Revisiting the immunocytochemical detection of Neurogenin 3 expression in mouse and man.

During embryonic development, endocrine cells of the pancreas are specified from multipotent progenitors. The transcription factor Neurogenin 3 (NEUROG3) is critical for this development and it has been shown that all endocrine cells of the pancreas arise from endocrine progenitors expressing NEUROG3. A thorough understanding of the role of NEUROG3 during development, directed differentiation of pluripotent stem cells and in models of cellular reprogramming, will guide future efforts directed at finding novel sources of ß-cells for cell replacement therapies. In this article, we review the expression and function of NEUROG3 in both mouse and human and present the further characterization of a monoclonal antibody directed against NEUROG3. This antibody has been previously been used for detection of both mouse and human NEUROG3. However, our results suggest that the epitope recognized by this antibody is specific to mouse NEUROG3. Thus, we have also generated a monoclonal antibody specifically recognizing human NEUROG3 and present the characterization of this antibody here. Together, these antibodies will provide useful tools for future studies of NEUROG3 expression, and the data presented in this article suggest that recently described expression patterns of NEUROG3 in human foetal and adult pancreas should be re-examined.

Neurogenin 3+ cells contribute to ß-cell neogenesis and proliferation in injured adult mouse pancreas.

We previously showed that injury by partial duct ligation (PDL) in adult mouse pancreas activates Neurogenin 3 (Ngn3)(+) progenitor cells that can differentiate to ß cells ex vivo. Here we evaluate the role of Ngn3(+) cells in ß cell expansion in situ. PDL not only induced doubling of the ß cell volume but also increased the total number of islets. ß cells proliferated without extended delay (the so-called 'refractory' period), their proliferation potential was highest in small islets, and 86% of the ß cell expansion was attributable to proliferation of pre-existing ß cells. At sufficiently high Ngn3 expression level, upto 14% of all ß cells and 40% of small islet ß cells derived from non-ß cells. Moreover, ß cell proliferation was blunted by a selective ablation of Ngn3(+) cells but not by conditional knockout of Ngn3 in pre-existing ß cells supporting a key role for Ngn3(+) insulin(-) cells in ß cell proliferation and expansion. We conclude that Ngn3(+) cell-dependent proliferation of pre-existing and newly-formed ß cells as well as reprogramming of non-ß cells contribute to in vivo ß cell expansion in the injured pancreas of adult mice.

Pancreatic islet and progenitor cell surface markers with cell sorting potential.

AIMS/HYPOTHESIS: The aim of the study was to identify surface bio-markers and corresponding antibody tools that can be used for the imaging and immunoisolation of the pancreatic beta cell and its progenitors. This may prove essential to obtain therapeutic grade human beta cells via stem cell differentiation. METHODS: Using bioinformatics-driven data mining, we generated a gene list encoding putative plasma membrane proteins specifically expressed at distinct stages of the developing pancreas and islet beta cells. In situ hybridisation and immunohistochemistry were used to further prioritise and identify candidates. RESULTS: In the developing pancreas seizure related 6 homologue like (SEZ6L2), low density lipoprotein receptor-related protein 11 (LRP11), dispatched homologue 2 (Drosophila) (DISP2) and solute carrier family 30 (zinc transporter), member 8 (SLC30A8) were found to be expressed in early islet cells, whereas discoidin domain receptor tyrosine kinase 1 (DDR1) and delta/notch-like EGF repeat containing (DNER) were expressed in early pancreatic progenitors. The expression pattern of DDR1 overlaps with the early pancreatic and duodenal homeobox 1 (PDX1)⁺/NK6 homeobox 1 (NKX6-1)⁺ multipotent progenitor cells from embryonic day 11, whereas DNER expression in part overlaps with neurogenin 3 (NEUROG3)⁺ cells. In the adult pancreas SEZ6L2, LRP11, DISP2 and SLC30A8, but also FXYD domain containing ion transport regulator 2 (FXYD2), tetraspanin 7 (TSPAN7) and transmembrane protein 27 (TMEM27), retain an islet-specific expression, whereas DDR1 is undetectable. In contrast, DNER is expressed at low levels in peripheral mouse and human islet cells. Re-expression of DDR1 and upregulation of DNER is observed in duct-ligated pancreas. Antibodies to DNER and DISP2 have been successfully used in cell sorting. CONCLUSIONS/INTERPRETATION: Extracellular epitopes of SEZ6L2, LRP11, DISP2, DDR1 and DNER have been identified as useful tags by applying specific antibodies to visualise pancreatic cell types at specific stages of development. Furthermore, antibodies recognising DISP2 and DNER are suitable for FACS-mediated cell purification.

Homozygous carriers of the G allele of rs4664447 of the glucagon gene (GCG) are characterised by decreased fasting and stimulated levels of insulin, glucagon and glucagon-like peptide (GLP)-1.

AIMS/HYPOTHESIS: The glucagon gene (GCG) encodes several hormones important for energy metabolism: glucagon, oxyntomodulin and glucagon-like peptide (GLP)-1 and -2. Variants in GCG may associate with type 2 diabetes, obesity and/or related metabolic traits. METHODS: GCG was re-sequenced as a candidate gene in 865 European individuals. Twenty-nine variants were identified. Four variants that were considered to have a likelihood for altered functionality: rs4664447, rs7581952, Ile158Val and Trp169Ter, were genotyped in 17,584 Danes. RESULTS: When examined in 5,760 treatment-naive individuals, homozygous carriers of the low frequency (minor allele frequency 2.3%) G allele of rs4664447, predicted to disrupt an essential splice enhancer binding site, had lower levels of fasting plasma glucose (mean ± SD, 4.8 ± 1.2 vs 5.5 ± 0.8 mmol/l, p = 0.004); fasting serum insulin (22 ± 14 vs 42 ± 27 pmol/l, p = 0.04); glucose-stimulated serum insulin (159 ± 83 vs 290 ± 183 pmol/l, p = 0.01) and adult height (165 ± 10 vs 172 ± 9 cm, p = 0.0009) compared with A allele carriers. During oral glucose tolerance and hyperglycaemic arginine stimulation tests, the plasma AUC for GLP-1 (730 ± 69 vs 1,334 ± 288 pmol/l × min, p = 0.0002) and basal and stimulated levels of serum insulin and plasma glucagon were ∼50% decreased (p < 0.001) among three homozygous carriers compared with nine matched wild-type carriers. rs7581952, Ile158Val and Trp169Ter (where 'Ter' indicates 'termination') variants of GCG did not significantly associate or co-segregate with the metabolic traits examined. CONCLUSIONS/INTERPRETATION: Re-sequencing of GCG revealed a low frequency intronic variant, rs4664447, and follow-up physiological studies suggest that this variant in homozygous form may cause decreased fasting and stimulated levels of insulin, glucagon and GLP-1. Overall, our findings suggest that variation in GCG has no major impact on carbohydrate metabolism in the study populations examined.

Foetal proglucagon processing in relation to adult appetite control: lessons from a transplantable rat glucagonoma with severe anorexia.

We have previously reported severe anorexia abruptly induced in rats 2-3 weeks after they have been transplanted subcutaneously with the glucagonoma MSL-G-AN. Vagotomy did not affect the time of onset and severity of anorexia, and the anorectic state resembles hunger with strongly elevated neuropeptide Y (NPY) mRNA levels in the nucleus arcuatus. We now show that circulating levels of bioactive glucagon-like peptide-1 (GLP-1) (7-36amide) start to increase above control levels exactly at the time of onset of anorexia. At this time-point, bioactive glucagon as well as total glucagon precursors and GLP-1 metabolites are already vastly elevated compared to controls. We further show that intravenous administration of very high concentrations of GLP-1 to hungry schedule-fed rats causes anorexia in a dose-dependent manner, which is blocked by the GLP-1 receptor antagonist exendin (9-39). GLP-1 (7-36amide) has a well-characterized anorectic effect but also causes taste aversion when administered centrally. The anorectic effect is blocked in rats treated neonatally by monosodium glutamate (MSG). We show that MSG treatment does not prevent the MSL-G-AN-induced anorexia, thereby suggesting a different type of anorectic function. We show a very strong component of taste aversion as anorectic rats, when presented to novel or known alternative food items, will resume normal feeding for 1 day, and then redevelop anorexia. We hypothetize that the anorexia in MSL-G-AN tumour-bearing rats correlates with a foetal processing pattern of proglucagon to both glucagon and GLP-1 (7-36amide), and is due to taste aversion. The sudden onset is characterized by a dramatic increase in circulating levels of biologically active GLP-1 (7-36amide), suggesting eventual saturation of proteolytic inactivation of its N-terminus.

Novel monoclonal antibodies against Pdx1 reveal feedback regulation of Pdx1 protein levels.

The aim of this study was to characterize two monoclonal antibodies (F6A11 and F109-D12) generated against Pdx1 (pancreatic and duodenal homeobox-1), a homeodomain transcription factor, which is critical for pancreas formation as well as for normal pancreatic beta cell function. For production of monoclonal antibodies, we immunized Robertsonian POSF (RBF)mice with a GST-Pdx1 fusion protein containing a 68-amino acid C-terminal fragment of rat Pdx1. These monoclonal antibodies detect Pdx1 by western blotting and allow immunohistochemical detection of Pdx1 in both mouse and rat tissue. F6A11 and F109-D12 produce IHC staining patterns indistinguishable from that obtained with highly specific polyclonal Pdx1 antisera raised in rabbits and goats, when applied to embryonic or adult mouse pancreatic tissue. In contrast to previously generated polyclonal anti-Pdx1 antisera, we also demonstrate that F6A11 works for intracellular fluorescence activated cell sorting (FACS) staining of Pdx1. By using F6A11, we characterize the induction of Pdx1 in the Doxycycline (DOX) inducible insulinoma cell line INSralphabeta-Pdx1 and follow the reduction of Pdx1 after removing Dox. Finally, we show that induction of exogenous Pdx1 leads to a reduction in endogenous Pdx1 levels, which suggests that a negative feedback loop is involved in maintaining correct levels of Pdx1 in the cell.

Improved glucose tolerance and acinar dysmorphogenesis by targeted expression of transcription factor PDX-1 to the exocrine pancreas.

The homeodomain protein PDX-1 is critical for pancreas development and is a key regulator of insulin gene expression. PDX-1 nullizygosity and haploinsufficiency in mice and humans results in pancreatic agenesis and diabetes, respectively. At embryonic day (e) 10.5, PDX-1 is expressed in all pluripotential gut-derived epithelial cells destined to differentiate into the exocrine and endocrine pancreas. At e15, PDX-1 expression is downregulated in exocrine cells, but remains high in endocrine cells. The aim of this study was to determine whether targeted overexpression of PDX-1 to the exocrine compartment of the developing pancreas at e15 would allow for respecification of the exocrine cells. Transgenic (TG) mice were generated in which PDX-1 was expressed in the exocrine pancreas using the exocrine-specific elastase-1 promoter. These mice exhibited a marked dysmorphogenesis of the exocrine pancreas, manifested by increased rates of replication and apoptosis in acinar cells and a progressive fatty infiltration of the exocrine pancreas with age. Interestingly, the TG mice exhibited improved glucose tolerance, but absolute beta-cell mass was not increased. These findings indicate that downregulation of PDX-1 is required for the proper maintenance of the exocrine cell phenotype and that upregulation of PDX-1 in acinar cells affects beta-cell function. The mechanisms underlying these observations remain to be elucidated.

Pax4 represses pancreatic glucagon gene expression.

The paired box and homeodomain containing transcription factors Pax4 and Pax6 are known to be essential for development of the pancreatic endocrine cells. In this report we demonstrate that stable expression of Pax4 in a rat glucagon-producing cell line inhibits the endogenously expressed glucagon gene completely. Furthermore, Pax4 represses Pax6 independent transcription of the insulin promoter, suggesting that Pax4 can actively repress transcription in addition to acting by competition with the transcriptional activator Pax6.

Independent development of pancreatic alpha- and beta-cells from neurogenin3-expressing precursors: a role for the notch pathway in repression of premature differentiation.

The nature and identity of the pancreatic beta-cell precursor has remained elusive for many years. One model envisions an early multihormonal precursor that gives rise to both alpha- and beta-cells and the other endocrine cell types. Alternatively, beta-cells have been suggested to arise late, directly from the GLUT2- and pancreatic duodenal homeobox factor-1 (PDX1)-expressing epithelium, which gives rise also to the acinar cells during this stage. In this study, we have identified a subset of the PDX1+ epithelial cells that are marked by expression of Neurogenin3 (Ngn3). Ngn3, a member of the basic helix-loop-helix (bHLH) family of transcription factors, is suggested to act upstream of NeuroD in a bHLH cascade. Detailed analysis of Ngn3/paired box factor 6 (PAX6) and NeuroD/PAX6 co-expression shows that the two bHLH factors are expressed in a largely nonoverlapping set of cells, but such analysis also suggests that the NeuroD+ cells arise from cells expressing Ngn3 transiently. NeuroD+ cells do not express Ki-67, a marker of proliferating cells, which shows that these cells are postmitotic. In contrast, Ki-67 is readily detected in Ngn3+ cells. Thus, Ngn3+ cells fulfill the criteria for an endocrine precursor cell. These expression patterns support the notion that both alpha- and beta-cells develop independently from PDX1+/Ngn3+ epithelial cells, rather than from GLU+/INS+ intermediate stages. The earliest sign of alpha-cell development appears to be Brain4 expression, which apparently precedes Islet-1 (ISL1) expression. Based on our expression analysis, we propose a temporal sequence of gene activation and inactivation for developing alpha- and beta-cells beginning with activation of NeuroD expression. Endocrine cells leave the cell cycle before NeuroD activation, but re-enter the cell cycle at perinatal stages. Dynamic expression of Notch1 in PDX+ epithelial cells suggests that Notch signaling could inhibit a Ngn-NeuroD cascade as seen in the nervous system and thus prevent premature differentiation of endocrine cells.

Transcription factor hepatocyte nuclear factor 6 regulates pancreatic endocrine cell differentiation and controls expression of the proendocrine gene ngn3.

Hepatocyte nuclear factor 6 (HNF-6) is the prototype of a new class of cut homeodomain transcription factors. During mouse development, HNF-6 is expressed in the epithelial cells that are precursors of the exocrine and endocrine pancreatic cells. We have investigated the role of HNF-6 in pancreas differentiation by inactivating its gene in the mouse. In hnf6(-/-) embryos, the exocrine pancreas appeared to be normal but endocrine cell differentiation was impaired. The expression of neurogenin 3 (Ngn-3), a transcription factor that is essential for determination of endocrine cell precursors, was almost abolished. Consistent with this, we demonstrated that HNF-6 binds to and stimulates the ngn3 gene promoter. At birth, only a few endocrine cells were found and the islets of Langerhans were missing. Later, the number of endocrine cells increased and islets appeared. However, the architecture of the islets was perturbed, and their beta cells were deficient in glucose transporter 2 expression. Adult hnf6(-/-) mice were diabetic. Taken together, our data demonstrate that HNF-6 controls pancreatic endocrine differentiation at the precursor stage and identify HNF-6 as the first positive regulator of the proendocrine gene ngn3 in the pancreas. They also suggest that HNF-6 is a candidate gene for diabetes mellitus in humans.

Control of endodermal endocrine development by Hes-1.

Development of endocrine cells in the endoderm involves Atonal and Achaete/Scute-related basic helix-loop-helix (bHLH) proteins. These proteins also serve as neuronal determination and differentiation factors, and are antagonized by the Notch pathway partly acting through Hairy and Enhancer-of-split (HES)-type proteins. Here we show that mice deficient in Hes1 (encoding Hes-1) display severe pancreatic hypoplasia caused by depletion of pancreatic epithelial precursors due to accelerated differentiation of post-mitotic endocrine cells expressing glucagon. Moreover, upregulation of several bHLH components is associated with precocious and excessive differentiation of multiple endocrine cell types in the developing stomach and gut, showing that Hes-1 operates as a general negative regulator of endodermal endocrine differentiation.

Beta-cell maturation leads to in vitro sensitivity to cytotoxins.

Pancreatic beta-cells are more sensitive to several toxins (e.g., streptozotocin, alloxan, cytokines) than the other three endocrine cell types in the islets of Langerhans. Cytokine-induced free radicals in beta-cells may be involved in beta-cell-specific destruction in type 1 diabetes. To investigate if this sensitivity represents an acquired trait during beta-cell maturation, we used two in vitro cultured cell systems: 1) a pluripotent glucagon-positive pre-beta-cell phenotype (NHI-glu) that, after in vivo passage, matures into an insulin-producing beta-cell phenotype (NHI-ins) and 2) a glucagonoma cell-type (AN-glu) that, after stable transfection with pancreatic duodenal homeobox factor-1 (PDX-1), acquires the ability to produce insulin (AN-ins). After exposure to interleukin (IL)-1beta, both of the insulin-producing phenotypes were significantly more susceptible to toxic effects than their glucagon-producing counterparts. Nitric oxide (NO) production was induced in both NHI phenotypes, and inhibition with 0.5 mmol/l N(G)-monomethyl-L-arginine (NMMA) fully protected the cells. In addition, maturation into the NHI-ins phenotype was associated with an acquired dose-dependent sensitivity to the toxic effect of streptozotocin. Our results support the hypothesis that the exquisite sensitivity of beta-cells to IL-1beta and streptozotocin is an acquired trait during beta-cell maturation. These two cell systems will be useful tools for identification of molecular mechanisms involved in beta-cell maturation and sensitivity to toxins in relation to type 1 diabetes.

Cloning and DNA-binding properties of the rat pancreatic beta-cell-specific factor Nkx6.1.

The homeodomain (HD) protein Nkx6.1 is the most beta-cell-specific transcription factor known in the pancreas and its function is critical for the formation of the insulin-producing beta-cells. However, the target genes, DNA-binding site, and transcriptional properties of Nkx6.1 are unknown. Using in vitro binding site selection we have identified the DNA sequence of the Nkx6.1 binding site to be TTAATTG/A. A reporter plasmid containing four copies of this sequence is activated by an Nkx6.1HD/VP16 fusion construct. Full-length Nkx6.1 fails to activate this reporter plasmid in spite of robust interaction with the binding site in vitro. Stable expression of Nkx6.1 in the glucagon-producing alpha-cell-like MSL-G-AN cells induces expression of the endogenous insulin gene in a subset of the cell population. The expression of other known beta-cell-specific factors such as Pax4, Pax6, Pdx1, GLUT2 and GLP1-R is unchanged by the introduction of Nkx6.1.

Gastric amylin expression. Cellular identity and lack of requirement for the homeobox protein PDX-1. A study in normal and PDX-1-deficient animals with a cautionary note on antiserum evaluation.

The gene encoding amylin is implicated in the generation of amyloid in the islets of Langerhans of diabetics and is believed to be regulated by the homeodomain transcription factor PDX-1. Although gastric mucosa also produces amylin, studies on its cellular site of production have yielded highly divergent results, localizing this peptide to either gastrin, serotonin, or somatostatin cells or to combinations thereof. Using region-specific amylin antisera in combination with reverse transcriptase-polymerase chain reaction, we now document that the majority of cells expressing amylin correspond to somatostatin cells. Only a small subpopulation of gastrin cells contained immunoreactive amylin. Studies of PDX-1-deficient mice, which fail to develop gastrin cells while possessing normal numbers of somatostatin cells, revealed no detectable change in gastric amylin expression. These data show that neither normal gastrin cell development nor PDX-1 expression is needed for gastric amylin expression.

The hypothalamic satiety peptide CART is expressed in anorectic and non-anorectic pancreatic islet tumors and in the normal islet of Langerhans.

The hypothalamic satiety peptide CART (cocaine and amphetamine regulated transcript) is expressed at high levels in anorectic rat glucagonomas but not in hypoglycemic insulinomas. However, a non-anorectic metastasis derived from the glucagonoma retained high CART expression levels and produced circulating CART levels comparable to that of the anorectic tumors. Moreover, distinct glucagonoma lines derived by stable HES-1 transfection of the insulinoma caused severe anorexia but retained low circulating levels of CART comparable to that of insulinoma bearing or control rats. Islet tumor associated anorexia and circulating CART levels are thus not correlated, and in line with this peripheral administration of CART (5-50 mg/kg) produced no effect on feeding behavior. In the rat two alternatively spliced forms of CART mRNA exist and quantitative PCR revealed expression of both forms in the hypothalamus, in the different islet tumors, and in the islets of Langerhans. Immunocytochemistry as well as in situ hybridization localized CART expression to the somatostatin producing islet D cell. A potential endocrine/paracrine role of islet CART remains to be clarified.

Pax6 and Cdx2/3 form a functional complex on the rat glucagon gene promoter G1-element.

Alpha-cell specific transcription of the glucagon gene is mainly conferred by the glucagon promoter G1-element, while additional elements G2, G3, and G4 have broad islet cell specificity. Transcription of the glucagon gene has been shown to be stimulated by Pax6 through binding to the glucagon gene promoter G3-element. In this report, we show that Pax6 additionally binds the glucagon gene promoter G1-element and forms a transcriptionally active complex with another homeodomain protein, Cdx2/3. Two distinct mutations in the G1-element, that both reduce promoter activity by 85-90%, is shown to eliminate binding of either Pax6 or Cdx2/3. Additionally, Pax6 enhanced Cdx2/3 mediated activation of a glucagon reporter in heterologous cells. We discuss how Pax6 may contribute to cell-type specific transcription in the pancreatic islets by complex formation with different transcription factors.

Pax6 and Pdx1 form a functional complex on the rat somatostatin gene upstream enhancer.

The somatostatin upstream enhancer (SMS-UE) is a highly complex enhancer element. The distal A-element contains overlapping Pdx1 and Pbx binding sites. However, a point mutation in the A-element that abolishes both Pdxl and Pbx binding does not impair promoter activity. In contrast, a point mutation that selectively eliminates Pdx1 binding to a proximal B-element reduces the promoter activity. The B-element completely overlaps with a Pax6 binding site, the C-element. A point mutation in the C-element demonstrates that Pax6 binding is essential for promoter activity. Interestingly, a block mutation in the A-element reduces both Pax6 binding and promoter activity. In heterologous cells, Pdx1 potentiated Pax6 mediated activation of a somatostatin reporter. We conclude that the beta/delta-cell-specific activity of the SMS-UE is achieved through simultaneous binding of Pdx1 and Pax6 to the B- and C-elements, respectively. Furthermore, the A-element appears to stabilise Pax6 binding.

Human islets in mixed islet grafts protect mouse pancreatic beta-cells from alloxan toxicity.

We have previously shown that human beta-cells are resistant to the toxic effects of alloxan. In order to further clarify this characteristic of human islets, we investigated whether these cells might transfer their alloxan resistance to alloxan-sensitive rat or mouse islets. Islets from two species (human-mouse or rat-mouse) were mixed into one graft, which was implanted into the subcapsular kidney space of nude mice. Alloxan or saline was injected intravenously two weeks after implantation and one week thereafter the mice were killed. The number of grafted and endogenous beta-cells were evaluated by a semi-quantitative method after immunohistochemistry. Human islet production of the scavenging enzymes extracellular superoxide dismutase and plasma glutathione peroxidase were analyzed with ELISA-techniques, and mouse and human islet hydrogen peroxide breakdown activity were monitored with a horseradish peroxidase-dependent assay. Mouse beta-cells transplanted together with human islets were protected against alloxan cytotoxicity. Rat islets did not protect mouse beta-cells against alloxan, suggesting that the mixing procedure as such did not impose the protection. Production of extracellular superoxide dismutase and plasma glutathione peroxidase by human islets was very low. Moreover, H2O2 breakdown in vitro, did not differ between human and mouse islets. Alloxan-insensitive human islets protect mouse beta-cells against alloxan-induced lesions, suggesting that yet to be identified extracellular factors are involved in human islet resistance to alloxan toxicity.