Human Epicardial Fat Expresses Glucagon-Like Peptide 1 and 2 Receptors Genes

Abstract

Epicardial adipose tissue (EAT) is an easily measurable visceral fat of the heart with unique anatomy, functionality, and transcriptome. EAT can serve as a therapeutic target for pharmaceutical agents targeting the fat. Glucagon-like peptide-1 (GLP-1) and GLP-2 analogues are newer drugs showing beneficial cardiovascular and metabolic effects. Whether EAT expresses GLP-1 and 2 receptors (GLP-1R and GLP-2R) is unknown. RNA-seq analysis and quantitative real-time polymerase chain reaction (qRT-PCR) were performed to evaluate the presence of GLP-1R and GLP-2R in EAT and subcutaneous fat (SAT) obtained from 8 subjects with coronary artery disease and type 2 diabetes mellitus undergoing elective cardiac surgery. Immunofluorescence was also performed on EAT and SAT samples using Mab3f52 against GLP-1R. Our RNA-sequencing (RNA-seq) analysis showed that EAT expresses both GLP-1R and GLP-2R genes. qRT-PCR analysis confirmed that GLP-1R expression was low but detected by 2 different sets of intron-spanning primers. GLP-2R expression was detected in all patients and was found to be 5-fold higher than GLP-1R. The combination of accurately spliced reads from RNA-seq and successful amplification using intron-spanning primers indicates that both GLP-1R and GLP-2R are expressed in EAT. Immunofluorescence clearly showed that GLP-1R is present and more abundant in EAT than SAT. This is the first time that human EAT is found to express both GLP-1R and GLP-2R genes. Pharmacologically targeting EAT may induce beneficial cardiovascular and metabolic effects.

Introduction

Epicardial adipose tissue (EAT) is the visceral adipose depot of the heart [1]. EAT has unique embryology, anatomy, functionality, and transcriptome. EAT is highly enriched with genes involved in inflammation, coagulation, immune signaling, potassium transport and apoptosis when compared to subcutaneous fat in subjects with coronary artery disease (CAD) [2]. Due to its peculiar anatomical contiguity to the myocardium and bio-genetic features, EAT is thought to play an active role in CAD and diabetes-related atherosclerosis [3]. Clinically, given its rapid metabolism, organ fat specificity and simple measurability, EAT can serve as a modifiable risk factor and therapeutic target for pharmaceutical agents targeting the fat [4].

Glucagon-like peptide-1 (GLP-1) analogues are indicated for the treatment of type 2 diabetes mellitus. In addition to the well-established glucose-lowering effect, GLP-1 analogues can induce a modest weight loss and exert cardio-protective effects. Liraglutide, a GLP-1 analogue, has shown to reduce the risk of cardiovascular events, as recently reported in the LEADER study, although the mechanisms are still unclear [5, 6]. We have recently found that liraglutide causes a substantial and rapid EAT reduction in subjects with diabetes and obesity [7]. Whether the beneficial cardiovascular effects of GLP-1 analogues could be attributed to a direct drug-effect on EAT is unclear. The presence of GLP-1 receptors (GLP-1R) in EAT would support the hypothesis of a direct effect on the adipose tissue of the heart, but this data is currently unknown and unexplored. Moreover, recent experimental data suggest that glucagon-like peptide 2 (GLP-2) may produce beneficial effects on glucose metabolism in obese animal models [8, 9]. GLP-2 activation has been suggested to increase glucose incorporation in adipocytes and improve insulin resistance [8, 9]. However, whether human adipose tissue and in particular EAT may express GLP-2 receptor (GLP-2R) is also unknown. To answer these questions, RNA-seq analysis and quantitative real time polymerase chain reaction (qRT-PCR) were performed to evaluate the presence of GLP-1R and GLP-2R (the genes that encode GLP-1R and GLP-2R) in the EAT obtained from subjects with CAD and type 2 diabetes mellitus. RNA-seq has several advantages over microarray analysis, such as covering all aspects of the transcriptome without any prior knowledge of it. Immunofluorescence was also performed to further evaluate the presence of GLP-1R on EAT and SAT samples.

Subjects and Methods

We examined RNA-sequencing data from EAT samples obtained during elective cardiothoracic surgery to evaluate the expression of GLP-1R and GLP-2R genes in subjects with type 2 diabetes and CAD (n = 5) as well as in patients without type 2 diabetes (n = 3).

EAT was collected during cardiothoracic surgeries of all 8 patients who signed an informed consent form were enrolled in the study. Subjects were divided in 2 groups based on the presence or absence of type 2 diabetes defined as known past medical history of diabetes, current use of antidiabetic agents (either orals or injectable medications), or hemoglobinA1c (HbA1c)≥42 mmol/mol at least 1 month before the enrollment in the study. Of the 8 participants, 5 patients (2M/3F) aged 60–75 (69.6 ± 5.81 years) were found to be diabetics whereas the other 3 patients (2M/1F), aged 51–66 (58.0 ± 7.55 years) had no past medical history of diabetes, a HbA1c≤5.7% and were not taking antidiabetic medications on pre-operative testing. The absence or presence of CAD was determined by the cardiologists and cardiac surgeons based on clinical history, pre-operative coronary angiography and other routine tests. Six patients (3M/3F) aged 60–75 years (69±5.4 years) were found to have CAD whereas 2 patients (1M/1F) aged 51–57 years (54±4.2 years) did not have CAD. The study was conducted in accordance with the Declaration of Helsinki and the protocol was approved by the University of Miami ethical committee (IRB # 20110879).

EAT samples were obtained near the proximal segment of the right coronary artery, deep to the visceral layer of the pericardium prior to the patients being placed on-pump. Each tissue sample was immediately frozen over dry ice and then stored at −80 °C until processing. RNA was extracted from the collected adipose tissues as previously reported [10]. The whole transcriptome sequencing was carried out at the Sequencing Core facility of John P. Hussman Institute of Human Genomics at the University of Miami. Briefly, after ribosomal RNA (rRNA) was depleted, sequencing libraries were constructed following the standard Illumina protocols and were subsequently processed by a HiSeq2500 sequencing system (125 bp paired-end reads). Raw reads were first run through quality control metrics using FastQC (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/). After quality control was checked, sequence reads were aligned to the human transcriptome (GRCh38, Ensembl.org) and quantified using the STAR aligner [11]. To confirm the expression of GLP-1R and GLP-2R, qRT-PCR was performed using 2 different sets of intron-spanning primers per gene. The qScript Flex cDNA kit (Quanta Biosciences, Beverly, MA, USA) was used for reverse transcription according to the manufacturer’s instructions. qRT-PCR was performed in QuantStudio 12 K Flex (Applied Biosystems, Foster City, CA, USA) using the PowerUp SYBR Green Master Mix (Applied Biosystems, Foster City, CA, USA) with 10 μl reactions. Primers were designed to span introns (▶ Table 1). The transcript amplification results were analyzed with the QuantStudio 12k software, and all values were normalized to the levels of the housekeeping gene Succinate Dehydrogenase (SDHA) using the 2^−ΔΔCt method. A set of housekeeping genes was specifically investigated and it was found that SDHA had low variance between samples and was highly expressed. Furthermore, other studies have investigated the stability of many housekeeping genes and found that SDHA is more stable than other commonly used reference genes [12]. Statistical significance of differences in expression levels was assessed by Student’s t-test, at α = 0.05. The amplified PCR products were gel purified for Sanger sequencing using the Zymoclea gel DNA recovery kit (Zymo research, Irvine, CA, USA) and bidirectional capillary sequencing (Eurofins Genomics, Louisville, KY, USA). Immunofluorescence was performed on EAT and SAT samples using Mab3f52 against GLP-1R. This mouse monoclonal antibody has been extensively validated to show its specificity and it is available at the Developmental Studies Hybridoma Bank (DSHB) [13, 14]. Due to the relatively low abundance of GLP-1R the immunofluorescence was performed using a tyramide signal amplification reagent (TSA kit # 26 from Molecular probes) according to the manufacturer instructions. In summary, cryosection of the samples was performed at −35 °C with 12 μm thickness using a Leica CM1850 UV cryostat (Wetzlar, Germany). The sections were fixed for 20 min with 4% PFA, washed with PBS and dried at room temperature overnight. Sections were washed 3 x with PBS and incubated with PBS with 3 % BSA and 0.4 % triton X for 1 h. They were then incubated with Mab3f52 (1:50 dilution) antibody overnight in PBS with 3 % BSA and 0.4 % triton X. Sections were then washed 4 times with PBS and incubated with biotinylated anti mouse IgG for 1 h at room temperature, washed 4 times with PBS, then incubated at room temperature with horseradish peroxidase-streptavidin for another hour, washed 4 times with PBS and then incubated at room temperature with alexa fluor 647 tyramide for 10 min. Sections were further washed 3 times and incubated with DAPI (Thermo Fisher, Waltham, MA, USA) for 30 min. The samples were washed with 3 times PBS and mounted on Mowiol 4–88 (Sigma-Aldrich, St Louis, MO, USA). The tyramide signal can amplify from 10 to 200 times the immunofluorescence signal from low expressed proteins (Molecular probes, Eugene, OR, USA). The fluorescence images were acquired using a Zeiss LSM 710 confocal microscope (Oberkochen, Germany).

▶ Table 1.

Primers used for quantitative RT-PCR.

Gene	Species	Forward (5′→3′)	Reverse (5′→3′)
GLP1R	Human	CAAATGCAGACTTGCCAAGTCCACG	CCAGCTGGACCTCATTGTTGACAAAG
GLP1R	Human	CACCTCCTTCCAGGGGCTGATGG	CTCAGGCTGCTGGTGGGACACTTG
GLP2R	Human	GGGCGTTCATGAGATCCTCTTCTCTTTCATC	CAGCTCAGCCTTCACCTCTCCATTG
GLP2R	Human	GATGCTCTGTGTAACAGTCAATTTCTTC	CCAATAAAGGAATGAGGACCAGTGT
SDHA	Human	GCCAGGGAAGACTACAAGGTGCG	GAATGGCTGGCGGGACGGTG

Results

Our RNA-seq analysis showed that both EAT and SAT express GLP-1R and GLP-2R based on the presence of accurately-spliced reads. qRT-PCR showed that GLP-1R and GLP-2R expression were relatively low but detected by 2 different sets of intron-spanning primers per gene (▶ Fig. 1). GLP-2R expression was found to be ~5-fold higher than GLP-1R in EAT and ~40-fold higher in SAT. Both genes had relatively low expression and no difference between diabetic and non-diabetic EAT or between diabetic or nondiabetic SAT was found. However, analysis of differences between SAT and EAT revealed that GLP-2R was significantly decreased in EAT compared to SAT (p = 0.0065, ▶ Fig. 2). To confirm the specificity of our primers for RNA we performed capillary sequencing on the products of the qRT-PCR reactions. Indeed, the capillary traces showed correctly spliced RNA indicating that the amplicons were generated from only RNA for both GLP-1R and GLP-2R (▶ Fig. 3). Immunofluorescence was performed on EAT and SAT samples using Mab3f52 against GLP-1R. This antibody has been extensively validated to show its specificity [13, 14]. Due to the low abundance of GLP-1R a tyramide signal amplification reagent (TSA kit # 26 from Molecular probes) was used to increase the sensitivity. Staining showed a definite higher signal in the EAT samples than in SAT samples and samples stained with no primary antibody, clearly indicating a more abundant presence of GLP-1R in EAT (▶ Fig. 4).

▶ Fig. 1.

qRT-PCR of GLP-1R and GLP-2R between diabetic and nondiabetic patients: qRT-PCR results of GLP-1R in a EAT and b SAT. qRT-PCR results of GLP-2R in c EAT, and d SAT. Values are relative to nondiabetic control expression.

▶ Fig. 2.

qRT-PCR of GLP-1R and GLP-2R between SAT and EAT: qRT-PCR results of a GLP-1R and b GLP-2R. Values are normalized to SAT expression. * * Indicates p < 0.01.

▶ Fig. 3.

Capillary sequencing of GLP-1R and GLP-2R cDNA: Capillary sequencing of the qRT-PCR amplicons shows correctly spliced RNA of both GLP-1R and GLP-2R. a GLP-1R exon 10-exon 11 splice junction. b GLP-1R exon 11-exon 12 splice junction. c GLP-1R exon 12-exon 13 splice junction. d GLP-2R exon 9-exon 10 splice junction. e GLP-2R exon 10-exon 11 splice junction.

▶ Fig. 4.

Immunofluorescence of GLP-1R in epicardial adipose tissue (EAT) and subcutaneous adipose tissue (SAT): Immunofluorescence was performed on EAT and SAT samples using Mab3f52 against GLP-1R In the upper quadrants a, the immunofluorescence images (starting from the left) show definite higher signal in GLP-1R antibody-treated EAT, irrespective of diabetes, when compared to GLP-1R antibody-treated SAT and to not treated with primary antibody EAT (last image on the right). Lower quadrants b images show the amplification of the GLP-1R signal using biotinylated anti mouse IgG, horseradish peroxidase-streptavidin and alexa fluor 647 tyramide. DAPI was used to label the nucleus and autofluorescence of the tissue was collected at 488 nm excitation.

Discussion

This is the first time that human EAT is found to express both GLP-1R and GLP-2R genes.

The combination of accurately-spliced reads from RNA-seq and successful amplification using intron-spanning primers indicates that both GLP-1R and GLP-2R are expressed in EAT. The immunofluorescence clearly indicates that GLP-1R protein is present and more abundant in EAT than in SAT.

In light of the emerging use and pleiotropic effects of the GLP analogues, we believe these data are novel and with interesting clinical application. Epicardial fat is an easily measurable risk factor [15, 16] that has shown to significantly respond to thiazolidinediones, insulin, dipeptidyl peptidase-4 inhibitors and statins [17-19]. Some of these drugs caused a significant change and reduction in the EAT inflammatory infiltrate [17]. Exenatide, and mostly liraglutide, 2 widely used GLP-1 analogues, have also shown to significantly reduce EAT [6, 7], independently of weight loss and glucose control, arising therefore the question of a possible specific drug-effect. The presence of GLP-1R gene and mRNA expression, for the first time reported here, may further corroborate this hypothesis and suggest the use of EAT as therapeutic target for GLP-1 activation. The lack of difference in the GLP-1R expression between diabetics and nondiabetics may allow for a broader use of GLP-1-analogues targeting the fat.

Excessive EAT accumulation has been largely associated with higher cardiovascular risk [1]. EAT exerts its atherogenic action through the paracrine secretion of pro-inflammatory cytokines, upregulated in this fat depot [2, 20]. The LEADER trial recently showed a lower rate of cardiac events in subjects on long-term treatment with liraglutide [5]. Remarkably, our recent study showed a substantial rapid shrinkage of the ultrasound-measured EAT in patients treated with liraglutide [7]. Given the presence of the GLP-1R, we can now speculate some of the beneficial cardiovascular effects of liraglutide being EAT-mediated, at least partially. We know that GLP-1R is expressed in the adipose tissue and mRNA and protein expression is higher in visceral fat [8, 21]. It has been also suggested that GLP-1 promotes preadipocyte differentiation and improves local insulin resistance [21]. GLP-1 activation may also stimulate browing thermogenesis of EAT [22] and therefore restore its physiological role of providing heat to the myocardium [23].

We recently demonstrated a rapid and large reduction (around 36%) of the ultrasound measured epicardial fat thickness in type 2 diabetic subjects treated with additional liraglutide as compared to subjects on metformin monotherapy [7]. GLP-1 agonistic effect may increase epicardial fat lipolysis and insulin sensitivity, decrease lipogenesis and therefore cause a fat mass reduction.

In this study, we also found that EAT expresses GLP-2R gene. Adipose tissue GLP-2R gene and mRNA expression was never reported before. Glucose and metabolic effects of GLP-2 activation are currently under investigation. GLP-2 is a proglucagon-derived peptide produced by intestinal enteroendocrine L-cells. GLP-2R knockout animals show impaired postprandial glucose tolerance and hepatic insulin resistance [24]. GLP-2 activation has been reported to provide beneficial effects on the glucose homeostasis and insulin sensitivity in experimental obesity [24]. The expression of GLP-2R gene in a fast-responder and measurable adipose fat depot, such as EAT, may be of importance for future applications and studies.

Study limitations

We recognize that these findings should be interpreted with caution, as the sample size was quite small, although statistically significant to detect differences, and GLP-R expression was present, but relatively low. Our observations would certainly warrant future and larger studies to be confirmed. Lipolytic experiments on cultures of EAT adipocytes freshly isolated after sample accrual from cardiac surgery patients would help to better elucidate the mechanisms behind the substantial EAT mass reduction, recently reported by our group [7].

In conclusion, by showing that human EAT expresses both GLP receptors, our study suggests novel pathways and mechanisms of action of GLP analogues. Pharmacologically targeting and potentially modulating organ-specific fat depots may produce beneficial cardiovascular and metabolic effects.