Genetic inactivation of ANGPTL4 improves glucose homeostasis and is associated with reduced risk of diabetes

Fig. 1. ANGPTL4 p.E40K associates with reduced risk of type 2 diabetes. The association between the p.E40K variant and type 2 diabetes was tested in each study using logistic or Firth logistic regression, coding genotypes according to an additive model. “Combined” effects were calculated using inverse variance weighted fixed-effects meta-analysis. For each study, the squares indicate the odds ratio and lines indicate 95% confidence intervals. The square size is proportional to the standard error of the estimate. CI confidence interval, CGPS Copenhagen General Population Studies, DECODE deCODE, DiscovEHR DiscovEHR Discovery Study, DiscovEHR DiscovEHR-30K, DiscovEHR 30K Replication Study, EINT-C EPIC interact–CoreExome, EINT-Q EPIC Interact–Quad660, ENOR EPIC Norfolk, HUNT the Nord-Trøndelag Health study, MDC Malmo Diet and Cancer Study, MGI the Michigan Genomics Initiative, TD2G/GT2D/DG combined analysis of T2D-GENES, GoT2D, and DIAGRAM studies, UKBB United Kingdom Biobank. The study populations are described in full in Supplementary Table 4 and in the Supplementary Note

Fig. 2. Plasma ANGPTL4 levels are reduced in p.G313fs carriers. ANGPTL4 plasma levels were measured in fasted serum from 86 heterozygous p.E40K, 42 heterozygous p.G313fs variant carriers, and 55 controls matched for age, sex, and body mass index. Statistics performed by unpaired t-test with Welch’s correction, comparing each variant carriers group to controls, ****p < 0.0001

Fig. 3. K40 and G313fs abolish ANGPTL4 effect on triglycerides. a Plasma triglycerides levels (4 h fasted) in C57Bl/6 mice before (Baseline) and 7 days after (Day 7) hydrodynamic delivery via tail vein injection of cDNA encoding human E40, K40, and Gly313fs ANGPTL4 variants. Control animals were injected with empty vector. b ANGPTL4 plasma levels were measured in the animals described in (a). All groups had five animals. Values are mean ± SEM. Statistics performed by two-way ANOVA with Bonferroni correction, ****p < 0.0001 vs control

Fig. 4. Angptl4−/− mice have improved glucose homeostasis. a Serum triglycerides, (b) total cholesterol, and (c) blood glucose levels in Angptl4−/− and littermate control mice on a high-fat diet for 9 weeks. d Oral glucose tolerance test and (e) insulin tolerance test in the animals described in (a–c). All groups had 9–11 animals. Values are mean ± SEM. Statistical analysis by Welch’s t-test (a) and 2-way ANOVA with Sidak’s post-test (d, e), **p < 0.001, ****p < 0.0001. The study was conducted in three different cohorts of mice, with qualitatively similar results in each replicate

Fig. 5. Angptl4−/− mice have reduced liver fat. a Body weight, (b) lean and (c) fat mass in Angptl4−/− and littermate control mice on a high-fat diet. d Liver and (e) epididymal white fat weights were evaluated in mice described in (a–c) at the time of killing. f Hepatic triglyceride levels and (g) neutral lipid staining in the livers described in (d), scale bar = 200 µm. h Respiratory exchange ratio (RER), (i) locomotor activity, (j) food intake, and (k) energy expenditure were measured during dark and light cycles in Angptl4−/− and littermate control mice on a high-fat diet. All values are mean ± SEM. Statistical analysis was conducted by Welch’s t-test; *p < 0.05; **p < 0.01. The study was conducted in two different cohorts of mice, with qualitatively similar results in each replicate