Exercise-responsive circTSN as a potential systemic biomarker during COPD rehabilitation

Abstract

Objective: Chronic Obstructive Pulmonary Disease (COPD) features both pulmonary and systemic manifestations. While exercise is a proven therapeutic strategy, the specific non-coding RNAs mediating the systemic adaptive response remain largely unknown. This study aims to identify circTSN as a key circular RNA (circRNA) involved in exercise-induced adaptation in COPD.

Methods: We performed a prospective cohort study using peripheral blood RNA sequencing in COPD patients (n = 4) before and after a 12-week exercise intervention, and validated the expression changes with RT-qPCR (n = 18). The mechanism of circTSN (hsa-circ_0003789) was investigated using a COPD mouse model, dual-luciferase reporter assays in 293T cells, and functional in vitro experiments in BEAS-2B airway epithelial cells.

Results: circTSN expression was significantly upregulated post-exercise in both COPD patients and the mouse model, and this change was decoupled from its host gene, establishing it as an independent molecular marker. Mechanistically, circTSN was found to be cytoplasmically enriched and functioned as a molecular sponge for miR-144-3p. This axis subsequently modulated the expression of key inflammatory genes in BEAS-2B airway epithelial cells, including GATA6, IL-6, IL-1β, and TNF-α, linking the airway epithelium to systemic inflammation. Furthermore, in vivo, circTSN and miR-144-3p were inversely correlated, establishing this ceRNA axis as critical for exercise-mediated lung adaptation.

Conclusion: circTSN may serve as a molecular sensor linking changes in the pulmonary environment to the body's broader adaptive responses during exercise therapy. This discovery offers crucial mechanistic insight into how exercise benefits individuals with COPD, highlighting circTSN as a valuable target for future molecular monitoring and therapeutic strategies.

Keywords: TSN (Translin); cardiopulmonary exercise test (CPET); chronic obstructive pulmonary disease; circ TSN; exercise training; inflammation; mean response time (MRT); pulmonary rehabilitation (PR).

Figure 1

Outcomes following pulmonary rehabilitation in COPD patients. (A–C) Serum cytokine levels before (Pre-PR) and after (post-PR) the 12-week pulmonary rehabilitation. Mean differences of Δchanges are shown on the right. (D–F) Mean Residence Time (MRT) during warmup phase, Peak VO₂, and FEV1%predicted before and after PR. p-values are from paired tests. (G) scRNA-seq showing cell subset redistribution. (H) Heatmap showing the top 20 differentially expressed circRNAs (log₂FC > 1, adjusted p < 0.05) between patients who benefited from PR (aPR) and their baseline (bPR).

Figure 2

Relative expression of circ TSN and miR-144-3p in peripheral leukocytes from COPD patients and lung tissues of COPD mice. (A–C) circTSN, miR-144-3p, and linear TSN relative expression in peripheral leukocytes from COPD patients before (Pre-PR) and after pulmonary rehabilitation (Post-PR). Data were analyzed using ΔCt values. For clarity, fold change (2^−ΔΔCt) is shown in the figures. (D) The scatter plot shows post–pre changes in circTSN versus cytokine IL-1β (r = −0.50, p = 0.033). (E,F) circTSN and miR-144-3p expression in lung tissue from the elastase-induced COPD mouse model, including control, COPD, and COPD + PR (COPD mice with pulmonary rehabilitation) groups. Data are presented as mean ± SD (n = 3 per group). (H,I) RNase R digestion confirming circularity of circTSN in BEAS-2B cells. (G) External validation (GSE240656, GSE268499) reveals a reduced trend of circTSN in COPD, with no significant change in TSN mRNA.

Figure 3

circTSN functions as a molecular sponge for miR-144-3p to regulate GATA6 expression in BEAS-2B cells. (A,B) Dual-luciferase reporter assays confirming direct binding of miR-144-3p to circTSN (A) and the GATA6 3′-UTR (B). Mutation of the predicted binding sites abolished the inhibitory effect of miR-144-3p on luciferase activity. (C–J) Expression analysis of circTSN, miR-144-3p, and GATA6 in BEAS-2B cells following circTSN overexpression (OE), knockdown (si-circTSN), or miR-144-3p inhibitor. (K) Fluorescence in situ hybridization (FISH) images showing subcellular co-localization of circTSN (green) and miR-144-3p (red) in BEAS-2B cells under NC, OE, and si-circTSN conditions. Nuclei are stained with DAPI (blue). Co-localization in the cytoplasm supports their interaction.