American Journal of Therapeutics 0, 1–7 (2016)
Conditional Tissue-Specific Foxa2 Ablation in Mouse Pancreas Causes Hyperinsulinemic Hypoglycemia Zengbin Wu, MM,1 Aihua Fei, PhD,1 Yingbin Liu, PhD,2 and Shuming Pan, PhD1*
The forkhead/winged helix transcription factor Foxa2 is a major upstream regulator of Pdx1, a transcription factor necessary for pancreatic development. In the present study, we conditionally knocked out Foxa2 in Pdx1-expressing domain and further analyzed the contribution of Foxa2 to a- and b-cell development and the effect of Foxa2 deletion on plasma insulin, glucagon, and glucose levels. Homozygous pdx1cre/+ Foxa2 FL/FL mice and heterozygous pdx1cre/+ Foxa2 FL/+ mice were generated by homologous recombination using a Foxa2 gene-targeting vector. a- and b-cell mass was examined by immunofluorescence microscopy. Plasma glucose, insulin, and plasma were measured at postnatal day 10. For pdx1 lineage tracing studies, heterozygous pdx1cre/+ Foxa2 FL/+ EYFP 6 and homozygous pdx1cre/+ Foxa2 FL/FL EYFP 6 mice were used. Our immunofluorescence analysis revealed that in the pancreas sections of the homozygous mutant mice, Foxa2 was virtually absent from non-b cells and its expression almost exclusively coincided with remnant b cells. The density of both a and b cells apparently decreased in the pancreas of the heterozygous mutant mice and in the pancreas of the homozygous mutant mice, a cells lost its predominance and b cells increased proportionally. Direct Pdx1 cell lineage tracing revealed that, on embryonic day 18.5, in the homozygous mutant mice, Pdx1 expression coincided almost exclusively with that of insulin-secreting b cells. Chemiluminescence assays revealed that heterozygous pdx1cre/+ Foxa2 FL/+ mice had significantly lower insulin levels than control mice (P , 0.01). However, no apparent difference was observed between homozygous pdx1cre/+ Foxa2 FL/FL mice and control mice (P . 0.05). Chemiluminescence assays also showed that Foxa2 deletion significantly depressed plasma glucagon levels in both homozygous pdx1cre/+ Foxa2 FL/FL mice and heterozygous pdx1cre/+ Foxa2 FL/+ mice (P , 0.01 vs. controls). Plasma glucose on postnatal day 10 was significantly lower in homozygous pdx1cre/+ Foxa2 FL/FL mice compared with control mice (P , 0.01). Our study demonstrates that homozygous Foxa2 ablation leads to an imbalance in b/a ratio, profound hypoglucagonemia, inappropriate hyperinsulinemia, and hypoglycemia in mice. Our conditional tissue-specific Foxa2 ablation mouse model will be useful in elucidating regulation of normal and abnormal a- and b-cell differentiation and pinpointing novel targets for diabetes control. Keywords: hyperinsulinemic hypoglycemia, Foxa2, a- and b-cell differentiation, diabetes control
INTRODUCTION 1
2
Departments of Emergency and General Surgery, Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China. Supported by the Science and Technology Commission of Shanghai Municipality Grants (13ZR1426500). Y. Liu and S. Pan contributed equally to this article. *Address for correspondence: Department of Emergency, Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, 1665 Kongjiang Road, Shanghai 200092, China. Email: [email protected]
The differentiation of a and b cells in the pancreas, which is critical for blood glucose homeostasis, is a highly complex and tightly regulated process. The differentiation of a and b cells and maintenance of their function are under the control of a hierarchically controlled transcription factor network in the developing and maturing islet. Pancreatic and duodenal homeobox 1 (Pdx1), also known as insulin promoter factor 1, is a transcription factor necessary for pancreatic development and
1075–2765 Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
www.americantherapeutics.com
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
2
b-cell maturation.1 The protein is initially expressed in the foregut endoderm at the sites from which the dorsal and ventral pancreatic buds emerge. In the insulinsecreting islet b cells, Pdx1 transactivates the insulin gene. In Pdx1 null mice, the pancreas fails to develop shortly after the emergence of the prepancreatic buds2 and in heterozygous Pdx1 gene knockout (Pdx1+/2) mice, Pdx1 expression was reduced to 68% of that of the wild-type (Pdx1+/+) mice.3 Conditional repression of Pdx1 in mice showed that PDX1 is not only necessary for early pancreatic development, but is also required to maintain the function and allow the regeneration of pancreatic b cells.4 The forkhead/winged helix transcription factor Foxa2 is a major upstream regulator of Pdx1.5,6 Endodermspecific or b-cell–specific Foxa2 deletion led to absence of mature a cells and a reduction of Pdx1 expression and b-cell differentiation.5,7 In the present study, we conditionally knocked out Foxa2 in Pdx1-expressing domain and further analyzed the contribution of Foxa2 to a- and b-cell development and the effect of Foxa2 deletion on plasma insulin, glucagon, and glucose levels.
MATERIALS AND METHODS
Wu et al
6-mm sections. Paraffin-embedded pancreatic tissue sections were dewaxed using xylene, rehydrated through gradient alcohol. Immunofluorescence microscopy was performed as previously depicted8 and the following antibodies were used at the indicated dilutions: guinea pig antiinsulin antibody (1:1000), goat anti-GFP antibody (1:500) and mouse anti-Foxa2 antibodies (1:50) (all from Abcam, Cambridge, MA), and rabbit antiglucagon (1:800, Millipore, Darmstadt, Germany). Secondary FITC-conjugated goat antimouse IgG, Alex Fluor 488 goat antirabbit IgG and Texas Red/mCherry/AlexaFluor 594, and CY5 goat antimouse IgG (Invitrogen, Carlsbad, CA) were used. Nuclear staining was performed using 49,6-diamidino-2-phenylindole (Vector Laboratories, USA). Images were obtained using a Leica TCS SP2 laser scanning confocal microscope (Nikon) and analyzed using the Image pro plus 5.1 image analysis software (Media Cybernetics, Silver Spring, MD). Three mice from each group were analyzed and evaluators were blinded to the genotypes of the mice. Three sections from each mouse were analyzed. For each section, the number of insulin positive b cells and glucagon positive a cells was determined in each islet and the ratio of b cells to a cells was calculated.
Animals
Biochemical determinations
Homozygous pdx1cre/+ Foxa2 FL/FL mice and heterozygous pdx1cre/+ Foxa2 FL/+ mice, which were generated by homologous recombination using a Foxa2 genetargeting vector with loxP sites flanking exon 3, were obtained from Dr Klaus H. Kastner at the Department of Medicine, University of Pennsylvania School of Medicine. Genotyping was performed by polymerase chain reaction analysis using genomic DNA isolated from the tail tip of fetal and newborn mice. +/+ Foxa2 FL/FL & +/+ Foxa2 FL/+ mice were used as control mice (Jackson Laboratory, Bar Harbor, ME). For pdx1 lineage tracing studies, pdx1cre/+ Foxa2 FL/+ transgenic mice were crossed to Foxa2 FL/FL ROSA26, EYFP2/2 mice to obtain heterozygous pdx1cre/+ Foxa2 FL/+ EYFP 6 and homozygous pdx1cre/+ Foxa2 FL/FL EYFP 6 mice (Jackson Laboratory). pdx1cre/+ EYFP 6 mice (Jackson Laboratory) were used as control mice. The study protocol was approved by the local ethics committee at the authors’ affiliated institution and animal study was performed in accordance with established institutional and state guidelines on the use of experimental animals.
Plasma glucose content at postnatal day 10 was measured using an automatic glucose monitor (Abbot). Plasma insulin was measured by chemiluminescence using commercially available kits as instructed by the manufacturer (Beckman). Plasma glucagon was determined by chemiluminescence using commercially available kit as instructed by the manufacturer Millipore.
Immunofluorescence microscopy Pancreatic tissues were fixed in 4% paraformaldehyde overnight at 4°C, embedded in paraffin, and cut to American Journal of Therapeutics (2016) 0(0)
Statistical analysis All values unless otherwise indicated are expressed as mean 6 SD. Statistical analyses were performed using a 2-tailed Student unpaired t test, and P values , 0.05 were considered statistically significant.
RESULTS Foxa2 is specifically deleted in pancreatic cells in mice In the present study, we used mice with homozygous (pdx1cre/+ Foxa2 FL/FL) and heterozygous (pdx1cre/+ Foxa2 FL/+) Foxa2 deletion in pancreatic b cells that were generated using the Cre-loxP recombination system. We analyzed totally 27 liters. Twelve (12/35, 34.3%) homozygous mutant mice died before postnatal day 10, whereas none of the heterozygous mutant mice died. www.americantherapeutics.com
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Foxa2 and Hyperinsulinemic Hypoglycemia
The pancreas sections from mice on embryonic day 18.5 were double-immunolabeled for Foxa2 and insulin. Our immunofluorescence analysis revealed that in control mice, Foxa2 was expressed in b cells and in some non-b cells (a cells and acinar cells) (data not shown). In the pancreas sections of the heterozygous mutant mice, Foxa2 was expressed in b cells where it coincided with that of insulin expression, whereas it was not expressed in the pancreas sections of the homozygous mutant mice (Figure 1, left panel). In the pancreas sections of the homozygous mutant mice, Foxa2 was virtually absent from b cells (Figure 1, right panel). These findings suggest that Foxa2 is specifically deleted in pancreatic cells in mice with homozygous Foxa2 deletion. Foxa2 deletion causes an imbalance of a and b cells in mouse pancreas We further examined the number of a and b cells in mice with homozygous and heterozygous Foxa2 deletion by immunofluorescence analysis of pancreas sections on postnatal day 10. Double immunostaining for insulin and glucagon showed that in control pancreas section a and b cells formed the core of the islet with non-a or b cells positioned on the periphery of the islet and a cells were the predominant cell type in the islet (Figure 2A, left panel). The density of both a and b cells apparently decreased in the pancreas of the heterozygous mutant mice (Figure 2A, mid panel), and in the pancreas of the homozygous mutant mice, a cells lost its predominance and b cells increased proportionally (Figure 2A, right panel). Compared with the heterozygous mice, the homozygous mutant mice had a significantly higher number of b cells (37.2 6 1.3/islet vs. heterozygous mice: 22.1 6 1.2/islet, P , 0.01), but
3
a comparable number of a cells (16.5 6 1.1 vs. heterozygous mice: 18.5 6 1.4, P . 0.05). The number of b cells was comparable in the homozygous mice to that of the control mice (34.0 6 1.5/islet, P . 0.05), but significantly lower in the heterozygous mice than that of the control mice (P , 0.05). Furthermore, the number of a cells in both the homozygous and heterozygous mutant mice was apparently lower than that of the control mice (27.6 6 1.2/islet, P , 0.01) (Figure 2B). These findings translated into a noticeably higher b/a ratio in homozygous pdx1cre/+ Foxa2 FL/FL mice (2.23 6 0.02) compared with heterozygous pdx1cre/+ Foxa2 FL/+ mice (1.24 6 0.22) and the control mice (1.24 6 0.14) (P , 0.01) (Figure 2C), suggesting that Foxa2 deletion caused a relative dose-dependent increase in b-cell mass in the mutant mice. Consistently, direct Pdx1 cell lineage tracing showed that 98% of b cells and 70% of a cells were EYFP+ in the control mice (pdx1cre/+ EYFP+/2) were EYFP+. In the heterozygous mutant (pdx1cre/+ Foxa2 FL/FL EYFP+/2) mice, 88% of b cells were EYFP+ and 71% of a cells were EYFP+. By contrast, the homozygous mice (pdx1cre/+ Foxa2 FL/FL EYFP+/2) showed a significant reduction in the number of a cells (3% EYFP+), whereas 96% of b cells were EYFP+. These findings together indicate that Foxa2 deletion caused a noticeable reduction in a-cell mass and a corresponding dose-dependent increase in b-cell mass in the mutant mice. Homozygous Foxa2 deletion shifts Pdx1 expression to b cells in mouse pancreas In the mouse pancreas, direct Pdx1 cell lineage tracing revealed that, on embryonic day 18.5, in control mice (pdx1cre/+ EYFP+/2), b and a cells formed the core of the islet with other cells positioned on the periphery of the
FIGURE 1. Foxa2 is deleted in pancreatic cells in mice. Pancreas sections were prepared from mice on postnatal day 10 and double-immunolabeled for Foxa2 (pink) and insulin (green) as described in methods. Left panel: Foxa2 expression in pancreas sections of the heterozygous mutant (pdx1cre/+ Foxa2 FL/+) mice. Foxa2 is still expressed in occasional b cells (arrow) and non-b cells. Right panel: Foxa2 expression in pancreas sections of the homozygous mutant (pdx1cre/+ Foxa2 FL/FL) mice. Foxa2 is virtually absent from non-b cells and its expression almost exclusively coincides with that of insulin expression in b cells. www.americantherapeutics.com
American Journal of Therapeutics (2016) 0(0)
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
4
Wu et al
FIGURE 2. Foxa2 deletion causes an imbalance of a and b cells in mouse pancreas. Pancreas sections were prepared from mice on postnatal day 10 and double-immunolabeled for insulin (red) and glucagon (green) as described in methods. (A) Left panel: In immunostained of control pancreas sections, a and b cells form the core of the islet with non-a or b cells positioned on the periphery of the islet. a cells are the predominant cell type in the islet. Mid panel: The density of both a and b cells decreases in the pancreas sections of the heterozygous mutant mice. Right panel: In the pancreas of the homozygous mutant mice, a cells lose its predominance and b cells become predominant. (B) The number of a and b cells in the homozygous and heterozygous mutant mice. *P , 0.01 versus heterozygous mutant. (C) The ratio of b tο a cells in the homozygous and heterozygous mutant mice. *P , 0.01 versus heterozygous mutant.
islet (Figure 3A). Pdx1 expression coincided with that of insulin and glucagon. In the heterozygous mutant (pdx1cre/+ Foxa2 FL/FL EYFP+/2) mice, Pdx1 was expressed coincident mostly with that of insulin and glucagon and occasionally with non-b or non-a cells (Figure 3B). In the homozygous mutant (pdx1cre/+ Foxa2 FL/FL EYFP+/2) mice, Pdx1 expression coincided almost exclusively with that of insulin-secreting b cells (Figure 3C). These findings suggested that homozygous Foxa2 deletion shifted Pdx1 expression to b cells in mouse pancreas. Foxa2 deletion causes severe hypoglycemia in mice We then investigated the effect of Foxa2 deletion on plasma insulin, glucagon, and glucose levels. Chemiluminescence assays revealed that heterozygous pdx1cre/+ Foxa2 FL/+ mice had significantly lower insulin levels (0.36 6 0.07 ng/mL) than control mice (1.22 6 0.09 ng/mL) (P 5 0.0002) and homozygous pdx1cre/+ Foxa2 FL/FL mice (1.01 6 0.35 ng/mL) (P , 0.0001) (Figure 4A). However, no apparent difference was American Journal of Therapeutics (2016) 0(0)
observed between homozygous pdx1cre/+ Foxa2 FL/FL mice and control mice (P 5 0.099). Chemiluminescence assays further showed that tissue-specific deletion of Foxa2 significantly depressed plasma glucagon levels in both homozygous pdx1cre/+ Foxa2 FL/FL mice (0.014 6 0.003 ng/mL) and heterozygous pdx1cre/+ Foxa2 FL/+ mice (0.015 6 0.002 ng/mL) compared with control mice (0.038 6 0.013 ng/mL) (P 5 0.033 for homozygous mice and P 5 0.038 for heterozygous mice vs. control) (Figure 4B). Examination of plasma glucose on postnatal day 10 showed significantly lower plasma glucose content in homozygous pdx1cre/+ Foxa2 FL/FL mice (77.60 6 11.94 mg/dL) compared with control mice (122.50 6 8.39 mg/dL) (P , 0.0001) and heterozygous pdx1cre/+ Foxa2 FL/+ mice (95.25 6 9.11 mg/dL) (P 5 0.019) (Figure 4C). The homozygous mutant mice also developed seizures because of severe hypoglycemia (data not shown). Although the plasma glucose content was lower in heterozygous pdx1cre/+ Foxa2 FL/+ mice than control mice, no statistical difference was achieved (P . 0.05). These findings indicated that www.americantherapeutics.com
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Foxa2 and Hyperinsulinemic Hypoglycemia
5
FIGURE 3. Homozygous Foxa2 deletion shifts Pdx1 expression to b cells in mouse pancreas. Mesp1 cre/+ Foxa2 FL/+ transgenic mice were crossed to Foxa2 FL/FL ROSA26, EYFP2/2 mice to obtain heterozygous pdx1cre/+ Foxa2 FL/+ EYFP 6 and homozygous pdx1cre/+ Foxa2 FL/FL EYFP 6 mice. pdx1cre/+ EYFP 6 mice were used as control mice. Pancreas sections were prepared from mice on embryonic day 18.5 and stained for insulin (red) and glucagon (pink). (A) In the control mice, b and a cells form the core of the islet with other cells positioned on the periphery of the islet. (B) In heterozygous mutant mice, Pdx1 is expressed coincident mostly with that of insulin and glucagon-secreting cells and occasionally with non-b or non-a cells. (C) In homozygous mutant mice, Pdx1 expression coincides almost exclusively with that of insulin-secreting b cells.
tissue-specific deletion of Foxa2 caused inappropriate hyperinsulinemia and hypoglucagonemia in homozygous mutant mice, which contributed to severe hypoglycemia in these mice.
DISCUSSION In the present study, we generated homozygous and heterozygous conditionally ablated Foxa2 mutant mice by homologous recombination using a Foxa2 genetargeting vector with loxP sites flanking exon 3. We www.americantherapeutics.com
demonstrated that condition knockout of Foxa2 in Pdx1 expressing domains caused profound hypoglucagonemia and inappropriate hyperinsulinemia in the homozygous mutant mice, contributing to profound hypoglycemia in the animals. Furthermore, conditional tissue-specific Foxa2 ablation is associated an imbalance of b/a ratio as a result of significant but disproportional reduction in pancreatic a- and b-cell mass, indicating that in our knockout model, a cells are more responsive to Foxa2 ablation than b cells. Oliver-Krasinski et al9 generated hypomorphic Pdx1DC/DC mice by truncating the C terminus of Pdx1 American Journal of Therapeutics (2016) 0(0)
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
6
Wu et al
FIGURE 4. Foxa2 deletion causes severe hypoglycemia in mice. Plasma insulin was measured by ELISA (A), and plasma glucagon was determined using RIA (B) as described in methods. Plasma glucose content at postnatal day 10 was measured using an automatic glucose monitor (C). *P , 0.05 versus control and **P , 0.01 versus control.
at amino acid 210, and they found a global reduction in endocrine lineages in Pdx1DC/DC mice, especially a and b cells. In Pdx1+/DC mice, only b-cell area was reduced, indicating a greater sensitivity of the b-cell lineage to Pdx1 action. Our study also showed that conditional ablation of Foxa2, a major upstream regulator of Pdx1, led to apparent reduction in the number of a and b cells in the pancreas at postnatal day 10. In contrast to the finding by Oliver-Krasinski et al,9 our immunofluorescent analysis and direct Pdx1 lineage tracing revealed a greater reduction in the number of a cells, suggesting that a cells are more amenable to the actions of Foxa2. Furthermore, we observed a dose effect of Foxa2 as a higher b/a ratio was demonstrated in the homozygous mutant mice and the heterozygous mutant mice. Our findings are consistent with the present study by Heddad Masson et al10 who showed that small interfering RNA knockdown of Foxa2 led to impaired a-cell differentiation and glucagon synthesis, a process that also involves Foxa1. American Journal of Therapeutics (2016) 0(0)
Insulin deficiency due to reduced pancreatic islet b cell number underlies the progression of both type 1 and type 2 diabetes. Heterozygous mutations in Pdx1 are known to be associated with maturity onset diabetes of the young, type 4 (MODY4) and type 2 diabetes,11 and Foxa2 has been demonstrated to be as an essential activator of genes that function in multiple pathways governing insulin secretion and also regulates differentiation of a and b cells in the pancreas. Consistent with a significant reduction in the a-cell mass in the pancreas, profound hypoglucagonemia was demonstrated in both the heterozygous and homozygous mutant mice. Although we observed a marked reduction in plasma insulin in the heterozygous mutant mice, we failed to note a significant reduction in plasma insulin in the homozygous mutant mice. The plasma insulin levels in the homozygous mutant mice are inappropriately high considering severe hypoglycemia in these animals. This may be due to changes in islet cell lineage allocation, www.americantherapeutics.com
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Foxa2 and Hyperinsulinemic Hypoglycemia
defective regulation of insulin secretion, or decline in a-cell mass leading to an increase in insulin-producing b cells as a result of Foxa2 ablation. Plasma glucagon was approximately 3-fold lower in the homozygous mutant mice than that of the control mice, which, together with the inappropriate hyperinsulinemia, helps explain profound hypoglycemia in these animals. Our findings are consistent with the study by Sund et al12 who also observed hypoglycemia due to inappropriate hyperinsulinemia in mice with tissuespecific deletion of Foxa2 in pancreatic b cells. In conclusion, our study demonstrates that Foxa2 is required in both a and b cells for maintaining appropriate glucagon and insulin levels and for keeping glucose homeostasis. Elucidation of the regulation of normal and abnormal a and b cell differentiation offers insight into the pathogenesis of diabetes, and pinpoints novel targets for diabetes control and may lead to clues for reprogramming a and b cells for replacement therapy. Our conditional tissue-specific Foxa2 ablation mouse model will provide a unique model for delineating the role of a and b cells in the pathogenesis of diabetes and conditions such as familial hyperinsulinism.
REFERENCES 1. Ohlsson H, Karlsson K, Edlund T. IPF1, a homeodomaincontaining transactivator of the insulin gene. EMBO J. 1993;12:4251–4259. 2. Ahlgren U, Pfaff SL, Jessell TM, et al. Independent requirement for ISL1 in formation of pancreatic mesenchyme and islet cells. Nature. 1997;385:257–260.
www.americantherapeutics.com
7 3. Brissova M, Shiota M, Nicholson WE, et al. Reduction in pancreatic transcription factor PDX-1 impairs glucosestimulated insulin secretion. J Biol Chem. 2002;277: 11225–11232. 4. Hale MA, Kagami H, Shi L, et al. The homeodomain protein PDX1 is required at mid-pancreatic development for the formation of the exocrine pancreas. Dev Biol. 2005; 286:225–237. 5. Lee CS, Sund NJ, Vatamaniuk MZ, et al. Foxa2 controls Pdx1 gene expression in pancreatic beta-cells in vivo. Diabetes. 2002;51:2546–2551. 6. Gao N, LeLay J, Vatamaniuk MZ, et al. Dynamic regulation of Pdx1 enhancers by Foxa1 and Foxa2 is essential for pancreas development. Genes Dev. 2008;22:3435–3448. 7. Lee CS, Sund NJ, Behr R, et al. Foxa2 is required for the differentiation of pancreatic alpha-cells. Dev Biol. 2005; 278:484–495. 8. Cheng N, Hu X, Tian T, et al. PKMzeta knockdown disrupts post-ischemic long-term potentiation via inhibiting postsynaptic expression of aminomethyl phosphonic acid receptors. J Biomed Res. 2015;29:241–249. 9. Oliver-Krasinski JM, Kasner MT, Yang J, et al. The diabetes gene Pdx1 regulates the transcriptional network of pancreatic endocrine progenitor cells in mice. J Clin Invest. 2009;119:1888–1898. 10. Heddad Masson M, Poisson C, Guerardel A, et al. Foxa1 and Foxa2 regulate alpha-cell differentiation, glucagon biosynthesis, and secretion. Endocrinology. 2014;155: 3781–3792. 11. Shih DQ, Stoffel M. Molecular etiologies of MODY and other early-onset forms of diabetes. Curr Diab Rep. 2002; 2:125–134. 12. Sund NJ, Vatamaniuk MZ, Casey M, et al. Tissue-specific deletion of Foxa2 in pancreatic beta cells results in hyperinsulinemic hypoglycemia. Genes Dev. 2001;15: 1706–1715.
American Journal of Therapeutics (2016) 0(0)
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Conditional Tissue-Specific Foxa2 Ablation in Mouse Pancreas Causes Hyperinsulinemic Hypoglycemia Zengbin Wu, MM,1 Aihua Fei, PhD,1 Yingbin Liu, PhD,2 and Shuming Pan, PhD1*
The forkhead/winged helix transcription factor Foxa2 is a major upstream regulator of Pdx1, a transcription factor necessary for pancreatic development. In the present study, we conditionally knocked out Foxa2 in Pdx1-expressing domain and further analyzed the contribution of Foxa2 to a- and b-cell development and the effect of Foxa2 deletion on plasma insulin, glucagon, and glucose levels. Homozygous pdx1cre/+ Foxa2 FL/FL mice and heterozygous pdx1cre/+ Foxa2 FL/+ mice were generated by homologous recombination using a Foxa2 gene-targeting vector. a- and b-cell mass was examined by immunofluorescence microscopy. Plasma glucose, insulin, and plasma were measured at postnatal day 10. For pdx1 lineage tracing studies, heterozygous pdx1cre/+ Foxa2 FL/+ EYFP 6 and homozygous pdx1cre/+ Foxa2 FL/FL EYFP 6 mice were used. Our immunofluorescence analysis revealed that in the pancreas sections of the homozygous mutant mice, Foxa2 was virtually absent from non-b cells and its expression almost exclusively coincided with remnant b cells. The density of both a and b cells apparently decreased in the pancreas of the heterozygous mutant mice and in the pancreas of the homozygous mutant mice, a cells lost its predominance and b cells increased proportionally. Direct Pdx1 cell lineage tracing revealed that, on embryonic day 18.5, in the homozygous mutant mice, Pdx1 expression coincided almost exclusively with that of insulin-secreting b cells. Chemiluminescence assays revealed that heterozygous pdx1cre/+ Foxa2 FL/+ mice had significantly lower insulin levels than control mice (P , 0.01). However, no apparent difference was observed between homozygous pdx1cre/+ Foxa2 FL/FL mice and control mice (P . 0.05). Chemiluminescence assays also showed that Foxa2 deletion significantly depressed plasma glucagon levels in both homozygous pdx1cre/+ Foxa2 FL/FL mice and heterozygous pdx1cre/+ Foxa2 FL/+ mice (P , 0.01 vs. controls). Plasma glucose on postnatal day 10 was significantly lower in homozygous pdx1cre/+ Foxa2 FL/FL mice compared with control mice (P , 0.01). Our study demonstrates that homozygous Foxa2 ablation leads to an imbalance in b/a ratio, profound hypoglucagonemia, inappropriate hyperinsulinemia, and hypoglycemia in mice. Our conditional tissue-specific Foxa2 ablation mouse model will be useful in elucidating regulation of normal and abnormal a- and b-cell differentiation and pinpointing novel targets for diabetes control. Keywords: hyperinsulinemic hypoglycemia, Foxa2, a- and b-cell differentiation, diabetes control
INTRODUCTION 1
2
Departments of Emergency and General Surgery, Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China. Supported by the Science and Technology Commission of Shanghai Municipality Grants (13ZR1426500). Y. Liu and S. Pan contributed equally to this article. *Address for correspondence: Department of Emergency, Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, 1665 Kongjiang Road, Shanghai 200092, China. Email: [email protected]
The differentiation of a and b cells in the pancreas, which is critical for blood glucose homeostasis, is a highly complex and tightly regulated process. The differentiation of a and b cells and maintenance of their function are under the control of a hierarchically controlled transcription factor network in the developing and maturing islet. Pancreatic and duodenal homeobox 1 (Pdx1), also known as insulin promoter factor 1, is a transcription factor necessary for pancreatic development and
1075–2765 Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
www.americantherapeutics.com
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
2
b-cell maturation.1 The protein is initially expressed in the foregut endoderm at the sites from which the dorsal and ventral pancreatic buds emerge. In the insulinsecreting islet b cells, Pdx1 transactivates the insulin gene. In Pdx1 null mice, the pancreas fails to develop shortly after the emergence of the prepancreatic buds2 and in heterozygous Pdx1 gene knockout (Pdx1+/2) mice, Pdx1 expression was reduced to 68% of that of the wild-type (Pdx1+/+) mice.3 Conditional repression of Pdx1 in mice showed that PDX1 is not only necessary for early pancreatic development, but is also required to maintain the function and allow the regeneration of pancreatic b cells.4 The forkhead/winged helix transcription factor Foxa2 is a major upstream regulator of Pdx1.5,6 Endodermspecific or b-cell–specific Foxa2 deletion led to absence of mature a cells and a reduction of Pdx1 expression and b-cell differentiation.5,7 In the present study, we conditionally knocked out Foxa2 in Pdx1-expressing domain and further analyzed the contribution of Foxa2 to a- and b-cell development and the effect of Foxa2 deletion on plasma insulin, glucagon, and glucose levels.
MATERIALS AND METHODS
Wu et al
6-mm sections. Paraffin-embedded pancreatic tissue sections were dewaxed using xylene, rehydrated through gradient alcohol. Immunofluorescence microscopy was performed as previously depicted8 and the following antibodies were used at the indicated dilutions: guinea pig antiinsulin antibody (1:1000), goat anti-GFP antibody (1:500) and mouse anti-Foxa2 antibodies (1:50) (all from Abcam, Cambridge, MA), and rabbit antiglucagon (1:800, Millipore, Darmstadt, Germany). Secondary FITC-conjugated goat antimouse IgG, Alex Fluor 488 goat antirabbit IgG and Texas Red/mCherry/AlexaFluor 594, and CY5 goat antimouse IgG (Invitrogen, Carlsbad, CA) were used. Nuclear staining was performed using 49,6-diamidino-2-phenylindole (Vector Laboratories, USA). Images were obtained using a Leica TCS SP2 laser scanning confocal microscope (Nikon) and analyzed using the Image pro plus 5.1 image analysis software (Media Cybernetics, Silver Spring, MD). Three mice from each group were analyzed and evaluators were blinded to the genotypes of the mice. Three sections from each mouse were analyzed. For each section, the number of insulin positive b cells and glucagon positive a cells was determined in each islet and the ratio of b cells to a cells was calculated.
Animals
Biochemical determinations
Homozygous pdx1cre/+ Foxa2 FL/FL mice and heterozygous pdx1cre/+ Foxa2 FL/+ mice, which were generated by homologous recombination using a Foxa2 genetargeting vector with loxP sites flanking exon 3, were obtained from Dr Klaus H. Kastner at the Department of Medicine, University of Pennsylvania School of Medicine. Genotyping was performed by polymerase chain reaction analysis using genomic DNA isolated from the tail tip of fetal and newborn mice. +/+ Foxa2 FL/FL & +/+ Foxa2 FL/+ mice were used as control mice (Jackson Laboratory, Bar Harbor, ME). For pdx1 lineage tracing studies, pdx1cre/+ Foxa2 FL/+ transgenic mice were crossed to Foxa2 FL/FL ROSA26, EYFP2/2 mice to obtain heterozygous pdx1cre/+ Foxa2 FL/+ EYFP 6 and homozygous pdx1cre/+ Foxa2 FL/FL EYFP 6 mice (Jackson Laboratory). pdx1cre/+ EYFP 6 mice (Jackson Laboratory) were used as control mice. The study protocol was approved by the local ethics committee at the authors’ affiliated institution and animal study was performed in accordance with established institutional and state guidelines on the use of experimental animals.
Plasma glucose content at postnatal day 10 was measured using an automatic glucose monitor (Abbot). Plasma insulin was measured by chemiluminescence using commercially available kits as instructed by the manufacturer (Beckman). Plasma glucagon was determined by chemiluminescence using commercially available kit as instructed by the manufacturer Millipore.
Immunofluorescence microscopy Pancreatic tissues were fixed in 4% paraformaldehyde overnight at 4°C, embedded in paraffin, and cut to American Journal of Therapeutics (2016) 0(0)
Statistical analysis All values unless otherwise indicated are expressed as mean 6 SD. Statistical analyses were performed using a 2-tailed Student unpaired t test, and P values , 0.05 were considered statistically significant.
RESULTS Foxa2 is specifically deleted in pancreatic cells in mice In the present study, we used mice with homozygous (pdx1cre/+ Foxa2 FL/FL) and heterozygous (pdx1cre/+ Foxa2 FL/+) Foxa2 deletion in pancreatic b cells that were generated using the Cre-loxP recombination system. We analyzed totally 27 liters. Twelve (12/35, 34.3%) homozygous mutant mice died before postnatal day 10, whereas none of the heterozygous mutant mice died. www.americantherapeutics.com
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Foxa2 and Hyperinsulinemic Hypoglycemia
The pancreas sections from mice on embryonic day 18.5 were double-immunolabeled for Foxa2 and insulin. Our immunofluorescence analysis revealed that in control mice, Foxa2 was expressed in b cells and in some non-b cells (a cells and acinar cells) (data not shown). In the pancreas sections of the heterozygous mutant mice, Foxa2 was expressed in b cells where it coincided with that of insulin expression, whereas it was not expressed in the pancreas sections of the homozygous mutant mice (Figure 1, left panel). In the pancreas sections of the homozygous mutant mice, Foxa2 was virtually absent from b cells (Figure 1, right panel). These findings suggest that Foxa2 is specifically deleted in pancreatic cells in mice with homozygous Foxa2 deletion. Foxa2 deletion causes an imbalance of a and b cells in mouse pancreas We further examined the number of a and b cells in mice with homozygous and heterozygous Foxa2 deletion by immunofluorescence analysis of pancreas sections on postnatal day 10. Double immunostaining for insulin and glucagon showed that in control pancreas section a and b cells formed the core of the islet with non-a or b cells positioned on the periphery of the islet and a cells were the predominant cell type in the islet (Figure 2A, left panel). The density of both a and b cells apparently decreased in the pancreas of the heterozygous mutant mice (Figure 2A, mid panel), and in the pancreas of the homozygous mutant mice, a cells lost its predominance and b cells increased proportionally (Figure 2A, right panel). Compared with the heterozygous mice, the homozygous mutant mice had a significantly higher number of b cells (37.2 6 1.3/islet vs. heterozygous mice: 22.1 6 1.2/islet, P , 0.01), but
3
a comparable number of a cells (16.5 6 1.1 vs. heterozygous mice: 18.5 6 1.4, P . 0.05). The number of b cells was comparable in the homozygous mice to that of the control mice (34.0 6 1.5/islet, P . 0.05), but significantly lower in the heterozygous mice than that of the control mice (P , 0.05). Furthermore, the number of a cells in both the homozygous and heterozygous mutant mice was apparently lower than that of the control mice (27.6 6 1.2/islet, P , 0.01) (Figure 2B). These findings translated into a noticeably higher b/a ratio in homozygous pdx1cre/+ Foxa2 FL/FL mice (2.23 6 0.02) compared with heterozygous pdx1cre/+ Foxa2 FL/+ mice (1.24 6 0.22) and the control mice (1.24 6 0.14) (P , 0.01) (Figure 2C), suggesting that Foxa2 deletion caused a relative dose-dependent increase in b-cell mass in the mutant mice. Consistently, direct Pdx1 cell lineage tracing showed that 98% of b cells and 70% of a cells were EYFP+ in the control mice (pdx1cre/+ EYFP+/2) were EYFP+. In the heterozygous mutant (pdx1cre/+ Foxa2 FL/FL EYFP+/2) mice, 88% of b cells were EYFP+ and 71% of a cells were EYFP+. By contrast, the homozygous mice (pdx1cre/+ Foxa2 FL/FL EYFP+/2) showed a significant reduction in the number of a cells (3% EYFP+), whereas 96% of b cells were EYFP+. These findings together indicate that Foxa2 deletion caused a noticeable reduction in a-cell mass and a corresponding dose-dependent increase in b-cell mass in the mutant mice. Homozygous Foxa2 deletion shifts Pdx1 expression to b cells in mouse pancreas In the mouse pancreas, direct Pdx1 cell lineage tracing revealed that, on embryonic day 18.5, in control mice (pdx1cre/+ EYFP+/2), b and a cells formed the core of the islet with other cells positioned on the periphery of the
FIGURE 1. Foxa2 is deleted in pancreatic cells in mice. Pancreas sections were prepared from mice on postnatal day 10 and double-immunolabeled for Foxa2 (pink) and insulin (green) as described in methods. Left panel: Foxa2 expression in pancreas sections of the heterozygous mutant (pdx1cre/+ Foxa2 FL/+) mice. Foxa2 is still expressed in occasional b cells (arrow) and non-b cells. Right panel: Foxa2 expression in pancreas sections of the homozygous mutant (pdx1cre/+ Foxa2 FL/FL) mice. Foxa2 is virtually absent from non-b cells and its expression almost exclusively coincides with that of insulin expression in b cells. www.americantherapeutics.com
American Journal of Therapeutics (2016) 0(0)
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
4
Wu et al
FIGURE 2. Foxa2 deletion causes an imbalance of a and b cells in mouse pancreas. Pancreas sections were prepared from mice on postnatal day 10 and double-immunolabeled for insulin (red) and glucagon (green) as described in methods. (A) Left panel: In immunostained of control pancreas sections, a and b cells form the core of the islet with non-a or b cells positioned on the periphery of the islet. a cells are the predominant cell type in the islet. Mid panel: The density of both a and b cells decreases in the pancreas sections of the heterozygous mutant mice. Right panel: In the pancreas of the homozygous mutant mice, a cells lose its predominance and b cells become predominant. (B) The number of a and b cells in the homozygous and heterozygous mutant mice. *P , 0.01 versus heterozygous mutant. (C) The ratio of b tο a cells in the homozygous and heterozygous mutant mice. *P , 0.01 versus heterozygous mutant.
islet (Figure 3A). Pdx1 expression coincided with that of insulin and glucagon. In the heterozygous mutant (pdx1cre/+ Foxa2 FL/FL EYFP+/2) mice, Pdx1 was expressed coincident mostly with that of insulin and glucagon and occasionally with non-b or non-a cells (Figure 3B). In the homozygous mutant (pdx1cre/+ Foxa2 FL/FL EYFP+/2) mice, Pdx1 expression coincided almost exclusively with that of insulin-secreting b cells (Figure 3C). These findings suggested that homozygous Foxa2 deletion shifted Pdx1 expression to b cells in mouse pancreas. Foxa2 deletion causes severe hypoglycemia in mice We then investigated the effect of Foxa2 deletion on plasma insulin, glucagon, and glucose levels. Chemiluminescence assays revealed that heterozygous pdx1cre/+ Foxa2 FL/+ mice had significantly lower insulin levels (0.36 6 0.07 ng/mL) than control mice (1.22 6 0.09 ng/mL) (P 5 0.0002) and homozygous pdx1cre/+ Foxa2 FL/FL mice (1.01 6 0.35 ng/mL) (P , 0.0001) (Figure 4A). However, no apparent difference was American Journal of Therapeutics (2016) 0(0)
observed between homozygous pdx1cre/+ Foxa2 FL/FL mice and control mice (P 5 0.099). Chemiluminescence assays further showed that tissue-specific deletion of Foxa2 significantly depressed plasma glucagon levels in both homozygous pdx1cre/+ Foxa2 FL/FL mice (0.014 6 0.003 ng/mL) and heterozygous pdx1cre/+ Foxa2 FL/+ mice (0.015 6 0.002 ng/mL) compared with control mice (0.038 6 0.013 ng/mL) (P 5 0.033 for homozygous mice and P 5 0.038 for heterozygous mice vs. control) (Figure 4B). Examination of plasma glucose on postnatal day 10 showed significantly lower plasma glucose content in homozygous pdx1cre/+ Foxa2 FL/FL mice (77.60 6 11.94 mg/dL) compared with control mice (122.50 6 8.39 mg/dL) (P , 0.0001) and heterozygous pdx1cre/+ Foxa2 FL/+ mice (95.25 6 9.11 mg/dL) (P 5 0.019) (Figure 4C). The homozygous mutant mice also developed seizures because of severe hypoglycemia (data not shown). Although the plasma glucose content was lower in heterozygous pdx1cre/+ Foxa2 FL/+ mice than control mice, no statistical difference was achieved (P . 0.05). These findings indicated that www.americantherapeutics.com
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Foxa2 and Hyperinsulinemic Hypoglycemia
5
FIGURE 3. Homozygous Foxa2 deletion shifts Pdx1 expression to b cells in mouse pancreas. Mesp1 cre/+ Foxa2 FL/+ transgenic mice were crossed to Foxa2 FL/FL ROSA26, EYFP2/2 mice to obtain heterozygous pdx1cre/+ Foxa2 FL/+ EYFP 6 and homozygous pdx1cre/+ Foxa2 FL/FL EYFP 6 mice. pdx1cre/+ EYFP 6 mice were used as control mice. Pancreas sections were prepared from mice on embryonic day 18.5 and stained for insulin (red) and glucagon (pink). (A) In the control mice, b and a cells form the core of the islet with other cells positioned on the periphery of the islet. (B) In heterozygous mutant mice, Pdx1 is expressed coincident mostly with that of insulin and glucagon-secreting cells and occasionally with non-b or non-a cells. (C) In homozygous mutant mice, Pdx1 expression coincides almost exclusively with that of insulin-secreting b cells.
tissue-specific deletion of Foxa2 caused inappropriate hyperinsulinemia and hypoglucagonemia in homozygous mutant mice, which contributed to severe hypoglycemia in these mice.
DISCUSSION In the present study, we generated homozygous and heterozygous conditionally ablated Foxa2 mutant mice by homologous recombination using a Foxa2 genetargeting vector with loxP sites flanking exon 3. We www.americantherapeutics.com
demonstrated that condition knockout of Foxa2 in Pdx1 expressing domains caused profound hypoglucagonemia and inappropriate hyperinsulinemia in the homozygous mutant mice, contributing to profound hypoglycemia in the animals. Furthermore, conditional tissue-specific Foxa2 ablation is associated an imbalance of b/a ratio as a result of significant but disproportional reduction in pancreatic a- and b-cell mass, indicating that in our knockout model, a cells are more responsive to Foxa2 ablation than b cells. Oliver-Krasinski et al9 generated hypomorphic Pdx1DC/DC mice by truncating the C terminus of Pdx1 American Journal of Therapeutics (2016) 0(0)
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
6
Wu et al
FIGURE 4. Foxa2 deletion causes severe hypoglycemia in mice. Plasma insulin was measured by ELISA (A), and plasma glucagon was determined using RIA (B) as described in methods. Plasma glucose content at postnatal day 10 was measured using an automatic glucose monitor (C). *P , 0.05 versus control and **P , 0.01 versus control.
at amino acid 210, and they found a global reduction in endocrine lineages in Pdx1DC/DC mice, especially a and b cells. In Pdx1+/DC mice, only b-cell area was reduced, indicating a greater sensitivity of the b-cell lineage to Pdx1 action. Our study also showed that conditional ablation of Foxa2, a major upstream regulator of Pdx1, led to apparent reduction in the number of a and b cells in the pancreas at postnatal day 10. In contrast to the finding by Oliver-Krasinski et al,9 our immunofluorescent analysis and direct Pdx1 lineage tracing revealed a greater reduction in the number of a cells, suggesting that a cells are more amenable to the actions of Foxa2. Furthermore, we observed a dose effect of Foxa2 as a higher b/a ratio was demonstrated in the homozygous mutant mice and the heterozygous mutant mice. Our findings are consistent with the present study by Heddad Masson et al10 who showed that small interfering RNA knockdown of Foxa2 led to impaired a-cell differentiation and glucagon synthesis, a process that also involves Foxa1. American Journal of Therapeutics (2016) 0(0)
Insulin deficiency due to reduced pancreatic islet b cell number underlies the progression of both type 1 and type 2 diabetes. Heterozygous mutations in Pdx1 are known to be associated with maturity onset diabetes of the young, type 4 (MODY4) and type 2 diabetes,11 and Foxa2 has been demonstrated to be as an essential activator of genes that function in multiple pathways governing insulin secretion and also regulates differentiation of a and b cells in the pancreas. Consistent with a significant reduction in the a-cell mass in the pancreas, profound hypoglucagonemia was demonstrated in both the heterozygous and homozygous mutant mice. Although we observed a marked reduction in plasma insulin in the heterozygous mutant mice, we failed to note a significant reduction in plasma insulin in the homozygous mutant mice. The plasma insulin levels in the homozygous mutant mice are inappropriately high considering severe hypoglycemia in these animals. This may be due to changes in islet cell lineage allocation, www.americantherapeutics.com
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Foxa2 and Hyperinsulinemic Hypoglycemia
defective regulation of insulin secretion, or decline in a-cell mass leading to an increase in insulin-producing b cells as a result of Foxa2 ablation. Plasma glucagon was approximately 3-fold lower in the homozygous mutant mice than that of the control mice, which, together with the inappropriate hyperinsulinemia, helps explain profound hypoglycemia in these animals. Our findings are consistent with the study by Sund et al12 who also observed hypoglycemia due to inappropriate hyperinsulinemia in mice with tissuespecific deletion of Foxa2 in pancreatic b cells. In conclusion, our study demonstrates that Foxa2 is required in both a and b cells for maintaining appropriate glucagon and insulin levels and for keeping glucose homeostasis. Elucidation of the regulation of normal and abnormal a and b cell differentiation offers insight into the pathogenesis of diabetes, and pinpoints novel targets for diabetes control and may lead to clues for reprogramming a and b cells for replacement therapy. Our conditional tissue-specific Foxa2 ablation mouse model will provide a unique model for delineating the role of a and b cells in the pathogenesis of diabetes and conditions such as familial hyperinsulinism.
REFERENCES 1. Ohlsson H, Karlsson K, Edlund T. IPF1, a homeodomaincontaining transactivator of the insulin gene. EMBO J. 1993;12:4251–4259. 2. Ahlgren U, Pfaff SL, Jessell TM, et al. Independent requirement for ISL1 in formation of pancreatic mesenchyme and islet cells. Nature. 1997;385:257–260.
www.americantherapeutics.com
7 3. Brissova M, Shiota M, Nicholson WE, et al. Reduction in pancreatic transcription factor PDX-1 impairs glucosestimulated insulin secretion. J Biol Chem. 2002;277: 11225–11232. 4. Hale MA, Kagami H, Shi L, et al. The homeodomain protein PDX1 is required at mid-pancreatic development for the formation of the exocrine pancreas. Dev Biol. 2005; 286:225–237. 5. Lee CS, Sund NJ, Vatamaniuk MZ, et al. Foxa2 controls Pdx1 gene expression in pancreatic beta-cells in vivo. Diabetes. 2002;51:2546–2551. 6. Gao N, LeLay J, Vatamaniuk MZ, et al. Dynamic regulation of Pdx1 enhancers by Foxa1 and Foxa2 is essential for pancreas development. Genes Dev. 2008;22:3435–3448. 7. Lee CS, Sund NJ, Behr R, et al. Foxa2 is required for the differentiation of pancreatic alpha-cells. Dev Biol. 2005; 278:484–495. 8. Cheng N, Hu X, Tian T, et al. PKMzeta knockdown disrupts post-ischemic long-term potentiation via inhibiting postsynaptic expression of aminomethyl phosphonic acid receptors. J Biomed Res. 2015;29:241–249. 9. Oliver-Krasinski JM, Kasner MT, Yang J, et al. The diabetes gene Pdx1 regulates the transcriptional network of pancreatic endocrine progenitor cells in mice. J Clin Invest. 2009;119:1888–1898. 10. Heddad Masson M, Poisson C, Guerardel A, et al. Foxa1 and Foxa2 regulate alpha-cell differentiation, glucagon biosynthesis, and secretion. Endocrinology. 2014;155: 3781–3792. 11. Shih DQ, Stoffel M. Molecular etiologies of MODY and other early-onset forms of diabetes. Curr Diab Rep. 2002; 2:125–134. 12. Sund NJ, Vatamaniuk MZ, Casey M, et al. Tissue-specific deletion of Foxa2 in pancreatic beta cells results in hyperinsulinemic hypoglycemia. Genes Dev. 2001;15: 1706–1715.
American Journal of Therapeutics (2016) 0(0)
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
