Survivin drives tumor-associated macrophage reprogramming: a novel mechanism with potential impact for obesity

PURPOSE

Recent studies point to adipose-derived stem cells (ASCs) as a link between obesity and cancer. We aimed to determine whether survivin, which is highly secreted by ASCs from subjects with obesity, might drive a pro-tumoral phenotype in macrophages.

METHODS

The effect of ASC conditioned medium on the macrophage phenotype was assessed by expression studies. Survivin intracellular localization and internalization were examined by subcellular fractionation and immunofluorescence, respectively. Loss- and gain-of-function studies were performed using adenoviral vectors, and gene expression patterns, migration and invasion capacities of cancer cells were examined. Heterotypic cultures of ASCs, macrophages and cancer cells were established to mimic the tumor microenvironment. Survivin-blocking experiments were used to determine the impact of survivin on both macrophages and cancer cells. Immunohistochemical analysis of survivin was performed in macrophages from ascitic fluids of cancer patients and healthy controls.

RESULTS

We found that obese-derived ASCs induced a phenotypic switch in macrophages characterized by the expression of both pro- and anti-inflammatory markers. Macrophages were found to internalize extracellular survivin, generating hybrid macrophages with a tumor-associated phenotype that included secretion of survivin. Exogenous expression of survivin in macrophages generated a similar phenotype and enhanced the malignant characteristics of cancer cells by a mechanism dependent on survivin phosphorylation at threonine 34. Survivin secreted by both ASCs from subjects with obesity and tumor-associated macrophages synergistically boosted the malignancy of cancer cells. Importantly, survivin was mainly detected in ascites-associated macrophages from patients with a malignant diagnosis.

CONCLUSION

Our data indicate that survivin may serve as a molecular link between obesity and cancer and as a novel marker for tumor-associated macrophages.

Adipose tissue and serum CCDC80 in obesity and its association with related metabolic disease

Coiled-coil domain-containing 80 (CCDC80) is an adipocyte-secreted protein that modulates glucose homeostasis in response to diet-induced obesity in mice. The objective of this study is to analyze the link between human CCDC80 and obesity. CCDC80 protein expression was assessed in paired visceral (VAT) and subcutaneous (SAT) adipose tissue from 10 subjects (BMI range 22.4-38.8 kg/m²). Circulating CCDC80 levels were quantified in serum samples from two independent cross-sectional cohorts comprising 33 lean and 15 obese (cohort 1) and 32 morbid obese (cohort 2) male subjects. Insulin sensitivity, insulin secretion and blood neutrophil count were quantified in serum samples from both cohorts. Additionally, circulating free IGF-1 levels and oral glucose tolerance tests (OGTT) were assessed in cohort 1 whereas C-reactive protein levels and degree of atherosclerosis and hepatic steatosis were studied in cohort 2. In lean subjects, total CCDC80 protein content assessed by immunoblotting was lower in VAT than in SAT. In obese patients, CCDC80 was increased in VAT (P<0.05), but equivalent in SAT compared with lean counterparts. In cohort 1, serum CCDC80 correlated negatively with the acute insulin response to glucose and IGF1 levels, and positively with blood neutrophil count, independently of BMI, but not with insulin sensitivity. In cohort 2, serum CCDC80 was positively linked to the inflammatory biomarker C-reactive protein (r=0.46; P=0.009), atherosclerosis (carotid intima-media thickness, r=0.62; P<0.001) and hepatic steatosis (ANOVA P=0.025). Overall, these results suggest for the first time that CCDC80 may be a component of the obesity-altered secretome in VAT and could act as an adipokine whose circulant levels are linked to glucose tolerance derangements and related to inflammation-associated chronic complications.

Insulin resistance induced by tumor necrosis factor-alpha in myocytes and brown adipocytes

Insulin resistance is an important contributor to the pathogenesis of type 2 diabetes, and obesity is a risk factor for its development, in part because adipose tissue secretes proteins, called adipokines, that may influence insulin sensitivity. Among these molecules, tumor necrosis factor (TNF)-alpha has been proposed as a link between obesity and insulin resistance because TNF-alpha is overexpressed in adipose tissues of obese animals and humans, and obese mice lacking either TNF-alpha or its receptor show protection against developing insulin resistance. Direct exposure to TNF-alpha induces a state of insulin resistance in terms of glucose uptake in myocytes and brown adipocytes because of the activation of proinflammatory pathways that impair insulin signaling at the level of the insulin receptor substrate (IRS) proteins. In this regard, the Ser(307) residue in IRS-1 has been identified as a site for the inhibitory effects of TNF-alpha in myotubes, with p38 mitogen-activated protein kinase and inhibitor kB kinase being involved in the phosphorylation of this residue. Conversely, Ser phosphorylation of IRS-2 mediated by TNF-alpha activation of mitogen-activated protein kinase was the mechanism found in brown adipocytes. Protein-Tyr phosphatase (PTP)1B acts as a physiological, negative regulator of insulin signaling by dephosphorylating the phosphotyrosine residues of the insulin receptor and IRS-1, and PTP1B expression is increased in muscle and white adipose tissue of obese and diabetic humans and rodents. Moreover, up-regulation of PTP1B expression was recently found in cells treated with TNF-alpha Accordingly, myocytes and primary brown adipocytes deficient in PTP1B are protected against insulin resistance induced by this cytokine. Furthermore, down-regulation of PTP1B activity is possible by the use of pharmacological agonists of nuclear receptors that restore insulin sensitivity in the presence of TNF-alpha. In conclusion, the lack of PTP1B in muscle and brown adipocytes increases insulin sensitivity and glucose uptake and could confer protection against insulin resistance induced by adipokines.

Liver X receptor agonists ameliorate TNFalpha-induced insulin resistance in murine brown adipocytes by downregulating protein tyrosine phosphatase-1B gene expression

AIMS/HYPOTHESIS

The nuclear receptors, including nuclear receptor subfamily 1, group H, member 3 (NR1HR, also known as liver X receptor [LXR]), are sensors of cholesterol metabolism and lipid biosynthesis that have recently been proposed as insulin sensitisers. TNFalpha has been described as a link between obesity and the development of insulin resistance, an important contributor to the pathogenesis of type 2 diabetes. Therefore, we decided to investigate the ability of NR1HR agonists to ameliorate TNFalpha-induced insulin resistance in brown adipocytes.

METHODS

Primary brown adipocytes from rat fetuses, and from wild-type neonate mice and neonate mice deficient in the gene encoding protein tyrosine phosphatase-1B (Ptpn1, also known as Ptp1b) were cultured in the absence or presence of TNFalpha and different nuclear receptor agonists. Among them, the unrelated NR1HR ligands T0901317, GW3965 and (22R)-hydroxycholesterol were tested. After insulin stimulation, glucose uptake and solute carrier family 2 (facilitated glucose transporter), member 4 (SLC2A4, formerly known as GLUT4) translocation were measured. Next the insulin signalling cascade was determined by submitting cells to lysis, immunoprecipitation and immunoblotting.

RESULTS

NR1HR agonists ameliorate TNFalpha-induced insulin resistance restoring completely insulin-stimulated glucose uptake and SLC2A4 translocation to plasma membrane. This effect is parallel to the recovery of the insulin cascade insulin receptor/IRS-2/phosphatidylinositol 3-kinase/protein kinase B, and could be due to the fact that T0901317 prevents the increase of PTPN1 production and phosphatase activity produced by TNFalpha. In this regard, Ptpn1-deficient brown adipocytes showed protection against insulin resistance by TNFalpha. Moreover, we observed that T0901317 produced in itself a significant increase over basal glucose uptake consistent with an increase of SLC2A4 protein content in plasma membrane, attributable to the activation of protein kinase zeta and/or the increase of Slc2a4 expression.

CONCLUSIONS/INTERPRETATION

Nuclear receptors NR1HR are interesting potential targets for drug treatment of insulin resistance.

TGF-β transcriptionally activates the gene encoding the high-affinity adenosine transporter CNT2 in rat liver parenchymal cells

The nucleoside transporter CNT2 is the highest-affinity adenosine transporter identified so far. Recent evidence suggests that CNT2 has functions other than salvage (i.e. modulation of purinergic responses). Here we identified TGF-beta1 as a potent inducer of CNT2 protein expression in liver parenchymal cells. By contrast, CNT1, which is a target of multifunctional cytokines involved in liver cell proliferation, does not respond to TGF-beta1 treatment. Cloning of a murine CNT2 gene sequence with promoter-like activity enabled us to demonstrate that this cytokine exerts this effect by transcriptionally activating the CNT2-encoding gene in a JNK-dependent manner. The evidence that CNT2 is not a target of multifunctional cytokines involved in hepatocyte proliferation, but instead, of a cytokine that plays major roles in differentiation and apoptosis, further supports the view that the main physiological role of this transporter protein is not nucleoside salvage.

The concentrative nucleoside transporter family (SLC28): new roles beyond salvage?

The concentrative nucleoside transporter (CNT) family (SLC28) has three members: SLC28A1 (CNT1), SLC28A2 (CNT2) and SLC28A3 (CNT3). The CNT1 and CNT2 transporters are co-expressed in liver parenchymal cells and macrophages, two suitable models in which to study cell cycle progression. Despite initial observations suggesting that these transporter proteins might contribute to nucleoside salvage during proliferation, their subcellular localization and regulatory properties suggest alternative roles in cell physiology. In particular, CNT2 is a suitable candidate for modulation of purinergic responses, since it is under the control of the adenosine 1 receptor. Increasing evidence also suggests a role for CNT2 in energy metabolism, since its activation relies on the opening of ATP-sensitive K+ channels. Animal and cell models genetically modified to alter nucleoside transporter expression levels may help to elucidate the particular roles of CNT proteins in cell physiology.

Role of the Human Concentrative Nucleoside Transporter (hCNT1) In the Cytotoxic Action of 5[Prime]-Deoxy-5-fluorouridine, an Active Intermediate Metabolite of Capecitabine, a Novel Oral Anticancer Drug

We attempt to identify the plasma membrane transporter involved in the uptake of 5'-deoxy-5-fluorouridine (5'-DFUR), an intermediate metabolite of capecitabine. This novel oral fluoropyrimidine is used in cancer treatments and is a direct precursor of the cytostatic agent 5'-fluorouracil. We also examine the role of the transporter in 5'-DFUR cytotoxicity. The human concentrative nucleoside transporter (hCNT1) was cloned from human fetal liver and expressed in Xenopus laevis oocytes. The two-electrode voltage-clamp technique was used to demonstrate that 5'-DFUR, but not capecitabine or 5'-FU, is an hCNT1 substrate. Then, hCNT1 was heterologously expressed in the mammalian cell line Chinese hamster ovary-K1. Functional expression was demonstrated by monitoring transport of radiolabeled substrates and by using a monospecific polyclonal antibody generated against the transporter. hCNT1-expressing cells were more sensitive to 5'-DFUR than vector-transfected or wild-type cells. The sensitivity of the three cell types to other agents such as cisplatin or 5'-FU was identical. In conclusion, this study shows that 1) the pharmacological profile of a nucleoside transporter can be determined by an electrophysiological approach; 2) the hCNT1 transporter is involved in 5'-DFUR uptake; and 3) hCNT1 expression may increase cell sensitivity to 5'-DFUR treatment. This study also reports for the first time the generation of an antibody against hCNT1, which may be useful in the elucidation of the relationship between hCNT1 expression and tumor response to capecitabine treatment.