Cover Page
The following handle holds various files of this Leiden University dissertation:
http://hdl.handle.net/1887/61829
Author: Hoeke, G.
Title: A fatty battle: towards identification of novel genetic targets to comBAT cardiometabolic diseases
Issue Date: 2018-05-03
Chapter
General discussion and future perspectives
Modified from:
Relevance of lipid metabolism for brown fat visualization and quantification Curr Opin Lipidol 2016; 27(3): 242-8.
Role of brown fat in lipoprotein metabolism and atherosclerosis development Circ Res 2016; 118(1): 173-82.
9
1 GENERAL DISCUSSION AND FUTURE PERSPECTIVES
Cardiovascular diseases (CVD), mainly caused by atherosclerosis, are the leading cause of morbidity and mortality in Western Society (1). An important risk factor for the development of atherosclerosis is hyperlipidemia. To date, statins are the most widely used drugs to treat hyperlipidemia and they lower cholesterol with approximately 30%
(2). However, statin treatment only prevents 25-45% of all cardiovascular events (1), illustrating the need for additional therapies. Energy-combusting brown adipose tissue (BAT) has recently been identified as a key player in lipid metabolism by mediating triglyceride (TG) clearance (3) as well as indirectly contributing to cholesterol metabolism (4-6). In the first part of this thesis, we aimed to further elucidate the contribution of BAT to lipid metabolism and assess its therapeutic potential to combat atherosclerosis.
Another important risk factor for the development of atherosclerosis is inflammation, characterized by progressive accumulation of various immune cells within the arterial wall that drive plaque formation (7). Cells from both the innate and adaptive immune system contribute to this pro-inflammatory process by secreting pro- and anti- inflammatory cytokines upon their activation. We therefore aimed in the second part of this thesis to elucidate the role of the immune system in atherosclerosis development, specifically addressing the contribution of the anti-inflammatory interleukin 37 (IL-37) and members of the C-type lectin receptor (CLR) family.
From this thesis, various novel perspectives on the mechanisms underlying the anti- atherogenic effect of BAT activation have arisen, which will be discussed below.
Furthermore, translational limitations and therapeutic implications for both BAT activation and the immune system in relation to the development of atherosclerosis will be addressed.
2 TRANSLATIONAL CHALLENGES IN BAT RESEARCH
2.1 Importance of activated BAT in triglyceride metabolism
Activated BAT is an important player in TG metabolism by being involved in the uptake of TG-derived fatty acids (FA) from the circulation. However, it remained to be elucidated whether BAT takes up TG-derived FA via whole TG-rich lipoprotein particle (TRL) uptake,
9
after lipoprotein lipase (LPL)-mediated lipolysis of TRLs, or via a combination of both. We investigated this matter in chapter 2 by injecting TRL-like emulsion particles double- labeled with glycerol tri[3H]oleate ([3H]TO) and [14C]cholesteryl oleate ([14C]CO) of various sizes (i.e. mimicking small VLDL, large VLDL and small chylomicrons) in normolipidemic wild-type mice. In this setting, we showed that BAT takes up TG-derived FA mainly after LPL-mediated lipolysis, rather than via the uptake of whole TRLs (Fig. 1). Moreover, this appeared true for all tested sizes of TRL-like particles and for minimally active (i.e. housing at thermoneutrality) and active (i.e. housing in cold) BAT, although the absolute amount of TG-derived FA uptake by activated BAT was much higher as compared to inactive BAT. Whole particle uptake did take place by BAT, although to a low extent, and was largest for the small chylomicron-like particles (e.g. the largest sized particles) upon cold exposure. It can be questioned whether these particles are taken up by brown adipocytes themselves or by other cell types present in BAT. For instance, these particles may have been taken up by endothelial cells that are surrounding brown adipocytes (3), e.g. via the LDL receptor (LDLR) or via transcytosis (8). It is feasible that, once in the interstitial space, TG-derived FA are liberated from TRLs and taken up by brown adipocytes, after which TRL remnants are released in the circulation. Another explanation would be that especially large particles are taken up by macrophages, which are present in BAT (9), as larger particles may be recognized as foreign bodies and are consequently phagocytised.
Indeed, larger particles were also taken up to a higher extent by the spleen, which has a high content of resident macrophages. A way to evaluate these options would be to inject labeled particles in (cold-exposed) mice, isolate the vascular stromal fraction and subsequently separate the various cell types from this fraction by cell sorting.
Determining the uptake of the labels by endothelial cells and/or macrophages will be instrumental to further evaluate which cell types are responsible for whole TRL uptake by BAT. It may also be questioned whether the TRL-like particles that we used for our studies are true mimics for endogenous TRLs. However, injection of APOE*3-Leiden.CETP (E3L.CETP) mice with [3H]TO- and [14C]CO-labeled autologous VLDL resulted in similar findings with mainly uptake of TG-derived FA after LPL-mediated lipolysis by BAT [Hoeke et al., unpublished]. This demonstrates that our findings from chapter 2 represent the physiological situation.
We thus identified increased LPL-mediated TG-derived FA uptake by BAT, rather than whole particle uptake, as an important feature of activated BAT. This was not only observed after cold exposure but also when BAT is activated by pharmacological
compounds, such as rimonabant (10), salsalate (11) and a β3-adrenergic receptor (AR) agonist (chapter 3). As a consequence of the increased TG-derived FA uptake, 2 weeks of treatment of hyperlipidemic E3L.CETP with a β3-AR agonist already reduced plasma TG with 31% [Hoeke et al., unpublished]. Taken together, BAT activation is thus a powerful tool to alleviate hypertriglyceridemia in mice by accelerating LPL-mediated lipolytic processing of TRL.
In chapter 5, we studied the effect of BAT activation, by means of short-term cooling (i.e. 2 hours), on lipoprotein metabolism in healthy lean men. Although cold exposure is the most important physiological activator of BAT, and BAT activation in mice thus reduces plasma TG, short-term cooling of humans increased rather than decreased plasma TG levels. This was accompanied by an increased number of large VLDL particles, which was most likely due to increased hepatic VLDL secretion. Indeed, cold exposure has been shown to increase sympathetic outflow to the liver thereby increasing hepatic VLDL secretion, al least in rodents (12). Most human studies evaluate the effect BAT activation on energy and lipid metabolism by means of short-term cold exposure (i.e. 2 hours). A very recent study was the first to investigate the contribution of activated BAT in lipid uptake after 4 weeks of cold exposure (2 h of cold/day). Subjects received a meal containing [18F]FTHA, which is subsequently incorporated in chylomicrons, after which the relative uptake of these lipids by various organs including BAT was determined.
Indeed, the uptake of TG-derived FA by BAT per gram organ was much higher as compared to WAT and skeletal muscle. TG-derived FA uptake by BAT was accompanied by increased oxidative metabolism and reduced lipid content in BAT (13), all in line with increased BAT activity. Of note, upon cold exposure (after 4 weeks of cold acclimatization) and meal ingestion, not only plasma chylomicron levels increased, but also plasma VLDL levels. The latter is completely in line with our study (chapter 5) and likely due to direct sympathetic innervation of the liver upon cold exposure. The observation that BAT takes up dietary fatty acids from TRLs, likewise to mice, also implies that our work in mice likely translates to the human situation.
The fact that plasma TG levels in mice are reduced upon cold exposure may be due to the higher contribution of BAT to lipid metabolism in mice as compared to the contribution of BAT in lipid metabolism in humans. Mice are small mammals and are therefore much more reliant on BAT and as such, also have relative more BAT as compared to humans.
9
Figure 1: Activated BAT takes up TG-derived fatty acids after lipolysis.
In response to cold, norepinephrine (NE) is released from the sympathetic nervous system and binds to the β3- adrenergic receptor on brown and beige adipocytes. This results in activation of adenylyl cyclase to produce cyclic adenosine monophosphate (cAMP) that activates protein kinase A (PKA). PKA phosphorylates and activates cAMP response element-binding protein (CREB) resulting in enhanced transcription of uncoupling protein 1 (Ucp-1) and peroxisome proliferator-activated receptor-gamma co-activator (Pgc-1α). Additionally, activation of PKA enhances the activity of lipolytic enzymes that are associated with the intracellular lipid droplets, leading to release of fatty acids (FA) that enter the mitochondria for β-oxidation. The produced acetyl-CoA is used in the citric acid cycle where reducing equivalents are formed that donate electrons to the electron transport chain. UCP-1 is a proton channel that uncouples respiration, which results in the generation of heat instead of ATP. Upon prolonged activation of brown and beige adipocytes, intracellular lipid droplets become depleted and are replenished via the uptake of triglyceride (TG)-derived FA, mainly via selective delipidation of TRLs through the hydrolyzing action of lipoprotein lipase (LPL). The released FA are taken up by cluster of differentiation 36 (CD36) and FA transport proteins (FATP) and stored in the lipid droplets as TG. Furthermore, glucose is taken up as well via glucose transporter (GLUT) 1 and 4 and used for, amongst others, de novo lipogenesis. ATGL indicates adipose triglyceride lipase; DAG, diacylglycerol; HSL, hormone sensitive lipase; MAG, monoacylglycerol; MGL, monoglyceride lipase. From (18).
The amount of murine BAT comprises approximately 0.4-1% of their body weight, while this is only about 0.02% in humans (14). However, a few grams of BAT is estimated to increase total energy expenditure by 6-20% (15), indicating that, despite lower BAT volume in humans, activated BAT contributes to energy expenditure and holds therapeutic potential.
Since the effects of cold exposure on hepatic VLDL production apparently mask the effects of BAT activity on lipid metabolism, more specific methods to activate BAT are needed in order to investigate its exact contribution to TG metabolism in humans. Several drugs that activate BAT in mice, such as salsalate (11) and rimonabant (10) also reduce plasma TG levels in humans (16, 17). Although it needs to be elucidated whether the TG-lowering effects in humans are due to activated BAT, these drugs may be promising methods to activate BAT more specifically. It would be even more ideal to develop specific activators of BAT, which will be further discussed in section 2.1. Overall, future studies are necessary to investigate the effect of different, more specific BAT activators on plasma TG levels in (hypertriglyceridemic) humans.
2.2 Importance of activated BAT in cholesterol metabolism
We observed that BAT activation not only improves plasma TG levels, but also plasma cholesterol levels, at least in mice. As a consequence of increased TG-derived FA uptake by activated BAT, the formation and subsequent hepatic uptake of cholesterol-enriched TRL-remnants is accelerated. Via this pathway, BAT activation by β3-AR agonism not only alleviates hypertriglyceridemia but also hypercholesterolemia and atherosclerosis development in E3L.CETP mice (Fig. 2). The importance of an intact apoE-LDLR pathway in these effects of BAT activation was demonstrated by the fact that β3-AR agonism in Apoe-/- and Ldlr-/- mice reduced hypertriglyceridemia, but neither influenced hypercholesterolemia nor atherosclerosis development. In chapter 3 we showed that concomitant statin treatment, which increases the hepatic uptake of cholesterol-enriched TRL remnants, potentiated the cholesterol-lowering and anti-atherogenic effect of BAT activation by β3-AR agonism. We found increased hepatic uptake of TRL remnants upon the combination treatment and this was likely related to increased LDLR functionality, since LDLR protein content was unaffected. This could be due to compensatory PCSK9- mediated LDLR degradation, since PCSK9 levels in plasma and liver were increased upon statin treatment. LDLR degradation can be prevented by combining a PCSK9 inhibitor with BAT activation. Therefore, the combination of a PCSK9 inhibitor and a β3-AR agonist will likely be an even more powerful strategy to lower hyperlipidemia and atherosclerosis and is an interesting topic of future studies. TRL remnants are not only taken up by the liver via the LDLR, but may also be trapped by heparan sulfate proteoglycans (HSPGs) followed by subsequent internalization (19, 20). The contribution of this pathway was supported by increased gene expression of the HSPG Syndecan1 and Lpl in livers of E3L.CETP mice that were treated with statin on top of β3-AR agonism [Hoeke et al.,
9
Figure 2: BAT activation reduces hypercholesterolemia and atherosclerosis development.
A. When BAT is inactive, triglyceride (TG)-rich lipoproteins (TRLs) in the circulation donate small amounts of TG-derived fatty acids (FA) to BAT, leading to slow formation of TRL-remnants. While circulating, these TRL-remnants can infiltrate the vessel wall and induce atherosclerosis development. B. Upon BAT activation, we propose a model in which BAT takes up substantial amounts of TRL-derived FA, resulting in accelerated formation of cholesterol-rich remnants that acquire abundant quantities of apolipoprotein E (E). As a consequence, these remnants are efficiently cleared by the liver via the LDLR, thereby reducing hypercholesterolemia and atherosclerosis development. B indicates apoB. From (18).
data unpublished]. Moreover, HSPGs preferentially clear smaller TRL remnants that are enriched in apoE (19), and the formation of these particles is likely increased due to fast lipolysis of TRLs by activated BAT. This pathway may thus be important for the uptake of TRL remnants when statin treatment is combined with β3-AR agonism.
BAT activation in humans, by means of short-term cooling, increased serum total cholesterol levels (chapter 5). This is likely, at least in part, due to the increased number of VLDL particles, that also had a higher cholesterol content upon cooling.
The development of (more) specific BAT activators that can be used in humans is thus also of importance to investigate the exact contribution of activated BAT to cholesterol metabolism. Subjects should likely be treated for a longer duration with a specific BAT activator since the cholesterol-lowering effects of activated BAT are indirect and are therefore not expected to be as acute as the TG-lowering effects. Likewise, plasma cholesterol levels are only lowered in hyperlipidemic E3L.CETP mice that were treated with a β3-AR agonist for 2 weeks [Hoeke et al., data unpublished], while 24 hours of cold exposure already reduces plasma TG levels with 43% in wild-type mice (chapter 2). The notion that prolonged BAT activation may lower plasma cholesterol levels in humans is supported by the observation that subjects with detectable BAT have lower plasma total cholesterol and LDL-cholesterol levels as compared to subjects without detectable BAT (21). Furthermore, daily cold exposure of 20 min for 90 days reduced total cholesterol, LDL-cholesterol and body mass in hypercholesterolemic individuals, without differences in physical activity and food intake (5). Moreover, high BAT activity is associated with a reduced risk of CVD events (22). Although causality still has to be demonstrated, long- term BAT activation may thus alleviate hypercholesterolemia and atherosclerosis in humans.
BAT activation also has pronounced effects on HDL metabolism, which confers atheroprotective properties. In hyperlipidemic E3L.CETP mice, BAT activation by means of β3-AR agonism increased plasma HDL levels. More importantly, HDL functionality was also improved as evidenced by increased reverse cholesterol transport (RCT) in mice that were coldexposed or treated with a β3-AR agonist (23). Similar as in mice, we showed in chapter 5 that cold exposure improved HDL metabolism in humans.
More specifically, in healthy lean men cold exposure increased small HDL particles, as well as their cholesterol content and functionality. The increased HDL functionality upon cold exposure may be explained by the following mechanism. Cold-induced BAT
9
activation increases LPL-mediated hydrolysis of TG in TRLs, resulting in the generation of phospholipid-rich surface remnants. These remnants can fuse with lipid-poor apoAI by phospholipid transfer protein (PLTP). The formed small HDL particles are very efficient acceptors of cholesterol (23) via the transporter ATP-binding cassette transporter 1 (ABCA1). Indeed, the specific ABCA1-mediated cholesterol efflux was increased after cold exposure (chapter 5). Cholesterol within HDL is subsequently transported to the liver and used for the production of bile acids (BAs). BA are secreted into the intestines, where approximately 95% is reabsorbed and transported to the liver, while approximately 5%
is excreted via the feces (24). Of note, cholesterol can also be directly excreted from the body via the bile. In line with the improved HDL functionality upon BAT activation, HDL- cholesterol exposure correlated negatively with the atherosclerotic lesion size in E3L.
CETP mice treated with a β3-AR agonist (23). BAT activation may thus not only protect from CVD by lowering plasma (V)LDL-cholesterol levels, but also by improving HDL functionality.
2.3 Human BAT, more beige than brown?
Rodents primarily exhibit BAT in the interscapular and subscapular region, and these depots mainly consist of classical brown adipocytes. Adult humans primarily exhibit BAT depots in the supraclavicular and neck region as well as along the spine, which was shown by using the glucose tracer [18F]fluorodeoxyglucose (FDG) followed by PET-CT scan (25). For a long time it remained elusive whether the adipocytes in these depots are brown, beige or both. Murine beige cells express high levels of Cd137, Tbx1 and Tmem26, which are practically not expressed in brown and white adipocytes (26). Using this beige signature, one study demonstrated that superficial BAT depots in humans are mainly composed of beige adipocytes while the deeper depots contain mostly brown adipocytes (27). Yet, biopsies from the supraclavicular region of humans showed that human BAT most closely resembles murine beige adipocytes (26). Likewise, the gene signature of human differentiated clonal brown adipocyte cultures also presented more similarity to mouse beige adipocytes (28). Most likely, human BAT thus consists of primarily beige adipocytes.
Mice and humans also possess WAT depots that can be browned in response to several stimuli including cold exposure and β3-AR agonism. In mice, especially the subcutaneous (s)WAT depot is susceptible to browning (4). Of note, the uptake of TG- derived FA by sWAT is increased upon browning as induced by β3-AR agonism (chapter
3). Inhibition of intracellular lipolysis in classical BAT depots does not influence whole- body thermogenesis in cold-exposed mice due to compensatory browning of WAT (29), suggesting that beige adipocytes have similar thermogenic capacity as classical brown adipocytes. Although the exact contribution of beige cells to whole-body FA combustion remains to be determined, browned WAT thus has the capacity to contribute to energy and lipid metabolism in mice. This is of clinical relevance since humans who will likely benefit most from BAT activation will be overweight or obese, and will thus have large amounts of WAT that have the potential to be browned. In this respect, it should be noted that, in contrast to mice, human subjects who were cold-exposed for 10 days did not show browning of their sWAT (30). Rather, perirenal fat (31, 32) and omental fat (33) are susceptible to browning. These beige adipocytes within browned WAT likely have a similar phenotype and genetic signature as beige cells in the supraclavicular region (i.e.
BAT). However, further characterization of human beige adipocytes in browned WAT is challenging, as the perirenal and omental WAT depots are difficult to access for biopsies.
2.4 Optimizing methods for brown and beige adipocyte detection
Methods to quantify BAT are mainly related to substrate utilization by brown or beige adipocytes. Besides TG-derived FA, activated BAT also takes up large amounts of glucose.
Therefore, PET-CT using the glucose tracer [18F]FDG is currently the golden standard to visualize and quantify (cold-) activated BAT in humans (25, 34, 35), and is also used to visualize and quantify BAT in mice (36, 37). In humans, BAT activation improves both glucose tolerance and insulin sensitivity (38, 39). In line with these findings, prolonged cold exposure in humans increases insulin sensitivity in BAT-positive, but not BAT-negative, subjects (40). Likewise, human subjects with a low TG content in the supraclavicular region (i.e. indicative for more active BAT) have lower fasting glucose levels and are more insulin sensitive (41). Direct evidence that BAT is an insulin-sensitive organ comes from studies that show that insulin enhances [18F]FDG uptake by BAT while the uptake of [18F]
FDG by BAT is impaired under insulin resistant conditions (42). Reduced glucose uptake by insulin resistant BAT is likely due to reduced translocation of the glucose transporter 4 (GLUT 4) to the plasma membrane, as shown in mice (43). This implies that PET-CT using [18F]FDG is not the most optimal method to determine BAT volume and activity, especially under insulin resistant conditions that develop as a consequence of weight gain and ageing.
Although [18F]FDG PET-CT is the current golden standard to assess BAT volume and
9
activity, it should be noted that BAT uses FA rather than glucose for oxidation (44).
Therefore, it can be reasoned that lipid tracers may serve as better measures of BAT volume and activity as compared to [18F]FDG, especially under insulin resistant conditions.
Indeed, insulin resistant conditions do not impair the uptake of the nonesterified FA (NEFA)-tracer [18F]fluoro-6-thia-heptadecanoic acid (FTHA) or oxidative metabolism of BAT as assessed by 11C-acetate (42). However, NEFA ([18F]FTHA) likely represent a minor portion of the plasma FA pool that is taken up by brown adipocytes, given that the uptake of FA is highly dependent on LPL (3). Therefore, further development of clinical tracers for visualization and quantification of the uptake of TG-derived FA from TRLs should be encouraged to provide a more robust measure of BAT activity that is probably independent of insulin sensitivity. In fact, in mice we use [3H]TO-labeled TRL-like particles to determine the uptake of TG-derived FA. The uptake of this radiolabel by various organs can be quantified and has been instrumental to reveal that activated murine BAT selectively takes up large amounts of TG-derived FA (chapter 2). Development of a similar method for humans, e.g. by esterification of [18F]FTHA into TG with subsequent incorporation into TRL-like particles as a novel tracer for PET-CT, may therefore lead to an improved method to assess BAT activity as compared to [18F]FDG or [18F]FTHA.
BAT activity is dependent on liberation of FA from the intracellular TG pool within brown adipocytes. This is demonstrated by the observation that inhibition of intracellular lipolysis in cold-exposed humans reduces BAT activity and increases shivering (45). As a consequence of intracellular lipolysis in BAT, the fat fraction reduces. Ideally, (lipid- based) PET-CT tracers should thus be combined with measures of the fat fraction within BAT. The fat fraction can be measured with magnetic resonance imaging (MRI), magnetic resonance spectroscopy (MRS) or computed tomography (CT) measurements. Current MRI techniques are able to visualize the cold-induced reduction in TG content of BAT and identified that increased TG content in BAT is associated with decreased whole body insulin sensitivity in humans (46). Moreover, the fat fraction measured with CT relates to many markers of metabolic health (body mass index, plasma lipid levels, and whole body insulin sensitivity) (46), which indicates that the fat fraction within BAT may even be used as a marker for metabolic health.
Another interesting approach would be to identify other (non-invasive) predictors of BAT and browned WAT activity. Although (bio)markers for the activity of browned WAT are not identified yet, some parameters have been identified that may function as
predictors for BAT activity. The first potential candidate is cold-induced skin temperature of the supraclavicular area, i.e. one of the main areas where BAT is present, which can easily be measured with ibuttons that are placed on the skin (47, 48) or with 1H MRS (49). With both methods, cold-induced BAT temperature correlated with [18F]FDG uptake by BAT, indicating that BAT temperature may be a good alternative for [18F]FDG PET-CT.
Secondly, plasma lysophosphatidylcholines (LysoPCs) may be used as a biomarker of BAT activity. LysoPCs are products of RCT by HDL, which is improved in mice (23) and humans (chapter 5). Following cholesterol efflux towards HDL, the effluxed cholesterol is esterified with a FA that is liberated from a phosphatidylcholine, a process that is mediated by lecithin:cholesterol acyltransferase (LCAT). This results in the formation of LysoPC that is released into the circulation (50-52). Indeed, plasma levels of LysoPC16:0 increased in lean subjects that were cold-exposed for 2 hours (53) or 2 days (23).
Moreover, LysoPC16:0 plasma levels correlated with BAT activity in subjects that were cold-exposed for 2 hours (53). Although more research is required to validate the relation between BAT activation and LysoPC16:0 in both healthy and obese subjects, LysoPC16:0 is a promising potential marker for BAT activity. Future studies are necessary to validate the above-described, and novel biomarkers, and find additional biomarkers by using broader platforms, for brown and beige adipocyte activity in men.
3. IMPLEMENTING BAT ACTIVATION IN THE CLINIC
3.1 Development of specific activators of brown and beige cells
Before BAT activation can be implemented in the clinic as a therapeutic strategy to improve cardiometabolic diseases, we first need specific agonists to activate brown and beige adipocytes. Research into compounds in mice and humans was initially only focused on activators of brown adipocytes, although these compounds often also induce browning of WAT. The mostly used pharmacological method to activate BAT in mice is via β3-AR agonism (chapter 3, 4), which mimics the action of noradrenalin. In humans, the potential of noradrenalin to activate BAT and increase energy expenditure was demonstrated in patients that suffer from a pheochromocytoma, a catecholamine- producing neuroendocrine tumour (54). Patients have increased uptake of [18F]FDG in the classical brown fat depots, as well as perirenal fat (31, 32) and omental fat (33), accompanied by highly increased energy expenditure. After removal of the tumour, [18F]
FDG uptake by browned WAT was abolished (31, 32). Specific pharmacological activation
9
of the β3-AR by a single administration of the β3-AR agonist mirabegron, that is currently on the market to treat overactive bladder, also increases BAT activity as well as energy expenditure in healthy subjects (55). The β3-AR thus certainly has therapeutic potential as drug target to activate BAT in humans. It should be mentioned that, although β3- AR agonism is generally regarded as a way to selectively activate BAT, the β3-AR is also present on WAT (56), where its activation induces intracellular lipolysis and browning, and on immune cells (57), for which the function has not been elucidated as yet.
Since human BAT has a more beige phenotype than that of mice and humans have high amounts of WAT that can be browned, attention has also focussed on the identification of compounds that induce browning. However, the identification and subsequent development of pharmacological compounds that specifically cause browning in humans may be more difficult than the development of compounds that activate human BAT. It is believed that beige adipocytes occur after proliferation of precursor cells into beige cells, or via transdifferentiation of white adipocytes into beige adipocytes. A prerequisite for the latter is that white adipocytes have the ability to differentiate into beige cells.
It was recently demonstrated in mice that aging induces a senescence-like phenotype which impairs the ability of white adipocytes to differentiate into beige adipocytes (58).
This senescence-like phenotype could be reversed by targeting the p38/MAPK pathway (58) which has also been successful in blunting age-induced skeletal muscle dysfunction and stimulating skeletal muscle renewal (59). Upon reversal, cells of aged mice and humans differentiated successfully towards beige adipocytes that were also functional (58). Since middle-aged subjects will likely benefit most from browning/BAT activation, these patients should ideally first be screened for this senescence-like phenotype in order to estimate the success rate of such a treatment. If necessary, patients should first receive therapeutics that delay or even reverse the process of aging so that browning can be achieved.
Possible agents that induce browning include thyroid hormones (60), BAs (61) and FGF21 (62). However, these conclusions are derived from preclinical studies and the relevance for humans still needs to be demonstrated. In order to develop more compounds that induce browning, future studies may use the genetic signature of beige cells to identify which genes are required for differentiation into a beige cell. A similar approach was previously used in mice. Treatment with bone morphogenetic protein 7 (BMP7), an inducer of both brown and beige adipocyte differentiation, increased BAT activity,
browning of WAT and improved plasma lipid and glucose levels (63). Although BMP7 treatment in humans has negative side effects related to excess bone formation and is therefore restricted in its clinical use (64), this example does show the potential of stimulating the presence of ‘beiging’ genes to induce browning.
The transition from bench to bedside requires initial testing in preclinical models, which is time-consuming and limits fast development of pharmacological compounds. The development of an in vitro high through-put system that also translates to human BAT would therefore be a big step forward. A promising accomplishment is the successful differentiation of human pluripotent stem cells into beige adipocytes, as evidenced by increased gene expression of beige adipocyte markers. Moreover, these cells can also be transplanted into mice (65), providing the opportunity to test compounds further in vivo. Advances that facilitate fast(er) development of drug targets should be further developed and implemented in research.
Not only the identification of pharmacological compounds is of importance, but also the route of administration. According to advertisements, BAT can be activated by wearing a cooling vest (e.g. Thin Ice). Although this likely activates BAT, it can be questioned whether patients will comply to wearing such clothing since cold exposure is experienced as unpleasant. It would thus be better to activate BAT in a pharmacological manner. The most easy method to activate human BAT is by systemic administration of pharmacological compounds. However, a disadvantage of this approach is that relatively higher dosages in the circulation are required to reach the optimal dose in the brown or beige adipocyte itself. A more direct method is by the use of microneedle patches that can be inserted in the subcutaneous adipose tissue. In mice, this approach effectively delivered agents that induce browning and improved their metabolic phenotype (66).
In humans, these microneedle patches can be incorporated in a vest that patients can wear during (periods of ) the day. Wearing such a vest in the morning will likely give the best results, since BAT is most active, and prone to further activation, in the beginning of the wakeful period [van den Berg and Kooijman et al., unpublished]. Such an approach holds therapeutic potential in humans, although it should first be established whether drug delivery to sWAT is sufficiently effective, since this depot may be less susceptible to browning.
9
3.2 Combining BAT activation with other treatments
BAT activation alone will likely not be sufficient for the treatment or prevention of CVD and its comorbidities, and should therefore ideally be combined with other therapies.
Combining BAT activation with strategies that increase the hepatic uptake of TRL remnants, such a statin treatment (chapter 3) or PCSK9 inhibitors (discussed in section 1.1), are promising options. Another approach would be to use a BA sequestrant, which is already implemented in the clinic for patients with hypercholesterolemia and indirectly increases the hepatic uptake of TRL remnants. BA sequestrants inhibit BA reabsorption in the intestines and therefore lead to more excretion of BAs via the feces. In order to compensate for the increased BA excretion, hepatic BA production is increased. Since cholesterol is the main substrate for BA production, the uptake of hepatic TRL remnants is increased. We showed in chapter 4 that combining BAT activation with the BA sequestrant colesevelam further reduces hepatic cholesterol and plasma nonHDL-C levels as compared to BAT activation alone in E3L.CETP mice. This indicates that, at least in mice, this combination may also further reduce atherosclerosis development, which is currently under investigation.
Another approach would be to combine BAT activation with drugs that reduce food intake, which will likely further reduce body fat and plasma lipid and glucose levels.
Activating the glucagon-like peptide-1 (GLP1) receptor or inhibition of the cannabinoid 1 receptor (CB1R) may be targets to reduce food intake. GLP1 is an intestinal hormone that is released after ingestion of a meal and stimulates meal-related insulin secretion, inhibits glucagon release, and delays gastric emptying. Various drugs are on the market that mimic GLP1, such as GLP1 receptor agonists (e.g. exenatide), or increase the amount of available endogenous GLP1, such as dipeptidyl peptidase-4 (DPP4) inhibitors (e.g.
sitagliptin) that inhibits degradation of GLP1. In addition to improved glucose tolerance, exenatide treatment also causes weight loss due to increased satiety. Moreover, studies in mice showed that central GLP1 receptor agonism also increases BAT activity (67).
GLP1R agonism is thus a promising avenue to pursue in the search for human BAT- activating strategies.
While GLP1 is located in the gastrointestinal tract, the CB1R is located both in the brain and periphery. The CB1R is a component of the endocannabinoid system, which controls food intake and energy metabolism. A decade ago, the potent CB1R antagonist rimonabant came on the market to treat obesity. This drug indeed successfully reduced
food intake and induced weight loss in humans (16, 68), but was removed from the market in 2008 due to psychiatric side effects, which were likely due to inhibition of central CB1R. Studies in mice showed that strict peripheral inhibition of the CB1R is sufficient to improve increase BAT activity and reduce weight gain and plasma lipid levels (10).
Therefore, peripheral CB1R inhibition may still be a promising strategy to activate BAT without psychiatric side effects. Whether GLP-1 receptor agonism and peripheral CB1R antagonism also increase BAT activity in humans remains the be elucidated. However, even if these strategies do not increase BAT activity, the combination of BAT activation by other tools with inhibition of satiety as induced by these strategies is a promising combination to improve cardiometabolic diseases.
Yet, pharmacological treatments should always be combined with guidelines for a healthier lifestyle, e.g. a healthy diet and daily exercise. Many studies showed that it is difficult for patients to comply to a diet, and even more difficult to maintain their body weight after weight loss. More efforts should be undertaken to guide patients to a healthier lifestyle, as treating patients with medication against CVD and comorbidities, while patients themselves continue an unhealthy lifestyle is certainly not the desired future perspective. One possible way is to organize support groups, including a dietician and psychologist, where patients together can talk about their experiences and feel supported by each other. This may increase compliance during weight loss and change behaviour in such a way that they can maintain a healthy lifestyle and lower body weight.
4. THE IMMUNE SYSTEM: TRANSLATIONAL CHALLENGES AND FUTURE DIRECTIONS
4.1 Translational challenges in immune research
It is widely accepted that inflammation is a major contributor to CVD and many studies aim at reducing the development of atherosclerosis by lowering the pro-inflammatory state. Inflammation is a multifactorial process due to numerous interactions between components of the immune system. Without doubt, we still have insufficient knowledge about these interactions. The complexity of the immune system is exemplified by the compensatory mechanisms that take place upon deletion or overexpression of components in the immune system. In chapter 7 and 8 we showed that deletion of hematopoietic Dectin-2 does not influence atherosclerosis development. Although we
9
cannot rule out that the role of Dectin-2 activation in atherosclerosis development can be neglected, this may be due by compensatory upregulation or activation of other receptors within the CLR family, such as Dectin-1 or Mincle.
The CANTOS trial is one of the first studies showing that attenuation of the pro- inflammatory response, via IL-1β antagonism, attenuates the development of CVD (69).
However, most of the compounds that were successful in preclinical models and have been tested in clinical trials are not implemented in clinic practice (70-72). There may be a few reasons for the poor translation of preclinical results to the human situation. First, in preclinical research publication bias occurs, leading to underrepresentation of neutral or negative studies (73). Positive effects of anti-inflammatory strategies may therefore be overestimated. Secondly, the inflammation-driven process of atherosclerosis is different between preclinical models and men. For example, mice do normally not develop atherosclerosis and the use of genetically modified models is therefore required. In addition, atherosclerosis development in these mouse models is mainly lipid-driven, i.e.
due to accumulation of (V)LDL-cholesterol, rather than inflammation-driven. To establish the low-grade inflammation in these models that drives atherosclerosis development, specific stimuli are used that are probably suboptimal. For example, in chapter 6 we established low-grade inflammation by using a Western-type diet with a high cholesterol content (i.e. 1%). However, this diet also dramatically increased plasma lipids which contributed to atherosclerosis development, and may have masked more subtle effects of low-grade inflammation on the development of atherosclerosis. Another model that may be used to increase (local) low-grade inflammation is cuff-induced atherosclerosis.
In this model, a non-constricting polyethylene cuff is placed around the femoral artery, which causes a local pro-inflammatory response. As a consequence, mice develop stenosis with atherosclerosis, which can be prevented by inhibition of inflammation (74).
However, these mice are also fed a cholesterol-rich diet and as such, have high cholesterol levels, which are required to induce atherosclerosis (75). Mouse models that sufficiently develop immune-, rather than lipid-, driven atherosclerosis are thus still lacking.
Thirdly, the phenotype of the atherosclerotic lesions is different between mice and men.
In humans, plaque rupture and the subsequent formation of a thrombus that stops the blood flow, and thus oxygen supply, are the actual causes of cardiovascular events. Since unstable plaques are more prone to rupture, research that focusses on unstable plaques has clinical relevance. The pro-inflammatory behaviour of immune cells is the primary
cause for these unstable plaques. For example, activated macrophages and T-cells within the atherosclerotic plaque release cytokines that inhibit cap formation and proteases that digest the collagen within the plaque, thereby reducing plaque stability (76). Plaque rupture also causes the formation of atherothrombosis, which is characterized by the formation of a superimposed luminal thrombus on a ruptured or eroded atherosclerotic plaque, and further increases the risk for CVD events (77). In contrast to humans, mice have very stable plaques due to a relative thick fibrous cap and a relative low number of immune cells within the plaque. As a consequence, mice do not develop spontaneous atherothrombosis and do not die of cardiovascular complications. In fact, there are currently no good mouse models that develop unstable plaques (78), illustrating the need for the development of better mouse models to allow better translation of preclinical research. A good start is the development of models for atherothrombosis.
Interestingly, a recent study showed that Apoe−/− mice treated with small interfering RNAs against protein C, an important component of the coagulation system, developed low-incident spontaneous atherothrombosis (79). This approach is currently also tested in E3L.CETP mice [Berbée et al., unpublished], which may yield a valuable research model in which the effects of immune modulation on top of lipid modulation can be evaluated in relation to occurrence of atherothrombosis
4.2 Future directions in immune research
The effects of immune modulators on atherosclerosis development in preclinical studies thus poorly translate to clinical practice. This not only illustrates the need for better preclinical models, but also the need for more research in humans. In humans, the risk for cardiovascular complications is determined by the plaque vulnerability.
Although unstable plaques are often accompanied by high inflammation in the plaque (76), this inflammatory state is not evaluated in clinical practice. This is likely due to the fact that we only have very general markers for inflammation, such as CRP. In addition, the inflammatory state in the circulation does not necessarily reflect the inflammatory state in the atherosclerotic plaque (80). It is therefore necessary to develop better non- invasive techniques that can image the plaque composition. This may firstly lead to a better understanding of atherosclerosis development in humans and better evaluation of novel drugs that lower CVD. Secondly, this may ultimately lead to better treatment of patients. For example, the use of tracers that can be visualized by MRI could serve as such a non-invasive technique. Ideally, various tracers should be developed that are taken up by different components of the plaque, such as cells of the immune system
9
(macrophages, neutrophils, T-cells), smooth muscle cells and collagen to determine the phenotype of the plaque. In the future, the type of treatment may be based on the phenotype of the atherosclerotic lesion, and this would be an important step towards personalized medicine.
Strategies that deliver drugs locally at the site of inflammation will be an improvement in clinical practice. Anti-inflammatory drugs are almost always administered systemically and the disadvantage is that it is difficult to reach the required dosage of the drug at the site of inflammation (i.e. the lesion in the case of atherosclerosis) without adverse side effects. A more specific target of drug delivery systems may be monocytes and macrophages, which are important drivers of atherosclerosis development.
HDL is important in the process of RCT, which involves the efflux of cholesterol from macrophages in the atherosclerotic lesion towards HDL. Subsequently, the cholesterol is esterified in HDL and transported to the liver after which the cholesterol is excreted from the body via the bile. HDL nanoparticles may improve CVD via either increasing the cholesterol efflux capacity (81), thereby lowering plasma cholesterol levels, or via delivery of anti-inflammatory drugs into atherosclerotic lesions. In fact, the HDL-mimetic CER001 successfully localized in human plaques (82). Future research is required to assess whether CER001 can really be used as a drug delivery system in humans and to evaluate the efficacy of such a method.
5. CONCLUDING REMARKS
Hyperlipidemia is one of the main risk factors for the development of atherosclerosis.
Activated BAT improves hyperlipidemia via the uptake of TG-derived FA, which accelerates the formation and hepatic uptake of cholesterol-enriched TRL remnants.
Via this mechanism activated BAT reduces circulating cholesterol and alleviates atherosclerosis development, at least in mice. [18F]FDG PET-CT is currently the golden method to visualize BAT in humans. However, given that BAT combusts FA rather than glucose, lipid tracers may be better as compared to glucose tracers to assess BAT volume and activity, and should ideally be combined with measures of fat fraction within BAT as a second measure of BAT activity. In addition, non-invasive measures of BAT volume and/
or activity, such as supraclavicular skin temperature and circulating biomarkers, need further evaluation. In humans, BAT is mostly comprised of beige adipocytes, and large
amounts of WAT are waiting to be browned. In order to successfully use BAT activation in the clinic, novel therapeutic strategies should not only aim at activating classical BAT depots, but also at increasing browning and activating beige adipocytes. Moreover, to reach maximal CVD risk reduction, compounds that activate BAT and/or induce browning should be combined with other therapies to treat CVD including statins, PCSK9 inhibitors and/or BA sequestrants. In conclusion, BAT activation is a promising therapeutic avenue to lower cardiometabolic diseases in humans.
Inflammation is also an important risk factor for the development of atherosclerosis.
However, development of atherosclerosis in currently available mouse models is not sufficiently inflammation-driven, which probably hampers the translational value of inflammation-targeted interventions in mice and urges the need for improved preclinical models. Inflammation-related research would also benefit from tracers that can be used to phenotype the atherosclerotic lesion in humans, which may result in personalized treatment. Also, the efficiency of drugs that lower inflammation should be improved.
The use of drug delivery systems, such as HDL mimetics, that specifically target the lesion may be an encouraging approach to enhance drug efficiency and reduce negative side effects. In conclusion, better preclinical models, and more research in humans is required in order to increase our understanding of the inflammatory processes that contribute to atherosclerosis development. However, the recent success of the CANTOS trial demonstrates that there is a future for immune modulators besides lipid-modulating agents in the battle against cardiometabolic diseases.
9
REFERENCES
1. Jukema, J. W., C. P. Cannon, A. J. De Craen, R. G. Westendorp, and S. Trompet, The controversies of statin therapy: weighing the evidence. J. Am. Coll. Cardiol. 2012. 60: 875-881.
2. Chan, D. C., P. H. Barrett, and G. F. Watts, The metabolic and pharmacologic bases for treating atherogenic dyslipidaemia. Best Pract. Res. Clin. Endocrinol. Metab. 2014. 28: 369-385.
3. Bartelt, A., O. T. Bruns, R. Reimer, H. Hohenberg, H. Ittrich, K. Peldschus, M. G. Kaul, U. I. Tromsdorf, H.
Weller, C. Waurisch, A. Eychmuller, P. L. Gordts, F. Rinninger, K. Bruegelmann, B. Freund, P. Nielsen, M. Merkel, and J. Heeren, Brown adipose tissue activity controls triglyceride clearance. Nat. Med.
2011. 17: 200-205.
4. Berbee, J. F., M. R. Boon, P. P. Khedoe, A. Bartelt, C. Schlein, A. Worthmann, S. Kooijman, G. Hoeke, I. M. Mol, C. John, C. Jung, N. Vazirpanah, L. P. Brouwers, P. L. Gordts, J. D. Esko, P. S. Hiemstra, L. M.
Havekes, L. Scheja, J. Heeren, and P. C. Rensen, Brown fat activation reduces hypercholesterolaemia and protects from atherosclerosis development. Nat. Commun. 2015. 6: 6356.
5. De Lorenzo, F., M. Mukherjee, Z. Kadziola, R. Sherwood, and V. V. Kakkar, Central cooling effects in patients with hypercholesterolaemia. Clin. Sci. (Lond.) 1998. 95: 213-217.
6. Geerling, J. J., M. R. Boon, G. C. Van Der Zon, S. A. Van Den Berg, A. M. Van Den Hoek, M. Lombes, H.
M. Princen, L. M. Havekes, P. C. Rensen, and B. Guigas, Metformin Lowers Plasma Triglycerides by Promoting VLDL-Triglyceride Clearance by Brown Adipose Tissue in Mice. Diabetes 2014. 63: 880- 891.
7. Ait-Oufella, H., S. Taleb, Z. Mallat, and A. Tedgui, Recent advances on the role of cytokines in atherosclerosis. Arterioscler. Thromb. Vasc. Biol. 2011. 31: 969-979.
8. Bian, F., J. Cui, T. Zheng, and S. Jin, Reactive oxygen species mediate angiotensin II-induced transcytosis of low-density lipoprotein across endothelial cells. Int. J. Mol. Med. 2017. 39: 629-635.
9. Wolf, Y., S. Boura-Halfon, N. Cortese, Z. Haimon, S. H. Sar, Y. Kuperman, V. Kalchenko, A. Brandis, E. David, Y. Segal-Hayoun, L. Chappell-Maor, A. Yaron, and S. Jung, Brown-adipose-tissue macrophages control tissue innervation and homeostatic energy expenditure. Nat. Immunol.
2017. 18: 665-674.
10. Boon, M. R., S. Kooijman, A. D. Van Dam, L. R. Pelgrom, J. F. Berbee, C. A. Visseren, R. C. Van Aggele, A. M. Van Den Hoek, H. C. Sips, M. Lombes, L. M. Havekes, J. T. Tamsma, B. Guigas, O. C. Meijer, J. W.
Jukema, and P. C. Rensen, Peripheral cannabinoid 1 receptor blockade activates brown adipose tissue and diminishes dyslipidemia and obesity. FASEB J. 2014. 28: 5361-5375.
11. Van Dam, A. D., K. J. Nahon, S. Kooijman, S. M. Van Den Berg, A. A. Kanhai, T. Kikuchi, M. M.
Heemskerk, V. Van Harmelen, M. Lombes, A. M. Van Den Hoek, M. P. De Winther, E. Lutgens, B.
Guigas, P. C. Rensen, and M. R. Boon, Salsalate activates brown adipose tissue in mice. Diabetes 2014. 64: 1544-1554.
12. Mantha, L. and Y. Deshaies, beta-Adrenergic modulation of triglyceridemia under increased energy expenditure. Am. J. Physiol. 1998. 274: R1769-R1776.
13. Blondin, D. P., H. C. Tingelstad, C. Noll, F. Frisch, S. Phoenix, B. Guerin, E. E. Turcotte, D. Richard, F. Haman, and A. C. Carpentier, Dietary fatty acid metabolism of brown adipose tissue in cold- acclimated men. Nat. Commun. 2017. 8: 14146.
14. Geisler, J. G., Targeting energy expenditure via fuel switching and beyond. Diabetologia 2011. 54:
237-244.
15. Towsend, K. L. and Y.-H. Tseng, Of mice and men: novel insights regarding constitutive and recruitable brown adipocytes. Int. J. Obes. Suppl. 2015. 5: S15-S20.
16. Despres, J. P., A. Golay, and L. Sjostrom, Effects of rimonabant on metabolic risk factors in overweight patients with dyslipidemia. N. Engl. J. Med. 2005. 353: 2121-2134.
17. Goldfine, A. B., V. Fonseca, K. A. Jablonski, L. Pyle, M. A. Staten, and S. E. Shoelson, The effects of salsalate on glycemic control in patients with type 2 diabetes: a randomized trial. Ann. Intern. Med.
2010. 152: 346-357.
18. Hoeke, G., S. Kooijman, M. R. Boon, P. C. Rensen, and J. F. Berbee, Role of Brown Fat in Lipoprotein Metabolism and Atherosclerosis. Circ. Res. 2016. 118: 173-182.
19. Foley, E. M., P. L. Gordts, K. I. Stanford, J. C. Gonzales, R. Lawrence, N. Stoddard, and J. D. Esko,
Hepatic remnant lipoprotein clearance by heparan sulfate proteoglycans and low-density lipoprotein receptors depend on dietary conditions in mice. Arterioscler. Thromb. Vasc. Biol. 2013.
33: 2065-2074.
20. Stanford, K. I., J. R. Bishop, E. M. Foley, J. C. Gonzales, I. R. Niesman, J. L. Witztum, and J. D. Esko, Syndecan-1 is the primary heparan sulfate proteoglycan mediating hepatic clearance of triglyceride-rich lipoproteins in mice. J. Clin. Invest. 2009. 119: 3236-3245.
21. Matsushita, M., T. Yoneshiro, S. Aita, T. Kameya, H. Sugie, and M. Saito, Impact of brown adipose tissue on body fatness and glucose metabolism in healthy humans. Int. J. Obes. (Lond.) 2014. 38:
812-817.
22. Takx, R., A. Ishai, Q. A. Truong, M. H. Macnabb, M. Scherrer-Crosbie, and A. Tawakol, Supraclavicular Brown adipose tissue FDG uptake and cardiovascular disease. J. Nucl. Med. 2016. 57: 1221-1225.
23. Bartelt, A., C. John, N. Schaltenberg, J. F. P. Berbee, A. Worthmann, M. L. Cherradi, C. Schlein, J.
Piepenburg, M. R. Boon, F. Rinninger, M. Heine, K. Toedter, A. Niemeier, S. K. Nilsson, M. Fischer, S. L. Wijers, W. Van Marken Lichtenbelt, L. Scheja, P. C. N. Rensen, and J. Heeren, Thermogenic adipocytes promote HDL turnover and reverse cholesterol transport. Nat. Commun. 2017. 8:
15010.
24. Kuipers, F., V. W. Bloks, and A. K. Groen, Beyond intestinal soap--bile acids in metabolic control. Nat.
Rev. Endocrinol. 2014. 10: 488-498.
25. Van Marken Lichtenbelt, W. D., J. W. Vanhommerig, N. M. Smulders, J. M. Drossaerts, G. J. Kemerink, N. D. Bouvy, P. Schrauwen, and G. J. Teule, Cold-activated brown adipose tissue in healthy men. N.
Engl. J. Med. 2009. 360: 1500-1508.
26. Wu, J., P. Bostrom, L. M. Sparks, L. Ye, J. H. Choi, A. H. Giang, M. Khandekar, K. A. Virtanen, P. Nuutila, G. Schaart, K. Huang, H. Tu, W. D. Van Marken Lichtenbelt, J. Hoeks, S. Enerback, P. Schrauwen, and B. M. Spiegelman, Beige adipocytes are a distinct type of thermogenic fat cell in mouse and human. Cell 2012. 150: 366-376.
27. Cypess, A. M., A. P. White, C. Vernochet, T. J. Schulz, R. Xue, C. A. Sass, T. L. Huang, C. Roberts-Toler, L. S. Weiner, C. Sze, A. T. Chacko, L. N. Deschamps, L. M. Herder, N. Truchan, A. L. Glasgow, A. R.
Holman, A. Gavrila, P. O. Hasselgren, M. A. Mori, M. Molla, and Y. H. Tseng, Anatomical localization, gene expression profiling and functional characterization of adult human neck brown fat. Nat.
Med. 2013. 19: 635-639.
28. Shinoda, K., I. H. Luijten, Y. Hasegawa, H. Hong, S. B. Sonne, M. Kim, R. Xue, M. Chondronikola, A. M. Cypess, Y. H. Tseng, J. Nedergaard, L. S. Sidossis, and S. Kajimura, Genetic and functional characterization of clonally derived adult human brown adipocytes. Nat. Med. 2015. 21: 389-394.
29. Shin, H., Y. Ma, T. Chanturiya, Q. Cao, Y. Wang, A. K. G. Kadegowda, R. Jackson, D. Rumore, B. Xue, H. Shi, O. Gavrilova, and L. Yu, Lipolysis in Brown Adipocytes Is Not Essential for Cold-Induced Thermogenesis in Mice. Cell Metab. 2017. 26: 764-777.
30. Van Der Lans, A. A., J. Hoeks, B. Brans, G. H. Vijgen, M. G. Visser, M. J. Vosselman, J. Hansen, J. A.
Jorgensen, J. Wu, F. M. Mottaghy, P. Schrauwen, and W. D. Van Marken Lichtenbelt, Cold acclimation recruits human brown fat and increases nonshivering thermogenesis. J. Clin. Invest. 2013. 123:
3395-3403.
31. Kuji, I., E. Imabayashi, A. Minagawa, H. Matsuda, and T. Miyauchi, Brown adipose tissue demonstrating intense FDG uptake in a patient with mediastinal pheochromocytoma. Ann. Nucl.
Med. 2008. 22: 231-235.
32. Yamaga, L. Y., A. F. Thom, J. Wagner, R. H. Baroni, J. T. Hidal, and M. G. Funari, The effect of catecholamines on the glucose uptake in brown adipose tissue demonstrated by (18)F-FDG PET/
CT in a patient with adrenal pheochromocytoma. Eur. J. Nucl. Med. Mol. Imaging 2008. 35: 446-447.
33. Frontini, A., A. Vitali, J. Perugini, I. Murano, C. Romiti, D. Ricquier, M. Guerrieri, and S. Cinti, White- to-brown transdifferentiation of omental adipocytes in patients affected by pheochromocytoma.
Biochim. Biophys. Acta 2013. 1831: 950-959.
34. Cypess, A. M., S. Lehman, G. Williams, I. Tal, D. Rodman, A. B. Goldfine, F. C. Kuo, E. L. Palmer, Y. H.
Tseng, A. Doria, G. M. Kolodny, and C. R. Kahn, Identification and importance of brown adipose tissue in adult humans. N. Engl. J. Med. 2009. 360: 1509-1517.
35. Virtanen, K. A., M. E. Lidell, J. Orava, M. Heglind, R. Westergren, T. Niemi, M. Taittonen, J. Laine, N. J.
Savisto, S. Enerback, and P. Nuutila, Functional brown adipose tissue in healthy adults. N. Engl. J.
Med. 2009. 360: 1518-1525.
9
36. Mirbolooki, M. R., S. K. Upadhyay, C. C. Constantinescu, M. L. Pan, and J. Mukherjee, Adrenergic pathway activation enhances brown adipose tissue metabolism: a [(1)(8)F]FDG PET/CT study in mice. Nucl. Med. Biol. 2014. 41: 10-16.
37. Park, J. W., K. H. Jung, J. H. Lee, C. H. Quach, S. H. Moon, Y. S. Cho, and K. H. Lee, 18F-FDG PET/CT Monitoring of beta3 Agonist-Stimulated Brown Adipocyte Recruitment in White Adipose Tissue.
J. Nucl. Med. 2015. 56: 153-158.
38. Hanssen, M. J., J. Hoeks, B. Brans, A. A. Van Der Lans, G. Schaart, J. J. Van Den Driessche, J. A.
Jorgensen, M. V. Boekschoten, M. K. Hesselink, B. Havekes, S. Kersten, F. M. Mottaghy, W. D. Van Marken Lichtenbelt, and P. Schrauwen, Short-term cold acclimation improves insulin sensitivity in patients with type 2 diabetes mellitus. Nat. Med. 2015. 21: 863-865.
39. Iwen, K. A., J. Backhaus, M. Cassens, M. Waltl, O. C. Hedesan, M. Merkel, J. Heeren, C. Sina, L.
Rademacher, A. Windjager, A. R. Haug, F. W. Kiefer, H. Lehnert, and S. M. Schmid, Cold-Induced Brown Adipose Tissue Activity Alters Plasma Fatty Acids and Improves Glucose Metabolism in Men. J. Clin. Endocrinol. Metab. 2017. 102: 4226-4234.
40. Chondronikola, M., E. Volpi, E. Borsheim, C. Porter, P. Annamalai, S. Enerback, M. E. Lidell, M. K.
Saraf, S. M. Labbe, N. M. Hurren, C. Yfanti, T. Chao, C. R. Andersen, F. Cesani, H. Hawkins, and L. S.
Sidossis, Brown adipose tissue improves whole-body glucose homeostasis and insulin sensitivity in humans. Diabetes 2014. 63: 4089-4099.
41. Raiko, J., M. Holstila, K. A. Virtanen, J. Orava, V. Saunavaara, T. Niemi, J. Laine, M. Taittonen, R. J.
Borra, P. Nuutila, and R. Parkkola, Brown Adipose Tissue Triglyceride Content Is Associated with Decreased Insulin Sensitivity Independent of Age and Obesity. Diabetes Obes. Metab. 2015. 17:
516-519.
42. Blondin, D. P., S. M. Labbe, C. Noll, M. Kunach, S. Phoenix, B. Guerin, E. E. Turcotte, F. Haman, D.
Richard, and A. C. Carpentier, Selective Impairment of Glucose but Not Fatty Acid or Oxidative Metabolism in Brown Adipose Tissue of Subjects With Type 2 Diabetes. Diabetes 2015. 64: 2388- 2397.
43. Garcia-Casarrubios, E., C. De Moura, A. I. Arroba, N. Pescador, M. Calderon-Dominguez, L. Garcia, L. Herrero, D. Serra, S. Cadenas, F. Reis, E. Carvalho, M. J. Obregon, and A. M. Valverde, Rapamycin negatively impacts insulin signaling, glucose uptake and uncoupling protein-1 in brown adipocytes. Biochim. Biophys. Acta 2016. 1861: 1929-1941.
44. Labbe, S. M., A. Caron, I. Bakan, M. Laplante, A. C. Carpentier, R. Lecomte, and D. Richard, In vivo measurement of energy substrate contribution to cold-induced brown adipose tissue thermogenesis. FASEB J. 2015. 29: 2046-2058.
45. Blondin, D. P., F. Frisch, S. Phoenix, B. Guerin, E. E. Turcotte, F. Haman, D. Richard, and A. C. Carpentier, Inhibition of Intracellular Triglyceride Lipolysis Suppresses Cold-Induced Brown Adipose Tissue Metabolism and Increases Shivering in Humans. Cell Metab. 2017. 25: 438-447.
46. Din, U., J. Raiko, T. Saari, V. Saunavaara, N. Kudomi, O. Solin, R. Parkkola, P. Nuutila, and K. A. Virtanen, Human Brown Fat Radiodensity Indicates Underlying Tissue Composition and Systemic Metabolic Health. J. Clin. Endocrinol. Metab. 2017. 102: 2258-2267.
47. Boon, M. R., L. E. Bakker, R. A. Van Der Linden, L. Pereira Arias-Bouda, F. Smit, H. J. Verberne, W.
D. Van Marken Lichtenbelt, I. M. Jazet, and P. C. Rensen, Supraclavicular skin temperature as a measure of 18F-FDG uptake by BAT in human subjects. PLoS One 2014. 9: e98822.
48. Haq, T., J. D. Crane, S. Kanji, E. Gunn, M. A. Tarnopolsky, H. C. Gerstein, G. R. Steinberg, and K. M.
Morrison, Optimizing the methodology for measuring supraclavicular skin temperature using infrared thermography; implications for measuring brown adipose tissue activity in humans. Sci.
Rep. 2017. 7: 11934.
49. Koskensalo, K., J. Raiko, T. Saari, V. Saunavaara, O. Eskola, P. Nuutila, J. Saunavaara, R. Parkkola, and K. A. Virtanen, Human Brown Adipose Tissue Temperature and Fat Fraction Are Related to Its Metabolic Activity. J. Clin. Endocrinol. Metab. 2017. 102: 1200-1207.
50. Eisenberg, S., Lipoproteins and lipoprotein metabolism. A dynamic evaluation of the plasma fat transport system. Klin. Wochenschr. 1983. 61: 119-132.
51. Glomset, J. A., The plasma lecithins: cholesterol acyltransferase reaction. J. Lipid Res. 1968. 9: 155- 167.
52. Magill, P., S. N. Rao, N. E. Miller, A. Nicoll, J. Brunzell, H. J. St, and B. Lewis, Relationships between the metabolism of high-density and very-low-density lipoproteins in man: studies of apolipoprotein
kinetics and adipose tissue lipoprotein lipase activity. Eur. J. Clin. Invest. 1982. 12: 113-120.
53. Boon, M. R., L. E. H. Bakker, C. Prehn, J. Adamski, M. J. Vosselman, I. M. Jazet, L. M. P. Arias-Bouda, W.
D. M. Van Lichtenbelt, K. W. Van Dijk, P. C. N. Rensen, and D. O. Mook-Kanamori, LysoPC-acyl C16:0 is associated with brown adipose tissue activity in men. Metabolomics 2017. 13: 48.
54. Lenders, J. W., G. Eisenhofer, M. Mannelli, and K. Pacak, Phaeochromocytoma. Lancet 2005. 366:
665-675.
55. Cypess, A. M., L. S. Weiner, C. Roberts-Toler, E. F. Elia, S. H. Kessler, P. A. Kahn, J. English, K. Chatman, S. A. Trauger, A. Doria, and G. M. Kolodny, Activation of Human Brown Adipose Tissue by a beta3- Adrenergic Receptor Agonist. Cell Metab. 2015. 21: 33-38.
56. Cannon, B. and J. Nedergaard, Brown adipose tissue: function and physiological significance.
Physiol. Rev. 2004. 84: 277-359.
57. Heidt, T., H. B. Sager, G. Courties, P. Dutta, Y. Iwamoto, A. Zaltsman, M. C. Von Zur, C. Bode, G. L. Fricchione, J. Denninger, C. P. Lin, C. Vinegoni, P. Libby, F. K. Swirski, R. Weissleder, and M.
Nahrendorf, Chronic variable stress activates hematopoietic stem cells. Nat. Med. 2014. 20: 754- 758.
58. Berry, D. C., Y. Jiang, R. W. Arpke, E. L. Close, A. Uchida, D. Reading, E. D. Berglund, M. Kyba, and J.
M. Graff, Cellular Aging Contributes to Failure of Cold-Induced Beige Adipocyte Formation in Old Mice and Humans. Cell Metab. 2017. 25: 166-181.
59. Cosgrove, B. D., P. M. Gilbert, E. Porpiglia, F. Mourkioti, S. P. Lee, S. Y. Corbel, M. E. Llewellyn, S. L.
Delp, and H. M. Blau, Rejuvenation of the muscle stem cell population restores strength to injured aged muscles. Nat. Med. 2014. 20: 255-264.
60. Lin, J. Z., A. J. Martagon, S. L. Cimini, D. D. Gonzalez, D. W. Tinkey, A. Biter, J. D. Baxter, P. Webb, J. A. Gustafsson, S. M. Hartig, and K. J. Phillips, Pharmacological Activation of Thyroid Hormone Receptors Elicits a Functional Conversion of White to Brown Fat. Cell Rep. 2015. 13: 1528-1537.
61. Fang, S., J. M. Suh, S. M. Reilly, E. Yu, O. Osborn, D. Lackey, E. Yoshihara, A. Perino, S. Jacinto, Y.
Lukasheva, A. R. Atkins, A. Khvat, B. Schnabl, R. T. Yu, D. A. Brenner, S. Coulter, C. Liddle, K. Schoonjans, J. M. Olefsky, A. R. Saltiel, M. Downes, and R. M. Evans, Intestinal FXR agonism promotes adipose tissue browning and reduces obesity and insulin resistance. Nat. Med. 2015. 21: 159-165.
62. Fisher, F. M., S. Kleiner, N. Douris, E. C. Fox, R. J. Mepani, F. Verdeguer, J. Wu, A. Kharitonenkov, J. S.
Flier, E. Maratos-Flier, and B. M. Spiegelman, FGF21 regulates PGC-1alpha and browning of white adipose tissues in adaptive thermogenesis. Genes Dev. 2012. 26: 271-281.
63. Boon, M. R., S. A. Van Den Berg, Y. Wang, J. Van Den Bossche, S. Karkampouna, M. Bauwens, M.
De Saint-Hubert, G. Van Der Horst, S. Vukicevic, M. P. De Winther, L. M. Havekes, J. W. Jukema, J. T.
Tamsma, G. Van Der Pluijm, K. W. Van Dijk, and P. C. Rensen, BMP7 activates brown adipose tissue and reduces diet-induced obesity only at subthermoneutrality. PLoS One 2013. 8: e74083.
64. Sreekumar, V., R. H. Aspera-Werz, G. Tendulkar, M. K. Reumann, T. Freude, K. Breitkopf-Heinlein, S.
Dooley, S. Pscherer, B. G. Ochs, I. Flesch, V. Hofmann, A. K. Nussler, and S. Ehnert, BMP9 a possible alternative drug for the recently withdrawn BMP7? New perspectives for (re-)implementation by personalized medicine. Arch. Toxicol. 2017. 91: 1353-1366.
65. Guenantin, A. C., N. Briand, E. Capel, F. Dumont, R. Morichon, C. Provost, F. Stillitano, D. Jeziorowska, J. P. Siffroi, R. J. Hajjar, B. Feve, J. S. Hulot, P. Collas, J. Capeau, and C. Vigouroux, Functional Human Beige Adipocytes From Induced Pluripotent Stem Cells. Diabetes 2017. 66: 1470-1478.
66. Zhang, Y., Q. Liu, J. Yu, S. Yu, J. Wang, L. Qiang, and Z. Gu, Locally Induced Adipose Tissue Browning by Microneedle Patch for Obesity Treatment. ACS Nano 2017. 11: 9223-9230.
67. Kooijman, S., Y. Wang, E. T. Parlevliet, M. R. Boon, D. Edelschaap, G. Snaterse, H. Pijl, J. A. Romijn, and P. C. Rensen, Central GLP-1 receptor signalling accelerates plasma clearance of triacylglycerol and glucose by activating brown adipose tissue in mice. Diabetologia 2015. 58: 2637-2646.
68. Pi-Sunyer, F. X., L. J. Aronne, H. M. Heshmati, J. Devin, and J. Rosenstock, Effect of rimonabant, a cannabinoid-1 receptor blocker, on weight and cardiometabolic risk factors in overweight or obese patients: RIO-North America: a randomized controlled trial. JAMA 2006. 295: 761-775.
69. Ridker, P. M., B. M. Everett, T. Thuren, J. G. Macfadyen, W. H. Chang, C. Ballantyne, F. Fonseca, J.
Nicolau, W. Koenig, S. D. Anker, J. J. P. Kastelein, J. H. Cornel, P. Pais, D. Pella, J. Genest, R. Cifkova, A.
Lorenzatti, T. Forster, Z. Kobalava, L. Vida-Simiti, M. Flather, H. Shimokawa, H. Ogawa, M. Dellborg, P. R. F. Rossi, R. P. T. Troquay, P. Libby, and R. J. Glynn, Antiinflammatory Therapy with Canakinumab for Atherosclerotic Disease. N. Engl. J. Med. 2017. 377: 1119-1131.
