Pleiotropic effects of IL-6 action, including those that promote muscle atrophy, increased muscle wasting and the mechanisms of muscle wasting, are mainly described in animal models
[32][30][33]. In mice affected by severe cachexia (Apc
Min/+ mice), IL-6 overexpression altered the expression of proteins regulating mitochondrial biogenesis and fusion in cultured myoblasts (C2C12 cells). The affected specimen could be rescued by the administration of an IL-6 receptor antibody, as well as exercise
[34]. Further studies proved that IL-6 overexpression in cachectic mice increases the levels of fasting insulin and triglycerides, with the possibility of their normalization through exercise, due to increased oxidative capacity, induction of Akt signaling, and down-regulation of AMP-activated protein-kinase (AMPK) signaling in muscles
[35][36].
Another mouse-model study showed a phenotype switch from WAT to brown fat (WAT browning) during early CC stages, even before muscle atrophy. This occurs due to an increase in the expression of uncoupling protein 1 (UPC1) in WAT. This protein affects the mitochondrial respiration, uncoupling it toward thermogenesis instead of ATP synthesis, which results in an increase in lipid mobilization and energy spending in cachectic mice. In turn, the increase in UPC1 production in WAT is caused by the effects of IL-6, as well as the presence of chronic inflammation
[37]. Similarly, Han et al. reported a correlation between chronic inflammation (especially that mediated by IL-6) and CC promotion. IL-6 concentration was correlated with serum-free fatty acids (FFA), both in early- and late-stage CC, while serum TNF-α was positively linked to FFA in early-stage but not late-stage CC. Furthermore, WAT lipolysis was increased in early- and late-stage CC, while WAT browning increased only in late-stage CC
[38]. The WAT browning, as an effect of systemic inflammation, has a particular role in high energy expenditure associated with CC
[37][38]. Recent studies showed augmented expression of pro-inflammatory cytokines, including IL-6, TNF-α, and IL-8 in CC patients as compared to the control group. Furthermore, IL-8 content was also higher in weight-stable cancer (WSC) patients compared to control. The plasma fatty-acid profile was also positively correlated with some of the pro-inflammatory cytokines expression in the CC patients (IL-8, IFN-γ, CCL2, and IL-1Ra)
[39].
A comparison of serum concentrations of four pro-inflammatory factors (e.g., IL-6, IL-1β, Chemokine (C-X-C Motif) Ligand 8(CXCL8)/IL-8, and TNF-α) in advanced-stage cancer patients showed that IL-1β levels correlated more strongly with clinical characteristics than those of IL-6
[40]. There are also some clinical reports of resectable pancreatic cancer patients, which noted low levels of the described pro-inflammatory cytokines (IL-6, IL-1β, IFN-γ, and TNF-α), with their concentrations not correlated with CC, even if more sensitive methods were used for the analyses. Among the 25 tested circulating factors, only the monocyte chemoattractant protein 1 (MCP-1) showed an increase in treatment-naïve cachectic patients, as compared to those unaffected by the condition. Hence, it was suggested as a potential CC biomarker
[41]. Recent studies by Cao et al. describe reporter cell lines that could detect factors associated with CC, such as myostatin, activin A, and TNF-α. This cell model could be a valuable tool of early-cachectic-state detection and differentiation of cancer-induced cachexia in humans and mice
[42].
3. Systemic Inflammatory Response in Colorectal Cancer-Associated Cachexia
A direct correlation between cancer cachexia and SIR was also confirmed in patients with CRC
[13][44][45]. Most of the available literature concerns the role of the inflammatory process in muscle-tissue changes during CC. In turn, recent studies show a correlation between muscle catabolism and systemic inflammation in CRC. The presence of myopenia (reduced muscle mass) in preoperative CRC patients was significantly correlated with a number of SIR markers, such as elevated CRP concentration, systemic immune-inflammation index (SII), neutrophil-platelet score, and a decreased lymphocyte:monocyte ratio (LMR), prognostic nutritional index, and serum albumin level. Among those, only the increase in CRP level was indicated as an independent risk factor for the presence of preoperative myopenia
[13]. Other reports indicate a correlation between elevated SIR, as measured by the high modified Glasgow prognostic score (mGPS), CRP, and neutrophil:lymphocyte ratio (NLR) with a low skeletal-muscle index in patients with primary operable CRC
[46]. In turn, Malietzis et al. reported that host SIR in resectable CRC patients is associated with not only myopenia but also myosteatosis (increased infiltration by inter- and intra-muscular adipose tissue). High neutrophil:lymphocyte ratios (NLR) and low albumin levels were independent predictors of myopenia, with the former being correlated with myosteatosis
[47]. Furthermore, Sirniö et al. noted correlations between serum levels of several amino acids (AAs) (low with glutamine and histidine and high with phenylalanine), SIR markers, and muscle catabolism in patients with CRC. Hence, the authors proposed several SIR indicators, e.g., mGPS, high blood NLR, and high serum levels of CRP, IL-6, and IL-8. Of the 13 cytokines tested, IL-6 most strongly associated with increased phenylalanine and lower histidine levels. The authors suggest that SIR is associated with glutamine consumption and muscle wasting in CRC
[44]. In turn, the studies of Ohmori et al. suggest that only TNF-α, the levels of which were negatively correlated with skeletal-muscle index (SMI) and SDS-soluble myosin light chain 1 (SDS-MYL1), is a good serum muscle-atrophy marker in CRC-associated cachexia
[45]. Other authors confirmed that higher TNF-α concentrations are usually associated with patients with more advanced CRC stages (stage III/IV vs. stage I/II tumors). In turn, patients with the greatest nutritional deficit exhibited higher levels of adipocytokines. However, these TNF-α variations were only significant in CRC vs. control when the nutritional status was evaluated by phase-angle tertiles
[48].
There was also an attempt to evaluate the role of gut barrier disruption, possibly resulting in persistent activation of the host immune response, as a cause of local and systemic inflammatory changes in CRC-associated cachexia. Comparative studies concerned cachectic and weight-stable CRC patients, analyzing the circulating profile and tissue expression of various cytokines (including chemokines), growth and differentiation factors, as well as morphological intestinal changes in CRC-associated cachexia. Mucosal biopsies were derived from the rectosigmoid region, 20 cm from the tumor margin. An increase in serum concentrations of IL-6 and IL-8 was observed, together with elevated tissue expression of IL-7, IL-13, and transforming growth factor β3 (TGF-β3), as well as rich lymphocyte and macrophage infiltration in the colon of CC patients, compared to those of normal body weight. These changes suggest the presence of repair mechanisms in the damaged large intestine, with expanded recruitment of immune cells and higher tissue production of IL-13 and TGF-β3
[6].
Animal models also allowed for the clarification of some of the mechanisms of heart-muscle damage in CRC-associated cachexia involving inflammation. Studies on CD2F1 mice inoculated with C26 cells presented an increase in the production of IL-6, IL-6R, and F4/80 (a marker for macrophages infiltration) in the heart of tumor-bearing vs. non-tumor-bearing mice. Furthermore, increased fibrosis, disrupted myocardial structure, and altered composition of contractile proteins, e.g., troponin I (decreased), myosin heavy-chains isoforms α (decreased), and β (increased), was detected in the cardiac muscle of tumor-bearing mice, resulting in muscle-function impairment
[49].
Inflammatory changes also play a role in metabolic disturbances observed in patients with CRC-associated cachexia. Mice bearing C26 carcinoma represented a well-established murine model of CC resulting in a high systemic level of IL-6
[50]. It was proven that these mice exhibit reduced adipose mass, increased AT lipolysis, and a five-fold increase in FFA plasma levels. These alterations were linked to the activation of IL-6 signaling in WAT through a three-fold increase in phosphorylated STAT3 and high suppressor of cytokine signaling 3 (SOCS3) gene expression levels
[51]. It was also recently reported that the leukemia inhibitory factor (LIF) secreted by the tumor induces changes in the expression of AT, as well as serum levels of IL-6 and leptin, acting in a JAK-dependent manner. This, in turn, results in cachexia-associated adipose wasting as well as anorexia. The authors noted that the use of JAK inhibitors in both in vitro and in vivo models of CC inhibits this process
[52].
Lipolysis is also a typical manifestation of CC. Some of the mechanisms of this process were elucidated, mainly in in vitro models, through the evaluation of the activity of enzymes involved in lipid degradation in cancer cachexia. One example of such is spermidine/Spermine N-1 acetyl transferase (SSAT), an important enzyme in polyamine metabolism, which plays other potential roles, e.g., decreasing lipid accumulation. Its activity is stimulated by a multitude of factors, including those associated with cachexia, such as several cytokines, hormones, and natural substances (reviewed in:
[53]). Constant SSAT activation leads to an increased demand for acetyl-CoA (a cofactor of SSAT), thereby restricting conversion of acetyl-CoA by acetyl-CoA carboxylase (ACC) to malonyl-CoA
[54]. Malonyl-CoA inhibits the rate-limiting step in the β-oxidation of fatty acids and is a substrate in fatty-acid synthesis. A study of the activity of enzymes responsible for increased lipolysis in a mouse model of cancer cachexia (C26) using a cachexigenic clone 20 (c20), and noncachexigenic clone 5 (c5), showed a significant increase in SSAT as well as a decrease in acetyl-CoA carboxylase (ACC) and malonyl-CoA in cachectic mice
[55]. The studies of Chiba et al. confirmed these observations, demonstrating a correlation between malonyl-CoA in the liver and symptoms of CC (c5). In turn, in mice bearing the c5 tumor, the levels of SSAT mRNA did not correlate with the severity of CC manifestations. Hence, malonyl-CoA in the liver, correlating with the weight of fat stores in mice, seems to be a good marker of CC. A decrease in malonyl-CoA resulted in lowered fatty-acid synthesis and accelerated lipolysis via activation of carnitine palmitoyltransferase 1 (CPT-1) and ketogenesis. According to the authors, a decrease in the level of malonyl-CoA activity, playing a major role in CC induction, can be also caused by factors other than SSAT. From the panel of investigated pro-inflammatory (e.g., IL-1β, IL-6, and TNF-α) and anti-inflammatory/immunomodulatory cytokines (e.g., IL-10), only IL-10 had a negative correlation with body weight and the weights of skeletal muscle and storage fat
[54].
In another study, Apc
Min/+ mice (carrying a dominant mutation in the APC gene) are a model of colon cachexia directly related to an intestinal tumor burden and subsequent chronic inflammation (with elevated IL-6 levels). These mice were classified as noncachectic (8 weeks of age), precachectic (14 weeks of age), and severely cachectic (20 weeks of age). These studies confirmed the disturbances in hepatic triglyceride metabolism in catechetic mice. Furthermore, a novel role of hepatic glycosylphosphatidylinositol-anchored high-density lipoprotein binding protein 1 (GPIHBP1) was confirmed in hypertriglyceridemia with marked liver steatosis in Apc
Min/+ mice. It was also reported that GPIHBP1 is involved in the NF-κB signaling pathway in the liver of cachectic mice. The authors found a decrease in the hepatic uptake of fatty acids and lipolysis, with no differences in fatty-acid β-oxidation
[56].
Clinical studies of CC showed that the increase in serum FFAs (an indicator of enhanced lipolysis) was positively correlated with serum pro-inflammatory cytokine levels (e.g., IL-6 and TNF-α)
[38]. In turn, disorders of lipid metabolism in cachexia (including mechanisms of lipolysis) were described in recent review papers. The involvement of both inflammatory factors, defects in energy utilization, and molecular mechanisms underlying the WAT dysfunction and browning in CC are all highlighted in the mentioned works
[26][57].
The role of direct adipose-cell involvement in the mechanisms of lipolysis in cachexia is described later in the paper (
Section 5.2). In turn, a brief summary of selected mediator action during systemic inflammatory response associated with CC is presented in .
Table 1. Summary of some mechanisms and effects of selected procachectic mediators involved in the cancer cachexia.