ObjectiveBone marrow adipose tissue (BMAT) represents > 10% total fat mass in healthy humans and further increases in diverse clinical conditions. Like white adipose tissue (WAT), BMAT can produce hormones such as adiponectin, suggesting that it may exert systemic metabolic effects. Unlike WAT, BMAT increases in conditions of caloric restriction (CR), in which it is a key source of increased circulating adiponectin. Thus, BMAT appears developmentally and functionally distinct from WAT. However, the impact of BMAT on health and disease remains poorly understood. Thus, the objective of my research is to determine the fundamental functions and clinical significance of BMAT.MethodsMy lab's research combines preclinical animal models, human clinical studies and data science approaches using the UK Biobank. Mouse models include CR studies in adiponectin knockout mice, to test if BMAT acts via adiponectin to influence the metabolic, skeletal and immunological effects of CR. Clinical studies include the use of positron emission tomography and computed tomography (PET-CT) to assess BMAT's metabolic functions in vivo. Finally, we are using deep learning to analyse BMAT in magnetic resonance imaging data from the UK Biobank, thereby allowing us to identify the physiological, pathological and genetic factors associated with altered BMAT.Results(1) Adiponectin KO mice: We have found, unexpectedly, that lack of adiponectin enhances, rather than impairs, CR-induced improvements in glucose tolerance. Compared to wild-type counterparts, KO mice have lower fasting glucose during CR. Adiponectin KO also alters the immunological effects of CR. (2) PET-CT studies: in mice we find that BMAT glucose uptake is not insulin or cold-responsive, demonstrating that BMAT is metabolically distin...