Twice upon a time: PI3K's secret double life exposed

Twice upon a time: PI3K's secret double life exposed
of 5
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
  Twice upon a time: PI3K’s secretdouble life exposed Emilio Hirsch, Laura Braccini, Elisa Ciraolo, Fulvio Morello and Alessia Perino Molecular Biotechnology Center, Via Nizza 52, 10126 Torino, Italy Class I phosphoinositide 3-kinases (PI3Ks) are heterodi-meric enzymes involved in signal transduction triggeredby growth factors and G-protein-coupled receptors. Thecatalytic function of PI3Ks is well known to promote awide variety of biological processes, including prolifer-ation, survival and migration, but a new layer of com-plexity in the function of PI3Ks has recently emerged,indicating that these proteins function not only askinases but also as scaffold proteins. Knockout micethat lack PI3K protein expression show a different phe-notypefromknock-inmiceexpressingPI3Kmutantsthathavelosttheirkinaseactivity,providingevidenceforthisnovel role of PI3Ks. We will discuss such findings, high-lighting the crucial scaffold function of PI3K g  in cAMPhomeostasis and PI3K b  in receptor recycling.Setting the stage The enzymatic activity of phosphoinositide 3-kinases(PI3Ks) is essential in eukaryotic cells to regulate many processes, such as cytoskeletal dynamics, transcription,protein synthesis, metabolic responses and membranetrafficking  [1 – 4]. PI3Ks are recruited and activated down-stream of tyrosine kinase receptors (RTKs) and G-protein-coupled receptors (GPCRs) and phosphorylate phospho-inositides (PIs) in their D3 position. Three differentPI3K classes can be distinguished based on theirsubstratespecificityandmolecularstructure(Table1).ClassIPI3Kstransduce responses to agonists by regulating the cellularlevels of phosphoinositide (3,4,5)-trisphosphate (PIP 3 ),whereas class II and III PI3Ks use PI as a substrate toproduce phosphoinositide 3-phosphate (PI3P) [5].In class I PI3Ks, the lipid kinase activity is crucially controlled by protein – protein interactions. These PI3Ksfunction as obligate heterodimers composed of a catalyticand an adaptor/regulatory subunit * . Mammals expressfour catalytic subunits (p110 a , - b , - g  and - d ) that sharestructural homology and that each contain an adaptor-binding domain, a Ras-binding domain (RBD), a C2domain potentially involved in membrane attachment, ahelical/accessory domain (PIK) and a kinase domain (cat-alytic domain). Each catalytic p110 subunit binds directly to a regulatory subunit (where p85 a , p85 b , p55 a , p55 g  orp50 a  can complex with either p110 a , - b  or - d ; and wherep101 or p84/87 y can complex with p110 g ) (Table 1) [5].  Associationofthecatalyticsubunittoanadaptor/regulatormodifies its subcellular localization and enzymaticactivity. For instance, binding of p110 to p85 inhibits itscatalytic activity in the cytoplasm. Conversely, through itsSH2 domain, the p85 isoforms tether the complex to acti- vated RTKs. This interaction not only releases the inhi-bition on p110 but also localizes this subunit next to itslipid substrates at the membrane [5]. It has been recently reported that a novel protein, PI3K-interacting protein 1(PIK3IP1), interacts with p110 a  and - b , acting as anotherregulator of p110 function [6]. Furthermore, p110 activity is modulated by its association with Ras, which interactswiththeRBDdomain.TheselectivebindingofGTP – Rastop110 a , - g  or - d  is thought to trigger a positive-feedback loop, generating a synergy between Ras and PI3K sig-naling  [7].Most of the structural details explaining the regulationof p110 catalytic function exerted by its association withp85 isoforms or Ras have been elucidated [8,9]. However,the abundance of solvent-exposed surfaces on p110 a  andp110 g  crystals suggests further protein – protein inter-actions [10,11]. For instance, recent evidence has beenprovided that p110 g  interacts with protein kinase C a (PKC a ) [12]. In this context, p110 g  phosphorylates PKC a and modulates its activity. This can be considered as akinase – substrate interaction, but it has recently emergedthat p110 subunits can also act, independently of theirkinase activity, as scaffold proteins. This indicates a novellayer of complexity in p110-mediated signaling, whichdepends on protein – protein interactions and does notinvolve the modulation of the catalytic function of PI3Ks.Indeed, a specific kinase-independent role was firstreported for p110 g , and a distinct selective scaffolding function has recently been described for p110 b  as well. Kinase-independent roles of p110 g p110 g  is activated by G bg  subunits of GPCRs. The bio-logical roles of this class I PI3K have been well character-ized in hematopoietic cells, where p110 g  activates adownstreamsignalingpathwayessentialfortheregulationofimmunityandinflammation[2].Accordingly,p110 g -nullmice present reduced leukocyte migration towards inflam-matory sites, which is phenocopied by mice that express akinase-dead p110 g  (p110 g KD/KD ) [13]. Furthermore, selec-tive p110 g  inhibitors are anti-inflammatory in multipledisease models [14,15]. A twist in our understanding of p110 g  signaling has come from studies on cardiomyocytebiology. Remarkably, p110 g  is expressed also in theheart, where it regulates  b -adrenergic signaling in akinase-dependent manner [16]. Both lipid and protein Review Corresponding author:  Hirsch, E. ( * p85 functions as both an adaptor and a regulator. . y The double name indicates different molecular weights found for the sameprotein. . 244  0968-0004/$  –  see front matter    2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.tibs.2009.02.003 Available online 17 April 2009  kinase activities of p110 have been shown to have pivotalroles in regulating   b -adrenergic receptor endocytosis [17].It was found that the p110 subunit interacts, through itsPIK domain, with the cytosolic enzyme  b  AR kinase-1( b  ARK1).Thisinteractionisfundamentalforthetargeting ofp110totheagonist-occupiedreceptor[17],hencetrigger-ing the localized PIP 3  production required for the organ-ization of clathrin-coated pits driving receptorinternalization [17].It was therefore unexpected to find that when subjectedto transverse aortic constriction leading to pressure over-load, cardiac remodeling and eventually failure, p110 g -null but not p110 g KD/KD mice developed a rapidly fatalcardiomyopathy [13].Infact,theadaptivecardiacresponseof p110 g KD/KD mice to aortic constriction was undistin-guishable from that of wild-type controls. These obser- vations clearly indicate that p110 g  is involved in cardiaccontractility and that it exerts this role mostly through akinase-independent mechanism. The action of p110 g  oncontractility depends on the formation of a multiproteincomplex containing phosphodiesterase 3B (PDE3B). Thep110 g -containing complex enhances the ability of PDE3Bto degrade cAMP, thus negatively regulating cardiac con-tractility  [13]. In the absence of the p110 g  protein, cardi-omyocytes, under basal conditions, present a higher cAMPlevel both in basal condition and after pressure overload,when this secondary messenger rises in an uncontrolledmanner and leads to rapid ventricular dilation and failure.By contrast, p110 g KD/KD mice show normal cAMP concen-tration both in resting conditions and in response to aorticconstriction. Furthermore, the phosphodiesterase activity that can be co-immunoprecipitated with p110 g  in pull-down experiments is comparable in p110 g KD/KD andwild-type control hearts, thus showing that the non-cata-lytic activity of p110 g  is indispensable in the activation of PDE3B. This is in keeping with a fundamental idea thatp110 g  acts as a scaffold protein that is likely to tetherPDE3B in the proximity of its activators. Although thespecific components of the PI3K  g -containing complex areyet to be identified, the adaptors p84/87 and p101 mightdirectly interact with PDE3B, thus providing a furtherlayer of action that modulates both p110 g -kinase-depend-ent and -independent functions [18] (Figure 1). The dual face of p110 g  signaling is also suggested by other observations in endothelial progenitor cells (EPCs)and platelets. In EPCs, p110 g  favors angiogenesis andsurvival. However, complementary actions could beexerted by kinase-dependent and -independent activitiesof p110 g . Indeed, p110 g -null mice present an impairmentin post-ischemic hindlimb angiogenesis, whereas p110 g KD/ KD mice normally develop vascular collaterals in ischemicmuscles, thus suggesting that a kinase-independentfunction of p110 g  is actually required for proper vascularrepair [19]. Althoughthesefindingsawaitfurther mechan-istic explanation, a phenotypic difference between p110 g -null and p110 g KD/KD EPCs underlines the dual nature of p110 g  signaling.In platelets, aggregation and thrombosis are regulatedby both p110 g  and p110 b  [20]. Inhibition of PI3K catalyticactivity with the pan-PI3K blocker wortmannin leads todefective platelet aggregation and impaired thrombus for-mation. However, p110 g -null platelets unexpectedly showed a stronger defect than wild-type controls treatedwith wortmannin. This shows that p110 g  can promoteplatelet aggregation independently of its kinase activity and that the scaffolding role of p110 g  is operational also inplatelets. However, both the catalytic activity and thescaffoldingfunctionofp110 g cooperateinpromotingaggre-gation, potentially through the orchestration of distinctpathways.Infact,whereaskinaseinhibitionmostlyaffectsthrombusstability,thescaffoldactivityofp110 g affectstheinitial thrombus growth process [20]. It remains unknownwhether the kinase-independent actions of p110 g  in EPCsand platelets share a similarity with those uncovered incardiomyocytes. Kinase-independent roles of p110 b p110 b  is ubiquitously expressed, and its kinase activity shows the unique ability among class I PI3Ks to be trig-gered by both RTKs and GPCRs [21,22]. The p110 b  cata-lytic subunit has a fundamental role in mammalian Table 1. Classification of mammalian PI3Ks Class Catalytic subunits Regulatory subunits Lipid substrate a I p110 a , p110 b , p110 d  p85 a , p85 b , p55 a , p55 g , p50 a  PI, PI4P, PIP 2 p110 g  p101, p84/87 PI, PI4P, PIP 2 II PI3K-C2 a , -C2 b , -C2 g  b PI, PI4PIII Vps34 p150 PI Abbreviations: PI4P, phosphoinositide 4-phosphate; PIP 2 , phosphoinositide (4,5)-bisphosphate; VPS34, vesicular protein sorting 34. a The indicated lipid substrate specificity refers to their respective  in vitro   lipid kinase activity. b For class II PI3Ks, the existence of a regulatory subunit has not been reported so far. Figure 1 . Kinase-dependent and kinase-independent roles of p110 g .  (a)  GPCR-dependent p110 g  activation triggers activation of Akt kinase through PIP 3 production, leading to different cellular responses such as cardiac remodeling,inflammation, EPC migration and thrombus formation.  (b)  The scaffolding role of p110 g  in modulating cAMP levels through PDE3B activation. p110 g  and p84/87bind to PDE3B in a macromolecular complex in which other members are yet to beidentified (as represented by a question mark). PDE3B degrades cAMP, thuscausing a decrease in heart contractility. Review  Trends in Biochemical Sciences   Vol.34 No.5 245  embryonic development, and its absence in mice causes anearlyembryoniclethalphenotype[23].Unexpectedly,how-ever, mice expressing a catalytically inactive p110 b  areborn and reach adulthood [21]. This discrepancy betweenphenotypes caused by the p110 b -null and kinase-deadmutants indicates that p110 b  exerts roles that are inde-pendent of its catalytic activity and that these functionsare essential for embryonic development. In line with thishypothesis, primary murine embryonic fibroblasts (MEFs)lacking p110 b  show a strong proliferation defect, whereasMEFs expressing a kinase-dead p110 b  grow normally   invitro [21,24].Consistentwiththisobservation,aninhibitorthat selectively blocks the kinase activity of the p110 b isoform[25]doesnotalterproliferationofwild-typecontrolMEFs [21]. Although proliferation and growth are cellular pro-cessesstrictlyregulatedbyRTKsignaling,therecruitmentof p110 b  by activated RTKs is still a matter of investi-gation. In multiple conditions, p110 b  is not directly boundto activated RTKs or to insulin receptor substrate 1 and 2(IRS-1/2) complexes [1,22]. Nonetheless, mice lacking p110 b  in the liver show a reduced response to insulin, aprototype RTK agonist. Although it has been previously reported that the acute response to insulin leading todownstreamphosphorylationofAkt(alsoknownasproteinkinase B [PKB]) relies only on p110 a  activation [1,26],these data might suggest the additional involvement of ap110 b -kinase-independent activity in insulin signaling [24]. Although p110 b  is not required for the short-termresponse to insulin, its kinase activity seems to be necess-ary for long term signaling because mice expressing akinase-dead p110 b , as well as those treated with p110 b selective inhibitors, lose their ability to sustain insulinsignaling  [21]. Therefore, the relative contributions of the kinase-dependent and -independent functions of p110 b ininsulinsignalingstillawaitfurtherinvestigation.Conversely, the involvement of p110 b  kinase activity downstream of RTKs is supported by the observation thatp110 b  has a role in cancer development [21]. Indeed, itskinase activity seems to be operational in conditions whenthe tumor-suppressor lipid phosphatase and tensin hom-ologue (PTEN) is lost [24,27] or RTK signaling is consti-tutively active [21].Recent data have highlighted the implication of p110 b in the endocytic pathway. In this respect, previous studieshave reported that the PI3K binding site of the activatedhuman platelet-derived growth factor receptor (PDGFR) isrequiredforits internalization [28,29]. However, theinvol- vementofspecificPI3Kisoformsandtheunderlyingmech-anisms have remained poorly understood. Recently,several lines of evidence prompted the consideration of the role of p110 b  in this context. For instance, the loss of p110 b  results in impaired internalization of the transfer-rinreceptor(TfR)andtheepidermalgrowthfactorreceptor(EGFR)[21,24].Importantly,thisdefectinreceptorintern-alization can be rescued by transfection of p110 b -null cellswith a kinase-dead p110 b  [24]. This finding implies thatthe participation of p110 b  in the endocytic process is notdue to p110 b  kinase activity but entails a protein  proteininteraction mechanism (Figure 2). However, the exactnature of p110 b  interactors in this process remainsunclear. A likely candidate is represented by the smallmonomeric GTPase Rab5 (Ras-related GTP-binding protein 5), which has a major role in the formation andfusion of clathrin-coated vesicles (CCVs). Indeed, it hasbeen reported that p110 b  interacts with Rab5 in mem-branes,althoughthefunctionalrelationshipbetweentheseproteins remains elusive [30]. The absence of p110 b expression causes a dramatic reduction in the number of CCVs forming beneath the plasma membrane, and thetransfection of constitutively active Rab5 into these cellscannot rescue their endocytic defect [21]. The incapacity of constitutively active Rab5 to rescue the impairment in vesicle formation resulting from p110 b  loss indicates that Figure 2 . Kinase-dependent and kinase-independent roles of p110 b .  (a)  Schematic model of kinase-dependent functions of p110 b , leading to activation of Akt kinasedownstream of GPCRs and RTKs. Upon the activation of RTKs and GPCRs, p110 b  is recruited to the plasma membrane in the proximity of its substrate, thus leading to PIP 3 production. PIP 3  acts as a second messenger that, through Akt activation, mediates many cellular responses, such as cancer and glucose homeostasis.  (b)  Proposedmechanism for the kinase-independent function of p110 b  in the endocytic route. The p110 b – p85 a  complex binds to Rab5 on the cytoplasmic surface of budding CCVs,hence leading to receptor endocytosis and vesicle fusion. Review  Trends in Biochemical Sciences   Vol.34 No.5 246  p110 b  does not act upstream of Rab5 but rather has a rolein an early event in the formation of CCVs [21]. In con-ditionswherep110 b isexpressed,ithasbeenreportedthatatripartitecomplex,p85 a – p110 b – Rab5,mightform,hencedirectly modulating Rab5 activity. In this context, p85 a could act as a GTPase-activating protein (GAP) for smallGTPases like Rab5 and Rab4, accelerating their catalyticrate and thus inactivating their function [31]. Given themajor role of p110 b  in promoting the early steps in endo-cytosis, it is conceivable that the latter process is only required for fine tuning. Nonetheless, our understanding ofp110 b functionisstillincompleteandfurtherstudiesarerequired to clarify this issue. The TRRAP-ing factor Where interactors bind p110s and how they trigger thep110 scaffolding function are not yet fully understood. Thesequences of p110 that are involved in any protein – proteininteractions seem to involve multiple domains. Regionspreviously thought to be only required for kinase activity (e.g. the PIK domain in substrate selection) are now alsoconsidered as preferential sites for the protein binding. Inline with this, even the kinase domain could be involved inprotein – protein interactions. Potential insight into thep110 binding mechanism could come from studies of thetransformation/transcription domain-associated protein(TRRAP), a distant p110 homologue that is present inalleukaryotesandthatprovidessupportforsuchahypoth-esis. Interestingly, TRRAP is a pseudokinase belonging tothe PI3K-related kinase (PIKK) superfamily, which alsoincludes the closely p110-related DNA-dependent proteinkinase (DNA-PK), ataxia telangiectasia mutated (ATM)andataxiatelangiectasiarelated (ATR)kinases(Figure3).TRRAP presents a C-terminal domain that is significantly homologous to the kinase domain of p110 subunits butlackstheaminoacidicresiduesresponsibleforthecatalyticactivity  [32]. Direct evidence has been provided thatTRRAP is kinase-defective, indicating that its functionsdo not rely on kinase activity. Different lines of evidencehave shown that TRRAP constitutes a fundamental com-ponent of the histone acetyltransferase (HAT) complexesand is responsible for their recruitment to chromatinduring transcription, replication and repair. Its role inthese functions is mirrored by embryonic lethality causedby its genetic deletion in mice [33]. Interestingly, deletionswithin the pseudokinase domain can blunt the binding of TRRAP to thehumantranscription factor suppressor ofTy 3 (hSPT3) and to the HAT complex, suggesting that thisdomain is necessary but not sufficient for this scaffolding function [34]. Although no direct evidence has beenreported so far for a similar function of the p110 kinasedomain, these findings suggest that this area is worthy of further investigation. Concluding remarks and future perspectives The differences in physiology between kinase-null miceandkinase-deadmutantshaveexposedapreviouslyuntoldstory for PI3K, one in which their catalytically activesubunits have been found to have an unconventional scaf-folding role. This chapter is still relatively new and, assuch, is ripe for speculation on its importance to the cell.Therefore, we must limit ourselves in the short term to thebasics of discovering more about the who and what of interaction before we can learn the why. Therefore, bio-chemical and genetic studies will be fundamental in theprecise mapping of the domains and residues involved inthe interaction of p110 with other proteins. Furthermore,the interaction partners identified to date provide only ahint of a potentially larger set of associations, and newproteomic studies will also be essential to help uncoveradditional binding partners of p110s. Once we have aclearer picture of both the sites of binding and of thepartners involved, we will then be in a position to betterdelineate the mechanisms for the kinase-independentfunctions of the PI3Ks and learn why they are significantto cellular function. Acknowledgements This work was supported by the European Union Sixth Framework Programme (FP6) EUGeneHeart project, Fondation Leducq and theTelethon Foundation. References 1 Foukas,L.C.  etal. (2006)Criticalroleforthep110 a phosphoinositide-3-OH kinase in growth and metabolic regulation.  Nature  441, 366 – 3702 Hirsch, E.  et al.  (2000) Central role for G protein-coupledphosphoinositide 3-kinase  g  in inflammation.  Science  287, 1049 – 10533 Shahbazian, D.  et al.  (2006) The mTOR/PI3K and MAPK pathwaysconvergeoneIF4Btocontrolitsphosphorylationandactivity.  EMBOJ. 25, 2781 – 27914 Jones, A.T. and Clague, M.J. (1995) Phosphatidylinositol 3-kinaseactivity is required for early endosome fusion.  Biochem. J.  311, 31 – 345 Leevers, S.J.  et al.  (1999) Signalling through phosphoinositide3-kinases: the lipids take centre stage.  Curr. Opin. Cell Biol.  11,219 – 225 Figure 3 . The evolutionary relationship among PIKK superfamily members. Thisphylogenetic tree has been constructed using the ClustalW (1.81 version) tool andis based on the sequence alignment of the kinase domains of PIKK proteins withthe Blosum matrix. Many PIKK components phosphorylate lipid and proteinsubstrates (in blue), whereas others use only protein substrates (in green). To date,TRRAP (in red) is the only pseudokinase that has been identified in thissuperfamily. TRRAP, p110 b  and p110 g  (highlighted with asterisks) have beendemonstrated to possess a scaffolding activity. Abbreviations: DNA-PKcs, DNA-dependent protein kinase catalytic subunit; FRAP, FK506 binding protein 12-rapamycin associated protein. Review  Trends in Biochemical Sciences   Vol.34 No.5 247  6 He, X.  et al.  (2008) PIK3IP1, a negative regulator of PI3K, suppressesthe development of hepatocellular carcinoma.  Cancer Res.  68, 5591 – 55987 Hawkins, P.T.  et al.  (2006) Signalling through Class I PI3Ks inmammalian cells.  Biochem. Soc. Trans.  34, 647 – 6628 Miled, N.  et al.  (2007) Mechanism of two classes of cancer mutations inthe phosphoinositide 3-kinase catalytic subunit.  Science  317, 239 – 2429 Pacold, M.E.  et al.  (2000) Crystal structure and functional analysis of Ras binding to its effector phosphoinositide 3-kinase  g .  Cell  103, 931 – 94310 Huang, C.H.  et al.  (2007) The structure of a human p110 a  /p85 a complex elucidates the effects of oncogenic PI3K  a  mutations.  Science  318, 1744 – 174811 Walker, E.H.  et al.  (1999) Structural insights into phosphoinositide 3-kinase catalysis and signalling.  Nature  402, 313 – 32012 Lehmann,K.  etal. (2008)PI3K  g controlsoxidativeburstinneutrophils via interaction with PKC a  and p47phox.  Biochem J  13 Patrucco, E.  et al.  (2004) PI3K  g  modulates the cardiac response tochronic pressure overload by distinct kinase-dependent and -independent effects.  Cell  118, 375 – 38714 Camps, M.  et al.  (2005) Blockade of PI3K  g  suppresses jointinflammation and damage in mouse models of rheumatoid arthritis.  Nat. Med.  11, 936 – 94315 Barber, D.F.  et al.  (2005) PI3K  g  inhibition blocks glomerulonephritisand extendslifespan in a mousemodelofsystemic lupus.  Nat.Med. 11,933 – 93516 Nienaber, J.J.  et al.  (2003) Inhibition of receptor-localized PI3K preserves cardiac  b -adrenergic receptor function and amelioratespressure overload heart failure.  J. Clin. Invest.  112, 1067 – 107917 Naga Prasad, S.V.  et al.  (2005) Protein kinase activity of phosphoinositide 3-kinase regulates  b -adrenergic receptorendocytosis.  Nat. Cell Biol.  7, 785 – 79618 Voigt,P.  etal. (2006)Characterizationofp87PIKAP,anovelregulatory subunitofphosphoinositide3-kinase g thatishighlyexpressedinheartand interacts with PDE3B.  J. Biol. Chem.  281, 9977 – 998619 Madeddu, P.  et al.  (2008) Phosphoinositide 3-kinase  g  gene knockoutimpairs postischemic neovascularization and endothelial progenitorcell functions.  Arterioscler. Thromb. Vasc. Biol.  28, 68 – 7620 Schoenwaelder, S.M.  et al.  (2007) Identification of a unique co-operative phosphoinositide 3-kinase signaling mechanism regulating integrin  a IIb b 3  adhesive function in platelets.  J. Biol. Chem.  282,28648 – 2865821 Ciraolo, E.  et al.  (2008) Phosphoinositide 3-kinase p110 b  activity: key role in metabolism and mammary gland cancer but not development.  Sci. Signal.  1, ra322 Guillermet-Guibert, J.  et al.  (2008) The p110 b  isoform of phosphoinositide 3-kinase signals downstream of G protein-coupledreceptors and is functionally redundant with p110 g .  Proc. Natl. Acad. Sci. U. S. A.  105, 8292 – 829723 Bi, L.  et al.  (2002) Early embryonic lethality in mice deficient in thep110 b  catalytic subunit of PI 3-kinase.  Mamm. Genome  13, 169 – 17224 Jia, S.  et al.  (2008) Essential roles of PI(3)K-p110 b  in cell growth,metabolism and tumsrcenesis.  Nature  454, 776 – 77925 Jackson, S.P.  et al.  (2005) PI 3-kinase p110 b : a new target forantithrombotic therapy.  Nat. Med.  11, 507 – 51426 Knight, Z.A.  et al.  (2006) A pharmacological map of the PI3-K family defines a role for p110 a  in insulin signaling.  Cell  125, 733 – 74727 Wee, S.  et al.  (2008) PTEN-deficient cancers depend on PIK3CB.  Proc. Natl. Acad. Sci. U. S. A.  105, 13057 – 1306228 Joly, M.  et al.  (1994) Disruption of PDGF receptor trafficking by mutation of its PI-3 kinase binding sites.  Science  263, 684 – 68729 Joly, M.  et al.  (1995) Phosphatidylinositol 3-kinase activity is requiredat a postendocytic step in platelet-derived growth factor receptortrafficking.  J. Biol. Chem.  270, 13225 – 1323030 Christoforidis, S.  et al.  (1999) Phosphatidylinositol-3-OH kinases areRab5 effectors.  Nat. Cell Biol.  1, 249 – 25231 Chamberlain, M.D.  et al.  (2004) The p85 a  subunit of phosphatidylinositol 3 0 -kinase binds to and stimulates the GTPaseactivity of Rab proteins.  J. Biol. Chem.  279, 48607 – 4861432 McMahon,S.B.  etal. (1998)ThenovelATM-relatedproteinTRRAPisanessential cofactor for the c-Myc and E2F oncoproteins.  Cell  94, 363 – 37433 Herceg, Z.  et al.  (2001) Disruption of Trrap causes early embryoniclethality and defects in cell cycle progression.  Nat. Genet.  29, 206 – 21134 Park,J.  etal. (2001)TheATM-relateddomainofTRRAPisrequiredforhistone acetyltransferase recruitment and Myc-dependentoncogenesis.  Genes Dev.  15, 1619 – 1624 Review  Trends in Biochemical Sciences   Vol.34 No.5 248
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!