Portal de Periódicos da CAPES

O-linked N-acetylglucosamine transferase (OGT) regulates pancreatic α-cell function in mice

2021; Elsevier BV; Volume: 296; Linguagem: Inglês

10.1016/j.jbc.2021.100297

1083-351X

Ahmad Essawy, Seokwon Jo, Megan Beetch, Amber Lockridge, Eric Gustafson, Emilyn U. Alejandro,

Carbohydrate Chemistry and Synthesis

The nutrient sensor O-GlcNAc transferase (OGT) catalyzes posttranslational addition of O-GlcNAc onto target proteins, influencing signaling pathways in response to cellular nutrient levels. OGT is highly expressed in pancreatic glucagon-secreting cells (α-cells), which secrete glucagon in response to hypoglycemia. The objective of this study was to determine whether OGT is necessary for the regulation of α-cell mass and function in vivo. We utilized genetic manipulation to produce two α-cell specific OGT-knockout models: a constitutive glucagon-Cre (αOGTKO) and an inducible glucagon-Cre (i-αOGTKO), which effectively delete OGT in α-cells. Using approaches including immunoblotting, immunofluorescent imaging, and metabolic phenotyping in vivo, we provide the first insight on the role of O-GlcNAcylation in α-cell mass and function. αOGTKO mice demonstrated normal glucose tolerance and insulin sensitivity but displayed significantly lower glucagon levels during both fed and fasted states. αOGTKO mice exhibited significantly lower α-cell glucagon content and α-cell mass at 6 months of age. In fasting, αOGTKO mice showed impaired pyruvate stimulated gluconeogenesis in vivo and reduced glucagon secretion in vitro. i-αOGTKO mice showed similarly reduced blood glucagon levels, defective in vitro glucagon secretion, and normal α-cell mass. Interestingly, both αOGTKO and i-αOGTKO mice had no deficiency in maintaining blood glucose homeostasis under fed or fasting conditions, despite impairment in α-cell mass and function, and glucagon content. In conclusion, these studies provide a first look at the role of OGT signaling in the α-cell, its effect on α-cell mass, and its importance in regulating glucagon secretion in hypoglycemic conditions. The nutrient sensor O-GlcNAc transferase (OGT) catalyzes posttranslational addition of O-GlcNAc onto target proteins, influencing signaling pathways in response to cellular nutrient levels. OGT is highly expressed in pancreatic glucagon-secreting cells (α-cells), which secrete glucagon in response to hypoglycemia. The objective of this study was to determine whether OGT is necessary for the regulation of α-cell mass and function in vivo. We utilized genetic manipulation to produce two α-cell specific OGT-knockout models: a constitutive glucagon-Cre (αOGTKO) and an inducible glucagon-Cre (i-αOGTKO), which effectively delete OGT in α-cells. Using approaches including immunoblotting, immunofluorescent imaging, and metabolic phenotyping in vivo, we provide the first insight on the role of O-GlcNAcylation in α-cell mass and function. αOGTKO mice demonstrated normal glucose tolerance and insulin sensitivity but displayed significantly lower glucagon levels during both fed and fasted states. αOGTKO mice exhibited significantly lower α-cell glucagon content and α-cell mass at 6 months of age. In fasting, αOGTKO mice showed impaired pyruvate stimulated gluconeogenesis in vivo and reduced glucagon secretion in vitro. i-αOGTKO mice showed similarly reduced blood glucagon levels, defective in vitro glucagon secretion, and normal α-cell mass. Interestingly, both αOGTKO and i-αOGTKO mice had no deficiency in maintaining blood glucose homeostasis under fed or fasting conditions, despite impairment in α-cell mass and function, and glucagon content. In conclusion, these studies provide a first look at the role of OGT signaling in the α-cell, its effect on α-cell mass, and its importance in regulating glucagon secretion in hypoglycemic conditions. In the pancreas, glucose homeostasis is mainly regulated by the islets of Langerhans, which consist of different endocrine cells including the β-cells and α-cells (1Da Silva Xavier G. The cells of the islets of Langerhans.J. Clin. Med. 2018; 7: 54Crossref Google Scholar). β-cells secrete insulin in response to high blood glucose. In contrast, α-cells secrete glucagon in response to low blood glucose levels. Together, these endocrine cells, secreting their counter regulatory hormones, work together to maintain blood glucose at a physiological level (2Folli F. La Rosa S. Finzi G. Davalli A.M. Galli A. Dick Jr., E.J. Perego C. Mendoza R.G. Pancreatic islet of Langerhans' cytoarchitecture and ultrastructure in normal glucose tolerance and in type 2 diabetes mellitus.Diabetes Obes. Metab. 2018; 20 Suppl 2: 137-144Crossref PubMed Scopus (22) Google Scholar). For both type 1 (T1D) and type 2 (T2D) diabetes, a clear role for absolute deficiency or insufficiency in insulin has been established (3Donath M.Y. Halban P.A. Decreased beta-cell mass in diabetes: Significance, mechanisms and therapeutic implications.Diabetologia. 2004; 47: 581-589Crossref PubMed Scopus (307) Google Scholar). In addition to altered insulin levels, dysregulation of glucagon levels in T1D and T2D contributes to the pathology of these diseases (4D'Alessio D. The role of dysregulated glucagon secretion in type 2 diabetes.Diabetes Obes. Metab. 2011; 13 Suppl 1: 126-132Crossref PubMed Scopus (156) Google Scholar, 5Wewer Albrechtsen N.J. Kuhre R.E. Pedersen J. Knop F.K. Holst J.J. The biology of glucagon and the consequences of hyperglucagonemia.Biomark. Med. 2016; 10: 1141-1151Crossref PubMed Scopus (44) Google Scholar). High glucagon levels in both T1D and T2D exacerbate hyperglycemia due to enhanced hepatic glucose output (6Lund A. Bagger J.I. Christensen M. Knop F.K. Vilsboll T. Glucagon and type 2 diabetes: The return of the alpha cell.Curr. Diab. Rep. 2014; 14: 555Crossref PubMed Scopus (65) Google Scholar). While the cause and result of functional β-cell deficiency have been studied heavily in the field, the mechanisms and impact of α-cell failure are less known. Thus, a greater understanding of processes that regulate α-cell function may present new avenues for optimal glucose control in T1D or advanced therapy for T2D patients. The hexosamine biosynthetic pathway (HBP) is a minor branch of glycolysis responsible for the production of the key substrate for protein glycosylation, UDP-GlcNAc (7Love D.C. Hanover J.A. The hexosamine signaling pathway: Deciphering the "O-GlcNAc code".Sci. STKE. 2005; 2005re13PubMed Google Scholar). Posttranslational addition of UDP-GlcNAc (known as O-GlcNAcylation) is a dynamic and reversible process analogous to phosphorylation and has been shown to affect the function, stability, and subcellular localization of many proteins (8Hanover J.A. Krause M.W. Love D.C. The hexosamine signaling pathway: O-GlcNAc cycling in feast or famine.Biochim. Biophys. Acta. 2010; 1800: 80-95Crossref PubMed Scopus (231) Google Scholar). O-GlcNAcylation is catalyzed solely by the enzyme O-linked GlcNAc transferase (OGT), by means of the addition of UDP-GlcNAc to the serine or threonine residues on nuclear and cytosolic proteins (9Hart G.W. Akimoto Y. The O-GlcNAc modification.in: Varki A. Cummings R.D. Esko J.D. Freeze H.H. Stanley P. Bertozzi C.R. Hart G.W. Etzler M.E. Essentials of Glycobiology Chapter 18. Cold Spring Harbor, NY2009Google Scholar). Conversely, the O-GlcNAc is removed by the enzyme O-linked β-N-acetyl hexosaminidase (O-GlcNAcase or OGA) (10Gloster T.M. Vocadlo D.J. Mechanism, structure, and inhibition of O-GlcNAc processing enzymes.Curr. Signal Transduct. Ther. 2010; 5: 74-91Crossref PubMed Scopus (42) Google Scholar). Approximately 3 to 5% of glucose entering the cell is shunted toward use by the HBP, and therefore, the degree of O-GlcNAcylation present in the cell is dependent on the amount of nutrients (glucose) present in the cellular environment (11Vasconcelos-Dos-Santos A. Oliveira I.A. Lucena M.C. Mantuano N.R. Whelan S.A. Dias W.B. Todeschini A.R. Biosynthetic machinery involved in aberrant glycosylation: Promising targets for developing of drugs against cancer.Front. Oncol. 2015; 5: 138Crossref PubMed Scopus (78) Google Scholar). OGT is richly expressed in the pancreas (12Akimoto Y. Kreppel L.K. Hirano H. Hart G.W. Localization of the O-linked N-acetylglucosamine transferase in rat pancreas.Diabetes. 1999; 48: 2407-2413Crossref PubMed Scopus (56) Google Scholar, 13Lubas W.A. Frank D.W. Krause M. Hanover J.A. O-Linked GlcNAc transferase is a conserved nucleocytoplasmic protein containing tetratricopeptide repeats.J. Biol. Chem. 1997; 272: 9316-9324Abstract Full Text Full Text PDF PubMed Scopus (375) Google Scholar). At the protein level, OGT appears to be more abundantly expressed in islets compared with the acinar tissue (14Alejandro E.U. Bozadjieva N. Kumusoglu D. Abdulhamid S. Levine H. Haataja L. Vadrevu S. Satin L.S. Arvan P. Bernal-Mizrachi E. Disruption of O-linked N-acetylglucosamine signaling induces ER stress and beta cell failure.Cell Rep. 2015; 13: 2527-2538Abstract Full Text Full Text PDF PubMed Scopus (23) Google Scholar). Within the islet, high OGT mRNA expression has been detected in both the α- and β-cells (12Akimoto Y. Kreppel L.K. Hirano H. Hart G.W. Localization of the O-linked N-acetylglucosamine transferase in rat pancreas.Diabetes. 1999; 48: 2407-2413Crossref PubMed Scopus (56) Google Scholar, 15Hanover J.A. Lai Z. Lee G. Lubas W.A. Sato S.M. Elevated O-linked N-acetylglucosamine metabolism in pancreatic beta-cells.Arch. Biochem. Biophys. 1999; 362: 38-45Crossref PubMed Scopus (107) Google Scholar). Recently, our group reported that O-GlcNAcylation is essential for cell survival and function. We reported that loss of O-GlcNAcylation in the β-cell leads to cell failure and diabetes in mice due to increased endoplasmic reticulum (ER) stress and apoptosis (14Alejandro E.U. Bozadjieva N. Kumusoglu D. Abdulhamid S. Levine H. Haataja L. Vadrevu S. Satin L.S. Arvan P. Bernal-Mizrachi E. Disruption of O-linked N-acetylglucosamine signaling induces ER stress and beta cell failure.Cell Rep. 2015; 13: 2527-2538Abstract Full Text Full Text PDF PubMed Scopus (23) Google Scholar). Ablation of OGT in pancreatic progenitors also leads to an increased apoptosis (16Baumann D. Wong A. Akhaphong B. Jo S. Pritchard S. Mohan R. Chung G. Zhang Y. Alejandro E.U. Role of nutrient-driven O-GlcNAc-posttranslational modification in pancreatic exocrine and endocrine islet development.Development. 2020; 147dev186643Crossref PubMed Scopus (4) Google Scholar). Independent of its role in cell survival, OGT also regulates insulin secretion at basal (14Alejandro E.U. Bozadjieva N. Kumusoglu D. Abdulhamid S. Levine H. Haataja L. Vadrevu S. Satin L.S. Arvan P. Bernal-Mizrachi E. Disruption of O-linked N-acetylglucosamine signaling induces ER stress and beta cell failure.Cell Rep. 2015; 13: 2527-2538Abstract Full Text Full Text PDF PubMed Scopus (23) Google Scholar) and in obesity conditions in part through SERCA2 (17Lockridge A.,J.S. Gustafson E. Damberg N. Mohan R. Olson M. Abrahante J. Alejandro E.U. Islet O-GlcNAcylation is required for lipid-potentiation of insulin secretion through SERCA2.Cell Rep. 2020; 31: 107609Abstract Full Text Full Text PDF PubMed Scopus (5) Google Scholar). Moreover, OGT regulates insulin processing via downstream target eIF4G1, a translation initiation factor (18Jo S. Lockridge A. Alejandro E.U. eIF4G1 and carboxypeptidase E axis dysregulation in O-GlcNAc transferase-deficient pancreatic beta cells contributes to hyperproinsulinemia in mice.J. Biol. Chem. 2019; 294: 13040-13050Abstract Full Text Full Text PDF PubMed Scopus (6) Google Scholar). However, as a nutrient sensor protein that is highly expressed in glucagon-secreting cells, the role of OGT in α-cells has not been explored. We hypothesize that OGT plays a key role in the maintenance of α-cell mass and proper function of secreting glucagon in response to hypoglycemia. It is unknown how nutrient-driven posttranslational O-GlcNAcylation of proteins impacts pancreas α-cell mass and function. In the currently study, through the characterization of mice lacking α-cell OGT, the only enzyme capable of adding O-GlcNAc modification onto proteins, we show that O-GlcNAcylation is necessary for the maintenance of α-cell mass and regulation of glucagon secretion. High expression of OGT mRNA has been reported in the pancreas (12Akimoto Y. Kreppel L.K. Hirano H. Hart G.W. Localization of the O-linked N-acetylglucosamine transferase in rat pancreas.Diabetes. 1999; 48: 2407-2413Crossref PubMed Scopus (56) Google Scholar). Within the islet, it is controversial whether glucagon-producing α-cells or insulin-producing β-cells express more OGT mRNA (12Akimoto Y. Kreppel L.K. Hirano H. Hart G.W. Localization of the O-linked N-acetylglucosamine transferase in rat pancreas.Diabetes. 1999; 48: 2407-2413Crossref PubMed Scopus (56) Google Scholar, 15Hanover J.A. Lai Z. Lee G. Lubas W.A. Sato S.M. Elevated O-linked N-acetylglucosamine metabolism in pancreatic beta-cells.Arch. Biochem. Biophys. 1999; 362: 38-45Crossref PubMed Scopus (107) Google Scholar). Therefore, we first sought to compare protein levels of OGT and OGA between α-cells and β-cells cell lines due to the limited number and difficulty of selecting α-cells in primary islets. Baseline levels of OGT and OGA protein were measured in αTC-1 and βTC-6 immortalized cell lines (Fig. 1A). OGT levels were similar in both cell types in cell lines (Fig. 1B, Fig. S1, A and B). However, there was a significantly lower level of OGA protein present in the αTC-1 cell line compared with the βTC-6 cell line (Fig. 1C, Fig. S1, A and C). Interestingly, the level of O-GlcNAcylation, measured by the RL2 antibody, was comparable between both cell lines (Fig. S1D). RL2 is a specific O-GlcNAc antibody that has been validated in our lab (18Jo S. Lockridge A. Alejandro E.U. eIF4G1 and carboxypeptidase E axis dysregulation in O-GlcNAc transferase-deficient pancreatic beta cells contributes to hyperproinsulinemia in mice.J. Biol. Chem. 2019; 294: 13040-13050Abstract Full Text Full Text PDF PubMed Scopus (6) Google Scholar) and others (19Tashima Y. Stanley P. Antibodies that detect O-linked beta-D-N-acetylglucosamine on the extracellular domain of cell surface glycoproteins.J. Biol. Chem. 2014; 289: 11132-11142Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar, 20Snow C.M. Senior A. Gerace L. Monoclonal antibodies identify a group of nuclear pore complex glycoproteins.J. Cell Biol. 1987; 104: 1143-1156Crossref PubMed Scopus (361) Google Scholar). Together, these data show that OGT and OGA are expressed in α-cells, and thus the role of O-GlcNAcylation in α-cell physiology should be tested in vivo. Therefore, we next tested the hypothesis that OGT is required for the maintenance of α-cell mass and for appropriate α-cell function in vivo. We utilized genetic manipulation to generate a mouse model with α-cell specific OGT deletion in vivo. αOGTKO mice were produced by crossing Gcg-Cre mice (constitutive KO) with OGTflox/flox mice. To facilitate lineage tracing, αOGTKO mice were also crossed with a fluorescent Cre reporter (either RFP or GFP) to mark all cells exhibiting the Cre activity. The Gcg-Cre recombination efficiency was previously calculated at 94 to 97% of α-cells, whereas it was detected in a negligible (∼0.2%) proportion of β-cells (21Shiota C. Prasadan K. Guo P. Fusco J. Xiao X. Gittes G.K. Gcg (CreERT2) knockin mice as a tool for genetic manipulation in pancreatic alpha cells.Diabetologia. 2017; 60: 2399-2408Crossref PubMed Scopus (9) Google Scholar). In our hands, we detected RFP, by immunofluorescent staining, colocalized with glucagon-expressing cells of the islet in Gcg-cre, OGTWT mice (Fig. 1D). We then compared the staining patterns in Gcg-cre, OGTWT mice with those in αOGTKO mice. In addition, we conducted RL2 staining to test whether genetic deletion of OGT in the α-cells had an effect on O-GlcNAcylation. At 1 month of age, αOGTKO mice showed reduced RL2 staining in glucagon-positive cells in comparison with the control (Fig. 1E). Importantly, the glucagon-positive cells with reduced RL2 staining were also positive for RFP, suggesting activation of the Gcg-Cre in these cells. These data confirmed that our αOGT deletion resulted in a decrease in α-cell O-GlcNAcylation. After confirming that OGT deletion reduced O-GlcNAcylation in α-cells, we next sought to assess the metabolic health of the αOGTKO mice in fed and fasted states, in order to determine what effect this deficit has on islet function. In nonfasted states, male and female αOGTKO mice showed normal blood glucose levels (Fig. 2, A and B) as well as no alterations in serum insulin levels, compared with controls (Fig. S2, A and B). When we analyzed serum for glucagon levels, however, male αOGTKO mice displayed significantly reduced levels of serum glucagon in the fed state (Fig. 2C). Comparable level of glucagon was observed in female αOGTHET and control (Fig. 2D). Female αOGTKO mice showed a trend toward reduction of serum glucagon levels in nonfasted state, but the results did not reach significance (Fig. 2D). Both male and female αOGTKO mice displayed normal glucose tolerance (Fig. 2, E and F). Additionally, neither group showed differences in insulin sensitivity via IP insulin tolerance test (Fig. 2, G and H). These results show that in spite of their reduced serum glucagon levels, αOGTKO mice manage to maintain normal glucose homeostasis. Next, we challenged the αOGTKO mice by performing a fasting experiment. Male control and αOGTKO mice showed similar decrease in blood glucose after 18 h of fasting (Fig. 3A). Control and αOGTKO blood glucose levels did level off and even begin to increase between 18 and 24 h of fasting, and there was a trend toward higher increase in the αOGTKO (Fig. 3A). Female control and partial or full αOGTKO mice also showed similar decrease in blood glucose after 16 h of fasting, with no marked differences in glucose level (Fig. 3B). During this fasting period, male αOGTKO circulating glucagon levels were significantly lower than the control (Fig. 3C). Female αOGTKO circulating glucagon levels were significantly lower at the 16-h fasted timepoint (Fig. 3D). These knockout mice failed to increased serum glucagon levels in response to the fasting. When subjected to an IP pyruvate tolerance test, which measures the gluconeogenesis capacity of the mice in a hypoglycemic condition, both male (Fig. 3E) and female (Fig. 3F) αOGTKO mice exhibited significantly reduced blood glucose in response to pyruvate injection compared with the control, indicating impaired gluconeogenesis. Four-month-old male αOGTKO mice exhibited normal glucagon sensitivity (Fig. 3G). Liver glycogen measurement in both male and female mice showed no significant differences (Fig. 3H). These results show that αOGTKO mice were able to maintain normal glucose homeostasis, despite evidence of reduced glucagon levels in the circulation. After observing reduced glucagon levels in αOGTKO mice, we determined whether loss of α-cell mass can contribute to this effect. Morphometric analysis of 1-month-old male αOGTKO pancreata demonstrated that α-cell mass at postweaning age was not statistically different between genotypes (Fig. 4, A and B). However, α-cell mass was significantly decreased in 6-month-old male αOGTKO mice with an average mass of approximately 0.08 mg in the αOGTKO, compared with an average of approximately 0.33 in the control (Fig. 4, C and D). Representative images of the whole pancreas were shown in Fig. S3, A and B. Based on these data, we hypothesized that α-cells were diminishing over time in αOGTKO mice. Due to the heterogeneity of islet cells, and limitations imposed by using primary islets from αOGTKO mice (reduced number of α-cells in αOGTKO islets), we used αTC-1 cells to assess the effect of OGT inhibition on α-cell death. We used the OGT inhibitor, OSMI-1 (50 μM), to reduce O-GlcNAcylation in αTC-1 cells. OSMI-1 treatment of αTC-1 cells decreased O-GlcNAcylation measured at both 8 and 24 h timepoints compared with control (DMSO) samples (Fig. S4A), demonstrating successful reduction of O-GlcNAcylation. We then assessed the protein level of apoptosis markers, cleaved-caspase-3 and cleaved-Poly (ADP-ribose) polymerase (PARP) (Fig. 4, E–G). During apoptosis, PARP, a nuclear DNA-binding protein, which detects DNA strand breaks and performs in base excision repair, is cleaved by caspase-3. Here, we observed a trend of increased levels of cleaved-PARP and a significant increase in cleaved-caspase-3 level after 24 h treatment with OSMI-1. We previously reported that β-cell OGT-deficient cells display ER stress-mediated apoptosis (14Alejandro E.U. Bozadjieva N. Kumusoglu D. Abdulhamid S. Levine H. Haataja L. Vadrevu S. Satin L.S. Arvan P. Bernal-Mizrachi E. Disruption of O-linked N-acetylglucosamine signaling induces ER stress and beta cell failure.Cell Rep. 2015; 13: 2527-2538Abstract Full Text Full Text PDF PubMed Scopus (23) Google Scholar), and we show in OGT inhibited αTC-1 cells (Fig. S4B), induction of Bip, a marker of ER stress (Fig. 4, H and I). These data suggest that increased ER stress may contribute to cell death in αTC-1 cells when O-GlcNAcylation is blocked, consistent with the reported effects of OGT inhibition in other pancreatic cells (14Alejandro E.U. Bozadjieva N. Kumusoglu D. Abdulhamid S. Levine H. Haataja L. Vadrevu S. Satin L.S. Arvan P. Bernal-Mizrachi E. Disruption of O-linked N-acetylglucosamine signaling induces ER stress and beta cell failure.Cell Rep. 2015; 13: 2527-2538Abstract Full Text Full Text PDF PubMed Scopus (23) Google Scholar, 16Baumann D. Wong A. Akhaphong B. Jo S. Pritchard S. Mohan R. Chung G. Zhang Y. Alejandro E.U. Role of nutrient-driven O-GlcNAc-posttranslational modification in pancreatic exocrine and endocrine islet development.Development. 2020; 147dev186643Crossref PubMed Scopus (4) Google Scholar) in vivo. After establishing the effect of OGT loss on α-cell mass, we next sought to determine the effect on α-cell function. To assess α-cell function, we subjected islets from 2-month-old male mice to an in vitro glucose and arginine-inhibited glucagon secretion test in vitro. Younger mice were used for this test in order to avoid major confounding effect of reduced α-cell number present per islet, since our data show a severe reduction of α-cell mass in 6-month-old mice. Islets isolated from 2-month-old male αOGTKO mice showed a significant impairment in glucagon secretion compared with control, as shown by the significantly lower amount of glucagon released when islets were treated with 1 mM glucose +20 mM arginine condition for 1 h (Fig. 5A). A trend toward a lower level of total islet glucagon content (normalized by islet DNA) was also observed (Fig. 5B). This indicated to us that the secretory defect in αOGTKO islets was not likely due to a defect in secretory mechanism, but rather could be attributed to lower glucagon content per single α-cell in the αOGTKO islets. Assessment of total pancreas glucagon content was not ideal due to the defect in α-cell mass. Therefore, to directly assess glucagon content at the single cell level, we dispersed islets from αOGTKO mice with GFP reporter for visualization and hand-picked GFP cells under the microscope (Fig. S5A). We determined that the averaged glucagon content per cell in αOGTKO mice was significantly reduced compared with that of the control cells (Fig. 5C). To test whether glucagon biosynthesis is altered in the islets, we assessed for glucagon transcript and observed a reduction in i-αOGTKO islets (Fig. S4C). We also treated αTC-1 cells acutely with OSMI-1 as before. We found decreased mRNA levels of α-cells transcription factors (Arx, Nkx2.2, and FOXA2, Fig. 5, D–F) as well as protein level of FOXA2 in OGT-inhibited αTC-1 cells (Fig. 5, H and I). However, no significant reduction in glucagon mRNA level was detected in the OSMI-1-treated αTC-1 cells (Fig. 5G). Then, we assessed the protein level of the prohormone convertase, Carboxypeptidase E (CPE), which was shown to regulate total glucagon content in αTC-1 cells (22McGirr R. Guizzetti L. Dhanvantari S. The sorting of proglucagon to secretory granules is mediated by carboxypeptidase E and intrinsic sorting signals.J. Endocrinol. 2013; 217: 229-240Crossref PubMed Scopus (16) Google Scholar), and found it decreased in OSMI-1 treated αTC-1 cells (Fig. 5, H and J). Decreased CPE protein was correlated with the elevated proglucagon levels by western blot (Fig. 5, H and K), which suggested a defect in glucagon processing. Although glucagon transcription was not altered in acute treatment of αTC-1 cells with OSMI-1, glucagon mRNA was reduced in islets of iαOGTKO mice. Next, we aimed to test whether FOXA2 was O-GlcNAc-modified in α-cells. FOXA2, a potent regulator of glucagon transcription, was projected to be O-GlcNAc by a prediction software (23Gupta R. Brunak S. Prediction of glycosylation across the human proteome and the correlation to protein function.Pac. Symp. Biocomput. 2002; : 310-322PubMed Google Scholar), at multiple potential O-GlcNAc sites. In αTC-1 cells, we confirmed that FOXA2 was indeed an OGT target, via IP pull-down of the protein (Fig. 5, L and M). Altogether, these findings illustrate the resultant deficiencies of αOGTKO, consisting of reduced α-cell mass in older mice and a deficiency in islet glucagon secretion and single α-cell content in younger mice. αOGTKO mice displayed reduced α-cell mass, reduced glucagon levels, and impaired gluconeogenesis. This, however, did not affect their ability to maintain normal blood glucose levels. Next, we sought to corroborate these findings in the inducible i-αOGTKO model, thus observing the effect of acute OGT deletion in adult mice. We generated i-αOGTKO animals by crossing male Gcg-CreERTM (21Shiota C. Prasadan K. Guo P. Fusco J. Xiao X. Gittes G.K. Gcg (CreERT2) knockin mice as a tool for genetic manipulation in pancreatic alpha cells.Diabetologia. 2017; 60: 2399-2408Crossref PubMed Scopus (9) Google Scholar) (inducible KO) and female with OGTflox/flox mice. Morphometric analysis of i-αOGTKO mouse pancreata was conducted at 15 weeks post-tamoxifen induction (Fig. 6A). Expression of RFP cells in glucagon cells demonstrated the efficiency of the tamoxifen in inducing Cre recombination (Fig. S5B). We assessed α-cell mass at 15 weeks post-tamoxifen, and we uncovered that i-αOGTKO and age-matched control mice show comparable α-cell mass (Fig. 6B). I-αOGTKO mice demonstrated normal blood glucose level (Fig. 6C), fasting and fed serum insulin (Fig. S6, A and B) compared with the control. In line with the constitutive αOGTKO model, i-αOGTKO also displayed a significantly reduced random blood glucagon levels (Fig. 6D). The defect in glucagon secretion from islets of i-αOGTKO in vitro was confirmed (Fig. 6E), as well as loss of glucagon content (Fig. S6C). Additionally, iαOGTKO mice exhibited significant pyruvate intolerance (Fig. 6F) and normal glucagon sensitivity (Fig. 6G). No body weight phenotype was observed in the iαOGTKO. Body weights of experimental mice (two different cohorts) at 15 or 24 weeks post-tamoxifen induction showed no significantly different increase in body weight than the age-matched controls (Fig. 6, H and I, Fig. S6, D and E). Glucose tolerance was also noted to be comparable between the control and iαOGTKO at 24 weeks post-tamoxifen induction (Fig. 6J, Fig. S6F). These findings demonstrated that acute deletion of OGT in adult mice had a similar effect on glucagon level but differed from αOGTKO through its milder effect on α-cell mass. Unlike the iαOGTKO model, the αOGTKO mice displayed a body weight phenotype. Weekly weighing of mice under normal chow diet revealed heavier body weight over time in male αOGTKO mice, which became significant at 2 months of age and continued to increase in until 3 months of age, body weight measurement from 1 month to 3-month-old mice is shown (Fig. 7A). In sharp contrast, αOGTKO and control females did not show differences in body weight (Fig. 7A). To analyze contributing body mass differences in these mice, we conducted an Echo-MRI imaging in young and old cohort of mice. MRI imaging of 1-month-old male αOGTKO mice showed no significant differences in lean mass or fat mass (Fig. 7B). In addition, metabolic cage analysis in these mice showed no significant differences in heat expenditure between αOGTKO and control mice (Fig. 7C). Food intake studies between 2 and 3 months of age revealed that αOGTKO mice, at this timepoint, were eating a significantly higher average amount of chow per month (Fig. 7D). We also performed the metabolic cages study at 6 months of age, where male αOGTKO mice had become significantly heavier than their control counterparts, with Echo-MRI analysis revealing significantly increased levels of both fat and lean mass (Fig. 7E). When corrected with body weight, there was a nonsignificant difference in heat expenditure between αOGTKO and control mice (Fig. 7F). Consistent with the progression of obesity, we found normal circulating leptin level in 2 to 3 months old mice, but significantly increased leptin levels in 6-month-old αOGTKO mice (Fig. S6, G and H). These findings show a phenotype of obesity in part due to hyperphagia and not due to defect in energy balance, and this was discovered only in male, but not in female αOGTKO or in male i-αOGTKO mice 3 months (Fig. 6H) or 5 months post-tamoxifen injection (Fig. 6I). Visual differences in fat deposition between male αOGTKO and control mice was observed in 6-month-old mice at the time of sacrifice (Fig. 7G). Next, we sought to determine the mechanisms behind the obesity in αOGTKO mice. Deletion of OGT in α-CaMKII-positive periventricular nucleus (PVN) neurons in the hypothalamus has been previously shown to cause obesity in part due to hyperphagia (24Lagerlof O. Slocomb J.E. Hong I. Aponte Y. Blackshaw S. Hart G.W. Huganir R.L. The nutrient sensor OGT in PVN neurons regulates feeding.Science. 2016; 351: 1293-1296Crossref PubMed Scopus (73) Google Scholar). After observing hyperphagia-induced obesity in the αOGTKO mice, we hypothesized that deletion of OGT was