Fig 2: Reintroduction of Nrf2 rescues the Mst1/2-null phenotype. a Flow cytometry analyzing of ROS levels in Mst1fl/flMst2fl/fl (WT) BMDMs, Mst1fl/flMst2fl/fl Lyz-Cre (DKO) BMDMs, or DKO BMDMs infected with adenovirus expressing GFP (Ad-GFP) or Nrf2 (Ad-Nrf2), with the CellRox dye and quantification of the relative fluorescence intensity in the indicated cell samples shown in the right panel. b–e RT-qPCR analysis of the mRNA levels of the antioxidant genes Ho-1 and Nqo-1 (b), fluorescence microscopy of p-H2A.X (red), DAPI-stained nuclei (blue) and GFP+ cells (green) (c), Annexin V/DAPI staining (d) and quantification of Annexin V+DAPI- cells (e) in WT BMDMs, DKO BMDMs, or DKO BMDMs infected with Ad-GFP or Ad-Nrf2, followed with or without H2O2 treatment as indicated. Scale bars, 20 µm. f Relative fluorescence intensities of telomere FISH of GFP+ peritoneal macrophages isolated from WT mice transplanted for 6 months with DKO bone marrow cells, which were transduced with retrovirus expressing Nrf2, or control GFP, respectively. g A proposed working model for kinases Mst1/2 sense ROS and maintain cellular redox balance by modulating the stability of Nrf2. Phagosomal or mitochondrial ROS release attracts Mst1/2 to cap around phagosome or mitochondrion from the cytosol and activates Mst1/2; Mst1/2 phosphorylate Keap1 to stabilize Nrf2 and regulate the expression of antioxidant enzymes to protect cell against oxidative damage. Data are from one experiment representative of three independent experiments with similar results. ns, not significant (P > 0.05); *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, compared with the control sample (Student’s t test). Data are from one experiment representative of three independent experiments with similar results (mean and s.d. of n = 3 (a, b, e); mean and s.e.m. of n = 50 (f); biological replicates (a, e); and experimental replicates (b, f))

Fig 3: Loss of Mst1/2 promotes ageing in macrophages. a RT-qPCR analysis of various Tlrs genes on peritoneal macrophages isolated from Mst1fl/flMst2fl/fl (WT) or Mst1fl/flMst2fl/fl Lyz-Cre (DKO) mice with indicated age. b Proliferation assay of CFSE-labeled CD4+ OT-II T cells co-cultured with WT and DKO BMDMs in presence of OVA protein. c Representative transmission electron micrographs (TEMs) of lipofuscin granules within peritoneal macrophages isolated from 12-month-old WT and DKO mice. ×25 magnification of areas outlined in the main images as indicated. Scale bars, 2 µm. d, e RT-qPCR analysis of Mst1 and Mst2 (d), and immunoblot analysis of Mst1, Mst2 and p-Mob (e) in peritoneal macrophages isolated from WT mice with indicated age. f–h The relative telomere length (T/S ratio) (f), representative fluorescence microscopy images of telomere FISH analysis (red) and nuclei (blue) (g), and relative fluorescence intensity of telomere FISH (h) in peritoneal macrophages isolated from 2-, 8-, or 12-month-old WT and DKO mice. Scale bars, 10 µm. i Relative fluorescence intensities of telomere FISH in peritoneal macrophages isolated from WT and DKO mice with or without NAC supplementation in drinking water for 7 months. ns, not significant (P > 0.05); **P < 0.01, ***P < 0.001, ****P < 0.0001, compared as indicated (Student’s t test). Data are from one experiment representative of three independent experiments with similar results (mean and s.d. of n = 3 (a, d, experimental replicates); n = 5 (f, biological replicates); mean and s.e.m. of n = 50 (h, i, experimental replicates))

Fig 4: Mst1/2-mediated Keap1 phosphorylation disrupts the Keap1-Nrf2 complex. a Immunoblot analysis of Cul3 and Keap1 in total lysates (bottom) and anti-HA IP (top) of 293T cells expressing various combinations of Flag-tagged Cul3, HA-tagged Keap1WT, Keap14A or Keap14D. b Immunoblot analysis of keap1 levels in total lysates (bottom) and anti-Flag IP (top) of 293T cells expressing various combinations of HA-tagged or Flag-tagged wild-type Keap1WT, mutated Keap14A or Keap14D. c, d Immunoblot analysis of dimerization of Keap1 in total lysates of 293T cells expressing Flag-tagged WT or mutated Keap1 (c) or various combinations of HA-tagged WT or kinase inactivated Mst2K/R with Flag-tagged Keap1 (d) using native or regular PAGE. e SIM of HeLa cells cotransfected to express HA-tagged Keap14A (red) with Myc-tagged Keap14A, Keap1WT or Keap14D (green); ×64 magnification of areas outlined in the main images are shown next to the main images. Scale bars, 20 µm. f Quantification of the percentage of foci (larger than 0.4 µM) on e.n.s. no significant, *P < 0.05, **P < 0.01, ***P < 0.001 (Student’s t test, mean and s.d. of n = 7 images). g Immunoblot analysis Nrf2 and Keap1 in total lysates (bottom) and anti-Flag IP (top) of 293T cells expressing various combinations of Flag-tagged Nrf2 and HA-tagged Keap1WT, Keap14A, or Keap14D in present of MG132. h Immunoblot analysis of Keap1 and Nrf2 in anti-IgG or anti-Nrf2 IP (top) and total lysates (bottom) of Mst1fl/flMst2fl/fl (WT) and Mst1fl/flMst2fl/fl Lyz-Cre (DKO) BMDMs treated with antimycin A (AA) for indicated times, probed with antibodies to Keap1, Nrf2, Mst1, Mst2, and GAPDH. i Immunoassay of the ubiquitination of Nrf2 in anti-IgG or anti-Nrf2 IP (top) and total lysates (bottom) of BMDMs infected with adenovirus expressing (Ad-) Keap1WT, Keap14A, Keap14D or control vector. j, k Immunoblot analysis of NQO1, HO-1, Nrf2, and Keap1 (j) or RT-qPCR analysis of Nqo1 and Ho-1 (k) in Raw264.7 cells expressing Flag-tagged Keap1WT, Keap14A, or Keap14D infected with E. coli (MOI: 100). ns, not significant (P > 0.05); *P < 0.05, **P < 0.01, ***P < 0.001, compared with the control sample (Student’s t test) (mean and s.d. of n = 3, experimental replicates). Data are from one experiment representative of three independent experiments with similar results

Fig 5: Mst1/2 sense phagosomal and mitochondrial ROS. a Flow cytometry analysis of cellular ROS levels in WT BMDMs pretreated with or without NAC, infected with E. coli (MOI: 100) and stained with CellRox for 30 min. b SIM of Mst1 staining (red) and DAPI-stained nuclei (blue) in WT BMDMs infected with GFP-E. coli (green) treated with or without NAC as indicated; ×25 magnification of areas outlined in the main images are shown next to the main images. Scale bars, 20 µm. c Immunoblot analysis of phosphorylated (p)-Mob1, Mob1, p-Mst1/2, Mst1, Mst2, and GAPDH in BMDMs pretreated with PBS or NAC (5 µM) and then infected with E. coli (MOI: 100). d Immunoblot analysis of Mst1, Mst2, ß-actin and Hsp60 in the cytoplasmic (Cyto) and mitochondrial (Mito) fractions of NAC-treated or non-treated BMDMs infected with E. coli (MOI: 100) for the indicated time. e SIM of Mst1 staining (red), Tomm20 (green) and DAPI-stained nuclei (blue) in WT BMDMs treated with DMSO or antimycin A, with or without NAC pretreatment, as indicated; ×49 magnification of areas outlined in the main images are shown next to the main images. Scale bars, 20 µm. f, g Immunoblot analysis of Mst1, Mst2, ß-actin, and Hsp60 in the cytoplasmic (Cyto) and mitochondrial (Mito) fractions of WT BMDMs treated with antimycin A (f) or rotenone (g), with or without NAC pretreatment, for the indicated time. h, i Immunoblot analysis of p-Mob1, Mob1, p-Mst1/2, Mst1, Mst2, and GAPDH in BMDMs treated with antimycin A (h) or rotenone (i) for the indicated time or with antimycin A (h) or rotenone (i) at the indicated dose for 30 min. j, k Immunoblot analysis of p-Mob1, Mob1, p-Mst1/2, Mst1, Mst2, and GAPDH in PBS or NAC pretreated BMDMs followed with 10 µM antimycin A (j) or 3 µM rotenone (k) treatment. l Flow cytometry analysis of mitochondrial ROS levels in THP1 cells treated with 10 µM antimycin A or 3 µM rotenone followed with or without Mst1/2 kinases inhibitor, XMU-MP-1, treatment for indicated times. Data are from one experiment representative of three independent experiments with similar results