Temporal response to bioenergetic stress in adipose tissue and skeletal muscle

Diabetes is a common disease—affecting more than 500 million people. Although type 2 diabetes maintains a genetic component—the implementation of lifestyle modification serves as a potent risk-reducing modality. Chronic exercise is among these lifestyle modifications, as it generates therapeutic effects at the tissue level. However, strategies to optimize exercise need further investigation. The aim of this doctoral work is to characterize the interplay between exercise, temporal variables, and tissue biology in the context of whole-body energy homeostasis.

In study I we evaluated the timing-specific effects of acute exercise on adipose tissue biology in mice through measurement of circulating factors in blood and analysis of the adipose tissue transcriptome. Only early active phase induced an immediate increase in serum non-esterified fatty acids and expression of thermogenic and angiogenic markers in inguinal adipose tissue. Synchronized 3T3-L1 adipocytes also showed a timing dependent difference in β2 adrenergic receptor (Adrb2) expression and a greater lipolytic activity. Rest phase mice lacked the transcriptomic response to exercise, regardless of feeding state, suggesting that timing supersedes feeding status as a modulator of the post exercise adipose tissue transcriptome. These findings highlight that adipose tissue responds to exercise in a time-of-day-dependent manner, which may be partly driven by the circadian clock.

In study II we followed up on the timing-specific effects of exercise with the additional context of obesity by feeding mice a 5-week high-fat diet. In both diet groups, acute exercise at the early active phase induced a shift towards oxidative metabolism for the duration of the phase and over 24 hours following. Effects were milder following rest phase exercise, with exercise leading to one hour of increased energy expenditure. Oxidation of glucose was reduced at the early active phase, regardless of diet; however, chow-fed mice showed a milder post-exercise shift in glucose oxidation, potentially indicating an exercise timing effect on substrate preference unique to lean mice. Upon measuring lipolysis in adipose tissue ex vivo, only adipose tissue from chow mice showed heightened lipolysis at the early active phase and in response to exercise. These data highlight the functional relevance of exercise timing for whole-body metabolism following exercise, and reveal an absence of diurnal lipolytic function in the adipose tissue of obese mice.

In study III we evaluated if the circadian clock alters the lipolytic metabolism in concert with AMP-activated protein kinase (AMPK) in human skeletal muscle cells. We observed that core clock genes Neuronal PAS domain protein 2 (NPAS2) and D-box binding PAR bZIP transcription factor (DBP) were decreased in cells where AMPKα was silenced. These AMPKα-knockdown cells also showed increased expression of genes associated with fatty acid metabolism (Fatty acid translocase, CD36; Carnitine palmitoyltransferase 1B, CPT1B; and Peroxisome proliferator-activated receptor gamma coactivator 1-alpha, PPARGC1A). Upon directly silencing expression of DBP and NPAS2, we observed a similar increase in expression of lipid metabolism markers (PPARGC1A, CD36, and Fatty acid binding protein 3, FABP3). Assessment of palmitate oxidation revealed that silencing of NPAS2 alone elicited an increase in lipid oxidation, revealing a role of the molecular clock on oxidative metabolism in human muscle cells. This work emphasizes the critical role of circadian function in the regulation of skeletal muscle lipid homeostasis, and may reveal NPAS2 as a therapeutic target to decrease ectopic lipid deposition in skeletal muscle.

In study IV we evaluated the temporal dynamics of adipose tissue transcription following exercise in individuals with normal glucose tolerance or type 2 diabetes. This analysis revealed that adipose tissue from people with type 2 diabetes displays a robust upregulation of inflammatory transcriptional activity at 3 hours following exercise. Based on magnitude of fold change after exercise, we selected Oncostatin-M (OSM) as a candidate gene in adipose tissue. Upon treating human adipocytes with OSM, lipolysis was increased. Further, we identified that OSM expression is greatest in the immune fraction of adipose tissue, and observed that human THP-1 macrophages show increased expression and secretion of OSM following treatment with exercise-like stimuli. These results reveal the influence of time following exercise on transcriptional regulation, which can be unique depending on disease status. Additionally, this work highlights the necessity to consider time of sample collection in evaluating metabolic responses to bioenergetic stress.

In conclusion, the works that comprise this thesis address the critical role of time in the response to exercise. In the preclinical rodent studies I and II, we discovered novel transcriptional and metabolic responses to exercise at the early active phase; these insights highlight potential for discoveries in human models, which may be used in the development of ideal individual exercise prescriptions. Identification of clock gene NPAS2 in study III as a regulator of basal metabolism in human muscle cells raises questions about the targeting of clock genes in therapeutic approaches to increase lipid clearance, which is particularly relevant in addressing ectopic lipid deposition. Lastly, characterizing the time course of post-exercise transcription in individuals with normal glucose tolerance or type 2 diabetes in study IV provided insights into the critical role of measurement timing in identifying gene candidates. The summative findings in these studies highlight critical metabolic factors, and emphasize the importance of future considerations of timing in exploratory research.

List of scientific papers

I. Pendergrast LA, Lundell LS, Ehrlich AM, Ashcroft SP, Schönke M, Basse AL, Krook A, Treebak JT, Dollet L, Zierath JR. Time of day determines postexercise metabolism in mouse adipose tissue. Proc Natl Acad Sci U S A. 2023 Feb 21;120(8):e2218510120.
https://doi.org/10.1073/pnas.2218510120

II. Pendergrast LA, Ashcroft SP, Ehrlich AM, Treebak JT, Krook A, Dollet L, Zierath JR. Metabolic plasticity and obesity-associated changes in diurnal postexercise metabolism in mice. Metabolism. 2024 Mar 11:155834.
https://doi.org/10.1016/j.metabol.2024.155834

III. Pendergrast LA*, Kamble P*, Björnholm M, Zierath JR. Skeletal muscle oxidative metabolism is mediated by clock gene NPAS2. *These authors contributed equally. [Manuscript]

IV. Dollet L, Lundell LS, Chibalin AV, Pendergrast LA, Pillon NJ, Lansbury EL, Elmastas M, Frendo-Cumbo S, Jalkanen J, Barbosa TC, Cervone DT, Caidahl K, Dmytriyeva O, Deshmukh AS, Ryden M, Wallberg-Henriksson H, Zierath JR, Krook A. Exercise-induced crosstalk between immune cells and adipocytes in humans: Role of oncostatin-M. Cell Rep Med. 2024 Jan 16;5(1):101348.
https://doi.org/10.1016/j.xcrm.2023.101348