Diabetes-associated dysregulated cytokines and cancer

Epidemiological data demonstrate that patients with diabetes have an augmented risk of developing various types of cancers, accompanied by higher mortality. A number of mechanisms for this connection have been hypothesized, such as insulin resistance, hyperinsulinemia, hyperglycemia, and increased inflammatory processes. Apart from these potential mechanisms, several diabetes-associated dysregulated cytokines might be implicated in the link between diabetes and cancer. In fact, some inflammatory cytokines, e.g. TNF-α, IL-6 and leptin, have been revealed to play important roles in both initiation and progression of tumor. Here, we depict the role of these cytokines in key events of carcinogenesis and cancer development, including their capability to induce oxidative stress, inflammation, their participation in epithelial mesenchymal transition (EMT), angiogenesis, and metastasis. Finally, we will also highlight the existing knowledge in terms of the involvement of these cytokines in different cancer types and comment on potential significances for future clinical applications. Abbreviations: AGC: Advanced Gastric Cancer, AMPK: AMPActivated Protein Kinase, AP-1: Activator Protein 1, BBL: Benign Breast Lesions, BC: Breast Cancer, BMI: Body Mass Index, CAC: ColitisAssociated Colon Cancer, CXCL16: C-X-C Chemokine Ligand 16, DM: Diabetes mellitus, EC: Endothelial Cell, EMT: Epithelial Mesenchymal Transition, FGF2: Fibroblast Growth Factor 2, HER2: Human Epidermal Growth Factor Receptor 2, HGF: Hepatocyte Growth Factor, IGF: Insulin-Like Growth Factor, IGT: Impaired Glucose Tolerance, IL-6: Interleukin 6, LINE-1: Long Interspersed Nuclear Element-1, MAPK: p38 Mitogen-Activated Protein Kinase, MIF: Macrophage Migration Inhibitory Factor, MMP: Matrix Metalloproteases, MS: Metabolic Syndrome, NSCLC: Non-Small Cell Lung Cancer, PAI-1: Plasminogen Activator Inhibitor-1, RNS: Reactive Nitrogen Species, ROS: Reactive Oxygen Species, SDF-1: Stromal Cell-Derived Factor-1, STAT3: Signal Transducer and Activator of Transcription 3, TLR4: Toll-like Receptor 4, TNF: Tumor Necrosis Factor , uPA: Urokinase Plasminogen Activator, VEGF: Vascular Endothelial Growth Factor, VN: Vitronectin. Introduction Diabetes mellitus (DM), the 12th leading cause of death worldwide [1], is a severe and chronic health problem worldwide that disturbs human body’s ability to utilize the energy in food. It can be classified into three major types: type 1, type 2, and gestational diabetes. DM can cause serious acute and chronic complications that adversely impact the quality of life and survival of the majority of people with this disease. Cancer is the 2nd primary cause of death globally [1]. A growing number of studies demonstrate a positive link between DM and the risk of cancer and cancer-related mortality [2]. Currently, the number of people lives with diabetes worldwide is 250 million and this figure anticipated to reach 380 million in 20 years. Accordingly, one can assume that even a slight increase in the cancer risk associated with diabetes may have significant consequences at the population level. The association between diabetes and cancer was first hypothesized more than 70 years ago and identified in the 1960s in population-based studies [3]. More recently, a number of studies [4] indicate that some type of cancers develop more frequently in patients with diabetes (primarily type 2 DM), whereas others such as prostate cancer happens less often in patients with diabetes. Diabetes imparts the highest relative risks (about 2-fold or higher) for liver, pancreas, and endometrium cancer, and lower risks (about 1.2–1.5-fold) for colon and rectum, breast, and bladder cancer [3]. Lung cancer does not seem to be associated with an augmented risk in diabetes, and the evidence for kidney cancer and non-Hodgkin lymphoma is indecisive. At present, few studies have investigated links between type 1 DM and cancer [3]. Despite their important clinical significance, the potential biologic links between these two diseases are poorly understood [3,5], which elicits a tremendous challenge for patient care. Some perplexing factors that have common or site-specific relevance make it hard to precisely evaluate cancer risk in patients with diabetes. These factors comprise diabetes duration, diverse therapeutic drugs, variable levels of metabolic control, and the potential existence of chronic complications. Currently, the mainstream view proposes that the relationship may be attributed to the direct effects of diabetes, i.e. hyperglycaemia [3,6] or alterations in fundamental metabolic conditions including insulin resistance and hyperinsulinaemia [3]. Hyperglycemia may contribute to increased cancer risk in diabetes via augmented oxidative stress and DNA damage [7,8]. Studies of the relationship between hyperglycemia and cancer risk have been conducted for a long time, nonetheless the Correspondence to: Yong Wu, Division of Cancer Research and Training, Charles R. Drew University of Medicine and Science, Los Angeles, CA 90059, USA, Tel: (323) 563-4885, Fax: (323) 563-4889; E-mail: yongwu@cdrewu.edu/yongwu@ucla.edu


Introduction
Diabetes mellitus (DM), the 12 th leading cause of death worldwide [1], is a severe and chronic health problem worldwide that disturbs human body's ability to utilize the energy in food. It can be classified into three major types: type 1, type 2, and gestational diabetes. DM can cause serious acute and chronic complications that adversely impact the quality of life and survival of the majority of people with this disease. Cancer is the 2nd primary cause of death globally [1]. A growing number of studies demonstrate a positive link between DM and the risk of cancer and cancer-related mortality [2]. Currently, the number of people lives with diabetes worldwide is 250 million and this figure anticipated to reach 380 million in 20 years. Accordingly, one can assume that even a slight increase in the cancer risk associated with diabetes may have significant consequences at the population level.
The association between diabetes and cancer was first hypothesized more than 70 years ago and identified in the 1960s in population-based studies [3]. More recently, a number of studies [4] indicate that some type of cancers develop more frequently in patients with diabetes (primarily type 2 DM), whereas others such as prostate cancer happens less often in patients with diabetes. Diabetes imparts the highest relative risks (about 2-fold or higher) for liver, pancreas, and endometrium cancer, and lower risks (about 1.2-1.5-fold) for colon and rectum, breast, and bladder cancer [3]. Lung cancer does not seem to be associated with an augmented risk in diabetes, and the evidence for kidney cancer and non-Hodgkin lymphoma is indecisive. At present, few studies have investigated links between type 1 DM and cancer [3].
Despite their important clinical significance, the potential biologic links between these two diseases are poorly understood [3,5], which elicits a tremendous challenge for patient care. Some perplexing factors that have common or site-specific relevance make it hard to precisely evaluate cancer risk in patients with diabetes. These factors comprise diabetes duration, diverse therapeutic drugs, variable levels of metabolic control, and the potential existence of chronic complications. Currently, the mainstream view proposes that the relationship may be attributed to the direct effects of diabetes, i.e. hyperglycaemia [3,6] or alterations in fundamental metabolic conditions including insulin resistance and hyperinsulinaemia [3]. Hyperglycemia may contribute to increased cancer risk in diabetes via augmented oxidative stress and DNA damage [7,8]. Studies of the relationship between hyperglycemia and cancer risk have been conducted for a long time, nonetheless the Previous studies have revealed augmented circulating levels of inflammatory cytokines TNF-and IL-6 in patients with diabetes [23][24][25][26][27]. The plasma TNF-and IL-6 are probably produced by activated leukocytes and adipocytes and vascular endothelial cells. TNF-and IL-6 are typical pro-inflammatory cytokines with a pro-tumorigenic effect.
It is well known that unsolved inflammation can result in cancer. TNF-α is a pro-inflammatory cytokine involved in carcinogenesis [28]. Persistent exposure to low concentration of TNF-α can prompt a tumor phenotype [29]. TNF-α facilitates tumorigenesis and cancer development via activating nuclear factor-kB [30] and increasing reactive oxygen species (ROS) and reactive nitrogen species (RNS) formation, which can trigger DNA damage [31]. TNF-α has more significant impact in the early stage of carcinogenesis, e.g. angiogenesis and invasion [19,30]

Leptin
Leptin is an imperative pro-inflammatory cytokine mainly produced by white adipose tissue and implicated in pathophysiological mechanisms associated with diabetes and its complications [58]. Leptin levels are elevated in overweight or obese people [59]. A study examining the link between plasma leptin levels and diabetes mellitus demonstrate that higher plasma leptin concentrations are associated with diabetes mellitus but not independently associated with diabetes mellitus after adjustment for body mass index (BMI) [60]. Leptin can adjust energy homeostasis through modulating food intake and energy expenditure via its effect on the hypothalamus; alternatively, it also induces cell growth, migration, and invasion [61]. Thus, leptin may play an important etiologic role in triggering malignant transformation or tumor development. For instance, preclinical evidence has revealed that leptin stimulates esophageal, colorectal, prostate and breast cancer cell proliferation [62][63][64][65][66] and is responsible for the association between diabetes and prostate cancer [67]. The leptin-receptor might predict poor prognosis in patient with advanced gastric cancer (AGC) [68]. In addition, a pooled analysis from three cohorts by Stolzenberg-Solomon et al. supports an relationship between enhancing leptin levels and pancreatic cancer [69]. The role of leptin in promoting cancer development has also been substantiated by in vivo studies. Actually, leptin-deficient and -resistant mice (ob/ob and db/db mice) do not develop transgene-triggered mammary cancers [70,71].
The mechanisms underlying leptin-related cancer development have been extensive studied. Leptin enhances the synthesis and release of cytokines by macrophages and proangiogenic factors, including fibroblast growth factor 2 (FGF2), vascular endothelial growth factor (VEGF), and matrix metalloproteases 2 and 9 (MMP-2/9) [72], which can prompt neoangiogenesis or further promote cancer cells [73]. Leptin has also been shown to significantly augment endothelial cell (EC) growth via a BCL2-dependent mechanism [74][75][76][77]. Leptin can elevate estrogen synthesis via activation of aromatase, thus promoting cancer growth in breast and endometrial cancer [78]. Recently, Chang et al. demonstrate that leptin stimulates STAT3 and G9a histone methyltransferase to silence miR-200c, a genetic program of epithelial homeostasis in breast cancer stem-like cells (CSC) that induces malignant development [79]. In addition, leptin is able to activate STAT3 in colitis-associated colon cancer (CAC) cells [80][81][82][83][84]. A recent study suggest that leptin can induce EMT in BC cells and this involves IL-8 activation via the PI3K/Akt signaling pathway [85].

Plasminogen activator inhibitor-1 (PAI-1)
Extracellular proteases regulate various physiological and pathological processes, e.g. organ development, inflammation, tissue injury/repair, and cancer [86,87]. The urokinase plasminogen activator (uPA)-plasmin is one of the most explored protease systems [88][89][90]. The serine protease uPA is activated when binding to its receptor uPAR on the cell surface [91]. The proteolytic activity of uPA is explicitly regulated by plasminogen activator inhibitor-1 (PAI-1), a glycoprotein of roughly 50 kDa, which upon combining with uPA facilitates the swift endocytosis of the trimolecular uPA/PAI-1/uPAR complex [92,93]. PAI-1 is synthesized by vasculature-surrounding cells such as endothelial cells and platelets. Its circulating active form is comparatively unstable. Nevertheless, PAI-1 is stabilized by vitronectin (VN) in blood circulation and extracellular matrix. This abundant glycoprotein is involved in thromboembolic diseases, atherosclerosis, fibrosis, cell migration, cell invasion, cell proliferation, cancer, and tissue remodeling [ In the last two decades, increasing evidence suggest that an elevated level of PAI-1 protein in human primary tumors represents one of the most helpful biochemical markers of an unfavorable prognosis in a number of human cancer types. This observation has given the motivation to substantial research on the role of PAI-1 in cancer growth, invasion, and metastasis [115]. Recently, Buta et al. propose that tumor size and PAI-1 can be used in combination as prognostic and predictive phenotypes in node-negative, postmenopausal BC patients bearing histological grade II tumors with ER/PR expression [116]. Positive stroma PAI-1 protein expression in the human epidermal growth factor receptor 2 (HER2)-negative patients is related to lower risk of death, hence it might identify a subgroup of HER2negative advanced BC patients who may benefit from trastuzumab treatment [117] and can be securely spared the toxicity and expenses of adjuvant chemotherapy [118]. Both PAI-1 and uPA stimulate cancer development and metastasis. Increased uPA and PAI-1 in BC tissue are independent and effective predictors of unfavorable outcome of BC patients, including patients with lymph node-negative disease. Apart from being prognostic biomarkers, overexpression of uPA and PAI-1 is capable of predicting benefit from adjuvant chemotherapy in early BC patients [118].

Resistin
Resistin, a signaling molecule secreted by adipocytes and monocytes, belongs to cysteine-rich protein family named ''resistinlike molecules'' [145,146]. It is up-regulated in obesity and participates in the pathogenesis of insulin resistance [146,147], type 2 diabetes [148][149][150] and metabolic syndrome (MS) [151]. Accumulating evidence shows that resistin plays an imperative controlling role in inflammatory disease [152,153]. Resistin expression at mRNA level is significantly augmented by pro-inflammatory cytokines [154,155]. Clinically, resistin concentrations are related to inflammatory markers obviously independent of BMI and can be used as a predictive factor for coronary atherosclerosis [152,156,157].
Sun et al. [158] and Dalamaga et al. [159] demonstrate that patients with breast cancer (BC) have strikingly enhanced resistin levels when compared with control subjects and patients with benign breast lesions (BBL). Additionally, the biological gradient of BC risk by plasma resistin concentrations still exists following adjustment for measurers of adiposity. The dose-dependent relationship between resistin levels and BC risk is remarkably prominent in female with superfluous exposure to estrogens. Thus, resistin might have an adiposity-independent role in BC pathogenesis. High resistin levels in BC tissue are related to a more malignant pathological status and unfavorable patient survival [160]. Thus, resistin might be potentially employed as a prognosis predictor for BC, a marker for hormone treatment stratification, as well as a possible therapeutic target. Moreover, resistin has been discovered to associate with other cancer types such as prostatic, colorectal, gastroesophageal and endometrial cancer [161][162][163][164][165][166][167][168]. In cancer patients, resistin is dramatically related to tumor markers, cancer stage, tumor size, grade and lymph node invasion [159,169].
Resistin's effects on cancer progression involve multiple mechanisms. Deshmukh et al. substantiate that resistin stimulates growth and aggressiveness of BC cells, and these effects are mediated by STAT3 activation [170]. In addition, resistin advances chondrosarcoma metastasis and expression of MMP-2 via activating the AMP-activated protein kinase (AMPK)/p38 mitogen-activated protein kinase (MAPK) pathway and downregulating miR-519d expression. Hence, resistin might represent a latent new therapeutic target in chondrosarcoma metastasis [171]. Hsieh YY et al. demonstrate that resistin promotes stromal cell-derived factor-1 (SDF-1) expression by Toll-like receptor 4 (TLR4) and stimulation of p38 MAPK/NFκB pathway in gastric cancer cells, which could explain a novel role of resistin in the association between obesity and gastric cancer [172].

Concluding comments
Diabetes, especially T2DM, and diabetes risk factors might be related to cancer. Robust and conceivable evidence propose associations between diabetes and cancer; nonetheless, the underlying mechanisms are poorly understood and there is little applicable clinical management of patients presenting with these diseases concomitantly. Accordingly, a multidisciplinary method is required to reveal the mechanisms underlying the links between diabetes and cancer, eventually improve clinical outcomes.
Currently, increasing attention has been given to the role of cytokine in cancer initiation and progression. Diabetes-associated cytokine dysregulation may be an important pathogenesis responsible for diabetes-related cancer. The important role of cytokines has been described, as a diagnostic or prognostic marker for tumor. For instance, the measurement of the serum concentrations of cytokines, e.g. resistin, IL-6 and IL-10, may have guiding significance for tumorigenic process or prognosis [40,208]. While various cytokines prompt carcinogenesis, their pro-tumoral roles rely on the equilibrium of distinct inflammatory mediators and the stage of cancer development. Therefore, investigating the roles of these mediators in various cancers or stages of development is indispensable for designing novel personalized managements by means of these latent therapeutic targets.
Although advancement has been made in the comprehension of the mechanisms of diabetes-associated dysregulated cytokines in tumorigenesis and cancer development, establishing a correlation between cytokines regulation and disease progression, as well as response to treatment still remains a challenge.

Authors' contributions
YW and YL contributed to the conception of the idea, literature search and writing the manuscript. YD and JV contributed to the critical evaluation and editing of the manuscript. All authors critically reviewed and accepted the final version of the manuscript.   115. Andreasen PA (2007)