Two ubiquitin-like conjugation systems essential for autophagy

Seminars in Cell &Developmental Biology 15(2004)231–236Two ubiquitin-like conjugation systems essential for autophagyYoshinori Ohsumi ∗,Noboru MizushimaDepartment of Cell Biology,National Institute for Basic Biology,Myodaiji-cho,Okazaki 444-8585,JapanKeywords:Autophagy;ATG genes;Atg12;Atg8;apg1.IntroductionThere are two major pathways of intracellular protein degradation.First,the ubiquitin-proteasome system in the cytosol is involved in degradation of short-lived,damaged or misfolded proteins [1,2].Target proteins to be degraded are first tagged with ubiquitin and then digested by the protea-some with strict recognition by the ubiquitin ligase system.Long-lived proteins are believed to be degraded within a specific compartment,the lysosome/vacuole.So far sev-eral delivery routes to this lytic compartment have been pro-posed.The process of degradation of the cell’s own intracel-lular constituents in lysosomes is generally called autophagy in contrast to heterophagy,degradation of extracellular ma-terials [3].Macroautophagy (hereafter autophagy)is a major pathway in autophagy,and initiates by enwrapping a por-tion of cytoplasm by a membrane sac called the isolation membrane,to form a double membrane structure,the au-tophagosome [4].The autophagosome then fuses with the lysosome,becoming the autolysosome and its inner mem-brane and contents are digested for reuse.Autophagy is characterized as being nonselective result-ing in the bulk degradation of cellular proteins.More than 90%of cellular proteins are long lived,thus the turnover of long-lived proteins is important to understand the phys-iology of the cell.Since the lysosome was discovered [5],electron microscopy has revealed that autophagy occurs in a variety of cells from different tissues and cultured cells,and it is now generally accepted that autophagy is a ubiquitous activity in eukaryotic cells.However,the molecular mechanism of autophagy has re-mained elusive.There was no specific monitoring marker or quantitative assay system to detect autophagy in mam-malian cells.Therefore,the genes involved in autophagy re-mained unidentified for a long time.Here,we will focus on the recent progress in the molecular dissection of autophagy.∗Correspondingauthor.Tel.:+81-564-75-7515;fax:+81-564-75-7516.E-mail address:yohsumi@nibb.ac.jp (Y .Ohsumi).Discoveries of two ubiquitin-like conjugation reactions have begun to unravel the mystery of autophagy.2.Discovery of autophagy in the yeast,Saccharomyces cerevisiaeIt has been postulated that the vacuole is a lytic organelle in yeast like the lysosome [6].It was shown that bulk pro-tein turnover is induced upon nitrogen starvation,which is dependent upon vacuolar enzyme activities [7,8].Obvious questions are what kinds of substrates are degraded in the vacuole and what is the mechanism of sequestering those substrates into the vacuole?In early 1990,we found that the yeast cell induces au-tophagy under various nutrient starvation conditions.When vacuolar proteinase-deficient mutants in a rich medium were shifted to a nitrogen-depleted medium,spherical structures appeared in the vacuole after 30–40min of lag time,accumulated and almost filled the vacuole within 10h [9].These structures,named autophagic bodies,are mostly single membrane-bound structures containing a portion of cytoplasm [9].Furthermore,double membrane structures of the same size as autophagic bodies were found in the cytosol of the starved cells.The autophagosome in yeast is about 300–900nm in diameter and contains cytoplasmic components nonselectively.Fusion of the outer membrane of the autophagosome with the vacuole was shown by freeze-fracture electron microscopy [10,11].The mem-brane dynamics of autophagy in yeast are topologically the same as that of autophagy in mammals,though the vacuole is much larger than lysosome.The whole scheme of the autophagy in yeast is shown in Fig.1.3.Genetic approaches to yeast autophagyGenetic approaches were taken to address the molecular mechanism of autophagy.The morphological change of the1084-9521/$–see front matter ©2003Published by Elsevier Ltd.doi:10.1016/j.semcdb.2003.12.004232Y.Ohsumi,N.Mizushima/Seminars in Cell&Developmental Biology15(2004)231–236Fig.1.Macroautophagy in the yeast Saccharomyces cerevisiae.A por-tion of cytoplasm,including organelle,isfirst enclosed in a double or multilamellar structure,the autophagosome.The outer membrane of the autophagosome then fuses with the vacuole,and the cytoplasm-derived materials in the inner membrane are rapidly disintegrated.vacuole under starvation,accumulation of autophagic bod-ies,and an immunoscreen of cells that retain a cytosolic enzyme(fatty acid synthase)after starvation,were used to obtain autophagy-defective mutants(apg and aut).Later, two hybrid screens using Apg proteins as bait identified two more APG genes[12,13].Klionsky’s group isolated mutants defective in the maturation of one of the vacuolar enzymes,aminopeptidase I(API)[14].These so called cvt mutants(cytoplasm to vacuole targeting pathway)over-lapped with autophagy-defective mutants[15,16]though the two pathways are quite different.Electron microscopic analyses showed that the Cvt pathway uses similar mem-brane dynamics to autophagy[17].These genetic analyses revealed that at least16genes(Table1)are required specif-Table1Atg proteins essential for autophagosome formationNew nomenclature Classical name Atg12conjugation systemAtg12Apg12Atg7∗Apg7Atg10Apg10Atg5Apg5Atg16Apg16Atg8lipidation systemAtg8Aut7/Apg8Atg4Aut2/Apg4Atg7∗Apg7Atg3Apg3Atg1protein kinase complexAtg1Apg1Atg13Apg13Atg17Apg17 Autophagy-specific PI3kinaseAtg6(Vps30)Apg6Atg14Apg14(Vps34)(Vps15)OthersAtg2Apg2Atg9Apg9Atg18Aut10∗Atg7functions in two conjugation systems.ically for autophagy in yeast.Recently the nomenclature ofautophagy-related genes was changed to a novel gene nameATG[18].Here,ATG x will be used instead of the previousAPG x nomenclature.4.Characterization of Atg proteinsAll the atg mutants grew normally in rich medium,butfailed to induce bulk protein degradation under nutrient-depletion conditions.A homozygous diploid with an atgmutation could not perform sporulation[19].Another fea-ture of autophagy-defective mutants is the loss of viabilityduring nitrogen starvation.These mutants start to die after2days of starvation and almost completely lose viabilityafter1week[19].Cloning and identification of autophagygenes revealed that almost all are novel genes except ATG6,allelic to VPS30[20].In yeast,autophagy is almost com-pletely shut off under growing conditions,but every ATGgene is expressed in the growing conditions.Several ATGgenes(ATG8and ATG14)were transcriptionally upregulatedby starvation or under the negative regulation of Tor kinases[21,22].Molecular biological and biochemical analyses ofthese gene products uncovered the genetic and physical in-teractions among the Atg proteins.One of the most remark-ablefindings amongst the Atg proteins was the discoveryof two ubiqutin-like conjugation systems[23].In fact,halfof the16former APG genes essential for autophagy are in-volved in these novel conjugation systems.5.The yeast Atg12conjugation systemAtg12,a hydrophilic small protein of186amino acidswith no apparent homology to ubiquitin,covalently links toAtg5[24].The mode of conjugation of Atg12to Atg5isquite similar to that of ubiquitination.The carboxy-terminalresidue of Atg12is a single glycine,which is activated byAtg7in an ATP-dependent manner[24].Then,Atg12formsa conjugate with Atg7through a thioester bond[25–27].Atg7shows restricted homology with the ubiquitin-E1en-zyme within the regions around the putative ATP-bindingsite and the active site cysteine[24].Atg12is then trans-ferred to Atg10through a thioester bond[28].The functionof Atg10is similar to that of E2enzymes,although Atg10shows no homology to a known E2.Finally,the carboxy-terminal glycine of Atg12forms an isopeptide bond with the ε-amino group of lysine149of Atg5[24].Formation of the conjugate is requisite for the progression of autophagy.For-mation of the Atg12–Atg5conjugate is a constitutive pro-cess,which is not regulated by nutrient conditions.Atg12and Atg5form a conjugate immediately after their synthesisand free forms of them are hardly detectable.The Atg12sys-tem has distinct features from other ubiquitin-like systems.Atg12is synthesized as an active form and does not requireC-terminal processing.Atg5is the only target of the Atg12Y.Ohsumi,N.Mizushima/Seminars in Cell&Developmental Biology15(2004)231–236233modification,and so far no protease activity to deconjugate this conjugate was found,suggesting that this conjugation reaction is irreversible.The Atg12–Atg5conjugate behaves as if it is a single protein.Yeast Atg12may consist of a C-terminal ubiquitin-fold and a long N-terminal region.The three-dimensional structure of Atg12has not been solved yet but recent analysis suggested that its essential function resides within the ubiquitin-fold.The Atg12–Atg5conju-gate then forms a complex with Atg16.Atg16binds to Atg5 but not to Atg12[12].Atg16has a coiled-coil region in its C-terminal half and likely forms a tetramer through this re-gion.Atg12–Atg5·Atg16forms a350-kDa multimeric com-plex essential for autophagy[29].6.Atg8conjugation systemThe second ubiquitin-like protein essential for autophagy is Atg8(Aut7/Apg8),a117-amino acid protein.Atg8was shown to be a good marker for membrane dynamics during autophagy because it resides on the membrane sac enwrap-ping the cytoplasm,autophagosome,and also the autophagic body[21].Cell fractionation studies showed that Atg8is mostly membrane bound;about half is peripherally bound to membrane but half behaves like an intrinsic membrane protein[30].Epitope tagging analyses of Atg8revealed processing of nascent Atg8near the C-terminal end[30].Atg4,a novel cys-teine protease,is responsible for processing Atg8by cleav-ing a single Arg residue,consequently exposing Gly at the C-terminus of Atg8.The processed form of Atg8is then activated by Atg7,and is transferred to a conjugating E2 enzyme,Atg3[31].Atg7is a unique enzyme that activates two different ubiquitin-like proteins,Atg12and Atg8,and assigns them to their proper E2enzymes,Atg10and Atg3, respectively.Atg3has no overall significant homology to E2 enzymes of the ubiquitin system but shows some homology to Atg10.Interestingly,Atg8forms a conjugate not with a protein,but with phosphatidylethanolamine(PE),an abun-dant membrane phospholipid[31].This lipidation reaction is necessary for the membrane dynamics of autophagy.For normal autophagosome formation Atg8-PE should be de-conjugated by the processing enzyme,Atg4,therefore,the cycle of conjugation and deconjugation is important for the normal progression of autophagy[31].The two conjugation reactions are somehow related,since the Atg8-PE levels be-come significantly reduced in the absence of the Atg12–Atg5 conjugate.7.Other factors required for autophagyRecent analyses of Atg proteins revealed that there are two kinase complexes in addition to ubiquitin-like conjuga-tion systems.One is the Atg1protein kinase associated with Atg13,Atg17,Cvt9[32,33].The N-terminus contains a pro-tein kinase domain,and kinase activity was detected in vitro.A kinase-negative atg1mutant is defective in autophagy,im-plying that the kinase activity is essential for the function [13,32].The kinase activity is upregulated during induction of autophagy,so the level of kinase activity seems to be important for the regulation of autophagosome formation. Atg13is highly phosphorylated under nutrient-rich condi-tions.Upon starvation,it is dephosphorylated[13].Atg13 functions as a positive regulator of Atg1kinase.Recent ge-netic analyses suggested that the Atg1complex functions at a rather late step of autophagosome formation[34].The Atg1complex may control membrane dynamics rather than act as a signal transducer.The fourth complex is the autophagy-specific phos-phatidylinositol3-kinase(PI3K)complex.It was found that Vps30forms two distinct protein complexes[35].One con-sists of Vps30,Atg14,Vps34,and Vps15,and the other of Vps30,Vps38,Vps34,and Vps15.Vps34is a sole PI3K in yeast and Vps15is a regulatory protein kinase of Vps34. The former PI3K complex is responsible for autophagy,and the latter is for vacuolar protein sorting.Vps30is a possi-ble coiled-coil protein and associated with the membrane through Vps15and Vps34.Atg14is a specific factor in the autophagy-specific PI3K complex[35].8.Preautophagosomal structure,the site of autophagsosome formationAll the former APG genes have a function before or dur-ing the formation step of the autophagosome.The mem-brane dynamics of autophagy are distinct from the classical vesicular membrane trafficking.There are many fundamen-tal questions relating to the molecular mechanism of au-tophagosome formation.For a long time,the origin of the autophagosome mem-brane was proposed to be the ER.Freeze-fracture of the autophagsome indicates that both membranes are quite dif-ferent from the ER,and only the outer membrane contains intramembrane particles.We proposed that autophagosome formation is not simply due to enwrapping by a pre-existing large membrane structure,such as the ER,but rather by as-sembly of new membrane from its constituents.As mentioned earlier,all former Apg proteins function quite closely together at the autophagosome formation step.It is therefore crucial to know the intracellular local-ization of Atg proteins.By usingfluorescent protein-fused Atg proteins,visualization of Atg proteins were performed. Atg8nicely stains autophagosomes,autophagic bodies in the vacuoles,and also the intermediate isolation mem-brane.However,Atg5shows a single bright dot structure next to the vacuole.Atg8also co-localizes with this struc-ture.We named it the preautophagosome structure(PAS). Recent studies indicated that almost all Atg proteins are co-localized in the PAS[34].This structure seems to be an organizing center of the autophagosome.The lipidation234Y.Ohsumi,N.Mizushima /Seminars in Cell &Developmental Biology 15(2004)231–236is requisite for the recruitment of Atg8to the PAS.In the mutants defective in Atg12system Atg8does not localize to the PAS.In atg14or atg6,Atg8and Atg5do not asso-ciate with PAS,indicating that the autophagy-specific PI3K complex may play an important role in the organization of PAS [34].Atg9also has a strong effect on organization of the PAS,whereas defects in the Atg1kinase complex show little effect on the PAS structure.9.The Atg conjugation systems in mammalian cells The two Atg conjugation systems are highly conserved in higher eukaryotic cells,suggesting that eukaryotes ac-quired the mechanism of autophagy at the beginning of its evolution.In both mouse and human,there is only one or-thologue for each component of the Atg12system.Atg12is conjugated to Atg5[36],which is catalyzed by Atg7[37]and Atg10[38,39].In mammalian cells,the Atg12–Atg5conjugate forms an ∼800-kDa protein complex.Character-ization of this complex revealed an additional protein that interacts with Atg5[53],a 63-to 74-kDa protein with sev-eral spliced isoforms.Since the N-terminal region of this novel protein exhibits several features in common with yeast Atg16,such as the Atg5-binding site and coiled-coil region,it was designated Atg16L.Atg16L,however,has a long C-terminal extension containing seven WD repeats.Atg16L homologues are found in all eukaryotes except two yeasts,Saccharomyces cerevisiae and Pichia pastoris ,which have Atg16lacking WD repeats.The role of the WD repeats in Atg16L has not been uncovered yet.In contrast to Atg12,at least three Atg8homologues were identified in mammals:microtubule-associated protein 1(MAP1)light chain 3(LC3)[40],Golgi-associated ATPase enhancer of 16kDa (GATE-16)[41],and ␥-aminobutyric acid (GABA)A -receptor-associated protein (GABARAP)[42].Among them,LC3was shown to localize on the au-tophagosome membrane [43].GABARAP was suggested to be involved in the GABA A receptor clustering [44]or trans-port [45].GATE-16has been suggested to be an intra-Golgi transport modulator that interacts with N -ethylmaleimide-sensitive factor (NSF)and Golgi v-SNARE GOS-28[46].However,further analyses are required to reveal thephys-Fig.2.Molecular mechanism of autophagy in mammalian cells.The Atg12–Atg5conjugate and Atg16L localize to the isolation membrane throughout its elongation process.LC3is recruited to the membrane in the Atg5-dependent manner.The Atg12–Atg5·Atg16L complex dissociates from the membrane upon completion of autophagosome formation,while LC3(-II)remains on the autophagosome membrane.Atg5and its modification by Atg12are required for elongation of the isolation membrane.PI3K activity is required for the formation of isolation membrane.iological roles of these three molecules and their modi-fied forms.All these Atg8homologues are processed by mammalian Atg4homologues to expose the conserved C-terminal glycine [47,48].They are modified by mammalian Atg7[37]and Atg3homologues [49].10.Role of Atg conjugation in mammalian autophagy The function of the mammalian Atg12system was shown using mouse embryonic stem (ES)cells.ES cells induce au-tophagy well by nutrient starvation and the size of the au-tophagosome in ES cells is larger than in other cell lines.First,the localization of Atg12–Atg5was examined in de-tail using GFP-fused Atg5[50].A small fraction of cy-tosolic Atg12–Atg5·Atg16L complex localizes to the isola-tion membrane throughout its elongation process (Fig.2).Atg12–Atg5initially associates with a small crescent-shaped vesicle evenly.As the membrane elongates,Atg12–Atg5shows asymmetric localization;most Atg12–Atg5associates with the convex surface of the isolation membrane.Finally,Atg12–Atg5dissociates from the membrane upon comple-tion of autophagosome formation.Atg16L is usually found complexed with Atg12–Atg5,suggesting that these proteins associate and dissociate from the membrane as the 800-kDa complex.LC3also targets to the isolation membrane in an Atg5-dependent manner,and remains on the autophagoso-mal membrane even after Atg12–Atg5dissociates (Fig.2)[43,50].The molecular basis of this transient association of Atp12–Atg5conjugates is not known yet.Visualization of GFP-Atg5in living cells directly demon-strated that autophagosomes are generated by elongation of small membrane structures,autophagosome precursors,not derived from pre-existing large membrane such as ER cister-nae [50].The nature of the precursor vesicle remains to be solved.Treatment of cells with 3-methyladenine,which is widely used as an autophagy inhibitor [51],abolishes the for-mation of any Atg5-positive structure [50].3-Methyladenine is now considered as a PI3K inhibitor [52].Indeed,the au-tophagosome precursors are not formed by treatment with wortmannin,a well-known PI3K inhibitor,suggesting that PI3K activity is required for early stage of autophagosome formation [50].Although it is not known whether a structureY.Ohsumi,N.Mizushima/Seminars in Cell&Developmental Biology15(2004)231–236235equivalent to the yeast PAS exists in mammalian cells,these results are quite consistent with those observed in yeast.A gene targeting study demonstrated that Atg12–Atg5 is essential for elongation of the isolation membrane in mammalian cells[50].LC3cannot target to the membrane in ATG5−/−cells.In wild-type cells,LC3is detected in two forms on immunoblotting at18kDa(LC3-I)and at 16kDa(LC3-II)[43].Since LC3-II is a tightly membrane-bound form,it may be a PE-conjugated form.Strikingly,in ATG5−/−cells,LC3-II is not generated at all[50].Thus,re-lationship between the Atg12and LC3system is quite sim-ilar to the yeast system.Using these Atg proteins as molecular marker,and com-bination of yeast and mammalian system,we will be able to unveil the membrane dynamics during autophagy in the near future.References[1]Hochstrasser M.Ubiquitin-dependent protein degradation.Annu RevGenet1996;30:405–39.[2]Hershko A,Ciechanover A.The ubiquitin system.Annu RevBiochem1998;67:425–79.[3]Mortimore GE,Poso AR.Intracellular protein catabolism and itscontrol during nutrient deprivation and supply.Annu Rev Nutr 1987;7:539–64.[4]Seglen PO,Bohley P.Autophagy and other vacuolar protein degra-dation mechanisms.Experientia1992;48:158–72.[5]de Duve C.In:Hayashi T,editor.Subcellular particles.New York:Ronald;1959.p.128–59.[6]Klionsky DJ,Herman PK,Emr SD.The fungal vacuole:composition,function,and biogenesis.Microbiol Rev1990;54:266–92.[7]Zubenko GS,Jones EW.Protein degradation,meiosis and sporulationin proteinase-deficient mutants of Saccharomyces cerevisiae.Genetics 1981;97:45–64.[8]Egner R,Thumm M,Straub M,Simeon A,Schuller HJ,Wolf DH.Tracing intracellular proteolytic pathways.Proteolysis of fatty acid synthase and other cytoplasmic proteins in the yeast Saccharomyces cerevisiae.J Biol Chem1993;268:27269–76.[9]Takeshige K,Baba M,Tsuboi S,Noda T,Ohsumi Y.Autophagy inyeast demonstrated with proteinase-deficient mutants and conditions for its induction.J Cell Biol1992;119:301–11.[10]Baba M,Takeshige K,Baba N,Ohsumi Y.Ultrastructural analysisof the autophagic process in yeast:detection of autophagosomes and their characterization.J Cell Biol1994;124:903–13.[11]Baba M,Osumi M,Ohsumi Y.Analysis of the membrane structureinvolved in autophagy in yeast by freeze-replica method.Cell Struct Funct1995;20:465–71.[12]Mizushima N,Noda T,Ohsumi Y.Apg16p is required for the functionof the Apg12p–Apg5p conjugate in the yeast autophagy pathway.EMBO J1999;18:3888–96.[13]Kamada Y,Funakoshi T,Shintani T,Nagano K,Ohsumi M,OhsumiY.Tor-mediated induction of autophagy via an Apg1protein kinase complex.J Cell Biol2000;150:1687–95.[14]Harding TM,Morano KA,Scott SV,Klionsky DJ.Isolation andcharacterization of yeast mutants in the cytoplasm to vacuole protein targeting pathway.J Cell Biol1995;131:591–602.[15]Harding TM,Hefner-Gravink A,Thumm M,Klionsky DJ.Geneticand phenotypic overlap between autophagy and the cytoplasm to vacuole protein targeting pathway.J Biol Chem1996;271:17621–4.[16]Scott SV,Hefner-Gravink A,Morano KA,Noda T,Ohsumi Y,Klion-sky DJ.Cytoplasm-to-vacuole targeting and autophagy employ thesame machinery to deliver proteins to the yeast vacuole.Proc Natl Acad Sci USA1996;93:12304–8.[17]Baba M,Osumi M,Scott SV,Klionsky DJ,Ohsumi Y.Two dis-tinct pathways for targeting proteins from the cytoplasm to the vac-uole/lysosome.J Cell Biol1997;139:1687–95.[18]Klionsky DJ,Cregg JM,Dunn Jr WA,Emr SD,Sakai Y,SandovalIV,et al.A unified nomenclature for yeast autophagy-related genes.Dev Cell2003;5:539–45.[19]Tsukada M,Ohsumi Y.Isolation and characterization of autophagy-defective mutants of Saccharomyces cerevisiae.FEBS Lett1993;333: 169–74.[20]Kametaka S,Okano T,Ohsumi M,Ohsumi Y.Apg14p andApg6/Vps30p form a protein complex essential for autophagy in the yeast,Saccharomyces cerevisiae.J Biol Chem1998;273:22284–91.[21]Kirisako T,Baba M,Ishihara N,Miyazawa K,Ohsumi M,Yoshi-mori T,et al.Formation process of autophagosome is traced with Apg8/Aut7p in yeast.J Cell Biol1999;147:435–46.[22]Chan TF,Bertram PG,Ai W,Zheng XF.Regulation of APG14expression by the GATA-type transcription factor Gln3p.J Biol Chem 2001;276:6463–7.[23]Ohsumi Y.Molecular dissection of autophagy:two ubiquitin-likesystems.Nat Rev Mol Cell Biol2001;2:211–6.[24]Mizushima N,Noda T,Yoshimori T,Tanaka Y,Ishii T,George MD,et al.A protein conjugation system essential for autophagy.Nature 1998;395:395–8.[25]Tanida I,Mizushima N,Kiyooka M,Ohsumi M,Ueno T,OhsumiY,et al.Apg7p/Cvt2p:a novel protein-activating enzyme essential for autophagy.Mol Biol Cell1999;10:1367–79.[26]Kim J,Dalton VM,Eggerton KP,Scott SV,Klionsky DJ.Apg7p/Cvt2p is required for the Cvt,macroautophagy,and peroxi-some degradation pathway.Mol Biol Cell1999;10:1337–51. [27]Yuan W,Stromhaug PE,Dunn Jr WA.Glucose-induced microau-tophagy of peroxysomes in Pichia pastoris requires a unique E1-like protein.Mol Biol Cell1999;10:1353–66.[28]Shintani T,Mizushima N,Ogawa Y,Matsuura A,Noda T,Ohsumi Y.Apg10p,a novel protein-conjugating enzyme essential for autophagy in yeast.EMBO J1999;18:5234–41.[29]Kuma A,Mizushima N,Ishihara N,Ohsumi Y.Formation of the∼350kDa Apg12–Apg5·Apg16multimeric complex,mediated by Apg16oligomerization,is essential for autophagy in yeast.J Biol Chem2002;277:18619–25.[30]Kirisako T,Ichimura Y,Okada H,Kabeya Y,Mizushima N,Yoshi-mori T,et al.The reversible modification regulates the membrane-binding state of Apg8/Aut7essential for autophagy and the cyto-plasm to vacuole targeting pathway.J Cell Biol2000;151:263–75.[31]Ichimura Y,Kirisako T,Takao T,Satomi Y,Shimonishi Y,IshiharaN,et al.A ubiquitin-like system mediates protein lipidation.Nature 2000;408:488–92.[32]Matsuura A,Tsukada M,Wada Y,Ohsumi Y.Apg1p,a novel pro-tein kinase required for the autophagic process in Saccharomyces cerevisiae.Gene1997;192:245–50.[33]Straub M,Bredschneider M,Thumm M.AUT3,a serine/threoninekinase gene,is essential for autophagocytosis in Saccharomyces cere-visiae.J Bacteriol1997;179:3875–83.[34]Suzuki K,Kirisako T,Kamada Y,Mizushima N,Noda T,Ohsumi Y.The pre-autophagosomal structure organized by concerted functions of APG genes is essential for autophagosome formation.EMBO J 2001;20:5971–81.[35]Kihara A,Noda T,Ishihara N,Ohsumi Y.Two distinct Vps34phosphatidylinositol3-kinase complexes function in autophagy and carboxypeptidase Y sorting in Saccharomyces cerevisiae.J Cell Biol 2001;152:519–30.[36]Mizushima N,Sugita H,Yoshimori T,Ohsumi Y.A new pro-tein conjugation system in human.The counterpart of the yeast Apg12p conjugation system essential for autophagy.J Biol Chem 1998;273:33889–92.236Y.Ohsumi,N.Mizushima/Seminars in Cell&Developmental Biology15(2004)231–236[37]Tanida I,Tanida-Miyake E,Ueno T,Kominami E.The human ho-molog of Saccharomyces cerevisiae Apg7p is a protein-activating enzyme for multiple substrates including human Apg12p,GATE-16, GABARAP,and MAP-LC3.J Biol Chem2001;276:1701–6. [38]Mizushima N,Yoshimori T,Ohsumi Y.Mouse Apg10as an Apg12conjugating enzyme:analysis by the conjugation-mediated yeast two-hybrid method.FEBS Lett2002;532:450–4.[39]Nemoto T,Tanida I,Tanida-Miyake E,Minematsu-Ikeguchi N,Yokota M,Ohsumi M,et al.The mouse APG10homologue,an E2-like enzyme for Apg12p conjugation,facilitates MAP-LC3mod-ification.J Biol Chem2003;278:39517–26.[40]Mann SS,Hammarback JA.Molecular characterization of light chain3.J Biol Chem1994;269:11492–7.[41]Sagiv Y,Legesse-Miller A,Porat A,Elazar Z.GATE-16,a membranetransport modulator,interacts with NSF and the Golgi v-SNARE GOS-28.EMBO J2000;19:1494–504.[42]Wang H,Bedford FK,Brandon NJ,Moss SJ,Olsen RW.GABA A-receptor-associated protein links GABA A receptors and the cytoskeleton.Nature1999;397:69–72.[43]Kabeya Y,Mizushima N,Ueno T,Yamamoto A,Kirisako T,Noda T,et al.LC3,a mammalian homologue of yeast Apg8p,is localized in autophagosome membranes after processing.EMBO J2000;19:5720–8.[44]Chen L,Wang H,Vicini S,Olsen W.The␥-aminobutyric acidtype A(GABA A)receptor-associated protein(GABARAP)promotes GABA A receptor clustering and modulates the channel kinetics.Proc Natl Acad Sci USA2000;97:11557–62.[45]Kneussel M,Haverkamp S,Fuhrmann JC,Wang H,Wassle H,Olsen RW,et al.The␥-aminobutyric acid type A receptor (GABA A R)-associated protein GABAPAR interacts with gephyrin but is not involved in receptor anchoring at the synapse.Proc Natl Acad Sci USA2000;97:8594–9.[46]Elazar Z,Scherz-Shouval R,Shorer H.Involvement of LMA1and GATE-16family members in intracellular membrane dynamics.Biochim Biophys Acta2003;1641:145–56.[47]Scherz-Shouval R,Sagiv Y,Shorer H,Elazar Z.The COOH terminusof GATE-16,an intra-Golgi transport modulator,is cleaved by the human cysteine protease HsApg4A.J.Biol.Chem.2003;278:14053–58.[48]Hemelaar J,Lelyveld VS,Kessler BM,Ploegh HL.A single pro-tease,Apg4B,is specific for the autophagy-related ubiquitin-like proteins GATE-16,MAP1-LC3,GABARAP,and Apg8L.J.Biol.Chem.2003;278:51841–50.[49]Tanida I,Tanida-Miyake E,Komatsu M,Ueno T,Kominami E.Hu-man Apg3p/Aut1p homologue is an authentic E2enzyme for multiple substrates,GATE-16,GABARAP,and MAP-LC3,and facilitates the conjugation of hApg12p to hApg5p.J Biol Chem2002;277:13739–44.[50]Mizushima N,Yamamoto A,Hatano M,Kobayashi Y,KabeyaY,Suzuki K,et al.Dissection of autophagosome formation us-ing Apg5-deficient mouse embryonic stem cells.J Cell Biol 2001;152:657–67.[51]Seglen PO,Gordon PB.3-Methyladenine:specific inhibitor of au-tophagic/lysosomal protein degradation in isolated rat hepatocytes.Proc Natl Acad Sci USA1982;79:1889–92.[52]Petiot A,Ogier-Denis E,Blommaart EF,Meijer AJ,Codogno P.Distinct classes of phosphatidylinositol3 -kinases are involved in signaling pathways that control macroautophagy in HT-29cells.J Biol Chem2000;275:992–8.[53]Mizushima N,Kuma A,Kobayashi Y,Yamamoto A,Matsubae M,Takao T,et al.Mouse Apg16L,a novel WD-repeat protein,targets to the autophagic isolation membrane with the Apg12–Apg5conjugate.J Cell Sci2003;116:1679–88.。