Dorrance H. Hamilton Research Laboratories, Division of Endocrinology, Diabetes and Metabolic Diseases, Department of Medicine, Jefferson Medical College, Thomas Jefferson University, Philadelphia, Pennsylvania 19107
Address all correspondence and requests for reprints to: Barry J. Goldstein, M.D., Ph.D., Director, Division of Endocrinology, Diabetes and Metabolic Diseases, Department of Medicine, Jefferson Medical College, Jefferson Alumni Hall, Suite 349, 1020 Locust Street, Philadelphia, Pennsylvania 19107-6799. E-mail: . Barry.Goldstein{at}mail.tju.edu
Introduction
Resistance to the action of insulin in its metabolic target tissues occurs in almost all patients with type 2 diabetes and forms the pathophysiologic basis of the visceral obesity-linked metabolic syndrome. Improving tissue sensitivity to insulin is a major clinical goal to help ameliorate not only abnormal glucose metabolism, but also some of the cardiovascular risk factors that accompany this syndrome. Protein-tyrosine phosphatases (PTPases) that function as negative regulators of the insulin signaling cascade have been identified as novel targets for the therapeutic enhancement of insulin action in insulin-resistant disease states. Recent studies have provided compelling evidence that one of the main functions of the intracellular enzyme PTPase 1B (PTP1B), and perhaps to a lesser extent the transmembrane PTPase leukocyte antigen-related (LAR), is to suppress insulin action. Reducing PTP1B abundance not only enhances insulin sensitivity and improves glucose metabolism but also protects against obesity induced by high-fat feeding. Inhibition of PTP1B in insulin target tissues using pharmaceutical agents and novel antisense oligonucoleotides has shown enhanced insulin signaling and glucose tolerance in preclinical models. PTPase inhibitors may eventually find an important clinical role as novel insulin sensitizers in the management of type 2 diabetes and the metabolic syndrome.
Insulin action and resistance
The pleiotropic actions of insulin on metabolic responses in a variety of cell types are initiated by an ordered series of reversible tyrosine phosphorylation events involving the insulin receptor and its cellular substrate proteins (1). Full activation of the insulin receptor requires the tyrosine autophosphorylation of three closely spaced tyrosine residues in the receptor kinase domain (2), and dephosphorylation of any one of these three activating phosphotyrosine residues dramatically reduces the receptor kinase activity (3, 4). The immediate postreceptor targets, most notably the insulin receptor substrate (IRS) proteins, are also tyrosine-phosphorylated by the receptor on multiple sites that mediate specific protein-to-protein associations and activate several downstream signaling proteins (5). In general, these distal proteins mediate a switch in reaction mechanisms from protein-tyrosine phosphorylation during the initial stages of insulin action to the activation of diverse signal transduction mechanisms downstream, ultimately resulting in altered cellular glucose and lipid metabolism as well as changes in gene expression.
While several potential underlying mechanisms causing insulin resistance in obesity and type 2 diabetes are currently under intensive study, it remains a plausible approach to enhance overall insulin action by promoting the tyrosine phosphorylation events involved in the initiation of the insulin signal. Recent work has been directed at reducing the negative regulation of the early steps in insulin action by manipulating the activity of cellular PTPases as targets for therapeutic intervention (6, 7). Because purified insulin receptors and IRS proteins retain their tyrosine phosphorylation state in vitro, cellular PTPases are clearly responsible for the rapid dephosphorylation of the receptor ß-subunit and its substrate proteins observed in intact cells following dissociation of insulin from its receptor (4, 8, 9). These data indicate that PTPases exert a negative regulatory "tone" to the insulin signaling pathway under conditions of normal physiology, and predict that inhibition of PTPase(s) that act on the receptor or its tyrosine phosphorylated substrate proteins would be expected to enhance insulin action.
Identification of PTPases involved in the regulation of insulin signaling
PTPases comprise an extensive family of homologous enzymes that regulate various events in cellular signal transduction and metabolism (10). The enzymes in the PTPase superfamily have in common a conserved domain that contains a reduced cysteine moiety that is required to catalyze phosphotyrosine hydrolysis by the formation of a cysteinyl-phosphate intermediate (11). PTPases have been divided into two broad categories (Fig. 1): Intracellular (nonreceptor type), which have a single PTPase domain and additional functional protein segments (e.g. PTP1B); and transmembrane (receptor-type), which have a general structure like a membrane receptor with an extracellular domain, a single transmembrane segment and one or two tandemly conserved PTPase catalytic domains (e.g. LAR).
|
Studies implicating LAR in insulin action
In the plasma membrane, where the insulin receptor is rapidly deactivated, LAR physically interacts with the receptor and promotes its dephosphorylation (8, 13, 14). The tandem intracellular PTPase homology domains of LAR appear to have complementary functions that may potentiate their catalysis of insulin receptor dephosphorylation, with receptor binding (substrate recognition) mediated by the second domain, and phosphotyrosine hydrolysis catalyzed by the first domain (15). These interactions may help explain the efficient dephosphorylation and inactivation of the autophosphorylated insulin receptor kinase domain demonstrated in vitro with the recombinant cytoplasmic segment of LAR (3).
Modulation of LAR protein abundance in intact cells using antisense RNA expression as well as overexpression of LAR cDNA constructs showed that LAR negatively regulates insulin receptor activation and signal transduction (13, 16). In mice, 2.5-fold overexpression of LAR in skeletal muscle results in whole body insulin resistance and reduced glucose disposal (17). Although insulin receptor and IRS-1 tyrosine phosphorylation were normal in the LAR overexpressing mice, IRS-2 tyrosine phosphorylation and IRS-associated phosphatidylinositol 3' kinase activity were decreased, suggesting that LAR overexpression resulted in enhanced dephosphorylation of specific postreceptor regulatory phosphotyrosines on IRS proteins. Studies in mice lacking expression of LAR, however, demonstrated a more complex picture (18). These mice exhibited apparent insulin sensitivity in the fasting state, but marked resistance to insulin action in suppression of hepatic glucose production and glucose disposal in clamp studies (18). Postreceptor defects in insulin signaling, likely due to secondary alterations in cellular glucose metabolism, may have occurred in the adult mice having lacked LAR expression since birth. Clearly, if LAR had no role in cellular insulin signaling, alterations in insulin action or glucose metabolism would not have been expected to occur. However, LAR appears to have multiple cellular roles, including important effects on neuronal migration and mammary gland development, and some of the metabolic alterations observed in the knockout animals may be mediated by changes in the central nervous system (19).
Studies implicating PTP1B in insulin action
PTP1B is a widely expressed PTPase, including in insulin-sensitive tissues. It is localized to an intracellular site, associated with the endoplasmic reticulum via a noncatalytic C-terminal domain, and is also found in the cytosol (20, 21, 22). In vitro, recombinant PTP1B is active against the autophosphorylated insulin receptor (23), and compared with other candidate PTPases, it has enhanced specific activity toward the in vitro dephosphorylation of IRS-1 (4, 24). Thus, PTP1B may exert its negative regulation at multiple sites in the insulin signaling cascade.
PTP1B forms a physical complex with the activated insulin receptor (25, 26, 27). This enzyme also has unique structural features that promote its interaction with the receptor, in particular a second phosphotyrosine binding site in the PTP1B catalytic region that strongly enhances its association with the tandem phosphotyrosine residues of the activated insulin receptor kinase domain (28, 29) (Fig. 2).
|
Recent studies in knockout mouse models have provided evidence that strongly implicates PTP1B as a specific regulator of the insulin action pathway and also that PTP1B influences the regulation of body weight and energy expenditure. Apparently healthy in phenotype, the PTP1B knockout strain reported by Elchebly et al. (31) exhibited lower fasting insulin and glucose levels and other alterations consistent with enhanced insulin sensitivity, including an accentuated drop in blood glucose during insulin tolerance testing. Studies of insulin signaling revealed enhanced insulin-stimulated phosphorylation of the insulin receptor and IRS-1 in skeletal muscle and liver, accounting for the changed dynamics in glucose metabolism in the fed state. Importantly, the evident specificity for insulin action in this knockout model was quite provocative because studies in a variety of cellular systems in vitro have indicated that PTP1B can influence a variety of other signaling pathways elicited by other hormones and cytokines (12). Perhaps unexpectedly, the knockout mice were also found to be resistant to weight gain when fed a high fat diet and did not demonstrate the typical increase in insulin resistance that occurs with high fat feeding (31). Another important observation in this study was that insulin receptor phosphorylation was unchanged in adipose tissue of the knockout mice, indicating that tissue-specific differences in the regulation of insulin sensitivity by PTP1B occur in vivo.
Independently, Klaman et al. (32) developed lines of PTP1B knockout mice and not only confirmed the main findings of the Montréal group but also further demonstrated that the general leanness in the knockout mice was associated with an increase in basal metabolic rate and total energy expenditure, and slightly increased core body temperature. The increase in insulin sensitivity in the knockout mice was also confirmed to be tissue specific, with increased glucose uptake into skeletal muscle but not into fat tissue in the knockout mice. These observations raised the possibility that inhibition of PTP1B could effectively two essential features of the metabolic syndrome, glucose intolerance, and obesity, making it an exciting target for potential drug development.
Relevance to human insulin-resistant disease states
The studies reviewed above in cellular and animal models show that reduction of basal activity or levels of PTP1B and LAR can alter insulin signaling in a positive manner. A related question has been whether PTP1B or LAR levels are actually increased in states of insulin resistance and contribute to the pathogenesis of this disorder. In this regard, heterogeneous results have been found in animal studies, especially with the use of different PTPase substrates. In skeletal muscle of obese, nondiabetic, or overtly diabetic Zucker rats, we found that PTPase activity toward the autophosphorylated insulin receptor kinase domain in the particulate fraction was increased 2-fold along with increases in the mass of both LAR and PTP1B (33). However, other studies using different animal models, substrates and subcellular fractions have reported different results for changes in tissue PTPase activities in obesity and diabetes in animal models (see Ref. 4 for review).
In human subjects, studies in obese, nondiabetic subjects have generally exhibited increases in PTPase activities in adipose tissue and skeletal muscle (34, 35, 36). In homogenates of sc adipose tissue from obese subjects, PTPase activity toward the insulin receptor was increased and strongly correlated in a positive manner with body mass index, and was also associated with a 2.0-fold rise in the abundance of LAR (35). Immunodepletion of LAR (but not of PTP1B or SHP-2, another widely expressed intracellular PTPase) from the homogenates resulted in normalization of the PTPase activity toward the insulin receptor, suggesting that the increase in LAR was responsible for the enhanced PTPase activity in the adipose tissue from obese subjects. After loss of 10% of body weight by dietary restriction, enhanced insulin sensitivity was accompanied by a significant decrease in overall PTPase activity and reduced abundance of both LAR and PTP1B by immunoblot analysis (37).
Similarly, in skeletal muscle from obese, nondiabetic human subjects, we found that PTPase activity in both the particulate fraction and cytosol was significantly increased above the level in lean controls (36). Of the enzymes studied, LAR and PTP1B were increased the most in the particulate fraction; immunodepletion of LAR normalized the PTPase activity when compared with control subjects. These findings provided further support for the potential involvement of both PTP1B and LAR in the negative regulation of insulin action in skeletal muscle insulin resistance.
Paradoxical decrease of PTPase activity in type 2 diabetes
In contrast to the findings in nondiabetic, insulin-resistant subjects, tissue PTPase activities in patients with overt type 2 diabetes have frequently been found to be decreased. PTPase activity toward the insulin receptor in cytosol and particulate fractions of skeletal muscle from obese subjects with type 2 diabetes was decreased to 39% of the level in control subjects, along with similar decreases in the abundance of several PTPases, including PTP1B and LAR (36). Using different assay substrates and methods, other groups have also reported reductions in skeletal muscle PTPase activity in diabetic subjects (38, 39).
Confounding effects of PTPase cysteine reduction
In essentially all of the previously mentioned studies, the enzyme activities have been measured in tissue lysates in the presence of high concentrations of biochemical reductants, to recover and maintain the enzymatic activity of the PTPase catalytic domain, which is dependent on the reduced state of the critical thiol residue in the active site. The need for biochemical reduction stems from handling of cell lysates in air, where the catalytic thiol is highly susceptible to stepwise oxidative inactivation (40, 41) (Fig. 3). However, the use of strong reducing conditions in the assay following air exposure obviates the characterization of the endogenous activity of the enzymes as isolated from the cellular environment. Also, the reduction may not be complete if a fraction of the enzyme pool has been irreversibly oxidized. As several recent studies have pointed out, PTPase protein levels may be dissociated from their specific enzyme activities, possibly as a result of these types of alterations in catalytic activity in vitro (42, 43).
|
Role of cellular reactive oxygen species in PTPase regulation
Because of the sensitivity of PTPases to oxidative inhibition, the potential role of cellular reactive oxygen species in PTPase regulation is an area of growing interest (47, 48, 49). This effect is particularly relevant to diabetes mellitus because both hyperglycemia and insulin itself are coupled to the generation of hydrogen peroxide and other reactive oxygen species in insulin-sensitive cells (50, 51, 52). We have recently shown that the rapid increase in reactive oxygen species following stimulation of adipocytes and hepatocytes with insulin plays an integral role in modulating cellular PTPase activities, including PTP1B, and directly influences both early and late responses in insulin signal transduction (52, 53).
Efforts to develop enzyme inhibitors that target specific PTPases
The discovery of agents targeted for specific PTPases has been an area of active investigation (6, 11). A variety of strategies have been employed to enhance the substrate specificity of the inhibitors, including the use of peptidomimetic structures incorporating noncleavable structures chemically related to phosphotyrosine (e.g. sulfotyrosyl or phosphonomethyl phenylalanine derivatives) by modeling small molecules predicted to have relative in vitro specificity for certain enzymes, or by traditional high-throughput screening strategies (6, 12).
As might have been expected from data presented for the knockout mice, a few of the recently developed PTP1B enzyme inhibitors have recently been shown to lower glucose levels in rodent models of obesity and type 2 diabetes (54, 55, 56). The observed reduction in fasting insulin along with glucose levels indicates that the inhibitors serve as novel insulin sensitizing agents in vivo in these insulin-resistant animal models.
A novel approach to reducing cellular levels of PTP1B using a proprietary antisense DNA oligonucleotide (ASO) has been developed by Isis Pharmaceuticals, Inc. These investigators reported in a series of abstracts presented at the American Diabetes Association Annual Scientific Sessions in June, 2001, that administration of the PTP1B-ASO to obese, insulin-resistant rodents, substantially reduced the mass of PTP1B protein in liver and adipose tissue, but not in skeletal muscle, apparently due to its poor penetration into muscle tissue (57). Nevertheless, reduction of PTP1B protein levels by ASO treatment resulted in a variety of enhancements in the insulin action cascade, including increased insulin receptor and IRS tyrosine phosphorylation in liver and fat. In ASO-treated animals, blood glucose and insulin levels were reduced, as were glucose excursions to a meal challenge test and HbA1c levels in diabetic animals (58). Finally, in the Zucker fatty rat model, ASO treatment caused a reduction in fat volume, further substantiating the role that PTP1B has in regulating energy metabolism and weight gain (59). These preliminary studies have documented that the PTP1B-ASO can serve as an insulin sensitizer and may provide an innovative approach to the management of clinical type 2 diabetes.
Perspective
One of the most pressing health problems affecting the United States and the developed world is the epidemic of obesity, accompanied by growing numbers of people with type 2 diabetes and the comorbid cardiovascular risk factors that constitute the metabolic syndrome (60). Because this epidemic shows no signs of abating, the development of additional pharmaceutical agents with novel mechanisms of action to help ameliorate the insulin resistance underlying these disorders is important not only for controlling the diabetes, but also the associated cardiovascular disease. PTPases, and in particular PTP1B, have been clearly shown to serve as key negative regulators of the insulin signaling pathway, and early studies with PTPase and PTP1B inhibitors have provided a proof-of-concept that this is a plausible approach for enhancing insulin action and possibly also limiting weight gain. Modulation of the cellular pathways that regulate the activity of specific cellular PTPases, either by affecting oxidative inhibition, thiol conjugation, or by altering other posttranslational modifications, such as phosphorylation, may provide additional means for the therapeutic manipulation of PTPase activity to enhance insulin signaling (12). Because many cellular systems, in addition to the insulin action cascade, have been shown to be regulated by PTP1B, the safety and potential adverse effects of aggressive PTP1B inhibition in human subjects remains to be determined. Nevertheless, the available data show that modulation of specific PTPases holds great promise in ultimately providing a novel treatment approach to assist in the clinical management of the prevalent insulin resistance/metabolic syndrome.
Acknowledgments
I regret that space limitations do not allow for the citation of the many important primary references that have contributed to the large body of work summarized in this review. The reader will find detailed reference lists in the review articles and book chapters cited in the text.
Footnotes
Work in the authors laboratory has been supported by NIH Grants RO1-43396 and RO1-53388 and a mentor-based fellowship grant from the American Diabetes Association.
Abbreviations: IRS, Insulin receptor substrate; LAR, leukocyte antigen-related; PTPases, protein-tyrosine phosphatases; PTP1B, PTPase 1B.
Received March 26, 2002.
Accepted April 1, 2002.
References