Deletion of the RabGAP TBC1D1 Leads to Enhanced Insulin Secretion and Fatty Acid Oxidation in Islets From Male Mice

Deletion of the RabGAP TBC1D1 Leads to Enhanced Insulin Secretion and Fatty Acid Oxidation in... Abstract The Rab guanosine triphosphatase–activating protein (RabGAP) TBC1D1 has been shown to be a key regulator of glucose and lipid metabolism in skeletal muscle. Its function in pancreatic islets, however, is not yet fully understood. Here, we aimed to clarify the specific impact of TBC1D1 on insulin secretion and substrate use in pancreatic islets. We analyzed the dynamics of glucose-stimulated insulin secretion (GSIS) and lipid metabolism in isolated islets from Tbc1d1-deficient (D1KO) mice. To further investigate the underlying cellular mechanisms, we conducted pharmacological studies in these islets. In addition, we determined morphology and number of both pancreatic islets and insulin vesicles in β-cells using light and transmission electron microscopy. Isolated pancreatic islets from D1KO mice exhibited substantially increased GSIS compared with wild-type (WT) controls. This was attributed to both enhanced first and second phase of insulin secretion, and this enhanced secretion persisted during repetitive glucose stimuli. Studies with sulfonylureas or KCl in isolated islets demonstrated that TBC1D1 exerts its function via a signaling pathway at the level of membrane depolarization. In line, ultrastructural analysis of isolated pancreatic islets revealed both higher insulin-granule density and number of docked granules in β-cells from D1KO mice compared with WT controls. Like in skeletal muscle, lipid use in isolated islets was enhanced upon D1KO, presumably as a result of a higher mitochondrial fission rate and/or higher mitochondrial activity. Our results clearly demonstrate a dual role of TBC1D1 in controlling substrate metabolism of the pancreatic islet. The two closely related Rab guanosine triphosphatase–activating proteins (RabGAPs), TBC1D1 and TBC1D4, represent essential signaling factors in the regulation of lipid and glucose metabolism (1–3). With the exertion of their respective function via the activation of small Rab guanosine triphosphatases, both molecules control vesicle trafficking processes within a diversity of cell types (4). However, the majority of studies, to date, concentrate on RabGAP function in skeletal muscle and white adipose tissue, where they mediate the translocation of the insulin- and contraction-regulated glucose transporter (GLUT) type 4 to the plasma membrane, thus directly influencing glucose influx (2, 5). Deficiency of one or two of the TBC1D proteins leads to severely impaired insulin- and contraction-stimulated glucose uptake into skeletal muscle and adipocytes, likely caused by missorting and subsequent depletion of the GLUT4 protein (2). Interestingly, lipid use is enhanced in RabGAP-deficient mice, clearly emphasizing the central role of the two RabGAPs in the regulation of substrate preference and metabolic flexibility, a major trait in the pathophysiology of insulin resistance and type 2 diabetes (6). In humans, a coding variant of TBC1D1 was linked to extreme obesity, and a mutation in TBC1D4 was shown to cause postprandial hyperglycemia (7–10). Presumably a result of compensatory mechanisms caused by the remaining RabGAP, deletion of both RabGAPs is necessary to considerably impair whole-body glycemia in mice, and single-knockout animals for each TBC1D protein show only mild disturbances in their glycemic control (2, 11). In addition, contradictory findings for plasma insulin levels in vivo have been demonstrated throughout different studies (2, 12, 13). There is recent evidence that the metabolic impact of the two RabGAPs is not limited to peripheral insulin-responsive tissues alone. So far, only little research has been conducted on TBC1D1 or TBC1D4 function in pancreatic islets. Despite the fact that both RabGAPs are expressed in this tissue and seem to influence insulin secretion (IS), their exact role in the regulation of glucose and lipid metabolism in β-cells is unclear. A previous publication showed that knockdown of Tbc1d4 in sorted mouse β-cells led to increased basal IS but reduced glucose-stimulated insulin secretion (GSIS) (14). In addition, decreased Tbc1d4 expression resulted in enhanced apoptosis and reduced proliferation in these cells (14). Two independent groups showed a mild phenotype regarding IS upon Tbc1d1 knockdown in sorted rat β-cells and isolated pancreatic islets of Tbc1d1-knockout (D1KO) rats, respectively (15, 16). The dynamics of GSIS follow a first and a second phase (17, 18). The triggering pathway triggers the stimulated secretion but does not determine its extent (19). The latter depends on the amplifying pathway, which involves the export of mitochondrial metabolites (20). Additionally, extracellular signals, such as incretins or free fatty acids, play a role in the movement and depletion (21) of insulin-containing granules, thereby relying on a precise regulation of vesicular transport via Rab proteins and RabGAPs (4). However, not much research has been performed, to date, addressing the underlying cellular mechanisms linking RabGAPs to basic islet physiology. The questions remain whether the two RabGAPs control substrate use in pancreatic β-cells in analogy to peripheral tissues and how TBC1D proteins exert their function on β-cell-specific processes. Our findings clearly demonstrate an important role for TBC1D1 in the regulation of IS, as well as lipid metabolism, in islets and suggest a contribution in whole-body glycemia. Materials and Methods Chemicals and buffer Chemicals and buffer ingredients are listed in Supplemental Table 1. Experimental animals Mice were kept in accordance with the US National Institutes of Health guidelines for the care and use of laboratory animals, and all experiments were approved by the Ethics Committee of the State Ministry of Agriculture, Nutrition and Forestry (State of North Rhine-Westphalia, Germany). Three to six mice per cage were housed at 22°C and a 12-hour light: 12-hour dark cycle with ad libitum access to food and water. After weaning, animals received a standard diet (Ssniff, Soest, Germany). Male mice were used at the age of 12 to 16 weeks. The generation of D1KO mice has been described previously (2, 6). To generate transgenic TG-RIP2-3xFLAG-Tbc1d1 (RIP2-D1) mice overexpressing Tbc1d1 under the control of the RIP2 promoter, a DNA fragment for RIP2-3xFLAG-Tbc1d1 was microinjected into the male nucleus of zygotes from an F1(C57BL/6JOlaHsd × SJL/JCrHsd) intercross (22). Embryos were transferred to the oviduct of pseudopregnant females, and hemizygous transgenic offspring were identified by Southern blotting. Several lines were backcrossed to the N5 generation to C57BL/6JCrl mice using marker-assisted genotyping until reaching 100% C57BL/6J background, based on the analysis of 108 microsatellite markers. Cloning of the RIP2-3xFLAG-Tbc1d1 gene construct We replaced the KasI/PmeI Cre-ER cassette in the plasmid pBKS-RIP2-Cre-ER (Addgene) (23) by blunt-end cloning of a 3xFLAG-Tbc1d1 (short isoform) (6), and the resulting construct was verified by Sanger sequencing. A 5.6-kbp FseI/PmeI fragment for RIP2-3xFLAG-Tbc1d1-polyA was gel isolated, purified, and used for oocyte injections. Control mice are wild-type (WT) littermates, not carrying the RIP2-3xFLAG-Tbc1d1 transgene. Intraperitoneal glucose tolerance test Mice were fasted for 6 hours. Basal blood glucose was determined at the tail tip, and additional blood was collected for plasma insulin measurements. Mice were injected intraperitoneally with 2 g/kg body weight glucose, and blood glucose and plasma insulin were measured at 15, 30, 60, and 120 minutes after injection. Plasma insulin was quantified with the Insulin (Mouse) Ultrasensitive ELISA kit (DRG, Marburg, Germany). Islet isolation and static GSIS Pancreatic islets were isolated by ductal collagenase perfusion of the pancreas, as previously described (24). Islets were allowed to regenerate overnight in CMRL (Connaught Medical Research Laboratories) islet medium. All incubation steps were conducted at 37°C with 5% CO2. GSIS assays were performed with eight islets per well in a 96-well plate. Islets were adjusted in Krebs-Ringer-HEPES (KRH) buffer with 2 mM glucose for 1 hour before 1-hour incubation with 2 mM glucose as basal condition. Thereafter, the same islets were incubated for 1 hour with KRH buffer containing the designated secretagogue (25 mM glucose, 1 µM glibenclamide, 30 mM KCl, or 5 µM A23187) and were lysed after the last stimulation. Insulin in both the supernatants and the lysates was measured using mouse insulin enzyme-linked immunosorbent assay (ELISA) kits. Total pancreatic insulin/proinsulin content Dissected pancreas was snap frozen in liquid nitrogen. The pancreas was homogenized with ice-cold acid ethanol (0.18 M HCl in 75% ethanol) and kept shaking overnight at 4°C. The homogenate was centrifuged for 15 minutes at 5000 relative centrifugal force at 4°C, and the clear supernatant was subjected to insulin and proinsulin ELISA measurements and bicinchoninic acid protein assay. Islet isolation and dynamic perifusion Mice were euthanized, and the pancreas was removed and manually chopped in KRH buffer with scissors. After 8.5 minutes of collagenase digestion in a 37°C shaking waterbath, the tissue was washed with KRH buffer, and islets were handpicked from the exocrine tissue and assayed within 1 hour. Perifusion of 50 WT and D1KO islets, respectively, was carried out according to the previously described protocols (25). Palmitate uptake and oxidation Assays were done essentially as described (6). In brief, for palmitate uptake, 100 islets were incubated in KRH containing 0.1% fatty acid-free bovine serum albumin (BSA) and 2.8 mM glucose for 2 hours, washed three times with KRH + 0.1% BSA, and incubated with KRH + 40 mM BSA and 0.5 mCi/mL [1-14C]palmitic acid at 37°C (20 minutes for uptake assays). The cells were then washed with ice-cold KRH + 0.1% BSA and lysed with sodium dodecyl sulfate, and radioactivity was determined by scintillation counting. For palmitate oxidation, 10 islets were seeded in separate 48 wells in CMRL islet medium and cultured overnight at 37°C/ 5% CO2. The plate was placed into a custom-made oxidation chamber; islet medium was supplemented with [14C]palmitic acid (0.3 µCi per well), fatty acid-free BSA (6.24 µM per well), and l-Carnitin (1 µM per well); and the procedure continued as previously described (26). Mitochondrial copy number and citrate synthase activity The ratio of islet mitochondrial DNA (mtDNA; mt-Nd2) and nuclear DNA (Rps18), as well as citrate synthase activity, was determined, as previously described (5). Measurement of oxygen consumption rate To determine islet oxygen consumption rate (OCR), isolated islets were collected after overnight regeneration and resuspended in Seahorse XF Base medium (Agilent, Wilmington, DE), supplemented with 1% fetal bovine serum, 3 mM glucose, and 0.5 mM HEPES (pH 7.4), and seeded with 70 islets per well of a 24-well islet capture microplate, as described elsewhere (27). OCRs were measured at baseline and following injection with 5 µM oligomycin and 5 µM antimycin A/rotenone (XF Cell Mito Stress Test Kit), respectively, using a Seahorse XF24e analyzer (Agilent). The OCRs were normalized to the baseline values, as previously described (27), and expressed as percent of WT baseline. Morphometric analysis of islets Dissected pancreas was immediately fixed with 4% paraformaldehyde for 24 hours and embedded in paraffin after dehydration, according to standard procedures. Sections of 5 µm were prepared on microscope slides and stained with hematoxylin/eosin. Total pancreas area was determined by point counting (28, 29). Finally, 14 to 21 sections of different levels per pancreas from three animals per genotype were analyzed with cellSense Dimensions Software (Olympus, Hamburg, Germany). Electron microscopy of islets Isolated islets from three WT and D1KO mice, respectively, were pooled and incubated with 2 or 25 mM glucose in KRH buffer for 1 hour at 37°C after overnight regeneration. Islets were washed with BSA-free KRH buffer and processed, as described previously (30). Mature granules were determined by grayscale threshold analysis and counted with cellSense Dimensions Software. Immature granules, as well as docked granules, were identified and counted manually. Finally, 10 β-cells of three to five different islets per condition were analyzed. Mitochondrial area was determined with cellSense Dimensions Software after manual identification and labeling. Complementary DNA synthesis, reverse transcription polymerase chain reaction, and quantitative polymerase chain reaction RNA was isolated using RNeasy Mini Kit, and complementary DNA (cDNA) was synthesized using the GoScript™ Reverse Transcription Kit (Promega) with 500 ng RNA for quantitative polymerase chain reaction (qPCR) and 1 µg RNA for reverse transcription PCR (RT-PCR) with random hexanucleotide primers (Roche). For qPCR, several primers were used with the GoTaq® qPCR Master Mix on a StepOne Plus device (Applied Biosystems). RT-PCR for Tbc1d1 isoform detection was realized with primers flanking exons 12 and 13 of Tbc1d1. The forward primer is located in exon 11; the reverse primer is located in exon 15. Primer sequences are shown in Supplemental Tables 2 and 3. Analysis was performed using the 2-ΔΔCT) method (31), with 36b4 as a reference gene. For the determination of the relative copy number of messenger RNA (mRNA) in freshly isolated pancreatic islets, ΔCt values for Tbc1d1 and Tbc1d4 genes were normalized using a calibration curve obtained from the amplification of plasmids containing the respective cDNA sequences (pcDNA3-CMV-3xFLAG-Tbc1d1, 9319 bp; pCR2.1-TOPO-Tbc1d4, 7916 bp; see Supplemental Fig. 1). Western blot analysis We used standard protocols to prepare total protein extracts from isolated islets and frozen tissues. Total islet protein (10 to 30 µg) was separated in an 8% to 12% sodium dodecyl sulfate–polyacrylamide gel electrophoresis, and tank blotting of the proteins onto polyvinylidene difluoride membrane was performed. Membranes were blocked and incubated with primary antibodies and secondary horseradish peroxidase–conjugated antibodies (Supplemental Table 4). Proteins were detected with enhanced chemiluminescence reagent (PerkinElmer) in a ChemiDoc device (Bio-Rad, Munich, Germany) and quantified with the ImageLab software (Bio-Rad). Statistical analysis All experiments were performed with at least n = 3 samples and shown as mean values ± standard error of the mean (SEM). Statistical significance was calculated with appropriate tests using GraphPad Prism 5 software (GraphPad, La Jolla, CA). Exact conditions and tests are depicted in the figure legends. Results TBC1D1 is predominant over TBC1D4 in mouse islets Pancreatic islets from mice express both RabGAPs: Tbc1d1 and Tbc1d4. Real-time qPCR indicates that Tbc1d1 transcripts are predominant over Tbc1d4 in isolated pancreatic islets from mice (Fig. 1A and Supplemental Fig. 1A). As illustrated in Fig. 1B, islets exclusively contain mRNA for the short variant of Tbc1d1 [1162 amino acid (aa)], whereas skeletal muscle and the heart also contain the long Tbc1d1 isoform (1255 aa). As expected, TBC1D1 protein was undetectable in islets from D1KO mice (Fig. 1C and 1D). We generated transgenic mice overexpressing the short isoform of Tbc1d1 under the control of the RIP2 promoter (RIP2-D1). Islets from RIP2-D1 mice showed a 2.6-fold increase in TBC1D1 protein abundance compared with islets from RIP2-WT littermates (Fig. 1C and 1D). However, TBC1D4 protein was equally abundant in islets from D1KO and RIP2-D1 mice compared with their respective controls (Fig. 1C and 1D). Figure 1. View large Download slide RabGAP expression in isolated islets of D1KO and RIP2-D1 mice. (A) Normalized mRNA expression of Tbc1d1 and Tbc1d4 in freshly isolated C57BL/6J mouse islets. ΔCt values (with 36b4 as a housekeeping gene) were measured by real-time qPCR, corrected and normalized using standard concentration curves (Supplemental Fig. 1A). Data are means ± SEM (n = 8). Student t test, two-tailed, unpaired, ***P < 0.001. (B) RT-PCR of Tbc1d1 isoforms in heart (HE), quadriceps muscle (SM), and pancreatic islets (PI), as well as negative control (N). Positions of isoform-detecting primers are indicated by arrows. PCR of the long isoform produces a 645-bp fragment; the short isoform product corresponds to a 366-bp fragment as a result of the lack of exons 12 and 13. (C and D) Abundance of TBC1D1 and TBC1D4 in islets of D1KO, RIP2-D1 mice, and respective WT littermates. RabGAPs were normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Data represent means ± SEM (n = 4 to 8), Mann-Whitney U test, two-tailed, *P < 0.05; ***P < 0.001. Figure 1. View large Download slide RabGAP expression in isolated islets of D1KO and RIP2-D1 mice. (A) Normalized mRNA expression of Tbc1d1 and Tbc1d4 in freshly isolated C57BL/6J mouse islets. ΔCt values (with 36b4 as a housekeeping gene) were measured by real-time qPCR, corrected and normalized using standard concentration curves (Supplemental Fig. 1A). Data are means ± SEM (n = 8). Student t test, two-tailed, unpaired, ***P < 0.001. (B) RT-PCR of Tbc1d1 isoforms in heart (HE), quadriceps muscle (SM), and pancreatic islets (PI), as well as negative control (N). Positions of isoform-detecting primers are indicated by arrows. PCR of the long isoform produces a 645-bp fragment; the short isoform product corresponds to a 366-bp fragment as a result of the lack of exons 12 and 13. (C and D) Abundance of TBC1D1 and TBC1D4 in islets of D1KO, RIP2-D1 mice, and respective WT littermates. RabGAPs were normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Data represent means ± SEM (n = 4 to 8), Mann-Whitney U test, two-tailed, *P < 0.05; ***P < 0.001. D1KO affects glucose-induced plasma insulin levels in vivo To determine the impact of D1KO on plasma insulin concentrations in vivo, blood glucose and plasma insulin levels were determined acutely after an intraperitoneal injection of glucose. In line with previous studies (2, 12, 13), we did not observe any differences in blood glucose between D1KO and WT mice (Fig. 2A and 2D). Compared with the 6-hour fasting plasma insulin levels, D1KO exhibited an elevated increment in plasma insulin upon an acute glucose stimulus (Fig. 2B and 2E). However, absolute plasma insulin levels were similar between D1KO mice and WT mice during the test period (Fig. 2C). Notably, D1KO mice had significantly reduced plasma insulin levels after 6 hours of fasting, indicating increased insulin sensitivity in peripheral tissues in the D1KO mice. Figure 2. View largeDownload slide Intraperitoneal glucose tolerance test in D1KO mice compared with WT mice. Mice were fasted for 6 hours and injected 2 mg/kg body weight glucose intraperitoneally. (A) Glucose tolerance as a measure of glucose clearance from peripheral tail blood within 120 minutes after glucose injection. (B) Plasma insulin levels expressed as the percentage of the basal values of D1KO and WT mice, respectively. (C) Absolute plasma insulin levels expressed as nanograms per milliliter of D1KO and WT mice. (D) Area under curve (AUC) of the blood glucose levels as shown in (A). (E) AUC of the normalized plasma insulin levels as shown in (B). Data are means ± SEM (n = 12 to 14). Student t test, two-tailed, unpaired, *P < 0.05; **P < 0.01. Figure 2. View largeDownload slide Intraperitoneal glucose tolerance test in D1KO mice compared with WT mice. Mice were fasted for 6 hours and injected 2 mg/kg body weight glucose intraperitoneally. (A) Glucose tolerance as a measure of glucose clearance from peripheral tail blood within 120 minutes after glucose injection. (B) Plasma insulin levels expressed as the percentage of the basal values of D1KO and WT mice, respectively. (C) Absolute plasma insulin levels expressed as nanograms per milliliter of D1KO and WT mice. (D) Area under curve (AUC) of the blood glucose levels as shown in (A). (E) AUC of the normalized plasma insulin levels as shown in (B). Data are means ± SEM (n = 12 to 14). Student t test, two-tailed, unpaired, *P < 0.05; **P < 0.01. D1KO increases GSIS in isolated islets We investigated GSIS from isolated islets in both static and dynamic approaches. Upon stimulation with 25 mM glucose, islets from D1KO mice exhibited a substantially increased GSIS (≈1.5-fold) compared with islets from WT controls, whereas IS at basal glucose concentration (2 mM) was unchanged (Fig. 3A). Islets from RIP2-D1 mice showed no genotype-dependent differences IS (Fig. 3A). Measurements of insulin content in the islet lysates after GSIS revealed no genotype-specific differences (Supplemental Fig. 2A). The increased GSIS in D1KO islets was even sustained after three repetitive glucose challenges (Fig. 3B). IS was additionally determined by perifusion of freshly isolated islets. After 60 minutes at substimulatory glucose concentrations (5 mM), glucose was raised to 30 mM. When normalized to the last prestimulatory secretion rate (60 minutes), the increase in response to 30 mM glucose of the D1KO islets was markedly pronounced compared with WT islets, although the difference failed to achieve statistical significance (Fig. 3C). The ratio of total pancreatic proinsulin/insulin was significantly decreased in D1KO mice compared with WT controls (Fig. 3D), which is a result of similar insulin (Fig. 3E) but decreased proinsulin content (Fig. 3F) in pancreas from random-fed D1KO mice. The gene-expression levels of prohormone convertases 1 and 2 (Pcsk1 and Pcsk2) were significantly increased in D1KO islets (Fig. 3G). Figure 3. View largeDownload slide GSIS and pancreatic insulin and proinsulin content. (A) Isolated islets of D1KO, RIP2-D1, and respective WT mice were incubated with 2 and 25 mM glucose for 1 hour at 37°C, and secreted insulin in the supernatants and islet lysates were measured with ELISA. Data are means ± SEM (n = 7), two-way analysis of variance (ANOVA) with Bonferroni post hoc test, *P < 0.05. (B) Isolated islets of D1KO mice were incubated with 2 and 25 mM glucose for 1 hour at 37°C in three subsequent repetitions. Secreted insulin in the supernatants from the first basal and the last stimulated samples [25 mM (3)] and islet lysates were measured with ELISA. Data are means ± SEM (n = 5), two-way ANOVA with Bonferroni post hoc test, *P < 0.05. (C) Fifty freshly isolated islets per genotype were perifused in a custom-made perifusion chamber with 5 mM glucose and 30 mM glucose for the designated time points. Insulin from the eluate fractions was measured with ELISA, and the values are expressed as percentage of the secretion rate at the end of the basal glucose period (WT = 8.66 ± 3.80 pg ⋅ minutes−1 ⋅ islet−1; D1KO = 5.78 ± 2.52 pg ⋅ minutes−1 ⋅ islet−1). Data are means ± SEM (n = 5). Mann-Whitney U test, two-tailed of the area under the curve: P = 0.151. (D–F) Pancreata from D1KO and WT mice were homogenized in acid ethanol and insulin, and proinsulin was measured with ELISA. (D) Proinsulin-to-insulin ratio, as calculated by the (E) total insulin and (F) proinsulin content, normalized to protein content. Data are means ± SEM (n = 6). Welch-corrected t test, two-tailed, unpaired, *P < 0.05. (G) mRNA expression of insulin-processing prohormone convertases 1 and 2 (Pcsk1 and Pcsk2) from isolated islets of D1KO mice and WT controls. Data are means ± SEM (n = 4 to 7). Welch-corrected Student t test, two-tailed, unpaired, *P < 0.05. n.s., not significant. Figure 3. View largeDownload slide GSIS and pancreatic insulin and proinsulin content. (A) Isolated islets of D1KO, RIP2-D1, and respective WT mice were incubated with 2 and 25 mM glucose for 1 hour at 37°C, and secreted insulin in the supernatants and islet lysates were measured with ELISA. Data are means ± SEM (n = 7), two-way analysis of variance (ANOVA) with Bonferroni post hoc test, *P < 0.05. (B) Isolated islets of D1KO mice were incubated with 2 and 25 mM glucose for 1 hour at 37°C in three subsequent repetitions. Secreted insulin in the supernatants from the first basal and the last stimulated samples [25 mM (3)] and islet lysates were measured with ELISA. Data are means ± SEM (n = 5), two-way ANOVA with Bonferroni post hoc test, *P < 0.05. (C) Fifty freshly isolated islets per genotype were perifused in a custom-made perifusion chamber with 5 mM glucose and 30 mM glucose for the designated time points. Insulin from the eluate fractions was measured with ELISA, and the values are expressed as percentage of the secretion rate at the end of the basal glucose period (WT = 8.66 ± 3.80 pg ⋅ minutes−1 ⋅ islet−1; D1KO = 5.78 ± 2.52 pg ⋅ minutes−1 ⋅ islet−1). Data are means ± SEM (n = 5). Mann-Whitney U test, two-tailed of the area under the curve: P = 0.151. (D–F) Pancreata from D1KO and WT mice were homogenized in acid ethanol and insulin, and proinsulin was measured with ELISA. (D) Proinsulin-to-insulin ratio, as calculated by the (E) total insulin and (F) proinsulin content, normalized to protein content. Data are means ± SEM (n = 6). Welch-corrected t test, two-tailed, unpaired, *P < 0.05. (G) mRNA expression of insulin-processing prohormone convertases 1 and 2 (Pcsk1 and Pcsk2) from isolated islets of D1KO mice and WT controls. Data are means ± SEM (n = 4 to 7). Welch-corrected Student t test, two-tailed, unpaired, *P < 0.05. n.s., not significant. Pancreatic islets of D1KO mice show increased IS upon treatment with sulfonylureas or KCl We further investigated different steps in the triggering pathway of GSIS. GLUT2 abundance was unchanged between islets of WT and D1KO mice (Supplemental Fig. 2B). Pharmacological approaches revealed that adenosine triphosphate–sensitive potassium (K+-ATP) channel inhibition substantially increased basal IS in D1KO islets compared with WT controls. This finding was demonstrated by additional perifusion experiments, where addition of 500 µM tolbutamide to 5 mM glucose significantly increased IS in D1KO islets compared with WT controls (Fig. 4A). Additional stimulation with 40 mM KCl resulted in a generally increased IS, independent from the genotype (Fig. 4A). In a static GSIS experiment using 1 µM glibenclamide (Fig. 4B), the increased GSIS in D1KO islets after K+-ATP channel inhibition was confirmed. In contrast to the perifusion experiment, 30 mM KCl-induced IS in the static approach was significantly higher in islets from D1KO mice (Fig. 4C). In contrast, after incubation of isolated pancreatic islets with 5 µM Ca2+ ionophore A23187 (Fig. 4D) at 2 mM glucose, respectively, no genotype-dependent changes in IS were detected. The increased sulfonylurea-induced IS was not attributed to changes in total mRNA expression of the K+-ATP channel-forming subunits Kcnj11 (Kir6.2) and Abcc8 (SUR1), as they remained unchanged between the genotypes (Fig. 4E). Figure 4. View largeDownload slide Pharmacological stimulation of IS at basal glucose concentrations. (A) Fifty freshly isolated islets per genotype were perifused in a custom-made perifusion chamber with 5 mM glucose and stimulated with 500 µM tolbutamide (Tol) and 40 mM KCl for the designated time points. Insulin from the eluate fractions was measured with ELISA, and the values are expressed as percentage of the secretion rate at the end of the basal glucose period (WT = 8.52 ± 4.84 pg ⋅ minutes−1 ⋅ islet−1; D1KO = 4.73 ± 2.15 pg ⋅ minutes−1 ⋅ islet−1), with the area under the curve (AUC) between 60 and 90 minutes of the tolbutamide stimulation as inset. Data are means ± SEM (n = 5), Mann-Whitney U test, two-tailed of the AUC. (B–D) Isolated islets of D1KO mice were incubated with 2 mM glucose, (B) with or without 1 µM glibenclamide (solubilized in dimethyl sulfoxide), (C) 30 mM KCl, or (D) 5 µM calcium-ionophore A23187 (D) for 1 hour at 37°C, respectively. Secreted insulin in the supernatants and islet lysates was measured with ELISA. Data are means ± SEM (n = 5 to 7), two-way analysis of variance with Bonferroni post hoc test, **P < 0.01. (E) mRNA expression of K+-ATP channel–forming subunits Kcnj11 (Kir6.2) and Abcc8 (SUR1) from isolated islets of D1KO mice and WT controls. Data are means ± SEM (n = 8). Figure 4. View largeDownload slide Pharmacological stimulation of IS at basal glucose concentrations. (A) Fifty freshly isolated islets per genotype were perifused in a custom-made perifusion chamber with 5 mM glucose and stimulated with 500 µM tolbutamide (Tol) and 40 mM KCl for the designated time points. Insulin from the eluate fractions was measured with ELISA, and the values are expressed as percentage of the secretion rate at the end of the basal glucose period (WT = 8.52 ± 4.84 pg ⋅ minutes−1 ⋅ islet−1; D1KO = 4.73 ± 2.15 pg ⋅ minutes−1 ⋅ islet−1), with the area under the curve (AUC) between 60 and 90 minutes of the tolbutamide stimulation as inset. Data are means ± SEM (n = 5), Mann-Whitney U test, two-tailed of the AUC. (B–D) Isolated islets of D1KO mice were incubated with 2 mM glucose, (B) with or without 1 µM glibenclamide (solubilized in dimethyl sulfoxide), (C) 30 mM KCl, or (D) 5 µM calcium-ionophore A23187 (D) for 1 hour at 37°C, respectively. Secreted insulin in the supernatants and islet lysates was measured with ELISA. Data are means ± SEM (n = 5 to 7), two-way analysis of variance with Bonferroni post hoc test, **P < 0.01. (E) mRNA expression of K+-ATP channel–forming subunits Kcnj11 (Kir6.2) and Abcc8 (SUR1) from isolated islets of D1KO mice and WT controls. Data are means ± SEM (n = 8). D1KO β-cells exhibit more insulin granules after glucose stimulation We assessed morphometric parameters in hematoxylin/eosin-stained pancreatic sections of D1KO mice and WT littermates. D1KO did not affect overall islet morphology (Fig. 5A). Total pancreas area, total islet number, amount of islets per millimeter-squared pancreas, and the mean islet area per millimeter-squared pancreas were unchanged (Fig. 5B), as well as the islet area distribution within the pancreas (Supplemental Fig. 3A). Figure 5. View largeDownload slide Morphometric and ultrastructural analysis of islets from D1KO mice. (A) Pancreas from D1KO and WT mice was dissected, and sections of the formalin-fixed and paraffin-embedded pancreas were stained with hematoxylin/eosin. (B) Fourteen to 21 sections of different levels from three pancreata per genotype, respectively, were analyzed for pancreas area by point counting, total amount of islets, amount of islets per millimeter-squared pancreas, and percent of islet area per pancreas area. Data are means ± SEM (n = 3). (C) Representative TEM images of β-cells from isolated WT and D1KO islets after 2 or 25 mM glucose stimulation with insulin-containing granules. (D) Representative magnification of an electron microscopic picture showing mature insulin granules in solid circles and immature insulin granules in dashed circles. (E) Quantification of mature insulin granules per β-cell area in islets of WT and D1KO mice after 2 and 25 mM glucose stimulation. Mature insulin granules were identified and quantified by grayscale threshold analysis. Data are means ± SEM of 10 β-cells from four to five different islets per condition, respectively. Two-way analysis of variance (ANOVA) with Bonferroni post hoc test, *P < 0.05; Student t test, two-tailed, unpaired, ###P < 0.001. (F) Immature insulin granules in β-cells of islets after 2 and 25 mM glucose stimulation were identified and counted manually. Data are means ± SEM of 10 β-cells from four to five different islets per condition, respectively. Student t test, two-tailed, unpaired, #P < 0.05; ###P < 0.001. (G) Docked insulin granules in β-cells of islets after 2 and 25 mM glucose stimulation were identified and counted manually. Granules with a distance ≤400 nm from the plasma membrane were considered as docked granules. Data are means ± SEM of 10 β-cells from four to five different islets per condition, respectively. Two-way ANOVA with Bonferroni post hoc test, **P < 0.01; Student t test, two-tailed, unpaired, ###P < 0.001. Figure 5. View largeDownload slide Morphometric and ultrastructural analysis of islets from D1KO mice. (A) Pancreas from D1KO and WT mice was dissected, and sections of the formalin-fixed and paraffin-embedded pancreas were stained with hematoxylin/eosin. (B) Fourteen to 21 sections of different levels from three pancreata per genotype, respectively, were analyzed for pancreas area by point counting, total amount of islets, amount of islets per millimeter-squared pancreas, and percent of islet area per pancreas area. Data are means ± SEM (n = 3). (C) Representative TEM images of β-cells from isolated WT and D1KO islets after 2 or 25 mM glucose stimulation with insulin-containing granules. (D) Representative magnification of an electron microscopic picture showing mature insulin granules in solid circles and immature insulin granules in dashed circles. (E) Quantification of mature insulin granules per β-cell area in islets of WT and D1KO mice after 2 and 25 mM glucose stimulation. Mature insulin granules were identified and quantified by grayscale threshold analysis. Data are means ± SEM of 10 β-cells from four to five different islets per condition, respectively. Two-way analysis of variance (ANOVA) with Bonferroni post hoc test, *P < 0.05; Student t test, two-tailed, unpaired, ###P < 0.001. (F) Immature insulin granules in β-cells of islets after 2 and 25 mM glucose stimulation were identified and counted manually. Data are means ± SEM of 10 β-cells from four to five different islets per condition, respectively. Student t test, two-tailed, unpaired, #P < 0.05; ###P < 0.001. (G) Docked insulin granules in β-cells of islets after 2 and 25 mM glucose stimulation were identified and counted manually. Granules with a distance ≤400 nm from the plasma membrane were considered as docked granules. Data are means ± SEM of 10 β-cells from four to five different islets per condition, respectively. Two-way ANOVA with Bonferroni post hoc test, **P < 0.01; Student t test, two-tailed, unpaired, ###P < 0.001. We further analyzed the ultrastructure of β-cells from D1KO mice and WT controls with regard to the amount and distribution of mature and immature insulin granules by transmission electron microscopy (TEM). In accordance with the ex vivo static GSIS measurements, isolated islets were incubated at 2 and 25 mM glucose for 1 hour (Fig. 5C and 5D) before TEM analysis. The density of mature granules was equal in β-cells from both genotypes at 2 mM glucose. However, after glucose stimulation, the density of mature granules decreased significantly in β-cells of WT but not D1KO β-cells. This led to a significantly increased, mature granule density in β-cells from D1KO islets after 25 mM glucose stimulation (Fig. 5E). The density of immature insulin granules was significantly decreased in D1KO β-cells after stimulation with 2 mM glucose but markedly increased in the high-glucose state. Consequently, D1KO β-cells demonstrated a highly significant increase of immature granules upon glucose stimulation, whereas the number of immature granules remained stable in WT β-cells (Fig. 5F). Likewise, the total density of granules, 400 nm beneath the plasma membrane (docked granules), was increased in β-cells of D1KO islets compared with WT after 25 mM glucose stimulation (Fig. 5G). Additional expression analysis of various genes relevant for islet physiology and vesicle dynamics showed no meaningful differences between D1KO and WT islets (Supplemental Fig. 4). D1KO increases palmitate uptake and oxidation in isolated islets In previous studies, D1KO mice exhibited enhanced lipid use in vivo, as well as increased uptake and oxidation of [3H]palmitate in isolated skeletal muscles (6). Therefore, we analyzed lipid handling in isolated islets of D1KO mice and WT controls. In line with previous results from skeletal muscle, islets of D1KO mice showed a substantially (40%) increased palmitate uptake (Fig. 6A) and a concomitant increase in palmitate oxidation (Fig. 6B). Of note, mRNA expression of genes involved in fatty acid signaling, uptake, and oxidation was unchanged in isolated islets from both genotypes (Supplemental Fig. 5A). However, D1KO islets demonstrated a substantially elevated mtDNA copy number compared with WT islets (Fig. 6C). In addition, expression of the mitochondrial fission gene Dnm1l (Drp1) was significantly increased in D1KO, whereas the expression of Opa1 that mediates mitochondrial fusion was unchanged (Fig. 6D). Nevertheless, we could not find a differential citrate synthase activity (Fig. 6E), alterations in the mitochondrial area in single β-cells (Fig. 6F), or a differential protein abundance of oxidative phosphorylation (OXPHOS) complexes between D1KO and WT islets (Fig. 6G). Seahorse measurements for OCRs showed a small but significant increase in baseline oxygen consumption in D1KO islets after baseline normalization. Injection of 5 µM oligomycin reduced OCRs by ∼40% in WT and 25% in D1KO islets compared with baseline OCRs. Remaining OCRs, after injection of oligomycin, were not different between D1KO and WT islets (P = 0.266). Injection of a 5-µM antimycin A/rotenone mixture reduced OCRs by ∼80% compared with baseline OCRs in both WT and D1KO islets (Fig. 6H). Figure 6. View largeDownload slide Lipid metabolism and mitochondrial function in isolated islets of D1KO mice. (A) Isolated islets of D1KO and WT mice were incubated with [14C]palmitic acid for 20 minutes at 37°C. Accumulated [14C]palmitic acid was measured by scintillation counting in the islet lysates, and cost per thousand impression (CPM) values normalized to islet numbers were expressed as fold over WT. Data are means ± SEM (n = 6), Mann-Whitney U test, two-tailed, **P < 0.01. (B) Isolated islets of D1KO and WT mice were incubated with [14C]palmitic acid for 4 hours at 37°C. The produced 14C-CO2 from oxidized palmitic acid was trapped in filter papers and measured by scintillation counting, and CPM values normalized to islet numbers were expressed as fold over WT. Data are means ± SEM (n = 6), Mann-Whitney U test, two-tailed, *P < 0.05. (C) Mitochondrial copy number. The ratio of mtDNA (mt-Nd2) to nuclear DNA (Rps18) was determined by real-time qPCR using SYBR Green with 40 ng DNA from isolated islets. Data are means ± SEM (n = 5 to 7), Student t test, two-tailed, unpaired, ***P < 0.001. (D) mRNA expression of Dnm1l and Opa1 in isolated islets. Data are means ± SEM (n = 7), Student t test, two-tailed, unpaired, *P < 0.05. (E) Citrate synthase activity was analyzed from isolated islets using 8 µg total protein with the Citrate Synthase Assay Kit (Sigma, Steinheim, Germany), according to the manufacturer’s instructions. Data are means ± SEM (n = 7 to 8). (F) Mitochondria of β-cells were analyzed from a small subset of TEM images (Fig. 5). The sum of the area of all mitochondria per β-cell was referred to the cell area and expressed as percentage. Data are means ± SEM of 10 β-cells per genotype. (G) Quantification with representative Western blot showing abundance of OXPHOS complexes (C)II, -III, and -V of isolated D1KO and WT islets against glyceraldehyde 3-phosphate dehydrogenase abundance. Data are means ± SEM (n = 3 to 4). (H) Measurement of OCR in isolated islets with the Seahorse XF24e analyzer. Seventy islets were seeded per well of a 24-well islet capture plate in duplicates, and baseline OCR was initially measured four times within 30 minutes. After addition of 5 µM oligomycin (Oligo) or 5 µM antimycin A/rotenone (AA/Rot) mixture, five measurements within 40 minutes were taken. After baseline correction, as recommended by Agilent and previous reports (27), the last baseline measure and the lowest of the five OCR measures of both oligomycin and antimycin A/rotenone were used for calculation. Data are means ± SEM (n = 4), Student t test, two-tailed, unpaired, *P < 0.05. Figure 6. View largeDownload slide Lipid metabolism and mitochondrial function in isolated islets of D1KO mice. (A) Isolated islets of D1KO and WT mice were incubated with [14C]palmitic acid for 20 minutes at 37°C. Accumulated [14C]palmitic acid was measured by scintillation counting in the islet lysates, and cost per thousand impression (CPM) values normalized to islet numbers were expressed as fold over WT. Data are means ± SEM (n = 6), Mann-Whitney U test, two-tailed, **P < 0.01. (B) Isolated islets of D1KO and WT mice were incubated with [14C]palmitic acid for 4 hours at 37°C. The produced 14C-CO2 from oxidized palmitic acid was trapped in filter papers and measured by scintillation counting, and CPM values normalized to islet numbers were expressed as fold over WT. Data are means ± SEM (n = 6), Mann-Whitney U test, two-tailed, *P < 0.05. (C) Mitochondrial copy number. The ratio of mtDNA (mt-Nd2) to nuclear DNA (Rps18) was determined by real-time qPCR using SYBR Green with 40 ng DNA from isolated islets. Data are means ± SEM (n = 5 to 7), Student t test, two-tailed, unpaired, ***P < 0.001. (D) mRNA expression of Dnm1l and Opa1 in isolated islets. Data are means ± SEM (n = 7), Student t test, two-tailed, unpaired, *P < 0.05. (E) Citrate synthase activity was analyzed from isolated islets using 8 µg total protein with the Citrate Synthase Assay Kit (Sigma, Steinheim, Germany), according to the manufacturer’s instructions. Data are means ± SEM (n = 7 to 8). (F) Mitochondria of β-cells were analyzed from a small subset of TEM images (Fig. 5). The sum of the area of all mitochondria per β-cell was referred to the cell area and expressed as percentage. Data are means ± SEM of 10 β-cells per genotype. (G) Quantification with representative Western blot showing abundance of OXPHOS complexes (C)II, -III, and -V of isolated D1KO and WT islets against glyceraldehyde 3-phosphate dehydrogenase abundance. Data are means ± SEM (n = 3 to 4). (H) Measurement of OCR in isolated islets with the Seahorse XF24e analyzer. Seventy islets were seeded per well of a 24-well islet capture plate in duplicates, and baseline OCR was initially measured four times within 30 minutes. After addition of 5 µM oligomycin (Oligo) or 5 µM antimycin A/rotenone (AA/Rot) mixture, five measurements within 40 minutes were taken. After baseline correction, as recommended by Agilent and previous reports (27), the last baseline measure and the lowest of the five OCR measures of both oligomycin and antimycin A/rotenone were used for calculation. Data are means ± SEM (n = 4), Student t test, two-tailed, unpaired, *P < 0.05. Discussion In this study, we investigated the role of the RabGAP TBC1D1 in mouse pancreatic islets. RabGAPs have been described to be key players of energy substrate use in skeletal muscle, but their role in other tissues remains rather unknown. In our study, we demonstrate that TBC1D1 is an important regulator of IS and lipid metabolism in mouse pancreatic islets by controlling glucose- and secretagogue-stimulated IS and lipid oxidation. We showed that expression of Tbc1d1 highly dominates over Tbc1d4 in isolated islets, emphasizing the bigger role for TBC1D1 in islet physiology. Furthermore, our study represents the description of a short isoform of Tbc1d1 in islets and the generation of the transgenic RIP2-D1 mice carrying this short isoform. In recent studies, knockdown of Tbc1d1 and Tbc1d4 in pancreatic β-cells has been shown to affect GSIS, but the effects were not conclusive: in sorted rat β-cells and in a global rat knockout model, D1KO led to a moderate increase in both basal IS and GSIS (15, 16). In contrast, knockdown of the related Tbc1d4 in primary mouse β-cells resulted in increased basal IS but decreased GSIS (14), suggesting an opposing role of both RabGAPs in GSIS. In the current study, we first identified Tbc1d1 as the predominant RabGAP in mouse islets compared with Tbc1d4. The short Tbc1d1 isoform, which is also abundant in adipocytes but not in muscle (1), lacks 94 aa (631–724) containing two 5-aminoimidazole-4-carboxamide ribonucleotide–regulated phosphorylation sites: Ser-660 and Ser-700 (1, 32). Whereas the specific functions of TBC1D1 isoforms are unknown, previous studies of TBC1D4 demonstrated altered biological activities of isoforms and isoform-specific mutations (33, 34). Consistent with data from rats (15), we found a relatively higher increase of IS in Tbc1d1 knockout islets than in islets from WT mice. Furthermore, TBC1D1 protein appears not to be rate limiting in islets, as β-cell-specific overexpression of Tbc1d1 neither changes basal IS nor GSIS compared with WT controls. The divergence from previous studies in rat β-cells (15, 16) might be most likely explained by a species-related diversity. In fact, D1KO rats exhibited no changes in GLUT4 abundance in skeletal muscle, which is considered a hallmark of all published D1KO mice so far (16). Importantly, increased GSIS in D1KO islets persists after three repetitive glucose challenges, indicating that the mechanism leading to increased GSIS upon D1KO is operating continuously and thus, reflects a physiologically relevant context. This is also supported by the markedly amplified first and second phase of IS in D1KO islets, demonstrating that both the triggering and amplifying pathways are enhanced upon D1KO. The higher degree of interindividual variation of IS in perifusion assays compared with the static GSIS assay may result from differences in islet handling and preincubation times but applied to both genotypes. The increase in both glibenclamide- and tolbutamide-stimulated IS at basal glucose concentrations (2 and 5 mM) was markedly potentiated in islets of D1KO mice. This effect is unlikely a result of altered adenosine triphosphate (ATP)/adenosine 5′-diphosphate ratios in β-cells but rather reflects enhanced K+-ATP channel signaling. Interestingly, the K+-ATP channels have been reported to localize not only to the plasma membrane of β-cells but are also present in considerable amounts on insulin granules (35). Moreover, sulfonylureas were found to stimulate IS via granule-localized channel inhibition without binding to plasma membrane K+-ATP channels (36) and can mediate the fusion of newly generated granules for IS (37). It is possible that D1KO increases K+-ATP channel trafficking and surface expression at basal glucose concentrations (38, 39). However, we found no relevant changes in total mRNA expression of the K+-ATP channel–forming subunits in islets. A previous study reported that Tbc1d1 knockdown prevents IS induced by KCl in sorted rat β-cells (15). In our study, we show that 60 minutes of 30 mM KCl stimulation results in increased IS in the D1KO islets. This discrepancy might be attributed to differences in the experimental protocol and the fact that Tbc1d1 knockdown in rat β-cells led to only 70% reduction of TBC1D1 protein (15). Together with the sulfonylurea-induced increase in IS, these findings suggest a role for TBC1D1 on cell-membrane depolarization. This is supported by the fact that KCl-induced IS in the perifusion approach was not different between the genotypes, as applied in addition to the already-present tolbutamide. Furthermore, with the bypassing of the downstream Ca2+ channel with the Ca2+ ionophore A23187, IS was the same in both WT and D1KO islets. Stimulation of IS with this ionophore is quite low compared with the stimulation with the other secretagogues. However, the stimulation fold with 5 µM A23187 is comparable with what was already found with INS-1 cells (40). Taken together, these data suggest that the extent of the cell-membrane depolarization might be one explanation for the differences in IS observed upon D1KO. Our results indicate that both synthesis of newly generated insulin vesicles and exocytosis of existing granules are increased by D1KO (41–44). Consistently, our ultrastructure analysis strongly suggests that β-cells of D1KO islets are capable of faster restoring of insulin granules after glucose stimulation to increase secretion in a time-dependent manner. This is also reflected by the dynamic perifusion experiment with 30 mM glucose. The measurements of proinsulin and insulin in the total pancreas also reflect the results of the insulin-granule analysis with the consideration of the basal 2 mM state. As immature granules may still contain more proinsulin than mature granules (45), the lower amount of proinsulin in the D1KO pancreas reflects the lower density of immature granules in D1KO β-cells at 2 mM glucose and therefore, the lower proinsulin-to-insulin ratio. In accordance, islets from D1KO mice show increased gene-expression levels of Pcsk1 and Pcsk2, suggesting a higher activity in the proinsulin-to-insulin processing compared with WT islets. Apart from proinsulin, total pancreatic insulin content, as well as insulin content of islet lysates at the end of the stimulation period from the GSIS, showed no differences between WT and D1KO mice. Whereas this observation is in line with our results from the morphometric analysis that showed no differences in islet size, density, and morphology, genotype-dependent alterations in insulin content per granule might be resolved in more detailed analyses. Several genes, related to islet cell identity, insulin exocytosis, and insulin signaling, were analyzed. Among these genes, only Mafa, Dnajc5, and Irs2 showed a differential mRNA expression in D1KO islets compared with WT controls. However, for all three genes, the respective protein abundance was found to be unchanged between WT and D1KO islets. In line with previous studies, D1KO mice showed normal glucose tolerance (2, 12, 13). However, we observed lower fasting insulin levels, implicating moderately elevated hepatic insulin sensitivity (46). Moreover, compared with WT, D1KO mice secreted relatively more insulin in response to the glucose bolus to achieve normoglycemia, which may reflect impaired glucose clearance as a result of reduced insulin-stimulated glucose transport into skeletal muscle. Measurements of plasma insulin after an intraperitoneal glucose injection may not exactly reflect in vivo IS. Further in-depths analyses, e.g., with the use of hyperglycemic clamp techniques, would be necessary to better characterize the impact of TBC1D1 on in vivo IS. Whereas we cannot rule out that the observed increase in GSIS might be a compensatory reaction to maintain clearance of glucose from the blood in D1KO mice, the data provide evidence for a direct role of TBC1D1 in vesicle dynamics in islets. Skeletal muscles from D1KO mice exhibit increased uptake and oxidation of long-chain fatty acids (6). Fatty acid metabolism is likely also to influence β-cell function and IS via the amplifying pathway (19, 47). In addition, ectopic accumulation of lipids is a well-known precursor in the development of insulin resistance (48, 49). Isolated pancreatic islets from D1KO mice displayed an increase in both uptake and oxidation of [14C]palmitic acid, which might reflect a more generalized role of TBC1D1 in substrate use. Although this effect seems not to be mediated by increased expression of genes regulating lipid metabolism, it might be a result of an enhanced mitochondrial function. Interestingly, the mitochondrial copy number was significantly increased in D1KO islets, which was in accordance with higher mRNA levels of the Dnm1l, a factor for mitochondrial fission processes. In contrast, no changes in mitochondrial fusion, citrate synthase activity, or mitochondrial area were observed between the genotypes. Of note, previous results in skeletal muscle showed an increased citrate synthase activity without a change in Nd2 amplification in D1KO mice (2). A recent report indicates an increased adenosine 5′-diphosphate-dependent mitochondrial respiration in skeletal muscle of D1KO rats in combination with increased fatty acid oxidation as well (50). However, abundance of OXPHOS complexes in the rat muscle was unchanged. This is in line with our results, where the abundance of OXPHOS complexes is also unaffected as a result of the D1KO. Basal oxygen consumption, as measured using the Seahorse XF24e analyzer, was increased in D1KO islets at baseline but was unchanged after addition of the oligomycin or antimycin A/rotenone mixture. As citrate synthase activity is unchanged, it is possible that anaplerotic reactions or the glycerol-3-phosphate dehydrogenase shuttle of the reduced form of nicotinamide adenine dinucleotide (51) contribute to the increased basal respiration without affecting the Krebs cycle. The increased expression of Dnm1l could also indicate an increased fission of peroxisomes (52). Thus, we cannot rule out that the increase in palmitate oxidation is a result of a bigger proportion of peroxisomal lipid oxidation in D1KO islets. Moreover, D1KO may not only affect mitochondrial function in β-cells but also in other islet cell types, also contributing to the increased palmitate oxidation. As fatty acid oxidation leads to a rise in cellular ATP levels, this TBC1D1-dependent process might also induce IS via the triggering pathway. These findings strengthen the hypothesis of a protective mechanism with the aim to maintain IS, despite a progressive insulin resistance in peripheral tissues by downregulation of TBC1D1 abundance or activity. In conclusion, we present evidence for a role of TBC1D1 as regulator of IS and lipid metabolism in pancreatic β-cells by enhanced lipid flux and glucose- and sulfonylurea-induced IS in the absence of TBC1D1. Appendix. Antibody Table Peptide/Protein Target  Antigen Sequence (if Known)  Name of Antibody  Manufacturer, Catalog No.  Species Raised in; Monoclonal or Polyclonal  Dilution Used  RRID  MafA  aa 300–359 of Mouse MafA  MafA antibody  Bethyl Laboratories, A300-611A  Rabbit; polyclonal  1:500  AB_2297116  IRS-2  Total IRS-2 protein  IRS-2 antibody  Cell Signaling, 4502  Rabbit; polyclonal  1:1000  AB_2125774  TBC1D1  Around V796 of mouse TBC1D1  TBC1D1 (V796)  Cell Signaling, 4629  Rabbit; polyclonal  1:1000  AB_1904162  TBC1D4  aa 1178–1189 of Rat AS160  Anti-AS160  Millipore, 07-741  Rabbit; polyclonal  1:1000  AB_492639  α-Tubulin  —  Monoclonal anti-α-tubulin  Calbiochem, T 6074  Mouse; monoclonal  1:1000  AB_477582  CSPα  aa 182–198 of Rat CSP  Anti-CSP  Synaptic Systems, 154 003  Rabbit; polyclonal  1:1000  AB_887710  GLUT2  aa 32–98 of Human GLUT2  GLUT2 (H-67)  Santa Cruz, sc-9117  Rabbit; polyclonal  1:1000  AB_641068  GAPDH  —  Anti-GAPDH  Ambion, AM4300  Mouse; monoclonal  1:4000  AB_437392  GAPDH  Near carboxy terminus of human GAPDH  Anti-GAPDH (14C10)  Cell Signaling, 2118  Rabbit; monoclonal  1:4000  AB_561053  OXPHOS    Total OXPHOS rodent WB antibody cocktail  Abcam, ab110413  Mouse; monoclonal  1:250  AB_2629281  Secondary HRP-conjugated anti-rabbit IgG    Goat anti-rabbit IgG (H + L)  Dianova, 111-035-003  Goat  1:20,000  AB_2313567  Secondary HRP-conjugated anti-mouse IgG    Rabbit anti-mouse IgG, Fγ  Dianova, 315-035-008  Rabbit  1:20,000  AB_2340063  Peptide/Protein Target  Antigen Sequence (if Known)  Name of Antibody  Manufacturer, Catalog No.  Species Raised in; Monoclonal or Polyclonal  Dilution Used  RRID  MafA  aa 300–359 of Mouse MafA  MafA antibody  Bethyl Laboratories, A300-611A  Rabbit; polyclonal  1:500  AB_2297116  IRS-2  Total IRS-2 protein  IRS-2 antibody  Cell Signaling, 4502  Rabbit; polyclonal  1:1000  AB_2125774  TBC1D1  Around V796 of mouse TBC1D1  TBC1D1 (V796)  Cell Signaling, 4629  Rabbit; polyclonal  1:1000  AB_1904162  TBC1D4  aa 1178–1189 of Rat AS160  Anti-AS160  Millipore, 07-741  Rabbit; polyclonal  1:1000  AB_492639  α-Tubulin  —  Monoclonal anti-α-tubulin  Calbiochem, T 6074  Mouse; monoclonal  1:1000  AB_477582  CSPα  aa 182–198 of Rat CSP  Anti-CSP  Synaptic Systems, 154 003  Rabbit; polyclonal  1:1000  AB_887710  GLUT2  aa 32–98 of Human GLUT2  GLUT2 (H-67)  Santa Cruz, sc-9117  Rabbit; polyclonal  1:1000  AB_641068  GAPDH  —  Anti-GAPDH  Ambion, AM4300  Mouse; monoclonal  1:4000  AB_437392  GAPDH  Near carboxy terminus of human GAPDH  Anti-GAPDH (14C10)  Cell Signaling, 2118  Rabbit; monoclonal  1:4000  AB_561053  OXPHOS    Total OXPHOS rodent WB antibody cocktail  Abcam, ab110413  Mouse; monoclonal  1:250  AB_2629281  Secondary HRP-conjugated anti-rabbit IgG    Goat anti-rabbit IgG (H + L)  Dianova, 111-035-003  Goat  1:20,000  AB_2313567  Secondary HRP-conjugated anti-mouse IgG    Rabbit anti-mouse IgG, Fγ  Dianova, 315-035-008  Rabbit  1:20,000  AB_2340063  Abbreviations: GAPDH, glyceraldehyde 3-phosphate dehydrogenase; H + L, heavy chain and light chain; HRP, horseradish peroxidase; IgG, immunoglobulin G; RRID, Research Resource Identifier. View Large Abbreviations: Abbreviations: aa amino acid ATP adenosine triphosphate BSA bovine serum albumin cDNA complementary DNA D1KO Tbc1d1-deficient ELISA enzyme-linked immunosorbent assay GLUT glucose transporter GSIS glucose-stimulated insulin secretion IS insulin secretion K+-ATP adenosine triphosphate–sensitive potassium KRH Krebs-Ringer-HEPES mRNA messenger RNA mtDNA mitochondrial DNA OCR oxygen consumption rate OXPHOS oxidative phosphorylation Pcsk1/2 prohormone convertase 1/2 qPCR quantitative polymerase chain reaction RabGAP Rab guanosine triphosphatase–activating protein RIP2-D1 TG-RIP2-3xFLAG-Tbc1d1 RT-PCR reverse transcription polymerase chain reaction SEM standard error of the mean TEM transmission electron microscopy WT wild-type Acknowledgments We thank Angelika Horrighs, Anette Kurowski, Barbara Bartosinska, Jennifer Schwettmann, Carolin Borchert, Ines Grüner, Verena Lier-Glaubitz, and Dr. Bengt-Frederik Belgardt for expert technical assistance, support, and consultation. Financial Support: This work was supported by the German Center for Diabetes Research of the Federal Ministry for Education and Research and the Ministry of Science and Research of the State North Rhine-Westphalia. T. Stermann received a stipend by the research training group “vivid” of the Heinrich-Heine University Düsseldorf. Additional project funding came from the German Diabetes Association (DDG) and the German Academic Exchange Service (DAAD). Author Contributions: T. Stermann, A.C., and H.A.-H. wrote the manuscript and analyzed and interpreted the data. T. Stermann, F.M., C.W., K.J., D.A., T.B., A.P., D.M.O, G.H.T., C.d.W., S.L. and T. Schallschmidt performed the experiments and analyzed data. J.W., F.B., I.R., M.K., E.L., A.C., and H.A.-H. were involved in the study design and contributed to data interpretation. Disclosure Summary: The authors have nothing to disclose. References 1. Roach WG, Chavez JA, Mîinea CP, Lienhard GE. Substrate specificity and effect on GLUT4 translocation of the Rab GTPase-activating protein Tbc1d1. Biochem J . 2007; 403( 2): 353– 358. Google Scholar CrossRef Search ADS PubMed  2. Chadt A, Immisch A, de Wendt C, Springer C, Zhou Z, Stermann T, Holman GD, Loffing-Cueni D, Loffing J, Joost HG, Al-Hasani H. “Deletion of both Rab-GTPase–activating proteins TBC1D1 and TBC1D4 in mice eliminates insulin- and AICAR-stimulated glucose transport [ published erratum appears in Diabetes. 2015:64(4):1492]. Diabetes . 2015; 64( 3): 746– 759. Google Scholar CrossRef Search ADS PubMed  3. Mîinea CP, Sano H, Kane S, Sano E, Fukuda M, Peränen J, Lane WS, Lienhard GE. AS160, the Akt substrate regulating GLUT4 translocation, has a functional Rab GTPase-activating protein domain. Biochem J . 2005; 391( 1): 87– 93. Google Scholar CrossRef Search ADS PubMed  4. Stenmark H. Rab GTPases as coordinators of vesicle traffic. Nat Rev Mol Cell Biol . 2009; 10( 8): 513– 525. Google Scholar CrossRef Search ADS PubMed  5. Dokas J, Chadt A, Nolden T, Himmelbauer H, Zierath JR, Joost HG, Al-Hasani H. Conventional knockout of Tbc1d1 in mice impairs insulin- and AICAR-stimulated glucose uptake in skeletal muscle. Endocrinology . 2013; 154( 10): 3502– 3514. Google Scholar CrossRef Search ADS PubMed  6. Chadt A, Leicht K, Deshmukh A, Jiang LQ, Scherneck S, Bernhardt U, Dreja T, Vogel H, Schmolz K, Kluge R, Zierath JR, Hultschig C, Hoeben RC, Schürmann A, Joost HG, Al-Hasani H. Tbc1d1 mutation in lean mouse strain confers leanness and protects from diet-induced obesity. Nat Genet . 2008; 40( 11): 1354– 1359. Google Scholar CrossRef Search ADS PubMed  7. Stone S, Abkevich V, Russell DL, Riley R, Timms K, Tran T, Trem D, Frank D, Jammulapati S, Neff CD, Iliev D, Gress R, He G, Frech GC, Adams TD, Skolnick MH, Lanchbury JS, Gutin A, Hunt SC, Shattuck D. TBC1D1 is a candidate for a severe obesity gene and evidence for a gene/gene interaction in obesity predisposition. Hum Mol Genet . 2006; 15( 18): 2709– 2720. Google Scholar CrossRef Search ADS PubMed  8. Dash S, Sano H, Rochford JJ, Semple RK, Yeo G, Hyden CS, Soos MA, Clark J, Rodin A, Langenberg C, Druet C, Fawcett KA, Tung YC, Wareham NJ, Barroso I, Lienhard GE, O’Rahilly S, Savage DB. A truncation mutation in TBC1D4 in a family with acanthosis nigricans and postprandial hyperinsulinemia. Proc Natl Acad Sci USA . 2009; 106( 23): 9350– 9355. Google Scholar CrossRef Search ADS PubMed  9. Dash S, Langenberg C, Fawcett KA, Semple RK, Romeo S, Sharp S, Sano H, Lienhard GE, Rochford JJ, Howlett T, Massoud AF, Hindmarsh P, Howell SJ, Wilkinson RJ, Lyssenko V, Groop L, Baroni MG, Barroso I, Wareham NJ, O’Rahilly S, Savage DB. Analysis of TBC1D4 in patients with severe insulin resistance. Diabetologia . 2010; 53( 6): 1239– 1242. Google Scholar CrossRef Search ADS PubMed  10. Meyre D, Farge M, Lecoeur C, Proenca C, Durand E, Allegaert F, Tichet J, Marre M, Balkau B, Weill J, Delplanque J, Froguel P. R125W coding variant in TBC1D1 confers risk for familial obesity and contributes to linkage on chromosome 4p14 in the French population. Hum Mol Genet . 2008; 17( 12): 1798– 1802. Google Scholar CrossRef Search ADS PubMed  11. Abou-Sabe’ MA. Stimulation of cAMP synthesis by glucose in a mutant of E. coli B-r. Nat New Biol . 1973; 243( 127): 182– 185. Google Scholar CrossRef Search ADS PubMed  12. Hargett SR, Walker NN, Keller SR. Rab GAPs AS160 and Tbc1d1 play non-redundant roles in the regulation of glucose and energy homeostasis in mice. Am J Physiol Endocrinol Metab . 2016: 310( 4): E276– E288. Google Scholar CrossRef Search ADS PubMed  13. Szekeres F, Chadt A, Tom RZ, Deshmukh AS, Chibalin AV, Björnholm M, Al-Hasani H, Zierath JR. The Rab-GTPase-activating protein TBC1D1 regulates skeletal muscle glucose metabolism. Am J Physiol Endocrinol Metab . 2012; 303( 4): E524– E533. Google Scholar CrossRef Search ADS PubMed  14. Bouzakri K, Ribaux P, Tomas A, Parnaud G, Rickenbach K, Halban PA. Rab GTPase-activating protein AS160 is a major downstream effector of protein kinase B/Akt signaling in pancreatic beta-cells. Diabetes . 2008; 57( 5): 1195– 1204. Google Scholar CrossRef Search ADS PubMed  15. Rütti S, Arous C, Nica AC, Kanzaki M, Halban PA, Bouzakri K. Expression, phosphorylation and function of the Rab-GTPase activating protein TBC1D1 in pancreatic beta-cells. FEBS Lett . 2014; 588( 1): 15– 20. Google Scholar CrossRef Search ADS PubMed  16. Paglialunga S, Simnett G, Robson H, Hoang M, Pillai RA, Arkell AM, Simpson JA, Bonen A, Huising M, Joseph JW, Holloway GP. The Rab-GTPase activating protein, TBC1D1, is critical for maintaining normal glucose homeostasis and beta-cell mass. Appl Physiol Nutr Metab . 2017; 42( 6): 647– 655. Google Scholar CrossRef Search ADS PubMed  17. Cerasi E. Potentiation of insulin release by glucose in man. Acta Endocrinol (Copenh) . 1975; 79( 3): 511– 534. Google Scholar PubMed  18. Cerasi E. Insulin secretion: mechanism of the stimulation by glucose. Q Rev Biophys . 1975; 8( 1): 1– 41. Google Scholar CrossRef Search ADS PubMed  19. Henquin JC. Triggering and amplifying pathways of regulation of insulin secretion by glucose. Diabetes . 2000; 49( 11): 1751– 1760. Google Scholar CrossRef Search ADS PubMed  20. Panten U, Willenborg M, Schumacher K, Hamada A, Ghaly H, Rustenbeck I. Acute metabolic amplification of insulin secretion in mouse islets is mediated by mitochondrial export of metabolites, but not by mitochondrial energy generation. Metabolism . 2013; 62( 10): 1375– 1386. Google Scholar CrossRef Search ADS PubMed  21. Henquin JC. The dual control of insulin secretion by glucose involves triggering and amplifying pathways in β-cells. Diabetes Res Clin Pract . 2011; 93( Suppl 1): S27– S31. Google Scholar CrossRef Search ADS PubMed  22. Valera A, Pujol A, Pelegrin M, Bosch F. Transgenic mice overexpressing phosphoenolpyruvate carboxykinase develop non-insulin-dependent diabetes mellitus. Proc Natl Acad Sci USA . 1994; 91( 19): 9151– 9154. Google Scholar CrossRef Search ADS PubMed  23. Dor Y, Brown J, Martinez OI, Melton DA. Adult pancreatic beta-cells are formed by self-duplication rather than stem-cell differentiation. Nature . 2004; 429( 6987): 41– 46. Google Scholar CrossRef Search ADS PubMed  24. Yesil P, Michel M, Chwalek K, Pedack S, Jany C, Ludwig B, Bornstein SR, Lammert E. A new collagenase blend increases the number of islets isolated from mouse pancreas. Islets . 2009; 1( 3): 185– 190. Google Scholar CrossRef Search ADS PubMed  25. Belz M, Willenborg M, Görgler N, Hamada A, Schumacher K, Rustenbeck I. Insulinotropic effect of high potassium concentration beyond plasma membrane depolarization. Am J Physiol Endocrinol Metab . 2014; 306( 6): E697– E706. Google Scholar CrossRef Search ADS PubMed  26. Lambernd S, Taube A, Schober A, Platzbecker B, Görgens SW, Schlich R, Jeruschke K, Weiss J, Eckardt K, Eckel J. Contractile activity of human skeletal muscle cells prevents insulin resistance by inhibiting pro-inflammatory signalling pathways. Diabetologia . 2012; 55( 4): 1128– 1139. Google Scholar CrossRef Search ADS PubMed  27. Wikstrom JD, Sereda SB, Stiles L, Elorza A, Allister EM, Neilson A, Ferrick DA, Wheeler MB, Shirihai OS. A novel high-throughput assay for islet respiration reveals uncoupling of rodent and human islets [ published erratum appears in PLoS One. 2013;8(12)]. PLoS One . 2012; 7( 5): e33023. Google Scholar CrossRef Search ADS PubMed  28. Weibel ER. Stereological principles for morphometry in electron microscopic cytology. Int Rev Cytol . 1969; 26: 235– 302. Google Scholar CrossRef Search ADS PubMed  29. Bock T, Pakkenberg B, Buschard K. Increased islet volume but unchanged islet number in ob/ob mice. Diabetes . 2003; 52( 7): 1716– 1722. Google Scholar CrossRef Search ADS PubMed  30. Jain D, Weber G, Eberhard D, Mehana AE, Eglinger J, Welters A, Bartosinska B, Jeruschke K, Weiss J, Päth G, Ariga H, Seufert J, Lammert E. DJ-1 protects pancreatic beta cells from cytokine- and streptozotocin-mediated cell death. PLoS One . 2015; 10( 9): e0138535. Google Scholar CrossRef Search ADS PubMed  31. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(−Delta Delta C(T)) method. Methods . 2001; 25( 4): 402– 408. Google Scholar CrossRef Search ADS PubMed  32. Taylor EB, An D, Kramer HF, Yu H, Fujii NL, Roeckl KS, Bowles N, Hirshman MF, Xie J, Feener EP, Goodyear LJ. Discovery of TBC1D1 as an insulin-, AICAR-, and contraction-stimulated signaling nexus in mouse skeletal muscle. J Biol Chem . 2008; 283( 15): 9787– 9796. Google Scholar CrossRef Search ADS PubMed  33. Baus D, Heermeier K, De Hoop M, Metz-Weidmann C, Gassenhuber J, Dittrich W, Welte S, Tennagels N. Identification of a novel AS160 splice variant that regulates GLUT4 translocation and glucose-uptake in rat muscle cells. Cell Signal . 2008; 20( 12): 2237– 2246. Google Scholar CrossRef Search ADS PubMed  34. Moltke I, Grarup N, Jørgensen ME, Bjerregaard P, Treebak JT, Fumagalli M, Korneliussen TS, Andersen MA, Nielsen TS, Krarup NT, Gjesing AP, Zierath JR, Linneberg A, Wu X, Sun G, Jin X, Al-Aama J, Wang J, Borch-Johnsen K, Pedersen O, Nielsen R, Albrechtsen A, Hansen T. A common Greenlandic TBC1D4 variant confers muscle insulin resistance and type 2 diabetes. Nature . 2014; 512( 7513): 190– 193. Google Scholar CrossRef Search ADS PubMed  35. Geng X, Li L, Watkins S, Robbins PD, Drain P. The insulin secretory granule is the major site of K(ATP) channels of the endocrine pancreas. Diabetes . 2003; 52( 3): 767– 776. Google Scholar CrossRef Search ADS PubMed  36. Geng X, Li L, Bottino R, Balamurugan AN, Bertera S, Densmore E, Su A, Chang Y, Trucco M, Drain P. Antidiabetic sulfonylurea stimulates insulin secretion independently of plasma membrane KATP channels. Am J Physiol Endocrinol Metab . 2007; 293( 1): E293– E301. Google Scholar CrossRef Search ADS PubMed  37. Nagamatsu S, Ohara-Imaizumi M, Nakamichi Y, Kikuta T, Nishiwaki C. Imaging docking and fusion of insulin granules induced by antidiabetes agents: sulfonylurea and glinide drugs preferentially mediate the fusion of newcomer, but not previously docked, insulin granules. Diabetes . 2006; 55( 10): 2819– 2825. Google Scholar CrossRef Search ADS PubMed  38. Chen PC, Bruederle CE, Gaisano HY, Shyng SL. Syntaxin 1A regulates surface expression of beta-cell ATP-sensitive potassium channels. Am J Physiol Cell Physiol . 2011; 300( 3): C506– C516. Google Scholar CrossRef Search ADS PubMed  39. Sivaprasadarao A, Taneja TK, Mankouri J, Smith AJ. Trafficking of ATP-sensitive potassium channels in health and disease. Biochem Soc Trans . 2007; 35( 5): 1055– 1059. Google Scholar CrossRef Search ADS PubMed  40. Cui X, Yang G, Pan M, Zhang XN, Yang SN. Akt signals upstream of L-type calcium channels to optimize insulin secretion. Pancreas . 2012; 41( 1): 15– 21. Google Scholar CrossRef Search ADS PubMed  41. Dehghany J, Hoboth P, Ivanova A, Mziaut H, Müller A, Kalaidzidis Y, Solimena M, Meyer-Hermann M. A spatial model of insulin-granule dynamics in pancreatic β-cells. Traffic . 2015; 16( 8): 797– 813. Google Scholar CrossRef Search ADS PubMed  42. Ohara-Imaizumi M, Fujiwara T, Nakamichi Y, Okamura T, Akimoto Y, Kawai J, Matsushima S, Kawakami H, Watanabe T, Akagawa K, Nagamatsu S. Imaging analysis reveals mechanistic differences between first- and second-phase insulin exocytosis. J Cell Biol . 2007; 177( 4): 695– 705. Google Scholar CrossRef Search ADS PubMed  43. Seino S, Shibasaki T, Minami K. Dynamics of insulin secretion and the clinical implications for obesity and diabetes. J Clin Invest . 2011; 121( 6): 2118– 2125. Google Scholar CrossRef Search ADS PubMed  44. Hoboth P, Müller A, Ivanova A, Mziaut H, Dehghany J, Sönmez A, Lachnit M, Meyer-Hermann M, Kalaidzidis Y, Solimena M. Aged insulin granules display reduced microtubule-dependent mobility and are disposed within actin-positive multigranular bodies [ published erratum appears in Proc Natl Acad Sci USA. 2015;112(16):E2114]. Proc Natl Acad Sci USA . 2015; 112( 7): E667– E676. Google Scholar CrossRef Search ADS PubMed  45. Borg LA, Westberg M, Grill V. The priming effect of glucose on insulin release does not involve redistribution of secretory granules within the pancreatic B-cell. Mol Cell Endocrinol . 1988; 56( 3): 219– 225. Google Scholar CrossRef Search ADS PubMed  46. Chen L, Chen Q, Xie B, Quan C, Sheng Y, Zhu S, Rong P, Zhou S, Sakamoto K, MacKintosh C, Wang HY, Chen S. Disruption of the AMPK-TBC1D1 nexus increases lipogenic gene expression and causes obesity in mice via promoting IGF1 secretion. Proc Natl Acad Sci USA . 2016; 113( 26): 7219– 7224. Google Scholar CrossRef Search ADS PubMed  47. Itoh Y, Kawamata Y, Harada M, Kobayashi M, Fujii R, Fukusumi S, Ogi K, Hosoya M, Tanaka Y, Uejima H, Tanaka H, Maruyama M, Satoh R, Okubo S, Kizawa H, Komatsu H, Matsumura F, Noguchi Y, Shinohara T, Hinuma S, Fujisawa Y, Fujino M. Free fatty acids regulate insulin secretion from pancreatic beta cells through GPR40. Nature . 2003; 422( 6928): 173– 176. Google Scholar CrossRef Search ADS PubMed  48. Grundy SM. Overnutrition, ectopic lipid and the metabolic syndrome. J Investig Med . 2016; 64( 6): 1082– 1086. Google Scholar CrossRef Search ADS PubMed  49. Kuzmenko DI, Klimentyeva TK. Role of ceramide in apoptosis and development of insulin resistance. Biochemistry (Mosc) . 2016; 81( 9): 913– 927. Google Scholar CrossRef Search ADS PubMed  50. Whitfield J, Paglialunga S, Smith BK, Miotto PM, Simnett G, Robson HL, Jain SS, Herbst EAF, Desjardins EM, Dyck DJ, Spriet LL, Steinberg GR, Holloway GP. Ablating the protein TBC1D1 impairs contraction-induced sarcolemmal glucose transporter 4 redistribution but not insulin-mediated responses in rats. J Biol Chem . 2017; 292( 40): 16653– 16664. Google Scholar CrossRef Search ADS PubMed  51. MacDonald MJ. High content of mitochondrial glycerol-3-phosphate dehydrogenase in pancreatic islets and its inhibition by diazoxide. J Biol Chem . 1981; 256( 16): 8287– 8290. Google Scholar PubMed  52. Itoyama A, Michiyuki S, Honsho M, Yamamoto T, Moser A, Yoshida Y, Fujiki Y. Mff functions with Pex11pβ and DLP1 in peroxisomal fission. Biol Open . 2013; 2( 10): 998– 1006. Google Scholar CrossRef Search ADS PubMed  Copyright © 2018 Endocrine Society http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Endocrinology Oxford University Press

Loading next page...
 
/lp/ou_press/deletion-of-the-rabgap-tbc1d1-leads-to-enhanced-insulin-secretion-and-qpRpiujzH8
Publisher
Endocrine Society
Copyright
Copyright © 2018 Endocrine Society
ISSN
0013-7227
eISSN
1945-7170
D.O.I.
10.1210/en.2018-00087
Publisher site
See Article on Publisher Site

Abstract

Abstract The Rab guanosine triphosphatase–activating protein (RabGAP) TBC1D1 has been shown to be a key regulator of glucose and lipid metabolism in skeletal muscle. Its function in pancreatic islets, however, is not yet fully understood. Here, we aimed to clarify the specific impact of TBC1D1 on insulin secretion and substrate use in pancreatic islets. We analyzed the dynamics of glucose-stimulated insulin secretion (GSIS) and lipid metabolism in isolated islets from Tbc1d1-deficient (D1KO) mice. To further investigate the underlying cellular mechanisms, we conducted pharmacological studies in these islets. In addition, we determined morphology and number of both pancreatic islets and insulin vesicles in β-cells using light and transmission electron microscopy. Isolated pancreatic islets from D1KO mice exhibited substantially increased GSIS compared with wild-type (WT) controls. This was attributed to both enhanced first and second phase of insulin secretion, and this enhanced secretion persisted during repetitive glucose stimuli. Studies with sulfonylureas or KCl in isolated islets demonstrated that TBC1D1 exerts its function via a signaling pathway at the level of membrane depolarization. In line, ultrastructural analysis of isolated pancreatic islets revealed both higher insulin-granule density and number of docked granules in β-cells from D1KO mice compared with WT controls. Like in skeletal muscle, lipid use in isolated islets was enhanced upon D1KO, presumably as a result of a higher mitochondrial fission rate and/or higher mitochondrial activity. Our results clearly demonstrate a dual role of TBC1D1 in controlling substrate metabolism of the pancreatic islet. The two closely related Rab guanosine triphosphatase–activating proteins (RabGAPs), TBC1D1 and TBC1D4, represent essential signaling factors in the regulation of lipid and glucose metabolism (1–3). With the exertion of their respective function via the activation of small Rab guanosine triphosphatases, both molecules control vesicle trafficking processes within a diversity of cell types (4). However, the majority of studies, to date, concentrate on RabGAP function in skeletal muscle and white adipose tissue, where they mediate the translocation of the insulin- and contraction-regulated glucose transporter (GLUT) type 4 to the plasma membrane, thus directly influencing glucose influx (2, 5). Deficiency of one or two of the TBC1D proteins leads to severely impaired insulin- and contraction-stimulated glucose uptake into skeletal muscle and adipocytes, likely caused by missorting and subsequent depletion of the GLUT4 protein (2). Interestingly, lipid use is enhanced in RabGAP-deficient mice, clearly emphasizing the central role of the two RabGAPs in the regulation of substrate preference and metabolic flexibility, a major trait in the pathophysiology of insulin resistance and type 2 diabetes (6). In humans, a coding variant of TBC1D1 was linked to extreme obesity, and a mutation in TBC1D4 was shown to cause postprandial hyperglycemia (7–10). Presumably a result of compensatory mechanisms caused by the remaining RabGAP, deletion of both RabGAPs is necessary to considerably impair whole-body glycemia in mice, and single-knockout animals for each TBC1D protein show only mild disturbances in their glycemic control (2, 11). In addition, contradictory findings for plasma insulin levels in vivo have been demonstrated throughout different studies (2, 12, 13). There is recent evidence that the metabolic impact of the two RabGAPs is not limited to peripheral insulin-responsive tissues alone. So far, only little research has been conducted on TBC1D1 or TBC1D4 function in pancreatic islets. Despite the fact that both RabGAPs are expressed in this tissue and seem to influence insulin secretion (IS), their exact role in the regulation of glucose and lipid metabolism in β-cells is unclear. A previous publication showed that knockdown of Tbc1d4 in sorted mouse β-cells led to increased basal IS but reduced glucose-stimulated insulin secretion (GSIS) (14). In addition, decreased Tbc1d4 expression resulted in enhanced apoptosis and reduced proliferation in these cells (14). Two independent groups showed a mild phenotype regarding IS upon Tbc1d1 knockdown in sorted rat β-cells and isolated pancreatic islets of Tbc1d1-knockout (D1KO) rats, respectively (15, 16). The dynamics of GSIS follow a first and a second phase (17, 18). The triggering pathway triggers the stimulated secretion but does not determine its extent (19). The latter depends on the amplifying pathway, which involves the export of mitochondrial metabolites (20). Additionally, extracellular signals, such as incretins or free fatty acids, play a role in the movement and depletion (21) of insulin-containing granules, thereby relying on a precise regulation of vesicular transport via Rab proteins and RabGAPs (4). However, not much research has been performed, to date, addressing the underlying cellular mechanisms linking RabGAPs to basic islet physiology. The questions remain whether the two RabGAPs control substrate use in pancreatic β-cells in analogy to peripheral tissues and how TBC1D proteins exert their function on β-cell-specific processes. Our findings clearly demonstrate an important role for TBC1D1 in the regulation of IS, as well as lipid metabolism, in islets and suggest a contribution in whole-body glycemia. Materials and Methods Chemicals and buffer Chemicals and buffer ingredients are listed in Supplemental Table 1. Experimental animals Mice were kept in accordance with the US National Institutes of Health guidelines for the care and use of laboratory animals, and all experiments were approved by the Ethics Committee of the State Ministry of Agriculture, Nutrition and Forestry (State of North Rhine-Westphalia, Germany). Three to six mice per cage were housed at 22°C and a 12-hour light: 12-hour dark cycle with ad libitum access to food and water. After weaning, animals received a standard diet (Ssniff, Soest, Germany). Male mice were used at the age of 12 to 16 weeks. The generation of D1KO mice has been described previously (2, 6). To generate transgenic TG-RIP2-3xFLAG-Tbc1d1 (RIP2-D1) mice overexpressing Tbc1d1 under the control of the RIP2 promoter, a DNA fragment for RIP2-3xFLAG-Tbc1d1 was microinjected into the male nucleus of zygotes from an F1(C57BL/6JOlaHsd × SJL/JCrHsd) intercross (22). Embryos were transferred to the oviduct of pseudopregnant females, and hemizygous transgenic offspring were identified by Southern blotting. Several lines were backcrossed to the N5 generation to C57BL/6JCrl mice using marker-assisted genotyping until reaching 100% C57BL/6J background, based on the analysis of 108 microsatellite markers. Cloning of the RIP2-3xFLAG-Tbc1d1 gene construct We replaced the KasI/PmeI Cre-ER cassette in the plasmid pBKS-RIP2-Cre-ER (Addgene) (23) by blunt-end cloning of a 3xFLAG-Tbc1d1 (short isoform) (6), and the resulting construct was verified by Sanger sequencing. A 5.6-kbp FseI/PmeI fragment for RIP2-3xFLAG-Tbc1d1-polyA was gel isolated, purified, and used for oocyte injections. Control mice are wild-type (WT) littermates, not carrying the RIP2-3xFLAG-Tbc1d1 transgene. Intraperitoneal glucose tolerance test Mice were fasted for 6 hours. Basal blood glucose was determined at the tail tip, and additional blood was collected for plasma insulin measurements. Mice were injected intraperitoneally with 2 g/kg body weight glucose, and blood glucose and plasma insulin were measured at 15, 30, 60, and 120 minutes after injection. Plasma insulin was quantified with the Insulin (Mouse) Ultrasensitive ELISA kit (DRG, Marburg, Germany). Islet isolation and static GSIS Pancreatic islets were isolated by ductal collagenase perfusion of the pancreas, as previously described (24). Islets were allowed to regenerate overnight in CMRL (Connaught Medical Research Laboratories) islet medium. All incubation steps were conducted at 37°C with 5% CO2. GSIS assays were performed with eight islets per well in a 96-well plate. Islets were adjusted in Krebs-Ringer-HEPES (KRH) buffer with 2 mM glucose for 1 hour before 1-hour incubation with 2 mM glucose as basal condition. Thereafter, the same islets were incubated for 1 hour with KRH buffer containing the designated secretagogue (25 mM glucose, 1 µM glibenclamide, 30 mM KCl, or 5 µM A23187) and were lysed after the last stimulation. Insulin in both the supernatants and the lysates was measured using mouse insulin enzyme-linked immunosorbent assay (ELISA) kits. Total pancreatic insulin/proinsulin content Dissected pancreas was snap frozen in liquid nitrogen. The pancreas was homogenized with ice-cold acid ethanol (0.18 M HCl in 75% ethanol) and kept shaking overnight at 4°C. The homogenate was centrifuged for 15 minutes at 5000 relative centrifugal force at 4°C, and the clear supernatant was subjected to insulin and proinsulin ELISA measurements and bicinchoninic acid protein assay. Islet isolation and dynamic perifusion Mice were euthanized, and the pancreas was removed and manually chopped in KRH buffer with scissors. After 8.5 minutes of collagenase digestion in a 37°C shaking waterbath, the tissue was washed with KRH buffer, and islets were handpicked from the exocrine tissue and assayed within 1 hour. Perifusion of 50 WT and D1KO islets, respectively, was carried out according to the previously described protocols (25). Palmitate uptake and oxidation Assays were done essentially as described (6). In brief, for palmitate uptake, 100 islets were incubated in KRH containing 0.1% fatty acid-free bovine serum albumin (BSA) and 2.8 mM glucose for 2 hours, washed three times with KRH + 0.1% BSA, and incubated with KRH + 40 mM BSA and 0.5 mCi/mL [1-14C]palmitic acid at 37°C (20 minutes for uptake assays). The cells were then washed with ice-cold KRH + 0.1% BSA and lysed with sodium dodecyl sulfate, and radioactivity was determined by scintillation counting. For palmitate oxidation, 10 islets were seeded in separate 48 wells in CMRL islet medium and cultured overnight at 37°C/ 5% CO2. The plate was placed into a custom-made oxidation chamber; islet medium was supplemented with [14C]palmitic acid (0.3 µCi per well), fatty acid-free BSA (6.24 µM per well), and l-Carnitin (1 µM per well); and the procedure continued as previously described (26). Mitochondrial copy number and citrate synthase activity The ratio of islet mitochondrial DNA (mtDNA; mt-Nd2) and nuclear DNA (Rps18), as well as citrate synthase activity, was determined, as previously described (5). Measurement of oxygen consumption rate To determine islet oxygen consumption rate (OCR), isolated islets were collected after overnight regeneration and resuspended in Seahorse XF Base medium (Agilent, Wilmington, DE), supplemented with 1% fetal bovine serum, 3 mM glucose, and 0.5 mM HEPES (pH 7.4), and seeded with 70 islets per well of a 24-well islet capture microplate, as described elsewhere (27). OCRs were measured at baseline and following injection with 5 µM oligomycin and 5 µM antimycin A/rotenone (XF Cell Mito Stress Test Kit), respectively, using a Seahorse XF24e analyzer (Agilent). The OCRs were normalized to the baseline values, as previously described (27), and expressed as percent of WT baseline. Morphometric analysis of islets Dissected pancreas was immediately fixed with 4% paraformaldehyde for 24 hours and embedded in paraffin after dehydration, according to standard procedures. Sections of 5 µm were prepared on microscope slides and stained with hematoxylin/eosin. Total pancreas area was determined by point counting (28, 29). Finally, 14 to 21 sections of different levels per pancreas from three animals per genotype were analyzed with cellSense Dimensions Software (Olympus, Hamburg, Germany). Electron microscopy of islets Isolated islets from three WT and D1KO mice, respectively, were pooled and incubated with 2 or 25 mM glucose in KRH buffer for 1 hour at 37°C after overnight regeneration. Islets were washed with BSA-free KRH buffer and processed, as described previously (30). Mature granules were determined by grayscale threshold analysis and counted with cellSense Dimensions Software. Immature granules, as well as docked granules, were identified and counted manually. Finally, 10 β-cells of three to five different islets per condition were analyzed. Mitochondrial area was determined with cellSense Dimensions Software after manual identification and labeling. Complementary DNA synthesis, reverse transcription polymerase chain reaction, and quantitative polymerase chain reaction RNA was isolated using RNeasy Mini Kit, and complementary DNA (cDNA) was synthesized using the GoScript™ Reverse Transcription Kit (Promega) with 500 ng RNA for quantitative polymerase chain reaction (qPCR) and 1 µg RNA for reverse transcription PCR (RT-PCR) with random hexanucleotide primers (Roche). For qPCR, several primers were used with the GoTaq® qPCR Master Mix on a StepOne Plus device (Applied Biosystems). RT-PCR for Tbc1d1 isoform detection was realized with primers flanking exons 12 and 13 of Tbc1d1. The forward primer is located in exon 11; the reverse primer is located in exon 15. Primer sequences are shown in Supplemental Tables 2 and 3. Analysis was performed using the 2-ΔΔCT) method (31), with 36b4 as a reference gene. For the determination of the relative copy number of messenger RNA (mRNA) in freshly isolated pancreatic islets, ΔCt values for Tbc1d1 and Tbc1d4 genes were normalized using a calibration curve obtained from the amplification of plasmids containing the respective cDNA sequences (pcDNA3-CMV-3xFLAG-Tbc1d1, 9319 bp; pCR2.1-TOPO-Tbc1d4, 7916 bp; see Supplemental Fig. 1). Western blot analysis We used standard protocols to prepare total protein extracts from isolated islets and frozen tissues. Total islet protein (10 to 30 µg) was separated in an 8% to 12% sodium dodecyl sulfate–polyacrylamide gel electrophoresis, and tank blotting of the proteins onto polyvinylidene difluoride membrane was performed. Membranes were blocked and incubated with primary antibodies and secondary horseradish peroxidase–conjugated antibodies (Supplemental Table 4). Proteins were detected with enhanced chemiluminescence reagent (PerkinElmer) in a ChemiDoc device (Bio-Rad, Munich, Germany) and quantified with the ImageLab software (Bio-Rad). Statistical analysis All experiments were performed with at least n = 3 samples and shown as mean values ± standard error of the mean (SEM). Statistical significance was calculated with appropriate tests using GraphPad Prism 5 software (GraphPad, La Jolla, CA). Exact conditions and tests are depicted in the figure legends. Results TBC1D1 is predominant over TBC1D4 in mouse islets Pancreatic islets from mice express both RabGAPs: Tbc1d1 and Tbc1d4. Real-time qPCR indicates that Tbc1d1 transcripts are predominant over Tbc1d4 in isolated pancreatic islets from mice (Fig. 1A and Supplemental Fig. 1A). As illustrated in Fig. 1B, islets exclusively contain mRNA for the short variant of Tbc1d1 [1162 amino acid (aa)], whereas skeletal muscle and the heart also contain the long Tbc1d1 isoform (1255 aa). As expected, TBC1D1 protein was undetectable in islets from D1KO mice (Fig. 1C and 1D). We generated transgenic mice overexpressing the short isoform of Tbc1d1 under the control of the RIP2 promoter (RIP2-D1). Islets from RIP2-D1 mice showed a 2.6-fold increase in TBC1D1 protein abundance compared with islets from RIP2-WT littermates (Fig. 1C and 1D). However, TBC1D4 protein was equally abundant in islets from D1KO and RIP2-D1 mice compared with their respective controls (Fig. 1C and 1D). Figure 1. View large Download slide RabGAP expression in isolated islets of D1KO and RIP2-D1 mice. (A) Normalized mRNA expression of Tbc1d1 and Tbc1d4 in freshly isolated C57BL/6J mouse islets. ΔCt values (with 36b4 as a housekeeping gene) were measured by real-time qPCR, corrected and normalized using standard concentration curves (Supplemental Fig. 1A). Data are means ± SEM (n = 8). Student t test, two-tailed, unpaired, ***P < 0.001. (B) RT-PCR of Tbc1d1 isoforms in heart (HE), quadriceps muscle (SM), and pancreatic islets (PI), as well as negative control (N). Positions of isoform-detecting primers are indicated by arrows. PCR of the long isoform produces a 645-bp fragment; the short isoform product corresponds to a 366-bp fragment as a result of the lack of exons 12 and 13. (C and D) Abundance of TBC1D1 and TBC1D4 in islets of D1KO, RIP2-D1 mice, and respective WT littermates. RabGAPs were normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Data represent means ± SEM (n = 4 to 8), Mann-Whitney U test, two-tailed, *P < 0.05; ***P < 0.001. Figure 1. View large Download slide RabGAP expression in isolated islets of D1KO and RIP2-D1 mice. (A) Normalized mRNA expression of Tbc1d1 and Tbc1d4 in freshly isolated C57BL/6J mouse islets. ΔCt values (with 36b4 as a housekeeping gene) were measured by real-time qPCR, corrected and normalized using standard concentration curves (Supplemental Fig. 1A). Data are means ± SEM (n = 8). Student t test, two-tailed, unpaired, ***P < 0.001. (B) RT-PCR of Tbc1d1 isoforms in heart (HE), quadriceps muscle (SM), and pancreatic islets (PI), as well as negative control (N). Positions of isoform-detecting primers are indicated by arrows. PCR of the long isoform produces a 645-bp fragment; the short isoform product corresponds to a 366-bp fragment as a result of the lack of exons 12 and 13. (C and D) Abundance of TBC1D1 and TBC1D4 in islets of D1KO, RIP2-D1 mice, and respective WT littermates. RabGAPs were normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Data represent means ± SEM (n = 4 to 8), Mann-Whitney U test, two-tailed, *P < 0.05; ***P < 0.001. D1KO affects glucose-induced plasma insulin levels in vivo To determine the impact of D1KO on plasma insulin concentrations in vivo, blood glucose and plasma insulin levels were determined acutely after an intraperitoneal injection of glucose. In line with previous studies (2, 12, 13), we did not observe any differences in blood glucose between D1KO and WT mice (Fig. 2A and 2D). Compared with the 6-hour fasting plasma insulin levels, D1KO exhibited an elevated increment in plasma insulin upon an acute glucose stimulus (Fig. 2B and 2E). However, absolute plasma insulin levels were similar between D1KO mice and WT mice during the test period (Fig. 2C). Notably, D1KO mice had significantly reduced plasma insulin levels after 6 hours of fasting, indicating increased insulin sensitivity in peripheral tissues in the D1KO mice. Figure 2. View largeDownload slide Intraperitoneal glucose tolerance test in D1KO mice compared with WT mice. Mice were fasted for 6 hours and injected 2 mg/kg body weight glucose intraperitoneally. (A) Glucose tolerance as a measure of glucose clearance from peripheral tail blood within 120 minutes after glucose injection. (B) Plasma insulin levels expressed as the percentage of the basal values of D1KO and WT mice, respectively. (C) Absolute plasma insulin levels expressed as nanograms per milliliter of D1KO and WT mice. (D) Area under curve (AUC) of the blood glucose levels as shown in (A). (E) AUC of the normalized plasma insulin levels as shown in (B). Data are means ± SEM (n = 12 to 14). Student t test, two-tailed, unpaired, *P < 0.05; **P < 0.01. Figure 2. View largeDownload slide Intraperitoneal glucose tolerance test in D1KO mice compared with WT mice. Mice were fasted for 6 hours and injected 2 mg/kg body weight glucose intraperitoneally. (A) Glucose tolerance as a measure of glucose clearance from peripheral tail blood within 120 minutes after glucose injection. (B) Plasma insulin levels expressed as the percentage of the basal values of D1KO and WT mice, respectively. (C) Absolute plasma insulin levels expressed as nanograms per milliliter of D1KO and WT mice. (D) Area under curve (AUC) of the blood glucose levels as shown in (A). (E) AUC of the normalized plasma insulin levels as shown in (B). Data are means ± SEM (n = 12 to 14). Student t test, two-tailed, unpaired, *P < 0.05; **P < 0.01. D1KO increases GSIS in isolated islets We investigated GSIS from isolated islets in both static and dynamic approaches. Upon stimulation with 25 mM glucose, islets from D1KO mice exhibited a substantially increased GSIS (≈1.5-fold) compared with islets from WT controls, whereas IS at basal glucose concentration (2 mM) was unchanged (Fig. 3A). Islets from RIP2-D1 mice showed no genotype-dependent differences IS (Fig. 3A). Measurements of insulin content in the islet lysates after GSIS revealed no genotype-specific differences (Supplemental Fig. 2A). The increased GSIS in D1KO islets was even sustained after three repetitive glucose challenges (Fig. 3B). IS was additionally determined by perifusion of freshly isolated islets. After 60 minutes at substimulatory glucose concentrations (5 mM), glucose was raised to 30 mM. When normalized to the last prestimulatory secretion rate (60 minutes), the increase in response to 30 mM glucose of the D1KO islets was markedly pronounced compared with WT islets, although the difference failed to achieve statistical significance (Fig. 3C). The ratio of total pancreatic proinsulin/insulin was significantly decreased in D1KO mice compared with WT controls (Fig. 3D), which is a result of similar insulin (Fig. 3E) but decreased proinsulin content (Fig. 3F) in pancreas from random-fed D1KO mice. The gene-expression levels of prohormone convertases 1 and 2 (Pcsk1 and Pcsk2) were significantly increased in D1KO islets (Fig. 3G). Figure 3. View largeDownload slide GSIS and pancreatic insulin and proinsulin content. (A) Isolated islets of D1KO, RIP2-D1, and respective WT mice were incubated with 2 and 25 mM glucose for 1 hour at 37°C, and secreted insulin in the supernatants and islet lysates were measured with ELISA. Data are means ± SEM (n = 7), two-way analysis of variance (ANOVA) with Bonferroni post hoc test, *P < 0.05. (B) Isolated islets of D1KO mice were incubated with 2 and 25 mM glucose for 1 hour at 37°C in three subsequent repetitions. Secreted insulin in the supernatants from the first basal and the last stimulated samples [25 mM (3)] and islet lysates were measured with ELISA. Data are means ± SEM (n = 5), two-way ANOVA with Bonferroni post hoc test, *P < 0.05. (C) Fifty freshly isolated islets per genotype were perifused in a custom-made perifusion chamber with 5 mM glucose and 30 mM glucose for the designated time points. Insulin from the eluate fractions was measured with ELISA, and the values are expressed as percentage of the secretion rate at the end of the basal glucose period (WT = 8.66 ± 3.80 pg ⋅ minutes−1 ⋅ islet−1; D1KO = 5.78 ± 2.52 pg ⋅ minutes−1 ⋅ islet−1). Data are means ± SEM (n = 5). Mann-Whitney U test, two-tailed of the area under the curve: P = 0.151. (D–F) Pancreata from D1KO and WT mice were homogenized in acid ethanol and insulin, and proinsulin was measured with ELISA. (D) Proinsulin-to-insulin ratio, as calculated by the (E) total insulin and (F) proinsulin content, normalized to protein content. Data are means ± SEM (n = 6). Welch-corrected t test, two-tailed, unpaired, *P < 0.05. (G) mRNA expression of insulin-processing prohormone convertases 1 and 2 (Pcsk1 and Pcsk2) from isolated islets of D1KO mice and WT controls. Data are means ± SEM (n = 4 to 7). Welch-corrected Student t test, two-tailed, unpaired, *P < 0.05. n.s., not significant. Figure 3. View largeDownload slide GSIS and pancreatic insulin and proinsulin content. (A) Isolated islets of D1KO, RIP2-D1, and respective WT mice were incubated with 2 and 25 mM glucose for 1 hour at 37°C, and secreted insulin in the supernatants and islet lysates were measured with ELISA. Data are means ± SEM (n = 7), two-way analysis of variance (ANOVA) with Bonferroni post hoc test, *P < 0.05. (B) Isolated islets of D1KO mice were incubated with 2 and 25 mM glucose for 1 hour at 37°C in three subsequent repetitions. Secreted insulin in the supernatants from the first basal and the last stimulated samples [25 mM (3)] and islet lysates were measured with ELISA. Data are means ± SEM (n = 5), two-way ANOVA with Bonferroni post hoc test, *P < 0.05. (C) Fifty freshly isolated islets per genotype were perifused in a custom-made perifusion chamber with 5 mM glucose and 30 mM glucose for the designated time points. Insulin from the eluate fractions was measured with ELISA, and the values are expressed as percentage of the secretion rate at the end of the basal glucose period (WT = 8.66 ± 3.80 pg ⋅ minutes−1 ⋅ islet−1; D1KO = 5.78 ± 2.52 pg ⋅ minutes−1 ⋅ islet−1). Data are means ± SEM (n = 5). Mann-Whitney U test, two-tailed of the area under the curve: P = 0.151. (D–F) Pancreata from D1KO and WT mice were homogenized in acid ethanol and insulin, and proinsulin was measured with ELISA. (D) Proinsulin-to-insulin ratio, as calculated by the (E) total insulin and (F) proinsulin content, normalized to protein content. Data are means ± SEM (n = 6). Welch-corrected t test, two-tailed, unpaired, *P < 0.05. (G) mRNA expression of insulin-processing prohormone convertases 1 and 2 (Pcsk1 and Pcsk2) from isolated islets of D1KO mice and WT controls. Data are means ± SEM (n = 4 to 7). Welch-corrected Student t test, two-tailed, unpaired, *P < 0.05. n.s., not significant. Pancreatic islets of D1KO mice show increased IS upon treatment with sulfonylureas or KCl We further investigated different steps in the triggering pathway of GSIS. GLUT2 abundance was unchanged between islets of WT and D1KO mice (Supplemental Fig. 2B). Pharmacological approaches revealed that adenosine triphosphate–sensitive potassium (K+-ATP) channel inhibition substantially increased basal IS in D1KO islets compared with WT controls. This finding was demonstrated by additional perifusion experiments, where addition of 500 µM tolbutamide to 5 mM glucose significantly increased IS in D1KO islets compared with WT controls (Fig. 4A). Additional stimulation with 40 mM KCl resulted in a generally increased IS, independent from the genotype (Fig. 4A). In a static GSIS experiment using 1 µM glibenclamide (Fig. 4B), the increased GSIS in D1KO islets after K+-ATP channel inhibition was confirmed. In contrast to the perifusion experiment, 30 mM KCl-induced IS in the static approach was significantly higher in islets from D1KO mice (Fig. 4C). In contrast, after incubation of isolated pancreatic islets with 5 µM Ca2+ ionophore A23187 (Fig. 4D) at 2 mM glucose, respectively, no genotype-dependent changes in IS were detected. The increased sulfonylurea-induced IS was not attributed to changes in total mRNA expression of the K+-ATP channel-forming subunits Kcnj11 (Kir6.2) and Abcc8 (SUR1), as they remained unchanged between the genotypes (Fig. 4E). Figure 4. View largeDownload slide Pharmacological stimulation of IS at basal glucose concentrations. (A) Fifty freshly isolated islets per genotype were perifused in a custom-made perifusion chamber with 5 mM glucose and stimulated with 500 µM tolbutamide (Tol) and 40 mM KCl for the designated time points. Insulin from the eluate fractions was measured with ELISA, and the values are expressed as percentage of the secretion rate at the end of the basal glucose period (WT = 8.52 ± 4.84 pg ⋅ minutes−1 ⋅ islet−1; D1KO = 4.73 ± 2.15 pg ⋅ minutes−1 ⋅ islet−1), with the area under the curve (AUC) between 60 and 90 minutes of the tolbutamide stimulation as inset. Data are means ± SEM (n = 5), Mann-Whitney U test, two-tailed of the AUC. (B–D) Isolated islets of D1KO mice were incubated with 2 mM glucose, (B) with or without 1 µM glibenclamide (solubilized in dimethyl sulfoxide), (C) 30 mM KCl, or (D) 5 µM calcium-ionophore A23187 (D) for 1 hour at 37°C, respectively. Secreted insulin in the supernatants and islet lysates was measured with ELISA. Data are means ± SEM (n = 5 to 7), two-way analysis of variance with Bonferroni post hoc test, **P < 0.01. (E) mRNA expression of K+-ATP channel–forming subunits Kcnj11 (Kir6.2) and Abcc8 (SUR1) from isolated islets of D1KO mice and WT controls. Data are means ± SEM (n = 8). Figure 4. View largeDownload slide Pharmacological stimulation of IS at basal glucose concentrations. (A) Fifty freshly isolated islets per genotype were perifused in a custom-made perifusion chamber with 5 mM glucose and stimulated with 500 µM tolbutamide (Tol) and 40 mM KCl for the designated time points. Insulin from the eluate fractions was measured with ELISA, and the values are expressed as percentage of the secretion rate at the end of the basal glucose period (WT = 8.52 ± 4.84 pg ⋅ minutes−1 ⋅ islet−1; D1KO = 4.73 ± 2.15 pg ⋅ minutes−1 ⋅ islet−1), with the area under the curve (AUC) between 60 and 90 minutes of the tolbutamide stimulation as inset. Data are means ± SEM (n = 5), Mann-Whitney U test, two-tailed of the AUC. (B–D) Isolated islets of D1KO mice were incubated with 2 mM glucose, (B) with or without 1 µM glibenclamide (solubilized in dimethyl sulfoxide), (C) 30 mM KCl, or (D) 5 µM calcium-ionophore A23187 (D) for 1 hour at 37°C, respectively. Secreted insulin in the supernatants and islet lysates was measured with ELISA. Data are means ± SEM (n = 5 to 7), two-way analysis of variance with Bonferroni post hoc test, **P < 0.01. (E) mRNA expression of K+-ATP channel–forming subunits Kcnj11 (Kir6.2) and Abcc8 (SUR1) from isolated islets of D1KO mice and WT controls. Data are means ± SEM (n = 8). D1KO β-cells exhibit more insulin granules after glucose stimulation We assessed morphometric parameters in hematoxylin/eosin-stained pancreatic sections of D1KO mice and WT littermates. D1KO did not affect overall islet morphology (Fig. 5A). Total pancreas area, total islet number, amount of islets per millimeter-squared pancreas, and the mean islet area per millimeter-squared pancreas were unchanged (Fig. 5B), as well as the islet area distribution within the pancreas (Supplemental Fig. 3A). Figure 5. View largeDownload slide Morphometric and ultrastructural analysis of islets from D1KO mice. (A) Pancreas from D1KO and WT mice was dissected, and sections of the formalin-fixed and paraffin-embedded pancreas were stained with hematoxylin/eosin. (B) Fourteen to 21 sections of different levels from three pancreata per genotype, respectively, were analyzed for pancreas area by point counting, total amount of islets, amount of islets per millimeter-squared pancreas, and percent of islet area per pancreas area. Data are means ± SEM (n = 3). (C) Representative TEM images of β-cells from isolated WT and D1KO islets after 2 or 25 mM glucose stimulation with insulin-containing granules. (D) Representative magnification of an electron microscopic picture showing mature insulin granules in solid circles and immature insulin granules in dashed circles. (E) Quantification of mature insulin granules per β-cell area in islets of WT and D1KO mice after 2 and 25 mM glucose stimulation. Mature insulin granules were identified and quantified by grayscale threshold analysis. Data are means ± SEM of 10 β-cells from four to five different islets per condition, respectively. Two-way analysis of variance (ANOVA) with Bonferroni post hoc test, *P < 0.05; Student t test, two-tailed, unpaired, ###P < 0.001. (F) Immature insulin granules in β-cells of islets after 2 and 25 mM glucose stimulation were identified and counted manually. Data are means ± SEM of 10 β-cells from four to five different islets per condition, respectively. Student t test, two-tailed, unpaired, #P < 0.05; ###P < 0.001. (G) Docked insulin granules in β-cells of islets after 2 and 25 mM glucose stimulation were identified and counted manually. Granules with a distance ≤400 nm from the plasma membrane were considered as docked granules. Data are means ± SEM of 10 β-cells from four to five different islets per condition, respectively. Two-way ANOVA with Bonferroni post hoc test, **P < 0.01; Student t test, two-tailed, unpaired, ###P < 0.001. Figure 5. View largeDownload slide Morphometric and ultrastructural analysis of islets from D1KO mice. (A) Pancreas from D1KO and WT mice was dissected, and sections of the formalin-fixed and paraffin-embedded pancreas were stained with hematoxylin/eosin. (B) Fourteen to 21 sections of different levels from three pancreata per genotype, respectively, were analyzed for pancreas area by point counting, total amount of islets, amount of islets per millimeter-squared pancreas, and percent of islet area per pancreas area. Data are means ± SEM (n = 3). (C) Representative TEM images of β-cells from isolated WT and D1KO islets after 2 or 25 mM glucose stimulation with insulin-containing granules. (D) Representative magnification of an electron microscopic picture showing mature insulin granules in solid circles and immature insulin granules in dashed circles. (E) Quantification of mature insulin granules per β-cell area in islets of WT and D1KO mice after 2 and 25 mM glucose stimulation. Mature insulin granules were identified and quantified by grayscale threshold analysis. Data are means ± SEM of 10 β-cells from four to five different islets per condition, respectively. Two-way analysis of variance (ANOVA) with Bonferroni post hoc test, *P < 0.05; Student t test, two-tailed, unpaired, ###P < 0.001. (F) Immature insulin granules in β-cells of islets after 2 and 25 mM glucose stimulation were identified and counted manually. Data are means ± SEM of 10 β-cells from four to five different islets per condition, respectively. Student t test, two-tailed, unpaired, #P < 0.05; ###P < 0.001. (G) Docked insulin granules in β-cells of islets after 2 and 25 mM glucose stimulation were identified and counted manually. Granules with a distance ≤400 nm from the plasma membrane were considered as docked granules. Data are means ± SEM of 10 β-cells from four to five different islets per condition, respectively. Two-way ANOVA with Bonferroni post hoc test, **P < 0.01; Student t test, two-tailed, unpaired, ###P < 0.001. We further analyzed the ultrastructure of β-cells from D1KO mice and WT controls with regard to the amount and distribution of mature and immature insulin granules by transmission electron microscopy (TEM). In accordance with the ex vivo static GSIS measurements, isolated islets were incubated at 2 and 25 mM glucose for 1 hour (Fig. 5C and 5D) before TEM analysis. The density of mature granules was equal in β-cells from both genotypes at 2 mM glucose. However, after glucose stimulation, the density of mature granules decreased significantly in β-cells of WT but not D1KO β-cells. This led to a significantly increased, mature granule density in β-cells from D1KO islets after 25 mM glucose stimulation (Fig. 5E). The density of immature insulin granules was significantly decreased in D1KO β-cells after stimulation with 2 mM glucose but markedly increased in the high-glucose state. Consequently, D1KO β-cells demonstrated a highly significant increase of immature granules upon glucose stimulation, whereas the number of immature granules remained stable in WT β-cells (Fig. 5F). Likewise, the total density of granules, 400 nm beneath the plasma membrane (docked granules), was increased in β-cells of D1KO islets compared with WT after 25 mM glucose stimulation (Fig. 5G). Additional expression analysis of various genes relevant for islet physiology and vesicle dynamics showed no meaningful differences between D1KO and WT islets (Supplemental Fig. 4). D1KO increases palmitate uptake and oxidation in isolated islets In previous studies, D1KO mice exhibited enhanced lipid use in vivo, as well as increased uptake and oxidation of [3H]palmitate in isolated skeletal muscles (6). Therefore, we analyzed lipid handling in isolated islets of D1KO mice and WT controls. In line with previous results from skeletal muscle, islets of D1KO mice showed a substantially (40%) increased palmitate uptake (Fig. 6A) and a concomitant increase in palmitate oxidation (Fig. 6B). Of note, mRNA expression of genes involved in fatty acid signaling, uptake, and oxidation was unchanged in isolated islets from both genotypes (Supplemental Fig. 5A). However, D1KO islets demonstrated a substantially elevated mtDNA copy number compared with WT islets (Fig. 6C). In addition, expression of the mitochondrial fission gene Dnm1l (Drp1) was significantly increased in D1KO, whereas the expression of Opa1 that mediates mitochondrial fusion was unchanged (Fig. 6D). Nevertheless, we could not find a differential citrate synthase activity (Fig. 6E), alterations in the mitochondrial area in single β-cells (Fig. 6F), or a differential protein abundance of oxidative phosphorylation (OXPHOS) complexes between D1KO and WT islets (Fig. 6G). Seahorse measurements for OCRs showed a small but significant increase in baseline oxygen consumption in D1KO islets after baseline normalization. Injection of 5 µM oligomycin reduced OCRs by ∼40% in WT and 25% in D1KO islets compared with baseline OCRs. Remaining OCRs, after injection of oligomycin, were not different between D1KO and WT islets (P = 0.266). Injection of a 5-µM antimycin A/rotenone mixture reduced OCRs by ∼80% compared with baseline OCRs in both WT and D1KO islets (Fig. 6H). Figure 6. View largeDownload slide Lipid metabolism and mitochondrial function in isolated islets of D1KO mice. (A) Isolated islets of D1KO and WT mice were incubated with [14C]palmitic acid for 20 minutes at 37°C. Accumulated [14C]palmitic acid was measured by scintillation counting in the islet lysates, and cost per thousand impression (CPM) values normalized to islet numbers were expressed as fold over WT. Data are means ± SEM (n = 6), Mann-Whitney U test, two-tailed, **P < 0.01. (B) Isolated islets of D1KO and WT mice were incubated with [14C]palmitic acid for 4 hours at 37°C. The produced 14C-CO2 from oxidized palmitic acid was trapped in filter papers and measured by scintillation counting, and CPM values normalized to islet numbers were expressed as fold over WT. Data are means ± SEM (n = 6), Mann-Whitney U test, two-tailed, *P < 0.05. (C) Mitochondrial copy number. The ratio of mtDNA (mt-Nd2) to nuclear DNA (Rps18) was determined by real-time qPCR using SYBR Green with 40 ng DNA from isolated islets. Data are means ± SEM (n = 5 to 7), Student t test, two-tailed, unpaired, ***P < 0.001. (D) mRNA expression of Dnm1l and Opa1 in isolated islets. Data are means ± SEM (n = 7), Student t test, two-tailed, unpaired, *P < 0.05. (E) Citrate synthase activity was analyzed from isolated islets using 8 µg total protein with the Citrate Synthase Assay Kit (Sigma, Steinheim, Germany), according to the manufacturer’s instructions. Data are means ± SEM (n = 7 to 8). (F) Mitochondria of β-cells were analyzed from a small subset of TEM images (Fig. 5). The sum of the area of all mitochondria per β-cell was referred to the cell area and expressed as percentage. Data are means ± SEM of 10 β-cells per genotype. (G) Quantification with representative Western blot showing abundance of OXPHOS complexes (C)II, -III, and -V of isolated D1KO and WT islets against glyceraldehyde 3-phosphate dehydrogenase abundance. Data are means ± SEM (n = 3 to 4). (H) Measurement of OCR in isolated islets with the Seahorse XF24e analyzer. Seventy islets were seeded per well of a 24-well islet capture plate in duplicates, and baseline OCR was initially measured four times within 30 minutes. After addition of 5 µM oligomycin (Oligo) or 5 µM antimycin A/rotenone (AA/Rot) mixture, five measurements within 40 minutes were taken. After baseline correction, as recommended by Agilent and previous reports (27), the last baseline measure and the lowest of the five OCR measures of both oligomycin and antimycin A/rotenone were used for calculation. Data are means ± SEM (n = 4), Student t test, two-tailed, unpaired, *P < 0.05. Figure 6. View largeDownload slide Lipid metabolism and mitochondrial function in isolated islets of D1KO mice. (A) Isolated islets of D1KO and WT mice were incubated with [14C]palmitic acid for 20 minutes at 37°C. Accumulated [14C]palmitic acid was measured by scintillation counting in the islet lysates, and cost per thousand impression (CPM) values normalized to islet numbers were expressed as fold over WT. Data are means ± SEM (n = 6), Mann-Whitney U test, two-tailed, **P < 0.01. (B) Isolated islets of D1KO and WT mice were incubated with [14C]palmitic acid for 4 hours at 37°C. The produced 14C-CO2 from oxidized palmitic acid was trapped in filter papers and measured by scintillation counting, and CPM values normalized to islet numbers were expressed as fold over WT. Data are means ± SEM (n = 6), Mann-Whitney U test, two-tailed, *P < 0.05. (C) Mitochondrial copy number. The ratio of mtDNA (mt-Nd2) to nuclear DNA (Rps18) was determined by real-time qPCR using SYBR Green with 40 ng DNA from isolated islets. Data are means ± SEM (n = 5 to 7), Student t test, two-tailed, unpaired, ***P < 0.001. (D) mRNA expression of Dnm1l and Opa1 in isolated islets. Data are means ± SEM (n = 7), Student t test, two-tailed, unpaired, *P < 0.05. (E) Citrate synthase activity was analyzed from isolated islets using 8 µg total protein with the Citrate Synthase Assay Kit (Sigma, Steinheim, Germany), according to the manufacturer’s instructions. Data are means ± SEM (n = 7 to 8). (F) Mitochondria of β-cells were analyzed from a small subset of TEM images (Fig. 5). The sum of the area of all mitochondria per β-cell was referred to the cell area and expressed as percentage. Data are means ± SEM of 10 β-cells per genotype. (G) Quantification with representative Western blot showing abundance of OXPHOS complexes (C)II, -III, and -V of isolated D1KO and WT islets against glyceraldehyde 3-phosphate dehydrogenase abundance. Data are means ± SEM (n = 3 to 4). (H) Measurement of OCR in isolated islets with the Seahorse XF24e analyzer. Seventy islets were seeded per well of a 24-well islet capture plate in duplicates, and baseline OCR was initially measured four times within 30 minutes. After addition of 5 µM oligomycin (Oligo) or 5 µM antimycin A/rotenone (AA/Rot) mixture, five measurements within 40 minutes were taken. After baseline correction, as recommended by Agilent and previous reports (27), the last baseline measure and the lowest of the five OCR measures of both oligomycin and antimycin A/rotenone were used for calculation. Data are means ± SEM (n = 4), Student t test, two-tailed, unpaired, *P < 0.05. Discussion In this study, we investigated the role of the RabGAP TBC1D1 in mouse pancreatic islets. RabGAPs have been described to be key players of energy substrate use in skeletal muscle, but their role in other tissues remains rather unknown. In our study, we demonstrate that TBC1D1 is an important regulator of IS and lipid metabolism in mouse pancreatic islets by controlling glucose- and secretagogue-stimulated IS and lipid oxidation. We showed that expression of Tbc1d1 highly dominates over Tbc1d4 in isolated islets, emphasizing the bigger role for TBC1D1 in islet physiology. Furthermore, our study represents the description of a short isoform of Tbc1d1 in islets and the generation of the transgenic RIP2-D1 mice carrying this short isoform. In recent studies, knockdown of Tbc1d1 and Tbc1d4 in pancreatic β-cells has been shown to affect GSIS, but the effects were not conclusive: in sorted rat β-cells and in a global rat knockout model, D1KO led to a moderate increase in both basal IS and GSIS (15, 16). In contrast, knockdown of the related Tbc1d4 in primary mouse β-cells resulted in increased basal IS but decreased GSIS (14), suggesting an opposing role of both RabGAPs in GSIS. In the current study, we first identified Tbc1d1 as the predominant RabGAP in mouse islets compared with Tbc1d4. The short Tbc1d1 isoform, which is also abundant in adipocytes but not in muscle (1), lacks 94 aa (631–724) containing two 5-aminoimidazole-4-carboxamide ribonucleotide–regulated phosphorylation sites: Ser-660 and Ser-700 (1, 32). Whereas the specific functions of TBC1D1 isoforms are unknown, previous studies of TBC1D4 demonstrated altered biological activities of isoforms and isoform-specific mutations (33, 34). Consistent with data from rats (15), we found a relatively higher increase of IS in Tbc1d1 knockout islets than in islets from WT mice. Furthermore, TBC1D1 protein appears not to be rate limiting in islets, as β-cell-specific overexpression of Tbc1d1 neither changes basal IS nor GSIS compared with WT controls. The divergence from previous studies in rat β-cells (15, 16) might be most likely explained by a species-related diversity. In fact, D1KO rats exhibited no changes in GLUT4 abundance in skeletal muscle, which is considered a hallmark of all published D1KO mice so far (16). Importantly, increased GSIS in D1KO islets persists after three repetitive glucose challenges, indicating that the mechanism leading to increased GSIS upon D1KO is operating continuously and thus, reflects a physiologically relevant context. This is also supported by the markedly amplified first and second phase of IS in D1KO islets, demonstrating that both the triggering and amplifying pathways are enhanced upon D1KO. The higher degree of interindividual variation of IS in perifusion assays compared with the static GSIS assay may result from differences in islet handling and preincubation times but applied to both genotypes. The increase in both glibenclamide- and tolbutamide-stimulated IS at basal glucose concentrations (2 and 5 mM) was markedly potentiated in islets of D1KO mice. This effect is unlikely a result of altered adenosine triphosphate (ATP)/adenosine 5′-diphosphate ratios in β-cells but rather reflects enhanced K+-ATP channel signaling. Interestingly, the K+-ATP channels have been reported to localize not only to the plasma membrane of β-cells but are also present in considerable amounts on insulin granules (35). Moreover, sulfonylureas were found to stimulate IS via granule-localized channel inhibition without binding to plasma membrane K+-ATP channels (36) and can mediate the fusion of newly generated granules for IS (37). It is possible that D1KO increases K+-ATP channel trafficking and surface expression at basal glucose concentrations (38, 39). However, we found no relevant changes in total mRNA expression of the K+-ATP channel–forming subunits in islets. A previous study reported that Tbc1d1 knockdown prevents IS induced by KCl in sorted rat β-cells (15). In our study, we show that 60 minutes of 30 mM KCl stimulation results in increased IS in the D1KO islets. This discrepancy might be attributed to differences in the experimental protocol and the fact that Tbc1d1 knockdown in rat β-cells led to only 70% reduction of TBC1D1 protein (15). Together with the sulfonylurea-induced increase in IS, these findings suggest a role for TBC1D1 on cell-membrane depolarization. This is supported by the fact that KCl-induced IS in the perifusion approach was not different between the genotypes, as applied in addition to the already-present tolbutamide. Furthermore, with the bypassing of the downstream Ca2+ channel with the Ca2+ ionophore A23187, IS was the same in both WT and D1KO islets. Stimulation of IS with this ionophore is quite low compared with the stimulation with the other secretagogues. However, the stimulation fold with 5 µM A23187 is comparable with what was already found with INS-1 cells (40). Taken together, these data suggest that the extent of the cell-membrane depolarization might be one explanation for the differences in IS observed upon D1KO. Our results indicate that both synthesis of newly generated insulin vesicles and exocytosis of existing granules are increased by D1KO (41–44). Consistently, our ultrastructure analysis strongly suggests that β-cells of D1KO islets are capable of faster restoring of insulin granules after glucose stimulation to increase secretion in a time-dependent manner. This is also reflected by the dynamic perifusion experiment with 30 mM glucose. The measurements of proinsulin and insulin in the total pancreas also reflect the results of the insulin-granule analysis with the consideration of the basal 2 mM state. As immature granules may still contain more proinsulin than mature granules (45), the lower amount of proinsulin in the D1KO pancreas reflects the lower density of immature granules in D1KO β-cells at 2 mM glucose and therefore, the lower proinsulin-to-insulin ratio. In accordance, islets from D1KO mice show increased gene-expression levels of Pcsk1 and Pcsk2, suggesting a higher activity in the proinsulin-to-insulin processing compared with WT islets. Apart from proinsulin, total pancreatic insulin content, as well as insulin content of islet lysates at the end of the stimulation period from the GSIS, showed no differences between WT and D1KO mice. Whereas this observation is in line with our results from the morphometric analysis that showed no differences in islet size, density, and morphology, genotype-dependent alterations in insulin content per granule might be resolved in more detailed analyses. Several genes, related to islet cell identity, insulin exocytosis, and insulin signaling, were analyzed. Among these genes, only Mafa, Dnajc5, and Irs2 showed a differential mRNA expression in D1KO islets compared with WT controls. However, for all three genes, the respective protein abundance was found to be unchanged between WT and D1KO islets. In line with previous studies, D1KO mice showed normal glucose tolerance (2, 12, 13). However, we observed lower fasting insulin levels, implicating moderately elevated hepatic insulin sensitivity (46). Moreover, compared with WT, D1KO mice secreted relatively more insulin in response to the glucose bolus to achieve normoglycemia, which may reflect impaired glucose clearance as a result of reduced insulin-stimulated glucose transport into skeletal muscle. Measurements of plasma insulin after an intraperitoneal glucose injection may not exactly reflect in vivo IS. Further in-depths analyses, e.g., with the use of hyperglycemic clamp techniques, would be necessary to better characterize the impact of TBC1D1 on in vivo IS. Whereas we cannot rule out that the observed increase in GSIS might be a compensatory reaction to maintain clearance of glucose from the blood in D1KO mice, the data provide evidence for a direct role of TBC1D1 in vesicle dynamics in islets. Skeletal muscles from D1KO mice exhibit increased uptake and oxidation of long-chain fatty acids (6). Fatty acid metabolism is likely also to influence β-cell function and IS via the amplifying pathway (19, 47). In addition, ectopic accumulation of lipids is a well-known precursor in the development of insulin resistance (48, 49). Isolated pancreatic islets from D1KO mice displayed an increase in both uptake and oxidation of [14C]palmitic acid, which might reflect a more generalized role of TBC1D1 in substrate use. Although this effect seems not to be mediated by increased expression of genes regulating lipid metabolism, it might be a result of an enhanced mitochondrial function. Interestingly, the mitochondrial copy number was significantly increased in D1KO islets, which was in accordance with higher mRNA levels of the Dnm1l, a factor for mitochondrial fission processes. In contrast, no changes in mitochondrial fusion, citrate synthase activity, or mitochondrial area were observed between the genotypes. Of note, previous results in skeletal muscle showed an increased citrate synthase activity without a change in Nd2 amplification in D1KO mice (2). A recent report indicates an increased adenosine 5′-diphosphate-dependent mitochondrial respiration in skeletal muscle of D1KO rats in combination with increased fatty acid oxidation as well (50). However, abundance of OXPHOS complexes in the rat muscle was unchanged. This is in line with our results, where the abundance of OXPHOS complexes is also unaffected as a result of the D1KO. Basal oxygen consumption, as measured using the Seahorse XF24e analyzer, was increased in D1KO islets at baseline but was unchanged after addition of the oligomycin or antimycin A/rotenone mixture. As citrate synthase activity is unchanged, it is possible that anaplerotic reactions or the glycerol-3-phosphate dehydrogenase shuttle of the reduced form of nicotinamide adenine dinucleotide (51) contribute to the increased basal respiration without affecting the Krebs cycle. The increased expression of Dnm1l could also indicate an increased fission of peroxisomes (52). Thus, we cannot rule out that the increase in palmitate oxidation is a result of a bigger proportion of peroxisomal lipid oxidation in D1KO islets. Moreover, D1KO may not only affect mitochondrial function in β-cells but also in other islet cell types, also contributing to the increased palmitate oxidation. As fatty acid oxidation leads to a rise in cellular ATP levels, this TBC1D1-dependent process might also induce IS via the triggering pathway. These findings strengthen the hypothesis of a protective mechanism with the aim to maintain IS, despite a progressive insulin resistance in peripheral tissues by downregulation of TBC1D1 abundance or activity. In conclusion, we present evidence for a role of TBC1D1 as regulator of IS and lipid metabolism in pancreatic β-cells by enhanced lipid flux and glucose- and sulfonylurea-induced IS in the absence of TBC1D1. Appendix. Antibody Table Peptide/Protein Target  Antigen Sequence (if Known)  Name of Antibody  Manufacturer, Catalog No.  Species Raised in; Monoclonal or Polyclonal  Dilution Used  RRID  MafA  aa 300–359 of Mouse MafA  MafA antibody  Bethyl Laboratories, A300-611A  Rabbit; polyclonal  1:500  AB_2297116  IRS-2  Total IRS-2 protein  IRS-2 antibody  Cell Signaling, 4502  Rabbit; polyclonal  1:1000  AB_2125774  TBC1D1  Around V796 of mouse TBC1D1  TBC1D1 (V796)  Cell Signaling, 4629  Rabbit; polyclonal  1:1000  AB_1904162  TBC1D4  aa 1178–1189 of Rat AS160  Anti-AS160  Millipore, 07-741  Rabbit; polyclonal  1:1000  AB_492639  α-Tubulin  —  Monoclonal anti-α-tubulin  Calbiochem, T 6074  Mouse; monoclonal  1:1000  AB_477582  CSPα  aa 182–198 of Rat CSP  Anti-CSP  Synaptic Systems, 154 003  Rabbit; polyclonal  1:1000  AB_887710  GLUT2  aa 32–98 of Human GLUT2  GLUT2 (H-67)  Santa Cruz, sc-9117  Rabbit; polyclonal  1:1000  AB_641068  GAPDH  —  Anti-GAPDH  Ambion, AM4300  Mouse; monoclonal  1:4000  AB_437392  GAPDH  Near carboxy terminus of human GAPDH  Anti-GAPDH (14C10)  Cell Signaling, 2118  Rabbit; monoclonal  1:4000  AB_561053  OXPHOS    Total OXPHOS rodent WB antibody cocktail  Abcam, ab110413  Mouse; monoclonal  1:250  AB_2629281  Secondary HRP-conjugated anti-rabbit IgG    Goat anti-rabbit IgG (H + L)  Dianova, 111-035-003  Goat  1:20,000  AB_2313567  Secondary HRP-conjugated anti-mouse IgG    Rabbit anti-mouse IgG, Fγ  Dianova, 315-035-008  Rabbit  1:20,000  AB_2340063  Peptide/Protein Target  Antigen Sequence (if Known)  Name of Antibody  Manufacturer, Catalog No.  Species Raised in; Monoclonal or Polyclonal  Dilution Used  RRID  MafA  aa 300–359 of Mouse MafA  MafA antibody  Bethyl Laboratories, A300-611A  Rabbit; polyclonal  1:500  AB_2297116  IRS-2  Total IRS-2 protein  IRS-2 antibody  Cell Signaling, 4502  Rabbit; polyclonal  1:1000  AB_2125774  TBC1D1  Around V796 of mouse TBC1D1  TBC1D1 (V796)  Cell Signaling, 4629  Rabbit; polyclonal  1:1000  AB_1904162  TBC1D4  aa 1178–1189 of Rat AS160  Anti-AS160  Millipore, 07-741  Rabbit; polyclonal  1:1000  AB_492639  α-Tubulin  —  Monoclonal anti-α-tubulin  Calbiochem, T 6074  Mouse; monoclonal  1:1000  AB_477582  CSPα  aa 182–198 of Rat CSP  Anti-CSP  Synaptic Systems, 154 003  Rabbit; polyclonal  1:1000  AB_887710  GLUT2  aa 32–98 of Human GLUT2  GLUT2 (H-67)  Santa Cruz, sc-9117  Rabbit; polyclonal  1:1000  AB_641068  GAPDH  —  Anti-GAPDH  Ambion, AM4300  Mouse; monoclonal  1:4000  AB_437392  GAPDH  Near carboxy terminus of human GAPDH  Anti-GAPDH (14C10)  Cell Signaling, 2118  Rabbit; monoclonal  1:4000  AB_561053  OXPHOS    Total OXPHOS rodent WB antibody cocktail  Abcam, ab110413  Mouse; monoclonal  1:250  AB_2629281  Secondary HRP-conjugated anti-rabbit IgG    Goat anti-rabbit IgG (H + L)  Dianova, 111-035-003  Goat  1:20,000  AB_2313567  Secondary HRP-conjugated anti-mouse IgG    Rabbit anti-mouse IgG, Fγ  Dianova, 315-035-008  Rabbit  1:20,000  AB_2340063  Abbreviations: GAPDH, glyceraldehyde 3-phosphate dehydrogenase; H + L, heavy chain and light chain; HRP, horseradish peroxidase; IgG, immunoglobulin G; RRID, Research Resource Identifier. View Large Abbreviations: Abbreviations: aa amino acid ATP adenosine triphosphate BSA bovine serum albumin cDNA complementary DNA D1KO Tbc1d1-deficient ELISA enzyme-linked immunosorbent assay GLUT glucose transporter GSIS glucose-stimulated insulin secretion IS insulin secretion K+-ATP adenosine triphosphate–sensitive potassium KRH Krebs-Ringer-HEPES mRNA messenger RNA mtDNA mitochondrial DNA OCR oxygen consumption rate OXPHOS oxidative phosphorylation Pcsk1/2 prohormone convertase 1/2 qPCR quantitative polymerase chain reaction RabGAP Rab guanosine triphosphatase–activating protein RIP2-D1 TG-RIP2-3xFLAG-Tbc1d1 RT-PCR reverse transcription polymerase chain reaction SEM standard error of the mean TEM transmission electron microscopy WT wild-type Acknowledgments We thank Angelika Horrighs, Anette Kurowski, Barbara Bartosinska, Jennifer Schwettmann, Carolin Borchert, Ines Grüner, Verena Lier-Glaubitz, and Dr. Bengt-Frederik Belgardt for expert technical assistance, support, and consultation. Financial Support: This work was supported by the German Center for Diabetes Research of the Federal Ministry for Education and Research and the Ministry of Science and Research of the State North Rhine-Westphalia. T. Stermann received a stipend by the research training group “vivid” of the Heinrich-Heine University Düsseldorf. Additional project funding came from the German Diabetes Association (DDG) and the German Academic Exchange Service (DAAD). Author Contributions: T. Stermann, A.C., and H.A.-H. wrote the manuscript and analyzed and interpreted the data. T. Stermann, F.M., C.W., K.J., D.A., T.B., A.P., D.M.O, G.H.T., C.d.W., S.L. and T. Schallschmidt performed the experiments and analyzed data. J.W., F.B., I.R., M.K., E.L., A.C., and H.A.-H. were involved in the study design and contributed to data interpretation. Disclosure Summary: The authors have nothing to disclose. References 1. Roach WG, Chavez JA, Mîinea CP, Lienhard GE. Substrate specificity and effect on GLUT4 translocation of the Rab GTPase-activating protein Tbc1d1. Biochem J . 2007; 403( 2): 353– 358. Google Scholar CrossRef Search ADS PubMed  2. Chadt A, Immisch A, de Wendt C, Springer C, Zhou Z, Stermann T, Holman GD, Loffing-Cueni D, Loffing J, Joost HG, Al-Hasani H. “Deletion of both Rab-GTPase–activating proteins TBC1D1 and TBC1D4 in mice eliminates insulin- and AICAR-stimulated glucose transport [ published erratum appears in Diabetes. 2015:64(4):1492]. Diabetes . 2015; 64( 3): 746– 759. Google Scholar CrossRef Search ADS PubMed  3. Mîinea CP, Sano H, Kane S, Sano E, Fukuda M, Peränen J, Lane WS, Lienhard GE. AS160, the Akt substrate regulating GLUT4 translocation, has a functional Rab GTPase-activating protein domain. Biochem J . 2005; 391( 1): 87– 93. Google Scholar CrossRef Search ADS PubMed  4. Stenmark H. Rab GTPases as coordinators of vesicle traffic. Nat Rev Mol Cell Biol . 2009; 10( 8): 513– 525. Google Scholar CrossRef Search ADS PubMed  5. Dokas J, Chadt A, Nolden T, Himmelbauer H, Zierath JR, Joost HG, Al-Hasani H. Conventional knockout of Tbc1d1 in mice impairs insulin- and AICAR-stimulated glucose uptake in skeletal muscle. Endocrinology . 2013; 154( 10): 3502– 3514. Google Scholar CrossRef Search ADS PubMed  6. Chadt A, Leicht K, Deshmukh A, Jiang LQ, Scherneck S, Bernhardt U, Dreja T, Vogel H, Schmolz K, Kluge R, Zierath JR, Hultschig C, Hoeben RC, Schürmann A, Joost HG, Al-Hasani H. Tbc1d1 mutation in lean mouse strain confers leanness and protects from diet-induced obesity. Nat Genet . 2008; 40( 11): 1354– 1359. Google Scholar CrossRef Search ADS PubMed  7. Stone S, Abkevich V, Russell DL, Riley R, Timms K, Tran T, Trem D, Frank D, Jammulapati S, Neff CD, Iliev D, Gress R, He G, Frech GC, Adams TD, Skolnick MH, Lanchbury JS, Gutin A, Hunt SC, Shattuck D. TBC1D1 is a candidate for a severe obesity gene and evidence for a gene/gene interaction in obesity predisposition. Hum Mol Genet . 2006; 15( 18): 2709– 2720. Google Scholar CrossRef Search ADS PubMed  8. Dash S, Sano H, Rochford JJ, Semple RK, Yeo G, Hyden CS, Soos MA, Clark J, Rodin A, Langenberg C, Druet C, Fawcett KA, Tung YC, Wareham NJ, Barroso I, Lienhard GE, O’Rahilly S, Savage DB. A truncation mutation in TBC1D4 in a family with acanthosis nigricans and postprandial hyperinsulinemia. Proc Natl Acad Sci USA . 2009; 106( 23): 9350– 9355. Google Scholar CrossRef Search ADS PubMed  9. Dash S, Langenberg C, Fawcett KA, Semple RK, Romeo S, Sharp S, Sano H, Lienhard GE, Rochford JJ, Howlett T, Massoud AF, Hindmarsh P, Howell SJ, Wilkinson RJ, Lyssenko V, Groop L, Baroni MG, Barroso I, Wareham NJ, O’Rahilly S, Savage DB. Analysis of TBC1D4 in patients with severe insulin resistance. Diabetologia . 2010; 53( 6): 1239– 1242. Google Scholar CrossRef Search ADS PubMed  10. Meyre D, Farge M, Lecoeur C, Proenca C, Durand E, Allegaert F, Tichet J, Marre M, Balkau B, Weill J, Delplanque J, Froguel P. R125W coding variant in TBC1D1 confers risk for familial obesity and contributes to linkage on chromosome 4p14 in the French population. Hum Mol Genet . 2008; 17( 12): 1798– 1802. Google Scholar CrossRef Search ADS PubMed  11. Abou-Sabe’ MA. Stimulation of cAMP synthesis by glucose in a mutant of E. coli B-r. Nat New Biol . 1973; 243( 127): 182– 185. Google Scholar CrossRef Search ADS PubMed  12. Hargett SR, Walker NN, Keller SR. Rab GAPs AS160 and Tbc1d1 play non-redundant roles in the regulation of glucose and energy homeostasis in mice. Am J Physiol Endocrinol Metab . 2016: 310( 4): E276– E288. Google Scholar CrossRef Search ADS PubMed  13. Szekeres F, Chadt A, Tom RZ, Deshmukh AS, Chibalin AV, Björnholm M, Al-Hasani H, Zierath JR. The Rab-GTPase-activating protein TBC1D1 regulates skeletal muscle glucose metabolism. Am J Physiol Endocrinol Metab . 2012; 303( 4): E524– E533. Google Scholar CrossRef Search ADS PubMed  14. Bouzakri K, Ribaux P, Tomas A, Parnaud G, Rickenbach K, Halban PA. Rab GTPase-activating protein AS160 is a major downstream effector of protein kinase B/Akt signaling in pancreatic beta-cells. Diabetes . 2008; 57( 5): 1195– 1204. Google Scholar CrossRef Search ADS PubMed  15. Rütti S, Arous C, Nica AC, Kanzaki M, Halban PA, Bouzakri K. Expression, phosphorylation and function of the Rab-GTPase activating protein TBC1D1 in pancreatic beta-cells. FEBS Lett . 2014; 588( 1): 15– 20. Google Scholar CrossRef Search ADS PubMed  16. Paglialunga S, Simnett G, Robson H, Hoang M, Pillai RA, Arkell AM, Simpson JA, Bonen A, Huising M, Joseph JW, Holloway GP. The Rab-GTPase activating protein, TBC1D1, is critical for maintaining normal glucose homeostasis and beta-cell mass. Appl Physiol Nutr Metab . 2017; 42( 6): 647– 655. Google Scholar CrossRef Search ADS PubMed  17. Cerasi E. Potentiation of insulin release by glucose in man. Acta Endocrinol (Copenh) . 1975; 79( 3): 511– 534. Google Scholar PubMed  18. Cerasi E. Insulin secretion: mechanism of the stimulation by glucose. Q Rev Biophys . 1975; 8( 1): 1– 41. Google Scholar CrossRef Search ADS PubMed  19. Henquin JC. Triggering and amplifying pathways of regulation of insulin secretion by glucose. Diabetes . 2000; 49( 11): 1751– 1760. Google Scholar CrossRef Search ADS PubMed  20. Panten U, Willenborg M, Schumacher K, Hamada A, Ghaly H, Rustenbeck I. Acute metabolic amplification of insulin secretion in mouse islets is mediated by mitochondrial export of metabolites, but not by mitochondrial energy generation. Metabolism . 2013; 62( 10): 1375– 1386. Google Scholar CrossRef Search ADS PubMed  21. Henquin JC. The dual control of insulin secretion by glucose involves triggering and amplifying pathways in β-cells. Diabetes Res Clin Pract . 2011; 93( Suppl 1): S27– S31. Google Scholar CrossRef Search ADS PubMed  22. Valera A, Pujol A, Pelegrin M, Bosch F. Transgenic mice overexpressing phosphoenolpyruvate carboxykinase develop non-insulin-dependent diabetes mellitus. Proc Natl Acad Sci USA . 1994; 91( 19): 9151– 9154. Google Scholar CrossRef Search ADS PubMed  23. Dor Y, Brown J, Martinez OI, Melton DA. Adult pancreatic beta-cells are formed by self-duplication rather than stem-cell differentiation. Nature . 2004; 429( 6987): 41– 46. Google Scholar CrossRef Search ADS PubMed  24. Yesil P, Michel M, Chwalek K, Pedack S, Jany C, Ludwig B, Bornstein SR, Lammert E. A new collagenase blend increases the number of islets isolated from mouse pancreas. Islets . 2009; 1( 3): 185– 190. Google Scholar CrossRef Search ADS PubMed  25. Belz M, Willenborg M, Görgler N, Hamada A, Schumacher K, Rustenbeck I. Insulinotropic effect of high potassium concentration beyond plasma membrane depolarization. Am J Physiol Endocrinol Metab . 2014; 306( 6): E697– E706. Google Scholar CrossRef Search ADS PubMed  26. Lambernd S, Taube A, Schober A, Platzbecker B, Görgens SW, Schlich R, Jeruschke K, Weiss J, Eckardt K, Eckel J. Contractile activity of human skeletal muscle cells prevents insulin resistance by inhibiting pro-inflammatory signalling pathways. Diabetologia . 2012; 55( 4): 1128– 1139. Google Scholar CrossRef Search ADS PubMed  27. Wikstrom JD, Sereda SB, Stiles L, Elorza A, Allister EM, Neilson A, Ferrick DA, Wheeler MB, Shirihai OS. A novel high-throughput assay for islet respiration reveals uncoupling of rodent and human islets [ published erratum appears in PLoS One. 2013;8(12)]. PLoS One . 2012; 7( 5): e33023. Google Scholar CrossRef Search ADS PubMed  28. Weibel ER. Stereological principles for morphometry in electron microscopic cytology. Int Rev Cytol . 1969; 26: 235– 302. Google Scholar CrossRef Search ADS PubMed  29. Bock T, Pakkenberg B, Buschard K. Increased islet volume but unchanged islet number in ob/ob mice. Diabetes . 2003; 52( 7): 1716– 1722. Google Scholar CrossRef Search ADS PubMed  30. Jain D, Weber G, Eberhard D, Mehana AE, Eglinger J, Welters A, Bartosinska B, Jeruschke K, Weiss J, Päth G, Ariga H, Seufert J, Lammert E. DJ-1 protects pancreatic beta cells from cytokine- and streptozotocin-mediated cell death. PLoS One . 2015; 10( 9): e0138535. Google Scholar CrossRef Search ADS PubMed  31. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(−Delta Delta C(T)) method. Methods . 2001; 25( 4): 402– 408. Google Scholar CrossRef Search ADS PubMed  32. Taylor EB, An D, Kramer HF, Yu H, Fujii NL, Roeckl KS, Bowles N, Hirshman MF, Xie J, Feener EP, Goodyear LJ. Discovery of TBC1D1 as an insulin-, AICAR-, and contraction-stimulated signaling nexus in mouse skeletal muscle. J Biol Chem . 2008; 283( 15): 9787– 9796. Google Scholar CrossRef Search ADS PubMed  33. Baus D, Heermeier K, De Hoop M, Metz-Weidmann C, Gassenhuber J, Dittrich W, Welte S, Tennagels N. Identification of a novel AS160 splice variant that regulates GLUT4 translocation and glucose-uptake in rat muscle cells. Cell Signal . 2008; 20( 12): 2237– 2246. Google Scholar CrossRef Search ADS PubMed  34. Moltke I, Grarup N, Jørgensen ME, Bjerregaard P, Treebak JT, Fumagalli M, Korneliussen TS, Andersen MA, Nielsen TS, Krarup NT, Gjesing AP, Zierath JR, Linneberg A, Wu X, Sun G, Jin X, Al-Aama J, Wang J, Borch-Johnsen K, Pedersen O, Nielsen R, Albrechtsen A, Hansen T. A common Greenlandic TBC1D4 variant confers muscle insulin resistance and type 2 diabetes. Nature . 2014; 512( 7513): 190– 193. Google Scholar CrossRef Search ADS PubMed  35. Geng X, Li L, Watkins S, Robbins PD, Drain P. The insulin secretory granule is the major site of K(ATP) channels of the endocrine pancreas. Diabetes . 2003; 52( 3): 767– 776. Google Scholar CrossRef Search ADS PubMed  36. Geng X, Li L, Bottino R, Balamurugan AN, Bertera S, Densmore E, Su A, Chang Y, Trucco M, Drain P. Antidiabetic sulfonylurea stimulates insulin secretion independently of plasma membrane KATP channels. Am J Physiol Endocrinol Metab . 2007; 293( 1): E293– E301. Google Scholar CrossRef Search ADS PubMed  37. Nagamatsu S, Ohara-Imaizumi M, Nakamichi Y, Kikuta T, Nishiwaki C. Imaging docking and fusion of insulin granules induced by antidiabetes agents: sulfonylurea and glinide drugs preferentially mediate the fusion of newcomer, but not previously docked, insulin granules. Diabetes . 2006; 55( 10): 2819– 2825. Google Scholar CrossRef Search ADS PubMed  38. Chen PC, Bruederle CE, Gaisano HY, Shyng SL. Syntaxin 1A regulates surface expression of beta-cell ATP-sensitive potassium channels. Am J Physiol Cell Physiol . 2011; 300( 3): C506– C516. Google Scholar CrossRef Search ADS PubMed  39. Sivaprasadarao A, Taneja TK, Mankouri J, Smith AJ. Trafficking of ATP-sensitive potassium channels in health and disease. Biochem Soc Trans . 2007; 35( 5): 1055– 1059. Google Scholar CrossRef Search ADS PubMed  40. Cui X, Yang G, Pan M, Zhang XN, Yang SN. Akt signals upstream of L-type calcium channels to optimize insulin secretion. Pancreas . 2012; 41( 1): 15– 21. Google Scholar CrossRef Search ADS PubMed  41. Dehghany J, Hoboth P, Ivanova A, Mziaut H, Müller A, Kalaidzidis Y, Solimena M, Meyer-Hermann M. A spatial model of insulin-granule dynamics in pancreatic β-cells. Traffic . 2015; 16( 8): 797– 813. Google Scholar CrossRef Search ADS PubMed  42. Ohara-Imaizumi M, Fujiwara T, Nakamichi Y, Okamura T, Akimoto Y, Kawai J, Matsushima S, Kawakami H, Watanabe T, Akagawa K, Nagamatsu S. Imaging analysis reveals mechanistic differences between first- and second-phase insulin exocytosis. J Cell Biol . 2007; 177( 4): 695– 705. Google Scholar CrossRef Search ADS PubMed  43. Seino S, Shibasaki T, Minami K. Dynamics of insulin secretion and the clinical implications for obesity and diabetes. J Clin Invest . 2011; 121( 6): 2118– 2125. Google Scholar CrossRef Search ADS PubMed  44. Hoboth P, Müller A, Ivanova A, Mziaut H, Dehghany J, Sönmez A, Lachnit M, Meyer-Hermann M, Kalaidzidis Y, Solimena M. Aged insulin granules display reduced microtubule-dependent mobility and are disposed within actin-positive multigranular bodies [ published erratum appears in Proc Natl Acad Sci USA. 2015;112(16):E2114]. Proc Natl Acad Sci USA . 2015; 112( 7): E667– E676. Google Scholar CrossRef Search ADS PubMed  45. Borg LA, Westberg M, Grill V. The priming effect of glucose on insulin release does not involve redistribution of secretory granules within the pancreatic B-cell. Mol Cell Endocrinol . 1988; 56( 3): 219– 225. Google Scholar CrossRef Search ADS PubMed  46. Chen L, Chen Q, Xie B, Quan C, Sheng Y, Zhu S, Rong P, Zhou S, Sakamoto K, MacKintosh C, Wang HY, Chen S. Disruption of the AMPK-TBC1D1 nexus increases lipogenic gene expression and causes obesity in mice via promoting IGF1 secretion. Proc Natl Acad Sci USA . 2016; 113( 26): 7219– 7224. Google Scholar CrossRef Search ADS PubMed  47. Itoh Y, Kawamata Y, Harada M, Kobayashi M, Fujii R, Fukusumi S, Ogi K, Hosoya M, Tanaka Y, Uejima H, Tanaka H, Maruyama M, Satoh R, Okubo S, Kizawa H, Komatsu H, Matsumura F, Noguchi Y, Shinohara T, Hinuma S, Fujisawa Y, Fujino M. Free fatty acids regulate insulin secretion from pancreatic beta cells through GPR40. Nature . 2003; 422( 6928): 173– 176. Google Scholar CrossRef Search ADS PubMed  48. Grundy SM. Overnutrition, ectopic lipid and the metabolic syndrome. J Investig Med . 2016; 64( 6): 1082– 1086. Google Scholar CrossRef Search ADS PubMed  49. Kuzmenko DI, Klimentyeva TK. Role of ceramide in apoptosis and development of insulin resistance. Biochemistry (Mosc) . 2016; 81( 9): 913– 927. Google Scholar CrossRef Search ADS PubMed  50. Whitfield J, Paglialunga S, Smith BK, Miotto PM, Simnett G, Robson HL, Jain SS, Herbst EAF, Desjardins EM, Dyck DJ, Spriet LL, Steinberg GR, Holloway GP. Ablating the protein TBC1D1 impairs contraction-induced sarcolemmal glucose transporter 4 redistribution but not insulin-mediated responses in rats. J Biol Chem . 2017; 292( 40): 16653– 16664. Google Scholar CrossRef Search ADS PubMed  51. MacDonald MJ. High content of mitochondrial glycerol-3-phosphate dehydrogenase in pancreatic islets and its inhibition by diazoxide. J Biol Chem . 1981; 256( 16): 8287– 8290. Google Scholar PubMed  52. Itoyama A, Michiyuki S, Honsho M, Yamamoto T, Moser A, Yoshida Y, Fujiki Y. Mff functions with Pex11pβ and DLP1 in peroxisomal fission. Biol Open . 2013; 2( 10): 998– 1006. Google Scholar CrossRef Search ADS PubMed  Copyright © 2018 Endocrine Society

Journal

EndocrinologyOxford University Press

Published: Apr 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial