Image Map

The Language of Cells – Part III
by: Wen Zhang

The first article in this series introduced two general classes of signaling molecules- hormones that bind to receptors on the surface membrane of cells (i.e. insulin, a “web” type signaling molecule), versus those that bind to intracellular or intranuclear receptors (i.e. steroid hormones, a “direct” signaling molecule). In that installment, we also mentioned we would be focusing on insulin’s signal transduction pathway. In the second issue, we finished reviewing general signal transduction concepts and began to delve more deeply into the specific details of cellular signaling. This month, we will step just past the receptor and look at what happens once a hormone binds the insulin receptor undergoes autophosphorylation. The role of docking proteins in signal amplification, diversity, and specificity will be discussed, with particular emphasis on the mechanisms surrounding the latter effect.

Before we get into the meat of the article, a few important points need to be discussed/reviewed: the insulin receptor is a member of the receptor tyrosine kinase family (RTK). These receptors and their respective ligands initiate signaling by autophosphorylation of the receptor upon binding, resulting in recruitment of downstream signaling components. There are two primary classes of these components- those that are active, and those that are passive. Active and passive refers to whether the component directly catalyzes a reaction (i.e. phosphorylation or dephosphorylation of another molecule), or serves as a docking site/scaffold to recruit and/or localize other signaling components, respectively. Therefore, when one hears the terms “adaptor, scaffold” or “docker,” one knows that the protein being referenced is passive, while the term “kinase,” or “phosphatase” means the protein is active. Active and passive signaling intermediates are equally important in signal transduction.

Fig 1. Blue trapezoid = receptor; yellow/grey rectangle = scaffold/adator/docking protein; red circles = inative enzymes; green circles = active enzymes. In this figure, an activated receptor recruits a passive docking protein (i.e. IRS-1), which in turn recruits passive enzymes and by various mechanisms (i.e. bringing all subunits together, localizing other proteins that activate the enzyme, etc), makes them active.

First, there are a few concepts that bear emphasis, not only for this article, but for all of signaling. Pay particularly close attention to the last three.

1. In the cell, there are many compartments of molecules with specific functions. Hence, we have the nucleus and its associated proteins, lysosomes, peroxisomes, the endoplasmic reticulum, the Golgi apparatus, mitochondria, etc. Each of these organelles has its own membrane, which serves both as physical barrier and as a potential depot for certain organic molecules which may be involved in signaling (i.e. phospholipids).

2. The cytoplasm, which is the medium in which all of the organelles and other cellular machinery resides, is reinforced by a protein-based cytoskeleton.

3. This cytoskeleton, composed mainly of actin, not only serves a structural role, but also helps to further compartmentalize cellular components and can even serve as a railing system for transport.

4. The cytoskeleton is dynamic and under constant remodeling in response to a variety of signals- chemical and mechanical.

5. The entire cell is encased in a bilayer lipid-and-protein based membrane (AKA the plasma membrane, plasmalemna, cell membrane).

6. The lipids that make up the plasma membrane are varied and dynamic, responding to the same types of input that the cytoskeleton does, albeit manifesting these changes in a different manner.

7. All lipid-containing membranes in the cell, exterior or interior, follow this rule to some extent.

8. Specially modified lipids, called phospholipids (lipids with a phosphate group and another associated moiety, i.e. an inositol molecule) are one class of phospholipids found in cellular membranes.

9. These modified lipids can undergo further modification, can be released from the membrane, and can participate in signaling events.

10. Receptors may be surrounded by specific types and quantities of lipids (lipid raft hypothesis), which effectively localizes necessary signaling components, rather conveniently, in one area. [Diet is one of the principal regulators of membrane lipid composition]

11. Various other signaling molecules (active and passive) are found in various cellular locations, and may need to be recruited to other locales- modified lipids in membranes are perfect candidates for this process, and are even more so if they are localized and coupled (directly or indirectly) to a receptor (see above).

12. Cytoskeletal remodeling and phospholipid modification work in conjunction to localize signaling events; this localization is proposed to be one mechanisms to induce specificity of signaling; in other words, subcellular localization is important.

13. Cell type matters when comparing the molecular effects of the same hormone on two or more different tissue/cell types.

14. Almost every signaling molecule encountered from here on out has several isoforms, even if they are coded by a single gene; isoforms of even very closely related molecules can have distinctly unique functions. Therefore, never assume that the functions of PKB-alpha and PKB-beta are interchangeable, even in the same cell type.

15. Close sequence homology (how many amino acids in the protein’s sequence are the same when comparing two or more proteins) does not guarantee functional homology- how the protein folds (structure) can be just as important, or even more so, than amino acid sequence. Thus, mutation of a single amino acid may have no effect at all, or may completely alter the function of a protein depending on what that mutation does to the structure and/or charge. This also means that protein domains with 90+ percent amino acid homology still retain unique substrate specificities.

Note: The following section discusses signaling proteins with a heavy emphasis on modular domains and how they are specifically recruited. For reference purposes, here is a brief overview of these domains:

Fig. 2. Nomenclature of a protein. Each colored box represents a different domain of the protein. In this particular example, the entire protein is 800 amino acids (residues) long.

A. PH domain: The PH, or Pleckstrin homology domain, is named after the protein Pleckstrin. It is a small domain composed of 100-200 amino acids, and binds preferentially to phosphorylated phospholipids, and in particular, PIP2 and PIP3. It binds other molecules and protein domains as well, but for our purposes, think of the PH domain as a means to localize/recruit proteins to membrane surfaces (because phospholipids are predominantly membrane structures).

B. SH2 domain: The SH2, or Src-homology-2 domain, is named after the Src protein. SH2 domains are ~100 amino acids long, and bind to/are attracted to phosphorylated tyrosine residues. SH2 domain-containing proteins are numerous, and phosphotyrosine residues are ubiquitous, so specificity is conferred by amino acids surrounding the phosphorylated tyrosine. The determining amino acids plus the phosphotyrosine residue make up an amino acid motif. Think of SH2 domains as a means to connect tyrosine phosphorylation events to downstream (SH2 domain-containing) effectors.

C. PTB domain: The PTB, or phosphotyrosine-binding domain, is 100-200 amino acids long, and like SH2 domains, also binds to phosphorylated tyrosine residues, and likewise, derives specificity from adjacent amino acids. However, there is evidence that PTBs actually prefer non-phosphorylated tyrosines, providing an important distinction from SH2 domains (7). Remember that phosphorylation of tyrosine residues is a reaction to a stimulus, and therefore, under basal conditions, the overwhelming majority of tyrosine residues are not phosphorylated. This gives PTB-containing proteins important implications in regulation of signaling under such states.

Fig. 3. Turquoise trapezoid = activated receptor; Turquoise vertical rectangles = membrane-bound phospholipids; Tri-colored rectangle = docking protein (IRS-1/2 in this case); Red circle = enzyme; blue box attached to red circle = SH2 domain of enzyme. The PH domain of IRS-1 is partially recruited by membrane phospholipids and partially recruited by phosphorylated tyrosine residue of the activated insulin receptor (phosphotyrosines are not explicitly shown for sake of clarity). The variable C-terminal domain of IRS-1/2 contains multiple tyrosine phosphorylaiton sites in the context of specific motifs, which recruit various SH2-domain containing proteins (enzymes mostly).

Post-Receptor Signaling- the Docking Protein: IRS and Shc

Immediately downstream of the insulin receptor are the insulin receptors substrate (IRS) and Shc proteins. The IRS proteins are a family of (thus far) six members that are absolutely essential for propagation of many of insulin’s metabolic signals (11-17). IRS proteins are found in all tissues of the body, and while all members share some similar characteristics, their functions differ widely according to isoform (8), receptor coupling (5, 6), and cell type (4). When dealing with insulin, IRS-1 and 2 are of primary interest due to their strong presence in the peripheral tissues as well as the metabolic abnormalities seen when their function is disrupted, either through insulin resistance or via gene manipulation. Shc is a family of docking proteins that mediates the majority of insulin’s mitogenic effects, and while it is often ignored in insulin signaling, it is discussed with regularity in the field of cancer biology (19). While our focus will be primarily the IRS proteins in this and later articles, comparing Shc and IRS binding and signaling has useful conceptual implications.

A Fork in the Path-Which Way to gGo?

Starting from the insulin receptor onwards, insulin has two main signaling branches- one arm deals with metabolic effects, i.e. nutrient regulation, while the other is dedicated to cell proliferation and survival, i.e. mitogenesis. Within each primary role, there are subdivisions. Then, there are subclasses within the subdivisions. Under metabolic effects fall the general processes of glucose, lipid, and amino acid metabolism. Glucose metabolism alone encompasses GLUT4 translocation, regulation of glycolysis, the Kreb’s cycle, glycogen synthesis, and gluconeogenesis. Then, there is regulation at the level of enzyme activity, translation, and transcription; the same trends are true for lipid and amino acid metabolism. Additionally, there is the issue of cell type and related signaling pathways that share intermediates. For one hormone to regulate all of these processes in a coordinated manner, great diversity is needed from the very top, which is exactly what the docking proteins accomplish. Think of these proteins as multifunctional outlet adaptors that one might use to expand the capacity and function of a single electrical wall socket- without compromising specificity.

When insulin binds to the insulin receptor, the conformation of the receptor changes and activates an intrinsic phosphorylation capacity found in a portion of the receptor (1). This results in several tyrosine residues of the insulin receptor being phosphorylated; as mentioned in the last article, phosphorylation of an amino acid alters not only its own properties, but by virtue of charge and steric (size) characteristics of the phosphate group, neighboring amino acids may change their conformation. This change may lead to recruitment of another protein, or open up a pocket in the original protein to allow access by an enzyme. In the case of the insulin receptor, tyrosine autophosphorylation occurs on several tyrosine residues, in different regions of the receptor. This leads to recruitment of different protein domains that are attracted by phosphotyrosines to the now-phosphorylated regions on the insulin receptor (18). The first proteins that are recruited are the IRS and Shc isoforms (17, 18) by virtue of their PH and PTB domains. The PH domain recruits the proteins to the cell membrane while the PTB domain binds directly to phosphotyrosine residues on the insulin receptor- or, more specifically, to three phosphotyrosines in the so-called “activation loop” of the insulin receptor (18-20).

As mentioned above, the IRS proteins mediate most of insulin’s metabolic effects, while Shc appears more important in mitogenic signaling (19). If partitioning of insulin’s two main signaling arms occurs at the level of docking proteins immediately downstream of the receptor, yet the two classes of docking proteins have the same type of domain (PTB) that binds the same region of the phosphorylated insulin receptor (activation loop), the obvious issue of specificity (i.e. choosing one pathway over the other) comes into being. Unfortunately, the precise signal that determines the magnitude of metabolic vs. mitogenic preference is currently unclear, although it has been proposed that the decision is based upon ligand binding kinetics (i.e. the affinity of the ligand for the receptor and how rapidly the ligand dissociates) to the insulin receptor (22) as well as differential binding thermodynamics between the IRS and Shc proteins (21, 22); IRSs have been referred to as having “fast” binding kinetics, while Shc is “slower” (22). Such considerations enter into a realm of biology and biophysics that the author is unfamiliar with, but it suffices to say that the structure of the ligand, receptor, and docking protein is largely responsible for the discrepancies in signaling specificity and strength.

Fig. 4.Yellow rectangle = PH domain; Purple rectangle = PTB domain; Grey rectangle = C-terminal region of IRS-1/2; Red rectangle = C-terminal region of SHC. Upon phosphorylaion of tyrosine residues the insulin receptor (blue trapezoid), two primary types of docking proteins are recruited- the IRS proteins and the SHC proteins; the former mediated primarily metabolic effects of insulin, while the latter is involved in mitogenic signaling.

The coordinated roles of ligand, receptor, and docking proteins should now start to become apparent in guiding signaling. It is often observed in vitro that ligands for different receptor tyrosine kinases, such as epidermal growth factor (EGF), platelet-derived-growth factor (PDGF), and insulin-like-growth factor-1 (IGF-1) bind to other RTKs in addition to their cognate receptor (23-26). When EGF or PDGF binds to the insulin receptor in an adipocyte model cell line, it can signal through the same molecules insulin uses, and can even mediate glucose transport; however, the effect is much weaker compared to a similar quantity of insulin (24). We can now postulate a mechanism for this discrepancy: if the binding kinetics of various growth factors with respect to the insulin receptor mediates post-receptor signaling (22) due to effects on receptor phosphorylation (affecting which sites are phosphorylated, the rapidity of the reaction, as well as its reversibility), then:

1. The type of docking protein recruited may vary2. The amount of docking protein recruited may vary3. The efficiency of phosphorylation of tyrosine residues in the docking proteins may vary4. The subcellular distribution of docking proteins may vary

5. Any combination of the above

Such exquisite specificity is possible because of (often) slight but significant differences in structure and sequence of molecules involved in signaling which result in alterations in protein-protein interactions.

The IRS Proteins in Insulin Signaling

IRS-1 was discovered in the mid 1980s by Kahn et al., who observed that upon treatment with insulin, the insulin receptor was not the only protein that underwent tyrosine phosphorylation; one particular protein separate from the insulin receptor was tyrosine phosphorylated within seconds after the receptor underwent autophosphorylation (1). Later on, this protein was named insulin receptor substrate-1, and its proposed role was as an adaptor/scaffold protein connecting the insulin receptor to phosphatidylinositol 3-kinase, or PI3K for short, a key mediator of insulin’s metabolic effects (we will discuss PI3K in future articles) (2). An isoform of IRS-1, anointed IRS-2, was purified and cloned by Sun et al. in 1995 (3); both IRS-1 and 2, and indeed, all members of the IRS family, have a closely conserved N-terminal PH (Pleckstrin homology) and PTB (phosphotyrosine-binding) domain. They differ most in their C-terminal region that has various sites (tyrosine and serine residues) that can be phosphorylated by certain kinase activities – including that of the activated insulin receptor- and then recruit different proteins with SH2 domains, such as PI3K (3).

Clearly, the C-terminal region of the IRS proteins governs the types of SH2 domain-containing molecules that are recruited to their vicinity and activated. However, even though various protein domains can be quite homologous in terms of amino acid sequence and/or tertiary structure (as is the case for the N-terminal PH and PTB domains), this does not mean that they recognize the same motifs (“motif” referring to a conserved sequence of amino acids). For example, the SH2 domain is characterized by an affinity for phosphorylated tyrosine residues; however, the amino acids neighboring the phosphorylated tyrosine residue determine the particular SH2 domain that is recruited. Thus, the YXXM or YMXM motif (where Y = tyrosine, X = an amino acid, and M = methionine) preferentially recruits the SH2 domain of PI3K, while the YVNI motif (where Y = tyrosine, V = valine, N = asparagine, and I = isoleucine) is specific for Grb2, another SH2 domain-containing protein (26, 27). What the cell has done here is to create a universal signal/reaction and make it unique by altering neighboring amino acids- a very efficient and effective mechanism by which specificity is conferred.

Fig. 5. Within the IRS protein are several tyrosine residues that can be phosphorylated; these then recruit the SH2 domains of different proteins. The many closely related SH2 domains can specifically recognize amino acids in the context of the phosphorylated tyrosine, conferring specificity of downstream effector recruitment and activation.

Comparing IRS-1 and 2, the overall amino acid sequence similarity between them is 45%; homology is >60% in the PH and PTB domains but <35% in the divergent carboxyterminal tail (4). As mentioned above, the variable C-terminal tail contains tyrosine residues capable of being phosphorylated, and these tyrosines are present in several different motifs to recruit different SH2-containing proteins. This means that the insulin receptor can bind one IRS molecule, but upon tyrosine phosphorylation by the interaction, that single IRS molecule can recruit roughly a dozen SH2-containing proteins (26). This is far more efficient than direct binding, and is potentially easier to regulate because if the signal needs to be amplified or shut down, the cell can synthesize more IRS as opposed to the actual receptor. Thus, the IRS proteins confer advantages with respect to specificity (through unique amino acid motifs on the IRS proteins), diversity (through different motifs), and efficiency (via many motifs).

Coming back to the closely related N-terminus of IRS-1 and 2, where the PH and PTB domains are located, we see that there are differences with respect to insulin receptor binding- after all, even if the domains were 95% homologous, that still leaves plenty of room for unique amino acid sequences, and thus, tertiary structure (5). While the PH domain of IRS-1 and 2 targets the protein to the plasma membrane (recognition of membrane phospholipids), the PTB domain of IRS proteins bind to a specific phosphotyrosine motif in the insulin receptor, the NPXpY sequence (N = asparagine, P = praline, X = another amino acid, pY = phosphotyrosine) (6, 18). However, IRS-2 has an additional interaction site with the insulin receptor in a region called the kinase regulatory loop binding domain (5, 6). This association may be responsible for some of the notable difference in IRS-1 and 2 signaling; knocking out the IRS-1 gene leads to impaired glucose tolerance and a 50% reduction in intrauterine growth, but not overt diabetes or obesity (8). On the other hand, knocking out IRS-2 causes overt diabetes, with insulin resistance in the skeletal muscle and liver, decreased pancreatic beta-cell compensation, and adiposity (9). Thus, despite the similarity in structure and name, IRS-1 and 2 have rather different functions depending on the tissue being examined.

To further stress the extent of diversity in function between IRS-1 and 2, we turn to gene manipulation studies. Through global knockout, tissue-specific knockout, and siRNA-mediated gene knockdown, IRS-2 has emerged as being necessary for central leptin signaling (11-16), regulation of hepatic lipid metabolism (11, 12), and mediation of the mitogenic pathway in skeletal muscle (11). IRS-1, on the other hand, appears to be primarily involved in glucose metabolism in the liver and skeletal muscle, as well as actin cytoskeleton remodeling in skeletal muscle (11, 12); this latter finding may also be related to the differential subcellular concentration of IRS-1 and 2, with IRS-1 being localized to intracellular membranes and IRS-2 preferring the cytosol (12). IRS-1 also seems to be important in IGF-1 mediated cellular growth, while IRS-2 is not (17). This brief summary is by no means exhaustive, and we will come back to some of these topics when all of the relevant and known signaling intermediates downstream of the IRS proteins have been discussed.


To summarize, IRS-1 and 2 are the two IRS isoforms that are critical components of insulin signaling in most peripheral tissues and even some central ones. Through the coordinated action of their PH and PTB domains, IRS-1 and 2 are recruited to the plasma membrane and bind the insulin receptor, which phosphorylates at least twelve tyrosine residues in the carboxyterminal tail. Each tyrosine is within a specific motif, and when phosphorylated, the tyrosine serves as a sign for the appropriate SH2-containing protein to be recruited from its subcellular location. The presence of different motifs in IRS-1 and 2 allows great variability in their signaling potentials. However, IRS-2 has an additional ability to bind the activated insulin receptor in a region called the kinase regulatory loop binding domain, and this interaction may govern either additional specificity, signal strength, or both. If one thinks of the insulin receptor as an empty electrical socket, then SH2-containing proteins are devices with the correct plugs to plug into the socket, be activated/powered up. These are then removed to further activate other devices (signaling events downstream of the IRS proteins which will be covered in the subsequent articles). However, if there are fifty devices that need to be plugged in and activated, using a multi-pronged adaptor is much more efficient. And, if we use a multi-prong adaptor with outlets of all shapes and sizes, we increase our efficiency and diversity without sacrificing specificity. That, simply stated, is the role of the IRS proteins, and indeed, all related docking proteins.


1. White, M. F., Maron, R. & Kahn, C. R. (1985) Insulin rapidly stimulates tyrosine phosphorylation of a Mr-185,000 protein in intact cells. Nature, 318(6042): 183-6

2. Ruderman, N. B. & Kelly, K. L. (1993) Insulin-stimulated phosphatidylinositol 3-kinase. Association with a 185-kDa tyrosine-phosphorylated protein (IRS-1) and localization in a low density membrane vesicle. J Biol Chem, 268(6): 4391-8

3. Sun, X. J., Wang, L. M., Zhang, Y., Yenush, L., Myers, M. G. Jr., Glasheen, E., Lane, W. S., Pierce, J. H. & White, M. F. (1995) Role of IRS-2 in insulin and cytokine signaling. Nature, 377(6545): 173-7

4. Sun, X. J., Pons, S., Wang, L. M., Zhang, Y., Wenhsu, L., Burks, D., Myers, M. G. Jr., Glasheen, E., Copeland, N. G., Jenkins, N. A., Pierce, J. H., White, M. F. (1997) The IRS-2 gene on murine chromosome 8 encodes a unique signaling adapter for insulin and cytokine action. Mol Endocrinol, 11(2): 251-62

5. Swaka-verhelle, D., Baron, V., Mothe, I., Filloux, C., White, M. F. & Van Obberghen, E. (1997) Tyr624 and Tyr628 in insulin receptor substrate-2 mediate its association with the insulin receptor. J Biol Chem, 272(26): 16414-20

6. Sawka-Verhelle, D., Tartare-Deckert, S. White, M. F. & Van Obberghen, E. (1996) Insulin receptor substrate-2 binds to the insulin receptor through its phosphotyrosine-binding domain and through a newly identified domain comprising amino acids 591-786. J Biol Chem, 271(11): 5980-3

7. Schlessinger, J. & Lemmon, M. A. (2003) SH2 and PTB domains in tyrosine kinase signaling. Science STKE, 2003(191): RE12

8. Araki, E., Lipes, M. A., Patti, M. E., Bruning, J. C., Haag, B. 3rd, Johnson, R. S. & Kahn, C. R. (1994) Alternative pathway of insulin signalling in mice with targeted disruption of the IRS-1 gene. Nature, 372(6502): 186-90

9. Withers, D. J., Gutierrez, J. S., Towery, H., Burks, D. J., Ren, J. M., Previs, S., Zhang, Y., Bernal, D., Pons, S., Shulman, G. I., Bonner-Weir, S. & White, M. F. (1998) Disruption of IRS-2 causes type 2 diabetes in mice. Nature, 391(6670): 900-4

10. Huang, C., Thirone, A. C., Huang, X. & Klip, A. (2005) Differential contribution of insulin receptor substrates 1 versus 2 to insulin signaling and glucose uptake in L6 myotubes. J Biol Chem, 280(19): 19426-35

11. Taniguchi, C. M., Ueki, K. & Kahn, R. (2005) Complementary roles of IRS-1 and IRS-2 in the hepatic regulation of metabolism. J Clin Invest, 115(3): 718-27

12. Tobe, K., Suzuki, R., Aoyama, M., Yamauchi, T., Kamon, J., Kubota, N., Terauchi, Y., Matsui, J., Akanmua, Y., Kimura, S., Tanaka, J., Abe, M., Ohsumi, J., Nagai, R. & Kadowaki, T. (2001) Increased expression of the sterol regulatory element-binding protein-1 gene in insulin receptor substrate-2 (-/-) mouse liver. J Biol Chem, 276(42): 38337-40

13. Suzuki, R., Tobe, K., Aoyama, M., Inoue, A., Sakamoto, K., Yamauchi, T., Kamon, J., Kubota, N., Terauchi, Y., Yoshimatsu, H., Matsuhisa, M., Nagasaka, S., Ogata, H., Tokuyama, K., Nagai, R. & Kadowaki, T. (2004) Both insulin signaling defects in the liver and obesity contribute to insulin resistance and cause diabetes in the IRS2 (-/-) mice. J Biol Chem, 279(24): 25039-49

14. Asilmaz, E., Cohen, P., Miyazaki, M., Dobryzn, P., Ueki, K., Fazikodjaeva, G., Soukas, A. A., Kahn, C. R., Ntambi, J. M., Socci, N. D. & Friedman, J. M. (2004) Site and mechanism of leptin action in a rodent form of congenital lipodystrophy. J Clin Invest, 113(3): 414-24

15. Burks, D. J., de Mora, J. F., Schubert, M., Withers, D. J., Myers, M. G., Towery, H. H., Altamuro, S. L., Flint, C. L. & White, M. F. (2000) IRS-2 pathways integrate female reproduction and energy homeostasis. Nature, 407(6802): 377-82

16. Bruning, J. C., Gautam, D., Burks, D. J., Gillette, J., Schbert, M., Orban, P. C., Klein, R., Krone, W., Muller-Wieland, D. & Kahn, C. R. (2000) Role of brain insulin receptor in control of body weight and reproduction. Science, 289(5487): 2066-7

17. Bruning, J. C., Winnay. J., Cheatham, B. & Kahn, C. R. (1997) Differential signaling by insulin receptor substrate 1 (IRS-1) and IRS-2 in IRS-1-deficient cells. Mol Cell Biol, 17(3): 1513-21

18. Mandiyan, V., O’Brien, R., Zhou, M., Margolis, B., Lemmon, M. A., Sturtevant, J. M. & Schlessinger, J. (1996) Thermodynamic studies of SHC phosphotyrosine interaction domain recognition of the NPXpY motif. J Biol Chem, 271(9): 4470-5

19. Sasaoka, T. & Kobayahshi, M. (2000) The functional significance of Shc in insulin signaling as a substrate of the insulin receptor. Endocrine J, 47(4): 373-81

20. Tartare-Deckert, S., Sawka-Verhelle, D., Murdaca, J. & Van Obberghen, E. (1995) Evidence for a differential interaction of SHC and the insulin receptor substrate-1 (IRS-1) with the insulin-like growth factor-1 (IGF-1) receptor in the yeast two-hybrid system. J Biol Chem, 270(40): 23456-60

21. Farooq, A., Plotnikova, O., Zeng, L. & Zhou, M. M. (1999) Phosphotyrosine binding domains of Shc and insulin receptor substrate 1 recognize the NPXpY motif in a thermodynamically distinct manner. J Biol Chem, 274(10): 6114-21

22. Shymoko, R. M., Dumont, E., De Meyts, P. & Dumont, J. E. (1999) Timing-dependence of insulin-recptor mitogenic versus metabolic signalling: a plausible model based on coincidence of hormone and effector binding. Biochem J, 399(Pt3): 675-83

23. Ricort, J. M., Tanti, J. F., Van Obberghen, E. & Le Marchand-Brustel, Y. (1996) Different effects of insulin and platelet-derived growth factor on phosphatidylinositol 3-kinase at the subcellular level in 3T3-L1 adipocytes. A possible explanation for their specific effects on glucose transport. Eur J Biochem, 239(1): 17-22

24. Conricode, K. M. (1995) Involvement of phosphatidylinositol 3-kinase in stimulation of glucose transport by growth factors in 3T3-L1 adipocytes. Biochem Mol Biol Int, 36(4): 835-43

25. Staubs, P. A., Nelson, J. G., Reichart, D. R. & Olefsky, J. M. (1998) Platelet-derived growth factor inhibits insulin stimulation of insulin receptor substrate-1 associated phosphatidylinositol 3-kinase in 3T3-L1 adipocytes without affecting glucose transport. J Biol Chem, 273(39): 25139-47

26. Summers, S. A., Whiteman, E., Cho, H., Lipfert, L. & Birnbaum, M. J. (1999) Differentiation-dependent suppression of platelet-derived growth factor signaling in cultured adipocytes. J Biol Chem, 274(34): 23858-67

27. Sun, X. J., Crimmins, D. L., Myers, M. G. Jr., Miralpeix, M. & White, M. F. (1993) Pleiotropic insulin signals are engaged by multisite phosphorylation or IRS-1. Mol Cell Biol, 13(12): 7418-28

28. Myers, M. G. Jr., Mendez, R., Shi, P., Pierce, J. H., Rhoads, R. & White, M. F. (1998) The COOH-terminal tyrosine phosphorylation sites on IRS-1 bind SHP-2 and negatively regulate insulin signaling. J Biol Chem, 273(41): 26908-14