Fig 2: IGFBP1 mediates the induction of lipolysis and NASH in Wtap-HKO mice.a Heatmap of top differentiated genes encoding secreted proteins in the livers of Wtapflox/flox and Wtap-HKO mice (Wtap-HKO fpkm > 500 and FoldChange > 4.5). Data were presented as log 10 fpkm (n = 3 for each group). b Relative Igfbp1 mRNA levels in the liver and eWAT (n = 8 for each group; Liver, P = 0.0014; eWAT, P = 0.547). c IGFBP1 protein levels in the liver and eWAT were measured by immunoblotting (n = 4 for each group). The samples were derived from the same experiment and the blots were processed in parallel. d Serum IGFBP1 protein levels were measured by ELISA (n = 11 for each group; P = 0.0011). e–p Wtap-HKO mice at 7 weeks old were intravenously injected with a control antibody (IgG) and an anti-IGFBP1 neutralizing antibody per day, respectively, for 5 days. Serum FFA levels were measured (e) (IgG, n = 7; Anti-IGFBP1, n = 8; P = 0.0455). p-HSL, HSL, p-PKA substrate, ATGL and GAPDH protein levels in eWAT were measured by immunoblotting and quantified by Image J (f) (n = 3 for each group; p-HSL/HSL, P = 0.0445; ATGL/GAPDH, P = 0.0335; p-PKA substrate/GAPDH, P = 0.0499). cAMP levels were measured by ELISA (g) (IgG, n = 7; Anti-IGFBP1, n = 8; P = 0.00998). The ADCY3, ADCY4 and ADCY6 protein levels in the eWAT were measured by immunoblotting and quantified by Image J (h) (n = 3 for each group; ADCY3, P = 0.0009; ADCY4, P = 0.0318; ADCY6, P = 0.00918). serum ALT activity were measured (i) (n = 6 for each group; P = 0.0031). Cleaved caspase 3 levels (j) (n = 3 for each group; P = 0.0012), liver TAG levels (k) (n = 6 for each group; P < 0.0001), hepatic lipid droplets (l), TUNEL-positive cells (m, n) (n = 5 for each group; P = 0.0129), and cytochemokine mRNA levels (o) (n = 5 for each group; Tnfa, P = 0.0294; Il1ß, P = 0.0133; Il6, P = 0.0267; iNos, P = 0.3452; Infg, P = 0.6684; Cd14, P = 0.0111; Csf1, P = 0.0292; Ccl2, P = 0.0172; Ccl3, P = 0.0824; Ccl5, P = 0.0025; Ccl22, P = 0.0722; Ccr2, P = 0.0969; Cxcl2, P = 0.0693; Cxcl5, P = 0.092; Cxcl10, P = 0.0747; Cx3cl1, P = 0.1067) were measured in the liver. Serum cytokine levels were measured by ELISA (p) (IgG, n = 7; Anti-IGFBP1, n = 8; TNFa, P = 0.000178; IL1ß, P = 0.0323; CCL2, P = 0.0456). For immunoblotting, the samples were derived from the same experiment and the blots were processed in parallel. n was the number of biologically independent mice. Data represent the mean ± SEM. Significance was determined by unpaired two-tailed Student’s t test analysis. *p < 0.05. **p < 0.01. Source data are provided as a Source Data file.

Fig 3: Decreased lipid accumulation due to adipose lipolysis dysfunction leads to compromised MIER1 regulation and impaired liver regeneration.a Immunoblot (HSL) in Control and Lipe-AKO adipose tissues. The liver triglycerides (TG) content (b) (n = 8, 7, 8, 6, 8 in Control group, and n = 8, 6, 6, 6, 8 in Lipe-AKO group, at different time points after PHx), liver/body weight ratio (%) (c) (n = 9, 8, 10, 10, 8 in Control group, and n = 9, 9, 8, 8, 9 in Lipe-AKO group, at different time points after PHx), immunoblots of liver MIER1 and EIF2S1 phosphorylation (d) in Control or Lipe-AKO animals before or after surgery. e Polysome profiles and qRT-PCR analysis of distributions of indicated transcripts in liver tissues collected before or 24 h after hepatectomy from Lipe-AKO animals. n = 3. The liver TG content (f) (Control, n = 10; Lipe-AKO, n = 7; Lipe-AKO; Mier1 sgRNA, n = 9), liver/body weight ratio (%) (g) (Control, n = 10; Lipe-AKO, n = 7; Lipe-AKO; Mier1 sgRNA, n = 9), liver Ki-67 immunostaining (h) (Control, n = 10; Lipe-AKO, n = 7; Lipe-AKO; Mier1 sgRNA, n = 9), and liver immunoblots (PCNA, Cyclin A2 and GAPDH) (i) in Control, Lipe-AKO or Lipe-AKO; Mier1 sgRNA animals at 36 h after hepatectomy. Values represent means with SEM. P values were assessed by unpaired, two-tailed Student’s t test (e), Two-Way ANOVA with post hoc Šídák’s multiple-comparison tests (b, c), or One-Way ANOVA with Dunnett’s multiple comparisons test (f, g, h). The experiments were repeated for three times with similar results (d, i). The average values of the quantification of western blots were indicated (d, i). Source data are provided as a Source Data file.

Fig 4: Functional analysis of CD8+ T cells from VAT between cachexia and non-cachexia patients.a Pseudo-time ordered CD8+ T cells from VAT overlying the inferred cell trajectories by using Monocle. Pseudo-time was depicted from dark to light blue (upper). Pseudo-time ordered cells overlying the cell trajectories were labeled by indicated CD8+ T-cell clusters (lower). b Pseudo-time ordered cells classified by the cell origin. c Volcano plot showing the DEGs of CD8+ T cells in VAT from patients with or without cachexia. Red dots represented the selected genes of interest with significant upregulation during CAC, whereas blue dots were downregulated genes. d Representative images of the immunohistochemistry analyses for GZMB in greater omentum (VAT) sections for another 8 patients with gastric cancer. Positive staining was pointed with arrowhead lines. Scale bars: 50 μm. e Scatter plot showing the negative correlation between the mean diameters of adipocytes in each section and the number of GZMB-positive cells. Linear regression and Pearson’s correlation test with 95% CI were performed and the line of the best fit was shown in the plot. f Bar plot showing the inferred percentage of cytotoxic CD8+ T cells in all immune cells for both cancer with weight stable (CWS) and cancer with weight loss (CWL) group. P-value was determined by Student’s t-test. *P < 0.05. g Heatmap plot depicting the row scaled expression values of indicated pro-catabolic cytokines within each cell type from VAT, separated by the disease condition. h Representative images of adipocytes co-cultured with murine primary splenic CD8+ T cells activated in vitro for 24 h. Scale bars: 25 μm. i Density plot showing the distribution of diameters of droplets within adipocytes in each group. The diameters were measured from three randomly selected field for each well, at 40× magnification (n = 3/group). j Supernatant glycerol levels of adipocytes under each co-culturing condition, normalized by the glycerol concentrations in medium under co-culturing with pre-stimulated CD8+ T cells. Bars indicate means ± SEM. P-values were determined by one-way ANOVA and post hoc Dunnett’s multiple-comparison correction tests. k Western blot analysis of the phosphorylation of HSL and the expression of ATGL in each co-culturing condition (n = 3/group). l Quantification of results from the penal k. Bars indicate means ± SEM. P-values were determined by one-way ANOVA with post hoc Dunnett’s multiple-comparison correction tests.

Fig 5: Relative mRNA and protein expression in abdominal fat of Pekin ducks in Met-deficient and Met-adequate groups. Results are presented as means with plus error bars (standard deviation). Results are presented as means with plus error bars (standard deviation). Differences were assessed by Student’s t test (n = 6) and denoted as follows: * P < 0.05; ** P < 0.01; *** P < 0.001. Met-D, Met-deficient; Met-A, Met-adequate; A-FABP (FABP4), Fatty acid-binding protein, adipose; ACBP, Acyl-CoA-binding protein; FATP5, Fatty acid transport protein 5; LPL, Lipoprotein lipase; ATGL, Adipose triacylglyceride lipase; FAT/CD36, Fatty acid translocase; HSL, Hormone-sensitive lipase; p-HSL-S563, Phosphor-hormone-sensitive lipase at Ser563