Age-induced BAT ADH5 reduction contributes to BAT senescence.

A. Representative images (63X) of protein S-nitrosylation staining (SNO) in BAT from young (3-month) or old mice (12- and 22-month). –AS: no ascorbate; negative control for SNO staining. Scale bar: 10 μm. B. Representative Western and densitometric analysis (normalized to ACTB input) of SNO proteins in BAT from 3- or 22-month-old wild-type mice. C&D. Adh5 expression (normalized to Gapdh) in BAT and iWAT respectively of Adh5fl and Adh5BKO mice. N= 3-7 mice/group. E. Levels of mRNAs in BAT of young (3-month) Adh5fl, Adh5BKO, or WT mice compared to 22-month-old WT by RT-qPCR. N= 3-10 mice/group. F. Representative images of H&E, IF of p16lNK4a, and β gal staining in BAT from age-matched Adh5fl and Adh5BKO mice. Scale bar: 10 μm in H&E and p16lNK4a, 50 μm in β gal staining. Right panel: quantified fluorescence intensities of p16lNK4a. 5 fields/mouse, n= 3 mice/group. Data are presented as means ± SEM. *indicates statistical significance relative to the 3-month groups, and #indicates genetic effects in mice at the same age; as determined by Student’s t-test in (B) and ANOVA followed by posthoc test in (C-E&F), p<0.05.

BAT Adh5 deletion accelerates age-related functional declines.

A. Body weight, and B. Body composition of Adh5fl and Adh5BKO mice at indicated age (n= 5-22 mice/group). Inset: representative picture of Adh5fl and Adh5BKO mice at 12 months. C&D. Whole-body energy balance measured in Adh5fl and Adh5BKO mice housed within a metabolic cage (n= 5-7 mice/group). E-F. GTT and AUC from Adh5fl and Adh5BKO mice (n= 14-30 mice/group). G. Grip strength, H. Rotarod assessment, I. Barnes maze assessment of velocity, J. percent of time in target/nontarget zone and K. primary errors to escape hole in 19-month-old Adh5fl and Adh5BKO mice. N = 7-9 mice/group. L&M. Heart mass and left ventricle thickness (LV) from age-matched Adh5fl and Adh5BKO mice measured by echocardiography; n = 5-7 mice/group. Data are presented as means ± SEM. *indicates genetic effects in same-age mice, and #indicates age effects in the same type of mice; as determined by Student’s t-test (A, C, D, and G-M), and ANOVA in (B) or ANOVA of AUC in (F), p<0.05.

BAT ADH5-deficiency augments age-associated BAT immuno-metabolic imbalances.

A. Gene expression of measured markers in SVF from BAT of Adh5fl or Adh5BKO mice. N= 3-5 mice/group. B. Plasma levels of IL-1β in 3-month-old Adh5fl or Adh5BKO mice measured by ELISA. N = 5-6 mice/group. C. Representative images (63X) and quantification of IF of F4/80 staining in BAT from Adh5fl or Adh5BKO mice at indicated age. N= 3 mice/group, 4-8 fields/mouse. Scale bar: 1 μm. D. F4/80 (63X, scale bar: 1 μm) staining and quantification in iWAT from Adh5fl or Adh5BKO mice at the indicated age. N= 3 mice/group, 4-8 fields/mouse. E&F. Representative Western blot and quantification of ETC proteins in BAT of 3-month-old Adh5fl or Adh5BKO mice (n= 3 mice/group). G. Representative TEM images of BAT from 3-month-old Adh5fl or Adh5BKO mice. Red arrow, the mitochondrion. Scale bar: 5 μm (I), 2 μm (II), and 1 μm (III). H. Quantification of mitochondrial area in TEM images of BAT in mice in (C). Quantification was performed using ImageJ software. I. Schematic summary of the BAT metabolomic analysis in Adh5fl or Adh5BKO mice (3-month-old) analyzed by using Metaboanalyst online software; n = 3 age-matched mice per group. Downregulated metabolites by ADH5 deletion are depicted in red; upregulated metabolites are depicted in green. Data are presented as means ± SEM. * indicates genetic effects in same-age mice as determined by Student’s t-test (A, B, F and H), and ANOVA (C&D), p<0.05.

HSF1 induces Adh5 and ameliorates senescence in BAT.

A. Hsf1 levels in BAT at indicated age (n= 6-12 mice/group) measured by qRT-PCR. Data were normalized to Gapdh. B. Left panel: representative HSF1 staining (63X, scale bar: 1 μm; zoomed images are on the right) from BAT of young and old wild-type mice. Right panel: quantification, n = 3-4 mice/age. 4-8 fields/mouse. C. Occupancy of Adh5 promoter regions by HSF1 from BAT treated with a pharmacological activator of HSF1 (HSF1A; Millipore, 20 μM, 24-hr) in controls or Adh5BKO mice (n= 3 mice/group). D. Levels of mRNAs encoding the indicated genes of BAT explants treated ex vivo with HSF1A (20 μM, 24-hr) or HSF1 inhibitor (KRIBB11; 2 μm, 24-hr). Data were normalized to Gapdh. N=4 mice/group. E. Levels of mRNAs of the indicated genes in SVF from 3- or 5-month-old BAT pad isolated from wild-type mice followed by treatment of HSF1A (20 μM, 24-hr). N=4 mice/age group. Data were normalized to Gapdh. Data are presented as means ± SEM. *indicates statistical significance compared to the 3-month groups (A&B), to the Adh5fl groups in (C), and to the control groups in (D-F) as determined by ANOVA in (A-D), and Student’s t-test in (E), p<0.05.

BAT HSF1 mitigates age-associated functional decline.

A. Comparison of release curves between Collogen-NC-HSF1A vs Collagen-HSF1A. B. Body weight measurement over time in 12-month-old wild-type mice after interscapular injection with Collagen-NC-Vehicle (NC-Vehicle) or Collagen-NC-HSF1A (NC-HSF1A; 0.1 mg). C. Heat, VO2, and energy expenditure (EE) in 12-month-old wild-type mice after one BAT injection of control (NC-Vehicle) or NC-HSF1A (0.1 mg; n= 5-6 mice/group). D. GTT and AUC of mice from (B). N= 19-37 mice/group. E. Latency (Barnes maze test) and F. AUC of escape measured from mice in (C) after one-month injection. N= 7-13 mice/group. G. Representative H&E (Scale bar: 20 or 10 μm) and IF of p16lNK4a (scale bar: 10 μm) in BAT from mice in (C). H. Quantitation of IF staining of p16lNK4a. N= 5 mice/group. 4-6 fields/images. I. Representative F4/80 staining of BAT from (C); quantitation is left of the image. Scale bar: 1 μm, n = 5 mice/group. 4-6 fields/images. J. Levels of mRNAs encoding genes of interest (normalized to Gapdh) in BAT from mice in (C). N = 3-9 mice/group. Data are presented as means ± SEM. *indicates statistical significance compared to the vehicle groups as determined by Student’s t-test, p<0.05.