Introduction

Fatty acids play a pivotal role in a variety of metabolic controls and cell signalling processes in various tissues [1]. In particular, short-term exposure of fatty acids to pancreatic beta cells augments glucose-stimulated insulin secretion (GSIS), a process in which fatty acid-derived metabolites such as long-chain fatty acyl-CoAs act as crucial effectors [2]. However, the entire mechanism whereby fatty acids acutely induce GSIS augmentation has not been fully elucidated [3]. In contrast, chronic fatty acid exposure causes marked deterioration of beta cell function, which is referred to as lipotoxicity [4, 5].

Several investigators have recently demonstrated that fatty acids act as ligands for membrane-bound G-protein-coupled receptors such as GPR40 [3, 6, 7], GPR41, GPR43 [8, 9] and GPR120 [10]. GPR40 is preferentially expressed in pancreatic beta cells in rodents and augments GSIS after acute exposure to mid- and long-chain fatty acids [6]. Silencing GPR40 with the small interfering RNA (siRNA) system suppresses long-chain fatty acid-induced GSIS augmentation in pancreatic beta cells [6]. A recent study of GPR40 knockout mice and beta-cell-specific GPR40 transgenic mice suggested a physiological and pathophysiological role for GPR40 in insulin secretion and diabetes mellitus [11]. Although these findings implicate GPR40 in insulin secretion and glucose metabolism in rodents, little is known about the physiological significance of GPR40 in humans.

In this context, we investigated GPR40 gene expression in the pancreas and in islet cell tumours collected during surgery. We also explored the potential role of GPR40 in beta cell function in humans.

Subjects and methods

Participants, tissue sampling and pancreatic islet preparation

Seventeen patients with pancreatic tumours provided written informed consent to participation in the present study, which was approved by the Ethical Committee on Human Research of Kyoto University Graduate School of Medicine (No. 508, 2003), and conducted according to the principles of the Declaration of Helsinki. Table 1 summarises the patient profiles. Patients who underwent pancreatectomy (patients 1–12) were numbered according to the GPR40 mRNA level in the pancreas. None of the 12 patients were treated with oral glucose-lowering agents or insulin. Pancreatic, intestinal and hepatic tissues free of tumour invasion as well as islet cell tumour tissues were obtained at the time of surgery (Table 1). Islet tissues from three patients (patients 9–11) were promptly isolated from the pancreas through the mince method [12]. Briefly, the pancreas was finely minced by hand for 15–30 min on ice and digested at 37°C with 600 IU/ml of type V collagenase (Sigma, St Louis, MO, USA) in Hanks’ Balanced Salt Solution (HBSS) containing 1% bovine serum albumin (Fraction V, Sigma) for 20 min. The digested tissue was washed three times in cold HBSS. After dithizone staining, pancreatic islets were manually collected using a stereo microscope (SZ-STB1; Olympus, Tokyo, Japan).

Table 1 Clinical profiles of patients and tissues
Full size table

Quantification of mRNA expression of GPR40 and other receptor genes

We measured mRNA expression of the GPR40 gene as well as genes encoding sulfonylurea receptor 1 (ABCC8, previously known as SUR1) [13], glucagon-like peptide 1 receptor (GLP1R) [14, 15] and somatostatin receptor (SSTR) 3 and 5 [16] using the following method. Total RNA was extracted using the Qiagen RNeasy Mini Kit (Qiagen, Hilden, Germany) [17]. First-strand cDNA was synthesised by random hexamer-primed reverse transcription using SuperScript II reverse transcriptase (Invitrogen, Carlsbad, CA, USA) [18]. The mRNA level was quantified by the TaqMan PCR method using an ABI Prism 7700 Sequence Detector (Applied Biosystems, Foster City, CA, USA) as described [19]. To calculate the copy number of each mRNA, standard curves were generated using synthesised oligo DNA fragments (Proligo Japan, Kyoto, Japan) containing the PCR amplicon region. The mRNA expression in each gene was normalised to that of GAPDH (i.e. the mRNA level for each gene was expressed as receptor/GAPDH [copy/copy]). Table 2 summarises the sequences of primers and probes used in the present study. The primers or probes were designed not to cover any reported single-nucleotide polymorphisms [20–23].

Table 2 Sequences of TaqMan primers and probes
Full size table

Data analysis on glucose homeostasis

The insulin AUC was calculated using the trapezoidal rule from OGTT data. We evaluated beta cell function and systemic insulin resistance using the insulinogenic index (n=7) and the homeostasis model assessment of insulin resistance (HOMA-IR) (n=10), respectively. The insulinogenic index was calculated as the ratio of the insulin concentration (pmol/l) increment to the glucose concentration (mmol/l) increment at 30 min into the OGTT (Δ30insulin/Δ30glucose) [24]. HOMA-IR was calculated in fasting conditions as plasma insulin (pmol/l)×blood glucose (mmol/l)/22.5 [25, 26]. The difference in the patient numbers for the two indices is based on the difference in data availability, such as blood glucose and insulin levels at 30 min during the OGTT.

Statistical analysis

The relationship between the GPR40 mRNA level in the pancreas and clinical or metabolic profiles was tested using Spearman’s rank correlation and a p value of less than 0.05 was considered significant. The statistical significance of differences in two groups was assessed using unpaired two-tailed t-test and a p value of less than 0.05 was considered significant (Statcel, Social Research Information, Tokyo, Japan).

Results

Expression of GPR40 mRNA in human pancreas and isolated islets

The expression of GPR40 mRNA in human tissues was assessed by TaqMan PCR using total RNA samples from patients who underwent pancreatectomy and/or other pertinent surgeries. GPR40 mRNA was detected in all human pancreases examined (n=12), and at higher levels than those in the duodenum, jejunum or liver (Fig. 1a). The inter-individual variability in the GPR40 mRNA levels was considerable in the human pancreas. The GPR40 mRNA level in the pancreas was not significantly different between sites (head vs body) of the pancreas or between men and women. The GPR40 mRNA level in the pancreas did not correlate significantly with age. The GPR40 mRNA level in fresh islets that were isolated from pancreatic tissues (n=3) was approximately 20-fold higher than that in the pancreas from the same patients (Fig. 1b).

Fig. 1
figure 1

TaqMan quantitative analyses of GPR40 mRNA expression in human pancreatic tissues. Total RNA extracted from various tissues was analysed. a GPR40 mRNA was detected in all human pancreas specimens examined, at higher abundance than in the duodenum, jejunum or liver. Patients 1–12 were numbered according to their relative level of expression of GPR40 mRNA. b GPR40 mRNA level in isolated islets was ∼20-fold higher than in pancreatic tissues from the same patients. The numbers in a and b correspond to those in Table 1. c Estimated mRNA copy number of the GPR40 gene was similar to or higher than those for genes encoding sulfonylurea receptor 1 (ABCC8), glucagon-like peptide 1 receptor (GLP1R) and somatostatin receptors 3 (SSTR3) and 5 (SSTR5) (n=3). Data are means±SEM. Open bars pancreas; closed bars isolated islets

Full size image

To gain further insight into the expression of the GPR40 gene in pancreatic islets, we analysed the expression of genes known to be expressed abundantly in the pancreatic islets. The estimated mRNA copy number of the GPR40 gene in isolated islets was comparable to or higher than those of genes encoding receptors for sulfonylurea, glucagon-like peptide 1 and somatostatin (Fig. 1c), suggesting that high levels of the GPR40 gene are expressed in pancreatic islets.

Expression of GPR40 mRNA in insulinoma tissues

We analysed the expression of GPR40 mRNA in islet cell tumours, including insulinoma (n=3), glucagonoma (n=1) and gastrinoma (n=1). GPR40 mRNA was detected in tissue extracts from three cases of insulinoma (Fig. 2), which was comparable to that in pancreatic islets (Fig. 2). GPR40 mRNA was below detectable levels in tissue extracts from glucagonoma or gastrinoma (Fig. 2).

Fig. 2
figure 2

TaqMan quantitative analyses of GPR40 mRNA expression in human islet cell tumours. Total RNA extracted from pancreases (n=12), pancreatic islets (n=3) and islet cell tumours (patients 4, 14, 15, 16, 17) was analysed. GPR40 mRNA was abundantly expressed in three cases of insulinoma among islet cell tumours. Open bar, pancreas; hatched bar, pancreatic islets; closed bars, islet cell tumours. The patient numbers in the figure correspond to those in Table 1. The GPR40 mRNA level in the pancreas and pancreatic islets is expressed as mean±SEM. INS insulinoma, GLU glucagonoma, GAS gastrinoma

Full size image

The correlation between the GPR40 mRNA level in the pancreas and insulinogenic index is positive

To understand the physiological role of GPR40 in humans, we examined the relationship between the GPR40 mRNA level in the pancreas and various metabolic parameters. Table 3 summarises the metabolic profiles of patients who underwent pancreatectomy. The GPR40 mRNA level in the pancreas did not correlate significantly with BMI, fasting plasma glucose, plasma glucose at 2 h under the OGTT (2h-PG) or insulin AUC. Furthermore, the GPR40 mRNA level in the pancreas did not correlate significantly with the HOMA-IR (n=10) (Fig. 3a). The significant positive correlation between the GPR40 mRNA level in the pancreas and the insulinogenic index was notable (n=7) (p=0.044, r=0.82) (Fig. 3b). To verify the significant association, correlation was tested between the GPR40 mRNA level in the pancreas and the HOMA-IR, using data from the same patients from whom data on the insulinogenic index were available (n=7), and we confirmed that there was no significant correlation between the GPR40 mRNA level and the HOMA-IR (p=0.86, r=0.07). The GPR40 mRNA level in the pancreas was not significantly associated with HbA1c or fasting triglyceride.

Table 3 Metabolic profiles of patients studied
Full size table
Fig. 3
figure 3

Positive correlation between GPR40 mRNA level in the pancreas and insulinogenic index. a Relationship between GPR40 mRNA level in the pancreas and the HOMA-IR (n=10). GPR40 mRNA level and the HOMA-IR did not correlate significantly (r=0.11, p=0.73). b Relationship between GPR40 mRNA level in the pancreas and the insulinogenic index (n=7). Correlation was marginally but significantly positive between GPR40 mRNA level and the insulinogenic index (r=0.82, p=0.044). Spearman’s rank correlation test was used to determine p and r values. The solid line is the regression line

Full size image

Discussion

GPR40 is abundantly expressed in murine pancreatic beta cells [3, 7], where it mediates the fatty acid-induced augmentation of GSIS in vitro [6]. Long-chain fatty acids act as ligands for human GPR40 in vitro [3, 7, 23]. Additionally, two studies suggest the possible involvement of GPR40 in the proliferation and cell function of breast cancer [27, 28]. Two laboratories reported the possible relationship between variation of the GPR40 single-nucleotide polymorphisms and insulin secretion in humans [22, 23], where the results were inconsistent. The physiological role of GPR40 in humans remains obscure.

We here demonstrate for the first time that a large amount of GPR40 mRNA is expressed in pancreatic islets in humans using expeditiously isolated islets from pancreatic tissue. TaqMan analysis revealed that levels of GPR40 mRNA expression in pancreatic islets were 20-fold higher than those in the pancreas in the same individuals. Notably, the mRNA level of the GPR40 gene in isolated islets was comparable to those of genes encoding GLP1R, ABCC8 (SUR1) and SSTRs, all of which are abundantly expressed in human pancreatic islets [13–16].

The present study demonstrates that a large amount of GPR40 mRNA is expressed solely in three cases of insulinoma among islet cell tumours. The finding suggests that the GPR40 mRNA detected in insulinoma is attributed to the type of islet cell tumour rather than the inclusion of endothelial, neuronal or other types of cells. Collectively, these data prompt us to speculate that GPR40 is expressed mainly in beta cells in the human pancreatic islet.

It is noteworthy that the mRNA levels of GPR40 and ABCC8 (SUR1) were comparable in human pancreatic islets, because ABCC8 (SUR1) is abundantly expressed in human pancreatic beta cells and the protein functions as a target of sulfonylurea agents [29]. These findings suggest that GPR40 has also some functional property in terms of insulin secretion in humans. In the present study, the GPR40 mRNA level in the pancreas significantly correlated with the insulinogenic index rather than the HOMA-IR, supporting the notion that GPR40 is involved in the regulation of insulin secretion in humans.

A recent study of GPR40 knockout mice and beta-cell-specific GPR40 transgenic mice provided evidence that GPR40 is involved in the pathophysiology of glucose intolerance and beta cell lipotoxicity [11]. In this context, it is important to note that patients enrolled in the present study were neither obese nor severely diabetic. Thus, clarification of the pathophysiological role of GPR40 in human diabetes must await further investigation in patients with a wider range of body weight, glucose intolerance or dyslipidaemia.

As pancreatic tissues are very vulnerable to postmortem autolysis, specimens obtained at operation have a great advantage for the precise analysis of the GPR40 mRNA level. Pancreatic biopsy is rarely conducted because of the risk of pancreatitis and is not justified in those without severe illness [30]. Thus, we analysed human pancreatic tissues collected during surgery. To our knowledge, specific antibody against human GPR40 has not been available, hence the lack of analyses of GPR40 protein expression in the present study.

In summary, the present study demonstrates that GPR40 mRNA is abundantly expressed in human pancreatic islets and insulinoma. The results provide evidence for GPR40 expression in pancreatic beta cells and its involvement in insulin secretion in humans.