Homocysteine Exacerbates Pulmonary Fibrosis via Orchestrating Syntaxin 17 Homocysteinylation of Alveolar Type II Cells

Abstract

Idiopathic pulmonary fibrosis (IPF) is a lethal interstitial lung disease, marked by progressive extracellular matrix deposition, for which there are no effective treatments to halt disease progression. Although hyperhomocysteinemia is implicated in multiple pathological processes, its role in IPF remains largely unexplored. Through multiomics profiling of IPF patients, significantly elevated homocysteine (Hcy) concentrations in plasma and bronchoalveolar lavage fluid are identified compared to healthy controls. Single-cell RNA sequencing and spatial transcriptomics reveal alveolar type 2 epithelial cells as the primary site of Hcy metabolism, with downregulation of Hcy-catabolizing enzyme methionine synthase reductase (MTRR) during fibrotic progression. Genetic perturbation studies in murine models demonstrate that MTRR knockdown exacerbates bleomycin-induced mortality and fibrosis, whereas MTRR overexpression exerts protective effects. Furthermore, Hcy supplementation initiates and accelerates pulmonary fibrosis development, while folate administration reduces pulmonary Hcy levels and alleviates fibrosis. Mechanistically, it is revealed that pathogenic hyperhomocysteinemia induces homocysteinylation-ubiquitination cascades that modify Syntaxin 17 (STX17) posttranslationally, leading to its proteasomal degradation and consequent impairment of autophagic flux. Notably, pharmacological folate administration reverses STX17 depletion, restoring autophagic flux and mitigating pulmonary fibrosis in mouse models. These findings collectively establish a Hcy-STX17-proteostasis axis wherein excess homocysteinylation creates a self-reinforcing loop of autophagy dysfunction and fibrogenesis.

Keywords: autophagy; folate; homocysteine; homocysteinylation; idiopathic pulmonary fibrosis.

Figure 1

MR and case control study reveals potential risk of Hcy on IPF. A) Analysis flowchart showing the steps of the experiment. B) Scatter plot showing positive relationship between Hcy and IPF. C) Forest plot showing positive MR effect size for Hcy on IPF. D) Heterogeneity of SNPs demonstrated by funnel plot. E) “Leave‐one‐out” sensitivity analysis results demonstrating by odds ratio. F) Computed tomography (CT) images showing IPF patients’ imaging characteristics. G) Clinical and pathological characteristics of cases with or without diagnosis of IPF. Statistical significance tests: gender, Pearson χ2 test; age, BMI and FVC (%), Two independent‐sample t‐test; Hcy and DLCO%, Welch's t‐test. H,I) BALF Hcy level and serum Hcy level in healthy control (n = 40) and IPF patients (n = 42). J,K) Correlation analysis of BALF Hcy level or serum Hcy level and FVC (%). L,M) Correlation analysis of BALF Hcy level or serum Hcy level and DLCO (%). N,O) Receiver operating characteristic (ROC) curve revealed prognostic value of Hcy level for IPF (cutoff value = 10.14 µmol L⁻¹) and ROC curve revealing robust diagnostic value of serum Hcy (cutoff value = 11.85 µmol L⁻¹). Data are presented as mean ± standard deviation (SD) or number (%). MR: Mendelian randomization; BMI: body mass index; BALF: bronchoalveolar lavage fluid; FVC: forced vital capacity; DLCO: diffusing capacity of the lung for carbon monoxide.

Figure 2

Bulk‐seq and scRNA‐seq analysis of human IPF samples showing weakened Hcy metabolism in IPF. A) Schematic diagram of Hcy metabolism. B) GSEA showed the enrichment of fibrosis and Hcy metabolism terms in IPF group. C,D) Heatmap of enzymes (MTR, MTRR, MTHFR, MTHFD1, CBS, BHMT) involved in Hcy metabolism and fibrosis markers (COL1A1, COL3A1, ACTA2, FN1, CDH2). Data source: GSE52463 (C), GSE199949 (D). E) Correlation analysis between the expression of fibrosis and Hcy metabolism related genes in IPF environment. F) Spatial feature plots showing the distinct expression patterns of MTRR in control and IPF samples. G) T‐distributed stochastic neighbor embedding (t‐SNE) plot showing 13 distinct clusters resulting from scRNA‐seq of cells derived from lung tissues harvested from IPF and non‐IPF group. H) Joint density plot of Hcy metabolism genes created by scCustomize R package. I) Comparison of MTRR expression in all cells between control and IPF group demonstrated by t‐SNE plot. J) Violin plot showing the expression of MTRR among different cell types. K) Violin plot with dots showing the expression of MTRR in AT2 cells in control and IPF group. L) Pseudotime trajectory analysis of lung cells (AT1: alveolar type 1 cell; AT2: alveolar type 2 cell; Endo: endothelial cell; Basal: basal cell; Club: club cell; Fibro: fibroblast). M) MTRR relative expression with the progression of pseudotime in lung cells. N) Violin plot showing MTRR expression in different sub‐type cells of AT2 in control and IPF group (data source: GSE122960, GSE190889, GSE132771).

Figure 3

Hcy metabolism is disrupted in the IPF state. A,B) Immunofluorescence costaining of COL1A1 (A) and vimentin (B) with MTRR in lung sample slices from IPF patients and normal control (n = 3 per group). Right‐hand panels display magnified areas from images indicated by dashed boxes. White arrows indicated AT2 cells with higher expression of MTRR stained into light blue. Scale bar = 100 µm. C) Quantitative analysis of mean fluorescence intensity of COL1A1, Vimentin, and MTRR (n = 5). D) Costaining of COL1A1 and vimentin with MTRR in mouse lung tissues (n = 5). Scale bar = 100 µm. E) Main Hcy metabolic enzymes (MTRR, MTR, MTHFR, MAT1A mRNA levels were quantified by qRT‐PCR (n = 5 per group). F) Lung tissues from mice treated with BLM (5.0 mg kg⁻¹ body weight) were subjected to western blotting, showing the expression level of MTRR (n = 3). G) Quantification of MTRR protein expression level (n = 3). H) MTRR in human AT2 cells treated with TGF‐β (10 ng mL⁻¹) were costained with COL1A1 and vimentin (n = 5 per group). Scale bar, 50 µm. I) Quantitative PCR analysis of MTRR, MTR, MTHFR, and MAT1A mRNA expression level in AT2 cells (n = 5). J) TGF‐β (10 ng mL⁻¹)‐treated‐AT2 cells were lysed for western blotting (n = 3 per group). K) Quantification of MTRR protein expression detected by western blot (n = 3). L) Hydroxyproline level measured in lung homogenates from BLM (5.0 mg kg⁻¹ body weight) challenged mice. M) Hcy level measured in lung homogenates from BLM (5.0 mg kg⁻¹ body weight) challenged mice. N) Correlation analysis between Hcy concentration and hydroxyproline level in mouse lung tissues. O) Hcy level measured in AT2 cell lysis after TGF‐β treatment (10 ng mL⁻¹). Data are presented as the mean ± standard error of the mean (SEM). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 by Student's t‐tests.

Figure 4

MTRR genetic editing impacted IPF progression. A) Representative axial (top row) and their corresponding coronal (bottom row) images from differently treated groups determined by micro‐computed tomography (micro‐CT) showing radiological features. Healthy lungs are black, and lungs with fibrosis were increasingly white (elevated density) (n = 5 per group). B) Masson staining of lung sections from different adenovirus treated mice. Images in the lower panels were magnified from the photomicroscopy images in the upper panels (n = 5). Upper scale bar = 1 mm; lower scale bar = 50 µm. C) Vimentin costaining with α‐SMA showing fibrosis severity. White arrows showed AT2 cells in EMT process which highly expressed vimentin and α‐SMA. Scale bar = 50 µm. D) Ashcroft score assessed the severity of pulmonary fibrosis (n = 10). E) Hcy levels (showed by fold changes) of lung homogenates in different groups were examined by Hcy test kit (n = 5 per group). F) Relative mRNA expression of fibrosis related genes (COL1A1, COL3A1, ACTA2, FN1, TGFB) by RT‐qPCR (n = 5 per group). G) Micro‐CT images showing BLM‐induced fibrosis phenotype. H) Masson trichrome staining of left lungs from mice at day 28 post‐BLM (5.0 mg kg⁻¹ body weight) or ‐saline (50 µL) administration. I) Lung fibrosis in BLM and Ad‐mMTRR, BLM, or control group was assessed by Ashcroft scoring (n = 10). J) Fold changes of Hcy level in mouse lung homogenates (n = 5 per group). K) Vimentin costaining with α‐SMA showing fibrosis severity. White arrows showed AT2 cells in EMT process which highly expressed vimentin and α‐SMA. Scale bar = 50 µm. L) Relative mRNA expression of genes (COL1A1, COL3A1, ACTA2, FN1, TGFB) by RT‐qPCR (n = 5 per group). Data are presented as the mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 by one‐way ANOVA with Tukey's multiple comparison tests (D, E, I, and J) and two‐way ANOVA with Tukey's multiple comparison tests (F and L). ns, not significant.

Figure 5

Hcy accelerates IPF progression. A) Axial and their corresponding coronal micro‐CT images were acquired after BLM administration. The lower panels show the representative 3D images drawn out from micro‐CT images, based on different tissues with varying density (n = 5). B) Survival curve showing percentage of differently treated mice over a 28 days period experiment (n = 15 per group). C,D) Lung function parameters including dynamic compliance (C _dyn) (C) and respiratory resistance (D) among different groups were compared 28 days after BLM challenge (5.0 mg kg⁻¹ body weight) (n = 5 per group). E) Masson's trichrome staining of lung sections from Hcy‐treated mice in the IPF mouse lung model. Scale bar = 1 mm. Images in the lower panels were enlarged from the photomicroscopy images in the upper panels. Scale bar = 50 µm. F) Immunofluorescence analysis of α‐SMA and vimentin expressions in lung sections. Scale bar = 100 µm. G) Ashcroft scores assessment of lung sections (n = 10). H) Hydroxyproline content of lungs from Hcy (100 mg kg⁻¹) supplemented or not mice after BLM injury (n = 5 per group). I) Quantitative real‐time PCR analysis of COL1A1, COL3A1, ACTA1, and FN1 mRNA levels in lung homogenates of BLM‐challenged (5.0 mg kg⁻¹ body weight) Hcy‐ (100 mg kg⁻¹) or saline‐treated mice (n = 5). J) Representative western blot results of fibrosis markers (FN1, COL1, COL3, α‐SM(A)). GAPDH was used as the loading control (n = 3 per group). K) Relative protein level quantified from western blot results (n = 3).Data are presented as the mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 by log‐rank (Mantel–Cox) test (B), one‐way ANOVA with Tukey's multiple comparison tests (C, D, G, H, and K) and two‐way ANOVA with Tukey's multiple comparison tests (I). ns, not significant.

Figure 6

Folate treatment attenuates fibrosis induced by BLM. A) Heatmap of folate transporters showing a significant lower folate transport in IPF lung tissues (data source: GSE110147). B) Single‐cell analysis indicating AT2 cells played the main role in folate transport and exhibited an inhibited transport of folate in IPF samples. C) Representative axial, coronal, and 3D reconstruction images determined by micro‐CT showing interstitial pathological changes. D) Percentages of surviving mice were plotted over a 28 days period post‐BLM (5.0 mg kg⁻¹ body weight) or Fol (0.5 mg kg⁻¹) administration (n = 15 at start). E,F) Lung function parameters including dynamic compliance (C _dyn) (E) and respiratory resistance (F) among different groups were compared 28 days after BLM (5.0 mg kg⁻¹ body weight) challenge or Fol treatment (0.5 mg kg⁻¹) (n = 5). G) Ashcroft score quantified fibrosis level of lung sections (n = 10). H) Hydroxyproline content of lungs from Fol‐ (0.5 mg kg⁻¹) or saline‐treated mice after BLM injury (5.0 mg kg⁻¹ body weight) (n = 5 per group). I) Masson trichrome staining of left lungs from Fol‐treated or not treated mice at day 28 post‐BLM or ‐saline administration. Scale bar = 1 mm. Images in the lower panels were magnified from the photomicroscopy images in the upper panels. Scale bar = 50 µm. J) Immunofluorescence analysis showing α‐SMA and vimentin positive AT2 cells in lung sections (indicated by white arrows). Scale bar = 100 µm. K) Fibrosis marker genes relative expression level was showed by RT‐PCR (n = 5). L,M) Representative western blots analyzing fibrosis proteins (FN1, COL1, COL3, α‐SMA) in lung samples and their densitometric values normalized to GAPDH (n = 3 per group). Data are presented as the mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 by log‐rank (Mantel–Cox) test (D), one‐way ANOVA with Tukey's multiple comparison tests (E–H, and M) and two‐way ANOVA with Tukey's multiple comparison tests (K). ns, not significant.

Figure 7

STX17 is downregulated via homocysteinylation and ubiquitination in IPF. A) Joint density plot of key autophagy genes created by scCustomize R package. B,C) Feature plots showing differences of MAP1LC3B (B) and STX17 (C) expression between control and IPF group. D,E) Representative Western blots analyzing autophagy proteins (Beclin1, p62, ULK1, LC3B) and SNARE protein component STX17 (D) and their densitometric values normalized to GAPDH (E) (n = 3). F,G) Representative western blot analyzing protein ubiquitination and homocysteinylation in lung tissues. H) Protein ubiquitination densitometric values normalized to GAPDH (n = 3). I) Protein homocysteinylation densitometric values normalized to GAPDH (n = 3). J) Immunoprecipitation of ubiquitinated and Hcy‐lated proteins and detection of STX17 protein levels by western blotting in the lung tissues. K) Ubiquitinated STX17 (STX17‐Ubq) and Hcy‐lated (STX17‐Hcy) densitometric values normalized to STX17 protein level in input samples. L,M) Representative Western blots analyzing autophagy proteins (Beclin1, p62, ULK1, LC3B) and SNARE protein component STX17 from Folate‐treated (0.5 mg kg⁻¹) or not treated BLM (5.0 mg kg⁻¹ body weight) challenged samples (L) and their densitometric values normalized to GAPDH (M) (n = 3). N,O) Representative western blot analyzing protein Ubiquitination and homocysteinylation in lung tissues. P) Protein ubiquitination densitometric values normalized to GAPDH (n = 3). Q) Protein homocysteinylation densitometric values normalized to GAPDH (n = 3). R) Immunoprecipitation of Ubq and Hcy and STX17 detection by Western blotting using lung tissue lysates as inputs. Representative Western blots are demonstrated. S) Ubiquitinated STX17 (STX17‐Ubq) and Hcy‐lated (STX17‐Hcy) densitometric values normalized to STX17 protein level in input (n = 3). Data are presented as the mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 by one‐way ANOVA with Tukey's multiple comparison tests (H, I, K, P, Q, and S) and two‐way ANOVA with Tukey's multiple comparison tests (E and M). ns, not significant.