In this study, we firstly revealed the differences of adiponectin, nesfatin-1, IL-6 and TNF-α between prediabetes and T2DM or healthy individuals in Chinese population. Our results demonstrated that adiponectin levels in prediabetes were distinctly lower than that in the healthy, while no significant difference was observed when compared to T2DM, which was consistent with the consequences of previous studies[30, 31]. Another research reported significant reduction of adiponectin in impaired fasting glucose group compared with hyperglycemic group[31]. Others proved mildly lower nesfatin-1 level in impaired fasting glycemia and impaired glucose tolerance in contrast to the healthy group[23]. The variation of the satiety molecule nesfatin-1 level in prediabetes hasn’t been widely probed, despite a study found slightly low nesfatin-1 levels in impaired fasting glycemia and impaired glucose tolerance compared to the healthy [23]. Our data revealed that nesfatin-1 levels in prediabetes were obviously higher than that in T2DM but no significant difference when compared to the healthy, which is the first delineation. In our study, remarkably increased IL-6 levels were detected in prediabetes in contrast to the healthy but that levels insignificantly dropped when compared to T2DM. Upadhyaya et al. reported that serum IL-6 levels in impaired fasting glucose group was dramatically higher than normoglycemic group and mildly lower than hyperglycemic group[31]. It has been reported that IL-6 protein expression at sections of pancreas and isolated islets from prediabetes were significantly upregulated compared to the normal persons in a recent study, where the prediabetes criteria only included impaired fasting glucose and HbA1c[32]. Here we added indexes of FBG, 2hFBG and HbA1c into the prediabetes criteria to make the evidences more sufficient.
Based on the experimental evidences that TNF-α concentration in prediabetes was significantly lower than in T2DM but mildly higher than in the healthy, Lainampetch et al. proved that TNF-α promoted developing prediabetes into T2DM[33]. In agreement with their findings, we also found declined TNF-α level in prediabetes when compared to T2DM and moderate elevation when compared to the healthy. Correlation test in our study uncovered that TNF-α levels were strongly relevant to adiponectin in prediabetes (R = 0.2939, P = 0.0105).
Our logistic regression analysis indicated that elevated TNF-α, HOMA-IR, TG, TC, TB and 2hBG and reduced insulin and eGFR were risk factors for facilitating the progression of pre-diabetic condition to diabetes. Recent studies revealed that combination of insulin resistance and compensatory hyperinsulinemia could give rise to dyslipidemia, which characterized by high levels of TG, TC, LDL-C and reduction of HDL-C concentration in serum[34, 35]. To better understand the roles of nesfatin-1, adiponectin, IL-6 and TNF-α in the development of prediabetes and T2DM, we introduced influence factors of HOMA-IR and HbA1c in the present study due to insulin resistant and hyper insulinemia have major impacts on prediabetes evolvement.
Our results suggested that falling nesfatin-1 levels was a risk factor for HOMA-IR rising (OR = 2.812, P = 0.01) while reduced adiponectin level raised the risk of HbA1c elevation (OR = 6.238, P = 0.019). Many studies focused on nesfatin-1 and insulin resistant, decreased serum nesfatin-1 levels of insulin resistant associated disease such as polycystic ovarian syndrome[36], non-alcoholic fatty liver disease[37] had covered. An in vivo study demonstrated that injecting nesfatin-1 into the third cerebral ventricle of high-fat diet-fed rats contributed to promoting insulin-induced glucose uptake and insulin signaling through upregulating the phosphorylation levels of insulin receptors[22]. Nesfatin-1 was also found to be augmented in gastric cancer and exert decent role for diagnosing gastric cancer(AUC = 0.857, 95%CI (0.769–0.946)) with a threshold of 1.075 ng/ml[38]. There is growing evidence on nesfatin-1’s versatility, but its receptor remains unknown.
In this study, adiponectin presented a significant positive correlation with TNF-α (R = 0.2939, P = 0.0105) or IL-6 (R = 0.3918, P = 0.0005) in prediabetes, and their relationship was notably enhanced in prediabetes with HOMA-IR (TNF-α: R = 0.7732, P < 0.0001, IL-6: R = 0.6663, P = 0.0005). As an endocrine organ, adipose tissue secretes numerous cytokines including TNF-α, IL-6, visfatin, leptin, and adiponectin. Adiponectin is an antiinflammation adipocytokine and macrophage is the primary target of adiponectin’s anti-inflammation effect [39]. Adiponectin inhibits the differentiation of myeloid progenitor cells, regulates the function of macrophage and decreases expression of Toll-like receptor 4[40]. Moreover, adiponectin suppresses the proinflammatory M1 macrophage activation and elevates the activation and proliferation of M2 macrophage, as well as modulates macrophage polarization from M1 phenotype towards M2 cells[41, 42]. It has been demonstrated that knock-out adiponectin in mice resulted in upregulates expression of M1 macrophage markers such as TNF-α, IL-6 and monocyte chemoattractant protein-1 and downregulation of M2 macrophage including arginase 1, macrophage galactose N-acetyl-galactosamine specific lectin-1 and IL-10[42]. It is worth noting that adiponectin appeared to have proinflammatory effects according to a few researches[43, 44]. Adiponectin induced human macrophage and T- cell differentiation to proinflammatory phenotype which resembles M1 other than M2[43], and in human adipose tissue and placenta, adiponectin promoted the release of TNF-α and IL-6 through NF-κB and ERK signaling[44]. In addition, it has been proven that TNF-α could damage adiponectin multimerization and consequently impairs adiponectin secretion by altering disulfide bond modification in the endoplasmic reticulum[45]. Collectively, adiponectin could function as an anti-inflammatory factor or proinflammatory molecule by inhibiting the polarization of macrophage M1[43], or eliciting the secretion of TNF-α and IL-6 in human adipose tissues[45]. It can be inferred that adiponectin participates in the progression of prediabetes toward diabetes and changes its roles of being anti-inflammatory or proinflammatory in response to different disease condition.
In summary, this research revealed that serum adiponectin, nesfatin-1, IL-6 and TNF-α levels in prediabetes were obviously change compared with T2DM or healthy individuals. We also found that adiponectin exerted its anti-inflammatory function in prediabetes phase, suggesting a significant but slow and gradual change during the progression of healthy condition toward diabetes via pre-diabetes.
This study also exists some limitations. Firstly, the research included only the Han nationality participants, thus racial differences should be noticed when apply the conclusion to other populations. Secondly, the size of the participants in this study is relatively small and more cases are needed to verify the variation of adiponectin, nesfatin-1, TNF-α and IL-6 in prediabetes and diabetes, as well as to make the conclusion more convincible. Thirdly, the cut-off HOMA-IR value of Chinese adults are vacant, studies are needed to definite the value. More importantly, the mechanisms of how the adiponectin, nesfatin-1, TNF-α and IL-6 factors are involved in the progression of prediabetes to diabetes are worth our in-depth exploration.