Beirne Carter Center for Immunology Research and the Department of Microbiology, University of Virginia, Charlottesville, Virginia, VA 22908, USA

Correspondence to: K S Ravichandran, Beirne Carter Center for Immunology Research and the Department of Microbiology, University of Virginia, Charlottesville, Virginia, VA 22908, USA. E-mail: Ravi@virginia.edu

The adapter protein Shc was initially identified as an SH2 containing proto-oncogene involved in growth factor signaling. Since then a number of studies in multiple systems have implicated a role for Shc in signaling via many different types of receptors, such as growth factor receptors, antigen receptors, cytokine receptors, G-protein coupled receptors, hormone receptors and integrins. In addition to the ubiquitous ShcA, two other shc gene products, ShcB and ShcC, which are predominantly expressed in neuronal cells, have also been identified. ShcA knockout mice are embryonic lethal and have clearly suggested an important role for ShcA in vivo. Based on dominant negative studies and mouse embryos deficient in ShcA, a clear role for Shc in leading to mitogen activated protein kinase (MAPK) activation has been established. However MAPK activation may not be the sole function of Shc proteins. Although Shc has also been linked to other signaling events such as c-Myc activation and cell survival, the mechanistic understanding of these signaling events remains poorly characterized. Given the apparently central role that Shc plays signaling via many receptors, delineating the precise mechanism(s) of Shc-mediated signaling may be critical to our understanding of the effects mediated through these receptors. Oncogene (2001) 20, 6322-6330.

Shc; Grb2; MAPkinase; SH2; PTB; SHIP

The adapter protein Shc is a prototype adapter protein that has been quite useful in our understanding of the function of adapter proteins in cellular signaling. Three shc genes have been identified in mammals and their gene products have been referred to as ShcA, ShcB and ShcC (Luzi et al., 2000; Pelicci et al., 1996). Homologues of shc have also been identified in Drosophila and C. elegans, suggesting an evolutionarily conserved role for Shc proteins (Lai et al., 1995; Luzi et al., 2000). In mammals shcA is ubiquitously expressed, while shcB and ShcC expression appear limited to neuronal cells (Pelicci et al., 1996; Sakai et al., 2000). ShcA is expressed as three isoforms of about 46, 52 and 66 kDa (all generated from the same message either through RNA splicing or alternative translational initiation) (Migliaccio et al., 1997; Pelicci et al., 1992).

All Shc isoforms possess two distinct domains that bind phosphotyrosine containing sequences (PTB and SH2) and a central region (denoted CH1) that contains critical tyrosine phosphorylation sites (Pelicci et al., 1996). The 66 kDa isoform of ShcA is expressed in most cells except in the hematopoietic lineage and contains an additional amino-terminal CH-like region (denoted CH2) (Migliaccio et al., 1997; Pelicci et al., 1992; Figure 1). The tyrosine phosphorylation of Shc has been noticed upon engagement of numerous cell surface receptors such as growth factor receptors (Gelderloos et al., 1998; Pelicci et al., 1992; Pronk et al., 1994; Rozakis-Adcock et al., 1992; Sasaoka et al., 1994; Stephens et al., 1994; Yokote et al., 1994); antigen receptors (Ravichandran et al., 1993; Saxton et al., 1994); cytokine receptors (Burns et al., 1993; Damen et al., 1993; Lioubin et al., 1994; Matsuguchi et al., 1994; Pratt et al., 1996; Ravichandran and Burakoff, 1994; Velazquez et al., 2000); G-protein coupled receptors (Chen et al., 1996; van Biesen et al., 1995) and hormone receptors (Erwin et al., 1995; Kousteni et al., 2001; Migliaccio et al., 1996; Morte et al., 1998). In addition, a role for Shc has been ascribed in transformation by the polyoma middle T antigen and BCR-ABL (Campbell et al., 1994; Goga et al., 1995; Mullane et al., 1998; Puil et al., 1994; Zhu et al., 1998). Moreover, hyper-phosphorylation of Shc has been seen in many different types of tumors (Biscardi et al., 1998; Pelicci et al., 1995; Stevenson and Frackelton, 1998). The importance of Shc in vivo has been demonstrated by the knockout of the shcA gene, which results in embryonic lethality at day 11.5 (Lai and Pawson, 2000). Although a role for Shc in activation of the Ras/MAPK pathway and its role in mitogenic signaling has been better described (Bonfini et al., 1996), what other roles it may play in signaling and why this protein is so often recruited/targeted for phosphorylation by many different types of receptors is not clear. In the following sections, I have attempted to provide an overview of the modular architecture of Shc proteins along with the known functions for these domains, the signaling pathways in which Shc has been implicated, the existing knowledge or the lack thereof about the mechanisms of Shc function, and some speculations on Shc function in vivo. I have restricted my discussion primarily to p52 ShcA, with reference to ShcB and ShcC as needed.

Modular domains of Shc

Shc-SH2 domain

The SH2 domain of Shc, which is located at its very C-terminus, was initially characterized as a domain responsible for its recruitment to the activated epidermal growth factor (EGF) receptor and was implicated in EGF-dependent tyrosine phosphorylation of Shc (Pelicci et al., 1992). However, more recent studies suggest that the SH2 may play a less critical role in recruitment of Shc to the EGFR (see below). Shc-SH2 also interacts with the T cell receptor chain (TCR-) (Ravichandran et al., 1993) and the platelet derived growth factor receptor (Gelderloos et al., 1998; Yokote et al., 1994). The nuclear magnetic resonance structure of the Shc-SH2 domain in complex with a TCR- derived peptide revealed that the Shc-SH2 folds overall very similar to other SH2 domains (Mikol et al., 1995; Zhou et al., 1995b). Shc-SH2 has a characteristic pocket for binding phosphotyrosine, and along with residues within its D loop showed a preference for peptides with a leucine or isoleucine at the +3 position relative to the phosphotyrosine (pY). Recently, a phosphotyrosine-independent interaction between Shc-SH2 and mPAL, a novel protein expressed in proliferating cells, has also been identified (Schmandt et al., 1999). The biological significance of the Shc:mPAL interaction remains to be determined. Although one study reported that Shc-SH2 could facilitate formation of dimers between Shc molecules at low pH conditions (Rety et al., 1996), we have thus far failed to observe any evidence for dimerization of native Shc molecules (based on multiple approaches).

It was initially presumed that recruitment of Shc to all the known receptors occurred via its SH2 domain. However, it quickly became apparent that the Shc-SH2 alone did not interact with some of the receptors, although full-length Shc was able to bind (such as the TrkA receptor or the polyoma middle T antigen) (Campbell et al., 1994; Stephens et al., 1994). Mutagenesis studies of the TrkA receptor and the polyoma middle T antigen also suggested a requirement for a -3 asparagine residue relative to the tyrosine for Shc binding, despite the then existing conclusion that the specificity of SH2 domain binding depended on residues C-terminal to the pY (Campbell et al., 1994; Stephens et al., 1994). Although this led to some debate on how Shc-SH2 may recognize target sequences, the identification of the phosphotyrosine-binding domain (PTB) at the N-terminus of Shc quickly resolved this issue.

Shc-PTB domain

Two groups identified the existence of a second domain capable of interacting with phosphorylated tyrosine residues independently (Blaikie et al., 1994; Kavanaugh and Williams, 1994). One of the unique features of the recognition by the Shc-PTB (also referred to as phosphotyrosine-interaction domain/PID) was that its binding to target sequences was determined by residues N-terminal to the pY, and was not influenced by residues C-terminal to the pY (Blaikie et al., 1997; Trub et al., 1995; Zhou et al., 1995a,c). Since then more than 160 proteins that contain PTB domains have been recognized. However, Pelicci and colleagues have recently noted that the Shc family of proteins is unique in having both the N-terminal PTB and a C-terminal SH2 domain (Luzi et al., 2000).

Phosphopeptide binding by the Shc-PTB domain: The structure of Shc-PTB domain in complex with the TrkA phosphopeptide revealed a number of features of Shc-PTB recognition (Zhou et al., 1995c). The salient observations were: (i) The Shc-PTB domain is a compact, tightly folded structure; (ii) The peptide itself forms a tight -bend when bound to the PTB domain. Previous observations by many groups demonstrating a requirement for the -3 asparagine and -5 hydrophobic residues (relative to the phosphotyrosine) were explained by the contacts made by the peptide with the PTB domain; (iii) Arg175 (R175) within the phosphotyrosine-binding pocket was critical for interaction with specific tyrosine-phosphorylated proteins in vitro and in vivo as determined by site-directed mutagenesis. Recently, a variant of Drosophila Shc (dShc) that carries a mutation at a site analogous to R175 was found to be null for Shc function (Luschnig et al., 2000).

The core Shc-PTB region that binds phosphopeptide appears to exist between amino acids 46-207 of p52Shc. The first 46 amino acids of the Shc-PTB may also play a role in regulating Shc-PTB binding to targets. This is based on the observation that the affinity of peptide binding to the PTB domain constructs that begin from amino acid 46 is about 10-fold lower than PTB constructs that begin at amino acid 1 (Trub et al., 1995) (MM Zhou, unpublished observations). Moreover, recent NMR studies suggest that there may be intramolecular regulation of the Shc-PTB domain that involves the N-terminal residues (MM Zhou and KS Ravichandran, unpublished observations). In this regard it is unclear what role p46 plays in vivo since it would be expected to bind less efficiently through its PTB compared to p52Shc. Interestingly, the p52Shc, but not the p46Shc, is efficiently phosphorylated by the insulin receptor and polyoma middle T antigen, both of which depend on PTB interactions; in contrast, the EGFR leads to efficient phosphorylation of both isoforms (Okada et al., 1995). Whether p46Shc may have some unique regulatory role on the functions of the other isoforms remains to be determined.

Structural homology to PH domains and phospholipid binding: One of the unexpected features of the structure of the Shc-PTB domain was its striking structural similarity to pleckstrin homology (PH) domains, despite any sequence homology between Shc-PTB and PH domains (Zhou et al., 1995c). PH domains have been shown to bind to acidic phospholipids such as PI(4,5)P2 and PI(3,4,5)P3 (Harlan et al., 1994, 1995). Consistent with the structural homology, the Shc-PTB domain was also found to bind acidic phospholipids such as PI (4,5)P2, PI(4)P (Zhou et al., 1995c) and also to PI(3,4,5)P3 (Rameh et al., 1997). The affinity of binding (K_D=10-50 M) is comparable to the affinity of lipid vesicles to myristoylated or farnesylated peptides (K_D=5-39 M), the latter being implicated in protein:lipid interactions (e.g. Ras). This suggested that the interaction of Shc with the membrane could occur independent of an interaction with tyrosine-phosphorylated receptors.

Consistent with the above observation, we have seen about 5% of endogenous Shc being localized in the membrane fractions in unstimulated cells (Ravichandran et al., 1997). Lotti et al. (1996) also observed that Shc proteins are localized to the membranes by immunofluorescence. When different domains of Shc were transiently expressed in COS cells, the PTB domain of Shc was sufficient for localization to the membrane fractions (Ravichandran et al., 1997). Moreover, using NMR-structure based mutagenesis, we were able to identify residues within Shc-PTB critical for phospholipid binding (and in turn membrane localization) that are distinct from the residues necessary for pTyr binding (and in turn, receptor-binding). These studies also indicated that both the phosphotyrosine and phospholipid interactions via the PTB domain are critical for Shc function in interleukin-3 signaling (Ravichandran et al., 1997). Although we had only tested PI(4,5)P2 and PI(4)P, other lipids such as PIP3 (whose levels rise upon cellular activation) could also regulate localization and signaling via Shc. The latter possibility has been supported by the observation that inhibition of PI3 kinase (by wortmannin) affected Shc phosphorylation by G-protein coupled receptors (Touhara et al., 1995).

Although the PTB domains were so named because the Shc-PTB domain (which was the first identified) binds phosphotyrosine sequences with very high affinity (Kavanaugh and Williams, 1994), many other PTB domains do not require a phosphorylated tyrosine or even a tyrosine for their binding to targets (Margolis, 1996; Margolis et al., 1999). Moreover, the structures of the other PTB domains that have been solved suggest that all PTB domains have a PH domain fold, despite the lack of obvious amino acid sequence similarity between these domains. However, not all PTB domains bind to phospholipids. For example, the IRS-1 PTB domain fails to bind phospholipids (Eck et al., 1996). It is interesting to note that many signaling proteins contain a PH domain (which is used for targeting or regulation) in addition to the PTB or SH2 domain. For example, IRS-1 contains a PTB domain along with an adjacent PH domain, both of which are critical for its function (Dhe-Paganon et al., 1999; Shoelson, 1997). The 'hybrid' nature of the Shc-PTB, and its ability to accomplish membrane localization and receptor binding via a single domain, is relatively uncommon among signaling proteins (note: seen also for Dab-2 PTB) (Howell et al., 1999; Ravichandran et al., 1997).

CH1 domain

The CH1 domain of Shc (about 52-167 amino acids in different shc genes) resides between the PTB and SH2 domains (Luzi et al., 2000). Although this region was originally named as 'collagen homology' (CH) due to the high number of glycines and prolines in this region (Pelicci et al., 1992), Shc does not have the typical collagen like repeats and unlike collagen, Shc is intracellular. This CH1 region overall shows the least conservation between Shc proteins, with short stretches of higher homology. It has been noted recently that short stretches of the CH1 region contain unique 'signature' sequences that are present only in ShcA, B or C (Luzi et al., 2000). The CH1 region of p52Shc contains three critical tyrosines, Y239, Y240 and Y317 that become phosphorylated upon engagement of a number of cell surface receptors. Y317 is conserved among mammalian proteins, but is not seen in lower organisms. Y239 and Y240 are seen in Drosophila Shc, but not in the worm protein. Thus, the phosphorylation of Y317 may be a more recently acquired feature of Shc (Luzi et al., 2000).

During in vitro studies, we observed some subtle, but reproducible differences in phosphorylation of Y239/240 versus Y317 by different tyrosine kinases (Walk et al., 1998). It is noteworthy that recently, using specific Src-family kinase inhibitors, Blake et al. (2000) have observed that the Y239/Y240 sites are targets for Src in vivo, while the Y317 is phosphorylated by another kinase, perhaps the PDGF receptor itself. It remains to be seen whether different kinases might be used for Shc phosphorylation by other receptors as well and whether any hierarchy of phosphorylation exists between these tyrosines.

Several lines of evidence suggest that these tyrosines are important, since mutants of either Y317 alone, Y239F/Y240F alone or Y239/240/317 combined have been shown to have dominant negative effects (Gotoh et al., 1996, 1997; Pratt et al., 1999; Salcini et al., 1994). Interestingly, both Y239 and Y317 have the canonical +2 asparagine that is critical for the binding of Grb2-SH2 domain. We and others have shown that the SH2 domain of Grb2 can bind to both the Y239 and Y317 (Walk et al., 1998). Velazquez et al. (2000) recently reported that Grb2 may bind preferentially to Y317, but Grb2 could bind equally well to the Y239 site when the Y240 was also phosphorylated. Besides Grb2, other proteins that bind differentially to these tyrosines have been observed using Shc-derived phosphopeptides (van der Geer et al., 1996). Gotoh et al. (1996, 1997) have suggested that Y239/Y240 residues may lead to activation of c-Myc, while the Y317 may lead to MAPK activation. The relative significance of these modes of signaling is discussed in greater detail later.

CH1 domain of Shc also contains several PXXP motifs that can serve as targets for SH3 domains on other proteins. In fact, an interaction between Shc and the SH3 domain of Src, Fyn and Lyn has been recognized, although the biological function of this interaction remains to be understood (Weng et al., 1994). Also, the PTB+CH or CH+SH2 fusion proteins appear to bind the respective targets better than either the PTB alone or SH2 alone in vitro. Thus, there appears to be an unexplained role for the CH domain in regulating the binding via the PTB domain (our unpublished observations).

Based on the published studies and our unpublished work, we suggest a two-step model for Shc recruitment to individual receptors and its subsequent tyrosine phosphorylation. A small fraction of Shc proteins would be basally localized to the membrane through PTB:phospholipid interactions. This would be a low affinity interaction, but would keep a fraction of Shc on the membrane. Upon phosphorylation of the target receptor, Shc would then translocate to the activated receptor, since the PTB : phospholipid affinity is lower than Shc-PTB : phosphotyrosine or Shc-SH2 : phosphotyrosine interactions. This would then lead to Shc tyrosine phosphorylation and downstream signaling.

CH2 domain

p66ShcA contains a second CH-like domain (denoted as CH2) (Migliaccio et al., 1997). This CH2 domain is also in the longer isoforms of ShcB and C, but not in the Drosophila Shc protein (Luzi et al., 2000). This has led Pelicci and colleagues to propose that the evolution of the CH2 domain may be more recently acquired (Luzi et al., 2000). Interestingly, the CH2 domain does not become tyrosine phosphorylated, but gets serine phosphorylated. This serine phosphorylation has been linked to the role of this p66Shc isoform in oxidative stress responses (Migliacchio et al., 1999). Migliacchio et al. also noted that the overexpression of the CH2 domain alone could function as a dominant negative and inhibit c-fos activation (Migliaccio et al., 1997). Precisely how the CH2 domain leads to the inhibition of the c-fos promoter, and how p66 functions during oxidative stress signaling are not known.

Signaling via Shc

Role of Shc in mitogenic signaling through Ras

Conversion of Ras from its inactive GDP-bound form to its active GTP-bound form is an essential early event in signaling through most receptors, and Ras activation is crucial for mitogenesis. Studies from various laboratories in different model systems have shown an important role for Shc in leading to Ras activation. The current model for Shc-mediated Ras activation is as follows: First, Shc is tyrosine-phosphorylated by receptor activation and it subsequently interacts with Grb2. Grb2, in turn, binds to Ras guanine nucleotide exchange factor, Sos. Second, the Shc : Grb2 : Sos complex gets localized to the membrane through the interaction of Shc with the phosphorylated receptor. This Shc : receptor interaction could be mediated either via the Shc-SH2 (Pelicci et al., 1992; Ravichandran et al., 1993; Yokote et al., 1994) or the Shc-PTB domain (Blaikie et al., 1994; Pratt et al., 1996; Ravichandran et al., 1996). In the case of the integrin family of receptors and G-protein coupled receptors, the precise mechanism by which the Shc:Grb2:Sos complex is localized to the membrane is not clear. Although some level of Sos is bound to Grb2 basally, in some systems, the Shc:Grb2 interaction also increases the level of Sos bound to Grb2 (Buday et al., 1995; Ravichandran et al., 1995). Sos has been found preferentially in complexes that also contain Shc (Pronk et al., 1994). Thus, besides simply translocating the Grb2:Sos complex, Shc may also influence the extent of Ras activation. A number of studies using dominant negative Shc proteins and mice lacking Shc expression have definitively established a role for Shc in MAPK activation (Gotoh et al., 1997; Lai and Pawson, 2000; Pratt et al., 1999; Salcini et al., 1994).

It appears that the requirement for Y317 versus Y239/Y240 in leading to Ras/MAPK activation differs between cell types and receptors. In EGF receptor signaling and IL-3 receptor signaling, the Y317F Shc appears critical for MAPK activation, while the Y239/Y240 appears to be less important or dispensable (Gotoh et al., 1997). However, we have observed that full-length Shc proteins with Y239F/240F mutation were far more potent than the Y317F as a dominant negative protein in inhibiting T-cell receptor mediated MAPK activation (Pratt et al., 1999). It is very likely that the phosphorylation on Y317 and Y239/Y240 is influenced by the tyrosine kinases that are activated by a given receptor and in turn could affect the phosphorylation on the different Shc tyrosines (Blake et al., 2000).

Role of Shc in c-Myc activation

Based on dominant negative expression of Shc proteins carrying mutation in the Y239/Y240 sites, Gotoh et al. (1996) have reported that Shc could regulate c-Myc activation in response to IL-3 stimulation in BaF cells. These authors also extended these studies with EGF signaling in NIH3T3 cells. This was the first demonstration that Shc may also have functions other than leading to MAPK activation (Gotoh et al., 1997). These studies also provided the evidence for differential downstream signaling via the Y317 versus Y239/Y240. Subsequently, others have also linked Shc to c-Myc activation (Lord et al., 1998). However, little is known about how Shc leads to c-Myc activation and what target genes are in turn affected by c-Myc. Defining this pathway may be crucial to our better understanding of signaling through Shc.

Role of Shc in survival signaling

Given the known role for Shc in Ras-dependent mitogenic signaling, it has often been difficult to distinguish between proliferation versus survival signals. However, in two instances this has become clearly evident that Shc can play a role in both types of signaling. Cattaneo and colleagues have shown that ShcA mRNA and protein levels are detectable in proliferating neuroblasts, but are progressively downmodulated in post-mitotic neurons (Cattaneo and Pelicci, 1998; Conti et al., 1997). In contrast, these post-mitotic neurons still express ShcB and ShcC proteins. This suggested that ShcA plays a role in neuronal proliferation, but that the ShcB/ShcC isoforms perhaps play a role in survival of these post-mitotic neurons. Quite interestingly, mice that are deficient in ShcB and/or ShcC have a loss of certain types of peptidergic and nociceptive neurons suggesting that the ShcB and ShcC isoforms do play a role in neuronal cell survival (Sakai et al., 2000). In tissue culture systems, a role for Shc in IL-3 dependent survival of BaF cells (Gotoh et al., 1996) and IL-2 dependent survival of T cells (Friedmann et al., 1996) have also been reported. Precisely how Shc leads to survival signals has been unclear. While the MAPK activation mediated through Ras activation itself could contribute to cell survival, Shc has also been linked to bcl-2 expression (Lord et al., 1998). Again little is known about the mechanism of Shc mediated regulation of Bcl-2 expression. Recently, a survival signal mediated via estrogen and androgen receptors that involves Shc has also been reported (Kousteni et al., 2001).

In addition to the above, the hyper-phosphorylation of Shc has been seen in many types of primary tumors and tumor cell lines (Pelicci et al., 1995). For example, Shc has been identified as one of the important molecules involved in BCR-Abl mediated transformation (Goga et al., 1995). Moreover, Shc has been found to be hyperphosphorylated in several breast cancer lines and prostate cancer cells, and in some cases the proliferation of these transformed cells could be inhibited by overexpression of dominant negative proteins (Agarwal, 2000; Biscardi et al., 1998; Dankort et al., 2001; Gresham et al., 1998; Qiu et al., 1998; Rauh et al., 1999; Stevenson and Frackelton, 1998; Stevenson et al., 1999). Whether Shc provides a proliferative or survival signal under these conditions is not known, but it was suggestive that enhanced Shc-mediated signaling was beneficial in the growth or maintenance of the tumorigenic state.

Shc and the cytoskeleton

Shc has been linked to cytoskeletal organization of cells under different contexts. The embryonic fibroblasts from ShcA knockout mice have a defect in cell spreading when plated on fibronectin (Lai and Pawson, 2000). In addition, Shc has been found to be localized to focal adhesions and can be a substrate for the focal adhesion kinase (FAK). However, other reports suggest distinct roles for Shc and FAK in focal adhesions and cell motility (Barberis et al., 2000; Gu et al., 1999). In one case, a direct binding between Shc and F-actin was seen in PC12 cells (Thomas et al., 1995). Precisely how Shc plays a role in cytoskeletal reorganization in other cells is not well understood. Given the number of SH3 domain proteins that are recruited to the focal adhesions, the PXXP motifs within the CH1 domain of Shc may also be involved in localization of Shc to focal contacts.

Lessons from the Shc knock-out mice

Lai and Pawson (2000) have provided the best evidence to date for the biological role of Shc in vivo. When exons 2 and 3 of ShcA (which encode the PTB domain) were ablated by gene targeting (ShcA^ex2/3), the expression of all three isoforms of Shc in homozygous mutants was abolished (although a smaller 40 kDa protein was expressed at very low levels). The homozygous mutant embryos died at day 11.5 with severe defects in heart development and establishment of mature blood vessels. There was a defect in angiogenesis and cell-cell contacts in the cardiovascular system. Consistent with this, Shc expression was predominantly detected in the cardiovascular system. While this clearly established a role for Shc in vivo, the embryonic lethality has not allowed analyses of Shc function in mature animals.

This study also established a role for Shc in MAPK activation in vivo. Whole mount immunostaining with phospho-specific Erk antibodies revealed a loss of MAPK activation within the cardiovascular system in ShcA^ex2/3 mutants compared to wild type embryos. Studies with the embryonic fibroblasts derived from homozygous mutants showed that at high concentration of EGF or PDGF there was no detectable difference in MAPK activation. In dose-response analyses, 50-fold and 25-fold greater concentrations of EGF or PDGF, respectively, were needed to attain MAPK phosphorylation comparable to wild type cells. These data established a critical role for Shc in MAPK activation in vivo. Also, cells from mutant mice showed significant changes in focal contact organization and actin stress fibers when plated on fibronectin-coated cover slips, suggesting a role for Shc in cytoskeletal organization.

In contrast, ShcB and ShcC knockout mice (either single or combined) are viable and fertile and show a defect in the survival of only a subset of neurons: peptidergic and nociceptive for ShcB^-/- and the loss of neurons in the superior cervical ganglia for ShcB^-/- ShcC^-/- double knockout mice (Sakai et al., 2000). The selective loss of a subset of neurons in the absence of ShcB or ShcC genes suggests that Shc-mediated signaling was required only for certain growth factor-mediated cell survival in the nervous system. p66 Shc knock-out mice have a longer life-span and this has been correlated with the role of p66 Shc in oxidative stress signaling (Migliacchio et al., 1999).

Role of Shc in 'negative' signaling?

One of the prominent proteins associated with ShcA (in hematopoietic cells) that has been seen after engagement of multiple cytokine receptors and antigen receptors is the SH2-containing inositol phosphatase, SHIP1 (Damen et al., 1996; Liu et al., 1994, 1997b). After insulin receptor signaling a complex between Shc and SHIP2, a second isoform of SHIP, has also been seen (Habib et al., 1998). Both SHIP1 and SHIP2 dephosphorylate the critical phospholipid PI(3,4,5)P3. Numerous studies from cell lines and knockout mice indicate that SHIP1 and SHIP2 are negative regulators of signaling, most likely due to their effect on PI(3,4,5)P3 (Erneux et al., 1998; Helgason et al., 1998). It is puzzling that the Shc:SHIP complex is readily detected and occurs rapidly after engagement of receptors that are mitogenic. The notion that positive signals through Shc and negative signals through SHIP would provide some type of balance of signaling/fine tuning is attractive, but it is unclear at present what role Shc would play when in a complex with SHIP. Shc could somehow remove the 'basal' inhibition due to SHIP, perhaps through sequestration or inhibition of SHIP enzymatic activity. However, we observe Shc and SHIP in similar membrane compartments and membrane microdomains and modulation of SHIP enzymatic activity through Shc binding has been very difficult to demonstrate in in vitro phosphatase assays (unpublished observations). A competition model, where SHIP would compete with Grb2 for binding to phosphorylated Shc and thereby downmodulate the positive signaling via Shc, has also been proposed (Liu et al., 1997a; Tridandapani et al., 1997). However, most of these studies were done in vitro with phosphopeptides, and we and others have not been able to reproduce these results in vivo (Harmer and DeFranco, 1999; Lamkin et al., 1997). Moreover, in Shc-deficient DT40 B cells, the FcgRIIB1-mediated inhibition of B cell signaling, which is absolutely dependent on SHIP, occurs normally (Aman et al., 2000). Precisely what role this complex plays remains to be determined.

Where do we go from here?

There are over 1300 citations on Shc (as of May 2001) describing its involvement in signaling via many different receptors. However, a number of issues related to Shc function remain unanswered. Perhaps one of the central questions that is puzzling is the Shc tyrosine phosphorylation that is detected in so many different receptor systems. There are at least two possible, not mutually exclusive, explanations. First, the signaling through Shc provides some type of a central and universal component of intracellular signaling that is being used by many different receptors. Second, the phosphorylation of Shc seen in many cases is a bystander effect (when many different kinases become activated), and Shc phosphorylation may not always be functionally relevant. I will discuss both of these possibilities below.

If the Shc tyrosine phosphorylation provides a crucial signal(s), what is/are such signals? The p52Shc mediated MAPK activation is well established and this may be one signal (Bonfini et al., 1996; Lai and Pawson, 2000). However, this is unlikely to be the sole reason for the use of Shc by many receptors, since most receptors use more than one mechanism to activate the Ras/MAPK pathway. Dominant negative Shc or loss of Shc expression often results in only a partial loss of MAPK activation. Moreover, the C. elegans and Drosophila Shc proteins do not carry the tyrosine 317 that is reported to be more favored to lead to MAPK activation in vivo (Lai et al., 1995; Luzi et al., 2000). The identification of the Y239/Y240 sites of tyrosine phosphorylation and its possible link to c-Myc activation provides an attractive candidate for that central second signal (Gotoh et al., 1996). However, this argument also has holes. c-Myc expression again can be induced by other signals. Moreover, these Y239/Y240 sites have been conserved in Drosophila, but are not seen in the worm. It is possible that the non-phosphotyrosine dependent interactions through other motifs within the CH domain might be important. Since the CH1 domain of Shc is the least conserved, this is difficult to discern simply by sequence comparisons. Further studies to identify non-phosphorylation dependent CH1 domain interacting proteins may hold some answers.

It is becoming increasingly clear that individual domains within adapter proteins do not function as independent units, but rather influence each other. Perhaps the binding of target receptors through either the PTB or SH2 or both, may cause changes to the rest of the Shc molecule that in turn recruit unique effectors that are critical. To our knowledge, a simultaneous binding of one Shc molecule via its SH2 and PTB domains to two different targets has not been documented. It is possible that the engagement of the ligand by either domain in vivo requires the second domain in a yet to be defined manner, or once one of the two domains is engaged, it may affect the second domain in some critical way. The structural organization of Shc with an N-terminal PTB and C-terminal SH2 domain is quite unique and is highly conserved through evolution (Luzi et al., 2000). Shc interacting proteins from simpler organisms may reveal some answers on Shc's evolutionarily conserved role in signaling.

The second explanation for Shc phosphorylation, seen in many receptor systems, may simply be an artifact of using too much ligand for stimulations. In other words, Shc phosphorylation in many cases might be a 'bystander' effect (since Shc is a very good substrate for many different kinases) without real consequences for signaling. For example, in fibroblasts derived from ShcA deficient embryos, there was no difference in MAPK activation seen at high concentration of growth factors and the requirement for Shc was best observed only when low concentrations of the growth factors were used (Lai and Pawson, 2000). Carpenter and colleagues had earlier observed that only a small subset of cellular proteins, including Shc, became phosphorylated when primary epithelial cells with endogenous amounts of EGFR were stimulated in limiting amounts of EGF (similar to in vivo levels) (Soler et al., 1994). Thus, several of the studies that used very high concentrations of growth factor or antigen may need to be revisited.

It is noteworthy that the defect in the Shc null mouse embryos is predominantly in the cardiovascular system, and the cells in the embryos are able to divide and develop until day 11.5. If Shc function were critical for signaling via many different receptors, embryos might have died earlier. Similarly, in Drosophila carrying a defective Shc protein, a selective defect in signaling via specific receptor tyrosine kinases was observed. It is noteworthy that there is a caveat to both of these studies. The Shc deficient mouse embryos still expressed a partial Shc protein of 40 kDa that still contained intact CH1 and SH2 domains (albeit at lower levels); similarly, there was only a point mutation in the PTB domain of the dShc that was studied, and may have only affected events that are absolutely PTB dependent. Despite this caveat, it is quite possible that at physiological concentrations of ligand, Shc would be critically needed only by a subset of receptors.

Addressing the role of Shc under in vivo conditions in different tissues would require generation of conditional knockout mice, which can then be induced to lose Shc expression in a tissue specific and temporal manner. A combination of such in vivo mouse studies along with ex vivo studies using cells lacking Shc (from particular tissues) may provide some definitive answers. Moreover, in vivo 'bypass' or 'rescue' experiments with putative downstream effectors, along with transfection studies in cell lines could be quite useful in delineating the mechanism. Similarly, expression of dominant negative forms of Shc as a transgene using tissue specific promoters may also provide some useful answers.

In summary, it is well established that the adapter protein Shc is a key intracellular signaling molecule. Our challenges ahead involve defining the downstream effectors of Shc other than the MAPK pathway, as well as determining whether Shc is universally needed or is required only for a subset of responses. A combination of genetic, structural and biochemical studies may provide some of these answers in the coming years.

Acknowledgements

I thank Hua-Poo Su, Li Zhang and other members of my laboratory for input and discussions. I apologize to many of my colleagues whose work could not be quoted due to space constraints. This work was supported by the NIH grant GM-55761.

Agarwal R. (2000). Biochem. Pharmacol. 60, 1051-1059. MEDLINE

Aman MJ, Walk SF, March ME, Su HP, Carver DJ, Ravichandran KS. (2000). Mol. Cell. Biol. 20, 3576-3589. MEDLINE

Barberis L, Wary KK, Fiucci G, Liu F, Hirsch E, Brancaccio M, Altruda F, Tarone G, Giancotti FG. (2000). J. Biol. Chem. 275, 36532-36540. MEDLINE

Biscardi JS, Belsches AP, Parsons SJ. (1998). Mol. Carcinog. 21, 261-272. Article MEDLINE

Blaikie P, Immanuel D, Wu J, Li N, Yajnik V, Margolis B. (1994). J. Biol. Chem. 269, 32031-32034. MEDLINE

Blaikie PA, Fournier E, Dilworth SM, Birnbaum D, Borg JP, Margolis B. (1997). J. Biol. Chem. 272, 20671-20677. MEDLINE

Blake RA, Broome MA, Liu X, Wu J, Gishizky M, Sun L, Courtneidge SA. (2000). Mol. Cell. Biol. 20, 9018-9027. MEDLINE

Bonfini L, Migliaccio E, Pelicci G, Lanfrancone L, Pelicci PG. (1996). Trends Biochem. Sci. 21, 257-261. Article MEDLINE

Buday L, Warne PH, Downward J. (1995). Oncogene 11, 1327-1331. MEDLINE

Burns LA, Karnitz LM, Sutor SL, Abraham RT. (1993). J. Biol. Chem. 268, 17659-17661. MEDLINE

Campbell KS, Ogris E, Burke B, Su W, Auger KR, Druker BJ, Schaffhausen BS, Roberts TM, Pallas DC. (1994). Proc. Natl. Acad. Sci. USA 91, 6344-6348. MEDLINE

Cattaneo E, Pelicci PG. (1998). Trends Neurosci. 21, 476-481. Article MEDLINE

Chen Y, Grall D, Salcini AE, Pelicci PG, Pouyssegur J, Van Obberghen-Schilling E. (1996). EMBO J. 15, 1037-1044. MEDLINE

Conti L, De Fraja C, Gulisano M, Migliaccio E, Govoni S, Cattaneo E. (1997). Proc. Natl. Acad. Sci. USA 94, 8185-8190. Article MEDLINE

Damen JE, Liu L, Cutler RL, Krystal G. (1993). Blood 82, 2296-2303. MEDLINE

Damen JE, Liu L, Rosten P, Humphries RK, Jefferson AB, Majerus PW, Krystal G. (1996). Proc. Natl. Acad. Sci. USA 93, 1689-1693. Article MEDLINE

Dankort D, Maslikowski B, Warner N, Kanno N, Kim H, Wang Z, Moran MF, Oshima RG, Cardiff RD, Muller WJ. (2001). Mol. Cell. Biol. 21, 1540-1551. MEDLINE

Dhe-Paganon S, Ottinger EA, Nolte RT, Eck MJ, Shoelson SE. (1999). Proc. Natl. Acad. Sci. USA 96, 8378-8383. MEDLINE

Eck MJ, Dhe-Paganon S, Trub T, Nolte RT, Shoelson SE. (1996). Cell 85, 695-705. MEDLINE

Erneux C, Govaerts C, Communi D, Pesesse X. (1998). Biochim. Biophys. Acta 1436, 185-199. MEDLINE

Erwin RA, Kirken RA, Malabarba MG, Farrar WL, Rui H. (1995). Endocrinology 136, 3512-3518. MEDLINE

Friedmann MC, Migone TS, Russell SM, Leonard WJ. (1996). Proc. Natl. Acad. Sci. USA 93, 2077-2082. Article MEDLINE

Gelderloos JA, Rosenkranz S, Bazenet C, Kazlauskas A. (1998). J. Biol. Chem. 273, 5908-5915. MEDLINE

Goga A, McLaughlin J, Afar DE, Saffran DC, Witte ON. (1995). Cell 82, 981-988. MEDLINE

Gotoh N, Tojo A, Shibuya M. (1996). EMBO J. 15, 6197-6204. MEDLINE

Gotoh N, Toyoda M, Shibuya M. (1997). Mol. Cell. Biol. 17, 1824-1831. MEDLINE

Gresham J, Margiotta P, Palad AJ, Somers KD, Blackmore PF, Wright Jr GL, Schellhammer PF, Wasilenko WJ. (1998). Int. J. Cancer 77, 923-927. MEDLINE

Gu J, Tamura M, Pankov R, Danen EH, Takino T, Matsumoto K, Yamada KM. (1999). J. Cell. Biol. 146, 389-403. MEDLINE

Habib T, Hejna JA, Moses RE, Decker SJ. (1998). J. Biol. Chem. 273, 18605-18609. Article MEDLINE

Harlan JE, Hajduk PJ, Yoon HS, Fesik SW. (1994). Nature 371, 168-170. MEDLINE

Harlan JE, Yoon HS, Hajduk PJ, Fesik SW. (1995). Biochemistry 34, 9859-9864. MEDLINE

Harmer SL, DeFranco AL. (1999). J. Biol. Chem. 274, 12183-12191. MEDLINE

Helgason CD, Damen JE, Rosten P, Grewal R, Sorensen P, Chappel SM, Borowski A, Jirik F, Krystal G, Humphries RK. (1998). Genes Dev. 12, 1610-1620. MEDLINE

Howell BW, Lanier LM, Frank R, Gertler FB, Cooper JA. (1999). Mol. Cell. Biol. 19, 5179-5188. MEDLINE

Kavanaugh WM, Williams LT. (1994). Science 266, 1862-1865. MEDLINE

Kousteni S, Bellido T, Plotkin LI, O'Brien CA, Bodenner DL, Han L, Han K, DiGregorio GB, Katzenellenbogen JA, Katzenellenbogen BS, Roberson PK, Weinstein RS, Jilka RL, Manolagas SC. (2001). Cell 104, 719-730. MEDLINE

Lai KM, Olivier JP, Gish GD, Henkemeyer M, McGlade J, Pawson T. (1995). Mol. Cell. Biol. 15, 4810-4818. MEDLINE

Lai KM, Pawson T. (2000). Genes Dev. 14, 1132-1145. MEDLINE

Lamkin TD, Walk SF, Liu L, Damen JE, Krystal G, Ravichandran KS. (1997). J. Biol. Chem. 272, 10396-10401. MEDLINE

Lioubin MN, Myles GM, Carlberg K, Bowtell D, Rohrschneider LR. (1994). Mol. Cell. Biol. 14, 5682-5691. MEDLINE

Liu L, Damen JE, Cutler RL, Krystal G. (1994). Mol. Cell. Biol. 14, 6926-6935. MEDLINE

Liu L, Damen JE, Hughes MR, Babic I, Jirik FR, Krystal G. (1997a). J. Biol. Chem. 272, 8983-8988. MEDLINE

Liu L, Damen JE, Ware M, Hughes M, Krystal G. (1997b). Leukemia 11, 181-184.

Lord JD, McIntosh BC, Greenberg PD, Nelson BH. (1998). J. Immunol. 161, 4627-4633. MEDLINE

Lotti LV, Lanfrancone L, Migliaccio E, Zompetta C, Pelicci G, Salcini AE, Falini B, Pelicci PG, Torrisi MR. (1996). Mol. Cell. Biol. 16, 1946-1954. MEDLINE

Luschnig S, Krauss J, Bohmann K, Desjeux I, Nusslein-Volhard C. (2000). Mol. Cell. 5, 231-241. MEDLINE

Luzi L, Confalonieri S, Di Fiore PP, Pelicci PG. (2000). Curr. Opin. Genet. Dev. 10, 668-674. MEDLINE

Margolis B. (1996). J. Lab. Clin. Med. 128, 235-241. MEDLINE

Margolis B, Borg JP, Straight S, Meyer D. (1999). Kidney Int. 56, 1230-1237. Article MEDLINE

Matsuguchi T, Salgia R, Hallek M, Eder M, Druker B, Ernst TJ, Griffin JD. (1994). J. Biol. Chem. 269, 5016-5021. MEDLINE

Migliaccio A, Di Domenico M, Castoria G, de Falco A, Bontempo P, Nola E, Auricchio F. (1996). EMBO J. 15, 1292-1300. MEDLINE

Migliaccio E, Giorgio M, Mele S, Pelicci G, Reboldi P, Pandolfi PP, Lanfrancone L, Pelicci PG. (1999). Nature 402, 309-313. Article MEDLINE

Migliaccio E, Mele S, Salcini AE, Pelicci G, Lai KM, Superti-Furga G, Pawson T, Di Fiore PP, Lanfrancone L, Pelicci PG. (1997). EMBO J. 16, 706-716. Article MEDLINE

Mikol V, Baumann G, Zurini MG, Hommel U. (1995). J. Mol. Biol. 254, 86-95. MEDLINE

Morte C, Iborra A, Martinez P. (1998). Mol. Reprod. Dev. 50, 113-120. MEDLINE

Mullane KP, Ratnofsky M, Cullere X, Schaffhausen B. (1998). Mol. Cell. Biol. 18, 7556-7564. MEDLINE

Okada S, Yamauchi K, Pessin JE. (1995). J. Biol. Chem. 270, 20737-20741. MEDLINE

Pelicci G, Dente L, De Giuseppe A, Verducci-Galletti B, Giuli S, Mele S, Vetriani C, Giorgio M, Pandolfi PP, Cesareni G, Pelicci PG. (1996). Oncogene 13, 633-641. MEDLINE

Pelicci G, Lanfrancone L, Grignani F, McGlade J, Cavallo F, Forni G, Nicoletti I, Pawson T, Pelicci PG. (1992). Cell 70, 93-104. MEDLINE

Pelicci G, Lanfrancone L, Salcini AE, Romano A, Mele S, Grazia Borrello M, Segatto O, Di Fiore PP, Pelicci PG. (1995). Oncogene 11, 899-907. MEDLINE

Pratt JC, van den Brink MR, Igras VE, Walk SF, Ravichandran KS, Burakoff SJ. (1999). J. Immunol. 163, 2586-2591. MEDLINE

Pratt JC, Weiss M, Sieff CA, Shoelson SE, Burakoff SJ, Ravichandran KS. (1996). J. Biol. Chem. 271, 12137-12140. MEDLINE

Pronk GJ, de Vries-Smits AM, Buday L, Downward J, Maassen JA, Medema RH, Bos JL. (1994). Mol. Cell. Biol. 14, 1575-1581. MEDLINE

Puil L, Liu J, Gish G, Mbamalu G, Bowtell D, Pelicci PG, Arlinghaus R, Pawson T. (1994). EMBO J. 13, 764-773. MEDLINE

Qiu Y, Ravi L, Kung HJ. (1998). Nature 393, 83-85. Article MEDLINE

Rameh LE, Arvidsson A, Carraway III KL, Couvillon AD, Rathbun G, Crompton A, VanRenterghem B, Czech MP, Ravichandran KS, Burakoff SJ, Wang DS, Chen CS, Cantley LC. (1997). J. Biol. Chem. 272, 22059-22066. Article MEDLINE

Rauh MJ, Blackmore V, Andrechek ER, Tortorice CG, Daly R, Lai VK, Pawson T, Cardiff RD, Siegel PM, Muller WJ. (1999). Mol. Cell. Biol. 19, 8169-8179. MEDLINE

Ravichandran KS, Burakoff SJ. (1994). J. Biol. Chem. 269, 1599-1602. MEDLINE

Ravichandran KS, Igras V, Shoelson SE, Fesik SW, Burakoff SJ. (1996). Proc. Natl. Acad. Sci. USA 93, 5275-5280. MEDLINE

Ravichandran KS, Lee KK, Songyang Z, Cantley LC, Burn P, Burakoff SJ. (1993). Science 262, 902-905. MEDLINE

Ravichandran KS, Lorenz U, Shoelson SE, Burakoff SJ. (1995). Ann. NY Acad. Sci. 766, 202-203. MEDLINE

Ravichandran KS, Zhou MM, Pratt JC, Harlan JE, Walk SF, Fesik SW, Burakoff SJ. (1997). Mol. Cell. Biol. 17, 5540-5549. MEDLINE

Rety S, Futterer K, Grucza RA, Munoz CM, Frazier WA, Waksman G. (1996). Protein Sci. 5, 405-413. MEDLINE

Rozakis-Adcock M, McGlade J, Mbamalu G, Pelicci G, Daly R, Li W, Batzer A, Thomas S, Brugge J, Pelicci PG. et al. (1992). Nature 360, 689-692. MEDLINE

Sakai R, Henderson JT, O'Bryan JP, Elia AJ, Saxton TM, Pawson T. (2000). Neuron 28, 819-833. MEDLINE

Salcini AE, McGlade J, Pelicci G, Nicoletti I, Pawson T, Pelicci PG. (1994). Oncogene 9, 2827-2836. MEDLINE

Sasaoka T, Rose DW, Jhun BH, Saltiel AR, Draznin B, Olefsky JM. (1994). J. Biol. Chem. 269, 13689-13694. MEDLINE

Saxton TM, van Oostveen I, Bowtell D, Aebersold R, Gold MR. (1994). J. Immunol. 153, 623-636. MEDLINE

Schmandt R, Liu SK, McGlade CJ. (1999). Oncogene 18, 1867-1879. MEDLINE

Shoelson SE. (1997). Curr. Opin. Chem. Biol. 1, 227-234. MEDLINE

Soler C, Alvarez CV, Beguinot L, Carpenter G. (1994). Oncogene 9, 2207-2215. MEDLINE

Stephens RM, Loeb DM, Copeland TD, Pawson T, Greene LA, Kaplan DR. (1994). Neuron 12, 691-705. MEDLINE

Stevenson LE, Frackelton Jr AR. (1998). Breast Cancer Res. Treat. 49, 119-128. MEDLINE

Stevenson LE, Ravichandran KS, Frackelton Jr AR. (1999). Cell Growth Differ. 10, 61-71. MEDLINE

Thomas D, Patterson SD, Bradshaw RA. (1995). J. Biol. Chem. 270, 28924-28931. MEDLINE

Touhara K, Hawes BE, van Biesen T, Lefkowitz RJ. (1995). Proc. Natl. Acad. Sci. USA 92, 9284-9287. MEDLINE

Tridandapani S, Kelley T, Cooney D, Pradhan M, Coggeshall KM. (1997). Immunol Today 18, 424-427. MEDLINE

Trub T, Choi WE, Wolf G, Ottinger E, Chen Y, Weiss M, Shoelson SE. (1995). J. Biol. Chem. 270, 18205-18208. MEDLINE

van Biesen T, Hawes BE, Luttrell DK, Krueger KM, Touhara K, Porfiri E, Sakaue M, Luttrell LM, Lefkowitz RJ. (1995). Nature 376, 781-784. MEDLINE

van der Geer P, Wiley S, Gish GD, Pawson T. (1996). Curr. Biol. 6, 1435-1444. MEDLINE

Velazquez L, Gish GD, van Der Geer P, Taylor L, Shulman J, Pawson T. (2000). Blood 96, 132-138. MEDLINE

Walk SF, March ME, Ravichandran KS. (1998). Eur. J. Immunol. 28, 2265-2275. Article MEDLINE

Weng Z, Thomas SM, Rickles RJ, Taylor JA, Brauer AW, Seidel-Dugan C, Michael WM, Dreyfuss G, Brugge JS. (1994). Mol. Cell. Biol. 14, 4509-4521. MEDLINE

Yokote K, Mori S, Hansen K, McGlade J, Pawson T, Heldin CH, Claesson-Welsh L. (1994). J. Biol. Chem. 269, 15337-15343. MEDLINE

Zhou MM, Harlan JE, Wade WS, Crosby S, Ravichandran KS, Burakoff SJ, Fesik SW. (1995a). J. Biol. Chem. 270, 31119-31123.

Zhou MM, Meadows RP, Logan TM, Yoon HS, Wade WS, Ravichandran KS, Burakoff SJ, Fesik SW. (1995b). Proc. Natl. Acad. Sci. USA 92, 7784-7788.

Zhou MM, Ravichandran KS, Olejniczak EF, Petros AM, Meadows RP, Sattler M, Harlan JE, Wade WS, Burakoff SJ, Fesik SW. (1995c). Nature 378, 584-592. MEDLINE

Zhu W, Eicher A, Leber B, Andrews DW. (1998). Oncogene 17, 565-576. MEDLINE

Figure 1 The schematic organization of ShcA isoforms is depicted. The PTB and SH2 domains of Shc bind to phosphotyrosine containing sequences, although the specificity of interaction is determined by residues N or C-terminal to the pY, respectively. Within the CH domain, three tyrosine-phosphorylation sites have been identified. The Y239/Y240 twin tyrosines have been linked to c-Myc activation; however, the mechanism by which Shc leads to c-Myc activation is not known. Y317 tyrosine has been established in leading to MAP kinase activation through Grb2 and Sos. In some cases, the Y239/Y240 can also lead to MAP kinase activation through Grb2 binding. The p46Shc lacks the first 46 amino acids within the PTB domain. The p66Shc possesses an additional CH2 domain that contains a serine phosphorylation site that has been implicated in oxidative stress signaling