Genetics, diet and obesity contribute to the current epidemic of human type 2 diabetes. A substantial proportion of the human population seems predisposed to a combination of these factors, perhaps analogous to different mouse strains that are resistant or susceptible to diet- and obesity-induced diabetes3
. Our findings show that the extent of disruption of beta cell glycosylation and glucose transport by diet and obesity directly contributes to disease onset and severity in susceptible mice, illuminating a pathogenic pathway that encompasses lipotoxicity and glucotoxicity32
. This pathway seems to be conserved in normal human islet cells and is activated in islets from donors with type 2 diabetes. A pathogenic tipping point in this pathway may occur when elevated free fatty acid (FFA) concentrations impair the expression and function of FOXA2 and HNF A transcription factors sufficiently in beta cells to deplete GnT-4a glycosylation and glucose transporter expression. The resulting dysfunction of beta cells leads to impaired glucose tolerance and failure of GSIS and further contributes to hyperglycemia, hepatic steatosis and systemic insulin resistance. Preservation of beta cell GnT-4a glycosylation and glucose transporter expression breaks this pathogenic cycle and its link to diet and obesity.
The nuclear exclusion and functional attenuation of transcription factors Foxa2 and Hnf a induced by high-fat diet in mice was similarly observed in islet cells from human donors with type 2 diabetes as well as in mouse and human islet cell cultures augmented with the FFA palmitic acid. The linkage of palmitic acid abundance to diminished GnT-4a activity and glucose transport can further explain how elevated FFA concentrations contribute to beta cell dysfunction and the onset of diabetes. FFAs induce the activation of one or more beta cell G protein–coupled FFA receptor such as GPR40, GPR119 and GPR120 (ref. 33
). Although FFAs seem to promote beta cell function in some contexts, the chronic elevation of FFAs increases mitochondrial oxidation and reactive oxygen species and has been observed in beta cell cultures with diminished GSIS and reduced glucose transporter expression34,35
. Our observations that the antioxidant N
-acetylcysteine inhibits FFA-induced attenuation of Foxa2 and Hnf a function are consistent with those findings. Nuclearcytoplasmic shuttling of Foxa2 has been observed in hepatocytes, where attenuation of Foxa2 contributes to the onset of steatosis and insulin resistance36
. In addition, Foxa2 inactivation has been biochemically linked to phosphorylation at Thr 56, independent of nuclear exclusion, and involving the Akt kinase, which can be activated in response to insulin or stimuli that induce oxidative stress in beta cells37,38
. Whereas both Foxa2 and Hnf a contribute to the expression of GnT-4a and the glucose transporters in beta cells, a complete absence of Foxa2 engineered by genetic ablation can alone cause failure of GSIS39,40
. Our findings support the role of Foxa2 and Hnf a as key metabolic regulators of energy homeostasis pathways that include beta cell responses to nutritional cues.
The connections established among beta cell HNF A function and the expression of GnT-4a and the glucose transporters suggest that the pathogenesis of diet- and obesity-associated type 2 diabetes may occur by a similar mechanism to that operating in mature onset diabetes of the young subtype 3 (MODY3), the most common form of human MODY41
. Hnf a transactivates the mouse Slc2a2
and we have found that expression of the human MGAT4A
gene is also maintained by HNF A function. The presence of multiple HNF A-binding sites in the promoter regions of human MGAT4A
genes is consistent with the possibility that singular or combined decreases in human GnT-4a and glucose transporter expression in beta cells may be responsible for the loss of GSIS in MODY3.
The altered expression of hundreds of genes has been reported in comparisons of human islet cells from normal donors and donors with type 2 diabetes43
. In those studies and in ours reported here, the MGAT4A
gene is substantially and similarly downregulated, although the decrease is relatively modest and might not garner attention when the expression of other genes is altered by tenfold or more. Yet this 60% reduction of MGAT4A
RNA expression was associated with a 0- to 50-fold decrease of core β1-4GlcNAc N-glycan linkages produced by GnT-4a activity. This outcome may reflect various mechanisms that can regulate glycosyltransferases, including altered intracellular localization, competition with endogenous acceptor substrates and variable access to donor substrates44
Multiple cell types express GnT-4a with high expression in the pancreas of normal rodents and humans. Nevertheless, beta cell–specific transgene expression markedly diminished signs of high-fat diet– induced diabetes in the presence of endogenous GnT-4a expression. In beta cells, multiple glycoprotein substrates of GnT-4a exist, including the insulin-like growth factor-1 receptor and insulin receptor-α subunit. However, the turnover of these substrates was unaffected by the deficiency of GnT-4a activity and core β1-4GlcNAc glycan linkages25
. The glucose transporters of beta cells are thus unique among glycoproteins analyzed in requiring GnT-4a glycosylation to sustain an extended half-life at the cell surface, suggesting a stabilizing role of lectin-ligand binding that may involve the galectins45,46
. This specificity of action may reflect a combinatorial role of protein and glycan sequences in determining biological function, analogous to enzymatic phosphorylation, which dictates different functional outcomes on different proteins44
The impact of enforced beta cell–specific expression of MGAT4A
on metabolic abnormalities including GSIS and insulin resistance was notable in this study. Constitutive beta cell expression of GnT-4a or Glut-2 preserved considerable systemic insulin sensitivity, indicating that beta cell function influences insulin action on these peripheral target tissues. Beta cells may accomplish this by one or more mechanisms. The retention of beta cell GSIS may achieve a more effective delivery of insulin, precisely timed to glucose excursions to meet tissue and body needs. Without such a ‘natural’ delivery system, chronic insulin exposure by some current treatment modalities may desensitize insulin-signaling pathways or provide improperly timed signals. For example, the optimization of insulin administration can better normalize blood glucose concentration and increase peripheral insulin sensitivity in subjects with type 2 diabetes, and the degree of glycemic control achieved by insulin treatment in subjects with type diabetes is a major determinant of the degree of insulin sensitivity retained in peripheral tissues in vivo47–49
. The marked reduction in hyperglycemia we observed from enforced expression of either GnT-4a or Glut-2 indicates how effective preservation of beta cell glucose transport can be in improving glycemia. In turn, normalizing glucose homeostasis could also promote peripheral insulin action by attenuating the effects of glucolipotoxicity, which impair insulin signaling. For instance, the severity of hyperglycemia is a factor in protein kinase C–mediated inhibition of insulin signaling50
Our findings do not address whether retention of GSIS is protective in the onset of diabetes; however, loss of GSIS is a marker of beta cell dysfunction in type 2 diabetes, and retention of GSIS is associated with disease protection. The beta cell could produce and secrete multiple glycoprotein factors that depend upon GnT-4a glycosylation for normal activity and half-life and that, along with insulin, directly contribute to glucose homeostasis and peripheral insulin action. Nevertheless, we predict that the severe decrease in glucose transport and intracellular glucose availability causes metabolic alterations in beta cells that may connect our findings to further downstream events that contribute to disease etiology. For example, the reduction of glucose transporter expression may reach a threshold at which glucose transport instead of glucokinase activity becomes rate limiting in the formation of glucose-6-phosphate needed for maintaining various beta cell functions.
Protection from disease conferred by GnT-4a and beta cell glucose transport was consistent with the assignment of beta cell GnT-4a glycosylation as a limiting factor. Both GnT-4a and the glucose transporters are highly regulated in pancreatic beta cells. Their expression is rapidly lost in primary beta cell cultures, and immortalized beta cell lines lose GSIS activity. Dietary stimuli leading to obesity trigger this process with the inactivation of transcription factors that normally sustain beta cell GnT-4a protein glycosylation and glucose transport, thereby promoting the onset of a disease pathway implicated in the diet- and obesity-associated component of type 2 diabetes mellitus. The molecules that impinge upon this pathway to sustain beta cell GnT-4a activity and glucose transporter expression may suggest new therapeutic targets to achieve effective prevention and treatment of diabetes.