The role of the adipocyte-derived factor visfatin in metabolism remains controversial, although some pancreatic β-cell-specific effects have been reported. This study investigated the effects of visfatin upon insulin secretion, insulin receptor activation and mRNA expression of key diabetes-related genes in clonal mouse pancreatic β-cells. β-TC6 cells were cultured in RPMI 1640 and were subsequently treated with recombinant visfatin. One-hour static insulin secretion was measured by ELISA. Phospho-specific ELISA and western blotting were used to detect insulin receptor activation. Real-time SYBR Green PCR array technology was used to measure the expression of 84 diabetes-related genes in both treatment and control cells. Incubation with visfatin caused significant changes in the mRNA expression of several key diabetes-related genes, including marked up-regulation of insulin (9-fold increase), hepatocyte nuclear factor (HNF)1β (32-fold increase), HNF4α (16-fold increase) and nuclear factor κB (40-fold increase). Significant down-regulation was seen in angiotensin-converting enzyme (−3.73-fold) and UCP2 (−1.3-fold). Visfatin also caused a significant 46% increase in insulin secretion compared to control (P<0.003) at low glucose, and this increase was blocked by co-incubation with the specific nicotinamide phosphoribosyltransferase inhibitor FK866. Both visfatin and nicotinamide mononucleotide induced activation of both insulin receptor and extracellular signal-regulated kinase (ERK)1/2, with visfatin-induced insulin receptor/ERK1/2 activation being inhibited by FK866. We conclude that visfatin can significantly regulate insulin secretion, insulin receptor phosphorylation and intracellular signalling and the expression of a number of β-cell function-associated genes in mouse β-cells.
Type 2 diabetes (T2D) is a condition with a multi-factorial aetiology. Obesity is one well-established risk factor for T2D, with the prevalence of obesity being >90% in sufferers of T2D in some western societies (Astrup & Finer 2000). Although the precise mechanisms by which increased adiposity predisposes so strongly for T2D are not fully understood, one emerging paradigm is an involvement of adipose tissue-secreted molecules commonly referred to as adipokines. These molecules include hormones, growth factors and cytokines such as resistin, visfatin, adiponectin, retinol-binding protein 4, leptin and tumour necrosis factor-α (Ahima 2006), which have been implicated in the modulation of both insulin sensitivity and β-cell function (Brown et al. 2002, 2007, Trayhurn & Wood 2005). Research has shown that the majority of these molecules circulate in increased concentrations in obese individuals.
One of the more recently described adipokines is visfatin (Berndt et al. 2005). Visfatin (also referred to as pre-B cell colony-enhancing factor) was initially reported as having insulin-mimetic properties. Subsequent studies have identified visfatin as a nicotinamide phosphoribosyltransferase (NAMPT) capable of producing nicotinamide mononucleotide (NMN), a precursor for the metabolic co-factor NAD+ (Kim et al. 2006). The circulating levels of visfatin have been reported to be elevated in both obesity and T2D (Li et al. 2006, Jin et al. 2008), suggesting that in these metabolic disorders, the pancreatic β-cell may be exposed to elevated visfatin levels. As the pancreatic β-cell is known to express the insulin receptor (Muller et al. 2006) and the insulin receptor is known to have an essential role in regulating β-cell function (Otani et al. 2004), any compound such as visfatin which may interact with insulin signalling could potentially elicit significant regulatory effects.
Interestingly, previous studies have given conflicting results regarding a role for visfatin in the β-cell with visfatin being reported to be not only a positive regulator of pancreatic β-cell function (Revollo et al. 2007) but also a negative one (Lopez-Bermejo et al. 2006), suggesting that the role of visfatin in diabetes is yet to be clearly elucidated.
This study therefore aimed to clarify the actions of visfatin in the pancreatic β-cell by investigating not only its effects on insulin secretion and insulin receptor phosphorylation, but also its regulation of the expression of a panel of diabetes-related genes in mouse pancreatic β-cells.
Materials and methods
Cell culture and treatment
All materials were purchased from Sigma unless otherwise stated.
β-TC6 cell line (BTC) cells were purchased from ATCC (LGC Promochem, Teddington, UK). Cells between passages 30–38 were cultured in standard 11 mmol/l glucose RPMI-1640 (supplemented with 2 mmol/l l-glutamine, 50 IU/ml penicillin and 50 μg/ml streptomycin, and 10% foetal bovine serum) prior to treatment.
For experimental treatments, cells were seeded into 6-well plates (gene and protein expression) or in 96-well plates (for insulin secretion, extracellular signal-regulated kinase (ERK)1/2 activation and cell viability assays). Cells were serum starved for 4 h and were subsequently incubated with serum-free RPMI containing either recombinant visfatin (Axxora, Nottingham, UK) NMN, U0126 (Promega) or FK866 (AxonChem Groningen, Netherlands) for appropriate time courses.
PCR array analysis and real-time PCR
For gene expression experiments, total RNA was extracted using SV Total RNA Isolation kit (Promega). RNA was DNase treated and 1 μg total RNA was reverse transcribed using a first-strand cDNA synthesis kit (PrimerDesign, Southampton, UK). Ninety-eight microlitres of the resulting cDNA were mixed with 1225 μl of SYBR Green PCR master mix (PrimerDesign) and 1127 μl of PCR water (Ambion, Warrington, UK) and were aliquoted out into a commercially available diabetes pathway-specific real-time PCR array (Tebu-Biosciences, Peterborough, UK). Samples were run as duplicates for 40 cycles using an iCycler thermal cycler (Bio-Rad). Cycle thresholds were measured, and the relative expression of genes was calculated by comparison of Ct values. The mean for all samples was calculated and subsequently normalized to the housekeeping genes β-glucoronidase, hypoxanthine phosphoribosyltransferase 1, heat shock 90 kDa protein 1β, β-actin and glyceraldehydes-3-phosphate dehydrogenase. Melt-curve analysis was used to confirm single amplicon production for each gene tested. Genomic contamination was assessed by non-reverse transcribed samples (RT-ve). Pre-validated SYBR Green PCR primers (PrimerDesign) were used to confirm PCR array results in a separate set of experiments (Supplementary Figure 1, see section on supplementary data given at the end of this article).
Cells were treated with a dose range (0–200 ng/ml) of visfatin for 1 h. Supernatants were collected and centrifuged to remove any cells, and insulin content was measured using a mouse insulin ELISA kit (Mercodia, Uppsala, Sweden).
Insulin receptor activation
To measure the effect of visfatin on insulin receptor activation, sandwich ELISA using phospho-specific anti-insulin receptor antibodies was employed (R&D, Abingdon, UK). Phospho-specific western blotting was used to qualitatively confirm the results. Cells were cultured as described above and were treated with visfatin in a concentration-dependent (0–200 ng/ml) and time-dependent (0, 2, 5, 15, 30 and 60 min) manner. Cell lysates were collected using RIPA buffer containing phosphatase and protease inhibitors for assay. Insulin treatment of 100 nmol/l was used as a positive control for insulin receptor activation.
ERK1/2 activation assay
To study the activation of intracellular p42–44 ERK signalling, a proprietary phospho-ERK1/2 ELISA was used (RayBiotech, Norcross, GA, USA). Cells were plated out in 96-well plates and treated over a time period (0, 2, 5, 15, 30 and 60 min) with 200 ng/ml visfatin before being fixed and assayed. Five-minute phorbol myristate acetate (PMA) treatment was used as a positive control for ERK1/2 activation, n=5.
Statistical analysis for PCR array was undertaken using a Mann–Whitney U test. For insulin secretion, ERK1/2 activation and cell viability experiments, ANOVA followed by Tukey's post-hoc test was used; P≤0.05 was taken as a point of significance.
Visfatin regulates mRNA expression of key diabetes genes in BTC cells
Real-time SYBR Green PCR array analysis of a diabetes-specific gene array was performed. Data analysis demonstrated that 24-h treatment of BTC cells with 50 ng/ml recombinant visfatin induced significant changes in expression (ranging from P=0.0031 to P=0.0434) compared to untreated cells in 12 of the 84 genes tested (Table 1).
Visfatin regulation of β-cell mRNA expression
|Fold change||P value|
|Compared to control|
|ATP citrate lyase||2.701||0.471|
|Adrenergic receptor 1a||4.925||0.518|
|Adrenergic receptor 3b||1.231||0.334|
|V-Akt murine thymoma viral oncogene homolog 2||24.251||0.002|
|Chemokine ligand 5||7.294||0.052|
|Chemokine receptor 2||3.175||0.330|
|CCAAT/enhancer-binding protein (C/EBP), α||4.387||0.407|
|Cytotoxic T-lymphocyte-associated protein 4||8.574||0.167|
|Dual specificity phosphatase 4||3.647||0.419|
|Ectonucleotide pyrophosphatase/phosphodiesterase 1||1.516||0.663|
|Forkhead box C2||1.741||0.831|
|Forkhead box G1||1.414||0.801|
|Forkhead box P3||−1.289||0.087|
|Glucose-6-phosphatase, catalytic subunit||4.925||0.075|
|Glucose-6-phosphate dehydrogenase 2||1.910||0.775|
|Glucose transporter type 4||2.701||0.624|
|Glucagon-like peptide 1 receptor||1.072||0.284|
|Glycerol-3-phosphate dehydrogenase 1 (soluble)||2.962||0.172|
|Glycogen synthase kinase 3β||−1.447||0.266|
|Heme oxygenase (decycling) 1||3.647||0.202|
|Intercellular adhesion molecule 1||−1.260||0.363|
|IGF-binding protein 5||2.764||0.844|
|Inhibitor of nuclear factor κ-B kinase β subunit||1.516||0.727|
|IL4 receptor, α||12.409||0.103|
|Inositol polyphosphate phosphatase-like 1||2.462||0.371|
|Insulin receptor substrate 1||4.287||0.320|
|Mitogen-activated protein kinase 14||2.639||0.771|
|Mitogen-activated protein kinase 8||−1.122||0.039|
|Neurogenic differentiation 1||1.122||0.118|
|Nuclear factor κ-B DNA binding subunit||40.317||0.019|
|Nitric oxide synthase 3||3.980||0.360|
|Nuclear respiratory factor 1||1.587||0.391|
|Pancreatic and duodenal homeobox 1||2.047||0.922|
|Poly (ADP-ribose) polymerase family member 1||1.176||0.169|
|Paired box gene 4||16.000||0.114|
|Phosphoenolpyruvate carboxykinase 1||3.482||0.108|
|Phosphatidylinositol-3-kinase catalytic δ polypeptide||2.095||0.967|
|Phosphoinositide-3-kinase, regulatory subunit 1 (α)||2.297||0.947|
|PPARγ, coactivator 1α||5.528||0.404|
|Protein tyrosine phosphatase, non-receptor type 1||2.297||0.904|
|Phosphorylase, glycogen, liver||2.000||0.777|
|RAB4A, member RAS oncogene family||1.176||0.284|
|Serine (or cysteine) peptidase inhibitor, clade E, member 1||1.289||0.553|
|Solute carrier family 14 (urea transporter), member 2||2.351||0.874|
|Synaptosome-associated protein, 23 kDa||−1.047||0.417|
|Synaptosome-associated protein, 25 kDa||−2.144||0.128|
|Superoxide dismutase 2||1.414||0.692|
|Sterol regulatory element binding transcription factor 1||1.721||0.813|
|Syntaxin-binding protein 1||−11.055||0.319|
|Syntaxin-binding protein 4||2.962||0.405|
|Transforming growth factor, β1||54.443||0.155|
|TNF receptor superfamily, member 1a||1.662||0.737|
|TNF receptor superfamily, member 1b||2.406||0.861|
|Tribbles homolog 3||2.351||0.953|
|Uncoupling protein 2||−1.289||0.005|
|VAMP-associated protein A||−30.555||0.254|
|Vascular endothelial growth factor A||−4.702||0.390|
SYBR Green real-time PCR array analysis of diabetes-related genes. Treatment with 50 ng/ml visfatin resulted in a significant up- or down-regulation of 12 genes compared to untreated cells. Data are representative of three independent experiments. HNF, hepatocyte nuclear factor; IL, interleukin; PPAR, peroxisome proliferator-activated receptor; TNF, tumour necrosis factor; VAMP, vesicle-associated membrane protein. Figures in bold represent statistically significant finding.
Visfatin but not NMN increases insulin secretion at low glucose
Visfatin has previously been suggested to have a role in glucose-induced insulin secretion. We therefore investigated the effect of a static 1-h incubation of BTC cells with a range of visfatin concentrations at low and high glucose levels. Incubation with 200 ng/ml visfatin for 1 h at 2.2 mmol/l (low) glucose caused a significant 46% increase in insulin secretion compared to control (P<0.001; Fig. 1). Lower concentrations of visfatin (0–100 ng/ml) caused no significant change in insulin secretion. Co-incubation of visfatin with the specific NAMPT inhibitor FK866 significantly blocked the insulinotropic effect of visfatin (P<0.05) compared to 200 ng/ml visfatin alone. No effect of visfatin on insulin secretion at high glucose was seen (data not shown). To investigate the involvement of ERK1/2 mitogen-activated protein kinase (MAPK) in visfatin-induced insulin secretion, cells were co-incubated with 200 ng/ml visfatin, and the ERK1/2 inhibitor U0126 and insulin secretion was measured. Co-incubation of cells with 200 ng/ml visfatin and 1 μmol/l U0126 caused no significant difference in insulinotropic effect compared with the treatment with visfatin alone. Incubation of cells with NMN alone resulted in a small non-significant increase in the levels of insulin secretion in the BTC cells (P=0.1).
Incubation with visfatin induces insulin receptor phosphorylation
Visfatin has previously been described as having the ability to activate the insulin receptor (Chan et al. 2007). We therefore investigated the activation of the insulin receptor in BTC by visfatin. Insulin receptor activation was assessed by phospho-specific insulin receptor sandwich ELISA. Using 50, 200 and 500 ng/ml visfatin, significant increases in the level of insulin receptor phosphorylation were seen after 15 min (P<0.001), with increases being greater at 200 and 500 ng/ml than at 50 ng/ml. This phosphorylation decreased over the subsequent 45 min. NMN alone also caused a significant increase in insulin receptor phosphorylation at the same time point (P<0.05). Co-incubation of cells with 200 ng/ml visfatin and the specific visfatin inhibitor FK866 caused a partial reduction in the level of insulin receptor phosphorylation compared to cells treated with 200 ng/ml visfatin alone (Fig. 2).
Visfatin and NMN cause activation of ERK1/2 but not of p38 MAPK
Visfatin has previously been shown to cause activation of the ERK1/2 and p38 MAPKs (Chen et al. 2006, Adya et al. 2008a,b). We therefore investigated the activation of these MAPKs by both visfatin and NMN. Treatment of cells with 200 ng/ml visfatin caused a significant 122% up-regulation in ERK1/2 phosphorylation after 15 min (P<0.001; Fig. 3). This result was mirrored by a similar 15-min increase in ERK1/2 phosphorylation of 105% by NMN (P<0.001; Fig. 3). Co-incubation with the specific visfatin inhibitor FK866 significantly reduced ERK1/2 activation by visfatin compared to visfatin alone (P<0.05; Fig. 3). Neither visfatin nor NMN caused any significant change in p38 activation compared to basal (Fig. 3). PMA positive control showed an expected significant increase in both p38 and ERK1/2 MAPK activation.
Despite significant interest and controversy over the effects of visfatin and insulin receptor signalling, there have been relatively little data published which have suggested precisely what the biological functions of visfatin are, and more importantly, how these functions might fit into the models of metabolic disorder. Here, we can report for the first time that visfatin is able to regulate the mRNA expression of a variety of genes which are key to maintaining normal pancreatic β-cell function. We found that 12 of the 84 genes tested were significantly up- or down-regulated by treatment with 50 ng/ml visfatin (a concentration only slightly higher than that reported in the circulation of obese/diabetic subjects (Chen et al. 2006); Table 1), including insulin and syntaxin 4 (a protein known to facilitate glucose-stimulated insulin secretion (Spurlin & Thurmond 2005)). We have subsequently confirmed visfatin regulation of mRNA expression of several of these genes using a separate real-time PCR using pre-validated primers (Supplementary Figure). Previously published reports have highlighted the importance of these genes in β-cell function, particularly hepatocyte nuclear factor (HNF)4α and HNF1β (Maestro et al. 2007), two genes which form single gene defects in forms of maturity onset diabetes of the young (MODY1 and MODY5 respectively). The marked up-regulation of these important transcription factors fits well with the observation here that visfatin causes an increase in insulin secretion (Fig. 2), as they are both thought to be involved in insulin gene transcription (Bernardo et al. 2008).
Significant increases were also observed in some unexpected genes, including CD28 (a 97-fold increase). CD28 is more commonly associated with type 1 diabetes and immune responses, and any potential role in T2D is still unclear. The increase we report in nuclear factor-κB (NF-κB) fits well with previously published data regarding visfatin and NF-κB activity in endothelial cells (Adya et al. 2008a,b). Visfatin has previously been suggested to have a role in the regulation of insulin secretion, where NMN was shown to be able to rescue insulin secretion in islets isolated from visfatin knockout mice (Revollo et al. 2007). This report, however, focussed on the intracellular actions of visfatin, as opposed to administered visfatin. Our observations suggest that visfatin therefore has an important role as both an intra- and an extracellular regulator of pancreatic β-cell function.
Interestingly, the significant increase in insulin secretion from BTC cells seen here was blocked by co-incubation with the specific NAMPT inhibitor FK866 (Fig. 1). This might initially suggest that the mechanism by which visfatin is regulating insulin secretion is through its ability to synthesize NMN (as FK866 non-competitively binds the enzymatic region of visfatin which normally binds nicotinamide; Kim et al. 2006). However, NMN itself did not significantly increase insulin secretion in the BTC cells, suggesting a more complex mechanism for this action of visfatin. It is possible that FK866 not only blocks the ability of visfatin to bind nicotinamide, but also blocks other as yet undescribed interactions, possibly those involving the insulin receptor. Clearly, further research is needed to fully assess the physiological relevance of these observations.
Curiously, we found that at a supraphysiological concentration (500 ng/ml) of visfatin caused a decrease in cell viability over 24 h (data not shown). Although most reports of the circulating levels of visfatin are well below this concentration, some have suggested far higher circulating concentrations of visfatin, potentially placing this within the physiological range (Chan et al. 2007), which suggests that visfatin may have a more complex role in the pancreas than our data suggest. One previous report has linked β-cell dysfunction to elevated serum visfatin levels (Lopez-Bermejo et al. 2006), and the potential exists that although a short-term increase in insulin secretion is seen, longer term effects of exposure to elevated visfatin might be more deleterious to the β-cell.
The role of visfatin in insulin receptor signalling is a controversial one. We report for the first time that in the pancreatic β-cell, visfatin has a definite role in insulin receptor signalling, with a significant up-regulation of insulin receptor phosphorylation occurring when BTC cells are exposed to a range of visfatin concentrations. This matches previous reports of an interaction of visfatin with the insulin signalling pathway in at least one other tissue (Xie et al. 2007). As insulin signalling has been shown to be essential for β-cell function (Otani et al. 2004), this confirms the importance of this role for visfatin in the pancreas, a role which exists potentially via multiple mechanisms (both intra- and extracellular) involving both insulin secretion and signalling.
In conclusion, these data suggest that visfatin has a role in multiple aspects of pancreatic β-cell biology, including a role in the regulation of insulin secretion and receptor signalling. These findings confirm the previously reported modulatory role of visfatin in the insulin signalling pathway, and suggest that these actions occur via its ability to synthesize NMN.
This is linked to the online version of the paper at http://dx.doi.org/10.1677/JME-09-0071.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
This research was partly funded by an Early Research Award Scheme (University of Wolverhampton).