Fig. 3. Enhanced heat sensitivity and adaptive thermogenesis in Y524S mice.

a Relationship of ambient housing temperature and the maxVO₂ during acute 37 °C heat challenge in WT (n = 129) and YS (n = 154) mice. b Effect of prior median ambient housing temperature on estimated survival probability of YS mice (n = 154) after acute heat challenge. c Kaplan–Meier analysis of the heat challenge survival rate of YS mice (n = 154) grouped by duration of prior ambient housing temperature at lower than 20 °C. d, e Core body temperature before (d) and after (e) acute 37 °C heat challenge in WT and YS mice preconditioned for 1 week at 4 °C (WT n = 23, YS n = 21), room temperature (WT n = 35, YS n = 40), and 30 °C (WT n = 6, YS n = 11) ambient environment. f, g Core body temperature before (f) and after (g) acute 37 °C heat challenge in YS mice preconditioned at fasted (n = 6) and refed state (n = 11, post-heat challenge temperature of one refed YS mice was not measured due to full body contracture). All mice were within the controlled age range (8.9 ± 0.9 week old) at the time of study. P values are indicated as analyzed by F-test for deviation from zero-slope of linear regression (a), Mantel–Cox log-rank test (c), ordinary one-way analysis of variance (ANOVA) with Dunnett’s multiple comparisons test (d, e) and Welch’s t test (f, g). All statistical tests are two-sided. R² values are indicated to quantify goodness-of-fit to nonlinear regression with variable slope (b). Survival probability of each mouse is estimated based on the survival rate from ten mice in the respective ambient temperature (b) subgroups. Effect of ambient housing temperature on heat challenge survival is measured by half-maximal effective ambient housing temperature (ET₅₀) on the estimated survival probability (b). Data are represented as asymmetrical 95% confidence intervals (CI) from linear best-fit line (a), or mean ± standard deviation (d–g). Source data are provided as a source data file.