HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 16, 13739-13744, April 19, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/16/13739    most recent M200385200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Tanida, I. Articles by Kominami, E. Search for Related Content PubMed PubMed Citation Articles by Tanida, I. Articles by Kominami, E. Social Bookmarking                      What's this?

Human Apg3p/Aut1p Homologue Is an Authentic E2 Enzyme for Multiple Substrates, GATE-16, GABARAP, and MAP-LC3, and Facilitates the Conjugation of hApg12p to hApg5p^*

Isei Tanida, Emiko Tanida-Miyake, Masaaki Komatsu, Takashi Ueno, and Eiki Kominami

From the Department of Biochemistry, Juntendo University School of Medicine, 2-1-1, Hongo, Bunkyo-ku, Tokyo 113-8421, Japan

Received for publication, January 14, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Autophagy is a process of bulk degradation of cytoplasmic components by the lysosome/vacuole and has a significant relationship to several neurodegenerative disorders and myopathies in mammals. One of APG gene products essential for autophagy in yeast, Apg3p, is a protein-conjugating enzyme for Apg8p lipidation (Ichimura, Y., Kirisako, T., Takao, T., Satomi, Y., Shimonishi, Y., Ishihara, N., Mizushima, N., Tanida, I., Kominami, E., Ohsumi, M., Noda, T., and Ohsumi, Y. (2000) Nature 408, 488-492). In this study, the cloning of a human Apg3p homologue (hApg3p) as an E2 enzyme essential for human Apg8p homologues (i.e. GATE-16, GABARAP, and MAP-LC3) is shown, and its unique characteristics are described. The predicted amino acid sequence of the isolated clone shows 34.1% identity and 48.1% similarity to yeast Apg3p. Site-directed mutagenesis revealed that Cys²⁶⁴ of hApg3p is an authentic active-site cysteine residue essential for the formation of hApg3p·hApg8p homologue intermediates. Overexpression of hApg7p enhances the formation of a stable E2-substrate complex between hApg3p^C264S and each of the hApg8p homologues, and MAP-LC3 is preferred as the substrate over the other two Apg8p homologues. These results indicate that hApg3p is an E2-like enzyme essential for three human Apg8p homologues. Co-immunoprecipitation of hApg7p with hApg3p indicates that hApg3p forms an E1·E2 complex with hApg7p as in the case of yeast Apg3p and Apg7p. Furthermore, hApg3p coimmunoprecipitates with hApg12p, and the overexpression of hApg3p facilitates the formation of the GFPhApg12p·hApg5p conjugate, suggesting that hApg3p cross-talks with the hApg12p conjugation system.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Autophagy is a process of bulk degradation of cytoplasmic components by the lysosomal/vacuolar system (1-3). In the initial step of macroautophagy, a cup-shaped membrane sac surrounds cytosolic components to form an autophagosome (4, 5), and the outer membrane of the autophagosome fuses with a lysosome/vacuole. The APG and AUT genes (autophagy or autophagy), which play indispensable roles in autophagy, have been identified and characterized in yeast, Saccharomyces cerevisiae (6, 7). The analyses of the gene products revealed two modifier-conjugation systems (the modifiers are Apg12p and Apg8p/Aut7p) that are essential for autophagy and the Cvt¹ pathway, and these modifiers have been shown to be processed by an enzymatic system similar to ubiquitylation (8-12, for review see Refs. 1, 3, 13, and 14).

In the yeast S. cerevisiae, Apg8p/Aut7p is unique among ubiquitin and other modifiers (9). Unlike the other modifiers, the target of Apg8p is a lipid, not a protein. Apg8p is activated by an E1 enzyme, Apg7p, and transferred to an E2 enzyme, Apg3p/Aut1p (9, 10, 15). In the last step, Apg8p is conjugated to phosphatidylethanolamine on a preautophagosomal membrane sac (9). After the formation of the autophagosome, luminal Apg8p is released into the luminal space and degraded in the vacuole (12, 16). At the same time, Apg8p on the cytosolic surface of the autophagosome is detached into the cytosol by Apg4p, a cysteine protease that is also essential for the activation of Apg8p (9). Apg8p/Aut7p also interacts with two ER-to-Golgi vacuole SNAREs (Bet1p and Sec22p) and vacuolar target SNARE and vacuole SNARE (Vam3p and Nyv1p) (17). For the reaction, the carboxyl-terminal Gly in Apg8p is essential as in ubiquitin and other modifiers. Furthermore, in addition to the Apg8p-lipidation system, the Apg12p-conjugation system is also essential for the Cvt pathway and autophagy (8). Apg12p is activated by the common E1 enzyme, Apg7p, transferred to a second E2 enzyme, Apg10p, and finally conjugates with Apg5p (8, 10, 11). The carboxyl-terminal Gly of Apg12p is also essential for conjugation. After the formation of the Apg12p·Apg5p conjugate, Apg16p attaches to Apg5p, forming an Apg12p·Apg5p·Apg16p complex for autophagy (18). Apg5p and the Apg12p·Apg5p conjugate localize to the membrane fraction but not the autophagosome, suggesting that they play a role in an initial step in the formation of autophagosomes and Cvt vesicles (8).

Unlike other modifier-conjugation systems, two conjugation systems play indispensable roles in the formation of membrane structures including autophagosomes and Cvt vesicles. Apg7p is a unique E1 enzyme for two substrates in two independent modification systems of autophagy and the Cvt-pathway and exists as a homodimer (19). More interestingly, Apg3p/Aut1p forms an E1·E2 enzyme complex with Apg7p, and Apg3p also interacts with Apg12p, which is a substrate for Apg7p and Apg10p but not Apg3p (19, 20). These results suggest that the E1·E2 complex is a key to the cooperative regulation of two modification systems.

In mammalian cells, two modification systems seem to be conserved. A human Apg12p homologue (hApg12p) conjugates with the human Apg5p homologue (hApg5p), which was first identified as an apoptosis-specific protein (21, 22). Recently, experiments using embryonic stem cells that knocked out the mouse APG5 gene demonstrated that a murine Apg5p homologue is essential for autophagy (23). With regard to mammalian Apg8p modification, there are three mammalian Apg8p/Aut7p homologue candidates, the Golgi-associated ATPase enhancer of 16 kDa (GATE-16), GABA_A receptor-associated protein (GABARAP), and microtubule-associated protein light chain 3 (MAP-LC3) (16, 24-27). GATE-16 interacts with N-ethylmaleimide-sensitive fusion protein and the 28-kDa Golgi SNARE protein (26). The mRNA for GATE-16 is expressed ubiquitously but at significantly higher levels in brain tissue. GABARAP interacts with GABA_A receptors, the cytoskeleton, and gephyrin, suggesting its functional importance in brain or neuronal cells (24, 28, 29). MAP-LC3 co-polymerizes with tubulin and is a component of the MAP-1 complex, which is composed of light chains 1, 2, and 3 and heavy chains (30, 31). Rat MAP-LC3 is localized on autophagosomal membranes, suggesting that rat MAP-LC3 is also a functional Apg8p homologue (25). These results suggest that mammalian Apg8p homologues have divergent functions in mammalian cells, especially in neuronal cells. Recently, we showed that the human Apg7p homologue (hApg7p) is an E1 enzyme essential for multiple substrates such as hApg12p, GATE-16, GABARAP, and MAP-LC3 and forms a homodimer (27). Considering the functional divergence of GATE-16, GABARAP, and MAP-LC3, it is assumed that a regulatory system for the conjugation system exists in mammalian cells. To reveal the three mammalian Apg8p homologue-modification systems, we are interested in the human Apg3p homologue (hApg3p). In this paper, we report the isolation and characterization of hApg3p as an E2 enzyme for mammalian Apg8p homologues. Furthermore, we showed a functional relationship between hApg3p- and the hApg12p-conjugation system.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Strains, Media, Materials, and Molecular Biological Techniques-- Escherichia coli strain DH5 cells, the host for plasmids and protein expression, were grown in Luria Broth medium in the presence of antibiotics as required (32). pGEM-T was purchased from Promega (Madison, WI), pCMV-Tag2B was from Stratagene, pEGFP-C1 and pIRES were from CLONTECH, and pGEX4T-1 was from Amersham Biosciences. A membrane blotted poly(A) RNA derived from human tissues for Northern analysis was purchased from CLONTECH, and Northern analysis was performed according to the manufacturer's protocol using the cDNA of the open reading frame of hAPG3 as a probe.

Cloning the cDNA of the Human APG3 Homologue-- Based on the DNA sequence of two EST clones (GenBank^TM accession numbers AI830763 and AI857615), two oligonucleotides were synthesized (hAPG3-GSP1, 5'-AAGTTCTCCCCCTCCTTCTG-3', and hAPG3-GSP2, 5'-TGCCGTTGCTCATCATAGCC-3'). Using these primers, 5'-RACE was performed by high fidelity PCR with human brain Marathon^TM-ready cDNA (normal whole brain from a 50-year-old Caucasian male) as a template according to the manufacturer's protocol (CLONTECH).

Plasmid Construction and Site-directed Mutagenesis-- Based on the obtained DNA sequence of the human APG3 homologue, we amplified an open reading frame of the human APG3 cDNA by PCR with high fidelity introducing a BglII site before the start codon, and a SalI site after the termination codon cloned the fragment into pGEM-T and designated the resultant plasmid as pGEMhAPG3. To express GFPhApg3p under the control of the cytomegalovirus promoter, a BglII-SalI fragment of the pGEMhAPG3 plasmid was introduced into a pEGFP-C1 vector (CLONTECH) and designated pEGFPhAPG3. To express FLAGhApg3p in COS7 cells, a BglII-SalI fragment of the pGEMhAPG3 plasmid was introduced into the BamHI-SalI site of pCMV-Tag2B vector (Stratagene) and designated pTag2BhAPG3. Mammalian expression vectors for each of the enhanced GFP modifier fusion proteins (EGFPhApg12p, EGFPhMAP-LC3, EGFPhGATE-16, and EGFPhGABARAP) have been described previously (27). To express both hApg7p and enhanced GFP modifier fusion proteins, we inserted the internal ribosome entry site sequence between the DNA sequences of each of the enhanced GFP modifiers and hApg7p using the pIRES vector. Cys²⁶⁴ within hApg3p was replaced by Ser, mutagenized by the Gene-Editor in vitro site-directed mutagenesis system (Promega) with an oligonucleotide (hAPG3CS, 5'-ATGTGTTCAGTTCACCCAAGCAGGCATGCTGA-3') according to the manufacturer's protocol. The expression plasmid for mutant hApg3p^C264S was constructed as in the case of pEGFP-hAPG3 and designated pEGFPhAPG3CS.

Antibodies and Immunoprecipitation-- A polyclonal antibody against a synthetic polypeptide corresponding to residues 550-571 of hApg7p has been described previously (27). For the preparation of antiserum against hApg3p, rabbits were immunized with a glutathione S-transferase-hApg3p fusion protein. The resultant antiserum against hApg3p recognized a recombinant green fluorescent protein (GFP)-hApg3p fusion protein in COS7 cells. The polyclonal and monoclonal anti-GFP antibodies were purchased from CLONTECH. The monoclonal anti-FLAG antibody (M2) was purchased from Sigma. Co-immunoprecipitation of interacting proteins has been described previously (27).

Expression of Human Apg Homologues in COS7 and HEK293 Cells-- COS7 and HEK293 cells were maintained in Dulbecco's modified Eagle's medium containing 10% fetal calf serum. For transfection, 2 × 10⁵ cells were seeded on 60-mm dishes. After incubation for 24 h at 37 °C, the cells were transfected with a mixture of 1 µg of plasmid DNA and 15 µl of FuGENE 6 (Roche Diagnostics). For co-transfection, 1 µg of each plasmid was used. The transfectant was harvested after incubation for an additional 48 h.

Other Techniques-- The subcellular fractionation of HEK293 cells has been described by Kabeya et al. (25).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Isolation of a Human Apg3p/Aut1p Homologue That Is Expressed Ubiquitously in Human Tissues-- To investigate factors in the modification system for mammalian hApg8p homologues, we isolated a cDNA to hApg3p. A BLAST search of the EST data base with the amino acid sequence of yeast Apg3p indicated candidate hApg3p homologues (GenBank^TM accession numbers AI830763 and AI857615). Based on the DNA sequences of two EST clones, a cDNA to hApg3p was amplified by 5'-RACE with a human brain cDNA library as a template. The DNA sequence of each of the four isolated clones contains a single open reading frame and a 5' sequence identical to that of another EST clone (GenBank accession number AW408464). The predicted amino acid sequence of the isolated clone (calculated molecular mass of 35.8 kDa) shows 34.1% identity and 48.1% similarity to yeast Apg3p, and the region corresponding to the predicted active-site cysteine residue is significantly conserved between the isolated clones and yeast Apg3p (Fig. 1A). Therefore, we conclude that the clone is a cDNA to hApg3p.

View larger version (51K): [in this window] [in a new window]   Fig. 1.   Homology and tissue distribution of the human Apg3p homologue. A, a comparison of the amino acid sequences of the human Apg3p homologue with that of yeast Apg3p. The amino acid sequence of hApg3p is compared with that of yeast Apg3p by the ClustalW program. Asterisks, identical amino acids; dots, similar amino acids. The region containing the predicted active-site cysteine is underlined. B, Northern analysis of human APG3 mRNA in human tissues. A DNA fragment of an open reading frame of the human APG3 homologue was used as a probe.

To investigate the expression of the human APG3 mRNA in human tissues, we performed Northern blot analysis using a multiple tissue-specific mRNA blot and a DNA fragment encoding the open reading frame of hApg3p. The human APG3 mRNA is expressed ubiquitously in all human tissues examined with especially high levels of expression in heart, skeletal muscle, kidney, liver, and placenta (Fig. 1B).

The Human Apg3p Homologue Is an Authentic E2-like Protein-conjugating Enzyme for Three Human Apg8p Homologues, MAP-LC3, GATE-16, and GABARAP-- If the isolated clone encodes an authentic E2 enzyme essential for hApg8p homologues, its gene product, hApg3p, will form E2·hApg8p intermediates. To investigate this possibility, we employed site-directed mutagenesis of the predicted active-site cysteine residue within hApg3p (Fig. 1A, Cys²⁶⁴). Wild type hApg3p forms an enzyme/substrate intermediate via a thiol ester bond. Because of the rapid turnover of the Apg3p reaction, it is difficult to recognize such an intermediate in sufficient quantity. If the active-site cysteine residue of Apg3p is replaced by serine, a stable O-ester bond instead of a thiol ester bond will be formed between the enzyme and substrate(s), and the formation of a high molecular mass Apg8p·Apg3p intermediate will be clearly detected in immunoblots as demonstrated previously with hApg7p (27). We changed Cys²⁶⁴ within hApg3p to serine by site-directed mutagenesis and expressed both the mutant GFPhApg3p^C264S and each GFPhApg8ps (GFPhGATE-16, GFPhGABARAP, and GFPhMAP-LC3) in COS7 cells (Fig. 2). Wild type GFPhApg3p, mutant GFPhApg3p^C264S, and GFP-hApg8ps were well expressed in COS7 cells; however, contrary to our expectations, no high molecular mass band corresponding to the predicted intermediate was recognized (Fig. 2, Short Exposure and Long Exposure, lanes 2, 4, and 6). We reasoned that because of the sparse content of endogenous Apg7p, the overexpressed Apg8p homologues could not be activated efficiently and could not be subsequently transferred to hApg3p. When hApg7p was expressed together with mutant GFPhApg3p^C264S and GFPhApg8ps, higher molecular mass bands consistent with stable GFPhApg8p·GFPhApg3p^C264S intermediates (~105 kDa) were recognized by immunoblot with anti-GFP antibody (Fig. 2, Long Exposure, lanes 8, 10, and 12). hMAP-LC3 preferentially forms an intermediate with hApg3p^C264S in COS7 cells (Fig. 2, Short Exposure, lane 12). The interactions of hApg3p with hApg8ps were also confirmed by co-immunoprecipitation (data not shown). These results indicate that hApg3p is an authentic E2 enzyme essential for hGATE-16, hGABARAP, and hMAP-LC3, and that the activation of the hApg8p homologue by hApg7p is essential for a further reaction mediated by hApg3p.

View larger version (43K): [in this window] [in a new window]   Fig. 2.   hApg7p-dependent formation of hApg3pC264S·hApg8p intermediates. Wild type GFPhApg3p (hApg3p, Wt) or mutant GFPhApg3pC264S (CS) was expressed together with GFPhGATE-16, GFPhGABARAP, or GFPhMAP-LC3 in COS7 cells. GFP fusion proteins in the cell lysates were recognized by immunoblot with anti-GFP antibody. hApg7p was recognized by immunoblot with anti-hApg7p antibody. Long exposure is 10 times longer than Short exposure. Asterisk indicate nonspecific bands.

To investigate the intracellular localization of hApg3p, we performed subcellular fractionation. HEK293 cells expressing GFPhApg3p and FLAGhMAP-LC3 were lysed and fractionated by ultracentrifugation at 100,000 × g for 1 h. Total proteins in the supernatant and pellet were analyzed by SDS-PAGE, and GFPhApg3p was recognized by immunoblot with anti-GFP antibody (Fig. 3, WB: anti-GFP). GFPhApg3p fractionated mainly in the supernatant, and the FLAGhMAP-LC3·GFPhApg3p^C264S intermediate also fractionated in the supernatant (Fig. 3, WB: anti-GFP, LC3-GFPhApg3p). The results suggest that hApg3p is present in the cytosol, and that the reaction of hMAP-LC3 mediated by hApg3p occurs predominantly in the cytosol.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Human Apg3p reacts with hMAP-LC3 in the cytosol. FLAGhMAP-LC3 and GFPhApg3p wild type (Wt), GFPhApg3pC264S mutant (CS), and vector control () were expressed in HEK293 cells. The cell lysate was fractionated by ultracentrifugation at 100,000 × g for 1 h. GFPhApg3p and FLAGhMAP-LC3 in the resultant supernatant and pellet were recognized by immunoblot with anti-GFP (WB: anti-GFP) and anti-FLAG (WB: anti-FLAG) antibodies, respectively.

The Human Apg3p Homologue Forms an E1·E2 Complex with the Human Apg7p Homologue-- In yeast, Apg3p forms an E1·E2 complex with Apg7p, which is one of its unique characteristics compared with other protein-conjugation systems. To investigate whether hApg3p interacts with hApg7p as in the case of yeast, co-immunoprecipitation was performed. We expressed FLAGhApg3p and hApg7p in COS7 cells (Fig. 4, Expression, lanes 1-3), and FLAGhApg3p in the lysate of the transfectant was immunoprecipitated well with anti-hApg3p antibody (Fig. 4, IP: anti-hApg3, WB: anti-FLAG, lanes 1 and 3). When both FLAGhApg3p and hApg7p were expressed in the cells, hApg7p co-immunoprecipitated with FLAGhApg3p by the anti-hApg3p antibody (Fig. 4, IP: anti-hApg3, WB: anti-hApg7, lane 3). When hApg7p alone was expressed in the cells, hApg7p was not immunoprecipitated by the anti-hApg3p antibody (Fig. 4, IP: anti-hApg3, WB: anti-hApg7, lane 2). The co-immunoprecipitation of hApg7p with hApg3p was confirmed using another hApg3p fusion protein (Fig. 4, GFPhApg3p, lanes 4-6). When GFPhApg3p was expressed in COS7 cells, GFPhApg3p in the lysate immunoprecipitated with anti-hApg3p antibody, whereas GFP itself did not immunoprecipitate with this antibody (Fig. 4, Expression, IP: anti-hApg3, WB: anti-GFP, lanes 4-6). Only when both GFPhApg3p and hApg7p were co-expressed in COS7 cells did hApg7p co-immunoprecipitate with GFPhApg3p using the anti-hApg3p antibody (Fig. 4, IP: anti-hApg3, WB: anti-hApg7, lane 6).

View larger version (34K): [in this window] [in a new window]   Fig. 4.   hApg3p interacts with hApg7p to form an E1·E2 complex. Plasmids pTag2BhAPG3, pGFPhAPG3, and/or pCMVhAPG7 were transfected into COS7 cells to express FLAGhApg3p, GFPhApg3p, and/or hApg7p, respectively. pCMV-Tag2B and pEGFP-C1 were used as controls. When pGFPhAPG3 and pTag2BhAPG3 were transfected into COS7 cells, the respective GFPhApg3p and FLAGhApg3p were expressed well (see Expression, WB: anti-hApg3p, anti-FLAG, anti-GFP). When pCMVhAPG7 was transfected into COS7 cells, hApg7p was also expressed well (see Expression, WB: anti-hApg7p). After both FLAGhApg3p and hApg7p were expressed in COS7 cells, FLAGhApg3p in the cell lysate was immunoprecipitated with anti-hApg3p antibody, and hApg7p co-immunoprecipitated with FLAGhApg3p (see IP: anti-hApg3p, WB: anti-FLAG, anti-hApg7, lane 3) and vice versa (see IP: anti-hApg7p, WB: anti-FLAG, anti-hApg7). When both GFPhApg3p and hApg7p were expressed in COS7 cells, hApg7p also co-immunoprecipitated with GFPhApg3p (see IP: anti-hApg3p, WB: anti-GFP, anti-hApg7, lane 6) and vice versa (see IP: anti-hApg7p, WB: anti-GFP, anti-hApg7).

Finally, we confirmed the interaction using an anti-hApg7p antibody. When hApg7p was expressed in COS7 cells, the hApg7p immunoprecipitated well with anti-hApg7p antibody (Fig. 4, IP: hApg7, WB: anti-hApg7, lanes 2, 3, 5, and 6). When FLAG-hApg3p or GFP-hApg3p was co-expressed with hApg7p in COS7 cells, they co-immunoprecipitated with hApg7p using the anti-hApg7p antibody (Fig. 4, IP: anti-hApg7, WB: anti-FLAG, WB: anti-GFP, lanes 3 and 6). Considering these results, we conclude that hApg3p forms an E1·E2 complex with hApg7p as is the case in yeast.

The Human Apg3p Homologue Interacts with the Human Apg12p Homologue-- Another characteristic feature of Apg3p is that it also interacts with a second modifier protein, Apg12p, which is a substrate for Apg7p but not for Apg3p in yeast (19, 20). We then investigated the interaction of hApg3p with hApg12p in mammalian cells by co-immunoprecipitation. We expressed GFPhApg12p and FLAGhApg3p in COS7 cells (Fig. 5, Expression, WB: anti-FLAG, anti-GFP, lanes 1-3). When FLAGhApg3p was expressed in the cells, FLAGhApg3p in the cell lysate immunoprecipitated well with the anti-hApg3p antibody (Fig. 5, IP: anti-hApg3, WB: anti-FLAG, lanes 1 and 3). When FLAGhApg3p and GFPhApg12p were expressed together in the cells, GFPhApg12p co-immunoprecipitated with FLAGhApg3p (Fig. 5, IP: anti-hApg3, WB: anti-GFP, lane 3). These results indicate that hApg3p interacts with hApg12p.

View larger version (47K): [in this window] [in a new window]   Fig. 5.   hApg3p interacts with hApg12p. Plasmids pTag2BhAPG3 and pGFPhAPG12 were transfected into COS7 cells to express FLAGhApg3p and GFPhApg12p, respectively. pCMV-Tag2B and pEGFP-C1 were used as controls. When pTag2BhAPG3 was transfected into COS7 cells, FLAGhApg3p was expressed well in the cells (see Expression, WB: anti-FLAG). When pGFPhAPG12 was transfected into COS7 cells, GFPhApg12p was also expressed (see Expression, WB: anti-GFP). When both FLAGhApg3p and GFPhApg12p were expressed in COS7 cells, FLAGhApg3p in the cell lysate was immunoprecipitated with anti-hApg3p antibody, and GFPhApg12p co-immunoprecipitated with FLAGhApg3p (see IP: anti-hApg3p, WB: anti-FLAG, anti-GFP, lane 3).

The Overexpression of hApg3p Facilitates the Formation of the hApg12p·hApg5p Conjugate-- Recent analyses using murine APG5 gene-deficient embryonic stem cells revealed that the Apg12p·Apg5p conjugate cooperates sequentially with MAP-LC3 in the formation of a preautophagosomal membrane sac (23). As described in the previous section, hApg3p interacts with both hApg7p and hApg12p (Figs. 4 and 5). Considering these interactions, it is interesting to see whether the hApg7p·hApg3p complex and/or the hApg3p·hApg12p complex play some roles in the two conjugation reactions. As reported previously (27), when both hApg7p and GFPhApg12p were expressed in COS7 cells, the formation of the GFPhApg12p-hApg5p conjugate was recognized (Fig. 6, GFPhApg12p-Apg5p, lane 5). We next examined the effects of the overexpression of hApg3p, hApg12p, and hApg7p in conjugate formation. When GFPhApg3p was expressed together with hApg7p and GFPhApg12p, the amount of the hApg12p·hApg5p conjugate increased significantly (Fig. 6, GFPhApg12p-Apg5p conjugate, lane 7). The carboxyl-terminal Gly in hApg12p is reported to be essential for the formation of the conjugate (22). We constructed a mutant hApg12pG with a deletion of the carboxyl-terminal Gly. When mutant GFPhApg12pG was expressed in cells instead of GFPhApg12p, no conjugate was formed even in the presence of both hApg7p and hApg3p (Fig. 6, lanes 6 and 8). The enhanced conjugate formation is dependent on hApg7p, because when hApg7p was not expressed, the conjugate was not recognized even in the presence of overexpressed hApg3p and hApg12p (Fig. 6, lane 9). These results indicate that the overexpression of hApg3p in addition to hApg12p and hApg7p further facilitates the formation of the hApg12p·hApg5p conjugate.

View larger version (25K): [in this window] [in a new window]   Fig. 6.   Formation of the hApg12p·hApg5p conjugate is facilitated by the overexpression of hApg3p. GFPhApg12p wild type hApg12p (Wt) or mutant hApg12pG (G)] and/or GFPhApg3p were expressed in COS7 cells in the presence (+) or absence () of hApg7p. hApg12pG, the carboxyl-terminal Gly within hApg12p, which is essential for the formation of the hApg12p·hApg5p conjugate (GFPhApg12p-hApg5p), was deleted by site-directed mutagenesis (G). GFPhApg12p and GFPhApg3p were recognized by immunoblot with anti-GFP antibody and confirmed with anti-mApg12p and anti-hApg3p antibodies. hApg7p was recognized by immunoblot with anti-hApg7p antibody.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In this study, we show that the human Apg3p homologue is an authentic E2 enzyme for the hApg8p-conjugation system(s), and that human GATE-16, GABARAP, and MAP-LC3, the three hApg8p homologues, are substrates for hApg3p (Table I). We show that hApg3p^C264S in which the active-site cysteine is changed to serine can bind to GATE-16, GABARAP, and MAP-LC3 to form stable enzyme-substrate intermediates. The overexpression of hApg7p facilitates this reaction. Overexpressed hApg7p may be required for the efficient activation of overexpressed hApg8p homologues, which is necessary for the accumulation of substantial quantities of hApg3p·hApg8p intermediate via an O-ester bond. Alternatively, overexpressed hApg7p may enhance the formation of the E1·E2 complex with hApg3p, which may facilitate the sequential E2 reaction after the activation of hApg8p homologues by hApg7p. Furthermore, the overexpressed hApg3p facilitates the formation of hApg12p·hApg5p, whereas hApg3p is not an E2 enzyme for hApg12p (Table I).

There are two unique characteristics of hApg3p. 1) It forms an E1-E2 complex with hApg7p. 2) It interacts with hApg12p. More importantly, we showed for the first time that the overexpression of hApg3p together with hApg7p and hApg12p enhances the formation of the hApg12p-hApg5p conjugate. The overexpression of hApg12p and hApg3p in the presence of endogenous hApg7p did not cause an enhancement of conjugate formation. Thus, the enhancement appears to be attributed to the formation of the hApg7p-hApg3p (E1-E2) complex rather than the hApg12p-hApg3p complex. These results strongly suggest that the formation of a complex between hApg7p and hApg3p, two indispensable members of the hApg8p-conjugation system, also plays an important role in the hApg12p-conjugation system. Mizushima et al. (23) reported that the formation of the mApg12p·mApg5p conjugate precedes the lipidation and subsequent targeting of MAP-LC3 to autophagosomal precursors. Our data on the facilitation of the hApg12p-conjugation reaction by hApg3p indicate that there is intimate cross-talk between hApg12p-conjugation and the hApg8p-modification systems in the formation of autophagosomes. To our knowledge, this is the first observation of a cooperative relationship between the two different conjugation systems. In Fig. 7, we present a hypothetical scheme based on our experimental results. Although the precise mechanism by which the hApg7p·hApg3p complex facilitates hApg12p·hApg5p conjugation is still unknown, it is reasonable to assume that it may be possible by the activation of some step(s) in the hApg12p-conjugation pathway. For example, hApg7p complexed with hApg3p may be more active as an E1 enzyme than uncomplexed hApg7p. It is also possible that the presumptive hApg10p may be directly or indirectly involved in this mechanism. So far, authentic hApg10p has not been identified (Table I). We are now attempting to isolate and characterize an hApg10p homologue. In summary, hApg3p, which interacts with hApg12p on the one hand, forms an E1-E2 complex with hApg7p on the other and functions as a facilitating factor in the hApg12p-conjugation system in addition to an authentic E2 enzyme for hApg8p homologues. An enhanced level of the hApg12p-hApg5p conjugate in turn promotes the recruitment of the lipidated form of MAP-LC3 onto autophagosomal membranes. In the end, the activation of hApg12p-conjugation by the hApg7p-hApg3p complex promotes hApg8p-conjugation reaction. Thus, the two autophagic conjugation systems, which comprise the one activation enzyme (E1, hApg7p) in common, two distinct E2 enzymes (hApg3p and hApg10p) and four different modifiers (hApg12p and three hApg8p homologues) interact and cooperate with each other.

View larger version (33K): [in this window] [in a new window]   Fig. 7.   Working hypothesis of the facilitation of the hApg12p-conjugation system by the hApg7p·hApg3p (E1·E2) complex in autophagy. In the hApg12p-conjugation system, hApg12p is activated by hApg7p (E1 enzyme), transferred to hApg10p (hypothetic E2 enzyme), and finally conjugated to hApg5p. The hApg12p·hApg5p conjugate plays an indispensable role in the formation of the autophagosomal precursor. The second modification system is essential for the next step in the formation of the autophagosome. MAP-LC3 is activated by the same E1 enzyme, transferred to hApg3p, and finally conjugated to phosphatidylethanolamine. The modification of MAP-LC3 is essential for the formation of the preautophagosomal membrane. Thereafter, the cup-shaped preautophagosome is closed to form the autophagosome. The hApg7p·hApg3p (E1·E2) complex facilitates the formation of the hApg12p·hApg5p conjugate to promote the cooperation of the hApg12p-conjugation system with the MAP-LC3-modification system for the formation of the autophagosome.

GATE-16, which interacts with NSF and 28 kDa Golgi SNARE protein, is localized in the Golgi and is expressed in the largest amounts in brain (26). GABARAP is a GABA_A-receptor-associated protein that co-localizes with the GABA_A receptor in cultured cortical neurons and interacts with gephyrin (24, 28). MAP-LC3 is localized on autophagosomal membranes (25). Considering the divergent functions and intracellular localizations of the three Apg8p homologues, it is surprising that all three hApg8p homologues, MAP-LC3, GATE-16, and GABARAP, are substrates for hApg3p. The affinity of hApg3p for these substrates differs from that of hApg7p. hApg3p preferentially conjugates with MAP-LC3 in COS7 cells, whereas hApg7p conjugates almost equally with each of the hApg8p homologues. What are the implications of the difference in substrate specificity between the E1 and E2 enzymes? The higher affinity of hApg3p for MAP-LC3 is coincident with the autophagosomal localization of MAP-LC3 considering the function of hApg3p in autophagy (25). With regard to GATE-16 and GABARAP, it is possible that there is another Apg3p homologue specific for GATE-16 and/or GABARAP. But no further candidate was obtained from further BLAST search on the EST data base and 5'-RACE, and genomic Southern analysis of the murine genome suggests that there are no further Apg3p homologues.² Another possibility is that there is a cofactor or E3-like complex that determines the specificity for substrates. In COS7 cells, MAP-LC3 is expressed, whereas little GABARAP or GATE-16 is expressed.³ The expression of each of the hApg8p homologues in mammalian tissues and cultured cell lines is diverse, whereas hApg7p and hApg3p are expressed ubiquitously.⁴ In some cases of ubiquitylation, an E3 complex specifies the target protein. It is probable that there is a regulatory factor(s) that facilitates the conjugation of hApg3p with MAP-LC3 in COS7 cells. We are now investigating whether there is a regulatory factor or not by co-immunoprecipitation using anti-hApg3p antibody.

Considering the divergent functions and intracellular localizations of the three Apg8p homologues, MAP-LC3, GABARAP, and GATE-16, it is necessary to determine the final target of each hApg8p homologue and to investigate the regulatory system. Recent findings suggest that the Apg machinery plays an important role at least in brain and in cardiac and skeletal muscles. Clinical and biochemical analyses of a group of severe inheritable neurodegenerative disorders and X-linked vacuolar cardiomyopathy, myopathy in humans, and lysosome-associated membrane glycoprotein-deficient mice have also suggested that lysosomal degradation via autophagy occurs actively during neuronal development and in normal mammalian bodies (33-36). We will investigate these problem using murine APG3 and APG7 homologue knock-out mice.

	ACKNOWLEDGEMENTS

We thank Drs. Y. Ohsumi, T. Yoshimori, T. Noda, Y. Ichimura, K. Kirisako (National Institute for Basic Biology, Okazaki, Japan), N. Mizushima (PRESTO, Kawaguchi, Japan), and D. J. Klionsky (University of California, Davis, CA) for significant discussion and information, and Drs. K. Ishidoh, J. Ezaki, and D. Muno (Juntendo University, Tokyo, Japan) for helpful discussion. We also thank Margaret Dooley-Ohto for editing language of the manuscript.

	FOOTNOTES

^* This work was supported in part by Grants-in-aid 12780543 (to I. T.), 09680629 (to T. U.), and 12470040 (to E. K.) for Scientific Research, Grants-in-aid 12146205 for Scientific Research on Priority Areas from the Ministry of Education, Science, Sports and Culture of Japan, and The Science Research Promotion Fund from the Japan Private School Promotion Foundation (to E. K.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank^TM/EBI Data Bank with accession number AB079384.

To whom correspondence should be addressed. Tel.: 81-3-5802-1031; Fax: 81-3-5802-5889; E-mail: kominami@med.juntendo.ac.jp.

Published, JBC Papers in Press, February 1, 2002, DOI 10.1074/jbc.M200385200

² M. Komatsu, I. Tanida, T. Ueno, and E. Kominami, unpublished results.

³ I. Tanida and E. Kominami, unpublished results.

⁴ I. Tanida, T. Ueno, and E. Kominami, unpublished results.

	ABBREVIATIONS

The abbreviations used are: Cvt, cytoplasm-to-vacuole targeting; E1, protein-activating enzyme; E2, protein-conjugating enzyme; MAP-LC3, microtubule-associated protein light chain 3; FLAGhApg3p, FLAG-tagged human Apg3p/Aut1p homologue; FLAGhMAP-LC3, FLAG-tagged human MAP-LC3; EST, expressed sequence tag; GABA_A, -aminobutyric acid, type A; GABARAP, GABA_A receptor-associated protein; GATE-16, Golgi-associated ATPase enhancer of 16 kDa; GFP, green fluorescent protein; GFPhApg3p, GFP-tagged hApg3p; GFPhApg12p, GFP-tagged hApg12p; mApg12p, murine Apg12p homologue; SNARE, soluble NSF attachment protein receptors; RACE, rapid amplification of cDNA ends.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				Y.-s. Sou, S. Waguri, J.-i. Iwata, T. Ueno, T. Fujimura, T. Hara, N. Sawada, A. Yamada, N. Mizushima, Y. Uchiyama, et al. The Atg8 Conjugation System Is Indispensable for Proper Development of Autophagic Isolation Membranes in Mice Mol. Biol. Cell, November 1, 2008; 19(11): 4762 - 4775. [Abstract] [Full Text] [PDF]

				K. Oh-oka, H. Nakatogawa, and Y. Ohsumi Physiological pH and Acidic Phospholipids Contribute to Substrate Specificity in Lipidation of Atg8 J. Biol. Chem., August 8, 2008; 283(32): 21847 - 21852. [Abstract] [Full Text] [PDF]

				Z. Xie, U. Nair, and D. J. Klionsky Atg8 Controls Phagophore Expansion during Autophagosome Formation Mol. Biol. Cell, August 1, 2008; 19(8): 3290 - 3298. [Abstract] [Full Text] [PDF]

				A. Melendez and T. P. Neufeld The cell biology of autophagy in metazoans: a developing story Development, July 15, 2008; 135(14): 2347 - 2360. [Abstract] [Full Text] [PDF]

				N. Fujita, T. Itoh, H. Omori, M. Fukuda, T. Noda, and T. Yoshimori The Atg16L Complex Specifies the Site of LC3 Lipidation for Membrane Biogenesis in Autophagy Mol. Biol. Cell, May 1, 2008; 19(5): 2092 - 2100. [Abstract] [Full Text] [PDF]

				M. Koike, M. Shibata, M. Tadakoshi, K. Gotoh, M. Komatsu, S. Waguri, N. Kawahara, K. Kuida, S. Nagata, E. Kominami, et al. Inhibition of Autophagy Prevents Hippocampal Pyramidal Neuron Death after Hypoxic-Ischemic Injury Am. J. Pathol., February 1, 2008; 172(2): 454 - 469. [Abstract] [Full Text] [PDF]

				Y. Fujioka, N. N. Noda, K. Fujii, K. Yoshimoto, Y. Ohsumi, and F. Inagaki In Vitro Reconstitution of Plant Atg8 and Atg12 Conjugation Systems Essential for Autophagy J. Biol. Chem., January 25, 2008; 283(4): 1921 - 1928. [Abstract] [Full Text] [PDF]

				M. P. Taylor and K. Kirkegaard Modification of Cellular Autophagy Protein LC3 by Poliovirus J. Virol., November 15, 2007; 81(22): 12543 - 12553. [Abstract] [Full Text] [PDF]

				G. Marino, N. Salvador-Montoliu, A. Fueyo, E. Knecht, N. Mizushima, and C. Lopez-Otin Tissue-specific Autophagy Alterations and Increased Tumorigenesis in Mice Deficient in Atg4C/Autophagin-3 J. Biol. Chem., June 22, 2007; 282(25): 18573 - 18583. [Abstract] [Full Text] [PDF]

				Z.-W. Chen, C.-S. S. Chang, T. A. Leil, and R. W. Olsen C-Terminal Modification Is Required for GABARAP-Mediated GABAA Receptor Trafficking J. Neurosci., June 20, 2007; 27(25): 6655 - 6663. [Abstract] [Full Text] [PDF]

				E. Fass, E. Shvets, I. Degani, K. Hirschberg, and Z. Elazar Microtubules Support Production of Starvation-induced Autophagosomes but Not Their Targeting and Fusion with Lysosomes J. Biol. Chem., November 24, 2006; 281(47): 36303 - 36316. [Abstract] [Full Text] [PDF]