CART is overexpressed in human type 2 diabetic islets and inhibits glucagon secretion and increases insulin secretion
Insufficient insulin release and hyperglucagonaemia are culprits in type 2 diabetes. Cocaine- and amphetamine-regulated transcript (CART, encoded by Cartpt) affects islet hormone secretion and beta cell survival in vitro in rats, and Cart −/− mice have diminished insulin secretion. We aimed to test if CART is differentially regulated in human type 2 diabetic islets and if CART affects insulin and glucagon secretion in vitro in humans and in vivo in mice.
CART expression was assessed in human type 2 diabetic and non-diabetic control pancreases and rodent models of diabetes. Insulin and glucagon secretion was examined in isolated islets and in vivo in mice. Ca2+ oscillation patterns and exocytosis were studied in mouse islets.
We report an important role of CART in human islet function and glucose homeostasis in mice. CART was found to be expressed in human alpha and beta cells and in a subpopulation of mouse beta cells. Notably, CART expression was several fold higher in islets of type 2 diabetic humans and rodents. CART increased insulin secretion in vivo in mice and in human and mouse islets. Furthermore, CART increased beta cell exocytosis, altered the glucose-induced Ca2+ signalling pattern in mouse islets from fast to slow oscillations and improved synchronisation of the oscillations between different islet regions. Finally, CART reduced glucagon secretion in human and mouse islets, as well as in vivo in mice via diminished alpha cell exocytosis.
We conclude that CART is a regulator of glucose homeostasis and could play an important role in the pathophysiology of type 2 diabetes. Based on the ability of CART to increase insulin secretion and reduce glucagon secretion, CART-based agents could be a therapeutic modality in type 2 diabetes.
KeywordsCART peptide Cocaine- and amphetamine-regulated transcript Glucagon Insulin Islets Type 2 diabetes
Cytoplasmic Ca2+ concentration
Acute insulin response
Cocaine- and amphetamine-regulated transcript
Glucose-stimulated insulin secretion
In situ hybridisation
Glucose tolerance index
Insufficient insulin secretion and exaggerated glucagon secretion are key features of type 2 diabetes [1, 2, 3]. The pancreatic islets are vital regulators of glucose homeostasis, and islet dysfunction is a primary cause of type 2 diabetes . Islet hormone secretion is controlled by the metabolic state, e.g. plasma levels of glucose, amino acids and fatty acids. Furthermore, the islets are influenced by regulatory peptides expressed in islet cells or released from nerve fibres innervating the islets .
The anorexigenic regulatory peptide cocaine- and amphetamine-regulated transcript (CART) [6, 7, 8, 9, 10] is expressed in islet cells and islet nerve fibres in animals [11, 12, 13, 14, 15, 16, 17]. CART regulates islet hormone secretion and protects beta cells from glucotoxicity-induced cell death in vitro in rat islets [12, 18]. Importantly, CART is required for normal islet function: Cart −/− mice exhibit islet dysfunction with diminished insulin secretion and reduced glucose elimination . On the other hand, expression of CART is augmented in beta cells in rat models of type 2 diabetes . Moreover, Cart −/− mice are obese [15, 19], and genetic variations in CARTPT associate with obesity in humans [20, 21].
It is not known if CART affects human islet function or if CART is expressed in human islets and is affected by type 2 diabetes. Furthermore, it is not known how CART affects glucose homeostasis in vivo. To address this, we examined islet CART expression in non-diabetic and type 2 diabetic individuals, as well as the effect of CART on islet hormone secretion in human islets. Furthermore, we studied the effect of exogenous CART in mice to pinpoint the mechanisms underlying the effect of CART on insulin secretion.
Human pancreases from nine type 2 diabetic patients (five men, four women; median age 54 years, BMI 28.9 ± 7 kg/m2) and ten controls (four men, six women; median age 56 years, BMI 25.8 ± 2 kg/m2) were used for morphometric analysis. Human islets were provided by the Nordic Network for Clinical Islet Transplantation, Uppsala University. All procedures were approved by Uppsala and Lund University ethics committees.
Rat CART55-102 peptide (American Peptide, Sunnyvale, CA, USA or gift from L. Thim [Novo Nordisk, Målöv, Denmark]) was used.
Immunocytochemistry, in situ hybridisation and image analysis
Immunocytochemistry was performed using characterised antibodies as detailed [13, 22]. The specificity of CART antibodies (Cocalico, Stevens, PA, USA) was verified in Cart −/− mice and with pre-absorption with CART55-102 . For in situ hybridisation (ISH), 30-mer oligodeoxyribonucleotide probes for Cart mRNA were used in human  and rat  specimens as described previously . Hybridisation in the presence of excess unlabelled probe on adjacent sections was used as a negative control. CART-immunoreactive alpha cells and beta cells, and CART ISH-labelled area/whole-islet area were quantified blinded in the indicated samples in all islets in three separate sections as described .
Western blot was performed as described previously  using CART primary antibody (Cocalico) diluted 1:1000 with CART55-102 (Novo Nordisk) as a positive control.
Animal models of type 2 diabetes
Experiments were approved by the regional animal ethics committee and the Guide for the care and use of laboratory animals (8th edn) was followed. Female Sprague Dawley rats (225 g) were used. Six rats received dexamethasone (DEX), 2 mg/kg/day i.p. for 10 days [12, 27]. Six rats received DEX and insulin detemir (Levemir, Novo Nordisk), 10 U/day for 10 days. Controls (n = 6) received NaCl 154 mmol/l. In addition, the genetically modified mouse strains (1) HNF1α dominant-negative (Hnf1α dn) mice  and (2) ob/ob mice with respectively matched control mice, as well as C57Bl/6J mice (Taconic, Bomholt, Denmark) fed a high-fat diet (HFD) or control diet for 6 months were used (all female mice, aged 10–14 weeks, n = 5–8 in each group) . The pancreases were processed as described previously .
Real-time quantitative PCR
Total RNA was isolated using All Prep DNA/RNA kit (Qiagen, Hilden, Germany). RNA, 1 μg, was reverse-transcribed to cDNA with QuantiTect Reverse Transcription kit (Qiagen). Quantitative RT-PCR was performed using the ABI Prism 7900 HT system (Applied Biosystems, Foster City, CA, USA) with TaqMan technology and assay Hs00182861_m1 for Cart and 4333768F for Hprt1, which served as an endogenous control. Gene expression was analysed with the ΔΔCt method.
In vitro islet studies
Secretion studies in human and mouse islets (1 h incubations) with insulin secretagogues, with and without CART55-102 as indicated, were performed as described previously [15, 30, 31]. One experiment represents one donor/animal, with 6–8 technical repeats for each experiment. For Cart mRNA expression studies, rat islets were cultured for 24 h in 5, 11.1 or 25 mmol/l glucose in RPMI 1640 medium (1% serum).
IVGTT was performed as previously described  in female C57Bl/6J mice with or without 150 nmol/kg CART55-102 and 4 nmol/kg glucagon-like peptide 1 (GLP-1) co-injected, i.v, with glucose, 0.75 g/kg. Controls received NaCl. Acute insulin response (AIR) was calculated as mean suprabasal insulin level of insulin at mins 1 and 5 (i.e. [ins5 min + ins1 min]/2 – ins0 min). The glucose elimination rate (KG) was calculated as the slope for 1-20 min of the logarithmic transformation of individual plasma glucose values .
Glucagon secretion in vivo
Female C57Bl/6J mice were fasted overnight and given i.v. CART55-102, 150 nmol/kg. Blood samples were taken from the retro-orbital plexus after 2 min. Controls received NaCl.
Depolarisation-evoked exocytosis was recorded by capacitance measurements using standard whole-cell configuration with or without CART55-102 as indicated . Islets from female C57Bl/6J mice were used. Exocytosis was induced by 500 ms depolarisation from −70 mV to 0 mV.
Ca2+ signalling and analysis of cell synchronisation
Changes in the cytoplasmic Ca2+ concentration ([Ca2+]i) were recorded as previously described . In brief, mouse islets or islet cells were loaded with Fura-2 LR acetoxymethyl ester (TEFLabs, Austin, TX, USA) and transferred to poly-l-lysine-coated coverslips in a 50 μl chamber on the stage of an Eclipse TE2000U microscope (Nikon, Tokyo, Japan) equipped for ratiometric epifluorescence recordings of [Ca2+]i . The chamber was superfused at 160 μl/min with 37°C buffer containing 138 mmol/l NaCl, 4.8 mmol/l KCl, 1.2 mmol/l MgCl2, 1.3 mmol/l CaCl2, 3 mmol/l glucose and 25 mmol/l HEPES with pH set to 7.40, and 1 mg/ml BSA. [Ca2+]i is expressed as the background-corrected Fura-2 LR 340/380 nm fluorescence excitation ratio.
To estimate the synchronisation of [Ca2+]i signals between different non-overlapping regions-of-interest (ROI) in the islet, we calculated the linear correlation (Pearson’s r) between all possible ROI pairs and displayed the correlation coefficients in a matrix. The correlation matrices were subsequently used to construct undirected graphs representing functional connectivity. These graphs show the geometric location of each ROI with lines added between the ROI pairs with significant correlation coefficients (p < 0.001) equal to or exceeding a threshold value of 0.8. Correlation analyses and the construction of correlation matrices and connectivity maps were performed with standard tools using Igor Pro software version 6.37 (Wavemetrics, Lake Oswego, OR, USA).
Glucose insulin and glucagon measurements
Glucose was measured using Infinity Glucose (Ox) (Thermo Scientific, Lexington, MA, USA) . Insulin and glucagon levels were determined with ELISA (Mercodia, Uppsala, Sweden).
Statistics and overall design for experimental studies
Human insulin secretion data were analysed using a Wilcoxon signed-rank test. All other data were analysed using one-way ANOVA, followed by Bonferroni’s test, or unpaired Student’s t test. Data are expressed as means (SEM). All experiments were performed and analysed in a randomised and blinded fashion when possible. Outliers were identified using Grubbs’ test for outliers.
CART is expressed in human beta cells and alpha cells
Islet CART expression is increased in type 2 diabetic patients and mouse models of diabetes
CART mRNA expression was threefold higher in islets from type 2 diabetic patients (n = 9; p < 0.009) (Fig. 1b, d) and type 2 diabetic patients had threefold more CART-positive beta cells (p < 0.008; Fig. 1a, e) and twofold more CART-positive alpha cells (p < 0.007) (Fig 1a, f) than controls (n = 10). CART-positive beta cells were also several fold more abundant in three mouse models of diabetes (Fig. 1h-j), whereas CART was absent in mouse alpha cells (ESM Fig. 2).
Beta cell CART is regulated by glucose in vivo and in vitro in rats
CART stimulates insulin secretion and inhibits glucagon secretion in human islets
CART stimulates insulin secretion, reduces glucagon secretion and improves glucose elimination in vivo in mice
To examine the effect of exogenous CART on whole-body glucose homeostasis, CART (150 nmol/kg) was given together with glucose (0.75 g/kg) during IVGTT in mice (n = 16). CART provoked a twofold increase in first-phase insulin secretion (AIR 783 ± 99 vs 365 ± 26 pmol/l in controls; p < 0.001; Fig. 3d, e) and improved KG (CART treated: 4.0 ± 0.2%/min; controls: 3.2 ± 0.2%/min; p < 0.05; Fig. 3f, g). Next, we tested the effect of CART on GLP-1-stimulated GSIS. CART alone provoked a similar response in insulin secretion to that of GLP-1 (AIR 2140 ± 382 vs 2349 ± 365 pmol/l; p = 0.28). Notably, CART caused a further 30% increase in insulin secretion above that achieved by GLP-1 alone (AIR 2871 ± 522 vs 2349 ± 365 pmol/l, p < 0.02) (n = 16; Fig. 3h). To test if CART affects glucagon secretion, fasted mice were given i.v. CART, 150 nmol/kg. This provoked a 40% reduction in glucagon secretion after 2 min (n = 8; p < 0.001) compared with controls (n = 7; Fig. 3i). In summary, CART increased both GSIS and GLP-1-stimulated GSIS, reduced glucagon secretion and enhanced glucose elimination in mice.
CART increases insulin and inhibits glucagon secretion in vitro in mice
CART enhances exocytosis of insulin and reduces exocytosis of glucagon in vitro in mice
To assess the effect of CART on exocytosis at the single cell level in beta and alpha cells, we used patch clamp in intact mouse islets. In control beta cells, the capacitance increase (ΔCtot) induced by a train of depolarisations amounted to 139 ± 23 fF. Beta cells treated with CART (10 nmol/l, 1 h) showed a slight, but non-significant, increase in exocytosis (Fig. 4f-i). When pre-treated for 24 h, insulin exocytosis increased twofold compared with controls (287 ± 102 fF, p < 0.05) (Fig. 4f-i). Control alpha cells responded with a capacitance increase of 338 ± 98 fF. Already after 1 h treatment, 10 nmol/l CART robustly reduced glucagon exocytotic capacity by 62% to 130 ± 22 fF (p < 0.05) (Fig. 4j-l).
CART alters the [Ca2+]i signalling pattern in intact mouse islets
Agents that stimulate insulin secretion and reduce glucagon secretion are attractive in the search for new treatments for type 2 diabetes. Here, we provide evidence for CART being a novel human islet peptide, which holds promise for use as diabetes therapy. This is based on our data showing that CART inhibits glucagon secretion and stimulates insulin secretion, even on top of GLP-1.
Importantly, CART increased insulin secretion in islets from type 2 diabetic patients, a finding that gains support from previous observations in GK rat islets . Furthermore, similar to incretins, the stimulatory effect of CART on insulin secretion was glucose dependent. Notably, and in agreement with our previous in vitro data , CART augmented GLP-1-stimulated GSIS even further. This, together with the fact that Cart −/− mice have diminished insulin secretion, suggests that CART is a physiologically important regulator of insulin secretion. Acute CART stimulation was not associated with alterations of beta cell [Ca2+]i level. However, CART promoted a change in the pattern of [Ca2+]i oscillations, with fast oscillations being replaced by slow ones, known to underlie the pulsatile release of insulin [33, 36]. Furthermore, the change in [Ca2+]i oscillation pattern was paralleled by improved synchronisation of the [Ca2+]i signal between different islet subregions. Synchronisation of beta cells is critical for normal insulin secretion kinetics  but it is not known how it is modulated by CART. The exact mechanism underlying CART amplification of insulin secretion remains to be established. We suggest that the short-term stimulatory action of CART on glucose-induced secretion may be related to its effect on the islet [Ca2+]i signalling pattern. This would explain why CART lacks effect when glucose is combined with high K+ or carbachol, known to induce pronounced, non-oscillatory islet [Ca2+]i increases [37, 38].
A factor complicating the mechanistic studies is that the CART receptor remains unknown [7, 39]. The majority of the effects of CART on islets were rapid, and evident within 1 h, however 24 h stimulation with CART was needed to increase exocytosis from single voltage-clamped beta cells. Whether this effect was secondary to transcriptional effects of CART, as reported previously , is not known.
We also showed that CART is an inhibitor of glucagon secretion in vivo in mice, in human islets and in mouse islets via a direct effect on the alpha cell. Of note, insulin , somatostatin  and GLP-1  are the only hormones reported so far to have such a direct effect. In human islets the glucagon-lowering effect of CART was glucose dependent and only evident at stimulatory glucose levels. Together with the insulinotropic actions and protective effects against beta cell death , this positions CART as a plausible target for diabetes therapy.
Furthermore, CART was found to be a novel constituent of human beta and alpha cells. It is of notice that there are large species differences in islet CART expression. Thus, in rats CART is mainly expressed in delta cells [11, 12], whereas in mice CART is expressed in a small subpopulation of beta cells. Pigs have no CART expression in islet cells . Notably, individuals with type 2 diabetes, as well as four different rodent models of diabetes, had increased islet CART expression. Our data suggest that beta cell CART is upregulated as a response to hyperglycaemia as the upregulation of CART in rat models of type 2 diabetes was prevented by glucose-normalising insulin therapy and Cart mRNA expression was enhanced in rat islets cultured at high glucose. Our data also show that islet CART expression is dynamic, which is supported by the fact that CART is transiently upregulated in rodent islets during development .
In view of the insulinotropic and glucagon-lowering effects of CART, the upregulation of CART in type 2 diabetic islets is most likely a homeostatic compensatory response trying to overcome hyperglycaemia via paracrine actions. Even though CART has been localised to beta and delta cell secretory granules , it is not known if islet CART contributes to circulating levels or if these are altered by type 2 diabetes.
The present study expands previous knowledge obtained in vitro in rodents and shows that CART affects islet hormone secretion in human islets and in vivo in mice. We conclude that CART is an important regulator of glucose homeostasis and, based on its properties of stimulating insulin secretion and inhibiting glucagon secretion, CART is a peptide that should be explored as a potential future therapy for diabetes.
We thank D. Persson, B.-M. Nilsson, L. Bengtsson, L. Kvist, A.-H. Thorén-Fischer, U. Voss, L. Wallgren, and A. Östlund at Lund University, Sweden for technical assistance.
This work was supported by grants from the Swedish Research Council (Project grants: 521-2012-2119, 522-2008-4216, K2009-55X 21111-01-4 (NW), 521-2010-3490 (LG), 325-2012-6778 (AT), 6834 (BA), Linneaeus grant to Lund University Diabetes Centre 349-2006-237, Strategic Research Area grant EXODIAB 2009-1039), from Novo Nordisk Foundation, European Foundation For the Study of Diabetes/Merck Sharp & Dohme, Royal Physiographic Society in Lund, Faculty of Medicine at Lund University, Region Skåne, Crafoord Foundation, Gyllenstiernska Krapperup, Tore Nilsson, Åke Wiberg, Lars Hierta, Fredrik and Ingrid Thuring, Magnus Bergwall, Albert Påhlsson, Knut & Alice Wallenberg foundation (KAW 2009.0243), Swedish Diabetes Foundation, Family Ernfors Fund, Heart and Lung Foundation and Åhlén Foundation. MJK: RR00165, DA15162, and DA15040. LG: ERC Advanced Researcher grant GENETARGET-T2D (GA-269045), Sigrid Juselius Foundation, Folkhälsan Research, Finnish Diabetes Research Foundation.
Duality of interest
The authors declare that there is no duality of interest associated with this manuscript.
Each author: (1) contributed substantially to the conception and design, acquisition of data and/or analysis and interpretation of data; (2) drafted the article and/or revised it critically for important intellectual content; and (3) gave final approval of the version to be published. NW is guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
- 20.del Giudice EM, Santoro N, Cirillo G, D’Urso L, Di Toro R, Perrone L (2001) Mutational screening of the CART gene in obese children: identifying a mutation (Leu34Phe) associated with reduced resting energy expenditure and cosegregating with obesity phenotype in a large family. Diabetes 50:2157–2160CrossRefPubMedGoogle Scholar
- 28.Hagenfeldt-Johansson KA, Herrera PL, Wang H, Gjinovci A, Ishihara H, Wollheim CB (2001) Beta-cell-targeted expression of a dominant-negative hepatocyte nuclear factor-1 alpha induces a maturity-onset diabetes of the young (MODY)3-like phenotype in transgenic mice. Endocrinology 142:5311–5320PubMedGoogle Scholar