Increased dosage of mammalian Sir2 in pancreatic β cells enhances glucose-stimulated insulin secretion in mice

A R T I C L E

Increased dosage of mammalian Sir2 in pancreatic β cells enhances glucose-stimulated insulin secretion in mice Kathryn A. Moynihan,1 Andrew A. Grimm,1 Marie M. Plueger,1 Ernesto Bernal-Mizrachi,2 Eric Ford,2 Corentin Cras-Méneur,2 M. Alan Permutt,2 and Shin-ichiro Imai1,* 1

Department of Molecular Biology and Pharmacology Department of Medicine, Division of Endocrinology, Metabolism, and Lipid Research Washington University School of Medicine, St. Louis, Missouri 63110 *Correspondence: [email protected] 2

Summary Sir2 NAD-dependent deacetylases connect transcription, metabolism, and aging. Increasing the dosage or activity of Sir2 extends life span in yeast, worms, and flies and promotes fat mobilization and glucose production in mammalian cells. Here we show that increased dosage of Sirt1, the mammalian Sir2 ortholog, in pancreatic β cells improves glucose tolerance and enhances insulin secretion in response to glucose in beta cell-specific Sirt1-overexpressing (BESTO) transgenic mice. This phenotype is maintained as BESTO mice age. Pancreatic perfusion experiments further demonstrate that Sirt1 enhances insulin secretion in response to glucose and KCl. Microarray analyses of β cell lines reveal that Sirt1 regulates genes involved in insulin secretion, including uncoupling protein 2 (Ucp2). Isolated BESTO islets also have reduced Ucp2, increased ATP production, and enhanced insulin secretion during glucose and KCl stimulation. These findings establish the importance of Sirt1 in β cell function in vivo and suggest therapeutic interventions for type 2 diabetes. Introduction Silent information regulator 2 (Sir2) proteins are an evolutionarily conserved family of NAD-dependent protein deacetylases (Imai et al., 2000; Landry et al., 2000; Smith et al., 2000). In a wide variety of experimental organisms, Sir2 proteins regulate aging and longevity (Bordone and Guarente, 2005). Genetic and chemical manipulations that increase the dosage or activity of Sir2 proteins dramatically extend the life spans of yeast, worms, and flies, while deletions or mutations of the Sir2 genes shorten their life spans (Astrom et al., 2003; Howitz et al., 2003; Kaeberlein et al., 1999; Rogina and Helfand, 2004; Tissenbaum and Guarente, 2001; Wood et al., 2004). Sir2 proteins also connect nutritional and hormonal cues to the regulation of life span. Caloric restriction, the most consistent regimen to retard aging and extend longevity in many different species, requires the Sir2 genes for life span extension in Saccharomyces cerevisiae and Drosophila melanogaster (Anderson et al., 2003; Lin et al., 2000; Lin et al., 2004; Lin et al., 2002; Rogina and Helfand, 2004). Additionally, the Sir2 ortholog SIR-2.1 in Caenorhabditis elegans promotes longevity via the forkhead transcription factor DAF-16 in the highly conserved insulin-like signaling pathway (Tissenbaum and Guarente, 2001). This pathway has been demonstrated to play a critical role in regulating aging in worms, flies, and mice (Guarente and Kenyon, 2000; Tatar et al., 2003). While it is not yet clear whether Sir2 similarly regulates aging and longevity in mammals, the Sir2 ortholog, Sirt1/Sir2α, has been shown to regulate metabolism in response to nutrient availability (Bordone and Guarente, 2005). Nutrient deprivation in mammalian cells results in an increase in Sirt1 expression dependent upon the forkhead transcription factor Foxo3 and the tumor suppressor p53 (Nemoto et al., 2004). Likewise, dur-

ing fasting and caloric restriction in rodents, Sirt1 mRNA and protein levels are upregulated in tissues such as muscle, brain, liver, and fat (Al-Regaiey et al., 2004; Cohen et al., 2004; Nemoto et al., 2004; Rodgers et al., 2005). Increased dosage or activity of Sirt1 in adipocytes triggers lipolysis and promotes free fatty acid mobilization by repressing PPAR-γ, a nuclear receptor that promotes adipogenesis (Picard et al., 2004). Sirt1 also regulates the gluconeogenic and glycolytic pathways in liver in response to fasting by interacting with and deacetylating PGC-1α, a key transcriptional regulator of glucose production in the liver (Rodgers et al., 2005). These results indicate that Sirt1 regulates metabolic responses to changes in nutrient availability in a wide variety of tissues. Pancreatic β cells play a central role in maintaining glucose homeostasis in mammals by secreting insulin in response to elevated blood glucose levels. β cells have a highly coordinated mechanism that senses glucose and transduces its signal to insulin secretion. Glucose, which enters the β cells via GLUT2 glucose transporters, is metabolized during glycolysis, resulting in an increase in the ATP/ADP ratio. Subsequent closure of ATP-dependent K+ (KATP) channels causes membrane depolarization, thereby opening voltage-gated Ca2+ channels. The inward Ca2+ flux ultimately triggers exocytosis of insulincontaining granules. Based on the fact that Sir2 proteins connect nutrient availability to longevity and our finding that β cells express Sirt1 in pancreatic islets, we hypothesized that Sirt1 plays an important role in the regulation of insulin secretion and that increasing the dosage of Sirt1 in β cells conveys beneficial effects on glucose homeostasis. Here we show that pancreatic beta cell-specific Sirt1-overexpressing (BESTO) transgenic mice exhibit improved glucose tolerance and enhanced insulin secretion when challenged with high glucose. This phenotype is maintained as BESTO mice age. In situ pancreatic perfusion

CELL METABOLISM : AUGUST 2005 · VOL. 2 · COPYRIGHT © 2005 ELSEVIER INC.

DOI 10.1016/j.cmet.2005.07.001

105

A R T I C L E

experiments further demonstrate enhanced insulin secretion in BESTO pancreata during glucose and potassium chloride (KCl) stimulation. Gene expression analyses on Sirt1-overexpressing and Sirt1-knockdown pancreatic β cell lines show that Sirt1 regulates a variety of genes involved in β cell function, including Ucp2, an uncoupling protein that negatively regulates insulin secretion in β cells. Consistent with these results, isolated BESTO islets have reduced Ucp2 and increased ATP levels and exhibit enhanced insulin secretion in response to both glucose and KCl. Together, these results establish that an increased dosage of Sirt1 has beneficial effects on mammalian physiology in vivo. Our findings also provide new insight into the physiological and molecular functions of Sirt1 in glucose metabolism and suggest therapeutic interventions for the prevention and treatment of age-associated metabolic disorders, such as type 2 diabetes mellitus. Results Sirt1 is expressed in pancreatic β cell lines and mouse islets To examine the connection between Sirt1 and β cell function, we first assessed the expression levels of the Sirt1 protein in two pancreatic β cell lines, MIN6 (mouse) and INS1 (rat), and in primary islets isolated from C57BL/6 mice (Figure 1A). Moderate levels of Sirt1 were detected in MIN6 and INS1 cell lines compared to a relatively high expression level of Sirt1 in the mouse NIH3T3 fibroblast cell line. An intermediate level of Sirt1 was also detected in whole pancreatic extracts compared to low and high levels of Sirt1 in brain and testis extracts, respectively (Figure 1A). Most importantly for our studies, Sirt1 was expressed in isolated mouse islets (Figure 1A). To examine the distribution of Sirt1 in the pancreas, we immunostained Sirt1 on mouse pancreatic sections with an antibody against the N-terminal portion of Sirt1. Strikingly, Sirt1 was expressed specifically in pancreatic islets and not in pancreatic exocrine tissues (Figure 1B), suggesting a role for Sirt1 in pancreatic endocrine function. Within the islets, β cells had a weak, punctate Sirt1 signal in the nucleus and a diffusely distributed signal in the cytoplasm, although the signal intensity varied from cell to cell (Figures 1B, 1C, and 1E). Immunostaining of testis sections showed that Sirt1 was expressed exclusively in the nuclei of spermatogenic cells (Figure 1D), a positive control consistent with previous reports (McBurney et al., 2003). An antibody against the C-terminal portion of Sirt1 also showed similar staining patterns in pancreatic islets (data not shown). Therefore, it is unlikely that the cytoplasmic staining in β cells is simply due to nonspecific crossreactivity of the antibody. Interestingly, glucagon-positive α cells exhibited strong, exclusive cytoplasmic staining of Sirt1 (Figures 1E–1G). The biological role of this cytoplasmic Sirt1 in α cells is currently unknown. The nuclear and cytoplasmic distribution of Sirt1 in β cells and the central role of β cells in glucose metabolism prompted us to investigate whether Sirt1 could play a role in the regulation of β cell function and glucose homeostasis in mammals. Production of BESTO transgenic mice We generated BESTO transgenic mice in order to address the hypothesis that increased Sirt1 dosage in β cells might convey beneficial effects on β cell function in mammals. In these mice, a 1.9 kb fragment of the human insulin promoter, which has

106

been demonstrated to mediate pancreatic β cell-specific transcription in mice (Hotta et al., 1998; Yamamoto et al., 2003), drives expression of the mouse Sirt1 cDNA (Figure 2A). Two independent BESTO transgenic lines, 431-2 and 431-4, were established on the C57BL/6 inbred background. Sirt1 was overexpressed 12- and 18-fold in islets isolated from mice from lines 431-2 and 431-4, respectively, compared to control nontransgenic siblings (Figure 2B). The β cell-specific overexpression of Sirt1 was confirmed by immunostaining pancreatic sections of BESTO mice (Figure 2C). Strong nuclear staining of Sirt1 was observed in insulin-positive β cells but not in glucagon-positive α cells of the BESTO mice, although the levels of Sirt1 expression varied from cell to cell and from islet to islet (Figure 2C). No obvious overexpression of Sirt1 was detected by Western blotting and immunostaining of other tissues, including brain, liver, kidney, adipose tissue, and muscle (data not shown). BESTO mice from both lines 431-2 and 431-4 did not show any gross abnormalities and did not differ in body weight from their nontransgenic siblings up to 18 months of age (data not shown). We also determined whether the increased dosage of Sirt1 affected islet structure by measuring the percent islet area and the β cell mass. BESTO mice and their controls did not differ significantly in either parameter (Figures 2D and 2E). Additionally, the pancreatic insulin content did not differ statistically between BESTO and control mice, although some BESTO mice showed slightly higher values (data not shown). BESTO mice exhibit improved glucose tolerance and enhanced insulin secretion in response to high glucose To examine the effects of increased Sirt1 dosage on β cell function in vivo, we first measured blood glucose and plasma insulin levels in the BESTO mice. At 3 months of age, fed and fasted glucose levels were similar between transgenic mice and their nontransgenic siblings in both the 431-2 and 431-4 lines (Figure 3A). Fed and fasted insulin levels also did not differ between transgenic mice and their controls in both BESTO lines (Figure 3B and 0 min time points in Figure 3E). However, when challenged with high glucose in an intraperitoneal glucose tolerance test (IPGTT), both male and female BESTO mice in both lines showed significantly improved glucose tolerance compared to their nontransgenic siblings (Figures 3C and 3D). We measured insulin levels at 0 and 30 min time points during the IPGTTs. Consistent with their improved glucose tolerance, plasma insulin levels 30 min after glucose injection were significantly higher in male BESTO mice compared to their nontransgenic siblings (Figure 3E, left panel). Females also showed a similar trend, although the difference was not statistically significant (Figure 3E, right panel). These findings suggest that an increased dosage of Sirt1 in pancreatic β cells enhances their ability to secrete insulin and reduce blood glucose levels in response to high glucose in vivo. In situ-perfused pancreata of BESTO mice show enhanced insulin secretion In order to further confirm that insulin secretion is enhanced in BESTO pancreata in response to high glucose, we conducted in situ pancreatic perfusion experiments in BESTO and control mice. Total insulin release in response to 25 mM glucose was significantly enhanced in BESTO pancreata compared to controls (Figures 4A and 4B, left panel). This increased insulin re-

CELL METABOLISM : AUGUST 2005

Sirt1 regulates pancreatic β cell function in mice

Figure 1. Expression of Sirt1 in pancreatic β cell lines and islets A) Western blot analysis of Sirt1 in extracts prepared from NIH3T3 mouse fibroblasts, MIN6 and INS1 pancreatic β cell lines, pancreas, brain, and testis. Cell and tissue extracts (30 and 45 ␮g) were blotted in two separate membranes, respectively. For mouse islets, Sirt1 was first immunoprecipitated from 270 ␮g of extract with the anti-Sirt1 antibody and immunoblotted. Actin was used as a loading control. B–D) (B) Immunostaining of Sirt1 with an N-terminal Sirt1 antibody (green) on mouse pancreatic sections. No signal was detected with a rabbit IgG control (data not shown). (C) Higher magnification of one of the islets shown in (B) showed weak, punctate signals (arrows) in the nucleus as well as diffusely distributed signals in the cytoplasm of β cells. (D) Immunostaining of Sirt1 with the N-terminal antibody in the testis served as a positive control for the nuclear localization of Sirt1. E–H) Sirt1 ([E], green) and glucagon ([F], red) were coimmunostained on the same mouse pancreatic section, and the merged image (G) is shown with Hoechst dye staining of the nuclei (blue). Insulin ([H], red) was stained on an adjacent section.

lease by BESTO pancreata was mainly due to an increase in first-phase insulin secretion (Figure 4B, left panel). We also examined the response of BESTO pancreata to KCl, a potent secretagogue of insulin secretion, in the same perfusion experiments (Figure 4A). In response to KCl, BESTO pancreata showed an even more pronounced increase in insulin secretion compared to controls (Figures 4A and 4B, right panel). These results further support the notion that the insulin-secretion capability of β cells is enhanced by Sirt1 in BESTO mice. BESTO mice maintain improved β cell function as they age Because Sir2 is considered to be an important link between metabolism and aging, we next examined whether the same cohorts of BESTO mice still maintain enhanced β cell function

CELL METABOLISM : AUGUST 2005

compared to controls at a later age. As at 3 months, fed and fasted glucose and insulin levels of BESTO mice were similar to those of controls at 8 months of age (Figures 5A and 5B, and 0 min time points in Figure 5E). Furthermore, 8-monthold BESTO mice still maintained improved glucose tolerance compared to controls (Figures 5C and 5D). Although male BESTO mice in line 431-4 did not show a statistically significant improvement, we suspect that this was due to the smaller number of available individuals in this cohort for the 8 month analysis. Plasma insulin levels were also higher 30 min after glucose injection in 8-month-old male BESTO mice compared to controls in both lines (Figure 5E, left panel, and data not shown). Female BESTO mice at 8 months of age did not show a statistically significant difference in plasma insulin levels in both lines, although the trend was similar to the 3 month data

107

A R T I C L E

Figure 2. Generation of BESTO transgenic mice A) Schematic representation of the pIns-Sirt1 transgene used to generate the BESTO mice. The transgene contains a 1.9 kb fragment of the human insulin promoter, rabbit β-globin introns, the 2.2 kb mouse Sirt1 cDNA, and a polyadenylation signal (pA) from the rabbit β-globin gene (Miyazaki et al., 1990). B) Western blot analysis of Sirt1 expression in transgenic and control islet extracts. Islets were pooled from three mice from lines 431-2 and 431-4. Actin was used as a loading control for normalization. C) Immunostaining of Sirt1 (green) in transgenic (upper panel) and control (lower panel) pancreatic sections from line 431-2. Equal exposure times were applied to the transgenic and control samples. Glucagon (red) was costained with Sirt1, and the merged image is shown with Hoechst dye staining the nuclei (blue). Insulin (red) was stained on adjacent sections. Similar overexpression was seen in line 431-4. D and E) (D) Percent islet area and (E) β cell mass in 4-month-old BESTO transgenic male mice from line 431-4 compared to controls (n = 4). All results are expressed as mean ± SE.

(Figure 5E, right panel, and data not shown). Taken together, these data indicate that increased dosage of Sirt1 in pancreatic β cells is able to convey beneficial effects on glucose metabolism at both 3 and 8 months of age by improving glucose tolerance and enhancing insulin secretion in response to high glucose.

108

Sirt1 regulates the expression of genes important for β cell function To dissect the molecular mechanisms by which Sirt1 regulates β cell function, we employed gene expression profiling to identify genes that are affected by changing the dosage of Sirt1 in pancreatic β cells. For this purpose, we established two inde-

CELL METABOLISM : AUGUST 2005

Sirt1 regulates pancreatic β cell function in mice

Figure 3. Blood glucose, plasma insulin, and IPGTTs in BESTO transgenic mice at 3 months of age A) Fed and fasted blood glucose levels in male (M) and female (F) BESTO transgenic mice (black bars) compared to controls (white bars) in lines 431-2 and 431-4 (n = 10–22). B) Fed plasma insulin levels in male (M) and female (F) BESTO transgenic mice (black bars) compared to controls (white bars) in lines 431-2 and 431-4 (n = 8–13). C and D) IPGTTs in male and female BESTO transgenic mice (closed circles) compared to controls (open circles) in lines 431-2 and 431-4. Following a 15 hr fast, 2 g of 50% glucose (w/v) per kg body weight was injected intraperitoneally, and blood glucose values were assessed at 0, 15, 30, 60, and 120 min post injection. (C) Line 431-2. BESTO males (n = 14), control males (n = 15), BESTO females (n = 8), control females (n = 9). (D) Line 431-4. BESTO males (n = 15), control males (n = 10), BESTO females (n = 11), control females (n = 11). E) Plasma insulin levels in male (left panel) and female (right panel) BESTO transgenic mice (black bars) compared to controls (white bars) in line 431-4 at 0 and 30 min time points in IPGTTs shown in Figure 3D. BESTO males (n = 13), control males (n = 8), BESTO females (n = 10), control females (n = 10). ‡, measurements in both transgenic and control female mice lower than the detectable limit of 0.156 ng/ml were averaged as 0.156 ng/ml. All results are expressed as mean ± SE. *p % 0.05, **p % 0.01, ***p % 0.001.

pendent MIN6 cell lines overexpressing Sirt1 (MIN6-Sir2OE1 and 2), as well as two control cell lines carrying the neomycinresistance gene only (MIN6-NEO1 and -2). MIN6-Sir2OE1 and -2 showed 7.6- and 8.5-fold higher expression levels of Sirt1 compared to controls (Figure 6A). We also established two independent MIN6 cell lines that constitutively expressed two different small interfering RNAs (siRNAs) against the Sirt1 gene. In these Sirt1-knockdown MIN6 cell lines (MIN6-Sir2KD1 and -2), Sirt1 expression levels were reduced to 28% and 53%, respectively, of the control cell line carrying the backbone vector (MIN6-mU6pro) (Figure 6B). Using these experimental and control cell lines, microarray experiments were conducted with dye swaps and strict filtering criteria (Figure 6C, see Supple-

CELL METABOLISM : AUGUST 2005

mental Experimental Procedures in the Supplemental Data available with this article online). Forty-one downregulated and 24 upregulated transcripts were identified in Sir2OE cell lines (see Table S1), while 24 upregulated and 18 downregulated transcripts were identified in Sir2KD cell lines (see Table S2). As expected, there was a statistically significant overlap in the genes that changed in Sir2OE and Sir2KD cell lines as determined by the hypergeometric distribution (p = 2.65 − 10−12, data not shown). Microarray results (deposited into the NCBI GEO database as series GSE2365) were confirmed by measuring transcript levels of several genes whose expression was changed by Sir2 overexpression, knockdown, or both (Figure 6D). Gene expression changes measured by quantitative real-

109

A R T I C L E

Figure 4. In situ pancreatic perfusion experiments in BESTO transgenic mice A) Insulin release from perfused pancreata of male BESTO (n = 5) and control (n =5) mice from line 431-4 in response to 25 mM glucose (black bar) and 25 mM KCl (gray bar). The two arrows indicate the start of first and second phase insulin secretion, respectively. All results are expressed as mean ± SE. B) Area under the curve, representing the amount of insulin release, was calculated for each time interval during glucose (left panel) and KCl (right panel) stimulation. The brackets in the left panel indicate total, first, and second phase insulin secretion during glucose stimulation. The bracket in the right panel indicates total insulin secretion during KCl stimulation. The Wilcoxon matched pairs test was employed to determine significance: **p % 0.01; ***p % 0.001; NS, not significant.

time RT-PCR were statistically significant and highly consistent with the microarray results (Figure 6D). The gene expression profiles revealed that Sirt1 regulates expression of genes involved in β cell physiology and glucose homeostasis. For example, the uncoupling protein 2 gene (Ucp2), the prolactin receptor gene (Prlr), and the urocortin 3 gene (Ucn3) were downregulated in Sir2OE cell lines and/or upregulated in Sir2KD cell lines (Figure 6D). Ucp2 is a member of the mitochondrial inner membrane proton transporters that uncouple respiration from ATP synthesis by leaking protons from the intermembrane space to the matrix side of the inner membrane (Chan et al., 2004). It has been demonstrated that Ucp2 negatively regulates glucose-stimulated insulin secretion and is a key component of β cell glucose sensing (Joseph et al., 2002; Zhang et al., 2001). Signaling from the prolactin receptor is also important in regulating islet mass and glucose

110

sensitivity during pregnancy (Amaral et al., 2004), and prolactin receptor-deficient islets show defects in insulin production and glucose-stimulated insulin secretion (Freemark et al., 2002). Ucn3, a high-affinity ligand for the type 2 corticotropin-releasing factor receptor, is secreted by pancreatic islets and MIN6 cells and stimulates insulin and glucagon secretion in vitro and in vivo (Li et al., 2003). The large conductance calcium-activated potassium channel gene (Kcnmb4) and the monocarboxylic acid transporter gene (Mct1) were upregulated in Sir2OE cell lines but not significantly changed in Sir2KD cell lines (Figure 6D). The large conductance calcium-activated potassium channels are speculated to cause the primary repolarizing current and serve to terminate glucose-initiated electrical bursting in β cells (Li et al., 1999). Mct1 transports pyruvate and lactate through the plasma membrane. It has been reported that increased Mct1 and lactate dehydrogenase activities result in

CELL METABOLISM : AUGUST 2005

Sirt1 regulates pancreatic β cell function in mice

Figure 5. Blood glucose, plasma insulin, and IPGTTs in BESTO transgenic mice at 8 months of age A) Fed and fasted blood glucose levels in male (M) and female (F) BESTO transgenic mice (black bars) compared to controls (white bars) in lines 431-2 and 431-4 (n = 10–22). B) Fed plasma insulin levels in male (M) and female (F) BESTO transgenic mice (black bars) compared to controls (white bars) in lines 431-2 and 431-4 (n = 6–16). C and D) IPGTTs in male and female BESTO transgenic mice (closed circles) compared to controls (open circles) in lines 431-2 and 431-4. These mice were the same mice used for the 3 month analyses shown in Figure 3. (C) Line 431-2. BESTO males (n = 11), control males (n = 11), BESTO females (n = 7), control females (n = 9). (D) Line 431-4. BESTO males (n = 7), control males (n = 5), BESTO females (n =11), control females (n =11). E) Plasma insulin levels in male (left panel) and female (right panel) BESTO transgenic mice (black bars) compared to controls (white bars) in line 431-2 at 0 and 30 min time points in IPGTTs shown in Figure 5C. BESTO males (n = 16), control males (n = 18), BESTO females (n = 12), control females (n = 16). All results are expressed as mean ± SE. *p % 0.05, **p % 0.01, ***p % 0.001.

pyruvate- and lactate-stimulated insulin secretion in isolated rat islets (Ishihara et al., 1999). It is still unclear to what extent each gene expression change contributes to the Sir2-dependent regulation of β cell function. Nevertheless, the fact that Sirt1 affects a wide variety of genes involved in the regulation of insulin secretion is consistent with our hypothesis that Sirt1 plays an important role in the regulation of β cell function and glucose homeostasis. Islets from BESTO mice show enhanced insulin secretion with decreased amounts of Ucp2 and increased ATP production Among the genes identified from the microarray analyses, we examined several gene products by Western blot analysis in

CELL METABOLISM : AUGUST 2005

islets isolated from BESTO and control mice. Based on the availability of appropriate antibodies, we chose Ucp2, Glut2, Kcnmb4, and Mct1 (Figure 7A). Three of the four gene products, Ucp2, Glut2, and Kcnmb4, changed in the same direction as in the microarray and quantitative real-time RT-PCR. Ucp2 decreased approximately 1.8-fold, Kcnmb4 increased approximately 1.5-fold, and Glut2 showed a marginal decrease compared to controls (Figure 7B). However, Mct1 decreased approximately 1.5-fold compared to controls, a change in direction opposite to that seen in the microarray and RT-PCR results (Figure 7B). The downregulation of Ucp2 in BESTO islets was particularly intriguing because the reduction in Ucp2 levels was consistent with the observed increase in insulin secretion in BESTO mice

111

A R T I C L E

Figure 6. Gene expression profiling in Sirt1-overexpressing and Sirt1-knockdown pancreatic β cell lines A) Western blot analysis of Sirt1 in MIN6 cell lines independently infected with pBabe-Neo (MIN6NEO1 and NEO2) or pBabe-Sirt1 (MIN6-Sir2OE1 and Sir2OE2) retrovirus. The asterisk indicates nonspecific crossreacting bands that serve as loading controls. B) Western blot analysis of Sirt1 in MIN6 cell lines independently cotransfected with the mU6pro plasmid (MIN6-mU6pro) or one of two different Sirt1siRNA expression vectors (MIN6-Sir2KD1 and Sir2KD2) and a plasmid carrying a neomycin-resistant gene. C) Scheme of microarray experiments. Four microarray hybridizations with dye swaps were conducted for each pairwise comparison. D) Validation of microarray results with quantitative real-time RT-PCR. OE1, MIN6-Sir2OE1; OE2, MIN6Sir2OE2; KD1, MIN6-Sir2KD1; KD2, MIN6-Sir2KD2; M, microarray; R, RT-PCR. Serpina1a, serine proteinase inhibitor 1-1 (α1-antitrypsin); Arid1b, AT-rich interactive domain 1B; Kcnmb4, large conductance calcium-activated potassium channel b4; Mct1, solute carrier family 16 member 1 (monocarboxylic acid transporter 1); Abcc3, ATP binding cassette transporter c3; Prlr, prolactin receptor; Ucn3, urocortin 3; Ucp2, uncoupling protein 2. All error bars represent standard deviations.

during IPGTTs and in situ pancreatic perfusion experiments. When Ucp2 decreases in β cells, the uncoupling between respiration and ATP synthesis also decreases. This in turn causes an increase in mitochondrial ATP production, which subsequently promotes insulin secretion (Zhang et al., 2001). Therefore, we examined the effect of the observed decrease in Ucp2 on ATP production in BESTO islets. Consistent with the observed decrease in Ucp2 levels, the BESTO islets had significantly higher ATP levels compared to controls when stimulated with 20 mM glucose (Figure 7C). We then examined insulin secretion in response to both high glucose and KCl in islets isolated from BESTO and control mice. In response to 20 mM glucose, isolated BESTO islets showed a statistically significant enhancement of insulin secretion compared to controls (Figure 7D), although there was a large variation in insulin secretion among the individual groups of BESTO islets (Figure S1). The BESTO islets also showed an increase in insulin content compared to controls when stim-

112

ulated with 20 mM glucose (Figure 7E). Strikingly, KCl stimulation of isolated BESTO islets elicited an even more pronounced increase in insulin secretion compared to controls (Figure 7D and Figure S1). Consistent with the results from the IPGTTs and the in situ pancreatic perfusion experiments, these findings demonstrate that increased dosage of Sirt1 in β cells enhances the insulin secretion capability of isolated BESTO islets. Discussion Increased dosage of Sirt1 in pancreatic β cells enhances glucose-stimulated insulin secretion in vivo In the present study, we have provided several lines of evidence demonstrating that the mammalian Sir2 ortholog Sirt1 regulates glucose-stimulated insulin secretion in pancreatic β cells. First, Sirt1 was expressed exclusively in pancreatic islets and not in the pancreatic exocrine tissue, and Sirt1 was pres-

CELL METABOLISM : AUGUST 2005

Sirt1 regulates pancreatic β cell function in mice

Figure 7. Ucp2, ATP, and insulin measurements in isolated BESTO islets A) Western blot analysis of Ucp2, Glut2, Kcmnb4, and Mct1 proteins in transgenic and control islet extracts from line 431-4. For each independent extract, islets were pooled from two to three mice, and two to five independent Western blots were conducted. The asterisk represents a crossreacting band. Actin was used as a loading control for normalization. B) Quantitation of the independent Western blot results, normalized to actin. C) ATP levels (pmol/␮g protein) in transgenic (black bars) and control (white bars) islets from line 431-4 in extracts collected after stimulation with the indicated concentrations of glucose in insulin secretion assays (n = 4). D) Insulin secreted (ng/ml/hr) from transgenic (black bars) and control (white bars) islets from line 431-4 at the indicated glucose or KCl concentrations (n = 8 for glucose stimulation; n = 4 for KCl stimulation). The secretion assays were performed on triplicate groups of ten islets for each mouse. E) Islet insulin content (ng/islet) of transgenic (black bars) and control (white bars) islets from line 431-4 in extracts collected after stimulation with the indicated concentrations of glucose in insulin secretion assays (n = 4). F) Proposed model for Sirt1-mediated regulation of insulin secretion in the β cell. All results are expressed as mean ± SE. *p % 0.05, **p % 0.01, ***p % 0.001.

ent in both the nucleus and cytoplasm of β cells (Figures 1B– 1H). Second, BESTO transgenic mice exhibited significantly improved glucose tolerance and enhanced insulin secretion when challenged with high glucose during IPGTTs at three

CELL METABOLISM : AUGUST 2005

months of age (Figures 3C–3E). Third, in situ pancreatic perfusion experiments further demonstrated that BESTO pancreata have enhanced insulin secretion in response to both glucose and KCl stimulation (Figure 4). Fourth, BESTO mice maintained

113

A R T I C L E

enhanced glucose-stimulated insulin secretion compared to controls at 8 months of age (Figures 5C–5E). Finally, isolated BESTO islets exhibited enhanced insulin secretion in response to both high glucose and KCl (Figure 7D). These data establish that increased dosage of Sirt1 conveys beneficial effects on mammalian physiology, particularly on glucose metabolism, in vivo. The phenotype observed in BESTO mice must primarily be due to an alteration of β cell function. In BESTO mice, the dosage of Sirt1 was increased exclusively in pancreatic β cells (Figure 2C). Additionally, islet area and β cell mass did not differ between BESTO mice and their controls (Figures 2D and 2E). Furthermore, results from the perfused pancreata and the isolated islet experiments demonstrated that the enhanced insulin secretion in BESTO mice was intrinsic to pancreatic islets and not secondary to altered insulin sensitivity in insulin target tissues such as liver or skeletal muscle (Figures 4 and 7D). Therefore, the enhanced insulin secretion and improved glucose tolerance of BESTO mice must be due to changes in β cell function rather than ectopic expression of Sirt1 in other tissues or alterations in islet structure. Genetic manipulations that overexpress proteins in pancreatic β cells can disrupt their structure and/or function. However, this is unlikely to be the case in BESTO mice. The increased dosage of Sirt1 improved β cell function in response to high glucose, and no physiological abnormalities were observed. One interpretation of this phenotype is that increasing Sirt1 dosage in β cells amplifies the normal Sirt1-mediated physiological response to nutrients in β cells. Sirt1 might maintain the robustness of β cell function in the face of dietary challenges, such as rapid changes in blood glucose, lipids, and amino acids after the ingestion of a nutrient-dense meal. Indeed, isolated BESTO islets appear to be resistant to high levels of free fatty acids (K.A.M. and S.-i.I., unpublished data). Therefore, it would be of great interest to test whether BESTO mice are more resistant to complications caused by high-carbohydrate or high-fat diets. Mechanisms of insulin secretion altered by Sirt1 in pancreatic β cells Sirt1 likely regulates β cell function through multiple mechanisms. Transcriptional profiling of Sirt1-overexpressing and Sirt1-knockdown β cell lines suggests that Sirt1 affects a variety of genes involved in the regulation of insulin secretion (Figure 6D). Western blot analyses of Ucp2, Glut2, Kcnmb4, and Mct1 in isolated islets confirmed these changes in BESTO mice (Figures 7A and 7B). Because minor changes in Glut2 expression are not sufficient to limit glucose utilization in islets (Guillam et al., 2000), it is unlikely that the marginal decrease in Glut2 protein levels in the BESTO islets would contribute to the phenotype of the BESTO mice. The changes in Kcnmb4 and Mct1 protein levels detected in isolated BESTO islets, as well as other genes identified by the microarray analyses, might contribute to the BESTO phenotypes. However, further investigation will be necessary to understand these potential connections. We were particularly interested in the decrease in Ucp2 observed in the BESTO islets, because Ucp2-deficient mice have improved glucose tolerance and enhanced insulin secretion during IPGTTs, phenotypes similar to those of BESTO mice (Zhang et al., 2001). Consistent with this observed decrease in

114

Ucp2, we also detected a significant increase in ATP levels in response to high glucose in isolated BESTO islets (Figure 7C). Glucose-induced ATP production initiates a prompt response of the insulin secretory machinery in β cells by closing KATP channels (see Figure 7F). While correlative, the observed increase in first-phase insulin secretion in the perfused BESTO pancreata is consistent with an enhancement of ATP production by Sirt1. The observed increase in ATP levels might also account for the increase in insulin secretion from isolated BESTO islets in response to glucose (Figure 7D). Additionally, Guarente and his colleagues independently found that a reduction in Sirt1 levels increases Ucp2 expression and decreases insulin secretion in β cells (L. Bordone and L. Guarente, personal communication). Taken together, the reduction of Ucp2 by Sirt1 likely contributes, at least in part, to the enhanced insulin secretion in response to high glucose in BESTO mice (Figure 7F). The observed increase in ATP levels could also explain the increase in insulin content in isolated BESTO islets exposed to high glucose (Figure 7E). It has been shown that static incubation of islets in high glucose induces proinsulin biosynthesis within 1 hr at the level of translation (Wicksteed et al., 2003). The mammalian target of rapamycin kinase, which is stimulated by ATP during glucose stimulation (McDaniel et al., 2002), has been suggested to be a potential downstream mediator of this effect (Wicksteed et al., 2003). Whether the increase in ATP levels causes the increase in insulin content and to what extent this change accounts for the increase in insulin secretion in vitro remain to be determined. Strikingly, KCl, which directly depolarizes β cells, elicited a more pronounced increase in insulin secretion in both BESTO pancreata and islets compared to the effect of glucose (Figures 4 and 7D). This suggests that Sirt1 also regulates insulin secretion downstream of β cell depolarization, independently of Ucp2 (Figure 7F). It has been proposed that nutrients, such as free fatty acids and amino acids, also stimulate insulin secretion through mechanisms downstream of depolarization (Yaney and Corkey, 2003). Although we have not yet identified the molecular target of Sirt1 that acts downstream of depolarization, this additional Sirt1-mediated mechanism might regulate insulin secretion in vivo in response to these nutrients. Indeed, we observed that the addition of free fatty acids during glucose stimulation appeared to further augment the difference in insulin secretion between BESTO and control islets (K.A.M. and S.-i.I., unpublished data). Therefore, it is of great interest to examine whether BESTO mice also show enhanced insulin secretion in response to other nutritional stimulants. Therapeutic potential for Sirt1 in pancreatic β cells Aging is one of the greatest risk factors for obesity, glucose intolerance, and type 2 diabetes (Chang and Halter, 2003; Moller et al., 2003). In addition to increased abdominal adiposity and decreased physical fitness, a progressive decline in β cell function in the elderly is a major contributing factor to the pathophysiology of type 2 diabetes (Basu et al., 2003; Iozzo et al., 1999). A similar age-associated impairment of β cell function has also been demonstrated in rodents (Muzumdar et al., 2004). The finding that Sirt1 improves β cell function at both 3 and 8 months of age provides insight into the development of new dietary and pharmacological interventions to treat type 2 diabetes.

CELL METABOLISM : AUGUST 2005

Sirt1 regulates pancreatic β cell function in mice

Based on this and other studies, Sirt1 likely functions to preserve a wide variety of cell-type- and tissue-specific functions. Therefore, it would be of great importance to further investigate the tissue-dependent effects of Sirt1 on mammalian physiology by altering Sirt1 dosage/activity in tissue-specific transgenic and knockout mice. Experimental procedures Generation of BESTO transgenic mice The pIns-Sirt1 transgene was constructed by converting the EcoRI site to SalI in the pIns-1 vector (Miyazaki et al. [1990], a gift from Dr. Jun-ichi Miyazaki, Osaka University, Japan) and inserting a 2.2 kb fragment of the mouse Sirt1 cDNA that contains the entire coding region of Sirt1 into the SalI site. In this vector, the Sirt1 cDNA is driven by the 1.9 kb human insulin promoter (Hotta et al., 1998; Yamamoto et al., 2003). The pIns-Sirt1 transgene was linearized with SphI and XhoI, purified, and microinjected into fertilized eggs of C57BL/6 inbred mice in the Washington University Mouse Genetics Core Facility. Transgenic mice were identified by PCR genotyping from tail DNA using forward and reverse primers 5#-GCGGCTTGAGGGTAATCAATAC CTGTTTGT-3# and 5#-TGATAGGCAGCCTGCACCTGAGGAGTGAAT-3#, respectively. Two transgenic founders (lines 431-2 and 431-4) were established and crossed to wild-type C57BL/6 mice (Jackson Laboratories). The F1 and F2 generations from these lines were used for analysis, and nontransgenic littermates were used as controls. All animal procedures were approved by the Washington University Animal Studies Committee. Islet morphometry analysis and total pancreatic insulin content Pancreata were fixed and processed as described in the Immunohistochemical Analysis section (see Supplemental Data). For each pancreas, insulin staining was performed in five sections (5 ␮m) separated by 200 ␮m, and the percent islet area and the β cell mass were quantitated as described previously (Bernal-Mizrachi et al., 2001) using the BioQuant image analyzer (BIOQUANT Image Analysis Corporation) and ImageJ (Girish and Vijayalakshmi, 2004) software. Total pancreatic insulin content was measured after acid-ethanol extraction of the whole pancreas, as described previously (Ma et al., 2004). Extraction was repeated twice, and the supernatants were pooled and stored at −20°C. The insulin concentration was measured by radioiummunoassay in the Washington University Radioimmunoassay (RIA) Core Facility. Fed and fasted glucose and insulin measurements Fed glucose and insulin levels were measured between 9 and 10 a.m., while fasted glucose and insulin levels were measured after 15 hr of overnight fasting. Glucose levels were determined using the Accu-Chek II glucometer (Roche Diagnostics) with blood collected from the tail vein. For determining insulin levels, blood was collected from the tail vein into chilled heparinized capillary tubes, and plasma was separated by centifugation and stored at −80°C. Insulin levels were determined on 10 ␮l aliquots using rat insulin ELISA kits with mouse insulin standards (Crystal Chem Inc.) at the Washington University RIA Core Facility.

min intervals, as indicated. The insulin concentration of each fraction was determined by radioimmunoassay. The first phase was defined from the earliest time point with an increased insulin secretion after the glucose stimulus until the time point when insulin secretion plateaued after the first peak. The second phase continued from this point until the end of glucose stimulation (Spinas et al., 1998). The area under the curve (AUC) at each time interval was calculated to determine the amount of insulin secreted during glucose and KCl stimulation, and the Wilcoxon matched pairs test was employed for statistical analyses. Glucose-stimulated insulin secretion and insulin content from isolated islets Islets were isolated by collagenase digestion. Briefly, pancreata were inflated with isolation buffer (10x HBSS, 10 mM HEPES, 1 mM MgCl2, 5 mM glucose [pH 7.4]) containing 0.375 mg/ml collagenase (Sigma) via the pancreatic duct after clamping off its entry site to the duodenum. The inflated pancreas was then removed, incubated at 37°C for 12–15 min, and shaken vigorously. Islets were separated from acinar tissue after a series of washes and passages through a 70 ␮m nylon BD Falcon Cell Strainer (BD Biosciences). Handpicked islets were cultured overnight in RPMI media containing either 5 or 11 mM glucose, 2 mM L-glutamine, penicillin/streptomycin, and 10% fetal bovine serum (FBS; GIBCO, Invitrogen). The islets were then preincubated in KRB buffer containing 2 mM glucose for 1 hr at 37°C. Islets of similar size were handpicked into groups of ten islets in triplicate and incubated with 1 ml KRB buffer containing either 2 mM glucose, 20 mM glucose, or 20 mM KCl plus 2 mM glucose for 1 hr at 37°C. The supernatant was stored at −20°C prior to insulin measurements. After the insulin secretion experiments, islets were washed two times with PBS and extracted with acid-ethanol overnight at 4°C in order to measure insulin content. Insulin levels of all samples were measured by radioimmunassay at the Washington University RIA Core Facility. ATP measurements from isolated islets Islets were pre-cultured in RPMI media containing 5 mM glucose, 2 mM L-glutamine, penicillin/streptomycin, 0.25% fatty acid-free BSA, 0.4 mM palmitate, and 5% FBS. After the insulin secretion experiments, islets were washed two times with PBS and then extracted with extraction buffer (0.1 M NaOH, 0.5 mM EDTA). After neutralizing the samples with 0.1 M HCl, ATP levels were measured using the ATP Bioluminescent Assay Kit (Sigma) according to the manufacturer’s instructions. Protein concentration of the islets was determined, and ATP was normalized to the protein content. Statistical analysis Unless otherwise indicated, all values are expressed as mean ± standard error (SE), and statistical analyses were carried out using an unpaired Student’s t test. Differences were considered to be statistically significant when p < 0.05. Additional experimental procedures are provided in Supplemental Data.

Supplemental data

Intraperitoneal glucose tolerance tests After mice were fasted for 15 hr, 50% dextrose (2 g/kg body weight) was injected intraperitoneally. Blood was collected from the tail vein at 0, 15, 30, 60, and 120 min after injection for glucose measurements. Plasma for insulin measurements was also collected at 0 and 30 min after glucose injection and submitted to the Washington University RIA Core facility.

Supplemental data include one figure, two tables, supplemental experimental procedures, and supplemental references and can be found with this article online at http://www.cellmetabolism.org/cgi/content/full/2/2/105/ DC1/.

Insulin secretion from in situ-perfused pancreata Pancreata were perfused via the aorta at the celiac artery in a humidified, temperature-controlled chamber, as described previously (Johnson et al., 2003). The perfusate consisted of oxygenated Krebs-Ringer bicarbonate buffer (KRB), containing 119 mM NaCl, 4.7 mM KCl, 25 mM NaHCO3, 2.5 mM CaCl2, 1.2 mM MgSO4, 1.2 mM KH2PO4, and 0.25% radioimmunoassay grade BSA with either 2 or 25 mM glucose or 25 mM KCl. The perfusion was maintained at a constant rate of 1 ml/min using Minipuls 3 peristaltic pumps. Prior to sample collection, the pancreas was perfused with 2 mM glucose/KRB for 45 min. The effluent was then collected at 1, 2, or 5

We thank Kenneth Polonsky for technical suggestions and helpful discussions about islet physiology, Laura Todt for transgene construction, Chris Welling and Mitsuru Ohsugi for help with IPGTTs and MIN6 cell culture, Sharon Travis for insulin measurements, Yiyong Zhou and Seth Crosby for help with microarray analysis, Jun-ichi Miyazaki for providing MIN6 cells and pIns-1, and David Turner for mU6pro. We also thank Irving Boime for encouragement, Laura Bordone and Leonard Guarente for sharing unpublished results, and members of the Imai and Permutt laboratories for their help and discussions. This work was supported by grants from NIA (AG024150); NIDDK through the Washington University Diabetes Research

CELL METABOLISM : AUGUST 2005

Acknowledgments

115

A R T I C L E

Training Center (DK20579); the National Center for Research Resources (C06 RR015502); and the Washington University Center for Aging (to S.-i.I.), the Lucille P. Markey Special Emphasis Pathway in Human Pathology (to K.A.M.), and the Glenn/AFAR Scholarship for Research in the Biology of Aging (to K.A.M. and A.A.G.).

Received: January 6, 2005 Revised: April 16, 2005 Accepted: July 1, 2005 Published: August 16, 2005

J.G., Zipkin, R.E., Chung, P., Kisielewski, A., Zhang, L.L., et al. (2003). Small molecule activators of sirtuins extend Saccharomyces cerevisiae lifespan. Nature 425, 191–196. Imai, S., Armstrong, C.M., Kaeberlein, M., and Guarente, L. (2000). Transcriptional silencing and longevity protein Sir2 is an NAD-dependent histone deacetylase. Nature 403, 795–800. Iozzo, P., Beck-Nielsen, H., Laakso, M., Smith, U., Yki-Jarvinen, H., and Ferrannini, E. (1999). Independent influence of age on basal insulin secretion in nondiabetic humans. European Group for the Study of Insulin Resistance. J. Clin. Endocrinol. Metab. 84, 863–868.

References

Ishihara, H., Wang, H., Drewes, L.R., and Wollheim, C.B. (1999). Overexpression of monocarboxylate transporter and lactate dehydrogenase alters insulin secretory responses to pyruvate and lactate in beta cells. J. Clin. Invest. 104, 1621–1629.

Al-Regaiey, K.A., Masternak, M.M., Bonkowski, M., Sun, L., and Bartke, A. (2004). Long-lived growth hormone receptor knockout mice: interaction of reduced IGF-1/insulin signaling and caloric restriction. Endocrinology 146, 851–860. Published online October 21, 2004. 10.1210/en.2004–1120.

Johnson, J.D., Ahmed, N.T., Luciani, D.S., Han, Z., Tran, H., Fujita, J., Misler, S., Edlund, H., and Polonsky, K.S. (2003). Increased islet apoptosis in Pdx1+/− mice. J. Clin. Invest. 111, 1147–1160.

Amaral, M.E., Cunha, D.A., Anhe, G.F., Ueno, M., Carneiro, E.M., Velloso, L.A., Bordin, S., and Boschero, A.C. (2004). Participation of prolactin receptors and phosphatidylinositol 3-kinase and MAP kinase pathways in the increase in pancreatic islet mass and sensitivity to glucose during pregnancy. J. Endocrinol. 183, 469–476. Anderson, R.M., Bitterman, K.J., Wood, J.G., Medvedik, O., and Sinclair, D.A. (2003). Nicotinamide and PNC1 govern lifespan extension by calorie restriction in Saccharomyces cerevisiae. Nature 423, 181–185. Astrom, S.U., Cline, T.W., and Rine, J. (2003). The Drosophila melanogaster sir2+ gene is nonessential and has only minor effects on position-effect variegation. Genetics 163, 931–937. Basu, R., Breda, E., Oberg, A.L., Powell, C.C., Dalla Man, C., Basu, A., Vittone, J.L., Klee, G.G., Arora, P., Jensen, M.D., et al. (2003). Mechanisms of the age-associated deterioration in glucose tolerance: contribution of alterations in insulin secretion, action, and clearance. Diabetes 52, 1738– 1748. Bernal-Mizrachi, E., Wen, W., Stahlhut, S., Welling, C.M., and Permutt, M.A. (2001). Islet β cell expression of constitutively active Akt1/PKBα induces striking hypertrophy, hyperplasia, and hyperinsulinemia. J. Clin. Invest. 108, 1631–1638. Bordone, L., and Guarente, L. (2005). Calorie restriction, SIRT1 and metabolism: understanding longevity. Nat. Rev. Mol. Cell Biol. 6, 298–305. Chan, C.B., Saleh, M.C., Koshkin, V., and Wheeler, M.B. (2004). Uncoupling protein 2 and islet function. Diabetes Suppl. 1 53, S136–S142. Chang, A.M., and Halter, J.B. (2003). Aging and insulin secretion. Am. J. Physiol. Endocrinol. Metab. 284, E7–12. Cohen, H.Y., Miller, C., Bitterman, K.J., Wall, N.R., Hekking, B., Kessler, B., Howitz, K.T., Gorospe, M., De Cabo, R., and Sinclair, D.A. (2004). Calorie restriction promotes mammalian cell survival by inducing the SIRT1 deacetylase. Science 305, 390–392. Freemark, M., Avril, I., Fleenor, D., Driscoll, P., Petro, A., Opara, E., Kendall, W., Oden, J., Bridges, S., Binart, N., et al. (2002). Targeted deletion of the PRL receptor: effects on islet development, insulin production, and glucose tolerance. Endocrinology 143, 1378–1385. Girish, V., and Vijayalakshmi, A. (2004). Affordable image analysis using NIH Image/ImageJ. Indian J. Cancer 41, 47. Guarente, L., and Kenyon, C. (2000). Genetic pathways that regulate ageing in model organisms. Nature 408, 255–262. Guillam, M.T., Dupraz, P., and Thorens, B. (2000). Glucose uptake, utilization, and signaling in GLUT2-null islets. Diabetes 49, 1485–1491. Hotta, M., Tashiro, F., Ikegami, H., Niwa, H., Ogihara, T., Yodoi, J., and Miyazaki, J.-I. (1998). Pancreatic β cell-specific expression of thioredoxin, an antioxidative and antiapoptotic protein, prevents autoimmune and streptozotocin-induced diabetes. J. Exp. Med. 188, 1445–1451. Howitz, K.T., Bitterman, K.J., Cohen, H.Y., Lamming, D.W., Lavu, S., Wood,

116

Joseph, J.W., Koshkin, V., Zhang, C.Y., Wang, J., Lowell, B.B., Chan, C.B., and Wheeler, M.B. (2002). Uncoupling protein 2 knockout mice have enhanced insulin secretory capacity after a high-fat diet. Diabetes 51, 3211– 3219. Kaeberlein, M., McVey, M., and Guarente, L. (1999). The SIR2/3/4 complex and SIR2 alone promote longevity in Saccharomyces cerevisiae by two different mechanisms. Genes Dev. 13, 2570–2580. Landry, J., Slama, J.T., and Sternglanz, R. (2000). Role of NAD+ in the deacetylase activity of the SIR2-like proteins. Biochem. Biophys. Res. Commun. 278, 685–690. Li, Z.W., Ding, J.P., Kalyanaraman, V., and Lingle, C.J. (1999). RINm5f cells express inactivating BK channels whereas HIT cells express noninactivating BK channels. J. Neurophysiol. 81, 611–624. Li, C., Chen, P., Vaughan, J., Blount, A., Chen, A., Jamieson, P.M., Rivier, J., Smith, M.S., and Vale, W. (2003). Urocortin III is expressed in pancreatic beta-cells and stimulates insulin and glucagon secretion. Endocrinology 144, 3216–3224. Lin, S.-J., Defossez, P.-A., and Guarente, L. (2000). Life span extension by calorie restriction in S. cerevisiae requires NAD and SIR2. Science 289, 2126–2128. Lin, S.-J., Kaeberlein, M., Andalis, A.A., Sturtz, L.A., Defossez, P.-A., Culotta, V.C., Fink, G.R., and Guarente, L. (2002). Calorie restriction extends Saccharomyces cerevisiae lifespan by increasing respiration. Nature 418, 344–348. Lin, S.-J., Ford, E., Haigis, M., Liszt, G., and Guarente, L. (2004). Calorie restriction extends yeast life span by lowering the level of NADH. Genes Dev. 18, 12–16. Ma, D., Shield, J.P., Dean, W., Leclerc, I., Knauf, C., Burcelin, R.R., Rutter, G.A., and Kelsey, G. (2004). Impaired glucose homeostasis in transgenic mice expressing the human transient neonatal diabetes mellitus locus, TNDM. J. Clin. Invest. 114, 339–348. McBurney, M.W., Yang, X., Jardine, K., Hixon, M., Boekelheide, K., Webb, J.R., Lansdorp, P.M., and Lemieux, M. (2003). The mammalian SIR2alpha protein has a role in embryogenesis and gametogenesis. Mol. Cell. Biol. 23, 38–54. McDaniel, M.L., Marshall, C.A., Pappan, K.L., and Kwon, G. (2002). Metabolic and autocrine regulation of the mammalian target of rapamycin by pancreatic β-cells. Diabetes 51, 2877–2885. Miyazaki, J.-I., Araki, K., Yamato, E., Ikegami, H., Asano, T., Shibasaki, Y., Oka, Y., and Yamamura, K. (1990). Establishment of a pancreatic beta cell line that retains glucose-inducible insulin secretion: special reference to expression of glucose transporter isoforms. Endocrinology 127, 126–132. Moller, N., Gormsen, L., Fuglsang, J., and Gjedsted, J. (2003). Effects of ageing on insulin secretion and action. Horm. Res. 60, 102–104. Muzumdar, R., Ma, X., Atzmon, G., Vuguin, P., Yang, X., and Barzilai, N.

CELL METABOLISM : AUGUST 2005

Sirt1 regulates pancreatic β cell function in mice

(2004). Decrease in glucose-stimulated insulin secretion with aging is independent of insulin action. Diabetes 53, 441–446.

Tatar, M., Bartke, A., and Antebi, A. (2003). The endocrine regulation of aging by insulin-like signals. Science 299, 1346–1351.

Nemoto, S., Fergusson, M.M., and Finkel, T. (2004). Nutrient availability regulates SIRT1 through a forkhead-dependent pathway. Science 306, 2105– 2108.

Tissenbaum, H.A., and Guarente, L. (2001). Increased dosage of a sir-2 gene extends lifespan in Caenorhabditis elegans. Nature 410, 227–230.

Picard, F., Kurtev, M., Chung, N., Topark-Ngarm, A., Senawong, T., Oliveira, R.M., Leid, M., McBurney, M.W., and Guarente, L. (2004). Sirt1 promotes fat mobilization in white adipocytes by repressing PPAR-γ. Nature 429, 771–776.

Wicksteed, B., Alarcon, C., Briaud, I., Lingohr, M.K., and Rhodes, C.J. (2003). Glucose-induced translational control of proinsulin biosynthesis is proportional to preproinsulin mRNA levels in islet beta-cells but not regulated via a positive feedback of secreted insulin. J. Biol. Chem. 278, 42080–42090.

Rodgers, J.T., Lerin, C., Haas, W., Gygi, S.P., Spiegelman, B.M., and Puigserver, P. (2005). Nutrient control of glucose homeostasis through a complex of PGC-1alpha and SIRT1. Nature 434, 113–118.

Wood, J.G., Rogina, B., Lavu, S., Howitz, K., Helfand, S.L., Tatar, M., and Sinclair, D. (2004). Sirtuin activators mimic caloric restriction and delay ageing in metazoans. Nature 430, 686–689.

Rogina, B., and Helfand, S.L. (2004). Sir2 mediates longevity in the fly through a pathway related to calorie restriction. Proc. Natl. Acad. Sci. USA 101, 15998–16003.

Yamamoto, K., Hashimoto, H., Tomimoto, S., Shintani, N., Miyazaki, J.I., Tashiro, F., Aihara, H., Nammo, T., Li, M., Yamagata, K., et al. (2003). Overexpression of PACAP in transgenic mouse pancreatic β-cells enhances insulin secretion and ameliorates streptozotocin-induced diabetes. Diabetes 52, 1155–1162.

Smith, J.S., Brachmann, C.B., Celic, I., Kenna, M.A., Muhammad, S., Starai, V.J., Avalos, J.L., Escalante-Semerena, J.C., Grubmeyer, C., Wolberger, C., and Boeke, J.D. (2000). A phylogenetically conserved NAD+-dependent protein deacetylase activity in the Sir2 protein family. Proc. Natl. Acad. Sci. USA 97, 6658–6663. Spinas, G.A., Laffranchi, R., Francoys, I., David, I., Richter, C., and Reinecke, M. (1998). The early phase of glucose-stimulated insulin secretion requires nitric oxide. Diabetologia 41, 292–299.

CELL METABOLISM : AUGUST 2005

Yaney, G.C., and Corkey, B.E. (2003). Fatty acid metabolism and insulin secretion in pancreatic beta cells. Diabetologia 46, 1297–1312. Zhang, C.Y., Baffy, G., Perret, P., Krauss, S., Peroni, O., Grujic, D., Hagen, T., Vidal-Puig, A.J., Boss, O., Kim, Y.B., et al. (2001). Uncoupling protein-2 negatively regulates insulin secretion and is a major link between obesity, beta cell dysfunction, and type 2 diabetes. Cell 105, 745–755.

117