Research Projects

Our research program is in the areas of signal transduction and membrane trafficking. It is aimed at elucidating, at the molecular level, the pathways of signaling from the insulin receptor and the trafficking of membrane proteins to the cell surface in response to insulin.

Signaling

During the past decade we have purified, cloned, and characterized three proteins that are rapidly tyrosine phosphorylated by the insulin receptor. These proteins are now known as insulin receptor substrates 1, 3, and 4 (IRS-1, 3, and 4). These three proteins, together with a fourth member of the family known as IRS-2, play a key role in insulin signaling. We and others have shown that the tyrosine phosphorylated IRS's act as docking/effector proteins for a number of SH2 domain-containing signaling proteins. Notable among these are the lipid kinase, phosphatidylinositol 3-kinase (PI3K), and the linker protein Grb2, which is also associated with the guanine nucleotide exchange protein for Ras. Association of PI3K with a tyrosine phosphorylated IRS activates the kinase, and the product of this kinase, phosphatidylinositol 3,4,5-trisphosphate, then causes the activation of several downstream serine kinases, including Akt (also known as protein kinase B). Similarly, association of Grb2 with an IRS activates guanine nucleotide exchange on Ras, and the GTP form of Ras causes activation of the mitogen-activated protein kinase cascade.

Subsequent to the discovery and cellular characterization of the IRS's, we and others investigated the physiological roles of each IRS by generating and characterizing mice lacking the IRS. We have generated mice null for IRS-3 and IRS-4, and others have done the same for IRS-1 and IRS-2. Measurement of several parameters of glucose homeostasis in these mice (concentrations of glucose and insulin in the blood, glucose and insulin tolerance tests, glucose transport into adipocytes) has revealed that IRS-1 and IRS-2 play key roles in insulin-regulated glucose homeostasis, whereas IRS-3 and IRS-4 do not. In order to determine the degree of functional redundancy among the IRS's, we and our collaborators have generated and characterized mice lacking pairs of the IRS's. This approach led to the discovery that mice lacking both IRS-1 and IRS-3 are severely deficient in fat cells (lipoatrophic) and exhibit diabetes. Treatment of these mice with the hormone leptin ameliorated the diabetes.

Our current research on insulin signaling has two directions. First, many of the targets for phosphorylation and thus regulation by insulin-activated Akt are unknown. We are attempting to identify some of these. Our present approach has been to isolate and identify substrate proteins for Akt by immunoprecipitation with phosphomotif-specific antibodies followed by microsequencing. This approach has led to the discovery of four novel substrates for Akt. One of these substrates is a Rab GTPase activating protein which we have recently shown participates in insulin regulation of glucose transport (see below). We are now characterizing each of the other three substrate proteins. Our goal is to decipher the function of each and to determine how phosphorylation affects the function.

Second, insulin treatment of cells activates a range of kinases and phosphatases, resulting in changes in phosphorylation at many sites on many proteins. Our current knowledge is limited to only a fraction of these. Consequently, in collaboration with Forest White, an expert in phosphoproteomics at MIT, we have initiated a more comprehensive approach, along the following lines. Total protein is isolated from untreated and insulin-treated cells; tryptic peptides are generated; classes of phosphopeptides are isolated with phosphomotif-specific antibodies; and then the sequences of these peptides, the sites of phosphorylation, and the change in extent of phosphorylation in response to insulin are determined by mass spectrometry. We have recently completed a study of this type with an antibody against phosphotyrosine. In it we identified 122 sites of tyrosine phosphorylation in

adipocytes, of which 86 increased in phosphorylation in response to insulin. Only 20 of these sites have previously been described. Hence, this study revealed a large number of novel insulin-elicited phosphorylations and has opened the way to determining the effect of such phosphorylations.

Membrane Trafficking

Insulin lowers blood glucose level in part by stimulating the transport of glucose into fat and muscle cells. The basis for this stimulation is an increase in the number of glucose transporters in the plasma membrane. The latter is brought about by generation of intracellular vesicles containing glucose transporters from the endosomes and/or trans-Golgi network, the movement of these vesicles to the plasma membrane, and the fusion of them with the membrane. Insulin stimulates one or more steps in this process. The glucose transporter isotype that is present in fat and muscle cells is known as GLUT4. Hence the intracellular vesicles are referred to as GLUT4 vesicles, and the overall process is known as GLUT4 translocation.

We are interested in the following questions: What is the protein machinery involved in GLUT4 vesicle formation, in GLUT4 vesicle movement, and in GLUT4 vesicle fusion with the plasma membrane? What steps does insulin stimulate? What is (are) the signal transduction pathway(s) from the insulin receptor that bring about the stimulation?

In order to answer some of these questions, we developed a method to purify the GLUT4 vesicles, and then examined the major proteins in these vesicles. This approach led to the identification of the fusion proteins VAMP-2 and cellubrevin (v-SNAREs) as components of the vesicles, and subsequent studies in other laboratories have shown that they are part of the machinery for fusion of the vesicles with the plasma membrane. In addition, we identified and cloned a novel membrane aminopeptidase present in the vesicles. This protein, now known as the insulin-regulated aminopeptidase, exhibits trafficking behavior identical to that of GLUT4. Thus, insulin increases the aminopeptidase activity, as well as the glucose transport activity, at the surface of the cell.

There is evidence that Akt is on the signaling pathway from the insulin receptor to GLUT4 translocation. Recently we discovered a key link between this signal transduction pathway and the trafficking machinery in GLUT4 translocation. We have found that Akt phosphorylates a novel GTPase-activating protein (GAP) for a Rab, and have shown that this phosphorylation is required for insulin-stimulated GLUT4 translocation. Rabs are small G proteins that in their active GTP form participate in vesicle formation, movement, and fusion. A Rab GAP catalyzes the hydrolysis of bound GTP to generate the inactive Rab in its GDP form. Thus, our present hypothesis is that insulin-stimulated phosphorylation of this novel GAP inactivates it. As a result, the active Rab GTP complex is generated, and GLUT4 translocation proceeds. One ongoing project is identification of the participating Rab(s). We have found several Rabs that are substrates for the GAP, from among the 60 Rabs in animal cells, and our recent evidence indicates that one of these, Rab10, is required for GLUT4 translocation. A second project is to elucidate the step(s) at which the Rab(s) function.

Relevance to Disease

The basic science upon which we are focused is very relevant to diabetes. Approximately 15,000,000 Americans suffer from adult-onset diabetes, which is also referred to as type 2 diabetes or non-insulin dependent diabetes. A major feature of this disease is that insulin is less effective in its action upon its target tissues of muscle, liver, and fat. This feature is referred to as insulin resistance. The basis of insulin resistance is not understood. Fundamental research of type in which we are engaging provides a framework within which to identify the changes in type 2 diabetes that cause insulin resistance, as well as to develop therapeutics with which to treat it.