We have generated unbiased proteomic and transcriptional profiles for our discovery platform to identify key enzymes and receptors controlling brown fat activity. We have validated some of our top hits in fat cells and have made, or are in the process of making, mouse models to study the physiological function of these candidates. We published two manuscripts on key enzymes in circadian and cold control of brown adipose metabolism with several more papers in revision or review. Collectively, our findings from aCROBAT made major in-roads into the understanding of adipose biology and the connection to metabolic disease.
In the first published study, we investigated an enzyme called NAMPT, which is involved in control of fat cell bioenergetic programs and circadian rhythm. We found that when we deleted Nampt only in fat tissue of mice, the mice did not become obese when given a diet high in fat. Even though the mice were fed a diet similar to eating only hamburgers, hot dogs, and pizza, they gained no more weight than mice fed a healthier, lower fat diet. Our research is the first to reveal that NAMPT is a critical circadian factor in expansion of fat tissue and is indispensable for the development of obesity. More broadly, NAMPT represents a novel paradigm by which mammals can specifically control consumption of dietary fat based on how much energy can be stored in the fat tissue.
The goal of the second published study was to investigate which enzymes were most critical in mediating environmental control of brown fat activity. Combined proteomic and lipidomic analyses revealed that a single lipid species known as cardiolipins and their obligate synthase, CRLS1, were not only indispensable drivers of brown fat energy expenditure but affected metabolic homeostasis or balance throughout the body. Depleting cardiolipin in brown fat caused mice to become insulin resistant and diabetic. These results showed for the first time that a lipid in the mitochondria of brown fat was capable of modulating whole body insulin sensitivity. Human CRLS1 expression in fat tissue of healthy or diabetic individuals suggests that our findings are translatable. This study involved the generation of several new mouse models, human cells, tissue, and genetic association studies, and collaborative efforts between basic researchers and clinicians in Germany, the US, and Denmark.
In the aCROBAT proposal, we outlined the goal of leveraging our basic research findings into bona fide innovation to fight metabolic disease. During the third reporting period we began realizing that goal when we spun out a new company, Embark Biotech, from the University of Copenhagen with the help of an ERC PoC grant. Embark is based on a novel receptor in brown adipose tissue that significantly increases calorie-burning and reduces bodyweight in obese mice. We made the discovery during the first stages of aCROBAT and have characterised the mechanism-of-action by generating tissue specific receptor knockout mice. In parallel, we are now optimising the pharmaco-kinetic properties of the target to develop a next-generation obesity therapeutic.
Additionally, we have uncovered another receptor, GPR3, that plays a critical role in the activation of human brown adipose. Depleting this receptor in fat cells isolated from seven different patients showed decreased thermogenic gene expression in all cases. We developed gain and loss-of-function animal models to study a causal role for this receptor in thermogenic adipose physiology. We found that activation of this receptor is fully sufficient to drive the calorie-burning function of brown adipose even in the absence of sympathetic signalling. This receptor and its agonists will eventually form the basis for a second spinout company my group based on the ERC starting grant. In the meantime, we are finalizing the revision for the manuscript.
Finally, we have continued on the work on the Nampt NAD+ biosynthetic pathway in adipose, focusing on the role of Nampt in the molecular clock. Previous dogma in the field argues that Nampt is an integral member of the clock in all cells. However, these studies were all performed in immortalised cel lines. We used our adipose specific and muscle specific Nampt KOs, combined with global gene profiling to uncover that Nampt dramatically affects the brown adipose clock but does not impinge on skeletal muscle rhythmicity at all. Moreover, Nampt knockdown has completely different effects on transcription in white adipose compared to brown. This work represents the first characterisation of Nampt's role in the molecular clock in an in vivo setting across different tissues.
