Fig 2: KCNJ2 reduced the ubiquitination degradation of HIF1a. A–B Expression of HIF1a and CA9 was detected by western blotting after KCNJ2-inhibition. C–D Expression of HIF1a and CA9 was detected by western blotting after KCNJ2-overexpression. E–F The degradation rate of HIF1a in 143B cells with empty vector and KCNJ2-overexpression. G–H Western blotting indicated that KCNJ2-inhibition markedly reduced HIF1a and CA9 expression in 143B cells, while the 26S protease inhibitor MG132 relieved this downregulation of HIF1a and CA9 that was induced by KCNJ2-inhibition. I Ubiquitination levels of HIF1a in 143B cells with empty vector and KCNJ2-overexpression. **, P < 0.01

Fig 3: KCNJ2 expression is associated with metastasis in osteosarcoma (OS). A Differentially expressed genes (DEGs) between OS cells with high and low metastatic ability in the GSE18947 cohort. B Differentially expressed genes (DEGs) between OS cells with high metastatic ability and low metastatic ability in the GSE49003 cohort. C Overlapped genes in two cohorts. D Change-fold and P value of overlapped DEGs. E Expression of KCNJ2 in OS tissues provided by patients diagnosed with stage I–II or stage III OS. F Kaplan–Meier survival analysis for OS patients with high and low KCNJ2 expression. G, H The protein levels of KCNJ2 in normal cells (hFOB1.19 and BMSC), OS cells with strong metastatic ability (KHOS, MNNG/HOS, SJSA-1, and 143B) and OS cells with low metastatic ability (MG63, Saos2, U2OS, and HOS). *, P < 0.05; **, P < 0.01

Fig 4: Kir2.1 repression alleviates inflammation triggered by pathogenic or danger signals in mouse models and human samples.a IL-1ß levels in serum from septic mice measured by ELISA. (n = 9; mean ± SEM) (b) IL-1a and IL-6 levels in serum from septic mice measured by ELISA. (n = 9, mean ± SEM) (c) Survival rates of sepsis model mice (n = 15, log-rank test [Mantel–Cox]). d IL-1ß levels in serum from septic mice measured by ELISA (Kcnj2f/f, n = 12; Lyz2-cre-Kcnj2f/f, n = 13; mean ± SEM). e Survival rates of sepsis model mice (n = 13; log-rank test [Mantel–Cox]). f IL-1ß levels in serum from septic mice measured by ELISA (DMSO, n = 8; ML133, n = 7; Kcnj2f/f, n = 12, Lyz2-cre-Kcnj2f/f, n = 11; mean ± SEM). g The expression of KCNJ2 in monocytes from healthy donors, sepsis patients, and those who recovered from sepsis with or without LPS63. Data sourced from the GEO database (GSE46955). (n = 6, 8, 8 respectively; mean ± SEM). h Heatmap showing the expression of Kir2.1 and inflammatory genes in monocytes from septic patients and those who recovered from sepsis in response to LPS63. Data sourced from the GEO database (GSE46955). (n = 8, 8 respectively; mean ± SEM). i Relative IL-1ß levels in the supernatant of cultured synovial fluid cells from two gouty patients analyzed by ELISA. Two-tailed unpaired Student’s t-test. Source data are provided as a Source Data files.

Fig 5: Kir2.1 is a critical regulator of macrophage Vm driving inflammation.a Expression profile of K+ channels in mouse bone marrow-derived macrophages (BMDMs) analyzed by RNA-seq (n = 3). The FPKM data of different genes are standardized to the Z-score by column in IBM SPSS Statistics. b–d Electrophysiological properties of freshly isolated WT (b–d) and Kir2.1-knockout (d) mouse peritoneal macrophages using patch-clamp. Thioglycolate-elicited mouse peritoneal macrophages were treated with (b) or without (c) 500 ng/ml LPS for 1 h before recording. The time-dependent current traces were recorded at a clamp voltage of –115 mV or +60 mV (red arrow, initiation of recording). The lower panel shows the I–V curves of the corresponding points for different doses of ML133 in the left panel. d Representative I–V curves of WT and Kir2.1-knockout mouse peritoneal macrophages. e Statistics of current amplitude recorded at a clamp voltage of –115 mV. Macrophages were treated and clamped as described above (Kcnj2f/f, n = 16, 18, 8, and 17 respectively; Lyz2-cre-Kcnj2f/f, n = 16, 9, 10, and 10 respectively; mean ± SEM). f Statistics of Vm recorded at 3 min after initiation of recording (Kcnj2f/f, n = 10, 7, 8, 9, 9, 8, 8, and 7, respectively; Lyz2-cre-Kcnj2f/f, n = 9, 8, 12, and 12, respectively; mean ± SEM). g, h RNA-seq analysis of inflammatory genes. GSEA analysis showing enrichment of the inflammatory response genes (g) and heatmap of inflammatory genes (h). NES, normalized enrichment score; FDR, false discovery rate. (n = 3). i Il1a, Il1b, and Tnf mRNA transcription assays. Mouse peritoneal macrophages from WT or Lyz2-cre-Kcnj2f/f mice treated with 500 ng/ml LPS for 6 h in the presence or absence of ML133 (25 µM) followed by qPCR analysis (n = 6, mean ± SEM). j Mouse peritoneal macrophages treated with or without 500 ng/ml LPS in the presence or absence of elevated [K+]e (50 mM), gramicidin (1.25 µM), ML133 (25 µM), PAP1 (Kv1.3 inhibitor, 1 µM), Tram34 (KCa3.1 inhibitor, 1 µM) for 6 h followed by qPCR analysis of mRNA transcription. Two-tailed unpaired Student’s t-test. Source data are provided as a Source Data file.