Positive interplay between FFAR4/GPR120, DPP-IV inhibition and GLP-1 in beta cell proliferation and glucose homeostasis in obese high fat fed mice

.


Introduction
Interest in the activation of free fatty acid sensing receptors (FFARs) as a therapeutic target has increased over the years together with research on their activation with FFA ligands, thereby revealing extra potential for new therapeutic treatments for type 2 diabetes [5,9,15,19,21,29].Activation of GPCRs expressed in pancreatic islets and intestinal cells triggers secretion of insulin and incretin hormones that are key players in aspects of glucose homeostasis including maintenance of islet morphology, beta cell regeneration and hormone release, thereby suggesting potential as new therapeutic agents [23,24,28,35,39].
Expression of FFAR4/GPR120 has been identified in the thymus, intestine, taste buds, macrophages, adipose tissue, brain, pancreas, lung, and pituitary [14].The FFAR4/GPR120 receptor plays a pivotal role in the maintenance of intestinal homeostasis and its expression in the GI tract increases the release of insulinotropic hormones such as GLP-1, GIP, and CCK from enteroendocrine cells [13,16,22].Interestingly, FFAR4/GPR120 knockout mice have demonstrated increased adiposity, insulin resistance and glucose intolerance compared to wild-type mice [30,32].
Alongside endogenous ligands of FFAR4/GPR120, several synthetic agonists have shown affinity for the activation of FFAR4/GPR120, such as GW9508, GSK137647, CpdA, PBI-4547, GPU-028 and TUG-891.TUG-891 was developed from GPR40/120 dual agonist GW9508 as a selective FFAR4/GPR120 agonist [5] and many studies develop selective ligands by mimicking endogenous LCFA agonists [25].Oh et al. 2014 designed CpdA to be high selective for FFAR4/GPR120 and lack specificity for FFA1 [31].Compound A is an orally administered small molecule agonist with high selectivity for GPR120 (logEC50 (M) = − 7.62 ± 0.11) and high potency (EC50 0.35 µM), as well as demonstrating negligible interactions with GPR40 (FFAR1) [31].An array of endogenous and synthetic ligands have been found with the ability to induce insulin release in a glucose dependent manner from clonal beta cells and isolated mice islets [28].Alpha-linolenic acid (ALA) activation of FFAR4/GPR120 has been extensively studied, with acute and long-term administration of this agonist resulting in elevated incretin hormone, GLP-1 secretion in vitro and in vivo [37].Palmitoleic acid and ALA also influenced the release of CCK in vivo and in mouse intestinal enteroendocrine cells [37].
The purpose of this study was to investigate the effects of activation of FFAR4/GPR120 with CpdA in combination with a DPP-IV inhibitor in obese mice to determine the role of FFAR4/GPR120-mediated effects in the maintenance of pancreatic and intestinal function using pre-clinical studies, islet cell architecture and downstream signalling analysis.

Chronic administration of FFAR4/GPR120 agonist CpdA in combination with sitagliptin in vivo HFD diabetic mice
In long term studies, obese, insulin-resistant, HFD mice received once daily oral administration of saline vehicle (0.9% [w/v] NaCl), FFAR4/GPR120 agonist CpdA (0.1µmol/kg bw), Sitagliptin (50 mg/kg bw), or CpdA and Sitagliptin in combinational therapy for 21 days [21].Mice were maintained on high fat diet throughout the study, with food intake, blood glucose, and plasma insulin and GLP-1 monitored at regular intervals.AUC of food intake was carried out based on weighed food at regular time points across the study duration, calculating food consumption since last time point.Following the 21-day treatment period, glucose tolerance (18 mmol/kg bw; oral; 18-hr fasted) and insulin sensitivity (40 U/kg bw bovine insulin; i.p; non-fasted) tests were conducted.Blood glucose, collected from the cut tip of tail, was measured using an Ascencia Contour blood glucose meter (Bayer Healthcare, UK).Plasma samples (collected during and after study for insulin and GLP-1, after study for cholesterol and triglycerides) were used to assess insulin by in-house radioimmunoassay [11], total GLP-1 by ELISA (Millipore, Watford, UK), and Triglyceride, LDL cholesterol and total cholesterol using a RX Daytona+ (Randox Laboratory, Co. Antrim, UK).At the end of the study, pancreas and small intestine were excised and processed as outlined below.Terminal analyses was also performed on all carcasses using Dual energy X-ray absorption (DEXA) for analysis of %fat [27].This was used to determine fat mass in raw weight.DEXA scanner was calibrated, and quality control performed with the aluminium/lucite phantom (0.069 g/cm 2 , 12.0% fat) using a PIXImus system (software version 1.4x).No adverse effects were noted with any of the treatment regimens.

Tissue distribution of FFAR4/GPR120 by immunohistochemistry
Pancreatic tissues were removed at 21 days for immunohistochemistry. Pancreatic tissue was fixed in 4% PFA/PBS, embedded in paraffin wax, sectioned at 8μm, and morphometric analysis performed on every tenth section throughout each pancreas.Sections were mounted onto polylysine-coated slides and dried on a hot plate.After incubation, wax was removed and tissue re-hydrated in ethanol (100%), ethanol (95%), ethanol (80%) and distilled water for 5 min each.Slides were incubated in 50 mM sodium citrate for 20 min at 90 • C for antigen retrieval.BSA (2.5%) was added to each slide (200 µl) for 45 min.Slides were incubated overnight at 4 C with guinea pig anti-insulin (1:500) or guinea pig anti-glucagon (1:500).Rabbit anti-Ki67 (1:200) was incubated at 37 C for 2 hr.After washing in PBS, sections were incubated with anti-guinea pig Alexafluor 488 nm, anti-guinea pig Alexafluor 594 nm and Anti-Rabbit Alexafluor 594 nm (1:400; Molecular Probes (Life Technologies Ltd, Paisley, UK) for 45 min at 37 C and DAPI nuclear stain for 15 min at 37 C. Finally, slides were washed in PBS, mounted and analysed using a BX51 Olympus microscope equipped with an Olympus XM10 digital camera.

Statistical Analysis
All data was analysed with Prism (v.5.0,GraphPad Software Inc. CA, USA) and expressed as mean ± S.E.M. Results were compared using the Student's t-test (non-parametric, with two-tailed P values and 95% confidence interval)), one-way analysis of variance (ANOVA) or twoway analysis of variance (ANOVA) as appropriate.Area under the curve (AUC) was calculated using trapezoidal rule with baseline correction.Differences in data were considered to be statistically significant for p<0.05.

Effects of CpdA and Sitagliptin alone, or in combination, on glucose tolerance and insulin sensitivity
The chronic effects of activation of FFAR4/GPR120 via a once-daily administration of CpdA on metabolic health were assessed by an oral glucose tolerance test (OGTT) and insulin sensitivity test.CpdA was assessed alone and in combination with the DPP-IV inhibitor (sitagliptin).Long-term administration of CpdA alone and in combination significantly improved glucose tolerance (Fig. 1A,B).CpdA lowered glucose concentrations by 34% in obese HFD mice (Fig. 1A).CpdA in combination therapy improved glucose tolerance by 33% (p<0.001).At 30 min, the glucose spike was reduced by CpdA alone and in combination therapy (p<0.05 and p<0.001, respectively) (Fig. 1A).

Effects of CpdA and Sitagliptin alone, or in combination on food intake, fat mass and lipids in HFD mice
Food intake was reduced over the 21 days of treatment with CpdA (19%, p<0.01) and when given with sitagliptin (33%, p<0.001), compared to sitagliptin alone (AUC Fig. 2A).Fat mass, measured by DEXA, was reduced by sitagliptin (p<0.05) and in combination therapy (p<0.05,Fig. 2B).LDL cholesterol was decreased by 50% (p<0.05) and 89% (p<0.01)following long-term treatment with CpdA and combinational therapy, respectively (Fig. 2E).Interestingly, combinational therapy had an additive effect, reducing LDL cholesterol significantly compared to its individual components sitagliptin (p<0.001) and Compound A (p<0.01).

Effects of CpdA and Sitagliptin alone, or in combination, on pancreatic islet morphology in HFD mice
Islet morphology was assessed by immunohistochemical staining.Islet size, β-cell mass and area were determined, as shown in Fig. 4 (A-C), and complimented by measurement of α-cell area (Fig. 4D) and mass (Fig. 4E).Representative images, shown in Fig. 5, illustrate insulin (red) and glucagon (green) localisation as well as DAPI (blue) and all three combined, in lean (A-D), HFD (E-H), Sitagliptin (I-L), CpdA (M-P), and combination therapy (Q-T).
HFD mice exhibited a 30% reduction in β-cell area (p<0.001Fig. 4B) and a significant expansion of α-cell area (p<0.01Fig. 4D) compared to lean mice.CpdA treatment resulted in a 41% increase in β-cell area compared to HFD mice (p<0.001),restoring levels to that of lean control mice.A high fat diet induced a 55% increase in islet size, which was normalised by CpdA alone (p<0.05), and in combination with sitagliptin (p<0.05) (Fig. 4A).HFD mice exhibited increased β-cell mass (p<0.01)compared to lean mice, and none of the treatment interventions had any effect on beta cell mass.Alpha-cell area (p<0.01) and mass (p<0.001) were increased in HFD mice (Fig. 4D,E).CpdA treatment restored α-cell area by 49% (p<0.001) and α-cell mass by 31% (p<0.05) to lean control mice levels.

Effects of CpdA and Sitagliptin alone, or in combination, on FFAR4/ GPR120 expression, ERK 1/2 expression and beta-cell proliferation in HFD mice
Consumption of a high fat diet in pre-clinical studies resulted in a decrease of GPR120/FFAR4 expression (p<0.01) in the pancreas of HFD mice, compared to lean controls (Fig. 6A).FFAR gene expression was markedly raised with combination therapy compared to HFD mice (p<0.001) and also compared to CpdA alone (p<0.01) (Fig. 6A).

Discussion
Ever since the deorphanisation of FFAR4/GPR120 in 2005 [13] and the discovery of its involvement in the regulation of GLP-1 secretion, there has been significant interest in elucidating the role of this receptor in type 2 diabetes as well as designing specific and potent agonists to target this receptor.Acute effects of endogenous and synthetic agonists on glucose homeostasis have been studied in normal and high fat fed mice, revealing notable improvements in glucose tolerance as well as insulin sensitivity [28,31].In the present study, single administration of CpdA improved glucose tolerance by 34%, and insulin sensitivity by up to 27%, suggesting that specific FFAR4/GPR120 activation over a longer period has potential to improve blood glucose control in diabetes.Indeed there is growing evidence suggesting that incretins, specifically GLP-1, play a large role in the regulation of FFAR4/GPR120 beneficial effects, which is further supported by the expression of FFAR4/GPR120 on L-cells in the GI tract [5,22].Although disputed by others [33], the present study extends these observations by showing that the antidiabetic effects of FFAR4/GPR120 are partly mediated by GLP-1, as circulating GLP-1 and intestinal Gcg gene expression were increased with CpdA treatment.
The activation of FFAR4/GPR120 by CpdA resulted in significant reductions in food intake and may involve GLP-1, particularly as appetite was further reduced when CpdA was combined with Sitagliptin.GLP-1 is well known to induce satiety, which is thought to be mediated via many mechanisms including neural pathways from periphery and direct actions at GLP-1 receptors in the brain, specifically the hypothalamus, ventral tegmental area and the nucleus accumbens [3].Indirect mechanisms involving these areas include stimulation by GLP-1 produced by neurons in the nucleus tractus solitarii, which is in turn activated by vagal afferents stimulated by nutrients, other gut hormones and gastric distension [3].Furthermore, FFAR4/GPR120 has been found to inhibit ghrelin secretion from the stomach, implying a further role in hunger repression potentially through removing the acceleratory effect of ghrelin on gastric emptying [39,40].These inhibitory effects have also been attributed to GPR120 binding of Gαq/11, leading to phosphorylation of ERK via PLC, which may reduce expression of prohormone convertase 1, which is involved in the conversion of proghrelin to ghrelin [40].GPR120 is highly expressed in adipocytes and adipose tissue and may have an important role in adipose tissue development and adipocyte differentiation [30].FFAR4/GPR120 has been shown to impact energy balance through its role in adipose tissue, and activation can stimulate processes such as adipose tissue browning which helps regulate energy expenditure [36].Brown adipocytes show high thermogenic capacity and dissipate the stored energy in the form of heat.In this study, we found an increase in energy expenditure when mice were fed a high fat-diet, and no change with CpdA (Data not shown).
One major risk factor of type 2 diabetes is the development of heart problems including cardiomyopathy, heart disease, dyslipidaemia and stroke [20].In this study, the high fat diet resulted in an increase in total cholesterol and notable (but not significant) increase in triglycerides.Treatment with CpdA alone and in combination therapy decreased LDL cholesterol, by 50% and 89% respectively.
Combination therapy of FFAR4/GPR120 agonism with a DPP-IV inhibitor also resulted in significant additive improvements in glucose handling and insulin secretion compared to the individual treatments alone, and this was supported by markedly raised GLP-1 when the two treatments were given together.Moreover, satiety effects were further enhanced when CpdA was administered with Sitagliptin and there was a marked reduction in LDL cholesterol with the combination.The gain in body fat mass with the high fat diet was significantly decreased by combination therapy.Importantly, there were no discernible changes of behaviour in these mice and further beneficial effects of the two treatments was observed with increases in FFAR4/GPR120, ERK1 and ERK2 gene expression.
Islet β-cell dysfunction and demise is a key factor in the onset of hyperglycaemia and diabetes [6,7,18].Activation of FFAR4/GPR120 was shown to have direct beneficial effects on pancreatic β-cell health, as demonstrated by significant increases in proliferating β-cells and β-cell area, thereby reversing the detrimental impact of the high fat diet.This action may involve GLP-1 which is known to enhance β-cell expansion, survival and differentiation [4,17] as well as to decrease β-cell to α-cell transdifferentiation [38].The observation of augmented beta-cell proliferation was supported by increased expression of ERK1 and ERK2 following combination therapy with CpdA and Sitagliptin.ERK1 and ERK2 are proliferative signaling molecules of FFAR4/GPR120, implicating this receptor in anti-apoptotic effects, independently of GLP-1 [40].It is well-established that a decrease in β-cell number and a deficiency in the function of existing β-cells contributes to the development and detrimental effects of diabetes.ERK1 and ERK2 signalling cascade has a pivotal role to play in the cellular signalling pathways involving β-cell proliferation and survival.Activation of the ERK1/2 cascade by FFAR4/GPR120 agonism offers promise in diabetes therapy particularly with the β-cell expansion and β-cell proliferation observed with CpdA and combination therapy.
In conclusion, this study has demonstrated that chronic activation of FFAR4/GPR120 has beneficial effects on islet health and glucose homeostasis in a well established animal model of obesity-diabetes.The effect is partly dependent on activation of GLP-1 pathways and is significantly augmented by conjoint DPP-IV inhibition.Further clinical studies are needed to investigate the potential benefit of this oral combination in man.