Visfatin Induces Inflammation and Insulin Resistance via the NF-κB and STAT3 Signaling Pathways in Hepatocytes

Background It has been suggested that visfatin, which is an adipocytokine, exhibits proinflammatory properties and is associated with insulin resistance. Insulin resistance and inflammation are the principal pathogeneses of nonalcoholic fatty liver disease (NAFLD), but the relationship, if any, between visfatin and NAFLD remains unclear. Here, we evaluated the effects of visfatin on hepatic inflammation and insulin resistance in HepG2 cells and examined the molecular mechanisms involved. Methods After treatment with visfatin, the inflammatory cytokines IL-6, TNF-α, and IL-1β were assessed by real-time polymerase chain reaction (RT-PCR) and immunocytochemical staining in HepG2 cells. To investigate the effects of visfatin on insulin resistance, we evaluated insulin-signaling pathways, such as IR, IRS-1, GSK, and AKT using immunoblotting. We assessed the intracellular signaling molecules including STAT3, NF-κB, IKK, p38, JNK, and ERK by western blotting. We treated HepG2 cells with both visfatin and either AG490 (a JAK2 inhibitor) or Bay 7082 (an NF-κB inhibitor); we examined proinflammatory cytokine mRNA levels using RT-PCR and insulin signaling using western blotting. Results In HepG2 cells, visfatin significantly increased the levels of proinflammatory cytokines, reduced the levels of proteins (e.g., phospho-IR, phospho-IRS-1 (Tyr612), phospho-AKT, and phospho-GSK-3α/β) involved in insulin signaling, and increased IRS-1 S307 phosphorylation compared to controls. Interestingly, visfatin increased the activities of the JAK2/STAT3 and IKK/NF-κB signaling pathways but not those of the JNK, p38, and ERK pathways. Visfatin-induced inflammation and insulin resistance were regulated by JAK2/STAT3 and IKK/NF-κB signaling; together with AG490 or Bay 7082, visfatin significantly reduced mRNA levels of IL-6, TNF-α and IL-1β and rescued insulin signaling. Conclusion Visfatin induced proinflammatory cytokine production and inhibited insulin signaling via the STAT3 and NF-κB pathways in HepG2 cells.


Introduction
Visfatin is an adipocytokine that is expressed predominantly in visceral adipose tissue [1] and also in skeletal muscle, bone marrow, and hepatocytes [2]. Visfatin was previously identified as a pre-B-cell colony-enhancing factor (PBEF), a cytokine expressed and secreted by lymphocytes [3], and also as nicotinamide phosphoribosyltransferase (NAMPT), an enzyme converting nicotinamide to nicotinamide mononucleotide (NMN), a precursor of nicotinamide adenine dinucleotide (NAD) [4]. Visfatin has been extensively studied in terms of its roles in obesity, insulin resistance, type 2 diabetes, and other metabolic disorders, but the findings have been contradictory [5][6][7].
Nonalcoholic fatty liver disease (NAFLD) is a spectrum of conditions ranging from fatty liver to nonalcoholic steatohepatitis (NASH), fibrosis, and cirrhosis [8]. Insulin resistance is considered to be the principal pathogenesis of NAFLD, and inflammation is also known to be involved in the disease progress [9]. Several adipocytokines have been suggested to play important roles in pathogenesis [10]. The role of visfatin in NAFLD has been evaluated, but the results were inconsistent [11][12][13], and it remains unclear whether visfatin contributes to hepatic insulin resistance and inflammation. Moreover, the effects of visfatin as an adipocytokine in hepatocytes have rarely been explored. We reported previously that visfatin stimulated gluconeogenesis in hepatocytes [14], suggesting a contribution of visfatin to hepatic insulin resistance. However, the effects of visfatin on insulin signaling in hepatocytes have not been reported previously.
Elevated levels of circulating cytokines may impair insulin signaling in peripheral organs [18]. It remains unclear how visfatin might affect liver insulin resistance and inflammation. Here, we evaluated the effects of visfatin on inflammation and insulin resistance in HepG2 cells and examined the molecular mechanisms involved.

Cell
Culture. HepG2 cells were obtained from the American Type Culture Collection and grown in minimum essential medium (MEM) supplemented with 10% (v/v) fetal bovine serum (FBS) and antibiotics (10 μg/mL streptomycin and 100 IU/mL penicillin) at 37°C in a humidified atmosphere of 95% air and 5% CO 2 (both v/v). Before treatment with 0.1% BSA/PBS or visfatin (visfatin dissolved in 0.1% BSA/PBS), media were replaced with serum-free DMEM. To measure insulin signaling, media containing visfatin were exchanged for serum-free DMEM followed by incubation for 1 hour and then treatment with insulin for 10 minutes.

RNA Isolation and Quantitative Real-Time Polymerase
Chain Reaction (PCR). Total cellular RNAs were isolated using the RNAiso Plus reagent (Takara Bio Inc., Otsu, Japan) according to the manufacturer's instructions. Briefly, HepG2 cDNA was prepared using avian myeloblastosis virus reverse transcriptase and random 9-mer primers. The cDNA was amplified by qPCR using primer sets specific for human TNF-α: TGA AAG CAT GAT CCG GGA CG (forward (F)) and TGA GGA ACA AGC ACC GCC TG (reverse (R)); human IL-6: TGT GTG GGG CGG CTA CAT CT (F) and GCC TTC GGT CCA GTT GCC TT (R); and human IL-1β: CCT TTG GTC CCT CCC AGG AA (F) and TGA GTC TGC CCA GTT CCC CA (R). Quantitative real-time PCR was performed using SYBR Green Master Mix (Takara Bio Inc.) on a Takara TP-815 instrument. All expression levels were normalized to those of GAPDH.
2.6. Enzyme-Linked Immunosorbent Assay (ELISA) for IL-6, TNF-α, and IL-1β. The IL-6, TNF-α, and IL-1β concentrations in culture supernatants were measured by sandwich ELISA. Monoclonal capture antibodies (4 μg/mL; R&D Systems) were added to 96-well plates (Nunc), and the plates were incubated for 2 hours at room temperature. The plates were incubated with blocking solution consisting of phosphate-buffered saline (PBS) containing 1% BSA and 0.05% Tween 20 for 2 hours at room temperature. The test samples and standard recombinant IL-6, TNF-α, and IL-1β (R&D Systems) were added to the plates, and the plates were incubated overnight at 4°C. The plates were washed four times with PBS containing Tween 20; then, 200 ng/mL of biotinylated detection monoclonal antibodies (R&D Systems) was added, and the plates were incubated for 2 hours at room temperature. The plates were washed, streptavidin-alkalinephosphatase (diluted 1 : 2000; Sigma-Aldrich) was added, and the reaction was allowed to proceed for 2 hours at room temperature. The plates were washed four times, and 1 mg/mL of p-nitrophenyl phosphate dissolved in diethanolamine (both from Sigma-Aldrich) was added to induce the color reaction, which was stopped by adding 50 μL of 1 N NaOH. The optical density at 405 nm was measured on an automated microplate reader (Bio-Rad, Hercules, CA, USA). Standard curves were drawn by plotting optical density versus the logs of IL-6, TNF-α, and IL-1β concentrations.

Statistical Analysis.
All experiments were performed two or three times. Data were compared using Student's t-test. A P value ≤ 0.05 was considered to reflect statistical significance.

Visfatin Induced Proinflammatory Cytokine Expression in HepG2
Cells. We hypothesized that visfatin might trigger an inflammatory response in hepatocytes. We treated HepG2 cells with visfatin and measured the levels of the proinflammatory cytokines IL-6, TNF-α, and IL-1β. We used visfatin at concentrations ranging from 100 to 400 ng/mL with reference to previous studies [14,19]. At 200 and 400 ng/mL, visfatin dramatically increased the mRNA levels of IL-6, TNF-α, and IL-1β (Figure 1(a)). Next, we immunocytochemically confirmed that the cytokines were produced at the protein level. Visfatin increased expression of IL-6, TNF-α, and IL-1β (Figure 1(b)). Stained cells were counted in at least 10 zones (100 random cells/zone) and quantified using ImageJ software.

Visfatin Induced Insulin Resistance in a Dose-Dependent
Manner in HepG2 Cells. We explored the effect of visfatin on Nuclei were counterstained with hematoxylin. (c) Staining intensities were analyzed using ImageJ software. The basal intensities of cells not exposed to visfatin were set to 100%, and the relative test intensities were then calculated. The data are means ± standard errors of those of three independent experiments. * p < 0 05 compared to the untreated control.

Discussion
In the present study, we found that, in HepG2 cells, visfatin significantly increased the expression levels of proinflammatory cytokines IL-6, TNF-α, and IL-1β and reduced the expression levels of the insulin-signaling pathway proteins phospho-IRS-1 (Tyr612) and phospho-AKT. Visfatin increased STAT3 and NF-κB pathway activities but not JNK, p38, or ERK pathway activities. A STAT3 or an NF-κB inhibitor blocked visfatin-induced proinflammatory cytokine synthesis and rescued insulin signaling. NAFLD is a complex spectrum of diseases ranging from simple steatosis to NASH and cirrhosis. NAFLD is strongly associated with insulin resistance and is regarded as the liver manifestation of metabolic syndrome [9]. However, the underlying mechanisms remain unclear. Inflammation is also known to be involved in the progress of NAFLD. IL-6 and TNF-α have been suggested to play key roles in NAFLD [24], as have several adipocytokines [10]. Adiponectin was reported to improve hepatic insulin resistance and exert anti-inflammatory effects [25]. Leptin augmented inflammatory and fibrogenic responses in a murine model [26].
Several studies have reported associations between circulating visfatin levels and NAFLD. Serum visfatin levels were higher in females with NAFLD than in controls [27]. Another report showed that serum visfatin levels reflected the extent of portal inflammation [2]. However, some studies reported contradictory results [11]. Moreover, the role, if any, of visfatin in NAFLD development and the mechanisms involved have been rarely studied; we evaluated the effects of visfatin on hepatic insulin resistance and inflammation using HepG2 cells.
It was initially suggested that visfatin acts as an insulin mimic by binding to the insulin receptor (IR) to induce phosphorylation of IRS-1, IRS-2, AKT, and MAPK. However, this report was later retracted [28]; now, IR is no longer considered to serve as a visfatin receptor. Several human studies found positive relationships between circulating visfatin levels and insulin resistance [29], whereas others did not [30]. Any role for visfatin in insulin receptor signaling remains controversial, although there were reports that visfatin upregulated IR phosphorylation in pancreatic β-cells and osteoblasts [31,32]. When we assessed the effects of visfatin in HepG2 cells, we found that visfatin dose-dependently decreased the levels of phospho-IRS-1 (Tyr612) and phospho-AKT but increased the phospho-IRS-1 (Ser307) level. To our knowledge, this is the first study to show that visfatin reduced insulin signaling in hepatocytes. Such inhibition of insulin signaling by visfatin can contribute to the pathogenesis of NAFLD.
Some studies have suggested that visfatin exerted proinflammatory effects, for example, in macrophages [33] and chondrocytes [34]. We evaluated the direct effects of visfatin  on proinflammatory cytokine production in hepatocytes and found that visfatin increased production and secretion of the proinflammatory cytokines IL-6, TNF-α, and IL-1β. The elevated levels of cytokines released from hepatocytes by visfatin may play a role in hepatic inflammation. Further studies of the interactions of visfatin with various cells in the liver and its role in the complex inflammatory process in the development of hepatic inflammation are required.
Visfatin was reported to enhance endothelial IL-6 production and angiogenesis via STAT3 activation [37]. As one of the STAT transcriptional factors activated by JAK, which is itself activated by IL-6 [38], STAT3 serves as both a mediator and a biomarker of endothelial activation. IL-6 triggered hepatic insulin resistance attributable to the "mammalian target of rapamycin" in a manner involving the STAT3-SOCS3 pathway [39]. We found that visfatin increased hepatocyte IL-6 production; thus, we explored whether the JAK-STAT3 pathway was involved in visfatininduced hepatic production of inflammatory cytokines. Addition of a STAT3 inhibitor followed by visfatin reduced IL-6, TNF-α, and IL-1β production, suggesting that visfatin exerted proinflammatory effects on hepatocytes via the STAT3 pathway. The JAK-STAT3 pathway was also involved in hepatic insulin signaling. When we treated hepatocytes with visfatin and a STAT inhibitor, the visfatin-induced deterioration of insulin signaling was rescued.
We found that the hepatocyte levels of phospho-JNK, phosphor-p38, and phosphor-ERK were not changed by visfatin, although previous studies reported that visfatin induced p38 and ERK activation in macrophages, enhancing endothelial angiogenesis via the MAPK pathway [40]. The JNK, p38, and ERK pathways did not seem to be involved in the visfatin-induced inflammation and insulin resistance of hepatocytes; the STAT3 and NF-κB pathways appear to be the main pathways in play. STAT and NF-κB signaling was measured using antibodies against phospho-STAT3, phospho-NF-κB, and actin. The maximum phosphoprotein intensities in visfatin-treated samples were set to 100%, and the relative intensities of test samples were then calculated. The data are means ± standard errors of those of three independent experiments. * p < 0 05, * * p < 0 01, and * * * p < 0 001 compared to the control.

Conclusion
Our study showed that visfatin induced proinflammatory cytokine production and inhibited insulin signaling via the STAT3 and NF-κB pathways in HepG2 cells. Further studies are needed to determine whether the insulin resistance and inflammation in hepatocytes induced by visfatin play roles in the development of NAFLD.  Figure 5: A JAK2/STAT3 inhibitor, AG490, and an NF-κB inhibitor, BAY11-7082, rescue visfatin-impaired insulin signaling. HepG2 cells were pretreated with AG490 (5 μM) or BAY11-7082 (5 μM) for 1 h before exposure to visfatin (200 ng/mL) for 24 h and then stimulated with 10 nM insulin for 10 min. (a) After the cells were harvested, insulin signaling was analyzed by immunoblotting using antibodies against phospho-IR (p-IR), phospho-IRS-1 (p-IRS-1), phospho-AKT (p-AKT), phospho-GSK-3α/β (p-GSK-3α/β), and actin. (b) The maximum phosphoprotein intensities in insulin-treated samples without visfatin were set to 100%, and the relative intensities of test samples were then calculated. The data are means ± standard errors of those of three independent experiments. * p < 0 05, * * p < 0 01, and * * * p < 0 001 compared to control.