FOXO1 and cell cold adaptation.

a Stacked bar plots showing genomic annotation of chromatin accessible regions in different groups. b Binding motifs of transcription factors (TF) enriched in H1 cells following 4-h incubation at 4 °C. c Up: Immunoblots of FOXO1 proteins in cytosolic (C.) and nuclear (N.) fractions, conditions annotated; down: FOXO1 protein levels (n = 5 experiments). d Confocal images of FOXO1, pluripotency marker OCT4 and nuclear DNA (DAPI) staining in human embryonic stem cells (H1 ESCs) at indicated conditions (n = 10 experiments). e Left: Stimulated emission depletion (STED) micrographs detailing FOXO1, Propidium Iodide (PI; to stain dead cells) and DAPI signals in H1 cells at indicated conditions; right: Nuc F_FOXO1 versus Cyto F_FOXO1, or versus F_PI (n = 7 and 8 experiments for DMSO and FOXO1-i, respectively); FOXO1-i, FOXO1 inhibitor. f Confocal images of FOXO1 and neuronal marker TUBB3 in TLGS iPSC-derived neurons at indicated conditions. g Left: STED images showing FOXO1 proteins and nuclear DNA (DAPI-stained) in TLGS iPSC-neurons at indicated conditions; right: the mean intensity of FOXO1 fluorescence (F_FOXO1) in the nucleus (Nuc) versus that in the whole cell (Cell) or the cytosol (Cyto) was analyzed (from left to right, n = 10, 10, 9 images from 5 experiments). Data are shown as mean and SEM. Statistics: two-tailed Student’s t test (c, e) and one-way ANOVA followed by Tukey’s test (g). Scale bars: 20 μm (d, f) and 2.5 μm (e, g). Source data are provided as a Source Data file.

Temperature-induced differential FOXO1-DNA binding and transcriptomic changes in H1 ESCs.

a Left: fold change vs. -log₁₀ FDR-corrected Q-value for differential FOXO1-binding at 37 °C or 4 °C 4 h in H1 ESCs revealed by CUT&Tag-seq; right: IGV (Integrative Genomics Viewer) snapshots showing FOXO1-binding signals around the DHX9 and EP300 loci (right); kb, kilobase pairs. b Line plots summarizing the mean of FOXO1 CUT&Tag signals within 3 kb up- and down-stream of gene transcription start sites (TSS) in H1 ESCs at 37 °C and 4 °C 4 h; kb, kilobase pairs. c Venn diagram indicating the overlap between differentially expressed genes (DEGs) upon rewarming and 4 °C-induced FOXO1 differentially bound genes in H1 ESCs. d Enrichment analysis on the overlapped genes from (c); left: Enrichment analysis on genes that had higher levels of transcripts at 4 °C 4 h compared to 4 °C 4 h_37 °C 2 h (rewarmed); right: Enrichment analysis on genes that had higher levels of transcripts at the rewarmed stage compared to 4 °C 4 h. P values in (d) were calculated in Metascape.

Temperature-mediated FOXO1 transport dependent on RANBP2, XPO1 and IPO7 SUMOylation.

a Left: Confocal images of H1 ESCs with annotated treatments; FOXO1, DAPI were co-stained with RANBP2, Importin-7 (IPO7) or Exportin-1 (XPO1); right: Nuclear F_FOXO1 versus cytosolic or whole-cell F_FOXO1 (n = 10, 10 and 12 images from 5 experiments for RANBP2, IPO7 and XPO1 RNA interference, respectively). b Confocal images of proximity ligation assay (PLA) on proteins of interest in H1 ESCs at indicated conditions. c Average counts of PLA puncta per cell from (b) (from left to right, n = 4, 4, 5, 7, 5 and 5 images from 3 experiments). d Up: immunoblots of XPO1, IPO7 and FOXO1 in total protein extracts (Input) and FOXO1 immunoprecipitated (IP-ed) fractions from H1 cells at annotated conditions; down: signal intensities normalized (n = 4 experiments). e Left: confocal images of PLA on proteins of interest in H1 ESCs at indicated conditions; right: average counts of PLA puncta per cell (from left to right, n = 4, 6, 6 and 6 experiments). f Up: immunoblots of XPO1 and IPO7 in Input and SUMO1 IP-ed fractions from H1 cells at annotated conditions; down: signal intensities normalized (n = 5 experiments). g Left: confocal images of PLA on FOXO1 interaction with endogenous IPO7 (n = 13 images from 3 experiments), or with overexpressed, HA-tagged IPO7 with a K-R mutation on its SUMOylation site in ARPE-19(WT) cells (n = 12 images from 3 experiments); WT, wild-type; right: average counts of PLA puncta per cell. Data are shown as mean and SEM. Statistics: one-way ANOVA followed by Tukey’s test (a, g) and two-tailed Student’s t test (c−f). Scale bars: 20 μm (a, b, e) and 2.5 μm (g). Source data are provided as a Source Data file.

FOXO1 nuclear export determined by FOXO1 SUMO-interacting motif (SIM) and XPO1 SUMOylation

a Left: confocal images of FOXO1 and DAPI in ARPE-19(WT) and ARPE-19(FOXO1-minus SIM) at annotated conditions; right: F_FOXO1 ratio analyzed (from left to right, n = 14, 12, 10 and 10 images from 5 experiments). b Up: Input and FOXO1 IP-ed fractions from ARPE-19(WT) and ARPE-19(FOXO1-minus SIM) cells at annotated conditions were immunoblotted against XPO1, IPO7 and FOXO1; down: signal intensities normalized (n = 5 experiments). c Left: confocal images of PLA on FOXO1 interaction with endogenous XPO1, or with overexpressed, HA-tagged XPO1 with both SUMOylation sites replaced with an arginine residue (K-R mutation) in ARPE-19(WT) and ARPE-19(FOXO1-minus SIM) cells at 37 °C (from left to right, n = 9, 7, 11 and 11 images from 3 experiments); right: average counts of PLA puncta per cell. d Confocal images of FLAG tag and DAPI staining in ARPE-19(WT) cells overexpressing FLAG-tagged, FOXO1 or mutant FOXO1 deleted of both NLS and NES domains (FOXO1-minus NLS/NES; in-frame deletion of amino acids 245-274 and 371-380) at annotated conditions (n = 3 experiments); NLS, nuclear localization sequence; NES, nuclear export signal. e Left: confocal images of PLA on indicated interacting pairs in ARPE-19(WT) cells overexpressing FLAG-tagged, FOXO1 or FOXO1-minus NLS/NES at annotated conditions; right: average counts of PLA puncta per cell (from left to right, n = 6, 8, 5, 7, 7, 7, 7 and 8 images from 3 experiments). Data are shown as mean and SEM. Statistics: one-way ANOVA followed by Tukey’s test (a−c) and two-tailed Student’s t test (e). Scale bars: 20 μm. Source data are provided as a Source Data file.

FOXO1 SIM as a Nuclear Exit Signal for FOXO3.

a Schema of FOXO3 mutant construct that contains the FOXO1 SIM. b Confocal images of FOXO3 and DAPI staining in ARPE-19(WT) cells at indicated conditions (n = 6 experiments). c Confocal images of FOXO3, HA tag and DAPI staining in ARPE-19 cells overexpressing FOXO3-plus SIM-HA at indicated conditions. dF_FOXO3 ratio (from left to right, n = 10, 14, 15 and 15 images from 5 experiments). e A model for temperature and FOXO1 SIM-mediated FOXO1 transportation. The illustration was based on a template created with BioRender.com. Data are shown as mean and SEM. Statistics: two-tailed Student’s t test (d). Scale bars: 20 μm (b, c). Source data are provided as a Source Data file.

Activating FOXO1 nuclear entry to enhance cold survival in obese pre-diabetic mice.

a Confocal images of XPO1, FOXO1 and DAPI staining in obese mice (12-14 mo) pancreatic sections from indicated conditions (n = 5 mice from 5 experiments); KPT-330, XPO1 inhibitor. b Confocal images of FOXO1 and DAPI in obese mice heart, liver and skeletal muscle sections from indicated conditions (n = 5 mice from 5 experiments). c Survival rate of obese mice at indicated conditions; KPT-330, XPO1 inhibitor; DMSO is the solvent control. d Left: from a representative mouse energy expenditure experiment, heat and oxygen consumption (VO₂) of obese mice with an intraperitoneal injection of KPT-330 (3 mg/kg) (n = 7) or DMSO (n = 5) prior to 4 °C exposure; right: average heat and VO₂ under basal and cold-exposed conditions. e Daily changes of mouse body weight during 4 °C exposure (KPT-330, n = 9; DMSO, n = 5). f Mouse body mass composition following 10-day cold exposure at indicated conditions (KPT-330, n = 8; DMSO, n = 6). g Venn diagram showing the numbers of overlapped DEGs among indicated tissues from obese mice after 6 h of 4 °C exposure; also see Supplementary Data 3. h Enrichment analysis on DEGs upregulated in brown adipose tissues (BAT), epididymal white adipose tissues (eWAT), skeletal muscles and liver at indicated conditions. Data are shown as mean and SEM. Statistics: two-tailed Student’s t test (d−f); P values in (h) were calculated in Metascape. Scale bars: 20 μm (a, b). Source data are provided as a Source Data file.

Mouse pancreatic islet long-term storage and transplantation.

a Left: confocal images of FOXO1, PI and DAPI staining in mouse islets at indicated conditions; right: statistics on cell death percentage (from left to right, n = 14, 11 and 8 images from 3 experiments); UW, University of Wisconsin Solution; HS, basal hibernation solution used in cell cold exposure experiments. b Left: dead cell percentage in cold-stored mouse islets at annotated conditions (from left to right, n = 14, 9, 6, 8 and 16 images from 3 experiments); 2-D08 and ginkgolic acid are SUMOylation inhibitors added to HS; Ins1-Foxo1-KD, pancreatic β cell-specific knock-down of Foxo1; right: normalized levels of secreted and intracellular insulin in rewarmed islets as indicated (n = 3 experiments). Normalized islet oxygen consumption rate (OCR) from indicated conditions, as an assessment of islet quality following 24-h (fresh and UW: n = 7 experiments; KPT-330: n = 8 experiments) (c) or 2-h (n = 6 and 3 experiments for 7 and 14 d, respectively) (d) rewarming in vitro; K&P, KPT-330 and protease inhibitors. Glucose-stimulated insulin secretion assays with mouse islets from indicated conditions following 24-h (n = 6 experiments) (e) or 2-h (n = 5 experiments) (f) rewarming in vitro. g Blood glucose levels in Streptozotocin (STZ)-treated diabetic mice with cold-stored islets at indicated conditions (mice with sham transplantation: n = 3; transplanted with fresh islets: n = 5; n = 3 each for UW 7 or 10 d; n = 5 and 6 for HS + K&P 10 d and 14 d, respectively). h Representative image showing the transplanted islets (arrowheads), and confocal images of insulin, Glucagon and DAPI staining on kidney sections of the diabetic mice transplanted with islets stored at 4 °C in HS + K&P for 14 days (n = 6 mice), or with islets stored in UW solution at 4 °C for 5 days (n = 3 mice). Data are shown as mean and SEM. Statistics: one-way ANOVA followed by Tukey’s test (a, b) and two-tailed Student’s t test (b−f). Scale bars: 20 μm (a, f). Source data are provided as a Source Data file.

Static cold storage of human pancreatic tissues and islets.

a Confocal images of FOXO1, insulin and DAPI staining in human pancreatic sections at indicated conditions (n = 9 donors). b Normalized levels of total insulin in human pancreatic tissues at annotated cold storage conditions (n = 3 samples from each of 9 donors). c GSIS assays with islets from human at indicated conditions (fresh ctrl: n = 6 donors; UW and K&P: n = 7 donors, respectively); K&P, KPT-330 and protease inhibitors. Data are shown as mean and SEM. Statistics: two-tailed Student’s t test (b, c). Scale bars: 20 μm (a). Source data are provided as a Source Data file.