Regulation of the Akt Kinase by Interacting Proteins

Regulation of the Akt kinase by interacting proteins

Keyong Du1 and Philip N Tsichlis*,1

1Molecular Oncology Research Institute, Tufts-New England Medical Center, Boston, MA 02111, USA

Ten years ago, it was observed that the Akt kinase revolutionized our understanding of cell function at is activated by phosphorylation via a phosphoinositide the molecular level.These technologies also allowed 3-kinase (PI-3K)-dependent process. This discovery gene- the identification of a large array of protein–protein rated enormous interest because it provided a link between interaction motifs that facilitate our understanding of PI-3K, an enzyme known to play a critical role in cellular signal transduction by providing clues on the potential physiology, and its downstream targets. Subsequently, it function of novel signaling proteins (Barnouin, 2004; was shown that the activity of the core components of the Miller and Stagljar, 2004). ‘PI-3K/Akt pathway’ is modulated by a complex network Akt1, also known as protein kinase Ba (PKBa) of regulatory proteins and pathways. Some of the Akt- (Bellacosa et al., 1991; Coffer and Woodgett, 1991; binding partners modulate its activation by external Jones et al., 1991), is the founding member of a protein signals by interacting with different domains of the Akt kinase family composed of three members, Akt1, Akt2, protein. This review focuses on the Akt interacting and Akt3.Akt family members regulate a diverse proteins and the mechanisms by which they regulate Akt array of cellular functions, including apoptosis, cellular activation. proliferation, differentiation, and intermediary meta- Oncogene (2005) 24, 7401–7409. doi:10.1038/sj.onc.1209099 bolism (Chan et al., 1999). Akt activation depends on PtdIns-3,4,5-P3, and to a lesser extent on PtdIns-3,4-P2, Keywords: Akt; PKB; protein–protein interaction; both of which are products of phosphoinositide 3-kinase phosphoinositide 3-kinase (PI-3K) (Franke et al., 1995). The interaction of PtdIns- 3,4,5-P3 with the Pleckstrin homology (PH) domain of Akt promotes the translocation of Akt to the plasma membrane, where it undergoes phosphorylation at Introduction two sites, one in the activation loop (Thr308) and the other in the carboxy-terminal (C-terminal) tail Intercellular and intracellular communications depend (Ser473).Phosphorylation at Ser473 appears to precede to a significant degree on the interaction between and facilitate phosphorylation at Thr308 (Sarbassov different signaling proteins.Such interactions may be et al., 2005 and references therein). The kinase that responsible for the initiation and progression of a phosphorylates Akt at Thr308 is PI-3K-dependent variety of cellular processes.For example, protein– kinase 1 (PDK1) (Chan et al., 1999). The kinase that protein interactions may induce conformational changes phosphorylates Akt at Ser473 (PDK2) had been elusive that stabilize, destabilize, or alter the enzymatic proper- for many years.Recent studies revealed that in 293T ties of the interacting partners (Goh et al., 2004). cells and several carcinoma cell lines, Akt phosphory- Protein–protein interactions may also contribute to the lation at this site depends on a complex that contains formation of large molecular complexes that allow the the mammalian target of rapamycin (mTOR), GbL, coordinate response of multiple molecules to incoming and rictor (Sarbassov et al., 2005). Earlier studies had signals (Stagljar, 2003).Finally, some proteins may shown that several additional kinases, including ILK-1, function as anchors that allow the association of their PDK1, Akt itself, and perhaps other members of the binding partners with certain cellular structures or may AGC kinase family, as well as DNA-PKcs, may have the facilitate the transport of their partners between cellular potential to phosphorylate Akt at this site (Chan and compartments (Fowler and Alarid, 2004).To accom- Tsichlis, 2001; Dong and Liu, 2005).Although the role plish these diverse outcomes, interacting proteins bind of the rictor–mTOR complex in the phosphorylation of each other either constitutively or in a signal-dependent Akt at Ser473 is beyond doubt, the potential involve- manner, and modify the activities of their interacting ment of these kinases in this process in some cells partners by providing either an adaptor or an enzymatic responding to specific signals has not been excluded function.The realization that protein–protein inter- today.Phosphorylated Akt may translocate from the actions play a critical role in cell signaling led to plasma membrane to the cytosol or the nucleus (Meier the establishment of technologies that allowed the et al., 1997; Meier and Hemmings, 1999). Activated Akt identification of networks of interacting proteins and ultimately undergoes dephosphorylation by phosphatases and returns to the inactive state (Meier et al., 1998) *Correspondence: PN Tsichlis; E-mail: [email protected] (Figure 1). Regulation of the Akt kinase K Du and PN Tsichlis 7402

External Signals.

Receptor

P P P P P P P P P P P P P P P P P Tyr376 P P P PDK1 P Thr308 P PIP3 P P Tyr373 PI3K P P P P Ser241 P Akt P Ser473 Inositol Poly phosphate phosphatases

Thr308 P Akt P Thr308 Ser473 P Akt Thr308 P Akt Ser473 Ser473 Thr308 Akt nucleus Ser473

Figure 1 Core components of the PI-3K/Akt activation pathway.Extracellular signals activate PI-3K, which in turn promotes the synthesis of D3 PPIs.Akt binds to the plasma membrane-associated D3 PPIs via its PH domain, and undergoes phosphorylation at Ser473 (by PDK2) and Thr308 (by PDK1).Phosphorylation at Ser473 precedes and facilitates phosphorylation at Thr308.Activated Akt translocates from the plasma membrane to the cytosol or to the nucleus, where it phosphorylates a wide array of downstream effector proteins.Signal termination via D3 PPI hydrolysis is mediated by several inositol polyphosphate phosphatases, including PTEN, SHIP, and SHIP2.The activation cycle ends with dephosphorylation of Akt by protein phosphatases.Akt interacting proteins modulate the activity of the PI-3K/Akt pathway by interacting with different domains of the Akt protein

Given the functional importance of Akt, the search Thr308 Ser473

for the identification of its binding partners has been Akt1 PH Kinase Domain C-Tail intense.Proteins interacting with Akt, and other TCL1 Cdc37/HSP90 CTMP kinases, include kinase substrates and kinase regulators. JIP1 TRB3 APE In this review, we will focus on the interaction of Akt Grb10 APPL1 Src with regulatory molecules.For the sake of clarity, Akt RasGAP Keratin 10 ArgBP2γ interacting proteins will be grouped based on the Akt Prohibitin2 domain they interact with (Figure 2). PDK1* Figure 2 Interacting proteins bind specifically to different domains of Akt and either promote or inhibit Akt activation. Proteins interacting with the PH domain of Akt Linear diagram of Akt illustrating the domain architecture of the protein.Akt interacting partners are listed under the domain they interact with.*The interaction of PDK1 with The Akt C-tail is The PH domain of Akt binds D3 PPIs and promotes the supported by a host of experimental data, but it has not been translocation of the kinase to the plasma membrane, a formally demonstrated process that is critical for Akt activation (Chan et al., 1999).However, the PH domain may also interact with Akt regulatory proteins.The earliest observation that encodes a 14-kDa protein whose crystal structure the PH domain of Akt functions as a protein–protein suggests that it may be involved in molecular transport, interaction motif was presented in a paper showing that a function that may be relevant to its role in the Akt oligomerizes and that oligomerization is PH regulation of the Akt kinase (Petock et al., 2002). The domain-dependent (Datta et al., 1995). Tcl1 gene is the founding member of a gene family that includes Tcl1 itself, MTCP-1, located on chromosome Xq28 (Gritti et al., 1997) and Tcl1b, located together Tcl1 and other members of the Tcl1 protein family with Tcl1, on chromosome 14q32.1 (Hallas et al., 1999; The Tcl1 oncogene on chromosome 14q32.1 is activated Pekarsky et al., 1999). The latter is represented by at by chromosomal inversion or translocation ((Inv(14) least five homologs in the mouse (Hallas et al., 1999). All (q11;q32) and t(14; 14) (q11;q32) or t(7; 14) (q35;q32)) three genes have been shown to be involved in the in human T-cell leukemia (Virgilio et al., 1994). Tcl1 induction of different forms of human T-cell leukemia

Oncogene Regulation of the Akt kinase K Du and PN Tsichlis 7403 (Teitell, 2005).Moreover, Tcl1 and MTCP1 have been of Tcl1.This illustrates that the interaction of Akt and shown to induce lymphoid tumors when overexpressed Tcl1 may have provided clues for the design of a in transgenic mice (Gritti et al., 1998; Virgilio et al., molecule that inhibits Akt by a wide range of signals. 1998). Tcl1 and Tcl1b are expressed normally in early T-cell progenitors, in pre-B, and in immature lgM- JNK interacting protein 1 (JIP1) expressing B cells (Pekarsky et al., 1999, 2004; Teitell, 2005). MTCP1 encodes 8- and 13-kDa proteins.Expres- JIP1 is a 660 amino-acid protein (707 amino acids (aa) sion of the 13-kDa isoform has been detected only in for JIP1b), which is involved in the assembly of T cells with a t(X;14) translocation (Madani et al., signaling complexes that contain JNK, MKK7, and 1996). MLK3 and is required for JNK activation (Yasuda All three Tcl1 family members interact with Akt1 and et al., 1999; Whitmarsh et al., 2001). It was recently Akt2.However, only Tcl1 interacts with all Akt shown that JIP1 also interacts specifically with the isoforms, including Akt3 (Laine et al., 2002). Tcl1 forms PH domain of Akt1.Through this interaction, Akt1 trimeric complexes and each of these complexes may sequesters JIP1 and makes it unavailable for the assem- interact with three molecules of Akt, resulting in Akt bly of JNK activation complexes.As a result, Akt1 oligomerization (French et al., 2002). Although Tcl1 inhibits JNK activation and apoptosis induced by the may promote oligomerization of Akt, it should be excitotoxin kainate in neuronal cells (Kim et al., 2002). remembered that Akt may oligomerize via PH domain Early studies had shown that the disruption of the binding in yeast, which lacks Tcl1 or other members of Akt1/JIP1 complex is not associated with a change in the Tcl1 family (Datta et al., 1995). the Akt1 phosphorylation status and that both the It is generally agreed that Tcl1-bound Akt in wild-type and the kinase-inactive Akt1 inhibit the acti- transfected cells exhibits higher stoichiometry of phos- vation of JNK by kainate.These data were challenged phorylation than unbound Akt (Laine et al., 2000; by more recent studies showing that overexpression of Pekarsky et al., 2000). This suggests, but does not prove, JIP1 enhances the catalytic activity of Akt (Kim et al., that binding to Tcl1 may promote the activation of 2003) or that JIP1 binding inhibits Akt (Song and Lee, Akt.One group has provided evidence that binding 2005).In either case, the interaction of JIP1 with Akt1 to Tcl1 indeed activates Akt both in vitro and in vivo represents a novel mechanism by which Akt may (Laine et al., 2000). Although technically convincing, the regulate apoptosis. in vitro data are somewhat difficult to interpret because they show that Tcl1-mediated Akt activation depends Growth factor receptor-binding protein-10 (Grb10) on the kinase activity of Akt and is independent of PDK1. Grb10 is a member of a superfamily of adaptor proteins. Based on information available today, we propose the Its closest relatives are Grb7 and Grb14.The proteins following plausible model regarding the role of Tcl1 in that belong to this family share a common overall struc- Akt activation.Oligomerization of Akt promoted by ture, that is characterized by a conserved N-terminal Tcl1 may facilitate Akt phosphorylation at Ser473 by proline-rich motif, a central PH domain, a C-terminal Akt itself, if one assumes that Akt can phosphorylate Src homology 2 (SH2) domain and a conserved region this site, or by another kinase such as mTOR (Sarbassov located between the PH and the SH2 domains (Between et al., 2005). This may in turn promote conformational PH and SH2 (BPS)) (Lim et al., 2004). Grb10 and Akt changes that facilitate phosphorylation at Thr308 by interact constitutively via their PH domains (Jahn et al., PDK1.This is consistent with the Akt2 structural 2002).Via its SH2 domain, Grb10 also binds phos- data (Yang et al., 2002) and with the finding that the phorylated tyrosine residues in several ligand-activated effects of Tcl1 are PI3K-dependent (Laine et al., 2000; tyrosine kinase receptors, including the insulin receptor Pekarsky et al., 2000). The notion that the interaction of (IR), the insulin-like growth factor-I receptor (IGF-IR) Tcl1 with the PH domain of Akt may abrogate the need and c-kit.The Grb10/Akt interaction contributes to Akt for D3 PPI binding is not consistent with the PI-3K activation by growth factors.This is suggested by dependence of the Tcl1 effect. several observations.First, Grb10 and c-kit act syner- In addition to activating Akt, Tcl1 also appears to gistically to activate Akt.Additionally, a constitutively promote the translocation of the activated Akt kinase active mutant of c-kit, which lacks the binding site for into the nucleus (Pekarsky et al., 2000). This suggests PI-3K (c-kitD814V/Y719F) and does not activate Akt, that Tcl1 may not only enhance the activity of the Akt can be fully complimented in Akt activation by over- kinase, but, by modulating its subcellular distribution, expression of Grb10.As a result, whereas overexpres- also affect the spectrum of Akt substrates. sion of c-kitD814V/Y719F does not render Ba/F3 cells Definition of the Tcl1 domain that interacts with the IL-3 independent, coexpression of c-kitD814V/Y719F PH domain of Akt allowed the synthesis of a peptide with wild type, but not an SH2 deletion mutant, of that binds Akt and inhibits its activation by PDGF in Grb10 does (Jahn et al., 2002). 293 cells.This peptide inhibits the in vivo growth of the fibrosarcoma cell line QRsP-11 in syngeneic C57BL/6 Ras GTPase-activating protein (RasGAP) mice (Hiromura et al., 2004). Since 293 and QRsP-11 cells do not express Tcl1, this peptide may inhibit the RasGAP, a negative regulator as well as a downstream activation of Akt by mechanisms that are independent effector of Ras (Scheffzek et al., 1997), interacts with the

Oncogene Regulation of the Akt kinase K Du and PN Tsichlis 7404 PH domain of Akt and promotes Akt activation Recent studies revealed that Hsp90 may exhibit higher by serum.The ability of RasGAP to promote Akt affinity to Hsp90 ATPase inhibitors in tumor cells than activation is independent of its GAP activity as in normal cells (Kamal et al., 2003; Vilenchik et al., overexpression of a RasGAP mutant lacking the 2004).This finding may explain the higher susceptibility GTPase-activation domain enhanced the phosphory- of tumor cells to these inhibitors (Bagatell and White- lation and basal activity of the Akt kinase and protected sell, 2004).The higher affinity of the Hsp90 inhibitors cardiomyocytes from staurosporin-induced apoptosis. toward tumor Hsp90 could be caused by an increase Knockdown of endogenous RasGAP expression, by in affinity of Cdc37 toward Hsp90 in tumor cells, which RasGAP shRNA, on the other hand, inhibited Akt in turn may be caused by tumor-specific changes in phosphorylation, despite the fact that it also upregu- the covalent state of Hsp90, or by modifications of lated RAS GTP binding (Yue et al., 2004). Given that Cdc37 or other Hsp90 co-chaperones.Structural studies RasGAP inhibits the activity of Ras, a small GTPase have shown that Cdc37 binds at the mouth of the ATP- that activates PI-3K (Chan et al., 2002), the activation binding pocket of Hsp90 (Roe et al., 2004). As a result, of Akt by RasGAP may depend on how the two Cdc37 binding may reduce the ‘off rate’ of inhibitory opposing functions of this protein are integrated in molecules bound to the pocket and may effectively different physiological settings. increase their apparent affinity toward Hsp90.One modification that may be more common in tumors than Other PH domain interacting proteins in normal cells is phosphorylation of Cdc37 at Ser13 by casein kinase II (CKII).Phosphorylation at this site has Two additional proteins have been shown to interact been shown to regulate the affinity of Cdc37 toward with the PH domain of Akt: lnosine-50 monophosphate Hsp90 (Shao et al., 2003; Miyata and Nishida, 2004). dehydrogenase (IMPDH), which interacts with the PH We speculate here that if Akt also phosphorylates domain of Akt1 (Ingley and Hemmings, 2000), and components of the Hsp90 complex, it may contribute to myosin II (Tanaka et al., 1999). The physiological signi- the observed tumor-specific changes in the properties of ficance of these interactions remains to be determined. the complex.We also suggest that modification of the complex by Akt will allow Akt to exert global effects on cell function. Proteins interacting with the kinase domain Since the Cdc37/Hsp90 complex binds proteins other Cdc37/Hsp90 than Akt, it is possible that it may facilitate the interaction of Akt with some of the Akt substrates and Heat shock protein-90 (Hsp90) is a molecular chaperone the subsequent phosphorylation of these substrates.One that facilitates protein folding and protects a variety of protein whose phosphorylation by Akt may be facili- proteins from degradation during stress (Neckers and tated by Hsp90 is endothelial nitric oxide synthase Ivy, 2003; Pearl, 2005).Hsp90 forms a complex with the (eNOS) (Fontana et al., 2002). co-chaperone Cdc37 and the complex binds a variety of cellular proteins, including tyrosine and serine–threo- TRB3: a Tribbles homolog nine protein kinases (Grammatikakis et al., 1999; Richter and Buchner, 2001; Pearl, 2005).One of these TRB3 belongs a family of at least three known proteins, kinases is Akt, which interacts with the Cdc37/Hsp90 TRB1, TRB2, and TRB3, all of which harbor a variant complex via its kinase domain (Basso et al., 2002a). kinase core motif, which lacks the ATP-binding site Disruption of this interaction with geldanamycin and (TRB3 175–182 aa; LRDLKLRR vs consensus: HRDL other ansamycin antibiotics promotes protein degrada- KPEN).As a result, TRB proteins lack kinase activity tion via the proteasome pathway, suggesting that the (Du et al., 2003). TRB proteins are similar to tribbles,a interaction of this complex with Akt protects Akt from Drosophila protein that inhibits mitosis early in deve- degradation (Basso et al., 2002a). It is interesting that, in lopment by binding to the CDC25 homolog String and tumor cell lines expressing a constitutively active by promoting String ubiquitination and proteasome- tyrosine kinase receptor, treatment with low doses of mediated degradation (Grosshans and Wieschaus, ansamycin antibiotics does not induce Akt degradation, 2000). but specifically inhibits its activity.This may be caused TRB3 (also known as NIPK) was originally cloned as by the ansamycin-induced degradation of the tyrosine a gene that is upregulated in neurons undergoing kinase receptor, which may be more sensitive to changes apoptosis induced by NGF withdrawal (Mayumi- in the stoichiometry of binding to the Cdc37/Hsp90 Matsuda et al., 1999). The interaction between TRB3 complex than Akt (Basso et al., 2002a, b). The specificity and Akt was first detected in the course of a yeast two- of the inhibitory effect of ansamycin antibiotics toward hybrid screen designed to identify Akt-binding proteins. Akt in these cell lines may be due to the sensitivity of TRB3 binds the kinase domain of Akt in the activation Akt to tyrosine kinase receptor signals.Inhibition of loop, and inhibits its phosphorylation at both Thr308 Akt by ansamycin antibiotics suppresses cell cycle and Ser473 by growth factors and insulin (Du et al., progression and promotes apoptosis.As a result, 2003).The physiological relevance of the interaction ansamycin antibiotics inhibit the growth of tumor cells between Akt and TRB3 is supported by recent findings both in culture and in animals, and represent promising showing that TRB3 expression is induced in normal chemotherapeutic agents. liver following fasting, and that TRB3 levels are

Oncogene Regulation of the Akt kinase K Du and PN Tsichlis 7405 significantly higher in the liver of db/db diabetic mice may be unique, although it is possible that they may than in the liver of wild-type mice.More important, also overlap.Thus, although mammalian TRB3 appears whereas hepatic overexpression of TRB3 in amounts to promote apoptosis and to inhibit cell growth and comparable to those in db/db mice induced glucose proliferation, the Xenopus TRB2 protein (xTRB2) intolerance and hyperglycemia (Du et al., 2003), knock- promotes cell proliferation.Moreover, xTRB2 cannot ing down TRB3 expression with RNAi in the liver substitute for dTribbles during Drosophila development of db/db mice, restored insulin responsiveness (Koo (Saka and Smith, 2004).Finally, whereas dTribbles et al., 2004). Owing to its involvement in the activation promotes proteosome-mediated degradation of its asso- of Akt by external signals, TRB3 also contributes to the ciated partners Cdc25 (Seher and Leptin, 2000) and regulation of cell growth and apoptosis.Thus, whereas C/EBP (Rorth et al., 2000), there is no evidence that overexpression of TRB3 in different cell types is mammalian TRBs are involved in protein degradation. associated with a decrease in cell size and an enhance- ment in apoptosis (Ohoka et al., 2005), downregulation Adaptor protein containing PH domain, PTB domain of its expression by RNAi is associated with the and Leucine zipper motif (APPL) inhibition of apoptosis induced by endoplasmic reticu- lum (ER) stress, growth factor withdrawal, and TNF-a APPL1, the founding member of a two-member family, (Wu et al., 2003; Ohoka et al., 2005). is a 710-aa protein that interacts with the kinase Mammalian TRB3 also interacts with several other domains of Akt1 and Akt2 and with the catalytic sub- proteins, including several members of the MAPK family unit of type 1A PI-3K (p110a) (Mitsuuchi et al., 1999). (Kiss-Toth et al., 2004) and the transcription factors Whether the second member of the family (APPL2) ATF4 (Bowers et al., 2003; Ord and Ord, 2003), CHOP interacts with Akt is currently unknown.APPL1 was (Ohoka et al., 2005), and NF-kB, while dTribbles also mapped to human chromosome 3p4.1–21.1 and is highly interacts with dC/EBP (Rorth et al., 2000). Binding of expressed in skeletal muscle, heart, ovary, and pancreas. TRB3 to these proteins generally inhibits their acti- As indicated by its name, it contains a PH domain, a vities.Thus, TRB3 inhibits the transcriptional activities PTB domain and a putative leucine zipper domain.Its of ATF4 and CHOP (Ohoka et al., 2005), as well as interaction with Akt2 is mediated by the PTB domain the PKA-induced transcriptional activity of NF-kB (Mitsuuchi et al., 1999; Yang et al., 2003). Recent (Wu et al., 2003). Regarding MAPKs, TRB3 may either studies suggested that overexpression of APPL enhances upregulate or downregulate their activities, depending on Akt activation by IGF1 and promotes Akt-mediated its expression levels (Kiss-Toth et al., 2004). suppression of androgen receptor transactivation (Yang TRB3 expression is induced by various stimuli, et al., 2003). The importance of APPL in the physio- including hypoxia (Bowers et al., 2003), TNF-a,and logical regulation of Akt remains to be determined. reagents that induce ER stress.In addition, it is induced APPL1 also interacts with the small GTPase Rab5, in the liver, in the course of pathological syndromes, such a key regulator of endocytosis, and with a deacetylase as fatty degeneration (Klingenspor et al., 1999) and multiprotein complex NuRD/MeCP1, which regulates diabetes (Du et al., 2003). In response to ER stress, ATF4 chromatin structure and gene expression (Miaczynska and CHOP bind to the TRB3 promoter and activate et al., 2004). APPL1 is localized in a subpopulation of TRB3 gene expression (Bowers et al.,2003;Ohokaet al., endosomes (Miaczynska et al., 2004). In response to 2005).In fatty and diabetic liver, upregulation of TRB3 is stimuli such as EGF and oxidative stress, APPL1 asso- mediated through PPARa.More recently, TRB3 expres- ciated with endosomes that contain both Rab5 and sion in liver was also linked to PGC-1, a coactivator for APPL1 translocates into the nucleus, where it interacts PPARa and PPARg.PGC-1 overexpression induced with NuRD/MeCP1.As a result, APPL1 is essential for TRB3 and hyperglycemia.Knocking down TRB3 by cell growth in response to EGF.Finally, APPL1 (also RNAi in PGC1-overexpressing livers partially inhibited known as DIP13a) also interacts with a region in the hyperglycemia, suggesting that the induction of hyper- cytoplasmic domain of Deletion in Colon Cancer (DCC) glycemia by PGC1- is TRB3-dependent.These data and promotes DCC-induced apoptosis (Liu et al., 2002). suggest that, by interfering with the activation of Akt, How the interaction of Akt with APPL1 modulates the TRB3 may contribute to insulin resistance in individuals effects of APPL1 on membrane trafficking, cell growth, with type II diabetes (Koo et al.,2004). gene expression, proliferation and apoptosis is currently As mentioned above, TRB3 belongs to a protein unknown. family that consists of three members.This raises several questions that remain to be addressed: (a) Do different Proteins interacting with the C-terminal tail members of the family interact with each other, and if they do, what is the biological significance of the C-terminal modulator protein (CTMP) resulting homodimers and heterodimers? (b) Is the interaction with Akt a unique property of TRB3, or The human CTMP gene encodes a 240-aa protein that is a common property of all the members of the family? expressed in skeletal muscle, testis uterus, brain and Another question that remains to be addressed is kidney with lower levels in heart liver and lung.In whether TRB proteins in different species are function- starved cells, CTMP is nonphosphorylated and is ally equivalent.Evidence to date favors the hypothesis associated with the plasma membrane.CTMP, in its that the functions of these proteins in different species nonphosphorylated membrane-associated form, binds

Oncogene Regulation of the Akt kinase K Du and PN Tsichlis 7406 the C-terminal tail of Akt.Bound Akt, colocalizing with kinase Src.Bound Src may phosphorylate Akt at Tyr315 CTMP at the plasma membrane, is inactive.Exposure and may promote its activation by external signals to external signals promotes the phosphorylation of (Chen et al., 2001). CTMP and the release and activation of Akt (Maira Although early studies failed to show that Akt under- et al., 2001). However, it is not known whether the pool goes tyrosine phosphorylation, more recent studies sug- of Akt undergoing activation is the same pool that is gested that it does undergo phosphorylation on tyrosine bound to CTMP prior to stimulation. in response to EGF stimulation.EGF-induced tyrosine Although CTMP clearly inhibits the activity of the phosphorylation of Akt is likely to be mediated by Src it Akt kinase, its physiological role in regulating Akt was singificantly reduced in SYF cells lacking Src, Yes and activation remains to be determined.It is generally Fyn, and a P424A/P427A double Akt mutant which does thought that, in unstimulated cells, the majority of Akt not bind Src, failed to undergo tyrosine phsphorylation is localized in the cytosol, where it is inactive, despite the upon stimulation with EGF.The same mutant failed to fact that it is not bound to CTMP.Translocation to the undergo activation in response to EGF, suggesting that membrane occurs following stimulation and it is phosphorylation by Src plays an important role in Akt associated with Akt phosphorylation and activation.It activation by external signals (Jiang and Qiu, 2003). is possible that CTMP assembles a pool of catalytically The other SH3 domain-containing protein that inactive Akt at the plasma membrane and that this pool interacts with Akt is ArgBP2g, an adaptor protein that may be transported rapidly to membrane microdo- belongs to a four-member protein family (Kioka et al., mains, where activation takes place following stimula- 2002).ArgBP2 ( a and b) was first identified as a protein tion.If CTMP indeed assembles a rapidly accessible Akt that interacts with the tyrosine kinases Arg and Abl pool, one would expect that CTMP would be required (Wang et al., 1997). ArgBP2g contains a SoHo (Sorbin for Akt activation by physiological signals.However, Homology) motif and three SH3 domains (Kioka et al., this remains to be determined. 2002).Through its SoHo motif, it binds to spectrin and may therefore regulate the actin cytoskeleton (Cestra Akt phosphorylation enhancer (APE), or Hook-related et al., 2005). protein-1 ArgBP2g binds the proline-rich C-terminal motifs of Akt via its first and second SH3 domains.In addition, APE, also known as Hook-related protein-1 (HKRP1), is it binds, the p21-activated protein kinase PAK1, via its a 1847-aa protein with a coiled-coil domain and a putative first and third SH3 domains.As a result, it may contri- microtubule-binding domain, which is believed to be bute to the formation of Akt–ArgBPg–PAK1 complexes. involved in endosome trafficking (Simpson et al., 2005). Although the interaction between ArgBP2g and Akt does APE specifically associates with the C-terminus of not affect the Akt kinase activity (Yuan et al., 2005), it Akt (Anai et al., 2005). Binding to APE enhances the may contribute to in the regulation of PAK1 by Akt. basal phosphorylation of Akt.However, APE-induced ArgBP2g contains four Akt phosphorylation consensus phosphorylation of Akt requires PI-3K, as wortmannin sites that can be phosphorylated by Akt both in vitro or LY294002 inhibits APE-induced Akt activation.In and in vivo.Phosphorylation of ArgBP2 g at these sites addition, APE enhances and/or prolongs the PI-3K- may contribute to the activation of PAK1 by Akt.Thus, dependent phosphorylation of Akt by external signals the nonphosphorylatable ArgBP2g-4A mutant and a (Anai et al., 2005). However, the molecular mechanism truncated mutant of ArgBP2g missing one of its SH3 by which APE enhances the activity of Akt is not clear. domains inhibit the activation of PAK1 by Akt.The APE behaves as a DNA replication licensing factor. same mutants inhibit the phosphorylation of Bad by Akt In cells in which APE was knocked down by RNAi, Akt and PAK1, and reduce the Akt and PAK1 antiapoptotic phosphorylation and DNA synthesis were both signifi- activities.Collectively, these findings indicate that the cantly reduced.In cells overexpressing both Akt and binding of ArgBP2g to both Akt and PAK1 and the APE, DNA replication was initiated and it was followed phosphorylation of ArgBP2g by Akt are required for by Chk2 activation.DNA rereplication and Chk2 the activation of PAK1 by Akt and for the integrity of activation promoted apoptosis, despite the APE- the Akt/PAK1 survival pathway (Yuan et al., 2005). induced marked increase in the basal phosphorylation et al of Akt (Anai ., 2005). The potential involvement Prohibitin 2 (PHB2) of Akt in the licensing of DNA replication by APE has not yet been determined.In addition, it is not yet clear Prohibitin 2, a protein that belongs to a two-member whether the activation of Akt by APE has the potential protein family, was originally identified as a 37-kDa to alter the proapoptotic effects of the APE protein. protein that is associated with the intracellular domain of the mlgM class of antigen receptors in B cells SH3 domain-containing proteins, Src and Arg-binding (BAP37) (Terashima et al., 1994). Prohibitin 2 is a protein 2g (ArgBP2g) highly conserved, ubiquitously expressed protein with homologs in bacteria, yeast, plants, Drosophila, and In the C-terminus of Akt, there is a proline-rich region mammals (Nijtmans et al., 2002). Prohibitins 1 and 2 are that harbors two PXXP motifs.Through these motifs, both components of a large complex that is associated Akt can bind SH3 domain-containing proteins.The first with the mitochondrial inner membrane (Nijtmans et al., such protein shown to interact with Akt is the tyrosine 2000).Moreover, prohibitin interacts with Rb and E2F,

Oncogene Regulation of the Akt kinase K Du and PN Tsichlis 7407 represses the transcriptional activity of E2F and inhibits epithelial cells.Keratin-10 is expressed in postmitotic cellular proliferation (Wang et al., 1999, 2002). differentiating keratinocytes in vivo and its expression Akt interacts with prohibitin 2 (120–232 aa) via its is severely reduced in proliferating cells, including C-terminal tail (413–480 aa).The interaction of the two neoplastic cells derived from cutaneous tumors (Siegel proteins was originally detected in a yeast two-hybrid and Howard, 2002).Keratin-10 binds PKCzeta and screen in which the bait was full-length Akt (Sun et al., Akt via its N-terminus (Paramio et al., 2001). However, 2004).Prohibitin also interacts with MyoD, and MEF2, the domain of Akt that interacts with keratin-10 is and represses their transcriptional activities.The bind- unknown. ing of Akt to prohibitin relieves the prohibitin-mediated Binding to keratin-10 sequesters Akt to the cytoske- repression of chromatin-associated transcriptional regu- leton and inhibits its translocation to the plasma lators.Thus, Akt promotes myogenic differentiation, membrane.As a result, the binding of keratin-10 to perhaps by relieving the prohibitin-mediated repression Akt inhibits Akt activation.Inhibition of Akt by of MyoD and MEF2 (Sun et al., 2004). keratin-10 interferes with the induction of cyclin D1 via E2F.Targeting a human keratin-10 transgene via the 3-Phosphoinositde dependent protein kinase-1 (PDK1) bovine keratin-5 promoter to the basal layer of the epidermis in mice resulted in impaired activation of Akt As mentioned in the introduction, phosphorylation of and in decreased proliferation and hypoplasia of the Akt at Ser473 appears to precede phosphorylation by epidermis (Santos et al., 2002). PDK1 at Thr308 (Sarbassov et al., 2005, and references therein).More important, the S473D mutant of Akt and Intramolecular Akt interactions Akt phosphorylated at Ser473 by the rictor–mTOR complex are better targets of PDK1 than nonpho- The crystal structure of Akt2 (Yang et al., 2002; Huang sphorylated Akt (Sarbassov et al., 2005). These ﬁndings et al., 2003) suggests a model that addresses the suggest that phosphorylation at Ser473 may provide a molecular mechanism by which phosphorylation pro- docking site for PDK1. motes the activation of Akt.According to this model, The interaction of PDK1 with the phosphorylated phosphorylation of Akt2 at Ser474 (equivalent to Ser473 hydrophobic motif of Akt is supported by a host of in Akt1) promotes the interaction of the hydrophobic earlier observations.Thr308 and Ser473 phosphorylation motif with the N-terminal lobe of the kinase and the sites are separated by 165 aa.The arrangement of two transition of the aC helix from a disordered to an phosphorylation sites, one in the activation loop and the ordered state (Yang et al., 2002). This transition, in turn, other in a hydrophobic motif in the kinase tail, is a facilitates the interaction of the aC helix with the common feature of the AGC kinase family, to which Akt phosphorylated Thr309 moiety (equivalent to Thr308 in belongs (Peterson and Schreiber, 1999).Some of these Akt1), and the restructuring of the activation loop of the kinases, including Protein Kinase C-related Kinase-2 kinase (Huang et al., 2003). Based on this model, (PRK2), contain a hydrophobic motif in which the hydrophobic motif phosphorylation should be critical phosphorylatable moiety has been substituted with for Akt activation.Experimental data however, indicate aspartic acid.PRK2 interacts via this motif with PDK1 that although hydrophobic motif phosphorylation con- (Balendran et al., 1999, 2000). A peptide that spans the tributes to, it is not required for Akt activation phosphomimetically altered hydrophobic motif of PRK2 (Bellacosa et al., 1998). One potential explanation for continues to bind PDK1 and it is known as the PDK1- this discrepancy is that Akt may interact with the interacting fragment (PIF) (Balendran et al., 1999). The phosphorylated or phosphomimetically altered hydro- PDK1 motif that binds PIF is known as the PIF-binding phobic motifs of other AGC kinases and that such pocket (Biondi et al., 2001; Collins et al., 2003). More interactions may be functionally equivalent with the recent studies demonstrated that, through this motif, intramolecular interaction between the hydrophobic PDK1 also binds AGC kinases other than PRK2, motif and the N-terminal lobe of the Akt kinase following phosphorylation of their hydrophobic motifs. described above.However, this remains to be deter- For two such kinases, p70S6K and SGK, phosphory- mined. lation at this site appears to be essential for activation loop phosphorylation (Collins et al., 2003; Mora et al., 2004).For Akt it is not essential, although it facilitates Summary Thr308 phosphorylation by PDK1 (Sarbassov et al., 2005, and references therein), suggesting that it may promote This review focuses on proteins that interact with Akt the Akt–PDK1 interaction (Balendran et al., 1999). and may modulate its activity.The Akt interacting proteins were grouped based on the Akt domain they interact with.For some of these proteins, their Proteins for which the Akt domain they interact with has role in Akt activation by physiological signals is not been determined clearly established, while for others it remains tentative. Keratin-10 The interaction of Akt with two of these proteins (Cdc37/Hsp90 and Tcl1) led to the discovery of Akt Keratin-10 is a component of a intermediate ﬁlament inhibitors, which are currently at various stages of cytoskeleton, which imparts mechanical integrity to development.

Oncogene Regulation of the Akt kinase K Du and PN Tsichlis 7408 Acknowledgements laboratory has been supported by PHS grant R01 CA57436. We thank Tung On Chan (Thomas Jefferson University) for KD is a recipient of Thomas R Lee Career Development helpful discussions.Work on PI-3K and Akt in the PNT Award from the American Diabetes Association.

References

Anai M, Shojima N, Katagiri H, Ogihara T, Sakoda H, Onishi Glazer RI.(1998). Curr. Pharm. Des., 4, 277–290. Y, Ono H, Fujishiro M, Fukushima Y, Horike N, Viana A, Goh CS, Milburn D and Gerstein M.(2004). Curr. Opin. Kikuchi M, Noguchi N, Takahashi S, Takata K, Oka Y, Struct. Biol., 14, 104–109. Uchijima Y, Kurihara H and Asano T.(2005). J. Biol. Grammatikakis N, Lin JH, Grammatikakis A, Tsichlis PN Chem., 280, 18525–18535. and Cochran BH.(1999). Mol. Cell Biol., 19, 1661–1672. Bagatell R and Whitesell L.(2004). Mol. Cancer Ther., 3, Gritti C, Choukroun V, Soulier J, Madani A, Dastot H, 1021–1030. Leblond V, Radford-Weiss I, Valensi F, Varet B, Sigaux F Balendran A, Biondi RM, Cheung PC, Casamayor A, Deak M and Stern MH.(1997). Oncogene, 15, 1329. and Alessi DR.(2000). J. Biol. Chem., 275, 20806–20813. Gritti C, Dastot H, Soulier J, Janin A, Daniel MT, Madani A, Balendran A, Casamayor A, Deak M, Paterson A, Gaffney P, Grimber G, Briand P, Sigaux F and Stern MH.(1998). Currie R, Downes CP and Alessi DR.(1999). Curr. Biol., 9, Blood, 92, 368–373. 393–404. Grosshans J and Wieschaus E.(2000). Cell, 101, 523–531. Barnouin K.(2004). Methods Mol. Biol., 261, 479–498. Hallas C, Pekarsky Y, Itoyama T, Varnum J, Bichi R, Basso AD, Solit DB, Chiosis G, Giri B, Tsichlis P and Rosen Rothstein JL and Croce CM.(1999). Proc. Natl. Acad. Sci. N.(2002a). J. Biol. Chem., 277, 39858–39866. USA, 96, 14418–14423. Basso AD, Solit DB, Munster PN and Rosen N.(2002b). Hiromura M, Okada F, Obata T, Auguin D, Shibata T, Oncogene, 21, 1159–1166. Roumestand C and Noguchi M.(2004). J. Biol. Chem., 279, Bellacosa A, Chan TO, Ahmed NN, Datta K, Malstrom S, 53407–53418. Stokoe D, McCormick F, Feng J and Tsichlis P.(1998). Huang X, Begley M, Morgenstern KA, Gu Y, Rose P, Zhao H Oncogene, 17, 313–325. and Zhu X.(2003). Structure (Camb.), 11, 21–30. Bellacosa A, Testa JR, Staal SP and Tsichlis PN.(1991). Ingley E and Hemmings BA.(2000). FEBS Lett., 478, Science, 254, 274–277. 253–259. Biondi RM, Kieloch A, Currie RA, Deak M and Alessi DR. Jahn T, Seipel P, Urschel S, Peschel C and Duyster J.(2002). (2001). EMBO J., 20, 4380–4390. Mol. Cell. Biol., 22, 979–991. Bowers AJ, Scully S and Boylan JF.(2003). Oncogene, 22, Jiang T and Qiu Y.(2003). J. Biol. Chem., 278, 15789–15793. 2823–2835. Jones PF, Jakubowicz T, Pitossi FJ, Maurer F and Hemmings Cestra G, Toomre D, Chang S and De Camilli P.(2005). Proc. BA.(1991). Proc. Natl. Acad. Sci. USA, 88, 4171–4175. Natl. Acad. Sci. U.S.A., 102, 1731–1736. Kamal A, Thao L, Sensintaffar J, Zhang L, Boehm MF, Fritz Chan TO, Rittenhouse SE and Tsichlis PN.(1999). Annu. Rev. LC and Burrows FJ.(2003). Nature, 425, 407–410. Biochem., 68, 965–1014. Kim AH, Sasaki T and Chao MV.(2003). J. Biol. Chem., 278, Chan TO, Rodeck U, Chan AM, Kimmelman AC, Ritten- 29830–29836. house SE, Panayotou G and Tsichlis PN.(2002). Cancer Kim AH, Yano H, Cho H, Meyer D, Monks B, Margolis Cell, 1, 181–191. B, Birnbaum MJ and Chao MV.(2002). Neuron, 35, Chan TO and Tsichlis PN.(2001). Sci. STKE, 2001, PE1. 697–709. Chen R, Kim O, Yang J, Sato K, Eisenmann KM, McCarthy Kioka N, Ueda K and Amachi T.(2002). Cell Struct. Funct., J, Chen H and Qiu Y.(2001). J. Biol. Chem., 276, 27, 1–7. 31858–31862. Kiss-Toth E, Bagstaff SM, Sung HY, Jozsa V, Dempsey C, Coffer PJ and Woodgett JR.(1991). Eur. J. Biochem., 201, Caunt JC, Oxley KM, Wyllie DH, Polgar T, Harte M, 475–481. O’Neill L A, Qwarnstrom EE and Dower SK.(2004). J. Biol. Collins BJ, Deak M, Arthur JS, Armit LJ and Alessi DR. Chem., 279, 42703–42708. (2003). EMBO J., 22, 4202–4211. Klingenspor M, Xu P, Cohen RD, Welch C and Reue K. Datta K, Franke TF, Chan TO, Makris A, Yang SI, Kaplan (1999). J. Biol. Chem., 274, 23078–23084. DR, Morrison DK, Golemis EA and Tsichlis PN.(1995). Koo SH, Satoh H, Herzig S, Lee CH, Hedrick S, Kulkarni R, Mol. Cell. Biol., 15, 2304–2310. Evans RM, Olefsky J and Montminy M.(2004). Nat. Med., Dong LQ and Liu F.(2005). Am. J. Physiol. Endocrinol. 10, 530–534. Metab., 289, E187–E196. Laine J, Kunstle G, Obata T and Noguchi M.(2002). J. Biol. Du K, Herzig S, Kulkarni RN and Montminy M.(2003). Chem., 277, 3743–3751. Science, 300, 1574–1577. Laine J, Kunstle G, Obata T, Sha M and Noguchi M.(2000). Fontana J, Fulton D, Chen Y, Fairchild TA, McCabe TJ, Mol. Cell, 6, 395–407. Fujita N, Tsuruo T and Sessa WC.(2002). Circ. Res., 90, Lim MA, Riedel H and Liu F.(2004). Front Biosci., 9, 866–873. 387–403. Fowler AM and Alarid ET.(2004). Sci. STKE, 2004, tr11. Liu J, Yao F, Wu R, Morgan M, Thorburn A, Finley Jr RL Franke TF, Yang SI, Chan TO, Datta K, Kazlauskas A, and Chen YQ.(2002). J. Biol. Chem., 277, 26281–26285. Morrison DK, Kaplan DR and Tsichlis PN.(1995). Cell, 81, Madani A, Choukroun V, Soulier J, Cacheux V, Claisse JF, 727–736. Valensi F, Daliphard S, Cazin B, Levy V, Leblond V, Daniel French SW, Shen RR, Koh PJ, Malone CS, Mallick P and MT, Sigaux F and Stern MH.(1996). Blood, 87, 1923–1927. Teitell MA.(2002). Biochemistry, 41, 6376–6382. Maira SM, Galetic I, Brazil DP, Kaech S, Ingley E, Thelen M Fukushima K, Miyamoto S, Komatsu H, Tsukimori K, and Hemmings BA.(2001). Science, 294, 374–380. Kobayashi H, Seki H, Takeda S and Nakano H.(2003). Mayumi-Matsuda K, Kojima S, Suzuki H and Sakata T. Biol. Reprod., 68, 1771–1778. (1999). Biochem. Biophys. Res. Commun., 258, 260–264.

Oncogene Regulation of the Akt kinase K Du and PN Tsichlis 7409 Meier R, Alessi DR, Cron P, Andjelkovic M and Hemmings Scheffzek K, Ahmadian MR, Kabsch W, Wiesmuller L, BA.(1997). J. Biol. Chem., 272, 30491–30497. Lautwein A, Schmitz F and Wittinghofer A.(1997). Science, Meier R and Hemmings BA.(1999). J. Recept. Signal 277, 333–338. Transduct. Res., 19, 121–128. Seher TC and Leptin M.(2000). Curr. Biol., 10, 623–629. Meier R, Thelen M and Hemmings BA.(1998). EMBO J., 17, Shao J, Prince T, Hartson SD and Matts RL.(2003). J. Biol. 7294–7303. Chem., 278, 38117–38120. Miaczynska M, Christoforidis S, Giner A, Shevchenko A, Siegel DH and Howard R.(2002). Curr. Opin. Pediatr., 14, Uttenweiler-Joseph S, Habermann B, Wilm M, Parton RG 419–425. and Zerial M.(2004). Cell, 116, 445–456. Simpson F, Martin S, Evans TM, Kerr M, James DE, Miller J and Stagljar I.(2004). Methods Mol. Biol., 261, Parton RG, Teasdale RD and Wicking C.(2005). Trafﬁc, 247–262. 6, 442–458. Mitsuuchi Y, Johnson SW, Sonoda G, Tanno S, Golemis EA Song JJ and Lee YJ.(2005). J. Biol. Chem., 280, 26845–26855. and Testa JR.(1999). Oncogene, 18, 4891–4898. Stagljar I.(2003). Sci. STKE, 2003, pe56. Miyata Y and Nishida E.(2004). Mol. Cell. Biol., 24, Sun L, Liu L, Yang XJ and Wu Z.(2004). J. Cell Sci., 117, 4065–4074. 3021–3029. Mora A, Komander D, van Aalten DM and Alessi DR. Tanaka M, Konishi H, Touhara K, Sakane F, Hirata M, Ono (2004). Semin. Cell Dev. Biol., 15, 161–170. Y and Kikkawa U.(1999). Biochem. Biophys. Res. Commun., Neckers L and Ivy SP.(2003). Curr. Opin. Oncol., 15, 419–424. 255, 169–174. Nijtmans LG, Artal SM, Grivell LA and Coates PJ.(2002). Teitell MA.(2005). Nat. Rev. Cancer, 5, 640–648. Cell Mol. Life Sci., 59, 143–155. Vilenchik M, Solit D, Basso A, Huezo H, Lucas B, He H, Nijtmans LG, de Jong L, Artal Sanz M, Coates PJ, Berden JA, Rosen N, Spampinato C, Modrich P and Chiosis G.(2004). Back JW, Muijsers AO, van der Spek H and Grivell LA. Chem. Biol., 11, 787–797. (2000). EMBO J., 19, 2444–2451. Virgilio L, Lazzeri C, Bichi R, Nibu K, Narducci MG, Russo Ohoka N, Yoshii S, Hattori T, Onozaki K and Hayashi H. G, Rothstein JL and Croce CM.(1998). Proc. Natl. Acad. (2005). EMBO J., 24, 1243–1255. Sci. USA, 95, 3885–3889. Ord D and Ord T.(2003). Exp. Cell Res., 286, 308–320. Virgilio L, Narducci MG, Isobe M, Billips LG, Cooper MD, Paramio JM, Segrelles C, Ruiz S and Jorcano JL.(2001). Mol. Croce CM and Russo G.(1994). Proc. Natl. Acad. Sci. USA, Cell. Biol., 21, 7449–7459. 91, 12530–12534. Pearl LH.(2005). Curr. Opin. Genet. Dev., 15, 55–61. Wang B, Golemis EA and Kruh GD.(1997). J. Biol. Chem., Pekarsky Y, Hallas C, Isobe M, Russo G and Croce CM. 272, 17542–17550. (1999). Proc. Natl. Acad. Sci. USA, 96, 2949–2951. Wang S, Nath N, Fusaro G and Chellappan S.(1999). Mol. Pekarsky Y, Koval A, Hallas C, Bichi R, Tresini M, Malstrom Cell. Biol., 19, 7447–7460. S, Russo G, Tsichlis P and Croce CM.(2000). Proc. Natl. Wang S, Zhang B and Faller DV.(2002). EMBO J., 21, Acad. Sci. USA, 97, 3028–3033. 3019–3028. Pekarsky Y, Zanesi N, Aqeilan R and Croce CM.(2004). Whitmarsh AJ, Kuan CY, Kennedy NJ, Kelkar N, Hematol. Oncol. Clin. North Am., 18, 863–879 ix. Haydar TF, Mordes JP, Appel M, Rossini AA, Jones SN, Peterson RT and Schreiber SL.(1999). Curr. Biol., 9, Flavell RA, Rakic P and Davis RJ.(2001). Genes Dev., 15, R521–R524. 2421–2432. Petock JM, Torshin IY, Wang YF, Du Bois GC, Croce Wu M, Xu LG, Zhai Z and Shu HB.(2003). J. Biol. Chem., CM, Harrison RW and Weber IT.(2002). Sci. World J., 2, 278, 27072–27079. 1876–1884. Xi W, Satoh H, Kase H, Suzuki K and Hattori Y.(2005). Richter K and Buchner J.(2001). J. Cell. Physiol., 188, Biochem. Biophys. Res. Commun., 332, 200–205. 281–290. Yang J, Cron P, Thompson V, Good VM, Hess D, Hemmings Roe SM, Ali MM, Meyer P, Vaughan CK, Panaretou B, BA and Barford D.(2002). Mol. Cell, 9, 1227–1240. Piper PW, Prodromou C and Pearl LH.(2004). Cell, 116, Yang L, Lin HK, Altuwaijri S, Xie S, Wang L and Chang C. 87–98. (2003). J. Biol. Chem., 278, 16820–16827. Rorth P, Szabo K and Texido G.(2000). Mol. Cell, 6, Yasuda J, Whitmarsh AJ, Cavanagh J, Sharma M and Davis 23–30. RJ.(1999). Mol. Cell Biol., 19, 7245–7254. Saka Y and Smith JC.(2004). Dev. Biol., 273, 210–225. Yuan ZQ, Kim D, Kaneko S, Sussman M, Bokoch GM, Kruh Santos M, Paramio JM, Bravo A, Ramirez A and Jorcano JL. GD, Nicosia SV, Testa JR and Cheng JQ.(2005). J. Biol. (2002). J. Biol. Chem., 277, 19122–19130. Chem., 280, 21483–21490. Sarbassov DD, Guertin DA, Ali SM and Sabatini DM.(2005). Yue Y, Lypowy J, Hedhli N and Abdellatif M.(2004). J. Biol. Science, 307, 1098–1101. Chem., 279, 12883–12889.

Oncogene