Glutathione-S-transferase P promotes glycolysis in asthma in association with oxidation of pyruvate kinase M2

Abstract

Background: Interleukin-1-dependent increases in glycolysis promote allergic airways disease in mice and disruption of pyruvate kinase M2 (PKM2) activity is critical herein. Glutathione-S-transferase P (GSTP) has been implicated in asthma pathogenesis and regulates the oxidation state of proteins via S-glutathionylation. We addressed whether GSTP-dependent S-glutathionylation promotes allergic airways disease by promoting glycolytic reprogramming and whether it involves the disruption of PKM2.

Methods: We used house dust mite (HDM) or interleukin-1β in C57BL6/NJ WT or mice that lack GSTP. Airway basal cells were stimulated with interleukin-1β and the selective GSTP inhibitor, TLK199. GSTP and PKM2 were evaluated in sputum samples of asthmatics and healthy controls and incorporated analysis of the U-BIOPRED severe asthma cohort database.

Results: Ablation of Gstp decreased total S-glutathionylation and attenuated HDM-induced allergic airways disease and interleukin-1β-mediated inflammation. Gstp deletion or inhibition by TLK199 decreased the interleukin-1β-stimulated secretion of pro-inflammatory mediators and lactate by epithelial cells. ¹³C-glucose metabolomics showed decreased glycolysis flux at the pyruvate kinase step in response to TLK199. GSTP and PKM2 levels were increased in BAL of HDM-exposed mice as well as in sputum of asthmatics compared to controls. Sputum proteomics and transcriptomics revealed strong correlations between GSTP, PKM2, and the glycolysis pathway in asthma.

Conclusions: GSTP contributes to the pathogenesis of allergic airways disease in association with enhanced glycolysis and oxidative disruption of PKM2. Our findings also suggest a PKM2-GSTP-glycolysis signature in asthma that is associated with severe disease.

Keywords: Allergic airways disease; House dust mite; Interleukin-1β; S-glutathionylation; Thymic stromal lymphopoietin.

Fig. 1

Genetic ablation of Gstp attenuates mucus metaplasia, subepithelial collagen, markers of airway remodeling, and airway hyperresponsiveness in mice with HDM-induced allergic airway disease. A, Assessment and quantification of mucus metaplasia by PAS staining intensity (original magnification, ×200). B, MUC5AC protein content in BAL fluid. n=5 per group. C, Assessment and quantification of collagen deposition by Masson's trichrome staining (original magnification, ×200). D, mRNA expression of Col1a1, Fn1, and Acta2 (α-smooth muscle actin) normalized to Ppia.E, Assessment of airway responsiveness to inhaled methacholine using a Flexivent small animal ventilator. F, Total, and G, Differential cell counts in BAL fluid. n=8–10 per group unless otherwise noted. * respective saline vs HDM group; # WT HDM vs Gstp^−/− HDM. */#p < 0.05 analyzed by two-way ANOVA.

Fig. 2

Genetic ablation of Gstp attenuates HDM-mediated lactate secretion. Measurements of lactate levels in A, BAL fluid and B, lung tissue. C, Representative Western blot analyses for HK2, PKM2, LDHA, TBK1, IKKε. β-actin is used as the loading control. D, Measurements of IL-1β, CCL20, and IL-33 in lung tissue homogenates by ELISA. n=9–10 per group. *p < 0.05 analyzed by two-way ANOVA.

Fig. 3

Gstp absence attenuates the release of pro-inflammatory cytokines and decreases airway neutrophilia following intranasal administration of IL-1β. A, Schematic depicting the intranasal administration of 1 μg of IL-1β for either 6 or 24 h. The total cell count and cell differentials in the BAL fluid reflect 24 h post IL-1β treatment, the other results shown are obtained 6 h post IL-1β administration. B, Representative western blots for total GSTP levels. β-actin; loading control. C, Total, and D, Differential cell counts in BAL fluid. E, Levels of the pro-inflammatory cytokines TSLP, GM-CSF, CXCL1 and CCL20 in lung tissue homogenates by ELISA. F, Measurements of lactate levels in BAL fluid. n=4–8 per group. *p < 0.05 analyzed by two-way ANOVA.

Fig. 4

Ablation or inhibition of Gstp, by TLK199, attenuates IL-1β-induced lactate and pro-inflammatory responses in primary MTE cells. A, Measurements of lactate and B, pro-inflammatory cytokines TSLP, GM-CSF, CXCL1, and CCL20 in cell culture supernatants of wild-type and Gstp^−/- MTE cells. n=3–9 per group. C, Wild-type MTE cells were pre-treated with 50 μM TLK199 for 1 h followed by stimulation of 1 ng/mL IL-1β for 24 h. Lactate levels, and D, levels of pro-inflammatory cytokines in cull culture supernatants. n=3–9 per group. E, Metabolomics data of total levels (unlabeled fraction and labelling with ¹³C glucose) of metabolites in the glycolysis pathway. n=5–6 per group. The heavy metabolites are presented with solid color panels and are depicted as M+3 (green/blue), M+5 (purple), and M+6 (orange). The unlabeled portion of each metabolite is depicted as M and presented with solid white color panels. *p < 0.05 analyzed by two-way ANOVA. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 5

GSTP-linked S-glutathionylaton of Pyruvate Kinase M2 attenuates its glycolysis activity. A, Total PSSG in lung tissue from saline- and HDM-exposed WT and Gstp^−/- mice, analyzed by DTNB assay. n=9–10 per group. B, Representative Western blot of PKM2-SSG in saline vs HDM-exposed mouse lungs. C, Pyruvate kinase enzymatic activity assay of 10 ng recombinant PKM2 incubated with 50 μM oxidized glutathione (GSSG). D, Complex models and interactions between dimeric or tetrameric PKM2 with a GSTP dimer by protein-protein docking. PDB structures used for modeling: 1T5A (PKM2 tetramer) and 1AQW (GSTP dimer). The most likely models are illustrated here, and all predicted models are shown in a cartoon in the supplemental information. E, Representative western blots of GSTP-PKM2 interaction via co-immunoprecipitation and total PKM2 levels in IL-1β-treated mouse lung tissues. F, Representative Western blot of PKM2-GSTP (IP: PKM2; IB; GSTP) and total PKM2 levels in wild type MTE cells pre-treated with 50 μM TLK199 for 1 h followed by stimulation with 1 ng/ml IL-1β for 2 h *p < 0.05 analyzed by two-way ANOVA.

Fig. 6

GSTP and PKM2 expression are increased in HDM-exposed BALF and in sputum of people with asthma with high levels of IL-1ß and lactate and correlates with disease severity. A, Representative western blots of extracellular GSTP and PKM2 in BALF from HDM-exposed WT and Gstp^−/- mice. B, Representative western blots of extracellular GSTP and PKM2 from induced sputum samples of 4 healthy subjects (HS), 4 participants with asthma (AL) with low sputum IL-1β (<20 pg/mL) and low sputum lactate levels (<33 μM), and 4 participants with asthma (AH) with high sputum IL-1β (>30 pg/mL) and sputum lactate (>80 μM) levels. Each lane represents an individual participant. C, Quantification of GSTP (left) and PKM2 (right) immunoreactivity from 20 HS, AL and AH subjects. Individual western blots of GSTP and PKM2 from the additional 16 HS, AL, and AH subject are shown in Fig. S5. D, Spearman correlation analysis of bronchial brushings transcriptomics, sputum transcriptomics and sputum proteomics datasets, derived from the U-BIOPRED cohort. Shown are Rho and p values of the correlations between glycolysis and PKM2, glycolysis and GSTP, as well as PKM2 and GSTP. E, Figure showing the correlation between GSTP and PKM2 in patients of the different datasets: bronchial brushing transcriptomics (left), sputum transcriptomics (middle), and sputum proteomics (right). HV: healthy volunteers, MMA: mild/moderate non-smoking asthma subjects, SA.ns: severe non-smoking participants with asthma, SA.S: current/ex-smokers with severe asthma. *p < 0.05 analyzed by one-way ANOVA with Tukey's multiple comparisons test.