements

Abstract

Pulmonary fibrosis is an interstitial scarring disease of the lung characterized by poor prognosis and limited treatment options. Tissue transglutaminase 2 (TG2) is believed to promote lung fibrosis by crosslinking extracellular matrix components and activating latent TGFβ. This study assessed physiologic pulmonary function and metabolic alterations in the mouse bleomycin model with TG2 genetic deletion. TG2-deficient mice demonstrated attenuated the fibrosis and preservation of lung function, with significant reduction in elastance and increases in compliance and inspiratory capacity compared to control mice treated with bleomycin. Bleomycin induced metabolic changes in the mouse lung that were consistent with increased aerobic glycolysis, including increased expression of lactate dehydrogenase A and increased production of lactate, as well as increased glutamine, glutamate, and aspartate. TG2-deficient mice treated with bleomycin exhibited similar metabolic changes but with reduced magnitude. Our results demonstrate that TG2 is required for a typical fibrosis response to injury. In the absence of TG2, the fibrotic response is biochemically similar to wild-type, but lesions are smaller and lung function is preserved. We also show for the first time that profibrotic pathways of tissue stiffening and metabolic reprogramming are interconnected, and that metabolic disruptions in fibrosis go beyond glycolysis.

Keywords: metabolism; metabolomics; pulmonary fibrosis; transglutaminase.

FIGURE 1

TG2 deficiency attenuates lung fibrosis in the bleomycin mouse model. C57BL/6 and TG2 KO mice were treated with 2 units/kg bleomycin and harvested on Day 21 as described. Lung tissue was inflated and fixed, and FFPE sections were prepared. (a) Representative whole slide images of H&E‐stained sections. (b) Collagen was detected in FFPE sections using SHG microscopy. Green pixels indicate collagen, red pixels indicate non‐collagen tissue, and black pixels indicate air spaces (airways, alveoli, or blood vessels). Representative regions are shown. White arrows highlight collagen present around airways. Orange boxes highlight collagen changes in the parenchyma. (c) SHG collagen was quantified removing airway collagen and expressed as fold change normalized to saline‐treated C57BL/6 mice. To quantify collagen in parenchymal tissue, an algorithm was used to exclude collagen around large airways and blood vessels, and collagen in the parenchyma (green pixels) was determined as a percentage of total tissue (red + green pixels). N = 9–13 mice per group. (d) Collagen as detected by hydroxyproline in total lung tissue N = 15 mice per group. p‐values from two‐way ANOVA with Sidak's post‐test.

FIGURE 2

TG2 deficiency attenuates the expression of fibrosis‐related genes. C57BL/6 and TG2 KO mice were treated with bleomycin and harvested as described above. RNA was extracted from the right middle lung lobe, and expression of (a) collagen 1, (b) collagen 3, and (c) fibronectin determined by RT‐PCR and normalized to GAPDH. N = 5 mice per group. p‐values are by one‐way ANOVA with Tukey's post‐test for multiple comparisons.

FIGURE 3

TG2 deficiency preserves lung function in bleomycin‐treated mice. C57BL/6 and TG2 KO mice were treated with bleomycin as described above. Immediately prior to euthanasia (Day 21), lung function was measured with a flexiVent ventilator. (a) Elastance, (b) Tissue elastance (H), (c) Tissue dampening (G), (d) dynamic compliance (C), (e) Quasi‐static compliance, (f) Resistance (R), (g) airway resistance (RN), and (h) Inspiratory capacity from PV loop Area. PV, pressure–volume. N = 9–13 mice per group. p‐values from two‐way ANOVA with Sidak's post‐test.

FIGURE 4

Lactate glycolytic reprogramming is reduced in TG2 deficient mice. (a) Serial sections were stained with Gomori's trichrome, or immunostained for LDHA, and isotype control. Control sections are in normal parenchymal area. Bleomycin sections are from fibrotic lesions. H&E images were taken with a 4× objective and Gomori's trichrome, LDHA, and isotype control images were taken with a 10× objective (4× scale bar = 500 μm, 10× scale bar = 100 μm). The dashed box in the 4× H&E images shows the regions that were photographed at 10× for the other stains. (b) Western blot of total lung tissue homogenates for LDHA normalized to loading control beta‐tubulin. (c) Quantification of densitometry for LDHA/βTubulin; n = 4–5 per group. (d) Lactate in total lung tissue homogenates. (e) Total TGFβ per upper right lung lobe.

FIGURE 5

Untargeted metabolomics reveals altered amino acid and energy metabolism. (a) Heatmap of significantly changing features (uncorrected p ≤ 0.05, log2 fold change>|1|). One hundred seventy‐eight total spectral features. Seventeen named compounds. (b) MetaboAnalyst pathway enrichment of known metabolites identified in the dataset. (c–e) Key metabolites in the glutamine pathway are shown, normalized to total detections. Points represent individual mice, p‐values are from two‐way ANOVA with Sidak's post‐test. (f) Glutamate metabolism can provide alternative energy inputs into the TCA cycle or resources for collagen synthesis.