|Home | About | Journals | Submit | Contact Us | Français|
Conceived and designed the experiments: SP MH DAC. Performed the experiments: SP AGN DAC. Analyzed the data: SP AGN MH DAC. Wrote the paper: SP DAC. Reviewed/edited the manuscript: MH.
Mutations in the human homolog of the Vhlh gene [encoding the von-Hippel Lindau (VHL) protein] lead to tumor development. In mice, depletion of Vhlh in pancreatic ß-cells causes perturbed glucose homeostasis, but the role of this gene in other pancreatic cells is poorly understood. To investigate the function of VHL/HIF pathway in pancreatic cells, we inactivated Vhlh in the pancreatic epithelium as well as in the endocrine and exocrine lineages. Our results show that embryonic depletion of Vhlh within the pancreatic epithelium causes postnatal lethality due to severe hypoglycemia. The hypoglycemia is recapitulated in mice with endocrine-specific removal of Vhlh, while animals with loss of Vhlh predominantly in the exocrine compartment survive to adulthood with no overt defects in glucose metabolism. Mice with hypoglycemia display diminished insulin release in response to elevated glucose. Significantly, the glucagon response is impaired both in vivo (circulating glucagon levels) as well as in an in vitro secretion assay in isolated islets. Hypoxia also impairs glucagon secretion in a glucagon-expressing cell line in culture. Our results reveal a novel role for the hypoxia/HIF pathway in islet hormone secretion and maintenance of the fine balance that allows for the establishment of normoglycemia.
The role of hypoxia in ß-cell formation and function has recently gained considerable interest . Central to the hypoxic response is the transcription factor hypoxia-inducible factor (HIF) whose activity is regulated by proteasomal degradation in the presence of oxygen, a process mediated by the von Hippel-Lindau (VHL) tumor suppressor-containing ubiquitin ligase complex . Several studies, including our own, have uncovered a critical role for the HIF hypoxia response pathway in glucose homeostasis , , . Deletion of Vhlh (the murine homolog of VHL) specifically in the pancreatic ß-cell lineage renders ß-cells unable to respond appropriately to elevated glucose, leading to severe glucose intolerance. Furthermore, hypoxia-responsive genes are upregulated in islets of pre-diabetic Zucker diabetic fatty (ZDF) and diabetic Goto-Kakizaki (GK) rats , . An increase in HIF1α is also observed in islets of diabetic Goto-Kakizaki (GK) rats.
These studies clearly suggest a critical role for VHL/HIF signaling in ß-cell function, raising the question of whether aberrant HIF activation might interfere with the function of other pancreatic cell types. Total pancreas inactivation of Vhlh using a Pdx-1-Cre (pancreatic and duodenal homeobox gene-1 promoter) transgenic mouse line causes glucose intolerance, similar to what has been observed in mice with deletion of Vhlh in ß-cells, but no other pancreatic abnormalities are observed . In contrast, Vhlh inactivation using a different Pdx-1-Cre strain causes neonatal lethality in mice. The cause of lethality in these mice is currently unknown since no obvious pancreatic abnormalities were observed . To resolve these contradictory findings and to further investigate the function of VHL/HIF pathway in pancreatic cells, we inactivated Vhlh using three different pancreatic Cre lines. Our data show that mice with a Vhlh-deficiency in all pancreatic cells die perinatally due to severe hypoglycemia, which appears to result from a defect in glucagon secretion. In agreement with the mouse studies, glucagon-producing cells cultured under hypoxic conditions fail to secrete glucagon in response to reduced glucose levels. Altogether, these results suggest a more general role of the VHL/HIF pathway in endocrine function.
Pdx-1-Creearly, a line with homogeneous Cre expression in all pancreatic cell types in our hands, was crossed into the VhlhloxP/loxP mouse , . In agreement with a previous report, approximately 70% of Pdx-1-Creearly;VhlhLoxP/LoxP pups died around weaning age (Figure S1A) . Gross morphological analyses of pancreata from two week old Pdx-1-Creearly;VhlhLoxP/LoxP pups did not reveal any apparent abnormalities (Figure 1A). Efficient excision of the VhlhloxP/loxP allele by Cre recombinase was evidenced by significant accumulation of HIF1α in islet, acinar and ductal compartments of the pancreata from Pdx-1-Creearly;VhlhLoxP/LoxP animals (Figure 1B). Increased expression of HIFα targets (a functional readout for HIF stabilization) was observed in pancreata from Pdx-1-Creearly;VhlhLoxP/LoxP pups (Figure 1C). Vhlh expression was also reduced in these samples, correlating well with a robust activation of the hypoxic response pathway (Figure 1C). Expression of endocrine (insulin, glucagon and somatostatin) and exocrine (amylase and mucin) markers appeared normal by immunostaining (Figure 1D-F). Importantly, gene expression of ß-cell maturity markers Pdx-1, MafA and Urocortin3 was reduced in mutant islets isolated from animals aged between p13 and p18 as compared to control littermates (Figure 1G). Thus, although there appeared to be delayed maturation of ß-cells, lethality was not due to an overt perturbation in pancreas formation.
Mice with Vhlh loss in ß-cells exhibit glucose intolerance , , , and the reduced expression of maturity markers led us to hypothesize that the perinatal lethality observed in Pdx-1-Creearly;VhlhLoxP/LoxP mice might result from perturbed glucose homeostasis due to compromised cellular function. Surprisingly, blood glucose measurements in the postnatal period revealed that Pdx-1-Creearly;VhlhLoxP/LoxP mice are severely hypoglycemic (Figure 2A). Two lines of evidence suggest that hypoglycemia is the cause and not the consequence of VHL-associated lethality. First, hypoglycemia is observed shortly after birth before the mutants display reduced weight gain compared to control littermates (Figure 2A). Second, the severity of hypoglycemia positively correlates with the poorest survival rates, as pups with milder hypoglycemia live longer (data not shown). To identify pancreatic cell type(s) responsible for the observed perinatal lethality and hypoglycemia in Pdx-1-Creearly;VhlhLoxP/LoxP mice, Vhlh was specifically inactivated in the endocrine lineage using the Ngn3-Cre mouse strain, as Ngn3 marks all endocrine progenitors during embryogenesis . Significantly, Ngn3-Cre;VhlhLoxP/LoxP mice displayed postnatal lethality similar to that observed in Pdx-1-Creearly;VhlhLoxP/LoxP mice, with >75% pups exhibiting blood glucose levels below 20 mg/dl between two and three weeks of age (Figure 2B and data not shown). Endocrine hormones were expressed appropriately in the pancreata of Ngn3-Cre;VhlhLoxP/LoxP mice (data not shown). Recently, concerns have been raised over the extra-pancreatic expression, especially in areas of the brain, of both Pdx-1-Creearly and Ngn3-Cre transgenic lines , . A third Cre-expressing mouse line was employed to tease out the effect of loss of Vhlh in the pancreas. Rfx6 is a transcription factor that is expressed in the gut endoderm during early development and is later restricted to islets in the pancreas, with no confounding brain expression . Rfx6-Cre;VhlhLoxP/LoxP pups exhibited severe postnatal hypoglycemia with associated lethality (Figure 2C). In fact, 50% of the Rfx6-Cre;Vhlh animals are dead by day 10 post-birth while the same is true closer to day 18 after birth for the Pdx-1-Creearly;Vhlh mice (Figure 2C and Figure S1A). Islet architecture was unaffected in the postnatal pancreas of Rfx6-Cre;VhlhLoxP/LoxP pups (Figure S1B). Summarily, our results indicate that VHL loss in endocrine cells during embryogenesis leads to neonatal hypoglycemia and perinatal lethality.
It has been proposed that reduced survival in Pdx-1-Creearly;VhlhLoxP/LoxP mice might be associated with exocrine defects . Cre expression in the Ptf1a-Cre line  is low in endocrine cells, but robust in acinar cells and a subset of ducts (Figure S2A) , thus providing a useful tool to inactivate Vhlh predominantly in the exocrine compartment. Immunohistochemistry for HIF1α confirmed efficient inactivation of Vhlh in the exocrine, but not in the endocrine, compartment (Figure S2B). Ptf1a-Cre;VhlhLoxP/LoxP mice survived to adulthood (>6 months) without any symptoms of compromised health (data not shown), demonstrating that exocrine ablation of VHL does not cause lethality. Notably, Ptf1a-Cre;VhlhLoxP/LoxP animals were normoglycemic at all stages analyzed (Figure S2C). Thus, Vhlh inactivation in endocrine cells, not in the exocrine compartment, causes perinatal lethality.
One obvious cause of hypoglycemia is hyperinsulinemia. Surprisingly, plasma insulin was lower in Pdx-1-Creearly;VhlhLoxP/LoxP and Ngn3-Cre;VhlhLoxP/LoxP pups than in control littermates (Figure 3A). Given the overlapping phenotypes of the three mouse models, we focused subsequent analyses on the Pdx-1-Creearly;VhlhLoxP/LoxP mice. Insulin content within the islet, measured by gene expression and total protein appeared reduced in the Vhlh deficient mice (Figure 3B-C). Insulin secretion from isolated islets showed that, while basal secretion was unaffected, there was a failure of hormone secretion upon exposure to increased glucose in Vhlh mutant mice (Figure 3D). In support of these data, mutant pups showed delayed glucose clearance in response to an intra-peritoneal glucose tolerance test, despite being hypoglycemic at the start of the assay (Figure 3E). Thus, overt hyperinsulinemia does not appear to cause hypoglycemia in Pdx-1-Creearly;VhlhLoxP/LoxP pups.
An alternative explanation for hypoglycemia could be compromised glucagon function. Mutant mice with defective α-cell formation or reduced glucagon production are severely hypoglycemic and display neonatal lethality , , . Robust nuclear HIF1α accumulation was observed in α-cells of Pdx-1-Creearly;VhlhLoxP/LoxP islets (Figure 4A). Quantification of HIF1α-Glucagon costaining revealed an ~41% overlap. However, this number is likely an underestimation of α-cells with active HIF pathway, as cells with cytoplasmic HIF1α accumulation were not scored. Pancreatic control samples were clearly negative for HIF1α staining. No overt defects in α-cell formation, glucagon expression or glucagon production were observed in Pdx-1-Creearly;VhlhLoxP/LoxP mice (Figure 1D and 4B-C). Surprisingly, serum glucagon levels in mutant mice under both fed conditions as well as after an overnight fast were similar to control littermates, pointing to a defective glucagon response, as persistent hypoglycemia should elicit an increase in circulating glucagon (Figure 4D-E). Administration of exogenous glucagon induced a comparable rise in blood glucose in mutant and control mice, indicating that glucagon sensitivity was normal in Pdx-1-Creearly;VhlhLoxP/LoxP mice (Figure S3) and that hypoglycemia was not due to impaired liver response to glucagon, but most likely a result of deficient glucagon secretion. In support of this hypothesis, islets isolated from Pdx-1-Creearly;VhlhLoxP/LoxP pups did not respond appropriately upon exposure to reduced glucose concentration in vitro (Figure 4F). Under conditions of high glucose, mutant islets appeared to secrete higher levels of glucagon when compared to the control islets. However, this difference was not statistically significant (p value=0.067). On the contrary, under conditions of low glucose exposure, the mutant islets secreted less glucagon as compared to the control islets (p value=0.04). Insulin plays an inhibitory role in glucagon secretion, and we tested the role that the reduction in insulin secretion from Vhlh mutant islets might play in glucagon secretion. Insulin was added to control and mutant islets upon incubation under low glucose conditions (Figure 4F). We find that upon incubation of control islets under low glucose with insulin, a suppression of glucagon secretion is detectable (p value=0.02). In the mutant islets, however, there is no significant change in the levels of glucagon secretion with or without insulin under low glucose (p value=0.141). These results indicate that insulin does not inhibit glucagon secretion in Vhlh-deficient islets, and argue that the defect in glucagon secretion in the Vhlh deficient mice is not an indirect effect due to decreased insulin secretion from Vhlh-deficient ß-cells.
Since Vhlh is deleted in all endocrine cell types in Pdx-1-Creearly;VhlhLoxP/LoxP animals, we asked whether the observed α-cell defect was cell-autonomous. Loss of Vhlh mimics a hypoxic state, thus to directly ask whether hormone secretion is blocked under conditions of hypoxia, glucagon-producing α-TC6.1 cells were cultured under normal (20%) and reduced (1%) oxygen levels and glucagon secretion assessed. Glucagon secretion in response to low glucose conditions was impaired when cells were incubated in a hypoxic environment (Figure 5A). Total glucagon content was not significantly altered in the cells incubated under hypoxia (Figure 5B). Cell viability was also determined and no significant change in cell death was observed (Figure 5C). As expected, HIF target genes were significantly upregulated upon hypoxia indicating a robust response of the cells to the low oxygen concentration (Figure 5D). Summarily, our results indicate that secretion of glucagon is impaired in islets lacking Vhlh and suggest a likely mechanism for the hypoglycemia in Pdx-1-Creearly;VhlhLoxP/LoxP, Ngn3-Cre;VhlhLoxP/LoxP and Rfx6;VhlhLoxP/LoxP mice.
Oxygen and HIF activity has been recently implicated in pancreas development ,  and previous work has shown that VHL elimination in pancreatic progenitor cells causes early postnatal lethality . Using the same pancreatic Pdx-1-Creearly line [Tg(Pdx1/cre)89.1Dam]  described in this study to inactivate Vhlh, we also observed significant postnatal death in Vhlh-deficient mice. In contrast, Vhlh inactivation in pancreatic progenitor cells using a different Pdx-1-Cre strain [Tg(Pdx1-cre)1Herr]  does not result in postnatal lethality . This apparent discrepancy might be due to differences in timing and/or level of Cre expression between the mouse strains. The Pdx-1-Creearly line used both by us and Shen et al  is active earlier during embryogenesis, while Cre expression occurs at a later stage of development in the Tg(Pdx1-cre)1Herr line , making it likely that the excision efficiency and cell types affected are distinct in the two models.
It has been suggested that reduced postnatal survival in Pdx-1-Creearly;VhlhLoxP/LoxP mice might be associated with exocrine defects . Although older surviving Pdx-1-Creearly;VhlhLoxP/LoxP mice display exocrine lesions, histological examination of pancreata in young mice do not reveal any apparent exocrine defects ( and our own results). Our results point to defects in endocrine function as the cause of death of Pdx-1-Creearly;VhlhLoxP/LoxP pups. Using exocrine- and endocrine-specific Cre-expressing mouse strains, our studies clearly demonstrate that VHL elimination in the endocrine, but not in the exocrine, lineage results in postnatal lethality.
Severe hypoglycemia appears to be the primary cause of death in Pdx-1-Creearly;VhlhLoxP/LoxP (and Rfx6-Cre;VhlhLoxP/LoxP and Ngn3-Cre;VhlhLoxP/LoxP) pups. These mice show a marked reduction in glucose levels since the first day of life. Abnormal feeding does not appear to be the cause for the low blood glucose levels, since milk was observed in the stomach of pups and they appeared to be properly nursed by the mother. In agreement with this, surviving mice do not show low blood glucose levels (data not shown). The hypoglycemic phenotype was somewhat surprising given our previous observation that ß-cell-specific knockout of Vhlh does not result in fatal hypoglycemia. Inactivation of Vhlh in islets results in defects in glucose-stimulated insulin secretion, similar to what has been described in ß-cell-specific Vhlh deficient mice, but an increase in insulin that might explain the hypoglycemic phenotype was not detected. These observations suggest that hypoglycemia in young mice with an islet-specific Vhlh deficiency is not primarily linked to defects in ß-cell function. We propose that aberrant glucagon secretion may underlie hypoglycemia in mice with VHL loss in neonatal islets. In agreement with this hypothesis, several studies have linked impaired α-cell formation or glucagon production to severe hypoglycemia and associated neonatal lethality , , . However, it should be noted that other studies have not observed hypoglycemia or postnatal lethality in mice with disrupted glucagon production , , . The reasons for these discrepancies are not clear but may be related to the different mouse models used.
Several lines of evidence from our study indicate that activation of the hypoxia/HIF pathway in glucagon-producing cells blocks glucagon secretion. First, no increase in serum glucagon levels was detected under fed or fasted conditions in islet-specific Vhlh mutant mice despite decreased blood glucose levels. Second, studies performed in isolated islets from islet-specific neonatal Vhlh mutant mice revealed defective glucagon secretion. Finally, glucagon-producing α-TC6.1 cells grown under hypoxic conditions failed to secrete glucagon in response to low glucose. Of note, under hypoxic conditions, canonical target genes of the HIF complex are upregulated in α-TC6.1 cells, similar to what we observe in VHL-depleted islets. Our results are in agreement with a recent study performed in isolated islets cultured under hypoxia . This study reported that islets under hypoxia exhibit inappropriate high basal glucagon release. Interestingly, we observed a trend towards increased basal glucagon secretion in Pdx-1-Creearly;VhlhLoxP/LoxP mutant islets. However, the difference in glucagon secretion between control and mutant islets was not statistically significant despite the relatively large sample size. Significantly, a failure to secrete glucagon was clear when the mutant islets were shifted to low glucose. In apparent contradiction to our hypothesis identifying glucagon-producing cells as the culprit for hypoglycemia and postnatal lethality of islet-specific Vhlh mutant mice, transgenic mice with VHL inactivation specifically in α-cells survive to adulthood . Our attempts at reproducing this study were unsuccessful, as the level of gene excision achieved with another Glucagon-Cre mouse line  was exceedingly low in our hands (data not shown). Nonetheless, it is important to note that the Glucagon-Cre line used by Shen et al efficiently targets α-cells only at adult stages, not during early postnatal stages . Thus, a possible explanation for this apparent discrepancy is that VHL inactivation affects glucagon secretion only during early stages. To this regard, it is interesting to note that neonatal pancreatic endocrine cells appear to be functionally immature . It is tempting to speculate that this immaturity can make endocrine cells more sensitive to increased HIF activity. In agreement with this notion, we observed that Vhlh inactivation in pancreatic progenitor cells resulted in delayed ß-cell differentiation. Nevertheless, we cannot formally rule out that a combined perturbation in α- and ß-cell function may cause the hypoglycemia observed in Pdx-1-Creearly;VhlhLoxP/LoxP mice. In the absence of glucagon counter-regulation, mutant mice would effectively have sufficient insulin to reduce glucose levels, thus providing a possible explanation for the hypoglycemia. Indeed, hypoglycemia has been observed in certain models of α- and ß-cell dysfunction. Mice deficient in a component of the voltage-gated sodium channel (VGSC) present with hypoglycemia associated with defective insulin and glucagon secretion  and intra-islet insulin signaling is known to play an essential role in the regulation of glucagon secretion in both normo- and hypoglycemic conditions . Interestingly, the inhibitory effect of insulin on glucagon secretion is not observed in Vhlh-deficient islets indicating that VHL loss might compromise intra-islet insulin signaling.
A potential concern of our study is the extra-pancreatic Cre expression of Pdx-1-Creearly and Ngn3-Cre mouse strains, including the brain and duodenum. Although we cannot exclude this possibility since no truly pancreas-specific Cre line is available, we consider the likelihood of a non-pancreatic effect causing the lethality as small. First, cell-autonomous defects are detected in islets of Pdx-1-Creearly;VhlhLoxP/LoxP mice in in vitro secretion assays (Figures 3 and and4).4). Second, no brain expression is observed in the Rfx6-Cre mouse strain . Third, a detailed analysis of the brain cell types led to the conclusion that distinct, non-overlapping neurons are marked by Pdx-1-Creearly and Ngn3-Cre lines . It is unknown whether there is overlap in the duodenal expression of Pdx-1-Creearly and Ngn3-Cre line. However, specific loss of VHL in the intestine does not result in any major defects in mice .
In contrast to insulin secretion, the mechanism of glucagon secretion is still poorly understood . Our results reveal hypoxia/HIF signaling as a novel regulator of glucagon secretion and further underscore the importance of finely balanced hormone signaling for proper glucose homeostasis.
The animal experiments described in this study were approved by the Committee on Animal Research at the University of California, San Francisco (approval number AN088473-01C). Mice were examined after birth daily for signs of ill health. Any animal displaying body condition score 2 (underconditioned, segmentation of vertebral column evident and dorsal pelvic bones palpable) or less was euthanized. Euthanasia in animals <10 days old were performed by cervical dislocation by trained personnel. In animals >10 days old euthanasia was performed by carbon dioxide inhalation followed by cervical dislocation according to UCSF Institutional Animal Care and Use Committee (IACUC) guidelines. Ins-Cre and Ptf1a-Cre mice were obtained from Dr. Herrera and Dr. Wright respectively , . Pdx-1-Creearly, Ngn3-Cre, Rfx6-Cre and VhlhloxP/loxP mice have been described previously , , . Control animals include Cre-positive animals that are heterozygous for the Vhlh floxed allele, and Cre-negative animals that are either heterozygous or homozygous for the Vhlh floxed allele. Both genders were included is all analyses.
Pancreata were processed for hematoxylin/eosin and immunostaining as described previously . The following primary antibodies were used: rabbit anti-amylase, 1:700; mouse anti-glucagon 18000 (Sigma Chemical Co, St. Louis, MO); guinea pig anti-insulin, 1300, rabbit anti-glucagon, 1300 (Linco Research Inc.); armenian hamster anti-Mucin-1, 1200 (Neomarkers, Fremont, CA); rabbit anti-Hif1α 1500 (Novus Biologicals, Littleton, CO); rabbit anti-somatostatin, 1200 (Dako, Carpinteria, CA). Primary antibodies were detected with Alexa-488, Alexa-546, and Alexa-633 conjugated secondary antibodies (Jackson Immunoresearch Laboratory, West Grove, PA). For anti-HIF1α detection, a tyramide Signal Amplification kit (PerkinElmer, Waltham, MA) was used following manufacturer’s instructions. Bright field images were acquired using a Zeiss Axio Imager D1 microscope. A Zeiss Axioscope2 widefield microscope was used to visualize and photograph fluorescence. Confocal images were acquired using an LTC SP2 confocal microscope (Leica).
RNA isolation, cDNA preparation, and quantitative PCR were performed as described . Gene expression was normalized to cyclophilin and actin expression. Primer sequences are available upon request.
Analyses on hypoglycemic mice were carried out between p13 and p18 after birth. After a 16hr fast for adults or 6-hour fast for pups (<21days), animals were weighed, blood glucose measured followed by an intraperitoneal injection of glucose as described previously . For in vivo hormone measurement, blood from the tail vein was centrifuged to collect serum, and insulin and glucagon concentration was calculated using the Insulin EIA kit (ALPCO) and Glucagon RIA kit (#GL-32K, Millipore) respectively. An overnight fast was performed in pups aged p15 for assessing fasted glucagon levels in serum.
The Islet Production Facility Core at the Diabetes Center at UCSF isolated mouse islets. Glucose stimulated insulin secretion from isolated islets and measurement of insulin content of islets have been described previously , . For glucagon secretion, islets (10 per replicate) were incubated in 500 µl of high (25 mM) or low (2.7 mM) glucose at 37°C for one hour. Supernatants were collected, proteinase inhibitors added (Aprotinin, Sigma Chemical Co, St. Louis, MO) and assayed for glucagon using a Glucagon Elisa Kit (Yanaihara Institute Inc, Shizouka, Japan). Total protein content was measured using a DC Protein Assay kit (Bio-Rad, Hercules, CA). To analyze the inhibitory effects of insulin on glucagon secretion from isolated islets, exogenous insulin was added at a final concentration of 17 nM. For the hypoxia experiments, α-TC6.1 murine cells (obtained from ATCC, catalog number CRL-2934) were grown in low glucose medium (2 g/l glucose) in a hypoxia incubator (1% O2 37°C and 5% CO2) for 12 hours. For the glucagon secretion assay, cells were incubated in KRB-high glucose (25 mM) at 37°C or KRB-low glucose (2.7 mM) for one hour. The hypoxia/normoxia condition was maintained during this incubation period. Supernatants were collected, proteinase inhibitors added and assayed for glucagon as described above. Cellular DNA was extracted with ATE (Ammonium Hydroxide/Triton X-100) solution and quantified using Nanodrop-2000 (Thermo Scientific).
Glucagon (Sigma) was injected intraperitoneally at 16 µg/kg body weight and blood glucose monitored for 2 hours using a Lifescan Glucometer.
Student’s t test was used for calculating significance for all data sets except for the glucagon secretion experiments. The values in these experiments were not normally distributed and were compared using the Mann-Whitney U test, calculated according to the Monte Carlo method. Differences were considered significant if p<0.05.
A. Survival curve of Pdx-1-Creearly;VhlhLoxP/LoxP mice (n=35) as compared to control littermates (n=40). B. Normal islet formation in Rfx6-Cre;VhlhLoxP/LoxP mice.
A. Ptf1a-Cre is not efficiently expressed in the pancreatic islet. The Z/AP reporter strain , which expresses alkaline phosphatase (AP) upon Cre-mediated recombination, was used to determine the Cre recombinase activity in Ptf1a-Cre mice. Staining for alkaline phosphatase marks cells that have undergone recombination in 4 week old Ptf1a-Cre;Z/AP mice. Histological sections were enzymatically stained for alkaline phosphatase activity (blue), DBA lectin to mark pancreatic ducts (brown) and nuclear fast red as a counterstain (pink). Exocrine tissue is efficiently targeted in Ptf1a-Cre;Z/AP pancreas as shown by the clear alkaline phosphatase. However, only scattered cells are marked by alkaline phosphatase in islets (dashed in yellow). B. Robust HIF1α accumulation in exocrine, but not endocrine, pancreatic tissue of Ptf1a-Cre;VhlhLoxP/LoxP mice (p15). Islets are outlined in yellow. C. Normal blood glucose levels in neonatal (left) and older (right) Ptf1a-Cre;VhlhLoxP/LoxP mice.
Glucagon stimulation in Pdx-1-Creearly;VhlhLoxP/LoxP mutant mice. Exogenous glucagon was administered to control (square, n=17) and Pdx-1-Creearly;VhlhLoxP/LoxP (triangle, n=7) animals and blood glucose measured over the next two hours. **p<0.01, ****p<10−6.
We would like to thank Dr. Mike German and D. Melton for providing the Rfx6-Cre and Pdx-1-Creearly line, respectively.
SP was supported by a grant from the Leona M. and Harry B. Helmsley Charitable Trust and a grant from the National Institutes of Health (NIH) (R01 DK60533). AGN was supported by a fellowship from the Andalusian Regional Government (P08-CVI-3727). DAC was supported by the “Ramón y Cajal“ program (RYC-2006-001071) and grants (SAF2008-02469 and SAF2011-26805) from the Spanish Ministry of Science and Innovation and the Andalusian Regional Ministry of Health (PI-0250/2008). Work in MH’s laboratory was supported by grants from the NIH (DK60533) and a Scholar Award from the Juvenile Diabetes Research Foundation. Image acquisition and islet isolation were supported by the imaging and islet cores of the UCSF Diabetes and Endocrinology Research Center (DERC) NIH grant P30DK63720. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.