Fig 2: Glufosinate promotes a rewiring of IL10 macrophages toward a M1‐like phenotype through HIF1α stabilization and abolishes immunosuppressive effect of hypoxia A–DEvaluation of M1 markers in macrophages by real‐time PCR. Fold change of TNFA, CD80, CXCL9, and CXCL10 mRNA in IL10, MSO‐ and glufosinate (10 and 20 μM)‐stimulated IL10 macrophages (n = 3).E–HEvaluation of M2 markers in macrophages by real‐time PCR. Fold change of MRC1, MSR1, CCL17, and CCL18 mRNA in IL10, MSO‐, and glufosinate (10 and 20 μM)‐stimulated IL10 macrophages (n = 3).I–MEvaluation of M1 markers in macrophages following HIF1α inhibition by real‐time PCR. Fold change of TNFA, CXCL10, CD86, CD80, and CXCL9 mRNA in IL10 alone or glufosinate (10 and 20 μM)‐ and acriflavine/glufosinate (10 and 20 μM)‐IL10 macrophages (n = 3).N–PEvaluation of M2 markers in macrophages following HIF1α inhibition by real‐time PCR. Fold change of MRC1, MSR1, and CCL18 mRNA in IL10, glufosinate (10 and 20 μM)‐treated, and acriflavine/glufosinate (10 and 20 μM)‐treated IL10 macrophages (n = 3).QQuantification of cancer cell motility through a matrigel‐coated membrane in presence of IL10, MSO/IL10, and glufosinate (10 and 20 μM)‐IL10-treated macrophages after 24 h of incubation (n = 6).REvaluation of the capillary network formation in presence of macrophages pretreated for 24 h with IL10 or MSO/IL10, and glufosinate (10 and 20 μM)/IL10 after 4 h of incubation with HUVEC cells (n = 6).SCD8+ T‐cell suppression by macrophages treated with IL10 or MSO/IL10, and glufosinate (10 and 20 μM)/IL10 for 24 h (n = 4). Proliferation was evaluated by reading radioactivity as cpm (counts per minute), after incubation with 1 μCi/well tritiated thymidine. The proliferation of T cells cultured without macrophages was used as control.TCD8+ T‐cell recruitment in a transwell system by macrophages treated with IL10 or MSO/IL10, and glufosinate (10 and 20 μM)/IL10 for 24 h versus macrophages treated with LPS/IFNγ; the migration of T cells cultured without macrophages (Mφ‐) in presence of CXCL10 was used as positive control (n = 4).URepresentative image of Western blotting analysis of HIF1α, REDD1, 4E‐BP1, S6, and P70S6K (in their phosphorylated and unphosphorylated form) to test mTOR activation in normoxic (NRX) and hypoxic (HYP) IL10 macrophages treated with glufosinate (20 μM), rapamycin (20 nM) and a combination of both (n = 3).VCD8+ T‐cell suppression by normoxic and hypoxic IL10 macrophages treated with glufosinate (20 μM), rapamycin, and a combination of both for 24 h (n = 4). Proliferation was evaluated by reading radioactivity as cpm (counts per minute), after incubation with 1 μCi/well tritiated thymidine. The proliferation of T cells cultured without macrophages was used as control.Data are reported as means ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 ****P < 0.0001.Exact P values and statistical tests are reported for each experiment in Appendix Table S2. Source data are available online for this figure.

Fig 3: Glufosinate treatment induces an M1‐like phenotype and inhibits metastasis AExperimental plan. Mice were subcutaneously injected with LLC cells. When the tumor reached 80–100 mm3 size, mice were treated by gavage with glufosinate (10, 20, and 40 mg/kg) for 14 days.B, CQuantification of glufosinate (B) and glutamine (GLN) (C) levels in wet tumor tissues of vehicle and glufosinate‐treated mice (n = 8), by liquid chromatography–mass spectrometry (LC‐MS).D, ESubcutaneous LLC tumor growth over the time (D) and number of lung metastatic nodules in vehicle and glufosinate (10 and 20 mg/kg)‐treated mice (E) (pool of 2 independent experiments; 10 mice per condition in total).F–HQuantification and representative images of F4/80+ CD11c+ cells in the lung metastasis (F), F4/80+ CD206+ (G) and F4/80 CD11c+ (H) cells infiltration in tumors of vehicle and glufosinate‐treated mice (n = 6).I, JRT–PCR quantification of M2 (Ccl22, Arg1, and Ccl17) (I) and M1 (Tnfa, Cxcl9, Nos2, Cd86, and Cd80) (J) markers in vehicle and glufosinate‐treated mice (n = 4).In each histological quantification, n represents the number of animals. Six images per tumor were analyzed. Scale bars: 20 μm (F), 50 μm (G‐H).Data are reported as means ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 ****P < 0.0001.Exact P values and statistical tests are reported for each experiment in Appendix Table S2. Source data are available online for this figure.

Fig 4: STAT3 restored the regulatory role of miR-140 in RA FLSs. STAT3 expression levels were detected in cells using (A) western blotting and (B) RT-qPCR following miR-140 and STAT3 transfections. All groups were normalized to the control group (100%). The proliferative ability of cells expressing STAT3 and miR-140, miR-140 alone, or control was measured at 48 h post-transfection using the (C) BrdU incorporation assays and (D) MTT assays. (E) The apoptotic rate of RA FLSs in each group was determined by annexin V-FITC/PI flow cytometry. Early apoptotic cells and later apoptotic cells are indicated in the top right and bottom right quadrants, respectively, in each plot. Quantitative analysis of the apoptotic rate of RA FLSs in two groups is displayed in the right panel. (F) Western blotting was carried out to assess the expression levels of pro-inflammatory cytokines regulated by the miR-140 and STAT3 plasmid transfections. (G) RT-qPCR was performed to examine the gene expression of cytokines from the RA FLSs after transfection. All groups were normalized to the control group (100%). Data represents the mean ± SD. *P<0.05, **P<0.01 vs. the control group; #P<0.05, ##P<0.01 vs. the miR-140 group. FLS, fibroblast-like synoviocyte; IL, interleukin; miR, microRNA; PI, propidium iodide; RA, rheumatoid arthritis; RT-qPCR, reverse transcription-quantitative PCR; TNF, tumor necrosis factor.

Fig 5: TUG1 overexpression led to decreased TNFA mRNA expression. The interaction between TUG1 and miR-27a was further explored by transfecting TUG1 expression vector or miR-27a mimic into AC16 cells. Overexpression of TUG1 and miR-27a was confirmed by qPCR at 24 hours post-transfection (a). The effect of TUG1 or miR-27a overexpression on the expression of each other was also analyzed by qPCR at 24 hours post-transfection (b). Effects of TUG1 and miR-27a on the expression of TNFA in AC16 cells were analyzed by qPCR and western blot at the mRNA (c) and protein (d) levels, respectively. Experiments were repeated three times and mean values are shown; error bars represent standard deviations. *P < 0.05. TNFA, tumor necrosis factor-α gene; TUG1, taurine-upregulated gene 1; AC16, human cardiomyocyte cell line; qPCR, real-time quantitative PCR.