Fig. 4: Hepatic LbNOX expression enhances capacities for exercise of both intensities by enhanced redox-dependent gluconeogenesis.

a,f, Blood glucose levels of Ad-LacZ and Ad-LbNOX mice after intraperitoneal administrations of lactate (a) and glycerol (f). n = 8 (in a) and n = 7 (in f) per group; repeated measures two-way ANOVA followed by Holm–Šídák post hoc analysis (two-sided); **P < 0.0001 vs Ad-LacZ (at 30 min) (in a); **P = 0.0003 vs Ad-LacZ (at 30 min), **P = 0.0097 vs Ad-LacZ (at 60 min), **P = 0.0257 vs Ad-LacZ (at 120 min) (in f). b,g, Exercise capacities for high-intensity (b) and low-intensity (g) exercise in male Ad-LacZ and Ad-LbNOX mice were determined based on the times until exhaustion. n = 8 per group; two-tailed unpaired t-test; *P = 0.0284 (in b); **P = 0.0037 (in g). c–e,h–j Concentrations of blood glucose (c,h), plasma lactate (d,i) and plasma glycerol (e,j) in male Ad-LacZ and Ad-LbNOX mice were measured before and after 20 min of high-intensity exercise or 60 min of low-intensity exercise. n = 8 per group; repeated measures two-way ANOVA followed by Holm–Šídák post hoc analysis (two-sided); *P = 0.0391 vs control (post), ^##P = 0.0050 vs pre (within Ad-LbNOX) (in c); **P < 0.0001 vs Ad-LacZ (post), ^##P < 0.0001 vs pre (within both groups) (in d). k,o, Male control and L-Pck1KO or L-GykKO mice were intravenously administered adenovirus vector expressing LacZ or LbNOX, as indicated, 1 week after tamoxifen treatment. Then, 1 week after the adenovirus injection, mice were subjected to high-intensity (k) or low-intensity (o) exercise and their times until exhaustion were recorded. n = 7 per group; one-way ANOVA followed by Tukey’s post hoc analysis (two-sided); **P = 0.0008 for Ad-LacZ vs Ad-LbNOX (within control), *P = 0.0240 for control vs L-Pck1KO (within Ad-LacZ), **P < 0.0001 for control vs L-Pck1KO (within Ad-LbNOX) (in k); **P = 0.0037 for Ad-LacZ vs Ad-LbNOX (within control), **P = 0.0033 for control vs L-GykKO (within Ad-LacZ), **P < 0.0001 for control vs L-GykKO (within Ad-LbNOX) (in o). l–n,p–r, Contents of glycogen (l,n), ATP (m,q) and Pi (n,r) in the gastrocnemius were measured in male Ad-LacZ and Ad-LbNOX mice under sedentary conditions and with the indicated intensity of exercise. n = 8 per group (in l–n); n = 7 per group in sedentary, n = 8 per group in exercise (in p–q); one-way ANOVA followed by Tukey’s post hoc analysis (two-sided); *P = 0.0479 for sedentary vs exercise (within Ad-LacZ) (in l); **P < 0.0001 for sedentary vs exercise (within Ad-LacZ), **P < 0.0001 for Ad-LacZ vs Ad-LbNOX (within exercise) (in n); *P = 0.0170 for sedentary vs exercise (within LacZ) (in p); **P < 0.0001 for sedentary vs exercise (within LacZ), **P = 0.0031 for Ad-LacZ vs Ad-LbNOX (within exercise) (in r). s, Summary of the inter-organ networks during high-intensity and low-intensity exercise. In high-intensity exercise, cross-talk between muscle and liver mediated by lactate-derived gluconeogenesis has a key role in energy production. In low-intensity exercise, the inter-organ network between adipose–liver–muscle by glycerol-derived gluconeogenesis is important for energy production. The liver serves as a hub linking these inter-organ networks, wherein the hepatic cytosolic [NADH]/[NAD⁺] ratio modulates gluconeogenesis from lactate or glycerol. The schematic diagram was created in BioRender.com. Experiments shown in a, b, c–e, f, g, h–j, k, l–n, o and p–r were conducted using distinct cohorts of biologically independent mice. All data are presented as means; error bars, s.e.m. Each plot on the bar graph shows raw data.

Source data