Matrix metalloproteinases (MMPs) are proteolytic enzymes involved in extracellular matrix remodeling and have been implicated in the pathogenesis of interstitial lung disease (ILD). However, the causal relationship between circulating MMPs and ILD risk remains uncertain, and the potential mediating role of immune cell phenotypes in this pathway has not been systematically investigated. This study aimed to evaluate the causal effects of six circulating MMPs (MMP1, MMP2, MMP3, MMP7, MMP9, and MMP12) on ILD risk using Mendelian randomization (MR) and to determine whether specific immune cell phenotypes mediate this relationship.

We conducted a comprehensive two-sample MR and mediation analysis utilizing publicly available genome-wide association study (GWAS) summary statistics from European-ancestry populations. Genetic instruments for six MMPs were selected at genome-wide significance (P < 5 × 10⁻⁸), and for 731 immune cell traits at a threshold of P < 1 × 10⁻⁵, followed by clumping to ensure independence (r² < 0.001 within 10,000 kb). The analysis proceeded through three sequential phases: (1) bidirectional MR to establish causal relationships between MMPs and ILD, (2) forward MR to identify immune cell phenotypes causally associated with ILD, and (3) mediation MR to quantify the proportion of MMP9 effect mediated through significant immune cells. The inverse-variance weighted (IVW) method served as the primary analysis, supplemented by MR-Egger, weighted median, simple mode, and weighted mode methods. Comprehensive sensitivity analyses included Cochran's Q test for heterogeneity, MR-Egger intercept test for horizontal pleiotropy, and MR-PRESSO for outlier detection. Mediation proportion was calculated using the product-of-coefficients method.

Among the six MMPs examined, genetically predicted MMP9 levels demonstrated a significant positive causal effect on ILD risk (IVW odds ratio [OR] = 1.034, 95% confidence interval [CI]: 1.006–1.064, p = 0.018), whereas reverse MR analysis showed no causal effect of ILD on MMP9 (OR = 1.017, 95% CI: 0.872–1.187, p = 0.826), supporting unidirectional causality. No significant associations were observed for MMP1, MMP2, MMP3, MMP7, or MMP12 (all p > 0.05). From 731 immune cell phenotypes, initial screening identified 35 traits nominally associated with ILD (p < 0.05). After rigorous filtering requiring consistent effect direction across five MR methods and absence of horizontal pleiotropy (MR-Egger intercept p > 0.05), eight immune traits remained significantly associated with ILD. These included effector memory double-negative T cell (EM DNT) absolute count (OR = 1.170, 95% CI: 1.063–1.287, p = 0.001), naive CD8 bright T cell absolute count (OR = 0.903, p = 0.001), percentage of naive CD8 bright T cells (OR = 0.930, p = 0.007), CD14⁺CD16⁺ monocyte absolute count (OR = 0.912, p = 0.007), percentage of CD8 dim natural killer T cells (OR = 1.133, p = 0.004), herpesvirus entry mediator on effector memory CD4⁺ T cells (OR = 0.928, p < 0.001), HVEM on central memory CD8 bright T cells (OR = 0.940, p = 0.003), and programmed death ligand 1 on CD14⁻CD16⁻ cells (OR = 1.059, p = 0.006). Subsequent MR analysis evaluating MMP9 as exposure and these eight immune traits as outcomes revealed that MMP9 significantly increased the abundance of EM DNT (IVW OR = 1.041, 95% CI: 1.009–1.074, p = 0.012), while no significant effects were observed on other immune phenotypes. Mediation analysis demonstrated that EM DNT mediated 18.5% of the total effect of MMP9 on ILD (mediated effect β₁₂ = 0.006; total effect β₀ = 0.034; direct effect = 0.028). Sensitivity analyses confirmed the robustness of these findings, with no evidence of significant heterogeneity (Cochran's Q p > 0.05 for all analyses) or horizontal pleiotropy (MR-Egger intercept p > 0.05 for all causal paths).

This study provides genetic evidence supporting a causal role of MMP9 in ILD development, with approximately one-fifth of this effect mediated through EM DNT. These findings identify MMP9 and EM DNT as potential therapeutic targets and offer novel insights into the immune-mediated mechanisms underlying ILD pathogenesis. Further experimental studies are warranted to validate these observations and elucidate the underlying biological mechanisms.