The role of triglyceride lipases in cancer associated cachexia.

Cancer associated cachexia (CAC) is a complex multiorgan syndrome frequently associated with various forms of cancer. Affected patients suffer from a dramatic loss of skeletal muscle and adipose tissue. Most cases are accompanied by anorexia, and nutritional supplements are not sufficient to stop or reverse its course. CAC impairs many forms of therapeutic interventions and accounts for 15-20% of all deaths of cancer patients. Recently, several studies have recognized the importance of lipid metabolism and triglyceride hydrolysis as a major metabolic pathway involved in the initiation and/or progression of CAC. In this review, we explore the contributions of the triglyceride lipases to CAC and discuss various factors modulating lipase activity.

Cancer associated cachexia: definition, symptoms, and mediators

Cachexia is a multifactorial syndrome that occurs in many chronic and end stage diseases. Patients suffering from cancer frequently show signs of mild to severe forms of cachexia, which is referred to as cancer associated cachexia (CAC). The severity of cachexia varies depending on the tumor type, site, and stage [1]. CAC is prevalent in patients suffering from gastrointestinal and pancreatic adenocarcinoma, occurring in 80–90% of patients. It also arises in patients suffering from other carcinomas, as well as sarcomas, malignant lymphoma, and acute leukemia, although to a lesser extent [2,3]. The occurrence of cachexia relies heavily on the host response to tumor progression, including the activation of an immune response and associated signaling pathways [4]. Patients suffering from CAC show anorexia and a loss of body weight, white adipose tissue (WAT), and skeletal muscle mass, which lead to physical impairment [5]. Patients suffering from CAC often suffer from profound inflammation which escalates with loss of WAT weight and progression of CAC [6]. A negative energy balance complicates therapeutic interventions and nutritional supplements can only partially ameliorate the symptoms of cachexia [7-9]. Thus, it should not be confused with anorexia. This devastating condition is estimated to be the cause of death in up to 15% of cancer patients [1,10,11].

Despite considerable advancements in understanding the syndrome, effective approaches for managing CAC have been hindered by the lack of an exact definition and reproducible diagnostic criteria. After a comprehensive and thorough review of the literature, a panel of experts recently published a consensus definition of CAC that states ‘cancer cachexia is a multi-factorial syndrome characterized by an ongoing loss of skeletal muscle mass (with or without loss of fat mass) that cannot be fully reversed by conventional nutritional support and leads to functional impairment’ [12]. The panel recognized cachexia as a progressive disorder rather than a single event and classified the manifestation of cachexia into three stages: pre-cachexia, cachexia, and refractory cachexia. A recently developed scoring system will also be a valuable tool to stage cachectic cancer patients [13], and the diagnostic criteria proposed in these two studies are a big step forward for diagnosing and managing CAC. However, a thorough understanding of metabolic mechanisms that lead to the occurrence and progression of cachexia is imperative for designing more effective therapies to prevent, or at least ameliorate, the syndrome.

Cancer-induced weight loss is attributed mostly to loss of WAT and skeletal muscle mass. Patients suffering from cachexia lose on average 32% of their body weight, 75% of their skeletal muscle, as well as 85% of total body fat [14]. The loss of muscle mass might be a result of decreased protein synthesis, increased protein catabolism, or both depending on the tumor or stage of CAC. Recently, Augustsson et al. reported that both visceral and subcutaneous WAT is already significantly reduced in newly diagnosed CAC patients although they did not observe changes in lean body mass at that time [15]. This is consistent with previous observations indicating that at the onset of cachexia WAT volume is reduced before a loss of lean mass occurs [16] and may help in diagnosing early stages of cachexia. Importantly, recent reports recognize a prominent role for lipid metabolism and triglyceride lipases in the initiation and/or progression of CAC [17-20]. Therefore, in this review we focus on alterations of lipid catabolism in CAC and explore specifically the role of triglyceride lipases and their modulators in CAC.

Altered lipid metabolism in CAC

WAT is regarded as the main reservoir of lipids that are stored mostly in the form of triacylglycerol (TAG) and serve as a major source of energy. TAG levels in WAT are maintained by the dynamic equilibrium of two opposing key processes, lipogenesis and lipolysis. In the anabolic process of lipogenesis, glucose, which is transported into adipocytes mainly via GLUT4, is catabolized to acetyl-CoA, which is then acted upon by a series of enzymes and cofactors to generate fatty acids (FAs). Three FAs are then esterified to a single glycerol molecule to create a molecule of TAG and stored as an energy reserve in WAT. In addition to de novo lipogenesis, WAT imports FAs by catabolizing circulating lipoproteins into FA via membrane-bound lipoprotein lipase (LPL). These FAs are imported in the cytoplasm via the action of FA transporters and FA-binding proteins and are subsequently re-esterified into TAG. Conversely, TAG molecules stored in WAT are broken down into FAs and glycerol by a highly regulated process known as lipolysis, a process of three subsequent steps catalyzed by three essential lipases. Adipose triglyceride lipase (ATGL) catalyzes the rate-limiting first step, converting TAG into diacylglycerol (DAG) and a free FA. DAG is then acted upon by hormone sensitive lipase (HSL) to remove the second FA to form monoacylglycerol (MAG). The third and final enzyme, monoglyceride lipase (MGL) converts MAG into glycerol and a third free FA. Thus, three molecules of FA are released in this process that are then either re-esterified to form new TAG or shuttled to metabolically active tissues where they undergo fatty acid oxidation to generate energy or might be used for anabolic processes. Lipolysis is influenced by several additional factors that regulate the activities of these essential lipases, and both lipoloysis and lipogenesis not only regulate the homeostasis of WAT but also direct systemic energy production.

Recent studies in animal models of CAC and observation of human WAT have confirmed that fat loss in cachexia is a consequence of TAG depletion and not due to cell death [21]. The processes leading to the depletion of TAG reserves involve decreased lipogenesis, decreased FA transport, and, most prominently, increased lipolysis [5]. Trew and Begg were the first to identify a severe deregulation of lipogenesis in WAT using an animal tumor model [22]. This observation was confirmed in many subsequent studies, which also identified declines in the activities of key lipogenic enzymes in WAT, such as fatty acid synthase, citrate cleavage enzyme, and malic enzyme, associated with tumor progression in murine cachexia models [23]. In stark contrast to these results, lipogenesis was reported to be either upregulated in WAT of mice bearing colonic adenocarcinoma [24] or no significant change was detected in WAT of tumor bearing patients [21,25]. However, Notarnicola et al. recently showed that both mRNA levels and the activity of FA synthase are reduced in WAT adjacent to the tumor, compared with distant WAT, in samples from 32 colorectal cancer patients [26]. Differences in the anatomical origin of WAT and/or differences in cancer types might explain these discrepancies.

Elevated levels of serum TAG and cholesterol are often detected in patients suffering from CAC [5], and a marked increase in FA and glycerol concentrations are observed in cancer patients suffering from cachexia as compared to weight-stable cancer patients [27-30]. Murine models of CAC also show increased serum levels of FA and glycerol, suggesting increased TAG catabolism and delipidation of fat reserves [18]. Ryden et al. have confirmed these findings and state that increased lipolysis and not cell death or impaired lipogenesis is involved in WAT loss in cachectic patients [21]. Similarly, TAG hydrolase activity in visceral WAT of cancer patients shows a strong negative correlation with body mass index (BMI), whereas such a correlation is absent in patients without tumor [18]. Observations made by several other groups, including us, in patient samples and animal models of cachexia reinforced the hypothesis that increased lipolysis is a major determinant of adipose tissue loss in CAC [18,21,28,29]. This observation is also supported by a comprehensive transcriptomics analysis of adipose tissue describing a significant upregulation of genes involved in the FA degradation pathway, as well as genes involved in the tricarboxylic acid cycle (TCA) cycle and the oxidative phosphorylation (OXPHOS) pathway in CAC patients [31]. In conclusion, cancer shifts lipid metabolism to a catabolic state, thus decreasing the fat mass in cachectic patients. As a result, systemic lipid metabolism is severely disturbed which, in turn, might cause lipotoxic effects in other tissues including skeletal muscle (Box 1). Indeed Stephens et al. observed that intramyocellular lipid droplets increased in size and number in the course of cachexia progression in cancer patients [32]. This is supported by our observations of increased TAG concentrations in Musculus gastrocnemius of Lewis lung carcinoma (LLC) bearing cachectic mice [18]. Moreover, cancer impedes adipogenesis, which is evident by downregulation of key adipogenic factors: CCAAT/enhancer binding protein β (C/EBPβ), C/EBPα, peroxisome proliferator-activated receptor γ (PPARγ), and sterol regulatory element-binding protein-1c (SREBP-1c) [33]. Figure 1 illustrates various tumor associated mechanisms that may influence lipid homeostasis in WAT and lead to CAC.

Figure 1

Altered lipid metabolism in cancer associated cachexia (CAC). Malignant tumors release various cytokines including interleukin (IL)-6 and tumor necrosis factor-α (TNF-α), lipid-mobilizing factor (LMF)/Zinc-a2 glycoprotein (ZAG), and an unknown cachexia inducing factor (CIF) in circulation. All these factors act on white adipose tissue (WAT) and deregulate lipid metabolism. TNF-α acts through the TNF receptor (TNF-r) and downregulates G0S2 (G0/G1 switch gene 2), which binds to and negatively regulates adipose triglyceride lipase (ATGL) activity. Therefore, TNF-α can increase ATGL activity. ATGL, along with its coactivator CGI-58 (comparative gene identification-58), catabolizes the first step of lipolysis by converting triacylglycerol (TAG) present in lipid droplets to diacylglycerol (DAG). DAG is further acted upon by activated (phosphorylated) hormone sensitive lipase (HSL) and is converted into monoacylglycerol (MAG). As the final step of the lipolytic process, monoglyceride lipase (MAGL) converts MAG into glycerol (G) which can be released into circulation. One molecule of free fatty acid (FFA) is generated in each of the steps. HSL is phosphorylated by protein kinase A (PKA) or cGMP-dependent protein kinase 1 (cGK1). PKA can be activated by catecholamine or LMF/ZAG through various mechanisms. Similarly, glucocorticoids activate cGK1 by various processes. FA generated by increased lipolysis can be transported out of the adipocytes or β-oxidized in mitochondria. CIF induces cell death activator (CIDEA), which, in turn, increases the level of pyruvate dehydrogenase complex (PDC) and, hence, assists in β-oxidation. Along with TNF-α, IL-6 also induces lipolysis, although the mechanisms are unknown. In addition to inducing lipolysis, TNF-α downregulates lipogenesis and impairs FA uptake by inhibiting lipoprotein lipase (LPL) activity, leading to decreased TAG concentration in adipocytes and increased very-low-density lipoprotein/low-density lipoprotein (VLDL/LDL) levels in circulation. Increased lipolysis also attracts macrophages and results in macrophage infiltration in WAT often seen in cachexia. Factors in red are upregulated in cachexia, those in green are downregulated in cachexia, and factors shown in black have no change or have not yet been determined in cachexia. Solid lines indicate pathways confirmed in CAC and dotted lines expected but unconfirmed ones.

Triglyceride lipase involvement in cancer cachexia

Lipoprotein lipase (LPL)

LPL is a member of the triglyceride hydrolase family and is attached to the surface of endothelial cells most abundantly in WAT, skeletal muscle, and heart. LPL hydrolyzes TAG present in circulating very-low-density lipoproteins (VLDLs) and chylomicrons (lipoprotein particles very rich in triglycerides). Free FAs are subsequently transported directly into parenchymal cells. Increased circulating TAG, low-density lipoproteins (LDLs), and VLDL in cancer patients as well as in animal models of CAC suggest the downregulation of LPL activity.

Obeid and Emery found decreased LPL activity early during progressive tumor growth in mice [34], and decreased LPL activity is also reported in patients suffering from gastric and colorectal carcinoma, but not in breast carcinoma patients [35]. Interestingly, circulating levels of interleukin (IL)-6 did not correlate with LPL activity in these patients. Similarly, a decrease in LPL activity was noted in later stages of tumor cachexia in mice bearing MAC16 adenocarcinoma [36]. Leukemia inhibitory factor (LIF) and tumor necrosis factor-α (TNF-α) are both reported to decrease the activity as well as mRNA expression of LPL in cultured adipocytes [37,38]. Thus, increased circulating levels of LIF and TNF-α in CAC patients and mice models might explain decreased LPL activities. In addition, Nara-Ashizawa et al. have reported the presence of a factor that seems to be different from the well-known cytokines involved in cachexia, IL-1β, IL-6, IL-1, TNF-α, transforming growth factor-β (TGF-β), and LIF [39].

Treatment with benzafibrate, an antihyperlipidemic drug, has been shown to enhance LPL activity in tumor bearing rats, which resulted in the preservation of epididymal WAT and partial prevention of hyperlipidemia [40]. Similar effects have been shown for NO1886 and pronalrestat in B16 melanoma-induced cachexia [41,42]. These promising effects have, unfortunately, not yet been translated into successful treatment strategies for CAC patients.

Hormone sensitive lipase (HSL)

Although fasting-induced lipolysis in WAT had been known for several decades, HSL was the first lipase to be discovered that is responsive to fasting and regulated by catabolic hormones [43]. HSL has a relatively broad substrate specificity, hydrolyzing DAG, cholesterol ester, retinoic ester, TAG, MAG, as well as short chain carbonic esters and ester substrates [44], although several studies indicate that DAG is a much better substrate for HSL than TAG [44,45]. HSL is activated by phosphorylation mediated by protein kinase A (PKA), extracellular signal-regulated kinase (ERK), glycogen synthase kinase 3β (GSK-3β; previously known as GSK-4), calmodulin-dependent kinase II (CAMKII), as well as AMP-activated protein kinase (AMPK) [46-48]. Lipolysis by HSL is also affected by PKA-mediated phosphorylation of perlipin-1 at serines 81, 222, and 276, which induces HSL binding and provides HSL access to lipid droplets, thus increasing lipolysis [49,50].

Thompson et al. reported a twofold increase in HSL mRNA in WAT and a similar elevation of free FAs in serum from cancer patients, compared with normal controls [25], an observation supported by subsequent studies reporting significantly increased HSL mRNA and protein expression in WAT of patients in the early phases of CAC [51,52]. Increased HSL expression also corresponds with increased hormone-stimulated lipolysis in isolated fat cells from subcutaneous WAT of CAC patients that could be prevented by an HSL-specific inhibitor, BAY [52].

Adipose triglyceride lipase (ATGL)

ATGL was discovered by three independent groups in 2004 [53-55]. It belongs to a large family of patatin domain-containing proteins, preferentially hydrolyzes TAG, and, thus, catalyzes the first step of lipolysis that generates DAG and a molecule of FA. Recently, Yang et al. showed that overexpression of ATGL can compensate for HSL loss during basal or TNF-α-stimulated lipolysis [56]. ATGL is expressed in most human tissues with the highest levels in WAT and BAT (brown adipose tissue). Recent studies have convincingly established that ATGL catalyzes the rate-limiting step of lipolysis in human as well as murine adipose tissue [45].

ATGL levels are regulated by PPARγ via the PPAR responsive element (PPRE) present in its promoter [57], ATGL is also regulated directly by its cofactor comparative gene identification-58 (CGI-58) [58], a member of the α/β hydrolase-fold containing subfamily. It activates ATGL by direct interaction with the N-terminal domain of ATGL. The C terminus of ATGL appears to have an inhibitory effect on enzyme activity; it is, however, important for interactions with lipid droplets. Full activation of ATGL by CGI-58 requires both protein–protein interaction with ATGL as well as binding to a lipid droplet [44]. In addition to CGI-58, Yang et al. recently showed that the protein G0S2 (G0/G1 switch gene 2) also regulates enzymatic activity of ATGL by direct interaction [59]. Similar to ATGL, G0S2 is ubiquitously expressed and localizes to mitochondria, lipid droplets, the endoplasmic reticulum, and the cytosol. The N-terminal domain of G0S2 interacts with the patatin domain of ATGL, thereby restricting access of the enzyme to lipid droplets and thus inhibiting ATGL activity [59].

Agustsson et al. have shown that Atgl expression is somewhat enhanced in WAT of cachectic cancer patients; however, the increase was not significant [52]. Given that ATGL activity is post-translationally regulated by several factors, mRNA expression does not always correspond to enzyme activity. Using two murine cachexia models, our group recently reported that ATGL activity increases in parallel with tumor growth and WAT mobilization [18]. Atgl mice were protected from tumor-induced lipolysis, which preserved WAT mass. Analysis of WAT from cachectic cancer patients showed significantly increased ATGL activity that was strongly negatively correlated with BMI [18]. In line with previous reports, wild type mice bearing tumors presented with lipolysis and depletion of WAT prior to skeletal muscle loss. Interestingly, in addition to WAT, skeletal muscle was also preserved in tumor bearing Atgl mice. Of note, ATGL is also expressed in skeletal muscle and has been reported to play an important role in regulating muscle lipid metabolism [60-62]. Therefore, it remains to be investigated if the preservation of muscle is an indirect effect of reduced levels of circulating free FA and/or other factors released from WAT or if it is due to loss of ATGL activity in muscle itself.

Modulators and regulators of triglyceride lipases in cancer cachexia

Tumor necrosis factor-alpha (TNF-α)

Kawakami and Cerami noticed that lipopolysaccharide (LPS) sensitive C3H/HeN mice, when injected with LPS, suppressed LPL activity in WAT. In addition, they observed that serum of endotoxin sensitive mice that had been administered LPS suppressed LPL activity, even in endotoxin-resistant mice [63]. Cerami and his group later termed the factor produced in vitro by macrophages primed with LPS, as well as other microbial substances that suppressed LPL activity in isolated adipocytes, ‘cachectin’ [64]. Purified cachectin displayed tumor necrosis activity in vitro, suggesting that cachectin was homologous to TNF-α; indeed, N-terminal protein and DNA sequencing revealed that these are the same, highly conserved protein [65].

The onset of anorexia, anemia, weight loss, and loss of body fat and muscle protein after administration of sublethal doses to healthy animals suggests that TNF-α plays a central role in the development of cachexia. TNF-α has been detected in serum from 36.5% of pancreatic carcinoma patients, and the level inversely correlated with patient BMI and body weight [66]. TNF-α is reported to suppress the expression of LPL mRNA and LPL activity, along with stimulating lipolysis in vitro and in vivo [67]. The mechanism by which TNF-α increases lipolysis is still unknown, although several hypotheses have been proposed. Compared with lipolysis induced by hormones, lipolysis induced by TNF-α is a slow process, taking 6–12 h to reach a detectable level and achieving the maximum effect after 48 h [56]. Interestingly, TNF-α downregulates the expression of both ATGL and HSL in rodent adipocytes [68-71]. Given that ATGL and HSL are the major lipases present in adipocytes and drive 90% of TAG hydrolysis, this indicates an indirect regulation of lipolysis by TNF-α. In addition to regulating lipase expression, TNF-α induces phosphorylation of perilipin-1, most likely by an increase in cAMP levels that activates PKA [72]. This process might also induce phosphorylation of HSL, which accelerates its translocation to lipid droplets and, hence, increases lipolysis. Recently, however, Yang et al. showed that ATGL downregulation almost completely abolished glycerol release induced by TNF-α in 3T3-L1 adipocytes, whereas ablation of HSL or the ATGL coactivator CGI-58 had only a partial influence [56], suggesting an essential role for ATGL in lipolysis induced by TNF-α. In addition, a sharp decrease in G0S2 mRNA and protein levels was observed that could be blocked by inhibiting proteasomal degradation using the protease inhibitor MG-132. G0S2 overexpression also significantly decreased lipolysis induced by TNF-α [56].

Intraperitoneal administration of TNF-α antibodies in two mice models of CAC, LLC and MCG-101, prevented cachexia-associated features including the loss of WAT, lean mass, and body weight and significantly reduced tumor size, thus confirming previous reports on TNF-α involvement in CAC progression [73]. However, antibody treatment had no effect on profound anemia, hypoalbuminemia, or the increase in serum amyloid P concentrations. TNF receptor type I knockout mice with exponentially growing LLC showed reduced WAT loss compared with wild type controls [74], but the fact that CAC was not completely prevented in TNF receptor type I knockout mice led the authors to consider the involvement of other cytokines or factors in the onset of cachexia. Mahony et al. failed to identify TNF-α in tumor or serum from MAC16 bearing cachectic mice [75]. Moreover, circulating TNF-α levels in cancer patients vary considerably and generally do not correlate with weight loss in terminal cancer patients [5]. Because of the very short half-life of TNF-α and variations in sensitivities of the detection methods, it might, however, be premature to draw a final conclusion. It is also conceivable that the role of TNF-α in the progression of cachexia might be limited to pre-cachexia and early cachexia stages, but resolving the exact roles of TNF-α will require further studies.

Interleukins

Similar to TNF-α, IL-6 also induces lipolysis in 3T3-L1 adipocytes as well as in primary mouse and human adipocytes [76]. In addition, IL-6 decreases LPL activity as well as perlipin A and PPARγ mRNA expression in 3T3-L1 adipocytes. IL-6 treatment stimulates lipolysis by activating ERK, but it does not elevate cAMP levels and a PKA inhibitor had no effect on IL-6 stimulated lipolysis [77].

Strassman et al. reported increased serum levels of IL-6, but not TNF-α, in cachectic mice bearing C-26 murine colon adenocarcinoma, which are reduced once the tumors are resected [78]. In addition, monoclonal antibodies against IL-6, but not TNF-α, improved body weight and conserved WAT mass in C-26 bearing mice [78]. Recently, Tsoli et al. confirmed previous reports and attributed deregulation of lipid metabolism in BAT to elevated circulating IL-6 levels in C-26 bearing cachectic mice [79]. Patients with non-small-cell lung carcinoma (NSCLC) displaying weight loss showed a significant increase in circulating IL-6 compared with weight-stable NSCLC patients [80]. However, another study did not find a correlation between TNF-α, IL-1, IL-6, and weight loss of 61 terminal cancer patients [81]. Soda et al. reported that a C-26 clone which did not induce cachexia also enhanced IL-6 to the same level as the C-26 clone that induced severe cachexia [82]. In agreement, Inadara et al. reported that the factor causing depletion of adipose tissue in C-26 tumor bearing mice is different from IL-6 or TNF-α [83], suggesting that IL-6 cannot be the only factor causing cachexia in tumor bearing mice.

Zinc alpha glycoprotein (ZAG)

Aiming to characterize and isolate the lipid-mobilizing factor(s) present in serum of cancer patients or tumor bearing animals, Tisdale and coworkers identified a factor identical to a known protein, zinc α2-glycoprotein (ZAG), that is present in serum of mice bearing the cachexia-inducing MAC16 tumor as well as in urine and serum of CAC patients [84,85]. When injected into mice, the factor caused increased lipolysis, imitating tumor-induced cachexia [86]. ZAG is expressed by tumors that induce severe cachexia in patients as well as animal models of CAC [86]. It induces lipolysis and loss of WAT by stimulating adenylyl cyclase in a GTP-dependent process, which in turn activates HSL, thus increasing TAG hydrolysis and mobilizing both FAs and glycerol [5]. However, the effect of ZAG on ATGL, or the involvement of ATGL in ZAG-stimulated lipolysis, remains unknown. We identified a significant increase of ZAG in tumor bearing wild type and Atgl mice, but only wild type mice showed cachexia [18]. This warrants a detailed investigation of the potential role of ZAG in ATGL-mediated lipolysis.

In addition to various tumors, several tissues such as WAT, BAT, heart, skeletal muscle, liver, and lung also express and release ZAG. Both mRNA and protein expression of ZAG in WAT is increased in MAC16 bearing mice and correlates with progressive loss of body weight [87]. Accordingly, Zag mice gained weight and WAT mass when put on a high-fat diet, confirming the importance of ZAG in regulating WAT metabolism. Basal lipolysis is not affected in the adipocytes of Zag mice; however, they showed decreased sensitivity towards isoprenaline-induced lipolysis [88]. The effects of cachexia-inducing tumors on WAT of Zag mice would be very interesting to study.

G0/G1 switch gene 2 (G0S2)

G0S2 was first described by Russell and Forsdyke in lymphocytes as a protein highly expressed between the G0 and G1 phases of the cell cycle [89]. To date, there is conflicting evidence of its role in cell-cycle regulation [90,91]. G0S2 expression is not restricted to cancer tissue; it is expressed in almost all normal human and murine tissues. Zandbergen et al. nicely showed that the expression of G0S2 is tightly regulated by PPARα and PPARγ [92], as well as being influenced by nutritional status, insulin, and TNF-α [59]. Liu and colleagues recently reported that G0S2 interacts with cytosolic ATGL and impedes its interaction with TAG, thus inhibiting TAG hydrolysis [59]. Downregulation of G0S2 significantly increased lipolysis and ATGL activity without affecting ATGL expression [93]. G0S2, however, does not affect the expression or activity of HSL [59]. It is interesting to note that TNF-α/cachectin, which is elevated in the serum of CAC patients, can reduce G0S2 expression and thus increase lipolysis via changes in ATGL activity [56]. This supports the finding that ATGL is a key player in mobilizing TAG reserves of adipose depots during CAC, despite the fact that the expression of ATGL is mostly unaltered in WAT of CAC patients or animals. It would, therefore, be of great interest to investigate G0S2 expression levels in WAT of CAC patients or in CAC animal models.

Concluding remarks

CAC is now recognized to be a devastating syndrome in cancer patients that results in a significant decrease of WAT as well as muscle mass which cannot be reversed with classical nutritional interventions. Increased circulating cytokine levels and activated catabolic processes underline the major pathological changes observed. Although several factors and mechanisms associated with CAC have been identified in the past decades, answers to the basic question of how and why tumors cause cachexia remain elusive (Box 1). In this review, we summarize and put into perspective recent reports that indicate the importance of lipolysis and TAG hydrolases in the development of the syndrome. In CAC, increased activities of major TAG hydrolases in WAT, namely, ATGL and HSL, are observed, thus boosting lipolysis as evidenced by increased circulatory glycerol and FAs that might, in turn, harm other tissues and organs, an effect termed lipotoxicity (Box 2). In addition, a significant reduction in WAT LPL activity is reported, which explains increased TAG levels in CAC patients. In summary, CAC reduces WAT mass by downregulating anabolic metabolism while significantly enhancing the catabolic rates of lipids.

Of the greatest clinical relevance, CAC severely reduces skeletal muscle mass in cancer patients leading to atrophy and loss of muscle function. Several factors released by tumors, including various cytokines, induce proteolysis and muscle catabolism, leading to atrophy. Of note, recent reports point to the importance of WAT–muscle interactions in CAC. WAT is affected earlier than the effects of CAC can be observed in skeletal muscle. This might reflect a sequential progression of cachexia rather than a stochastic phenomenon. Both WAT and muscle have recently been recognized to have important endocrine functions secreting various factors commonly termed adipokines and myokines (Box 3). Fat–muscle crosstalk has been investigated intensively in metabolic syndromes such as obesity and diabetes, and reports frequently show lipotoxic cell death as well as insulin resistance as a result (Boxes 2 and 3). The existence and importance of fat–muscle crosstalk in the initiation and/or progression of CAC remains to be uncovered. Figure 2 summarizes the possible WAT–muscle crosstalk leading to skeletal muscle atrophy.

Figure 2

Importance of white adipose tissue (WAT)–skeletal muscle crosstalk in cancer associated cachexia (CAC) and/or lipotoxicity. Factors released by tumors induce lipolysis in WAT (Figure 1) and alter adipokine secretion by WAT. As outlined in Figure 1, tumor necrosis factor-α (TNF-α), through the TNF receptor and G0S2 (G0/G1 switch gene 2), regulates the activity of the lipases adipose triglyceride lipase (ATGL), hormone sensitive lipase (HSL), and monoglyceride lipase (MAGL). Leptin and adiponectin released by WAT act through the leptin receptor (LR) or the adiponectin receptor (ADIPOR) present on myocytes and activates AMP-activated protein kinase (AMPK), which, in turn, activates ATGL. Catecholamines signal through the β-adrenoceptor and phosphorylate HSL. Fatty acids (FAs) released by adipocytes as a result of increased lipolysis are transported into myocytes by FA transporters such as FATP1 and CD36. FAs are β-oxidized in mitochondria, generating reactive oxygen species (ROS), and, in turn, cause cell death by apoptosis often seen in lipotoxicity and cachexia. Both FAs and ceramides induce stress in the endoplasmic reticulum (ER stress). ER stress activates PERK, which phosphorylates the ribosome binding protein elF2a and thus downregulates protein synthesis, as is often seen in lipotoxicity and cachexia. FAs are also converted into diacylglycerol (DAG), which along with ceramides activate protein kinase C (PKC). PKC activates the nuclear factor (NF)-κB pathway. This downregulates MyoD and increases the ubiquitin ligases, MURF1 and MAFBx, thus leading to protein degradation through activation of the proteasome complex. PKC also phosphorylates JNK (c-Jun N-terminal kinase). Activated JNK along with PKC phosphorylates IRS-1 at serine residue 1101, which leads to insulin resistance because of decreased PI3K/Akt activity. Decreased Akt and increased TNF-α signaling increases expression of MAFBx and MURF1 through FOXO, which also leads to increased protein degradation. Factors in red are upregulated in cachexia, those in green are downregulated in cachexia, and factors shown in black have no change or have not yet been determined in cachexia. Solid lines indicate pathways confirmed in CAC and dotted lines expected but unconfirmed ones.

It is very likely that the diversity of genetic and epigenetic changes occurring in different tumors results in diverse effects on tumor metabolism. Therefore, it is not that surprising that great variety is observed in the induction, progression, and severity of CAC, and it is equally likely that several different mechanisms contribute to CAC development. An exact delineation of these processes will be necessary to devise novel and improved treatment strategies to combat this devastating condition.