Legacy Effects of Early β-Adrenergic Stimulation Program Adipose Plasticity and Confer Metabolic Resilience in Obesity

Figures

• 

  Extended Data Figure 1:. “Browning” and “Re-whitening” of Adipose Tissues a) Adipose depot mass following 2 weeks of cold exposure (CE) (Post-Intervention, left) and after the 4-week recovery period (Post-Recovery, right). n= 5–6. * denotes p <0.05, ** denotes p <0.01 by unpaired t-test. Bars represent mean + SEM. Each data point represents a unique animal. Exact measurements and statistical data are provided in Source Data Extended Data Figure 1. b) Representative images of H&E-stained epididymal WAT (eWAT) and inguinal WAT (iWAT) following 2 weeks of the indicated intervention or after the 4-week recovery period at RT. c) Volcano plots depicting number of differentially expressed genes (padj ≤0.05; absolute log2FC ≥1) in interscapular BAT following 2 weeks of CL 316,243 (top) or after the recovery period (bottom). Numbers of genes enriched in vehicle (negative fold change) or CL 316,243 (positive fold change) groups are shown. d) EnrichR pathway analysis of genes enriched in eWAT (log2 FC >0.25; padj <0.05) after CL 316,243 treatment. e) Same as panel d, after CL 316,243 recovery. Full datasets for panels c-e are in Supplementary Tables 3, 6–8.

• 

  Extended Data Figure 2:. Cellular landscapes of eWAT after β3AR agonist withdrawal a) Integrated Uniform Manifold Approximation and Projection (UMAP) of 14,851 CD45+ cell transcriptomes obtained from pooled eWAT depots of 4 mice previously treated with vehicle or 4 mice previously treated with CL 316,243. Data were filtered with the criteria of having more than 500 unique molecular identifiers (UMIs) per gene and more than 250 genes that consisting less than 10% in cells. The median number of UMIs detected per cell was 3,484 in the Vehicle group and 3,544 in the CL group. The mean reads per cell was 91,408 in the Vehicle group and 93,425 in the CL. b) Bubble plot of gene expression defining indicated immune cell types. c) UMAP of 12,504 nuclei transcriptomes obtained from pooled eWAT depots of 4 mice previously treated with vehicle or 4 mice previously treated with CL 316,243. Data were filtered with the criteria of having more than 500 unique molecular identifiers (UMIs) per gene and more than 250 genes that consisting less than 10% in cells. The median number of UMIs detected per cell was 1,882 in the Vehicle group and 1,810 in the CL group. The mean reads per cell was 28,820 in the Vehicle group and…

• 

  Extended Data Figure 3:. Macrophage transcriptional remodeling in eWAT following β3AR stimulation and withdrawal. a) UMAP of 8,313 eWAT macrophages/monocytes in vehicle-treated or CL-trained mice (Post-Recovery) (sub-clustered from CD45+ single cell RNA-seq shown in Extended Data Fig. 2). b) Bubble plot of gene expression defining indicated macrophage/monocyte subclusters. Full dataset of cluster markers is in Supplementary Tables 15. c) Frequency of total macrophages (CD45+ CD11b+ F4/80+) and CD163+ perivascular macrophages in eWAT following two weeks of vehicle or CL 316,243 treatment (Post-Intervention) and four weeks after cessation of the intervention (Post-Recovery). Bars represent mean + SEM. Each data point represents a unique animal. Exact measurements and statistical data are provided in Source Data Extended Data Figure 3. d) Violin plot of indicated genes associated with efferocytosis and lipid associated macrophages. Full datasets of differentially expressed genes (CL vs Veh.) and pathway analyses are in Supplementary Tables 14–28.

• 

  Extended Data Figure 4.. Prior cold exposure or β3AR agonism enhances future capacity for de novo adipogenesis during high-fat diet challenge. a) mRNA levels of indicated transcripts in whole eWAT or iWAT of control and CL-trained animals after 10 weeks of HFD feeding. n=11 mice per group. * denotes p<0.05, ** denotes p<0.01 by unpaired t-test. b) Pulse-Chase lineage tracing strategy: PDGFRβ+ adipose progenitor cells were labeled with mGFP expression in male PdgfrbrtTA; TRE-CRE; Rosa26RmT/mG mice with the addition of doxycycline (DOX) during the last week of recovery following CL 316,243 treatment. Animals were then placed on HFD (no DOX) (“Chase”) for 10 weeks prior to harvest. c) Representative confocal immunofluorescence images of PERILIPIN and mGFP expression in eWAT and iWAT after HFD feeding. mGFP+ PERILIPIN+ adipocytes represent newly formed adipocytes.

• 

  Extended Data Figure 5.. Adaptive adipose remodeling is linked to gene signatures of EBF, PPARγ, and PRDM16 activity. a) Top experimentally defined gene signatures of indicted transcription factor perturbations overrepresented (p adj. <0.05) in eWAT of CL-trained mice following high fat diet feeding (Post-HFD). b) Top experimentally defined gene signatures of indicted transcription factor perturbations enriched (p adj. <0.05) in visceral adipocytes of individuals with MHO compared to BMI-matched individuals with MUO. EnrichR analysis was performed using list of differentially expressed genes (MHO vs. MUO) described in Reinisch et al. Complete EnrichR analysis for panels a,b can be found in Supplementary Tables 38,40.

• 

  Figure 1:. β3AR agonism and withdrawal remodel adipose tissue morphology and transcriptional programs a) Experimental Design: 6-weeks-old male C57BL/6N mice were either 1) single-housed at 6°C (cold exposure, CE) or at room temperature (RT), or 2) treated daily with the β3AR agonist CL 316,243 (CL) (1 mg/kg body weight) or saline (vehicle; Veh) at RT for 2 weeks. Following the 2-week intervention period, animals were returned to RT housing for 4 weeks (Recovery). Image created in BioRender. b) Adipose depot mass following 2 weeks of vehicle or CL 316,243 treatment. n= 5–6. * denotesp <0.05 by unpaired t-test. c) Representative images of H&E-stained epididymal WAT (eWAT) and inguinal WAT (iWAT) fromanimals harvested following 2 weeks of vehicle or CL 316,243 treatment. Scale bar = 100 µm. d) Volcano plots depicting number of differentially expressed genes (padj ≤0.05; absolute log2FC ≥1) in whole eWAT or iWAT following 2 weeks of CL 316,243. Numbers of genes enriched in vehicle (negative fold change) or CL 316,243 (positive fold change) groups are shown. e) Routine oxygen consumption rate measured from primary fractionated adipocytes collected following 2 weeks of CL 316,243. Bars represent…

• 

  Figure 2:. β3AR agonism induces durable transcriptional reprogramming in epididymal adipocytes a) UMAP of single epididymal adipocyte nuclei transcriptomes obtained following the recovery from CL 316,243, revealing three molecularly distinct adipocyte (Ad) subclusters (Ad1–3). b) Heatmap depicting expression of most enriched transcripts defining Ad1–3. c) Overrepresented WikiPathways amongst genes enriched in Ad 1 d) same as panel c) for Ad 2 e) same as panel c) for Ad 3 f) Experimentally defined gene signatures of indicted transcription factor perturbations overrepresented in Ad 2. g,h) Violin plots of indicated gene expression in adipocyte subclusters. EnrichR was used for pathway analysis, with defining transcripts of each cluster (log2FC >0.25 and padj. p<0.05) used as input. Full datasets for each panel are in Supplementary Tables 30–34.

• 

  Figure 3:. Early cold or β3AR training ameliorates insulin resistance during future obesity a) Experimental Design: 4 weeks after cold exposure or CL316,243 treatment, animals were switched to a high fat diet (60% kcal from fat) and maintained at RT for ≥ 10 weeks (“HFD Challenge”). b) Average body weights of animals following the cold intervention (Post-INT), the recovery period (Post-REC), and after 12 weeks of HFD feeding (Post-HFD). n= 9–10. c) Same as panel B, for animals treated with vehicle or CL 316,243. n= 11–12. d) (left) Intraperitoneal glucose tolerance tests (GTT) of control (RT) or cold-trained mice (CE) after 11 weeks of HFD feeding. (right) Area under the GTT curve (AUC) measurements. n=9–10 per group. e) Same as panel D, for animals treated with vehicle or CL 316,243. n=7–9 per group. For GTT, * denotes p<0.05, ** p<0.01 by two-way ANOVA followed by Šídák’s multiple comparisons test. *** denotes p <0.001 by unpaired t-test test for AUC measurement. f) Plasma insulin levels in control and cold-trained mice after HFD feeding. Insulin measurements were made at indicated time points following single oral glucose bolus. n= 11 * denotes p <0.05, ** denotes p<0.01 by two-way ANOVA…

• 

  Figure 4:. Early cold or β3AR training promotes adaptive WAT expansion during future obesity a) Mass of indicated tissues in cold-trained (top) or CL-trained mice (bottom) after 12 weeks of HFD feeding. Top, n= 9–10; bottom, n = 11–12. * denotes p<0.05, ** p<0.01, *** p<0.001 by unpaired t-test. b) Histological analyses of eWAT from cold-trained (left) or CL-trained (right) animals after 12weeks of HFD feeding. Top row: hematoxylin and eosin (H&E) staining; scale bar = 200 µm. Middle row: indirect immunofluorescence of PERILIPIN and MAC2 expression. Bottom row: picrosirius red staining of collagen deposition. scale bar = 200µm. c) Frequency of MAC2+ crown-like immune structures (CLS) in eWAT of cold-trained (left) or CL-trained (right) after 12 weeks of HFD feeding. n=6 mice per group. * p<0.05 by unpaired t-test. d) Mean adipocyte size in eWAT of cold-trained (left) or CL-trained (right) after 12 weeks of HFD feeding. n=6 mice per group. ** denotes p<0.01 by unpaired t-test. e) Volcano plot depicting differentially expressed genes (padj ≤0.05; absolute log2FC ≥1) in eWAT of CL-trained mice after 12 weeks of HFD feeding. n=4–6 mice for each group. Numbers of genes enriched in vehicle…