Defects in Insulin Secretion and Action in the Pathogenesis of Type 2 Diabetes Mellitus
- First Online:
- Cite this article as:
- Tripathy, D. & Chavez, A.O. Curr Diab Rep (2010) 10: 184. doi:10.1007/s11892-010-0115-5
- 418 Views
Type 2 diabetes mellitus (T2DM) is characterized by defects in insulin action and insulin secretion. Although insulin resistance manifests early during the prediabetic state, a failing β-cell function unable to overcome insulin resistance at target tissues determines the onset of T2DM. This review focuses on recent advances in the molecular mechanisms of insulin resistance and β-cell dysfunction. The role of mitochondrial dysfunction, impaired regulation of the enteroinsular axis, and endoplasmic reticulum stress are currently the subjects of intensive research. In addition, the adipose tissue has emerged as a major endocrine organ that secretes a growing list of adipocytokines with diverse central and peripheral metabolic effects. The role of a growing number of candidate genes and transcription factors regulating insulin action and secretion is also discussed.
KeywordsType 2 diabetes mellitusInsulin resistanceInsulin secretory defectsMitochondrial dysfunctionFree fatty acidsAdipokinesIncretins
The pathogenesis of type 2 diabetes mellitus (T2DM) has been studied extensively and it is well established that although genetic factors predispose, it is the environmental factors (ie, overnutrition and a sedentary lifestyle) that pull the trigger to manifest the disease. Type 2 diabetes is characterized by two fundamental defects: impaired insulin action in skeletal muscle, liver, and adipocytes, and impaired β-cell function [1••]. The relative contribution of insulin resistance versus pancreatic β-cell dysfunction to the pathogenesis of T2DM is still a cause of debate. The present review focuses on the recent advances in the understanding of T2DM pathogenesis, including the underlying molecular and metabolic basis of insulin resistance, and the mechanisms responsible for impaired β-cell function.
Normal Glucose Metabolism
In healthy individuals, plasma glucose levels are maintained within a very narrow range. Any given concentration of glucose in plasma is the result of simultaneous release of glucose into circulation and uptake of glucose from plasma by the cells. The liver is the primary source of endogenous glucose production (EGP), although kidneys contribute approximately 5% to 10% of EGP in the basal state. Following glucose or meal ingestion, a rise in plasma glucose leads to insulin release from β cells and the hyperinsulinemia stimulates glucose uptake by splanchnic (liver and gut) and peripheral tissue (primarily muscle). From a physiologic perspective, it is clear that any defects in insulin secretion in response to a meal or defects in insulin action in peripheral tissue will lead to increased plasma glucose. Although impaired insulin action has been reported in other tissues such as the brain and myocardium, we limit our discussion to insulin action in the skeletal muscle, liver, and the adipose tissue.
Defects in Insulin Action
The notion that people with T2DM have impaired insulin action due to insulin resistance has been known for several decades. Using euglycemic insulin clamp combined with tracer studies, it is possible to quantify skeletal muscle and hepatic insulin sensitivity .
Sites of Insulin Resistance
Skeletal muscle is the primary site of insulin action in the postprandial state. During euglycemic hyperinsulinemia (80–100 μU/mL), hepatic glucose production is suppressed, and about 75% to 80% of glucose uptake occurs in skeletal muscle . Several studies have unequivocally demonstrated reduced insulin-mediated glucose uptake in type 2 diabetes. Studies using femoral artery/vein and forearm catheterization techniques and positron emission tomography shown reduced insulin-stimulated muscle glucose uptake . The skeletal muscle of lean T2DM and obese normal glucose tolerant (NGT) individuals is resistant to insulin [5-7]. Collectively, there is enough evidence that skeletal muscle is the site of primary defect in insulin action. This diminished muscle glucose uptake, in concert with impaired suppression of hepatic glucose production by insulin, accounts for the excessive rise in plasma glucose after glucose ingestion.
By the time subjects develop hyperglycemia, it is difficult to determine whether metabolic/hormonal abnormalities are the cause or consequence of the hyperglycemia. Because the majority of T2DM subjects are also obese, they also have elevated plasma free fatty acid (FFA) concentration and increased cytokines and it is difficult to ascertain whether these are genetic or acquired defects. One way to circumvent this problem is to study NGT subjects who are at high risk for developing diabetes or to follow NGT subjects over a long time while they develop impaired glucose tolerance (IGT) and subsequently T2DM. Several investigators have demonstrated that NGT offspring and first-degree relatives of patients with T2DM exhibit moderate to severe insulin resistance [8, 9]. The impaired glucose uptake is primarily due to reduced nonoxidative glucose metabolism, which reflects impaired glycogen synthase activity in skeletal muscle. As NGT subjects progress to IGT their insulin sensitivity declines markedly, but their β cells are able to compensate for insulin resistance by increasing insulin secretion, which offsets the insulin resistance in liver, and skeletal muscle and blood glucose values are still maintained within normal range or mildly elevated [10-12].
The molecular mechanisms leading to insulin resistance in skeletal muscle continue to be unraveled with the use of newer techniques. Defects in insulin signaling, glucose transport/phosphorylation, glycogen synthesis, and glucose oxidation all contribute to the muscle insulin resistance. Earlier studies showed impaired insulin signaling, particularly impaired insulin receptor substrate-1 (IRS-1) tyrosine phosphorylation and phosphatidylinositol 3 (PI3)-kinase activity in the skeletal muscle of subjects with T2DM and NGT offspring of T2DM. Recent studies have implicated the role of inflammatory pathways in skeletal muscle insulin resistance. Sriwijitkamol et al.  examined skeletal muscle from T2DM subjects and demonstrated a 60% decrease in IkB-β protein content, an indicator of increased activation of the IkB-β/nuclear factor-κB (NF-κB) pathway. The IkB-β content in skeletal muscle correlated with insulin-mediated glucose disposal during a hyperinsulinemic-euglycemic clamp, suggesting that increased IkB-β/NF-kβ pathway activity is associated with muscle insulin resistance . Recently, the same group also showed that insulin-resistant obese and T2DM subjects had elevated toll-like receptor-4 (TLR4) gene expression and protein content in the skeletal muscle .
Mass spectrometry-based proteomic analysis of skeletal muscle has revealed differential expression of novel proteins in skeletal muscle of subjects with T2DM. Whereas carbonic anhydrase, 3-hydroxyisobutyrate dehydrogenase, and enolase were decreased, monoglyceride lipase, adenylate kinase, and Cu/Zn superoxide dismutase were noted to be increased . The role of these proteins in skeletal muscle insulin resistance is yet to be identified.
In the postabsorptive state, liver is the main source of EGP. Several studies have consistently shown that patients with T2DM have higher basal EGP compared with healthy individuals. A major action of insulin is to suppress EGP, which occurs mainly (80% to 90%) in the liver. Most patients with T2DM have higher fasting insulin levels and yet modestly elevated EGP rates, implying that there is considerable resistance to the suppressive effect of insulin on liver. Tripathy et al.  examined hepatic glucose production rates in 200 men of similar age, but with varying degrees of glucose tolerance; and showed that although in absolute terms the EGP was not high until late in the stages of glucose intolerance, hepatic insulin resistance was already evident even in subjects with IGT.
Because liver tissue is not readily accessible in humans, the exact molecular mechanisms leading to hepatic insulin resistance remain a speculation. Several animal studies imply increased activity of two key enzymes involved in gluconeogenesis: phosphoenolpyruvate carboxykinase (PEPCK) and glucose-6-phosphatase. It has been proposed that although insulin loses its ability to inhibit PEPCK, it retains its ability to stimulate fatty acid synthesis. A recent study in rodents showed that insulin has potent inhibitory effect on the PEPCK and stimulatory effect on SREBP-1, a transcription factor that activates lipogenesis . They showed a divergent effect of mTORC inhibitors on insulin signaling distal to the PI3-kinase and AKT and suggest that mTORC1 could explain the paradoxical effect of insulin on lipogenesis and gluconeogenesis in the liver.
Increased fat deposition in the liver or nonalcoholic fatty liver disease is a broad spectrum of abnormalities that include accumulation of simple triglyceride in the hepatocytes (fatty liver) to hepatosteatosis characterized by inflammation (nonalcoholic steatohepatitis [NASH]). The prevalence of NASH is remarkably high, affecting approximately 20% of the overweight and obese population. Although it is not clear whether increased fat accumulation in the liver is a primary event, or secondary to insulin resistance, it has been associated with several metabolic abnormalities, including impaired insulin-mediated inhibition of hepatic glucose production, reduced inhibition of lipolysis by insulin, and skeletal muscle insulin resistance. Chronic infiltration of hepatocytes by triglycerides and excessive fatty acids has been proposed to lead to accumulation of toxic metabolites diacylglycerol, fatty acyl-coenzyme As (CoAs), and induce inflammatory pathways, particularly NF-ĸB, and result in hepatic insulin resistance.
For many years adipose tissue has been viewed as a site where excessive fat is stored and FFAs were the only secretory products. However, in addition to FFAs, adipocytes synthesize and secrete a host of proteins that collectively are designated as adipocytokines. These adipokines have local autocrine/paracrine effects, as well as systemic effects. Tumor necrosis factor-α (TNF-α) was the first proinflammatory cytokine shown to be constitutively expressed in the adipocytes of obese insulin-resistant animals, and neutralization of TNF-α with anti-TNF-α antibody decreased insulin resistance . Subsequently, a number of other adipocytokines have been identified, including interleukin-6, plasminogen activator inhibitor-1, resistin, interleukin-1, leptin, adiponectin, resistin, and visfatin [19, 20]. All of these molecules exhibit positive (adiponectin) or negative (TNF-α, resistin, interleukin-6) effects on insulin sensitivity in a paracrine or endocrine manner. Activation of inflammatory pathways by adipocytokines has been shown to inhibit insulin signal transduction by causing serine/threonine phosphorylation of IRS-1 in multiple insulin-sensitive tissues, including muscle, liver, and adipocytes. Thus, expression and secretion into plasma of these and other cytokines could provide a link between insulin resistance and chronic inflammation in obesity .
In the postprandial state, adipose tissue accounts for less than 4% to 5% of whole-body glucose uptake, whereas skeletal muscle accounts for about 80% to 90%. From the quantitative standpoint, it is clear that resistance to effect of insulin in adipose tissue with respect to glucose metabolism cannot explain the systemic (whole body, which primarily reflects muscle) insulin resistance observed in obesity and type 2 diabetes. This raises an important question: how does inflammation in the adipose tissue lead to systemic (muscle and liver) insulin resistance in humans? There are several possible explanations: 1) decreased insulin responsiveness in adipose tissue leads to increased lipolysis and elevated plasma FFA concentrations, which are known to cause insulin resistance in muscle and liver; 2) adipokines released by adipocytes and resident macrophages are released into the circulation and act at distal sites (ie, skeletal muscle and liver), and activate inflammatory (c-Jun N-terminal kinase [JNK], NF-ĸB) pathways, to impair insulin signaling.
Although the glucose uptake and utilization by the central nervous system is not dependent on insulin, several recent studies in rodent models have suggested that impaired glucose metabolism and obesity might result from dysregulation of hypothalamic insulin signaling . There is evidence to suggest that insulin can regulate expression of neuropeptides in the hypothalamus involved in food intake and can regulate glucose metabolism via central nervous system connections that regulate hepatic glucose production. High-energy or high-fat diet has been shown to impair hypothalamic insulin signaling in rodents. Using the magnetoencephalographic approach during a two-step hyperinsulinemic-euglycemic clamp, an impaired cortical neuronal response to insulin was observed in obese individuals. These studies suggest that cerebral insulin resistance may also occur in humans . Although a relatively new concept, further studies are still needed in humans to understand the consequences of cerebral insulin resistance and its contribution to the pathogenesis of T2DM.
Mitochondrial Dysfunction and Insulin Resistance
Recently, mitochondrial dysfunction has been reported in the skeletal muscle in patients with T2DM as well as NGT relatives of patients with T2DM. Mitochondria in diabetic subjects has been shown to exhibit structural as well as functional abnormalities. Reduced insulin-stimulated adenosine triphosphate (ATP) synthesis and reduced subsarcolemmal mitochondria content have been reported in subjects with T2DM. Elevated FFA concentrations seen in T2DM and obesity have been linked to mitochondrial dysfunction. Prolonged (48-hour) physiologic elevation of FFA downregulated genes involved in mitochondrial biogenesis . In a recent study, we demonstrated that short-term (6-hour) physiologic (600–800 µM) increase in plasma FFA concentration in lean healthy individuals reduced skeletal muscle inner-mitochondrial membrane potential, but did not have any effect on genes regulating mitochondrial function or any structural abnormalities in mitochondria . This may be an important mechanism by which elevated FFAs induce skeletal muscle insulin resistance in subjects with type 2 diabetes. Hwang et al.  used mass spectrometry–based quantification of proteins and demonstrated decreased abundance in mitochondrial proteins and reduced expression of proteins involved with cytoskeleton structure, confirming the reduction of mitochondrial proteins in insulin-resistant states.
Endoplasmic Reticulum Stress and Insulin Resistance
Endoplasmic reticulum (ER) also plays a crucial role in the cellular response of insulin. Recent reports by Ozcan et al.  describe the link between obesity, ER stress, and insulin action in diabetes. In response to ER stress, cytokines, and fatty acids, JNK is activated whereupon it phosphorylates IRS-1 on Ser307, resulting in impaired insulin-mediated tyrosine phosphorylation. Treatment of ob/ob mice with a chemical chaperone, 4-phenylbutyric acid, and endogenous bile acid ursodeoxycholic acid and its taurine-conjugated derivative (TUDCA) resulted in normoglycemia within 4 days of treatment . Boden et al.  reported increased ER stress (phospho JNK-1 expression) in adipose tissue in obese individuals. Although ER stress appears to be a novel mechanism, a paucity of data still exist on the role of ER stress in skeletal muscle insulin resistance in humans.
FFA and Insulin Resistance
Elevated plasma FFA concentration has been proposed to be the link between insulin resistance and obesity and type 2 diabetes. Patients with type 2 diabetes and obese nondiabetic subjects have elevated day-long plasma FFA concentrations. Experimental elevation of the plasma FFA concentration in healthy subjects induces insulin resistance in a dose-dependent manner [30, 31]. Conversely, lowering of plasma FFA levels with acipimox or thiazolidinediones leads to improved insulin sensitivity. It has been proposed that increased FFA leads to the accumulation of triglycerides and toxic FFA metabolites (FACoA, DAG, and ceramide), which impair insulin signaling and cause insulin resistance (ie, “lipotoxicity”). In addition, experimental elevation of FFA in healthy individuals has been shown to increase reactive oxygen species generation and activate NF-ĸB in mononuclear cells . Human myotubes when cultured in the presence of palmitate are shown to activate the TLR4 gene and protein expression . These are some of the novel mechanisms by which elevated FFA could cause insulin resistance in skeletal muscle.
Defects in β-Cell Function in the Pathogenesis of T2DM
Although insulin resistance is considered the initiating event in the pathogenesis of T2DM, pancreatic β-cell dysfunction is a sine qua non for the development of the disease and hyperglycemia does not become manifest until there is severe β-cell dysfunction. During the initial stages in the natural history of T2DM when minimal variations in fasting plasma glucose (FPG) can be documented in the face of insulin resistance, hyperinsulinemia is already present as a compensatory mechanism. As the insulin resistance worsens with progression from NGT to an impaired fasting glucose (IFG)/IGT state, there is further decline in β-cell function and this eventually leads to overt diabetes. The decline in β-cell function is particularly apparent when insulin secretion is expressed in relation to ambient insulin sensitivity . Compared with healthy NGT subjects, subjects with IFG/IGT already have a 50% to 75% decline in β-cell function [33, 34]. Over the ensuing years following the diagnosis of diabetes, there is a further progressive decline in β-cell function [35-38]. Several acquired and inherited factors have been identified to affect β-cell function.
Lipotoxicity and Glucotoxicity in the Pathogenesis of β-Cell Failure and T2DM
Chronic elevation of FFA and glucose has also been implicated as an acquired cause of β-cell dysfunction. Short-term exposure of β cells to physiologic increases in FFA has been shown to stimulate insulin secretion; conversely, chronic β-cell exposure to elevated fatty acyl-CoA inhibits insulin secretion. Similarly, elevation of glucose for 48 h impairs insulin secretion, and correction of hyperglycemia with insulin or phlorizin restores β-cell function. The molecular mechanisms of lipotoxicity and glucotoxicity have been reviewed recently .
Endoplasmic Reticulum Stress
The efficient functioning of the ER is essential for proper cellular activities and survival. Components of the unfolded protein response play a dual role in β cells, acting as beneficial regulators under physiologic conditions or as triggers of β-cell dysfunction and apoptosis in response to chronic stress. Accumulating evidence suggests that ER stress plays a role in the pathogenesis of diabetes by contributing to pancreatic β-cell loss [40, 41]. Recently, it has been suggested that the expression of the transcription factor C/EBPβ in pancreatic β cells contributes to β-cell failure by enhancing susceptibility to ER stress .
Islet Amyloid Polypeptide and Dysfunctional Islet Remodeling
Hyperglycemia and islet amyloidosis have been implicated in the progressive loss of β-cells in T2DM. In normal conditions, a tight balance exists between the rate of apoptosis and proliferation of β cells, a process that is dysregulated in obesity and T2DM. In obese humans and primates, islet amyloid polypeptide (IAPP) is deposited to cause amyloidosis of the islet [43, 44]. The amyloid deposits are toxic for islet cells, inducing an accelerated rate of apoptosis. While the β cells are being replaced by amyloid deposit, there is a parallel decrease in glucose tolerance [45••]. Because there is controversy as to whether the main pathogenic factor in T2DM is β-cell dysfunction versus absolute loss of β-cell mass, recent studies in primates have been carried out to address this question. Interestingly, as the FPG increases and the percentage of islets occupied by amyloid increases, an increase in β-cell apoptosis and α-cell proliferation is observed, explaining in part the hyperglucagonemia observed in T2DM. These findings suggest that in addition to IAPP, differential regulation of islet α-cell and β-cell play an important role in deficient insulin secretion and insulin action [45••].
Recently, the incretin hormones glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP) have been shown to play a major role in regulating insulin secretion as well as in the control of postprandial blood glucose levels [46, 47]. Decreased levels and/or sensitivity of the β cells to incretins have been suggested to play a major role in the pathogenesis of T2DM. GLP-1 and GIP account for approximately 90% of the incretin effect, and this is severely impaired in T2DM. A reduced incretin effect could be secondary to impaired incretin (GLP-1 and GIP) hormone secretion and/or action. In T2DM, not only are the GLP-1 levels decreased, but the response to exogenous GLP-1 also is impaired. In addition to potentiating insulin secretion, GLP-1 also inhibits glucagon secretion, suppresses the appetite leading to a decrease in body weight, and slows gastric emptying .
The majority of studies have reported a modest decrease in plasma GLP-1 levels and no change or slight increase in plasma GIP levels in T2DM. An increased GIP level with decreased insulin secretion in subjects with T2DM implies β-cell resistance to the action of GIP. Studies that have assessed the insulinotropic effect of GIP in T2DM subjects consistently have demonstrated a marked decrease in GIP action [49, 50]. Unlike GIP, the insulinotropic effect of GLP-1 appears to be retained in subjects with T2DM. Intravenous administration of GLP-1 and GLP-1 agonists enhances insulin secretion in T2DM subjects in a dose-dependent manner, although there is decreased β-cell sensitivity to these incretin hormones [51, 52].
Genetic Determinants of Type 2 Diabetes
High concordance rate of T2DM (70% to 90%) in monozygotic twins
Increased concordance rates of T2DM in monozygotic versus dizygotic twins
High prevalence of T2DM (30% to 40%) in first-degree relatives of diabetic patients
Prior to the complete sequencing of the human genome, the identification of T2DM genes relied heavily upon linkage and candidate gene approaches. One of the genes identified using this approach is the Pro12Ala substitution in the peroxisome proliferator-activated receptor 2 (PPARG) gene [53, 54]. Whereas individuals with the rare Ala allele were associated with increased insulin sensitivity and protection from T2DM, individuals homozygous for the Pro12 are about 20% more likely to develop T2DM. Other genes consistently associated with T2DM in different populations were KCJN11 and TCF7L2. The KCJN1 gene encodes the islet ATP-sensitive inward rectifier potassium channel 11 (Kir6.2, KCNJ11) and affects insulin secretion.
Of note, PPARG is the only gene linked to increased diabetes risk by affecting insulin sensitivity. Even though insulin resistance is almost universal in subjects with T2DM, genes associated with increased risk of T2DM are mostly linked to insulin secretion. One possible explanation is that insulin secretion is more heritable than insulin sensitivity, or that insulin resistance is more likely to be affected by environmental factors than insulin secretion.
In recent years, with the advent of high-throughput technology and genome-wide association scans (GWAS), several other candidate genes have been identified. In the first GWAS for T2DM, Sladek et al.  confirmed the association of TCF7L2 with diabetes. The association of TCF7L2 with T2DM appears to be the most reproducible and has the largest effect on disease susceptibility. Individuals with the high-risk T allele are associated with impaired insulin secretion, and its effects on insulin secretion are mediated through its effect on the enteroinsular axis [56••].
Another recent study demonstrated the association of a common variant in the melatonin receptor 1B (MTNR1B) gene and impaired insulin secretion and increased risk for T2DM . In two recent papers by the MAGIC trial (Meta-Analysis of Glucose and Insulin-Related Traits), a genetic variant in the GIP receptor was noted to influence postprandial glucose and insulin secretion . In conclusion, although about 20 possible candidate genes have been identified, collectively they explain only a small fraction of overall risk for developing T2DM. However, better characterization of functional aspects of these genes could play a potential role in developing newer therapy for T2DM.
Cellular and molecular defects in insulin action and insulin secretion in the pathogenesis of T2DM
• Impaired insulin signaling: ↓ IRS-1 phosphorylation and PI3-kinase activity
• Intramyocellular lipid and ceramide accumulation with lipotoxicity
• Mitochondrial dysfunction: decreased mitochondria number and oxidative capacity
• Polymorphisms in PPAR-γ gene
• Inflammatory pathways: ↑ NF-ĸB/↓IkB-β and ↑ TLR4 expression
Impaired insulin signaling:
• ↓ PEPCK
• ↑ SREBP-1
• ↑ FFA: lipotoxicity
• ↓ Adiponectin: ↓ sensitivity
• ↑ Leptin
• Other adipocytokines: resistin, RBP4, TNF-α, interleukin-6
• Endoplasmic reticulum stress
Decreased CNS response to insulin:
• Impaired insulin/leptin signaling in hypothalamus
• Dysregulation of food intake/satiety and body weight: ↑ PKC-θ activity
• Neuropeptides controlling EGP
↓ β-cell function
• Lipotoxicity: chronic FFA exposure
• Genes associated with impaired insulin secretion: TCF7L2, KCJN1, MTNR1B
• Dysregulation of β-cell apoptosis/α-cell proliferation
• Endoplasmic reticulum stress
• Decreased incretin effect
↓ β-cell mass
• IAPP deposition
• Dysregulation of β-cell apoptosis
• Genetic factors?
• Incretin biology
• ↓ GLP-1 secretion
• ↑ GIP resistance
Dr. Devjit Tripathy has received research support from Takeda Pharmaceuticals. No other potential conflicts of interest relevant to this article were reported.